INTERACTION OF FERRITIN WITH TRANSITION METAL IONS AND CHELATES

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

http://repository.upenn.edu?utm_source=repository.upenn.edu%2Fedissertations%2F321&utm_medium=PDF&utm_campaign=PDFCoverPages

http://repository.upenn.edu/edissertations?utm_source=repository.upenn.edu%2Fedissertations%2F321&utm_medium=PDF&utm_campaign=PDFCoverPages

http://repository.upenn.edu/edissertations?utm_source=repository.upenn.edu%2Fedissertations%2F321&utm_medium=PDF&utm_campaign=PDFCoverPages

http://network.bepress.com/hgg/discipline/137?utm_source=repository.upenn.edu%2Fedissertations%2F321&utm_medium=PDF&utm_campaign=PDFCoverPages

http://repository.upenn.edu/edissertations/321?utm_source=repository.upenn.edu%2Fedissertations%2F321&utm_medium=PDF&utm_campaign=PDFCoverPages

http://repository.upenn.edu/edissertations/321

mailto:[email protected]

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

http://repository.upenn.edu/edissertations/321?utm_source=repository.upenn.edu%2Fedissertations%2F321&utm_medium=PDF&utm_campaign=PDFCoverPages

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).

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


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


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


0 50 100 150 200

0.2

0.4

0.6

0.8

1.0

1.2


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.

Documents

INTERACTION OF FERRITIN WITH TRANSITION METAL IONS AND CHELATES