University of PennsylvaniaScholarlyCommons
Publicly Accessible Penn Dissertations
Spring 5-16-2011
Interaction of Ferritin With Transition Metal Ionsand ChelatesLei [email protected]
Follow this and additional works at: http://repository.upenn.edu/edissertations
Part of the Inorganic Chemistry Commons
This paper is posted at ScholarlyCommons. http://repository.upenn.edu/edissertations/321For more information, please contact [email protected].
Recommended CitationZhang, Lei, "Interaction of Ferritin With Transition Metal Ions and Chelates" (2011). Publicly Accessible Penn Dissertations. 321.http://repository.upenn.edu/edissertations/321
Interaction of Ferritin With Transition Metal Ions and Chelates
AbstractIn this thesis, interaction of ferritin with transition metal ions and chelates is studied. First, a simple methodfor synthesizing gold nanoparticles stabilized by horse spleen apoferritin (HSAF) was reported using NaBH4or 3-(N-morpholino)propanesulfonic acid (MOPS) as the reducing agent. The average particle diameterswere 3.6 and 15.4 nm for NaBH4-reduced and MOPS-reduced Au-HSAF, respectively. NaBH4-reduced Au-HSAF was much more effective than MOPS-reduced Au-HSAF in catalyzing the reduction of 4-nitrophenolby NaBH4, based on the greater accessibility of the NaBH4-reduced gold particle to the substrate. Methodsfor studying ferritin-gold nanoparticle assemblies may be readily applied to other protein-metal colloidsystems. Second, the binding interactions between HSAF and three recombinant apoferritins and twocommonly employed gadolinium magnetic resonance imaging (MRI) contrast agents were investigated. Theanionic Gd(DTPA)2- complex was undetectable in apoferritin solutions after dialysis, indicating little protein-binding interaction. However, the non-ionic Gd(DTPA-BMA) complex bound to all apoferritins tested,producing remarkable relaxivity enhancements as well as inhibition of iron mineralization. Serum ferritinbinding and dysregulation of iron mineralization may have medical significance, particularly in patients withimpaired kidney function.
Degree TypeDissertation
Degree NameDoctor of Philosophy (PhD)
Graduate GroupChemistry
First AdvisorIvan J. Dmochowski
Keywordsferritin, gold, gadolinium, nanoparticle, MRI
Subject CategoriesInorganic Chemistry
This dissertation is available at ScholarlyCommons: http://repository.upenn.edu/edissertations/321
INTERACTION OF FERRITIN WITH TRANSITION METAL IONS AND CHELATES
Lei Zhang
A DISSERTATION
in
Chemistry
Presented to the Faculties of the University of Pennsylvania in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
2011
__________________________________ Professor Ivan J. Dmochowski Supervisor of Dissertation
__________________________________ Professor Gary A. Molander Graduate Group Chair Committee Members: So-Jung Park, Assistant Professor of Chemistry (Chair) Christopher B. Murray, Richard Perry University Professor of Chemistry Larry G. Sneddon, Blanchard Professor of Chemistry
ii
Dedication
To people who light up my life.
iii
Acknowledgement
A few days before the Christmas of 2004, a plane landed at Philadelphia International
Airport. It was said that day was the coldest day in 2004, but I got the warmest welcome
by the Dmochowski group at Penn Chemistry Department. I stared my research journey
in the Dmochowski lab in January 2005. Springs and autumns passed unheeded. I
suddenly realized that it is about time to write this acknowledgement of my Ph.D thesis!
If my parents are credited for opening the door of this world for me, Ivan is credited
for opening the door of science for me. The nanoscience and biotechnology I learned
from this lab will form the foundation of my lifelong research. One thing makes me
excited is the research diversity in this lab! As I see, Ivan initiated the ideology of this lab
from a central topic in chemistry -- molecular interactions, while quickly expanding the
actual research into many fields of bioinorganic, bioorganic, and biophysical chemistry.
Through the years, I not only did research on ferritin and nanomaterials, but also gained
knowledge of xenon biosensing and light-activated oligonucleotides. This would not be
made possible without Ivan’s scientific enthusiasm and masterful supervision. I truly
enjoyed working with Ivan! Besides research, Ivan’s good personality is among the best I
have ever met. Ivan is understanding and patient, kind and willing to help. When first-
year graduate students asked me for advice in choosing a research lab, I always said,
“Want to do wonderful research? Choose the Dmochowski lab!”
I would like to take this opportunity to thank all the group members, whom I have been
working wonderfully with since 2005! They are graduate students (year of joining lab)
Joe Swift and Aru Hill (2004), Chris Butts, Julia Richards, and Garry Seward (2005),
iv
Jenny Chambers (2006), Najat Khan and Julie Griepenburg (2007), Jasmina Cheung-Lau
and Brittani Ruble (2008), Zhengzheng Liao and Yubin Bai (2009), Brittany Riggle
(2010), and postdoctoral scholars Xinjing Tang, William Wehbi, Qian Wei, Dage Liu,
Olena Taratula, and Ashley Fiamengo. Particular appreciation will be given to Team
Ferritin members Joe, Jasmina, and Dage, who have shared a lot of knowledge and
research techniques with me through the years!
Considering life outside the lab, I am actually not a very social person. But I still
experienced many happy moments brought to me by my friends at Penn. The following
names deserve to be mentioned: Qiqi Wang and Dan Yang of GSE, Xin Cao and Hualei
Zhang of SEAS, Yirui Huang and Xin Ge of Design, Chen Zhou and his friends. When I
look at my pictures taken through the years, I always feel happy to see their faces here
and there! On this occasion, I would like to say “thank you” to three more people:
Hanqing Liu of our department, Yongan Xu of SEAS, and Dawei Zhu of SOM.
I have been in U.S. since 2003. Eight years is a long time, but it doesn’t weaken the
connections between me and my family/friends in China. I want to particularly mention
my honey Xiaoyu Zheng. She is unique to me.
Last but not least, I would like to express my sincere gratitude to Profs. So-Jung Park,
Christopher Murray, and Larry Sneddon. As my Dissertation Committee members, they
have witnessed my growth over the years at Penn. Their profound insight and kind
encouragement will always be treasured and appreciated!
v
ABSTRACT
INTERACTION OF FERRITIN WITH TRANSITION METAL IONS AND CHELATES
Lei Zhang
Dr. Ivan J. Dmochowski
In this thesis, interaction of ferritin with transition metal ions and chelates is studied. First,
a simple method for synthesizing gold nanoparticles stabilized by horse spleen apoferritin
(HSAF) was reported using NaBH4 or 3-(N-morpholino)propanesulfonic acid (MOPS) as
the reducing agent. The average particle diameters were 3.6 and 15.4 nm for NaBH4-
reduced and MOPS-reduced Au-HSAF, respectively. NaBH4-reduced Au-HSAF was
much more effective than MOPS-reduced Au-HSAF in catalyzing the reduction of 4-
nitrophenol by NaBH4, based on the greater accessibility of the NaBH4-reduced gold
particle to the substrate. Methods for studying ferritin-gold nanoparticle assemblies may
be readily applied to other protein-metal colloid systems. Second, the binding interactions
between HSAF and three recombinant apoferritins and two commonly employed
gadolinium magnetic resonance imaging (MRI) contrast agents were investigated. The
anionic Gd(DTPA)2- complex was undetectable in apoferritin solutions after dialysis,
indicating little protein-binding interaction. However, the non-ionic Gd(DTPA-BMA)
complex bound to all apoferritins tested, producing remarkable relaxivity enhancements
as well as inhibition of iron mineralization. Serum ferritin binding and dysregulation of
iron mineralization may have medical significance, particularly in patients with impaired
kidney function.
vi
Table of Contents
Dedication ii
Acknowledgement iii
Abstract v
Chapter 1: Introduction 1
1. Prologue 1
2. Ferritin: Structure and Function 2
2.1. Apoferritin 4
2.2. Ferritin assembly 6
2.3. The iron core 6
3. Ferritin as the Complexation Host 8
3.1. Catalysis within the large ferritin central cavity 10
3.2. Therapy 13
3.3. Bionanotechnology 15
4. References 16
Chapter 2: Apoferritin-Stabilized Gold Nanoparticles 31
1. Introduction 32
2. Experimental 35
2.1. Reagents 35
2.2. Sample preparation 35
2.3. Physical measurements 36
2.4. Fast protein liquid chromatography (FPLC) 36
vii
2.5. Catalytic properties 37
2.6. Effect of reaction time on catalysis 38
3. Results and Discussion 38
4. Conclusions 49
5. Future Directions 49
6. References 50
Chapter 3: Binding of Gd-Chelates to Ferritin 75
1. Introduction 76
2. Experimental 78
2.1. Reagents 78
2.2. Gd-chelates 78
2.3. Recombinant proteins 78
2.4. Instrumentation 79
3. Results and Discussion 80
4. Conclusions 87
5. Future Directions 87
6. References 89
Appendix 1: Developing Strategies for Creating Ferritin-Gold Nanoparticle
Dimers Using Streptavidin-Biotin Junctions 113
Experimental Section 118
References 119
Appendix 2: DNA-Based Streptavidin-Biotin Complexation System 128
viii
Experimental Section 131
References 132
ix
List of Tables
Table 2.1 62 Table 3.1 96 Table 3.2 97
x
List of Figures
Figure 1.1 30 Figure 2.1 63 Figure 2.2 64 Figure 2.3 65 Figure 2.4 66 Figure 2.5 67 Figure 2.6 68 Figure 2.7 69 Figure 2.8 70 Figure 2.9 71 Figure 2.10 72 Figure 2.11 73 Figure 2.12 74 Figure 3.1 98 Figure 3.2 99 Figure 3.3 100 Figure 3.4 101 Figure 3.5 102 Figure 3.6 103 Figure 3.7 104 Figure 3.8 105 Figure 3.9 106 Figure 3.10 107 Figure 3.11 108 Figure 3.12 109 Figure 3.13 110 Figure 3.14 111 Figure 3.15 112 Figure A1.1 123 Figure A1.2 124 Figure A1.3 125 Figure A1.4 126 Figure A1.5 127 Figure A2.1 135 Figure A2.2 136 Figure A2.3 137 Figure A2.4 138
1
Chapter 1: Introduction
1. Prologue
In the history of human civilization, curing diseases and common ailments by means of
herbal remedies is among the earliest endeavors that might be classified as medical
science. The ancient Greeks and Chinese actively sought out these gifts of nature, using
all manner of herbs to heal sickness and boost health.† One widely known example is tea.
The health benefits of tea have been examined ever since the first infusions of Camellia
sinensis almost 5000 years ago.1 Modern research has revealed an inverse correlation
between green tea consumption and levels of serum markers including ferritin, implying
that green tea may act protectively against various cardiovascular diseases.2 More
recently, researchers at Cornell studied the effect of tea polyphenols on iron uptake by
ferritin in Caco-2 cells suggesting possible interactions between ferritin and tea
polyphenols, which could affect iron uptake.3 These are vivid examples showing how
experiential learning is linked to and explained by modern science. Now it is time to start
our own research journey, which is also relevant to the star molecule ferritin. In this
thesis, we use ferritin as a nanoscale biomolecular platform for complexation of particles
and molecules.
________________________________________________________________________
† I have developed a natural interest in this field, probably because my late grandfather was a doctor in the Hospital of Shandong University of Traditional Chinese Medicine (In Memoriam: http://wh.sdzydfy.com/book4/2008321919455353.htm#wps).
2
2. Ferritin: Structure and Function
Ferritin was first discovered in 1937 by the French scientist Laufberger, who isolated a
new protein from horse spleen that contained up to 23% by dry weight of iron.4 Since
then, it has been found in most living organisms, such as bacteria, fungi, plants, mollusks,
insects and vertebrates. Ferritin is synthesized in response to iron; much is now known
that the interactions of iron-regulatory proteins and iron-responsive elements play a
pivotal role in controlling iron homeostasis in invertebrates and vertebrates via regulation
of ferritin gene expression.5 Its universal occurrence as well as its fascinating ability to
reversibly store and release iron ions to the living cells has attracted the interest of
researchers worldwide.6-12
Over the past several years, ferritins have been classified as maxiferritins and
miniferritins.8,13 These two classes have distinctly different inner and outer diameters,
and molecular weights: (1) maxiferritins, 12 nm in diameter with 8 nm diameter cavities,
formed from 24 subunits (MW ≈ 480 kDa); and (2) miniferritins, also called DNA
protection during starvation proteins, 8 nm in diameter with 5 nm diameter cavities,
formed from 12 subunits (MW ≈ 240 kDa).14,15 In this thesis, we focus solely on
maxiferritins, and maxiferritin in this context is always referred to as ferritin. A core of
hydrous ferric oxide [Fe(III)O∙OH] is stored in the central cavity. Up to 4500 iron atoms
can be included in the core along with variable amounts of phosphate.16
Ferritin is an enzyme. It is a good place here to introduce the enzymatic properties of
ferritin. Published studies relevant to the enzymatic properties of ferritin can be classified
into two categories: (1) fundamental studies of reactions catalyzed by native ferritins; (2)
3
applications of reactions catalyzed by engineered ferritins. Research relevant to (2) will
be introduced in Section 3.1 of this chapter. Reactions in (1) can be further classified into
two sub-categories: (1.1) reactions catalyzed by the ferroxidase centers of ferritin; (1.2)
reactions catalyzed by the iron core of ferritin.
The catalytic activity of native ferritins is associated with the H-chain subunits that
contain the dinuclear ferroxidase center.17-19 At low iron loading (below or equal to that
required to fill the ferroxidase centers), the dominant reaction can be described by the
following net equation:20,21
2Fe2+ + O2 + 4H2O 2FeOOH + H2O2 + 4H+
The above oxidation reaction has been shown to proceed via a μ-1,2-peroxodiferric
intermediate giving a characteristic blue color (λmax = 650 nm),22-25 which subsequently
decays to a more stable μ-1,2-oxodiferric species with the release of H2O2.26
When intermediate numbers of iron is added (100~500 Fe2+ per protein), some of the
hydrogen peroxide produced in the above reaction will be consumed in an Fe2+ oxidation
reaction:27
2Fe2+ + H2O2 + 2H2O 2FeOOH + 4H+
When large amounts of iron are added (>800 Fe2+ per protein), auto-oxidation
catalyzed by the growing core surface becomes the dominant reaction:20,27
4Fe2+ + O2 + 6H2O 4FeOOH + 8H+
In (1.2), ferritin plays an important role in the decomposition of noxious derivatives of
oxygen, thus reducing exposure of cells to potentially damaging species such as peroxide,
4
superoxide, and hydroxyl radical.28 The iron core of ferritin acts as a catalyst in the
Haber-Weiss reaction which generates the extremely injurious hydroxyl radical.29
The iron core can be removed by reduction of Fe3+ and chelation of Fe2+.30 Ferritin can
be reconstituted from apoferritin upon addition of Fe2+ and an oxidizing agent.31 Watt and
co-workers found that iron loading into ferritin can be stimulated in the presence of
phosphate and its closely related analogs arsenate, vanadate, and molybdate.32 Cations
(Li+, Na+, K+, Rb+, Cs+, NH4+) were shown to slow iron loading into ferritin by a general
positive charge screening effect. Trigonal planar anions do not inhibit iron loading into
ferritin. Sulfate, iodide, and fluoride were shown to slow the rate of iron loading but the
same final absorbance is achieved suggesting that the mineral core is unchanged. Here we
begin this section with the review of the structure and function of apoferritin.
2.1. Apoferritin
The apoferritin structure, first reported in 1975, was from horse spleen, which contains
85% light (L) and 15% heavy (H)-type chains.33-35 Comparative studies of the
morphology of crystals of apoferritin from mammals and E. coli indicate great similarity
of packing.36 Horse spleen apoferritin (HSAF) serves as the prototypal apoferritin in this
thesis due to its commercial availability and natural combination of H- and L-chain
subunits. HSAF crystallizes readily as a Cd2+ salt, and its crystal structure has been
reported by several research groups.15,35,37 To date, the highest resolution available for
HSAF is 1.15 Å (Protein Data Bank ID: 2V2P). Each subunit of HSAF contains 174
amino acids folded into a bundle of two pairs of α-helixes joined by a long loop (Figure
1.1). The four long α-helixes are comprised of residues T10-Y36 (A), E45-R72 (B), T92-
5
Al19 (C), and P123-V155 (D), respectively, while the fifth shorter helix extends from
residues A159 to L169 (E). A long loop connecting two helixes contains residues G73-
T91. At each of the eight corners of the HSAF 24-mer cubic unit cell, threefold
symmetric interactions between three nearest subunits form a channel with highly
conserved hydrophilic character. The demonstration of various metal-binding sites in
these channels and the presence of conserved residues, D127 and E130, at the base of the
channels suggest that the channels could be important for transporting metal in and out of
the protein.15,34,35 Mutagenesis study of conserved residues inside these channels also
supported this conclusion.38 Interactions of four subunits form six fourfold symmetric
channels lined by hydrophobic residues such as L165. Each of these six channels is about
1.0 nm in diameter at the surface of the molecule, constricts to about 0.7 nm, and then
widens again to about 1.6 nm near the inner core of the protein.15,34,35 It has been
suggested that the six fourfold channels provide a pathway for transferring protons during
iron mineralization.39
It is noteworthy that metal ion bridging plays an important role in crystallizing certain
types of ferritins. As mentioned above, HSAF (H / L subunits ≈ 1 : 5) crystallizes as a
Cd2+ salt; HSAF molecules are linked by double Cd2+ bridges in the crystals. These
double Cd2+ bridges are situated near the crystal twofold axes formed by pairs of Asp and
Gln side chains. However, adding Cd2+ to H ferritin results in the formation of
amorphous precipitates. This is because in all H ferritins, Asp-84 is conserved, whereas
Gln-86 is replaced by another amino acid, usually Lys. Therefore, in a 1991 Nature paper,
Lawson et al. replaced the wild-type Lys-86 in human H ferritin (HuHF) with Gln-86,
6
and successfully crystallized it as a Ca2+ salt.17 This eliminates problems that were
initially encountered with trying to crystallize HuHF and its mutants without substitution
of the wild-type Lys. More recently, using modern crystallization screening kits, Butts et
al. successfully crystallized two HuHF mutants that lacked this LysGln point
mutation.40 But it was determined that adding Ca2+ to the ferritin solution was still
necessary to achieve crystals.
2.2. Ferritin assembly
Ferritin is known to disassemble into subunits at pH 2 and reassemble into the 24-mer
at pH 7.41 However, this pH-driven disassembly-assembly process often causes loss of
protein activity due to the near-denaturing conditions at low pH.
In 2005, Johnson and co-workers reported the crystal structures of a tetrahedral open
pore ferritin from the hyperthermophilic archaeon Archaeoglobus fulgidus, which
represents the first structure of a ferritin from an archaeon, or a hyperthermophilic
organism.42 The exceptional property of this A. fulgidus ferritin (the wild-type protein is
hereafter abbreviated as T0) is its salt-mediated, fully reversible assembly/disassembly: at
low salt concentrations, the protein exists as dimers of four-helix bundles. As the salt
concentration is increased to approximately 600 mM, all of the protein dimers self-
assemble into 24-mers. Swift et al. encapsulated gold nanoparticles within the cavity of
T0 via particle-directed assembly of T0 subunits in high yield.43 Employing this
methodology, Appendix 1 of this thesis will present the preparation and characterization
of a T0 mutant-based streptavidin-biotin complexation system.
2.3. The iron core
7
Various techniques have been used to study the process of mineral core formation in
ferritin.44 The most widely used technique, also the one used in Chapter 3 of this thesis, is
the aerobic oxidative iron uptake assay, or ferroxidase assay, which involves the aerobic
addition of Fe2+ to a suitably buffered protein solution. This technique relies on the
measurement of the absorbance increase at 340 to 410 nm, which is associated with the
formation of a Fe-O mineral whose brown color arises from a O Fe3+ charge transfer.45
A major proposal for iron core formation is the crystal growth mechanism described by
Clegg et al.36 In their proposal, starting with the empty protein shell, incoming Fe2+ is
oxidized on the inner surface of the protein. These oxidized Fe2+ form polynuclear iron
clusters which can act as nucleation centers for mineral growth (the nucleation stage). At
a certain time point one of the clusters becomes the dominant nucleation center and
growth of the mineral core occurs from this site (the core formation stage). In 2009, by
using high angle annular dark field STEM, Brown and co-workers performed 3D
morphological studies of the human hepatic ferritin mineral core and provided new
insights into the core formation mechanism.46 They found out that during the early stage
of iron deposition, the sites near the ends of the eight threefold channels are favorable for
the incoming Fe2+ to be oxidized and mineralized. As this process goes on, more Fe2+
enters into ferritin and is further oxidized and mineralized on the surface of any existing
Fe3+ deposits near the entry channels. With higher iron-filling, a cubic-like core structure
with eight subunits develops. Oxidation of additional incoming Fe2+ results in the early
deposited Fe3+ diffusing inwards, which forms closely packed crystalline structures of
ferrihydrite.
8
Although the mineral core of ferritin was traditionally described as ferrihydrite, more
recent studies have revealed a polyphasic structure consisting of “classic” ferrihydrite and
other phases, including a magnetite-like phase.47,48 Gálvez et al. demonstrated that the
native ferritin iron core consists of a polyphasic structure including ferrihydrite (FeOOH),
magnetite (Fe3O4), and hematite (α-Fe2O3).49 TEM, SAXS, and XANES findings suggest
a ferrihydrite-enriched core with a surface predominantly formed by magnetite, and the
ferrithydrite phase is more labile to chemical reduction and mobilization compared with
the magnetite phase. As iron was removed from ferritin, the relative amount of
ferrihydrite fell from ~60% to ~20%, and the percentage of magnetite rose steadily from
~20% to ~70%. The percentage of hematite remained in the range of ~20% to ~10%
during this process. These results confirm the heterogeneous structure of the ferritin core.
It may be noted that certain publications refer the composition of the native ferritin iron
core as ferrihydrite, magnetite, and maghemite (γ-Fe2O3).50 A better understanding of the
structural nature of the iron core is beneficial to the extended investigation on the
formation and complexation of various particles within ferritin. Studies relevant to the
formation of mineral core of different metals in ferritin will be discussed in the following
section.
3. Ferritin as the Complexation Host
Biomolecules, such as DNA51-54 and protein,55-58 have also been developed as host
templates for constructing nanomaterials toward various applications. Kelley and co-
workers reviewed recent studies that exploit biomolecules as templates and scaffolds for
9
the synthesis and manipulation of inorganic nanomaterials including metals,
semiconductors, and carbon nanotubes.51 These precedents suggested that imparting a
more complex structure at the nanoscale, such as core-shell nanoparticles,59 would offer
significant improvements in the performance of traditional semiconductor devices.
Matsui reviewed the research progress in the field of self-assembled peptide-based
nanostructures.58 The application of these molecular self-assemblies of synthetic peptides
as nanometer-scale building blocks in devices is robust, practical, and affordable due to
their advantages of reproducibility, large-scale production ability, monodispersity, and
simpler experimental methods. Overviews of protein cages as multifunctional
nanoplatforms,60 together with special summaries of research works relevant to the
nanoparticle synthesis in protein cages,56 the application of thermostable protein cages,61
the utilization of plant viruses as templates in bionanotechnology,57,62 have been
presented by various authors.
My thesis is associated with various topics of using biomolecules as the templates to
construct complexed systems. Research works in Chapters 2 and 3 and Appendix 1 of
this thesis are all related to ferritin-based complexation systems. Appendix 2 of this thesis
will investigate the preparation and properties of a DNA-based streptavidin-biotin
complexation system.
There has been a rich literature record of using ferritin as a host platform for
complexation of particles and molecules. It is interesting to note that Young and Douglas
proposed in a few recent review articles that the complexation modes of guest molecules
with the protein host can be classified into three categories: the interior, the interface, and
10
the exterior interactions of the guest species with the host.60-63 All publications in this
field can be virtually sorted into one of these three categories. A literature survey of
ferritin complexing with particles and molecules will be given in Chapters 2 and 3 of this
thesis, respectively. Here in the Introduction, we highlight instead the potential
applications of ferritin-based complexation systems.
3.1. Catalysis within the large ferritin central cavity
The history of performing catalyzed reactions inside the large central chamber of
ferritin dates back to 2004, when the size-selective olefin hydrogenation reaction
catalyzed by Pd nanoclusters inside apoferritin was carried out.64 For sample preparation,
apoferritin was treated with 500 equiv of K2PdCl4 at pH 8.5. The Pd2+ ions were reduced
by 5 equiv of NaBH4 to yield Pd clusters. Olefin hydrogenation reactions were carried
out in aqueous solution at 7 °C (pH 7.5) with 30 μM Pd. The hydrogenation turnover
frequency (TOF = [product (mol)] per Pd atom per hour) for CH2=CHCONH-iPr was
8.10 times larger than that of CH2=CHCOOH catalyzed by Pd nanoclusters inside
apoferritin. When catalyzed by pure Pd particles (under the same conditions, but without
apoferritin), the TOF ratio drastically decreased to 1.25. This result can be explained by
the observation that it is less favorable to diffuse negatively charged molecules
(CH2=CHCOO-, Gd(DTPA)2-, etc.) into ferritin than neutral molecules (CH2=CHCONH-
iPr, Gd(DTPA-BMA), etc.). This result is consistent with a previous investigation by
Chasteen and co-workers, who used EPR spectroscopy and gel permeation
chromatography to study the molecular diffusion of small nitroxide spin probes (7-9 Å
diameter) into the ferritin cavity.65,66 They found that the positively charged and neutral
11
probe molecules penetrated the cavity, whereas the negatively charged probe molecules
did not. An important implication from the above catalytic study is that the ferritin shell
can be used to control the relative rates of catalytic reactions with different substrates,
based on size or charge of the substrate. It may be noteworthy the observation by Watt
and co-workers that chloride anion is transported into ferritin during reduction of the iron
core.67 This may help us to realize that size is important factor in addition to charge that
governs whether or not a molecule can be diffused into ferritin. Cl- (from the Watt work),
CH2=CHCOO- (from the above catalytic work), and 7-9 Å nitroxide (from the Chasteen
work) are all negatively charged molecules. But their abilities of diffusing into ferritin
decrease drastically, apparently correlated with size.
The olefin hydrogenation reaction catalyzed by Au/Pd core-shell nanoparticles inside
apoferritin was similarly examined.68 The catalytic activity of Pd located on the surface
of [Au](Pd) can be improved due to the electronic interaction between the Au core and
the Pd shell, which could provide an uneven distribution of electrons. The Pd in the
surface layer becomes poorer in electron density than in the Au core, which could make
Au/Pd core-shell nanoparticles more active than Pd clusters since the substrate olefin
having a double bond favors the electron-deficient surface.69,70 Most recently, the
reduction of 4-nitrophenol in the presence of NaBH4 by Au-Ag alloy nanoparticles inside
HSAF was reported.71 Exponential increase of the rate constant of the reduction was
observed when increasing the Au content in the Au-Ag alloy.
Not only metal nanoparticles, but also organometallic compounds can be loaded into
apoferritin to catalyze reactions in confined spaces. Organometallic chemistry in confined
12
spaces adds a new dimension to traditional surface organometallic chemistry as verified
by highly efficient nanovessel reactors and quantum-confined materials.72 Performing
organometallic compound-catalyzed reactions will help us not only understand the
interaction between proteins and organic molecules, but also find new applications in
catalytic reactions. Contributions in this field include redox reactions catalyzed by
ferrocene derivatives,73 the Suzuki coupling reaction catalyzed by Pd(allyl),74 and
phenylacetylene polymerization reactions catalyzed by Rh(nbd), all inside apoferritin.75 It
is noteworthy that the binding of these metal chelates inside apoferritin has been studied
by single-crystal X-ray crystallography. I want to particularly mention the ferrocene
work,73 because in this study ferrocene interacted noncovalently with the interior of the
protein. Three monosubstituted ferrocene derivatives (ferrocene carboxylic acid, 1;
dimethylaminomethylferrocene, 2; 1-dimethylaminoethylferrocene, 3) were loaded into
recombinant horse L-chain apoferritin. ICP-OES spectroscopy together with protein
concentration determination revealed that 1.9, 2.4, and 0.8 equiv ferrocene per subunit
were incorporated into 1-, 2-, and 3-apoferritin, respectively. The number of ferrocene
per subunit was quite high in all cases, suggesting that there were many weak interactions
with the ferritin interior. Crystal structures were determined with resolutions of 2.0, 1.6,
2.0 Å for 1-, 2-, and 3-apoferritin, respectively. The anomalous Fourier difference map
calculation for 2- and 3-apoferritin showed the existence of a new anomalous peak at the
twofold symmetry axis of the apoferritin molecule, which located between the
monomeric subunits of the apoferritin shell. Although the structures of the ferrocene
derivatives within the protein shell could not be determined, the X-ray analysis indicated
13
the binding of the ferrocenyl amines 2 and 3 at this position. Curiously, a similar
anomalous peak for 1-apoferritin was not observed; one possible explanation is that the
negative charge of the carboxylate lowered the propensity of 1 to permeate the ferritin
protein shell.
3.2. Therapy
Systems of organic molecules or inorganic particles trapped inside apoferritin have
shown promising therapeutic applications. In iron chelation therapy, it is known that
excessive iron in the human body will accumulate in the liver and other organs, which
can lead to the damage to tissue and organ.76 Complications associated with elevated iron
levels can be largely avoided by the use of molecules that can chelate and remove a
sufficient amount of iron, maintaining the body iron load at a non-toxic level.
Desferrioxamine B (DFOB) is the current drug for iron chelation therapy used via
injection. However, DFOB has short plasma half-life and must be administered slowly
over a period of time to be effective.77,78 One potential strategy then for improving the
efficacy of DFOB, is to trap DFOB within a biocompatible, slow-release delivery system.
The dissociation of apoferritin into subunits at pH 2 followed by its reassembly at pH 7.4
trapped ca. three DFOB inside.79 Investigation showed that these trapped DFOB
molecules are able to react with Fe3+ to form the corresponding complex. Trapping
molecules of this kind inside apoferritin provides a useful proof-of-concept for a mode of
drug delivery. Due to the lack of information (what timescale, how much, where, etc.)
relevant to the release of DFOB, at this moment it is hard to estimate if this drug delivery
system can be truly used in therapy.
14
Functional guest species trapped within apoferritin are also finding potential
applications in cancer therapy. Cancer can be treated by various means including
chemotherapy and radiation therapy. Doxorubicin is a chemotherapy drug used in the
treatment of a variety of cancers. In order to enhance potentially its therapeutic index and
achieve specific delivery to defined cancer cells, doxorubicin was trapped inside
apoferritin following essentially the same procedure as used for trapping DFOB.41
Radiation therapy is the use of penetrating beams of radiation to treat disease. However,
long-term experience of cancer treatment has shown that certain tumor types become
radioresistant and are very difficult to kill.80 Under this situation, neutron therapy is a
promising choice, which specializes in treating inoperable, radioresistant tumors
occurring anywhere in the body.81 Uranium is a promising metal to be used in neutron
therapy. However, its extensive use is limited by the toxicity of uranium compounds, the
low uptake by common metal chelators, and the lack of methodology to attach hundreds
of atoms per antibody. To solve these problems, Hainfeld loaded apoferritin with 800
uranium atoms, attached Fab’ antibody to the loaded product, and demonstrated the
retention of immunoreactivity by this complex.82 To achieve the high loading of
apoferritin required for potential therapy (> 300 uranium atoms per apoferritin), the
author developed a strategy to actively concentrate uranium within the central cavity of
apoferritin. This strategy relied on controlling the solubility of the uranyl ion by pH:
uranyl acetate is soluble at pH ≤ 4.5, but becomes insoluble uranyl complexes at higher
pH. By soaking apoferritin with uranyl acetate at pH 4 and then raising the pH to ≈ 7, the
author successfully nucleated and grew uranium crystals inside apoferritin.
15
3.3. Bionanotechnology
Bionanotechnology is emerging as an interdisciplinary field for chemists, material
scientists, and biomedical researchers. Biomolecules are attractive nanoreactors for the
artificial synthesis of homogeneous metal nanoparticles, semiconductor nanoparticles,
and nanofunctional bioconjugates.83 One typical example is the construction of a two-
dimensional crystal array of PbS nanoparticle-loaded chaperonin protein on the channel
part of MOS transistors, which worked successfully as charge storage nodes opening
potential applications in flash memory technology.84 Nanoparticles are fundamental
elements in nanotechnology. The inner cavity of apoferritin provides an ideal spatially
restricted chemical reaction chamber for nanoparticle synthesis.
Mann and co-workers have done pioneering work in applying ferritin to different
problems in nanotechnology. In their earliest contribution, iron sulfide, manganese oxide,
uranyl oxyhydroxide were respectively synthesized in apoferritin.85 Research in this field
has since attracted worldwide attention; examples will be summarized in the next chapter.
It may be noted that in the synthetic procedure, apoferritin is typically mixed with a
few thousand equiv of metal ions. It had been generally assumed that all the metal ions
were sucked into apoferritin during the incubation. Therefore, no dialysis was carried out
before the mineralization step.86-89 However, in a 2004 paper reporting the photocatalytic
synthesis of copper particles from Cu2+ by the ferrihydrite core of ferritin, the authors
found out that the Cu particle dimensions as measured by TEM and DLS were larger than
the 12-nm ferritin, which suggested that the formation of Cu nanoparticles could either
occur outside the protein cage, or that the protein cage was disrupted during formation of
16
the Cu particle within the protein.90 Later in 2007 the study of gold nanoparticles
synthesized in the presence of apoferritin revealed that it is not correct to assume that all
the metal ions can concentrated inside apoferritin after incubation.91 Two 2010 papers
figured out the importance of adding a dialysis step in the synthetic procedure. In the gold
nanoparticle paper, HSAF was mixed with HAuCl4 in the 1-to-1000 molar ratio. Before
adding NaBH4, the pale-yellow solution was applied to a desalting column to remove free
HAuCl4.92 The same column and essentially the same procedure were also used for the
Au-Ag alloy nanoparticle paper.71
An intriguing field of ferritin research toward bio-nanotechnological applications is the
fabrication of two-dimensional (2-D) ferritin assemblies. A remarkable work was done by
Yoshimura et al., who developed a simple method for making 2-D ferritin arrays using a
new spreading technique.93 In this paper, the addition of 10 mM Cd2+ to the aqueous
subphase is a critical factor in optimizing the 2-D array formation. In the absence of Cd2+,
only random adsorption of ferritin molecules to the unfolded protein film was observed.
Cd2+ should play the same role as explained in Section 2.1 of this chapter. In a later
publication, Yoshimura and co-workers fabricated iron and indium nanoparticle arrays on
carbon film and silicon wafer using a protein crystal lattice, which has potential use in a
wide variety of electronic and semiconductor devices.94 Self-assembly methods,
combined with protein design and semiconductor organic/inorganic interfaces, are
expected to herald the future of nanotechnology-based advanced manufacturing.
4. References
17
1. Sharangi, A. B. Medicinal and therapeutic potentialities of tea (Camellia sinensis
L.) - A review. Food Res. Int. 42, 529-535 (2009).
2. Imai, K. & Nakachi, K. Cross-sectional study of effects of drinking green tea on
cardiovascular and liver-diseases. Br. Med. J. 310, 693-696 (1995).
3. Laparra, J. M., Glahn, R. P. & Miller, D. D. Effect of tea phenolics on iron uptake
from different fortificants by Caco-2 cells. Food Chem. 115, 974-981 (2009).
4. Laufberger, M. V. Sur la cristallisation de la ferritine. Bull. Soc. Chim. Biol. 19,
1575-1582 (1937).
5. Thomson, A. M., Rogers, J. T. & Leedman, P. J. Iron-regulatory proteins, iron-
responsive elements and ferritin mRNA translation. Int J. Biochem. Cell Biol. 31,
1139-1152 (1999).
6. Wang, W., Knovich, M. A., Coffman, L. G., Torti, F. M. & Torti, S. V. Serum
ferritin: past, present and future. Biochim. Biophys. Acta-Gen. Subj. 1800, 760-
769 (2010).
7. Knovich, M. A., Storey, J. A., Coffman, L. G., Torti, S. V. & Torti, F. M. Ferritin
for the clinician. Blood Rev. 23, 95-104 (2009).
8. Theil, E. C., Liu, X. F. S. & Tosha, T. Gated pores in the ferritin protein nanocage.
Inorg. Chim. Acta 361, 868-874 (2008).
9. Theil, E. C. Iron, ferritin, and nutrition. Annu. Rev. Nutr. 24, 327-343 (2004).
10. Arosio, P. & Levi, S. Ferritin, iron homeostasis, and oxidative damage. Free
Radic. Biol. Med. 33, 457-463 (2002).
18
11. Torti, F. M. & Torti, S. V. Regulation of ferritin genes and protein. Blood 99,
3505-3516 (2002).
12. Proulxcurry, P. M. & Chasteen, N. D. Molecular aspects of iron uptake and
storage in ferritin. Coord. Chem. Rev. 144, 347-368 (1995).
13. Liu, X. F. & Theil, E. C. Ferritin reactions: direct identification of the site for the
diferric peroxide reaction intermediate. Proc. Natl. Acad. Sci. U. S. A. 101, 8557-
8562 (2004).
14. Liu, X. F. & Theil, E. C. in Cooley's Anemia Eighth Symposium 136-140 (New
York Acad Sciences, New York, 2005).
15. Granier, T., Gallois, B., Dautant, A., Destaintot, B. L. & Precigoux, G.
Comparison of the structures of the cubic and tetragonal forms of horse-spleen
apoferritin. Acta Crystallogr. Sect. D-Biol. Crystallogr. 53, 580-587 (1997).
16. Mann, S., Bannister, J. V. & Williams, R. J. P. Structure and composition of
ferritin cores isolated from human spleen, limpet (Patella vulgata) hemolymph
and bacterial (Pseudomonas aeruginosa) cells. J. Mol. Biol. 188, 225-232 (1986).
17. Lawson, D. M., Artymiuk, P. J., Yewdall, S. J., Smith, J. M. A., Livingstone, J. C.,
Treffry, A., Luzzago, A., Levi, S., Arosio, P., Cesareni, G., Thomas, C. D., Shaw,
W. V. & Harrison, P. M. Solving the structure of human H-ferritin by genetically
engineering intermolecular crystal contacts. Nature 349, 541-544 (1991).
18. Lawson, D. M., Treffry, A., Artymiuk, P. J., Harrison, P. M., Yewdall, S. J.,
Luzzago, A., Cesareni, G., Levi, S. & Arosio, P. Identification of the ferroxidase
center in ferritin. FEBS Lett. 254, 207-210 (1989).
19
19. Levi, S., Luzzago, A., Cesareni, G., Cozzi, A., Franceschinelli, F., Albertini, A. &
Arosio, P. Mechanism of ferritin iron uptake: activity of the H-chain and deletion
mapping of the ferro-oxidase site. J. Biol. Chem. 263, 18086-18092 (1988).
20. Yang, X. K., Chen-Barrett, Y., Arosio, P. & Chasteen, N. D. Reaction paths of
iron oxidation and hydrolysis in horse spleen and recombinant human ferritins.
Biochemistry 37, 9743-9750 (1998).
21. Sun, S. J., Arosio, P., Levi, S. & Chasteen, N. D. Ferroxidase kinetics of human
liver apoferritin, recombinant H-chain apoferritin, and site-directed mutants.
Biochemistry 32, 9362-9369 (1993).
22. Pereira, A. S., Small, W., Krebs, C., Tavares, P., Edmondson, D. E., Theil, E. C.
& Huynh, B. H. Direct spectroscopic and kinetic evidence for the involvement of
a peroxodiferric intermediate during the ferroxidase reaction in fast ferritin
mineralization. Biochemistry 37, 9871-9876 (1998).
23. Moenne-Loccoz, P., Krebs, C., Herlihy, K., Edmondson, D. E., Theil, E. C.,
Huynh, B. H. & Loehr, T. M. The ferroxidase reaction of ferritin reveals a diferric
μ-1,2 bridging peroxide intermediate in common with other O2-activating non-
heme diiron proteins. Biochemistry 38, 5290-5295 (1999).
24. Bou-Abdallah, F., Zhao, G. H., Mayne, H. R., Arosio, P. & Chasteen, N. D.
Origin of the unusual kinetics of iron deposition in human H-chain ferritin. J. Am.
Chem. Soc. 127, 3885-3893 (2005).
25. Treffry, A., Zhao, Z., Quail, M. A., Guest, J. R. & Harrison, P. M. Iron(II)
oxidation by H-chain ferritin: evidence from site-directed mutagenesis that a
20
transient blue species is formed at the dinuclear iron center. Biochemistry 34,
15204-15213 (1995).
26. Bou-Abdallah, F., Papaefthymiou, G. C., Scheswohl, D. M., Stanga, S. D., Arosio,
P. & Chasteen, N. D. μ-1,2-Peroxobridged di-iron(III) dimer formation in human
H-chain ferritin. Biochem. J. 364, 57-63 (2002).
27. Zhao, G. H., Bou-Abdallah, F., Arosio, P., Levi, S., Janus-Chandler, C. &
Chasteen, N. D. Multiple pathways for mineral core formation in mammalian
apoferritin. The role of hydrogen peroxide. Biochemistry 42, 3142-3150 (2003).
28. Bolann, B. J. & Ulvik, R. J. Decay of superoxide catalyzed by ferritin. FEBS Lett.
318, 149-152 (1993).
29. Halliwell, B. & Gutteridge, J. M. C. Oxygen toxicity, oxygen radicals, transition
metals and disease. Biochem. J. 219, 1-14 (1984).
30. Granick, S. & Michaelis, L. Ferritin: II. apoferritin of horse spleen. J. Biol. Chem.
147, 91-97 (1943).
31. Harrison, P. M., Fischbach, F. A., Hoy, T. G. & Haggis, G. H. Ferric
oxyhydroxide core of ferritin. Nature 216, 1188-1190 (1967).
32. Cutler, C., Bravo, A., Ray, A. D. & Watt, R. K. Iron loading into ferritin can be
stimulated or inhibited by the presence of cations and anions: a specific role for
phosphate. J. Inorg. Biochem. 99, 2270-2275 (2005).
33. Hoare, R. J., Harrison, P. M. & Hoy, T. G. Structure of horse spleen apoferritin at
6-Å resolution. Nature 255, 653-654 (1975).
21
34. Banyard, S. H., Stammers, D. K. & Harrison, P. M. Electron density map of
apoferritin at 2.8-Å resolution. Nature 271, 282-284 (1978).
35. Rice, D. W., Ford, G. C., White, J. L., Smith, J. M. A. & Harrison, P. M. The
spatial structure of horse spleen apoferritin. Adv. Inorg. Biochem. 5, 39-50 (1983).
36. Clegg, G. A., Fitton, J. E., Harrison, P. M. & Treffry, A. Ferritin: molecular-
structure and iron-storage mechanisms. Prog. Biophys. Mol. Biol. 36, 53-86
(1980).
37. Harrison, P. M., Andrews, S. C., Artymiuk, P. J., Ford, G. C., Guest, J. R.,
Hirzmann, J., Lawson, D. M., Livingstone, J. C., Smith, J. M. A., Treffry, A. &
Yewdall, S. J. Probing structure-function relations in ferritin and bacterioferritin.
Adv. Inorg. Chem. 36, 449-486 (1991).
38. Levi, S., Santambrogio, P., Corsi, B., Cozzi, A. & Arosio, P. Evidence that
residues exposed on the three-fold channels have active roles in the mechanism of
ferritin iron incorporation. Biochem. J. 317, 467-473 (1996).
39. Takahashi, T. & Kuyucak, S. Functional properties of threefold and fourfold
channels in ferritin deduced from electrostatic calculations. Biophys. J. 84, 2256-
2263 (2003).
40. Butts, C. A., Swift, J., Kang, S. G., Di Costanzo, L., Christianson, D. W., Saven, J.
G. & Dmochowski, I. J. Directing noble metal ion chemistry within a designed
ferritin protein. Biochemistry 47, 12729-12739 (2008).
41. Simsek, E. & Kilic, M. A. Magic ferritin: a novel chemotherapeutic encapsulation
bullet. J. Magn. Magn. Mater. 293, 509-513 (2005).
22
42. Johnson, E., Cascio, D., Sawaya, M. R., Gingery, M. & Schroder, I. Crystal
structures of a tetrahedral open pore ferritin from the hyperthermophilic archaeon
Archaeoglobus fulgidus. Structure 13, 637-648 (2005).
43. Swift, J., Butts, C. A., Cheung-Lau, J., Yerubandi, V. & Dmochowski, I. J.
Efficient self-assembly of Archaeoglobus fulgidus ferritin around metallic cores.
Langmuir 25, 5219-5225 (2009).
44. Lewin, A., Moore, G. R. & Le Brun, N. E. Formation of protein-coated iron
minerals. Dalton Trans., 3597-3610 (2005).
45. Le Brun, N. E., Wilson, M. T., Andrews, S. C., Guest, J. R., Harrison, P. M.,
Thomson, A. J. & Moore, G. R. Kinetic and structural characterization of an
intermediate in the biomineralization of bacterioferritin. FEBS Lett. 333, 197-202
(1993).
46. Pan, Y. H., Sader, K., Powell, J. J., Bleloch, A., Gass, M., Trinick, J., Warley, A.,
Li, A., Brydson, R. & Brown, A. 3D morphology of the human hepatic ferritin
mineral core: new evidence for a subunit structure revealed by single particle
analysis of HAADF-STEM images. J. Struct. Biol. 166, 22-31 (2009).
47. Cowley, J. M., Janney, D. E., Gerkin, R. C. & Buseck, P. R. The structure of
ferritin cores determined by electron nanodiffraction. J. Struct. Biol. 131, 210-216
(2000).
48. Quintana, C., Cowley, J. M. & Marhic, C. Electron nanodiffraction and high-
resolution electron microscopy studies of the structure and composition of
physiological and pathological ferritin. J. Struct. Biol. 147, 166-178 (2004).
23
49. Galvez, N., Fernandez, B., Sanchez, P., Cuesta, R., Ceolin, M., Clemente-Leon,
M., Trasobares, S., Lopez-Haro, M., Calvino, J. J., Stephan, O. & Dominguez-
Vera, J. M. Comparative structural and chemical studies of ferritin cores with
gradual removal of their iron contents. J. Am. Chem. Soc. 130, 8062-8068 (2008).
50. Brem, F., Stamm, G. & Hirt, A. M. Modeling the magnetic behavior of horse
spleen ferritin with a two-phase core structure. J. Appl. Phys. 99 (2006).
51. Ma, N., Sargent, E. H. & Kelley, S. O. Biotemplated nanostructures: directed
assembly of electronic and optical materials using nanoscale complementarity. J.
Mater. Chem. 18, 954-964 (2008).
52. Seeman, N. C. An overview of structural DNA nanotechnology. Mol. Biotechnol.
37, 246-257 (2007).
53. Gothelf, K. V. & LaBean, T. H. DNA-programmed assembly of nanostructures.
Org. Biomol. Chem. 3, 4023-4037 (2005).
54. Shih, W. M., Quispe, J. D. & Joyce, G. F. A 1.7-kilobase single-stranded DNA
that folds into a nanoscale octahedron. Nature 427, 618-621 (2004).
55. Kentsis, A. & Borden, K. L. B. Physical mechanisms and biological significance
of supramolecular protein self-assembly. Curr. Protein Pept. Sci. 5, 125-134
(2004).
56. Yoshimura, H. Protein-assisted nanoparticle synthesis. Colloid Surf. A-
Physicochem. Eng. Asp. 282, 464-470 (2006).
57. Steinmetz, N. F. & Evans, D. J. Utilisation of plant viruses in bionanotechnology.
Org. Biomol. Chem. 5, 2891-2902 (2007).
24
58. Gao, X. Y. & Matsui, H. Peptide-based nanotubes and their applications in
bionanotechnology. Adv. Mater. 17, 2037-2050 (2005).
59. Dabbousi, B. O., RodriguezViejo, J., Mikulec, F. V., Heine, J. R., Mattoussi, H.,
Ober, R., Jensen, K. F. & Bawendi, M. G. (CdSe)ZnS core-shell quantum dots:
synthesis and characterization of a size series of highly luminescent
nanocrystallites. J. Phys. Chem. B 101, 9463-9475 (1997).
60. Uchida, M., Klem, M. T., Allen, M., Suci, P., Flenniken, M., Gillitzer, E.,
Varpness, Z., Liepold, L. O., Young, M. & Douglas, T. Biological containers:
protein cages as multifunctional nanoplatforms. Adv. Mater. 19, 1025-1042
(2007).
61. Wiedenheft, B., Flenniken, M. L., Allen, M. A., Young, M. & Douglas, T.
Bioprospecting in high temperature environments; application of thermostable
protein cages. Soft Matter 3, 1091-1098 (2007).
62. Young, M., Willits, D., Uchida, M. & Douglas, T. Plant viruses as biotemplates
for materials and their use in nanotechnology. Annu. Rev. Phytopathol. 46, 361-
384 (2008).
63. Douglas, T. & Young, M. Viruses: making friends with old foes. Science 312,
873-875 (2006).
64. Ueno, T., Suzuki, M., Goto, T., Matsumoto, T., Nagayama, K. & Watanabe, Y.
Size-selective olefin hydrogenation by a Pd nanocluster provided in an apo-
ferritin cage. Angew. Chem. Int. Ed. 43, 2527-2530 (2004).
25
65. Yang, X. K. & Chasteen, N. D. Molecular diffusion into horse spleen ferritin: a
nitroxide radical spin probe study. Biophys. J. 71, 1587-1595 (1996).
66. Yang, X. K., Arosio, P. & Chasteen, N. D. Molecular diffusion into ferritin:
pathways, temperature dependence, incubation time, and concentration effects.
Biophys. J. 78, 2049-2059 (2000).
67. Hilty, S., Webb, B., Frankel, R. B. & Watt, G. D. Iron core formation in horse
spleen ferritin: magnetic susceptibility, pH, and compositional studies. J. Inorg.
Biochem. 56, 173-185 (1994).
68. Suzuki, M., Abe, M., Ueno, T., Abe, S., Goto, T., Toda, Y., Akita, T., Yamadae,
Y. & Watanabe, Y. Preparation and catalytic reaction of Au/Pd bimetallic
nanoparticles in apo-ferritin. Chem. Commun., 4871-4873 (2009).
69. Scott, R. W. J., Wilson, O. M., Oh, S. K., Kenik, E. A. & Crooks, R. M.
Bimetallic palladium-gold dendrimer-encapsulated catalysts. J. Am. Chem. Soc.
126, 15583-15591 (2004).
70. Toshima, N., Harada, M., Yamazaki, Y. & Asakura, K. Catalytic activity and
structural-analysis of polymer-protected Au-Pd bimetallic clusters prepared by the
simultaneous reduction of HAuCl4 and PdCl2. J. Phys. Chem. 96, 9927-9933
(1992).
71. Shin, Y. S., Dohnalkova, A. & Lin, Y. H. Preparation of homogeneous gold-silver
alloy nanoparticles using the apoferritin cavity as a nanoreactor. J. Phys. Chem. C
114, 5985-5989 (2010).
26
72. Anwander, R. SOMC@PMS. Surface organometallic chemistry at periodic
mesoporous silica. Chem. Mater. 13, 4419-4438 (2001).
73. Niemeyer, J., Abe, S., Hikage, T., Ueno, T., Erker, G. & Watanabe, Y.
Noncovalent insertion of ferrocenes into the protein shell of apo-ferritin. Chem.
Commun., 6519-6521 (2008).
74. Abe, S., Niemeyer, J., Abe, M., Takezawa, Y., Ueno, T., Hikage, T., Erker, G. &
Watanabe, Y. Control of the coordination structure of organometallic palladium
complexes in an apo-ferritin cage. J. Am. Chem. Soc. 130, 10512-10514 (2008).
75. Abe, S., Hirata, K., Ueno, T., Morino, K., Shimizu, N., Yamamoto, M., Takata,
M., Yashima, E. & Watanabe, Y. Polymerization of phenylacetylene by rhodium
complexes within a discrete space of apo-ferritin. J. Am. Chem. Soc. 131, 6958-
6960 (2009).
76. Weatherall, D. J. in: Martel, A. E., Anderson, W. F., Badman D. G. (Eds.),
Developement of iron chelators for clinical use. Elsevier, New York (1981).
77. Summers, M. R., Jacobs, A., Tudway, D., Perera, P. & Ricketts, C. Studies in
Desferrioxamine and Ferrioxamine metabolism in normal and iron-loaded
subjects. Br. J. Haematol. 42, 547-555 (1979).
78. Lee, P., Mohammed, N., Marshall, L., Abeysinghe, R. D., Hider, R. C., Porter, J.
B. & Singh, S. Intravenous-infusion pharmacokinetics of Desferrioxamine in
thalassemic patients. Drug Metab. Dispos. 21, 640-644 (1993).
79. Dominguez-Vera, J. M. Iron(III) complexation of Desferrioxamine B
encapsulated in apoferritin. J. Inorg. Biochem. 98, 469-472 (2004).
27
80. Cohen, L. & Hendrickson, F. R. Neutron therapy for nonresectable radioresistant
tumors. Arch. Surg. 124, 294-300 (1989).
81. Soloway, A. H., Tjarks, W., Barnum, B. A., Rong, F. G., Barth, R. F., Codogni, I.
M. & Wilson, J. G. The chemistry of neutron capture therapy. Chem. Rev. 98,
1515-1562 (1998).
82. Hainfeld, J. F. Uranium-loaded apoferritin with antibodies attached: molecular
design for uranium neutron-capture therapy. Proc. Natl. Acad. Sci. U. S. A. 89,
11064-11068 (1992).
83. Yamashita, I., Iwahori, K. & Kumagai, S. Ferritin in the field of nanodevices.
Biochim. Biophys. Acta-Gen. Subj. 1800, 846-857 (2010).
84. Sarkar, J., Tang, S., Shahrjerdi, D. & Banerjee, S. K. Vertical flash memory with
protein-mediated assembly of nanocrystal floating gate. Appl. Phys. Lett. 90
(2007).
85. Meldrum, F. C., Wade, V. J., Nimmo, D. L., Heywood, B. R. & Mann, S.
Synthesis of inorganic nanophase materials in supramolecular protein cages.
Nature 349, 684-687 (1991).
86. Okuda, M., Iwahori, K., Yamashita, I. & Yoshimura, H. Fabrication of nickel and
chromium nanoparticles using the protein cage of apoferritin. Biotechnol. Bioeng.
84, 187-194 (2003).
87. Yamashita, I., Hayashi, J. & Hara, M. Bio-template synthesis of uniform CdSe
nanoparticles using cage-shaped protein, apoferritin. Chem. Lett. 33, 1158-1159
(2004).
28
88. Iwahori, K., Yoshizawa, K., Muraoka, M. & Yamashita, I. Fabrication of ZnSe
nanoparticles in the apoferritin cavity by designing a slow chemical reaction
system. Inorg. Chem. 44, 6393-6400 (2005).
89. Mayes, E., Bewick, A., Gleeson, D., Hoinville, J., Jones, R., Kasyutich, O.,
Nartowski, A., Warne, B., Wiggins, J. & Wong, K. K. W. Biologically derived
nanomagnets in self-organized patterned media. IEEE Trans. Magn. 39, 624-627
(2003).
90. Ensign, D., Young, M. & Douglas, T. Photocatalytic synthesis of copper colloids
from Cu(II) by the ferrihydrite core of ferritin. Inorg. Chem. 43, 3441-3446
(2004).
91. Zhang, L., Swift, J., Butts, C. A., Yerubandi, V. & Dmochowski, I. J. Structure
and activity of apoferritin-stabilized gold nanoparticles. J. Inorg. Biochem. 101,
1719-1729 (2007).
92. Fan, R. L., Chew, S. W., Cheong, V. V. & Orner, B. P. Fabrication of gold
nanoparticles inside unmodified horse spleen apoferritin. Small 6, 1483-1487
(2010).
93. Yoshimura, H., Scheybani, T., Baumeister, W. & Nagayama, K. Two-
dimensional protein array growth in thin-layers of protein solution on aqueous
subphases. Langmuir 10, 3290-3295 (1994).
94. Okuda, M., Kobayashi, Y., Suzuki, K., Sonoda, K., Kondoh, T., Wagawa, A.,
Kondo, A. & Yoshimura, H. Self-organized inorganic nanoparticle arrays on
protein lattices. Nano Lett. 5, 991-993 (2005).
29
95. DeLano, W. L. The PyMOL molecular graphics system. DeLano Scientific LLC,
Palo Alto, CA, USA. (2006).
30
Figure 1.1. Ribbon diagram of the α-carbon backbone of a HSAF subunit. Image was generated using PyMol.95
31
Chapter 2: Apoferritin-Stabilized Gold Nanoparticles
† The work in this chapter was done in collaboration with Joe Swift, Christopher A. Butts, and Vijay Yerubandi. We thank Doug Yates and Neelima Shah for TEM studies, Feng Gai and Jeffery Saven for access to biophysical instrumentation.
‡ This work is from Zhang, L.; Swift, J.; Butts, C. A.; Yerubandi, V.; Dmochowski, I. J. Structure and Activity of Apoferritin-Stabilized Gold Nanoparticles. J. Inorg. Biochem. 2007, 101, 1719-1729.
Au3+
+ NaBH4 NaBH4
NO2
O-
NH2
OH
HSAF
Au
32
1. Introduction
Biomolecule-inorganic nanoparticle assemblies are the subject of growing research,
and have become important for advances in catalysis, sensing, optoelectronics and
biomedical applications. Biomolecules such as oligonucleotides,1-8 peptides,9-12 and
proteins13-17 offer rich structural and functional diversity for synthesizing inorganic
nanoparticles of different compositions, sizes, and shapes, and controlling their
arrangement into complex architectures.18,19 The need for more sensitive bioanalytical
assays have made gold nanoparticles a particular focus of current research.20-29 Gold
nanoparticles are useful for applications such as colorimetric sensing,30-32 contrast-
enhanced optical imaging,33 and photothermal therapy,34 based on strong absorption from
the gold surface plasmon resonance (SPR) band, which is sensitive to local
environment.35-37 For uses of gold nanoparticles in aqueous solutions, it is important to
enhance physicochemical stability and solubility through passivation of the gold surface.
We chose HSAF as the passivating agent, because unlike commonly used stabilizers,
apoferritin is comprised of twenty-four four-helix bundles that self assemble to form a
12-nm spherical protein with an 8-nm diameter cavity. We show in this work that gold
nanoparticles form on the outer protein surface, leaving the cavity empty for subsequent
reactions. The differential reactivity of HSAF interior and exterior surfaces, and the
ability to store small molecules within the protein cavity,38-40 provide opportunities to
fabricate multifunctional nanomaterials with controlled molecular features. It is well
known that HSAF can react with Fe2+ and dioxygen to store as many as 4000 Fe3+ ions as
a ferrihydrite core within its interior.41 Apoferritin also templates the synthesis of various
33
inorganic particles, including metals, metal oxides, and semiconductors.42 Growth of
particles inside apoferritin requires addition of metal ions together with a reductant. The
protein shell is perforated by narrow hydrophilic channels at the threefold symmetry axes
that are believed to promote the transport of metal ions into the cavity.15,43,44 Previous
studies have added metal ranging from a few hundred to several thousand ions per HSAF
holoprotein, as summarized in Table 2.1. Transmission electron microscopy (TEM) data
indicated that the nanoparticles typically grew inside the apoferritin cavity. Importantly,
the metal ion sources were divalent cations, with two exceptions being Cr3+ and Au+.45,46
Although the formation of Au2S nanoparticles inside HSAF was recently described,45,46
the present study is the first detailed report on the addition and reduction of Au3+ to form
Au0 nanoparticles outside of HSAF. The main goal in the present manuscript is not to
prepare efficient biocatalysts but to get a fundamental understanding about the HSAF-
gold system, how HSAF stabilizes gold particles, and how one of the two different
synthetic routes used gives better surface passivation/better access to reactants so that
gold can still participate in catalysis. We elucidated the structure and activity of ferritin-
gold nanoparticles formed under various reaction conditions. Methods for characterizing
these protein-gold assemblies may be applied to other bio-inorganic systems.
Several groups have sought to develop methods for preparing spherical gold particles
in a range of diameters,37,47-50 as important physical properties vary with the dimensions
of the gold.35 The SPR maximum is normally between 520 and 530 nm for spherical gold
particles less than 20 nm in diameter, but this absorption feature can vary widely,
depending on the size, aspect ratio, and aggregation state of the particles.19 Whetten and
34
colleagues studied the absorption spectra of a series of nanocrystalline gold particles, and
found that the SPR band became identifiable only for structures greater than 2.0 nm in
diameter.35 Researchers have employed capping agents, such as alkanethiols, surfactants,
peptides, or proteins, in order to confine gold particle growth on the nanoscale and
passivate the surface.19 One example is the globular protein bovine serum albumin
(BSA).51-53 The José-Yacamán group reported the synthesis of gold nanoparticles smaller
than 2 nm through NaBH4 reduction in the presence of BSA. This eliminated the need for
stabilizing agents, such as citrate, or linker molecules, such as glutaric dialdehyde.51
Using a different approach, Singh et al. produced gold nanoparticles by in situ reduction
of gold ions in a foam generated by passing N2 gas through a BSA solution.52 The protein
played a role as both foaming and stabilizing agent. The photoreduction of gold
nanoparticles in the presence of BSA has also been described, and it was noted that in
addition to BSA, several other proteins can stabilize the gold.53 Besides the use of
proteins, water-soluble, biocompatible gold nanoparticles of different sizes have also
been synthesized using dendrimers,54 and short peptides such as tiopronin37 and
glutathione as capping agents,55 as well as a dodecapeptide identified from a phage
display library.56
In a broad context, apoferritin serves as a useful paradigm for studying protein-metal
nanoparticle interactions. We found that upon reduction of several thousand equiv of
AuCl4- in the presence of HSAF, apoferritin stabilized the gold particles in aqueous
solution through exterior surface interactions. This is similar to copper-HSAF systems,
where Cu0 particles larger than the ferritin cavity were stabilized by the protein in
35
water.57 We report several methods for probing protein-gold assemblies and compare the
properties of Au-HSAF particles synthesized under different conditions. Key features of
the gold-apoferritin complexes, such as stability and particle morphology, were affected
by the nucleation, growth, and final passivation of the particles, which varied with the
reaction conditions. We begin to consider these factors in this work.
2. Experimental
2.1. Reagents
All chemicals were of analytical reagent grade and used without further purification,
unless otherwise stated. Sigma: horse spleen apoferritin (obtained as a 53 mg/mL, 0.15 M
NaCl solution); horse spleen ferritin (HSF, obtained as an 85 mg/mL, 0.15 M NaCl
solution); 3-(N-morpholino)propanesulfonic acid (MOPS). Fisher: hydrogen tetrachloro-
aurate(III) hydrate (HAuCl4·xH2O, p.a.). Acros: gold(III) chloride (AuCl3), and 4-
nitrophenol. Aldrich: sodium borohydride (NaBH4). All solutions were prepared using
deionized water purified by Ultrapure Water Systems.
2.2. Sample preparation
AuCl4- was used as the gold ion source, while NaBH4 or MOPS was used as the
reducing agent. AuCl3 was also tested as the gold ion source following the same synthetic
procedure. HSAF (3.79 μL, 0.11 mM) was dissolved in a 2-mL buffer solution (0.4 M
NaCl, 0.1 M KH2PO4, pH 7.5). 16.66 μL of 0.1 M AuCl4- (4000 equiv per HSAF) was
added and the mixture was stirred for 1 h at rt. 50 μL of 0.1 M NaBH4 (3 equiv per
36
AuCl4-) or 83.3 μL of 1 M MOPS (50 equiv per AuCl4
-) was then added to reduce the
gold ions and the resulting solution was stirred for 1 h. This solution was centrifuged at
5000 rpm for 10 min (Eppendorf Centrifuge 5415D). The Au-HSAF product was
subsequently passed through a gel filtration column (Econo-Pac 10DG column, Bio-Rad;
eluent: 0.4 M NaCl, 0.1 M KH2PO4, pH 7.5).
2.3. Physical measurements
UV-visible (UV-Vis) absorption spectra and kinetics monitoring absorption changes
were recorded on an Agilent 8453 UV-Vis spectrophotometer with temperature controller
(25 °C) and magnetic stirrer (300 rpm, Agilent 89090A), using a quartz cuvette with 1-
cm path length. Chemical reactions were monitored at 530 nm to track the SPR band
associated with gold particle formation. Circular dichroism (CD) measurements were
obtained on a polarimeter (AVIV, model 62A DS) in the temperature range of 4-96 °C
with a 1-mm path length optical cell. High-resolution TEM data and images were
collected using a JEOL 2010F TEM with 0.23 nm point-to-point resolution at the Penn
Regional Nanotechnology Facility. Low-resolution TEM data and images were collected
at the Penn Biomedical Imaging Facility. Inductively coupled plasma optical emission
spectroscopy (ICP-OES) was performed by Galbraith Laboratories, Inc. to determine the
gold stoichiometry of Au-ferritin samples in solution. The Bio-Rad Protein Assay, based
on the Bradford method,58 was used to determine the concentration of HSAF.
2.4. Fast protein liquid chromatography (FPLC)
HSAF solution (2 mL, 0.208 μM) was stirred with AuCl4- (41.6 μL, 0.1 M) before
adding NaBH4 (124.8 μL, 0.1 M). Following reduction for 1 h, the sample was divided
37
into four equal portions and additional HSAF solution was added to three of those
portions, to achieve a final HSAF concentration of 0.417 μM, 0.833 μM and 1.875 μM.
The samples were then left at 4 °C overnight, centrifuged at 5000 rpm for 10 min, and
passed through a Superdex S200 column (Akta FPLC system, GE Healthcare) running
0.15 M NaCl, 0.02 M Tris, pH 7.4 buffer (TBS) and monitoring absorption at 280 nm.
In a parallel experiment, HSAF solution (2 mL, 0.208 μM) was stirred with AuCl4-
(41.6 μL, 0.1 M) before adding MOPS (208 μL, 1 M). Following reduction for 1 h, the
sample was divided into three equal portions and more HSAF was added to two of these
portions, to achieve a final HSAF concentration of 0.417 μM and 0.833 μM. The samples
were then left at 4 °C overnight, centrifuged at 5000 rpm for 10 min, and passed through
a Superdex S200 column running TBS and monitoring absorption at 280 nm. The
Superdex S200 column was calibrated by using high molecular weight standards as
described in the manufacturer’s protocol.
2.5. Catalytic properties
The reduction of 4-nitrophenol by NaBH4 in the presence of Au-HSAF catalyst was
studied in 2-mL solutions, with an identical procedure for NaBH4-reduced and MOPS-
reduced Au-HSAF. In a standard quartz cell with a 1-cm path length, 50 μL of 0.1 M
NaBH4 was mixed with 0.5 mL of 0.2 mM 4-nitrophenol. 1.25 mL of H2O was then
added. UV-Vis measurements were initiated with the addition of 0.2 mL of 50 nM Au-
HSAF solution. For experiments measuring the effects of extra HSAF on catalytic
activity, 3, 10, or 30 μL of 10 μM HSAF were premixed with Au-HSAF solution and
then added together to the solution mixture of 4-nitrophenol and NaBH4. The kinetics of
38
4-nitrophenol reduction were monitored by UV-Vis spectroscopy, with spectra recorded
at 15-s intervals in a scanning range of 200-800 nm at 25 °C.
2.6. Effect of reaction time on catalysis
NaBH4- or MOPS-reduced Au-HSAF samples were prepared following the synthetic
procedure specified in Section 2.2., the only difference being that after adding NaBH4 or
MOPS, the resulting solution was stirred for up to 24 h. The concentrations of the Au-
HSAF samples were adjusted to reach the same absorption intensity at 530 nm (O.D.530nm
= 0.66). To study the effect of reaction time on catalysis, in a standard quartz cell with a
1-cm path length, 50 μL of 0.1 M NaBH4 was mixed with 0.5 mL of 0.2 mM 4-
nitrophenol. 1.25 mL of H2O was then added. UV-Vis measurements were initiated with
the addition of 0.2 mL of the above Au-HSAF solution. The kinetics of 4-nitrophenol
reduction were monitored by UV-Vis spectroscopy, with spectra recorded at 15-s
intervals in a scanning range of 200-800 nm at 25 °C.
3. Results and Discussion
During sample preparation, HSAF was first stirred for 1 h with 4000 equiv AuCl4- at
25 °C to afford a clear, pale yellow solution. No gold was observed to precipitate during
this mixing process. The gold ions were then reduced by NaBH4 (3 equiv per AuCl4-) or
MOPS (50 equiv per AuCl4-) to yield Au0 particles, and the solution became reddish-
purple during a 1 h reaction. Reduction times greater than 1 h led to decreased SPR band
intensity. Au-HSAF was purified by gel filtration chromatography in ca. 30% (NaBH4-
reduced sample) and 50% (MOPS-reduced sample) overall yields, based on the starting
39
quantity of HSAF, as determined by the Bio-Rad Protein Assay. ICP-OES measurements
revealed that the average Au atom-to-HSAF 24-mer ratio was 1500:1.
High-resolution TEM was used to characterize the structure of the Au-HSAF samples.
The TEM images (Figures 2.1A and 2.1B) and the particle size histograms (Figures 2.1C
and 2.1D) showed broad particle size distributions with mean diameters of 3.6 ± 1.2 and
15.4 ± 4.5 nm for NaBH4-reduced and MOPS-reduced Au-HSAF, respectively. Energy
dispersive spectroscopy (EDS) positively identified gold in the nanoparticle samples
(Figure 2.1E). Particle atomic spacings were 0.245 nm (Figure 2.1F). The lack of
correlation between particle morphology (mean diameter and the range of sizes) and the
8-nm diameter apoferritin interior suggested that particle growth and stabilization did not
occur within the protein cavity. The average diameter of MOPS-reduced gold particles
was approximately twice the size of the 8-nm HSAF cavity, whereas the NaBH4-reduced
gold particles were less than half the diameter of the HSAF cavity.
Figure 2.2A shows a high-resolution TEM image of NaBH4-reduced Au-HSAF, which
was stained with 2% phosphotungstic acid (PTA). PTA was largely excluded from the
HSAF interior, and therefore provided negative contrast for the protein. This TEM image
provided strong evidence that even the smaller NaBH4-reduced gold nanoparticles (black
circles) did not form inside the cavity and were forced to reside at the outer HSAF
surface (larger white circles). Similar results were obtained from low-resolution TEM
images of MOPS-reduced Au-HSAF (Figure 2.2B).
We postulate that the gold is binding with highest probability to amino acids with
sulfur-, nitrogen-, and oxygen-containing side chains at pH 7.5, all of which are
40
positioned prominently on the outer surface of HSAF. These residues are expected to
bind gold ions, where gold is subsequently reduced and gold particles nucleated on the
outer protein shell. The anionic protein channels of ferritin promote the transport of
divalent cations into the cavity (Table 2.1), but are not well suited to transport larger gold
complexes that are neutral (AuCl3) or monoanionic (AuCl4-). Thus, in a competition
between Au3+ binding and reduction at the outer ferritin surface and metal ion transport to
the protein interior, it appears that most of the gold is reduced at residues on the protein
exterior. Mutagenesis experiments are ongoing in our lab to probe gold-ferritin
interactions more directly.
In addition to electrostatic considerations, we hypothesized that the different gold
particle size distributions should relate to differences in the kinetics of gold particle
formation. NaBH4 is a strong reducing agent that rapidly reduced AuCl4- and likely
nucleated numerous particles, thereby depleting the gold salt before larger particles were
able to form. MOPS is a much milder reducing agent, yielding slower gold particle
nucleation and growth that was strongly mediated by the protein.
Figure 2.3A shows the kinetics of AuCl4- reduction by NaBH4 in the presence and
absence of HSAF, monitored at 530 nm. In both samples, the SPR band reached a
maximum in five seconds, indicating kinetics at least as fast as the mixing time of our
apparatus. In the absence of HSAF, the absorption was observed to decrease from its
initial maximum due to the precipitation of gold colloid. However, the SPR band
observed in the presence of HSAF remained unchanged, indicating that the gold
nanoparticles were stabilized by the protein in solution. As a second control experiment,
41
AuCl4- was stirred with HSAF without the addition of the reducing agent NaBH4 (Figure
2.3A) or MOPS (Figure 2.3B). AuCl4- was not reduced on the same timescale of seconds-
to-minutes, as evidenced by the lack of an SPR band. In fact, with only HSAF as the
reductant, the formation of gold nanoparticles required weeks of stirring in the dark. In
parallel with these studies, we attempted to form gold nanoparticle-protein assemblies via
a stepwise process of first reducing Au3+ (16.66 μL, 0.1 M) in solution with NaBH4 (50
μL, 0.1 M) or MOPS (83.3 μL, 1 M) for 1 hour and then adding HSAF (3.79 μL, 0.11
mM). The solution mixture was stirred for 1 hour. Bulk gold precipitation was
unambiguously observed, and the sequential addition of HSAF failed to produce soluble
gold nanoparticles under these conditions.
Figure 2.3B shows the kinetics of AuCl4- reduction by MOPS in the presence and
absence of HSAF, which occurred on the timescale of minutes. Interestingly, AuCl4-
reduction was measurably faster in the absence of HSAF, but the absorption signal
quickly decreased in this case, indicating the formation and precipitation of insoluble
gold colloids. Reduction of AuCl4- in the presence of HSAF produced a stable SPR band
in aqueous solution, again indicating the stabilizing role of the protein. When stored at 4
°C in buffered solution, Au-HSAF samples remained stable for several months.
It is noteworthy that the formation of gold nanoparticles via the reaction between
AuCl4- and MOPS in the absence of other stabilizing agents has been recently
reported.59,60 The reduction of Au3+ by Good’s buffers such as MOPS was shown to
occur in 1-electron steps, Au(III) Au(II) Au(I) Au(0), with the buffer serving as
a mild reducing agent. Electrochemical experiments identified a single anodic peak for
42
MOPS ~ +800 mV vs. Ag/AgCl electrode. The production of soluble gold nanoparticles
in these reports instead of bulk gold relates to differences in their synthetic procedures,
including: (i) lower concentrations of reagents in solution, (ii) lack of stirring, and (iii)
lack of concentrated phosphate salt buffer. In the present study, MOPS is expected to
reduce Au3+ by similar 1-electron transfer steps. However, the bound protein may affect
the redox potential of the gold and in some cases block inner-sphere electron transfer
from MOPS to the gold ions, thereby slowing gold nanoparticle formation. It is of
considerable benefit that HSAF solubilizes the gold particles formed by MOPS over a
much wider range of pH than is possible with just buffer in solution.59,60
Figure 2.4A shows the absorption spectra of Au-HSAF samples reduced by NaBH4 or
MOPS, which were characterized by the gold SPR band.19,61 The SPR band of MOPS-
reduced Au-HSAF (λmax = 535 nm) showed a 5-nm red shift compared to NaBH4-reduced
Au-HSAF, with absorption centered at 530 nm. The effect of reducing conditions on the
SPR maximum may reflect small differences in particle morphology35,62 and aggregation
state.8 Light scattering was observed for all NaBH4-reduced gold samples, as evidenced
by an elevated baseline in the UV-Vis spectra (Figure 2.4). To probe the effects of
possible aggregation, we centrifuged the MOPS-reduced gold particles more rigorously,
at 13,000 rpm for 10 min. Figure 2.5A shows the absorption spectra of MOPS-reduced
Au-HSAF products centrifuged at 5000 rpm (solid line) and 13,000 rpm (dashed line). A
9-nm blue shift of the SPR maxima from 535 nm to 526 nm was observed when the
sample was centrifuged at the higher rate. The blue shift appeared to be caused entirely
by a loss of aggregation, not the isolation of smaller gold particles. A representative TEM
43
image and histogram of the particle sizes of MOPS-reduced Au-HSAF centrifuged at
13,000 rpm are shown in Figures 2.5B and 2.5C, which showed virtually identical
particle size distributions for samples centrifuged at 5000 and 13,000 rpm. These data
indicated that for these Au-HSAF particles, the positions of the SPR bands were more
sensitive to particle aggregation state than size.
To test whether the exterior surface of ferritin was sufficient for forming stable gold
particles in solution, AuCl4- was reduced with MOPS under the same conditions using
HSF instead of HSAF. We hypothesized that the presence of the HSF iron hydroxide
mineral core should have little effect on gold nanoparticle formation. Consistent with this
hypothesis, an SPR band of Au-HSF centered at 535 nm was recorded for the MOPS-
reduced sample, which agreed with Au-HSAF. However, for reasons that are still under
investigation, HSF proved to be worse at stabilizing gold nanoparticles in aqueous
solution.
Exploring the gold-protein interactions further, we tested whether the negatively
charged threefold symmetric channels of HSAF were important for gold nanoparticle
formation using neutral AuCl3 as the Au3+ source. Figure 2.4B shows the absorption
spectra of Au-HSAF samples reduced by NaBH4 or MOPS, which gave essentially the
same results for AuCl3 and AuCl4-. This indicated that the gold ion source, whether
neutral or negatively charged, had little influence on particle formation. These results,
together with the TEM data and HSF experiments, supported a model in which gold
particles were nucleated on the exterior protein surface.
44
We also varied gold ion concentrations at three pH values and monitored the intensity
of the SPR band, in order to probe a range of conditions for gold nanoparticle formation.
At a pH of 6.5, 7.5, and 8.5, we added 1000, 4000, or 7000 equiv AuCl4- per HSAF and
monitored the SPR band upon reduction with either NaBH4 or MOPS. A pH of 7.5 was
chosen for subsequent experiments because it represented the most biologically relevant
conditions while retaining the ability to synthesize Au0 nanoparticles in an efficient
manner. For both reducing agents, it was found that the SPR intensities were enhanced by
increasing from 1000 to 4000 equiv AuCl4-. For 7000 equiv AuCl4
-, the SPR intensities
decreased in general, more for NaBH4 than when MOPS was the reducing agent.
Therefore, 4000 equiv AuCl4- per HSAF was chosen for synthetic experiments in this
work.
To determine if the gold surface was completely passivated by the protein after
reduction, samples of Au-HSAF were prepared with excess gold (10,000 equiv) and
purified by FPLC. The pure Au-HSAF sample was subsequently incubated with
additional HSAF, and the resulting samples were repurified by size-exclusion
chromatography. The elution profiles of NaBH4-reduced and MOPS-reduced Au-HSAF
samples are shown in Figure 2.6. In both cases, FPLC traces monitoring absorbance at
280 nm showed elution peaks at 7.8 mL and 10.6 mL. The solution eluting at 7.8 mL was
red but did not precipitate, indicating the presence of gold nanoparticles stabilized by
associated HSAF. The peak at 10.6 mL did not show any absorbance at 530 nm and was
found by column calibration to correspond to the molecular weight of pure HSAF (~ 480
kD).41 The majority of the initial NaBH4-reduced Au-HSAF sample precipitated in the
45
column, as evidenced by the small absorption peaks of the eluted material (Figure 2.6A),
as well as the red residue that remained on the column. However, addition of extra HSAF
to these NaBH4-reduced samples greatly increased their stability on the column. The
absorption peaks in the case of the MOPS-reduced Au-HSAF were much larger, but were
not greatly enhanced by post-reduction addition of HSAF (Figure 2.6B). Therefore, we
determined that HSAF, with perhaps some contribution from MOPS, was able to
passivate the gold surface under slow reduction conditions. Under conditions of rapid
NaBH4 reduction, HSAF capped the gold nanoparticles much less effectively.
The thermal stability of Au-HSAF was investigated by circular dichroism
measurements in which the temperature was increased from 4 to 96 °C at a rate of 0.5 °C
min-1. Figure 2.7 shows the first differential of the ellipticity signal monitoring α-helical
content of HSAF at 222 nm. Denaturation or precipitation of the protein at higher
temperatures caused a reduction in signal. The melting temperature (Tm) was determined
from the midpoint of the melting curve, where the first derivative was the maximum.
With melting temperatures of 56 °C (NaBH4-reduced) and 62 °C (MOPS-reduced), Au-
HSAF had lower thermal stability than HSAF, Tm = 71 °C.
We also examined the stability of Au-HSAF over a wide range of pH values. Figure
2.8 shows the pH stability of HSAF and Au-HSAF reduced by NaBH4 or MOPS.
Buffered solutions at pH 7.5 were titrated with 1 M HCl and NaOH to achieve pH 2-12,
and centrifuged at 5000 rpm for 10 min to remove any precipitate before spectral
characterization. The stability of the Au-HSAF sample as a function of pH was
determined by measuring the change of the absorption intensity at 530 nm, while in a
46
control experiment HSAF alone in solution was monitored at 280 nm. The absorbance
intensities of all three samples remained generally unchanged at pH values above 7,
which indicated comparable stability in alkaline solutions. HSAF effectively stabilized
the gold particles in aqueous solution from pH 6-12.
When the pH of the solution was gradually lowered from 7 to 2 in increments of 1 pH
unit, the absorbance (at 280 nm) decreased for HSAF and even more significantly (at 530
nm) for both Au-HSAF samples. At pH 2, neither Au-HSAF sample showed an SPR
band, indicating the absence of soluble gold nanoparticles, whereas they both showed
protein absorption at 280 nm. Calculations based on the UV-Vis spectra revealed that
when lowering the pH from 7 to 2, most of the protein remained in solution for HSAF,
NaBH4-reduced Au-HSAF, and MOPS-reduced Au-HSAF samples. Protonation of the
exterior protein ligands at low pH should help to dissociate gold from HSAF and yield
selective gold precipitation. It is also worth noting that by pH 2, the HSAF 24-mer has
dissociated almost completely into its constituent 4-helix bundles,63 which are likely to be
less effective at stabilizing gold particles in solution.
As an indirect method for elucidating the protein-metal nanoparticle interactions, we
examined the Au-HSAF-catalyzed reduction of 4-nitrophenol by NaBH4. We first tested
the catalytic properties of NaBH4-reduced Au-HSAF by adding this sample to an aqueous
solution containing 4-nitrophenol and NaBH4. The rapid conversion of the 4-nitrophenol
to 4-aminophenol was monitored at 400 nm (Figure 2.9). To exclude the possibility that
this reduction might be activated by the protein alone, HSAF was added to 4-nitrophenol
and NaBH4 as a control. No change in the phenol absorbance was observed, as shown in
47
Figure 2.10. In the absence of the Au-HSAF catalyst, the absorbance at 400 nm also
remained unaltered, indicating that NaBH4 was unable to reduce 4-nitrophenol directly.
These results proved that NaBH4-reduced Au-HSAF catalyzed the reduction of 4-
nitrophenol by NaBH4. This reaction has been attributed to a gold particle-mediated
electron transfer from BH4- to 4-nitrophenol.64,65
Figure 2.10 shows the plot of 400-nm absorption versus time for the Au-HSAF-
catalyzed reduction of 4-nitrophenol by NaBH4 in the presence of different
concentrations of additional HSAF. Because the concentration of NaBH4 added to this
system was very high relative to 4-nitrophenol, it was reasonable to assume that the
concentration of NaBH4 remained constant during the reaction. Data for ln(A400) were
plotted versus time and linear fitting gave a reaction rate constant, k = 1.66 × 10-2 s-1.
With added HSAF (15, 50, and 150 nM), the reaction was slowed (k = 1.23 × 10-2, 1.04 ×
10-2, and 0.97 × 10-2 s-1), which suggested that the extra protein blocked the access of the
substrate to the gold particles, either by increasing the surface coverage or the
aggregation state of the particles. This indicated an association process between free
HSAF and Au-HSAF in the NaBH4-reduced gold samples, which was consistent with the
results of the FPLC study using the same samples.
The catalytic effects of the MOPS-reduced Au-HSAF sample were also examined.
Figure 2.10 shows a plot of 400-nm absorbance intensity versus time for the Au-HSAF-
catalyzed reduction of 4-nitrophenol by NaBH4. In contrast to the previous study,
substrate reduction was extremely slow, and adding more HSAF had no noticeable effect
on the reaction. The catalyst was presumed to be less accessible to 4-nitrophenol, based
48
on greater coverage of the gold surface with protein, likely in combination with greater
particle aggregation. Another contributing factor may have been the lower total surface
area of the larger MOPS-reduced gold particles, which numbered fewer in solution. The
catalysis results corroborated data from UV-Vis spectroscopy, reduction kinetics, FPLC,
thermal and pH stability studies, and provided strong evidence for MOPS-reduced Au-
HSAF samples being much better passivated at neutral pH than NaBH4-reduced Au-
HSAF.
As an indirect measure for whether the Au-HSAF reaction went to completion in the 1
h standard synthetic protocol, the gold-HSAF reaction solution was stirred for between 1
and 24 h after adding either NaBH4 or MOPS. Figure 2.11 shows the plots of 400-nm
absorption versus time for the reduction of 4-nitrophenol by NaBH4 using NaBH4-
reduced Au-HSAF particles that were reacted for 1, 6, 12, or 24 h. Nanoparticles formed
in the longer reactions gave virtually identical kinetics for the catalytic reduction of 4-
nitrophenol. Au-HSAF prepared by reduction with MOPS for up to 24 h exhibited
similarly little catalytic activity (Figure 2.11). This indicated that Au-HSAF particle
formation was indeed complete on the 1-hour time scale, under both sets of reducing
conditions. Data from the previous FPLC experiments showed that for NaBH4-reduced
samples, the gold surface remained unpassivated and reactive to additional HSAF. Thus,
the catalysis results helped to confirm that the difference in catalytic activity between
NaBH4- and MOPS-reduced Au-HSAF was due to intrinsic differences in surface
passivation between the two samples, and all of the HSAF in solution was reacted within
1 h during gold reduction and particle formation.
49
4. Conclusions
In summary, we report a practical method for synthesizing spherical gold nanoparticles
using horse spleen apoferritin as the stabilizing agent. Au3+ reduction occurred much
more rapidly with NaBH4 than MOPS. Gold nanoparticles were shown by TEM to reside
on the protein outer surface. A 5-nm red-shifted absorbance of MOPS-reduced gold
particles was based mostly on their aggregation in solution; more passivated and less
aggregated particles could be obtained through additional centrifugation. Several studies
indicated that NaBH4-reduced Au-HSAF was generally less stable than MOPS-reduced
samples. This was probably due to rapid reduction by NaBH4, which resulted in fewer
stabilizing interactions between the gold and protein. Incomplete gold surface coverage
explained why NaBH4-reduced Au-HSAF effectively catalyzed the reduction of 4-
nitrophenol by NaBH4. The conjugation of ferritin proteins to metal nanoparticles
provides new routes to further particle functionalization and assembly by labeling surface
amino acids.66,67 The techniques employed in this work for studying ferritin-gold
nanoparticle assemblies may be readily applied to other protein-metal colloid systems.
5. Future Directions
This chapter reports the stabilization of gold nanoparticles by HSAF. In this work,
AuCl4- was mixed with HSAF, and then the reducing agent (NaBH4 or MOPS) was
added. There was no dialysis before adding the reducing agents. And reduction was
performed in only one step, instead of achieving nucleation, followed by particle growth.
50
TEM showed that gold nanoparticles resided on the protein outer surface. This result
indicated that the majority part of AuCl4- was not loaded into HSAF during the mixing
step.
Inspired by two recent publications,68,69 I performed the dialysis before adding the
reducing agent ascorbic acid. In the experiment, HSAF (100 μM, 1 μL) was mixed with
AuCl4- (10 mM, 133 μL) in 0.5-mL phosphate buffer. After dialyzing overnight, ascorbic
acid (0.1 M, 40 μL) was added. TEM images of these samples were measured (Figure
2.12), which revealed that gold nanoparticles of a diameter smaller than the cavity of
HSAF were synthesized. We expect that this synthetic method is applicable to other
ferritin proteins, such as human ferritin and thermophilic ferritin.
6. References
1. Gugliotti, L. A., Feldheim, D. L. & Eaton, B. E. RNA-mediated metal-metal bond
formation in the synthesis of hexagonal palladium nanoparticles. Science 304,
850-852 (2004).
2. Gugliotti, L. A., Feldheim, D. L. & Eaton, B. E. RNA-mediated control of metal
nanoparticle shape. J. Am. Chem. Soc. 127, 17814-17818 (2005).
3. Liu, D. G., Gugliotti, L. A., Wu, T., Dolska, M., Tkachenko, A. G., Shipton, M.
K., Eaton, B. E. & Feldheim, D. L. RNA-mediated synthesis of palladium
nanoparticles on Au surfaces. Langmuir 22, 5862-5866 (2006).
4. Monson, C. F. & Woolley, A. T. DNA-templated construction of copper
nanowires. Nano Lett. 3, 359-363 (2003).
51
5. Becerril, H. A., Stoltenberg, R. M., Wheeler, D. R., Davis, R. C., Harb, J. N. &
Woolley, A. T. DNA-templated three-branched nanostructures for nanoelectronic
devices. J. Am. Chem. Soc. 127, 2828-2829 (2005).
6. Becerril, H. A., Ludtke, P., Willardson, B. M. & Woolley, A. T. DNA-templated
nickel nanostructures and protein assemblies. Langmuir 22, 10140-10144 (2006).
7. Storhoff, J. J., Elghanian, R., Mirkin, C. A. & Letsinger, R. L. Sequence-
dependent stability of DNA-modified gold nanoparticles. Langmuir 18, 6666-
6670 (2002).
8. Mirkin, C. A., Letsinger, R. L., Mucic, R. C. & Storhoff, J. J. A DNA-based
method for rationally assembling nanoparticles into macroscopic materials.
Nature 382, 607-609 (1996).
9. Whaley, S. R., English, D. S., Hu, E. L., Barbara, P. F. & Belcher, A. M.
Selection of peptides with semiconductor binding specificity for directed
nanocrystal assembly. Nature 405, 665-668 (2000).
10. Lee, S. W., Mao, C. B., Flynn, C. E. & Belcher, A. M. Ordering of quantum dots
using genetically engineered viruses. Science 296, 892-895 (2002).
11. Stevens, M. M., Flynn, N. T., Wang, C., Tirrell, D. A. & Langer, R. Coiled-coil
peptide-based assembly of gold nanoparticles. Adv. Mater. 16, 915-918 (2004).
12. Djalali, R., Samson, J. & Matsui, H. Doughnut-shaped peptide nano-assemblies
and their applications as nanoreactors. J. Am. Chem. Soc. 126, 7935-7939 (2004).
13. Ueno, T., Koshiyama, T., Tsuruga, T., Goto, T., Kanamaru, S., Arisaka, F. &
Watanabe, Y. Bionanotube tetrapod assembly by in situ synthesis of a gold
52
nanocluster with (Gp5-His6)3 from bacteriophage T4. Angew. Chem. Int. Ed. 45,
4508-4512 (2006).
14. Blum, A. S., Soto, C. M., Wilson, C. D., Cole, J. D., Kim, M., Gnade, B.,
Chatterji, A., Ochoa, W. F., Lin, T. W., Johnson, J. E. & Ratna, B. R. Cowpea
mosaic virus as a scaffold for 3-D patterning of gold nanoparticles. Nano Lett. 4,
867-870 (2004).
15. Takahashi, T. & Kuyucak, S. Functional properties of threefold and fourfold
channels in ferritin deduced from electrostatic calculations. Biophys. J. 84, 2256-
2263 (2003).
16. Srivastava, S., Verma, A., Frankamp, B. L. & Rotello, V. M. Controlled assembly
of protein-nanoparticle composites through protein surface recognition. Adv.
Mater. 17, 617-621 (2005).
17. Yang, J., Mayer, M., Kriebel, J. K., Garstecki, P. & Whitesides, G. M. Self-
assembled aggregates of IgGs as templates for the growth of clusters of gold
nanoparticles. Angew. Chem. Int. Ed. 43, 1555-1558 (2004).
18. Katz, E. & Willner, I. Integrated nanoparticle-biomolecule hybrid systems:
synthesis, properties, and applications. Angew. Chem. Int. Ed. 43, 6042-6108
(2004).
19. Daniel, M. C. & Astruc, D. Gold nanoparticles: assembly, supramolecular
chemistry, quantum-size-related properties, and applications toward biology,
catalysis, and nanotechnology. Chem. Rev. 104, 293-346 (2004).
53
20. Lu, Y. & Liu, J. W. Functional DNA nanotechnology: emerging applications of
DNAzymes and aptamers. Curr. Opin. Biotechnol. 17, 580-588 (2006).
21. Liu, J. W. & Lu, Y. Colorimetric biosensors based on DNAzyme-assembled gold
nanoparticles. J. Fluorescence 14, 343-354 (2004).
22. Park, S. J., Taton, T. A. & Mirkin, C. A. Array-based electrical detection of DNA
with nanoparticle probes. Science 295, 1503-1506 (2002).
23. Park, S. J., Lazarides, A. A., Storhoff, J. J., Pesce, L. & Mirkin, C. A. The
structural characterization of oligonucleotide-modified gold nanoparticle
networks formed by DNA hybridization. J. Phys. Chem. B 108, 12375-12380
(2004).
24. Haes, A. J., Zou, S. L., Schatz, G. C. & Van Duyne, R. P. A nanoscale optical
biosensor: the long range distance dependence of the localized surface plasmon
resonance of noble metal nanoparticles. J. Phys. Chem. B 108, 109-116 (2004).
25. Hao, E., Schatz, G. C. & Hupp, J. T. Synthesis and optical properties of
anisotropic metal nanoparticles. J. Fluorescence 14, 331-341 (2004).
26. Haes, A. J., Haynes, C. L., McFarland, A. D., Schatz, G. C., Van Duyne, R. R. &
Zou, S. L. Plasmonic materials for surface-enhanced sensing and spectroscopy.
MRS Bull. 30, 368-375 (2005).
27. Yonzon, C. R., Zhang, X., Zhao, J. & Van Duyne, R. P. Surface-enhanced
nanosensors. Spectroscopy 22, 42-56 (2007).
28. Kohli, P., Wirtz, M. & Martin, C. R. Nanotube membrane based biosensors.
Electroanalysis 16, 9-18 (2004).
54
29. Wirtz, M., Yu, S. F. & Martin, C. R. Template synthesized gold nanotube
membranes for chemical separations and sensing. Analyst 127, 871-879 (2002).
30. Liu, J. W. & Lu, Y. A colorimetric lead biosensor using DNAzyme-directed
assembly of gold nanoparticles. J. Am. Chem. Soc. 125, 6642-6643 (2003).
31. Liu, J. W. & Lu, Y. Adenosine-dependent assembly of aptazyme-functionalized
gold nanoparticles and its application as a colorimetric biosensor. Anal. Chem. 76,
1627-1632 (2004).
32. Reynolds, R. A., Mirkin, C. A. & Letsinger, R. L. Homogeneous, nanoparticle-
based quantitative colorimetric detection of oligonucleotides. J. Am. Chem. Soc.
122, 3795-3796 (2000).
33. Chen, J., Saeki, F., Wiley, B. J., Cang, H., Cobb, M. J., Li, Z. Y., Au, L., Zhang,
H., Kimmey, M. B., Li, X. D. & Xia, Y. Gold nanocages: bioconjugation and
their potential use as optical imaging contrast agents. Nano Lett. 5, 473-477
(2005).
34. Chen, J. Y., Wiley, B., Li, Z. Y., Campbell, D., Saeki, F., Cang, H., Au, L., Lee,
J., Li, X. D. & Xia, Y. N. Gold nanocages: engineering their structure for
biomedical applications. Adv. Mater. 17, 2255-2261 (2005).
35. Alvarez, M. M., Khoury, J. T., Schaaff, T. G., Shafigullin, M. N., Vezmar, I. &
Whetten, R. L. Optical absorption spectra of nanocrystal gold molecules. J. Phys.
Chem. B 101, 3706-3712 (1997).
55
36. Tkachenko, A. G., Xie, H., Coleman, D., Glomm, W., Ryan, J., Anderson, M. F.,
Franzen, S. & Feldheim, D. L. Multifunctional gold nanoparticle-peptide
complexes for nuclear targeting. J. Am. Chem. Soc. 125, 4700-4701 (2003).
37. Templeton, A. C., Cliffel, D. E. & Murray, R. W. Redox and fluorophore
functionalization of water-soluble, tiopronin-protected gold clusters. J. Am. Chem.
Soc. 121, 7081-7089 (1999).
38. Aime, S., Frullano, L. & Crich, S. G. Compartmentalization of a gadolinium
complex in the apoferritin cavity: a route to obtain high relaxivity contrast agents
for magnetic resonance imaging. Angew. Chem. Int. Ed. 41, 1017-1019 (2002).
39. Dominguez-Vera, J. M. Iron(III) complexation of Desferrioxamine B
encapsulated in apoferritin. J. Inorg. Biochem. 98, 469-472 (2004).
40. Simsek, E. & Kilic, M. A. Magic ferritin: a novel chemotherapeutic encapsulation
bullet. J. Magn. Magn. Mater. 293, 509-513 (2005).
41. Theil, E. C. Ferritin: structure, gene regulation, and cellular function in animals,
plants, and microorganisms. Annu. Rev. Biochem. 56, 289-315 (1987).
42. Yoshimura, H. Protein-assisted nanoparticle synthesis. Colloids Surf. A 282, 464-
470 (2006).
43. Treffry, A., Bauminger, E. R., Hechel, D., Hodson, N. W., Nowik, I., Yewdall, S.
J. & Harrison, P. M. Defining the roles of the threefold channels in iron uptake,
iron oxidation and iron-core formation in ferritin: a study aided by site-directed
mutagenesis. Biochem. J. 296, 721-728 (1993).
56
44. Pead, S., Durrant, E., Webb, B., Larsen, C., Heaton, D., Johnson, J. & Watt, G. D.
Metal ion binding to apo, holo, and reconstituted horse spleen ferritin. J. Inorg.
Biochem. 59, 15-27 (1995).
45. Okuda, M., Iwahori, K., Yamashita, I. & Yoshimura, H. Fabrication of nickel and
chromium nanoparticles using the protein cage of apoferritin. Biotechnol. Bioeng.
84, 187-194 (2003).
46. Yoshizawa, K., Iwahori, K., Sugimoto, K. & Yamashita, I. Fabrication of gold
sulfide nanoparticles using the protein cage of apoferritin. Chem Lett 35, 1192-
1193 (2006).
47. Turkevich, J., Stevenson, P. C. & Hillier, J. A study of the nucleation and growth
processes in the synthesis of colloidal gold. Discuss. Faraday Soc. 11, 55-74
(1951).
48. Frens, G. Controlled nucleation for regulation of particle-size in monodisperse
gold suspensions. Nature Phys. Sci. 241, 20-22 (1973).
49. Brust, M., Walker, M., Bethell, D., Schiffrin, D. J. & Whyman, R. Synthesis of
thiol-derivatized gold nanoparticles in a 2-phase liquid-liquid system. J. Chem.
Soc., Chem. Commun., 801-802 (1994).
50. Templeton, A. C., Chen, S. W., Gross, S. M. & Murray, R. W. Water-soluble,
isolable gold clusters protected by tiopronin and coenzyme A monolayers.
Langmuir 15, 66-76 (1999).
57
51. Burt, J. L., Gutierrez-Wing, C., Miki-Yoshida, M. & Jose-Yacaman, M. Noble-
metal nanoparticles directly conjugated to globular proteins. Langmuir 20, 11778-
11783 (2004).
52. Singh, A. V., Bandgar, B. M., Kasture, M., Prasad, B. L. V. & Sastry, M.
Synthesis of gold, silver and their alloy nanoparticles using bovine serum albumin
as foaming and stabilizing agent. J. Mater. Chem. 15, 5115-5121 (2005).
53. Chiu, T. C., Chiou, S. H., Hsieh, M. M., Chen, Y. T. & Chang, H. T.
Photosynthesis of gold nanoparticles in presence of proteins. J. Nanosci.
Nanotechnol. 5, 2128-2132 (2005).
54. Kim, Y. G., Oh, S. K. & Crooks, R. M. Preparation and characterization of 1-2
nm dendrimer-encapsulated gold nanoparticles having very narrow size
distributions. Chem. Mater. 16, 167-172 (2004).
55. Schaaff, T. G., Knight, G., Shafigullin, M. N., Borkman, R. F. & Whetten, R. L.
Isolation and selected properties of a 10.4 kDa gold:glutathione cluster
compound. J. Phys. Chem. B 102, 10643-10646 (1998).
56. Slocik, J. M., Stone, M. O. & Naik, R. R. Synthesis of gold nanoparticles using
multifunctional peptides. Small 1, 1048-1052 (2005).
57. Ensign, D., Young, M. & Douglas, T. Photocatalytic synthesis of copper colloids
from Cu(II) by the ferrihydrite core of ferritin. Inorg. Chem. 43, 3441-3446
(2004).
58
58. Bradford, M. M. Rapid and sensitive method for quantitation of microgram
quantities of protein utilizing principle of protein-dye binding. Anal. Biochem. 72,
248-254 (1976).
59. Xie, J., Lee, J. Y. & Wang, D. I. C. Seedless, surfactantless, high-yield synthesis
of branched gold nanocrystals in HEPES buffer solution. Chem. Mater. 19, 2823-
2830 (2007).
60. Habib, A., Tabata, M. & Wu, Y. G. Formation of gold nanoparticles by Good's
buffers. Bull. Chem. Soc. Jpn. 78, 262-269 (2005).
61. Link, S. & El-Sayed, M. A. Spectral properties and relaxation dynamics of
surface plasmon electronic oscillations in gold and silver nanodots and nanorods.
J. Phys. Chem. B 103, 8410-8426 (1999).
62. Mandal, S., Arumugam, S. K., Adyanthaya, S. D., Pasricha, R. & Sastry, M. Use
of aqueous foams for the synthesis of gold nanoparticles of variable morphology.
J. Mater. Chem. 14, 43-47 (2004).
63. Crichton, R. R. & Bryce, C. F. A. Subunit interactions in horse spleen apoferritin:
dissociation by extremes of pH. Biochem. J. 133, 289-299 (1973).
64. Hayakawa, K., Yoshimura, T. & Esumi, K. Preparation of gold-dendrimer
nanocomposites by laser irradiation and their catalytic reduction of 4-nitrophenol.
Langmuir 19, 5517-5521 (2003).
65. Liu, J. C., Qin, G. W., Raveendran, P. & Kushima, Y. Facile "green" synthesis,
characterization, and catalytic function of beta-D-glucose-stabilized Au
nanocrystals. Chem. Eur. J. 12, 2132-2138 (2006).
59
66. Li, M. & Mann, S. DNA-directed assembly of multifunctional nanoparticle
networks using metallic and bioinorganic building blocks. J. Mater. Chem. 14,
2260-2263 (2004).
67. Flenniken, M. L., Willits, D. A., Harmsen, A. L., Liepold, L. O., Harmsen, A. G.,
Young, M. J. & Douglas, T. Melanoma and lymphocyte cell-specific targeting
incorporated into a heat shock protein cage architecture. Chem. Biol. 13, 161-170
(2006).
68. Fan, R. L., Chew, S. W., Cheong, V. V. & Orner, B. P. Fabrication of gold
nanoparticles inside unmodified horse spleen apoferritin. Small 6, 1483-1487
(2010).
69. Shin, Y. S., Dohnalkova, A. & Lin, Y. H. Preparation of homogeneous gold-silver
alloy nanoparticles using the apoferritin cavity as a nanoreactor. J. Phys. Chem. C
114, 5985-5989 (2010).
70. Ueno, T., Suzuki, M., Goto, T., Matsumoto, T., Nagayama, K. & Watanabe, Y.
Size-selective olefin hydrogenation by a Pd nanocluster provided in an apo-
ferritin cage. Angew. Chem. Int. Ed. 43, 2527-2530 (2004).
71. Galvez, N., Sanchez, P. & Dominguez-Vera, J. M. Preparation of Cu and CuFe
Prussian Blue derivative nanoparticles using the apoferritin cavity as nanoreactor.
Dalton Trans., 2492-2494 (2005).
72. Galvez, N., Sanchez, P., Dominguez-Vera, J. M., Soriano-Portillo, A., Clemente-
Leon, M. & Coronado, E. Apoferritin-encapsulated Ni and Co superparamagnetic
nanoparticles. J. Mater. Chem. 16, 2757-2761 (2006).
60
73. Douglas, T., Dickson, D. P. E., Betteridge, S., Charnock, J., Garner, C. D. &
Mann, S. Synthesis and structure of an iron(III) sulfide-ferritin bioinorganic
nanocomposite. Science 269, 54-57 (1995).
74. Meldrum, F. C., Heywood, B. R. & Mann, S. Magnetoferritin: in vitro synthesis
of a novel magnetic protein. Science 257, 522-523 (1992).
75. Meldrum, F. C., Wade, V. J., Nimmo, D. L., Heywood, B. R. & Mann, S.
Synthesis of inorganic nanophase materials in supramolecular protein cages.
Nature 349, 684-687 (1991).
76. Hainfeld, J. F. Uranium-loaded apoferritin with antibodies attached: molecular
design for uranium neutron-capture therapy. Proc. Natl. Acad. Sci. USA 89,
11064-11068 (1992).
77. Meldrum, F. C., Douglas, T., Levi, S., Arosio, P. & Mann, S. Reconstitution of
manganese oxide cores in horse spleen and recombinant ferritins. J. Inorg.
Biochem. 58, 59-68 (1995).
78. Zhang, B., Harb, J. N., Davis, R. C., Kim, J. W., Chu, S. H., Choi, S., Miller, T. &
Watt, G. D. Kinetic and thermodynamic characterization of the cobalt and
manganese oxyhydroxide cores formed in horse spleen ferritin. Inorg. Chem. 44,
3738-3745 (2005).
79. Douglas, T. & Stark, V. T. Nanophase cobalt oxyhydroxide mineral synthesized
within the protein cage of ferritin. Inorg. Chem. 39, 1828-1830 (2000).
80. Wong, K. K. W. & Mann, S. Biomimetic synthesis of cadmium sulfide-ferritin
nanocomposites. Adv. Mater. 8, 928-932 (1996).
61
81. Yamashita, I., Hayashi, J. & Hara, M. Bio-template synthesis of uniform CdSe
nanoparticles using cage-shaped protein, apoferritin. Chem. Lett. 33, 1158-1159
(2004).
82. Iwahori, K., Yoshizawa, K., Muraoka, M. & Yamashita, I. Fabrication of ZnSe
nanoparticles in the apoferritin cavity by designing a slow chemical reaction
system. Inorg. Chem. 44, 6393-6400 (2005).
62
Table 2.1. Literature survey of inorganic complexes formed with HSAF. ________________________________________________________________________
Material Metal ion source Metal:HSAF ratioa Reference ________________________________________________________________________
metals Pd PdCl4
2- 500 70 Cu Cu2+ 225 71 Co Co2+ not specified 72 Ni Ni2+ not specified 72
metal oxides, hydroxides, and oxyhydroxides FeOOH Fe2+ 4500 73 Fe3O4 Fe2+ 3000 74 U oxide UO2
2+ 4000 75 U oxo complex UO2
2+ 800 76 MnOOH Mn2+ 3000 77 MnOOH Mn2+ 1200 78 CoOOH Co2+ 2250 79 CoOOH Co2+ 1500 78 Cr hydroxide Cr3+ 4800 45 Ni hydroxide Ni2+ 8000 45
semiconductors CdS Cd2+ 55 ~ 275 80 CdSe Cd2+ 1000 81 ZnSe Zn2+ 1500 82 Au2S Au+ 3000 46
________________________________________________________________________ a: Number indicates initially added metal ions per HSAF 24-mer protein.
63
C D
E Figure 2.1. (A) TEM image of Au-HSAF reduced by NaBH4. (B) TEM image of Au-HSAF reduced by MOPS. (C) Histogram of the particle sizes of the sample in (A), 222 particles analyzed. (D) Histogram of the particle sizes of the sample in (B), 110 particles analyzed. (E) Energy dispersive spectrum of representative Au-HSAF sample placed on copper specimen grid. (F) The atomic gold lattice spacings of an Au-HSAF sample.
0 1 2 3 4 5 6 7 80
1020304050607080
No.
of P
artic
les
Diameter / nm3 6 9 12 15 18 21 24 27
05
1015202530
No.
of P
artic
les
Diameter / nm
64
Figure 2.2. TEM images showing gold nanoparticles (dark black circles) associated with the exterior surface of HSAF (white rings, examples outlined in red). AuCl4
- was reduced by (A) NaBH4 or (B) MOPS, resulting in gold particles with different size distributions. (A) was imaged at 200 kV using phosphotungstic acid (PTA) stain to create negative contrast for protein. (B) was imaged at 120 kV using an uranyl acetate stain; the larger white circles are an artifact caused by the stain drying process.
65
A
B Figure 2.3. (A) Kinetics traces of AuCl4
- reduction by NaBH4 in the presence () or absence () of HSAF. Kinetics trace of AuCl4
- stirred with HSAF without adding NaBH4 (); no particle formation was observed to occur on this timescale. (B) Kinetics traces of AuCl4
- reduction by MOPS in the presence () or absence () of HSAF. Kinetics trace of AuCl4
- stirred with HSAF without MOPS (). Reactions were monitored at 530 nm.
0 200 400 600 8000.0
0.5
1.0
1.5
2.0
2.5
no reducing agents
in the absence of HSAF
in the presence of HSAF
O.D
. 530n
m
Time / s0 200 400 600 800
0.0
0.5
1.0
1.5
2.0
2.5
0 200 400 600 8000.0
0.5
1.0
1.5
2.0
2.5
0 10 20 30 400.0
0.5
1.0
1.5
2.0
2.5
O.D
. 530n
m
Time / s0 10 20 30 40
0.0
0.5
1.0
1.5
2.0
2.5
in the absence of HSAF
in the presence of HSAF
0 10 20 30 400.0
0.5
1.0
1.5
2.0
2.5
no reducing agents
66
A
B Figure 2.4. (A) Absorption spectra of Au-HSAF reduced by NaBH4 (solid line) or MOPS (dashed line) using AuCl4
- as the gold ion source. (B) Absorption spectra of Au-HSAF reduced by NaBH4 (solid line) or MOPS (dashed line) using AuCl3 as the gold ion source.
200 400 600 8000.0
0.5
1.0
Abso
rptio
n / a
.u.
Wavelength / nm
reduced by NaBH 4
200 400 600 8000.0
0.2
0.4
0.6
0.8
1.0
reduced by MOPS
200 300 400 500 600 700 800 9000.0
0.2
0.4
0.6
0.8
1.0
Abso
rptio
n / a
.u.
Wavelength / nm
reduced by NaBH 4
200 400 600 8000.0
0.2
0.4
0.6
0.8
1.0
reduced by MOPS
67
A
× 5
B C Figure 2.5. (A) Absorption spectra of MOPS-reduced Au-HSAF products centrifuged at 5000 rpm (solid line) and 13,000 rpm (dashed line), respectively. (B) TEM image of MOPS-reduced Au-HSAF centrifuged at 13,000 rpm. (C) Histogram of the particle size of the sample in (B) (192 particles analyzed).
0 7 14 21 280
102030405060
No.
of P
artic
les
Diameter / nm
200 300 400 500 600 700 800 9000.0
0.2
0.4
0.6
0.8
1.0
Abs
orpt
ion
/ a.u
.
Wavelength / nm
centrifuged at 13000 rpm
200 400 600 8000.0
0.2
0.4
0.6
0.8
1.09 nm
centrifuged at 5000 rpm
68
A
B
Figure 2.6. (A) FPLC elution profiles of NaBH4-reduced Au-HSAF (0.208 μM) with 0 μM HSAF (solid line), 0.417 μM HSAF (dashed line), 0.833 μM HSAF (dotted line) and 1.875 μM HSAF (dashed dotted line). (B) FPLC elution profiles of 0.208 μM MOPS-reduced Au-HSAF with 0 μM HSAF (solid line), 0.417 μM HSAF (dashed line) and 0.833 μM HSAF (dotted line). The traces were scaled such that the area of the gold-protein peak (at 7.8 mL) was proportional to the initial UV-Vis absorbance measurement recorded at 530 nm. FPLC traces were monitored at 280 nm, where both the gold and protein absorb.
6 8 10 12 140
20
40
60
80
100
120
140
160
180
200
Abs
orpt
ion
/ a.u
.
Elution Volume / ml
0 µM HSAF
6 8 10 12 140
20
40
60
80
100
120
140
160
180
200
0.417 µM HSAF
6 8 10 12 140
102030405060708090
100110120130140150160170180190200
0.833 µM HSAF
5 6 7 8 9 10 11 12 13 14 150
40
80
120
160
200
1.875 µM HSAF
6 8 10 12 140
20
40
60
80
100
Abs
orpt
ion
/ a.u
.
Elution Volume / ml
0 µM HSAF
6 8 10 12 140
20
40
60
80
100
0.417 µM HSAF
5 6 7 8 9 10 11 12 13 14 150
20
40
60
80
100
0.833 µM HSAF
69
Figure 2.7. Thermal stability studies of HSAF (solid line), Au-HSAF reduced by NaBH4 (dashed line) and MOPS (dotted line). The curves were normalized to the same intensity.
30 40 50 60 70 800.0
0.2
0.4
0.6
0.8
1.0
Firs
t Der
iv. o
f Ellip
ticity
Temperature / oC
HSAF
30 40 50 60 70 800.0
0.2
0.4
0.6
0.8
1.0
NaBH 4-reduced Au-HSAF
30 40 50 60 70 800.0
0.2
0.4
0.6
0.8
1.0
MOPS-reduced Au-HSAF
70
Figure 2.8. pH stability of HSAF () and Au-HSAF reduced by NaBH4 () and MOPS (). For HSAF, the normalized absorbance intensities were monitored at 280 nm; for Au-HSAF, the normalized absorbance intensities were monitored at 530 nm. All three curves were normalized to the highest band intensities of each sample to facilitate comparisons of pH stability.
2 4 6 8 10 120.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
Frac
tion
of O
rigin
al O
.D.
pH
HSAF
2 4 6 8 10 12
0.0
0.2
0.4
0.6
0.8
1.0
NaBH 4-reduced Au-HSAF
2 4 6 8 10 120.0
0.2
0.4
0.6
0.8
1.0
MOPS-reduced Au-HSAF
71
Figure 2.9. UV-Vis absorption spectra for the reduction of 4-nitrophenol (50 μM) by NaBH4 (2.5 mM) in the presence of NaBH4-reduced Au-HSAF (5 nM protein) as catalyst. Solid line, initial spectrum; dashed line, final spectrum.
200 250 300 350 400 450 500 550 600 6500.0
0.2
0.4
0.6
0.8
1.0A
bsor
ptio
n
Wavelength / nm
200 250 300 350 400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
NO2-O
NH2O
Au-HSAF
+NaBH4
H
200 300 400 500 6000.0
0.2
0.4
0.6
0.8
1.0
72
Figure 2.10. Plots of 400-nm absorption versus time for the reduction of 4-nitrophenol (50 μM) by NaBH4 (2.5 mM) using NaBH4-reduced Au-HSAF (5 nM protein) as catalyst in the absence () and presence of different concentrations of additional HSAF: 15 nM (); 50 nM (); 150 nM (); and, using MOPS-reduced Au-HSAF (5 nM protein) as catalyst (). Reaction of 4-nitrophenol and NaBH4 in the presence of HSAF alone (5 nM protein, no gold) was included as a control (). Values in parentheses represent the final reaction concentrations.
0 100 200 300 4000.0
0.2
0.4
0.6
0.8
1.0O
.D. 40
0nm
Time / s
HSAF: 0 nM
0 100 200 300 4000.0
0.2
0.4
0.6
0.8
1.0
HSAF: 15 nM
0 100 200 300 4000.0
0.2
0.4
0.6
0.8
1.0
NaBH 4-reduced Au-HSAF:
HSAF: 50 nM
0 100 200 300 4000.0
0.2
0.4
0.6
0.8
1.0
HSAF: 150 nM
0 100 200 300 4000.0
0.2
0.4
0.6
0.8
1.0
MOPS-reduced Au-HSAF
0 100 200 300 4000.0
0.2
0.4
0.6
0.8
1.0
HSAF alone
73
Figure 2.11. Plots of 400-nm absorption versus time for the reduction of 4-nitrophenol (50 μM) by NaBH4 (2.5 mM) using NaBH4-reduced Au-HSAF, which was synthesized with different reduction times: 1 h (); 6 h (); 12 h (); 24 h (); and using MOPS-reduced Au-HSAF (), which was synthesized with 24-h reduction time.
0 50 100 150 200
0.2
0.4
0.6
0.8
1.0
1.2
MOPS-reduced Au-HSAF:
NaBH 4-reduced Au-HSAF:
O.D
. 400n
m
Time / s
reduction time: 1 h
0 50 100 150 200
0.2
0.4
0.6
0.8
1.0
1.2
reduction time: 6 h
0 50 100 150 200
0.2
0.4
0.6
0.8
1.0
1.2
reduction time: 12 h
0 50 100 150 200
0.2
0.4
0.6
0.8
1.0
1.2
reduction time: 24 h
0 50 100 150 200
0.2
0.4
0.6
0.8
1.0
1.2
reduction time: 24 h
74
Figure 2.12. Two representative TEM images of HSAF-encapsulated gold nanoparticles reduced by ascorbic acid.
75
Chapter 3: Binding of Gd-Chelates to Ferritin
† The work in this chapter was done in collaboration with Jasmina Cheung-Lau and David Hao. We thank Paolo Arosio for providing HuHF and HuLF cDNA, Elizabeth Theil for providing the FHF-L134P cDNA, Feng Gai for use of the CD spectrometer, Jeffery Saven for use of the fluorometer, Andrew Tsourkas for use of the relaxometer, and Steven Pickup for the MRI measurement.
‡ This work is from Zhang, L.; Cheung-Lau, J.; Hao, D.; Dmochowski, I. J. Gd(DTPA-BMA) Binds Apoferritin, Slowing Iron Mineralization. J. Am. Chem. Soc. 2011, submitted.
76
1. Introduction
Several transition metal complexes have been found to serve useful roles as medical
therapeutic and diagnostic agents.1-5 Nevertheless, the rational design of inorganic drugs
remains an elusive challenge. The ligands bound to the metal ion (i.e., the primary
coordination sphere) can tune the kinetic and thermodynamic stability of the complex, as
well as reactions with specific protein or nucleic acid targets. Evidence is growing that
many metallodrugs also interact with common human serum proteins in vivo.6-9 This
interaction alters drug bioavailability in unpredictable ways, and may produce unwanted
drug side effects.10 Such interactions particularly impact the design of metallo-diagnostic
agents, where high clearance rates (via the kidney or liver) are typically desirable. Prime
examples are gadolinium chelates used for magnetic resonance imaging (MRI), which are
delivered i.v. to patients. Roughly half of the ~30 million MRI procedures performed in
the U.S. annually involve contrast-enhancing agents, e.g., Gd(DTPA)2- (gadopentetate
dimeglumine) or Gd(DTPA-BMA) (gadodiamide) (Figure 3.1). Since the development of
Gd(DTPA)2- and Gd(DTPA-BMA) more than 20 years ago, most reports have suggested
that these drugs are inert in vivo.11 However, Caravan et al. recently provided conclusive
evidence of weak binding to many different serum proteins, including human serum
albumin (HSA) and lysozyme.12 Here, we investigate whether these commonly employed
Gd-chelates also bind the serum enzyme apoferritin and can affect its iron mineralization
activity.
Renewed interest in the biological interactions of Gd-chelates has arisen from a
cutaneous, sclerodoma-like disease named nephrogenic systemic fibrosis (NSF), first
77
recognized in 2000.13 Figure 3.2 shows the sclerotic plaque with calcinosis on the back
arm of the patient as an example of NSF.14 NSF has been observed in a small percentage
of patients with severe or end-stage renal failure who were administered Gd-chelates,
including Gd(DTPA)2- and Gd(DTPA-BMA).15 Skin biopsies from NSF patients indicate
high concentrations of Gd in the diseased tissue.16 A hazard for patients who suffer from
renal insufficiency is that Gd-chelates circulate in the body for an extended time (34 ± 23
h), compared to 1.3 ± 0.3 h elimination half-life in healthy volunteers.17 Long circulation
times are likely to promote serum protein interactions. A retrospective study18 of ten
patients administered gadodiamide found an increase of serum iron from 650 to 990
ng/mL and an increase of serum ferritin from 428 to 1640 ng/mL (data are median
values). This suggested to us a possible interaction between Gd-chelates and ferritin,
which to our knowledge has not been investigated previously.
For this first study, we examined the interaction of Gd(DTPA)2- and Gd(DTPA-BMA)
with HSAF and three recombinant apoferritins: human H-chain (HuHF), human L-chain
(HuLF), and frog H-chain apoferritin with a proline substitution for conserved leucine
134 (FHF-L134P). Possessing similar 9-coordinate water-bound structures, Gd(DTPA-
BMA) is non-ionic whereas Gd(DTPA)2- has a -2 charge.19,20 HSAF served as the
prototypal apoferritin in this study, due to its commercial availability and natural
combination of H- and L-chain subunits (in a ~1:5 ratio). In humans, both H- and L-chain
ferritin subunits are present in serum, and these interact non-covalently in different
stoichiometries to form intact 24-mer assemblies.21
78
2. Experimental
2.1. Reagents
In our experiments, all starting materials were purchased and used without further
purification, unless otherwise specified. Acros: GdCl3; TbCl3·6H2O. Sigma: HSAF.
Sigma-Aldrich: (NH4)2Fe(SO4)2·6H2O. Fe2+ solutions were freshly prepared from
(NH4)2Fe(SO4)2·6H2O. Tb3+ solutions were prepared by dissolution of TbCl3·6H2O in 1
mM HCl. Unreacted Tb3+ was separated from the apoferritin-Tb3+ samples by filtration
on Centricon® Centrifugal Filter Units. Unless otherwise specified, all experiments were
performed in the same buffer solution: 50 mM Tris, 50 mM NaCl, pH = 7.0.
2.2. Gd-chelates
Gd-chelates were synthesized by the reaction of GdCl3 and corresponding ligands
following literature precedent.22,23 50 mM stock solutions of Gd(DTPA)2- and Gd(DTPA-
BMA) were prepared by dissolving these solid powders in aqueous solution. Gd(DTPA)2-
: ESI-MS m/z: calcd for M + Na+ C14H18N3O10GdNa, 568.6; found, 568.8. Gd(DTPA-
BMA): ESI-MS m/z: calcd for M + Na+ C16H26N5O8GdNa, 596.7; found, 597.0. Stock
solutions of Gd(DTPA)2- and Gd(DTPA-BMA) were sent to Galbraith Laboratories, Inc.
for elemental analysis. The concentrations of elemental Gd, C, and N were analyzed by
ICP-OES (for Gd) and combustion (for C and N). Gd(DTPA)2-: Anal. Calcd: Gd : C : N =
1 : 14 : 3. Found: Gd : C : N = 1.00 : 14.81 : 3.06. Gd(DTPA-BMA): Anal. Calcd: Gd : C
: N = 1 : 16 : 5. Found: Gd : C : N = 1.00 : 16.21 : 4.91.
2.3. Recombinant proteins
79
HuHF and HuLF cDNA was obtained from Paolo Arosio, and these ferritins were
expressed and purified following published protocols.24,25 The FHF-L134P cDNA was
provided by Elizabeth Theil, and the protein was expressed and purified following
published protocols.26,27 BL21(DE3) pLysS cells containing FHF-L134P plasmid were
cultured with betaine and sorbitol to maximize correct folding of the expressed FHF-
L134P subunits. Differences in the protocols for synthesizing HuHF, HuLF, and FHF-
L134P are summarized in Table 3.1. HSAF and the three recombinant proteins were
purified by eluting through a Superdex 200 10/300 GL column (Akta FPLC system, GE
Healthcare) and collecting fractions at the center of the main peak that correspond to the
correctly assembled 24-mer. This procedure was performed 3-4 times for each protein in
order to remove all other proteins and DNA. Protein purity of > 95 % was confirmed by
gel electrophoresis. HuHF: MALDI-MS m/z: calcd, 21,094.45; found, 21,382.10. HuLF:
MALDI-MS m/z: calcd, 19,888.48; found, 19,880.64. FHF-L134P: MALDI-MS m/z:
calcd, 20,460.96; found, 20,537.11.
2.4. Instrumentation
Far-UV circular dichroism measurements were performed on an Aviv Model 410
Circular Dichroism Spectrometer at 25 °C. Experimental conditions: scan rate: 12
nm/min; data interval: 1.0 nm; protein concentration: 0.2 mg/mL. Tryptophan
fluorescence was measured at 25 °C with a Varian Cary Eclipse spectrophotometer
operated with the Cary Eclipse Bio software package. Experimental conditions:
excitation wavelength: 295 nm; scan rate: 30 nm/min; data interval: 0.5 nm; protein
80
concentration: 0.2 mg/mL. T1 relaxation times were determined using a Bruker mq60 MR
relaxometer operating at 1.41 T (60 MHz) and at 40 °C.
MRI studies were performed on a 40 cm bore, 4.7 T horizontal bore magnet (Varian,
Palo Alto, CA) equipped with a 12 cm I.D. gradient insert with a maximum gradient
strength of 25 G/cm (Magnex Scientific, Abingdon, UK). For this experiment, 0.5 mL of
each sample was added into a 3-well plate. Following generation of scout images,
inversion recovery images were generated using the TOMROP (T One by Multiple Read
Out Pulses) pulse sequence.28 The TOMROP protocol is an imaging implementation of
the Look-Locker29 experiment in which the magnetization recovery is sampled many
times (NTI) using low flip angle read pulses at evenly spaced intervals (TI) following
inversion. The following imaging parameters were used: slices: 1; slice thickness: 1.0
mm; FOV: 3×3 cm; averages: 1; flip angle: 10°; TE: 4 ms; TI: n*50 ms; n: 1~120;
recovery delay: 8 s. The TOMROP data were fitted using a non-linear least squares
analysis implemented in the Interactive Data Language (Research Systems Inc., Boulder,
CO) programming environment as described.30 T1 map using the fitted parameters was
then saved.
3. Results and Discussion
Procedures for synthesizing Gd-chelates and isolating recombinant proteins are given
in the experimental section. Gd-chelates were characterized by elemental analysis and
high-resolution mass spectrometry; apoferritins were purified by FPLC and their identity
confirmed by MALDI mass spectrometry. Figure 3.3 shows the FPLC elution traces of
81
HuHF, HuLF, and FHF-L134P monitoring absorbance at 280 nm. HSAF (50 μM, 100
μL) was incubated overnight at 25 °C with varied concentrations of Gd-chelates (1:1 to
20:1 Gd-to-HSAF subunit ratios). Recombinant HuHF, HuLF, or FHF-L134P was
incubated with Gd-chelates in a single Gd-to-apoferritin subunit ratio of 5:1. Samples
were dialyzed to remove unbound Gd-chelates (Figure 3.4). Far-UV circular dichroism
(Figure 3.5) and tryptophan fluorescence (Figure 3.6) measurements confirmed that
apoferritin samples exposed to Gd chelates in this way maintained wild-type α-helical
structure as well as native subunit assembly. To determine the stoichiometry of Gd-
chelate binding to apoferritin, dialyzed samples were digested in a 1:1 volume of
concentrated HNO3 at 80 °C for 1 h, followed by analysis with inductively coupled
plasma optical emission spectroscopy (ICP-OES) to determine Gd concentrations.
Apoferritin concentrations were determined by UV-Vis spectroscopy using the Bradford
assay and bovine serum albumin as a standard.31
Figure 3.7 shows the stoichiometry of binding between Gd-chelates and HSAF after
24-h dialysis to remove any weakly associated Gd. The HSAF-Gd-chelate samples were
stable in solution, with precipitation occurring only when Gd-to-HSAF subunit ratios
exceeded 20 to 1. Binding interactions were identified and quantified in dialyzed HSAF-
Gd(DTPA-BMA) solutions, for all Gd(DTPA-BMA)/HSAF ratios. The Gd-to-HSAF
subunit binding ratio increased in a linear fashion from 0.15 to 1.18 as the Gd-to-HSAF
subunit mixing ratio increased from 1 to 20. Lack of binding saturation over this
concentration range indicated one or more low-affinity sites on HSAF for Gd(DTPA-
BMA). Similar non-saturation behavior was observed previously for Cr(TREN) binding
82
to three frog apoferritins (wt H-chain, wt L-chain, mutant L-chain).4 With addition of 5
equiv Gd(DTPA-BMA), the stoichiometries of Gd binding to HSAF, HuHF, HuLF, and
FHF-L134P were 0.33, 0.37, 0.22, and 0.68 per apoferritin subunit, respectively. The
greater binding capacity of FHF-L134P for Gd(DTPA-BMA) has also been seen
previously for Cr(TREN), and can be attributed to the greater conformational flexibility
and higher accessibility of the FHF mutant.4 The fact that significantly more binding was
observed for FHF-L134P and HuHF suggests that Gd can interact with the di-iron
ferroxidase sites found in ferritin H subunits. These metal coordination sites are lacking
in HuLF, while found partially in HSAF (~4 per 24-mer), and exclusively in HuHF and
FHF-L134P.
In contrast to the apoferritin-Gd(DTPA-BMA) samples, the dialyzed apoferritin-
Gd(DTPA)2- solutions yielded no bound Gd, as analyzed by ICP-OES. This indicated
either a lack of association of Gd(DTPA)2- with apoferritin or that any bound
Gd(DTPA)2- dissociated from apoferritin during the 24-h dialysis procedure. The lack of
binding between Gd(DTPA)2- and apoferritin is likely a result of electrostatic forces: the
Chasteen lab showed that positively charged organic radicals penetrate into apoferritin
much more efficiently than do negatively charged radicals.32 It is established that cations
such as hydrated Fe2+ gain access to the ferritin interior via threefold pores lined with
negatively charged residues.
To investigate whether more transient interactions occur between Gd(DTPA)2- and
HSAF, we performed relaxivity-based binding measurements. Gd-chelates were
characterized by their longitudinal relaxivity, r1 = (Δ1/T1)/[CA], where Δ1/T1 is the
83
change in relaxation rate (s-1) of solvent water protons upon addition of contrast agent,
[CA] at mM concentration.11 Figure 3.8 shows T1 relaxation times for Gd(DTPA)2- and
Gd(DTPA-BMA) in the absence and presence of HSAF (Gd-to-HSAF subunit ratio = 1),
which were determined using a Bruker mq60 MR relaxometer operating at 1.41 T (60
MHz). As a useful comparison, T1 values were measured in the presence of the same
concentration of HSA (Figure 3.9), which is known to interact only very weakly with
Gd(DTPA)2- and Gd(DTPA-BMA).11,33 Linear fitting of the data gave the r1 values
shown in Table 3.2. The results with these undialyzed samples indicated slightly more
association of Gd(DTPA)2- with HSAF than with HSA, but much less binding than that
found for Gd(DTPA-BMA) and HSAF.
Relaxivity measurements were also used to probe dialyzed samples of apoferritin
bound to Gd(DTPA-BMA). Linear fitting of T1 versus Gd3+ concentration (Figure 3.10)
gave the “Gd-bound” r1 values shown in Table 3.2. The r1 values of Gd(DTPA-BMA)
increased more than 20-fold when bound to either horse spleen or human H apoferritin
(77.8 and 73.4 mM-1 s-1), as compared to its unbound form (3.4 mM-1 s-1). The Gd-chelate
MS-325 was shown previously to increase proton relaxivity ninefold upon binding HSA
(r1 = 51 mM-1 s-1), due to the increase in the rotational correlation time (τR) of the Gd-
chelate upon binding to protein (10.1 ns bound vs. 115 ps free).34 Here, the even slower
rotation of 12-nm diameter apoferritin (τR ≈ 200 ns)35 increased the relaxivity of protein-
bound Gd(DTPA-BMA).
The design of high-relaxivity gadolinium agents has been the focus of considerable
research,11,34 based on the potential for improving contrast in MRI. It is noteworthy that
84
the large relaxivities measured for apoferritin-Gd(DTPA-BMA) samples have been rarely
seen in other synthetic or biological Gd3+ conjugates. Tsourkas and co-workers increased
Gd payloads by first conjugating Gd-chelates to dendrimers and then either encapsulating
dendrimers within polymersomes36,37 or cross-linking dendrimers to form nanoclusters,38
and achieved relaxivities as high as 12.3 mM-1 s-1. In a few cases, relaxivities greater than
100 mM-1 s-1 per Gd have been posited in the literature.39,40 A relevant biological
example comes from Douglas et al. who covalently bound Gd3+ to cowpea chlorotic
mottle virus and recorded the highest relaxivity values (~ 200 mM-1 s-1) for paramagnetic
materials to date.39 In the present study, MRI data were collected for pure HSAF (5.3
μM) as a control, for pure Gd(DTPA-BMA), and for HSAF-1.18 equiv Gd(DTPA-BMA),
both containing 150 µM Gd3+. Figure 3.11 shows the T1 map. The binding of Gd(DTPA-
BMA) to HSAF caused a decrease in T1 values and produced a higher-signal image. This
MRI experiment confirms a method for increasing contrast by complexing Gd-chelates to
apoferritin.
The effect of Gd-chelate binding on iron oxidation and mineralization was studied with
all four apoferritins. UV-Vis spectroscopy measured the increase in absorbance at 350
nm immediately following the addition of 500 equiv Fe2+ to each apoferritin solution
(0.33 μM) (Figure 3.12). Kinetics traces of the dialyzed apoferritin-Gd(DTPA)2- samples
generally overlapped with those of pure apoferritin. One exception was the observation of
enhanced rates of iron oxidation and mineralization with dialyzed HuHF-Gd(DTPA)2-
samples. ICP-OES indicated less than 0.1 Gd per HuHF 24-mer, which was considered to
be no Gd in this work. In contrast, the presence of Gd(DTPA-BMA) caused a slower rate
85
of iron oxidation in all apoferritins studied. One possible explanation is that Gd(DTPA-
BMA) obstructs the routes of Fe2+ uptake, for example by binding in the threefold
channels of apoferritin. A second and concurrent possibility is that Gd(DTPA-BMA)
inhibits iron oxidation and mineralization by disrupting the early recognition of Fe2+
inside apoferritin, e.g., at the di-iron ferroxidase sites that are contained in H-ferritin
subunits. Support for the second explanation comes from the observation of least
inhibition in HuLF (Figure 3.12C), and is consistent with less Gd binding to HuLF, as
shown by ICP-OES.
To investigate the possibility that Gd(DTPA-BMA) blocks Fe2+ uptake in apoferritin,
we set out to determine whether the Gd-chelate binding to HSAF inhibits the subsequent
binding of Tb3+. Tb3+ has been shown previously to provide a luminescent probe for
measuring metal ion uptake by HSAF.41,42 In a different study, competition for Tb3+
binding sites in HSAF provided evidence of Cr(TREN) occupancy at the threefold
pores.43
Here, solutions of Tb3+ and HSAF, dialyzed HSAF-Gd(DTPA)2-, or HSAF-1.18 equiv
Gd(DTPA-BMA) were stirred at 25 °C overnight; the amount of Tb3+ added ranged from
1 to 8 equiv per HSAF. Unreacted Tb3+ was separated from the apoferritin-Tb3+ samples
by centrifugal filtration (100 kDa exclusion size). A control experiment was made by
mixing Tb3+ in protein-free solution. Filtrate (2.6 mL) that contained only reacted Tb3+
was mixed with 1,10-phenanthroline (phen, 0.4 mL) in a 4 : 1 phen-to-Tb3+ ratio,
producing quantifiable luminescence from the Tb3+-phen complex. Figure 3.13 presents
Tb3+-phen luminescence as a function of the total Tb3+ added per HSAF subunit. The
86
slope of luminescence intensity for the different filtrate solutions is equal to that of the
control solution, beginning at approximately 3 equiv of Tb3+ per HSAF subunit, which
indicated that saturation of all Tb3+ binding sites was achieved. No difference was
observed between the filtrate solutions originating from the different HSAF samples,
confirming that sites occupied by Gd(DTPA-BMA) did not inhibit the subsequent
binding of Tb3+. We conclude that Gd(DTPA-BMA) must bind sites outside the Tb3+-
delineated route of metal uptake, which includes the threefold pores.
To investigate more directly whether Gd(DTPA-BMA) competes with iron for the
apoferritin ferroxidase sites, a ferric intermediate test was performed. Previous
investigation of FHF revealed a millisecond-timescale diferric peroxo intermediate
formed in the iron oxidation process. This diferric intermediate is much longer-lived in
FHF-L134P,44-46 which allows convenient spectroscopic monitoring. In our study, just 48
equiv Fe2+ (two per protein subunit) were added in order to determine whether the H-
chain-specific diferric intermediate was formed during iron oxidation. An absorption
band at ca. 525 nm was immediately recorded upon mixing FHF-L134P (1.0 μM) and
Fe2+ (Figure 3.14A). This absorbance feature was not observed after mixing FHF-L134P-
0.68 equiv Gd(DTPA-BMA) and Fe2+ (Figure 3.14B). Furthermore, the presence of
Gd(DTPA-BMA) delayed the rise in UV absorption. This was in agreement with the
prior 500 equiv Fe2+ per apoferritin ferroxidase experiments (Figure 3.12), further
evidencing that Gd(DTPA-BMA) inhibits iron mineralization. Previously, loading
Cr(TREN) into FHF-L134P slowed formation of the ferric intermediate, even with less
than one Cr(TREN) bound per subunit.4 We conclude, as Theil et al. had in the study, that
87
the binding site(s) of Gd(DTPA-BMA) should be at or near the ferroxidase centers of the
H-chain subunit. Efforts are underway to characterize crystallographically the apoferritin-
bound Gd species.
4. Conclusions
In summary, we studied the interaction of four apoferritins with two widely used
gadolinium MRI contrast agents. The non-ionic Gd(DTPA-BMA) was found to bind
apoferritins in varying stoichiometries, in some cases above one Gd3+ complex per
subunit, while the anionic Gd(DTPA)2- was not detectable in apoferritin solutions after
dialysis. Remarkable relaxivity enhancement and iron mineralization inhibition were
observed as a result of Gd(DTPA-BMA) binding to apoferritins. This study suggests
possible ferritin-Gd(DTPA-BMA) interactions and dysregulation of iron storage in vivo,
which will require further investigation, and may be particularly relevant in patients with
impaired kidney function who clear the Gd agent slowly from the body.
5. Future Directions
This chapter reports the binding interactions between four different apoferritins and
two commonly employed gadolinium chelates. The use of apoferritin is particularly
helpful for determining the binding site(s) of gadolinium chelates. Results show that the
binding site(s) of Gd(DTPA-BMA) should be at or near the ferroxidase centers of the H-
chain subunit. However, in the human body, holoferritin (loaded with iron) is the
predominant form of the protein. Therefore, the study of the binding interactions between
88
holoferritins and gadolinium chelates will be a new research direction with biomedical
implications.
As a preliminary investigation, I chose horse spleen ferritin (HSF) as the holoferritin
and Gd(DTPA-BMA) as the gadolinium chelate. HSF-Gd(DTPA-BMA) complex was
prepared following essentially the same procedure as described for the apoferritin
experiments. After proper dialysis, the Gd and protein concentrations of this sample were
determined. The data showed that 63.8 equiv of Gd(DTPA-BMA) were associated with 1
equiv HSF, which is a much higher ratio than the 28.3 equiv Gd(DTPA-BMA) that were
determined to bind HSAF. The reason for this higher ratio remains to be determined, but
suggests that the Gd chelate may be interacting with ligands on the surface of the iron
mineral. The effect of Gd(DTPA-BMA) binding on iron oxidation and mineralization
was also studied. UV-Vis spectroscopy measured the increase in absorbance at 470 nm
immediately following the addition of 500 equiv Fe2+ to HSF solution (1.0 μM) (Figure
3.15). The absorption signal at 350 nm, which was used to monitor the kinetics in
apoferritin experiments, was too noisy in HSF experiments. Therefore, 470 nm was used
as the monitoring wavelength.
The kinetics trace of HSF-Gd(DTPA-BMA) reacting with 500 equiv Fe2+ was nearly
identical to that of pure HSF. It is noteworthy that sometimes the absorbance change
measured by UV-Vis spectroscopy with HSF-Gd(DTPA-BMA) was slightly lower than
pure HSF. Careful analysis revealed that this intensity difference started at t = 0. To
calibrate this difference, the absorption intensities at t = 0 were set as zero as shown in
Figure 3.15. Such a calibration is necessary in HSF experiments due to the strong
89
background absorption of the iron core. Future work will include human holoferritin and
experiment on both Gd(DTPA)2- and Gd(DTPA-BMA).
6. References
1. Gray, H. B. Biological inorganic chemistry at the beginning of the 21st century.
Proc. Natl. Acad. Sci. U. S. A. 100, 3563-3568 (2003).
2. Louie, A. Y. & Meade, T. J. Metal complexes as enzyme inhibitors. Chem. Rev.
99, 2711-2734 (1999).
3. Holm, R. H., Kennepohl, P. & Solomon, E. I. Structural and functional aspects of
metal sites in biology. Chem. Rev. 96, 2239-2314 (1996).
4. Barnes, C. M., Theil, E. C. & Raymond, K. N. Iron uptake in ferritin is blocked
by binding of a slow dissociating. Proc. Natl. Acad. Sci. U. S. A. 99, 5195-5200
(2002).
5. Liepold, L. O., Abedin, M. J., Buckhouse, E. D., Frank, J. A., Young, M. J. &
Douglas, T. Supramolecular protein cage composite MR contrast agents with
extremely efficient relaxivity properties. Nano Lett. 9, 4520-4526 (2009).
6. Neault, J. F. & Tajmir-Riahi, H. A. Interaction of cisplatin with human serum
albumin. Drug binding mode and protein secondary structure. Biochim. Biophys.
Acta-Protein Struct. Molec. Enzym. 1384, 153-159 (1998).
7. Bernersprice, S. J., Johnson, R. K., Giovenella, A. J., Faucette, L. F., Mirabelli, C.
K. & Sadler, P. J. Antimicrobial and anticancer activity of tetrahedral, chelated,
90
diphosphine silver(I) complexes: comparison with copper and gold. J. Inorg.
Biochem. 33, 285-295 (1988).
8. Will, J., Wolters, D. A. & Sheldrick, W. S. Characterisation of cisplatin binding
sites in human serum proteins using hyphenated multidimensional liquid
chromatography and ESI tandem mass spectrometry. ChemMedChem 3, 1696-
1707 (2008).
9. Dhubhghaill, O. M. N., Sadler, P. J. & Tucker, A. Drug-induced reactions of
bovine serum albumin: proton NMR studies of gold binding and cysteine release.
J. Am. Chem. Soc. 114, 1118-1120 (1992).
10. Zemlickis, D., Klein, J., Moselhy, G. & Koren, G. Cisplatin protein binding in
pregnancy and the neonatal period. Med. Pediatr. Oncol. 23, 476-479 (1994).
11. Caravan, P., Ellison, J. J., McMurry, T. J. & Lauffer, R. B. Gadolinium(III)
chelates as MRI contrast agents: structure, dynamics, and applications. Chem.
Rev. 99, 2293-2352 (1999).
12. Wang, Y., Spiller, M. & Caravan, P. Evidence for weak protein binding of
commercial extracellular gadolinium contrast agents. Magn. Reson. Med. 63, 609-
616 (2010).
13. Cowper, S. E., Robin, H. S., Steinberg, S. M., Su, L. D., Gupta, S. & LeBoit, P. E.
Scleromyxoedema-like cutaneous diseases in renal-dialysis patients. Lancet 356,
1000-1001 (2000).
14. Ting, W. W., Stone, M. S., Madison, K. C. & Kurtz, K. Nephrogenic fibrosing
dermopathy with systemic involvement. Arch. Dermatol. 139, 903-906 (2003).
91
15. Thomsen, H. S., Marckmann, P. & Logager, V. B. Update on nephrogenic
systemic fibrosis. Magn. Reson. Imaging Clin. N. Am. 16, 551-560 (2008).
16. Boyd, A. S., Zic, J. A. & Abraham, J. L. Gadolinium deposition in nephrogenic
fibrosing dermopathy. J. Am. Acad. Dermatol. 56, 27-30 (2007).
17. Joffe, P., Thomsen, H. S. & Meusel, M. Pharmacokinetics of gadodiamide
injection in patients with severe renal insufficiency and patients undergoing
hemodialysis or continuous ambulatory peritoneal dialysis. Acad. Radiol. 5, 491-
502 (1998).
18. Swaminathan, S., Horn, T. D., Pellowski, D., Abul-Ezz, S., Bornhorst, J. A.,
Viswamitra, S. & Shah, S. V. Nephrogenic systemic fibrosis, gadolinium, and
iron mobilization. N. Engl. J. Med. 357, 720-722 (2007).
19. Weinmann, H. J., Brasch, R. C., Press, W. R. & Wesbey, G. E. Characteristics of
gadolinium-DTPA complex: a potential NMR contrast agent. Am. J. Roentgenol.
142, 619-624 (1984).
20. Cacheris, W. P., Quay, S. C. & Rocklage, S. M. The relationship between
thermodynamics and the toxicity of gadolinium complexes. Magn. Reson.
Imaging 8, 467-481 (1990).
21. Harrison, P. M. & Arosio, P. Ferritins: molecular properties, iron storage function
and cellular regulation. Biochim. Biophys. Acta-Bioenerg. 1275, 161-203 (1996).
22. Hernandez, G., Tweedle, M. F. & Bryant, R. G. Proton magnetic relaxation
dispersion in aqueous glycerol solutions of Gd(DTPA)2- and Gd(DOTA)-. Inorg.
Chem. 29, 5109-5113 (1990).
92
23. Masui, T., Saeed, M., Wendland, M. F. & Higgins, C. B. Occlusive and
reperfused myocardial infarcts: MR imaging differentiation with nonionic Gd-
DTPA-BMA. Radiology 181, 77-83 (1991).
24. Butts, C. A., Swift, J., Kang, S. G., Di Costanzo, L., Christianson, D. W., Saven,
J. G. & Dmochowski, I. J. Directing noble metal ion chemistry within a designed
ferritin protein. Biochemistry 47, 12729-12739 (2008).
25. Levi, S., Salfeld, J., Franceschinelli, F., Cozzi, A., Dorner, M. H. & Arosio, P.
Expression and structural and functional properties of human ferritin L-chain
from Escherichia coli. Biochemistry 28, 5179-5184 (1989).
26. Fetter, J., Cohen, J., Danger, D., Sanders-Loehr, J. & Theil, E. C. The influence of
conserved tyrosine 30 and tissue-dependent differences in sequence on ferritin
function: use of blue and purple Fe(III) species as reporters of ferroxidation. J.
Biol. Inorg. Chem. 2, 652-661 (1997).
27. Takagi, H., Shi, D. S., Ha, Y., Allewell, N. M. & Theil, E. C. Localized unfolding
at the junction of three ferritin subunits. J. Biol. Chem. 273, 18685-18688 (1998).
28. Brix, G., Schad, L. R., Deimling, M. & Lorenz, W. J. Fast and precise T1 imaging
using a TOMROP sequence. Magn. Reson. Imaging 8, 351-356 (1990).
29. Look, D. C. & Locker, D. R. Time saving in measurement of NMR and EPR
relaxation times. Rev. Sci. Instrum. 41, 250-251 (1970).
30. Pickup, S., Wood, A. K. W. & Kundel, H. L. A novel method for analysis of
TOMROP data. J. Magn. Reson. Imaging 19, 508-512 (2004).
93
31. Bradford, M. M. Rapid and sensitive method for quantitation of microgram
quantities of protein utilizing principle of protein-dye binding. Anal. Biochem. 72,
248-254 (1976).
32. Yang, X. K. & Chasteen, N. D. Molecular diffusion into horse spleen ferritin: a
nitroxide radical spin probe study. Biophys. J. 71, 1587-1595 (1996).
33. Henrotte, V., Laurent, S., Gabelica, V., Elst, L. V., Depauw, E. & Muller, R. N.
Investigation of non-covalent interactions between paramagnetic complexes and
human serum albumin by electrospray mass spectrometry. Rapid Commun. Mass
Spectrom. 18, 1919-1924 (2004).
34. Caravan, P., Cloutier, N. J., Greenfield, M. T., McDermid, S. A., Dunham, S. U.,
Bulte, J. W. M., Amedio, J. C., Looby, R. J., Supkowski, R. M., Horrocks, W. D.,
McMurry, T. J. & Lauffer, R. B. The interaction of MS-325 with human serum
albumin and its effect on proton relaxation rates. J. Am. Chem. Soc. 124, 3152-
3162 (2002).
35. Gossuin, Y., Roch, A., Lo Bue, F., Muller, R. N. & Gillis, P. Nuclear magnetic
relaxation dispersion of ferritin and ferritin-like magnetic particle solutions: a pH-
effect study. Magn. Reson. Med. 46, 476-481 (2001).
36. Cheng, Z. L. & Tsourkas, A. Paramagnetic porous polymersomes. Langmuir 24,
8169-8173 (2008).
37. Cheng, Z. L., Thorek, D. L. J. & Tsourkas, A. Porous polymersomes with
encapsulated Gd-labeled dendrimers as highly efficient MRI contrast agents. Adv.
Funct. Mater. 19, 3753-3759 (2009).
94
38. Cheng, Z. L., Thorek, D. L. J. & Tsourkas, A. Gadolinium-conjugated dendrimer
nanoclusters as a tumor-targeted T1 magnetic resonance imaging contrast agent.
Angew. Chem. Int. Ed. 49, 346-350 (2010).
39. Allen, M., Bulte, J. W. M., Liepold, L., Basu, G., Zywicke, H. A., Frank, J. A.,
Young, M. & Douglas, T. Paramagnetic viral nanoparticles as potential high-
relaxivity magnetic resonance contrast agents. Magn. Reson. Med. 54, 807-812
(2005).
40. Hanaoka, K., Lubag, A. J. M., Castillo-Muzquiz, A., Kodadek, T. & Sherry, A. D.
The detection limit of a Gd3+-based T1 agent is substantially reduced when
targeted to a protein microdomain. Magn. Reson. Imaging 26, 608-617 (2008).
41. Lawson, D. M., Artymiuk, P. J., Yewdall, S. J., Smith, J. M. A., Livingstone, J.
C., Treffry, A., Luzzago, A., Levi, S., Arosio, P., Cesareni, G., Thomas, C. D.,
Shaw, W. V. & Harrison, P. M. Solving the structure of human H-ferritin by
genetically engineering intermolecular crystal contacts. Nature 349, 541-544
(1991).
42. Treffry, A., Bauminger, E. R., Hechel, D., Hodson, N. W., Nowik, I., Yewdall, S.
J. & Harrison, P. M. Defining the roles of the threefold channels in iron uptake,
iron oxidation and iron-core formation in ferritin: a study aided by site-directed
mutagenesis. Biochem. J. 296, 721-728 (1993).
43. Barnes, C. M., Petoud, S., Cohen, S. M. & Raymond, K. N. Competition studies
in horse spleen ferritin probed by a kinetically inert inhibitor,
95
[Cr(TREN)(H2O)(OH)]2+, and a highly luminescent Tb(III) reagent. J. Biol.
Inorg. Chem. 8, 195-205 (2003).
44. Waldo, G. S., Ling, J. S., Sandersloehr, J. & Theil, E. C. Formation of an Fe(III)-
tyrosinate complex during biomineralization of H-subunit ferritin. Science 259,
796-798 (1993).
45. Treffry, A., Zhao, Z., Quail, M. A., Guest, J. R. & Harrison, P. M. Iron(II)
oxidation by H-chain ferritin: evidence from site-directed mutagenesis that a
transient blue species is formed at the dinuclear iron center. Biochemistry 34,
15204-15213 (1995).
46. Treffry, A., Zhao, Z. W., Quail, M. A., Guest, J. R. & Harrison, P. M. How the
presence of three iron binding sites affects the iron storage function of the ferritin
(EcFtnA) of Escherichia coli. FEBS Lett. 432, 213-218 (1998).
96
Table 3.1. Differences in the protocols for synthesizing HuHF, HuLF, and FHF-L134P. ________________________________________________________________________
HuHF HuLF FHF-L134P ________________________________________________________________________
broth (1 L) tryptone: 10.0 g tryptone: 10.0 g tryptone: 10.0 g yeast extract: 5.0 g yeast extract: 5.0 g yeast extract: 5.0 g
NaCl: 10.0 g NaCl: 10.0 g NaCl: 10.0 g 1 M NaOH: 1.0 mL 1 M NaOH: 1.0 mL 1 M NaOH: 5.0 mL
sorbitol: 182.2 g betaine: 0.293 g
incubation time ~ 3 h ~ 4 h ~ 8 h needed to reach O.D.600 nm = 0.9~1.0
inducing agent IPTGa 3-IAAb IPTGa and amount 240 mg/L 50 mg/L 240 mg/L
incubation time 4 h overnight overnight and temperature 37 °C 20 °C 20 °C
________________________________________________________________________ a: IPTG = isopropyl β-D-1-thiogalactopyranoside b: 3-IAA = 3β-indoleacrylic acid
97
Table 3.2. Summary of r1 values. ________________________________________________________________________
Gd(DTPA)2- Gd(DTPA-BMA) ________________________________________________________________________
undialyzed samples pure Gd chelates 2.9 mM-1 s-1 3.4 mM-1 s-1 HSA-Gd chelates 2.9 mM-1 s-1 3.7 mM-1 s-1 HSAF-Gd chelates 3.2 mM-1 s-1 12.1 mM-1 s-1
dialyzed samples HSAF-1.18 equiv Gd(DTPA-BMA) 77.8 mM-1 s-1 HuHF-0.37 equiv Gd(DTPA-BMA) 73.4 mM-1 s-1
________________________________________________________________________
98
Figure 3.1. Molecular structures of Gd(DTPA)2- and Gd(DTPA-BMA).
Gd
NN N
O OOO
H2OO O
OOO
O3+Gd
NN N
O OOH2O
O OO
OO
3+HN NH
Gd(DTPA)2- Gd(DTPA-BMA)
99
Figure 3.2. Sclerotic plaque with calcinosis on the back arm of the patient.
100
Figure 3.3. FPLC elution traces of HuHF (red), HuLF (black), and FHF-L134P (blue) monitoring absorbance at 280 nm.
-5 0 5 10 15 20 25 30 35 400.0
0.2
0.4
0.6
0.8
1.0
1.2-5 0 5 10 15 20 25 30 35 40
0.0
0.5
1.0
-5 0 5 10 15 20 25 30 35 40
0.0
0.5
1.0
0 10 20 30 400.0
0.2
0.4
0.6
0.8
1.0
1.2
Abso
rptio
n / a
.u.
Time / min
101
Figure 3.4. Preparation of apoferritin-Gd-chelate complexes.
102
Figure 3.5. Far-UV circular dichroism measurements for samples (A) HSAF and (B) HuHF.
190 200 210 220 230 240 250 260-40
-20
0
20
40
60
∆ε /
M-1 c
m-1 re
sidu
e-1
Wavelength / nm
HSAF HSAF-Gd(DTPA) 2- dialyzed HSAF-1.18 equiv Gd(DTPA-BMA)
(A)
190 200 210 220 230 240 250 260-40
-20
0
20
40
60
∆ε /
M-1 c
m-1 re
sidu
e-1
Wavelength / nm
HuHF HuHF-Gd(DTPA) 2- dialyzed HuHF-0.37 equiv Gd(DTPA-BMA)
(B)
103
Figure 3.6. Tryptophan fluorescence measurements for samples (A) HSAF and (B) HuHF.
300 350 400 450 500 55050
100
150
200
250
300
350
400
Fluo
resc
ence
Inte
nsity
Wavelength / nm
HSAF HSAF-Gd(DTPA) 2- dialyzed HSAF-1.18 equiv Gd(DTPA-BMA)
(A)
300 350 400 450 500 5500
100
200
300
400
500
600
700
Fluo
resc
ence
Inte
nsity
Wavelength / nm
HuHF HuHF-Gd(DTPA) 2- dialyzed HuHF-0.37 equiv Gd(DTPA-BMA)
(B)
104
Figure 3.7. Binding stoichiometry of Gd-chelates to HSAF vs. number equiv Gd added, after 24-h dialysis.
0 5 10 15 200.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Gd-
chel
ates
boun
d / H
SAF
subu
nit
Gd-chelates added / HSAF subunit
Gd(DTPA) 2-
Gd(DTPA-BMA)
105
Figure 3.8. Relaxivity measurements for (A) pure Gd(DTPA)2-, (B) pure Gd(DTPA-BMA), (C) undialyzed HSAF-Gd(DTPA)2-, and (D) undialyzed HSAF-Gd(DTPA-BMA).
0.6 0.8 1.0 1.22.0
2.5
3.0
3.5
4.0 (A) Pure Gd(DTPA) 2-
1/T 1 /
s-1
[Gd 3+] / mM0.6 0.8 1.0 1.2
2.5
3.0
3.5
4.0
4.5(B) Pure Gd(DTPA-BMA)
1/T 1 /
s-1
[Gd 3+] / mM
0.6 0.8 1.0 1.2
1.5
2.0
2.5
3.0
3.5(C) HSAF-Gd(DTPA) 2- undialyzed
1/T 1 /
s-1
[Gd 3+] / mM0.6 0.8 1.0 1.2
6
8
10
12
14(D) HSAF-Gd(DTPA-BMA) undialyzed
1/T 1 /
s-1
[Gd 3+] / mM
106
Figure 3.9. Relaxivity measurements for (A) undialyzed HSA-Gd(DTPA)2- and (B) undialyzed HSA-Gd(DTPA-BMA).
0.6 0.8 1.0 1.2
1.0
1.5
2.0
2.5
3.0 (A) HSA-Gd(DTPA) 2- undialyzed
1/T 1 /
s-1
[Gd 3+] / mM0.6 0.8 1.0 1.2
1.5
2.0
2.5
3.0
3.5
4.0 (B) HSA-Gd(DTPA-BMA) undialyzed
1/T 1 /
s-1
[Gd 3+] / mM
107
Figure 3.10. Relaxivity measurements for (A) HSAF-1.18 equiv Gd(DTPA-BMA) and (B) HuHF-0.37 equiv Gd(DTPA-BMA).
0.4 0.5 0.6 0.7 0.8
30
40
50
60 (A) HSAF-1.18 equiv Gd(DTPA-BMA)
1/T 1 /
s-1
[Gd 3+] / mM0.08 0.10 0.12 0.14
4
6
8
10(B) HuHF-0.37 equiv Gd(DTPA-BMA)
1/T 1 /
s-1
[Gd 3+] / mM
108
sample: HSAF Gd(DTPA-BMA) HSAF-1.18 equiv Gd(DTPA-BMA) T1 values: 1424 ms 720 ms 307 ms Figure 3.11. T1 map of pure HSAF (5.3 μM), pure Gd(DTPA-BMA) (150 μM), and HSAF-1.18 equiv Gd(DTPA-BMA) ([Gd3+] = 150 μM).
109
Figure 3.12. Kinetics of iron oxidation and mineralization upon addition of 500 equiv Fe2+ to four apoferritins (A-D). Dotted line (black), Fe2+; dashed line (red), apoferritin + Fe2+; solid line (green), dialyzed apoferritin-Gd(DTPA)2- + Fe2+; dashed dotted line (blue), apoferritin-x equiv Gd(DTPA-BMA) + Fe2+, where x = 1.18 in (A), 0.37 in (B), 0.22 in (C), and 0.68 in (D).
0 5 10 15 20 25 300.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 300.0
0.1
0.2
0.3
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 300.0
0.1
0.2
0.3
0 10 20 300.0
0.1
0.2
0.3
O.D
.(350
nm)
Time / min
(B) HuHF
0 5 10 15 20 25 300.0000.0050.0100.0150.0200.0250.0300.0350.0400.0450.0500.0550.0600.0650.070
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 300.00
0.02
0.04
0.06
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 300.00
0.02
0.04
0.06
0 10 20 300.00
0.02
0.04
0.06
O.D
.(350
nm)
Time / min
(C) HuLF
0 5 10 15 20 25 300.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 300.0
0.1
0.2
0.3
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 300.0
0.1
0.2
0.3
0 10 20 300.0
0.1
0.2
0.3
O.D
.(350
nm)
Time / min
(D) FHF-L134P
0 5 10 15 20 25 300.0000.0050.0100.0150.0200.0250.0300.0350.0400.0450.0500.0550.0600.0650.0700.0750.080
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 300.00
0.02
0.04
0.06
0.08
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 300.00
0.02
0.04
0.06
0.08
0 10 20 300.00
0.02
0.04
0.06
0.08
O.D
.(350
nm)
Time / min
(A) HSAF
110
Figure 3.13. Luminescence measured upon addition of 1,10-phenanthroline to samples containing varying amounts of Tb3+ per HSAF subunit. Excitation wavelength: 290 nm. (green), Tb3+; (cyan), HSAF + Tb3+; | (red), dialyzed HSAF-Gd(DTPA)2- + Tb3+; — (black), HSAF-1.18 equiv Gd(DTPA-BMA) + Tb3+.
0 2 4 6 8
0
5
10
15
20
25
30
35
Em(5
44nm
)
Tb3+(total) / HSAF subunit
111
Figure 3.14. Absorption spectra on adding Fe2+ to (A) FHF-L134P or (B) FHF-L134P-0.68 equiv Gd(DTPA-BMA).
400 600 8000.00
0.02
0.04
0.06
0.08
0.10 (A)
Abso
rptio
n / a
.u.
Wavelength / nm
Time after adding Fe(II) 0 s 30 s 60 s 90 s 120 s
400 600 8000.00
0.02
0.04
0.06
0.08
0.10 (B)
Abso
rptio
n / a
.u.
Wavelength / nm
Time after adding Fe(II) 0 s 30 s 60 s 90 s 120 s
112
Figure 3.15. Kinetics of iron oxidation and mineralization upon addition of 500 equiv Fe2+ to HSF.
0 10 20 30 40 50 600.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07 HSF + Fe(II) HSF-Gd(DTPA-BMA) + Fe(II)
O.D
. 470n
m
Time / min
113
Appendix 1: Developing Strategies for Creating Ferritin-Gold
Nanoparticle Dimers Using Streptavidin-Biotin Junctions
Gold nanoparticle dimers have shown important applications in both physical and
biological sciences.1,2 In physical sciences, gold nanoparticle dimers are of special
interest because of their application as substrates in surface-enhanced Raman
spectroscopy (SERS).1 This enhancement effect is attributed to the random formation of
localized plasmons at the junctions between nanoparticles, giving rise to large
enhancements that enable SERS detection at or near single-molecule sensitivity.3
Theoretical investigation by Nordlander et al. showed that a pair of closely spaced
nanoparticles supports dimer plasmons that can be viewed as the bonding and
antibonding combinations, i.e., hybridization of the individual nanoparticle plasmons, in
analogy with eigenstates of homonuclear diatomic molecules.4 In biological sciences, the
distance change between two gold nanoparticles can be used to monitor biological
processes in real time.2 Alivisatos and co-workers proposed pairs of gold nanoparticles as
“plasmon rulers” based on the dependence of their light scattering on the interparticle
distance.5 In their later work, such a plasmon ruler has been used to study the cleavage of
DNA by the restriction enzyme EcoRV at the single-molecule level.2 Ringler et al.
synthesized nanoparticle dimers by linking biotin-functionalized gold nanoparticles with
streptavidin, and consequently changed optically the distance between two gold
nanoparticles by Raman-induced motion of the linker protein.6
114
During the past several years, preparing nanoparticle dimers using DNA as the linker
has witnessed substantial progress.7-9 Alivisatos and co-workers have succeeded in
putting a fixed number of thiolated ss-DNA molecules onto the surface of a gold
nanoparticle. Various structures, including DNA-linked nanoparticle dimers and pyrimids,
have been constructed following established techniques.7-9 However, when nanoparticle
dimers are prepared using methods other than the above DNA conjugation method,
several issues are revealed, and need to be summarized to benefit future investigation.
Issue 1: there are no specific binding sites on the nanoparticle monomers. In the paper
published by Feldheim and co-workers, nanoparticle dimers were formed by using
citrate-capped gold nanoparticles and sulfur-containing phenylacetylene oligomers as the
linkers.10 Because there are no specific binding sites on the gold nanoparticles, the
mixing of nanoparticles with linker molecules could easily lead to the formation of
aggregates. To avoid forming aggregates, the authors accomplished the dimer preparation
by producing a locally high concentration of gold nanoparticles with respect to the linkers.
As a result, a large amount of monomers were mixed with dimers (monomer-to-dimer
ratio = 2.3 : 1).
Issue 2: the preparation of nanoparticle dimers was not accomplished in a single phase.
To incorporate specific binding sites in the nanoparticles for dimer preparation, various
groups have introduced the idea of using the combination of two phases to mediate the
nanoparticle functionalization. Zhang and co-workers took advantage of a toluene-water
interface to attach tiopronin to one side of gold nanoparticles, and then prepared dimers
using ethylenediamine as the linker.11 Sardar et al. asymmetrically functionalized gold
115
nanoparticle via a solid-phase approach and then synthesized dimers in a solution
phase.12 The involvement of organic phase or solid phase not only makes the reaction
system complicated, but also limits the use of biomaterials in these samples.
DNA itself lacks various functional groups. To explore the diversity of nanoparticle
synthesis, we are interested in preparing nanoparticle dimers using a non-DNA-related
method. Systems that can solve the above two problems will (1) have specific binding
sites on the nanoparticle monomers, and (2) accomplish the nanoparticle dimer
preparation in a single phase. Here, we choose a particular thermophilic ferritin that has
the capability of disassembling at low salt concentrations and reassembling in the
presence of gold nanoparticles (Figure A1.1).13 We plan to label each ferritin with two
biotins and use streptavidin to link two ferritins together.
The wt thermophilic ferritin (hereafter abbreviated as T0) used in this work is from the
hyperthermophilic Archaeon Archaeoglobus fulgidus.14 To attach biotins to T0, the
commonly used method is to perform a biotinylation reaction with the cysteines of the
protein. However, there are no native cysteines anywhere (surface, interior, etc.) in T0.
Therefore, the cysteine should be introduced to T0 via site-directed mutagenesis. In fact,
having no native cysteines in T0 serves as an advantage, because in this case Cys labeling
can only occur at the engineered sites. After careful examination of the crystal structure
of T0, we decided to engineer mutated ferritin containing an S80C mutation (hereafter
abbreviated as T1eC) due to two reasons: (1) the S80C mutation site, as shown in Figure
A1.2A, is surface exposed providing convenience in Cys labeling; (2) the S80C mutation
site is in a loop of the 4-helix bundle, which can largely minimize the effect of mutation
116
on the subsequent subunit assembly. Figures A1.2A, A1.2B, and A1.2C present the
structures of T1eC dimer, the biotinylation reagent, and streptavidin complexed with four
biotins, respectively, together with relevant distance information. Figure A1.2D reveals
that our design is geometrically feasible.
In our design as illustrated in Figure A1.3, T1eC subunits are first labeled with biotins
and are then mixed with T0 subunits to encapsulate gold nanoparticles. Streptavidin is a
tetrameric protein that is remarkable for its ability to bind up four biotins with unusually
high affinity (dissociation constant KD = 10-15 M).15 In this work, streptavidin is used to
link two ferritins together.
T0 can be isolated from E. coli in high yield, whereas the production efficiency of
T1eC was much lower. Therefore in this work, I first tried to optimize the conditions for
growing T1eC. Two main factors that may affect the protein growth were examined: (i)
incubation time after adding IPTG; (ii) selection of bacterial colonies. To address (i), four
incubation times (2, 3, 4, and 5 hours) were tested. Running gels revealed that 4-hour
incubation time gave the best results. In (ii), different colonies from the same agar plate
were picked for protein growth. By running PAGE gels, it was found that the variation of
colonies had a profound influence on protein production. Figure A1.4 shows the effect of
colony on protein production. Besides two protein ladders, six protein lanes are shown in
this figure; the left three correspond to three flasks of LB broth solution right before
adding IPTG, while the right three correspond to the solution in the same flasks 4 hours
after adding IPTG. Although three colonies were grown in the same agar plate, different
117
colonies exhibited distinct protein production capability. In this work, only the medium
that contained the largest amount of protein was used in the protein synthesis.
T1eC subunits were conjugated to the maleimide-PEG2-biotin linkers at the S80C
mutation site. The assembly of T1eC subunits and T0 subunits around 10-nm gold
nanoparticles was done using a protocol developed by Swift et al.13 Biotin-labeled gold
nanoparticles were mixed with streptavidin in the ratio of 2:1. A control experiment
without the use of gold nanoparticles was also done. T1eC subunits and T0 subunits in
the 1:11 ratio were mixed in the high salt solution. Subsequently, biotin-labeled
apoferritin and streptavidin were mixed in the 2:1 ratio.
An attempt of purifying nanoparticle dimers by FPLC using a Superdex 200 10/300
size exclusion column was made. However, these ferritin-encapsulated gold nanoparticles
stuck to the packing material at the top of the column. Therefore, during the time I carried
out these experiments, samples without FPLC purification were used in following TEM
study. But it is noteworthy that later my colleagues found that stirring at room
temperature for 48 h could be a way alternative to the thermocycling protocol for the
assembly of thermophilic ferritin subunits around gold nanoparticles; more importantly,
these gold nanoparticles can pass the FPLC column. This implies the likelihood that gold
nanoparticles were exposed or protein was denatured in the thermocycling process.
TEM images of gold nanoparticle dimers and apoferritin dimers are shown in Figures
A1.5A and A1.5B, respectively. Results support the effectiveness of using streptavidin to
link two biotin-labeled ferritins. However, examination of the whole TEM grid revealed
the presence of both the nanoparticle dimers and unreacted nanoparticle monomers. This
118
implies that the production efficiency of the nanoparticle dimers from monomers needs to
be improved. Plans for further experiments on this project will include the use of 5-nm
gold nanoparticles as the metal core during the assembly of thermophilic ferritins, and
performing the assembly process via the room temperature stirring method.
Experimental Section
The pAF0834 vector containing T0 gene was provided by the laboratory of Dr. Eric
Johnson at Caltech.14 The S80C mutant was engineered using the Stratagene QuikChange
Site-Directed Mutagenesis Kit. The mutated cDNA was transformed into XL1-Blue
Supercompetent E. coli cells (Stratagene) according to the manufacturer’s protocol. DNA
was isolated using a QIAprep Spin Miniprep Kit (Qiagen).
The plasmid for growing T1eC was transformed into BL21-CodonPlus (DE3)-RP
competent cells of E. coli. Cell cultures were grown in 1-L LB broth supplemented with
200 μg/mL ampicillin and incubated (37 °C, 225 rpm) until O.D.600nm was between
0.9~1.0. Protein growth was induced by IPTG. After incubation at 37 °C, cells were
harvested by centrifugation (5k rpm, 15 min) and resuspended in 40-mL buffer (10 mM
Tris, 10 mM NaCl, pH 8.4, one protease inhibitor cocktail tablet). The cells were
sonicated on ice (3 × 2 min) and then centrifuged (5 krpm, 10 min). The lysis supernatant
was heat shocked (80 °C, 10 min) and ultracentrifuged (25 krpm, 90 min). The
supernatant was purified on a 5 mL HiTrap FastFlow Q anion exchange column (GE
Healthcare) (10 mM Tris, pH 8.4 with a 10 mM NaCl to 600 mM NaCl elution gradient;
5 mL/min flow rate). The protein was equilibrated in 1 M NaCl for 24 h, concentrated
119
(100k MWCO Centricon, Millipore) and purified by size exclusion chromatography
using a Superdex 200 10/300 GL column (GE Healthcare) (20 mM sodium phosphate,
800 mM NaCl, pH 7.6 buffer; 0.5 mL/min flow rate).
Prior to conjugation, T1eC was dissolved in 20 mM sodium phosphate, 20 mM NaCl,
pH 7.6 buffer for unfolding. Disulfide bonds were reduced using 5 mM TCEP for 30
minutes at room temperature, followed by TCEP removal using a 10DG desalting column
(20 mM sodium phosphate, 20 mM NaCl, pH 7.0 buffer). Biotin was conjugated to T1eC
subunits via the reaction of maleimide with the mutated cysteine. The biotinylation
reagent was added in 20-fold excess. To determine the biotin labeling efficiency, the
Ellman’s Reagent (Pierce # 22582) was used. In a T1eC solution (87.0 μM calculated
based on the subunits), the free –SH concentration was found to be 17.9 μM. The
percentage of unreacted –SH is 17.9 μM / 87.0 μM = 20.6%. Thus the biotin labeling
efficiency is ~80%. To After removing the excess biotinylation reagent, T1eC subunits
(10 nM) and T0 subunits (110 nM) were mixed in the presence of 10-nm gold
nanoparticles (10 nM). The reaction mixture was annealed with a thermal cycling
program comprised of 50 cycles of 2.5 min periods at 55 °C followed by 12 min periods
at 4 °C. Biotin-labeled ferritin-encapsulated gold nanoparticles were mixed with
streptavidin in the ratio of 2:1.
References
1. Talley, C. E., Jackson, J. B., Oubre, C., Grady, N. K., Hollars, C. W., Lane, S. M.,
Huser, T. R., Nordlander, P. & Halas, N. J. Surface-enhanced Raman scattering
120
from individual Au nanoparticles and nanoparticle dimer substrates. Nano Lett. 5,
1569-1574 (2005).
2. Reinhard, B. M., Sheikholeslami, S., Mastroianni, A., Alivisatos, A. P. &
Liphardt, J. Use of plasmon coupling to reveal the dynamics of DNA bending and
cleavage by single EcoRV restriction enzymes. Proc. Natl. Acad. Sci. U. S. A.
104, 2667-2672 (2007).
3. Kneipp, K., Wang, Y., Kneipp, H., Itzkan, I., Dasari, R. R. & Feld, M. S.
Population pumping of excited vibrational states by spontaneous surface-
enhanced Raman scattering. Phys. Rev. Lett. 76, 2444-2447 (1996).
4. Nordlander, P., Oubre, C., Prodan, E., Li, K. & Stockman, M. I. Plasmon
hybridizaton in nanoparticle dimers. Nano Lett. 4, 899-903 (2004).
5. Sonnichsen, C., Reinhard, B. M., Liphardt, J. & Alivisatos, A. P. A molecular
ruler based on plasmon coupling of single gold and silver nanoparticles. Nat.
Biotechnol. 23, 741-745 (2005).
6. Ringler, M., Klar, T. A., Schwemer, A., Susha, A. S., Stehr, J., Raschke, G.,
Funk, S., Borowski, M., Nichtl, A., Kurzinger, K., Phillips, R. T. & Feldmann, J.
Moving nanoparticies with Raman scattering. Nano Lett. 7, 2753-2757 (2007).
7. Sheikholeslami, S., Jun, Y. W., Jain, P. K. & Alivisatos, A. P. Coupling of optical
resonances in a compositionally asymmetric plasmonic nanoparticle dimer. Nano
Lett. 10, 2655-2660 (2010).
121
8. Mastroianni, A. J., Claridge, S. A. & Alivisatos, A. P. Pyramidal and chiral
groupings of gold nanocrystals assembled using DNA scaffolds. J. Am. Chem.
Soc. 131, 8455-8459 (2009).
9. Claridge, S. A., Liang, H. Y. W., Basu, S. R., Frechet, J. M. J. & Alivisatos, A. P.
Isolation of discrete nanoparticle: DNA conjugates for plasmonic applications.
Nano Lett. 8, 1202-1206 (2008).
10. Brousseau, L. C., Novak, J. P., Marinakos, S. M. & Feldheim, D. L. Assembly of
phenylacetylene-bridged gold nanocluster dimers and trimers. Adv. Mater. 11,
447-449 (1999).
11. Yim, T. J., Wang, Y. & Zhang, X. Synthesis of a gold nanoparticle dimer
plasmonic resonator through two-phase-mediated functionalization.
Nanotechnology 19, 435605 (2008).
12. Sardar, R., Heap, T. B. & Shumaker-Parry, J. S. Versatile solid phase synthesis of
gold nanoparticle dimers using an asymmetric functionalization approach. J. Am.
Chem. Soc. 129, 5356-5357 (2007).
13. Swift, J., Butts, C. A., Cheung-Lau, J., Yerubandi, V. & Dmochowski, I. J.
Efficient self-assembly of Archaeoglobus fulgidus ferritin around metallic cores.
Langmuir 25, 5219-5225 (2009).
14. Johnson, E., Cascio, D., Sawaya, M. R., Gingery, M. & Schroder, I. Crystal
structures of a tetrahedral open pore ferritin from the hyperthermophilic archaeon
Archaeoglobus fulgidus. Structure 13, 637-648 (2005).
122
15. Weber, P. C., Ohlendorf, D. H., Wendoloski, J. J. & Salemme, F. R. Structural
origins of high-affinity biotin binding to streptavidin. Science 243, 85-88 (1989).
123
Figure A1.1. Cartoon showing hyperthermophilic ferritin’s capability of disassembling at low salt concentrations and reassembling in the presence of gold nanoparticles.
124
(A) Structure of T1eC dimer distance between two S80C mutation sites on a T1eC dimer = 2.9 nm
(B) Structure of the biotinylation reagent length of the maleimide-PEG2-biotin linker = 2.9 nm
(C) Structure of streptavidin complexed with four biotins distance between adjacent biotin binding sites on streptavidin = 2.1 nm
(D) Summary of the distance information
Figure A1.2. Structures of (A) T1eC dimer, (B) the biotinylation reagent, and (C) streptavidin complexed with four biotins. (D) Summary of distance information.
N
O
O
NH
OO
O
HN
O
S
NHHN
O
Cys
Cys
Biotin
Biotin
2.9 nm
2.9 nm
2.9 nm
2.1 nm
125
Figure A1.3. Cartoon illustrations of the biotinylation reaction (top) and formation of the streptavidin-biotin complexation system (bottom). In these illustrations, an assumption that Cys remains solvent exposed after nanoparticle incorporation is made.
N
O
O
NH
OO
OHN
O
S
NHHN
O
N
O
O
NH
OO
O
HN
O
S
NHHN
O
N
O
O
HN
OO
ONH
O
S
NHHN
O
S
S
SH
SH
126
1 2 3 4 5 6 7 8 Figure A1.4. PAGE gel showing the effect of colony on protein production. Experimental conditions: voltage = 200 V; time = 35 min; stained by Coomassie. Lanes 1 and 8: protein ladders; lanes 2~4: LB broth solution in three flasks right before adding IPTG; lanes 5~7: LB broth solution in the same three flasks 4 hours after adding IPTG. LB broth solution in 1.5-mL microcentrifuge tubes has been kept in 80 °C water bath for 10 min before running the gel.
127
A
B Figure A1.5. TEM images of (A) gold nanoparticle dimers (red circles) and (B) apoferritin dimers (red circles).
128
Appendix 2: DNA-Based Streptavidin-Biotin Complexation
System
There are two basic types of nanotechnological construction: (i) ‘top-down’
construction, where atoms or molecules are microscopically manipulated to form elegant
patterns; and, (ii) ‘bottom-up’ construction, where molecules self-assemble in parallel
steps as a function of their molecular recognition properties. As a chemically based self-
assembly system, double-stranded DNA plays a key role in bottom-up nanotechnology
because of its attractive molecular recognition capability and well-predicted duplex
conformation.1,2 These nanometer-scale DNA building blocks provide the foundation for
a particular research field termed structural DNA nanotechnology -- building DNA
nanoarchitectures from the molecular level for a broad range of applications from new
materials to devices.
Structural DNA nanotechnology was first introduced in 1982 when Seeman proposed
the idea of combining the sticky-end cohesion and branched DNA junctions to make
geometric objects and periodic two- or three-dimensional lattices.3 More specifically, he
theoretically predicted the self-assembly of six-arm branched DNA junctions into an
ordered three-dimensional lattice. Later Seeman experimentally constructed such an
object topologically equivalent to a cube.4 A general rule of the sequence design for the
DNA building blocks is to minimize the sequence symmetry in the branched structure to
avoid possible undesired pairing between participating strands and mobility of the
junction points. Also, the DNA assemblies are based on the precise 1:1 association
129
between the strands involved, thus elaborate control of the purity and the relative
stoichiometry of the strands are required to ensure a high yield of the desired assembled
structure.
Addressing functional groups on DNA nanostructures has been a burgeoning research
area in structural DNA nanotechnology. Incorporating molecular components in the DNA
motif is necessary so that the DNA nanostructure can serve as a template for assembling
secondary components. A variety of methods have been used to attach molecules to
specific building blocks, or tiles, in the DNA nanostructure. These methods include the
incorporation of an additional double-stranded DNA hairpin which projects
perpendicularly from the plane of the array,5,6 the use of biotin-labeled DNA to bind
streptavidin,5,7 the use of biotin-labeled polyamides to bind streptavidin,8,9 the
conjugation of DNA to gold nanoparticles,10,11 the insertion of the thrombin-binding
aptamer and subsequent binding of thrombin,12 the insertion of DNA aptamers that can
be recognized by single-chain antibodies,13 and the recognition of a DNA-peptide fusion
by an antibody.14
It should be noted that DNA nanostructures are prone to dissociation as a result of the
change of the solution condition, including variation of the media and increase in
temperature. We aim at enhancing the stability of DNA nanostructure in this contribution.
We borrow the idea of incorporating molecular components in the DNA motif and
thereby introduce a streptavidin-biotin complexation unit to our DNA nanostructure. As a
prototypal study, a DNA complex featuring a grid structure is chosen.
130
The construction of the nanogrid is the basis of this work. LaBean and co-workers
designed and constructed a two-dimensional DNA nanogrid structure, which was
composed of nine strands, with one strand participating in all four junctions.7 This work
inspired Mao and co-workers to develop the concept of “sequence symmetry”. Utilizing
the fourfold symmetry of cross-shaped tiles, they provided a new method of designing
symmetric tile motifs.15 Sequence symmetry reduces the number of unique DNA strands
needed, thus minimizing both cost and experimental errors. Sequence symmetry also
helps to minimize the required unique sequence space, and thus greatly simplifies the
sequence design. In our work, the DNA nanogrid was constructed following Mao’s
design method.
Figure A2.1 depicts the structure of the DNA nanogrids, which are two-dimensional
and are composed of repeating DNA tiles. The red, yellow, and green strands in this
figure correspond to strands i, ii, and iii, respectively (see Experimental Section for the
sequences). Bridging these DNA tiles relies on the designed 5′-end (CGCG) and 3’-end
(CATG) in iii. In this bridging mode, four biotin molecules (designed in strand iii) reside
at each splint. Therefore, when the streptavidin is added after the formation of the grid,
two streptavidin molecules are expected to bind to each splint via a two-biotin-one-
streptavidin binding mode.
Figure A2.2 shows the AFM images of the DNA nanogrids and streptavidin-bound
DNA nanogrids. To test if such a design can truly increase the binding strength of the
splint and thus enhance the whole stability of the DNA nanogrid, two tests are performed,
namely, gel test and melting temperature test.
131
A crucial point in the stability test is to examine the DNA nanogrid before and after the
splints dissociate. To perform an important control experiment, one more DNA strand
was also used as a replacement of iii. Designated as iv, its sequence is 5′-/biotin/CGT
TAC CGT GTG GTT GCA TAG T/biotin/-3′. When using iv instead of iii, only discrete
DNA tiles are formed, and its property should be intrinsically comparable to the grid
dissociated at the splint sites. Figure A2.3 shows result from an agarose gel test. It is
apparent that the splints dissociated when running the gel (lane 3). But when biotins were
bound with streptavidin, the DNA nanogrid was still held up and behaved as a high
molecular weight species (lane 4).
Figure A2.4A shows the results from melting temperature measurement; Figure A2.4B
is the first derivative analysis of the data in Figure A2.4A. It clearly reveals that the
splints in the grid dissociated at ca. 52 °C (curve 3), while no dissociation was observed
at this temperature when the grid is bound with streptavidin (curve 4). The discrete DNA
tiles were dissociated beyond 60 °C in all four cases.
Both gel test and melting temperature test support that the stability of the DNA
nanogrids can be improved after streptavidin complexation. A high resolution AFM
image of streptavidin-bound DNA nanogrids is needed to publish this work. Future
applications of this DNA nanogrid include the use of this material as the substrate in
photo activation light microscopy and on solid-supported fluid membrane surfaces to
probe cell-substrate interactions.
Experimental Section
132
All oligonucleotides were purchased from IDT, Inc. and purified by denaturing PAGE.
Strand i: 5′-AGG CAC CAT CGT AGG TTT TCT TGC CAG GCA CCA TCG TAG
GTT TTC TTG CCA GGC ACC ATC GTA GGT TTT CTT GCC AGG CAC CAT CGT
AGG TTT TCT TGC C-3′; strand ii: 5′-ACT ATG CAA CCT GCC TGG CAA GCC
TAC GAT GGA CAC GGT AAC G-3′; strand iii: 5′-/biotin/CGC GCG TTA CCG TGT
GGT TGC ATA GTC ATG/biotin/-3′; strand iv: 5′-/biotin/CGT TAC CGT GTG GTT
GCA TAG T/biotin/-3′.
To prepare the DNA nanogrids: strands i (1 μM), ii (4 μM), and iii (4 μM) were
combined in Tris/acetate/EDTA/Mg2+ (TAE-Mg2+) buffer, which consisted of Tris (40
mM, pH 8.0), CH3COOH (20 mM), EDTA (2 mM), and Mg(CH3COO)2 (12.5 mM).
DNA nanogrids were formed by slowly cooling the DNA solution from 95 °C to 5 °C
over 40 h. Solution containing DNA nanogrids was stored at 4 °C and samples were
normally used within 24 hours. DNA nanogrid solution was slowly added to streptavidin
TAE-Mg2+ solution at 4 °C. Streptavidin-bound DNA nanogrid solution was stored at
4 °C and samples were normally used within 24 hours.
References
1. Seeman, N. C. DNA in a material world. Nature 421, 427-431 (2003).
2. Ke, Y. G., Lindsay, S., Chang, Y., Liu, Y. & Yan, H. Self-assembled water-
soluble nucleic acid probe tiles for label-free RNA hybridization assays. Science
319, 180-183 (2008).
133
3. Seeman, N. C. Nucleic-acid junctions and lattices. J. Theor. Biol. 99, 237-247
(1982).
4. Chen, J. H. & Seeman, N. C. Synthesis from DNA of a molecule with the
connectivity of a cube. Nature 350, 631-633 (1991).
5. Winfree, E., Liu, F. R., Wenzler, L. A. & Seeman, N. C. Design and self-
assembly of two-dimensional DNA crystals. Nature 394, 539-544 (1998).
6. Liu, F. R., Sha, R. J. & Seeman, N. C. Modifying the surface features of two-
dimensional DNA crystals. J. Am. Chem. Soc. 121, 917-922 (1999).
7. Yan, H., Park, S. H., Finkelstein, G., Reif, J. H. & LaBean, T. H. DNA-templated
self-assembly of protein arrays and highly conductive nanowires. Science 301,
1882-1884 (2003).
8. Cohen, J. D., Sadowski, J. P. & Dervan, P. B. Addressing single molecules on
DNA nanostructures. Angew. Chem. Int. Ed. 46, 7956-7959 (2007).
9. Cohen, J. D., Sadowski, J. P. & Dervan, P. B. Programming multiple protein
patterns on a single DNA nanostructure. J. Am. Chem. Soc. 130, 402-403 (2008).
10. Le, J. D., Pinto, Y., Seeman, N. C., Musier-Forsyth, K., Taton, T. A. & Kiehl, R.
A. DNA-templated self-assembly of metallic nanocomponent arrays on a surface.
Nano Lett. 4, 2343-2347 (2004).
11. Sharma, J., Chhabra, R., Liu, Y., Ke, Y. G. & Yan, H. DNA-templated self-
assembly of two-dimensional and periodical gold nanoparticle arrays. Angew.
Chem. Int. Ed. 45, 730-735 (2006).
134
12. Liu, Y., Lin, C. X., Li, H. Y. & Yan, H. Protein nanoarrays: aptamer-directed self-
assembly of protein arrays on a DNA nanostructure. Angew. Chem. Int. Ed. 44,
4333-4338 (2005).
13. Li, H. Y., LaBean, T. H. & Kenan, D. J. Single-chain antibodies against DNA
aptamers for use as adapter molecules on DNA tile arrays in nanoscale materials
organization. Org. Biomol. Chem. 4, 3420-3426 (2006).
14. Williams, B. A. R., Lund, K., Liu, Y., Yan, H. & Chaput, J. C. Self-assembled
peptide nanoarrays: an approach to studying protein-protein interactions. Angew.
Chem. Int. Ed. 46, 3051-3054 (2007).
15. He, Y., Tian, Y., Chen, Y., Deng, Z. X., Ribbe, A. E. & Mao, C. D. Sequence
symmetry as a tool for designing DNA nanostructures. Angew. Chem. Int. Ed. 44,
6694-6696 (2005).
135
Figure A2.1. Illustration of the designed DNA nanogrids. The red, yellow, and green strands correspond to strands i, ii, and iii, respectively.
136
A B Figure A2.2. AFM images of (A) DNA nanogrids and (B) streptavidin-bound DNA nanogrids.
137
a b Figure A2.3. Agarose gel test. Lanes a and b are lane markers. The box on the right-hand side explains the materials loaded in lanes 1, 2, 3, and 4. Strand i: 5′-AGG CAC CAT CGT AGG TTT TCT TGC CAG GCA CCA TCG TAG GTT TTC TTG CCA GGC ACC ATC GTA GGT TTT CTT GCC AGG CAC CAT CGT AGG TTT TCT TGC C-3′; strand ii: 5′-ACT ATG CAA CCT GCC TGG CAA GCC TAC GAT GGA CAC GGT AAC G-3′; strand iii: 5′-/biotin/CGC GCG TTA CCG TGT GGT TGC ATA GTC ATG/biotin/-3′; strand iv: 5′-/biotin/CGT TAC CGT GTG GTT GCA TAG T/biotin/-3′.
138
30 40 50 60 70 80 90
0.36
0.40
0.44
0.48
0.52
0.56
0.60
O.D
. 260n
m
Temperature / oC
(1) (2) (3) (4)
30 40 50 60 70 80 900.0
0.2
0.4
0.6
0.8
1.0
Firs
t Der
ivat
ive
Temperature / oC
(1) (2) (3) (4)
A B Figure A2.4. (A) Melting temperature measurement result. (B) First derivative analysis of the data in (A). In (A) and (B), the curves 1- 4 correspond to the same materials loaded in lanes 1-4 as specified in the caption of Figure A2.3.