Citation
Unmediated cyclic voltammetry of ferritin at a polycrystalline gold electrode modified with self-assembled monolayers

Material Information

Title:
Unmediated cyclic voltammetry of ferritin at a polycrystalline gold electrode modified with self-assembled monolayers
Creator:
Martin, Todd Darren
Publication Date:
Language:
English
Physical Description:
ix, 93 leaves : illustrations ; 29 cm

Subjects

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

Notes

Bibliography:
Includes bibliographical references (leaves 88-93).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Science, Department of Chemistry
Statement of Responsibility:
by Todd Darren Martin.

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:
31508838 ( OCLC )
ocm31508838
Classification:
LD1190.L46 1994m .M37 ( lcc )

Full Text
UNMEDIATED CYCLIC VOLTAMMETRY
OF FERRITIN AT A POLYCRYSTALLINE GOLD
ELECTRODE MODIFIED WITH
SELF-ASSEMBLED MONOLAYERS
by
Todd Darren Martin
B. A., Creighton University, 1992
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Chemistry
1994


1994 by Todd Darren Martin
All rights reserved.


This thesis for the Master of Science
degree by
Todd Darren Martin
has been approved for the
Department of
Chemistry
by


Martin, Todd Darren (M. S., Chemistry)
Unmediated Cyclic Voltammetry of Ferritin at a
Polycrystalline Gold Electrode Modified with Self-
Assembled Monolayers
Thesis directed by Assistant Professor Donald C. Zapien
ABSTRACT
Horse spleen ferritin has been investigated
voltammetrically by using a polycrystalline gold electrode
modified with self-assembled monolayers. Two particular
monolayers produce appreciable redox waves: 3-mercapto-l-
propane sulfonic acid (3-M-l-PSA) and 3-mercapto propionic
acid (3-MPA). Both adsorbed layers show redox chemistry that
occurs in the range of the iron core in ferritin. 3-M-l-PSA, an
adsorbed layer, is used to promote electron transfer in a 4
mg/mL solution of horse spleen ferritin. The resulting redox
wave shows a peak potential difference of 157 mV and a peak
current ratio of 0.895 (ipa/ipC). The peak current ratio
approaches that of a chemically reversible system. The peak
potential difference is more than twice that of theory for an
electrochemically reversible one electron transfer. However,
the peak potential difference of Fe(II)(CN)64', generally
iii


accepted as electrochemically reversible, is 194 mV using our
electrochemical cell and electrolyte system. Apoferritin is also
studied using cyclic voltammetry. The resulting cyclic
voltammetric scans show no redox activity as expected,
supporting the postulate that the iron core in horse spleen
ferritin is responsible for the observed electroactivity. The
cathodic potential is reduced sequentially by 100 mV resulting
in a marked decreased in the anodic peak current, suggesting a
response by a chemically reversible system. The iron core was
reconstituted in the apoferritin shell, but no redox waves were
observed in the cyclic voltammetric scans. However, the
reconstitution occurred in a phosphate free environment which
does not imitate the biological environment of the horse spleen.
Phosphate is thought to be responsible for the catalytic
oxidation of ferrous iron to ferric iron during the proliferation
of the iron core. The absence of phosphate may also affect the
electron transfer during a cyclic voltammetric scan.
This abstract accurately represents the content of the
candidate's thesis. I recommend its publication.
IV


I
CONTENTS
CHAPTER
1. INTRODUCTION 1
Cellular Ferritin: Structure and Function 1
Regulation of Cellular Ferritin 5
Serum Ferritin: How Does it Differ from
Cellular Ferritin 7
The Various Types of Cellular Ferritin 8
Horse Spleen Ferritin Structure 1 0
Proposed Mechanisms of Ferrous
Incorporation into the Iron Core 1 6
Cyclic Voltammetry of Proteins 1 8
Research Objectives 2 6
2. EXPERIMENTAL METHODS 2 9
Suppliers of All Chemicals Used 2 9
Equipment 3 0
Electrochemical Cell Configuration 3 0
The Combination of Electrodes Used 3 2
General Voltammetric Procedure for Proteins 3 6
Removing Iron from the Protein Shell 3 9
v


Reconstituting the Iron Core in Apoferritin 4 0
3. RESULTS 44
Unmediated Voltammetry of Cytochrome C
at a Promoter Modified Electrode 4 4
Voltammetry of Ferritin at the Bare Gold
Electrode, No Promoter 4 6
Various Promoters That Did Not Facilitate
a Reversible Electron Transfer 4 9
Unmediated Electron Transfer Using an
Adsorbed Layer of 3-Mercapto-l-Propane
Sulfonic Acid 5 5
Other Promoting Systems that Catalyze the
Electron Transfer of Ferritin 6 1
The Voltammetry of Commercially Supplied
Apoferritin 7 0
The Voltammetry of Ferritin with the Iron
Core Removed 7 5
CV Scans of the Reconstituted Iron Core,
both in Commercial and Laboratory
Apoferritin 7 5
4. DISCUSSION 8 3
BIBLIOGRAPHY 8 8
vi


FIGURES
Figures
1.1 Fenton reaction 3
1.2 Stem loop stucture 6
1.3 Ferritin's primary sequence 1 1
1.4 Ribbon diagram of apoferritin subunit 12
1.5 24 subunit ferritin protein shell 14
1.6 2 subunit symmetry 15
1.7 The redox reaction of hydroquinone 2 0
1.8 CV scan of hydroquinone 2 1
1.9 Reduction of cytochrome c using methyl viologen 2 4
1.10 Cytochrome c orientation with the gold electrode 2 7
2.1 Electrochemical cell 3 1
2.2 Glass sheathed electrode 3 3
2.3 CV scan of clean gold electrode 3 5
2.4 CV scan of adsorbed iodine 3 7
2.5 Calibration curve for the spectrophotometric assay 4 2
2.6 Calibration curve for the atomic absorption assay 4 3
3.1 CV scan of cytochrome c 4 5
vii


3.2 CV scan of Fe(II)S04
47
3.3 CV scan of ferritin at bare gold electrode 4 8
3.4 CV scan of 2-mercaptoethylamine adsorbed layer 5 0
3.5 CV scan of ferritin promoted with 2-
mercaptoethylamine 51
3.6 CV scan of octanethiol adsorbed layer 5 3
3.7 CV scan of ferritin promoted with octanethiol 5 4
3.8 CV scan of adsorbed 3-mercapto-l-propane sulfonic
acid 5 6
3.9 CV scan of ferritin promoted with 3-mercapto-l-propane
sulfonic acid 5 7
3.10 CV scan of adsorbed 3-mercapto-l-propane sulfonic
acid 5 9
3.11 CV scan of ferritin promoted with 3-mercapto-l-propane
sulfonic acid 6 0
3.12 CV scan of adsorbed 2-mercaptoethane sulfonic
acid 6 2
3.13 CV scan of ferritin promoted with 2-mercaptoethane
sulfonic acid 6 3
3.14 CV scan of adsorbed 3-mercaptopropionic acid 6 5
3.15 CV scan of ferritin promoted with 3-mercapto propionic
acid 6 6
Vlll


3.16 CV scan of ferritin promoted with 3-mercapto propionic
acid 6 7
3.17 CV scan of ferritin promoted with 3-mercapto propionic
acid 6 8
3.18 CV scan of adsorbed 3-mercapto-1-propane sulfonic
acid 7 1
3.19 CV scan of apoferritin promoted with 3-mercapto-1-
propane sulfonic acid 7 2
3.20 CV scan of adsorbed 3-mercapto propionic acid 7 3
3.21 CV scan of commercial apoferritin promoted with 3-
mercapto propionic acid 7 4
3.22 CV scan of laboratory apoferritin promoted with 3-
mercapto propionic acid 7 6
3.23 CV scan of adsorbed 3-mercapto propionic acid 7 8
3.24 CV scan of the reconstituted iron core in commercial
apoferritin promoted with 3-mercapto propionic
acid 7 9
3.25 CV scan of reconstituted iron core in laboratory
apoferritin promoted with 3-mercapto propionic
acid 8 0
3.26 CV scan of filtered ferritin promoted with 3-mercapto
propionic acid 8 1
3.27 CV scan of adsorbed 3-mercapto propionic acid 8 2
IX


CHAPTER 1
INTRODUCTION
Cellular Ferritin: Structure and Function
Cellular ferritin is a protein that is responsible for the
storage of iron. This protein is found in most cell types of
humans and other vertebrates, invertebrates, plants, fungi, and
bacteria (Crichton et al., 1982). Cellular ferritin is a large
protein, 12.0 nm in diameter, formed from a spherical protein
coat (apoferritin). The protein shell is 1.0 nm thick with a
mass of 450,000 Da ( Harrison et al., 1967). Ferritin contains
twenty four subunits composed of varying ratios of two
distinct subunits, H(eavy) which has a molecular mass of
21,099 Da (182 residues) and L(ight) which has a mass of
19,766 Da (174 residues) (Theil, 1987). In mammals, H and L
subunits are encoded by separate genes on chromosomes 11 and
19 respectively (Caskey et al., 1983; Worwood et al.,1985).
Ferritin's twenty four subunits are associated in 4:3:2
symmetry resulting in eight hydrophilic and six hydrophobic
channels which link the exterior with the interior of the
molecule (Theil, 1987).
1


Iron, although essential for life, can be very toxic as an
oxidizing and reducing agent. This is illustrated in the Fenton
reaction where the super oxide radical (O2), a normal product
of metabolism, is converted to hydrogen peroxide (H2O2) by an
enzyme called super oxide dismutase (SOD). Hydrogen peroxide
is normally converted to water by either catalase or
glutathione peroxidase (G.P.). But it can also be responsible for
the oxidation of iron, from the ferrous (Fe2+) to ferric (Fe3+)
form (Fig. 1.1). In the process a highly toxic peroxide radical
(OH) is generated which can readily oxidize membrane lipids,
altering membrane permeability. Ferritin is crucial for
sequestering iron and maintaining it in a form that seems to be
non toxic.
Cellular ferritin has a storage capacity of up to 4500
atoms of iron (III) per molecule (Clegg et al., 1980) as a
crystalline core of hydrous ferric oxide (FeO(OH)) with variable
amounts of attached phosphate (Ford et al., 1984; Treffry et al.,
1987). In humans 25% of the total body iron is contained in
ferritin, which is found predominantly in the liver, spleen, and
the bone marrow (Gutteridge et al., 1983).
Cellular ferritin has also been shown to have the ability
to bind other toxic metal ions. Beryllium binds to nucleic
acids, proteins and inhibits a number of enzymes. J. G. Joshi
2


02
Ft
Fig. 1.1 Fenton reaction.
3


was investigating the inhibitory effects of Be (II) on
phosphoglucomutase (PGM) when he noticed that in vivo ferritin
was able to bind and detoxify up to 1000 atoms of Be (Joshi et
al., 1985). Cellular ferritin can also bind Cu 2+, Zn 2+, Cd 2+, Tb
3+, U02 2+, Cr3+, or VO 2+(Price & Joshi, 1983; Treffry &
Harrison, 1982). The binding of various metal ions and the cell
specific variations in the structure of apoferritin emphasizes
the physiological importance of understanding in more detail
the formation of ferritin.
It has also been shown that iron and cellular ferritin
play a role in the pathogenesis of Parkinson's disease. In
Parkinson's, iron accumulates in the substantia nigra in
association with a decrease in ferritin concentrations,
paralleled with an increase in lipid peroxidation (Dexter et
al.,1990; Dexter et al., 1991). A comparison of the
concentrations of ferritin and iron in the parietal cortex from
normal and Alzheimer's patients shows a 38.2% increase in
ferritin concentrations and a 44.8% increase in the non-heam
iron of Alzheimer's disease patients (Dedman et al., 1992)
However, the increase in cellular ferritin concentrations and in
non-heme iron may reflect a pathological response rather than
a primary cause for the degeneration. A greater understanding
of the intricate iron balance that is held within the body and
4


the specific role that ferritin plays is essential to the
understanding of many disease states.
Regulation of Cellular Ferritin
The production of ferritin and the degradation of
transferrin are controlled at the level of translation. At low
iron concentrations a binding protein identified as aconitase
(Klausner et al., 1993), binds to a stem-loop structure known
as an iron responsive element (IRE), located in the 5'
untranslated region (UTR) of ferritin mRNA. Aconitase, bound
to the 5' IRE, inhibits the translation of ferritin mRNA; the
exact mechanism has not been established. There are also
several iron responsive elements at the 3' end of transferrin
receptor mRNA (Fig. 1.2). When aconitase binds to the
transferrin receptor's IRE at low iron concentrations,
endonuclease degradation of the mRNA is inhibited (Klausner et
al., 1993). This results in the continuous production of
transferrin receptors in order to import more iron to the cell.
In contrast, at high iron concentrations aconitase releases
from the respective mRNA sequences. At this point the
translation of ferritin and the degradation of transferrin mRNA
may begin. The reduced production of transferrin receptors
will result in a reduced amount of iron imported into the cell.
5


FERRITIN mRNA
ACONITASE
AOS

AAAAAA
AUG
IRE occupied by ACONITASE
Inhibiting translation Initiation.
ACONITASE
One or more IREs occupied by ACONITASE
protoctlng mRNA from rnle-determlning
stop In mRNA degradation.
Fig. 1.2 Schematic representation of the stem loop structure
occurring at the 5' end of ferritin mRNA and the 3' end of
transferrin receptor mRNA (adapted from Klausner et al., 1993).
6


In conjunction, ferritin concentrations will increase and thus
more iron will be sequestered.
Serum Ferritin: How Does It Differ from Cellular Ferritin
In 1982 Cragg et al. demonstrated the existence of a
unique type of ferritin subunit, glycosylated serum ferritin
(Cragg et al., 1982). In 1991, Rosenzweig and coworkers
showed that the heart functions as an endocrine gland; atrial
natriuretic factor (ANF) is synthesized, glycosylated, and
secreted by the heart (Rosenzweig, 1991). It was postulated
that the tissue of origin for serum ferritin was also the heart
(Campbell et al., 1993). To this end glycosylated heart ferritin
was recently characterized and compared to cellular ferritin
and serum ferritin. It was shown to be smaller in diameter (3-
5 nm) than cellular ferritin and it contained different size
subunits (66, 60.5, 53.5, 43.5, and 29.5 kDa) (Campbell et al.,
1993). Antisera raised against serum ferritin was shown to
cross-react with glycosylated heart ferritin but did not
significantly react with cellular ferritin (Campbell et al.,
1993). This work strongly supports the postulation that
glycosylated heart ferritin and serum ferritin are one in the
same.
Iron and serum ferritin are implicated as risk factors
in many epidemiological studies. A study by J. T. Salonen in
7


Finland tested the hypothesis that high serum ferritin
concentrations and high dietary intake of iron are associated
with an elevated risk of acute myocardial infarction. Their
results show that high stored iron levels, as tested by serum
ferritin concentration, is a risk factor for coronary heart
disease (Salonen et al., 1992). This study raises the question
of whether the increase in serum ferritin concentration is a
cause or an effect in response to the increase in iron or can it
be implicated as cause of coronary heart disease. It is
postulated that the iron associated with serum ferritin may be
able to oxidize low density lipoproteins (LDL) resulting in
further proliferation of atherosclerotic plaque.
The Various Types of Cellular Ferritin
Comparisons have been made between mammalian
ferritin and ferritin contained in the hemolymph of several
species of marine invertebrates which actively deposit iron
biominerals. Major points of difference regarding marine
invertebrate ferritin includes: the iron cores are larger but less
crystalline, these ferritin's take up Fe 2+ more rapidly and they
are immunologically distinct from mammalian ferritin (Webb et
al., 1985). The available protein sequence for E. coli ferritin
shows no homology with horse spleen ferritin; the structural
8


similarities may be a result of convergent evolution (Theil,
1987). Later work by Andrews et al. shows that known
sequences from invertebrates, bacteria and plants maintain a
closer homology with mammalian H chain than to L chain
(Andrews et al., 1992). Azotobacteria vinelandii (AVBF)
ferritin contains one protoporphyrin IX heme group per pair of
Subunits (12 hemes per molecule) where mammalian ferritin
does not (Watt et al., 1992). AVBF also has a more structurally
disordered iron core with phosphate/iron ratios near 1, in
contrast to native mammalian ferritin which maintains a ratio
closer to 0.1 (Watt et al.,1986; Frankel et al., 1987; Roher et
al.,1990).
There are also a number of distinctions among
mammalian ferritins. All of the cellular ferritins are
composed of virtually the same two subunits, L and H chains,
but their ratios vary depending upon the task at hand. A 55%
amino acid sequence similarity is seen when comparing L and H
subunits, but comparing members of L and H subunit types from
different mammals (e.g.. humans, rats, and mice) show 85% and
90% homology, respectively (Andrews et al., 1992; Bauminger
et al., 1993). Ferritin isolated from the cerebral cortex and
cerebellum has a H subunit content of approximately 65% and
60% respectively (Dedman et al., 1992) as does the heart. In
g


contrast ferritin high in L subunits is typical of the spleen and
liver (Arosio et al., 1978). The liver and spleen are known as
major iron storage tissues and Cozzi suggest that ferritins
high in L subunit content are associated with long term iron
storage whereas H rich ferritins are associated with iron
detoxification (Cozzi et al., 1990).
Horse Spleen Ferritin Structure
Horse spleen ferritin is one of the most characterized
ferritins and its primary sequence is often used to determine
homology with other mammalian apoferritins. It was chosen
for this study because essentially all (>90%) of the 24 subunits
are the L chain (Theil et al., 1985). Horse spleen apoferritin's L
subunit primary sequence is given in Figure 1.3.
Each subunit of horse spleen apoferritin is composed of
four closely packed helices A, B, C, and D with a shorter helix E
lying at an acute angle to the bundle axis (Fig. 1.4). The study
of horse spleen ferritin at a 2.8 A resolution shows that 129 of
the 174 subunit residues are maintained in these five helices
(Heusterspreute & Crichton, 1981). Helix A (residues 10-39), B
(45-72), C (92-120) and D (124-155) are closely packed due to
strong hydrophobic interactions. There is a long loop L


5 10 15
NAC-SER-SER-GLN-ILE-ARG-GLN-ASN-TYR-SER-THR-GLU-VAL-GLU-ALA-ALA-
20 25 30
VAL-ASN-ARG-LEU-VAL-ASN-LEU-TYR-LEU-ARG-ALA-SER-TYR-THR-TYR-
35 40 45
LEU-SER-LEU-GLY-PHE-TYR-PHE-ASP-ARG-ASP-ASP-VAL-ALA-LEU-GLU-
50 55 60
GLY-VAL-CYS-HIS-PHE-PHE-ARG-GLU-LEU-ALA-GLU-GLU-LYS-ARG-GLU-
65 70 75
GLY-ALA-GLU-ARG-LEU-LEU-LYS-MET-GLN-ASN-GLN-ARG-GLY-GLY-ARG-
80 85 90
ALA-LEU-PHE-GLN-ASP-LEU-GLN-LYS-PRO-SER-GLN-ASP-GLU-TRP-GLY-
95 100 105
THR-THR-LEU-ASP-ALA-MET-LYS-ALA-ALA-ILE-VAL-LEU-GLU-LYS-SER-
110 115 120
LEU-ASN-GLN-ALA-LEU-LEU-ASP-LEU-HIS-ALA-LEU-GLY-SER-ALA-GLN-
125 130 135
ALA-ASP-PRO-HIS-LEU-CYS-ASP-PHE-LEU-GLU-SER-HIS-PHE-LEU-ASP-
140 145 150
GLU-GLU-VAL-LYS-LEU-ILE-LYS-LYS-MET-GLY-ASP-HIS-LEU-THR-ASN-
155 160 165
ILE-GLN-ARG-LEU-VAL-GLY-SER-GLN-ALA-GLY-LEU-GLY-GLU-TYR-LEU-
170
PHE-GLU-ARG-LEU-THR-LEU-LYS-HIS-ASP-
Fig. 1.3 Primary sequence for the L subunit of horse spleen
ferritin (adapted from Crichton et al., 1982).
11


Position of 4 Subunit Symmetry
Fig. 1.4: Ribbon diagram of the a-carbon backbone of an
apoferritin subunit (adapted from Ford et al., 1984).


(residues 73-91) that connects the C terminus of helix B with
the N terminus of helix C (Ford et al.,1984). Figure 1.4 depicts
a cross section through the protein shell with helices B and D
facing the interior and helices A and C as well as loop L face
towards the exterior of the shell (Ford et al., 1984).
The twenty four subunits pack together in 4 3 2
symmetry as viewed down the four-fold molecular axis (Fig.
1.5). Each subunit is represented by the sausage-shaped
building block with N terminus labeled and the position of the E
helix is labeled accordingly. Two subunits interact along loop L
and L1 and helix A and A1 to form a dimmer, this results in a
grove 1,4 micrometers wide between helix B and B' (Fig. 1.6)
(Ford et al., 1984; White et al., 1983). Harrison has identified a
large number of hydrogen bonds and salt bridges within each
subunit and between subunits that may account for ferritin's
surprizing thermostability (Harrison et al., 1985). The
formation of eight hydrophilic channels per ferritin molecule is
due to interactions of three subunits near the N terminus of
each subunit (Fig. 1.5). The hydrophilic channels have been
detected to have metal binding sites (Harrison et al., 1985) and
highly conserved residues, Asp 127 and Glu 130, positioned in
the channels. This suggests the passage of iron in and out of
the core through these channels (Theil, 1987). The interaction


8 Hydrophilic Channels
6 Hydrophobic Channels
Fig. 1.5: Schematic representation of the 24 subunit ferritin
protein shell (adapted from Ford et al., 1984).
1 4


Exterior
Hydrophobic
Interactions
Fig. 1.6: Schematic representation of the symmetry of two
subunits, showing grove (adapted from Ford et a|., 1984).


of four subunits at the E helix (Fig. 1.5) forms six hydrophobic
channels lined with residues such as Leu 165 (Theil, 1987). It
has been postulated that these six hydrophobic channels serve
as a passage way for oxygen and phosphate (Harrison et al.,
1985).
Proposed Mechanisms of Ferrous Incorporation
into the Iron Core
The mechanism for iron uptake and nucleation has been a
major focus of ferritin research, especially for Pauline M.
Harrison and her group. In 1972 the uptake of iron was
recorded spectrophotometrically, starting with apoferritin, or
ferritin with a low iron content, free Fe2+ and an oxidizing
agent. Iron was oxidized at a greater rate in the presence of
ferritin than free iron alone, thus ferritin can be considered an
enzyme (Macara et al., 1972). It was also postulated that iron
incorporation occurs in two stages: nucleation, in which iron
binds to sites on the protein; followed by a faster crystalline
growth stage. In 1973 the catalytic binding site was further
illuminated by inhibition studies with zinc, the binding site
was presumed to be internal and to involve histine (Macara et
al., 1973). Further work produced the following model: Fe2+
enters apoferritin through inter-subunit channels, attaches to
the binding sites and is oxidized to Fe3+. Crystallites of


hydrous ferric oxide begin to grow with in the protein shell,
further additions of Fe2+ can be deposited and oxidized directly
on the surface of these crystals. The rate of core growth is
now dependent on the surface area of hydrous ferric oxide
crystallite (Hoy et al., 1974).
The discovery and further classification of a binding and
oxidation site on the H chain known as the ferroxidase center
has answered many questions (Lawson et al., 1989; Bauminger
et al., 1991 Treffry et al., 1992). Seven highly conserved
residues on the H chain that are thought to be responsible for
the binding and the formation of Fe3+-oxo-bridged dimmer
include Glu-27, Tyr-34, Glu-61, Glu-62, His-65, Glu-107 and
Gln-141 (Bauminger et al., 1993). The Fe3+ dimmer formation
could be the product of the oxidation of two Fe2+ atoms by one
oxygen molecule, catalyzed by the ferroxidase activity (Treffry
et al., 1992). Studies involving site-directed mutagenesis of
the H chain and recombinant horse L chain ferritin (all 24
subunits are the L chain) have concluded that significant
oxidation of Fe2+ occurs predominantly at the H chain
ferroxidase centers (Treffry et al., 1993). This is further
supported by the comparison of horse spleen ferritin, which
contains 1-4 H chains per protein, with recombinant horse L


chain ferritin. The rate of Fe2+ oxidation is much higher for
the horse spleen (Treffry et al., 1993).
The proposal that H and L chains function in a
cooperative manner (Levi et al., 1992) seems to be well
supported by mammalian physiology. The liver and the spleen,
known as iron storage organs, posses L-chain-rich ferritins
often containing large ferrihydrite cores (Mann et al., 1986).
The L chain appears to be more stable, due in part to the L-
chain-specific salt bridge (Santambrogio et al., 1992), and it is
proposed that L chains have a greater ability for ferrihydrite
nucleation (Wade et al., 1991). This could be due to differences
in distribution of potential metal binding amino acids (e.g.
carboxy groups) (Levi et al., 1992). The presence of ferritin
high in H subunit content, in the heart (Arosio et al., 1978) and
in the brain (Fleming & Joshi, 1987), could be significant if, in
this case, ferritin's role is the detoxification of iron rather
than its storage.
Cyclic Voltammetry of Proteins
For the purpose of discussing the theory behind the
electron transfer of proteins at a solid electrode, hydroquinone
(HQ) will be used to demonstrate an electrochemically
reversible system. Hydroquinone, also called 1,4 dihydroxy


benzene the reduced form is the reduced form of this redox
couple. The oxidized form is called benzoquinone (BQ). The
reduction involves the removal of two electrons from aromatic
ring (Fig. 1.7).
The Nernst equation relates the electrode potential (E)
to analyte concentration. The symbol E0' is the formal
potential, R is the molar gas constant (R = 8.31441 J/mol K),
the temperature (T), the number of electrons being transferred
in reduction/oxidation reaction (n), Faraday's constant (F =
9.64846 x 104 C/Equiv.), and the bulk concentration of the
oxidized species (Co*) and the reduced species (Cr*) (Eqn. 1.1).
E = E' + RT/nF In C0*/CR* (1.1)
Upon examination of this equation, it can be seen that
the electrode potential (E) will change throughout the reaction
as the bulk concentrations of the reduced and oxidized species
change. Cyclic voltammetry involves the continuous variation
of externally applied potential and measuring the current
response. In this way, one can study the ratio of electron
transfer at an electrode. The cyclic voltammetric curve of
hydroquinone is shown in Figure 1.8.


BQ HQ
Fig. 1.7 The redox reaction of hydroquinone, an
electrochemically reversible system.
20


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 1.8: Cyclic voltammetry scan of 0.1 mM Hydroquinone in 1M
H2S04. Scan rate, 10 mV/s; electrode, platinum; electrode area,
10.0 cm2.
21


It can be seen, that after the applied potential reaches
0.35 V the current rises rapidly until it reaches a peak (Fig.
1.8). This rapid rise is due to the increasing rate at which HQ
is oxidized to BQ as the potential is scanned anodically, and for
the reduction of BQ to HQ as the potential scans negative. The
rate of the redox reaction reaches a maximum at 0.48 V. At
this potential the rate of the reaction is controlled by the rate
at which hydroquinone diffuses to the electrode surface. The
current-potential relationship is mathematically described by
the following equation:
i = nFAk0[Co(0>t)e''aTlf(E-Eo,) CR(0,t)e(1'a)Tlf(E'Ec,,)] (1.2)
For an electrochemically reversible system, that is, a system
which exhibits fast electron-transfer kinetics, this equation
reduces to the Nernst equation.
The peak current is a function of the number of
electrons transferred during the redox reaction (n), the area of
the electrode (A), the diffusion characteristics of a particular
species in a particular solvent, known as the diffusion
coefficient (D0), the rate at which the potential is scanned (v)
22


and the bulk concentration of the reactive species (Co*) and
(CR). This is represented by the following equation:
i = (2.69 x 105) n3/2 A D01/2 v"2 C0* (1.3)
Historically, the electron transfer of proteins has been
studied with the use of mediators such as methyl viologen (Fig.
1.9). Poising the potential at the reduction potential the
mediator is reduced at the electrode surface and in turn will
reduce the protein in solution. The mediator will return to the
electrode surface to again be reduced and continue the cycle.
This method of mediated coulometry can only yield the value of
n, it can tell us little about the electron transfer kinetics.
Electrochemical techniques have been used to study
redox proteins and the environment surrounding the
electroactive centers. The use of a bare metal electrode to
study redox proteins, however poses some difficulties: (1) The
redox centers are often contained in groves within the globular
mass of the proteins; thus, even if the protein is in contact
with the electrode, the redox center may not be in a suitable
orientation permitting electron transfer to occur at
appreciable rates. (2) Proteins absorb strongly and irreversibly
to most metal electrodes (such as gold), forming an insulating
23


ELECTRODE SURFACE
Fig. 1.9 Reduction of cytochrome c using methyl viologen (MV)
as a mediator.


layer of denatured proteins that "poison" the electrode
(Hitchens, 1989). Promoters bound to the surface of the
electrode are used to enhance the kinetics necessary for a
reversible redox reaction. A promoter can draw the reactive
species (redox center) close to the electrode in a suitable
orientation for electron transfer to occur (Hitchens, 1989).
Ideally the reacted protein is quickly released so that
unreacted proteins may diffuse to the electrode and continue
the cycle. The manner in which the promoter brings about a
change in the adsorption behavior of the protein is not clear.
The chemical characteristics of this surface bound promoter
may illuminate the properties of the protein, specifically the
region surrounding the redox center.
Cytochrome c has been studied extensively using cyclic
voltammetry at a gold electrode. Gold is used for the electrode
material because platinum is too reactive a surface, even with
a promoter, and may extensively denature the protein.
Taniguchi and coworkers used 4,4' dithiodipyridine, that had
been adsorbed to the electrode surface, to produce an almost
reversible redox wave for cytochrome c (Taniguchi et al.,
1982). The 4,4' dithiodipyridine is able to form strong bonds to
the electrode surface, via the sulfur atom, leaving the pyridyl
ring pendent, interacting with cytochrome c. The heme group of
25


cytochrome c is exposed on one side of the protein and it is
surrounded by many positively charged lysine residues (NH3+)
(Ferguson-Miller et al., 1979). It is suggested that the weakly
basic nitrogen lone pair is able to interact electrostatically
with the positively charged lysine groups (Hitchens, 1989). As
a result, the heme group is brought in close proximity to the
electrode in an orientation suitable for electron transfer to
occur (Fig. 1.10). A layer of 1,2 bis (4-pyridyl) ethylene had
also been adsorbed onto the electrode surface but the current
of the resulting scan was decreased by 20% (Taniguchi et al.,
1982), suggesting that 4,4' dithiodipyridine was more
effective.
Research Objectives
Ferritin is not a redox protein but the iron in the core
has been shown to be electroactive in mediated coulometric
experiments (Watt et al., 1985). Exploiting the electroactivity
of the iron core using cyclic voltammetry can yield information
as to the stability of the iron core as well as the nature of the
protein surface. As with most metalloproteins, an electron
transfer promoter had to be found. By varying the end group of
an adsorbed short chain alkyl thiol, we attempted to vary the
26


Fig. 1.10 Schematic representation of the orientation of
cytochrome c to the gold electrode surface (adapted from
Taniguchi et al., 1982).
27


properties of the adsorbed layer (e.g. acidic, basic, non-polar,
polar, positively charged, and negatively charged).
The primary goal in this work is to find an adsorbate
which behaves as a suitable promoter. By metting this
research objective, it is expected that the nature of the protein
surface can begin to be characterized depending upon how the
protein responds to the adsorbed layer. The redox potentials
will be used to learn more about the stability of the chelated
iron core, and the current-potential behavior should lead to a
better understanding of the electron transfer kinetics of
ferritin. The experimental approaches to these ends and
results will be discussed.
28


CHAPTER 2
EXPERIMENTAL METHODS
Suppliers of All Chemicals Used
Electrochemical methods were used to accomplish the
purposes stated in the previous section. The chemicals,
equipment and procedures used are described below.
The following compounds were purchased from Sigma
Chemical Company (St. Louis, MO): 3-mercapto propionic acid
(HOOCCH2CH2SH); apoferritin: from horse spleen; cytochrome C:
from horse heart; thioglycolic acid; ferritin: type I: from horse
spleen, used without further purification.
Ammonium iron (II) sulfate hexahydrate (NH^
Fe(S04)2'H20; imidazole buffer; potassium iodate (Kl); 3-
mercapto-1-propane sulfonic acid, sodium salt (Na+
-3OSCH2CH2CH2SH); 2-mercaptoethane sulfonic acid, sodium
salt (Na+ -3OSCH2CH2SH) were all purchased from Aldrich
Chemical Company (Milwaukee, Wl), and were used without
further purification.
Sodium phosphate (NaH2P04.H20) was purchased from J.
T. Baker Chemical Company (Phillipsburg, NJ), sodium sulfate
(anhydrous) was purchased from Mallinckrodt, Science Products
29


Division (McDaw, IL), and the Sephedex PD-10 columns were
purchased from Pharmacia Chemical Company (Hercules, CA).
Equipment
In order to prepare uniform adsorbed layers one must
start with a clean electrode surface. To maintain surface
cleanliness, there must be no measurable quantities of
surface-active species coming from the electrolyte solution or
the cell components. A chromic acid cleaning solution was
used to assure that all of the glassware used, including the
electrochemical cells, was clean. The chromic acid cleaning
solution is prepared by dissolving 92 g of sodium dichromate
(Na2Cr204-H20) is dissolved into 458 mL of water. Finally, 800
mL of sulfuric acid is added while stirring. Once the glassware
has soaked in chromic acid for fifteen minutes it is removed
from the solution and rinsed with pyrolitically triply distilled
water (PTDW) (Conway et al., 1973).
Electrochemical Cell Configuration
The electrochemical cells that are used in all cyclic
voltammetry experiments are composed of two chambers (Fig.
2.1). A side chamber that contains the reference and auxiliary
electrodes and the main chamber contains the electrochemical
30


Main Chamber
Reference/Auxillary
Chamber
1
Nitrogen Outlet
\ (oi i
\ i
Fritted Disc
Fig. 2.1 Electrochemical cell used for all cyclic voltammetric
experiments.


solution under study. The two chambers are connected by a
fritted disc which allows ions to flow between the
compartments while keeping contaminants from the side
compartment from entering the main chamber. The nitrogen
inlet allows for the deaeration of the solutions; the nitrogen
flows in the bottom of the main chamber and can escape
through the outlet at a position higher than the solution
surface. The main chamber is fitted for a 24/40 ground glass
stopper in order to cover solutions when not in use.
The working electrode is contained in a glass cylinder
which is 10 mm outer diameter which is fitted with its own
nitrogen inlet (Fig. 2.2). The nitrogen is controlled by a Teflon
valve, which enables the experimenter to transfer the electrode
to and from different solutions while maintaining the electrode
and any absorbed layers under a nitrogen atmosphere. This
measure prevents the oxidation of adsorbed promoters by
atmospheric oxygen.
The Combination of Electrodes Used
A gold foil electrode is used as the working electrode
for all of the studies, its surface area being 8.48 cm2. The gold
electrode consists of a flag measuring 10 mm high, 28 mm in
circumference and 0.1 mm in thickness. A gold wire, 0.1
32


Teflon Stopcock
Electrode Lead
Nitrogen Inlet
6 mm 10 mm
Fig. 2.2 Glass sheathed electrode. A flow of nitrogen is used to
purge the solution surrounding the electrode.
33


mm diameter, is attached by weaving it through three holes at
the top edge of the flag. The wire and flag were subsequently
fused using a hammer and flat-pointed awl.
The reference electrode is Ag/AgCI in 1 M KCI. The
silver wire is plated with solid silver chloride, rinsed, and
excess silver chloride is place in the bottom of the electrode
container. The reference electrode is composed of a glass
cylinder 6 mm outside diameter closed at the bottom by a glass
cup fitted with a ground glass joint. This glass joint allows
for the conduction of ions. The auxiliary electrode is a
platinum electrode that is wrapped around the glass cylinder of
the reference electrode.
The working electrode is cleaned by first subjecting the
electrode gently in a natural gas/air flame, then rinsing with
PTDW. The electrode apparatus is then assembled and placed in
an H cell containing 1 M sulfuric acid, where it is cleaned by
cycling the potential between the anodic (1.500 V) and the
cathodic (-0.350 V) limits of the solution (Rodriguez et al.,
1987). Repeating this procedure for seven cycles assures that
all absorbed species will be removed from the surface, of the
electrode. Surface cleanliness is verified by a cyclic
voltammetric scan in the same 1M H2SO4 solution (Fig. 2.3).
The surface area of the electrode is determined by
34


CURRENT (|iA)
Fig. 2.3 Cyclic voltammetric scan of a clean polycrystalline gold electrode in 1M H2SO4
electrode area, 8.48 cm2.


measuring the anodic charge of an absorbed layer of iodine
(Rodriguez et al., 1987).
I (ads) + 3H2O I03 (aq) + 6H+ (gq) +5e_ (2.1)
A clean electrode is placed in a deaerated 1mM
potassium iodide solution at open potential for five minutes.
The electrode is then transferred under a nitrogen atmosphere
to 1M H2S04, purging the solution surrounding the electrode
fifteen times, then scanning a cyclic voltammogram of the
adsorbed layer (Fig. 2.4). The oxidation peak of the absorbed
iodine (peak 1) is cut and weighed. The weight of the back
ground (surface oxide) is subtracted and the area is calculated
using the following relationship:
A = (Q Qb) ox i /5 F Ti Caic IT caic = 104 nmol/cm2
General Voltammetric Procedure for Proteins
Cytochrome c is investigated using 4,4 dithiodipyridine
as a promoter (Taniguchi et al., 1982). All solutions are
constantly deaerated with nitrogen. A clean electrode is
immersed in a 1mM solution of 4,4 dithiodipyridine at open
potential for five minutes. The electrode is removed from the
36


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 2.4 Cyclic voltammetric scan of an adsorbed layer of iodine at a polycrystalline gold
electrode; electrode area, 8.48 cm2. Peak 1 is the current resulting from the oxidation of
adsorbed iodine. Peak 2 is the reduction of dissolved iodate.


promoter solution under a nitrogen atmosphere and the outside
of the electrode cylinder is rinsed with PTDW. The electrode is
then transferred to 0.05 M phosphate buffer (pH=7) with 0.5 M
Na2S04 which is added to provide the necessary conductivity.
In this solution the electrode cavity is purged fifteen times to
remove any promoter not bound to the surface. A voltammetric
scan is performed in the buffer solution to obtain a background
scan of the promoter. The electrode is then cleaned in 1 M
H2S04 and a new promoter layer is adsorbed as noted above. The
outside of the electrode cylinder is rinsed, and then the excess
promoter is rinsed into the buffer solution. The electrode is
then transferred to the cytochrome c solution and purged ten
times to assure that a known concentration of protein is
surrounding the electrode.
Solutions of ferritin and the promoters used to study
ferritin are prepared in the same manner. Ferritin solutions
are prepared at 4 mg/mL in 0.05 M phosphate buffer with 0.5 M
Na2S04. The promoters used include: 3-mercapto propionic
acid, 3-mercapto-1 -propane sulfonic acid, 2-mercaptoethane
sulfonic acid, 2-mercaptoethyl amine, and 4,4 Dithiodipyridine.
All of the promoters are 1 mM in concentration dissolved in
0.05 M phosphate buffer with 0.5 M Na2SC>4. The de-aeration
time for ferritin is only five minutes due to the viscosity of
38


the solution and that nitrogen bubbles carry the ferritin out of
the cell.
All electrodes (working, reference and auxiliary
electrodes) are connected to a potentiastat which will scan the
preset potential range. The resulting data is recorded using an
X/Y recorder.
Removing Iron from the Protein Shell
The iron core is removed, resulting in apoferritin, by the
addition of 1% thioglycolic acid to 0.05 M sodium phosphate
buffer (pH=7). This solution is maintained at 5 C for five
minutes and then filtered through a Sephadex G25 PD-10
column equilibrated with phosphate buffer (adapted from Levi
et al., 1992). This filter will separate large molecules, e.g..
ferritin, from electrolytes (e.g. free iron). The addition of 1%
thioglycolic acid was repeated two more times followed by a
solution of 0.1% thioglycolic acid, the solution is filtered
through the PD-10 column after each addition. The solution is
filtered through PD-10 column once again to assure that all of
the free iron is removed. Sodium sulfate (0.5 M) is dissolved in
the resulting apoferritin solution before any voltammetric
scans are performed.
39


Reconstituting the Iron Core in Apoferritin
The incorporation of iron into apoferritin is achieved by
preparing 500 mL of 1 mg/mL apoferritin in 100 mM imidazole
buffer (pH=7). Potassium iodate (0.05 M) is added as an
oxidizing agent and sodium thiosulfate (0.2 M) is also added
(Macara et al., 1972). Iron incorporation begins with the
addition of ferrous ammonium sulfate (0.1 M). The solution is
allowed to stand overnight at 4 C to assure maximum iron
incorporation. The solution is then filtered through a Sephadex
PD-10 column to remove any iron not contained in the iron core,
e.g. free iron and iron bound to the exterior of the protein shell.
The solution is concentrated to approximately 4 mg/mL by
using a centriprep cylinder.
The concentration of the ferritin protein is calculated
using a Bio-Rad Protein Assay. The Bio-Rad dye binds to
protein in solution to form a complex which absorbs at 595 nm.
Bovine serum albumin is used to establish a calibration curve.
Two apoferritin solutions were investigated, apoferritin
obtained commercially (Sigma apo) and that which was
prepared in the laboratory (prep apo); as well as ferritin which
was filtered through the PD-10 columns as a control (Sigma
Ft). The protein samples are diluted by a factor of eight and


the absorbance measured (Fig. 2.5). The resulting concentration
of the stock proteins solutions are as follows: sigma ferritin
(filtered) is 2.761 mg/mL, reconstituted apoferritin from
sigma is 1.511 mg/mL and reconstituted prepared apoferritin is
1.291 mg/mL. The iron content of each of the three ferritin
solutions; Sigma Ft, Sigma apo and prep apo, is determined
using a Varian AA-575 atomic absorption spectrophotometer.
The spectrophotometer is set at 249.3 nm and the flame is set
at a 63%/37% mixture of compressed air and acetylene,
respectively. Ammonium iron (II) sulfate is used to develop a
calibration curve and each ferritin solution is diluted by a
factor of 250. The resulting iron concentrations are 2168 Fe3+
atoms/ Ft molecule for sigma Ft, 1626 Fe3+ atoms/ Ft molecule
for reconstituted sigma apo and 2112 Fe3+ atoms/ Ft molecule
for the reconstituted lab apo (Fig. 2.6). Sodium sulfate (0.5 M)
is dissolved in the resulting ferritin solution before any
voltammetric scans are performed.
41


Ft BIORAD ASSAY w/ UNKNOWNS
Fig. 2.5 Calibration curve for the spectrophotometric assay of
protein. The solid diamonds represent the absorbance of the
reconstituted commercial apoferritin, the hollow squares
represent the absorbance of the commercial ferritin sample,
and the hollow diamonds represent the absorbance of the
prepared apoferritin.
42


IRON CALIBRATION CURVE w/ UNKNOWNS
Fig. 2.6 The calibration curve established with
(NH4)2Fe(ll)(S04)2- The solid diamonds represent the
absorbance of the iron content of commercial ferritin. The
hollow squares represent the iron content of the reconstituted
commercial apoferritin and the hollow diamonds represent
reconstituted apoferritin prepared in the laboratory.
43


CHAPTER 3
EXPERIMENTAL RESULTS
Unmediated Voltammetry of Cytochrome C at
a Promoter Modified Electrode
The cytochrome c results obtained by Taniguchi and
coworkers are reproduced as a check to see whether the
apparatus and the technique being used to modify the electrode
surface gives the same result (Taniguchi et al., 1982). The
promoter, 4,4' dithiodipyridine, and the methodology used for
forming the layer on the electrode surface is discussed in the
methods section. The resulting scan (Fig. 3.1) has a formal
potential (E0'), which is the midpoint measured between the
cathodic and anodic peak potential, is 0.28 V vs. Ag/AgCI
reference electrode. The peak separation (AEp) is 78 mV which
is close to the value stated by Taniguchi (70 mV) (Taniguchi et
al., 1982). This is also close to the theoretical value accepted
as the peak separation (60 mV) for an electrochemically
reversible one electron transfer (Bard & Faulkner, 1980). In
addition the peak currents of the reversible wave are similar.
This peak separation is cross-referenced with the peak
44


Fig. 3.1 Cyclic voltammetric scan of cytochrome c, 0.25 M in
0.05 M NaH2P04 (pH = 7) with 0.5 M Na2S04- Adsorbed layer,
4,4 Dithiodipyridine, from 1mM; electrode, polycrystalline gold;
electrode area, 8.48 cm^; scan rate, 5.0 mV/s.
45


separation of an iron electrolyte solution (1.0 mM Fe(ll)(CN)6-4)
which is known to be a reversible system (Fig. 3.2). The peak
currents of the cyclic voltammogram of free iron also appears
to be similar. The peak separation of the iron solution is 194
mV which indicates some cell resistance resulting in slow
electron transfer kinetics. The uncompensated resistance (iRu)
is classified as the potential drop existing between the
reference probe and the working electrode; this results in an
error in the accuracy of potential control (Kissinger &
Heineman, 1984). Methods for correcting the iRu are rarely
significant in reactions which generate or draw small currents.
It appears that the promoting system used to study
cytochrome c catalyzes the electron transfer at the heme
center resulting in a peak separation closely resembling an
electrochemically reversible system. Thus, after comparing
the above data it is concluded that the H-cell system performs
sufficiently and that the promoter/adsorption techniques are
also sufficient.
Voltammetry of Ferritin at the Bare Gold
Electrode. No Promoter
Unlike cytochrome c, ferritin exhibited an appreciable
voltammetric current using a bare gold electrode (Fig. 3.3). The
46


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.2 Cyclic voltammetric scan of (Fe(ll)(CN)6'4), 1.0 mM in
0.05 M NaH2P04 (pH = 7) with 0.5 M Na2S04. No promoter used;
electrode, polycrystalline gold; electrode area, 8.48 cm2; scan
rate, 5.0 mV/s.
47


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.3 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05
M NaH2PC>4 (pH = 7) with 0.5 M Na2S04. No promoter used;
electrode, polycrystalline gold; electrode area, 8.48 cm2; scan
rate, 5.0 mV/s.
48


cathodic current is quite pronounced, with a peak maximum at
-0.257 V. The anodic current is barely noticeable, thus a peak
potential can not be determined. The chemical irreversibility
of this system is made quite evident by qualitatively comparing
the anodic and cathodic peak currents. The cathodic peak
current is overwhelmingly large with respect to the anodic
current; a chemically reversible system will yield peak of the
same magnitude.
Various Promoters That Did Not Facilitate a Reversible
Electron Transfer
Another surface modifier, 2-Mercaptoethylamine was
tried. The absorbed layer is formed at the electrode surface
and a background voltammogram is scanned in the buffer
solution containing the sodium sulfate electrolyte (Fig. 3.4).
This back ground scan can be compared to the scan performed
with the adsorbed promoter in the ferritin solution (Fig. 3.5).
There is a notable increase in the reduction wave current in the
ferritin scan, and the oxidation wave is shifted slightly
negative compared to the oxidation wave of the promoter. The
redox peak separation in the ferritin scan is estimated at 232
mV, although the redox peaks are greatly masked by the peaks
of the adsorbed layer alone. The cathodic peak for the cyclic
49


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.4 Cyclic voltammetric scan of 2-mercaptoethylamine,
adsorbed layer, from 1mM; in 0.05 M NaH2P04 (pH = 7) with 0.5
M Na2S04. Electrode, polycrystalline gold; electrode area, 8.48
cm2; scan rate, 5.0 mV/s.
50


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.5 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05
M NaH2P04 (pH = 7) with 0.5 M Na2S04- Adsorbed layer, 2-
Mercaptoethylamine, from 1mM; electrode, polycrystalline gold;
electrode area, 8.48 cm2; scan rate, 5.0 mV/s.
51


voltammetric (CV) scan of the adsorbed layer and the cathodic
peak of the promoter in a ferritin solution occur at
approximately the same potential. The peak currents for the
adsorbed layer and ferritin can not be distinguished, thus this
adsorbed layer is not suitable as a promoter. The same anodic
peak current displays that this adsorbed layer does not
catalyze a reversible redox wave.
Octanethiol is used to investigate the effect of a highly
non-polar adsorbed layer. It can be seen that the scan of this
layer shows no voltammetric current (Fig. 3.6). The same
effect is seen when octanethiol is absorbed onto the electrode
and a scan is performed in ferritin (Fig. 3.7). These data
suggest that charged groups are lining the exterior surface of
the ferritin protein shell and that the highly non-polar
promoter does not interact favorably with the protein. This
result prompted the use of a promoter with a formal negative
charge that is exposed to the solution.
52


CURRENT (jiA)
Fig. 3.6 Cyclic voltammetric scan of octanethiol, adsorbed layer, from ImM; in 0.05 M
NaH2PC>4 (pH = 7) with 0.5 M Na2S04. Electrode, polycrystalline gold; electrode area, 8
cm2; scan rate, 5.0 mV/s.


CURRENT (pA)
/ 1- 2.96 pA
\ \- 1 1 1 1 1 1 1
-0.6 -0.4 -0.2 0 0.2 0.4 q.6
POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.7 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05 M NaH2p04 (pH = 7) with
0.5 M Na2S04. Absorbed layer, octanethiol, from ImM; electrode, polycrystalline gold;
electrode area, 8.48 cm2; scan rate, 5.0 mV/s.


Unmediated Electron Transfer Using an Adsorbed Laver of 3-
Mercapto-1-Propane Sulfonic Acid
The use of a negatively charged sulfonate group dangling
out in solution should act as a suitable modifier. A layer of 3-
mercapto-1 -propane sulfonic acid (3-M-1-PSA) is adsorbed
onto the electrode surface, as stated in the methods, and a CV
scan is performed in the buffer solution (Fig. 3.8). The
background scan shows the redox activity of the promoting
system alone, most prominent is the sharp reductive spike
which occurs at -450 mV. The scan of ferritin using an
adsorbed layer of 3-M-1-PSA still exhibits a sharp reductive
spike due to the promoting system, but reduction and oxidation
peaks for the iron core are evident (Fig. 3.9). The peak
potential separation for the redox reaction of the iron core is
157 mV which is not reversible according to theory. However,
when the potential peak separation for ferritin is compared to
the peak separation of Fe(ll)(CN)6'4, 194 mV, the redox
reactions of ferritin can at least be considered quasireversible.
The peak separation for cytochrome c, 78 mV, is much closer to
a theoretically reversible electron transfer compared to
ferritin. It is postulated that the promoting system used for
cytochrome c is more effective at catalyzing a reversible
electron transfer. The peak current ratio for ferritin can be
55


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.8 Cyclic voltammetric scan of 3-mercapto-1 -propane
sulfonic acid, adsorbed layer, from 1 mM; in 0.05 M NaH2P04
(pH = 7) with 0.5 M Na2S04. Temperature, 26.0 C; electrode,
polycrystalline gold; electrode area, 8.48 cm2; scan rate, 5.0
mV/s.
56


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.9 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05
M NaH2P04 (pH = 7) with 0.5 M Na2S04. Adsorbed layer, 3-
mercapto-1-propane sulfonic acid, from 1 mM; electrode,
polycrystalline gold; electrode area, 8.48 cm2. Temperature,
26.0 C; scan rate, 5.0 mV/s.
57



estimated by extrapolating base lines for the current (dashed
lines shown on CV scan) by subtracting the baseline current
from the respective electrolytic currents and dividing the
anodic peak current by the cathodic peak current. A truly
chemically reversible electron transfer will yield a peak
current ratio of unity; the ratio for ferritin is calculated to be
0.895. It is postulated that as a result of the formal negative
charge, the sulfonic acid group is able to draw ferritin close
enough to the electrode and position the iron core in an
orientation to facilitate a fast electron transfer. The CV scan
is reproducible and figure 3.9 is representative of many scans.
There are CV scans that are obtained at higher temperatures
which support the postulate that the iron core of ferritin can
produce quasi reversible redox waves. 3-M-1-PSA is still the
o
promoting system, now at 36 C, the background scan yields a
sharp reductive spike at -419 mV (Fig. 3.10). The
corresponding scan in ferritin, performed at 39 oC, shows an
oxidation peak almost identical to figure 3.9; but the reduction
peak is not clearly defined. The peak potential difference is
estimated to be 131 mV (Fig. 3.11).
58


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.10 Cyclic voltammetric scan of 3-mercapto-1-propane
sulfonic acid, adsorbed layer, from 1 mM; in 0.05 M NaH2P04
(pH = 7) with 0.5 M Na2SC>4. Temperature, 36.0 C; electrode,
polycrystalline gold; electrode area, 8.48 cm2; scan rate, 5.0
mV/s.
59


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.11 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05
M NaH2P04 (pH = 7) with 0.5 M Na2S04. Adsorbed layer, 3-
mercapto-1-propane sulfonic acid, from 1 mM, electrode,
polycrystalline gold; electrode area, 8.48 cm2. Temperature,
39.0 C; scan rate, 5.0 mV/s.
60


Other Promoting Systems That Catalyze the Electron Transfer
of Ferritin
2-Mercaptoethane Sulfonic Acid (2-MESA) is also used
to produce a reversible wave. This promoter involves the same
functional groups but the chain is shorter by one methylene.
The shorter chain may result in ferritin being brought closer to
the electrode surface. The adsorbed promoter layer on the
electrode surface still exhibits a sharp reductive spike that
partially masks the reduction wave of ferritin. An absorbed
layer of 2-MESA cycled in phosphate buffer at 25.8 C shows
this spike at -432 mV (Fig. 3.12). The corresponding scan in a
ferritin solution, at 26.2 C, shows a peak separation of 150
mV (Fig. 3.13). A peak current ratio can not be calculated
because a reliable base line can not be extrapolated for the
anodic peak. Essentially the length of the chain (C2 C3)
makes no difference in the catalytic properties of the layer.
The use of 3-Mercapto Propionic Acid (3-MPA) also
produces a reversible wave. This promoter is three carbons in
length but the third carbon is a carboxylic acid functional
group. In the adsorbed state, 3-MPA is expected to be 99%
deprotonated at pH=7, calculated using the pKa which is
estimated to be about five. Consequently, this promoter also
has a negatively charged end-group directed into the open
61


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.12 Cyclic voltammetric scan of 2-mercaptoethane
sulfonic acid, adsorbed layer, from 1 mM; in 0.05 M NaH2P04
(pH = 7) with 0.5 M Na2S04. Temperature, 25.8 C; electrode,
polycrystalline gold; electrode area, 8.48 cm2; scan rate, 5.0
mV/s.
62


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.13 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05
M NaH2PC>4 (pH = 7) with 0.5 M Na2S04. Adsorbed layer, 2-
mercaptoethane sulfonic acid, from 1 mM; electrode,
polycrystalline gold; electrode area, 8.48 cm2. Temperature,
26.2 C; scan rate, 5.0 mV/s.
63


solution. However, there is a broad peak associated with the
reduction and oxidation of this promoter system which, at 35.3
C, occurs at -450 mV and -388 mV, respectively (Fig. 3.14).
The corresponding CV scan in a ferritin solution, at 31.5 C,
shows a well defined anodic peak. Because the cathodic peak of
the adsorbed layer is shifted to more negative values the
cathodic peak of ferritin is much more visible (Fig. 3.15). The
peak separation is estimated to be 178 mV.
In theory, the oxidative current should decrease as the
quantity of reduced species is decreased. A way to investigate
whether the anodic and cathodic peaks are related is to
sequentially decrease the reduction limit by 100 mV. Using
adsorbed 3-MPA as the promoter, and starting with an initial
reductive potential of -400 mV and solution temperature of
26.2 C, it can be seen that the anodic peak current is virtually
unchanged (Fig. 3.16). By reducing the reduction potential by an
additional 100 mV and maintaining the temperature at 26.0 C,
it is made evident that the oxidation peak current is
significantly reduced (Fig. 3.17). These scans strongly support
the postulate that the electron transfer of ferritin using
adsorbed 3-MPA is chemically reversible.
The evidence given above shows a significant difference
between the CV scan of ferritin using a bare gold
64


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.14 Cyclic voltammetric scan of 3-mercapto propionic
acid, adsorbed layer, from 1 mM; in 0.05 M NaH2P04 (pH = 7)
with 0.5 M Na2S04. Temperature, 35.3 C; electrode,
polycrystalline gold; electrode area, 8.48 cm2; scan rate, 5.0
mV/s.
65


AEd = 178 mV
Fig. 3.15 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05
M NaH2P04 (pH = 7) with 0.5 M Na2S04. Adsorbed layer, 3-
mercapto propionic acid, from 1 mM; electrode, polycrystalline
gold; electrode area, 8.48 cm2. Temperature, 31.5 C; scan
rate, 5.0 mV/s.
66


0
<
=L
H
Z
w
Pi
Pi
p
u
----1-------1-------1
-0.4 -0.2 0
POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.16 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05
M NaH2P04 (pH = 7) with 0.5 M Na2S04. Adsorbed layer, 3-
mercapto propionic acid, from 1 mM; electrode, polycrystalline
gold; electrode area, 8.48 cm2. Temperature, 26.2 C; scan
rate, 5.0 mV/s.
67


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.17 Cyclic voltammetric scan of ferritin, 4 mg/mL in 0.05
M NaH2P04 (pH = 7) with 0.5 M Na2S04. Adsorbed layer, 3-
mercapto propionic acid, from 1 mM; electrode, polycrystalline
gold; electrode area, 8.48 cm2. Temperature, 26.0 C; scan
rate, 5.0 mV/s.
68


electrode and the CV scan using a promoter that is capable of
drawing ferritin close to the electrode. There is a formal
negative charge on the end group on all three of these
promoters. We propose that it is the negative charge that is
attracted to positive charges on the exterior of the ferritin
protein shell. This electrostatic attraction draws the iron core
close enough to the electrode surface in order for quasi
reversible electron transfer to occur. A negatively charged
electrode would not catalyze a quasireversible electron
transfer primarily because of the strong interaction of the
protein with the surface. The protein would denature and
poison the electrode preventing any further electron transfer.
The adsorbed layer prevents the denaturing of the protein and
perhaps promotes a favorable orientation of the redox center
with respect to the electrode surface. The fluidity of the
adsorbed layer could also allow the protein to rotate and
achieve the correct orientation. At a solid electrode (no
adsorbed layer), the protein will stick to the electrode surface
and rotation would be precluded. A longer alkyl thiol chain
would be more fluid at the electrode surface; future studies
will investigate the effect this might have upon the promotion.
69


The Voltammetry of Commercially Supplied Apoferritin
By using apoferritin in our investigative studies we can
further support the proposal that the iron core is the redox
center responsible for the voltammetric behavior. Using both
3-M-1-PSA and 3-MPA as promoting systems, apoferritin is
investigated voltammetrically. Using 3-M-1-PSA as the
promoter a back ground scan is performed at 36 C, yielding a
cathodic spike at -441 mV (Fig. 3.18). The corresponding CV
scan in an apoferritin solution is seemingly unchanged (Fig.
3.19). The cathodic spike is at -444 mV and the blunt oxidation
peaks that are present in the buffer solution are absent. The
most striking feature of these scans is that the peaks seen
with ferritin are not seen with apoferritin.
3-MPA is also used as a promoting system yielding
similar results. The back ground scan of the promoter,
performed at 24.0 C, shows a reduction peak at -438 mV and
an oxidation peak at -394 mV (Fig. 3.20). The corresponding
apoferritin scan, at 23.7 C, shows the same reduction peak,
now at -441 mV and the oxidation peak is diminished (Fig.
3.21). Again, the absence of redox peaks strongly supports that
the redox peaks seen with ferritin are due to the
oxidation/reduction of the iron core.
70


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.18 Cyclic voltammetric scan of 3-mercapto-1-propane
sulfonic acid, adsorbed layer, from 1 mM; in 0.05 M NaH2P04
(pH = 7) with 0.5 M Na2S04. Temperature, 36.0 C; electrode,
polycrystalline gold; electrode area, 8.48 cm2; scan rate, 5.0
mV/s.
71


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.19 Cyclic voltammetric scan of apoferritin, 4 mg/mL in
0.05 M NaH2P04 (pH = 7) with 0.5 M Na2S04. Adsorbed layer, 3-
mercapto-1-propane sulfonic acid, from 1 mM; electrode,
polycrystalline gold; electrode area, 8.48 cm2. Temperature,
36.0 C; scan rate, 5.0 mV/s.
72


Fig. 3.20 Cyclic voltammetric scan of 3-mercapto propionic
acid, adsorbed layer, from 1 mM; in 0.05 M NaH2P04 (pH = 7)
with 0.5 M Na2S04. Temperature, 24.0 C; electrode,
polycrystalline gold; electrode area, 8.48 cm2; scan rate( 5.0
mV/s.
73


0
I--------H----------1----h-
-0.5 -0.3 -0.1 0
POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.21 Cyclic voltammetric scan of commercially available
apoferritin, 4 mg/mL in 0.05 M NaH2P04 (pH = 7) with 0.5 M
Na2S04. Adsorbed layer, 3-mercapto propionic acid, from 1
mM; electrode, polycrystalline gold; electrode area, 8.48 cm2.
Temperature, 23.7 C; scan rate, 5.0 mV/s.
74


I
I
The Voltammetry of Ferritin with the
Iron Core Removed
The iron core of ferritin can be removed by using
thioglycolic acid, as stated in the methods section, to produce
apoferritin. This technique is performed on a sample of horse
spleen ferritin that was obtained commercially. The resulting
apoferritin product should illustrate the same redox properties
as the commercially available apoferritin. Adsorbed 3-MPA is
used as the promoter in a this apoferritin solution at 23.0 C.
The CV scan obtained shows a reduction peak at -431 mV and
j
virtually no oxidation peak (Fig. 3.22). This is quite similar to
both the background scan (Fig. 3.20) and the scan using
commercial apoferritin (Fig. 3.21). These two CV scans
!
involving apoferritin further support the postulate that the iron
core of ferritin is responsible for the reversible vqltammetric
wave seen with ferritin.
i
CV Scans of the Reconstituted Iron Core. ,
both in Commercial and Laboratory Apoferritin
The iron core can be reconstituted using ammonium iron
(II) sulfate, as stated in the methods section. This technique is
performed using a sample of apoferritin that was obtained
commercially, and apoferritin that is prepared in the
i
75


<
H
Z
w
PS
pi
p
u
-0.5 -0.3 -0.1 0
POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.22 Cyclic voltammetric scan of laboratory prepared
apoferritin, 4 mg/mL in 0.05 M NaH2PC>4 (pH = 7) with 0.5 M
Na2SC>4. Adsorbed layer, 3-mercapto propionic acid, from 1
mM; electrode, polycrystalline gold; electrode area, 8.48 cm2
Temperature, 23.0 C; scan rate, 5.0 mV/s.
76


laboratory. The protein concentration is determined and the
iron content is analyzed and calculated also as stated in the
methods section. The back ground scan of 3-MPA is performed
at 27.0 C in phosphate buffer, yielding a scan quite similar to
other background scans (Fig. 3.23). There is an oxidation peak
at -388 mV and a reduction peak at -441 mV. The
corresponding scans for the reconstituted iron core in both the
Sigma (26.0 C) and the prepared apoferritin (24.3 C) are
rapidly descending reduction spikes (Fig. 3.24 and 3.25). These
CV scans are not as expected, as the anodic and cathodic waves
are not present. Moreover, the scans of the reconstituted iron
core does not resemble those scans of the background.
As a control, commercial horse spleen ferritin is run
through the same type of column as the reconstituted ferritins
(PD-10). The resulting CV scan, performed at 26.5 C, shows an
estimated peak separation of 142 mV (Fig. 3.26) when compared
to its background scan (Fig. 3.27). The ferritin that is filtered
through the PD-10 column does not appear to be as reversible
as the same promoting system previously mentioned (Fig. 3.15).
It is clear that the PD-10 column does not adversely affect the
redox chemistry of ferritin iron.
77


0
I-------I --------1---h-
-0.5 -0.3 -0.1 0
POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.23 Cyclic voltammetric scan of 3-mercapto propionic
acid, adsorbed layer, from 1 mM; in 0.05 M NaH2P04 (pH = 7)
with 0.5 M Na2S04. Temperature, 27.0 C; electrode,
polycrystalline gold; electrode area, 8.48 cm2; scan rate, 5.0
mV/s.
78


Fig. 3.24 Cyclic voltammetric scan of a reconstituted iron
core, 1626 Fe3+ atoms/ molecule, in commercially available
apoferritin, 1.511 mg/mL in 0.05 M NaH2P04 (pH = 7) with 0.5 M
Na2SC>4. Adsorbed layer, 3-mercapto propionic acid, from 1
mM; electrode, polycrystalline gold; electrode area, 8.48 cm2.
Temperature, 26.0 C; scan rate, 5.0 mV/s.
79


0
---1------1-----1---1-
-0.5 -0.3 -0.1 0
POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.25 Cyclic voltammetric scan of a reconstituted iron
core, 2112 Fe3+ atoms/ molecule, in laboratory prepared
apoferritin, 1.291 mg/mL in 0.05 M NaH2P04 (pH = 7) with 0.5 M
Na2S04- Adsorbed layer, 3-mercapto propionic acid, from 1
mM; electrode, polycrystalline gold; electrode area, 8.48 cm2.
Temperature, 24.3 C; scan rate, 5.0 mV/s.
80


Fig. 3.26 Cyclic voltammetric scan of filtered ferritin, 2.761
mg/mL in 0.05 M NaH2P04 (pH = 7) with 0.5 M Na2S04. Iron
core, 2168 Fe3+ atoms/ molecule. Adsorbed layer, 3-mercapto
propionic acid, from 1 mM; electrode, polycrystalline gold;
electrode area, 8.48 cm2. Temperature, 26.5 C; scan rate, 5.0
mV/s.
81


POTENTIAL (Volts vs. Ag/AgCl)
Fig. 3.27 Cyclic voltammetric scan of 3-mercapto propionic
acid, adsorbed layer, from 1 mM; in 0.05 M NaH2P04 (pH = 7)
with 0.5 M Na2SC>4. Temperature, 27.0 C; electrode,
polycrystalline gold; electrode area, 8.48 cm2; scan rate, 5.0
mV/s.
02


CHAPTER 4
DISCUSSION
The primary goal of this research has been
accomplished, a suitable surface electrocatalyst has been
found to facilitate the electron transfer of ferritin. A
polycrystalline gold electrode modified with 3-M-1-PSA has
produced a quasireversible redox wave for the iron core in
ferritin (Fig. 3.9). The peak potential difference, 131 mV, is
about twice the difference of a theoretical chemically
reversible one electron transfer. An accepted chemically and
electrochemically reversible system (Fe(ll)(CN)6'4) has a peak
potential difference of 194 mV, using the same
electrochemical cell and electrolyte system (Fig. 3.2).
Although the peak potential difference for cytochrome c is 78
mV using an adsorbed layer of 4,4' dithiodipyridyl (Fig. 3.1), it
is proposed that the adsorbed layer used for cytochrome c is
very efficient at promoting fast electron transfer kinetics.
The adsorbed layer used for ferritin does not seem to be as
efficient at catalyzing rapid electron transfer kinetics under
the present conditions. The comparison to Fe(ll)(CN)64 does
83


support that the electron transfer for ferritin is at least
quasireversible. Future studies will involve changing the
solvent or the electrolyte used for charge transfer in order to
try to bring the peak potential difference for Fe(ll)(CN)64 much
closer to a theoretically reversible value. By first establishing
the reversibility of an accepted redox system such as
Fe(ll)(CN)64 in our electrochemical apparatus, the reversibility
of ferritin can then be quantified.
The thermodynamic stability of the iron core can be
discussed by comparing cytochrome c's heme group with the
octahedral configuration of Fe(ll)(CN)64. The six cyanide
groups strongly back donate electrons to the iron, resulting in a
very stable compound. The stability is indicated in the E' of
0.240 V. In cytochrome c, the iron is maintained in a heme
group. The iron is bonded equatorially to a porphyrin ring and a
histidine at an axial position. The E0' obtained for cytochrome c
in this study is 0.040 V. The iron maintained in ferritin's core
is proposed to be in a hydrous ferric oxide (Fe(lll)O-OH) form
with variable amounts of attached phosphate. The E0' for
ferritin is -0.312 V, as calculated in this study (Fig. 3.9). It is
quite evident from this comparison that the iron core of
ferritin is very thermodynamically stable. Recalling that
ferritin composed of primarily L subunit has a greater ability
84


to stabilize core formation. Future voltammetric studies can
test the hypothesis that the H subunit does not have as great an
ability to stabilize the iron core. In this case, cellular heart
ferritin which is 60-65 % H subunit should be used.
A possible explanation for the failure to see a
reversible wave involves the conditions under which the iron
core is deposited. The reconstitution experiment occurs in a
phosphate free environment, a condition which does not imitate
the conditions present in the horse spleen. It is documented
that in the proliferation of the micelle, Fe3+ undergoes a
reversible electron transfer with the core iron to from Fe3+,
presumably bound to the mineral core (Watt et al., 1992). Watt
also states that the phosphate that is present as part of the
core is involved in catalyzing this electron transfer reaction.
We propose that the absence of phosphate during the
reconstitution resulted in an iron core devoid of any
incorporated phosphate. The absence of incorporated phosphate
perhaps has shifted the potential of the reversible redox wave
of reconstituted ferritin substantially more negative than that
of the naturally occurring iron core, to potentials more
negative than the hydrogen overpotential of gold. Potassium
iodate is used as an oxidizing agent in the reconstitution
experiments. Any low-valent species left after the filtering
85


process will displace the surface promoter and block ferritin's
electron transfer at the electrode. The CV scans that are
obtained appear similar to those expected for adsorbed iodine.
Future research will investigate these proposals by 1)
reconstituting the iron core using sodium chlorate as the
oxidizing agent, then running the CV scan, 2) reconstituting the
iron core in the presence of phosphate, then running the scan, or
3) using a static mercury drop electrode whose hydrogen
overpotential is well beyond that of gold. Time precludes
inclusion of the results in this publication.
All of the adsorbed layers that resulted in observable
redox waves project a formal negatively .charged group into
solution. It is proposed that these groups on the electrode
surface interact with positively charged residues on the
surface of ferritin. Watt postulates that during core
proliferation, electrons are transferred through the core to a
redox center (personal communication, 29 March 1994). The
ferridoxase center, maintained in the H subunit, and this redox
center may be one and the same. The existence of the
ferridoxase center, possibly a redox center, in the protein shell
supports the postulation of two possible mechanisms for
electron transfer with a modified electrode. 1) The adsorbed
layer electrostatically interacts with positive residues near
86


the hydrophilic channel, bring the core in a suitable orientation
for electron transfer to occur at the gold electrode. 2) The
negatively charged electrode surface layer interacts with
residues near the ferridoxase or redox centers, which may bring
these centers close enough to the electrode surface for
electron transfer to occur through the protein shell. Future
studies will test this postulate by performing CV scans on
cellular heart ferritin. If the electron transfer is occurring
through the ferridoxase center, the cellular heart ferritin
might give rise to a larger current and faster electron transfer
kinetics.
The adsorbed layers 3-M-1-PSA and 3MPA are suitable
for promoting electron transfer, but the redox activity of these
layers and the resulting masking effects pose obvious
problems. Although a cathodic shoulder for ferritin is often
quite evident, a truly reliable peak separation and peak current
ratio can not be calculated. The discovery of an adsorbed layer
that does not result in any redox activity in the range for
ferritin is desireable. Such a promoter could be used for
temperature and pH dependence studies.
87


BIBLIOGRAPHY
Andrews, S. C., Smith, J. M. A., Yewdall, S. J., Guest, J. R. &
Harrison, P. M. (1992) FEBS Lett. 293: 164-168
Arosio, P., Adelman, T. G. & Drysdale, J. W. (1978) J. Biol. Chem.
253: 4451-4458
Bard, A. J. & Faulkner, L. R. (1980) "Electrochemical Methods:
Fundamentals and ApplicationsJohn Wiley & Sons, New York,
342
Bauminger, E. R., Harrison, P. M., Hechel, D., Hodson, N. W., Nowik,
I., Treffry, A. & Yewdall, S. J. (1993) Biochem. J. 296:709-719
Bauminger, E. R., Harrison, P. M., Hechel, D., Nowik, I. & Treffry,
A. (1991) Biochem. Biophys. Acta. 1118: 48-58
Biedermann, G. & Schindler, P. (1957) Acta Chem. Scand 4: 731-
740
Campbell, C. H., Crocker, D., Gruntmeir, J. J., Head, M., Kelly, T.,
Langfur, M. I. & Leimer, A. H. (1993) J. Cell. Biochem. 53: 420-
432
Caskey, J. H., Jones, J. H., Miller, Y. E. & Seligman, P. A. (1983)
Proc. Nat'l. Acad. Sci. U.S.A. 80: 482-486
Clegg, G. A., Fitton, J. E., Harrison, P. M. & Treffry, A. (1980)
Prog. Biophys. Mol. Biol. 36: 56-86
Conway, B. E., Angerstein-Kozlowska, H., Sharp, W. B. A. &
Criddle, E. E. (1973) Anal. Chem. 45: 1331
88


Cozzi, A., Santambrogio, P., Levi, S. & Arosio, P. (1990) FEBS
Lett. 277: 119-122
Cragg, S. J., Wagstaff, M. & Worwood, M. (1982) "The
Biochemistry and Physiology of Iron"] Saltman, P. & Hegenauer,
J., Eds., Elsevier North Holland, Inc., 365-372
Crichton, R. R., Heusterspreute, M., Mathijs, J., Wustefeld, C.,
Charlier, G. & Chen, M. (1982) "The Biochemistry and Physiology
of Iron"; Saltman, P. & Hegenauer, J., Eds., Elsevier North
Holland, Inc. 353-358
Dedman, D. J., Treffry, A., Candy, J. M., Taylor, G. A. A., Morris, C.
M., Bloxham, C. A., Perry, R. H., Edmardson, J. A. & Harrison, P. M.
(1992) Biochem. J. 287: 509-514
Dexter, D. T., Carayon, A., Javoy-Agid, F., Agid, Y., Wells, F. R.,
Daniel, S. E., Lees, A. J., Jenner, P. & Marsden, C. D. (1991) Brain
114: 1953-1957
Dexter, D. T., Carayon, A., Vidailhet, M., Ruberg, M., Agid, F.,
Agid, Y., Lees, A. J., Wells, F. R., Jenner, P. & Marsden, C. D.
(1990) J. Neurochem. 55:16-20
Ferguson-Miller, S., Brautigan, D. L. & Margoliash, E. (1979) J.
Biol. Chem. 253: 149-159
Fleming, J. & Joshi, J. G. (1987) Proc. Natl. Acad. Sci. U.S.A. 84:
7866-7870
Ford, G. C., Harrison, P. M., Rice, D. W., Smith, J. M. A., Treffry,
A., White, J. L. & Yariv, J. (1984) Philos. Trans. R. Soc. London
Ser. B 304: 551-565
Frankel, R. B., Papaefthymiou, G. C. & Watt, G. D. (1987)
Hyperfine Interact. 33: 233
89


Gutteridge, J. M. C., Halliwell, B., Treffry, A., Harrison, P. M. &
Blake, D. (1983) Biochem. J. 209: 557-560
Harrison, P. M., Fischbaoh, F. A., Hoy, T. G. & Haggis, G. H. (1967)
Nature 216: 1188
Harrison, P. M., White, J. L., Smith, J. M. A., Farrants, G. W., Ford,
G. C., Rice, D. W., Addison, J. M. & Treffry, A. (1985) "Proteins of
Iron Storage and TransportSpik, G. Montreuil, J. Crichton, R. R.
& Mazurier, J., Eds., Elsevier Science Publishers, B. V., 67-79
Heusterspreute, M. & Crichton, R. R. (1981) Fedn. Eur. Biochem.
Soc. Lett. 129: 322-327
Hitchens, G. D. (1989) TIBS 14: 152-155
Hoy, T. G., Harrison, P. M. & Shabbier, M. (1974) Biochem. J. 139:
603-607
Joshi, J. G., Goodman, S., Deshpande, V. V. & Price, D. J. (1985)
Proteins of Iron Storage and TransportSpik, G., Montreuil, J.,
Crichton, R. R. & Mazurier, J., Eds., Elsevier Science Publishers,
B. V., 93-96
Kissinger, P. T. & Heineman, W. R. (1984) "Laboratory
Techniques in Electroanalytical ChemistryMarcel Dekker, Inc.,
New York, 339-340
Klausner, R. D., Rouault, T. A. & Harford, J. B. (1993) Cell 72:
19-28
Lawson, D. M., Treffry. A., Artymiuk, P. J., Harrison, P. M.,
Yewdall, S. J., Luzzago, A., Cesareni, G., Levi, S. & Arosio, P.
(1989) FEBS Lett. 254: 207-210
90