LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

AUG, 852005 t
x 3 ~

o

'41"; El

 

NOV 0 4 2007 e

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDquISS-sz

 

ELECTRON TRANSFER IN CYTOCHROME C
OXIDASE: CYTOCHROME C DOCKING &
ELECTRON/PROTON COUPLING

By

Jun Yang

A THESIS

Submitted to
Michigan State University
in partial firlfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Biochemistry and Molecular Biology

2002

all

Abstract

ELECTRON TRANSFER IN CYTOCHROME C OXIDASE:
CYTOCHROME C DOCKING & ELECTRON/PROTON COUPLING

By

Jun Yang

As the last complex in the respiratory chain, cytochrome c oxidase (CcO) reduces
02 to water and is a key player in, and regulator of, the electron transfer process. Docking
of its substrate, the electron donor cytochrome c (Cc), involves a large protein-protein
interface, but is a transient interaction. Certain positively charged lysine residues on Cc
and negatively charged carboxyl groups on CcO have been identified as critical to
successful docking and electron transfer. A computer model has been generated for the
Cc/CcO complex. To test the model, we have conducted research with engineered
disulfide bonds between Cc and CcO. The results show evidence of a covalent Cc/CcO
complex in support of the data from computer modeling. Further studies to characterize
the complex and use it to assist in crystallization are underway.

To determine where the products of the electron transfer reaction, protons and
waters, exit from the protein, mutants of Asp27l in subunit I were made. This residue
interacts with Arg234 from subunit II and is the first residue in a conserved sequence,
D(/N)PxxY, found to be important for signal transduction in G-protein coupled receptors.
The mutants show that the charge on Asp27l is important for holding the whole protein
together; hence the disaggregation of mutant forms of the oxidase prevented us from

analyzing the possible water/proton exit function of this residue.

To

My Family in China

iii

 

Shl
Yu
all:
CR.
me
Sint

Und

ACKNOWLEDGMENTS

To say that this thesis is “by Jun Yang” overstates the case. Without the
significant contribution from other people, this thesis would not exist.

At the top of the list are my advisor, Dr. Shelagh F ergson-Miller, my committee
members, and members of my research lab. Shelagh not only has offered me her special
scientific acumen, but also has tried her best to give me support and encouragement. Her
professionalism and considerate efforts have made a remarkable impact in my
professional and personal life. Also my committee members, Dr. Honggao Yan & Dr.
Zachary Burton, have been very informative and supportive by offering me their
scientific insights. I would like to extend a special thanks to Dr. Burton for reading
through my thesis and providing valuable suggestions.

No word can express my gratitude to Dr. Carrie Hiser and Dr. Denise Mills. They
showed extraordinary patience whenever I needed their help. I would like to thank Dr.
Yuejun Zhen for helping me start my project, and continuously giving me advice even
after he left the lab. I really enjoyed collaborating with Dr. Yasmin Hilmi on the
cytochrome c docking project. Thanks to Bryan Schmidt for conducting EPR analyses for
me and for improving my English and introducing me to American culture. I am
sincerely thankful for the assistance from Ling Qin & Namjoon Kim. Many
undergraduate students, including Erica Westley and Jamie Slater, have also helped

accomplish the experiments in this thesis.

iv

 

Oil l

unti

act:

\l'ar

 

I would like to thank Dr. Frank Millett at the University of Arkansas for his help
on having fast kinetics data acquired for the Y159 mutants and Dr. Jon Hosler for sharing
his important research results.

Special thanks are to my family back in China for their unconditional support and
understanding. Their love and sacrifice constitute an irreplaceable part of my
achievement.

Last, to my fi'iends, especially, Christine Mikkola, John Mikkola and Yuhui

Wang, for helping me through the hardest moments of my life.

TABLE OF CONTENTS

CHAPTER 1

 

LITURATURE REVIEW

Part I The Structure of CcO

 

 

 

I. The Respiratory Chain
11. Complex IV (CcO)

 

 

III. Heme-Copper Oxidase Family
IV. The Structure of (103 Oxidase

 

l. Subunit I

 

2. Subunit II

 

 

3. Subunit III
4. Other Subunits

 

5. How CcO Performs Its Function

 

 

5.1 Oxygen Reactions

 

5.2 Proton Pumping

\OOO\)\IO\

l6
16
17
17
18

20

 

Part II Dockigg of Cytochrome c on Its Redox Partners

1. Chemical Modification

 

II. Crystallographic Approaches

 

 

III. Mutagenesis Approaches

 

1.Cytochrome c Peroxidase
2.Cytochrome c Oxidase

 

 

IV. Computational Modeling Approaches

 

V. Summary

Part III c-Type Cytochromes in Rhodobacter sphaeroides

20
22
23
23
24
27
30

32

 

 

1. Properties of c-Type Cytochromes
11. Which Cytochrome c Is the Physiological
Substrate for aa3-type Oxidase in R. sphaeroides?

 

 

1. Cytochrome C554
2. Cytochrome C2

 

3. Membrane Bound Cytochrome c

 

 

3.1 Cytochrome C552
3.2 Cytochrome cy

 

 

III. Summary

CHAPTER 2
THE WATER AND/OR PROTON EXIT PATHWAY AND ITS

vi

32

33
35
35
39
40
41
43

 

p.—
l”
- 1

i":

 

 

 

RELATIONSHIP TO RESPIRATORY CONTROL

 

 

 

 

 

 

44

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Introduction 45
Materials and Methods 53
Materials 53
Construction of the D271 Mutants 54
Conjugation 54
Cell Growth 55
Preparation of Cytoplasmic Membranes 55
Ni2+ - NTA Purification of CcO 56
FPLC-DEAE Chromatography 56
Activity Assay 56
Spectrophotometn'c Analysis 57
Results 57
Design of Mutants 57
Construction of Mutants 59
Optical Spectral Characteristics of Mutants 59
The Activities of Mutant Enzymes 6O
SDS-Page Analysis 60
Discussion 73
Subunit III 73
The Lyle3 in Subunit III 74
What Forces Are Important in Holding the Protein Together? 74
Signal Motif 78
Disulfide Bonds 79
Conclusion 80
CHAPTER 3
THE CROSS-LINKIN G ANALYSIS OF CYTOCHROME C
AND CYTOCHROME C OXIDASE 81
Introduction 82
Materials and Methods 88
Materials 88
Polymerase Chain Reaction (PCR) 88
Construction of the Y159 and A213 Mutants 89
Purification of R. sphaeroides Cytochrome c; 89
Activity Assay 90
Fast Electron Transfer Measurements by Flash Photolysis 90
Electron Paramagnetic Resonance (EPR) Spectroscopy 91

 

vii

Reconstitution of Cytochrome c Oxidase 91

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Proton Pumping Assay 91
Cross-Linking Assay 92
Results 93
Design of Mutants 93
Optical Characteristics of Mutants 98
CuA/Heme EPR Analysis 101
the Activities of Mutant Enzymes 101
Laser Flash Photoreduction Study of Kinetics of Y159C Mutants ----------- 108
Proton-Pumping Assay 108
Complex Formation 112
Discussion 1 19
Are the Changes Introduced by the Mutation at Position
Y-159 Disruptive to the Structure and Function of CcO? 119
Cation-Pi Interaction 122
Covalently Bound Complex
and Anion Exchange Chromotography 123
Conclusion 124

 

 

BIBLIOGRAPHY 125

viii

 

 

LIST

C HA1

Table

C HA

Table

CHA

Tablt

 

LIST OF TABLES

 

 

CHAPTER 1
Table 1.1 Characters of the Cytochromes c in R. sphaeroides 34
CHAPTER 2
Table 2.1 The Activity Comparison of D271 Mutants

of Cytochrome c Oxidase from R. sphaeroides 61
CHAPTER 3

Table 3.1 Activity Characteristics of Y159 Mutants 109

 

ix

 

 

 

CH.

FIQL
V.

Fig

Figr

Figi

LIST OF FIGURES

CHAPTER 1

 

Figure 1.1 Respiratory Chain Complexes 4

Figure 1.2 Crystal Structure of Rhodobacter sphaeroides
Cytochrome c Oxidase and Its Metal Centers 10

 

Figure 1.3 Crystal Structure of Rhodobacter sphaeroides
Cytochrome c Oxidase and Its Subunits 13

 

Figure 1.4 Crystal Structure Comparison for Horse Cytochrome c (A)
and R. sphaeroides Cytochrome C2 (B) 37

 

CHAPTER 2

Figure 2.1 Crystal Structure of Rhodobacter sphaeroides Cytochrome c Oxidase
and the Possible H20 and/or H+ Exit Pathway 46

 

Figure 2.2 Possible H20 and/or H+ Exit Pathway
of Cytochrome c Oxidase 48

 

Figure 2.3 The Overall Structure of Helix IV
and Possible Proton Pathways 51

 

Figure 2.4 Visible Spectra for
Ni2+-NTA Purified D271N and D271H Mutants 62

 

Figure 2.5 Comparison Between Visible Spectra for Solubilized Membrane
and Ni2+-NTA Purified 027m Mutant 64

 

Figure 2.6 SDS-Page (14% Separation and 8% stacking)
Analysis of Ni2+-NTA Purified D271N and D271H Mutants -------- 67

Figure 2.7 FPLC Profile for D271N Purification
with Tandem DEAE-SPW Column 69

 

Figure 2.8 SDS-Page Analysis of
F PLC Purified D271N Fractions 71

 

Figure 2.9 Conformation Comparison of Possible H20 and/or H+ Exit
for R. sphaeroides (A) and
Bovine (B) Cytochrome c Oxidase 75

 

 

 

 

 

 

 

Fig;

Figu

 

 

FIE:

CHAPTER 3

Figure 3.1 The Interacting Lysine Residues from Horse Cc and
Carboxyl Residues from R. sphaeroides Cc0
in Proposed Cc/Cc0 complex 83

 

Figure 3.2 Space-Filling Models of Subunit II Soluble Domain
of C00 (A), Horse Cc (B), and R. sphaeroides C02 (C) -------------- 85

Figure 3.3 Overall Structure of
Proposed Horse Cc / R. sphaeroides Cc0 Complex 94

 

Figure 3.4 Interaction Interface of a Possible Complex of Horse Cc
and R. sphaeroides CcO and the Residues that Possibly
Form a Covalent Bond between Them 96

 

Figure 3.5 Tyrosine Alignment at the Possible Interface

 

 

 

 

 

 

of CcO with Its Redox Partner Cc 99
Figure 3.6 Absolute Visible Spectra for Reduced Y159 Mutants 102
Figure 3.7 Visible Spectra for A213 Mutants in Solubilized Membrane -------- 104
Figure 3.8 Cu)», EPR Spectra of Y159 Mutants 106
Figure 3.9 Ionic Strength Dependent Electron Transfer

from Ru-39-Cc to Cytochrome c Oxidase 110
Figure 3.10 Proton Pumping Analysis For Y159C Mutant 113
Figure 3.11 Proton Pumping Analysis For Y159T Mutant 115

Figure 3.12 Gel-filtration for Cytochrome c Oxidase Cysteine Mutant (Y159C)
with Cytochrome c Cysteine Mutant (N 12C) 117

 

Figure 3.13 Gel-filtration for Cytochrome c Oxidase Cysteine Mutant (Y159C)
with Cytochrome c Cysteine Mutant (N12C/Q13K) 120

 

xi

 

 

Cc
Cu
Ct‘i
Cd
CPI
Cq(
Dli.
ED
EPl
FPI
GP
L.\1
Ni:
PC

SD
T.\'
T.\

Cc
Cc(H)
Cc(Y)
CcO
CcP
Cq0
DEAE
EDTA
EPR
FPLC
GPCR
LM
NizI-NTA
PCR
R. s.
SDS
TMPD
TMS 7

LIST OF ABBREVIATIONS

Cytochrome c

Cytochrome c from horse heart

yeast Cytochrome c

Cytochrome c Oxidase

Cytochrome c Peroxidase

Quinol Oxidase

diethylaminoethyl
ethylenediarninetetraacetic acid

Electron Paramagnetic Resonance spectroscopy
fast protein liquid chromatography
G-Protein-Coupled Receptors

lauryl maltoside

Nickel-nitrilotriacetic acid

polymerase chain reaction

Rhodobacter sphaeroides

sodium dodecyl sulfate
N-N-N’N’-tetramethyl-p-phenylenediamine

TransMembrane Segment 7

xii

Chapter 1

Literature Review

 

 

act;

for

CDC?

elc.

“h;

In
‘ (4
*1

pro:
cor:
imn
oxid

eleci

 

As most of the living organisms are not able to live photosynthetically, they
acquire energy through metabolizing foodstuffs and storing the energy obtained in the
form of fat and carbohydrate. The energy from foodstuffs is changed to ATP, the major
energy form that living organisms can use directly. This process involves feeding
electrons from NADH into a chain of membrane-bound complexes, the respiratory chain,
where electrons are used to generate a transmembrane potential, which drives ATP
synthesis (Figure 1.1) (Mitchell, 1976; Mitchell & Moyle, 1965).

Cytochrome c oxidase (Cc0) is the last complex in the respiratory chain. It
provides the sink for the electrons from NADH by reducing oxygen to water. It also
contributes to the transmembrane potential that can drive the synthesis of ATP. The
immediate electron donor to CcO is cytochrome c. It must react with a specific site on the
oxidase to deliver electrons and then dissociate to pick up more because for every 02, 4
electrons are required.

My thesis has involved the study of mutant forms of cytochrome c oxidase from
Rhodobacter sphaeroides designed to address two questions: 1) Where does cytochrome
c dock on cytochrome c oxidase to deliver electrons and 2) where do the protons and
water, that are the products of the electron transfer reaction, exit from the protein? The
studies were aimed at testing the accuracy of a computational prediction of the docking
site, and the importance of a conserved sequence, D(N)PxxY, found in G-protein coupled
receptors and oxidase, in controlling proton and/or water exit.

To provide background for this work, I will review the following issues:

1. The structure of CcO

2. Docking of cytochrome c on its redox partners

3. C-type cytochromes in Rhodobacter sphaeroides

 

 

Figi
and

X Al

 

repr

proé

lBu
61 a

(..\'r

Figure 1.1 Respiratory Chain Complexes. The three-dimensional crystal structures of
complexes 11, III, IV, V and cytochrome c are shown. Because the crystal structure of
NADqubiquinone oxidoreductase (complex I) is not available, green boxes are used to
represent this complex. This figure was created by Dr. Denise Mills using the computer
program Rasmol with the coordinates from fumarate reductase (Complex II) (Iverson et
al., 1999), bovine bcl complex (Complex 111) (Iwata et al., 1998), horse cytochrome c
(Bushnell et al., 1990), R. sphaeroides cytochrome c oxidase (Complex IV) (Svensson-Ek
et al., 2002) and yeast ATP synthase (Complex V) (Stock et al., 1999).

(Note: Images in this thesis are presented in color.)

> “SE—=60 >— 53830 n: 53:80 = 5.950 _ 5.9.80

  
     

- G<Z +ED<Z
Swami—£82695

 

ON: .6 uEeEuSzU

mas—955 55:0 beta—Emom

 

 

 

elec
pot

C}IO

83m

( Ha
con
mitt
ene
5.01:

“"h:

Part I The Structure of C00

1. The Respiratory Chain

0f the three steps (glycolysis, citric acid cycle, oxidative phosphorylation) by
which the organism converts foodstuffs to ATP, oxidative phosphorylation produces the
most energy. In this step, electrons from NADH are first transferred to complex I
(N ADH:ubiquinone oxidoreductase or NADH dehydrogenase) and complex 11
(succinatezubiquinone oxidoreductase or succinate dehydrogenase). Then electrons are
further transferred to complex III (cytochrome bcl complex), from where Cc shuttles
electrons to complex IV (Cc0) to reduce oxygen. In this process, a transmembrane
potential is built up by pumping protons from the inside (matrix of mitochondria or
cytoplasm of bacteria) to the outside (interrnembrane space of mitochondria or
periplasmic space in bacteria). The resulting electrochemical potential drives ATP
synthesis.

The four complexes in the respiratory chain have been found in mitochondria
(Hatefi, 1985) and many bacteria (Trumpower & Gennis, 1994) (Figure 1.1). As the
complexes from prokaryotes are structurally and functionally homologous to their
mitochondrial counterparts but with simpler structures, they are widely used to study the
energy conservation in the respiratory chain. In our research, we use C00 from R.
sphaeroides as a model system for the mitochondrial Cc0, which has thirteen subunits,

while the bacterial form has only four.

 

 

 

 

11.

0x3.

Ill

oxi.

bin.

her:

cm»
of \\
Oxic
Oxid
TeQe

alSQ

and}

Emb

 

II. Complex IV (CcO)

Cc0 takes four electrons sequentially from reduced Cc and uses them to reduce
oxygen to water. The overall reaction catalyzed by Cc0 follows equation I:

4 Cc ( re'1+ ) + 02 + 8H+in —> 4 Cc ( Fe‘"+ ) + 2 H2O + 4 trout (Eq. I)
For each oxygen molecule reduced, four protons are used to form water and four others

are pumped out of the bacterial cytoplasm (or the mitochondria matrix).

III. Heme-Copper Oxidase Family

The terminal oxidases found in eukaryotes and many prokaryotes are heme-copper
oxidases. In these oxidases, all the subunits I are homologous and contain a heme-copper
binuclear center, where the oxygen molecules are reduced to water (Garcia-Horsman et
al., 1994a). The heme in this center is a high spin heme. There is usually another low spin
heme situated near to the high spin heme in subunit I as well.

Although mammalian mitochondria have only one type of oxidase, the M3-
cytochrome c oxidase (Cc0), bacteria normally have additional terminal oxidases, some
of which use ubiquinol instead of cytochrome c as substrate. In R. sphaeroides, three
oxidases have been found. They are My type oxidase (Hosler et al., 1992), cbb3-type
oxidase and bb3-quinol oxidase (Garcia-Horsman et al., 1994b; Yun et al., 1994). In the
recently compeleted genome sequence for R. s. , genes, caa3- and bd- type oxidases have
also been located (J 01 website).

All the heme-copper terminal oxidases catalyze the conversion of oxygen to water
and build up an electrochemical gradient across the membrane in which they are

embedded. While some oxidases, such as quinol oxidases (Cq0) use membrane-bound

quinol as their electron donor, some use soluble or membrane-bound Cc. Most heme
copper oxidases of the Cc0 and Cq0 type have been shown to pump protons.

So far, four types of heme groups have been identified in the heme-copper oxidase
family. They are A-, B-, C-, and 0-type hemes. All of these heme groups have the same
basic tetrapyrrole porphyrin rings in their structure with differences in the substituents on
the ring. No C-type heme has been found in the binuclear center. However, it is found in
caa3- and cbb3-type oxidases in which the covalently bound c-type hemes can transfer
electrons to the binuclear center (Garcia-Horsman et al., 1994b; Mather et al., 1993).
Besides the heme differences, some heme-copper oxidases have a Cu site in their
structures while some do not (Yun et al., 1994). This copper site, found in all CcO
oxidases, aa3-, caa3- and ba3-type CcO, is generally agreed to be the initial electron

acceptor from Cc.

IV. The structure of aa3 oxidase

Before bacterial oxidases were discovered, the aa3-type cytochrome c oxidase
from bovine heart was the one that was studied most thoroughly. It has 13 subunits with
the three largest ones highly homologous to the three largest ones from bacteria. Its
crystal structure has been resolved, as well as the four-subunit oxidases from P.
denitrificans and R. sphaeroides. The Cc0 from bacteria is also functionally similar to its
counterpart from mitochondria. This makes the bacterial CcO an ideal system to study
using molecular genetic techniques.

Although mammals and prokaryotes are evolutionarily distantly related, the

largest three subunits of the oxidase structures are almost identical in their secondary

 

 

SII

al.

thy

IESF
Peri
den;

431

 

structure and metal sites. The crystal structures confirm this similarity. I will discuss the
three largest subunits and their similarities between three crystal structures (Ostermeier et

al., 1997; Yoshikawa et al., 1998; Svensson-Ek etal., 2002). The residue numbering in

this section is based on the R. sphaeroides CcO.

l. Subunit I

Subunit I has twelve transmembrane helices, with five metal centers in it (Figure
1.2). These metal centers are a low spin heme a, a high spin heme a3, the CuB site, a Mg
site and a Ca site.

0f the five metal centers in Subunit I, heme a, heme a3 and CuB site are redox
active. The two hemes are in close proximity, with an edge to edge distance between the
two hemes of 4.7A, 4.7A and 5.6A for P. denitrificans, bovine, and R. sphaeroides,
respectively. The iron to iron distance is 13 to 14 A. The heme planes are almost
perpendicular to each other with the interplanar angles of 108° and 104° for P.
denitrificans and bovine oxidases. The low spin heme a is ligated by His-102 and His-
421 while the high spin heme a3 is ligated by His-419. The CuB site is close to heme a3
and they are strongly electronically coupled, usually making both of them EPR silent. The
distance between the heme a3 center and Curt is 4.5A, 4.5 A, 5.2 A for bovine, R.
sphaeroides and P. denitrificans respectively in the oxidized form of the enzymes. The

ligands of CuB are His-284, -333 and —334.

 

 

 

 

Fig

Its

 

Figure 1.2 Crystal Structure of Rhodobacter sphaeroides Cytochrome c Oxidase and
Its Metal Centers. Helices of subunit 1, II, III and IV are shown in blue, magenta, cyan
and orange respectively. Yellow balls represent CuA and CuB, orange for magnesium and
black for Calcium. The figure was created using Insight 11 with coordinates for R.

sphaeroides CcO from (Svensson-Ek et al., 2002).

 

 

 

 

 

0C:

Gl.

ox.
ox;

all:

a

mi:-

19(-

 

and

 

The non-redox active Mg site is situated at the interface of subunit I and II. It is
suggested to be important to stabilize the overall structure and modulate the redox
potentials of the neighboring metal centers (Florens et al., 1998). A role in the
proton/water exit channel is also proposed (Ferguson-Miller & Babcock, 1996). Mg has
octahedral coordination in which it is ligated to His-411 and Asp-412 from subunit I and
Glu254 from subunit 11, along with three water molecules. The Ca site was not found
until two to three years ago when the refined structures of P. denitrificans and bovine
oxidases became available. It is situated at the periplasmic side of subunit I with five
oxygen ligands from side chains and backbone. The metal to metal distances from this Ca
site to Cut, and the iron of heme a are about 18A for the bacterial oxidases. A shift in the
spectrmn of heme a has been observed with binding of calcium in both Paracoccus and
mitochondrial cytochrome aa3 (Saari et al., 1980; Mkrtchyan et al., 1990; Riistama et al.,

1999; Pfitzner et al., 1999).

2. Subunit II

The structure of subunit II is like a hammer with a big head (C-terrninal domain)
and a handle formed by two N-terminal transmembrane helices anchoring the subunit in
the membrane. The soluble C-terminal domain forming a globular “head” consists of ten
B-strands, similar to the blue copper protein structure (Figure 1.3). Many studies have
suggested that part of the negatively charged surface of this domain, close to the Cup, site,

is where Cc binds.

12

 

 

 

 

OTC:

5p};

 

Figure 1.3 Crystal Structure of Rhodobacter sphaeroides Cytochrome c Oxidase and
Its Subunits. Helices of subunit 1, II, III and IV are shown in blue, magenta, cyan and
orange respectively. Yellow balls represent Cu and Cug, orange for magnesium and
black for calcium. The figure was created using Insight 11 with coordinates for R.

sphaeroides Cc0 from (Svensson-Ek et al., 2002).

13

 

CuA is a binuclear center with the distances between the two copper atoms in the
range of 2.5 to 2.8A. This short distance implies strong electronic coupling. The two
copper atoms have a mixed-valence state [Cu(1.5)-Cu(l .5)] ligated by two histidine
residues (His-217 and -260), two shared cysteines (Cys-252 and -256), one methionine
(Met-263), and one backbone carbonyl oxygen of Glu-254. The carboxyl side chain of
Glu-254 is also ligated to the Mg site (Iwata et al., 1995; Tsukihara et al., 1995;
Wilmanns et al., 1995). This binuclear site is the electron acceptor from Cc heme.

Right around the Cu is a stretch of well-conserved aromatic residues (Tyr-14 l ,
Trp-l43, Tyr—l44, Trp-145, Tyr-147, and Tyr-149). In all the crystal structures, Trp143 is
in van der Waals contact with one of the Cut ligands, Met-263, and is also hydrogen-
bonded to His-217, which places it in a favorable position for possibly mediating electron
transfer from Cc to CuA. Mutations of other CuA ligands confirm the importance of the
dinuclear copper in electron transfer but not the proton pumping function (Zhen et al.,
2002). The detailed information about computational docking analysis is shown in
Section IV of Part II.

The Cu site and accompanying aromatics are surrounded by several negatively-
charged residues. These residues are suggested to be important for the binding of Cc on
the CcO from the mutagenesis, binding, and kinetics studies. The computational
modeling also substantiates the idea that these carboxyl residues (Glu—l48, Asp214,
Glu195 and Glu157) interact electrostatically with corresponding lysine residues on the

surface of Cc to make the transient complex for electron transfer.

15

 

 

 

4.(

anc

“l1

 

3. Subunit III

Subunit 111 forms a V shape with helices 1 and H forming one arm and helices III
to VII forming the other, with space in the middle occupied by phospholipid. In the space-
filling model of the oxidase, Glu-286, a conserved residue close to the Cu]; site, can be
seen through a tunnel in the subunit I from the cleft of the two arms. If a bulky group
blocks this channel, the oxygen diffusion is hindered (Riistama et al., 1996). This is
consistent with the structural analysis suggesting that subunit III and its associated lipid
may provide a reservoir for efficient oxygen diffusion from the membrane to the CuB site
(Lappalainen et al., 1995). When the gene of subunit 1H is deleted from R. sphaeroides,
the two-subunit oxidase is active and pumps protons (although with lowered
stoichiometry, i.e. < 1 H+/e'), but undergoes rapid “suicide” inactivation during turnover,
suggesting that subunit 111 may be important for the protein assembly and stability

(Bratton et al., 1999).

4. Other subunits

The fourth subunit has been resolved in the crystal structure of P. denitrificans
and R. sphaeroides Cc0 (Figure 1.3). This subunit is a one-transmembrane-helix subunit
which contacts mainly with tightly bound lipids in Rs Cc0 (Svensson-Ek et al., 2002). It
has no homology to the ten small subunits in the bovine oxidase but there is 56%
homology between the P. denitrificans and R. sphaeroides version (personal
communication with Dr. Carrie Hiser). Because deletion of this subunit from P.
denitrificans has no obvious effect, it may not play a role in the catalytic function (Witt &

Ludwig, 1997) but may be important for the assembly and stability for the whole enzyme.

16

SI."

51'...

ha

19

 

 

 

Clc.

The ten other small subunits of Cc0 from bovine heart surround the three core
subunits. Some subunits may be important for the dimer structure of the protein. A zinc
site has been found on the matrix side of the protein. Two possible ADP binding sites
have also been located, occupied by cholic acid in the bovine structure, consistent with
the result that ADP and ATP regulate allosterically the activity of the Cc0 (Arnold etal.,

1999).

5. How CcO performs its function
5.1 Oxygen reactions

The overall reaction catalyzed by Cc0 is shown in Equation I in Section II of Part
I. As oxygen has a high positive redox potential, reduction of oxygen is associated with a
large negative free energy. The energy is efficiently stored as a transmembrane
electrochemical proton gradient. The reduction of one 02 molecule requires four electrons
and four protons to generate two water molecules. Concomitantly, four protons are
pumped across the membrane.

The process of oxygen reduction involves electron transfer, oxygen bond
cleavage, and protonation. In Cc0, proton transfer is the rate-limiting step (Ferguson-
Miller & Babcock, 1996; Riistama et al., 1996, Karpefors et al., 1998). Protons are
apparently loaded at certain sites before an electron transfer event takes place. There is
still controversy about the oxygen reaction; however, a consensus of the reaction scheme
has begun to emerge (Proshlyakov, et al., 2000; Babcock & Wikstrfim, 1992). Oxygen
first binds to CuB when the heme a3 — Cug site of oxidase is in the reduced state. Then the

oxygen binds to heme 03 in an end-on configuration to form an initial oxy form, as in

17

 

 

O:

in

pd

h)

:Jl
l .4

mt"
prt“
SL1;
a L

.\1

LI.

 

COp

uh.

D05

an 4

k

 

oxygen binding to hemoglobin. Subsequently, the oxy form is reduced to the peroxy (“P”)
form which gives a characteristic absorbance at 608nm. The oxygen bond is already
broken in the P form in a 4-electron transfer event of which one electron is believed to be
from Tyr 280. An additional electron and one substrate proton are transferred to the
binuclear center to make the “F” form (absorbance 580nm) and one or two protons are
pumped across the membrane. Finally, an electron and proton is used to make the
hydroxy form and the cycle is complete. This step pumps protons as well, but the
stoichiometry of pumping at each step is still debated. In the cycle, two waters are
produced at the active site and presumably go out into the bulk solvent.

5.2 Proton Pumping

The proton pumping activity of the oxidase is tightly coupled with the electron
transfer activity. An electroneutrality principle has been advanced suggesting that the
movement of negative electrons into a region of low dielectric must be compensated by
proton movement to decrease the energy requirement (Rich, 1995). This theory is
supported by the fact that many of the electron transfer steps are pH dependent and show
a D/H isotope effect (Héllen & Nilsson, 1992; Oliveberg et al., 1991; Brzezinski &
Malmstrom, 1987).

Two coupling mechanisms of proton pumping have been proposed, direct and
indirect. The indirect mechanism proposes that electron transfer causes an overall
conformational change that causes protons to be pumped at some sites at a distance from
where the oxygen reduction occurs. On the other hand, direct coupling mechanisms
postulate that one or more of the ligands of the redox active metal centers are protonated

and deprotonated to transfer protons. Two proton uptake paths, D and K, have been well

18

defined but as yet it is uncertain where the proton exit paths are situated. One of the
possibilities is that protons could exit with the product water. So far, there is no
compelling evidence to support any of the models of the CcO proton pumping

mechanism.

19

 

 

iii

\\ i
stl
Il’it
int
co

pri

crj.

[lit

are

Sir;

 

Part II Docking of Cytochrome c on Its Redox Partners

The role of cytochrome c in mitochondria is to transfer electrons from the bet
complex to cytochrome c oxidase in respiration. In bacteria, various cytochromes c
transfer electrons to many different oxidases and to the reaction center in the
photosynthesis system. Efficient electron transfer requires rapid complex formation but
also rapid dissociation of oxidized cytochrome c to prevent inhibition of further electron
transfer. However, the chemistry of the surface of cytochrome c that is used to interact
with its redox partners, and whether different surfaces are used for different partners, is
still not fully defined. Determination of the structure of the Cc/CcO complex could reveal
the types of interactions important for this transient interaction and also for efficient
intermolecular electron transfer within the complex. A crystal structure of the docked
complex with Cc and CcO would be extremely useful. However, as CcO is a membrane
protein and Cc is soluble, it has proven difficult to obtain the crystal structure of the
complex. Therefore, some indirect approaches such as chemical modification,
crystallography for CcO, mutagenesis and computer modeling have been used to explain
the interaction of the Cc/Cc0 complex. This section gives a brief review of these

approaches.

I. Chemical Modification

Decreased strength of interactions between Cc and its partners by increasing ionic

strength of the reaction medium is a proof that an electrostatic component is significant in

20

 

 

 

Ill

SP

C0

cf:

TEX

ll‘l‘.

u i-

 

PR

bi:
1h:

In

 

 

these interactions. Because conserved lysine residues in Cc were thought to be important
in these interactions, chemical modification studies were done with chemicals such as
CNDP (Brautigan et al., 1978; Osheroff et al., 1980), TFA (Staudenmayer etal., 1976),
and TFC (Smith et al., 1977) to define which specific residues were important for the
interaction.

After these single—residue—modified cytochromes c were made and purified,
spectroscopic analyses showed that these modifications had little effect on the overall
conformation and redox properties of the protein. The steady-state kinetics were analyzed
spectrophotometrically or polarographically (Smith et al., 1977; Ferguson-Miller et al.,
1978; Osheroff et al., 1980). The results showed that Lysines-l3, 86/87, 72, and 8 in
horse Cc had the most significant effect on the binding of Cc to the oxidase. These results
accompanied by the results that these five derivatives and Lys—27 also showed the biggest
effects on their binding to the bcl complex (Smith et al., 1981), beef mitochondrial
reductase (Speck et al., 1979) and CcP (Smith & Millett; Kang et al., 1977) suggesting
that Cc uses the same area of its surface to bind to its different redox partners. This area is
where the heme edge is exposed. The other 13 lysine residues showed little effect on the
binding.

A differential protection technique, in which modification was carried out in the
presence of the complex, was also used to address the same issue (Rieder & Bosshard,
1980). The same residues suggested from chemical modification studies as important for
binding also showed their significance with this technique. Now it is generally agreed that
the area where these lysine residues are situated in Cc is involved in the docking of Cc to

most of its redox partners.

21

 

[1. Crystallographic approaches

A number of the physiological partners of Cc, for example, the oxidases and the
bcr complexes, are membrane-bound proteins. The insolubility of these proteins and the
transient nature of the binding of Cc make them hard to crystallize by themselves and
even harder to crystallize with soluble Cc. As an exception, CcP from yeast is a small
soluble protein with a molecular weight of 34.2 kDa and a single heme group. The crystal
structures of CcP itself (Poulos et al., 1980), horse Cc/CcP complex [CcP:Cc(H)] and
yeast Cc/CcP [CcP:Cc(Y)] complex (Pelletier & Kraut) were all obtained. This makes the
peroxidase system a good model for studying long-range electron transfer and the Cc
docking mechanism.

Even though the crystal of CcP:Cc(Y) and CcP:Cc(H) were obtained under
different ionic strengths, 5mM and lSOmM, the complex structures are similar, especially
in that CcP and Cc use similar domains to interact. The predominant interactions between
these two complexes are hydrogen bonds and van der Waals interactions. Both of the
complex structures showed that lysine residues 8, 13, 72, and 87 are important for the
interaction between Cc and its redox partner. These residues are the same residues that
were suggested from chemical modification studies (Kang et al., 1978). They were also
determined to be important for the interaction of Cc and Cc0 by chemical
modification(Smith et al., 1977; Ferguson-Miller et al., 1978; Osheroff et al., 1980). The
heme methyl groups of both horse and yeast Cc are all in van der Waals contact with Ala-

193 and Ala-194 from CcP. An electron transfer pathway, which involves the backbone

22

 

 

C0

\\L

[I

CO

h}

 

 

chains of Trp-191, Gly-192, Ala-193, and Ala-194 has been identified. This pathway is
actually the shortest straight line between the two heme groups (Pelletier & Kraut, 1992).

The most noticeable difference between the two complexes is the slight rotation-
translation of Cc as related to the CcP. This slight difference made a gap of 7A between
Ala-193 and194 and the methyl group of the heme from Cc in CcP:Cc(H) complex,
suggesting a less efficient electron transfer. This low efiiciency of electron transfer is
confirmed by some kinetic studies (Geren et al., 1991; Hahm et al., 1992).

There is still a question about the extent to which the crystallographic studies
faithfully represent the conformation of the protein when it is under normal physiological
conditions. With the differences between the two complex structures, another question
was whether differences arise from the crystal growth conditions or from intrinsic
differences between the two cytochromes c. Mutagenesis approaches were used to help

answer this question.

III. Mutagenesis approaches

1. Cytochrome c peroxidase

In the crystal structure of the complexes, a region on the surface of CcP is in close
contact with Cc, including Glu—32, Asp-33, Asp-34 and Glu-35, with Glu-35 forming a
hydrogen-bond with Lys-87 of Cc. Glu-29O in CcP hydrogen-bonds with Lys-72 from
Cc(H) or Lys-73 from Cc(Y). Mutants have been made of these residues to convert the
carboxyl groups to amide groups. These mutants include E320, D34N, E3 SQ, E290N and
E291Q. Ala-193 has also been changed into phenylalanine and cysteine. The cysteine

mutant was labeled with a bulky group of 3-(N-ma1eimidylpropionyl)biocytin (MPB) to

23

introduce a bulky group at the interface (Fishel et al., 1987; Miller et al., 1994; Miller et
al., 1996). The electron transfer rates were measured with the ruthenium-Cc photolysis
technique. For D34N, E290N and A193F, the rate constants of kg, were found to decrease
2 to 4 fold. The binding constants (KD) decreased 6 to 15 fold. Mutants E320 and E291Q
had no significant effect on the electron transfer and binding (Wang et al., 1996). These
results are consistent with the results derived from the crystal structures. The A193C-
MPB which introduced a bulky group into the interface had a ~150 fold decrease in the
ka. This is also consistent with the crystal structure of the interface. For both Cc(H) and
Cc(Y), the mutants had similar effects, indicating that they bind to the same region on
CcP.

The controversy about whether the conformational differences in the two complex
structures are due to the different crystal growth conditions or to the intrinsic difference
between the two cytochromes c was addressed by using the mutant A193C-MPB(Miller
et al., 1996). This bulky group caused 20 to 100 fold decrease in electron transfer rates for
Cc(Y) while only 2 to 3 decrease for Cc(H). This result shows that the conformational
difference in the crystal structure is present in solution too. So the interface resolved in

the crystal structures represents the kinetically competent interaction in solution.

2. Cytochrome c oxidase

The mutagenesis analysis on the binding of Cc and Cc0 has also been done. From
chemical modification/protection (Millett et al., 1982; Millett et al., 1983), Asp-112
(Asp-151 in R. sphaeroides), Glu-l l4 (Glu-152 in R. sphaeroides), and Glu-198 (Glu-

254 in R. sphaeroides) in subunit ll of bovine CcO have been identified to be protected

24

 

 

Sip/1

19g
mu

obS

 

from modification by Cc binding. Cross-linking studies also identified a cross-link
between His-161 (His-217 in R. sphaeroides) and Lys-13 from horse Cc (Bisson et al.,
1982). Time resolved electron transfer kinetic studies (Hill, 1991; Pan et al., 1993;
Malatesta et al., 1998) and crystal structures of Cc from both bovine and Paracoccus
(Iwata et al., 1995; Tsukihara et al., 1995; Tsukihara et al., 1996) show results consistent
with the modification study with the exception of Glu—l98 (Glu-254 in R. sphaeroides).
These residues (Asp-112 & Glu-114) are all in a domain close to where Cu is situated

(although not in the precise binding domain for Cc as determined by mutagenesis; see

 

below) but Glu-198 (Glu-254 in R. sphaeroides) is a metal ligand shared by Cu), and Mg E
and therefore unable to participate. Time resolved electron transfer kinetic studies have
also shown that Cup, is the initial electron receptor from Cc.

Using site-directed mutagenesis studies on Paracoccus denitrificans CcO, Glu-
148, Glu-157, Asp-195, Asp-214, Asp-229, and Glu—254 (R. sphaeroides numbering)
have been shown to be important for the binding of Cc (Lappalainen et al., 1995; Witt et
al., 1995; Witt et al., 1998). Subsequently published crystal structures showed that Asp-
229 and Glu-254 are buried inside the protein, and therefore do not interact with Cc. The
mutational results must be attributed to general disruption of the Cc0 protein.

In parallel with the P. denitrificans, similar studies were done on the R.
sphaeroides CcO; mutants E1480, D151N/E1520, E157Q, D188N/E189Q, D195N and
D214N were made on the subunit II to test their importance in Cc binding (Zhen et al.,
1999). The turnover numbers with both horse Cc and R. sphaeroides CC2 for these
mutants vary from 20% to almost 100% of the wild type oxidase. To determine if these

observed changes in activity reflected different contributions of these carboxyl residues to

25

 

 

th-

“1

l'lt.’

x:

“Z

ll‘i t

C01

Ch.

it,

 

the binding of Cc, complete spectral analysis was carried out by UV-visible and EPR,
showing no structural effects of mutations. In the ionic strength dependent activity studies
with horse Cc, the maximum activities are achieved at lower ionic strengths for the
mutants, suggesting that Asp-214, Glu-148, Glu-157 and Asp-195 contribute to Cc
binding. Asp-151, Glu-152, Asp-188, and Glu-189 mutants have no significant effect on
the ionic strength dependence activity, consistent with little or no involvement in the
docking. These conclusions were confirmed by time-resolved kinetic assays that showed
no change in the intrinsic electron transfer rate within the Cc-CcO complex for the latter
set of mutants, but significantly inhibited rates in the former. Direct binding assays
further substantiated the conclusions (Wang et al., 1999).

The turnover number of wild-type Cc0 with horse Cc is about 6-fold higher than
with R. sphaeroides C02; this was attributed to the different surface pattern of charge on
the cytochromes. In particular, R. sphaeroides CC2 lacks Lys-13 and Lys-72. In the case of
D214N and D195N, they were found to have different relative reactivities with the horse
and R. s. cytochromes. D214N and D195N retain 24% and 71% of wide-type activity with
horse Cc respectively, whereas they retain 67% and 19% of wide-type activity with R. 3.
C02. Further ionic strength assays showed Asp-214 did not contribute to the binding with
R. s. CC2. This suggested that Asp-214 in Cc0 may interact with Lys-72 in horse Cc. In
contrast, Asp-195 contributes more to the formation and electron transfer capacity of the
R. sphaeroides CC2:Cc0 complex. These results imply a specific complementarity of
charge between Cc0 and Cc.

There is a set of highly conserved aromatic residues Trp(l43)-Tyr-Trp-X-Tyr-X-

Tyr(l49) in the CcO subunit II soluble domain which have been postulated to be involved

26

 

 

in ll]:

Sl‘l0\\

\Vl-l?

The [1'

 

ICSpL".

inhib

mUILli

IS abt‘

 

aPPCL‘
10 Cu
Stroni
disru
a hut
Entr}

bind
C>To
and

inlet.

it;

in the electron transfer from cytochrome c to Cu or Cu to heme a. The crystal structure
shows Trp-143 in close proximity to Cu on the surface of the soluble domain. Mutants
W143F and W143A were made to analyze the role of this tryptophan in electron transfer.
The turnover rates for both mutants are 2% and 7% of wild type for W143A and W143F
respectively when measured with both horse Cc and R. sphaeroides CC2. This similar
inhibition suggests that both horse Cc and R. sphaeroides CC2 use the same residue, Trp-
143, as the electron conduit to Cc0, because the Cu, center is spectrally unaltered and the
mutants have the same binding strength as the wild-type. In the crystal structure, Trp-143
is about SA above the Cup, center and in the center of the conserved carboxyl residues that
appear to be involved in the binding. Indeed, the intrinsic electron transfer rates from Cc
to Cu», are decreased up to IOOO-fold in the rapid kinetic assay. This very large decrease
strongly suggests that the pathway for electron transfer between heme c and Cu has been
disrupted. With the report that the mutation W143Q in P. denitrificans Cc0 also leads to
a large loss in steady-state electron transfer activity, Trp-143 is identified as the electron
entry site to Cu. This also places the two cytochromes on approximately the same
binding area on CcO but with slightly different orientation and guiding electrostatics.

The above analyses suggest a picture of the interface of the docking of
cytochromes on the CcO. Lys-8, Lys-13, and Lys-86/87 interact with Asp195, Glu157,
and Glu148, respectively. However, other interactions are possible and the exact

interaction can only be guessed at from these data.

IV. Computational Modeling Approaches

27

 

 

com:
line:
I
the l‘
box i:
qum
mole

six bl

 

 

 

indie
cents
oxid;

feprc

The transient binding of Cc to Cc0 make this redox system amenable to
computational modeling based on static crystallographic structures, because the fast
kinetics of these interactions suggest that large conformational changes do not occur upon
the binding of the electron transfer partner. The electron-transfer complex between
bovine CcO and horse Cc has been modeled with the docking program DOT (Roberts &
Pique, 1999), which provides a complete search of all orientations between the two
molecules by systematically rotating and translating one molecule about the other. Thirty-
six billion configurations of the two molecules were calculated. The free-energy map
indicates that the positively charged front face of Cc, with the exposed heme edge in the
center, is preferentially oriented toward the large negative region on the cytochrome c
oxidase surface formed by the soluble domain of subunit II. In this interaction region, a
representative configuration from the tight cluster of lowest energy solutions was further
analyzed to determine the specific interface contacts.

In this most-probable model complex, the exposed heme edge of heme c is within
4A of the indole ring of Trp-l43'. This position and proximity imply involvement of Trp-
143 in the electron transfer, as observed experimentally. A severe drop in intrinsic
electron transfer rate is measured when Trp-143 is mutated to Phe and Ala in R.
sphaeroides Cc0 (Zhen et al., 1999). A small, hydrophobic contact region surrounds the
Trp-l43 and the contacting heme-edge, which is in turn surrounded by electrostatically
complementary hydrophilic interactions. The hydrophobic region of Cc consists of the
exposed heme edge, the adjacent Gln-16 and Cys-l7 side chains, and residues lie-81,

Phe-82 and Ala-83, which form a B-strand paralleling one heme face. The corresponding

28

 

regui
loop
C CO
durn
regh
front

Ccl

rend

subu

resid
ASp-
hOmt
b0\ii

blndl

.All:

 

region on CcO consists of residues Tyr-141, Gln-l42, Trp-143 and Tyr-144 forming a
loop over the Cult, site. Among the lysine residues from Cc and the carboxyl groups from
CcO, the interaction of Lys-l 3 with Glu-157 and Lys-72 with Gln-142 and Asp-214 are
the most important hydrophilic interactions due to their proximity to the hydrophobic
region and exclusion from bulk solvent. Glu-l48 and Ala-213 (Glu-157 for bovine Cc0)
from Cc0 are slightly farther from the hydrophobic core of the interface. They are near
Cc Lys-86 and Lys-87. The DB] of Asp-195 is about 8A from NZ of Lys-8. The two
residues might interact with each other through a water bridge.

The experimental and computational results agree that the soluble domain of C00
subunit II is involved in the interaction with Cc. The main chains of subunit II of both
bovine and R. sphaeroides are highly similar. Sequence identity is highest surrounding
the Cu site and on the Cc-binding surface. A one-to-one correspondence is present for
most of the acidic residues surrounding the hydrophobic loop, even though bovine CcO
has one additional carboxyl residue on subunit II, Asp-157(Ala-213 in R. sphaeroides).
Mutagenesis studies in Paracoccus denitrificans Cc0 also implicates another carboxyl
residue on Paracoccus denitrificans CcO subunit 1, Asp-257 (Gln-265 in R. sphaeroides,
Asp-221 in bovine) in Cc binding (Roberts & Pique, 1999; Witt et al., 1998). The highly
homologous crystal structures of both oxidases suggests that the docking studies with the
bovine Cc0 should apply to its counterpart in R. sphaeroides Cc0, and the mutagenesis,
binding, and kinetic studies on R. sphaeroides Cc0 should apply to the bovine Cc0.

The predicted structure is consistent with the mutagenesis, binding and time-

resolved kinetics studies. The properties of the calculated interface strongly resemble

 

' All residue numbering in Cc0 is according to R. sphaeroides sequence.

29

 

 

 

 

th0>t

(PeH

“in:
agrcr
alxrr
lfcrd
the}:

OIlCi‘

COR}
Ofien
dock
hto~
SUpe
COHU

OUICI

 

those found in the crystallographic structure of the complex of yeast CcP with yeast Cc
(Pelletier & Kraut, 1992).

Finally, the DOT calculation suggests that there may be a second binding site
which is further from the membrane for Cc in the large negative region. This is in
agreement with the evidence that supports a second, weaker binding site of C60 in which
a bound Cc molecule could interact with a Cc molecule in the primary binding site
(Ferguson-Miller et al., 1976). This low-affinity binding could be more disordered than
the high-affinity one and may need a bound Cc at the high-affinity site to define its
orientation.

Another computer modeling using program FTDOCK(F16ck & Helms, 2002) was
conducted recently on complexes of cytochrome c oxidase from Paracoccus denitrificans
with horse cytochrome c, and with its physiological counterpart cytochrome c552. The
complex obtained from docking with horse heart cytochrome c has a very similar
orientation to the complex found by DOT modeling, in which horse cytochrome c was
docked against the bovine crystal structure of Cc0. The position and orientation of these
two complexes differ only by a few Angstroms and degrees, respectively. If one
superimposes the corresponding amino acids of the two Cc0 structures in both Cc/CcO
complexes, the RMSD between the two cytochromes c coordinates is 3.8A. The close

outcomes of these two computational approaches add strength to both.

V. Summary

30

 

andc
onu)
undd
ph}s

\\ or},

 

 

The results from chemical modification, crystallography for CcO, mutagenesis
and computer modeling have provided consistent information about the docking of Cc
onto Cc0. Of these methodologies, computer modeling offered a more detailed
understanding of the docking. However, does the complex model truthfully reflect the
physiological binding of Cc onto CcO? In order to test the accuracy of the model, further

work needs to be done.

31

 

2 IO \\

elect

sequ
bind

hem

puri?
Infoi
1982

[EVIL

Whi L

I-P

DCV;

 

Part III c-gpe cytochromes in Rhodobacter sghaeroides

Rhodobacter sphaeroides, a purple nonsulfer bacterium, can grow
chemoheterotrophically in the dark or photoheterotrophically in the light. Some variants
can use N20 as electron acceptor to grow anaerobically. To accommodate these
diversified growing conditions, several electron transfer chains are used under different
growth conditions and several c-type cytochromes play important roles in transferring
electrons in these pathways.

Almost all the c-type cytochromes have a CXXCH heme binding motif in their
sequences (Meyer & Kamen, 1982). These two cysteines of the apoprotein covalently
bind to the vinyl side chain of the heme via two thioether bonds. So far, no other types of
heme have been found to be covalently bound.

As cytochromes c are small, stable, mostly soluble and relatively easy to be
purified, they are favorite proteins in many research areas. This has led to abundant
information about these proteins (Margoliash, 1966; Bartsch, 1978; Meyer & Kamen,
1982; Pettigrew & Moore, 1987; Moore & Pettigrew, 1990; Scott & Mauk, 1996). This
review will address cytochromes c in Rhodobacter sphaeroides. The question regarding

which one is the physiological electron donor for the am oxidase is also discussed.

1. Properties of c-type cytochromes

Table 1.1 (Yu, Dong, & Yu, 1986; Trumpower, 1990; Overfield, Wraight &

Devault, 1979; Chory, Donohue, Varga, Staehelin & Kaplan, 1984; Donohue, McEwan,

32

 

Van
C us:

‘nIt
”Kym

C} It‘s
oxid
sphu

hast

used

II. ‘

Oxii

 

 

Van Doren, Crofts & Kaplan, 1984; Rott & Donohue, 1990; Cusanovich, 1971; Meyer &
Cusanovich, 1985) shows the characteristics of most of the cytochromes c found in the R.
sphaeroides.

Besides the cytochromes c covered in Table 1.1, there are three other c~type
cytochromes that are part of the cytochrome cbb3 oxidase complex. As a heme-copper
oxidase, cytochrome cbb3 oxidase has been found in many species, including R.
sphaeroides (Garcia-Horsman et al., 1994b). Instead of having a CuA site in the protein, it
has two covalently bound c-type cytochromes (Garcia-Horsman et al., 1994; de Gier, et
al., 1996) at the C-terrninus of subunit II and one covalently bound c-type cytochrome in
a subunit basically homologous to subunit III of Cc0. All of these cytochromes c have
predominantly negatively charged residues with all the conserved lysine residues in
mammalian Cc missing. This suggests that other forces than electrostatic interaction are

used to stabilize the cytochromes and the oxidase.

11. Which cytochrome c is the physiological substrate for aa3-type

oxidase in R. sphaeroides?

The aa3—type CcO of R. sphaeroides has been crystallized and characterized, but
its physiological substrate is still unknown. Knowing the real substrate can help to
understand the electron flow in the respiratory chain and the regulatory mechanisms in
cells. As aa3-type Cc0 is at higher levels when the cells are grown aerobically in the
dark, its physiological substrate should be present at relatively higher levels in the same

growing situation. To transfer electrons efficiently, a reasonable amount of c-type

33

r1~\\\oI-\olh\i\ I‘ I‘.‘ ‘1‘
:0. .\ Unintcg
gafiai
‘irK Q-nxa‘c Uktxpwnxkn
. EL ~.~
pvtsfiurfi

 

 

03238 0_ 002000028. .m 00 0020.. 2000.0 2800 0 I 05030800 000003000$~ 80b .0 08020030 00 0080800 .8200 02.0 .0

.220 00 0305. 82 .3

880000 30:00 0002000 2 .0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

owe 2% $2 “Em
03200
<2 <2 $2 $2 em 8038 as
.0080;
68-6 62 an a a; 0. _ 00 .600. 565.5 32 2 a. sad
..0 08020008
9 3082 :00 as an 6328
2602652 a a a: 400 as am a. $2 :2 08 as
30
000028000 28000 €003 8.0
-0800 0m :02 0803 0200a .0382 :0 03 03200
0E. 8083080: 20m mmmz on 83053: 6<\Z mm 05 . 0
N0 08020030 020.00 0000.00
00 @2802 2.3; G 0800 2 N0 08020030 03200 8
2602682 $02 0:. ~00 am 06 28 as ea a: $2 :2 0am - 62
00800 0030000 05 00
23000.23000020 08—0800 .03 08020003
30% 000 000 05 8000 23020.8 03200
0280 0000—00 0: omm 0mm 00800—0 00.0080. 300.862 3 ONE 0N0
003800 00 03200 00 003800 03200
03 05 00 080300 0 \r :0 Wmmm mam 63 80¢ 80800—0 000000; 32 cm 8080000 0 0
as 9.5 09.0
8:: .030.— .28000m 0.800005 3203 03200:—
808800 them .. 0 0800,: 003085 .30EU 02:00.02 \0320m 080 Z

 

 

0030;00:80. .2 E 0 008020030 0.: .«0 000000.820 a; 030k.

34

 

 

CWO

onh

1.(?

CODC
char
C}t(i
acre
("554

C}Io

and

teas

[hin-

Split

 

cytochromes should be present in the cell, too. Of all the cytochromes C, until recently,

only cytochrome C2 and cytochrome C554 meet these criteria.

1. Cytochrome C554

When R. sphaeroides is switched from photosynthetic to aerobic growth
conditions, cytochrome C554 is five to eight fold more abundant than it was before the
change of the grth condition. The level of the protein is also about three times that of
cytochrome C2 under aerobic growth. As the most abundant C-type cytochrome under
aerobically growing condition (Bartsch et al., 1989; Flory & Donohue, 1995), cytochrome
C5 54 is a prospective candidate as the substrate for CcO. Besides the sequence of
cytochrome 6554, however, there is limited information about the protein so far.

As the oxidases from both R. sphaeroides and bovine are very similar structurally
and functionally, and as the mammalian CC is a good substrate for both oxidases, it is
reasonable to speculate that the physiological substrates should be similar. From sequence
analysis, there is little similarity between cytochrome 6554 from R. s. and mammalian CC,
especially in the region around the exposed heme edge where interactions are known to
occur. Given these observations, cytochrome C554 is an unlikely candidate for the real

substrate of R. sphaeroides CCO.

2. Cytochrome C2
Cytochrome C2 can transfer electrons to dag-type oxidase with one-fourth to one-
third the turnover rate of bovine CC. The binding constant Km of cytochrome C2 and R.

sphaeroides CCO is about 10 fold that for bovine CC (Zhen et al. , 1999; Kituchi et al.,

35

 

 

CC

\\
or.
o t

di

Ill:
ha
str

Cc

Bo

LG I'

1965). When cytochrome C2 is deleted from the strain, there is no significant effect on the
cell’s aerobic growth rate although there is an abolition of the strain’s photosynthetic
growth capability (Donohue et al., 1988; Rott & Donohue, 1990). Isocytochrome C2 may
take the role of cytochrome C2 to transfer electrons to oxidase, but cells can still grow
aerobically even when isocytochrome C2 and cytochrome C2 are deleted from the strain.
When cytochrome C2 is used to react with bovine CCO, the steady-state turn-over rate was
only 1% of that of bovine CC with bovine CCO (Errede & Kamen, 1978). This difference
of the reactivity suggests that the protein structures and heme electronic structures are
different.

The low spin R. sphaeroides cytochrome C2 (Rs. CC2) has a highly similar crystal
structure to that of bovine Cc. In their structures, the heme is surrounded by 5 helices and
the N-terminal helix interacts with the one at the C-terminal in both structures. R. s. CC2
has all the residues that are conserved in bovine CC that are critical to maintain the overall
structures (Figure 1.4). However, the low p] value of 5.5 for R. s. CC2 versus 10 for bovine
Cc (Meyer & Cusanovich, 1985) tells us that they have different surface charges. Most
importantly, the counterparts of Lys-l 3 and Lys-72 are absent in Rs. CC2 (Figure 1.4).
Cytochrome C2 does nevertheless have a dipole moment, with most of the positively
charged residues located at the front face, where the heme edge is exposed. This front
face is used to interact with electron transfer partners (Long et al., 1989).

On the other hand, R. s. CC2 has a longer sequence because of some insertions.
Bovine CC has nine lysine residues (8, 13, 25, 27, 72, 73, 79, 86, 87), which are important

for its interaction with its electron transfer partners (Smith et al., 1977; Ferguson-Miller

36

Figure 1.4 Crystal Structure Comparison for Horse Cytochrome C (A) and R.
sphaeroides Cytochrome C2 (B). The Lysine residues conserved for both cytochromes C
are shown in green. The Lysine residues that are in horse CC but are missing in Rs. CC2
are shown in matching colors. The Nitrogen and Oxygen atoms are shown in blue and red
respectively. The figure was created using Insight 11 with coordinates for horse CC (A)

from (Bushnell et al., 1990), and for R. sphaeroides CC2 (B) from (Axelrod et al., 1994).

37

N.0 080.500rm0 .0.“ .m

 

0 080.800th 0000M .<

38

 

 

“l
idt
lac

COI

tha

COI‘r

fror

YEqi

COIT

the

 

bou

 

et al., 1978; Rieder & Bosshard, 1980). The counterpart of Lys-l3 is a glutamine residue in
Rs. CC2. This Lys-l 3 is one of the most important charges for the interaction of bovine CC
with its electron transfer partner (Ferguson-Miller et al., 1978). The Lys-72, which is also
identified as very critical in horse CC binding to its partner, is also missing in R. s. CC2. The
lack of Lys-13 and Lys-72 in R. s. CC2 may change the orientation of binding and, as a
consequence, lower the binding affinity and the electron transfer efficiency (Zhen et al.,
1999). Rs. CC2 has a high redox potential of 356mV (versus 260mV for bovine CC) meaning
that it has a different heme environment than that of bovine Cc. This difference may
contribute to the lower reactivity too.

Interestingly, R. s. CC2 has similar reactivity to bovine CC when it transfers electron
from the [)6] complex (Errede & Kamen, 1978). This evidence suggests that R. s. CC2 has the
required characteristics for it to transfer electron from the be] complex rapidly, whereas
some characteristics are missing for it to transfer electrons to oxidase. This is also supported
by the fact that cytochrome C2 is crucial for photosynthesis but not for aerobic respiration. It
may use different residues to interact with the b6] complex and the oxidase.

The above evidence suggests that R. s. CC2 may not transfer electrons from the b6]
complex to oxidase physiologically. It is more likely to mediate the electron transfer from
the b6] complex to the reaction center.

Besides these soluble cytochromes C, recent evidence suggests that some membrane

bound cytochrome C may be the real substrate for aa3-oxidase in Rs.

3. Membrane Bound Cytochromes C

39

 

 

31.

{ill

3.]
(, T1
jar;
(TL
mo
ami
tern
Iran
SU’U
C}l(
has
Mo
plat

OVe

 

din
flex
this

19:;

 

Besides cytochrome C], which is a part of the bcl complex and an electron donor to
cytochrome C, a membrane bound cytochrome C552 was found in P. denitrificans (Turba et
al., 1995), and a homologous cytochrome cy in R. capsulatus (Jenney Jr. & Daldal, 1993)
and in R. sphaeroides (Myllykallio et al., 1999). P. denitrificans and R. capsulatus are
closely related to R. sphaeroides.

3.1 Cytochrome c552 The genes for cytochrome c552 have been cloned from P. denitrificans
(Turba et al., 1995), R. capsulatus (Jenney Jr. & Daldal, 1993) and Bradyrhizobium
japonicum (Bott et al., 1991). Cytochrome C552 has been purified from P. denitrificans
(Turba et al., 1995) and its soluble domain has been crystallized (Harrenga et al., 2000). The
molecular weights of these homologous proteins are about 20 to 22 kDa. Analysis of the
amino acid sequences show that there is a hydrophobic N-terminal domain and a globular C -
terminal domain. The N-terminal domain (30 to 40 residues) is predicted to form a
transmembrane helix in order to anchor the protein to the membrane. In general, the
structure of the soluble C-terminal domain has the same features as that of known eukaryotic
cytochrome C. Most of the important residues in the C-domain are conserved. This domain
has the typical “cytochrome C fold” structure with the heme surrounded by five helices.
Most of the lysine residues on the front face of the bovine Cc are also present in a similar
place in the structure of C552 from P. denitrificans, but with a larger negative surface charge
overall, C552 has a smaller positively charged area around the solvent-exposed heme atoms.
A neck region between the N-terminal domain and the C-terminal domain may have
flexibility due to a high proportion of alanine and negatively charged residues. C5 52 may use
this flexibility to transfer electrons between the [)6] complex and the M3 oxidase(Bott et al.,

1991).

40

 

 

ubit

(E3e

den

C}tc

cm

this

oxi.

Cjc

Ftu'

C}t.

CCR

C Q;

Cor

10*

nu:

 

The isolated complex of cytochrome C552, bcl complex and aa; oxidase can oxidize
ubiquinone rapidly without cytochrome C2 and adding horse CC cannot increase the activity
(Berry & Trumpower, 1985). An anti-C552 antibody inhibits the electron transfer in P.
denitrificans (Turba et al., 1995; Smith & Davies, 1991; Steinrficke et al., 1991). The
cytochrome C oxidase from P. denitrificans oxidizes horse heart cytochrome C and the
cytochrome C552 with similar rates (Reincke et al., 1999). All of this evidence suggests that
this membrane anchored cytochrome C552 shuttles electrons from the bCl complex to the M3
oxidase (Bakeret et al., 1998).

2.3.2 Cytochrome Cy Another membrane anchored C-type cytochrome, cytochrome
Cy, has been found in R. capsulatus (Jenney Jr. & Daldal). It is found that either cytochrome
C2 or cytochrome Cy is essential for phototrophic growth only when the other is absent.
Furthermore, the cytochrome C2 deleted R. sphaeroides strain can be complemented to grow
photosynthetically by the cytochrome Cy gene from R. capsulatus. This suggests that
cytochrome Cy can mediate electron transfer between the be. complex and the reaction
center.

Until recently, R. sphaeroides was thought to have only soluble cytochrome C2.
Recent work on R. sphaeroides found a gene (Myllykallio et al., 1999), cyCYR‘, which is
highly homologous to cytochrome Cy of R. capsulatus and C552 from P. denitrificans, too.
Alignment of the amino acid sequences of cytochromes Cy from both R. sphaeroides and R.
capsulatus shows that they have a conserved bipartite domain structure. Each of them
contains a C-terminal cytochrome C subdomain and N-terminal stretch anchoring the protein
to the membrane. Several lysine residues (Lys—8, 13, 72, and 87 in horse cytochrome C

numbering) which are thought to be important for the interactions of cytochrome C with its

41

 

Cup
pho

COV

pro
Cfit

oxi-

 

oxi

“i1

C0 1‘

 

r63

 

redox partners are well conserved, whereas several others (Lys-7, 25, 27, 73, 86, and 88) are
not. The linker portion of cytochrome Cy from R. capsulatus is longer than those of R.
sphaeroides and P. denitrificans.

It was found that cytochrome Cy cannot participate in photosynthetic electron transfer
in R. sphaeroides, although it is active in respiratory electron transfer, unlike its R.
capsulatus counterpart. Chimeric constructs showed that the inability to grow
photosynthetically comes from, at least in part, its redox active subdomain, which carries the
covalently bound heme. There is a big difference between the calculated p1 of the different
cytochromes Cy of 4.74 and 7.76 for R. sphaeroides and R. capsulatus respectively. It seems
that the precise location of the charged residues determine whether cytochrome Cy can form a
productive electron transfer complex with its partners. The photosynthetic inability of
cytochrome Cy in R. sphaeroides suggests that its interactions with the reaction center and
oxidase are different. Even though the question of whether this cy is the real substrate for M3
oxidase is complicated by the presence of Cbb3 oxidase in R. sphaeroides, further studies
will present more information to clarify this ambiguity.

Lately, R. sphaeroides mutants lacking either cytochrome C2 or cytochrome Cy and
either the aa3-oxidase or the Cbb3-oxidase were obtained (Daldal et al., 2001). The data
shows that both mobile cytochrome C2 and membrane-anchored cytochrome Cy can transfer
electrons efficiently from the bcl complex to either the Cbb3-oxidase or aa5-oxidase. The
concurrent operation of multiple electron carriers and cytochrome C oxidases implies that the
respiratory electron transfer pathways of R. sphaeroides are more complicated than those of

R. capsulatus.

42

III. Summary

Several C-type cytochromes are involved in the branched electron transfer chains in
R. sphaeroides. Cytochrome C2 transfers electron from the b6] complex to the reaction
center. It can also transfer electrons to the day, oxidase at one-fourth the rate of bovine CC.

The finding of membrane bound cytochrome C5 52 and Cy in several bacteria and their
properties suggest that Cy may be the physiological substrate for 003 oxidase in R.
sphaeroides. If it is one of the substrates, it will anchor itself to the membrane and move
between the bCl complex and the M3 substrate using its flexible neck region between the N-
tenninal transmembrane domain and the C-terminal soluble domain. This may require the
bcl complex, the membrane bound cytochrome c and the M3 oxidase to be in proximity on
the cytoplasmic membrane to guarantee efficient electron transfer(Berry & Trumpower,
1985). Using analysis of soluble cytochromes and modeling of cytochrome C, a good
understanding of the important interactions has been obtained even in the absence of a

verified, physiological electron donor to Rs. CCO.

43

Chapter 2

The Water and/or Proton Exit Pathway

and Its Relationship to Respiratory Control

44

 

1—

If.

If.

Ol'

pm

31.

 

pat

am

the
has
Stu

COE

be
fro]

and

Introduction

Cytochrome C oxidase (CCO) is the terminal oxidase in the mitochondrial electron
transfer chain. In each reaction cycle, it takes four electrons from soluble cytochrome C
(CC) and reduces one oxygen molecule into two water molecules. Using the energy
available, four protons are translocated across the membrane (Wikstrom, 1977). This
translocation results in an electrochemical gradient across the membrane, which can be
used by ATP synthase to make ATP (Mitchell, 1966). ATP can be used as energy by
organisms directly. A similar process using homologous enzymes is carried out in many
prokaryotic as well as eukaryotic organisms.

With the availability of crystal structures of CCO from P. denitrificans (Iwata et
al., 1995), R. sphaeroides (Svensson-Ek etal., 2002), and bovine heart (Tsukihara et al.,
1995), several proton translocation pathways in the protein have been proposed. These
pathways include two proton entry routes, the D and K channels, and a possible water
and/or proton exit, immediately above the binuclear active site (Figure 2.1). This channel
is at the interface of subunits I and II, with the non-redox-active Mg metal center right at
the bottom of this channel (Figure 2.2). The octahedral coordination of this metal center
has 3 amino acid residues and 3 waters as its ligands. With highly resolved crystal
structures of oxidase, water molecules are seen in this region of the protein, which is
consistent with the proposal that it could provide a proton and/or water exit channel.

Two residues, Arg-234 from subunit II and Asp-271 from subunit I are found to
be at a possible external mouth of the channel. In the Rs. CCO structure, the distances
from the closest N atom of Arg-234 to the carboxyl oxygen atoms of Asp271 are 3.46A

and 3.05A, respectively. These distances are in the range of ionic strength interaction.

45

Figure 2.1 Crystal Structure of Rhodobacter sphaeroides Cytochrome C Oxidase and
the Possible H20 and/or H+ Exit Pathway. Helices of subunit 1, II, and III are shown in
yellow, purple and cyan respectively. Yellow balls represent Cu and Cug, orange ball
for magnesium and magenta ball for calcium. The green, blue and red arrows represent
the suggested K- and D- proton uptake pathways and H20 and/or H+ Exit Pathway. The
figure was created using the Insight 11 program with coordinates from Svensson-Ek et al. ,

2002.

46

 

Figure 2.2 Possible H20 and/or H” Exit Pathway of Cytochrome C Oxidase. Purple,
orange and cyan represent residues involved with the exit pathway from subunit 1, II, and
111 respectively. Yellow balls represent CuA and Cu}; and orange ball for magnesium. The
oxygen and nitrogen atoms involved with the pathway are shown in red and blue
respectively. The green arrows represent the suggested H2O and/or H+ Exit Pathway. The
figure was created using Insight II with coordinates for R. sphaeroides CCO from

Svensson-Ek et al. , 2002.

48

.HHmesnam

moi

Hzaaasm

E 3:595

   

. Mao

49

Further, Asp-271 is the first residue in a conserved DPxxY region, which is part
of a loop, situated at the periplasmic end of the helix VI. This (N/D)PxxY motif is also
highly conserved in the transmembrane segment 7(TMS 7) of G-protein-coupled
receptors (GPCRs) (Wess et al., 1993; Zhang & Ewinstein, 1993; Lou, Zhang &
Weinstein, 1994; Van Rhee & Jacobson, 1996). TMS 7 has been shown to play an
important role in the signal transduction function of GPCRs. Instead of being an alpha-
helix, as most of the transmembrane segments are, the sequence of TMS 7 is inconsistent
with it being an ideal alpha-helix and (N/D)PxxY motif has been identified as the major
determinant for deviation of TMS 7 from ideal helicity (Konvicka et al., 1998). The
deviation includes an Asn/Asp turn and a flexible “hinge” region. The flexibility
introduced by the specific structural perturbation produced by the (NP/DP) motif in TMS
7 is proposed to have a significant role in transmitting the signal from ligand binding to
the receptor, across the membrane.

In CCO, the DPxxY motif is situated at the periplasmic end of helix VI of subunit
1. Helix VI contains a number of essential, highly conserved residues (H284, Y288, E286,
and S299), which are important for the oxygen reduction at the heme a3 — CuB site, and
for proton pumping (Figure 2.3). His-284 is a ligand of CuB (Shapleigh et al., 1992;
Hosler et al., 1993); Tyr-288 is covalently bound to His-284 (Ostermeier et al., 1997;
Yoshikawa et al., 1998); Glu-286 is a buried charge that is part of a proton uptake
pathway, the D-path (Konstantinov et al., 1997; Junemann et al., 1999); and Set-299 is
part of another proton uptake path, the K-path. Also close to this loop region are the heme

propionates or nearby protonatable groups that are also proposed to trap protons during

50

 

TC

5;

Figure 2.3 The Overall Structure of Helix IV and Possible Proton Pathways. Purple
and yellow represent residues involved with the exit pathway from subunit I and II
respectively. Yellow balls represent Cu and CuB and the orange ball represents
magnesium. The oxygen and nitrogen atoms involved with the pathway are shown in red
and blue respectively. K-path and residues involved with it are shown in purple, D-path
and residues involved with it are shown in green and red represents the possible H2O
and/or H+ Exit Pathway. The figure was created using Insight 11 with coordinates for R.

sphaeroides CCO from Svensson-Ek et a1. , 2002.

51

 

52

coupled electron transfer (Michel, 1999). Due to its relationship to all these critical
groups, it is proposed (F erguson-Miller, S., NIH proposal) that the DPxxY motif may
undergo conformational changes during the catalytic cycle and mediate the interaction
between subunits I and 11, possibly controlling the release of the pumped protons. This
externally situated, charged region may also respond to the pH or electrical gradient to
mediate respiratory control, the inhibition of oxidase activity in the presence of a
membrane potential.

Arg-234 in subunit I has been converted to His and Cys mutants. These mutants
had ~100% and 50% of the wild type activity. The Cys mutant shows perturbed proton
pumping behavior and low respiratory control (Florens et al., unpublished data; personal
communication, Bryan Schmidt). As Asp-271 from subunit 1 strongly interacts with
Arg234 and has the proposed role in proton pumping as stated above, Asn and His
mutants were made to test the nature of the interaction and its importance in the fimction

of the protein. The data of these two mutants is reported.

Materials and Methods

Materials All restriction endonucleases were from Boehringer Mannheim
(Indianapolis, IN); Plasmid pRK415 and E. coli strain S-l7-1 were obtained from the
sources indicated in Shapleigh and Gennis (Shapleigh & Gennis, 1992). Plasmid pYJ l 00
containing the sequence of coxII and coxIII was originally prepared by Cao et a1. (Cao et
al., 1992). The gel purification kit and Ni2+-NTA resin were from Qiagen (Santa Clarita,
CA). Ultrafree-l 5 centrifugal filters were from Millipore (Bedford, MA), lauryl

maltoside from Anatrace Inc. (Maumee, OH). Horse heart Cc (Sigma type VI) was

53

purchased from Sigma and was purified on a carboxymethyl cellulose column before use
(Brautigan et al., 1978).

Construction of the D271 mutants Site-directed mutagenesis at residue Asp-271
was performed by the overlap extension PCR method (Ho et al., 1989). In the PCR
reactions, the high fidelity Pfu polymerase (Stratagene, La J olla, CA) was used, and all
PCR products were sequenced to make sure that no secondary mutations were
introduced.

The PCR products were digested with Kpn I and Bgl II restriction enzymes. These
digested products were cloned into the pSL1180 vector. Then the fragments were inserted
into pJS3-X6H2 vector, which has the whole COX I gene. At this step, the sequences
were checked to make sure that the mutations were introduced. After cutting out the
mutated COX 1, the fragments were inserted into pUCl9 vector to get rid of the unknown
fragment in pJS3-X6H2. This plasmid is called pJS3-SH. Then COX II and III were
inserted into pJS3-SH to make mutated plasmid pYJ123H. At this step, the genes for all
three subunits were together into one plasmid. The whole genes were then inserted into
pRK415 so that the protein could be overexpressed in R. sphaeroides. This was followed
by transformation into the E. coli host strain Sl7-1, then the plasmid was conjugated into
Rhodobacter sphaeroides strain J 8100, a COX I — deleted strain.

Conjugation Plasmid pRK415 and its derivatives were conjugated into R.
sphaeroides strain J 8100 with E. Coli strain S-17-l as the donor. Basically, J 8100 cells in
late exponential phase were mixed with S-17-1 cells in early exponential phase before the

mixture was spun for 1min. The pellet was resuspended in 100ul LB medium and spotted

onto a nitrocellulose filter on a LB plate. After incubation at 30°C for 6 hours, the filter

54

was washed with 5m] Sistrom’s media. The cells were spun down and resuspended in
1ml Sistrom’s medium. Then the cells were plated on Sistrom’s agar plates containing
Streptomycin/Spectinomycin antibiotics and were grown at 30°C for 2-3 days. A single
colony was picked and restreaked on a new Sistrom’s plate with the same antibiotics.
Cells were grown at 30°C for 3 days, then inoculated into two 100 mL volumes of
Sistrom’s media with the same antibiotics and grown at 30°C for about 2 days.

Cell growth E. coli strains S-l 7-1 were grown in LB medium at 37°C. Plasmids
pUC 19, pRK415 and their derivatives were maintained in the presence of the following
antibiotics: ampicillin (Ap) (50ug/ml), streptomycin (Sm) (SOug/ml), spectinomycin (Sp)
(SOug/ml) and tetracycline (Tc) (lSug/ml).

R. sphaeroides strain J 8100 and its conjugated derivatives were grown aerobically
at 30°C in 2.8 liter Fembach flasks containing 800ml of Sistrom’s minimal medium A
(Cohen-Bazire et al., 1957) with rapid swirling. All the strains used here were maintained
in the presence of Sm (SOug/ml) and Sp (50ug/ml). For the conjugated strains, Tc of
lug/ml was added.

Preparation of cytoplasmic membranes Membranes were prepared as described
in Zhen et a1. (1999). Approximately 40g of wet weight cells were resuspended in 200ml
of 50mM KH2PO4, 1mM EDTA, 50mg/ml DNasel and RNase. The resuspended cells
were broken by two passages through a French pressure cell at 20,000psi. Unbroken cells
and debris were removed by centrifugation at 30,000g for 30 minutes. The supernatant
was centrifuged at 200,000g for 1.5 hours at 4°C to get the pellets of the membranes. The
membrane pellets were resuspended in lOmM Tris-HCl, 40mM KCl at pH8.0. Lauryl

maltoside (LM) was added to a final concentration of 2% for solubilization. The

55

membranes were stirred at 4 ° C for 30 min and then centrifuged at 200,000g for 30 min
to remove unsolubilized material.

Niz+-NTA purification of CCO The solubilized membranes were mixed with
Ni2+-NTA resin after washing the resin with water in a ratio of lml of packed resin for
every 2.5mg of oxidase. Ten mM imidazole was added to prevent nonspecific binding.
The mixture was stirred at 4 ° C for 1b and then poured into a gravity column. The
column was washed with 10mM Tris-HCl, pH8.0, 40mM KCl, 10mM imidazole, 0.1%
LM until the flow-through was clear of color and turbidity. The column was further
washed with 5 bed volumes of 10mM Tris-HCI, pH8.0, 40mM KCl, 0.1% LM to wash
off the imidazole. The enzyme was eluted slowly with 100mM histidine, 10mM Tris-
HCl, pH8.0, 40mM KCl, 0.1% LM. The green colored fractions containing oxidase were
pooled together and washed 3 times with 10mM Tris-HCl, pH8.0, 40mM KCl, 0.1% LM,
by centrifugal filtration in Ultrafree-l 5 centrifugal filters with molecular weight cutoff of
50kD to reduce the histidine level to less than 1mM.

FPLC-DEAE chromatography The tandem DEAE-SPW Fast Performance
Liquid Chromatography (F PLC) columns (8mm x 7.5cm, TosoHaas) were attached to an
F PLC system (Pharrnacia LKB Biotechnology Inc.). The columns were preequilibrated
with 10mM KH2PO4, pH7.6, 1mM EDTA, 0.2% lauryl maltoside. The Ni2+-NTA
purified CCO was applied to the columns. The samples were eluted with the same buffer
but having gradient of 0.01-0.35M KCl at a flow rate of 0.5 ml/min. The fractions
containing oxidase were concentrated and stored at -80°C.

Activity assay The maximum turnover number (molecular activity) of CCO from

R. sphaeroides was measured polarographically using a Gilson model 5/6H oxygraph at

56

 

 

it

 

25°C in a reaction medium containing 50mM KH2PO4, pH6.5, in the presence of 0.05%
LM, 2.8mM ascorbate, 0.55mM TMPD and 30uM of horse CC. The oxidase
concentrations were in the range of 5 to 50nM.

SpectroPhotometric analysis The solubilized membrane or oxidase was diluted
in 50mM KH2PO4, pH6.5, 2% LM and the spectra were taken on a UV/visible
spectrometer (Perkin Elmer, Larnda 40P). Dithionite and ferricyanide were used to fully
reduce or oxidize the samples, respectively. The extinction coefficients used were A8606-
(,.5()=37cm"mM'l for absolute spectra of the purified dithionite-reduced enzyme, and
At:(506.630=28.9cm'1mM'l for difference spectra of the dithionite-reduced minus

ferricyanide-oxidized enzyme or membranes(Callahan & Babcock, 1983).

Results

Design of Mutants

From the crystal structure of Rhodobacter sphaeroides oxidase, we can see that
Asp271 in subunit I is in close proximity to Arg234 from subunit II and Lyle3 from
subunit III. Arg234 has been mutated to histidine, cysteine and serine (F lorens eta1.,
unpublished data). Even though all of these mutants have lower activity than wild type,
R234H mutant shows behavior close to wild type oxidase. There is evidence that all the
mutants have a tendency to dissociate and lose subunit II and III during purification of
the protein and possibly during assay procedures. This indicates that the salt bridge
between Arg234 and Asp271 may be important for holding subunit 1 together with 11 and

III. Mutation of Lyle3 also caused subunit III to be lost, but does not result in

57

significant loss of activity when the fraction still containing subunit III is purified by a
his-tag-III version of CCO (personal communication with Dr. John Hosler).

Because Asp-271 is also in the motif (N/D)PxxY, which has been shown to play
an important role in the signal transduction function of GPCR (Wess et al., 1993; Zhang
& Ewinstein, 1993; Lou, Zhang & Weinstein, 1994; Van Rhee & Jacobson, 1996),
changing Asp-271 to Asn should show us whether Asn can substitute for Asp at this
position for the function of signaling. Also, changing the carboxyl group to amide
removes the negative charge. This might demonstrate the significance of the charge on
Asp-271.

Histidine has a normal pKa of 6.0 so that under physiological conditions of pH
7.2 for R. sphaeroides, histidine should be neutral. However, the pKa of an amino acid
depends on its microenvironment in the protein. Also, histidine is a good metal ligand.
Changing Asp to His can give us further options to put a metal site at this exit of the
presumed water channel to reversibly block it.

The substitution of cysteine for Asp 271 was chosen because it can form a
disulfide bond with another cysteine residue if the two residues are close enough in
space. By changing both Arg234 from subunit I and Asp 271 from subunit II to cysteines,
they could form a disulfide bond , assuming some flexibility in the region, to shut down
the suspected water channel in a way different from the metal ligand method. We can

also attach a bulky and/or fluorescent group onto the cysteine residue.

58

Construction of Mutants
The mutants D271N and D271H were successfully made. However, the mutant
for D271C could not be made when constructing the pYJ123H plasmid. Therefore, only

two mutants were expressed and characterized.

Optical spectral characteristics of mutants

In the reduced form, wild-type CCO from R. sphaeroides has a characteristic
absorbance at 445nm and 606nm for the Soret and a-peaks, respectively, with a Soret to

a-peak ratio (A445490/A606-63o) of 5.6. Both mutants, D271H/N, have shown shifted 0L-
peaks in the visible spectrum. Specifically, both (Jr-peaks shified to about 605nm in the
solubilized membranes before the Ni2+-NTA column. After Ni2+-NTA purification, the or-
peaks shifted to 600nm and 597nm for D271N and D271 H, respectively (Figure 2.4 &
2.5). This indicates structural alteration at the heme centers in the proteins. Asp271 was
revealed by X-ray structures to be at the interface of subunits I, II and 111, not internally
located to ligate with the metal centers (Ostermeier et al., 1997; Yoshikawa et al., 1998;
Svensson-Ek et al., 2002). The observation of severely disrupted heme centers might
indicate the disaggregation of the subunits and exposure of the heme centers to a more
aqueous environment. It is also possible that the environment of the heme centers has
been changed with the introduction of the mutants because of the movement of helix VI.
Because the (it-peak shifted from 605nm to 600nm during Ni2+-NTA purification for
D271N, the mutant could have gone through further dissociation during the purification

process.

59

The Activities of mutant enzymes

CCO from R. sphaeroides shows high activity with horse Cc; as a result, horse CC
is used routinely in kinetics assays. The maximum turnover numbers of mutated CCO
have been measured using horse CC as the substrate in polarographic assays. Normally,
wild-type oxidase shows a maximum turnover rate of 1700 s'1 in 50mM KH2PO4, pH6.5,
in the presence of detergent. Under the same condition, the turnover rates for D271N and
D271H mutants are 5.9% and 4.5% of the rate for wild-type oxidase. These are shown in
Table 2.1. With the spectral data observed, the low activity might be the result of the

disaggregation or metal center disruption.

SDS-Page Analysis

During SDS-page, the protein is denatured with detergent and urea so that the subunits
are dissociated and the peptides are in an extended conformation. Then the subunits can
be separated dependent on their different molecular weights. Under this analysis, CCO of
R. sphaeroides mainly shows 3 bands, 49kD for subunit 1, 37kD for subunit II and 22kD
for subunit III. The real molecular weights for the three subunits are 62.6kD, 32.9kD and
30.1kD for subunit 1, subunit II and subunit 111 respectively. The observation of lower
molecular weight bands of two of the subunits after the SDS-page indicates that the
hydrophobic peptides of oxidase bind more of the negatively charged SDS and thus
migrate faster during electrophoresis. Subunit IV at ~ 6 kD (Sevensson-Ek et al., 2002),

is only clearly seen in a higher resolving gel (18% acrylamide).

60

Table 2.1 The Activity Comparison of D271 Mutants of Cytochrome C Oxidase from

R. sphaeroides

 

 

 

 

 

Enzymes a—peak (nm) Turnover" (SJ)
Wild-type 606 1700

D27 IN 5 99 1 00

D27 1 H 597 77

 

 

 

*The turnover rate is measured with 30 uM horse CC in 50mM KH2PO4, pH 6.5.

61

 

Figure 2.4 Visible Spectra for NiZI-NTA Purified D271N and D271H Mutants. The
proteins (purified with Ni2+-NTA column) are reduced with dithionite and measured in a
Perkin-Elmer Lambda 4.0 UV/V IS Spectrophotometer. These measurements were

conducted in 100mM KH2PO4, 0.1% LM with pH 7.2.

62

 

— q

I

one

oov

 

 

‘\ ‘‘‘‘‘‘
I

 

.5500

I. I\ \v Ila“,

‘0‘! v c I Illll o’l’!‘!|l\
’-

I I,

l‘II'l!l l
‘ '09..
|\ 0 I I

zrnmo .............. -
Irnmo ...................
83 2:5 I

’
01/
II

I ’0

.
12'.

n 00.0

L mod

I
O
‘—
O

l m...o

l 0N6

 

I mNd

SOUBQJOSqV

63

Figure 2.5 Comparison Between Visible Spectra for Solubilized Membrane and
Ni2+-NTA Purified D271N Mutant. The membranes are reduced with dithionite and
measured with membranes oxidized with ferricyanide as a reference in order to obtain the
difference spectra in a Perkin-Elmer Lambda 4.0 UV/V IS Spectrophotometer. The Ni”-
NTA purified proteins are reduced with dithionite. These measurements were conducted
in 100mM KH2PO4, 0.1% LM with pH 7.2. The absorbance for solubilized membrane

was magnified by 5 fold.

64

 

cozmoESa Loam. 2 F Rn. IIII
95525.: 85.528 2 END

090.0

I owed

 

r nKod

eoueqmsqv

65

Figure 2.6 shows the SDS-page gel analysis for D271 mutants after the Nib-NTA
purification. From the gel we can see that the purified mutant proteins have a greater than
stoichiometric ratio of subunit I compared to wild type. Ni2+—NTA column binds oxidase
by interacting with the His-tag, which is constructed on subunit I. If the mutated oxidase
cannot hold all the subunits together, Ni2+-NTA purification could result in extra subunit
I in the purified mixture.

To avoid getting extra subunit 1 after the Ni2+-NTA purification, F PLC-DEAE
chromatography was conducted for the D271N mutant without prior NiZI-NTA
purification. The tandem DEAE-SPW column separates proteins by their negative amino
acid charges on the surface of the 3-D structure and also somewhat by their molecular
weights. Figure 2.7 shows the FPLC profile. Figure 2.8 shows the SDS-page analysis for
the FPLC results. Because subunit II has a lot more negative amino acids on its surface, it
is responsible for much of the binding to the column. The whole oxidase that is attached
to subunit II is separated from other impurities, but will co-purify with subunit II alone.
From the SDS gel, we can see that the purified mutant has a lot of extra subunit II. This
further demonstrates that the protein is dissociating.

Due to the overall disruption of the structure of the mutant forms, further

functional characterization was not attempted because it would be difficult to interpret.

66

Figure 2.6 SDS-Page (14% Separation and 8% stacking) Analysis of Ni2+-NTA

Purified D271N and D271H Mutants. ~20 pg of proteins were added in each lane.

67

M
w
mzs
3 7
864

SubI

SubII

 

32fi5

25

I
nu
b

u
S

 

165

68

Figure 2.7 F PLC Profile for D271N Purification with Tandem DEAE-SPW Column.
The blue line shows the 280nm spectrum for the fractions eluted from the column.
Fractions are labeled I-VII. The SDS-page analysis of some of the fractions is shown in

Fig. 2.8.

69

 

 

 

  

 

 

                 
  

 

 
   

 

 

 

 

 

 

 

o. . . ..._.... in...» sag... .2. . .
.......: ....._.._......._.. .... ... .
a ........, . H.
.. 2mg... ,. :9:
. 2. .2. _. .
9v. . .. . .. . . u _ w ..
.. ._. .... 2...... .
. 5.. . g .
_ ... ...... é... .. .3.
3 2...... 3....
,... x... x H.
.
S. * econ
.
m.
30? M
. 9.?
E25. :43:
£555.- n25.59253. -l n-.- againiigzrms .
......n..me.w...58952.3: 950....meececlfifla. . >2-._..2..§52:.S Kil-.. .

 

 

70

Figure 2.8 SDS-Page Analysis of F PLC Purified D271N Fractions. The Roman

numbering refers the fractions in Figure 2.7.

71

16.5

 

72

Discussion

Subunit III

The catalytic core of CCO of the non-sulfur purple bacteria Paracoccus
denitrificans and Rhodobacter sphaeroides is made up of three subunits: I, II and III. The
three subunits are highly similar to those of the CCO of mitochondria (Saraste, 1990; Cao
etal., 1991; Cao et al., 1992; Hosler et al., 1992; Shapleigh & Gennis, 1992). Indeed, the
structures of the three subunits from the two bacteria mentioned above are almost
identical to the three-subunit core of the bovine oxidase (Iwata et al., 1995; Tsukihara et
al., 1996; Ostermeier et al., 1996; Svensson-Ek et al., 2002).

As a large hydrophobic peptide, subunit III binds to the transmembrane face of
subunit I opposite the transmembrane helices of subunit II (Iwata et al., 1995; Tsukihara
et al., 1996). It holds no metal centers and hence has no direct role in electron transfer
and oxygen reduction. However, its role in the catalytic core cannot be neglected because
it is highly conserved in different kinds of cytochrome C oxidase and is situated right
against the region of subunit I that contains the active site. The primary structure of
subunit III has 71% similarity between R. sphaeroides and human (Cao et al.; 1991).
Subunit III binds to subunit I in a relatively weak way, because it can be removed from
the enzyme with lauryl maltoside or polyoxyethylene detergents at high pH (Solioz et al.,
1982; Thompson et al., 1985; Gregory & Ferguson-Miller, 1988(1); Gregory &
Ferguson-Miller, 1988(2); Haltia et al., 1989; Haltia et al., 1991; Nicoletti et al., 1998; ).

How subunit III functions still needs to be determined.

73

The Lys103 in subunit III

Lys-103 from subunit III sits right next to D271 from subunit I and Arg 234 from
subunit II, making a charged triad (Figure 2.2). The mutation of Lyle3 to Ala results in
loss of subunit III (Personal communication with Dr. Jon Hosler). However, the fraction
of the mutated protein containing stoichiometric amounts of subunit III (isolated by a his-
tag on subunit III) is as active as wild type in terms of oxygen reduction and proton
pumping. Nor does the mutated protein suicide inactivate. Comparing Arg234 mutants of
subunit II and D271 mutants of subunit I to the Lysl 03A, the latter has a less dramatic
effect on activity although all of these residues appear to play a role in protein stability
(Personal communication, Bryan Schmidt, Dr. Jon Hosler). From subunit 111 sequence
alignments, we found that Lysine at 103 is not highly conserved. In some other CCOs
such as in bovine and human, this residue is substituted by a histidine. This substitute
could be positively charged and play a similar role but it suggests that Lyle3 has less
significance in the function of CCO, compared to Asp271 and Arg234, which are very
highly conserved. The importance of these residues in holding the protein together can be

assigned as Asp271(I) > Arg234(II) > Lyle3(III).

What Forces Are Important in Holding the Protein Together?

In the R. sphaeroides crystal structure, the distances from the closest 2 nitrogens
of Arg234(II) to the 2 carboxyl oxygen atoms of Asp271(I) are 3.0513. and 3.03A
respectively (Figure 2.2 & 2.9). The two residues are not facing each other head-to-head;

the side-on arrangement allows for a stronger interaction. The distance from the closest

74

Figure 2.9 Conformation Comparison of Possible H2O and/or H+ Exit for R.
sphaeroides (A) and Bovine (B) Cytochrome C Oxidase. Purple, orange and cyan
represent residues involved with the exit pathway from subunit 1, II, and 111 respectively.
The figure was created using Insight II with coordinates for R. sphaeroides CCO (A) from

Svensson-Ek et al., 2002 and for bovine CCO (B) fi'om Tsukihara et al., 1995.

75

A, lisp/lamina:

R234

   

1.702
‘J{ KMB

Sub II

Bovine

4.7721,- 24.62A

d:
9 i =.
i.
.31. ,1 I
‘ . 4" ... T:
. ,.-“' VHF ' ;
t. .2 .. ' \-
-. ‘\ i“
\
. 5‘

 

76

oxygen atom of Asp271(I) and e-nitrogen of Lyle3(III) is 2.67A. These latter two
residues are oriented in a way that is favorable for hydrogen bonding, by aligning the O-
atom of Asp27] and the N-atom of Lyle3 in a head-to-head conformation. Because of
relatively strong opposite charges and the short distance among them, the ionic
interaction can be very strong and may be the dominant force to hold the subunits
together. This does not exclude the existence of water molecules connecting the residues
together, although they are not resolved in the 2.3 A resolution X-ray structure
(Svensson-EK et al., 2002).

From the crystal structure of bovine CCO (Tsukihara et al., 1996), we can see that
Arg234(II) (R. sphaeroides numbering) is in a conformation to form a hydrogen bond
with Asp271(R sphaeroides numbering) (Figure 2.9). However, the distances from the
closest N-atom of Arg234(II) to the carboxyl oxygen atoms of Asp271(I) are 4.77A and
4.62A respectively, too far for a direct hydrogen bond. Even though the crystal structure
does not reveal water molecules, we can assume that either a water molecule or ionic
interactions are playing an important role at this position.

In both structures, the interaction between Lyle3 and Asp27] should be stronger
than the interaction between Asp271(II) and Arg234(I) because of the shorter distance.
From the data shown above, we know that changing the charge on any of the three
residues to neutral, by site-directed mutagenesis, leads to dissociation of the protein.
Thus, this charged triad is clearly important in the stability of the subunit 1, II, III

association.

77

Signaling Motif

As previously discussed, Asp271 is the first residue in the conserved DPxxY
motif, which is also highly conserved in the transmembrane segment 7 of G-protein-
coupled receptors. (Wess et al., 1993; Zhang & Ewinstein, 1993; Lou, Zhang &
Weinstein, 1994; Van Rhee & Jacobson, 1996). This motif has been shown to play an
important role in the signal transduction function of G-protein coupled receptors.

In CCO, the DPxxY motif is located at the periplasmic end of helix VI of subunit
1. This helix is critical in the function of CCO, containing several important residues for
oxygen reduction and proton pumping (Figure 2.3). These residues include His-284, Tyr-
288, Glu-286 & Ser-299. Mutagenesis studies and crystal structures have shown that His-
284 is one of the Cug center ligands in the active site (Shapleigh et al., 1992; Iwata et al.,
1995; Tsukihara et al., 1995). This is involved in an unusual covalent linkage with Tyr-
288 whose hydroxyl side chain is located within hydrogen-bonding distance of the
hydroxyl group of the farnesyl side chain of heme C13. The tyrosine residue is highly
conserved among the heme-copper oxidases and has been postulated to participate in the
oxygen reduction process by donating an electron and proton to form a neutral tyrosine
radical (Konstantinov et al., 1997; Proshlyakov et al., 2000). Reprotonation of the
tyrosine is thought to occur via the K-channel, which is postulated to be a proton intake
channel. G1u-286 is situated in the interior of the protein, the likely end point of the D-
channel for protons. It has been mutated to aspartate and glutamine (J unemann etal.,
1999). E286D retained 50% of wild-type catalytic activity, while E286Q retained less
than 0.01%. The observation implies a “glutamate trap” model for proton translocation.

Ser-299, toward the inside end of helix V1, is near the entry point of K-channel. Mutants

78

at this residue inhibit activity and help explain the function of Lys-362 and oxygen
reduction (Branden M., 2001; Sigurdson H., 2002). Therefore, because of the importance
of residues on helix VI in the function of the oxidase, both in proton movement and at
the heme a3 — CuB binuclear site, it is reasonable that slight conformational movements of
helix VI could cause drastic changes in the function of oxidase.

Because the crystal structures of both bovine and R. sphaeroides reveal different
conformations for Arg-234(II) and Asp-271(1) (Figure 2.9), this could show the
flexibility of the interaction between the two residues. This flexibility could be essential
for them to act as the gate of the water channel and/or possible proton exit. A signal from
the chemical changes at the active site could be transmitted through helix VI to induce a
conformational change at the external loop region, opening or closing the channel.
Unfortunately, the instability of the mutants produced at Asp27] do not allow us to test

this model.

Disulfide Bonds

Because changing Asp-271 causes dissociation of the subunits of the protein,
further research can be conducted by making a double mutant in which Asp-271 and Arg-
234 are both made into cysteines. This might allow a disulfide bond to form between
them. If the interaction between them is important in the structural stability of wild-type
oxidase, binding them together should help hold the protein together, but the loss of
flexibility might lead to inhibition of the protein. By connecting this interface of subunit I
and subunit II, the function of the area in both regulation and the water exit channel may

be elucidated.

79

Conclusion

Even though the dissociation of the protein after neutralizing the charge on Asp
271 prevents us from further analysis of the functionality of the residue, the experimental
results still helped us to understand the following:

Because the protein disaggregates with the removal of any one of the charges on
Asp 271 from subunit 1, Arg 234 from subunit II and Lys 103 from subunit 111, these
charges are of critical importance in holding the protein together. Because these charges
are relatively strong, ionic interactions could be the major force to hold the protein
together. This is an unexpected result, given the large hydrophobic interfaces between the
three subunits that would be expected to be more significant.

As the first residue of the highly conserved motif D(/N)PxxY, Asp 271 could play
a role in conformational transmission along helix VI of subunit I to regulate the events in
the protein, but the mutants examined so far do not provide any information on this

question.

80

Chapter 3

The Cross-Linking Analysis of Cytochrome C

and Cytochrome C Oxidase

81

Introduction

The electron transfer process between C-type cytochrome and CCO in the
respiratory chain is generally understood to require the transient formation of a CC/CCO
complex. There is considerable evidence that the formation of the complex is guided by
electrostatic interactions among some positively-charged residues on CC and some
negatively-charged residues on subunit II of CCO, facing the periplasmic side of the
membrane (Figure 3.1 & 3.2). This is discussed in detail in chapter 1.

Most of the C-type cytochromes are small (12kD), stable, soluble and easy to
handle. For mitochondria of higher eukaryotes, a single soluble and highly conserved CC
operates as the electron carrier between complex III (cytochrome [)6] complex) and
complex IV (cytochrome C oxidase). This is not the case in the bacterial system. Under
different growth conditions, bacteria normally use several aerobic and anaerobic electron
transfer chains. This requires variety and specificity of the electron transfer carriers under
different circumstances. Therefore, there is more than one cytochrome C which function
in almost the same way but with different electron transfer partners.

R. sphaeroides, our major research organism, has mainly two growth conditions.
Under photosynthetic growth conditions, cytochrome C2 or isocytochrome C transfers
electrons between the photochemical reaction center and b6] complex. Under respiratory
growth, a membrane bound cytochrome Cy has been found to function as the electron
carrier between the 061 complex and cytochrome C oxidase. This cytochrome Cy, unlike
that of R. capsulatus, cannot function effectively as an electron carrier between the

photochemical reaction center and the b0] complex. (Myllykallio et al., 1999). A recent

82

Figure 3.1 The Interacting Lysine Residues from Horse CC and Carboxyl Residues
from R. sphaeroides CCO in Proposed CC/CCO complex. CC and CCO are shown in
purple and yellow respectively. The oxygen and nitrogen atoms involved with the
docking are shown in red and blue respectively. The figure was created using the Insight

11 program with coordinates from (Dr. Brandon Hespenheide, personal communication).

83

 

OoU

F

 

84

Figure 3.2 Space-Filling Models of Subunit II Soluble Domain of CCO (A), Horse CC
(B), and R. sphaeroides CC2 (C). All of the CCO residues that are proposed in complex
formation are highlighted in color. The residues in CCO and CC that are proposed to
interact with each other are highlighted in the same color. The structure of CCO is from
P. denitrificans (1ar1.pdb) with R. sphaeroides numbering, horse CC from 1hrc.pdb, and

R. sphaeroides CC2 from 1cxa.pdb. This figure was made by Dr. Yuejun Zhen.

85

 

NQU hmfimfikw§~mmh .z .0 .90 09-6: .m—

oco 3...: £5
a. 528: 2.38 .<

86

study indicated that cytochrome C2 also participate in the electron transfer process under
respiratory growth as well (Daldal et al., 2001).

The low spin R. sphaeroides cytochrome C2 has a highly similar crystal structure
to that of bovine CC. Even though it has almost all of the residues that are conserved in
bovine CC which are critical to keep the overall structure, it has a lower pI value of 5.5
compared to 10 for bovine CC (Meyer & Cusanovich, 1985), and residues Lys-13 and
Lys-72 (bovine CC numbering), which have been shown to be important charges for
bovine CC to bind to CCO, are absent in the C2 cytochrome (Figure 3.2). Also, R.
sphaeroides cytochrome C2 is larger, 14kD versus 12.5kD, because of some insertions.
The absence of the two lysine residues may change the orientation of binding and likely,
as a consequence, have lowered the binding affinity and electron transfer efficiency
(Zhen et al., 1999). Its high redox potential of 356mV, versus 260mV for bovine CC, may
also contribute to the lower reactivity toward R. sphaeroides CCO, about one third the
turnover rate of horse heart CC.

Although there is evidence that R. spheroides cytochrome C2 is the electron carrier
between the bcl complex and the reaction center (Errede & Kamen, 1978), cytochrome C2
may not be the physiological electron donor for CCO. However, because cytochrome C2 is
soluble and can be overexpressed, while the possible physiological electron donor,
cytochrome Cy, is membrane-bound and not available yet, our research is based on the use
of R. sphaeroides cytochrome C2. As horse heart CC is a very efficient electron donor to
both bovine and R. sphaeroides CCO, it has also been used.

With its 3 core subunits highly similar to the 3 core subunits of its mitochondria

counterpart, bacterial CCO from R. sphaeroides has been used to observe the chemical

87

and biological fimction of CCO because of its smaller size, 4 subunits compared to 13
subunits in bovine CCO, and accessibility to mutational analysis.

As CCO is a membrane protein and CC is soluble, it has proven difficult to obtain
the crystal structure of the complex for CCO and CC (Yoshikawa, personal
communication). Therefore some indirect approaches have been taken to understand its
nature. Here, using a prediction from the computational modeling for docking horse CC
onto bovine CCO (Roberts & Pique, 1999), research was conducted to bind the electron
donor-acceptor partners together using engineered disulfide bonds. This could confirm
the computational model of the docking of CC on CCO, and possibly assist the

crystallization process.

Materials and Methods

Materials Refer to Materials section in chapter 2.

Polymerase Chain Reaction (PCR) In this method, double-stranded DNA from
plasmid pYJ 006 for Y159 mutants or pYJ 302 for A213 mutants was used as the template
for the reaction. For each mutant, four primers were made, two of which were
complementary to the flanking pUC19 sequences located at the end of the plugged-in
CCO sequence, equivalent to the “M13 universal and reverse” primers. The other two
primers are partially complementary to each other and to the template since they carry
mutations. High fidelity Pfu polymerase was used to extend the primers in the presence
of nucleotides and templates. The PCR product was cloned into pUC19 and sequenced
before cloning back to the coxII/III operon. The following primers were used to generate

the corresponding mutants, with the mutated codons underlined:

88

Y 1 59C: TCGTTCGAGAGCLQQATGATC

Y1 59L: TCGTTCGAGAGCflATGATC

Y1 59T: TCGTTCGAGAGCAQQATGATC

A2 1 3C: GGTGACCGGTLQQGACGTGATCC
A21 3 E: GGTGACCGGTQAAGACGTGATCC
A2 1 3 Y: GGTGACCGGTIA_C_GACGTGATCC
A21 3T: GGTGACCGGTAQAGACGTGATCC
A21 3 S: GGTGACCGGTIQQGACGTGATCC

Construction of the Y159 and A213 mutants All the mutated oxidases were
constructed using the overlapping extension PCR method (Ho, et al., 1989) addressed
above. After the sequencing confirms the insertion of the mutant, a procedure is followed
to put the mutated gene into overexpression plasmid pRK-pYJ 123 H. The mutated
oxidases were overexpressed and purified using Ni2+-NTA affinity column
chromatography as described in chapter 11.

Purification of R. sphaeroides cytochrome C2 The his-tagged R. sphaeroides
cytochrome C2 was isolated from a cytochrome C2 overexpression strain, pC2P404.1
(Brandner, et al., 1989). The cells were grown photosynthetically to maximize
cytochrome C2 expression, and harvested at late exponential phase. The harvested cells
were resuspended in 50mM KH2PO4, 1mM EDTA, 50mg/ml DNaseI and 50mg/ml
RNase. The resuspended cells were broken by two passages through a French pressure
cell at 20,000psi. Unbroken cells and debris were removed by centrifugation at 30,000g
for 30 minutes. The supernatant was centrifuged at 200,000g for 1.5 hours to remove the

pellets of the membranes. The supernatant was mixed with Ni2+-NTA resin in a ratio of

89

lml of packed resin for every 2.5mg of CC. Ten mM imidazole was added to prevent
nonspecific binding. The mixture was stirred at 4 ° C for lb and then poured into a
column. The column was washed with 10mM Tris-HCl, pH8.0, 40mM KCl, 10mM
imidazole, until the flow-through was clear of color and turbidity. The enzyme was eluted
slowly with 50mM histidine, 10mM Tris-HCl, pH8.0, 40mM KCl. The pink colored
fractions containing CC552 were pooled together and washed 3 times with 10mM Tris-
HCl, pH8.0, 40mM KCl by centrifugal filtration in Ultrafree-l 5 centrifugal filters to
reduce the imidazole level to less than 1mM.

Activity Assay Refer to the Activity assay section of chapter 2.

Fast Electron Transfer Measurements by Flash Photolysis These experiments
were conducted by Dr. Frank Millett’s lab at University of Arkansas. In order to measure
the rate of electron transfer with a 1:1 complex between CC and CCO, the transient
absorbance measurements were conducted in a 1 cm glass microcuvette by flash
photolysis of 300 pl solutions containing about IOuM CCO and equal amounts of
ruthenium labeled CC (Millett & Durham, 1991). Ten mM of aniline and 1mM of 3-
carboxyl-2,2,5,5-tetramethyl-l-pyrrolidinyloxy free radical (3-CP) were used. Ru-39-CC,
a horse CC derivative labeled with the inorganic complex ruthenium trisbipyridine
(Ru(bpy)3) at position 39, was used in the experiments. Once the laser is flashed on the
sample, electron transfer from CC to the Cu», center of CCO was measured by tracking the
decrease of absorbance at 550nm for Ru-39-CC, as well as 830nm for CuA reduction, and
the increase of absorbance at 605 nm for the heme a reduction. By varying the ionic
strength of the solution from 5mM to 300mM, the ionic strength dependencies for

electron transfer were determined.

90

Electron paramagnetic resonance (EPR) spectroscopy These experiments were
done as described in Hosler et al., 1992. The exception is that the solutions of the samples
were 50mM KH2PO4, 1mM EDTA, 0.1% LM, pH 7.2. The high Mg/low Mn sample was
made by growing R. sphaeroides cells in a high Mg/low Mn Sistrom’s media (Hosler et
al., 1995). The EPR spectra were taken with a Bruker ESP-300E spectrometer. This
experiment was conducted by Bryan Schmidt.

Reconstitution of Cytochrome C Oxidase All glassware was rinsed with ethanol
and de-ionized water to remove any residues of detergent. Asolectin was suspended to
40mg/ml by sonicating in 1% cholate, 75mM Hepes, pH 7.4 at 0°C under the protection
of argon. The suspension was centrifuged at 10,000 rpm for 15 min at 4°C to remove
titanium particles. After 4% cholate was added to the purified CCO, the enzyme was
added to the above resuspension buffer to a final concentration of 0.65nmol/mL and
dialyzed at 4°C with rapid stirring in 100 volumes of 75mM Hepes, 14mM KCl, 0.1%
cholate, pH 7.4 for 6 hours; 100 volumes of 75mM Hepes, 14mM KCl, pH7.4 for 12
hours; 100 volumes of 50mM Hepes, 24mM KCl, 15mM sucrose, pH7.4 for 12 hours;
and 500 volumes of SOuM Hepes, 44.6mM KCL, 45 mM sucrose for 12 hours.
Measurement of oxygen consumption was determined for the reconstituted enzymes in
order to determine the respiratory control ratio (RCR), which is a test of whether the
vesicles are intact and the enzymes are inserted correctly (Qian et al., 1997; Fetter et al.,
1995)

Proton Pumping Assay This assay was conducted on an Olis Rapid-Scanning
Monochromator (RSM). Horse CC was fully reduced by adding dithionite, and purified

with Sephadex G-75 column in 1mM Hepes, and 44.6mM KCl, pH 7.4. CCO vesicles

91

 

were diluted to about 200nM. Five uM of horse CC (final concentration) was used for
each assay (about 10 turnovers). Two hundred uM phenol red was added to detect the
proton pumping at 557nm. Measurements were taken at 1000 scans/sec. Cytochrome C
oxidation was followed at 550nm and the reduced minus oxidized spectra were fit by
Global analysis to obtain the first-order rate. The reaction between CC and CCO were
carried out under three conditions:

1) Without ionophore (controlled)

2) With 2 uM valinomycin

3) With 2 uM valinomycin and 10 uM CCCP (uncontrolled)

The isosbestic point of the reduced and oxidized spectra of cytochrome C was
determined for each reaction. This wavelength is used for the proton pumping assay. The
acidification and alkalinization rates were analyzed with a non-linear curve fitting in
Microcal Origin program. The small mixing artifact was subtracted and the backleak in
the kinetic trace with valinomycin (proton pumping) was fitted and subtracted from the
individual scan to eliminate the contribution of proton leakage through the membranes
which competes with the proton release.

Cross-Linking Assay Wild-type and cysteine mutants of cytochrome C2 and CCO
of R. sphaeroides were purified as mentioned above. After combining CC and CCO, the
mixture (final concentration of 20uM each with total volume of 50111) was passed through
a Sephadex G-75 0.5cm by 20cm column. The flow-through fractions were scanned with
visible spectrometry to detect the elution volumes for CC, CCO and possible CC/CCO
complex. With individual cysteine mutants, 1mM dithiothreitol was present to keep a

disulfide bond from forming. After mixing CC & CCO, the mixture went through either

92

dialysis or a G10 spin column to get rid of the DTT before being loaded on the G75

column.

Results

Design of Mutants

The analysis of the model complex of CczCCO to predict where a disulfide might
be inserted was conducted by Brandon Hespenheide. Using the original computationally
modeled complex for horse CC and bovine CCO (Roberts & Pique, 1999), the R.
sphaeroides CCO was superimposed onto the crystal structure of bovine CCO in the
complex. The 3-D structural superposition of subunit I and II from R. sphaeroides and
bovine CCO were conducted independently using the DALI server (www2.ebi.ac.uk/dali).
Combining the alignment information from the server, superposition was performed
again using InsightII’s superposition command. The RMSD for all backbone atoms in
this superposition was 1.246A (Figure 3.3).

With this superimposed complex model, search for any pair of Cu atoms between
the R. sphaeroides CCO and the docked CC whose distance is between 4.0A and 7.5A was
conducted. By inspecting visually for potential appropriate Ca-Cg vector direction,

Cc: 12-CCO:159(subunit II) has been identified. Even though other sites do not have the
proper geometry, disulfide bond formation is still considered possible for the following 3
pairs: Cc:77-CCO:183(subunit I), CC283-CCO:213(subunit II), Ccz83-CCO:214(subunit II)
(Figure 3.4). As Asp214 is highly conserved and observed to be a significant charge for

docking of Cc, the pair Ccz83-CCOz214(subunit II) was not considered a good candidate

93

Figure 3.3 Overall Structure of Proposed Horse CC I R. sphaeroides CCO Complex.

Cyan, yellow, blue and orange represent subunit 1, II, III and IV of CCO respectively. CC
is shown in magenta. The figure was created using the Insight II program with

coordinates from (Dr. Brandon Hespenheide, personal communication).

94

 

95

Figure 3.4 Interaction Interface of a Possible Complex of Horse CC and R.
sphaeroides CCO and the Residues that Could Possibly Form a Covalent Bond
between Them. CC and Subunit I & II of CCO are shown in magenta, cyan and yellow
respectively. The residues in CCO and CC that are proposed to be sufficiently proximal if
changed to cysteine to link covalently are highlighted in the same color. The figure was
created using the Insight 11 program with coordinates from (Dr. Brandon Hespenheide,

personal communication).

96

 

97

for cross-linking. Because the pair Ccz77-CCO:183(subunit I) of CCO is at the edge of the
speculated docking site, the possibility of the cross-link might not be as high as the pairs
closer to the center of the docking site. Therefore, 2 pairs, Cc:12-CCO:159(subunit II) and
Cc:83-CCO:213(subunit II) was chosen to perform the cross-linking experiment.

This research is a cooperative project with another graduate student in Dr.
Shelagh F erguson-Miller’s lab, Yasmin Hilmi. My focus was to make mutants in R.
sphaeroides CCO and Yasmin made mutants in cytochrome C2 of R. sphaeroides.
Therefore, this thesis concentrates on mutagenesis on the CCO and characterization of
these mutants. Further cross-linking experiments are also reported.

Because Tyr159 is in a region where a stack of aromatic residues, including
Tyrl44 & Tyr141 was found (Figure 3.5), threonine and leucine mutants were also made
in addition to the cysteine mutant. If this aromatic alignment is important functionally,
the Thr and Len mutants will help us to understand whether the aromaticity of the
benzene ring or the hydroxy group are important.

Pro83 (Pro71 in horse CC) in CC2 is a highly conserved residue and it is next to
Thr84 (Lys72 in horse CC). In horse Cc, Lys72 plays a significant role for the docking of
CC onto CCO. Because of these observations, a glutamate mutant was made along with
the cysteine mutant to change the charge distribution in this region. This might give us
more information about the docking mechanism.

Several mutants, including the cysteine mutant, were also designed at the Ala213

position to test the functional significance of this residue.

Optical characteristics of mutants

98

Figure 3.5 Tyrosine Alignment at the Possible Interface of CCO with Its Redox
Partner CC. The highly conserved Tyr-141, 144, and 159 in CCO are shown in blue.
Other highly conserved aromatic residues are shown in purple. Trp 143 is shown in red.
Yellow balls are Cu», atoms. The figure was created using Insight II with coordinates for

R. sphaeroides CCO from Svensson-Ek et al. , 2002.

99

0

m3?

I Smm
35M 3:

film V\

mva
5:.»
aw~> N
mafia
52
9.7%

100

All of the mutants have been successfully expressed and their spectra in the R.
sphaeroides membranes are shown in Figure 3.6 & 3.7. From the height and position of
the or-peak, we can tell that all mutants are well expressed and similar to wild-type
spectrally. Because of time constraints, we focused our research on the Y159C and

Y159T mutants. More data is presented below.

CuA/heme EPR analysis

For the X-band EPR spectrum of R. sphaeroides CCO, the signals at g = 2.83,
2.31 and 1.62 arise from the low spin heme a, the signal at g = 6.0 from heme a and the
signals at g = 2.0 and 2.19 from CuA (Figure 3.8) (Hosler et al., 1992). The EPR spectra
of Y159C and Y159T mutants are similar to that of wild-type, except that both mutants
have a peak at g = 2.35. This could be some other type of heme C than the type in wild-
type CCO or due to some impurities in the sample. The sample is only Ni2+-NTA column
purified. As the His-tag used to purify CCO is in subunit 1, the purified oxidase could
contain some extra subunit I due to its binding to the NizI-NTA. This subunit I could
have lost its heme 03 as sometimes observed (Jon Hosler, personal communication),
making the signal for heme a different. The EPR spectra indicate that the metal centers in

these mutants are mainly intact. This is consistent with the result from the visible spectra.
The activities of mutant enzymes

It is important that the mutants used for cross-linking should not affect the

conformational state of CCO. Additionally, in order to see if the cross-linked CcOzCC

101

Figure 3.6 Absolute Visible Spectra for Reduced Y159 Mutants. The proteins
(purified with Ni2+-NTA column) are reduced with dithionite and measured in a Perkin-

Elmer Lambda 4.0 UV/V IS Spectrophotometer. These measurements were conducted in

100mM KH2PO4, 0.1% LM with pH 7.2.

102

A£5 £mc2m>m>>

cow

 

l,
”
I
I

T1

'fl11l

I

I'le1

1

 

- .ull'll‘ltl‘n‘y \I.I..I‘\I-\I(li|l.l"v \ I I,
.II.

I.

I‘lll‘
\ l

.- m m P > ...-.-----l
0% Cr ..........
0 of»? U: >> i

oo.o
I Pod

. mod

I #06
I no.0
I mod
1 No.0
I mod
I mod
I 070
I _._..o
I N70

 

 

GOUBQJOsqV

103

Figure 3.7 Visible Spectra for A213 Mutants in Solubilized Membrane. These are
reduced-minus-oxidized R. sphaeroides membrane spectra. The membranes are reduced
with dithionite and measured with membranes oxidized with ferricyanide as a reference
in order to obtain the difference spectra in a Perkin-Elmer Lambda 4.0 UVN IS

Spectrophotometer. These measurements were conducted in 100mM KH2PO4, 0.1% LM

with pH 7.2.

104

0N0
_

A85 £mc2¢>m>>

com owm omm ovm ONm
. _ _ . _

 

 

P Dr L D b h
\_l
s
\
.\
.s
x I
x
,. \
x
\
x
\ T
\
. \
x c
\ \
\ \.l x. \
\«\ I \ \\ l
I l. I I 1.1 I I a. I, .x \x
-1. c .\
. I \ x
\ I I. ‘ \ I I z \. .
. \ .rn/. — \ I ll.’ \ \ I
it: /. It} a .1. 1 . ..
\ / I: \ \ / . /
it .- \r r J . . . z . \ k
I I .’ .\. x x z. .. . , I \ \
I . \.. \ x.’ . .\ z ( \
.. I I . \ z .
. Lu.-. \ \ a) q... z \
1’.- x x .D ... trlfl- \
z I . \ J ‘ x
1., .\ x f n. ,, x
1., \x J ‘. /< \
/. . \x “a. mm
or .. \\ .u. s.
d \ 3 u. I
.z N .w ..
f x. .w _
. x .
a, x. .s .-
. x 1— _
if \. A. . ~
.. \\ J. ...
m \ x. .b. ._ l
.. \\ ...o ._
p \ ...a . . ~
w \ \ —.. .' s _
. c \ r... \ l. . _
x ‘ .po.’ .i l \ ~
. . . . .
Fm _.N< ------- .. . 3.: . . 1
x \ z I\ \ I ~
\ x. I c 3 .
'''''''''' I (I | u I \ I \
.s x \ / .
c /\
........... . l
>m rN< .
~
0m _.N( I .
s
.\ I
.\
r'

NOAV

vOAY

OOAU

:90

NOAU

mogu

eoueqiosqv

105

Figure 3.8 Cu EPR Spectra of Y159 Mutants. EPR spectra of 30 uM oxidases in
50mM KH2PO4, 1mM EDTA, 0.1% LM, pH 7,2 were recorded at X-band using a Bruker
EP=3000E spectrometer. The related g values are indicated in the plots. The EPR spectra
were recorded at 10K with the following settings: microwave frequency 9.4831GHz for
Y159T and 9.4809 GHz for Y159C; microwave power 2mW; modulation amplitude
12.5G; sweep time 405; The cells were grown in high Mg / low Mn media. The data

shown is the average of three scans.

106

a.

 

 

 

 

000V 000” OOON COOP
u .. _ t _ . _ oooom-
..
oo.~ .. -
.. me _.> .---...-.----.-.. I COCON-
Omm _.> i
T. 0000 F:

 

 

 

 

 

 

 

 

107

eouquosqv

complex is correct, it is important to assess the activity of the mutant under normal CC
binding conditions. Since R. sphaeroides CCO shows high activity with horse CC, horse
Cc has been used in this part of the experiment. The maximum turnover numbers of wild-
type and Y159 mutants have been measured using an excess of horse CC as the substrate
in polarographic assays (Table 3.1). With the turnover rate of 1,700e'/sec/aa3 for wild-
type CCO, Y159C and Y159T have rates of only about 396 and 520 e'/sec/aa3. These
slower turnover rates may be the result of subtle changes in the docking of Cc on CCO, or

to the presence of some interfering contaminant forms.

Laser fiash photoreduction study of kinetics of Y159C mutant

The flash-induced photoreduction technique measures the electron transfer
activity from CC to Cup, of CCO within the complex, as well as the second order reaction
between CC and CCO from solution. Of the intracomplex and 2“d order reaction phases,
the latter phase normally shows a bell-shaped dependency on the ionic strength of the
solution. From the result shown in Figure 3.9, the two mutants show similar ionic
strength dependence as the wild type CCO but with lower activity, with Y159T and
Y159C showing about 1/6 and 1/8 of the wild type activity for the fast phase electron
transfer, respectively. This is consistent with, and may account for, the activity

measurement and steady-state kinetics result.

Proton-pumping assay
A normally functioning CCO is capable of pumping one proton per electron

transferred. The stopped-flow analysis gives the proton pumping results for both Y159T

108

 

mmm Emma?

 

 

 

 

cam Damtw
85 2:9 2;»
Ana-“Bouncy

3a.. 3.55:9

 

 

 

3:33). am; a. 8.2232226 £33. 3. 2.3.

109

Figure 3.9 Ionic Strength Dependent Electron Transfer from Ru-39—CC to
Cytochrome c Oxidase. The rates (kc-,5) are measured from the ruthenium-kinetics

assays, with rates for both the fast and slow phases plotted together.

110

1‘

 

 

 

 

 

.82 2:.
8a SN 8" c2 2: on a
_ _ p _ p L P
. 2
mg“ 2:
m 22
n- 82:
cacao,

 

 

can; .25 82> 3:22..
.25 was... 83.5 .n_.m .23 oo 83m 83..

(Peas) are:

111

and Y159C mutants. (Figure 3.10 & 3.11) The results show that both mutants pump
protons in a similar way to that wild-type oxidase, with the normal 1:1 stoichiometry 1
We when valinomycin is added. The usual increase in rates of cytochrome c oxidation
are also seen as the membrane potential is removed with valinomycin and the pH gradient

is relieved with CCCP as an uncoupler.

Complex Formation If

The CCO mutant was designed to test the accuracy of the predicted CC/CCO
complex by formation of a covalent bond between the C12 mutant of R. sphaeroides
cytochrome C2 (CC12) and the Y159C mutant of R. sphaeroides cytochrome C oxidase
(CC0159). The experiment was conducted as described in the Materials and Methods
section. Both dialysis and spin column techniques were used to remove DTT from the
samples. Four other experiments were done with (1) Cc0159 and Cc12; (2) CC0159 and
wild-type cytochrome C2; (3) wild-type CCO and CC12; and (4) Cc0159 and CC12 with
DTT in the sample throughout the experiment.

Sephadex G-75 column retains molecules of 3 — 80kD. This works well when
separating CC2 (14kDal) from CCO (126kDa1) as shown in Figure 3.12. If a complex
(l40kDal) of CC and CCO was formed, it would be expected to be resolved on the column
from CC but not from CCO.

The results show no evidence for formation of complexes in any of these samples,
including the case where the two Cys mutants are mixed in the absence of DTT (Figure

3.12).

112

Figure 3.10 Proton Pumping Analysis For Y159C Mutant. Kinetic traces for

 

cytochrome C oxidation at 550nm (A) and phenol red changes at 557nm (isosbestic point
of cytochrome c) (B) were collected. The measurement is conducted according to Hiser et

al., 2001.

113

Absorbaneo (557nm)

1.0 q
0.8 - Controlled
A --------- Valinomycin
g Y159C-1 -------------- Uncontrolled
§ 0.6 - H
g... i
O . I:
5 .
0.0 -
1 I I I ' l ' I fl
5 10 15 20
Time (Sec)
001 _ —-—-Valinomycin
----------- Uncontrolled
Y1 59C-1
0.00 '
H+le' ~1
-0.01 -
-002
I v I v I v I v
0.0 0.2 0.4 0.6 0.8 1.0
Time (Sec)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

114

Figure 3.11 Proton Pumping Analysis For Y159T Mutant. See Figure 3.10 legend.

115

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.24
Y159T
g 0.8-
.. —— Controlled
3 q --------- Valinomycin
5 Uncontrolled
E 0.4-
0
2
.
0.0-
u l ' I ' l I I I
5 10 15 20
Time (Sec)
0-02" Y159T
o 01 Valinomycin
E . - ., ~Uncontrolled
,5 .
3
000
é
a 0
+ -
3 -0.01- H /e N]

 

 

 

 

 

 

-0.02 -+ r
wVV‘V‘» "M\/%,~’\~J\JV\/.«m«j AV\_,’~MM”\VA-“\/AVWN
I ' I ' V ' 1 I
0.0 0.2 0.4 0.6 0.8
Time (Sec)

116

1.0

Figure 3.12 Gel-filtration for Cytochrome c Oxidase Cysteine Mutant (Y159C) with

Cytochrome c Cysteine Mutant (N 12C). The absorbances for cytochrome c and
cytochrome c oxidase were plotted against elution volume (pl). Each point on the plot
represents the actual absorbance at 550nm for cytochrome c and 606 nm for cytochrome

c oxidase of the specific elution fraction (50 pl).

117

 

23 0E2o> cozém
comm ooom oomw coop

 

 

 

 

 

album/0 i» \ \\I

D
f D

ONFO 0cm Ommr> E 000 +1
ONFO new 000 _.> E 00 I'le b,

33:00 E 000 +

.9950 E 00 '0'

 

 

000.0

N000

v00.0

000.0

000.0

0_.0.0

118

However, in all these cases the Ccz mutant forms did not have the missing key
Lysl 3 residue put into the CC2 to make it resemble the horse Cc upon which the model
was based. Yasmin Hilmi has made the double mutant of N12C/Ql 3K of R. sphaeroides
C02 and conducted a similar crosslinking experiments. The result is shown in Figure 3.13.

From this result we can see that there is a significant amount of Cc eluting from the

column with CCO. Compared to the result in Figure 3.12, it suggests that covalent fl
complex was formed when the Ly313 positive charge was present to properly orient the :4»
complex.

Discussion

 

Are the Changes Introduced by the Mutation at Position Y-159 Disruptive to the
Structure and/or Function of CCO?

The data acquired from visible spectra, EPR, activity, flash photolysis and proton-
pumping assay have shown that except for the lowered activity compared with that for
wild type CcO, the Y159C and Y159T mutants are not significantly different from wild
type in their metal and heme sites, proton pumping mechanism and binding behavior to
their redox partner, Ccz.

Because Y159 is the third tyrosine residue in the highly conserved Tyr14l-
Tyrl44-Tyr159 triad (Figure 3.5), which is part of the hydrophobic stretch surrounding
the Cup, entry site, it could be important for electron transfer from Cc to the Cup, within
the complex. This is consistent with the slower electron transfer rate observed from the

flash photolysis assay. Yl44 in this triad has also been mutated into more conservative

119

Figure 3.13 Gel-filtration for Cytochrome c Oxidase Cysteine Mutant (Y 159C) with

Cytochrome c Cysteine Mutant (N 12C/Q13K). The absorbances for cytochrome c and

cytochrome c oxidase were plotted against elution volume (pl). Each point on the oxidase
plot represents the absorbance at 444 nm for cytochrome c oxidase of the specific elution
fraction (50 pl). For the cytochrome c plot, data were corrected and the absorbance of the

oxidase at 416 nm was subtracted using this formula:

Cytochrome c absorbance at 416 = absorbance at 416 — (absorbance at 444 nm X 0.32)

where 0.32 = absorbance at 416nm / Absorbance at 444 nm for oxidase alone.

This Figure was made by Dr. Yasmin Hilmi.

120

 

:5 2.. 2.5 Seam
oomw ooor com com 00%
J/.. 0 _ _
[In/oflr/
/ \

a: e: a 6220 8 IT. .<

a: 3... a 6&5 03 lol

 

 

   

 

 

 

0.0

v.0

Nd

md

10

md

eoueqxosqv

121

residues such as phenylalanine, histidine and serine. These mutants also show lower
electron transfer rates from Cc to Cu (Personal communication with Dr. Yuejun Zhen).
These data suggest that this triad could be more important in controlling the electron
transfer than in the actual docking of Cc (For a detailed discussion, please refer to chapter
1). Since the Y159C mutant has a cysteine residue externally exposed, self-dimerization
could also contribute to the slow turnover rate of the protein.

Because the mutants do not show a significant change in Cc binding, as indicated
by the ionic strength dependence of the second order reaction, the potential formation of
a disulfide bond between Cc and CCO should reflect the true docking conformation of
wild-type CcOch. If a covalently-bound complex can be obtained and even crystallized,
it would support the conclusion based on computational modeling and further reveal the

binding behavior of both redox partners.

Cation-Pi Interaction

Contrary to the fact that electrostatic interactions play a major role in the complex
of cytochrome c and cytochrome c oxidase, binding of Cc and bcl is mainly stabilized by
hydrophobic interactions between amino acid residues around the heme crevices. These
residues form a trapezoid-shaped contact area (Lange & Hunte, 2002; Hunte et al., 2002).
In the center of the binding site a planar stacking cation-1t interaction has been found
important to stabilize the complex of the redox partners. This interaction is between
Arg18 of Cc and Phe230 of cytochrome c. of be] complex, with the guanidine group

oriented parallel to the aromatic plane. This kind of stacking pattern has been found at

122

molecular surfaces of proteins (e.g. Tyr 39 0f CcP with Arg 19 of Cc) (Pelletier & Kraut,
1992) in formation of enzyme substrate complexes.

Remarkably, at position 19 (position of Lysl3 in bovine Cc) of Cc a positive
charge is highly conserved. Lysl3 in horse Cc has been shown to be important for
binding of CcO (Rieder & Bosshard, 1980; Ferguson-Miller et al., 1978; Smith et al.,
1977) and be, complex (Yu et al., 1973). From the computational modeling of the
complex of horse Cc and bovine CcO (Roberts & Pique, 1999), we found that Lysl3 of
Cc is only 3.9334. away fromTyr144 of CcO. Tyrl44 mutants were made and further
analysis can be done to explore the importance of this cation-1r interaction. With the
results from these future experiments, we can achieve a better understanding of the role

of Lys-13 in the docking mechanism.

Covalently Bound Complex and Anion Exchange Chromatography

From the cross-linking experiments, we have evidence that only when a mutation
is carried out in CC; to put a lysine at position 13 (bovine numbering), is a significant
amount of complex detected (Yasmin Hilmi, personal communication). This helps to
validate the computational docking model and it further supports the important role of
Lys-13 in the binding of Cc onto CCO.

Even though we have proved that a complex can be formed between Cc and CCO,
separation of this complex from the two proteins, Cc and CcO has not been satisfactorily
achieved. Further experiments can be conducted with F PLC-DEAE chromatography.

Because the presumed complex would have a different charge distribution on the surface

123

 

of CCO, we should be able to get a better separation of the combined complex from the

other proteins in the sample.

Conclusion:

Introduction of Lys-13 into R. sphaeroides cytochrome c; improves the inter-
molecule disulfide bond formation and thus confirms the computer model results as well
as experimental data, that emphasizes the significant role of Lys-1 3 in docking of Cc onto
CcO. If the formation of the inter-molecule disulfide bond between the two molecules
can be made more efficient and gives an active complex, the covalent complex could

assist in crystallization efforts.

124

BIBLIOGRAPHY

125

BIBLIOGRAPHY

Arnold, S, Goglia, F, & Kadenbach, B. (1998) 3,5-Diiodothyronine binds to subunit Va
of cytochrome-c oxidase and abolishes the allosteric inhibition of respiration by ATP.
Eur J Biochem. 252:325-30.

Axelrod, H. L., Feher, G., Allen, J. P., Chirino, A. J., Day, M. W., Hsu, B. T. & Rees, D.
C. (1994) Crystallization and X-ray structure determination of cytochrome-C(Z) from
Rhodobacter-sphaeroides in 3 crystal froms. Acta Crystallogr. D. Biol. Crystallogr.
50:596-602.

Babcock, G. T. & Wikstrom, M. (1992) Oxygen activation and the conservation of
energy in cell respiration. Nature 356:301-9.

Baker, 8. C., Ferguson, S. J., Ludwig, B., Page, M. D., Richter, O.-M. H. & van
Spanning, R. J. M. (1998). Molecular genetics of the genus Paracoccus: metabolically
versatile bacteria with bioenergetic flexibility. Microbiol. Mol. Biol. Rev. 62:1046-1078.

Bartsch, R. G., Ambler, R. P., Meyer, T. E. & Cusanovich, M. A. (1989) Effect of
aerobic growth conditions on the soluble cytochrome content of the purple phototrophic

bacterium Rhodobacter sphaeroides: induction of cytochrome c554. Arch. Biochem.
Biophys. 271:433-440.

Bartsch, R. G. in The Photosynthetic Bacteria. Clayton, R. K. & Sistrom, W.R., Eds.
(Phenum Press, New York, 1978) pp. 249-280.

Berry, E. A. & Trumpower, B. L. (1985) Isolation of ubiquinol oxidase from Paracoccus
denitrificans and resolution into cytochrome bc; and cytochrome c-aa3 complexes. .1.
Biol. Chem. 260:2458-2467.

Bisson, R., Steffens, G.C.M., Capaldi, R. A. & Buse, G. (1982) Mapping of the
cytochrome c binding site on cytochrome c oxidase. FEBS Lett. 1442359-63.

Bott, M., Ritz, D. & Hennecke, H. (1991) The Bradyrhizobiumjaponicum cycM gene
encodes a membrane-anchored homolog of mitochondrial cytochrome c. J. Bacteriol.
173 :6766-6772.

Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dry binding. Anal. Biochem.
72:248-254.

Branden, M., Sigurdson, H., Namslauer, A., Gennis, R.B., Adelroth, P. & Brzezinski, P.

(2001) On the role of the K-proton transfer pathway in cytochrome c oxidase. Proc. Natl.
Acad. Sci. USA 98:5013-18.

126

Brandner, J. P., McEwan, A. G., Kaplan, S. & Donohue, T. J. (1989) Expression of the
Rhodobacter sphaeroides cytochrome c; structure gene. J. Bact. 171:360-368.

Bratton, M.R., Pressler, M.A., & Hosler JP. (1999) Suicide Inactivation of Cytochrome c
Oxidase: Catalytic Turnover in the Absence of Subunit III Alters the Active Site.
Biochemistry 38: 16236-45.

Brautigan, D. L., Ferguson-Miller, S. & Margoliash, E. (1978) Mitochondrial cytochrome
c: Preparation and activity of native and chemically modified cytochromes c. Methods
Enzymol. 53:128-164.

Brzezinski, P. & Malmstrom, B. G. (1987) The mechanism of electron gating in proton
pumping cytochrome c oxidase: the effect of pH and temperature on internal electron
transfer. Biochim. Biophys. Acta 894:29—38.

Bushnell, G., W., Louie, G. V., Brayer, G. D. (1990) High-resolution three-dimensional
structure of horse heart cytochrome c. J. Mol. Biol. 214: 585-95.

Callahan, P.M. & Babcock, G.T. (1983) Origin of the Cytochrome a Absorption Red
Shift: A pH-Dependent Interaction between Its Heme a Formyl and Protein in

Cytochrome Oxidase. Biochemistry 222452-61.

Cao, J., Hosler, J ., Shapleigh, J ., Gennis, R., Revzin, A. & Ferguson-Miller, S. (1992)
Cytochrome aa3 of Rhodobacter sphaeroides as a model for mitochondrial cytochrome c
oxidase. J. Biol. Chem. 267:24273-8.

Cao, J ., Shapleigh, J ., Gennis, R., Revzin, A., & Ferguson-Miller, S. (1991) The gene
encoding cytochrome c oxidase subunit II from Rhodobacter sphaeroides; comparison of

the deduced amino acid sequence with sequences of corresponding peptides from other
species. Gene 101:133-137.

Chory, J ., Donohue, T. J ., Varga, A. R., Staehelin, L. A. & Kaplan, S. (1984) Induction of
the photosynthetic membranes of Rhodopseudomonas sphaeroides: biological and
morphological studies. J. Bacteriol.159:540-554.

Cohen-Bazire, G., Sistrom, W. R. & Stanier, R. Y. (1957) Kinetic studies of pigment
synthesis by non-sulfur purple bacteria. J. Cell Comp. Physiol. 49:25-68.

Cusanovich, M. A. (1971) Molecular weight of some cytochrome c’. Biochim. Biophys.
Acta 236:238-241.

de Gier, J. L., Schepper, M., Reijinders, W. N. M., van Dyck, S. J ., Slotboom, D. J .,
Warner, A., Saraste, M., Krab, K., F inel, M., Stouthamer, A. H., van Spanning, R. J. M.
& van der Oost, J. (1996) Structural and functional analysis of aa3-type and cbb3-type
cytochrome c oxidases of Paracoccus denitrificans reveals significant differences in
proton-pumping design. Mol. Microbiol. 20:1247-1260.

127

Donohue, T. J ., McEwan, A. G., Van Doren, S., Crofts, A. R. & Kaplan, S. (1988)
phenotypic and genetic characterization of cytochrome c; deficient mutants of
Rhodobacter sphaeroides. Biochemistry 27: 1918-1925.

Daldal, F ., Mandaci, S., Winterstein, C., Myllykallio, H., Duyck, K. & Zannoni, D.
(2001) Mobile cytochrome C2 and Membrane-Anchored cytochrome cy are both efficient

electron donors to the cbb3- and aa3-type cytochrome c oxidase during respiratory growth
of Rhodobacter sphaeroides, .1. Bacterial. 183:2013-24.

Das, T.K., Pecoraro, C., Tomson, F.L., Gennis, R.B. & Rousseau D.L. (1998) The post-
translational modification cytochrome c oxidase is required to establish a functional
environment of the catalytic site. Biochemistry 37: 14471-6.

Errede, B. & Kamen, M. D. (1978) Comparative kinetic studies of cytochromes c
reactions with mitochondrial cytochrome c oxidase and reductase. Biochemistry 1721015-
1027.

Ferguson-Miller, S. & Babcock, G. (1996) Heme/copper terminal oxidases. Chem. Rev.
96:2889-2907.

Ferguson-Miller, S., Brautigan, D. L. & Margoliash, E. (1978) Definition of cytochrome
c binding domains by chemical modification. J. Biol. Chem. 253:149-159.

Ferguson-Miller, S., Brautigan, D. & Margoliash, E. (1976) Correlation of the kinetics of
electron transfer activity of various eukaryotic cytochrome c with binding to
mitochondrial cytochrome c oxidase. J. Biol. Chem. 251:1104-1115.

Fetter, J.R., Qian, J ., Shapleigh, J ., Thomas, J.W., Garcia-Horsman, J .A., Schmidt, H,
Hosler, J ., Babcock, G.T., Gennis, R.B. & Ferguson-Miller, S. (1995) Possible Proton
Relay Pathways in Cytochrome c Oxidase. Proc. Natl. Acad. Sci. USA. 92: 1604-8.

Fishel, L. A., Villafrance, J. E., Mauro, J. M. & Kraut, J. (1987) Yeast cytochrome c
peroxidase: Mutagenesis and expression in Escherichia coli show tryptophan-51 is not
the radical site in compound 1. Biochemistry 26:351-60.

Flock, D., & Helms, V. (2002) Protein-Protein Docking of Electron Transfer Complexes:

Cytochrome c Oxidase and Cytochrome c. Proteins: Structure, Function, and Genetics.
47:75-85.

Florens, L., Hoganson, C., McCracken, J ., Fetter, J .R., Mills, D. A., Babcock, G. T. &
Ferguson-Miller, S. (1998) The Role of Magnesium and Its Associated Water Channel in
Activity and Regulation of Cytochrome c Oxidase in The Phototropic Prokaryotes
Preschek, G., Eds. (Kluwer Academic/Plenum Publishers N.Y.), pp. 329-339

128

Flory, J. E. & Donohue, T. J. (1995) Organization and expression of the Rhodobacter
sphaeroides cycFG operon. J. Bateriol. 177:4311-4320.

Garcia-Horsman, J ., Barquera, B., Rumbley, J ., Ma, J. & Gennis, R. B. (1994a) The
superfamily of heme-copper respiratory oxidases. J. Bacteriol. 176:5587-5600.

Garcia-Horsman, J. A., Berry, E., Shapleigh, J. P., Alben, J. O. & Gennis, R. B. (1994b)
A novel cytochrome c oxidase from Rhodobacter sphaeroides that lacks CuA.
Biochemistry 33 :31 13-31 19.

Geren, L., Hahm, 8., Durham, B., Millett, F. (1991) Photoinduced electron transfer
between cytochrome c peroxidase and yeast cytochrome 0 labeled at Cys 102 with (4-
bromomethyl-4'-methylbipyridine)[bis(bipyridine)]ruthenium2+. Biochemistry 30:9450.

Gregory, L. C., & F erguson-Miller, S. (1988) (1) Effect of subunit 111 removal on control
of cytochrome c oxidase activity by pH. Biochemistry. 27:6307-14.

Gregory, L. C., & Ferguson-Miller, S. (1988) (2) in Advances in Membrane Biochemistry
and Bioenergetics (Kim, C.H., Tedeschi, H., Diwan, J. J ., & Salerno, J. C., eds.) pp 301-
309, Plenum Publishing.

Hahm, 8., Durham, B., Millett, F. (1992) Photoinduced electron transfer between
cytochrome c peroxidase and horse cytochrome c labeled at specific lysines with
(dicarboxybipyridine)(bisbipyridine)ruthenium(II) Biochemistry 31 :3472.

Haltia, T., Saraste, M., & Wikstrom, M. (1991) Subunit III of cytochrome c oxidase is
not involved in proton translocation: a site-directed mutagenesis study. EMBO J.
10:2015-21.

Haltia, T., F inel, M., Harms, N., Nakari, T., Raitio, M., Wikstrom, M., & Saraste, M.
(1989) Deletion of the gene for subunit III leads to defective assembly of bacterial
cytochrome oxidase. EMBO J. 8:3571-9.

Hallen, S. & Nilsson, T. (1992) Proton transfer during the reaction between fully reduced

cytochrom oxidase and dioxygen: pH and deuterium isot0pe effects. Biochemistry
31:11853-11859.

Harrenga, A., Reincke, B., Riiterjans, H., Ludwig, B. & Michel, H. (2000) Structure of
the Soluble Domain of Cytochrome c552 from Paracoccus denitrificans in the Oxidized

and Reduced States. J. Mol. Biol. 295:667-678.

Hatefi, Y. (1985) The mitochondrial electron transport and oxidative phosphorylation
system. Annu. Rev. Biochem. 54:1015-1069.

Hill, B. C. (1991) The reaction of the electrostatic cytochrome c-cytochrome oxidase
complex with oxygen. J. Biol. Chem. 266:2219-2226.

129

Hiser, C., Mills, D. A., Schall, M., & Ferguson-Miller, S. (2001) C-terminal truncation
and histidine-tagging of cytochrome c oxidase subunit [1 reveals the native processing
site, shows involvement of the C-terminus in cytochrome 0 binding, and improves the
assay for proton pumping. Biochemistry. 40(6): 1606-15.

Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K. & Pease, L. R. (1989) Site-directed
mutagenesis by overlap extension using the polymerase chain reacton. Gene 77: 51-59.

Hosler, J .P., Espe, M.P., Zhen, Y., Babcock, G.T. & Ferguson-Miller, S. (1995) Analysis

of site-directed mutants locates a non-redox-active metal near the active site of
cytochrome c oxidase of Rhodobacter sphaeroides. Biochemistry 34:75 86-92.

Hosler, J.P., Ferguson-Miller, S., Calhoun, M.W., Thomas, J.W., Hill, J ., Lemieux, L.
Ma, J .X., Georgiou, C., Fetter, J ., Shapleigh, J ., Techlenburg, M.M.J., Babcock, G.T.,
Gennis, RB. (1993) Insight into the active-site structure and function of cytochrome-
oxidase by analysis of site-directed mutants of bacterial cytochrome — aa3 and cychrome-
bo. J. Bioenergetics and Biomembranes 25:121-136.

Hosler, J. P., Fetter, J., Teckkenburg, M. M. J ., Espe, M., Lerma, C. & Ferguson-Miller,
S. (1992) Cytochrome aa3 of Rhodobacter sphaeroides as a model for mitochondrial
cytochrome c oxidase. J. Biol. Chem. 267:24264-24272.

Hunte, C ., Solmaz, S. & Lange C. (2002) Electron transfer between yeast cytochrome be;
complex and cytochrome c: a structural analysis. Biochim. Biophys. Acta. 1555:21-8.

Iwata, S., Ostermeier, C., Ludwig, B. & Michel, H. (1995) Structure at 2.8 A resolution
of cytochrome c oxidase from Paracoccus denitrificans. Nature 376:660-669.

Jenney In, F. E. & Daldal, F. (1993) A novel membrane-associated c-type cytochrome
cyt Cy, can mediate the photosynthetic growth of Rhodobacter capsulatus and
Rhodobacter sphaeroides. EMBO J. 12: 1283-1292.

J unemann, D., Meunier, B., Fisher, N. & Rich, PR. (1999) Effects of mutation of the
conserved glutarnic acid — 286 in subunit I of cytochrome c oxidase from Rhodobacter
sphaeroides. Biochemistry. 38:5248-55.

Kang, C. H., Brautigan, D. L., Osheroff, N. & Margoliash, E. (1978) Definition of
cytochrome c binding domains by chemical modification. Reaction of

carboxydinitrophenyl- and trinitrophenyl-cytochromes c with baker’s yeast cytochrome c
peroxidase. J. Biol. Chem. 253:6502-10.

Kang, C.H., F erguson-Miller, S., Margoliash, E. (1977) Steady state kinetics and binding

of eukaryotic cytochromes c with yeast cytochrome c peroxidase. J Biol Chem. 252:919-
26.

130

Karpefors, M., Adelroth, P., Zhen, Y., Ferguson-Miller, S. & Brzezinski, P. (1998)
Proton uptake controls electron transfer in cytochrome c oxidase. Proc. Natl. Acad. Sci.
USA 95(23):]3606-11.

Kituchi, G., Sairo, Y. & Motokawa, Y. (1965) On cytochrome oxidase as the terminal
oxidase of dark respiration of non-sulfur purple bacteria. Biochem. Biophys. Acta 94:1-
14.

Konstantinov, A. A., Siletsky, A., Mitchell, D., Kaulen, A., & Gennis R. B.. (1997) The
roles of the two proton input channels in cytochrome c oxidase from Rhodobacter
sphaeroides probed by the effects of site-directed mutations on time-resolved
electrogenic intraprotein proton transfer. Proc. Natl. Acad. Sci. USA 94:9085-90.

Konvicka K., Guamieri F., Ballesteros J. A., & Weinstein, H. (1998) A proposed
structure for transmembrane segment 7 of G protein-coupled receptors incorporating an
Asn-Pro/Asp-Pro motif. Biophys. J. 75(2):601-l 1.

Lange, C. & Hunte, C. (2002) Crystal structure of the yeast cytochrome be. complex with
its bound substrate cytochrome c. Proc. Natl. Acad. Sci. USA 99: 2800-2805.

Lappalainen, P., Watrnough, N. J., Greenwood, C. & Saraste, M. (1995) Electron transfer
between cytochrome c and the isolated CuA domain: identification of substrate-binding
residues in cytochrome c oxidase. Biochemistry 34: 5824-5830.

Long, J. E., Durham, B., Okamura, M. & Millett, F. (1989) Role of specific lysine
residues in binding cytochrome c; to the Rhodobacter sphaeroides reaction center in
optimal orientation for rapid electron transfer. Biochemistry 28:6970-6974.

Luo, X., Zhang, D. & Weinstein H. (1994) Ligand-induced domain motion in the
activation mechanism of a G-portein-coupled receptor. Protein Eng. 7: 1441-1448.

Malatesta, F ., Nicoletti, F ., Zickermann, V., Ludwig, B., & Brunori, M. (1998) Electron
entry in a CuA mutant of cytochrome c oxidase from Paracoccus denitrificans.
Conclusive evidence on the initial electron entry metal center. FEBS Lett. 434, 322-324.

Margoliashi, E. & Schejterm A. in Cytochrome c:A multidisplinary approach Acott, R.
A. & Mauk, A. G., Eds. (University Science Book, Sausalito, California, 1996)

Margoliash, E. & Schejter, A. (1966) Cytochrome c. Adv. Protein Chem. 21:113-286.
Mather, M., Springer, P., Hensel, S., Buse, G. & Fee, J. A. (1993) Cytochrome oxidase
genes from T hermus thermophilus: nucleotide sequence of the fused gene and analysis of

the deduced primary structures for subunits I and III of cytochrome caa3. J. Biol. Chem.
268:5395-5408.

I31

Meyer, T. E. & Cusanovich, M. A. (1985) Soluble cytochrome composition of the purple
phototrophic bacterium Rhodopseudomonas sphaeroides ATCC 17023. Biochim.
Biophys. Acta 807:308-391.

Meyer, T.E. & Kamen, MD. in Advances in protein chemistry (Academic Press, Inc.,
1982), vol.25, pp. 105-212

Michel, H. (1999) Cytochrome c oxidase: catalytic cycle and mechanisms of proton
pumping — a discussion. Biochemistry 38: 15129-40.

Miller, M. A., Geren, L., Han, G. W., Saunders, A., Beasley, J ., Pielak, G. J ., Durham, B.,
Millett, F. & Kraut, J. (1996) Identifying the physiological electron transfer site of
cytochrome c peroxidase by structure-based engineering. Biochemistry 35:667-73.

Miller, M. A., Liu, R. Q., Hahm, S., Geren, L., Hibdon, S., Kraut, J ., Durham, B. &
Millett, F. ( 1994) Interaction domain for the reaction of cytochrome c with the radical

and the oxyferryl heme in cytochrome c peroxidase compound 1. Biochemistry 3328686-
93.

Millett, F. & Durham, B. in Metal Ions in Biological Systems Sigel, H. & Siegel, A., Eds.
(Marcel Dekker, Inc., Basel, Switzerland, 1991), vol. 27, pp. 223-64.

Millett, F ., de Jong, C., Paulson, L. & Capaldi, R. A. (1983) Identification of specific
carboxylate groups on cytochrome c oxidase that are involved in binding cytochrome c.
Biochemistry 22:546-552.

Millett, F ., Darley-Usmar, V. & Capaldi, R. A. (1982) Cytochrome c is cross-linked to
subunit II of cytochrome c oxidase by a water-soluble carbodiimide. Biochemistry
21:3857-3862.

Mitchell, P. (1976) Possible Molecular Mechanisms of the Protonmotive Function of
Cytochrome Systems. J. T heor. Biol. 62:327-367.

Mitchell, P. (1966) Chemiosmotic Coupling in Oxidastive and Photosynthetic
Phosphorylation. Glynn Research Ltd, Bodmin.

Mitchell, P. & Moyle, J. (1965) Stoichiometry of proton translocation through the

respiratory chain and adenosine triphosphatase systems of rat liver mitochondria. Nature
208: 147-51 .

Mkrtchyan, H., Vygodina, T. & Konstantinov, A. (1990) Na(+)-induced reversal of the
Ca2(+)-dependent red-shift of cytochrome a. Is there a hydronium output well in
cytochrome c oxidase? Biochem. Int. 20:183-9.

Moore, G. R. & Pettigrew, G. W., Eds., Cytochrome c: evolutionary, structural and
physiochemical aspects (Springer Verlag, Berlin, 1990).

132

Myllykallio, H., Zannoni, D. & Daldal F. (1999) The membrane-attached electron carrier
cytochrome cy from Rhodobacter sphaeroides is functional in respiratory but not in
photosynthetic electron transfer. Proc. Natl. Acad. Sci. USA 96:4348-4353.

Nicoletti, F., Witt, H., Ludwig, B., Brunori, M. & Malatesta, F. (1998) Paracoccus
denitrificans cytochrome c oxidase: a kinetic study on the two- and four-subunit
complexes. Biochim. Biophys. Acta 1365:393-403.

Oliveberg, M., Hallén, S. & Nilsson, T. (1991) Uptake and release of protons during the
reaction between cytochrome c oxidase and molecular oxygen: A flow-flash
investigation. Biochemistry 30: 436-440.

Osheroff, N., Brautigan, D. L. & Margoliash, E. (1980) Definition of enzymic interaction
domains on cytochrome c. Purification and activity of singly substituted
carboxydinitrophenyl-lysine 7, 25, 73, 86, and 99 cytochromes c. J. Biol. Chem.
255:8245-51.

Ostermeier, C., Harrenga, A., Ermler, U. & Michel, H. (1997) Structure at 2.751
resolution of the Paracoccus denitrificans two-subunit cytochrome c oxidase complexed
with an antibody Fv fragment. Proc. Natl. Acad. Sci. USA 94:10547-10553.

Ostermeier, C ., Iwata, S., & Michel, H. (1996) Cytochrome c oxidase. Curr. Opin. Struct.
Biol. 6:460-466.

Overfield, R. E., Wraight, C. A. & Devault, D. (1979) Microsecond photooxidation
kinetics of cytochrome c; from Rhodopseudomonas sphaeroides: In vivo and solution
studies. FEBS Lett. 105: 1 37-142.

Pan, L. P., Hibdon, S., Liu, R.-Q., Durham, B. & Millett, F. (1993) Intracomplex electron
transfer between ruthenium-cytochrome c derivatives and cytochrome c oxidase.
Biochemistry 32:8492-8498.

Pelletier, H. & Kraut, J. (1992) Crystal structure of a complex between electron transfer
partners, cytochrome c peroxidase and cytochrome c. Science 258:1748-1755.

Pettigrew, G. W. & Moore, G. R., Eds, Cytochrome c: biological aspects (Springer
Verlag, Berlin, 1987)

Pfitzner, U., Kirichenko, A., Konstantinov A.A., Mertens, M., Wittershagen, A.,
Kolbesen, B.O., Steffens, G.C., Harrenga, A., Michel, H. & Ludwig, B. (1999) Mutations
in the Ca2+ binding site of the Paracoccus denitrificans cytochrome c oxidase. FEBS Lett.
456:365-9.

133

Poulos, T., Freer, S., Alden, R., Edwards, S., Skogland, U., Takio, K., Eriksson, B.,
Xuong, N., Yonetani, T. & Kraut, J. (1980) The crystal structure of cytochrome c
peroxidase. J. Biol. Chem. 255(2):575-580.

Proshlyakov, D.A., Pressler, M.A., DeMaso, C., Leykam, J.F., DeWitt, D.L. & Babcock,
G.T. (2000) Oxygen activation and reduction in respiration: involvement of redox-active
tyrosine 244. Science 290: 1588-91.

Qian, J ., Shi, W., Pressler, M., Hoganson, C., Mills, D., Babcock, G.T. & Ferguson-
Miller S. (1997) Aspartate-407 in Rhodobacter sphaeroides cytochrome c oxidase is not
required for proton pumping or Mn binding. Biochemistry 36:2539-43.

Rieder, R. & Bosshard, H. R. (1980) Comparison of the binding sites on cytochrome c for
cytochrome c oxidase, cytochrome bcl, and cytochrome C]. J. Biol. Chem. 255:4732-
4739.

Rott, M. A. & Donohue, T. J. (1990) Rhodobacter sphaeroides spd mutations allow
cytochrome cz-independent photosynthetic growth. J. Bacterial. 172:1954-1961.

Reincke, B., Thdny-Meyer, L., Dannehl, C., Odenwald, A., Aidim, M., Witt, H.,
Riiterjans, H. & Ludwig,, B. (1999). Heterologous expression of soluble fragments of
cytochrome c552 acting as electron donor to the Paracoccus denitrificans cytochrome c
oxidase. Biochim. Biophys. Acta 1411:114-120.

Rich, P. R. (1995) Toward understanding the chemistry of oxygen reduction and proton
translocation in iron-iron respiratory oxidases. Aust. J. Plant Physiol. 22:479-484.

Riistama, S., Laakkonen, L., Wikstrbm, M., Verkhovsky, M.I., & Puustinen, A. (1999)
The Calcium Binding Site in Cytochrome M; from Paracoccus denitrificans.
Biochemistry 3 8: 10670-10677.

Riistama, S., Puustinen, A., Garcia-Horsman, A., Iwata, 8., Michel, H. & Wikstrdm, M.

(1996) Channeling of dioxygen into the respiratory enzyme. Biochim. Biophys. Acta
127521-4.

Roberts, V. A. & Pique, M. E. (1999) Definition of the interaction domain for
cytochrome c on cytochrome c oxidase. III. Prediction of the docked complex by a
complete, systematic search. J. Biol. Chem. 274:38051-38060.

Saari, H., Penttila, T. & Wikstrom, M. (1980) Interactions of Ca2+ and H+ with heme A in
cytochrome oxidase. J Biaenerg Biomembr. 12:325-38.

Saraste, M. (1990) Structural features of cytochrome oxidase. Q. Rev. Biophys. 23:331-
366.

134

Scott, R. A. & Mauk, A. G., Eds., Cytochrome c: A multidisciplinary approach
(University Science Books, Sausalito, California, 1996).

Shapleigh, J. P. & Gennis, R. B. (1992) Cloning, sequencing, and deletion from the

chromosome of the gene encoding subunit I of the aa3-type cytochrome c oxidase of
Rhodobacter sphaeroides. Mal. Microbial. 6:635-642.

Shapleigh, J.P., Hosler, J .P., Techlenburg, M.M.J., Kim, Y.Y., Babcock, G.T., Gennis,
R.B. & Ferguson-Miller, S. (1992) Definition of the Catalytic Site of Cytochrome-c-
Oxidase-Specific Ligands of Heme-a and the Heme-a3-Cu3 Center. Proc. Natl. Acad. Sci.
USA. 89:4786-90.

Sigurdson, H., Branden, M., Namslauer, A., & Brzezinski, P. (2002) Ligand binding
reveals protonation events as the active site of cytochrome c oxidase; is the K-pathway
used for the transfer of H+ or 0H? J. Inorganic Biochem. 88:335-42.

Smith, H. T., Ahmed, A. J. & Millett, F. (1981) Electrostatic interaction of cytochrome c
with cytochrome c1 and cytochrome oxidase. J. Biol. Chem. 256:4984-90.

Smith, H. T., Staudenmayer, N. & Millett, F. (1977) Use of specific lysine modifications

to locate the reaction site of cytochrome c with cytochrome oxidase. Biochemistry
16:4971-4974.

Smith, L. & Davies, H. C. (1991) The reactions of the oxidase and reductase of
Paracoccus denitrificans with cytochromes c. J. Bioenerget. Biamemb. 23:303-319.

Smith, M. B. & Millett, F. (1980) A 19F nuclear magnetic resonance study of the
interaction between cytochrome c and cytochrome c peroxidase. Biochim. Biophys. Acta
626:64-72.

Solioz, M., Carafoli, E., & Ludwig, B. (1982) The cytochrome c oxidase of Paracoccus
denitrificans pumps protons in a reconstituted system. J. Biol. Chem. 257: 1579-1582.

Speck, S.H., F erguson-Miller, S., Osheroff, N., & Margoliash, E. (1979) Definition of
cytochrome c binding domains by chemical modification: kinetics of reaction with beef

mitochondrial reductase and functional organization of the respiratory chain. Proc. Natl.
Acad. Sci. USA 76(1):155-9.

Speck, S.H. & Margoliash, E. (1984) Characterization of the interaction of cytochrome c
and mitochondrial ubiquinol-cytochrome c reductase. J. Biol. Chem. 259: 1064-72.

Staudenmayer, N., Smith, M. B., Smith, H. T., Spies, F. K., Jr. & Millett, F. (1976) An

enzyme kinetics and 19F nuclear magnetic resonance study of selectively
trifluoroacetylated cytochrome c derivatives. Biochemistry 15:3198-205.

135

Steinrticke, P., Gerhus, E., Jetzek, M., Turba, A. & Ludwig, B. (1991) The cytochrome c

reductase/oxidase respiratory pathway of Paracoccus denitrificans: Genetic and
functional studies. J. Bioenerget. Biamemb. 23:227-239.

Svensson-Ek, M., Abramson, J., Larsson, G., Tdmroth, S., Brzezinski, P. & Iwata, S.
(2002) The X-ray crystal structure of wild-type and EQ(I-286) mutant cytochrome c
oxidases from Rhodobacter sphaeroides. J. Mal. Biol. 321 :329-339.

Taha, T. S. M. &Ferguson-Miller, S. (1992) Interaction of cytochrome c with cytochrome
c oxidase studied by monoclonal antibodies and a protein modifying reagent.
Biochemistry 31 :9090-9097.

Thompson, D.A., Gregory, L., & Ferguson-Miller, S. (1985) Cytochrome c oxidase
depleted of subunit III: proton-pumping, respiratory control, and pH dependence of the
midpoint potential of cytochrome a. J. Inorg. Biochem. 23:357-364.

Trumpower, B. L. & Gennis, R. B. (1994) Energy transduction by cytochrome complexes
in mitochondrial and bacterial respiration: the enzymology of coupling electron transfer
reactions to transmembrane proton transloction. Annu. Rev. Biochem. 63: 675-716.

Trumpower, B. L. (1990) The protonmotive Q cycle. J. Biol. Chem. 265:11409-11412.

Tsukihara, T., Aoyama, H., Yamashita, E., Tomizaki, T., Yamaguchi, H., Shinzawaitoh,
K., Nakashima, R., Yaono, R. & Yoshikawa, S. (1996) The whole structure of the 13-
subunit oxidized cytochrome c oxidase at 2.8 A. Science 272:1136-1144.

Tsukihara, T., Aoyama, H., Yarnashita, E., Tomizaki, T., Yamaguchi, H., Shinzawaitoh,
K., Nakashima, R., Yaono, R. & Yoshikawa, S. (1995) Structure of metal sites of
oxidized bovine heart cytochrome c oxidase at 2.8 A. Science 269:1069-1074.

Turba, A., Jetzek, M. & Ludwig, B. (1995) Purification of Paracoccus denitrificans
cytochrome c552 and sequences analysis of the gene. Eur. J. Biochem. 231:259-265.

Van Rhee, A. M. & Jacobson, K. A. (1996) Molecular architecture of G protein-coupled
receptors. Drug Dev. Res. 3721-38.

Wang, K., Zhen, Y., Sadoski, R., Grinnell, 8., Geren, L., Ferguson-Miller, 8., Durham, B.
& Millett, F. (1999) Definition of the interaction domain for cytochrome c on cytochrome
c oxidase. II. Rapid kinetic analysis of electron transfer from cytochrome c to
Rhodobacter sphaeroides cytochrome oxidase surface mutants. J. Biol. Chem.
274:38042-38050.

Wess, J ., Nanavati, S., Vogel, Z., & Maggio, R. (1993) Functional role of proline and

tryptophan residues highly conserved among G protein-coupled recptors studied by
mutational analysis of the m3 muscarinic receptor. EMBO J. 12:331-338.

136

Wikstrdm, M. (1977) Proton Pump Coupled to Cytochrome c Oxidase in Mitochondria.
Nature (London) 266:271-273.

Wilmanns, M., Lappalainen, P., Kelley, M., Sauer-Eriksson, E. & Saraste, M. (1995)
Crystal structure of the membrane exposed domain from a respiratory quinol oxidase

complex with an engineered dinuclear copper center. Proc. Natl. Acad Sci. (USA)
92:11955-11959.

Witt, H., Malatesta, F., Nicoletti, F., Brunori, M. & Ludwig, B. (1998) Cytochrome-c-
binding site on cytochrome oxidase in Paracoccus denitrificans. Eur. J. Biochem.
251:367-73.

Witt, H. & Ludwig, B. (1997) Isolation, analysis, and deletion of the gene coding for
subunit IV of cytochrome c oxidase in Paracoccus denitrificans. J. Biol. Chem.
272:5514-7.

Witt, H., Zichermann, V. & Ludwig, B. (1995) Site-directed mutagenesis of cytochrome

c oxidase reveals two acidic residues involved in the binding of cytochrome c. Biochim.
Biophys. Acta 1230: 74-6.

Yoshikawa, S., Tera, T., Takahashi, Y., Tsukihara, T. & Caughey, W. S. (1998)
Crystalline cytochrome c oxidase of bovine heart mitochondrial membrane: Composition
and x-ray diffraction studies. Proc. Natl. Acad. Sci. USA 85:1354-1358.

Yoshikawa, S., Shinzawa-Itoh, K. Nakashima, R., Yaono, R. Yamashita, E., Inoue, N.
Yao, M., Fei, M. J ., Libeu, C. P., Mizushima, T., Yamaguchi, H., Tomizaki, T. &
Taukihara, T.. (1998) Redox-Coupled Crystal Structural Changes in Bovine Heart
Cytochrome c Oxidase. Science (280): 1723-9.

Yoshikawa, S., Mochizuki, M., Zhao, X.J. & Caughey, W.S. (1995) Effects of Overal
Oxidation State on Infrared Spectra of Heme a3 Cyanide in Bovine Heart Cytochrome c
Oxidase — Evidence of Novel Mechanistic Roles for CuB. J. Biol. Chem. 270:4270-79.

Yu, C .A., Yu, L. & King, TE. (1973) Kinetics of electron transfer between cardiac
cytochrome c 1 and c. J. Biol. Chem. 248(2):528-33.

Yu, L., Dong, J .-H. & Yu, C.-A. (1986) Characterization of purified cytochrome c. from
Rhodobacter sphaeroides R-26. Biochim. Biophys. Acta 852: 203-211.

Yun, C.-H., Barquera, B., Iba, K., Takamiya, K.-I, Shapleigh, J ., Crofts, A. R. & Gennis,
R. B. (1994) Deletion of the gene encoding cytochrome b562 from Rhodobacter
sphaeroides. FEMS Microbia. Letts. 120:105-110.

Zhang, D. & Ewinstein, H. (1993) Signal transduction by a 5-HT2 receptor: a

mechanistic hypothesis from molecular dynamics simulations of the three-dimensional
model of the receptor complexed to ligands. J. Med. Chem. 36:934-938.

137

Zhen, Y., Schmidt, B., Kang, U.G., Antholine, W.,& Ferguson-Miller, S. (2002) Mutants
of the CuA Site in Cytochrome c Oxidase of Rhodobacter sphaeroides: 1. Spectral and
Functional Properties. Biochemistry 41 :2288-97.

Zhen, Y., Hoganson, C. W., Babcock, G. T. & Ferguson-Miller, S. (1999) Definition of
the interaction domain for cytochrome c on cytochrome c oxidase. 1. Biochemical,

spectral, and kinetic characterization of surface mutants in subunit II of Rhodobacter
sphaeroides cytochrome aa3. J. Biol. Chem. 274:38032-41.

Zhen, Yuejun (1998) The role of subunit II in cytochrome c oxidase in cytochrome c
binding and electron transfer. Ph. D. Dissertation.

138