Citation
The study of ferritin bare gold electrodes by cyclic voltammetry and infrared spectroelectrochemistry

Material Information

Title:
The study of ferritin bare gold electrodes by cyclic voltammetry and infrared spectroelectrochemistry
Creator:
Johnson, Michael Alfred
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
88 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Ferritin ( lcsh )
Voltammetry ( lcsh )
Infrared spectroscopy ( lcsh )
Charge exchange ( lcsh )
Charge exchange ( fast )
Ferritin ( fast )
Infrared spectroscopy ( fast )
Voltammetry ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 82-88).
Thesis:
Department of Chemistry
Statement of Responsibility:
by Michael Alfred Johnson.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
45210843 ( OCLC )
ocm45210843
Classification:
LD1190.L46 2000m .J64 ( lcc )

Downloads

This item has the following downloads:


Full Text
THE STUDY OF FERRITIN ON BARE GOLD ELECTRODES BY
CYCLIC VOLTAMMETRY AND
INFRARED SPECTROELECTROCHEMISTRY
by
Michael Alfred Johnson
B.S., United States Air Force Academy, 1988
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Chemistry
2000
VSTi-M


This thesis for the Master of Science
degree by
Michael Alfred Johnson
has been approved
by
q-z Date


Johnson, Michael Alfred (M.S., Chemistry)
The Study of Ferritin on Bare Gold Electrodes by Cyclic Voltammetry and
Infrared Spectroelectrochemistry
Thesis directed by Assistant Professor Donald C. Zapien
ABSTRACT
This research focused on the study of ferritin on a bare gold electrode
using cyclic voltammetry and infrared reflection-absorption spectroscopy
(IRRAS). The cyclic voltammetry of ferritin on bare gold suggested that a
reversible change in orientation of the ferritin molecule might occur.
Additionally, our data suggested that ferritin tends to oligomerize on bare
gold, which is in contrast to its behavior on gold modified by
mercaptopropionate groups. From this data we postulated that ferritin is
anchored to the electrode by the promoter layer. The relationship
between ferritin packing density and scan rate was found to be linear,
suggesting that the signals on our voltammetric traces were from ferritin
adsorbed onto the gold electrode and not ferritin in solution. Potential
dependence data suggested that ferritin is adsorbed in an electroactive
state only when it is first reduced. Additionally, the data suggest that a
change in quaternary structure occurs during the reduction process that
allows the protein to be adsorbed onto the gold surface while maintaining
electroactivity. Studies of packing density at various ionic strengths,
ferritin concentrations, and adsorption times were conducted. The ionic
strength data suggested that hydrophobic interactions predominate at the
ferritin-gold interface. The packing density of ferritin increased with
dissolved ferritin concentration. The adsorption data suggested that the
adsorption rate was limited more by the diffusion of ferritin onto the gold
surface than by structural rearrangements or changes in the orientation
following initial attachment to the electrode surface. An infrared
reflection-absorption apparatus was designed and constructed. This
apparatus included an airtight enclosure and reflection optics. Testing of
the apparatus revealed that the cell had the ability to exhaustively
electrolyze an analyte in fairly short time and that the infrared spectra can
III


be recorded in situ. More development will be required before the cell
can obtained oxidized minus reduced infrared difference spectra. The
infrared spectrum of ferritin was recorded and analyzed. Peak
assignments for amide I and amide II were made. The infrared spectrum
of ferritin confirmed the presence of cr-helical segments as well as some
disordered secondary structure.
This abstract accurately represents the content of the candidates thesis.
I recommend its publication.
Signed
Donald C. Zapie
IV


DEDICATION
I dedicate this work to my wife Monica for her loving support,
encouragement, and understanding throughout my graduate career.
V


ACKNOWLEDGEMENTS
Professor Donald C. Zapien, for his never-ending guidance, support,
patience, and enthusiasm both in and out of the lab.
Professor Larry G. Anderson, for his kind patience and advice in
obtaining this degree.
Mr. Fedri Marrugo, for machining the stainless steel flange and
constructing the external optics bench, and for assistance with the cell
design.
Research Corporation (Cotrell Research Grant Number C-3735), for
funding this work.
VI


CONTENTS
Figures...............................................................x
Chapter
1. Introduction......................................................1
1.1 Ferritin.....................................................1
1.1.1 Structure....................................................2
1.1.2 Function....................................................10
1.2 Electron Transfer..............................................16
1.2.1 Fundamentals of Electron Transfer..........................16
1.2.2 The Three-Electrode Cell....................................17
1.2.3 Voltammetry.................................................19
1.2.4 Mediated Electron Transfer..................................25
1.2.5 Direct Electron Transfer....................................25
1.3 Solution Infrared Spectroelectrochemistry...................29
1.4 Research Objectives.........................................34
2. Experimental
2.1 Electrochemistry of Ferritin on Bare Gold...................37
2.1.1 Materials...................................................37
vii


2.1.2 Methods......................................................38
2.2 Infrared Spectroelectrochemistry.............................45
2.2.1 Materials....................................................45
2.2.2 Methods......................................................46
3. Direct Electron Transfer of Ferritin Adsorbed on
Bare Gold Electrodes...............................................48
3.1 Adsorption and Redox Activity of Ferritin...................48
3.2 Potential Dependence of Ferritin Adsorption..................55
3.3 Dependence of Packing Density on Ionic Strength,
Ferritin Concentration, and Adsorption Time........................57
3.3.1 Ionic Strength...............................................57
3.3.2 Ferritin Concentration.......................................59
3.3.3 Adsorption Time..............................................59
4. Design of the Infrared Reflection Absorption Spectroscopy
(IRRAS) Apparatus..................................................63
4.1 Spectroelectrochemical Cell..................................63
4.1.1 Description..................................................63
4.1.2 Design.......................................................63
4.2 Enclosure...................................................68
4.2.1 Description..................................................68
4.2.2 Design.......................................................70
4.3 Optics.......................................................71
viii


4.3.1 Internal Optics..............................................71
4.3.2 External Optics..............................................73
4.3.3 Detector.....................................................73
4.4 Testing and Evaluation of the Apparatus......................74
4.4.1 Reduction of Ferricyanide....................................74
4.4.2 Spectrum of Ferritin.........................................76
5. Summary and Conclusion.........................................79
5.1 Summary...................................................... 79
5.2 Conclusion....................................................80
References.........................................................81
IX


FIGURES
1.1 Schematic Drawing of Ferritin Molecule......................3
1.2 Schematic Depiction of Ferritin Shell and Iron Core.........4
1.3 Primary Sructure of Ferritin................................6
1.4 Schematic Diagram of Ferritin Subunit.......................8
1.5 The Three-Electrode Cell...................................18
1.6 Cyclic Voltammetry.........................................21
1.7 Cyclic Voltammogram of Ferrocyanide........................22
1.8 Mediated Electron Transfer.................................26
1.9 Direct Electron Transfer of Cytochrome c...................28
1.10 Thin-layer Infrared Transmittance Cell...................32
1.11 Thin-layer Infrared Reflectance Cell.....................36
2.1 Electrochemical Cell.......................................39
2.2 Teflon Electrochemical Cell................................41
2.3 Cyclic Voltammogram of Clean Bare Gold.....................42
2.4 Cyclic Voltammogram of Adsorbed Iodide.....................43
3.1 Cyclic Voltammogram of Ferritin............................49
3.2 Consecutive Cycling of Ferritin............................51


3.3 Effect of Scan Rate on Peak Current..........................54
3.4 Dependence of Ferritin Adsorption on Potential................56
3.5 Effect of Ionic Strength on Packing Density..,...............58
3.6 Effect of Ferritin Concentration on Packing Density..........60
3.7 Effect of Adsorption Time on Packing Density.................62
4.1 Diagram of the Infrared Reflection Absorption Spectroscopy .
(IRRAS) Cell......................................................64
4.2 Beam Path Through the Dove Prism...........................67
4.3 Plexiglas Enclosure."........................................69
4.4 Optics Setup.................................................72
4.5 Infrared Spectrum of Ferricyanide and Ferrocyanide...........75
4.6 Infrared Spectrum of Ferritin................................77
XI


1. Introduction
1.1 Ferritin
Iron is essential for almost all forms of life. This need for iron is a
result of the varied roles it plays in biological systems, especially as a
catalyst in numerous intracellular and extracellular reactions (1). One of
the evolutionary challenges of developing life forms has been to maintain a
usable supply of iron. Approximately 2.5 billion years ago, when
photosynthetic organisms began using H20 as its source of hydrogen,
oxygen became the major by-product (2). This prevalence of oxygen
resulted in the oxidation of Fe(ll) to Fe(lll) (3); however, because Fe(lll) at
neutral pH forms insoluble Fe(OH)3 profound restrictions were placed on
its availability for use in biological systems (4). This insolubility in oxygen-
rich environments necessitated the evolution of iron storage methods in
which oxygen-dependent organisms could store iron in a soluble form
while accommodating and using oxygen. Ferritin, one outcome of this
process, maintains an ample supply of iron in a soluble form for use in
various biological functions, including oxygen transfer, electron transfer,
nitrogen fixation, and DNA synthesis (5).
1


1.1.1 Structure
1.1.1.1 Introduction
Cellular ferritin is a protein that modulates the availability of iron in
nearly all forms of living organisms, including most bacteria (1). The
ferritin molecule has a molecular weight of 450,000 Da, a diameter of 12.0
nm, and a spherical protein coat, known as apoferritin, which encases an
iron-containing core (6) (Figure 1.1). The apoferritin shell consists of two
types of subunits, denoted H (heavy) and L (light). A total of 24 subunits
make up this protein shell, which is in contact with the iron core at various
points, forming an interface between the protein and the iron core (7)
(Figure 1.2). Horse spleen apoferritin is often used for studies because
most of its subunits are identical (approximately 85% are L subunits) and
the location of essentially all amino acids has been elucidated by Harrison
and coworkers using X-ray crystallography (8,9). One ferritin subunit
consists of approximately 175 amino acids (1,10).
2


Figure 1.1. Schematic drawing of a ferritin molecule. The protein
contains a symmetrical arrangement of 24 equivalent subunits.
Adapted from (1).
3


Core
Channel
Figure 1.2. Schematic depiction of the ferritin shell and iron core.
The 24 subunits contact the iron core at various points to form an
interface between the shell and the core. Adapted from (1).
4


The protein shell encases an iron core that is limited by the size of
the inside cavity (approximately 8 nm). A complete iron core consists of
4000 to 4500 Fe(lll) atoms; however, most molecules in a ferritin
preparation contain little or no iron-core, and a large number of molecules
contain less than the full compliment of iron. The core forms as a complex
of small microcrystalline particles having the formula
(Fe00H)8(Fe0P03H2). Because the number and consistency of the
particles is variable, one or several particles may be present in addition to
some disordered material (7,11,12).
1.1.1.2 Protein Coat
The primary structure of a horse spleen ferritin subunit as well as
the primary structures of ferritin subunits from other species is given in
Figure 1.3 (13-22). An interesting feature of these sequences is that the L
sequences between species are more similar (85 to 90% identity) than H
and L sequences for the same species (human, 55% identity). The same
trend holds true for H subunits (>90% identity in three complete
sequences) (23).
5


5 10 15 20 25 30 35 40 45 50 55 60
HoS-L SSQIRQNYSTEVEAAVNRLVNLYLRASYTYLSLGFYFDRDDVALEGVCHFFRELAEEKRE
HuS-L ----------D------Y-------Y------------BY--------S------------
HuL-L ----------D------S-------Q----------------------S------------
HuL-H TV------HQDS----1Ql---EY--V---MSY---------KNFAKYLHQSH-E
RaL-L T--------------------H--------------f-----------G------------
RaL-H PV------HQDS----1Ql---EY--V---MSC---------KNFAKYLHQSH-E
65 70 75 80 85 90 95 100 105 110 115 120
HoS-L GAERLLKMQNQRGGRALFQDLQKPSQDEWGTTLDAMKAAIVLEKSLNQALLDLHALGSAQ
HuS-L -Y-------------------IK--AE-------K-P----MAK----------------R
HuL-L -Y-------------------IK--AE-------K-P----MAK----------------R
HuL-H HKM-L----------IFL-IK--DCDESG-NECLH-----NVSEKATDK
RaL-L--------M E-----------V------------KEE LA N-----------------R
RaL-H HKM-L----------IFL-IK--DRDESG-NRCLH-------VSEKATDK
125 130 135 140 145 150 155 160 165 170 175
HoS-L ADPHLCDFLESHFLDEEVKLIKKMGDHLTNIQRLVGSQAGLGEYLFERLTLKHD
HuS-L T---------T-------------------LRK-GPE-----------------
HuL-L T---------T-------------------LHGPE------------------
HuL-H N------|_TY-N-QAEL------VLRKM GAPESA----DKHGDSDNES
RaL-L T--------------K---------N----LR-WQ-PS----------------
RaL-H N---------1T'Y-N-QSEL--VLRKMGAPES-MA----DKHGGDES
Sequences are from the following sources. Dashes indicate the same amino acid as in HoS-L.
HoS-L (protein) Horse spleen L subunit
HuS-L (protein) Human spleen L subunit
HuL-L (cDNA, DNA) Human liver L subunit
HuL-H (cDNA, DNA) Human liver H subunit (N-terminal T T A S not shown)
RaL-L (DNA) Rat liver L subunit (insertion QPAQTGVA between residues 157 and 158 not
shown) RaL-H (DNA) Rat liver H subunit (N-terminal T T A S not shown)
Figure 1.3. The primary structure of ferritin from various sources. Adapted
from (1).
6


Variations in subunit primary structure may affect other properties of
ferritin in a variety of ways. For example, amino acid sequence, chain
length (172 to 183 residues), and mass may cause deviations in surface
charge (24-26), hydrophobicity (27,28), and size (29,30). Despite these
differences, X-ray diffraction studies have shown that the higher order
structure between subunits of different molecules is not greatly affected
(31,32).
Four a-helices connected to another small helix through a series of loops
make up the secondary structure of a ferritin subunit. Of the 175 residues,
129 are incorporated into these five a-helices, denoted A-E (Figure 1.4).
The four longer a-helices are twisted into a bundle while the shorter a-
helix is located at the end of this bundle. This small helix seems to play an
important role in how the subunits pack to form the quaternary structure of
the protein. A long loop lies on the outside of the twisted bundle and
The ferritin subunits interact via trimeric and tetrameric interactions
to form hydrophilic channels (with conserved carboxylate side chains
between residues 120-127) and hydrophobic channels (with conserved
nonpolar side chains between residues 160-165) (2,5). Along the 3-fold
axis of the molecule are located eight
7


N-terminus
Figure 1.4. Schematic diagram of a ferritin subunit. Helices A, B, C, and D
and the short helix E are depicted as rods. Adapted from (1).
8


hydrophilic channels which are 0.3 to 0.4 nm wide. Three glutamate
residues are located on the inside of the channel while three more
glutamate residues line the outside. The outside of the channel flares out
and is lined with primarily hydrophilic residues--six serines, six histidines,
and three cysteines-allowing water molecules are bound to this funnel. It
has been suggested that, due to this structure, iron may flow in and out of
these ports and that the funnel shape may provide chelaters and
reductants access to metal on the outer site (33).
A ferritin molecule also contains six hydrophobic channels located
along the 4-fold axis. These channels are 0.3 to 0.4 nm wide and 1.2 nm
long and are lined with 12 leucine residues, three from each of the four
subunits. The function of these channels is not fully understood, but
recent work has suggested that they facilitate the rapid exchange of water
due to the lack of hydrogen bonding sites (33).
1.1.1.3 Iron Core
The inorganic, micellular nature of the iron core was first
established by Granick, Michaelis, and Hahn, who discovered that the core
consists of a phosphate-containing ferric oxyhydroxide complex.
Interestingly, they found that treating this complex with 1 N NaOH (37)
could induce release of the iron core from the protein shell; however, they
g


incorrectly believed the iron micelles to cover the exterior of the protein
molecule. In 1954, Farrant, using electron microscopy, demonstrated that
the iron core filled the inside of an apoferritin protein shell, showing the
core to be an electron-dense mass of 5.5 to 6.0 nm in diameter (38).
Despite extensive study, information on the core is still quite limited, and
the exact atomic structure, while a matter of conjecture, likely has an
irregular arrangement and composition (33). What is certain, however,
that the iron core plays a role central to the function of ferritin.
1.1.2 Function
Ferritin functions as an iron storage protein and therefore requires
an efficient mechanism of iron uptake and release. One purpose of this
storage function is to protect the cell from oxidative damage from reactive
oxygen species. For example, Fe(ll) can catalyze the decomposition of
hydrogen peroxide to form highly reactive hydroxyl radical:
H202 + Fe2+ 2-OH + Fe3+
This reaction, known as the Fenton reaction, is prevented in part by the
uptake of Fe(ll) by ferritin. Another function of ferritin is the binding of
other potentially harmful metal ions. While investigating how
10


phosphoglucomutase is inhibited by Be2+, Joshi and coworkers found that
ferritin can bind up to 1000 atoms of Be2+ in vivo (46). Additionally, ferritin
has been shown to bind Cu2+, Zn2+, Cd2+, Tb3*, U022+, Cr3*, and V02+ (47,
48).
1.1.2.1 Iron Uptake
The loading of iron into the molecule is pivotal to the function of
ferritin as an iron storage protein. Research has suggested that entry into
apoferritin occurs via the 3-fold channels (39). The channels broaden
toward the shell exterior and contain numerous hydrophilic regions, which
may aid in tunneling Fe(ll) into the inner, narrow part of the channel.
These inner channels contain the highly conserved Glu 130 and Asp 127
residues that, due to the flexibility of the side chains, may effectively bind
Fe(ll) or Fe(lll). This idea was strengthened by studies that uncovered
evidence that at least two residues exposed on the three-fold channels
play active roles in the mechanism of iron incorporation into apoferritin.
These residues are metal binding sites that actively bind and transfer iron
into the ferritin cavity (50). One view is that iron ions approaching the
inner channel displace other iron ions inward from Glu 130 to Asp 127
(39).
11


The entry of iron into the ferritin molecule first requires its oxidation
from Fe(ll) to Fe(lll). Because iron is known to enter through the 3-fold
channels, the possibility of an oxidative mechanism associated with these
channels has been extensively studied (85-87). Researchers have found
from site-directed mutagenesis studies that a ferroxidase center, an area
responsible for the oxidation of Fe(ll) to Fe(lll), is present in the H-chain
subunits along the 3-fold channels (87,88). One proposed mechanism of
iron oxidation through this ferroxidase site suggests that nearby glutamate
residues in the ferroxidase center first chelate two Fe2+ atoms. Because
they are so close to one another, the two atoms simultaneously transfer
two electrons to molecular oxygen, generating Fe3+-0-Fe3+ and hydrogen
peroxide. This complex then dissociates and releases iron into the protein
cavity. From this point, the iron atoms go on to form the ferrihydrite core.
As the core becomes more developed, the core surface facilitates Fe(ll)
oxidation(90). This oxidation of Fe(ll) makes electrochemical methods
ideal for the study of ferritin.
1.1.2.2 Iron Release
Iron release is also a crucial function of ferritin because iron is an
essential nutrient for life in almost every living organism. Examples of iron
proteins and enzymes that require iron include the iron-tetrapyrrole
12


complexes such as those proteins that contain heme prosthetic groups, the
iron-sulfur proteins, and the iron proteins that contain an iron coordinated
to nitrogen and oxygen ligands (49).
One way in which ferritin releases iron is through degradation of the
protein(39). It is thought that ferritin may lose as much as one third of its
iron stores daily through this process, while it is still not clear how much is
lost from intact protein (39).
The reduction of iron from Fe(lll) to Fe(ll) is a pivotal step in the
normal release of iron from the ferritin core. Laboratory approaches that
have been used to remove iron include reducing ferritin to induce the
release of iron as Fe2+, with or without chelators, and using chelators that
have a high affinity for Fe3+ to remove Fe3+ from the ferritin molecule. Both
methods depend on the physical state of the cores. For example, the core
size, the degree of aggregation, and the proximity of protein or membrane
may greatly affect how readily iron is removed from ferritin (39).
In the cell three modes of release by this mechanism have been
proposed (39). The first model suggests that a reductant diffuses through
the protein shell and reduces Fe3* to Fe2+, which leaves the protein via the
three-fold channels. The second model proposes that Fe3+ is reduced in
the three-fold channels and the reduced iron diffuses out of the ferritin
13


shell through these channels. The third model involves the reductant
acting on the iron core by a long-range electron transfer through the ferritin
shell. The Fe^ ions, reduced to Fe2+ as a result of this electron tunneling,
would exit the ferritin molecule via the three-fold channels (39).
Additionally, chelators have been used in conjunction with reductants, and
could effect the removal of iron from holo-ferritin in a number of ways (39).
Although the in vivo release of iron from ferritin is still under study, it has
been shown that an enzyme in vertebrate livers catalyzes the reduction of
ferritin iron from Fe3* to Fe2+ when reduced nicotine adenine dinucleotide
(NADH) and flavin mononucleotide (FMN) are present and oxygen
concentrations are below 3 pM (40). It was subsequently found that
flavins, reduced by exogenous mitochondria, mediate the release of ferritin
iron (41), as shown below.
Ferritin Fe(lll)
Free Fe(ll)
14


Therefore, it was theorized that ferritin participates in heme synthesis by
donating iron directly and also that ferritin aids in transporting iron from the
plasma membrane by functioning as an intermediate (42-44).
Another proposed mechanism of ferritin iron release involves the
interaction of flavo-proteins with the outer ferritin coat. This mechanism
calls for the tunneling of electrons through the apoferritin shell to reduce
the iron core and the release of iron, which is then scavenged by a small
chelator (45).
1.1.2.3 Summary
As mentioned above, redox reactions are key to the uptake and
release of iron. After Fe(ll) enters the protein shell, it is oxidized to Fe(lll)
as it is incorporated into the iron core. The release of iron from the protein
shell first requires the reduction of Fe(lll) to Fe(ll). Electrochemical
methods are well-suited to the study of these electron transfer reactions,
and may yield clues to the electrochemical behavior of ferritin.
1.2 Electron Transfer
1.2.1 Fundamentals of electron transfer
In electrochemical methods, one electrode serves as a reference
electrode, while the another electrode serves as a working electrode. The
makeup of the reference electrode is fixed, making the potential on the
15


electrode fixed. If the potential on the working electrode is adjusted to a
low enough (negative) potential, then the energy for transfer of electrons
from the working electrode to the analyte in solution will be favored. As a
result, the species in solution will be reduced. If the potential of the
working electrode is adjusted to a high enough (more positive) potential,
then eventually the energy of the electrons on the analyte molecules will
be at a higher energy than those on the electrode. As a result, electrons
will tend to migrate from the analyte in solution to the electrode, resulting
in an oxidation (91).
1.2.2 The Three-Electrode Cell
One limitation of the two-electrode system described above is that
electrolysis would have to occur at the reference electrode as well as the
working electrode, resulting in a change of the composition of the
reference electrode, and thus its potential. A configuration better suited
for these types of measurements is the three-electrode cell (Figure 1.5).
This cell is composed of three electrodes: a working electrode, a counter
electrode, and a reference electrode. The potential between the working
and counter electrode is increased until it becomes great enough to induce
electron transfer with the species at the working electrode surface. The
reaction results in the flow of electrons, or a current, which is measured
16


with a suitable current-measuring device. The potential of the working
electrode is measured with respect to the reference electrode potential. In
this configuration, the amount of current flowing through the reference
electrode is negligible, thus its composition and therefore its potential
remain constant (92).
1.2.3 Voltammetry
Voltammetry refers to those electroanalytical techniques in which
current is measured as a function of applied potential. By analysis of the
current-potential plot, information regarding the desired analyte may be
derived. Voltammetric methods rely on complete polarization of the
working electrode, and for this reason, the working electrode is kept as
small as practical (92).
1.2.3.1 Linear Scan Voltammetry
One of the most widely used electrochemical methods that makes use of
the three-electrode system is linear sweep voltammetry. In a typical linear
sweep voltammetry experiment, the potential of the working electrode is
increased or decreased at a rate of 2 to 5 mV/s. The resulting current
between the counter and working electrodes is measured and plotted
versus applied potential (92).
17


Figure 1.5. The three-electrode cell. Adapted from (91).
18


1.2.3.2 Cyclic Voltammetry
1.2.3.2.1 Introduction
Cyclic voltammetry, a technique well-suited for the study of electron
transfer kinetics, involves the application of a triangular wave potential in
which the potential of the indicator electrode is swept linearly through a
specified range and then swept the opposite direction to the original
potential (Figure 1.6).
Shown in Figure 1.7 is the cyclic voltammogram of ferrocyanide in 1.0 M
NaCI. The potential applied to the working electrode is swept positive to +
0.7 V, reversed, and then scanned to -0.1 V. As the potential is
increased, an anodic current flows as electrons are transferred from
ferrocyanide to the gold electrode, then reaches a maximum at 0.04 V.
The half reaction for this process is given as: Fe(CN)4' - Fe(CN)3' + e\
Though the potential continues to increase, the current decreases
because the surface concentration of available ferrocyanide becomes
smaller. The current gradually levels off and the direction of the potential
scan is reversed at +0.7 V. As the potential approaches approximately
+0.04 V, the cathodic current increases as ferricyanide in solution is
reduced. This reduction is represented by the half reaction: Fe(CN)3' + e'
-+ Fe(CN)4' (92).
19


Figure 1.6. Cyclic voltammetry shown as a function of various parameters:
(a) potential vs time; (b) current vs time; (c) current vs potential. Adapted
from (60).
20


Figure 1.7. Cyclic voltammogram of ferrocyanide. The molecule is
oxidized to ferricyanide on the anodic sweep (top peak), and reduced back
to ferrocyanide on the cathodic sweep (bottom peak).
21


When obtaining measurements of species in solution, the current is
limited by how fast a species approaches the electrode surface, transfers
electrons with the electrode, then finally leaves the electrode surface. The
current due to the electrolysis of an adsorbed molecule is, of course, not
limited by the diffusion of the species to the electrode surface. However, it
is often useful to study the electrochemistry of adsorbed species, as the
electrochemical data may yield important information about the structure
as well as electron transfer processes of the adsorbate.
When electroactive species adsorb to an electrode, the amount of
material adsorbed, not diffusion, limit current. In the case of reversible
electron-transfer processes, the surface concentrations are then related by
the Nernst equation:
To(t)/ rR(t) = exp{nF[E(t) ES]/RT] = 0(t)
F0(t) and rR(t) are the surface concentrations of oxidized and reduced
species, respectively. Es is the formal potential of the adsorbed couple
and E(t) describes the potential as it varies with time. If the total surface
concentration (T) of both oxidized and reduced species remains constant,
22


the individual surface concentrations are dependent upon time and
potential and therefore an expression for the current can be derived:
i = n2F2AT*v0/RT(1+0)2
where 0 = E(t)/Es
According to this equation, the maximum current will occur when E = Es (0
= 1). This peak current is also related to the number of moles of adsorbed
species as well as the scan rate (v) (60).
1.2.4 Mediated Electron Transfer
Much of the early electrochemical studies on proteins involved the
use of electrochemical mediators. Mediators work by acting as an electron
transporter between the working electrode and the protein (Figure 1.8). At
reducing potentials, the mediator would be reduced directly at the
electrode and in turn would reduce the protein in solution. The mediator
then returns to the electrode to become reduced again, and the cycle is
repeated.
Mediated electrochemical studies on ferritin have yielded valuable
information, such as number of electrons transferred per molecule, redox
potentials, and rate of reduction of the iron core (93). Mediated
23


Reduced Mediator Oxidized Mediator
Figure 1.8. Mediated electron transfer. The mediator is reduced at the
electrode surface, then reduces the protein electroactive site. The
oxidized mediator is then reduced again at the electrode surface to
complete the cycle.
24


electrochemistry has been shown to be useful, other capabilities such as
the probing of electron transfer kinetics and investigating possible
chemical reactions coupled with electron transfer reactions are possible
only if the protein exchanges electrons directly with the electrode.
1.2.5 Direct Electron Transfer
The direct electron transfer of proteins has often encountered
difficulties not encountered with small molecules. Proteins have the
propensity to irreversibly adsorb to the bare electrode surface
forming an insulating layer that would often hinder the electron transfer of
the protein in solution. In addition, the electroactive centers) of proteins is
sometimes deeply embedded in the globular matrix. This results in slow
electron transfer kinetics, sometimes to the point where no currents are
observed.
To address these issues, electrodes with surface moieties or
electrodes modified by the application of adsorbed monolayers have been
effectively used to catalyze the electron transfer of proteins. Cytochrome c
is a protein that has been extensively characterized by electrochemical
methods. When one observes the cyclic voltammogram of cytochrome c in
solution using bare gold as the working electrode, the current response is
relatively small and requires higher concentrations of cytochrome c to
25


observe a well-defined current potential curve (53,62). One of the most
common modifiers used in the electrochemical study of cytochrome c is
4,4-dithiodipyridyl (DTDP). The molecules of DTDP dissociate, forming a
layer of 4-thiopyridine.
The voltammogram of cytochrome c at the Au-4-thiopyridine
modified surface is shown in Figure 1.9. Several factors may account for
the facilitated electron transfer. First, the promoter
(DTDP) layer may prevent protein denaturation by anchoring the protein
to the electrode surface. In its native state, cytochrome c may exhibit a
stronger current response than in its denatured state. Second, the
promoter layer may allow the protein to adsorb so that its heme prosthetic
group is closer to the electrode. Third, the attraction between the lysine
groups of cytochrome c and the nitrogen atom of pyridine may aid in
aligning the protein in a more favorable position for direct electron transfer
(62).
Interestingly, ferritin did not give appreciable currents at gold
electrodes modified by 4,4-dithiodipyridyl, octadecyl mercaptan, or
cysteamine. Ferritin has exhibited well-defined voltammograms at gold
electrodes modified by 3-mercaptopropionic acid and 2-mercaptosulfonic
26


Figure 1.9. Direct electron transfer of cytochrome c on gold enhanced by
modification of the gold electrode by 4,4-dithiodipyridyl (DTPT).
27


acid. However, the observed currents were much smaller than would be
predicted by theory for ferritin in solution. Further study later indicated that
the current was produced from ferritin adsorbed on the electrode surface
(94).
1.3 Solution Infrared Spectroelectrochemistry
While voltammetric techniques are useful in the examination of the
electron transfer of species, both adsorbed and unadsorbed, it has only
limited use in determining the structural properties and reaction
mechanisms of electroactive species on a microscopic scale (68). In order
to study these types of properties, spectroscopic methods of analysis are
needed. A number of methods have been used to study in detail the
interactions within the electrode/electrolyte interface. These methods have
included techniques such as electron spin microscopy (ESR), Mossbauer
spectroscopy, and various optical spectroscopies. Of these techniques, the
optical spectroscopies, due to their convenience, have been the most
widely used (68). Several factors contribute to the effectiveness of optical
spectroscopies. First, these techniques have the capacity to be highly
selective because molecules absorb only at frequencies that match their
difference in energy levels. Second, some methods are extremely
sensitive, allowing the observation of minute changes in surface coverage
28


(as little as a few percent of a monolayer). Finally; spectroscopic
techniques typically have a very flat response time, allowing kinetic studies
(68).
One spectroscopic technique well-suited for the study of
electrochemical processes both in solution and on a surface is infrared
spectroelectrochemistry. Two general approaches to this technique utilize
either a transmission cell or a reflectance cell. A transmission cell, also
known as an optically transparent thin-layer electrochemical (OTTLE) cell,
is based on a design in which an optically transparent electrode is
sandwiched between two IR-transparent windows (Figrure 1.10) (66). The
beam passes directly through the cell, exposing the sample to infrared
radiation. A principal advantage of this method is simplicity often times,
the cell may be placed directly into the sample chamber of the instrument,
eliminating the requirement for additional external optics (67).
Transmission cells have great utility in observing the
spectroelectrochemical parameters of an analyte in solution. One example
illustrating the usefulness of this technique is the study conducted by
Mantele and co-workers, who adapted the technique to detect redox-linked
conformational changes in cytochrome c (69), then later expanded the
technique to examine these redox-linked conformational changes at
29


various temperatures, pHs, and electrode surfaces (70). This technique
involved examination of the infrared difference spectrum (reduced minus
oxidized states). From the spectral data they were able to determine that,
at neutral pH, the redox transition involves modifications in the /Mum and
y9-sheet segments of cytochrome c.
A reflectance cell utilizes the surface of the electrode to reflect the
infrared beam radiation through a solution layer. Also known asinfrared
reflection-absorption spectroscopy (IRRAS), this technique has the
disadvantage of requiring additional optics to direct the beam from the
instrument to the cell and then to direct the beam from the cell to the
detector. Radiant power is dissipated over the longer beam path, resulting
in a decrease in the signal. Other disadvantages include the difficulty of
aligning the optics and the added expense of the optics (67). The primary
advantage of a reflective cell over a transmission apparatus is the ability to
perform spectroelectrochemical studies on adsorbed layers.
External infrared reflectance spectroscopy is a technique in which the
beam passes through an infrared transparent medium and a sample
solution, reflects off of a reflective electrode, and exits through the solution
and infrared-transparent medium (68). This in situ technique was first
developed and used by Bewick, Kunimatsu, and Pons, who in 1980,
30


Figure 1.10. Thin-layer infrared transmittance cell. Adapted from (67).
31


examined the vibrational spectrum of adsorbed indole on a platinum
electrode (71). Seki and coworkers designed a cell in which the thickness
of the solution layer above the electrode surface could be controlled. This
greatly reduced the interference from the electrolyte (Figure 1.11) (72).
Bockris and Yang designed an IRRAS system in which a stream of fresh
electrolyte continuously flowed over the working electrode, reducing the
optical interference from evolved gaseous species (73). While a
considerable body of work has been accomplished on studying the infrared
spectroelectrochemical properties of small molecules, similar analyses for
proteins has been limited to the use of transmission cells
1.4 Research Objectives
The central theme of this work was the characterization of ferritin
through cyclic voltammetry and infrared spectroscopy. The objectives of
this research were two-fold. The first objective was to investigate the
direct electron transfer of ferritin at a bare gold electrode. Because the
oxidation and reduction of the iron core of ferritin is crucial to its function in
iron uptake and release, cyclic voltammetry is ideally suited to this study.
To characterize the direct electron transfer, ferritin was adsorbed onto a
bare gold electrode and the potential scanned using a range of ionic
strengths, ferritin concentrations, scan rates, pH values, and adsorption
32


times. The second research objective was to design, build, and test an
infrared reflection absorption electrochemical cell capable of analyzing the
electrochemical and spectroscopic properties of ferritin. To accomplish
this, an infrared spectroelectrochemical cell and optical bench enclosure
apparatus was constructed, and the optics aligned.
33


Pt Counter
CaF2 Window Electrode
IR Beam
Working Electrode
Surface
Figure 1.11. A thin-layer electrochemical cell with a CaF2 window and an
adjustable working electrode. Adapted from (72).
34


2. Experimental
2.1. Electrochemistry of Ferritin on Bare Gold
2.1.1 Materials
Polycrystalline gold foil (99.99%) was purchased from Alfa-Johnson
Matthey, Danvers, MA. The liquid chromatography column was purchased
from Kontes Glass Co. (Vineland, NJ), G-200 Sephadex from Pharmacia
(Alameda, CA), and Bio-Rad Protein Assay Dye from Bio-Rad
Laboratories (Hercules, CA). PMSF (phenylmethyl-sulfonyl fluoride)
(>99%) was supplied by Aldrich Chemical Company (Milwaukee, Wl).
Sodium hydroxide (Analytical Reagent), sulfuric acid (98%), and sodium
phosphate monobasic (Analytical Reagent) were purchased from
Mallinkrodt Specialty Chemicals Co. (Pari, KY). Bovine serum albumin
(fraction 5 powder) and horse spleen ferritin (>85%) was supplied by
Sigma Chemical Company (St. Louis, MO). UV measurements were
measured using a Perkin-Elmer (Norwalk, CT) Model 552 UV-Visible
Spectrophotometer. Cyclic voltammetry and electrochemistry on the
infrared cell was conducted using a Cypress Model Omni 90 potentiostat
(Lawrence, KS) and a BioAnalytical Systems Model RXY recorder (West
Lafayette, IN).
37


2.1.2 Methods
2.1.2.1 Electrode Assembly and the Electrochemical Cell
Polycrystalline gold foil was fused to a 0.125 mm gold wire and then
shaped into a 3 mm diameter cylinder. The electrode was then inserted
into a 5 mm inner diameter glass sheath.
The cell in which cyclic voltammetry of the adsorbed layer of ferritin
was conducted consisted of two glass chambers separated by a glass frit
(Figure 2.1). The glass sheath was housed in a Teflon stopcock, which
held the electrode assembly in place using a 24/25 ground glass joint
(Figure 2.1). When the stopcock was opened, a stream of pressurized
nitrogen was allowed to blanket the electrode, shielding it from atmospheric
oxygen. The cell in which the ferritin was adsorbed was machined from
Teflon. The adsorption cell contained the reference and auxiliary
electrodes (encased in separate polyethylene tubes, plugged at one end
with Vycor disks) and the working electrode (Figure 2.2). The cell was
machined by D. Zapien.
38


N2 Gas
Teflon Stopcock
Glass Sheath
Gold Working
Electrode
Ag/AgCI
Reference
Electrode
Platinum
Counter
Electode
Glass Frit
Figure 2.1. Typical electrochemical ceil used for conducting cyclic
voltammetry experiments of adsorbed layers of ferritin, including glass
sheath and Teflon fitting. Both chambers contain electrolyte solution.
The larger chamber contains the ferritin to be adsorbed onto the gold
working electrode.
39


The electrode was cleaned by annealing in a natural gas/air flame,
soaking in chromic acid cleaning solution, rinsing with distilled water, and
finally by electrochemical cycling in 1M sulfuric acid between -0.35 V and
1.50 V (versus Ag/AgCI electrode). Electrochemical cycling was repeated
until a clean scan of bare gold was obtained (Figure 2.3).
2.1.2.2 Measurement of the Electrode Surface Area
The surface area was determined by measuring the anodic charge
of a monolayer of iodine adsorbed onto the electrode. The electrode was
cleaned in 1M H2SO4, and a cyclic voltammogram scanned between -0.35
V and 1.5 V. The electrode was submerged in an aqueous solution of 0.1
M potassium iodide for five minutes at open circuit potential to effect the
adsorption of iodide, then transferred to a solution of 1M H2S04 where the
electrode was rinsed free of unadsorbed iodide. A cyclic voltammogram
was scanned between -0.35 V and 1.5 V (Figure 2.4). The half-reaction
for the oxidation of adsorbed iodide is:
Uds + 3H20 - IO3' (aq) + 6H+ (aq) + 5e'
The anodic peak of adsorbed iodine was cut and weighed, and the
electrolytic charge represented by the peak was calculated. The surface
area determined using the relationship
A = Qox, i / 5 F Ti where T| = 1.04 nmol/cm2.
40


Gas flow valve
Counter electrode1
<

-Adapter
Working electrode lead
Glass sheath
Teflon adapter
Glass adapter
Teflon cell
Reference electrode
Working electrode
Figure 2.2. Teflon electrochemical cell used in ferritin adsorption step.
41


CURRENT (nA)
Figure 2.3. Cyclic voltammogram of clean bare gold, scanned from
-0.35V to 1.50V. The scan was conducted in the presence of 1M H2S04.
42


CURRENT (nA)
Figure 2.4. Cyclic voltammogram of adsorbed iodide, used in determining
electrode surface area.
43


2.1.2.3 Ferritin Current-Potential Curves
To obtain the current-potential curves for ferritin, a clean gold
electrode was immersed into a ferritin solution at 0.00 V. The potential was
scanned negative at 100 mV/s, and the potential held at -0.500 V. Under
nitrogen, the electrode was transferred to a cell of deaerated phosphate
buffer, where the electrode was rinsed free of unadsorbed ferritin at -0.500
V. The electrode was then placed in another cell containing fresh,
deaerated phosphate buffer solution, and the potential was cycled between
-0.50 V and 0.20 V.
To determine the charge of the anodic peak, the current of the
anodic peak of adsorbed ferritin was integrated by cutting and weighing.
The experimental packing densities were determined by the Faraday law:
T = Q/nFA
where Q is the integrated charge of the anodic peak and A is the electrode
area in cm2.
All measurements of packing density and peak current were
accomplished in three independent experiments. Error bars represent the
standard deviation of these triplicate measurements.
2.1.2.4 Ferritin Purification
Ferritin was eluted through a Sephadex G-200 size-exclusion gel
(protein fractionation range 5 600 kD) to remove lower molecular weight
44


proteins. The column (2.5 cm diameter x 20 cm length) was equilibrated
with 500 ml of G-200 Buffer (20 mM sodium dihydrogen phosphate, 0.9%
NaCI, 0.2 mM PMSF, 0.02% NaNs). The total protein concentration was
determined by the method of Bradford (81). A standard curve was
constructed by measuring the absorbances (595 nm) of a series of bovine
serum albumin (BSA) standards complexed with G-250 Coomassie blue
dye. The ferritin sample was complexed with the dye, and its absorbance
was projected onto the concentration axis of the standard curve.
2.2 Infrared Spectroelectrochemistry
2.2.1 Materials
Infrared measurements were conducted using a Nicolet 550 FTIR
spectrometer. The polycrystalline gold disk (99.99%), used as the working
electrode in the infrared cell, was purchased from Corning Corporation
(Saddlebrook, NJ). The spectroelectrochemical cell block and lid was
constructed from a Teflon block, purchased from Cadillac Plastics
(Denver, CO). The plunger assembly was machined from polyimide,
purchased from Regal Plastics (Denver, CO). The CaF2 dove prism used
on the infrared cell was purchased from Infrared Optics, Inc. Precision
machining of the block and plunger was conducted by ADAM Instrument
Company, Inc. (Cincinnati, OH). Machining of the Teflon lid and all other
cell attachments and supports was accomplished in the laboratory. The
45


differential micrometer, used for adjustment of the plunger, was purchased
from Shop Tools, Inc. (Aurora, CO). The spectroelectrochemical cell and
reflection optics were protected by an enclosure made from Plexiglas,
obtained from Regal Plastics (Denver, CO.
2.2.2 Methods
Measurements were conducted at room temperature using aqueous
solutions. Infrared scans of ferritin were accomplished using a
conventional transmission cell (International Crystal Laboratories) in the
sample chamber of the instrument at open circuit potential. In situ
ferricyanide reduction was conducted using the spectroelectrochemical
cell. Ferricyanide solution was injected into the cell injection port through
Teflon tubing, using a syringe. The potential was stepped from 0.200 V
to -0.100 V, and single-beam infrared spectra were taken at 2 minute
intervals.
46


3. Direct Electron Transfer of Ferritin
Adsorbed on Bare Gold Electrodes
3.1 Adsorption and Redox Activity of Ferritin
A clean gold electrode was immersed in ferritin solution at 0.20
V. The potential was then ramped from 0.20 V to -0.50 V at a scan rate of
100 mV/s. After rinsing and immersion into phosphate buffer electrolyte
solution at a potential of-0.50 V, the potential was scanned from -0.50 V
to 0.20 V. From this scan, a well-defined current potential curve was
observed (Figure 3.1). The anodic peak appeared at -0.13 V and the
cathodic peak was at 0.00 V. Of interest, is the fact that the currents
produced by the oxidation and reduction of cytochrome c on bare gold are
very small, while the current potential curves generated by a monolayer of
ferritin are easily measured (76). Ferritin exhibits intense peaks relative to
cytochrome c due to the greater number of electrons transferred between
the electrode and the ferritin molecule. This permits the use of
conventional cyclic voltammetry to study current-potential behavior. As
shown in Figure 3.2, the anodic peak of the second cycle is smaller than
that of the first cycle. This suggests that either some ferritin desorbs, or
47


-0.50 -0.10 0.30
POTENTIAL (Volts vs. Ag/AgCl)
Figure 3.1. Cyclic voltammogram of ferritin. The potential was cycled from
-0.50 V to 0.20 V. The dashed portion of the curve indicates a successive
scan under identical conditions
48


that the ferritin molecule has become less electroactive with each cycle. It
is not clear why the cathodic peak is smaller than the anodic peak. The
mere desorption of ferritin is unlikely given that the anodic peak on the
second sweep is larger than the previous cathodic peak. One possible
explanation for this phenomenon is that the ferritin molecule changes its
orientation on the metal surface when switching between reduced and
oxidized forms. This reversible change in orientation has been proposed
for cytochrome c by Sagara and co-workers (77). The packing density for
adsorbed ferritin was estimated to be 0.40 0.05 pmol/cm2 by the
integrated anodic charge and the Faraday law. Consecutive cycling of the
i-E curve, shown in Figure 3.2, causes the peak area to diminish,
indicating that the layer is not stable to repeated cycling. The reason for
this behavior is not clear. However, this result may indicate that with
multiple cycles, the protein gradually assumes a new orientation that is
electroinactive. This behavior is observed for cytochrome c3 at a bare
mercury electrode.
The voltammogram of ferritin (Figure 3.1) contains an anodic
shoulder at 0.11 V and a small cathodic peak at 0.06 V, which possibly
belong to a common redox couple. A similar phenomenon was reported in
the current potential curve of cytochrome c by Szucs and Novak, who
49


Figure 3.2. Cyclic voltammogram demonstrating consecutive cycling of
ferritin from -0.50 V to +0.20 V.
50


postulated that a small peak at a potential more positive than that of the
principal peak was due to oligomerized cytochrome c (51,52).
Oligomerization would alter the globular network, which in turn would
change the energy of the electroactive center, resulting in a change in the
potential. The extra feature does not appear in cytochrome c adsorbed at
promoter modified electrodes. One explanation why this phenomenon
does not occur on these electrodes is that the adsorbed promoter layer
may be anchoring the protein, limiting its lateral movement on the electrode
surface, while a bare electrode limits protein movement to a lesser extent,
allowing aggregation. This postulate may also explain ferritin's behavior on
gold and modified gold electrodes. Ferritin adsorbed onto an electrode
modified with mercaptopropionate groups does not exhibit peaks other than
those of the principally adsorbed species even though the scanning
conditions are the same. Ferritin may be anchored to the electrode by the
promoter layer in a fashion similar as cytochrome c.
In order to demonstrate that the electrochemical signals were due to
ferritin adsorbed onto the electrode surface and not ferritin in solution, the
current-potential curves were examined at scan rates of 50,100, 200, and
400 mV/sec. A plot of anodic peak current versus scan rate gives a linear
relationship, with a linear regression slope of 0.384 and a correlation
51


coefficient of 0.998 (Figure 3.3). According to theory, this correlation
indicates that the voltammetric signal is due to adsorbed ferritin. The
relationship between peak current and scan rate for adsorbed ferritin is
given by the following equation (91):
ip = (n2F2T/4RT) v
If the observed current arose from free iron in solution, a plot of scan rate
versus peak current would likely yield an upsloping curve given by the
equation (91):
lp = (2.69x1 O^n^ADo^V^Co*
Ferritin is a complex system in that potentially 1500 electrons are
transferred. If we assumed n = 1500, the calculated slope should be 0.85,
roughly 2000 times the experimental slope. The reason for the
discrepency is that the equation applies strictly to those systems for which
n is known and whose voltammetry behaves according to the known n-
value. If the system behaved as though n = 1500, however, the anodic and
cathodic branches would be extremely narrow, and the peak currents very
large. The large currents would then give a large slope, as expected. The
52


Peak Current (*/A)
Scan Rate (mV/sec)
Figure 3.3. Plot Illustrating how the peak current of ferritin varies with scan
rate. Scans were carried out in the presence of 1 M phosphate buffer, pH
7.0
53


voltammetry, however, resembles that of a one-electron transfer system;
as though 1500 individual one-electron transfer reactions are occurring.
There are several examples in the literature in which electrons are
transferred to and from molecules containing multiple redox centers (up to
1200 per molecule in some electroactive polymers) while the voltammetry
has the appearance of a one-electron transfer system (95-98).
3.2 Potential Dependence of Ferritin Adsorption
The degree to which ferritin adsorbs onto a bare gold electrode is
highly dependent upon potential. For example, when a clean gold
electrode is immersed in ferritin solution at 0.20 V, rinsed with phosphate
buffer, and the potential scanned between -0.50 and 0.20 V, no Faradaic
current is produced (Figure 3.4, panel a). However, when the electrode is
immersed in ferritin solution at potentials of-0.10 V -0.30 V, and -0.50 V,
it was demonstrated that the amount of faradaic charge from adsorbed
ferritin is dependent upon potential (Figure 3.4, panels a-d). Assuming that
all ferritin molecules on the anode are completely electroactive, the packing
densities for adsorption potentials of 0.0, -0.10, -0.30, and -0.50 V are 0.0,
0.0, 0.18 0.02, and 0.26 0.03 pmol/cm2, respectively. This data
suggests that ferritin is adsorbed in an electroactive state only when it is
first reduced. It is known that conformational changes in protein structure
may occur with changes in oxidation state of the reaction center (78,79).
55


-0.30 V c
____I-----------------1____1
-0-5*0 +0*2.0
Po*hg Figure 3.4. Dependence of ferritin adsorption on potential. Cyclic
voltammetry was carried out in the presence of 1M phosphate buffer, pH
7.0.
56


For example, as adsorbed ferrocytochrome c is converted to
ferricytochrome c, the reaction center entropy change becomes more
negative, indicating that the globular sphere is less rigid when the
electroactive center is in the oxidized form. It is possible that
conformational changes in quaternary structure, induced by the reduction
of the protein, allow adsorption onto the gold surface while maintaining
electroactivity.
3.3 Dependence of Packing Density on Ionic Strength, Ferritin
Concentration, and Adsorption Time
3.3.1 Ionic Strength
Of interest in characterizing the interactions between proteins and
bare electrodes is whether the interactions are primarily hydrophobic or
hydrophilic. One method of characterization involves observing changes in
packing density (the density of ferritin molecules adsorbed to the electrode
surface) with changes in ionic strength. Higher ionic strength favors
hydrophilic conditions; therefore, an increase in relative packing density at
higher ionic strengths may indicate that hydrophobic interactions dominate
at the protein-electrode interface. In the case of ferritin, as the ionic
strength is increased from 0.30 to 1.5 M, the relative packing density (the
absolute packing density normalized with respect to the greatest packing
density) increases to a maximum value beginning at about 1.0 M (Figure
57


Ionic Strength (Molar)
Figure 3.5. Variation of ferritin packing density with ionic strength. Scans
were carried out in phosphate buffer, pH 7.0. Ionic strength was varied
from 0.3 M to 1.5 M.
58


3.5). Apparently, the hydrophobic regions of ferritin interact with the gold
surface while the electroactivity of the protein is preserved. This result
suggests that the ferritin-gold interface is primarily hydrophobic in nature.
Similar behavior is observed in the adsorption of ferritin at tin-doped indium
oxide (TIO) electrodes, in which ferritin exhibits a limiting packing density at
about 1 M ionic strength (55).
3.3.2 Ferritin Concentration
Ferritin concentration has a direct effect on packing density.
Throughout this study, the ferritin concentration was held constant for each
experiment. An adsorption isotherm for ferritin at bare gold is shown in
Figure 3.6. For the range of concentrations studied, the packing density
showed a direct linear relationship with ferritin concentration. Globular
proteins exhibit well-defined plateau regions for dilute to semidilute
regimes. If higher concentrations of ferritin were available, the ferritin
isotherm would likely show a plateau at the higher concentration. Although
the packing density for the highest concentration of ferritin is approximately
0.40 pmol/cm2, the maximum packing density for a ferritin monolayer is
about 1.1 pmol/cm2, calculated from its footprint area.
59


Figure 3.6. Ferritin adsorption isotherm. The packing density varied
linearly with ferritin concentration throughout the concentration range
tested.
60


3.3.3 Adsorption Time
Adsorption is a key step in observing the behavior of proteins on
bare electrodes. Adsorption and desorption involve several distinct steps:
transport to the electrode surface, anchoring to the surface, structural
rearrangement, disengagement, and transport from the electrode. Protein
crowding on the surface of an electrode may change the structure or
orientation of a protein (80). In order to characterize the adsorption
process of ferritin to a bare gold surface, the adsorption time of ferritin at
-0.50 V was varied (Figure 3.7). After 90 seconds of adsorption time, a
limiting packing density was achieved. These data suggest that the
adsorption rate is limited more by the diffusion of ferritin to the gold surface
than by structural rearrangements or changes in orientation as the protein
adsorbs to the electrode. These results are in contrast to the long periods
of time required for ferritin to reach tin-doped indium oxide electrodes (55).
Because horse spleen ferritin has an isoelectric point of 4.6, it has a net
negative charge at neutral pH. The reasons for the large differences are
not clear; however, it appears that the negative charge of the surface
oxides on the tin-doped indium oxide electrode is not solely responsible.
Using gold electrodes, potentials negative of the potential of zero charge
were used; therefore, some electrostatic repulsion is also expected
between the gold surface and the negatively charged ferritin molecules.
61


(pmol/cm2)
Adsorption Time (seconds)
Figure 3.7. Relationship between ferritin packing density and adsorption
time. Adsorption was carried out at -0.50 V in 1M phosphate buffer, pH
7.0.
62


4. Design of the Infrared Reflection Absorption
Spectroscopy (IRRAS) Apparatus
4.1 Spectroelectrochemical Cell
4.1.1 Description
The infrared reflection-absorption spectroscopy (IRRAS) cell is an
external reflectance cell designed to simultaneously serve as an
electrochemical cell and infrared spectroscopic cell (Figure 4.1).
Features of the cell include Teflon construction for chemical inertness, a
CaF2 dove prism that serves as an infrared-transparent window as well
as a barrier for the cell solution, a gold disk embedded in a plunger
assembly serving as the working electrode and IR reflector, a differential
micrometer for adjusting the solution layer thickness, and a stainless-
steel base for mounting purposes.
4.1.2 Design
A fundamental design consideration is the type of cell used to
carry out reflection absorption measurements. Weighing potential
63


MOUNTING HOLES
Figure 4.1. Diagram of the infrared reflection-absorption spectroscopy
(IRRAS) cell.
64


advantages versus disadvantages is critical in choosing a specific cell
design. The principal advantages of external reflectance cells include the
ability to control the solution thickness above the electrode surface and to
provide a uniform flow of solution over the electrode (67). The cell was
designed with several considerations in mind, including chemical
inertness, effective collection of the infrared beam, reproducible sample
thickness, and mechanical stability of the cell. Chemical inertness was
important because any reactions between the ceil material and
components in the solution might generate species that interfere with
spectroscopic and electrochemical methods. For optimal resistance to
such reactions, the cell was machined from a 4 in3 Teflon cube, while
the plunger, to which the gold disk was mounted, was constructed from a
polyimide rod. Polyimide also offered the advantage of being resistant to
cold-flowing, a consequence of high pressures on a soft substance, such
as Teflon.
Another factor critical to the sensitivity of the measurements is the
proper direction of the infrared beam to the gold working electrode. This
is important because of the loss of any portion of the infrared beam would
result in reduced sensitivity and decreased signal to noise ratio. To
minimize reflection losses, the infrared beam approached the dove prism
at normal incidence. After passing through the prism, the beam impinges
65


on the working electrode at an angle of 60 degrees with respect to the
electrode surface normal (Figure 4.2). A dove prism made of CaF2 was
used because of its transparency in the mid-infrared region (4000-1100
cm'1) and its insolubility in aqueous media.
A 0.25 inch channel extended from the gold disk to the back of the
polyimide plunger. A copper wire was fed into the channel through a hole
drilled into the side of the plunger, located about one-inch from the end of
the plunger. A Teflon rod was inserted into the channel through the back
of the plunger, pushing the copper wire against the back of the gold disk.
A spring was inserted behind the Teflon rod, followed by a pipe-threaded
plug screwed into the back of the plunger. Two 2-inch lengths of 0.125
inch polyethylene tube were plugged on one end with 0.125 inch Vycor
disks. One of the tubes was filled with a 1 M NaCI solution, followed by
insertion of a 0.1 mm diameter gold wire; this assembly served as the
auxiliary electrode. The other tube was used to make a Ag/AgCI
reference electrode. A 1-mm channel linked the solution in front of the
gold disk with the sink located on the top of the reflection cell, allowing for
all three electrodes to be in electrical contact through the solution.
To provide reproducible sample thickness between
measurements, a differential micrometer (non-rotating) with 0.1 pm
resolution was installed on the backside of the cell using a four-poled
66


Figure 4.2. Schematic diagram illustrating how the beam passes through
the dove prism, reflects off of the gold disk electrode, and passes through
the dove prism on the other side.
67


stainless steel mount. The micrometer tip pushed against the stainless
steel pipe thread plug located on the back of the plunger. Adjustment of
the micrometer allowed for a variable distance between the gold
electrode and the dove prism.
4.2 Enclosure
4.2.1 Description
An air-tight enclosure was constructed from aluminum and
Plexiglas to provide a moisture-free environment for the optics, the cell,
and the detector, while preventing the entry of external sources of
radiation (Figure 4.3). An optical bench was constructed from 5/16 inch
thick anodized aluminum re-inforced by 1.5 inch wide skirts bolted
underneath the perimeter of the base plate. Four 1-inch square rods,
located at each corner, raised the bench to the height of 6 inches. The
optical bench was constructed by Mr. Fedri Marrugo. The enclosure
rested on top of the optical bench while the bench itself was bolted to the
FTIR spectrometer using 3/8 inch bolts. Finally, an aluminum flange,
housing a 2 inch diameter CaF2 window, was bolted on one side to the
enclosure, and on the other side to the FTIR enclosure. The window
flange was also machined by Mr. Fedri Marrugo.
68


Figure 4.3. Plexiglas enclosure containing the optics, the
spectroelectrochemical cell, and the detector.
69


4.1.3 Design
Important considerations in the design of the enclosure include
durability, accessibility to the interior components, and airtightness.
Plexiglas was chosen because of its strength and ease with which it is
machined. The Plexiglas sheets were joined with epoxy resin to
provide strong seams. Angled aluminum provided additional strength at
the bottom rim of the enclosure while providing a means of bolting it to
the bench. To enhance accessibility to the interior optical components,
the enclosure was built with a removable top attached to the enclosure
body with stainless steel cap screws. A utility plate, machined from
Plexiglas, was attached to the side opposite of the FTIR instrument. This
device afforded a common entrance for all solutions and electrode leads,
and also provided added accessibility to the internal components. To
make the apparatus airtight, all seams were sealed with silicone caulking.
Foam strips provided the seal between the enclosure base and optics
bench, and between the removable top and the enclosure, while a rubber
gasket provided a seal between the enclosure and the utility plate.
4.3 Optics
4.3.1 Internal Optics
The internal optics include those components within the FTIR
enclosure: the source, source optics, interferometer components, and
70


the mirror directing the beam out of the spectrometer. The beam
originates from a heated silicon carbide rod (Glowbar), and is directed
into the interferometer by a collimating mirror. The oscillating translation
of the moving plane mirror generates an interference pattern of radiant
power with time, called an interferogram. The interferogram is directed
through the sample, and into a detector. The analog signal produced by
the detector is digitized, and inputted into the processor where it is
transformed into the spectrum in the frequency domain.
Because the spectroelectrochemical cell was located on the external
optics bench, it was necessary for the beam to be routed outside of the
instrument box to the external optics. This was accomplished by changing
the way one of the off-axis parabolic mirrors was originally used (Figure
4.4). An aluminum mount of the proper height was machined and
mounted at the exit of the interferometer. The mirror was inverted,
attached to the mount, and aligned to focus the beam directly through the
exit port. After passing through the flanged CaF2 window, the beam
continued a path defined by the external optics.
4.3.2 External Optics
The external optics include a collimating mirror, a focusing mirror,
the spectroelectrochemical cell, and a detector mirror (Figure 4.4). Once
71


Figure 4.4. Schematic diagram illustrating the optical setup of the
interferometer, the optics mirrors, the spectroelectrochemical cell, and the
detector.
72


the beam leaves the FTIR spectrometer, a mirror collimates the
beam and directs it to a focusing mirror, which in turn focuses the beam
onto the surface of the gold working electrode. The beam is then
reflected from the electrode and is collected by the detector mirror, where
it is focused onto the detector.
4.3.3 Detector
A triglycine sulfonate detector, originally installed in the FTIR
spectrometer, was removed from the instrument and installed on the
external bench. A longer cable was constructed to permit the operation
of the detector in the remote location.
4.4 Testing and Evaluation of the Apparatus
4.4.1 Reduction of Ferricyanide
To validate the ability of the instrument to collect data, we tested
the instrument on a simpler system. Areas of concern regarding the cell
included 1) whether the cell would work to exhaustively electrolyze the
sample, 2) whether the infrared spectra of the analyte could be
measured, and 3) whether the cell design was adequate to prevent
appreciable diffusion of analyte from the cell.
To investigate the soundness of the cell design with regard to the
aforementioned concerns, the reduction of ferricyanide in the
spectroelectrochemical cell was conducted. Prior to the experiment,
73


infrared spectra of aqueous solutions of ferrocyanide and ferricyanide
were taken. The ferricyanide peak was present at 2104 cm'1 while the
ferrocyanide peak absorbed strongly at 2033 cm'1 (Figure 4.5a). Upon
stepping the potential from 0.200 V to -0.100 V, a gradual decrease in
the intensity of the ferricyanide peak was observed. After 7.5 minutes,
the peak had completely disappeared (Figure 4.5b). A simultaneous
increase in peak intensity of the ferrocyanide peak had also occurred.
The above data suggest that the exhaustive electrolysis readily occurs
with the spectroelectrochemical system and that diffusion of analyte away
from the cell does not readily occur.
4.4.2 Spectrum of Ferritin
Infrared spectra were taken at ferritin concentrations of 51.0 mg/ml
and 102 mg/ml (Figure 4.6). The data obtained indicate that ferritin
chromatographed over Sephadex G-200 is of sufficient concentration to
obtain infrared data. Additionally, some limited structural information may
be inferred from the spectrum. The peaks at 1650 cm'1 and 1553 cm'1
are known as the amide I and the amide II bands, respectively. The
amide I band arises from the C=0 stretching vibrations of the backbone
and peptide linkages while the amide II band arises from out-of-phase
combinations of N-H in-plane bending and C-N stretch vibrations of
peptide linkages (81).
74



(b)
Figure 4.5 (a) Infrared spectra of ferricyanide and ferrocyanide. (b) Reduction
of ferricyanide with respect to time. The potential was stepped from 0.200 V
to -0.100 V.
75


Absorbance
51 mg/mL
I------------1------------1------------i
1800 1700 1600 1500
Wavenumbers (cm'1)
Figure 4.6. Infrared spectrum of ferritin at 51 mg/ml and at 102 mg/ml.
76


The predominant secondary structure of a protein is a strong
determinant of the absorbance maxima of the amide I band. The amide I
band of a protein with a predominant a-helical structure is found near
1656 2 cm'1 whereas a protein having primarily a /5-sheet structure will
have a maximum between 1643 and 1631 cm'1 (81). In the case of
ferritin, the amide I band maxima is indicative of an a-helix secondary
structure. This is consistent with the extensive a-helical structure of
ferritin H and L subunits (5,10,39). Also supporting this deduction is the
lack of any shoulder between 1643 and 1631 cm'1, indicating a lack of fi-
sheet structure (81).
Another point of interest is the peak assignments of both maxima
of the amide I peak Earlier studies have established that bands at 1648
2 cm'1 may be assigned to disordered secondary structure (81). Given
this fact, the maximum at 1650 cm'1 may arise from disordered secondary
structure within the H and L subunits. Contradicting this notion, however,
is the predominance of a-helices and the absence of a significant amount
of disordered secondary structure upon visual inspection of the crystal
structure. Another possible explanation is that each of the two types of
subunits are responsible for each of the maxima. Additional studies will
77


be required to definitively ascertain whether the additional maximum is
due to disordered structure or the two different subunit types.
4.4.3 Problem Areas
The IRRAS apparatus had several problems which prevented the
collection of reduced minus oxidized spectra of ferritin. First, the cell is
not suitable for electrolyzing dissolved ferritin. Second, because of the
additional optics required to use the reflection absorption apparatus, a
significantly reduced signal to noise ratio would result. Finally,
atmospheric water vapor gives strong absorption bands in the regions of
Amide I and Amide II. We were unable to remove these signals by
subtraction purging alone. The development of a mathematical approach
to subtract these bands will be required.
78


5. Summary and Conclusion
5.1 Summary
The uptake and release of iron is central to the function of ferritin.
Additionally, redox reactions are crucial to this uptake and release.
Therefore, to study the structure and function of ferritin, we characterized
its electrochemical behavior on bare gold. We were able to obtain data
suggesting that ferritin oligomerizes on bare gold, but not on a gold
electrode modified with mercaptopropionate groups. We therefore
postulate that ferritin is anchored to the electrode by the promoter layer.
Additionally, our data suggest that a reversible change in orientation of
the ferritin molecule may be occurring on the electrode surface. The
relationship between ferritin packing density and scan rate was found to
be linear. From these data, it was determined that the voltammetric
signals were from ferritin adsorbed onto the gold electrode and not ferritin
in solution.
A study of the potential dependence on ferritin adsorption
suggested that ferritin is adsorbed in an electroactive state only when it is
first reduced. Additionally, the data suggest that it is possible that a
change in quaternary structure occurs during the reduction process,
79


allowing the protein to be adsorbed onto the gold surface while
maintaining its electroactivity.
Additionally, studies of packing density at various ionic strengths,
ferritin concentrations, and adsorption times were conducted. From the
ionic strength data, it was concluded that hydrophobic interactions
predominate at the ferritin-gold surface. The packing density of ferritin
increased with dissolved ferritin concentration. From the adsorption time
data, it was determined that the adsorption rate is limited more by the
diffusion of ferritin onto the gold surface rather than by structural
rearrangements or changes in orientation following initial attachment to
the electrode surface.
An infrared reflection-absorption spectroelectrochemical cell was
designed and constructed, along with an optical bench enclosure, and
mounts for reflection optics. Testing of the spectroelectrochemical cell
using ferricyanide revealed that the cell had the ability to exhaustively
electrolyze an analyte in fairly short time and that infrared spectra can be
recorded in situ.
5.2 Conclusion
As shown in this study, the electrochemical data of ferritin
adsorbed on bare gold are consistent with what has been suggested in
other studies: that the protein molecules may be laterally mobile, forming
80


oligomers. Perhaps more important, the data suggest that the structure
of the protein sheath is linked to the oxidation state of the core iron.
The spectroelectrochemical cell can be used to obtain in situ
infrared spectra. However, the cell is still not suitable for obtaining the
reduced-minus-oxidized difference spectrum of ferritin. There are several
reasons for this. First, the direct electron transfer of dissolved ferritin
does not occur appreciably at bare gold electrodes. The ferritin in
solution must be exhaustively electrolyzed in order for the difference
spectrum to be meaningful. An electrode system capable of effecting the
direct electron transfer of ferritin must first be found. Secondly, the use of
a reflection-absorption cell to obtain transmission spectra typically results
in a reduced signal to noise ratio due to the additional optics required to
transmit the infrared beam outside of the FTIR to the external bench. A
transmission cell which can be mounted inside the spectrometer,
resulting in a higher signal-to-noise ratio, will be needed to observe the
small difference bands. Lastly, since water absorbs strongly in the
regions of the Amide I and Amide II bands, a method must be developed
in order to eliminate the atmospheric water absorption bands from the
difference spectrum.
81


References
1. P. M Harrison and T. H Lilley (1989) Ferritin. in Iron Carriers and Iron
Proteins, ed. Thomas M. Loehr. (New York: VCH Publishers, Inc.) 123-238.
2. J. C. G. Walker (1977) The Evolution of the Atmosphere. (New York:
MacMillan) 262-264.
3. Biedermann, G., Schindler, Acta. Chem. Scand. 11 (1957) 731-740.
4. B. F. Matzanke, G. M. Matzanke, K. N. Raymond (1989) Siderophore-
Mediated Iron Transport. in Iron Carriers and Iron Proteins, ed. Thomas M.
Loehr. (New York: VCH Publishers, Inc.), 3.
5. E. C. Theil, Ann. Rev. Biochem. 56 (1987) 290.
6. P. M Harrison, F. A. Fischbach, T. G. Hoy, and G. H. Haggis, Nature 216
(1967) 1188.
7. W. H. Massover, J. M. Cowley, Proc. Nat. Acad. Sci. USA 70 (1973)
3847-3851.
8. J. L. White, J. M. A. Smith, P. M. Harrison, Adv. Inorg. Biochem. 5 (1983)
39-50.
9. S. H. Banyard, D. K. Stammers, P. M. Harrison, Nature 288 (1980) 298-
300.
10. E. C. Theil, Advances in Enzymology 63 (1990) 423.
11. G. H. Haggis, J. Mol. Biol. 14 (1965) 598.
12. P. M. Harrison, F. A. Fishbach, T. G. Hoy, G. H. Haggis, Nature
(London) 216 (1967) 1188.
13. F. Constanzo, M. Colombo, S. Staemplfli, C. Santoro, M. Marone, R.
Frank, H. Delius, R. Cortese, Nucleic Acid Res. 14 (1986) 721.
82


14. M. Heusterspreute, R. R. Crichton, FEBS Lett. 129 (1981) 322.
15. D. Boyd, C. Vecoli, D. M. Belcher, S. K. Jain, J. W. Drysdale, J. Biol.
Chem. 260(1985) 11755.
16. C. Santoro, M. Marone, M. Ferrone, F. Constanzo, M. Colombo, C.
Minganti, R. Cortese, L. Silengo, Nucleic Acid Res. 14 (1986) 2863.
17. C. Wustefeld, R. R. Crichton, FEBS Lett. 150 (1982) 43.
18. M. T. Murray, K. White, H. N. Munro, Proc. Nat. Acad. Sci. USA 84
(1987) 7438.
19. E. A. Leibold, H. N. Munro, J. Bioi. Chem. 262 (1987) 7335.
20. L. F. Dickey, S. Sreedharan, E. C. Theil, J. R. Didsbury, Y. -H. Wang, R.
E. Kaufman, J. Biol. Chem. 262 (1987) 7901.
21. P. W. Stevens, J. B. Dodgson, J. D. Engel, Mol. Cell. Biol. 7 (1987)
1751.
22. J. F. Collawn, L. K. Gowan, H. Crow, C. Schwabe, W. W. Fish, Arch.
Biochem. Biophys. 259 (1987) 105.
23. P. M. Harrison, T. H. Lilley (1989) Ferritin. in Iron Carriers and Iron
Proteins, ed. Thomas M. Loehr. (New York: VCH Publishers, Inc.), 126.
24. S. Otsuka, H. Maruyama, I. Listowsky, Biochemistry 20 (1981) 5226-
5232.
25. J. M. Addison, J. E. Fitton, W. G. Lewis, F. May, P. M. Harrison, FEBS
Lett. 64 (1983) 139-144.
26. D. J. Lavoie, K. Ishikawa, I. Listowsky, Biochemistry 17 (1978) 5448-
5454.
27. J. F. Collawn, Jr., W. W. Fish, Comp. Biochem. Physiol. B 78 (1984)
653-656.
28. J. F. Collawn, Jr., P. L. Lau, S. L. Morgan, A. Fox, W. W. Fish, Arch.
Bioch. Biophys. 233 (1984) 260-266.
83


29. W. H. Massover, Biochim. Biophys. Acta, 829 (1985) 377-386.
30. M. C. Linder, G. M. Nagel, M. Roboz, D. M. Hungerford, Jr., J. Biol.
Chem. 256 (1981) 9104-9110.
31. G. A. Clegg, J. E. Fitton, P. M. Harrison, A. Treffry, Prog. Biophys. Mol.
Biol. 36 (1981) 53-86.
32. D. W. Rice, B. Day, J. M. A. Smith, J. L. White, G. C. Ford, etal. FEBS
Lett. 181 (1985) 165-168.
33. P. M Harrison and T. H Lilley, (1989) Ferritin." in Iron Carriers and Iron
Proteins, ed. Thomas M. Loehr. (New York: VCH Publishers, Inc.) 182-189.
34. J. R. Mertz, E. C. Theil, J. Biol. Chem. 258 (1983) 11719-11726.
35. G. C. Ford, P. M. Harrison, D. W. Rice, J. M. A. Smith, A. Treffry, J. L.
White, and J. Yariv, Phil. Trans. Roy. Soc. Lond. B304 (1984) 551.
36. D. W. Rice, G. C. Ford, J. L. White, J. M. A. Smith, and P. M. Harrison,
Protides of the Biological Fluids 31 (1984)177.
37. S. Granick, Physiol. Rev. 31 (1951) 489.
38. J. L. Farant, Biochim. Biophys. Acta 13 (1954) 679.
39. P. M. Harrison, T. H. Lilley (1989) Ferritin. in Iron Carriers and Iron
Proteins, ed. Thomas M. Loehr. (New York: VCH Publishers, Inc.), 215-221.
40. S. Osaki and S. Sirivech, Fed. Proc. 30 (1971) 1292.
41. R. Ulvik and J. Romslo, Biochim. Biophys. Acta 541 (1978) 251.
42. R. J. Ulvik and I. Romslo, Biochim. Biophys. Acta 588 (1979) 256.
43. R. J. Ulvik, Bioenerg. Biomemb. 15 (1983) 151.
44. B. E. Speyer and J. Fielding, Brit. J. Haematol. 42 (1979) 255.
45. G. D. Watt, D. Jacobs, and R. B. Frankel, Proc. Nat. Acad. Sci. USA 85
(1988)7457.
84


46. J. G. Joshi, S. Goodman, V. V. Deshpande, and D. J. Price (1985)
Proteins of Iron Storage and Transport. Ed. G. Spik, J. Montreuil, R. R.
Crichton, and J. Mazunier (Elsevier Science Publishers) 93-96.
47. D. J. Price and J. G. Joshi, J. Biol. Chem. 258 (1983) 10873.
48. A. Treffry and P. M. Harrison (1982) The Biochemistry and Physiology
of Iron. Ed. P. Saltman and J. Hegenauer (Elsevier North Holland, Inc.) 459-
462.
49. T. M. Loehr, ed. (1989) Iron Carriers and Iron Proteins (New York: VCH
Publishers, Inc.) ix.
50. S. Levi, P. Santambrogio, B. Corsi, A. Cozzi, P. Arosio, Biochem. J. 317
(1996) 467-473.
51. A'. Szucs and M. Novak, J. Electroanal. Chem. 383 (1995) 75-84.
52. A'. Szucs and M. Novak, J. Electroanal. Chem. 384 (1995) 47-55.
53. I. Carelli, I. Chiarotto, A. Curulli, G. Marrosu, Anal. Lett. 28 (1995) 2325-
2345.
54. M. J. Tarlov and E. F. Bowden, J. Am. Chem. Soc. 113 (1991) 1847-
1849.
55. R. J. Cherry, A. J. Bjornsen, and D. C. Zapien, Langmuir 14, 8 (1998)
1971-1973.
56. P. Yeh andT. Kuwana, Chem. Lett. (1977) 1145-1148.
57. D. S. Cohen, F. M. Hawkridge, H. N. Blount, C. R. Hartzell, Charge Field
Eff. Biosyst. [Pap Int. Symp. Bioelectrochem. Bioenerg.] (1984), Meeting
Date 1983,19-31. Ed. J. M. Allen, P. N. R. Usherwood (Abacus Press,
Tunebridge Wells, UK).
58. E. F. Bowden, F. M. Hawkridge, J. F. Chlebowski, E. E. Bancroft, C.
Thorpe, and H. N. Blount, J. Am. Chem. Soc. 104 (1982) 7641-7644.
85


59. G. W. Ewing (1985) Instrumental Methods of Chemical Analysis, 5th ed.
(McGraw-Hill, New York) 291-292.
60. P. H.Rieger (1994) Electrochemistry, 2nd ed. (Chapman and Hall, New
York) 156-190.
61. T. D. Martin, S. A. Monheit, R. J. Niichel, S. C. Peterson, C. H.
Campbell, D. C. Zapien, J. Electroanal. Chem. 420 (1997) 279
62. T. Sagara, H. Murakami, S. Igarashi, H. Sato, and K. Niki, Langmiur, 7
(1991)3190.
63. S. Song, R. A. Clark, E. F. Bowden, M. J. J. Tarlov, Phys. Chem., 97
(1993) 6564.
64. T. Sagara, H. Kawamura, K. Ezoe, and N. Nakashima, J. Electroanal.
Chem., 445 (1998) 171-177.
65. L. S. Wang, V. L. Vilker, W. T. Yap, and V. Reipa, Langmuir., 11 (1995)
4818-4822.
66. W. R. Heineman, J. N. Burnett, and R. W. Murray, Anal. Chem., 40
(1968)1974.
67. K. Ashley, Talanta, 38 (1991) 1211.
68. B. Bedenend-C. Lamy (1988) "Infrared Reflectance Spectroscopy, In
Spectroelectrochemistry: Theory and Practice, ed. R. J. Gale (New York:
Plenum Press), 189-207.
69. D. Moss, E. Naberdryk, J. Breton, and W. Mantele, Eur. J. Biochem.,
187 (1990) 565.
70. D. D. Schlereth and W. Mantele, Biochemistry, 32 (1993) 1118.
71. A. Bewick, K. Kunimatsu, and B. S. Pons, Electrochimica Acta, 25
(1980)465.
72. H. Seki, K. Kunimatsu, and W. G. Golden, Applied Spectroscopy, 39
(1985) 437.
86


73. J. O. Bockris and B. Yang, J, Electroanal. Chem., 252 (1988) 209-210.
74. H. O. Finklea and D. D. Hanshew, J. Am. Chem. Soc., 114 (1992) 3173-
SI 81.
75. C. E. D. Chidsey, Science 251 (1991) 919-922.
76. C. Hinnen, R. Parsons, and K. Niki, J. Electroanal. Chem. Interfacial
Electrochem 147 (1983) 329-337.
77. T. Sagara, S. Nakajima, H. Akutsu, K. Niki, and G. S. Wilson, J.
Electroanalytical Chem. Interfacial Electrochem 297 (1991) 271-282.
78. K. B. Koller, F. M. Hawkridge, J. Am. Chem. Soc. 107 (1985) 7412-7417.
79. K. B. Koller, F. M. Hawkridge, J. Electroanal. Chem. Interfacial
Electrochem. 239 (1988) 291-306.
80. W. Norde, Adv. Coll. Interface Sci. 25 (1986) 267-340.
81. M. M. Bradford, Anal. Biochem. 72 (1976) 248-254.
82. Z. Q. Feng, S. Imabayashi, T. Kalciuchi, and K. Niki, J. Electroanal.
Chem. 408 (1996) 15-20.
83. J. M. Cooper, K. R. Greenough, and C. J. McNeil, J. Electroanal. Chem.
347 (1993) 267-275.
84. W. J. Albery, M. J. Eddowes, H. A. O. Hill, and A. R. Hillman, J. Am.
Chem. Soc. (1981)3904.
85. A. Treffry, J. Hirzmann, S.J. Yewdall, and P.M. Harrison, FEBS Letters
302 (2) (1992) 108-112.
86. D M. Lawson, et al., FEBS Letters 254 (1989) 207-210.
87. A. Treffry, P.M. Harrison, A. Luzzago, and G. Cesareni, FEBS Letters
247 (1989) 268-272.
88. V.J. Wade, et al., J Mol B/o/221 (1991) 1443-1452.
87


89. D.M. Lawson, etal., Nature 349 (1991) 541-544.
90. B. Xu and D. Chasteen, J. Biol. Chem. 266 (1991) 19965-19970.
91. A.J. Bard and L.R. Faulkner, Electrochemical Methods: Fundamentals
and Applications (1980) 1-26 (Johns Wiley and Sons).
92. D.A. Skoog and J.J. Leary, Principles of Instrumental Analysis (1992)
536-564 (Saunders College Publishing: New York).
93. N.D. Chasteen, I.M. Ritchie, and J. Webb, Anal. Bioch. 195 (1991) 296-
302.
94. T.D. Martin, S.A. Monheit, R.J. Niichel, S.C. Peterson, C.H. Campbell,
and D.C. Zapien, J. Electroanal. Chem. 420 (1997) 279-290.
95. F. Ammar and J. Saveant, J. Electroanal. Chem. 47(1973) 155.
96. T. Saji, N.F. Pasch, S.E. Webber, and A.J. Bard, J. Phys. Chem. 82
(1978)1101-1105.
97. T.W. Smith, J.E. Kuder, and D. Wychick, J. Polym. Sci. 14 (1976) 2433.
98. J.B. Flanagan, M. Schlomo, A.J. Bard, and F. Anson, J. Am. Chem.
Soc., 100 (1978) 4248-4253.
88


Full Text

PAGE 1

THE STUDY OF FERRITIN ON BARE GOLD ELECTRODES BY CYCLIC VOL TAMMETRY AND INFRARED SPECTROELECTROCHEMISTRY by Michael Alfred Johnson B.S., United States Air Force Academy, 1988 A thesis submitted to the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Master of Science Chemistry 2000

PAGE 2

This thesis for the Master of Science degree by Michael Alfred Johnson has been approved by q-zc;--oo Date

PAGE 3

Johnson, Michael Alfred (M.S., Chemistry) The Study of Ferritin on Bare Gold Electrodes by Cyclic Voltammetry and Infrared Spectroelectrochemistry Thesis directed by Assistant Professor Donald C. Zapien ABSTRACT This research focused on the study of ferritin on a bare gold electrode using cyclic voltammetry and infrared reflection-absorption spectroscopy (IRRAS). The cyclic voltammetry of ferritin on bare gold suggested that a reversible change in orientation of the ferritin molecule might occur. Additionally, our data suggested that ferritin tends to oligomerize on bare gold, which is in contrast to its behavior on gold modified by mercaptopropionate groups. From this data we postulated that ferritin is anchored to the electrode by the promoter layer. The relationship between ferritin packing density and scan rate was found to be linear, suggesting that the signals on our voltammetric traces were from ferritin adsorbed onto the gold electrode and ferritin in solution. Potential dependence data suggested that ferritin is adsorbed in an electroactive state only when it is first reduced. Additionally, the data suggest that a change in quaternary structure occurs during the reduction process that allows the protein to be adsorbed onto the gold surface while maintaining electroactivity. Studies of packing density at various ionic strengths, ferritin concentrations, and adsorption times were conducted. The ionic strength data suggested that hydrophobic interactions predominate at the ferritin-gold interface. The packing density of ferritin increased with dissolved ferritin concentration. The adsorption data suggested that the adsorption rate was limited more by the diffusion of ferritin onto the gold surface than by structural rearrangements or changes in the orientation following initial attachment to the electrode surface. An infrared reflection-absorption apparatus was designed and constructed. This apparatus included an airtight enclosure and reflection optics. Testing of the apparatus revealed that the cell had the ability to exhaustively electrolyze an analyte in fairly short time and that the infrared spectra can Ill

PAGE 4

be recorded in situ. More development will be required before the cell can obtained oxidized minus reduced infrared difference spectra. The infrared spectrum of ferritin was recorded and analyzed. Peak assignments for amide I and amide II were made. The infrared spectrum of ferritin confirmed the presence of a-helical segments as well as some disordered secondary structure. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. Signed iv

PAGE 5

DEDICATION I dedicate this work to my wife Monica for her loving support, encouragement, and understanding throughout my graduate career. v

PAGE 6

ACKNOWLEDGEMENTS Professor Donald C. Zapien, for his never-ending guidance, support, patience, and enthusiasm both in and out of the lab. Professor Larry G. Anderson, for his kind patience and advice in obtaining this degree. Mr. Fedri Marrugo, for machining the stainless steel flange and constructing the external optics bench, and for assistance with the cell design. Research Corporation (Cotrell Research Grant Number C-3735), for funding this work. VI

PAGE 7

CONTENTS Figures ................. ................................................ ........................ ........... x Chapter 1. Introduction ..... ...................... ........ .............. ....... .................. ........... 1 1 1 Ferritin ................................ ........ ... .... ......... . .... .... ............. .. .. ..... 1 1.1.1 Structure ........... .... ...... .... ....... .... ..... ............. .......... ..... ....... .... 2 1.1.2 Function . ... ........................................... .... ................................. 10 1.2 Electron Transfer .... ... ... ................ .............. ....... ...... .......... ........ .... 16 1.2 1 Fundamentals of Electron Transfer .......... ........ ...... ... ..... ......... 16 1.2.2 The Three-Electrode Cell ................... . .... .................... .... ......... 17 1.2.3 Voltammetry ....... .......... .... ... ... .... .................................... ....... 19 1.2.4 Mediated Electron Transfer ...... .... .... ... .... ..... . .... ...... ... ..... . 25 1.2 5 Direct Electron Transfer.. . . ... .. ............. . ... ................... .... ..... ... 25 1 3 Solution Infrared Spectroelectrochemistry . . ...................... ..... 29 1.4 Research Objectives ........ .............. .... ..... ........ ........ ... .... ....... 34 2. Experimental 2.1 Electrochemistry of Ferritin on Bare Gold . .... ................. ........... 37 2.1.1 Materials ... ............. . ... ...... ....................... . ... ...... ......... ........ .... 37 VII

PAGE 8

2.1.2 Methods ..... .................... ......................... ................................... 38 2 2 Infrared Spectroelectrochemistry ........... ............................ .... ...... 45 2.2.1 Materials ..... ...................... ........................ ......................... .......... 45 2.2.2 Methods .......... ..... .................................... ............. ......... ... .......... 46 3. Direct Electron Transfer of Ferritin Adsorbed on Bare Gold Electrodes .................... ................................................ .......... 48 3.1 Adsorption and Redox Activity of Ferritin .... .......... .... ......... ........... .48 3.2 Potential Dependence of Ferritin Adsorption ... ................... ... ........ 55 3.3 Dependence of Packing Density on Ionic Strength, Ferritin Concentration, and Adsorption Time ....... .... ............ ..... .............. 57 3.3.11onic Strength .. .......... ................. ........................ .......... ....... .......... 57 3.3.2 Ferritin-Concentration .......... .. ......................... ........................ ........ 59 3.3 3 Adsorption Time ......... ............ ......... .......... ........................ ........... 59 4. Design of the Infrared Reflection Absorption Spectroscopy (IRRAS) Apparatus ........ ........ . ........................... ............ ...... ................ 63 4.1 Spectroelectrochemical Cell ....... ....................... ............... .. ..... ..... 63 4. 1.1 Description ............ ....... ........... .......... ......... ........ .......... . ............. 63 4 1.2 Design ........ ............... ........... .... ............... .... ..................... ...... ........ 63 4.2 Enclosure ...... ............ ......... .. ...... ....... ....... .... ........ ..... ........ .......... 68 4.2 1 Description ............. ......... ............ .................. .............. ....... ...... ..... 68 4 2.2 Design .................... ............. .... ............... ....................................... 70 4.3 Optics .... ........ .............. ............ .................. ............ ............. ...... .... 71 vi ii

PAGE 9

4.3.1 Internal Optics ................................................................................ 71 4.3.2 External Optics .................. ............................................................ 73 4.3.3 Detector .... .... ...................... ........................ .................................. 73 4.4 Testing and Evaluation of the Apparatus ........................................ 74 4.4.1 Reduction of Ferricyanide .............................................................. 7 4 4.4.2 Spectrum of Ferritin ........................................................................ 76 5. Summary and Conclusion ................................................................... 79 5.1 Summary ......................... ................................................................. 79 5.2 Conclusion ........................................................................................ 80 References ............................................................................................... 81 ix

PAGE 10

FIGURES 1 1 Schematic Drawing of Ferritin Molecule ...... ............ ..... ...... . ... . ...... 3 1.2 Schematic Depiction of Ferritin Shell and Iron Core .............. ............ .4 1.3 Primary Sructure of Ferritin ... ................ ...... ................. ........ .......... 6 1.4 Schematic of Ferritin Subunit.. ............... .... ....... .... .... ......... 8 1. 5 The Three-Electrode Cell ... ... ................. . ... . . .................. ....... ......... 18 1.6 Cyclic Voltammetry ............................. ............. .... . ................ .... .... 21 1. 7 Cyclic Voltammogram of Ferrocyanide ... ...... .... ......... ....... ..... . ....... 22 1.8 Mediated Electron Transfer ... ...... ................. ....... . ........... . . .... ....... 26 1.9 Direct Electron Transfer of Cytochrome c .... ... ...... ....... ... ........ ........ 28 1.10 Thin-layer Infrared Transmittance Cell .............. ....... ...... .... ... ....... 32 1.11 Thin-layer Infrared Reflectance Cell ........... ........................... ..... .... 36 2.1 Electrochemical Cell .................................... .... .................... .... ........ 39 2.2 Teflon Electrochemical Cell ........ . . . . ... ... ... ..... .......... ......... ... ...... . . 41 2.3 Cyclic Voltammogram of Clean Bare Gold ... ..... ........ .... ...... .... ......... 42 2.4 Cyclic Voltammogram of Adsorbed Iodide ... ..... ................. ............ 43 3.1 Cyclic Voltammogram of Ferritin . ... .................................. ... .......... 49 3 2 Consecutive Cycling of Ferritin ..... .......... ....... .... ............ ....... ..... ...... 51 X

PAGE 11

3.3 Effect of Scan Rate on Peak Current ........... .... ................................. 54 3.4 Dependence of Ferritin Adsorption on Potential. ... ... ... ..................... 56 3.5 Effect of Ionic Strength on Packing Density ................................. .... 58 3.6 Effect of Ferritin Concentration on Packing Density .......................... 60 3. 7 Effect of Adsorption Tirr1e on Pagking Density ............................ ..... 62 4.1 Diagram of the Infrared Reflection Absorption Spectroscopy (IRRAS) Cell. .......... ............................................. ........................ .......... 64 4.2 Beam Path Through the Dove Prism ............. ................. ... .... ....... ... 67 4.3 Plexiglas EnCJosure .. : ............ ................................................ .......... 69 4.4 Optics Setup .... ............................................. ....................... ... ......... 72 4 5 Infrared Spectrum of Ferri cyanide and Ferrocyanide ....... .......... ...... 75 4.6 Infrared Spectrum of Ferritin ............... .... ...... .......... ........ . ... .......... 77 xi

PAGE 12

1. Introduction 1.1 Ferritin Iron is essential for almost all forms of life This need for iron is a result of the varied roles it plays in biological systems, especially as a catalyst in numerous intracellular and extracellular reactions (1 ). One of the evolutionary challenges of developing life forms has been to maintain a usable supply of iron. Approximately 2.5 billion years ago, when photosynthetic organisms began using H20 as its source of hydrogen, oxygen became the major by-product (2}. This prevalence of oxygen resulted in the oxidation of Fe(ll) to Fe(lll) (3); however, because Fe(lll) at neutral pH forms insoluble Fe(OHh, profound restrictions were placed on its availability for use in biological systems (4). This insolubility in oxygen rich environments necessitated the evolution of iron storage methods in which oxygen-dependent organisms could store iron in a soluble form while accommodating and using oxygen. Ferritin, one outcome of this process, maintains an ample supply of iron in a soluble form for use in various biological functions, including oxygen transfer, electron transfer, nitrogen fixation, and DNA synthesis (5). 1

PAGE 13

1.1.1 Structure 1.1.1.1 Introduction Cellular ferritin is a protein that modulates the availability of iron in nearly all forms of living organisms, including most bacteria (1 ). The ferritin molecule has a molecular weight of 450,000 Da, a diameter of 12.0 nm, and a spherical protein coat, known as apoferritin, which encases an iron-containing core (6) (Figure 1.1 ). The apoferritin shell consists of two types of subunits, denoted H (heavy) and L (light). A total of 24 subunits make up this protein shell, which is in contact with the iron core at various points, forming an interface between the protein and the iron core (7) (Figure 1.2). Horse spleen apoferritin is often used for studies because most of its subunits are identical (approximately 85% are L subunits) and the location of essentially all amino acids has been elucidated by Harrison and coworkers using X-ray crystallography (8,9). One ferritin subunit consists of approximately 175 amino acids (1, 1 0). 2

PAGE 14

Hydrophobic channel Figure 1.1. Schematic drawing of a ferritin molecule. The protein contains a symmetrical arrangement of 24 equivalent subunits. Adapted from (1 ). 3

PAGE 15

Channel Figure 1.2. Schematic depiction of the ferritin shell and iron core. The 24 subunits contact the iron core at various points to form an interface between the shell and the core. Adapted from (1 ). 4

PAGE 16

The protein shell encases an iron core that is limited by the size of the inside cavity (approximately 8 nm). A complete iron core consists of 4000 to 4500 Fe( Ill) atoms; however, most molecules in a ferritin preparation contain little or no iron-core, and a large number of molecules contain less than the full compliment of iron. The core forms as a complex of small microcrystalline particles having the formula (FeOOH)8(FeOP03H2). Because the number and consistency of the particles is variable, one or several particles may be present in addition to some disordered material (7,11,12). 1.1.1.2 Protein Coat The primary structure of a horse spleen ferritin subunit as well as the primary structures of ferritin subunits from other species is given in Figure 1.3 (13-22). An interesting feature of these sequences is that the L sequences between species are more similar (85 to 90% identity) than H and L sequences for the same species (human, 55% identity). The same trend holds true for H subunits (>90% identity in three complete sequences) (23). 5

PAGE 17

5 10 15 20 25 30 35 40 45 50 55 60 HoS-L SSOIRONYSTEVEAA VNRLVNL YLRASYTYLSLGFYFDRDDVALEGVCHFFRELAEEKRE HuS L D Y Y BY S-----Hul-L D s o s.------HuL-H T -V--HODS-1-01-E-Y-V-MSY KNFAKY-LHOSH-E-Ral-L T H f G------Ral-H P-V--HQDS-1-01-E-Y-V-MSC KNFAKY-LHOSH-E-65 70 75 80 85 90 95 100 105 110 115 120 HoS-L GAERLLKMONORGGRALFODLQKPSODEWGTTLDAMKAAIVLEKSLNQALLDLHALGSAQ HuS-L -Y IK-AE--K-PMA-K R Hul-L -Y IK-AE--K-PMA-K R Hul-H H-K-M-L IFL-IK-DC-D-ESG-N-EC-LH-NV-8-E-K-ATDK Ral-L ---M-E V K-E-E-LA-N R Ral-H H -K-M-L IFL--IK-DR-D-ESG-N-RC-LH--V-8-E-K-ATDK 125 130 135 140 145 150 155 160 165 170 175 HoS-L ADPHLCDFLESHFLDEEVKLIKKMGDHL TNIORLVGSOAGLGEYLFERL TLKHD HuS-L T T LRK-G-PE-------Hul-L T T LH-G--PE--------Hul-H N I T-YN -0-A-EL--V-LRKMGAPES-A--DKH-GDSDNES Ral-L T K N--LR-WO-P-8-------Ral-H N1T -Y-N-0-S-EL--V-LRKMGAPES-MA--DKH-G-GDES Sequences are from the following sources. Dashes indicate the same amino acid as in HoS-L. HoS-L HuS-L HulL Hul-H Ral-L shown) Ral-H (protein) (protein) (eDNA, DNA) (eDNA, DNA) (DNA) (DNA) Horse spleen L subunit Human spleen L subunit Human liver L subunit Human liver H subunit (N-term inal T T A S not shown) Rat liver L subunit (insertion 0 P A 0 T G V A between residues 157 and 158 not Rat liver H subunit (Nterminal T T A S not shown) Figure 1.3 The primary structure of ferritin from various sources Adapted from (1 ) 6

PAGE 18

Variations in subunit primary structure may affect other properties of ferritin in a variety of ways. For example, amino acid sequence, chain length (172 to 183 residues), and mass may cause deviations in surface charge (24-26), hydrophobicity (27,28), and size (29,30).Despite these differences, X-ray diffraction studies have shown that the higher order structure between subunits of different molecules is not greatly affected (31 ,32) Four a-helices connected to another small helix through a series of loops make up the secondary structure of a ferritin subunit. Of the 175 residues, 129 are incorporated into these five a.-helices, denoted A-E (Figure 1.4). The four longer a-helices are twisted into a bundle while the shorter ahelix is located at the end of this bundle. This small helix seems to play an important role in how the subunits pack to form the quaternary structure of the protein. A long loop lies on the outside of the twisted bundle and The ferritin subunits interact via trimeric and tetrameric interactions to form hydrophilic channels (with conserved carboxylate side chains between residues 120-127) and hydrophobic channels (with conserved nonpolar side chains between residues 160-165) (2,5) Along the 3-fold axis of the molecule are located eight 7

PAGE 19

N-terminus Figure 1.4. Schematic diagram of a ferritin subunit. Helices A, 8, C, and 0 and the short helix E are depicted as rods. Adapted from (1 ). 8

PAGE 20

hydrophilic channels which are 0.3 to 0.4 nm wide. Three glutamate residues are located on the inside of the channel while three more glutamate residues line the outside The outside of the channel flares out and is lined with.primarily hydrophilic residues-six serines, six histidines, and three cysteines--allowing water molecules are bound to this funnel. It has been suggested that, due to this structure, iron may flow in and out of these ports and that the funnel shape may provide chelaters and reductants access to metal on the outer site (33) A ferritin molecule also contains six hydrophobic channels located along the 4-fold axis. These channels are 0 3 to 0.4 nm wide and 1.2 nm long and are lined with 12 leucine residues, three from each of the four subunits. The function of these channels is not fully understood, but recent work has suggested that they facilitate the rapid exchange of water due to the lack of hydrogen bonding sites (33). 1.1.1.31ron Core The inorganic, micellular nature of the iron core was first established by Granick, Michaelis, and Hahn, who discovered that the core consists of a phosphate-containing ferric oxyhydroxide complex. Interestingly, they found that treating this complex with 1 N NaOH (37) could induce release of the iron core from the protein shell; however, they 9

PAGE 21

incorrectly believed the iron micelles to cover the exterior of the protein molecule In 1954, Farrant, using electron microscopy, demonstrated that the iron core filled the inside of an apoferritin protein shell, showing the core to be an electron-dense mass of 5.5 to 6.0 nm in diameter (38). Despite extensive study, information on the core is still quite limited, and the exact atomic structure; while a matter of conjecture, likely has an irregular arrangement and composition (33). What is certain, however, that the iron core plays a role central to the function of ferritin. 1.1.2 Function Ferritin functions as an iron storage protein and therefore requires an efficient mechanism of iron uptake and release. One purpose of this storage function is to protect the cell from oxidative damage from reactive oxygen species. For example, Fe(ll) can catalyze the decomposition of hydrogen peroxide to form highly reactive hydroxyl radical: This reaction, known as the Fenton reaction, is prevented in part by the uptake of Fe( II) by ferritin. Another function of ferritin is the binding of other potentially harmful metal ions. While investigating how 10

PAGE 22

phosphoglucomutase is inhibited by Be2+, Joshi and coworkers found that ferritin can bind up to 1 000 atoms of Be2+ in vivo ( 46). Additionally, ferritin has been shown to bind Cu2+, Zn2+, Cd2+, Tb3+, UO/+, Cr3+, and V02+ (47, 48). 1.1.2.1 Iron Uptake The loading of iron into the molecule is pivotal to the function of ferritin as an iron storage protein. Research has suggested that entry into apoferritin occurs via the 3-fold channels (39). The channels broaden toward the shell exterior and contain numerous hydrophilic regions, which may aid in funneling Fe( II) into the inner, narrow part of the channel. These inner channels contain the highly conserved Glu 130 and Asp 127 residues that, due to the flexibility of the side chains, may effectively bind Fe( II) or Fe(lll). This idea was strengthened by studies that uncovered evidence that at least two residues exposed on the three-fold channels play active roles in the mechanism of iron incorporation into apoferritin. These residues are metal binding sites that actively bind and transfer iron into the ferritin cavity (50). One view is that iron ions approaching the inner channel displace other iron ions inward from Glu 130 to Asp 127 (39) 11

PAGE 23

The entry of iron into -the ferritin molecule first requires its oxidation from Fe( II) to Fe( II I). Because iron is known to enter through the 3-fold channels, the possibility of an oxidative mechanism associated.with these channels has been extensively studied (85-87). Researchers have found from site-directed mutagenesis studies that a "ferroxidase" center, an area responsible for the oxidation of Fe( II) to Fe( II I), is present in the H-chain subunits along the 3-fold channels (87,88). One proposed mechanism of iron oxidation through this ferroxidase site suggests that nearby glutamate residues in the ferroxidase center first chelate two Fe2+ atoms. Because they are so close to one another, the two atoms simultaneously transfer two electrons to molecular oxygen, generating Fe3 + -O-Fe3 + and hydrogen peroxide. This complex then dissociates and releases iron into the protein cavity. From this point, the iron atoms go on to form the ferrihydrite core. As the core becomes more developed, the core surface facilitates Fe( II) oxidation(90) This oxidation of Fe( II) makes electrochemical methods ideal for the study of ferritin. 1.1.2.2 Iron Release Iron release is also a crucial function of ferritin because iron is an essential nutrient for life in almost every living organism. Examples of iron proteins and enzymes that require iron include the iron-tetrapyrrole 12

PAGE 24

complexes such as those proteins that contain heme prosthetic groups, the iron-sulfur proteins, and the iron proteins that contain an iron coordinated to nitrogen and oxygen ligands {49). One way in which ferritin releases iron is through degradation of the protein(39). It is thought that ferritin may lose as much as one third of its iron stores daily through this process, while it is still not clear how much is lost from intact protein (39). The reduction of iron from Fe( Ill) to Fe(ll) is a pivotal step in the normal release of iron from the ferritin core. Laboratory approaches that have been used to remove iron include reducing ferritin to induce the release of iron as Fe2+, with or without chelators and using chelators that have a high affinity for Fe3+ to remove Fe3+ from the ferritin molecule. Both methods depend on the physical state of the cores For example, the core size, the degree of aggregation, and the proximity of protein or membrane may greatly affect how readily iron is removed from ferritin (39). In the cell three modes of release by this mechanism have been proposed (39) The first model suggests that a reductant diffuses through the protein shell and reduces Fe3+ to Fe2+ which leaves the protein via the three-fold channels. The second model proposes that Fe3+ is reduced in the three-fold channels and the reduced iron diffuses out of the ferritin 13

PAGE 25

shell through these channels. The third model involves the reductant acting on the iron core by a long-range electron transfer through the ferritin shell. The Fe3+ ions, reduced to Fe2+ as a result of this electron tunneling, would exit the ferritin molecule via the three-fold channels (39). Additionally, chelators have been used in conjunction with reductants, and could effect the removal of iron from halo-ferritin in a number of ways (39). Although the in vivo release of iron from ferritin is still under study, it has been shown that an enzyme in vertebrate livers catalyzes the reduction of ferritin iron from Fe3+ to Fe2+ when reduced nicotine adenine dinucleotide (NADH) and flavin mononucleotide (FMN) are present and oxygen concentrations are below 3 (40). It was subsequently found that flavins, reduced by exogenous mitochondria, mediate the release of ferritin iron (41 ), as shown below. NADH FMN Free Fe(ll) Enzyme Ferritin Fe(lll) 1.4

PAGE 26

Therefore, it was theorized that ferritin participates in heme synthesis by donating iron directly and also that ferritin aids in transporting iron from the plasma membrane by functioning as an intermediate (42-44). Another proposed mechanism of ferritin iron release involves the interaction of flavo-proteins with the outer ferritin coat. This mechanism calls for the tunneling of electrons through the apoferritin shell to reduce the iron core and the release of iron, which is then scavenged by a small chelator (45) 1.1.2.3 Summary As mentioned above, redox reactions are key to the uptake and release of iron. After Fe( II) enters the protein shell, it is oxidized to Fe( II I) as it is incorporated into the iron core. The release of iron from the protein shell first requires the reduction of Fe(lll) to Fe( II). Electrochemical methods are well-suited to the study of these electron transfer reactions, and may yield clues to the electrochemical behavior of ferritin. 1.2 Electron Transfer 1.2.1 Fundamentals of electron transfer In electrochemical methods, one electrode serves as a reference electrode, while the another electrode serves as a working electrode. The makeup of the reference electrode is fixed, making the potential on the 15

PAGE 27

electrode fixed. If the potential on the working electrode is adjusted to a low enough (negative) potential, then the energy for transfer of electrons from the working electrode to the analyte in solution will be favored. As a result, the species in solution will be reduced. If the potential of the working electrode is adjusted to a high enough (more positive) potential, then eventually the energy of the electrons on the analyte molecules will be at a higher energy than those on the electrode. As a result, electrons will tend to migrate from the analyte in solution to the electrode, resulting in an oxidation (91 ). 1.2.2 The Three-Electrode Cell One limitation of the two-electrode system described above is that electrolysis would have to occur at the reference electrode as well as the working electrode, resulting in a change of the composition of the reference electrode, and thus its potential. A configuration better suited for these types of measurements is the three-electrode cell (Figure 1.5) This cell is composed of three electrodes: a working electrode, a counter electrode, and a reference electrode. The potential between the working and counter electrode is increased until it becomes great enough to induce electron transfer with the species at the working electrode surface. The reaction results in the flow of electrons, or a current, which is measured 16

PAGE 28

with a suitable current-measuring device. The potential of the working electrode is measured with respect to the reference electrode potential. In this configuration, the amount of current flowing through the reference electrode is negligible, thus its composition and therefore its potential remain constant (92) 1.2.3 Voltammetry Voltammetry refers to those electroanalytical techniques in which current is measured as a function of applied potential. By analysis of the current-potential plot, information regarding the desired analyte may be derived. Voltammetric methods rely on complete polarization of the working electrode, and for this reason, the working electrode is kept as small as practical (92). 1.2.3.1 Linear Scan Voltammetry One of the most widely used electrochemical methods that makes use of the three-electrode system is linear sweep voltammetry. In a typical linear sweep voltammetry experiment, the potential of the working electrode is increased or decreased at a rate of 2 to 5 mV/s. The resulting current between the counter and working electrodes is measured and plotted versus applied potential (92). 17

PAGE 29

Power Supply working electrode counter electrode reference electrode Figure 1.5. The three-electrode cell. Adapted from (91 ). 18

PAGE 30

1.2.3.2 Cyclic Voltammetry 1.2.3.2.1 Introduction Cyclic voltammetry, a technique well-suited for the study of electron transfer kinetics, involves the application of a triangular wave potential in which the potential of the indicator electrode is swept linearly through a specified range and then swept the opposite direction to the original potential (Figure 1 6). Shown in Figure 1. 7 is the cyclic voltammogram of ferrocyanide in 1.0 M NaCI. The potential applied to the working electrode is swept positive to+ 0. 7 V, reversed, and then scanned to -0.1 V As the potential is increased, an anodic current flows as electrons are transferred from ferrocyanide to the gold electrode, then reaches a maximum at 0.04 V The half reaction for this process is given as: Fe(CNt Fe(CN)3 -+ e-. Though the potential continues to increase, the current decreases because the surface concentration of available ferrocyanide becomes smaller. The current gradually levels off and the direction of the potential scan is reversed at +0 7 V. As the potential approaches approximately +0.04 V, the cathodic current increases as ferricyanide in solution is reduced. This reduction is represented by the half reaction: Fe(CN)3+ e Fe(CNt (92). 19

PAGE 31

c ..... ::J 0 -c Q) ..... ..... ::J 0 time potential Figure 1.6. Cyclic voltammetry shown as a function of various parameters: (a) potential vs time; (b) current vs time; (c) current vs potential. Adapted from (60). 20

PAGE 32

I -O.lV I +0 7V Figure 1.7. Cyclic voltammogram offerrocyanide The molecule is oxidized to ferricyanide on the anodic sweep (top peak) and reduced back to ferrocyanide on the cathodic sweep (bottom peak). 21

PAGE 33

When obtaining measurements of species in solution, the current is limited by how fast a species approaches the electrode surface, transfers electrons with the electrode, then finally leaves the electrode surface. The current due to the electrolysis of an adsorbed molecule is, of course, not limited by the diffusion of the species to the electrode surface. However, it is often useful to study the electrochemistry of adsorbed species, as the electrochemical data may yield important information about the structure as well as electron transfer processes of the adsorbate. When electroactive species adsorb to an electrode, the amount of material adsorbed, not diffusion, limit current. In the case of reversible electron-transfer processes, the surface concentrations are then related by the Nernst equation: r o(t)/ r R(t) = exp{nF[E(t)-Es0]/RT] = 8(t) r o(t) and r R(t) are the surface concentrations of oxidized and reduced species, respectively. E5 is the formal potential of the adsorbed couple and E(t) describes the potential as it varies with time. If the total surface concentration (r) of both oxidized and reduced species remains constant 22

PAGE 34

the individual surface concentrations are dependent upon time and potential and therefore an expression for the current can be derived: where 8 = E(t)/E5 According to this equation, the maximum current will occur when E = Es0 (8 = 1 ). This peak current is also related to the number of moles of adsorbed species as well as the scan rate (v) (60). 1.2.4 Mediated Electron Transfer Much of the early electrochemical studies on proteins involved the use of electrochemical mediators. Mediators work by acting as an electron transporter between the working electrode and the protein (Figure 1 .8). At reducing potentials, the mediator would be reduced directly at the electrode and in turn would reduce the protein in solution. The mediator then returns to the electrode to become reduced again, and the cycle is repeated. Mediated electrochemical studies on ferritin have yielded valuable information, such as number of electrons transferred per molecule, redox potentials, and rate of reduction of the iron core (93). Mediated 23

PAGE 35

Oxidation Reduced Mediator Oxidized Mediator Reduction Electrode Surface Figure 1.8. Mediated electron transfer. The mediator is reduced at the electrode surface, then reduces the protein electroactive site. The oxidized mediator is then reduced again at the electrode surface to complete the cycle. 24

PAGE 36

electrochemistry has been shown to be useful, other capabilities such as the probing of electron transfer kinetics and investigating possible chemical reactions coupled with electron transfer reactions are possible only if the protein exchanges electrons directly with the electrode. 1.2.5 Direct Electron Transfer The direct electron transfer of proteins has often encountered difficulties not encountered with small molecules Proteins have the propensity to irreversibly adsorb to the bare electrode surface forming an insulating layer that would often hinder the electron transfer of the protein in solution. In addition, the electroactive center( s) of proteins is sometimes deeply embedded in the globular matrix. This results in slow electron transfer kinetics, sometimes to the point where no currents are observed. To address these issues, electrodes with surface moieties or electrodes modified by the application of adsorbed monolayers have been effectively used to catalyze the electron transfer of proteins. Cytochrome c is a protein that has been extensively characterized by electrochemical methods. When one observes the cyclic voltammogram of cytochrome c in solution using bare gold as the working electrode, the current response is relatively small and requires higher concentrations of cytochrome c to 25

PAGE 37

observe a well-defined current potential curve (53,62). One of the most common modifiers used in the electrochemical study of cytochrome c is 4,4'-dithiodipyridyl (DTDP). The molecules of DTDP dissociate, forming a layer of 4-thiopyridine. The voltammogram of cytochrome c at the Au-4-thiopyridine modified surface is shown in Figure 1. 9. Several factors may account for the facilitated electron transfer. First, the promoter (DTDP) layer may prevent protein denaturation by anchoring the protein to the electrode surface. In its native state, cytochrome c may exhibit a stronger current response than in its denatured state. Second, the promoter layer may allow the protein to adsorb so that its heme prosthetic group is closer to the electrode. Third, the attraction between the lysine groups of cytochrome c and the nitrogen atom of pyridine may aid in aligning the protein in a more favorable position for direct electron transfer (62). Interestingly, ferritin did not give appreciable currents at gold electrodes modified by 4,4'-dithiodipyridyl, octadecyl mercaptan, or cysteamine. Ferritin has exhibited well-defined voltammograms at gold electrodes modified by 3-mercaptopropionic acid and 2-mercaptosulfonic 2 6

PAGE 38

-< :::!.. ........ I 2.1311A -0.1 0 0.1 0.2 POTENTIAL (Volts vs Ag/AgCl) Figure 1.9. Direct electron transfer of cytochrome c on gold enhanced by modification of the gold electrode by 4' ,4-dithiodipyridyl (DTPT) 27

PAGE 39

acid. However, the observed currents were much smaller than would be predicted by theory for ferritin in solution. Further study later indicated that the current was produced-from ferritin adsorbed on the electrode surface (94) 1.3 Solution Infrared Spectroelectrochemistry While voltammetric techniques are useful in the examination of the electron transfer of species, both adsorbed and unadsorbed, it has only limited use in determining the structural properties and reaction mechanisms of electroactive species on a microscopic scale (68). In order to study these types of properties, spectroscopic methods of analysis are needed. A number of methods have been used to study in detail the interactions within the electrode/electrolyte interface. These methods have included techniques such as electron spin microscopy (ESR), Mossbauer spectroscopy, and various optical spectroscopies Of these techniques, the optical spectroscopies, due to their convenience, have been the most widely used (68). Several factors contribute to the effectiveness of optical spectroscopies. First, these techniques have the capacity to be highly selective because molecules absorb only at frequencies that match their difference in energy levels. Second, some methods are extremely sensitive, allowing the observation of minute changes in surface coverage 28

PAGE 40

(as little as a few percent of a monolayer). Finally,spectroscopic techniques typically have a very flat response time, allowing kinetic studies (68). One spectroscopic technique well-suited for the study of electrochemical processes both in solution and on a surface is infrared spectroelectrochemistry. Two general approaches to this technique utilize either a transmission cell or a reflectance cell. A transmission cell, also known as an optically transparent thin-layer electrochemical (OTTLE) cell, is based on a design in which an optically transparent electrode is sandwiched between two IR-transparent windows (Figrure 1.1 0) (66). The beam passes directly through the cell, exposing the sample to infrared radiation. A principal advantage of this method is simplicity-often times, the cell may be placed directly into the sample chamber of the instrument, eliminating the requirement for additional external optics (67). Transmission cells have great utility in observing the spectroelectrochemical parameters of an analyte in solution. One example illustrating the usefulness of this technique is the study conducted by Mantele and co-workers, who adapted the technique to detect redox-linked conformational changes in cytochrome c (69), then later expanded the technique to examine these redox-linked conformational changes at 29

PAGE 41

various temperatures, pHs, and electrode surfaces (70). This technique involved examination of the infrared difference spectrum (reduced minus oxidized states). From the spectral data they were able to determine that, at neutral pH, the redox transition involves modifications in the and segments of cytochrome c. A reflectance cell utilizes the surface of the electrode to reflect the infrared beam radiation through a solution layer. Also known asinfrared reflection-absorption spectroscopy (IRRAS), this technique has the disadvantage of requiring additional optics to direct the beam from the instrument to the cell and then to direct the beam from the cell to the detector Radiant power is dissipated over the longer beam path, resulting in a decrease in the signal. Other disadvantages include the difficulty of aligning the optics and the added expense of the optics (67). The primary advantage of a reflective cell over a transmission apparatus is the ability to perform spectroelectrochemical studies on adsorbed layers. External infrared reflectance spectroscopy is a technique in which the beam passes through an infrared transparent medium and a sample solution, reflects off of a reflective electrode, and exits through the solution and infrared-transparent medium (68). This in situ technique was first developed and used by Bewick, Kunimatsu, and Pons, who in 1980, 30

PAGE 42

port for Pt electrode Figure 1.1 0. Thin-layer infrared transmittance cell. Adapted from (67). 31

PAGE 43

examined the vibrational spectrum of adsorbed indole on a platinum electrode (71 ). Seki and coworkers designed a cell in which the thickness of the solution layer above the electrode surface could be controlled. This greatly reduced the interference from the electrolyte (Figure 1.11) (72). Bockris and Yang designed an IRRAS system in which a stream of fresh electrolyte continuously flowed over the working electrode, reducing the optical interference from evolved gaseous species (73). While a considerable body of work has been accomplished on studying the infrared spectroelectrochemical properties of small molecules, similar analyses for proteins has been limited to the use of transmission cells 1.4 Research Objectives The central theme of this work was the characterization of ferritin through cyclic voltammetry and infrared spectroscopy. The objectives of this research were two-fold. The first objective was to investigate the direct electron transfer of ferritin at a bare gold electrode. Because the oxidation and reduction of the iron core of ferritin is crucial to its function in iron uptake and release, cyclic voltammetry is ideally suited to this study. To characterize the direct electron transfer, ferritin was adsorbed onto a bare gold electrode and the potential scanned using a range of ionic strengths, ferritin concentrations, scan rates, pH values, and adsorption 32

PAGE 44

times. The second research objective was to design, build, and test an infrared reflection absorption electrochemical cell capable of analyzing the electrochemical and spectroscopic properties of ferritin. To accomplish this, an infrared spectroelectrochemical cell and optical bench enclosure apparatus was constructed, and the optics aligned. 33

PAGE 45

CaF2 Window IR Beam Pt Counter Electrode Ag/AgCI Reference Electrode Working Electrode Surface Solvent Well Micrometer Lead to Working Electrode Figure 1.11. A thin-layer electrochemical cell with a CaF2 window and an adjustable working electrode Adapted from (72) 34

PAGE 46

2. Experimental 2.1. Electrochemistry of Ferritin on Bare Gold 2.1.1 Materials Polycrystalline gold foil (99 99%) was purchased Alta-Johnson Matthey, Danvers, MA. The liquid chromatography column was purchased from Kontes Glass Co. (Vineland, NJ), G-200 Sephadex from Pharmacia (Alameda, CA), and Bio-Rad Protein Assay Dye from Bio-Rad Laboratories (Hercules, CA). PMSF (phenylmethyl-sulfonyl fluoride) (>99%) was supplied by Aldrich Chemical Company (Milwaukee, WI). Sodium hydroxide (Analytical Reagent), sulfuric acid (98%), and sodium phosphate monobasic (Analytical Reagent) were purchased from Mallinkrodt Specialty Chemicals Co. (Pari, KY). Bovine serum albumin (fraction 5 powder) and horse spleen ferritin (>85%) was supplied by Sigma Chemical Company (St. Louis, MO). UV measurements were measured using a Perkin-Elmer (Norwalk, CT) Model 552 UV-Visible Spectrophotometer. Cyclic voltammetry and electrochemistry on the infrared cell was conducted using a Cypress Model Omni 90 potentiostat (Lawrence, KS) and a BioAnalytical Systems Model RXY recorder (West Lafayette, IN). 37

PAGE 47

2.1.2 Methods 2.1.2.1 Electrode Assembly and the Electrochemical Cell Polycrystalline gold foil was fused to a 0.125 mm gold wire and then shaped into a 3 mm diameter cylinder. The electrode was then inserted into a 5 mm inner diameter glass sheath. The cell in which cyclic voltammetry of the adsorbed layer of ferritin was conducted consisted of two glass chambers separated by a glass frit (Figure 2.1 ). The glass sheath was housed in a Teflon stopcock, which held the electrode assembly in place using a 24/25 ground glass joint (Figure 2.1). When the stopcock was opened, a stream of pressurized nitrogen was allowed to blanket the electrode, shielding it from atmospheric oxygen The cell in which the ferritin was adsorbed was machined from Teflon. The adsorption cell contained the reference and auxiliary electrodes (encased in separate polyethylene tubes, plugged at one end with Vycor disks) and the working electrode (Figure 2.2). The cell was machined by D. Zapien. 38

PAGE 48

Teflon Stopcock Glass Sheath Gold Working Electrode Ag/AgCI Reference Electrode Platinum Counter Electode Glass Frit Figure 2.1. Typical electrochemical cell used for conducting cyclic voltammetry experiments of adsorbed layers of ferritin, including glass sheath and Teflon fitting. Both chambers contain electrolyte solution. The larger chamber contains the ferritin to be adsorbed onto the gold working electrode. 39

PAGE 49

The electrode was cleaned by annealing in a natural gas/air flame, soaking in chromic acid cleaning solution, rinsing with distilled water, and finally by electrochemical cycling in 1M sulfuric acid between -0.35 V and 1.50 V (versus Ag/AgCI electrode). Electrochemical cycling was repeated until a clean scan of bare gold was obtained (Figure 2.3). 2.1.2.2 Measurement of the Electrode Surface Area The surface area was determined by measuring the anodic charge of a monolayer of iodine adsorbed onto the electrode. The electrode was cleaned in 1M H2S04 and a cyclic voltammogram scanned between -0.35 V and 1.5 V. The electrode was submerged in an aqueous solution of 0.1 M potassium iodide for five minutes at open circuit potential to effect the adsorption of iodide, then transferred to a solution of 1M H2S04 where the electrode was rinsed free of unadsorbed iodide. A cyclic voltammogram was scanned between -0.35 V and 1.5 V (Figure 2.4). The half-reaction for the oxidation of adsorbed iodide is: lads+ 3H20 I03(aq) + 6H+ (aq) + 5eThe anodic peak of adsorbed iodine was cut and weighed, and the electrolytic charge represented by the peak was calculated. The surface area determined using the relationship A = Oox, 1/ 5 F fr where fr = 1. 04 nmollcm 2 40

PAGE 50

-----Adapter Working electrode lead -----Glass sheath -----Teflon adapter ---Glass adapter ; ---Teflon cell Figure 2.2. Teflon TM electrochemical cell used in ferritin adsorption step. 4 1

PAGE 51

1100 J.LA -0.35 -0.2 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.5 POTENTIAL (Volts vs. Ag/AgCI) Figure 2.3. Cyclic voltammogram of clean bare gold, scanned from -0.35V to 1.50V. The scan was conducted in the presence of 1M H2S04 42

PAGE 52

<: :t -E-4 :z; == u 0 J I 100 !!A -0.35 -0.2 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.5 POTENTIAL (Volts vs. Ag/AgCl) Figure 2.4. Cyclic voltammogram of adsorbed iodide, used in determining electrode surface area. 43

PAGE 53

2.1.2.3 Ferritin Current-Potential Curves To obtain the current-potential curves for ferritin, a clean gold electrode was immersed into a ferritin solution at 0.00 V. The potential was scanned negative at 100 mV/s, and the potential held at -0.500 V. Under nitrogen, the electrode was transferred to a cell of deaerated phosphate buffer, where the electrode was rinsed free of unadsorbed ferritin at -0.500 V. The electrode was then placed in another cell containing fresh, deaerated phosphate buffer solution, and the potential was cycled between -0;50 V and 0.20 V. To determine the charge of the anodic peak, the current of the anodic peak of adsorbed ferritin was integrated by cutting and weighing. The experimental packing densities were determined by the Faraday law r = Q/nFA where Q is the integrated charge of the anodic peak and A is the electrode area in cm2 All measurements of packing density and peak current were accomplished in three independent experiments. Error bars represent the standard deviation of these triplicate measurements. 2.1.2.4 Ferritin Purification Ferritin was eluted through a Sephadex G 200 size-exclusion gel (protein fractionation range 5-600 kD) to remove lower molecular weight 44

PAGE 54

proteins. The column (2.5 em diameter x 20 em length) was equilibrated with 500 ml of G-200 Buffer (20 mM sodium dihydrogen phosphate, 0.9% NaCI, 0.2 mM PMSF, 0.02% NaN3). The total protein concentration was determined by the method of Bradford (81). A standard curve was constructed by measuring the absorbances (595 nm) of a series of bovine serum albumin (BSA) standards complexed with G-250 Coomassie blue dye. The ferritin sample was complexed with the dye, and its absorbance was projected onto the concentration axis of the standard curve. 2.2 Infrared Spectroelectrochemistry 2.2.1 Materials Infrared measurements were conducted using a Nicolet 550 FTIR spectrometer. The polycrystalline gold disk (99.99%), used as the working electrode in the infrared cell, was purchased from Corning Corporation (Saddlebrook, NJ). The spectroelectrochemical cell block and lid was constructed from a Teflon block, purchased from Cadillac Plastics (Denver, CO). The plunger assembly was machined from polyimide, purchased from Regal Plastics (Denver, CO). The CaF2 dove prism used on the infrared cell was purchased from Infrared Optics, Inc. Precision machining of the block and plunger was conducted by ADAM Instrument Company, Inc. (Cincinnati, OH). Machining of the Teflon lid and all other cell attachments and supports was accomplished in the laboratory. The 45

PAGE 55

differential micrometer, used for adjustment of the plunger, was purchased from Shop Tools, Inc. (Aurora, CO). The spectroelectrochemical cell and reflection optics were protected by an enclosure made from Plexiglas TM, obtained from Regal Plastics (Denver, CO. 2.2.2 Methods Measurements were conducted at room temperature using aqueous solutions. Infrared scans of ferritin were accomplished using a conventional transmission cell (International Crystal Laboratories) in the sample chamber of the instrument at open circuit potential. In situ ferricyanide reduction was conducted using the spectroelectrochemical cell. Ferricyanide solution was injected into the cell injection port through Teflon TM tubing, using a syringe. The potential was stepped from 0.200 V to -0.100 V, and single-beam infrared spectra were taken at 2 minute intervals. 46

PAGE 56

3. Direct Electron Transfer of Ferritin Adsorbed on Bare Gold Electrodes 3.1 Adsorption and Redox Activity of Ferritin A clean gold electrode was immersed in ferritin solution at 0.20 V. The potential was then ramped from 0.20 V to -0.50 Vat a scan rate of 100 mV/s. After rinsing and immersion into phosphate buffer electrolyte solution at a potential of -0.50 V the potential was scanned from -0.50 V to 0 20 V. From this scan, a well-defined current potential curve was observed (Figure 3.1 ). The anodic peak appeared at -0.13 V and the cathodic peak was at 0.00 V. Of interest, is the fact that the currents produced by the oxidation and reduction of cytochrome c on bare gold are very small, while the current potential curves generated by a monolayer of ferritin are easily measured (76). Ferritin exhibits intense peaks relative to cytochrome c due to the greater number of electrons transferred between the electrode and the ferritin molecule. This permits the use of conventional cyclic voltammetry to study current-potential behavior. As shown in Figure 3.2, the anodic peak of the second cycle is smaller than that of the first cycle. This suggests that either some ferritin desorbs or 47

PAGE 57

0 10 11A 1 -0.50 -0.1 0 0.30 POTENTIAL (Volts vs. Ag/ AgCI) Figure 3.1. Cyclic voltammogram of ferritin. The potential was cycled from -0. 50 V to 0.20 V. The dashed portion of the curve indicates a successive scan under identical conditions 48

PAGE 58

that the ferritin molecule has become less electroactive with each cycle. It is not clear why the cathodic peak is smaller than the anodic peak. The mere desorption of ferritin is unlikely given that the anodic peak on the second sweep is larger than the previous cathodic peak. One possible explanation for this phenomenon is that the ferritin molecule changes its orientation on the metal surface when switching between reduced and oxidized forms. This reversible change in orientation has been proposed for cytochrome c by Sagara and co-workers (77). The packing density for adsorbed ferritin was estimated to be 0.40 0.05 pmol/cm2 by the integrated anodic charge and the Faraday law. Consecutive cycling of the i-E curve, shown in Figure 3.2, causes the peak area to diminish, indicating that the layer is not stable to repeated cycling. The reason for this behavior is not clear. However, this result may indicate that with multiple cycles, the protein gradually assumes a new orientation that is electroinactive. This behavior is observed for cytochrome c3 at a bare mercury electrode. The voltammogram of ferritin (Figure 3.1) contains an anodic shoulder at 0.11 V and a small cathodic peak at 0.06 V, which possibly belong to a common redox couple. A similar phenomenon was reported in the current potential curve of cytochrome c by Szucs and Novak, who 49

PAGE 59

0 10 {JA t -0.50 1 0 0.30 POTENTIAL (Volts vs. Ag/ AgCI) Figure 3.2. Cyclic voltammogram demonstrating consecutive cycling of ferritin from -0.50 V to +0.20 V. 50

PAGE 60

postulated that a small peak at a potential more positive than that of the principal peak was due to oligomerized cytochrome c (51 ,52). Oligomerization would alter the globular network, which in turn would change the energy of the electroactive center, resulting in a change in the potential. The extra feature does not appear in cytochrome c adsorbed at promoter modified electrodes. One explanation why this phenomenon does not occur on these electrodes is that the adsorbed promoter layer may be anchoring the protein, limiting its lateral movement on the electrode surface, while a bare electrode limits protein movement to a Jesser extent, allowing aggregation. This postulate may also explain ferritin's behavior on gold and modified gold electrodes. Ferritin adsorbed onto an electrode modified with mercaptopropionate groups does not exhibit peaks other than those of the principally adsorbed species even though the scanning conditions are the same. Ferritin may be anchored to the electrode by the promoter layer in a fashion similar as cytochrome c. In order to demonstrate that the electrochemical signals were due to ferritin adsorbed onto the electrode surface and not ferritin in solution, the current-potential curves were examined at scan rates of 50, 100, 200, and 400 mV/sec. A plot of anodic peak current versus scan rate gives a linear relationship, with a linear regression slope of 0.384 and a correlation 51

PAGE 61

coefficient of 0.998 (Figure 3.3). According to theory, this correlation indicates that the voltammetric signal is due to adsorbed ferritin. The relationship between peak current and scan rate for adsorbed ferritin is given by the following equation (91 ): If the observed current arose from free iron in solution, a plot of scan rate versus peak current would likely yield an upsloping curve given by the equation (91 ): Ferritin is a complex system in that potentially 1500 electrons are transferred. If we assumed n = 1500, the calculated slope should be 0.85, roughly 2000 times the experimental slope. The reason for the discrepancy is that the equation applies strictly to those systems for which n is known and whose voltammetry behaves according to the known n value. If the system behaved as though n = 1500, however, the anodic and cathodic branches would be extremely narrow, and the peak currents very large The large currents would then give a large slope, as expected. The 52

PAGE 62

160 """"' ........ 120 c Q) L.. 80 L.. :l u C'O Q) 40 c.. 0 100 200 300 400 Scan Rate (mV/sec) Figure 3.3. Plot Illustrating how the peak current of ferritin varies with scan rate. Scans were carried out in the presence of 1 M phosphate buffer, pH 7.0 53

PAGE 63

voltammetry, however, resembles that of a one-electron transfer system; as though 1500 individual one-electron transfer reactions are occurring. There are several examples in the literature in which electrons are transferred to and from molecules containing multiple redox centers (up to 1200 per molecule in some electroactive polymers) while the voltammetry has the appearance of a one-electron transfer system (95-98). 3.2 Potential Dependence of Ferritin Adsorption The degree to which ferritin adsorbs onto a bare gold electrode is highly dependent upon potential. For example, when a clean gold electrode is immersed in ferritin solution at 0.20 V, rinsed with phosphate buffer, and the potential scanned between -0.50 and 0.20 V, no Faradaic current is produced (Figure 3.4, panel a). However, when the electrode is immersed in ferritin solution at potentials of -0.10 V -0.30 V, and -0.50 V, it was demonstrated that the amount of faradaic charge from adsorbed ferritin is dependent upon potential (Figure 3.4, panels a-d). Assuming that all ferritin molecules on the anode are completely electroactive, the packing densities for adsorption potentials of 0.0, -0.10, -0.30, and -0.50 V are 0 .0, 0.0, 0.18 0.02, and 0.26 0.03 pmol/cm2 respectively. This data suggests that ferritin is adsorbed in an electroactive state only when it is first reduced. It is known that conformational changes in protein structure may occur with changes in oxidation state of the reaction center (78, 79). 55

PAGE 64

-o.30 v c 0.20 v a -o.s-o +oA.o Figure 3.4. Dependence of ferritin adsorption on potential. Cyclic voltammetry was carried out in the presence of 1M phosphate buffer, pH 7.0. 56

PAGE 65

For example, as adsorbed ferrocytochrome cis converted to ferricytochrome c, the reaction center entropy change becomes more negative, indicating that the globular sphere is less rigid when the electroactive center is in the oxidized form. It is possible that conformational changes in quaternary structure, induced by the reduction of the protein, allow adsorption onto the gold surface while maintaining electroactivity. 3.3 Dependence of Packing Density on Ionic Strength, Ferritin Concentration, and Adsorption Time 3.3.1 Ionic Strength Of interest in characterizing the interactions between proteins and bare electrodes is whether the interactions are primarily hydrophobic or hydrophilic. One method of characterization involves observing changes in packing density (the density of ferritin molecules adsorbed to the electrode surface) with changes in ionic strength. Higher ionic strength favors hydrophilic conditions; therefore, an increase in relative packing density at higher ionic strengths may indicate that hydrophobic interactions dominate at the protein-electrode interface. In the case of ferritin, as the ionic strength is increased from 0.30 to 1.5 M, the relative packing density (the absolute packing density normalized with respect to the greatest packing density) increases to a maximum value beginning at about 1.0 M (Figure 57

PAGE 66

>-iii c G) c 01 c l2 CJ as 0.. 0.6 G) > 'P as a; 0:: 0.30 0.60 0.90 1.2 1.5 Ionic Strength (Molar) Figure 3.5. Variation of ferritin packing density with ionic strength. Scans were carried out in phosphate buffer, pH 7.0. Ionic strength was varied from 0.3 M to 1.5 M. 58

PAGE 67

3.5). Apparently, the hydrophobic regions of ferritin interact with the gold surface while the electroactivity of the protein is preserved. This result suggests that the ferritin-gold interface is primarily hydrophobic in nature. Similar behavior is observed in the adsorption of ferritin at tin-doped indium oxide (TIO) electrodes, in which ferritin exhibits a limiting packing density at about 1 M ionic strength (55). 3.3.2 Ferritin Concentration Ferritin concentration has a direct effect on packing density. Throughout this study, the ferritin concentration was held constant for each experiment. An adsorption isotherm for ferritin at bare gold is shown in Figure 3.6. For the range of concentrations studied, the packing density showed a direct linear relationship with ferritin concentration. Globular proteins exhibit well-defined plateau regions for dilute to semidilute regimes. If higher concentrations of ferritin were available, the ferritin isotherm would likely show a plateau at the higher concentration Although the packing density for the highest concentration of ferritin is approximately 0.40 pmollcm2 the maximum packing density for a ferritin monolayer is about 1.1 pmollcm2 calculated from its "footprint" area. 59

PAGE 68

O.Z5 -o.zo N 1 0.15 0.10 0.05 0 0.50 1.0 1.5 z.o z.s 3.0 Ferritin Concentration (mg/ml) Figure 3.6. Ferritin adsorption isotherm. The packing density varied linearly with ferritin concentration throughout the concentration range tested. 60

PAGE 69

3.3.3 Adsorption Time Adsorption is a key step in observing the behavior of proteins on bare electrodes. Adsorption and desorption involve several distinct steps: transport to the electrode surface, anchoring to the surface, structural rearrangement, disengagement, and transport from the electrode Protein crowding on the surface of an electrode may change the structure or orientation of a protein (80). In order to characterize the adsorption process of ferritin to a bare gold surface, the adsorption time of ferritin at -0.50 V was varied (Figure 3. 7). After 90 seconds of adsorption time, a limiting packing density was achieved These data suggest that the adsorption rate is limited more by the diffusion of ferritin to the gold surface than by structural rearrangements or changes in orientation as the protein adsorbs to the electrode. These results are in contrast to the long periods of time required for ferritin to reach tin-doped indium oxide electrodes (55). Because horse spleen ferritin has an isoelectric point of 4.6, it has a net negative charge at neutral pH. The reasons for the large differences are not clear; however, it appears that the negative charge of the surface oxides on the tin-doped indium oxide electrode is not solely responsible. Using gold electrodes, potentials negative of the potential of zero charge were used; therefore, some electrostatic repulsion is also expected between the gold surface and the negatively charged ferritin molecules. 61

PAGE 70

0.25 0.20 I 0.15 0.10 0 30 60 90 120 Adsorption Time (seconds) Figure 3.7. Relationship between ferritin packing density and adsorption time Adsorption was carried out at -0.50 V in 1M phosphate buffer, pH 7.0. 62

PAGE 71

4. Design of the Infrared Reflection Absorption Spectroscopy (IRRAS) Apparatus 4.1 Spectroelectrochemical Cell 4.1.1 Description The infrared reflection-absorption spectroscopy (IRRAS) cell is an external reflectance cell designed to simultaneously serve as an electrochemical cell and infrared spectroscopic cell (Figure 4.1 ). Features of the cell include Teflon construction for chemical inertness, a CaF2 dove prism that serves as an infrared-transparent window as well as a barrier for the cell solution, a gold disk embedded in a plunger assembly serving as the working electrode and IR reflector, a differential micrometer for adjusting the solution layer thickness, and a stainlesssteel base for mounting purposes. 4.1.2 Design A fundamental design consideration is the type of cell used to carry out reflection absorption measurements. Weighing potential 63

PAGE 72

MOUNTING HOLES FOR THE LID ---RESERVOIR (AT EACH CORNER) 0 0 ._-GOLD DISC '-----SOLUTION HOLE L-----O-RING .__ ___ HOLES FOR DOVE PRISM MOUNT .._-----DOVE PRISM 0-RING GROOVE Figure 4.1. Diagram of the infrared reflection-absorption spectroscopy (IRRAS) cell. 6 4

PAGE 73

advantages versus disadvantages is critical in choosing a specific cell design. The principal advantages of external reflectance cells include the ability to control the solution thickness above the electrode surface and to provide a uniform flow of solution over the electrode (67). The cell was designed with several considerations in mind, including chemical inertness, effective collection of the infrared beam, reproducible sample thickness and mechanical stability of the cell. Chemical inertness was important because any reactions between the cell material and components in the solution might generate species that interfere with spectroscopic and electrochemical methods For optimal resistance to such reactions, the cell was machined from a 4 in3 Teflon TM cube, while the plunger, to which the gold disk was mounted, was constructed from a polyimide rod. Polyimide also offered the advantage of being resistant to cold-flowing, a consequence of high pressures on a soft substance, such as Teflon Another factor critical to the sensitivity of the measurements is the proper direction of the infrared beam to the gold working electrode. This is important because of the loss of any portion of the infrared beam would result in reduced sensitivity and decreased signal to noise ratio To minimize reflection losses, the infrared beam approached the dove prism at normal incidence. After passing through the prism, the beam impinges 65

PAGE 74

on the working electrode at an angle of 60 degrees with respect to the electrode surface normal (Figure 4.2). A dove prism made of CaF2 was used because of its transparency in the mid-infrared region (4000-1100 cm-1 ) and its insolubility in aqueous media. A 0.25 inch channel extended from the gold disk to the back of the polyimide plunger. A copper wire was fed into the channel through a hole drilled into the side of the plunger, located about one-inch from the end of the plunger. A Teflon rod was inserted into the channel through the back of the plunger, pushing the copper wire against the back of the gold disk. A spring was inserted behind the Teflon rod, followed by a pipe-threaded plug screwed into the back of the plunger. Two 2-inch lengths of 0.125 inch polyethylene tube were plugged on one end with 0.125 inch Vycor disks. One of the tubes was filled with a 1 M NaCI solution, followed by insertion of a 0.1 mm diameter gold wire; this assembly served as the auxiliary electrode. The other tube was used to make a Ag/AgCI reference electrode. A 1-mm channel linked the solution in front of the gold disk with the sink located on the top of the reflection cell, allowing for all three electrodes to be in electrical contact through the solution. To provide reproducible sample thickness between measurements, a differential micrometer (non-rotating) with 0.1 J.lm resolution was installed on the backside of the cell using a four-poled 66

PAGE 75

I I -----PLUNGER I ---GOLD DISC I .,. I ELECTRODE -t---CaF2 DOVE PRISM \ \ Figure 4.2. Schematic diagram illustrating how the beam passes through the dove prism, reflects off of the gold disk electrode, and passes through the dove prism on the other side. 67

PAGE 76

stainless steel mount. The micrometer tip pushed against the stainless steel pipe thread plug located on the back of the plunger. Adjustment of the micrometer allowed for a variable distance between the gold electrode and the dove prism. 4.2 Enclosure 4.2.1 Description An air-tight enclosure was constructed from aluminum and Plexiglas TM to provide a moisture-free environment for the optics, the cell, and the detector, while preventing the entry of external sources of radiation (Figure 4.3). An optical bench was constructed from 5/16 inch thick anodized aluminum re-inforced by 1.5 inch wide skirts bolted underneath the perimeter of the base plate. Four 1-inch square rods, located at each corner, raised the bench to the height of 6 inches The optical bench was constructed by Mr. Fedri Marrugo. The enclosure rested on top of the optical bench while the bench itself was bolted to the FTIR spectrometer using 3/8 inch bolts. Finally, an aluminum flange, housing a 2 inch diameter CaF2 window, was bolted on one side to the enclosure, and on the other side to the FTIR enclosure. The window flange was also machined by Mr. Fedri Marrugo. 68

PAGE 77

LID UTILITY PLATE Figure 4.3. Plexiglas enclosure containing the optics, the spectroelectrochemical c e ll, and the detector 69

PAGE 78

4.1.3 Design Important considerations in the design of the enclosure include durability, accessibility to the interior components, and airtightness Plexiglas TM was chosen because of its strength and ease with which it is machined. The Plexiglas sheets were joined with epoxy resin to provide strong seams. Angled aluminum provided additional strength at the bottom rim of the enclosure while providing a means of bolting it to the bench. To enhance accessibility to the interior optical components, the enclosure was built with a removable top attached to the enclosure body with stainless steel cap screws. A utility plate, machined from Plexiglas, was attached to the side opposite of the FTIR instrument. This device afforded a common entrance for all solutions and electrode leads, and also provided added accessibility to the internal components. To make the apparatus airtight, all seams were sealed with silicone caulking. Foam strips provided the seal between the enclosure base and optics bench, and between the removable top and the enclosure, while a rubber gasket provided a seal between the enclosure and the utility plate. 4.3 Optics 4.3.1 Internal Optics The internal optics include those components within the FTIR enclosure: the source, source optics, interferometer components, and 70

PAGE 79

the mirror directing the beam out of the spectrometer. The beam originates from a heated silicon carbide rod (Giowbar), and is directed into the interferometer by a collimating mirror. The oscillating translation of the moving plane mirror generates an interference pattern of radiant power with time, called an interferogram. The interferogram is directed through the sample, and into a detector. The analog signal produced by the detector is digitized, and inputted into the processor where it is transformed into the spectrum in the frequency domain. Because the spectroelectrochemical cell was located on the external optics bench, it was necessary for the beam to be routed outside of the instrument box to the external optics. This was accomplished by changing the way one of the off-axis parabolic mirrors was originally used (Figure 4.4 ). An aluminum mount of the proper height was machined and mounted at the exit of the interferometer. The mirror was inverted, attached to the mount, and aligned to focus the beam directly through the exit port. After passing through the flanged CaF2 window, the beam continued a path defined by the external optics. 4.3.2 External Optics The external optics include a collimating mirror, a focusing mirror, the spectroelectrochemical cell, and a detector mirror (Figure 4.4). Once 71

PAGE 80

fNT Interferometer F 1 First focusing Mirror P Wire Grid Polarizer W Window Flange C r.1 Collimating Mirror f 2 Second F oc.using R C Reflectance Cell F 3 Oetcc [Or Jo4irror 0 Hg C dTe Octector I II Figure 4.4. Schematic diagram illustrating the optical setup of the interferometer, the optics mirrors, the spectroelectrochemical cell, and the detector. 72

PAGE 81

the beam leaves the FTIR spectrometer, a mirror collimates the beam and directs it to a focusing mirror, which in turn focuses the beam onto the surface of the gold working electrode. The beam is then reflected from the electrode and is collected by the detector mirror, where it is focused onto the detector. 4.3.3 Detector A triglycine sulfonate detector, originally installed in the FTIR spectrometer, was removed from the instrument and installed on the external bench. A longer cable was constructed to permit the operation of the detector in the remote location. 4.4 Testing and Evaluation of the Apparatus 4.4.1 Reduction of Ferricyanide To validate the ability of the instrument to collect data, we tested the instrument on a simpler system. Areas of concern regarding the cell included 1) whether the cell would work to exhaustively electrolyze the sample, 2) whether the infrared spectra of the analyte could be measured, and 3) whether the cell design was adequate to prevent appreciable diffusion of analyte from the cell. To investigate the soundness of the cell design with regard to the aforementioned concerns, the reduction of ferricyanide in the spectroelectrochemical cell was conducted. Prior to the experiment, 73

PAGE 82

infrared spectra of aqueous solutions of ferrocyanide and ferricyanide were taken. The ferricyanide peak was present at 2104 cm1 while the ferrocyanide peak absorbed strongly at 2033 cm1 (Figure 4.5a). Upon stepping the potential from 0.200 V to -0.100 V, a gradual decrease in the intensity of the ferricyanide peak was observed. After 7.5 minutes, the peak had completely disappeared (Figure 4.5b). A simultaneous increase in peak intensity of the ferrocyanide peak had also occurred. The above data suggest that the exhaustive electrolysis readily occurs with the spectroelectrochemical system and that diffusion of analyte away from the cell does not readily occur 4.4.2 Spectrum of Ferritin Infrared spectra were taken at ferritin concentrations of 51.0 mg/ml and 102 mg/ml (Figure 4.6). The data obtained indicate that ferritin chromatographed over Sephadex G-200 is of sufficient concentration to obtain infrared data. Additionally, some limited structural information may be inferred from the spectrum The peaks at 1650 cm1 and 1553 cm1 are known as the amide I and the amide II bands, respectively. The amide I band arises from the C=O stretching vibrations of the backbone and peptide linkages while the amide II band arises from out-of-phase combinations of N-H in-plane bending and C-N stretch vibrations of peptide linkages (81 ). 7 4

PAGE 83

(a) 2033 em' F erricyanide Ferrocyanide 2400 2200 2000 1800 1600 Wavenumbers (em"') 2200 2000 2100 1 Wavenumbers (em) Figure 4.5 (a) Infrared spectra of ferricyanide and ferrocyanide. (b) Reduction of ferricyanide with respect to time. The potential was stepped from 0.200 V to -0.100 V 75

PAGE 84

c .n .... 0 (/) 1800 51 mg/ml 1655 cm-1 \1650 cm-1 1 / em I -.....--,___.._,;---v_./ """'"'-v -.../"...__,-..... 102 mg/mL p \ I \ (' I \ I \ / \J \, -------. J '-.__,/ ._ 1700 1600 1500 Wavenumbers (cm-1 ) Figure 4.6. Infrared spectrum of ferritin at 51 mg/ml and at 102 mg/ml. 76

PAGE 85

The predominant secondary structure of a protein is a strong determinant of the absorbance maxima of the amide I band. The amide I band of a protein with a predominant a-helical structure is found near 1656 2 cm1 whereas a protein having primarily a p.sheet structure will have a maximum between 1643 and 1631 cm1 (81). In the case of ferritin, the amide I band maxima is indicative of an a-helix secondary structure. This is consistent with the extensive a-helical structure of ferritin H and L subunits (5, 1 0,39). Also supporting this deduction is the lack of any shoulder between 1643 and 1631 cm-1 indicating a lack of p. sheet structure (81) Another point of interest is the peak assignments of both maxima of the amide I peak Earlier studies have established that bands at 1648 2 cm1 may be assigned to disordered secondary structure (81 ). Given this fact, the maximum at 1650 cm1 may arise from disordered secondary structure within the H and L subunits. Contradicting this notion, however, is the predominance of a-helices and the absence of a significant amount of disordered secondary structure upon visual inspection of the crystal structure. Another possible explanation is that each of the two types of subunits are responsible for each of the maxima. Additional studies will 77

PAGE 86

be required to definitively ascertain whether the additional maximum is due to disordered structure or the two different subunit types. 4.4.3 Problem Areas The IRRAS apparatus had several problems which prevented the collection of reduced minus oxidized spectra of ferritin. First, the cell is not suitable for electrolyzing dissolved ferritin. Second, because of the additional optics required to use the reflection absorption apparatus, a significantly reduced signal to noise ratio would result. Finally, atmospheric water vapor gives strong absorption bands in the regions of Amide I and Amide II. We were unable to remove these signals by subtraction purging alone. The development of a mathematical approach to subtract these bands will be required. 78

PAGE 87

5. Summary and Conclusion 5.1 Summary The uptake and release of iron is central to the function of ferritin. Additionally, redox reactions are crucial to this uptake and release. Therefore, to study the structure and function of ferritin, we characterized its electrochemical behavior on bare gold. We were able to obtain data suggesting that ferritin oligomerizes on bare gold, but not on a gold electrode modified with mercaptopropionate groups. We therefore postulate that ferritin is anchored to the electrode by the promoter layer. Additionally, our data suggest that a reversible change in orientation of the ferritin molecule may be occurring on the electrode surface. The relationship between ferritin packing density and scan rate was found to be linear. From these data, it was determined that the voltammetric signals were from ferritin adsorbed onto the gold electrode and not ferritin in solution A study of the potential dependence on ferritin adsorption suggested that ferritin is adsorbed in an electroactive state only when it is first reduced. Additionally, the data suggest that it is possible that a change in quaternary structure occurs during the reduction process, 79

PAGE 88

allowing the protein to be adsorbed onto the gold surface while maintaining its electroactivity. Additionally, studies of packing density at various ionic strengths, ferritin concentrations, and adsorption times were conducted. From the ionic strength data, it was concluded that hydrophobic interactions predominate at the ferritin-gold surface. The packing density of ferritin increased with dissolved ferritin concentration. From the adsorption time data, it was determined that the adsorption rate is limited more by the diffusion of ferritin onto the gold surface rather than by structural rearrangements or changes in orientation following initial attachment to the electrode surface. An infrared reflection-absorption spectroelectrochemical cell was designed and constructed, along with an optical bench enclosure, and mounts for reflection optics. Testing of the spectroelectrochemical cell using ferricyanide revealed that the cell had the ability to exhaustively electrolyze an analyte in fairly short time and that infrared spectra can be recorded in situ. 5.2 Conclusion As shown in this study, the electrochemical data of ferritin adsorbed on bare gold are consistent with what has been suggested in other studies: that the protein molecules may be laterally mobile, forming 80

PAGE 89

oligomers. Perhaps more important, the data suggest that the structure of the protein sheath is linked to the oxidation state of the core iron. The spectroelectrochemical cell can be used to obtain in situ infrared spectra. However, the cell is still not suitable for obtaining the reduced-minus-oxidized difference spectrum of ferritin. There are several reasons for this. First, the direct .electron transfer of dissolved ferritin does not occur appreciably at bare gold electrodes. The ferritin in solution must be exhaustively electrolyzed in order for the difference spectrum to be meaningful. An electrode system capable of effecting the direct electron transfer of ferritin must first be found. Secondly, the use of a reflection-absorption cell to obtain transmission spectra typically results in a reduced signal to noise ratio due to the additional optics required to transmit the infrared beam outside of the FTIR to the external bench. A transmission cell which can be mounted inside the spectrometer, resulting in a higher signal-to-noise ratio, will be needed to observe the small difference bands. Lastly, since water absorbs strongly in the regions of the Amide I and Amide II bands, a method must be developed in order to eliminate the atmospheric water absorption bands from the difference spectrum 81

PAGE 90

References 1. P. M Harrison and T. H Lilley (1989) Ferritin." in Iron Carriers and Iron Proteins. ed. Thomas M. Loehr. (New York: VCH Publishers, Inc.) 123-238. 2 J. C G. Walker (1977) The Evolution of the Atmosphere. (New York: MacMillan) 262-264. 3. Biedermann, G., Schindler, Acta. Chern. Scand. 11 ( 1957) 731-7 40. 4. B F. Matzanke, G. M. Matzanke, K. N. Raymond (1989) "Siderophore Mediated Iron Transport." in Iron Carriers and Iron Proteins. ed. Thomas M. Loehr. (New York: VCH Publishers, Inc.), 3. 5. E. C. Theil, Ann. Rev. Biochem 56 (1987) 290. 6. P M Harrison, F. A. Fischbach, T. G. Hoy, and G. H. Haggis, Nature 216 (1967) 1188. 7. W. H. Massover, J. M. Cowley, Proc. Nat. Acad. Sci. USA 70 (1973) 3847-3851. 8. J. L. White, J. M.A. Smith, P. M. Harrison, Adv. lnorg. Biochem. 5 (1983) 39-50. 9. S H. Banyard, D K. Stammers, P.M. Harrison, Nature 288 (1980) 298300. 10. E. C. Theil, Advances in Enzymology 63 (1990) 423. 11. G. H. Haggis, J. Mol. Bioi. 14 (1965) 598. 12. P.M. Harrison, F. A. Fishbach, T. G. Hoy, G. H. Haggis, Nature (London) 216 (1967) 1188. 13. F. Constanzo, M. Colombo, S. Staemplfli, C. Santoro, M. Marone, R. Frank, H. Delius, R. Cortese, Nucleic Acid Res. 14 (1986) 721. 82

PAGE 91

14. M. Heusterspreute, R R Crichton, FEBS Lett. 129 (1981) 322. 15. D. Boyd, C. Vecoli, D. M. Belcher, S. K. Jain, J. W Drysdale, J. Bioi. Chern. 260 (1985) 11755. 16. C. Santoro, M. Marone, M. Ferrone, F. Constanzo, M. Colombo, C. Minganti, R Cortese, L Silengo, Nucleic Acid Res. 14 (1986) 2863. 17. C. Wustefeld, R R Crichton, FEBS Lett. 150 (1982) 43. 18. M. T. Murray, K. White, H. N. Munro, Proc. Nat. Acad. Sci. USA 84 (1987) 7 438. 19. E. A. Leibold, H N. Munro, J. Bioi. Chern. 262 (1987) 7335. 20. L F. Dickey, S. Sreedharan, E. C. Theil, J. R Didsbury, Y. -H. Wang, R E. Kaufman, J. Bioi. Chern. 262 (1987) 7901. 21. P. W. Stevens, J. B. Dodgson, J. D. Engel, Mol. Cell. Bioi. 7 (1987) 1751. 22. J. F. Collawn, L K. Gowan, H. Crow, C. Schwabe, W. W. Fish, Arch. Biochem. Biophys. 259 (1987) 105. 23. P. M. Harrison, T. H. Lilley (1989)" Ferritin." in Iron Carriers and Iron Proteins. ed. Thomas M. Loehr. (New York: VCH Publishers, Inc.), 126. 24. S. Otsuka, H. Maruyama, I. Listowsky, Biochemistry 20 (1981) 52265232. 25. J. M. Addison, J. E. Fitton, W. G. Lewis, F. May, P.M. Harrison, FEBS Lett. 64 (1983) 139-144. 26. D. J. Lavoie, K. Ishikawa, I. Listowsky, Biochemistry 17 (1978) 54485454. 27. J. F. Collawn, Jr., W. W. Fish, Comp. Biochem. Physiot. B 78 (1984) 653-656. 28. J. F. Collawn, Jr., P. L. Lau, S. L Morgan, A. Fox, W. W Fish, Arch. Bioch. Biophys. 233 (1984) 260-266 .. 83

PAGE 92

29. W. H. Massover, Biochirn. Biophys. Acta, 829 (1985) 377-386. 30. M. C. Linder, G. M. Nagel, M. Roboz, D. M. Hungerford Jr., J. Bioi. Chern. 256 (1981) 9104-9110. 31. G. A Clegg, J. E. Fitton, P.M. Harrison, A Treffry, Prog. Biophys. Mol. Bioi. 36 (1981) 53-86. 32. D. W. Rice, B. Day, J. M. A Smith, J L White, G. C. Ford, eta/. FEBS Lett. 181 (1985) 165-168. 33. P. M Harrison and T. H Lilley, (1989)" Ferritin." in Iron Carriers and Iron Proteins. ed. Thomas M. Loehr. (New York: VCH Publishers, Inc ) 182-189. 34. J. R. Mertz, E. C. Theil, J. Bioi. Chern. 258 (1983) 11719-11726. 35. G. C. Ford, P. M. Harrison, D W. Rice, J. M. A Smith, A. Treffry, J. L White, and J Yariv, Phil Trans Roy. Soc. Lond._ B304 (1984) 551. 36. D. W. Rice, G. C. Ford J. L White, J. M. A Smith, and P M. Harrison, Protides of the Biological Fluids 31 (1984) 177. 37. S. Granick, Physiol. Rev. 31 (1951) 489. 38. J. L Farant, Biochirn. Biophys. Acta 13 (1954) 679. 39. P M. Harrison, T. H. Lilley (1989)" Ferritin." in Iron Carriers and Iron Proteins. ed. Thomas M. Loehr. (New York : VCH Publishers, Inc.) 215-221. 40. S. Osaki and S Sirivech, Fed. Proc. 30 (1971) 1292 41. R. Ulvik and J. Romslo, Biochirn. Biophys Acta 541 (1978) 251. 42. R. J. Ulvik and I. Romslo, Biochirn. Biophys. Acta 588 (1979) 256. 43. R. J. Ulvik, Bioenerg. Biornernb 15 (1983) 151. 44. B. E. Speyer and J. Fielding, Brit. J. Haernatol 42 (1979) 255. 45. G. D. Watt, D Jacobs, and R. B. Frankel, Proc. Nat. Acad. Sci. USA 85 (1988) 7457. 8 4

PAGE 93

46. J. G. Joshi, S. Goodman, V. V. Deshpande, and D. J. Price (1985) "Proteins of Iron Storage and Transport." Ed. G. Spik, J. Montreuil, R. R. Crichton, and J. Mazunier (Elsevier Science Publishers) 93-96. 47. D. J. Price and J. G. Joshi, J. Bioi. Chern. 258 (1983) 10873 48. A. Treffry and P. M. Harrison (1982) "The Biochemistry and Physiology of Iron." Ed. P. Saltman and J. Hegenauer (Elsevier North Holland, Inc.) 459462. 49. T. M. Loehr, ed. (1989) Iron Carriers and-Iron Proteins (New York: VCH Publishers, Inc.) ix 50. S. Levi, P. Santambrogio, B. Corsi, A. Cozzi, P. Arosio, Biochem J. 317 (1996) 467-473. 51. A'. Szucs and M. Novak, J. Electroanal. Chern. 383 (1995) 75-84. 52. A'. Szucs and M. Novak, J. Electroanal. Chern. 384 (1995) 47-55. 53. I. Carelli, I. Chiarotto, A. Curulli, G. Marrosu, Anal. Lett. 28 (1995) 23252345. 54. M. J. Tarlovand E. F. Bowden, J Am. Chern Soc. 113 (1991) 18471849. 55. R. J. Cherry, A. J. Bjornsen, and D. C. Zapien, Langmuir 14, 8 (1998) 1971-1973. 56. P. Yeh and T. Kuwana, Chern. Lett. ( 1977) 1145-1148. 57. D. S. Cohen, F. M. Hawkridge, H. N. Blount, C R. Hartzell, Charge Field Eff. Biosyst. [Pap -Int. Symp. Bioelectrochem. Bioenerg.) (1984), Meeting Date 1983, 19-31. Ed J. M. Allen, P. N. R. Usherwood (Abacus Press, Tunebridge Wells, UK). 58. E. F. Bowden, F. M. Hawkridge, J. F. Chlebowski, E. E. Bancroft, C Thorpe, and H. N. Blount, J. Am. Chern. Soc. 104 (1982) 7641-7644. 85

PAGE 94

59. G. W. Ewing (1985) Instrumental Methods of Chemical Analysis, 5th ed. (McGraw-Hill, New York) 291-292 60. P. H.Rieger (1994) Electrochemistry, 2"d ed. (Chapman and Hall, New York) 156-190. 61. T. D. Martin, S. A Monheit, R. J. Niichel, S. C Peterson, C. H. Campbell, D. C Zapien, J. Electroanal. Chern. 420 (1997) 279 62. T. Sagara, H. Murakami, S. Igarashi H. Sato, and K. Niki, Langmiur, 7 (1991) 3190. 63. S. Song, R. A Clark, E. F. Bowden, M. J. J. Tarlov, Phys. Chern., 97 (1993) 6564. 64. T. Sagara, H. Kawamura, K. Ezoe, and N. Nakashima, J. Electroanal. Chern., 445 (1998) 171-177. 65. L. S. Wang, V. L. Vilker, W. T. Yap, and V. Reipa, Langmuir., 11 (1995) 4818-4822. 66. W. R. Heineman, J. N. 'Burnett, and R. W. Murray, Anal. Chern., 40 (1968) 1974. 67. K. Ashley, Talanta, 38 (1991) 1211. 68. B Baden -and -c. Lamy (1988) "Infrared Reflectance Spectroscopy," In Spectroelectrochemistry: Theory and Practice. ed. R. J. Gale (New York: Plenum Press), 189-207 69. D. Moss, !;. Naberdryk, J. Breton, and W. Mantele, Eur. J. Biochem., 187 (1990) 565. 70. D. D. Schlereth and W. Mantele, Biochemistry, 32 (1993) 1118. 71. A Bewick, K. Kunimatsu, and B S. Pons, Electrochimica Acta, 25 (1980) 465. 72. H. Seki, K. Kunimatsu, and W G. Golden, Applied Spectroscopy, 39 (1985) 437. 8 6

PAGE 95

73. J. 0. Bockris and B. Yang, J. Electroanal. Chem., 252 (1988) 209-210. 74. H. 0. Finklea and D. D. Hanshew, J. Am. Chem. Soc., 114 (1992) 31733181 75. C. E. D. Chidsey, Science 251(1991) 919-922 76. C. Hinnen, R. Parsons, and K. Niki, J. Electroanal. Chem. Interfacial Electrochem 147 (1983) 329-337. 77. T. Sagara, S. Nakajima, H. Akutsu, K. Niki, and G. S. Wilson, J. Electroanalytical Chem. Interfacial Electrochem 297 (1991) 271-282. 78. K. B. Koller, F. M. Hawkridge, J. Am. Chem. Soc. 107 (1985) 7412-7417. 79. K. B. Koller, F. M. Hawkridge, J. Electroanal. Chem Interfacial Electrochem. 239 (1988) 291-306 80. W. Norde, Adv. Coli. Interface Sci 25 (1986) 267-340. 81. M. M Bradford, Anal Biochem. 72 (1976) 248-254. 82. Z. Q. Feng, S lmabayashi, T. Kalciuchi, and K. Niki J. Electroanal. Chem. 408 (1996) 15-20. 83. J. M Cooper, K. R. Greenough, and C J McNeil, J. Electroanal. Chem 347 (1993) 267-275. 84. W. J. Albery, M. J. Eddowes, H A. 0. Hill, and A. R. Hillman, J. Am. Chem. Soc (1981) 3904. 85 A. Treffry, J. Hirzmann, S.J Yewdall, and P .M. Harrison, FEBS Letters 302 (2) (1992) 108-112. 86. D.M Lawson, eta/., FEBS Letters 254 (1989) 207-210. 87. A. Treffry, P .M. Harrison, A. Luzzago, and G. Cesareni, FEBS Letters 247 (1989) 268-272 88 V J W a d e e t a/., J Mol Biol221 (1991) 1443 1452. 87

PAGE 96

89. D.M. Lawson, eta/., Nature 349 (1991) 541-544. 90. B. Xu and D. Chasteen, J. Bioi Chern. 266 (1991) 19965-19970. 91. A.J. Bard and L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications (1980) 1-26 (Johns Wiley and Sons). 92. D.A. Skoog and J.J. Leary, Principles of Instrumental Analysis (1992) 536-564 (Saunders College Publishing: New York) 93. N.D. Chasteen, I.M. Ritchie, and J. Webb, Anal. Bioch. 195 (1991) 296302. 94. T.D. Martin, S.A. Monheit, R.J Niichel, S.C. Peterson, C.H. Campbell and D.C. Zapien, J. Electroanal. Chern. 420 (1997) 279-290. 95 F. Ammar and J. Saveant, J. Electroanal. Chern. 47(1973) 155. 96. T. Saji, N.F. Pasch, S.E. Webber, and A.J. Bard, J. Phys. Chern. 82 (1978) 1101-1105. 97. T.W. Smith, J.E. Kuder, and D. Wychick, J. Polym. Sci. 14 (1976) 2433. 98. J.B. Flanagan, M. Schlomo, A.J. Bard and F. Anson, J. Am. Chern. Soc 100 (1978) 4248-4253 88