Gold Nanoparticles for Efficient Tumour Targeting ... · Representative TEM images, absorption spectra and DLS data of five gold nanoparticle sizes synthesized for testing blood pharmacokinetics

Gold Nanoparticles for Efficient Tumour Targeting: Materials, Biology & Application

by

Steven David Perrault

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Institute of Biomaterials & Biomedical Engineering University of Toronto

© Copyright by Steven David Perrault, 2010

ii

Nanoparticles for Efficient Tumour Target: Materials, Biology & Application

Steven David Perrault

Doctor of Philosophy

Institute of Biomaterials & Biomedical Engineering University of Toronto

2010

Abstract

As of 2010, cancer remains the leading cause of death in Canada11, 1, and

second in the United States of America2. This is despite decades of research into

development of chemotherapeutics and diagnostics. A number of major challenges

have prevented new discoveries from translating into a reduction in mortality rates. One

challenge is the poor efficiency with which anti-cancer agents (diagnostic contrast

agents and therapeutics) reach deregulated cells in the body3. Therefore, development

of new methods and technologies for improving efficiency of delivery has been a focus

of research. Nanoparticles are leading candidates for improving the efficiency of

delivery because they can act as payload vehicles for anti-cancer agents, because it is

possible to mediate their interaction with biological systems and thus their

pharmaockinetics, and because they can exploit inherent vulnerabilities of tumours4, 5.

This thesis describes the results from a series of research projects designed to progress

our understanding of how nanoparticles behave in vivo, and how their design can be

optimized to improve tumour targeting.

iii

Dedication To Heather, for her support and friendship.

To Shirley and Dennis, for everything.

iv

Acknowledgments First and foremost, I would like to thank Dr. Warren Chan for giving me a chance

to join his group after an unfortunate start to my PhD. I am happy to say that my PhD experience has been extremely positive since making that move. Moreover, I was fortunate to end up with an academic advisor who was able and willing to provide the training that I needed to succeed, and who didn’t give up while I worked to improve my writing style. If some day I have a laboratory and my own students to work with, I’ll know that I have an excellent mentor to look to.

As well, I would like to sincerely thank:

• Dr. Chris Yip and Dr. Gang Zheng for being a part of my thesis committee and providing constructive, helpful feedback. They helped me to see my data and experiments from a different perspective, resulting in better studies and publications than I would have otherwise achieved.

• Dr. Marianna Foldvari at the School of Pharmacy, University of Waterloo, for participating in my defense as external examiner.

• Heather for her patience and support when I needed both, and her friendship throughout my PhD and beyond.

• My parents for all of the support and encouragement that they’ve given me over the years, without which I would not have had the confidence to pursue a career in science.

• Carl Walkey for his friendship and always appropriate sense of humor.

• Past members of the Chan lab: Dr. Travis Jennings, Hans Fischer, Tanya Hauck, Robin Anderson, Barb Alexander, and Arezou Ghazani were all incredibly welcoming and made my start in the INBS laboratory easy and fun.

• Current members of the Chan lab, who have been enthusiastic team members.

• John, for encouraging me to finish being a student. Dave, for his sense of humor and company. Gundars and Jenn, for good friendship, meals and their encouragement.

v

Table of Contents

Dedication ....................................................................................................................... iii

Acknowledgments ........................................................................................................... iv

Table of Contents ............................................................................................................. v

List of Tables .................................................................................................................... x

List of Figures ................................................................................................................. xi

Chapter 1 ......................................................................................................................... 1

General Introduction ........................................................................................................ 1

1.1 Nanomaterials and Devices .................................................................................. 1

1.1.1 Definition and General Description ............................................................. 1

1.1.2 Types of Nanomaterials .............................................................................. 5

1.1.3 Nanomaterial Synthesis and Modification ................................................... 8

1.1.4 Current Applications of Gold Nanoparticles ................................................ 9

1.2 Nanomedicine ..................................................................................................... 11

1.2.1 Nanotechnology, Disease Diagnosis and Treatment ................................ 11

1.2.2 Nanotechnology and Cancer .................................................................... 12

1.3 Summary ............................................................................................................. 13

Chapter 2 ....................................................................................................................... 16

Synthesis of Gold Nanoparticles with Controlled Size and Surface Chemistry .............. 16

2.1 Introduction ......................................................................................................... 16

2.2 Materials and Methods ........................................................................................ 18

2.2.1 General Colloidal Gold Synthesis Technique ........................................... 18

2.2.2 Nanoparticle Characterization .................................................................. 18

2.2.3 Optimization and Characterization of Particle Pegylation ......................... 19

vi

2.3 Results ................................................................................................................ 21

2.3.1 Colloidal Gold Nanoparticle Synthesis ...................................................... 21

2.3.2 Optimization of Particle Pegylation: Buffer pH .......................................... 24

2.3.3 Optimization of Particle Pegylation: Loading Density ............................... 24

2.3.4 Synthesis and Pegylation of Various Gold Nanoparticle Designs ............. 27

2.4 Discussion ........................................................................................................... 29

2.5 Chapter Author Contributions .............................................................................. 32

Chapter 3 ....................................................................................................................... 33

A Novel Technique for Synthesizing Highly Monodispersed, Spheroidal Gold Nanoparticles of 50-200 nm ....................................................................................... 33

3.1 Introduction ......................................................................................................... 33


3.2.1 Hydroquinone Synthesis of Gold Nanoparticles ....................................... 36

3.2.2 Particle Pegylation .................................................................................... 37

3.2.3 Particle Characterization .......................................................................... 37

3.2.4 Statistics ................................................................................................... 38

3.3 Results ................................................................................................................ 39

3.3.1 Nanoparticle Growth Using Hydroquinone ............................................... 39

3.3.2 Kinetic Growth of Nanoparticle Seeds ...................................................... 43

3.3.3 Impact of Sodium Citrate Concentration on Particle Shape ...................... 43

3.3.4 Comparison of Citrate- and Hydroquinone-Synthesized AuNP ................ 44

3.3.5 Surface Modification of Hydroquinone Particles ....................................... 44

3.4 Discussion ........................................................................................................... 50


Chapter 4 ....................................................................................................................... 55

Mediating Tumour Targeting Efficiency of Nanoparticles Through Design .................... 55

vii

4.1 Introduction ......................................................................................................... 55


4.2.1 Cell Culture ............................................................................................... 57

4.2.2 Nanoparticle Synthesis and Pegylation .................................................... 57

4.2.3 Particle Characterization .......................................................................... 57

4.2.4 Examination of Particle Stability ............................................................... 58

4.2.5 Quantitative Sandwich ELISA for Detection of mPEG-AuNP ................... 58

4.2.6 Mapping Pharmacokinetic Dependence on Particle and mPEG Size ....... 59

4.2.7 Tumour Xenografts, In Vivo Biodistribution and Tumour Accumulation .... 60

4.2.8 Histology and Silver Enhancement ........................................................... 61

4.2.9 Quantification of Particle Permeation ....................................................... 61

4.2.10 Statistics ................................................................................................... 62

4.3 Results ................................................................................................................ 63

4.3.1 Particle Synthesis, Pegylation and Characterization ................................ 63

4.3.2 Particle Stability and ELISA Validation ..................................................... 63

4.3.3 Design-Dependent Blood Pharmacokinetics of mPEG-AuNP .................. 64

4.3.4 Design-Dependent Biodistribution and Tumour Accumulation of mPEG-AuNP in Tumour-Bearing Mice ................................................................. 71

4.3.5 Particle Size-Dependent Permeation of the Tumour Interstitium .............. 75

4.4 Discussion ........................................................................................................... 78

4.4.1 Particle Design and Synthesis .................................................................. 78

4.4.2 Blood Pharmacokinetics and Clearance of Sub-100 nm mPEG-AuNP ... 78

4.4.3 Tumour Accumulation ............................................................................... 79

4.4.4 Particle Size Dependence of Tumour Permeation .................................... 81

4.4.5 General Discussion .................................................................................. 81


viii

Chapter 5 ....................................................................................................................... 85

In Vivo Assembly of Nanoparticle Device Components to Improve Targeted Cancer Imaging ...................................................................................................................... 85

5.1 Introduction ......................................................................................................... 85


5.2.1 Synthesis of Gold Nanoparticles............................................................... 87

5.2.2 Characterization and Pegylation of AuNP ................................................ 87

5.2.3 Cell Culture ............................................................................................... 88

5.2.4 Tumour Xenograft ..................................................................................... 89

5.2.5 Biodistribution Study and ICP-MS ............................................................ 89

5.2.6 Förster Resonance Energy Transfer ........................................................ 90

5.2.7 Optical Imaging Data Collection and Analysis .......................................... 91

5.2.8 Histology, Silver Enhancement & Fluorescence Microscopy .................... 92

5.2.9 Statistics ................................................................................................... 92

5.3 Results ................................................................................................................ 93

5.3.1 Nanoparticle Anchor Component Synthesis ............................................. 93

5.3.2 In Vivo Behavior of the Anchor Component .............................................. 93

5.3.3 Imaging of In Vivo Kinetic Behavior: 2 Hours Post-Injection ..................... 96

5.3.4 Imaging of In Vivo Kinetic Behavior: 24 Hours Post-Injection ................. 105

5.4 Discussion ......................................................................................................... 109

5.5 Chapter Author Contributions ............................................................................ 112

Chapter 6 ..................................................................................................................... 113

Conclusions and Future Work ...................................................................................... 113

6.1 Statement of Major Conclusions ....................................................................... 113

6.2 Future Work ...................................................................................................... 115

6.2.1 Nanoparticle Synthesis ........................................................................... 115

ix

6.2.2 Future Work: Continuing to Improve Our Understanding of Nanoparticle Design and In Vivo Behavior. ............................................ 118

6.2.3 Future Work: Improving and adapting In Vivo Assembly. ....................... 121

Curriculum Vitae of Steven D. Perrault ........................................................................ 128

A. Personal Information ......................................................................................... 129

ISI ResearcherID: .................................................................................................... 129

Mailing Address: ...................................................................................................... 129

Laboratory: .............................................................................................................. 129

Fax: .......................................................................................................................... 129

Email: ....................................................................................................................... 129

Date & Place of Birth: .............................................................................................. 129

Citizenship: .............................................................................................................. 129

B. Education .......................................................................................................... 130

C. Employment and Training History ..................................................................... 130

D. Certifications ..................................................................................................... 131

E. Community Involvement & Volunteering ........................................................... 131

F. Scholarships, Fellowships and Awards ............................................................. 131

G. Publications ....................................................................................................... 132

H. Technical Reports ............................................................................................. 133

I. Book Chapters .................................................................................................. 134

K. Teaching Experience ........................................................................................ 135

Bibliography ................................................................................................................. 137

x

List of Tables

Table 1. Characterization of five batches of AuNP synthesized using different volumes

of 1% sodium citrate…………………………………………………………………………...22

Table 2. Particle Sets A and B, their design parameters, and resulting blood half-life

and tumour accumulation……………………………………………………………………65

xi

List of Figures

Figure 1. Particle size-dependent emission spectra of fluorescent quantum dots. ..........3

Figure 2. Biological molecules and structures on the nano scale. . ................................4

Figure 3. Examples of nanomaterial types that can be synthesized.. ..............................6

Figure 4. Various shapes and sizes of nanomaterials.. ...................................................7

Figure 5. Leading Causes of death in Canada, 2005.. .................................................. 14

Figure 6. Characterization of 5 batches of AuNP synthesized by the Frens technique

and using varying volumes of citrate solution for gold reduction .................................... 23

Figure 7. pH dependence of particle pegylation and AuNP-mPEG stability in 2M NaCl.

....................................................................................................................................... 25

Figure 8. mPEG-SH-5000 loading density and AuNP-mPEG stability in 1.5M NaCl. .... 26

Figure 9. Hydrodynamic diameter and ζ-potential of gold nanoparticle and mPEG-SH

combinations. ................................................................................................................. 28

Figure 10. Ultraviolet-visible absorption spectra of five nanoparticle batches produced

with varying quantities of nanoparticle seed .................................................................. 40

Figure 11. Particle size and polydispersity dependence on reaction seed quantity....... 42

Figure 12. Reaction kinetics of hydroquinone-mediated nanoparticle growth in the

presence and absence of nanoparticle seeds ................................................................ 45

Figure 13. Transmission electron microscopy images of particles produced with varying

concentrations of sodium citrate .................................................................................... 46

Figure 14. Size and shape dispersity of citrate- and hydroquinone-synthesized AuNP..

....................................................................................................................................... 47

xii

Figure 15. Comparison of citrate- and HQ-synthesized particles, and pegylation of HQ

synthesized AuNP .......................................................................................................... 48

Figure 16. Pre-and post-filtering absorption spectra of pegylated AuNP ....................... 49

Figure 17. Representative TEM images, absorption spectra and DLS data of five gold

nanoparticle sizes synthesized for testing blood pharmacokinetics ............................... 66

Figure 18. Representative TEM images, absorption spectra and DLS data of five gold

nanoparticle sizes synthesized to examine biodistribution, tumour accumulation and

tumour permeation ......................................................................................................... 67

Figure 19. Standard curve of optical absorbance versus percent of injected dose for the

61.3 nm HD particles from particle set B ........................................................................ 68

Figure 20. Testing particle Stability and protein interference for detection of mPEG-

AuNP by ELISA .............................................................................................................. 69

Figure 21. Map of mPEG-AuNP blood half-life as a function of particle size and mPEG-

SH molecular weight ...................................................................................................... 70

Figure 22. Comparison of mPEG-AuNP half-lives in CD1 versus CD1 athymic nude

mice, and the pharmacokinetic profiles of particle set B in CD1 athymic nude tumour

bearing mice. ................................................................................................................. 73

Figure 23. Reticuloendothelial system organ uptake, tumour accumulation, and

correlation of particle design to tumour accumulation .................................................... 74

Figure 24. Particle size-dependent permeation of the tumour interstitial space ............ 77

Figure 25. Characterization of the gold nanoparticles used as the anchor component . 94

Figure 26. Optimization of particle pegylation for streptavidin-Alexa fluor 750 binding. 95

Figure 27. Biodistribution, tumour localization and assembly of the two components ... 97

xiii

Figure 28. Förster Resonance Energy Transfer between an Alexa fluor dye pair bound

to the biotin-AuNP .......................................................................................................... 99

Figure 29. Region-of-interest analysis from fluorescence data ................................... 102

Figure 30. Region-of-Interest analysis of non-target (mouse body) normalized to the

initial signal (Io) for 100 minutes post-injection of strept-A750 ..................................... 103

Figure 31. Optical detection of fluorescence and analysis of tumour accumulation over

2 HPI of assembling contrast agent ............................................................................. 104

Figure 32. Signal Intensity (I) in mouse bodies normalized to the initial signal (Io) over

24 HPI of strept-A750 .................................................................................................. 106

Figure 33. Optical detection of fluorescence and analysis of tumour accumulation over

24 HPI of streptavidin-Alexa fluor 750 .......................................................................... 107

Figure 34. Schematic of nanoparticles assembling with contrast agent in vivo ........... 110

1

Chapter 1

General Introduction Scientists have made enormous strides in their ability to engineer materials on

the nanometer scale over the past few decades. This can be contributed to

improvements in instrumentation required for their characterization such as electron

microscopy and dynamic light scattering. It has also been driven by the motivation that

useful devices can be produced using the unique properties of nanometer scale

materials. Many of the proposed applications for nanomaterials are in biomedical

research or medicine. Improving delivery of anti-cancer agents to tumours is one of the

most promising4-7. In fact, several ground-breaking nanoparticle chemotherapeutics are

now in clinical trial4. For these advances to reach their utmost potential, researchers

must demonstrate that materials can be rationally designed and synthesized,

systematically investigate the interaction of nanomaterials with biological systems, and

demonstrate how such knowledge can be applied to overcome barriers to tumour

targeting.

1.1 Nanomaterials and Devices

1.1.1 Definition and General Description

Nanomaterials are engineered structures having all of their physical dimensions

between 1-100 nm8. Their physical structure is therefore larger than the atomic but

smaller than the bulk scale of materials. On this scale, some materials gain interesting

optical, electronic and catalytic properties. Some metal nanoparticles display intense

surface plasmons that result in very strong optical absorbance9, 10. Semiconductor

nanoparticles with dimensions below a material-specific threshold undergo quantum

confinement of their electrons, and convert ultraviolet absorbance into fluorescence

with narrow emission bands in the visible to near-infrared region11 (Figure 1).

2

Catalytic properties are observed in nanoparticles made of reactive elements (e.g.

silver) because their high surface-to-volume ratio exposes a large proportion of atoms

to the surface12. These properties can be exploited and have obvious biomedical

applications. A common feature of nanotechnology applications is their dependence

on the size, shape and composition of the material. This dependence of function on

structure provides engineers with a mechanism to tune or optimize the behavior of

materials towards specific applications.

A second advantage of nanomaterials is that their scale overlaps with that of

biologically relevant molecules (i.e. deoxyribonucleic acid and proteins) and structures

(i.e. ribosomes, viruses, Figure 2). The integration of synthesized nanomaterials with

biomolecules can therefore have profound effects on either or both, providing an

additional basis for novel applications. For example, small (i.e. 2 nm diameter)

metallic particles can be conjugated to antibodies (approximate hydrodynamic

diameter (HD) of 15 nm) and provide a tag for sensing their presence in an assay13.

Alternatively, large (100 nm) particles could have their surface covered with antibodies,

increasing the antibody’s binding kinetics through higher avidity and providing a

mechanism for the particle to interact with a target, or to be captured on a substrate14.

The integration of nanotechnology with biomaterials also provides a means for

predictable assembly of synthetic materials into multi-component devices.

Nanotechnology-based devices can be considered as complexes of either a

single or various types of nanomaterials, assembled or co-operating in a manner that

provides some emergent function8. For example, an engineered implant for drug

delivery could include multiple nanomaterials integrated together, with the whole

providing an improved drug release profile versus a single nanoparticle design15. This

relatively new concept has emerged from the rationale that biomedical knowledge of

DNA hybridization and protein binding can be applied to allow predictable assembly of

inorganic structures16-18.

The unique properties of nanomaterials, their capacity for integration with

biomolecules, and the functionality that can be achieved through device assembly

form the basis of nanomedicine and the research described in this thesis.

3

Figure 1. Particle size-dependent emission spectra of fluorescent quantum dots. Quantum dot fluorescence emission wavelength is a function of the particle’s size, with

larger particles emitting at longer wavelengths. Modified with permission from Walkey,

Sykes and Chan, 201019.

4

Figure 2. Biological molecules and structures on the nano scale. Nanomaterials

are defined as synthetic, engineered materials having all dimensions between 1 and

100 nm. This scale overlaps with that of biomedically relevant classes of biomolecules,

viruses and small cells. Unpublished image was provided by Leo Chou of the Chan lab.

5

1.1.2 Types of Nanomaterials

Nanomaterial engineers have successfully produced an enormous variety of

materials, each with many possible iterations and modifications. A comprehensive

description of nanomaterial classes is therefore difficult, although there have been

recent attempts to establish a system of nomenclature20. Examples of commonly used

material types include polymer, semiconductor, and metal particles, as well as carbon

nanotubes (Figure 3). All of these are typically modified for use in specific applications,

resulting in hybrid materials that complicate categorization. One must also consider that

for many materials, a range of sizes and shapes can be synthesized (Figure 4).

Indeed, some synthesis methods may lead to products that are highly polydispersed,

further complicating classification.

Classification of nanomaterials might also consider their function or application.

For example, whereas quantum dots would generally be used as fluorescent probes,

they could also act as platforms to modify behavior of surface-bound ligands. As well,

many different nanomaterials are fluorescent, but not all offer the same advantages. In

general, quantum dots may be most useful as fluorescent detection reagents for in vitro

biological assays11, metallic particles such as colloidal silver or gold have very strong

optical absorption and could be used for sensing9, 10, as platforms for functional

molecules21, 22, or simply as model nanoparticles systems23-25. Gold nanorods translate

their strong near-infrared optical absorption into heat, and may be useful as localized

therapeutics in tissue26. Polymer nanoparticles do not display the interesting properties

common to those composed of transition metals, but have shown promise as carriers

for drugs or contrast agents27, 28.

The research described herein makes use of colloidal gold nanoparticles (AuNP).

We describe in detail the methods commonly used for their synthesis and

characterization, and a new preparative technique developed in our laboratory. We

then use colloidal gold as a model system to test a biologically relevant hypothesis, and

to develop a novel application for use in tumour surveillance.

6

Figure 3. Examples of nanomaterial types that can be synthesized. Many different

types and classes of nanomaterials can be synthesized. A) Transmission electron

image of carbon Nanotubes, B) Solutions of colloidal quantum dots, C) transmission

electron image of various colloidal nanoparticle shapes and sizes and D) transmission

electron image of colloidal gold. Image was reproduced by Dr. Tanya Hauck, University

of Toronto.

7

Figure 4. Various shapes and sizes of nanomaterials. This figure illustrates only a

few of the numerous shapes and sizes of metallic nanomaterials that can be

synthesized, including rods, spheroids, and prisms. Image reproduced with permission

from Dr. Tanya Hauck, University of Toronto.

8

1.1.3 Nanomaterial Synthesis and Modification

Nanomaterials are synthesized using a top-down or bottom-up approach, each

suitable for different materials and applications8. Top-down synthesis methods first

determine the final structure of the material (or device), then “prints” this, typically

through lithography techniques. This approach is promising for chemically modifying or

patterning surfaces, for example to produce arrays of biomolecules or polymers.

However, a current lack of high throughput methods limits the capacity of scale-up with

this approach. As well, it is not yet possible to print complex materials such as

nanocircuits or multi-component devices using a top-down approach.

In contrast to this, bottom-up synthesis uses nucleation of atoms in solution to

produce materials whose final geometry or composition is controlled through regulation

of the reaction conditions and constituents. This approach allows for rapid and

relatively high-throughput batch synthesis. There are many recent examples where

efforts to understand and improve control of reaction kinetics have resulted in more

homogenous batch preparations29-32. Nevertheless, bottom-up synthesis methods

using batch-production still suffer from some degree of heterogeneity, such that no two

batches of nanomaterials can be considered as exactly identical.

Once synthesized, nanomaterials intended for biomedical applications need to be

functionalized for the application of interest. This usually involves addition of a surface

coating to the particles, such as a biomolecule22, 33-35 or polymer coating21, 36, 37. In the

case of particles synthesized in organic solvents (i.e. quantum dots), the surface may

be passivated by addition of a hydrophilic polymer to allow transfer into an aqueous

solvent11. Colloidal gold synthesized in aqueous solution can be directly modified to

provide functionality. This involves exchanging the stabilizing agent present on the

surface after synthesis with some other moiety21, 24. Proteins24, DNA38, and polymers

such as poly-(ethylene glycol) (PEG)21, 36 are examples of commonly used surface

coatings for biomedical applications of colloidal gold. Because the surface and surface-

bound moieties define the material’s interaction with target molecules, cells and tissues,

the quality and success of surface coating will greatly impact that interaction.

9

The research described herein uses bottom-up synthesis methods to produce

colloidal AuNP. Where possible, we aimed to use single batches of particles for

collection of all experimental data within a study to remove error from synthesis

variability. We optimized the addition of a PEG brush layer onto the nanoparticle

surface, as well as modified PEG that provides additional functionality. The particles

were then used for various biomedical applications.

1.1.4 Current Applications of Gold Nanoparticles

Perhaps owing to centuries old fascination, or to the variety of potential

applications that can be imagined, gold nanoparticles have become one of the most

common scientifically used nanomaterials. Gold nanoparticles can be synthesized by

different protocols to have a variety of shapes and sizes, including spheroids39-41,

nanorods30 and many additional shapes42. This matters because the properties of gold

nanoparticles that make them interesting and scientifically useful result from their

surface-to-volume ratio, their conductance band electron properties (surface plasmon),

and from the interaction of this with their dielectric environment, all of which are

determined by their size and shape.

One of the first major demonstrations of how the intense optical properties of

spheroidal gold nanoparticles can be usefully applied was by Mirkin et al. in 1996, who

showed that DNA-capped gold nanoparticles can assemble into controlled aggregates

via DNA hybridization, resulting in a distinct shift in surface plasmon properties

detectable by the naked eye43, 44. This concept has been developed by Nanosphere

Inc. into a commercialized product for detection of nucleic acids, and could potentially

replace the polymerase chain reaction as a sensitive method for detection of pathogen

nuclear material in clinical samples. Similarly, the same team discovered that the

melting properties of hybridized nucleic acids strands show sharper transitions and

greater discrimination of single base-pair mismatches when bound to gold nanoparticles

than free in solution45, 46. This phenomenon can be used to characterize relevant single

nucleotide polymorphisms, for example those implicated in patient response to a

pharmacologic, or genetic susceptibility to cancer.

10

Gold nanoparticle-DNA conjugates have also been tested as vehicles for gene

silencing DNA or RNA transcripts47-50. Gene silencing is well known for its ability to

specifically abrogate a molecular pathway of interest, and successful delivery in vivo

could make a major impact on genetic diseases such as cancer. Naked nucleic acid

has no ability to associate with cell membranes and undergo endocytosis, such that

gene silencing requires the therapeutic transcript to be delivered through a vehicle or

agent, for example through transfection with lipofectamine. Mirkin et al. have shown

that DNA gold conjugates are readily taken up by cells, and that the particle bound DNA

can act as anti-sense transcripts to down-regulate target genes47, 49. While this

technology has yet to be demonstrated in vivo, it clearly demonstrates the advantage of

using rationally designed gold nanoparticles as a reagent for biomedical applications.

Another intersting use for gold particles is molecular sensing based on Raman

spectroscopy, and enhancement of Raman signatures by nanoparticle surface

plasmons9, 51, 52. Here, the presence of an analyte causes sequestration or aggregation

of particles, resulting in electron dense “hotspots” upon irradiation with an incident

wavelength. Dye molecules attached to the particles and localized in the hotspots

experience enhancement of their vibrational state orders of magnitude larger than

normal. This spectroscopic method of detection has the advantage of ultra-high

sensitivity53, and has also been applied for in vivo sensing22.

Gold nanorods30, 54 and nanoshells55 are recent additions to the nanoparticle

toolbox that show promise as therapeutic agents. These variations on particle design

result in a second (nanorod) or shifted (nanoshell) absorption peak in the near-infrared

region. In both cases, the strong absorption is translated into a significant amount of

thermal energy. Both nanorods and nanoshells have been used for therapeutic thermal

ablation26, 55-59 or as part of a multi-component nanodevice in animal tumor studies56, 57.

Finally, gold nanoparticles make an excellent model system for testing a wide

variety of biological hypothesis. This is because they are easy to synthesize,

characterize and modify. As well, the breadth of sizes and shapes that can be

synthesized for a single material make it possible to test how these design parameters

impact the interaction of nanomaterials and biological systems. As examples, Chan et

11

al. used gold nanoparticles to demonstrate that size and shape23, 24, as well as ligand

density25 have a major impact on endocytosis. As well, Tamarkin used gold

nanoparticles as passive targeting vehicles to deliver therapeutic agents to tumors in a

murine cancer model21.

Gold nanoparticles have a long history of being used in biomedical research.

The striking and intense visible range absorbance resulting from their surface plasmon

has been used in a variety of diagnostic and therapeutic applications. Moreover, the

ability to synthesize and modify the particles into a diverse range of designs makes

them an ideal model system for studying the nano-bio interface.

1.2 Nanomedicine

1.2.1 Nanotechnology, Disease Diagnosis and Treatment

The past twenty years have seen a convergence of nano-engineering and

biomedical research, giving rise to many new technologies and ideas about what can be

achieved in medicine. Although few nanomaterial-based technologies have been

translated into clinical practice as of 2010, there is hope that they will cause a paradigm

shift in how medicine is administered, improving its effectiveness and efficiency.

Outside of prevention, diseases are managed through diagnosis and therapy.

Nanotechnology will make major contributions to both of these. For many diseases,

accurate diagnostic methods exist but tend to be slow, expensive, and labour-intensive.

For other diseases, particularly cancer, methods for accurate and sensitive diagnosis

remain elusive. Opportunities for nanotechnology to improve on current diagnostic

technologies are therefore vast. Various research teams have demonstrated systems

that use nanomaterials to achieve accurate, sensitive, rapid, and often multiplexed

detection of disease markers5, 60-62. As well, genome and proteome knowledge gained

from cancer biology research has been combined with micro- and nanotechnology in an

attempt to discover accurate methods for cancer diagnosis and tumour surveillance5.

The goal of reducing cost is perhaps a comparable challenge to accurate cancer

12

diagnosis, but technologies that are faster, multiplexed and more portable could achieve

efficiency in this area as well.

Nanotechnology will also allow new therapeutics or formulations of therapeutics

to be developed. Examples of nanomaterials with therapeutic properties include gold

nanorods for hyperthermia-based treatment26, 63 and silver nanoparticles as an anti-

microbial agent. In other cases, nanoparticles may act as platforms for therapeutic

biomolecules21, or vehicles for chemotherapeutics64, 65. Therapeutic biomolecules

engineered to alter a disease’s molecular pathway can be incorporated onto a

nanoparticle surface25 to increase potential for modulation of target cells.

Nanotechnology is poised to drive significant improvements in disease diagnosis

and treatment. There is hope that its greatest impact may be in improving how cancer

is managed, with a corresponding decrease in mortality. In this thesis, we describe a

series of studies that develop novel materials for use in vivo¸ systematically test how

nanomaterial design may impact usefulness in tumour targeting, and demonstrate a

novel strategy for delivering anti-cancer agents to tumours.

1.2.2 Nanotechnology and Cancer

Cancer remains the leading cause of death in Canada, claiming approximately

200 lives per day (Figure 5)1. Current technologies used in the war against cancer are

generally ineffective because they do not specifically address its complexity or the

known barriers to effective detection and therapy5. Although these obstacles are

scientifically and technologically challenging, nanomedicine may provide the necessary

solutions.

Defeating cancer requires major improvements in early detection, as this is a

strong indicator of a successful treatment outcome66. As described in the preceding

sections, nanotechnology can contribute to this in various ways. The inherent

heterogeneity of cancer means that molecular characterization of biomarkers or biopsy

samples must be highly multiplexed60. Quantum dots are therefore an ideal reagent for

in vitro molecular profiling of cancer. Diagnostic imaging and contrast agents used for

13

surveillance must become more sensitive to allow detection of small tumours. This

might be achievable by delivering contrast agents to tumours using nanoparticle

vehicles6. Multiplexed in vivo molecular imaging is a longer-term goal, but such a

breakthrough would allow non-invasive characterization of deregulated cells.

Small anticancer compounds have failed to reduce mortality for a number of

reasons. Such therapeutics are generally discovered based on toxicity to mitotic cells in

vitro, but display poor pharmacokinetic behavior and tumour accumulation in vivo.

Nanoparticle vehicles can overcome this as their pharmacokinetics are design-

dependent and can be optimized for tumour accumulation3, 36, 67. As well as increasing

efficacy against cancer cells, this may also reduce the drastic side-effects caused to

healthy tissue. Deregulated cells that are not detected early become less differentiated

and more drug resistant. Biomolecules present on a nanoparticle can alter its path of

uptake into cells and recent studies have shown that this can reduce exocytosis by drug

resistant cells68, 69. Further, it may be possible to engineer the intracellular distribution

of a payload to increase toxicity to the nucleus.

Nanotechnology will have a major impact on our ability to detect and treat cancer

and could achieve significant reductions in mortality rates. Molecular biologists are

working towards unraveling the disease’s molecular networks5. This is allowing

discovery of diagnostic biomarkers and therapeutic targets that can be incorporated with

nanomaterials. To take advantage of this potential, we must gain an understanding of

how nanomaterial design impacts behavior in vivo, specifically in a context of cancer.

1.3 Summary The following chapters detail improved methods for synthesizing and modifying

gold nanoparticles, a systematic study of how particle design impacts in vivo behavior

and tumor accumulation, and a novel application for nanotechnology-based detection of

tumours. In Chapter 2, we revisit the original gold synthesis method and carefully

characterize the quality of batches that can be produced. In particular, we focus on the

resulting size and shape properties of the particle batches, as this becomes important in

14

Figure 5. Leading causes of death in Canada, 2005. Cancer remains the leading

cause of death in Canada, and was responsible for 29.3% or 67,343 deaths in 20051.

15

later chapters. Finally, in this chapter we optimize protocols for surface modification of

the gold nanoparticles with poly-(ethylene glycol), important for their use in vivo.

Chapter 3 describes a novel method for synthesizing gold nanoparticles using a seeded

growth procedure and hydroquinone as a reducing agent. The experiments within were

motivated by the realization in Chapter 2 that the citrate synthesis method is

fundamentally flawed by its inability to produce monodispersed, spheroidal batches of

particles outside of a narrow size range. Following the same analytical approach used

in Chapter 2, we characterize the gold nanoparticles produced using hydroquinone and

find them to be better than citrate synthesized particles in their size and shape

parameters. As well, we demonstrate that the hydroquinone particles can be modified

in the same manner as those produced by citrate reduction. The aim of the work in

Chapter 4 was to systematically test and describe how nanoparticle design impacts in

vivo behavior and tumour accumulation. Pegylated gold nanoparticles are used as a

model system, and we find that particle size and surface chemistry have a major impact

on tumour accumulation. The results from this work were used to design the study

described in Chapter 5, in which we use in vivo assembly of nanoparticle components to

improve delivery of a diagnostic contrast agent to tumours. While this approach still

faces some challenges in optimization, the concept overcomes some limitations of

conventional targeting strategies, and should therefore have an impact on our ability to

efficiently deliver contrast agents and detect tumours.

Nanotechnology has demonstrated enormous promise for improving the

detection and treatment of cancer. The last decade has seen exponential growth in

investment by governments as well as private industry into nanotechnology, suggesting

that we may now be at leading edge of a major breakthrough in biomedical science.

Before nanomaterial-based technologies are common in hospital clinics and

laboratories, researchers must focus on carefully and systematically understanding not

only the material properties themselves, but how these interact with biological systems.

This is absolutely necessary to allow the potential efficiency gains of nanotechnology to

be realized before public interest and funding wanes. It is hoped that the following

chapters contribute to this goal in a meaningful way.

16

Chapter 2

Synthesis of Gold Nanoparticles with Controlled Size and Surface Chemistry

2.1 Introduction Colloidal AuNP and methods of synthesis have been extensively studied for

several decades. Wet synthesis of AuNP involves reducing ionic gold to metallic gold in

solution, resulting in nucleation of the atoms into clusters. The growth process can be

controlled and the particles stabilized by a capping agent present in the reaction that

associates with the particle surface. Many different synthesis methods have been

described that result in particles of various sizes, shapes, and surface properties70. The

first method to be described used phosphorous to reduce gold salt71, but the resulting

product was not optimized due to the obviously poor availability of analytical

instrumentation in 1857. A second method pioneered by Turkevich in 195139 and

optimized by Frens in 197340, 40, 41 uses sodium citrate as both reducing agent and

stabilizer. This method can produce spherical particles of approximately fifteen to one

hundred nanometers, although size and shape dispersity increase dramatically with

larger sizes. As other examples, the Brust method can produce particles of

approximately 2 nm that are soluble in non-polar solvents72, and Murphy et al. have

described techniques for producing gold nanorods30 and other interesting shapes42.

Colloidal AuNP have many potential biomedical applications. For example,

optical excitation of their resonant surface plasmons is both sensitive and is tunable

through changes in geometry and composition, making them versatile probes for high-

17

throughput sensing assays9, 73. AuNP have also been used for detection of22, and

carrying therapeutics to21, 74 tumours. Finally, they are a valuable model system for

testing hypothesis about the interaction of nanomaterials with biological systems23, 24.

Their usefulness as a model system can be attributed to the ease of preparative

techniques, the capacity to make large and geometrically monodispersed batches, and

the range of sizes that can be produced. As well, methods for addition of polymers21, 75,

deoxyribonucleic acids (DNA)52, 76 or proteins23, 24, 77, 78 to the particle surface have been

well characterized. Formation of self-assembling monolayers (SAM), for example with

poly-(ethylene glycol)-SH (PEG-SH), onto the nanoparticle surface can be achieved by

co-ordination of the thiol moiety to the gold surface79. Formation of a stable PEG layer

on the particle surface is important for in vivo applications because it restricts the

interaction of immune factors with the particle and reduces clearance by the

reticuloendothelial system80-82. The PEG chain length appears to impact this effect36, 83,

but the effect has not been systematically tested over a broad range of particle designs.

Particle design can have major implications for how a nanoparticle will interact

with biological systems. Chithrani and Chan first demonstrated that protein-coated

AuNP of 50 nm diameter were taken up by cells to a greater degree than smaller (20

nm) or larger (100 nm) sizes23. The efficiency of nanoparticle-based tumour targeting

may also be dependent on design, particularly on geometry and PEG molecular weight.

In this chapter, we use the Turkevich-Frens method of colloidal gold synthesis to

produce particles of various sizes and modify these with thiolated-PEG of various

molecular weights. Establishing these methods and characterizing the resulting

products will provide a basis for the following chapters, in which we improve on the

Turkevich-Frens synthesis method, and then employ colloidal AuNP in vivo for tumour

targeting.

18

2.2 Materials and Methods

2.2.1 General Colloidal Gold Synthesis Technique

AuNP were synthesized following the technique described by Turkevich39 and

optimized by Frens40, 41. Batches of 30 mL total volume were prepared. A stir plate

(VWR) was set to 315˚C and the temperature was allowed to stabilize. 30 ml of milli-Q

filtered H2O was added to a 125 mL Erlenmeyer flask with a 3/4-inch stir bar. A 1%

solution (W/V) of gold chloride (Sigma-Aldrich P/N 50778) was prepared by adding 50

mg of salt to 5 ml of H2O in a Falcon tube. A 1% solution (W/V) of sodium citrate

tribasic dihydrate (Sigma-Aldrich P/N C0909) was prepared in a similar manner. 300 µL

of the gold chloride solution was added to the flask, which was placed on the heated stir

plate. The solution was stirred vigorously, and a volume of the citrate solution was

added when it reached a vigorous boil. The volume added was varied between 175 and

600 µL for different batches, with higher volumes producing smaller particles. The flask

was removed from heat 3-5 minutes after no further colour transition was visible by eye,

and was allowed to cool to room temperature. The synthesis was then filtered using a

60 ml syringe (BD P/N 309653) and 0.22 µm filter (Millipore P/N SLGP033RS) into a 50

mL falcon tube. Water was added to bring the total synthesis volume to 20 mL.

2.2.2 Nanoparticle Characterization

The particles were then characterized for HD, zeta-potential (ζ-potential), and

ultra-violet and visible (UV-VIS) light absorption. Briefly, 500 µL of the synthesis was

added to a plastic disposable cuvette (10x4x45, Sarstedt PN 67.742). Hydrodynamic

diameter and ζ-potential were measured by dynamic light scattering on a Zetasizer

(Malvern NanoZS) using Dispersion Technology Software (version 5.0), and standard

operating program parameters of sample refractive index = 0.2, absorbance=3.32,

dispersant of H2O at 25˚C, viscosity of 0.8872 cP, and refractive index=1.330.

Measurements were made in triplicate for each sample. After this, 500 µL of H2O was

added for a total volume of 1 ml. Optical absorption was measured between

wavelengths of 350-750 nm on a spectrophotometer (Shimadzu UV-1601PC) and using

19

UVProbe software (V. 2.21). Core diameter of the particles was determined from

images obtained by transmission electron microscopy (TEM). 10 µL of a dilute particle

solution (approximately 1-to-5 dilution of stock) was dropped onto carbon-coated copper grids

(Ted Pella P/N 01813-F) to dry. Images were collected on a Hitachi HD2000 STEM (Hitachi

Corp). Particle sizes were measured in Gatan Micrograph V3.11.2 (Gatan Inc) by drawing a line

across the particle’s longest axis (n=30).

Concentration of nanoparticles in the stock was determined using the Beer-

Lambert law. The extinction coefficient was calculated from the maximum extinction

wavelength and the empirically-derived equation

(10^(1.0643*(LOG(3/2*3.141592654*HD))+4.0935))84, where HD is the HD of the

particles (nm). Based on these values, the surface area of individual particles and

within a working volume was calculated.

2.2.3 Optimization and Characterization of Particle Pegylation

Thiol-modified methoxy-PEG (mPEG-SH) of various molecular weights (MW)

was used to modify synthesized AuNP, including mPEG-SH 5K (MW=5000 Da, NOF

Corp P/N ME-050-SH), and mPEG-SH 1K, 2K and 10K (MW=1000, 2000, 10,000 Da,

Laysan Bio, Inc P/N mPEG-SH-1000, mPEG-SH-2000, mPEG-SH-10,000). The thiol

group present on the various PEG reagents provides a co-ordination group between the

polymer and the gold surface, causing formation of a SAM.

To test the pH dependence of SAM formation, 1 mL aliquots of 37.2 nm (HD)

particles were centrifuged at 7500x g for 10 minutes in microcentrifuge tubes and were

resuspended in 980 µL 5 mM Na3PO4 buffer varying in pH between 5 and 11, 3 aliquot

repetitions for each pH. 0.66 µg mPEG-SH-5000 was solubolized in 20 µL H2O and

added into each solution, which was immediately vortexed and incubated 5 minutes at

room temperature. This amount of mPE-SH-5000 was previously found to provide

moderate resistance to NaCl aggregation for 1 mL of these particles. Then, 250 µL of

each reaction was added to 750 µL of 2 M NaCl. This was incubated for 10 minutes to

allow any aggregation to occur. Optical absorption spectra were obtained for each

20

sample, and SAM formation was assessed by stability of the particles, quantified by a

ratio of 530/580 nm absorbance. Statistical differences in absorption ratio were

determined using an ANOVA and post-hoc Tukey’s tests.

A similar approach was used to examine the loading density of mPEG-SH-5000

on the particle surface. 20 µL of various concentrations of mPEG-SH-5000 were added

to 980 µL of 37.2 nm (HD) AuNP in 5 mM Na3PO4 buffer, pH 10.0 (n=3 for each

concentration). The quantity of mPEG-SH-5000 added was calculated to add between

1 molecule/0.1 nm2 and 1 molecule/6.0 nm2 of particle surface area. The solutions were

vortexed immediately, and incubated 30 minutes at room temperature. 250 µL of each

sample was mixed with 750 µL 2 M NaCl, and stability was assessed and statistically

tested as described above.

21

2.3 Results

2.3.1 Colloidal Gold Nanoparticle Synthesis

Particle synthesis proceeded as expected. After addition of citrate into the

boiling gold chloride solution, there was a short time lapse before initiation of a colour

transition. The solution then turned from clear to dark black to purple, and finally to red.

The time of transition was not consistent between batches; however 5 minutes was an

approximate time for the colour transition to reach completion. We produced five

batches of AuNP (n=1), each with a different volume of citrate solution added. Details

of the citrate volume used, HD, and mean core diameter are found in Table 1.

Addition of decreasing volumes of citrate was found to produce larger particles,

as expected based on the report by Frens40. Particles were spheroidal or near-

spheroidal in shape, with shape consistency decreasing as batch particle size increased

(Figure 6 A-E). The batch of largest particle sizes were found to be quite elongated

and asymmetric, whereas the batch of smallest sizes were very homogenous and quite

spheroidal. The plasmon maximum absorption red-shifted, with the smallest particles

displaying a λmax of 519 nm and the largest of 545 nm (Figure 6 F-J). The absorption

peak was also observed to broaden significantly with increasing particle sizes, which is

suggestive of an increasing dispersion of particle sizes within a batch. The HD of

particles was found to be from three to ten nm larger than core diameters measured by

TEM, with the exception of the largest size (Figure 6 K-O). This exception may be due

to the quality of the dynamic light scattering (DLS) data that could be obtained from this

batch, which displayed higher size dispersion than batches of smaller sizes. Indeed,

the geometric standard deviation generally increased as batch particle diameter

increased, suggesting that quality may be highest with large citrate volumes and small

nanoparticle sizes.

22

Table 1. Characterization of five batches of AuNP (n=1) synthesized using different volumes of 1% sodium citrate.

Citrate Volume

(µL) Ζ-potential

(mV) Hydrodynamic Diameter (nm)

Geometric Mean TEM (nm, n=30)

Geometric Standard Deviation

(nm, n=30)

600 -41.3 23.58 17.66 1.09

450 -35.9 37.27 30.91 1.17

300 -46.3 46.79 44.78 1.11

200 -38.3 58.74 58.12 1.20

160 -37.2 76.07 86.39 1.29

23

Figure 6. Characterization of 5 batches of AuNP synthesized by the Frens technique and using varying volumes of citrate solution for gold reduction. A-E)

Representative TEM images of particles from the 5 batches. F-J) Optical absorption

spectra of the 5 batches, showing a red-shifting plasmon maximum with larger particle

sizes. K-O) Raw HD data of the 5 batches, each measured in triplicate.

24

2.3.2 Optimization of Particle Pegylation: Buffer pH

The stability of particles after pegylation in Na3PO4 buffer of varying pH was

assessed by examining the stability of the particles in high NaCl concentration. Stability

was quantified by measuring the absorption spectra of the particles and calculating a

ratio of absorbance at 530/580 nm (Figure 7). The 530/580 nm absorbance ratio was

found to increase as the buffer pH increased from pH 5.0 to 10.0. The 530/580 nm ratio

of the three higher mPEG-SH-5000 quantities were statistically significant (p<0.01) from

the 3 lower quantity treatments. The absorption spectra and therefore the 530/580 nm

ratio of the highest two mPEG-SH-5000 quantity reactions (1 mPEH-SH-5000 per 1 or 2

nm2) were identical to that of the raw AuNP synthesis (data not shown).

2.3.3 Optimization of Particle Pegylation: Loading Density

The stability of particles after pegylation with varying quantities of mPEG-SH-

5000 per nm2 of surface area in 5mM Na3PO4 pH 10.0 was assessed by examining the

stability of AuNP-mPEG in a high NaCl concentration. Stability was quantified by

measuring the absorbance spectra and calculating a 530/580 nm absorbance ratio

(Figure 8). The ratio (and particle stability) was found to increase as the quantity of

mPEG-SH-5000 added into the reaction increased, reaching a maximum at 1.0 mPEG-

SH-5000 molecule per 0.1 nm2. The standard deviation was generally greater than in

the pH optimization, but there was a statistically higher ratio (stability) at higher values

than at the lowest value (p<0.05). The 530/580 nm ratio of the 0.1 - 2 nm2 per molecule

treatments was identical to that of the raw AuNP synthesis (data not shown).

25

Figure 7. pH dependence of particle pegylation and AuNP-mPEG stability in 2M NaCl. Particles pegylated in lower pH Na3PO4 buffer (pH 5.0 – 7.0) had a significantly

lower 530/580 nm ratio than those pegylated in pH 8.5 – 11.0 buffers (*A vs. *B,

p<0.01).

26

Figure 8. mPEG-SH-5000 loading density and AuNP-mPEG stability in 1.5 M NaCl. Gold particles were pegylated in 5 mM Na3PO4 pH 10.0 with varying quantities of

mPEG-SH-5000 per nm2 surface area. Particle stability in 1.5 M NaCl quantified by

530/580 nm absorbance ratio showed increasing stability with higher quantities of

mPEG-SH-5000. 0.1 to 2 nm2 per molecule had significantly higher ratios than 1

molecules per 6 nm2 per molecule (*A vs. *B, p<0.05)

27

2.3.4 Synthesis and Pegylation of Various Gold Nanoparticle Designs

Based on the results above, the five batches of AuNP sizes were then pegylated

with mPEG-SH-1000, -2000, -5000, and -10,000 Da (n=3). Following this, HD and ζ-

potential were measured, allowing trends in these parameters to be examined as a

function of particle design.

Pegylation of AuNP cores of average diameter 44.78, 66.31 and 86.39 nm with

mPEG-SH-1000, and 86.73 with mPEG-SH-2000 Da was unsuccessful and produced

aggregates. These treatments were therefore excluded from further analysis. All other

combinations of mPEG-SH and AuNP core sizes produced stable particles.

Hydrodynamic diameter increased fairly consistently with core size and PEG molecular

weight (Figure 9A) and did not appear to reach a maximum that would suggest a

different type of mPEG-SH configuration on the particle surface. mPEG-SH-1000 and -

2000 increased HD by approximately 15-20 nm, or 5-10 nm per side for a single mPEG

molecule. mPEG-SH-5000 increased HD by 30 nm, and mPEG-SH-10000 by 40 nm.

Good quality measurements of ζ-potential could not be obtained for the 17.66 nm

particle core and mPEG-SH-5000 and -10000. Analysis of the remaining samples

showed that ζ-potential of the particles increased and became more positive with

increasing mPEG-SH molecular weight (Figure 9B). For a given molecular weight

mPEG-SH, ζ-potential of the smaller size cores tended to be more positive than larger

core sizes.

28

Figure 9. Hydrodynamic diameter and ζ-potential of gold nanoparticle and mPEG-SH combinations. A) Hydrodynamic diameter of 17.72 (♦), 31.28 (▲), 45.03 (●),

66.54 (■), and 86.73 (Ӿ) nm core sizes increased when combined with the 4 molecular

weight mPEG-SH. The raw AuNP stabilized with citrate are the points closest to the

vertical axis. B) ζ -potential became more positive with larger molecular weight mPEG-

SH.

29

2.4 Discussion Gold nanoparticles have many applications in biomedical research. They have

been used as one component in diagnostic21, 22, 63 and therapeutic21, 26, 63 cancer

applications, and as a model nanoparticle system for examining the interaction of

nanoparticles with biological systems23, 24. These examples demonstrate some

common qualities and advantages provided by AuNP, including the capability of current

preparative techniques to synthesize a variety of sizes and shapes, and to modify their

surface towards specific applications.

Gold nanoparticles can be synthesized using many different methods. Wet-

synthesis methods involve reducing ionic gold to metallic gold in solution and controlling

the size and shape of the nucleating aggregates through inclusion of a surface

stabilizer. In this chapter, we synthesized and modified a number of AuNP sizes. The

particles were synthesized using the common Turkevich-Frens method, where sodium

citrate is applied to a heated gold salt solution and acts as both reducing and

nanoparticle surface-stabilizing agent.

Particle size was controlled during synthesis by varying the volume of citrate

solution added, and was measured from images captured using TEM. As expected

based on Frens’ description40, larger volumes of citrate were found to produce smaller

particle sizes and vice-versa. The smallest particles had a diameter of 17.66 nm and

the largest 86.39 nm. These results represent the approximate limits that can be

produced with the Turkevich-Frens method. In our work, we found that addition of

increasing volumes of citrate can produce particles no smaller than 15 nm. Particles of

100 nm diameter can be produced, but the size and shape dispersity of these becomes

significant.

The polydispersity of the batches increased as particle diameter increased,

measured as the geometric standard deviation. Whereas the smallest particles appear

very homogenous in size and shape, few of the largest size can be considered spherical

and the geometric standard deviation increases from 1.09 to 1.29 nm. This finding was

validated by examining the UV-VIS spectral absorbance of the particles. The

absorption peaks of the smaller sizes were sharp and narrow, whereas the largest sizes

30

displayed very broad absorption peaks. Because the absorption spectra of colloidal

AuNP result from their surface plasmon resonance and are dependent on and sensitive

to size and shape, broad absorption spectra indicate polydispersed batches. The

Turkevich-Frens method may therefore be capable of producing particles of 15-100 nm,

but can only produce batches of high quality particles up to approximately 30 nm.

Following synthesis, we optimized methods for surface modification of AuNP with

mPEG-SH of various molecular weights which would be beneficial for in vivo

applications. Producing stable and ordered polymer layers on gold surfaces can be

achieved through thiol co-ordination85. Thiol-gold co-ordination depends on

deprotonation of the thiol group86, having an approximate pKa of 8.3. The efficiency of

pegylation should therefore be highly pH dependent. We examined this by reacting

mPEG-SH-5000 with citrate-stabilized AuNP over a range of pH, and quantifying

pegylation efficiency as particle stability in high NaCl concentration. Stability increased

significantly as the buffer pH increased, and was significantly better when above the

pKa of thiol versus below it. This suggests that AuNP-pegylation reactions would be

most efficient and the resulting products most stable if carried out at high pH, above the

pKa of the thiol group.

We also examined the loading efficiency of mPEG-SH on the AuNP surface.

This was done only for one nanoparticle size (mean TEM diameter=30.91 nm) and one

mPEG-SH molecular weight (5000 Da). However, the results provide a starting point for

understanding the size of the PEG footprint on the gold surface and the density of the

mPEG-SH SAM. Using particle stability in high NaCl concentration as a measure of

loading density, we observed that stability reached a maximum when one or more

mPEG-SH molecule was provided for each nm2 of particle surface area. There was a

small but non-significant increase in stability above 1 molecule/nm2, suggesting that

SAM density may continue to increase, however our detection method was not sensitive

enough to resolve the difference. Nevertheless, considering that mPEG-SH-5000 was

subsequently found to add 30 nm to the HD and therefore 15 nm per PEG molecule, a

loading density of 1 molecule per nm2 suggests a very dense PEG layer on the particle

surface.

31

Using conditions discovered to be efficient for mPEG-SH SAM formation, an

array of particle designs was produced. We combined 5 nanoparticle core sizes with 4

mPEG-SH molecular weights to allow characterization of HD and ζ-potential of the

products. The mPEG-SH-1000 and -2000 were not capable of forming stable sols with

the largest nanoparticle sizes, suggesting that there is a minimum molecular length

required to stabilize a given particle size. Particle HD was found to increase in step with

mPEG-SH molecular weight. The smallest mPEG-SH added 5-10 nm per molecule,

whereas the largest added approximately 20 nm. ζ -potential of citrate stabilized AuNP

was found to be very negative at -30 to -40 mV in 5 mM phosphate buffer pH 7.0.

Replacement of the citrate with mPEG-SH increased the ζ-potential in an mPEG-SH

molecular weight- and particle size-dependent manner. ζ -potential became more

positive and approached zero with larger molecular weight mPEG-SH layers, and

smaller particle core sizes.

Prior studies have applied gold nanoparticles of different sizes and surface

coatings to biological systems21, 23, 24, demonstrating the importance of design on

outcome. Poly-(ethylene glycol) surface coatings are currently the best means for

reducing the immune clearance of injectable particulate systems36, 80, 83, 87. In this

chapter, we investigated a common method of AuNP synthesis and the quality of

particles produced. We then established a protocol for modification of the particle

surface with PEG. These results should prove useful for downstream tumour-targeting

applications.

32

2.5 Chapter Author Contributions

All experiments were designed, performed and analyzed by Steve Perrault.

33

Chapter 3

A Novel Technique for Synthesizing Highly Monodispersed, Spheroidal Gold Nanoparticles of 50-200 nm

3.1 Introduction

Recent studies have focused on elucidating the impact of nanoparticle size and

shape on their interaction with biological systems, with implications to understanding

nanotoxicity and the rational design of nanoparticles for medical use23, 24, 88, 89. It is

important that such studies use particles with homogeneous sizes and shapes so that

the resulting data are representative of a specific particle design. Despite this need, a

model inorganic particle system in the 50-200 nm size range is currently unavailable.

Gold nanoparticles are frequently chosen as a model system for biological

studies due to their ease of synthesis and modification. Most AuNP wet-synthesis

methods involve the reduction of gold chloride in the presence of a stabilizer or capping

agent, and particle formation occurs over several steps90. The reduction of ionic to

metallic Au produces a supersaturated solution. This results in some degree of

nucleation, producing clusters of gold atoms. Particle growth proceeds as additional

Au is reduced and becomes deposited on the surface of formed nuclei. The resulting

nanoparticle size depends on the rate of Au reduction, and therefore on the reducing

agent used. Strong reducing agents such as phosphorous71 or sodium borohydride72

result in 2-10 nm particles, while the relatively weaker reducing strength of sodium

citrate results in 15-100 nm particles40. The variants of this approach can produce high

quality particles with diameters up to 30 nm. Methods for synthesizing larger diameter

particles with a comparably high quality are unavailable or are not appropriate for most

biological applications.

34

A promising alternative method is to reduce ionic gold in the presence of AuNP

seeds30, 91-100. By separating the nucleation and growth stages of synthesis, it is possible

to use high quality seeds as templates for growth, resulting in improved batches of large

diameter particles. The drawback of this approach is that Au reduction results in new

nucleation as well as growth of seeds, creating two populations31, 97. New particle

formation can be minimized by a slow addition of reducing agent, but this produces

abnormal particle shapes31. A recent study overcame these limitations using 2-

mercaptosuccinic acid (MSA) as a reducing and capping agent, which produced highly

monodispersed and quasi-spherical particles of 50 – 150 nm. They hypothesized that

Au reduction and addition to the nanoparticle surface was carefully controlled because

MSA co-ordinates tightly to the nanoparticle surface via its thiol group. Despite the

success of this approach in producing high quality nanoparticles over a broad size

range, displacing the surface-capping MSA with other polymers or biomolecules to

functionalize the particles for downstream applications is extremely difficult.

Jana and Murphy compared several other reducing agents for their ability to

selectively reduce Au onto 12 nm AuNP seeds rather than cause new particle

formation31. Ascorbic acid, a mild reducing agent, resulted in a large population of newly

formed nanoparticles. Sodium citrate produced new nucleation, as well as particle

aggregation and precipitation. A combination of hydrazine as reducing agent and

sodium dodecyl sulphate as a stabilizing agent resulted in only new nucleation. The

strongest reducing agent examined, sodium borohydride, caused both new nucleation

and seed growth, resulting in a heterogeneous population. These three reducing

agents therefore failed to catalyze seed growth over new nucleation and to result in

monodispersed populations, but alternative reducing agents capable of meeting these

requirements may exist.

Hydroquinone (HQ) has not been tested as a selective reducing agent for seeded

growth of gold nanoparticles, but is a potential candidate. Hydroquinone has a weak

reduction potential (E°=-0.699, vs. NHE)100, and is commonly used in film development

for its ability to selectively reduce silver onto preformed silver grains (i.e. seeds) without

causing new nucleation. This arises from the difference in reduction potential of

isolated silver ions in solution (Ag+/Ag0, Eº=-1.8 V) versus silver ions in the presence of

Ag0 clusters or nanoparticles (Eº=+0.799 V)101, 102.

35

The reduction of gold is more complex because an aqueous gold chloride

solution includes a variety of species. Wang et al. have hypothesized that AuI may

predominate in solution due to its stability103. As well, the reduction potential for Au in

the presence versus absence of metal “grains” is not as well characterized as for Ag,

but is estimated to be +1.002104 to +1.6831 vs. NHE near clusters compared to -1.50 vs.

NHE when isolated in aqueous solution31, 104. Hydroquinone should therefore not cause

Au reduction unless nanoparticle seeds are present, favoring growth over nucleation.

In this chapter, we examine hydroquinone as a selective reducing agent for seed-

mediated growth of AuNP. We find that it favors particle growth over new nucleation,

resulting in particle sizes of 50-200 nm, in line with estimates based on Au content in

the reaction. Characterization of the resulting batches by UV-VIS optical absorbance,

TEM image analysis, and DLS verifies that the hydroquinone method results in

homogenous populations of quasi-spherical particles. A direct comparison to the

classic sodium citrate method40, 41 is then used to quantify the improvement in batch

quality. Finally, we show that the hydroquinone-synthesized particles can be easily

modified for downstream applications, making this a reasonable approach for synthesis

of particles in the 50-200 nm size range.

36


3.2.1 Hydroquinone Synthesis of Gold Nanoparticles

A AuNP seed solution was prepared by sodium citrate reduction. 100 mL of

ultra-pure H2O and 1 mL of 1% (W/V) HAuCl4 was added to a 250 ml Erlenmeyer flask

with a clean stir rod. This was rapidly brought to a boil while stirring at the maximum

speed possible that did not cause splashing of the solution. As soon as the solution

was boiling, 3 mL of 1% (W/V) sodium citrate tribasic dihydrate solution was added.

The flask was removed from heat after 10 minutes, once nanoparticle maturation was

complete as indicated by no further colour transition. Large diameter AuNP used to

compare citrate reduction against hydroquinone were synthesized in a similar manner

but for a 30 mL total volume using 300 µL 1% HAuCl4, then addition of 180 µL of 1%

sodium citrate.

Seed-mediated particle growth with hydroquinone was tested in 10 mL batches.

Hydroquinone (Sigma-Aldrich P/N -17902) and sodium citrate solutions were prepared

fresh each day in ultra-pure H2O. Sodium citrate was not included in the reactions used

to generate the kinetic data, but was otherwise included. For a given synthesis, 100 µL

of a 1% (W/V) HAuCl4 solution was added to 9.4 - 9.8 mL of ultra-pure H2O (depending

on balance of water added with seeds) in a 20 mL scintillation vial. The determined

volume of particle seeds was then added (minimum of 7.7 µL and maximum of 406 µL).

The solution was then stirred rapidly at room temperature. A volume (11 to 132 µL) of

1% (W/V) sodium citrate solution was then added to the reaction, followed immediately

by 100 µL of 0.03 M hydroquinone solution. The reaction was stirred for 30 minutes to

overnight, depending on the speed of the reaction. Smaller particle sizes reached

completion very rapidly, as indicated by the colour transition, whereas growth of very

large particles required many hours to complete.

37

3.2.2 Particle Pegylation

The 50 nm particle batch produced in the presence of 22 µL 1% sodium citrate

solution was used to test particle pegylation. The concentration of the particles was

determined using Beer’s Law and the equation described in Section 2.2.2 to determine

the particle’s extinction coefficient. Based on concentration and particle size, we

calculated the available particle surface area (nm2/mL). We then prepared pegylation

reactions. A 1.0 and 0.1 mg/ml solution of mPEG-SH-5000 was prepared in ultra-pure

H2O. The appropriate number of PEG molecules for 1 mL of particles, and providing

various densities of PEG (from 10 to 0.08 molecules/nm2) was then added into 8

microcentrifuge tubes. 1 mL of the particles was then added on top of the PEG volume,

and was mixed immediately by vortexing. This was incubated 30 minutes at room

temperature, then centrifuged at 8000x g for 5 minutes to pellet the particles. 950 µL of

supernantant was removed, and the particles were resuspended in the ~50 µL

remaining volume. 20 µL was then transferred to a 0.6 mL microcentrifuge tube and

mixed with 5 µL 40% sucrose as a loading buffer. The samples were then loaded into

wells of a 0.5% (W/V) agarose (Sigma-Aldrich P/N A5093) gel prepared in 0.5X Tris-

borate buffer (0.5X TBE), and separated by electrophoresis at 25 volts for 15 minutes

then 100 volts for 45 minutes. The remaining particles were resuspended in 1 mL of 5

mM sodium phosphate buffer pH 7.0, which was then used to measure HD and ζ-

potential.

3.2.3 Particle Characterization

The absorption spectra of synthesized particles were measured as 1:2 dilutions

(in H2O) using a UV-1601PC spectrophotometer. Transmission electron microscopy

images were obtained by adding 10 µL of a 1:5 (stock:H2O) dilution of the particle

solution onto carbon-coated copper grids, then imaged on a Hitachi HD2000 STEM

(Hitachi Corp). Hydrodynamic diameter and ζ-potential was measured on a Nano-ZS

Zetasizer with particles resuspended in a 5 mM phosphate buffer solution pH 7.0.

38

Particle diameters and length-width ratios (n=100 per treatment) were

determined in ImageJ using the “Particle Analyzer” function. Briefly, images were

imported, and the scale determined by drawing a line across the image scale bar, then

“Analyze, Set Scale” was chosen. Images were then cropped to exclude the caption.

We then performed the following steps. “1. Image, Adjust, Threshold. 2. Process,

Binary, Watershed. 3. Analyze, Set Measurements (including Ferret’s Diameter and Fit

Ellipse). 4. Analyze Particles, setting variables Size=0-100000, Circularity=0-1,

Show=Outline, Display Results (Yes) and Exclude on Edges (Yes). We then examined

the resulting outline images to ensure proper analysis and excluded only

unrepresentative outlines.

3.2.4 Statistics

Pearson’s correlation was used to test for a linear correlation between the

resulting particle size from HQ seeded growth and the expected size based on gold

content in the reaction.

39

3.3 Results

3.3.1 Nanoparticle Growth Using Hydroquinone

We first examined hydroquinone as a reducing agent in five reactions, each

having a varied number of nanoparticle seeds added. Reactions containing no

nanoparticle seeds remained clear after addition of hydroquinone (Figure 10, inlay).

Otherwise, reduction of the gold chloride was apparent immediately after addition of

hydroquinone by a change in solution colour from transparent to dark purple or black.

The colour transition was noticeably slower when fewer seeds were present. In

reactions containing the highest number of seeds, the solution transitioned from

transparent to black to a red colour characteristic of smaller nanoparticles within

seconds. With low numbers of seeds, the colour transition was slower and reached a

dark blue colour within several minutes, which then transitioned to a red-brown colour

typical for large nanoparticles over tens of minutes or hours.

Ultra-violet-visible absorption measurements revealed spectra typical for AuNP

(Figure 10). Inclusion of many nanoparticle seeds resulted in a narrow absorption peak

with a maximum at 530 nm. In contrast, inclusion of very few seeds resulted in a broad

absorption spectra with two peaks at approximately 560/660 nm. With the exception of

this reaction, which contained very few nanoparticle seeds, all other reactions displayed

only a single, sharp absorption maximum between 530 nm (many seeds) and 580 nm

(few seeds).

Dynamic light scattering of the five reactions showed populations of increasing

diameter with lower reaction seed quantities (Figure 11A). Analysis of particle core

size on TEM images (n=100) cross-validated the DLS measurements (Figure 11A),

showing larger diameters when fewer seeds were included in the reaction. The

smallest particle population produced from inclusion of 7.8*1011 seeds had a HD of 53.8

nm and a core diameter of 50.0 nm. The largest particle population produced with

1.5*1010 seeds had a HD of 168.2 nm and a core size of 175.7 nm.

The predicted particle size of each batch was calculated based on the quantity of

Au atoms within each reaction, the atomic density of gold, and the number of particle

40

Figure 10. Ultraviolet-visible absorption spectra of five nanoparticle batches (n=1) produced with varying quantities of nanoparticle seed. The maximum

absorption wavelength increased as fewer seeds were added (from red, orange, yellow,

purple, to brown). The inlay photograph shows (from left to right) a reaction without

hydroquinone, without seeds, then from highest to lowest seed number reaction

products.

41

seeds in five different reactions that this Au volume would be divided between. The

predicted particle size could then be plotted against the actual core sizes measured

from TEM images, showing a strong linear and proportional relationship described by

y=1.067x-3.6312 (Figure 11b). This relationship was also found to be statistically

significant by testing predicted and resulting particle sizes using the Pearson’s

correlation test (p=0.001).

The polydispersity of the populations measured by dynamic light scattering

increased in proportion to the number of seeds included, and therefore inversely to the

resulting particle size (Figure 11C). The largest nanoparticles of 168.2 nm HD were

measured to have a polydispersity index value of 0.086, whereas the smallest

nanoparticles of 53.8 nm HD were measured at 0.200.

42

Figure 11. Particle size and polydispersity dependence on reaction seed quantity. A) Hydrodynamic diameter (―) and core diameter (− −) measured from TEM images.

B) Predicted vs. measured particle core size from five reactions with varying seed

number, revealing a significant linear correlation (p=0.001). C) Polydispersity measured

by DLS increased with higher seed number.

43

3.3.2 Kinetic Growth of Nanoparticle Seeds

To test the specificity of Au reduction by hydroquinone, we measured UV-VIS

absorption spectra at 1, 5 and 10 minutes after addition of hydroquinone to a gold

chloride solution with and without seeds (Figure 12). Particle formation was observed

in the absence of seeds in some repetitions. However a gold chloride solution may

contain a fraction of Au0 clusters because ionic gold can be reduced by UV radiation.

We therefore centrifuged the stock gold chloride solution at high speed for 60 minutes

and examined the kinetics again, this time finding no particle formation until seeds were

added. When AuNP seeds were present in the solution and hydroquinone was added,

there was a rapid growth in absorption, having a single peak between 500 and 600 nm

characteristic of large diameter AuNP.

3.3.3 Impact of Sodium Citrate Concentration on Particle Shape

Inclusion of sodium citrate as a particle stabilizer was examined over various

reaction concentrations between 37 µM to 450 µM. There was no obvious trend in HD

or batch polydispersity (measured by DLS) as the citrate concentration was altered

(data not shown). However, TEM images revealed intriguing differences in nanoparticle

shape (Figure 13). The lowest citrate concentration (37 µM) produced particles that

were quasi- spheroidal but many of the particles contained “knobs” on their surface

(Figure 13A). On closer examination, many of these appeared as small particles.

Whether these were “grown” as facets on the larger particle’s surface or were

independently grown particles or seeds that became associated with the particle surface

was unclear. However, the number of small nanoparticles observed that were

unassociated with larger particles was small enough to be unquantifiable by TEM, and

was roughly estimated at 0.001-0.1% of the population. No population of smaller

nanoparticle sizes was revealed by DLS. A high citrate concentration (450 um) also

produced interesting nanoparticle shapes (Figure 13E). Magnification of TEM images

showed that the knobs present on the particles did not appear as smaller particles, but

simply as focal points of growth. Moderate citrate concentrations of 75 – 300 µM

produced a homogenous population of quasi-spheroidal particles (Figure 13B-D).

44

Magnification of these images showed highly regularly nanoparticle surfaces with few

abnormalities (Figure 13F).

3.3.4 Comparison of Citrate- and Hydroquinone-Synthesized AuNP

We synthesized large diameter AuNP by the citrate method to allow a comparison

against particles synthesized by hydroquinone. Monodispersity of citrate particles was

relatively poorer (Figure 14A). With a geometric mean diameter of 82.51 nm, their

geometric standard deviation was 1.27. Monodispersity of 84.95 nm particles

synthesized by hydroquinone had a geometric standard deviation of 1.09. We

determined the ratio of longest and shortest axis lengths as a measure of shape. The

citrate-synthesized population had a mean ratio of 1.37 and coefficient of variation of

0.15, compared to a ratio of 1.13 and COV of 0.08 for HQ-synthesized particles (Figure

14B).

3.3.5 Surface Modification of Hydroquinone Particles We tested the ease of surface ligand exchange by pegylation of 50.0 nm

hydroquinone-synthesized particles. Efficiency of particle pegylation (mPEG-SH-5000)

was tested by incubating 50 nm AuNP aliquots with various PEG-to-surface area ratios.

Separating the products by gel electrophoresis showed a shift from negative to positive

charge as PEG density increased (Figure 15A). The hydrodynamic diameter increased

until saturation at 5-10 PEG/nm2 (Figure 15B), and, ζ-potential increased from -31.1 mV

(citrate) to -7.2 mV (5 PEG/nm2) (Figure 15B).

Following pegylation, the particles were easily concentrated by centrifugation into

a pellet, and were easy to re-disperse in solution. We tested particle dispersion by

passing them through a 0.22 μm filter before and after pegylation and centrifugation

(Figure 16). The absorption λmax decreased from 0.619 to 0.599, a 3.23% loss of

nanoparticle content.

45

Figure 12. Reaction kinetics of hydroquinone-mediated nanoparticle growth in the presence and absence of nanoparticle seeds. Hydroquinone was added to solutions

of gold chloride with (―) and without (− −) AuNP seeds. Absorption spectra were

obtained at 1, 5 and 10 minutes after addition of hydroquinone. Inclusion of AuNP

seeds in the reaction causes hydroquinone-mediated growth and emergence of an

absorption profile typical for AuNP with a maximum at 575 nm.

46

Figure 13. Transmission electron microscopy images of particles produced with varying concentrations of sodium citrate. Sodium citrate concentration was varied

from A) 37.4, B) 74.8, C) 150.0, D) 299.2, E) 448.8 µM. F) Shows a magnification of

particles boxed in in image C to show the regularity of their surface. Scale bar = 50 nm.

47

Figure 14. Size and shape dispersity of citrate- and HQ-synthesized AuNP. A)

Hydroquinone-synthesized particles (■) had a more narrow size distribution than citrate

synthesized particles (■). B) Hydroquinone-synthesized particles were more spheroidal

in shape than citrate synthesized particles, measured as a ratio of longest and shortest

particle axis.

48

Figure 15. Comparison of citrate- and HQ-synthesized particles, and pegylation of HQ synthesized AuNP. A) Electrophoretric separation of 10x-PEG to citrate particles

(left to right) shows pegylation-dependent migration. B) Bar graph (left axis) shows

increasing hydrodynamic diameter as more PEG molecules are added. Line graph

(right axis) shows a more positive ζ-potential with more PEG added.

49

Figure 16. Pre-and post-filtering absorption spectra of pegylated AuNP. Absorption spectra were obtained before (―) and after (―) pegylation, centrifugation

and re-dispersion into water, with minimal loss of particle content from aggregation.

50

3.4 Discussion Gold nanoparticles have been used as a diagnostic agent22 and therapeutic

delivery vehicle21 against cancer, as a model system for examining particle size-

dependence on endocytosis23, 24, and in many other applications. This illustrates the

broad potential of colloidal gold which can be attributed in part to its ease of synthesis,

the range of sizes and shapes that can be produced, and the quality of the product.

Many different protocols exist for synthesis of gold nanoparticles. Differences in

reaction conditions, reducing agent, and inclusion of other reagents (i.e. stabilizers or

surfactants) allow a broad variety of particle shapes and sizes to be produced. One of

the most common methods is that demonstrated by Turkevich39 and optimized by

Frens39-41. Their technique uses sodium citrate as a reducing and particle stabilizing

agent and can produce particles of 15-100 nm, but quality decreases significantly as

particle diameter increases. To improve on this, studies have examined seed-mediated

growth in which the nucleation and growth phases of synthesis are separated30, 91-100.

Although seeded growth could give rise to more homogenous size and shape

distributions, most reducing agents tested thus far have been unsuccessful in favoring

growth of seeds over new nucleation31. The only exception is the approach described

by Niu et al.99, but the high quality nanoparticles produced using MSA as a reducing

agent are difficult to modify and are therefore not useful in biological applications.

In this chapter, we described for the first time the use of hydroquinone as a

reducing agent for seed-mediated growth of gold nanoparticles. Hydroquinone is

commonly used as a reducing agent for film development because of its ability to

selectively add reduced silver onto pre-existing grains, rather than forming new

nuclei100. We reasoned that this may also be the case for ionic Au and gold

nanoparticles.

Hydroquinone has a weak reduction potential (E°=-0.699 vs. NHE)100 insufficient

to reduce gold salt that is isolated in solution (estimated at E°= -1.50 vs. NHE). But if

nanoparticle seeds are present in the solution, they can catalyze the reduction process

and the redox value of AuI approaches its bulk electrode redox value (E°=+1.6 vs.

NHE)104. If the reduced Au0 is then added onto the surface of nanoparticle seeds rather

51

than forming new metal clusters, the size of the resulting products should depend on the

number of seeds present, with fewer seeds giving rise to larger particles. To examine

this, we prepared five reactions of gold chloride with hydroquinone, and varied the

number of nanoparticle seeds present. The resulting products had an appearance

characteristic for gold nanoparticles. Their absorption spectra were narrow and

inclusion of fewer seeds resulted in a red-shifting absorption maximum, indicative of

increasingly larger particles. This was confirmed by characterizing the particle’s

hydrodynamic diameter using DLS, and core size from images obtained by TEM. Both

parameters followed a similar trend where inclusion of few seeds resulted in large

(175.7 nm) particles and many seeds resulted in small (50.0 nm) particles. Dynamic

light scattering did not reveal any smaller diameter particle population within the

reactions, and small particles were observed only very rarely in TEM images. Taken

together the characterization data suggested that hydroquinone successfully caused

growth of nanoparticle seeds, and that new nucleation was at most a rare event.

To further cross-validate this we compared the resulting size of reaction products

to the predicted size based on the quantity of gold chloride included in the reaction. If

the gold chloride is completely reduced by hydroquinone and contributes to seed growth

rather than new nucleation, the sizes of reaction products should be in line with

predicted sizes. If a significant amount of new nucleation occurs, product sizes should

fall short of predicted sizes. A line-of-best-fit through the measured and predicted size

data showed a strong correlation with a slope of 1.0 and was offset from the origin by

only 3.6 nm. This suggests that all or most of the gold chloride was consumed and

contributed to seed growth rather than new nucleation.

Kinetic analysis of reaction products also supported this argument. We

measured absorption spectra of gold chloride solutions including or lacking particle

seeds at several time points after addition of hydroquinone. Solutions containing

nanoparticle seeds rapidly produced an absorption spectra characteristic for large gold

nanoparticles, which was absent for the reaction lacking seeds. Hydroquinone may

therefore not be capable of reducing ionic gold and forming particles in the absence of

seed particles.

52

Jana and Murphy tested a number of reducing agents for their ability to promote

seed growth over new nucleation, including ascorbic acid, sodium borohydride,

hydrazine and sodium citrate31. None of these showed the specificity that we observed

with hydroquinone, instead causing new nucleation or particle aggregation. This is

surprising because ascorbic acid (E°=+0.125V vs. NHE) has a lower redox potential

than hydroquinone, and should therefore be less likely to reduce gold salt when

catalyzing seeds are absent. Hydroquinone may therefore provide some additional

property to the process. We found that seed-mediated growth of gold nanoparticle

seeds by hydroquinone did not aggregate even when no additional stabilizing agent was

added to the reaction. It may therefore have a high enough affinity for the particle

surface that reduction of AuI occurs within close proximity to the seed and increases the

likelihood of the Au0 atom contributing to seed growth. The mechanism by which

hydroquinone promotes seed growth over new nucleation could be of enormous interest

to nanomaterial engineering and should be investigated further.

Despite the advantages of the citrate reduction method for gold nanoparticle

synthesis, it is only able to produce relatively monodispersed and quasi-spheroidal

particles from approximately 15-30 nm. Larger particles can be produced but their size

and shape becomes detrimentally polydispersed. We carried out a direct comparison of

80 nm hydroquinone-synthesized particles to those produced by citrate reduction,

showing the former to be superior. The hydroquinone particles had a small geometric

standard deviation of 1.09 versus 1.27 for citrate. Particle shape of the two methods

also showed major differences, with sodium citrate producing irregular shapes having a

mean length-to-width axis ratio of 1.37, versus 1.13 for hydroquinone. Although not

perfectly spheroidal, the hydroquinone method can produce quasi-spheroidal

nanoparticles that are quite homogenous in size.

The ability to modify the particle surface in a controlled manner is also

enormously important for any nanomaterial that is destined for use in a biological

application. We found that the hydroquinone synthesized particles could be easily

modified by mixing with mPEG-SH-5000, suggesting that the reaction surface ligands

can be displaced. The pegylated products were non-aggregated and highly stable.

53

In summary, AuNP synthesized by hydroquinone have improved size and shape

distributions compared to citrate synthesized particles, and can be produced over a

broad size range. Furthermore, we have synthesized nanoparticles as small as 30 nm

and larger than 200 nm. With this strategy and those of previous methods, we can now

synthesize gold nanoparticles over a range of 2-200 nm. This large range allows us to

use AuNP as a model system to elucidate the nano-bio interface.

54

3.5 Chapter Author Contributions The hydroquinone synthesis method was conceived of, experiments performed and

data analyzed by Steve Perrault. The manuscript was written by Steve Perrault and

edited by Warren Chan.

55

Chapter 4

Mediating Tumour Targeting Efficiency of Nanoparticles Through Design

4.1 Introduction A central focus in nanomedicine research is the development of sub-100nm

structures as contrast agents, delivery vehicles or therapeutics for improving the

diagnosis and treatment of cancer. Recent studies have successfully demonstrated

selective targeting of engineered nanostructures (e.g., quantum dots35, 105, fullerenes106,

and gold nanoparticles21, 22) to tumours. For such nanostructures to advance toward

clinical use there is a need to understand how they interact with the biological systems

in question, in order to optimize their diagnostic sensitivity, payload, or therapeutic

efficiency, respectively.

The central premise of targeting nanoparticles to tumours emerged from the

discovery of the interesting pathology and morphology of tumours. Tumours have

immature and porous vasculature, which provides access to circulating particles107-109.

This finding initiated an interest in developing nanometer-sized vehicles to more

efficiently deliver diagnostic and therapeutic agents into tumours. Currently, only a

handful of therapeutic formulations have been designed to exploit tumour vessel

hyperpermability even after two decades of research4, 110, 111. The engineering

parameters to build efficient delivery vehicles for tumour targeting have not been

thoroughly identified since few studies have systematically evaluated the influence of

the dimensional, physical, or chemical properties of engineered nanostructures on

tumour targeting behavior.

The recent focus of nanotechnology applications in cancer biology has begun to

provide a fundamental understanding of how nanoparticles interact with biological

56

systems. This fundamental information is necessary to allow for a rational design of

nano-delivery systems. Various in vitro studies have demonstrated that nanoparticle

size and surface chemistry greatly impacts how they interact with plasma proteins88,

cellular uptake23, 75, 112, 113, toxicity113-115, and molecular response25. However, the in vivo

environment is far more complex than in vitro model systems and few studies have

investigated the in vivo behavior of engineered nanostructures. Discher et al. examined

blood clearance and tumour accumulation of filamentous versus spherical micelles,

which were found to influence both the delivery and therapeutic efficiency of the anti-

cancer drug Paclitaxal116. Because they did not test a wide array of sizes and aspect

ratios, it would be difficult to extrapolate their findings to other nanostructures.

In this chapter we systematically examine how particle design can be optimized

towards improving the efficiency of tumour targeting. Pegylated AuNP are an ideal

system to map in vivo behaviour of sub-100nm particles as they are non-toxic117, are

produced as monodispersed batches, and can be easily modified to present different

surface chemistries. Steric stabilization of particles with a methoxy-poly(ethylene

glycol) (mPEG) surface brush layer is commonly utilized to create long-circulating

particles, which reduces the adsorption of reticuloendothelial system (RES) factors in

the blood to the particle surface and the rate of clearance of particles by cells of the

monophagocytic system87. In the case of gold nanoparticles, thiol-terminated mPEG

molecules provide a stable brush layer through co-ordination of the thiol functional

group with the gold particle surface21.

Using an array of particle sizes and surface chemistries, we first test how these

parameters impact blood compartment pharmacokinetics. We then measure particle

size-dependent tumour accumulation and demonstrate a correlation to blood

pharmacokinetics. Finally, the particle size-dependent permeation of the tumour mass

is examined, revealing a striking difference in behavior over the 20-100nm size range.

Through these systematic studies, we demonstrate that particle design has tremendous

consequences on tumour targeting behavior, which will allow for an optimized approach

to controlling delivery.

57


4.2.1 Cell Culture

MDA-MB-435 (American Type Tissue Culture) cells were maintained in RPMI

medium supplemented with 10% fetal bovine serum (FBS, Sigma P/N F1051) and 1%

Penicillin/Streptomycin (Sigma P/N P4333). Cultures were incubated at 37ºC under 4%

CO2 and atmosphere.

4.2.2 Nanoparticle Synthesis and Pegylation

Various sizes of AuNPs were synthesized according to the method developed by

Frens40 and described in Chapter 1. Note that the hydroquinone method described in

the previous chapter was not optimized in time for this study. Briefly, 900 µL of 1%

chloroauric acid (Sigma-Aldrich) was added to 90mL ultra-purified water, which was

brought to a boil. Immediately upon rapid boiling, 2.25 to 0.75 mL of a 1% sodium

citrate solution was added to synthesize the various nanoparticle core sizes. For

pegylation, the pH of a AuNP solution was adjusted to approximately 9.0-10.0 by

addition of 0.1 M NaOH. A 20x molar ratio of thiolated-mPEG to particle surface area

(molecules PEG/nm2) was added to each AuNP synthesis while stirring, and incubated

12 hours. mPEG-AuNP were collected by centrifugation at 18, 000x g in an Avanti

Series centrifuge. The pellet was washed 5x with ultra-purified water, resuspended in 5

mL of water and filtered using a 0.22 µm syringe filter. Particles were stored at 4°C.

4.2.3 Particle Characterization

Synthesized particles were characterized by ultra-violet and visible range

absorption spectroscopy using a UV-1601PC spectrophotometer (Shimadzu). TEM

images were obtained by deposition of a dilute particle solution onto carbon-coated

copper grids and using a Hitachi HD2000 STEM (Hitachi Corp). Particle sizes (n=50)

58

were measured from TEM generated images in Gatan Micrograph V3.11.2 (Gatan Inc).

Particle HD was measured by DLS using a Nano-ZS Zetasizer (Malvern).

4.2.4 Examination of Particle Stability

The stability of the pegylated particles is essential to their use in vivo, and to their

detection by our sandwich enzyme-linked immunosorbent assay (ELISA) assay, and

was therefore examined by incubating one injection dose of each pegylated size in a 2M

NaCl solution, a range of pH-adjusted phosphate buffered saline (PBS, pH 4.0 – 9.0)

solutions, and in a range of serum concentrations diluted with PBS (25% - 100% serum)

at 37°C for 72 hours. The particles were then re-analyzed by UV-VIS absorption

spectra, by examining their HD, and by ELISA detection signal.

4.2.5 Quantitative Sandwich ELISA for Detection of mPEG-AuNP

ELISA detection plates were prepared on Nunc-Immuno Maxisorp 96 well plates

(VWR P/N CA62407). 100 µL of 0.1 M sodium bicarbonate (pH 8.5) containing 1 µg/mL

of anti-mPEG antibody (Epitomics P/N 2061-1) was incubated in each well overnight at

4°C. The antibody solution was removed, and the plate was blocked using PBS + 5%

fetal bovine serum (FBS, Sigma-Aldrich) for two hours at RT. Tissues for analysis were

weighed and homogenized using a PRO250 homogenizer (PROScientific). Sample

dilutions were prepared in dilution buffer (DB), consisting of 1x PBS with 2.5% FBS

such that a specific weight of tissue (e.g. 5 mg) was measured per well. An appropriate

standard curve of particles was prepared in DB. The ELISA plate was washed 1x with

wash buffer (WB); 1x PBS with 0.05% Tween20 (Sigma-Aldrich P/N P379). 100 µL of

each sample and standard was added per well, with triplicate measurements. A DB

negative control, tissue negative control (diluted in DB), as well as a tissue positive

control (diluted in DB) spiked with a known amount of pegylated particles were included

with each run. The plate was covered with parafilm and samples were incubated for 1

hour at 37°C. Samples were removed and the plate was washed 3x with WB for 2

minutes each. When assaying blood samples only, the detection antibody was added

59

next (see below). When assaying tissue samples, it was necessary to block

endogenously expressed biotin. For this, 0.1 µM avidin (Invitrogen P/N A887) in WB

was added to the wells, and incubated for 15 minutes at RT. The plate was then

washed 5x for 5 minutes each with WB. 1 mM biotin (Sigma-Aldrich P/N B4501) in WB

was added, incubating for 15 minutes at RT followed by 5x washes for 5 minutes each

with WB. The anti-mPEG-biotin antibody (Epitomics P/N 2137-1) was then added at 0.5

µg/mL in WB for 1 hour at RT, followed by 3x washes with WB for 5 minutes each.

Neutravidin-HRP (1:10 000 dilution, Pierce P/N 31001) in WB was added for 30 minutes

at RT, followed by 5x washes with WB for 5 minutes each. Finally, particles were

detected by addition of 100 µL 1X TMB substrate (eBioscience P/N 00-4201-56) and

incubated until the signal was of appropriate strength (10-30 minutes). The substrate

reaction was stopped with 100 µL of 2M sulphuric acid (EM Science P/N 00-4201-56).

Signals were read on a 96-well plate reader (Tecan Group) at 450 nm with correction at

570 nm. The DB negative control signal was subtracted from all samples and

standards, and a standard curve was generated based on optical density versus

percent of the injected dose (%ID). Particle content of samples was then determined by

comparison to the standard curve, correcting for dilutions and subtraction of each

organs blood volume particle content, as determined from measurement of particle

content immediately after injection.

4.2.6 Mapping Pharmacokinetic Dependence on Particle and mPEG Size

Five different sizes of AuNP were synthesized as described above. Each

synthesis of particles was divided into 3 equal portions for pegylation, carried out as

described above using thiolated-mPEG of molecular weight 2, 5, and 10kDa (2K and

10K from Laysan Bio, Inc, 5K from NOF Corp.). Injection doses were prepared by

diluting stock solutions in a 154 mM sodium chloride solution (1X normal saline) such

that a 150 µL bolus injection would contain 2.5 x 1015 nm2 of surface area based on

hydrodynamic diameter of the particles. Single bolus injections of the 14 particle

designs were injected intravenously (IV) into CD1 mice (n=4) by tail vein. Following

this, small volumes of blood (approximately 30 µL) were obtained from the saphenous

60

vein at specified timepoints into Microvette capillaries with EDTA (Sarstedt P/N 16-443-

100). A final 500 µL blood sample was obtained by cardiac puncture after euthanization

by CO2 asphyxiation followed by cervical dislocation. Blood half-life for each particle

design using 4 mice, and 4-5 blood samples were obtained from each. Particle content

of blood samples was then measured by ELISA, allowing for calculation of

pharmacokinetic parameters.

4.2.7 Tumour Xenografts, In Vivo Biodistribution and Tumour Accumulation

All animal use protocols were approved by the University of Toronto Animal Care

Committee, and met the requirements of the Ontario and Canadian Animal Use

legislation. MDA-MB-435 growth medium was replaced 24 hours prior to tumour

initiation to allow for conditioning by the cells. Cells were collected using a cell scraper,

concentrated by centrifugation, counted using a haemocytometer and resuspended in

an appropriate volume of the conditioned growth medium. 2 x 106 MDA-MB-435 cells in

200 µL of growth media were injected sub-cutaneously on the back of 6-8 week old CD1

athymic nude mice (Charles River Laboratories). Tumour growth was monitored until

volume reached 1 cm3, determined by measurement with digital callipers and calculated

using (length*width* height*(π/6)). Mice bearing tumours of appropriate size were

randomized between different particle design groups, and a 150 µL dose was injected

IV into the tail vein. Each particle size and time point included 3 independent

repetitions. At specified timepoints, the mice were euthanized by CO2 asphyxiation,

followed by cervical dislocation. 500 µL of blood was immediately obtained by cardiac

puncture and mixed with 50 µL of 50 mM EDTA. Organs were harvested, rinsed with

PBS, and blotted dry on clean filter paper. Tissues were weighed, then stored at -20°C

until processed for ELISA analysis. Particle content of the tissues was calculated by

subtracting the blood volume particle content from the total measured by ELISA. Blood

volume was determined for each organ by measuring particle content at 5 minutes post-

injection.

61

4.2.8 Histology and Silver Enhancement

Histological samples obtained at time of tissue harvest were immediately fixed in

10% buffered formalin (Sigma-Aldrich P/N HT501128). Sample processing, mounting

and haematoxylin staining were carried out by Toronto Medical Laboratories using

standard procedures. Silver enhancement of histology sections was carried out on de-

waxed mounted sections prior to haematoxylin staining, using the Silver Enhancement

Kit for Light and Electron Microscopy (Ted Pella P/N 15718). Tissue sections were

rinsed well with double-distilled water. An equal volume of the kit’s two reagents were

mixed in a 1.5 mL microcentrifuge tube, and 50 µL was added to completely cover the

tissue section. Enhancement was allowed to proceed for 20-30 minutes before

stopping by rinsing with double-distilled water.

4.2.9 Quantification of Particle Permeation

Images for analysis were obtained on a Axiovert 135 microscope (Zeiss) with a

CoolSNAP-Pro camera (Media Cybernetics). Quantification of particle distribution

around blood vessels in silver enhanced tissue sections was carried out using ImageJ

version 1.39 (NIH), and the “Radial Profile Plot” plug-in available at

http://rsbweb.nih.gov/ij/plugins/index.html. Briefly, images were converted to 8-bit

grayscale, and the scale was set appropriately. Images were then adjusted using the

“threshold” function, leaving only the particles in the image. The Radial Profile Plot

“produces a plot of normalized integrated intensities around concentric circles as a

function of distance from a point in the image”. The center point of the blood vessel was

chosen by using the line tool to draw across the longest inner axis of a blood vessel.

We ran the plug-in using the spatial calibration that had been applied earlier, out to a

distance of 50 µm from the vessel center. The data was then “binned”, by averaging the

values over 10 µm intervals, and normalized to the signal at 0-10 µm. This analysis

was applied to 10 images per particle size.

62

4.2.10 Statistics

Statistical analysis was carried out using SPSS version 15.0 (SPSS, Inc).

Average half-life differences of particle designs (ie mPEG 5 versus mPEG 10 kDa) were

compared using a Student’s t-test. Half-lives of the two particle sets used for

pharmacokinetic mapping and tumour targeting were tested by Pearson’s correlation.

Total tumour accumulation was calculated as area-under-the-curve (AUC) using the

trapezoidal rule. Linear regression analysis was used to test the dependence of total

tumour accumulation on blood compartment half-life and particle size.

63

4.3 Results

4.3.1 Particle Synthesis, Pegylation and Characterization

We synthesized two sets of mPEG-AuNP to use throughout this study (Table 2).

The first set, used to map blood pharmacokinetics, is an array of 5 particle sizes and 3

different thiol-terminated mPEG molecular weight brush layers. The largest particle

diameter and lowest molecular weight mPEG combination aggregated during pegylation

and therefore only 14 particle designs in total were examined. The results from this first

study were used to design a second set of particles, used to test size-dependent tumour

accumulation and permeation. This second set of particle designs included 5

combinations of particle diameter and mPEG molecular weight having final HDs of

approximately 20, 40, 60, 80, and 100 nm. All of the synthesized particles were

characterized by transmission electron microscopy and DLS (Figure 17 and 18). ζ-

potential was measured for all particle designs before and after addition of the mPEG

brush layer.

4.3.2 Particle Stability and ELISA Validation

To allow for a highly sensitive detection of the mPEG-AuNP in tissues, we optimized a

quantitative sandwich ELISA for measuring the concentration of nanoparticles in tissue

homogenates. Our assay is highly sensitive, having a lower limit of detection of 0.01%

of the injected dose in 5 mg of tissue. It is also highly specific, detecting only pegylated

particles and not free mPEG in solution. A typical standard curve of signal from mPEG-

AuNP detection is shown in Figure 19. Desorption of the thiol-mPEG brush layer can

occur over time, which would greatly affect particle stability in vivo as well as our ability

to detect them via an ELISA. We therefore incubated mPEG-AuNP in PBS pH 4.0 to

pH 9.0 for 48 hours, verifying the stability of their mPEG brush layer and the ELISA

detection signal for a length of time comparable to our in vivo incubations (Figure 20A).

We also verified that non-specific protein adsorption would not interfere with particle

detection by comparing detection of particles incubated in 100% fetal bovine serum for

48 hours to those incubated in PBS (Figure 20B).

64

4.3.3 Design-Dependent Blood Pharmacokinetics of mPEG-AuNP

The array of 14 particle designs (particle set A, Table 2 and Figure 13) was

applied to CD1 mice (n=4) to determine how blood half-life changes over a range of

particle size and surface mPEG molecular weight. We obtained small volumes of blood

at set times post-injection for measurement of particle content by ELISA. We then

calculated pharmacokinetic parameters and mapped each design’s blood half-life

against design parameters of core size and mPEG molecular weight (Figure 18).

Particle half-life in circulation improved with smaller particle diameters and larger mPEG.

Half-life was 8 times greater when the same mPEG (10 kDa) was applied to 17nm

particles (t1/2=51.1 hours), compared to 86nm particles (t1/2=6.6 hours). Half-life also

increased for each particle size as mPEG molecular weight increased. The 14-times

average improvement between particles with mPEG 5 versus 2 kDa was much greater

than the 1.5-times improvement between mPEG 10 versus 5 kDa (Students t-test,

p<0.05).

65

Table 2. Particle sets A and B, their design parameters, and resulting blood half-life and tumour accumulation.

Core Diameter (nm) (±SD)

mPEG Weight

(Da) Hydrodynamic Diameter (nm)

Blood Half-Life

(h) Tumour AUC (particle

mass(ug)*hour/g) Tumour AUC (%ID*hour/g)

Part

icle

Set

A

17.72 (1.47) 2000 25.2 4.0

5000 34.8 29.7

10 000 63.5 51.1

31.28 (4.83) 2000 42.8 2.4

5000 62.6 19.3

10 000 77.5 22.7

45.03 (4.83) 2000 53.8 0.4

5000 60.4 14.1

10 000 83.9 16.1

66.54 (5.69) 2000 75.7 1.0

5000 85.6 9.2

10 000 105.8 11.3

86.73 (7.74) 5000 105.4 3.3

10 000 118.9 6.6 Pa

rtic

le S

et B

16.6 (1.80)) 2000 22.4 2.5 0.45 0.3

22.6 (2.68) 2000 39.6 4.0 18.89 15.8

32.5 (5.16) 5000 61.3 16.5 39.05 26.5

43.3 (5.08) 10 000 82.6 11.6 39.33 20.4

83.5 (8.29) 10 000 99.4 7.2 170.42 17.9

66

Figure 17. Representative TEM images, absorption spectra and DLS data of five gold nanoparticle sizes synthesized for testing blood pharmacokinetics. A-E)

Representative TEM images of particles from the 5 batches. F-J) Optical absorption

spectra of the 5 batches, showing a red-shifting plasmon maximum with larger particle

sizes. K-O) Raw HD data of the 5 batches.

67

Figure 18. Representative TEM images, absorption spectra and DLS data of five gold nanoparticle sizes synthesized to examine biodistribution, tumour accumulation and tumour permeation. A-E) Representative TEM images of particles

from the 5 batches in Set B. F-J) Optical absorption spectra of the 5 batches, showing

a red-shifting plasmon maximum with larger particle sizes. K-O) Raw HD data of the 5

batches.

68

Figure 19. ELISA Standard curve of optical absorbance versus percent of injected dose for the 61.3 nm HD particles from particle set B. The optical absorbance

increased as expected with an increase in particle concentration, as expected by on the

assay.

69

Figure 20. Testing particle stability and protein interference for detection of mPEG-AuNP by ELISA. A) Aliquots of the 61.3 nm mPEG-AuNP from particle set B

were incubated in PBS with pH adjusted between 4-9 for 24 hours. They were then

measured using the ELISA and optical absorbance is reported. B) Aliquots of the 61.3

nm mPEG-AuNP from particle set B were incubated in blends of FBS and PBS pH 7.0

for 24 hours. The solutions were then measured using the ELISA. The negative control

in both experiments was a PBS solution, pH 7.0.

70

Figure 21. Map of mPEG-AuNP blood half-life as a function of particle size and mPEG-SH molecular weight. Particles were injected IV into CD1 mice, blood

collected at various timepoints, and the half-life calculated. Blood half-life increased

with smaller nanoparticle diameter, and larger mPEG-SH molecular weight.

71

4.3.4 Design-Dependent Biodistribution and Tumour Accumulation of mPEG-AuNP in Tumour-Bearing Mice

We next aimed to examine the design-dependent tumour targeting capacity of

sub-100nm particles. Based on our results from pharmacokinetic mapping, we

designed, synthesized and characterized a second set of particles having hydrodynamic

diameters of approximately 20, 40, 60, 80, and 100 nm (Particle set B, Table 2 and

Figure 18). Athymic nude CD1 mice bearing 1 cm3 subcutaneous MDA-MB-435

xenograft tumours were randomly sorted between particle design groups (n=3) and

were then injected IV with a single bolus of mPEG-AuNP. At various timepoints up to

48 hours, 3 animals for each particle design were euthanized, and tissues were

collected for analysis. This second set of mPEG-AuNP had half-lives consistent with

similar designs used in our pharmacokinetic study (p<0.01, Figure 22A). The mean

blood half-lives of the 60 nm particles was 16.5 hours, 6.5-times longer than that of the

20 nm particles, due to its larger mPEG brush layer (5 kDa versus 2 kDa). The 80 nm

(11.5 hours) and 100 nm (7.2 hours) mPEG 10 kDa designs had half-lives of 70% and

44% of the 60 nm, respectively (Figure 22B).

Particle content in the liver (Figure 23A) and spleen (Figure 23B) increased over

time for all particle sizes, with the spleen filtering a greater proportion than liver by

weight (per gram of tissue). We observed low levels of particle uptake in non-target

tissue of heart and lung without any consistent trend between particle sizes and over

time. The 20 nm particles were cleared rapidly from the blood (Figure 22B) but without

a corresponding accumulation in the liver (Figure 23A) and spleen (Figure 23B). Particle accumulation in the tumour was calculated as the area-under-the-curve (AUC),

both by the measured percent injected dose per gram (%ID/g) over time and particle

mass/g over time (see Table 2 and Figure 23C). The various design sizes showed

different degrees of accumulation, with the 60nm particles providing the greatest AUC

by %ID at 26.47 %ID*h/g, 87.3-times greater than the 0.3 %ID*h/g measured for the 20

nm particles. The largest sizes of 80 and 100 nm also had significant accumulation at

20.4 and 17.0 %ID*h/g, respectively. Examining accumulation by particle mass, the 100

nm particles achieved the greatest accumulation at 170 µg*hour/g. This is 4.3-times

72

greater than that achieved by both the 60 and 80 nm particles, 9-times greater than the

40 nm, and 38-times greater than the 20 nm.

Regression analysis revealed that total tumour accumulation could be predicted

by blood half life for particles between 40 - 100 (Figure 23D, p<0.02), and by both blood

half-life and size for the entire 20 – 100 nm range of particle diameters (p<0.015).

y = 794.907 + 0.768a – (7.053*106) / (b-3)

where a is the half-life in minutes and b is the hydrodynamic diameter in nm. This

suggests that accumulation of 40-100nm particles is exclusively dependent on blood

half-life, whereas the accumulation of particles in the 20nm range depends on size and

half-life.

73

Figure 22. Comparison of mPEG-AuNP half-lives in CD1 versus CD1 athymic nude mice, and the pharmacokinetic profiles of particle set B in CD1 athymic nude tumour bearing mice. A) Blood half-lives of particle set B after injection into CD1

athymic nude tumour bearing mice was compared to similar designs of particle set A

obtained in CD1 mice, showing a statistically similar behavior (p<0.01). B)

Pharmacokinetic profiles of 20 (♦), 40 (□), 60 (▲), 80 (○), 100 (●) nm particles were

determined from measurements of blood particle content at various timepoints after

injection into tumour-bearing mice.

.

74

Figure 23. Reticuloendothelial system organ uptake, tumour accumulation, and correlation of particle design to tumour accumulation. A-C) Shows organ

accumulation of 20, 40, 60, 80 and 100 nm (left-to-right) mPEG-AuNP in A) liver, B)

spleen and C) Tumour at 1, 4, 8 and 24 hours post-injection (%ID/g). A detection signal

equal to or below the negative control, indicating no particle content, is shown with

double strikes through the x-axis. Uptake of particles was generally higher in spleen

than liver when normalized per gram of tissue. There was a trend of increasing

accumulation in these organs over time. Total tumour accumulation was calculated as

area-under-the-curve using a trapezoidal method. D) Linear regression analysis

revealed that total tumour accumulation (%ID*h/g) of the five mPEG-GNP designs is a

function of their blood half-life (hours) and of size (nm3).

75

4.3.5 Particle Size-Dependent Permeation of the Tumour Interstitium

We measured the size-dependent permeation of the interstitial space using a

novel approach to visualize and quantify particle localization within a histology sample.

Silver enhancement of tumour histological preparations causes particle growth by

deposition of reactive silver onto the surface of particles118. They can then be visualized

under brightfield microscopy, and quantified by contrast-enhanced densitometry

analysis (Figure 24). 1 hour post-injection, all particle sizes were localized primarily in

perivascular spaces, with little penetration into the tumour mass (Figure 24A-C). By 8

hours post-injection distinct tumour permeation trends appear between the different

particles sizes (Figure 24D-F). The 20 nm particles permeated far from vessel centers,

the 60 nm particles less so, and the 100 nm particles remained localized in perivascular

spaces. This trend was further exaggerated at 24 hours post-injection (Figure 24G-I). Quantitative analysis of all 5 particle sizes at 8 hours post-injection (HPI) revealed a

significant difference (n=10, p<0.05) in the migration of 20 nm particles versus all larger

diameters (see Figure 24J-K). There was a general trend of decreased permeation as

particle size increased. However, the difference between the 4 largest particle sizes

was not statistically significant at any distance away from the blood vessel centre.

76

77

Figure 24. Particle size-dependent permeation of the tumour interstitial space. A-

I) Representative images from histology samples obtained for 20, 60 and 100 nm

particle sizes at 1 hr, 8 hrs, and 24 hrs post-injection. Silver enhancement causes

growth of the AuNP, allowing their distribution relative to blood vessels (marked by

arrows) to be visualized under bright-field microscopy (scale bar in A corresponds to 40

µm in all images). A-C) 1 hour post-injection (HPI), all particle sizes are localized in the

perivascular space. D-F) 8 HPI, a size-dependent trend emerges with 20 nm particles

migrating far into the interstitial space, 60 nm particles showing limited migration and

100 nm particles localized in perivascular spaces. G-I) 24 hours post-injection, few 20

nm particles remain in the region surrounding the blood vessel, the 60 nm particles are

seen migrating moderately far away from the blood vessel and 100 nm particles have

limited migration. J) ImageJ software was used to generate contrast-enhanced images

for densitometry analysis. K) Densitometry signal was quantified at 10 µm distances

away from blood vessel centers 8 HPI, and was normalized to the signal at 0-10 µm. 20

nm particles had a significantly higher signal at distances further away from the vessel

center, demonstrating a higher extent of migration through the interstitial space (n=10,

p<0.05).

78

4.4 Discussion

4.4.1 Particle Design and Synthesis

Aspects of particle design and synthesis were covered in Chapter 2 of the thesis.

The method described in Chapter 3, using hydroquinone to synthesize high quality gold

nanoparticles over the size range of interest in this study, was not developed when the

current study was carried out. Therefore, the citrate method was used for AuNP

synthesis.

In deciding on how to normalize the quantity of AuNP injected within a single

bolus, we reasoned that interactions between the particles and biological systems

(blood components or cells) would occur at the particle surface, and therefore that a

constant surface area across particle designs would provide the best basis for

comparison. As a basis for deciding the quantity of particles, prior research has shown

that clearance by the RES cells in the liver and spleen of rats becomes saturated when

more than 1015 particles are injected in a single bolus119. All the injection doses used

here contain at least 1000-times fewer particles and the organ weight of mice is

approximately 1/10th that of rats, such that clearance should not be compromised by

saturation.

4.4.2 Blood Pharmacokinetics and Clearance of Sub-100 nm mPEG-AuNP

The blood pharmacokinetic trends that we observed and their dependence on

nanoparticle size and surface chemistry are similar to what has been reported for larger

particle systems36, 83. Fang et al. described pharmacokinetic parameters for a system of

particles ranging from 80 nm to 240 nm, where pegylated 80 nm particles had a similar

half-life (T1/2=11.33 hours) to a comparable AuNP design tested here (T1/2=9.24 hours).

From these results, it follows that the particle size- and surface chemistry-dependent

pharmacokinetic trends that previous studies have described for larger particles can be

79

extended to the sub-100nm region. This also suggests that blood half-life can be

maximized by the use of sub-100nm particles. This has implications for particle design.

Half-lives can be improved by larger mPEG molecular weight brush layers, but this also

increases hydrodynamic diameter which could impact their ability to extravasate into

and through tumours. From a design perspective, small hydrodynamic diameter and

long half-lives appear to converge when particles below 50 nm are protected with an

mPEG layer of moderate molecular weight (MW 5000 Da in our case).

Accumulation of particles and macromolecules in the tumour compartment

competes against uptake and clearance by monophagocytic cells. We found that

spleen had a greater clearance capacity than liver when normalized per gram of tissue,

suggesting a greater filtering efficiency. Surprisingly, the 20 nm particles were cleared

from circulation very rapidly but were not measured at high levels in either spleen or

liver tissue. They are also unlikely to be cleared via the kidney which has a

hydrodynamic diameter cut-off of approximately 6 nm120. These particles could have

accumulated in peripheral monophagocytic cells, in the lymph following extravasation

into non-specific tissues, or could have been cleared into the intestines via the liver’s

bile duct. A thorough examination of how particle design impacts non-specific

biodistribution may be of interest and merits further investigation in follow-up studies.

4.4.3 Tumour Accumulation

Tumour targeting vehicles must be small enough to access the tumour through

transvascular pores and fenestrations, making particle diameter central to design. Pore

cut-off sizes have been measured for a limited number of tumour models, with the

smallest reported at 100-200 nm89, 121. Our study focused on tumour accumulation of

20-100 nm particles, and we scaled the ratio of mPEG and particle size to limit

differences in tumour accumulation to blood pharmacoknetics and hydrodynamic

diameter.

Regression analysis revealed that blood half-life was a strong predictor of total

tumour accumulation. The exception to this was the lower accumulation of 20 nm

80

particles than would be predicted based on their blood half-life. Evidently, the entire 20

– 100 nm size range can access the tumour microenvironment in the MDA-MB-435

xenograft model. This is expected, given that pore cut-off sizes previously reported are

no smaller than 100 nm to 200 nm. As well, the MDA-MB-435 human cancer cell line

used here has been characterized as having very high vascular permeability relative to

other breast cancer cell lines, although its specific pore cut-off size has not been

described.

Tumour accumulation is a function of both the rate of extravasation from the

blood to the tumour space and also the rate of clearance from the tumour. Hobbs et al.

showed that the rate of extravasation of bovine serum albumin (BSA) was independent

of pore size over a variety of tumour models89. This demonstrates that for a 7 nm

molecule, much smaller than the transvascular pore size, extravasation is not

dependent on pore size but is instead a diffusive process that will depend on the

concentration gradient between blood and tumour. In our model, the absence of size-

dependent effects might be reflected by a dependence of total accumulation only on the

particle concentration gradient between blood and tumour, and therefore on blood half-

life over time67.

The poor accumulation of 20 nm particles is intriguing. Hydrostatic pressure in a

tumour mass typically decreases from the center to the periphery, the region where

porous neovasculature is most dense67. Because of this, particles which enter the

tumour through leaky vasculature may be carried by convection past the tumour

periphery and into the surrounding tissue, where they will likely enter lymph drainage of

host tissue. This loss from the tumour will be decreased if the particles have restricted

movement through the tumour extra-cellular matrix (ECM) 67, 122, which we observed to

occur in a size-dependent manner. The relatively high permeation rate of 20 nm

particles may therefore have caused their clearance into surrounding tissues, which

could account for their lower degree of accumulation. In contrast, the larger particle

diameters may have benefited from a slower migration rate through the interstitial

space, achieving greater accumulation.

81

Further experimentation is required to fully understand the nature of these size-

dependent effects. Nonetheless, these results have a clear implication for particle

design, namely that tumour accumulation of particles in the range of 40-100nm is highly

dependent on blood half-life and may allow accumulation to be optimized through

parameters that determine blood pharmacokinetics.

4.4.4 Particle Size Dependence of Tumour Permeation

To our knowledge, no study has yet examined the size-dependent permeation of

nanoparticles through a tumour interstitial space. Dreher et al. made use of

fluorescently labeled dextrans to demonstrate that vascular permeability and tumour

permeation decrease as macromoleculer weight increases (3.3 kDa to 2 mDa)123.

Although the size and behavior of colloidal metal particles may be quite different from

that of macromolecules, there are some interesting parallels with our results. They

found that tumour accumulation increased with larger molecular weight (40-70 kDa or

11.2-14.6nm), and attributed the poor accumulation of low molecular weight dextran

(3.3-10kDa) to a rapid rate of interstitial space permeation and clearance. It is difficult

to draw conclusions across these studies, owing to the differences in macromolecule

versus nanoparticle, and variability between tumour models. However, both studies

have provided strong evidence that small particles and macromolecules are unsuitable

for passive accumulation in the tumour compartment owing to a high rate of interstitial

space permeation.

4.4.5 General Discussion

We have systematically tested and described the in vivo behavior and tumour

targeting capacity of sub-100 nm particles. This size range appears to be ideally suited

to the design of tumour targeting vehicles as it allows for a broad range of blood

pharmacokinetics, and tumour accumulation and permeation. The ability to tune and

control the behavior of vehicles toward specific behaviors should provide a basis for

major improvements in tumour targeting efficiency.

82

Specific tumour targeting behaviors for different applications can be achieved

through particle design. For instance, we saw a striking size-dependent difference in

tumour permeation that could be very useful towards different outcomes. If, for

instance, the goal is to improve diagnostic sensitivity by maximizing the amount of a

contrast agent delivered into the tumour compartment but localization within the tumour

mass is unimportant, moderate particle cores protected with a large mPEG (5 or 10

kDa) brush layer having a final hydrodynamic diameter of 60-100 nm would provide

excellent blood pharmacokinetics and tumour accumulation, and could be used to

localize leaky vasculature.

The design of vehicles for delivery of therapeutics is more complex. Large

particles may behave similarly to our 100 nm particles, with accumulation restricted to

the perivascular space and little permeation into the tumour. This could be very useful

for anti-angiogenic therapy, but therapeutics aimed at the majority of the mass would be

less effective. As a consequence, the greater volume provided by larger vehicles will be

increasingly restricted from impacting cells far from leaky vasculature. This compromise

makes a rational design of particle vehicles more complicated, especially in light of the

variable nature of tumour growth. Nevertheless, we have observed a broad range of

possible behavior, from extremely restricted permeation (100 nm) to rapid permeation

leading to poor accumulation (20 nm). 20 – 100nm particles therefore offer a broadly

tunable range for controlling vehicle behavior.

The trends described here will need to be tested in a wide range of tumour

models and states of tumour growth before they can be applied universally. We did not

observe a size-dependent restriction on sub-100 nm particle extravasation from the

blood to the tumour, consistent with earlier reports. Sub-100 nm particles may be able

to completely avoid such a restriction in tumours with highly permeable vasculature.

Instead, the ability of a vehicle to either permeate the tumour interstitial space or remain

at the perivascular space may be the most important factor that can be optimized

through design. Future studies could aim to describe the variability in tumour extra-

cellular matrix, its impact on particle permeation and ultimately on effectiveness of a

therapeutic or diagnostic agent. The fundamental knowledge provided here should

83

provide a basis for such studies and for the rationale design of tumour targeting

nanomaterials.

84

4.5 Chapter Author Contributions The study was conceptualized by Warren Chan and designed by Steve Perrault

and Warren Chan. Experiments were designed by Steve Perrault and performed by

Steve Perrault, Carl Walkey, Travis Jennings, and Hans Fischer. Data was analyzed by

Steve Perrault and Carl Walkey. The Manuscript was written by Steve Perrault, Carl

Walkey and Warren Chan. Warren Chan also contributed enormously to the editing of

the manuscript. Carl Walkey also produced the Table of Contents figure included with

the manuscript.

85

Chapter 5

In Vivo Assembly of Nanoparticle Device Components to Improve Targeted Cancer Imaging

5.1 Introduction Determining the correct prognosis and therapeutic options for cancer requires

accurate staging and surveillance of tumours. Current detection strategies typically

combine sensitive imaging modalities with contrast agents6, 124. Yet these approaches

fail to detect lesions in many cases, typically because poor imaging contrast is

achieved124. This occurs because the low molecular weight of many contrast agents

rapid clearance from circulation and poor retention within tumour interstitium. This can

be overcome through tumour targeting strategies that link contrast agents to a polymer

or nanoparticle. As discovered in Chapter 4, nanoparticles are well suited to act as

tumour targeting vehicles because their circulatory kinetics are determined by design,

and they are able to leak into and accumulate in tumours via the enhanced permeability

and retention effect4, 36, 121, 125, 126. Despite these advantages, several obstacles limit

effective tumour detection with nanoparticle-based targeting strategies. Passive

targeting requires large diameter particles, but this simultaneously restricts transport

into tumours and accumulation occurs only after many hours in circulation67, 89, 127.

Actively targeting nanoparticle designs can achieve faster accumulation62, 105, 128, but

may not be appropriate for detecting lesions whose antigens are uncharacterized or are

heterogeneous and therefore unreliable. It is thus vital to develop novel targeting

strategies that can rapidly accumulate contrast agents into tumours without depending

on tumour antigen characterization.

The movement of nanoparticles through tumour extracellular matrix is primarily

dependent on diffusion67. We discovered in Chapter 4 that diffusive transport is limited

with larger particle diameters and becomes negligible by 100 nm. Particles having 80

86

nm diameters were found to penetrate slowly into the interstitium and 24 hours after

injection were localized within several cell lengths of leaky vasculature. Based on this

finding we hypothesized that size-dependent localization of particles near leaky blood

vessels would make them highly accessible to molecular contrast agents in circulation.

Synthetic antigens present on the particle surface would then provide a universal

anchor for active targeting of a second component. This in vivo assembly strategy

might be highly effective because transport of a lower molecular weight contrast agent

into the tumour would be rapid (i.e. high Kin), at which point its retention would be

determined by the large nanoparticle diameter (i.e. low Kout).

In this chapter we demonstrate a novel targeting device consisting of a molecular

contrast agent and engineered nanoparticle that undergo in vivo assembly within

tumour interstitium. Using an optical detection platform and fluorescent contrast agent,

we compare the rate and extent of accumulation from this strategy to that achieved by

passive accumulation of a macromolecule and of a nanoparticle. We also demonstrate

that the assembly process can be competitively inhibited, attenuating its effect on

tumour accumulation. Intra-tumour assembly was found to provide the fastest

accumulation kinetics and highest signal-to-noise. Our results describe a new strategy

for nanoparticle-based tumour targeting, and provide a conceptual demonstration of

how in vivo assembly can be applied to overcome in vivo barriers to targeting.

87


5.2.1 Synthesis of Gold Nanoparticles

The AuNP were synthesized according to the method developed by Frens40.

Briefly, 300 μL of chloroauric acid (Sigma-Aldrich) was added to 300 mL ultra-purified

H2O. This was brought to a boil, and 3 mL of 1% sodium citrate tribasic dihydrate

(Sigma-Aldrich) was added while stirring rapidly. The solution was removed from heat

after 5 minutes, cooled to room temperature, and filtered through a 0.22 μm 500 mL

filter (Nalgene).

5.2.2 Characterization and Pegylation of AuNP

Synthesized particles were characterized by ultra-violet and visible range

absorption spectroscopy using a UV-1601PC spectrophotometer (Shimadzu). TEM

images were obtained by adding 10 μl of a 1:5 (stock:H2O) dilution of the particle

solution onto carbon-coated copper grids, then imaged on a Hitachi HD2000 STEM

(Hitachi Corp). ImageJ (version 1.42q, National Institutes of Health, USA) was used to

measure the particle sizes, by first setting the scale according to the TEM image scale

bar, then Adjust-Threshold to create a binary image, and finally Analyze-Analyze

Particles to determine Feret’s diameter. Several images were measured and the first

200 results were used to determine particle diameter.

The AuNP concentration was determined by ICP-MS (see below), allowing us to

calculate the available surface area per mL of AuNP stock. Then, we examined the

PEG-blend dependence on binding of streptavidin-Alexa fluor 750 (Invitrogen, P/N

S21384) to the particles. Based on work described in Chapter 2, addition of

approximately 0.3x molecules of mPEG-SH (M.W.=5000 g/mol, NOF Corp.) per nm2 of

surface area is enough to stabilize AuNP of 15-50 nm. Keeping this constant, we added

a serial dilution of biotin-PEG-SH (M.W.=10,000 g/mol, Nanocs, custom reagent) at a

ratio of 10, 5, 2.5, 1.25, 0.6, 0.3, 0.15, and 0x per nm2, to 1 mL aliquots of the AuNP

stock. After immediate mixing by vortexing, the particles were incubated at room

88

temperature (RT) for 30 minutes on an inversion mixer. 10x (10 molecules/nm2) mPEG-

SH 5000 was then added to fully saturate the particle surface, incubating a further 30

minutes at RT and mixing by inversion. The 8 treatments were then centrifuged at 15,

000x g for 5 minutes, the supernatant was removed, and 1 mL of H2O was added to

wash. This was repeated 5x total, and finally the particles were resuspended in 100 μL

of PBS. 20 μL of the pegylated particle concentrates was then mixed with an excess of

streptavidin-Alexa fluor 750. The products were separated on a 0.7% agarose gel in

0.5x Tris-Borate-EDTA Buffer (TBE) at 100 volts for 20 minutes. Particle-bound

streptavidin was determined by measuring mean fluorescence in the background –

corrected gel image using ImageJ. A ratio of 0.6x biotin-PEG-SH was chosen based on

a compromise between high streptavidin-Alexa fluor binding while leaving most of the

particle surface covered with mPEG to stealth the particles from immune recognition.

This should provide approximately 2000 biotins per particle, or 1 biotin per nm2 of

surface area. Based on the tetramer structure of streptavidin and its HD (7 nm), we

assume that one biotin-AuNP can bind 500 streptavidin molecules.

The remaining particle stock was pegylated by scaling up the mPEG-SH and

biotin-PEG-SH according to the above method. Control mPEG particles were

synthesized identically, replacing the biotin-PEG-SH (M.W.=10,000 g/mol) with mPEG-

SH (M.W.=10,000 g/mol, Laysan Bio.). Hydrodynamic diameter of the particles was

measured by DLS using a Nano-SZ Zetasizer (Malvern). The final pegylated stock was

centrifuged at 12, 000x g for 30 minutes in an Avanti Series centrifuge (Bckman-

Coulter), the supernatant removed and the pellet resuspended in 1 mL H2O in a

microcentrifuge tube. This was centrifuged at 12, 000x g for 10 minutes, the

supernatant removed, and 1 mL H2O added to wash. After 5 washes, the pellet was

resuspended in 1 mL H2O and stored at 4°C until use.

5.2.3 Cell Culture

MDA-MB-435 cells (American Type Tissue Culture) were maintained in RPMI

medium (Sigma-Aldrich) supplemented with 10% fetal bovine serum (Sigma-Aldrich),

89

and 1% penicillin/streptomycin (Sigma-Aldrich). Cultures were maintained at 37°C and

4% CO2 atmosphere.

5.2.4 Tumour Xenograft

The animal use protocol (protocol #2000-7675) was approved by the University

of Toronto Animal Care Committee and met the requirements of animal use legislation

in Ontario. MDA-MB-435 growth medium was replaced 24 hours prior to seeding

tumours. Cells were removed from culture flasks using a cell scraper, concentrated by

centrifugation and quantified using a haemocytometer. Cells were resuspended in

conditioned media and Matrigel (BD Biosciences, P/N 354234) at a 1:1 ratio. 5x106

cells were injected in a 200 μL volume sub-cutaneously onto the back of 6-8 week old

CD1 athymic nude mice (Charles River Laboratories). Tumour growth was monitored

until the mass reached 0.75 cm in length in any direction. Tumour-bearing mice were

randomized before being used in either the biodistribution study or for imaging. AuNP

and streptavidin-Alexa fluor was administered to the mice in saline by intravenous

injection of 150 μL into the tail vein. Mice were euthanized by CO2 asphyxiation,

followed by cervical dislocation, or cervical dislocation while anaesthetized. Tissues

harvested for analysis were rinsed with PBS and blotted dry on clean filter paper.

5.2.5 Biodistribution Study and ICP-MS

Mice bearing MDA-MB-435 tumours were injected with a single bolus of 1x1012

particles in 150 μL normal saline. At 6 and 24 hours post-injection (HPI), mice were

euthanized and tissue was collected for analysis. 250 μL of blood was obtained by

cardiac puncture with a 25G needle tip and 1 mL syringe. Organs were rinsed with

PBS, blotted dry on clean filter paper, and weighed before storing at -20ºC until

digestion. Organ samples were then digested in 10% HCl (Fluka P/N 96208), and Au

content was quantified with ICP-MS to determine the total Au in the original organ.

90

5.2.6 Förster Resonance Energy Transfer

To produce Förster Resonance Energy Transfer (FRET) on biotin-AuNP, we

used a variant of the pegylated AuNP synthesis method described above. We

synthesized 24 nm AuNP using the citrate method, stabilized these with 500 biotin-PEG

(m.w.=10,000 g/mol) per nanoparticle, then saturated the remaining surface area with

mPEG-SH (m.w.=10,000 g/mol). These particles were found to produce a higher

energy transfer between the streptavidin-Alexa fluor 568 and 750 conjugates than the

design used for remaining imaging studies. After synthesis, the biotin-AuNP were

washed 5x with PBS+0.05% Tween20 before in vitro and in vivo experiments.

For in vitro characterization, we determined the concentration of the biotin-AuNP,

then prepared solutions of the particles and each of the fluorophore conjugates. These

were then mixed with biotin-AuNP either individually, to produce AuNP-A568 and AuNP-

A750, or simultaneously, to produce AuNP-A568/750. The products were then loaded

into wells of a 0.7% agarose gel and separated by electrophoresis for 45 minutes at 100

volts. The gel was then imaged on a Kodak In Vivo Multispectral Imaging System

(Carestream Health) with the following filter combinations: A. excitation at 570 and

emission 600, B. excitation 750 and emission 790, C. excitation 570 emission 790.

Fluorescence data was acquired for 5 minutes on a 4 megapixel camera, using 4x4

binning. 16-bit images were then exported for processing in ImageJ. We performed a

background subtraction, then applied a colour table to each image before producing a

montage of the 3 channels.

For in vivo measurements of FRET in harvested tissue, mice were injected with

1x1012 biotin-AuNP 24 hours before injection of fluorophore-streptavidin conjugates. A

solution of the two streptavidin-Alexa fluor conjugates was prepared in saline, with a 1:1

molar ratio and each at 2 μM. 150 μL was then injected IV by tail vein. After a 2-hour

incubation, the animals were sacrificed by cervical dislocation while anaesthetized, a

blood sample was collected by cardiac puncture, and organs were harvested. These

were imaged (50 µl of blood and whole organs) on the Kodak system using the same

settings described above for the in vitro analysis. The 16-bit images were exported and

analyzed in ImageJ using identical settings and methods as for the gel images.

91

5.2.7 Optical Imaging Data Collection and Analysis

For the biotin-AuNP and mPEG-AuNP treatment groups, mice were injected with

1x1012 AuNP 24 hours prior to injection with streptavidin-Alexa fluor 750. Mice in the

pre-assembled AuNP-A750 group were injected with assembled construct immediately

prior to imaging. Competitive inhibition controls were injected with 150 μL of a 5 µM

bolus of streptavidin 30 minutes before injection of streptavidin-Alexa fluor 750. In all

cases, imaging commenced as soon as possible (within 5 minutes). Mice were

anaesthetized with 5% isofluorane, transferred to a Kodak In Vivo Multispectral Imaging

System (Carestream Health), and maintained with 3.5% isofluorane. 10 image stacks

were acquired sequentially using Kodak Molecular Imaging Software version 5.0.0.86,

by excitation at 710, 720, 730, 750, 770 nm and emission at 830 nm. Data was

acquired for 2 minutes at each filter combination, approximately 11 minutes per image

stack, on a 4 megapixel camera with 4x4 binning and applying a field normaliser. After

completion of imaging experiments, the animals were sacrificed by cervical dislocation

while anaesthetized, a blood sample was obtained by cardiac puncture, and organs

were harvested for imaging using the same settings as for in vivo imaging. Organ

fluorescence data was expressed as a percent of the sum of fluorescence measured in

all tissues, normalizing the signal in the volume of blood measured (50 ul) to total blood

volume in a mouse (assumed to be 2 ml) for a 20 g mouse.

Image stacks were spectrally unmixed using Kodak Multispectral software

version 1.1 (Carestream Health). Unmixed 16-bit images were then background-

subtracted in ImageJ using a Rolling Ball radius of 150 pixels (50 for images of

harvested organs). Image intensity data was then obtained by region-of-interest (ROI)

analysis, tracing either the entire animal minus the tumour, the tumour outline on the

animal, or the outer edge of the harvested organs using the “Freehand Selection” tool,

and determining the mean intensity. Colour tables were applied to mouse (“Smart”) and

organ (“Red Hot”) images in ImageJ, and brightness was calibrated appropriately for

presentation.

92

5.2.8 Histology, Silver Enhancement & Fluorescence Microscopy

Samples for histology were fixed in 10% buffered formalin (Sigma-Aldrich).

Sample processing, mounting and haematoxylin staining were carried out by the

University Health Network laboratories using standard procedures. Silver enhancement

was carried out on de-waxed mounted sections prior to haematoxylin staining, using the

Silver Enhancement Kit for Light and Electron Microscopy (Ted Pella). Tissue sections

were rinsed with double distilled H2O, an equal volume of the kit’s two reagents were

mixed in a microcentrifuge tube, and 50 μL was deposited on top of the section.

Enhancement was allowed to proceed for 20-30 minutes before being stopped by

rinsing with double distilled H2O.

Fluorescence histology images were obtained on an Olympus IX51 (Olympus)

with excitation by a X-Cite120 light source (Exfo), and collection by a Retiga EXi CCD

camera (QImaging). Brightfield colour histology images were collected on a Axiovert

135 (Zeiss) and Coolsnap-Procf camera (Media Cybernetics).

5.2.9 Statistics

Values of percent injected-dose (%ID) for biodistribution of AuNP were compared

between 6 hours (n=3) and 24 hours (n=3) post-injection using a two-way Student’s t-

test. An increase in signal from tumours of biotin-AuNP treated mice were compared

against mPEG-AuNP controls (n=3) using a two-way Student’s t-test. Differences in

contrast agent biodistribution within harvested organs 2 HPI were tested by ANOVA and

post-hoc Dunnet’s test, comparing tissues from in vivo assembly group animals group

against the 3 controls (n=3). Differences in tissues harvested at 24 HPI were examined

with a two-way Student’s t-test. Signal-over-noise was calculated by ((target ROI - non-

target ROI)/(standard deviation of non-target ROI)). We tested for a higher signal-over-

noise and larger area-under-the-curve from in vivo assembly vs. AuNP-A750 passive

targeting nanoparticles using a two-way Student’s t-test (n=3).

93

5.3 Results

5.3.1 Nanoparticle Anchor Component Synthesis

Previous research and results in Chapter 4 provided guidelines for designing the

nanoparticle anchor component36, 125, 129. We used gold nanoparticles (AuNP) stabilized

with a thiolated-poly-(ethylene glycol) surface brush layer, consisting of both biotin- and

methoxy-terminating (mPEG) molecules (see Figure 25 for particle characterization).

The particles were stable and no aggregation was evident. Biotin-PEG provides ligands

for assembling streptavidin-fluorophore conjugates onto the biotin-gold nanoparticle

(biotin-AuNP) surface. The biotin-AuNP anchors were optimized to bind a large number

of streptavidin conjugates (approximately 500 streptavidin constructs per particle.

Figure 26A, B), and we verified that the metal nanoparticle did not quench the dye’s

fluorescence upon binding (Figure 26C,D), as has been reported for other noble metal

particle fluorophore conjugates130, 131.

5.3.2 In Vivo Behavior of the Anchor Component

We engineered the nanoparticle geometry and surface chemistry towards

specific in vivo behavioral criteria (i.e. pharmacokinetic and intratumour distribution),

validated by profiling its biodistribution (Figure 27A). By 24 hours post-injection (HPI),

the biotin-AuNP had a significant accumulation in tumour tissue (0.72±0.09 vs.

0.06±0.03 %ID/g at 6 HPI), and was nearly cleared from circulation (0.97±0.68 vs.

4.67±0.86 %ID/g at 6 HPI). Its 70-80 nm hydrodynamic diameter permitted

extravasation into the tumour interstitium but restricted permeation through the tumour

extracellular matrix, causing the biotin-AuNP to localize in perivascular spaces in close

proximity to leaky blood vessels (Figure 27B)118, 125. The biotin-streptavidin system was

chosen because it undergoes near-irreversible binding, but also because prior antibody-

based targeting studies had clearly demonstrated the binding functionality of this system

within the tumour interstitium132-137.

94

Figure 25. Characterization of the gold nanoparticles used as the anchor component. A) Ultraviolet-visible absorption profile of the synthesized AuNPs

(λmax=526.5). B) Hydrodynamic diameter of the Biotin-PEG-AuNPs, with the primary

peak maximum at 74.1 nm (TEM validated that peak at 10 nm is a measurement

artifact). Control particles (mPEG-GNPs) had a comparable hydrodynamic diameter

(76.2 nm). C) Transmission electron microscopy images of the synthesized GNP (scale

bar=20 nm). D) Histogram showing the biodistribution of sizes synthesized, with

analysis in ImageJ finding a mean diameter of 26.3 nm.

95

Figure 26. Optimization of particle pegylation for streptavidin-Alexa fluor 750 binding. A) Aliquots of citrate stabilized AuNP were incubated with a low amount of

mPEG-SH (M.W.=5,000) at 0.3x molecules per nm2 surface area, and various amounts

of biotin-PEG-SH (M.W.=10,000). Washed particles were then incubated with a fixed

quantity of streptavidin-Alexa fluor 750 for 10 minutes while shaking, and the products

were separated by electrophoresis through a 0.7% agarose gel. Assembled AuNP-

Alexa fluor 750 are seen migrating very slowly from the wells (top band), and free dye

(multiple species) are seen migrating down. B) Assembly of the components was

quantified by mean signal intensity (ImageJ), graphed versus biotin-PEG molecules

added. C) Enhancement of the Alexa fluor signal due to the nanoparticle surface was

tested by measuring the fluorescent signal of AuNP-A750 complexes over a range of

different Strept-A750 to biotin-AuNP ratios, compared to controls in which Strept-A750

was first pre-blocked by saturating with biotin-PEG prior to mixing with biotin-AuNP.

The signal from assembled AuNP-A750 (•) was higher than the control (•) by D) 10-20%

over the range of ratios tested. Error bars indicate standard error.

96

Assuming the molecular component chases the biotin-AuNP anchor into the

tumour interstitium, assembly should occur upon co-localization. The spatial distribution

of both components was visualized by injecting a single 2.5 µM bolus of streptavidin

Alexa fluor 568 (strept-A568) into tumour-bearing mice 24 HPI of AuNP-biotin and

collecting tissue for histology after a further 6 hours. There was a similar pattern of

contrast agent (Figure 27C) and biotin-AuNP (Figure 27D) distribution away from blood

vessels within the tumour interstitium. To further validate assembly of the two

components in vivo, we tested whether two streptavidin Alexa fluor dye conjugates

(strept-A568 and -A750) could simultaneously bind biotin-AuNP, causing Förster

Resonance Energy Transfer (FRET) between the two dyes by bringing them within

close proximity. After achieving FRET on biotin-AuNP in vitro (Figure 28), we injected

tumour-bearing mice with a single strept-A568/750 bolus 24 HPI of biotin-AuNP.

Organs were harvested and imaged 2 HPI of contrast agents for FRET-specific

fluorescence (Figure 27E). Autofluorescence-subtracted signal in tumours from biotin-

AuNP mice was more than twice as high as in control mPEG-AuNP mice (Figure 27F).

This and the co-localization data strongly supports that assembly of the two

components can occur within tumours.

5.3.3 Imaging of In Vivo Kinetic Behavior: 2 Hours Post-Injection

We were then motivated to test whether in vivo assembly could alter a contrast

agent’s tumour accumulation kinetics in a favourable manner. Streptavidin-Alexa fluor

750 was injected 24 HPI of biotin-AuNP, and fluorescence data was collected for nearly

two hours. We opted for single-fluorophore detection over FRET because longer

wavelength excitation light provides better tissue penetration6, 138. We compared the

kinetics of our assembly strategy against a non-assembling control (mPEG-AuNP

injection followed 24 hours later by strept-A750), and against a passive targeting

fluorescent nanoparticle. This was constructed by assembling biotin-AuNP and strept-

A750 prior to injection. As a final control, we competitively inhibited our system by

injecting non-fluorescent streptavidin 30 minutes before injection of strept-A750.

97

Figure 27. Biodistribution, tumour localization and assembly of the two components. A) Biodistribution of the biotin-AuNP anchor at 6 (■) and 24 (□) HPI

showing a decrease in blood concentration and an increase in tumour, liver and kidney

tissue. B) Silver enhancement staining of tumour tissue 24 HPI of biotin-AuNP showing

perivascular localization of the anchor component. Red arrows mark examples of leaky

blood vessels. C) Fluorescence microscopy showing an overlapping distribution of

98

streptavidin-Alexa568 (red) with the D) biotin-AuNP anchors at 6 HPI of contrast agent.

Nuclei are stained with DAPI (green), and yellow arrows mark a blood vessel. E) FRET

images of tissues harvested from mice 2 HPI of strept-A568 + 750 (excitation

wavelength of 570, emission wavelength of 790). Control mPEG-AuNP organs on the

left are marked with boxes (left to right and top to bottom: blood, tumour, liver, spleen,

heart, lung, kidneys). Green arrows mark an aliquot of the streptavidin conjugates

mixed with mPEG-AuNP (left) and biotin-AuNP (right), clearly showing a higher signal

from FRET in the latter. Calibration bar=500 to 7500 arbitrary unit (a.u.). F)

Quantification of fluorescence found a higher signal in tumours of biotin-GNP vs.

mPEG-GNP injected mice. Data points are the mean ± s.e.m. from n=3 animals.

Asterisk, p<0.05 students t-test between 6 and 24 hours (biodistribution data), p=0.050

1-tailed Student’s t-test vs. mPEG control (FRET data).

99

Figure 28. Förster Resonance Energy Transfer between an Alexa fluor dye pair bound to the biotin-AuNP. Streptavidin-Alexa fluor 568 and 750 were mixed

individually (Lane 1= 568, Lane 2= 750), or simultaneously (1:1, Lane 3) with the biotin-

AuNP component, incubated 10 minutes, then separated via electrophoresis in a 0.5%

agarose gel A) Fluorescence imaging by excitation at 570 and emission at 600. B)

Fluorescence imaging by excitation at 750 and emission at 790. C) Fluorescence

imaging for FRET by excitation at 570 and emission at 790. D) Contrast image by

excitation at 570 and no emission filter.

100

Region-of-interest (ROI) analysis (method shown in Figure 29) revealed

intriguing differences in contrast agent behavior between in vivo assembly and control

mice. To examine the non-target (background) signal within the mice, we applied an

ROI that encompassed the whole animal side profile excluding the tumour ROI. The

mean signal intensity (I) change over time (I/Io) in the non-target ROI was identical in

biotin-AuNP and mPEG-AuNP injected mice (Figure 30). Competitive inhibition of

assembly also appeared to compete for uptake into the kidneys. Mice injected with

passive targeting AuNP-streptA750 displayed the lowest rate of background signal,

which may reflect differences in nanoparticle vs. macromolecule diffusion into peripheral

vasculature.

We then examined the net tumour accumulation of the contrast agent by subtracting

mean non-target ROI signal from tumour signal (Figure 31A,B). Tumours of biotin-

AuNP injected animals displayed a 265.8% net increase within 30 minutes, a rate 16.0-,

8.1- and 4.6-times that of the mPEG-AuNP, AuNP-streptA750, and competitive

inhibition controls, respectively. By 90 minutes, tumour signal in biotin-AuNP animals

was 2.83-, 3.67- and 2.32-times that in mPEG-AuNP, AuNP-A750 and competitive

inhibition controls.

The kinetic data was then cross-validated against biodistribution of the contrast

agent in harvested organs (Figure 31C,D). With the exception of tumours,

biodistribution of strept-A750 was identical in biotin-AuNP and mPEG-AuNP treated

animals. Mice that were used for competitive inhibition of assembly showed a non-

significant reduction in liver, spleen, and kidney clearance. Passively targeted

fluorescent AuNP-A750 showed higher spleen (p<0.05) and lower kidney (p<0.01)

uptake versus the assembly strategy.

The tumours of biotin-AuNP treated mice contained 3.4% of the total

fluorescence measured across all tissues. This is higher than the 2.1, 0.9 and 2.1% in

mPEG-AuNP, passive targeting AuNP-A750 and competitive inhibition controls,

respectively. Accumulation in mPEG-AuNP and competitive inhibition controls was

identical. Accumulation in biotin-AuNP tumours was 1.3% higher than in corresponding

mPEG-AuNP and competitive inhibition controls. We can estimate whether this

101

increase can be attributed to assembly by calculating the capture potential of the biotin-

AuNP localized in tumour tissue. Based on a capture potential of 500 streptavidin per

particle and accumulation of 0.72% of the total biotin-AuNP injected dose, we would

expect an increase in signal up to 1.93%. The measured increase is therefore within the

expected range.

102

Figure 29. Region-of-interest (ROI) analysis from fluorescence data. ROI’s were

drawn around a) the body of the mouse and b) the tumour to determine the mean

fluorescence signal. c) Tumour location was verified on RGB composite images, and is

marked by arrows.

103

Figure 30. Region-of-Interest analysis of non-target (mouse body) normalized to the initial signal (Io) for 100 minutes post-injection of strept-A750. Mean intensity

values over time were normalized to the initial mean intensity. The biotin-AuNP (―)

and mPEG-AuNP (―) treated mice had similar body fluorescence kinetics, showing a

rapid initial increase, followed by a drop due to clearance. Pre-assembled AuNP-A750

constructs (―) had lower overall signal that increased linearly over approximately 100

minutes, and competitive inhibition (―) of biotin-AuNP assembly by injection of non-

fluorescent streptavidin 30 minutes prior to injection of strept-A750 showed a rapid

increase in signal over the entire time of acquisition.

104

Figure 31. Optical detection of fluorescence and analysis of tumour accumulation over 2 HPI of assembling contrast agent. A) Representative

fluorescence image at 2 HPI of biotin-AuNP (top) and mPEG-AuNP (bottom) injected

mice. Arrows mark tumour location, and calibration bar = 500 to 6000 a.u. B) Net

tumour accumulation (tumour-body) of contrast agent over time in biotin-AuNP (―)

injected mice, as well as control mPEG-AuNP (―), AuNP-A750 (―), and competitive

inhibition (―) injected mice. C) Representative fluorescence image of organs harvested

from biotin-AuNP (top) and mPEG-AuNP (bottom) injected mice 2 HPI of strept-A750.

Arrows mark the tumours, and calibration bar = 100 to 4500 a.u. D) Quantification of

tissue fluorescence from biotin-AuNP (■), mPEG-AuNP (■), AuNP-A750 (■) and

competitive inhibition (■) mice. Data points are the mean ± s.e.m. from n=3 animals.

Asterisk, p<0.05 for biotin-AuNP vs. mPEG-AuNP and competitive inhibition controls,

and p<0.01 for biotin-AuNP vs. AuNP-A750 using ANOVA and Dunnet’s post-hoc test.

105

5.3.4 Imaging of In Vivo Kinetic Behavior: 24 Hours Post-Injection

Finally, we compared the accumulation kinetics of in vivo assembly to that of

passively targeting nanoparticles (AuNP-A750) over a 24 hour period. One noticeable

difference was that the contrast agent’s time in circulation was greatly extended through

assembly onto biotin-AuNP prior to injection vs. injecting the components in sequence

(Figure 32). We also observed that a large amount of assembly occurred in drainage

tissue around tumours (see top mouse, Figure 33A). In vivo assembly provided a

nearly 200-fold higher rate of tumour accumulation than passive targeting AuNP-A750

within the first 3 HPI, reaching a maximum of 5.1-times higher net signal (p<0.05,

Figure 33B). The level of contrast agent then remained constant between 3 and 24

hours, during which time the passive targeting AuNP-A750 increased linearly until

reaching a statistically similar level at 24 HPI (p=0.31). In vivo assembly therefore

achieved comparable signal to a passively targeting nanoparticle, but 8-times faster.

As a measure of diagnostic sensitivity, the signal-over-noise achieved by in vivo

assembly was triple that of passive targeting nanoparticles by 3 HPI (p<0.05, Figure 33C). As well, the total AuC in the tumour achieved by in vivo assembly was calculated

to be 2.46-times that of passive targeting, offering a greater window of opportunity for

detecting tumours (Figure 33D).

We again cross-validated the kinetic data by profiling contrast agent

biodistribution in organs harvested 24 HPI (Figure 33E, F). Mice injected with pre-

assembled AuNP-A750 had higher signal remaining in the blood (p<0.001) and lower

uptake in the liver (p<0.05) and kidneys (p<0.01) at 24 HPI. Tumour signal was

comparable by 24 HPI (p<0.5), supporting what we observed in vivo. Attachment of

fluorescent streptavidin onto pegylated AuNP can therefore maintain it within the

circulation for long enough to achieve high levels of tumour accumulation via the

enhanced permeability and retention effect 24 hours after administration. However, a

comparable level of signal was achieved by in vivo assembly at 3 HPI.

106

Figure 32. Signal Intensity (I) in mouse bodies normalized to the initial signal (Io) over 24 HPI of strept-A750. Mean intensity values over time were normalized to the

initial mean intensity. The biotin-AuNP (―) body fluorescence kinetics reached a

maximum between 6-12 hours and then dropping off due to clearance. Signal in mice

treated with pre-assembled AuNP-A750 (―) continuously increased, and was at a

maximum at 24 HPI.

107

Figure 33. Optical detection of fluorescence and analysis of tumour accumulation over 24 HPI of streptavidin-Alexa fluor 750. A) Representative

fluorescence image at 24 HPI of biotin-AuNP (top) and AuNP-A750 (bottom) injected

mice. Arrows mark tumour location, and calibration bar = 500 to 6000 a.u.. B) Tumour

accumulation of contrast agent over time in biotin-AuNP (―) and AuNP-A750 (―)

injected mice. C) Signal-over-noise (S/N) at 3 HPI and D) total AuC achieved with in

vivo assembly (■) vs. in vitro assembly (■) of components. E) Representative

fluorescence image of organs harvested from mice injected with the in vivo assembly

108

system (top) and AuNP-A750 construct (bottom) at 24 HPI. Arrows mark tumours, and

calibration bar = 50 to 3000 a.u. F) Quantification of tumour fluorescence from mice

injected with the in vivo assembly system (■) vs. AuNP-A750 (■). Data points are the

mean ± s.e.m. from n=3 animals. Asterisks, 2-tailed Student’s t-test for in vivo

assembly vs. passive targeting AuNP-A750 of S/N (p=0.048) and AUC (p=0.036). Note

that animal and organ panels (A,E) are products of separate images that underwent

identical processing.

109

5.4 Discussion Strategies for nanoparticle-based tumour targeting involve either passive particle

accumulation via the EPR effect, or active targeting by inclusion of biomolecular

targeting agents on the particle surface. Active targeting is able to achieve a faster and

higher level of contrast agent accumulation than passive accumulation and is therefore

considered to be the preferred method. Yet active targeting suffers from a major

drawback in that it would require profiling of each individual patient’s antigens prior to

administration of an appropriately targeted agent. Alternative strategies that achieve

rapid contrast agent accumulation without targeting tumour antigens would therefore be

advantageous. In this chapter, we described a novel targeting strategy whereby

synthetic antigens on passively targeting nanoparticles are delivered to tumour tissue

via the EPR effect, providing a recognition moiety for an actively targeting contrast

agent (Figure 34).

A simplified model of tumour targeting might consider that the rate and extent of

an agent’s accumulation is defined by its rate of transfer from circulation into the tumour

mass (Kin), and rate of clearance from the tumour (Kout) either into circulation or

peripheral host tissue. Our rationale for in vivo assembly was that a high Kin would be

provided by the relatively small molecular contrast agent, and that assembly of this onto

the large diameter nanoparticle anchor would greatly reduce Kout and maximize

accumulation. Although we were not able to directly measure these rates individually

due to a lack of required instrumentation, we used whole animal optical imaging to

observe bulk differences in accumulation. Supporting our rationale for designing this

strategy, in vivo assembly achieved a faster rate of accumulation after injection of the

contrast agent relative to control groups, as high as 16-times. Remarkably,

measurements of fluorescent signal in tumour tissue harvested 2 HPI of the contrast

agent revealed an increase in contrast agent accumulation due to assembly that was

approximately in line with the binding potential of the particles. This finding greatly

suggests that further modulation of accumulation could be achieved by altering the dose

of anchor or contrast agent administered.

110

Figure 34. Schematic of nanoparticles assembling with contrast agent in vivo. Gold nanoparticles stabilized with biotinylated PEG (denoted as biotin-PEG anchor) are

injected as a first step. These enter tumours through leaky vasculature and passively

accumulate in the extracellular matrix over 24 hours. Fluorescently labeled streptavidin

is injected, leaks into tumours and interacts with biotin on the gold nanoparticles in the

interstitium. This favourably alters the contrast agent’s tumour accumulation kinetics

and leads to enhancement of fluorescent signal.

111

The difference in net signal that we observed 24 HPI of contrast agent was

interesting. There is little data available to describe how different targeting strategies

and agents may accumulate in tumours. We measured the accumulation of

streptavidin-Alexa fluor assembling onto synthetic antigens, in comparison to a

passively targeting 80 nm fluorescent AuNP. Passive nanoparticle targeting is well

documented as being a slow process that requires many hours of circulation. We found

that the fluorescent nanoparticle avoided RES clearance well enough to remain in

circulation 24 HPI, and that this allowed for a gradual, linear increase in net tumour

signal. In contrast to this behavior, our in vivo assembly strategy achieved a rapid

accumulation, reaching a maximum 3 HPI and remaining at that level for the remaining

duration of imaging. It should be noted that in vivo imaging and ex vivo organ analysis

found a statistically similar level of accumulation by the two strategies at 24 HPI, but

that in vivo assembly achieved a higher signal-over-noise, area-under-the-curve, and

net tumour signal within 3 HPI.

The nanoparticle design that we used for our targeting strategy was based on

fundamental knowledge provided in earlier research36, 125. We used the biotin-

streptavidin binding system because of its strong binding affinity. Future studies on

alternative molecular binding systems could aim to improve the overall signal-over-noise

ratio, or to enable alternative contrast agents (i.e. gadolinium, radioisotopes) or

therapeutics to assemble on nanoparticle anchors in vivo. The results presented here

clearly demonstrate the concept and advantage of assembling targeting components in

vivo to improving tumour delivery of diagnostic agents.

112

5.5 Chapter Author Contributions The study was conceived of by Steve Perrault and Warren Chan. The

experiments were designed, carried out and data analyzed by Steve Perrault. The

manuscript was written by Steve Perrault and Warren Chan.

113

Chapter 6

Conclusions and Future Work

6.1 Statement of Major Conclusions

The hypotheses tested and science described in completion of this thesis were

undertaken because nanotechnology has shown enormous promise for reducing

mortality rates due to cancer. The author hopes that the work described herein will

contribute to this aim, as well as in generally advancing our understanding of how the

design and synthesis of nanomaterials can be tuned towards specific biomedical

applications.

Nanomedicine emerged as a discipline of applied engineering science based on

discoveries made in materials science over the past several decades. The pace of new

nanomaterial discovery has grown, as do new methods of modifying them for specific

applications. The work described in Chapters 2 and 3 contribute to this effort. Through

a scientific and analytical approach to materials synthesis in Chapter 2, we adopted

known methods and developed comprehensive protocols to produce pegylated gold

nanoparticles for use in passive tumour targeting. These can now be applied by others

to validate the findings described in this thesis, or adapted to test alternative hypothesis

regarding the interaction of nanomaterials with biological systems.

In Chapter 3 we described a novel preparative technique for colloidal gold

nanoparticles that overcomes limitations of the method pioneered by Turkevich 60 years

ago39. Our hydroquinone-based synthesis method has a number of major advantages.

It produces batches of colloidal gold having significantly improved monodispersity and

quasi-spheroidal shape consistency. As well, our protocol has expanded the range of

sizes that can be produced beyond what was previously possible. Finally and

importantly, it shows much greater consistency and predictability in production than the

114

Turkevich/Frens method. For these reasons and due to the ubiquity of gold

nanoparticles as a reagent in nanotechnology, we expect our method to have

widespread use within nanomedicine and nanotechnology in general.

Our in vivo tumour targeting study in Chapter 4 is not the first to investigate the

design-dependence of nanomaterials on tumour accumulation. Dr. Jain first described

such effects in a seminal work published in 199889. However, we advanced that and

other research teams’ findings in several respects. No research group had

systematically investigated the sub-100 nm particle size range for design-dependent

effects. Our findings show the importance of this range. We discovered that sub-100

nm pegylated particles have a broad range of circulation half-lives, dependent on their

design. Moreover this was a strong predictor of total tumour accumulation for a given

design. As well, we observed that the 20-100 nm range displays a complete spectrum

of tumour permeation behavior, with 20 nm particles diffusing rapidly and accumulating

poorly, and 100 nm particles having no capacity to permeate the tumour extracellular

matrix. Further studies will need to build on this finding across different tumour models,

yet the implications are significant.

We were also able to show how such fundamental information might be applied.

Using the limited tumour-permeation capacity of large diameter nanoparticles, we

developed a novel contrast agent delivery strategy that overcomes limitations of prior

strategies. Early tumour detection is a strong indicator of a successful outcome

because an early stage malignancy is less likely to have metastasized and become

drug resistant66. Methods that improve success of tumour surveillance are therefore

very likely to make an impact on cancer mortality rates. Our strategy is also a first

demonstration that assembling multiple components of a nanoparticle device within

tumour interstitium is possible. The ability to combine kinetics and functions of multiple

materials should allow a variety of obstacles to be overcome. We therefore expect this

conceptual demonstration to be a stepping stone for many future applications using in

vivo assembly of nano devices.

As is typical in research, the work in this study provides few complete and final

answers to problems in nanomedicine. Instead, the data often supported an opposing

115

answer to our initial hypotheses and gave rise to many additional questions. Some

potential future directions for follow-up studies are described in the next few sections.

6.2 Future Work

6.2.1 Nanoparticle Synthesis

New nanomaterials and improved synthesis techniques for “old” materials will

continue to be described. There were several obvious challenges and unanswered

questions that we encountered in completing our biological studies. We found the

quality of materials produced by the Turkevich/Frens method generally lacking.

Synthesis of batches over 30-40 nm showed a high degree of size dispersion and

elongated, irregular shapes. Although we were able to overcome this through

development of a new synthesis method, it was not in time to use this in our study

systematically testing passive nanoparticle accumulation in tumours. Follow-up studies

may find that more monodispersed pegylated particles with more spheroidal shapes

have a greater ability to reject immune recognition and RES clearance than those used

in Chapter 4. I would expect and anticipate that future studies should aim to use the

highest quality materials possible, and perhaps to test the impact of size and shape

irregularities on in vivo behavior. As well, we did not investigate the long-term stability

of thiol-PEG brush layers on the particles. It would be worthwhile to examine how

storage conditions impact PEG stability, and how this in turn can impact

pharmacokinetics in vivo.

6.2.1.1 Material synthesis study 1: measuring the impact of imperfections on pharmacokinetics of quasi-spheroidal gold nanoparticles.

Hypothesis: Deviations in spheroidal shape consistency will result in imperfect mPEG-

thiol layers on gold nanoparticles, leading to a reduced blood half-life.

Approach:

116

1. Synthesize 3-5 batches of large (i.e. 60 nm) gold nanoparticles having

quantifiably different shapes. This could include differences in the length-to-

width axis ratio, as well as differences in the smoothness of the surface. Note in

Chapter 3 that changing the concentration of citrate in hydroquinone-synthesis

reactions results in particles with knobbed surfaces. Variations of the

hydroquinone and Turkevich/Frens synthesis methods should therefore allow

batches of different shape qualities to be produced and characterized.

2. The particles should then be pegylated following a consistent protocol, for

example all batches should be pegylated to saturation with mPEG-thiol having

molecular weight 5000 Da.

3. The particles should then be concentrated, washed, and characterized.

Following this, injection doses can be prepared. Intravenous injection of the

particles into either a wildtype mouse or rat model will allow blood samples to be

obtained at various timepoints post-injection. Quantification of blood particle

content will allow pharmacokinetic trends to be examined and parameters

calculated.

4. Statistical testing of specific pharmacokinetic parameters (i.e. half-life) based on

different nanoparticle batch qualities should then allow conclusions to be made

regarding the impact of particle shape qualities on in vivo behavior.

Expected Results: Rejection of immune factors by PEG will be found dependent on

the density and arrangement of PEG molecules on the particle surface. Irregularities

in particle shape will likely result in imperfect PEG brush layers, which will impact

particle pharmacokinetics in vivo. We would then expect particles with near-

spheroidal shapes and few shape imperfections (i.e. bending, knobs) to produce the

most consistent mPEG brush layers and to have the longest pharmacokinetic

profiles.

117

6.2.1.2 Material synthesis study 2: stability of the mPEG brush layer and dependence on storage conditions.

Hypothesis: Oxidation or loss of mPEG brush layers over time will result in decreased

particle stability. Optimizing storage conditions will improve this and provides guidelines

for future studies and commercialization.

Approach:

1. Synthesize and pegylate particles. Characterize their stability in a 1M NaCl

solution after characterization.

2. Split the particle stock into various conditions, including freezing, 4°C and room

temperature in water as well as in low salt buffer (i.e. 5 mM sodium bicarbonate),

and “powdered” particles obtained by freeze drying over the same range of

temperatures. For each condition, produce a number of aliquots that can be

characterized at various time points after starting (T0).

3. Determine other potential methods to quantify particle stability and PEG brush

layer quality, including perhaps oxidation or degradation of PEG molecules.

4. At pre-determined timepoints providing at least 6 months of data, characterize

aliquots from each condition for stability in salt, hydrodynamic diameter (looking

for aggregation), and absorption spectra. For the freeze-dried particles, test their

capacity to be resuspended in water or low salt buffer as well.

5. Analyze results to determine differences in particle stability based on storage

conditions.

Expected results: Stability of the particles and PEG brush layer may improve if the

particles are powdered after pegylation, and stored at either -20 or 4°C. Pegylated

particles would be expected to resuspend well after freeze-drying. This should

extend their shelf-life from weeks to months or longer. The guidelines produced by

this study will be enormously helpful for future studies utilizing pegylated gold

nanoparticles.

118

6.2.2 Future Work: Continuing to Improve Our Understanding of Nanoparticle Design and In Vivo Behavior.

The interaction of nanomaterials with biological systems is a highly complex

phenomenon, and a truly comprehensive understanding of the process may be

unachievable. This is in part because the variety of nanomaterials and their

modifications that can be produced is near-infinite. Nevertheless, I anticipate that a

systematic examination of particle shape and in vivo behavior may discover a design

that shows significant advantages over spheroidal nanoparticles. For example,

nanorods may display excellent pharmacokinetic and tumour permeation behavior, and

can contribute therapeutic functionality. Future studies could therefore aim to test a

variety of different nanomaterial shapes. The major challenge of such a study would be

to synthesize and then modify the surface of many different particle shapes in a manner

that makes them equivalent or comparable for in vivo behavior. Otherwise, a similar

approach to in vivo analysis could be used to that described in Chapter 4.

A drawback of the study conducted in Chapter 4 is the size of the animal cohort

required by the analysis methods used. This will have increased the experimental error,

in addition to slowing down the discovery of useful knowledge and requiring

euthanization of more animals. It may therefore be advantageous to develop alternative

methods to investigate the paradigm of nanomaterial design and biological impact.

High-throughput methods based on multiplexed imaging would be a faster and less

animal-intensive approach to advance our understanding. As well, it would significantly

reduce the error associated with aggregating pharmacokinetic data obtained from a

large number of animals. Considering the heterogeneous nature of cancer and tumours

and the number of variables involved in design of nanomaterials, a multi-variate

analysis approach in which a single outcome of interest is identified (e.g. maximum total

tumour accumulation or animal survival), then many different particle design

combinations tested to reveal statistically which is best at achieving that outcome might

be most powerful.

119

6.2.2.1 Multiplexed detection of nanoparticle pharmacokinetics and tumour accumulation.

Hypothesis: Multiplexed detection of two or three fluorescent nanoparticles will allow

for greatly improved accuracy in profiling of how particle design impacts in vivo behavior

and tumour accumulation. Development and proof-of-concept demonstration of this

device could therefore allow for faster advances in improving targeting efficiency.

Approach:

1. Develop methods for fluorescent labeling of pegylated gold nanoparticles. This

could include conjugation of fluorescent dye to a PEG-thiol prior to particle

pegylation, or labeling after pegylation. Characterize the product for

fluorescence and overall stability.

2. Using the methods optimized in “1”, design three different particles for use in

vivo. This could include passive versus active targeting, different sizes of a

single particle shape, different particle shapes, or a single size/shape with

different surface chemistries. Synthesize the three designs and characterize

them in full.

3. Seed tumours on one hind leg of CD1 nude mice using MDA-MB-435 human

cancer cell lines. When tumours have reached approximately 1 cm length,

administer particles and track pharmacokinetics for imaging. Imaging

experiments should first test each nanoparticle design alone, then duplex

multiplexing, and finally triplex multiplexing. Imaging of pharmacokinetics for 4-6

hours post-injection (high resolution) as well as for 24 hours post-injection (low

resolution) should be performed.

4. Analyze data in two ways. First, to compare the quality of data achieved when

tracking probes alone versus multiplexing. Use standard deviation as a metric.

Second, compare pharmacokinetics of the probes to determine major differences

in behavior and tumour accumulation.

120

5. Produce guidelines from the results in “4”. First, guidelines on whether

multiplexing can produce high quality data that represents what is observed

when probes are measured alone. Second, guidelines on particle design for

efficient tumour accumulation.

Expected Results: Standard deviation of tumour accumulation data will show

significant improvement (decreased error) versus data sets produced from

aggregating biodistribution analysis of tissues collected from many mice. Single

probe analysis will produce the smallest standard deviation. Increased multiplexing

will increase error but this approach may still be an improvement over analyzing

particle content in tissues by other methods. Otherwise, design-dependent tumour

accumulation will generally reflect what was described in Chapter 4 of this thesis.

121

6.2.3 Future Work: Improving and adapting In Vivo Assembly.

The in vivo assembly concept is new and may be enormously useful. Future

studies could aim to improve the approach and application that we have described. For

example, our study made use of the biotin-streptavidin molecular binding system

because it is well characterized and is known to function in vivo. However it is not an

ideal system because streptavidin’s molecular weight is above the renal clearance

threshold (of mice). In order to maximize signal-over-noise by this strategy, it would be

advantageous to replace this with an alternative binding system that allows the second

component (molecular contrast agent) to be cleared through renal filtration. Nucleic

acid-based assembly offers a number of obvious advantages, most notably reversibility,

but the probability of successfully designing a DNA or RNA hybridization system for use

in vivo might be remote.

Future studies could also use the system described here to model how

optimizing the quantity of contrast agent administered relative to available antigens in

tumours can be useful. For example, in Chapter 5 we calculated that approximately 2%

of the injected contrast agent dose could be captured by nanoparticles in the tumour.

Administering less contrast agent could allow a greater proportion of the dose to be

captured by the particles, and may provide a higher signal-over-noise. The gold

nanoparticle design could also be optimized to increase or decrease the number of

antigens present on the surface. Testing the impact of antigen variability on diagnostic

sensitivity would be extremely difficult using a purely biological model, but the synthetic

universal antigen system presented in this thesis provides a useful tool.

Many additional and useful applications of in vivo assembly can be envisioned.

For example, a multi-component drug delivery system consisting of a drug-laden

nanoparticle and molecular tether could be used to overcome the binding site barrier

and achieve greater drug efficacy against tumour cells. Alternatively, assembly of a

multi-particle system using plasmonic nanoparticles might allow for major increases in

signal-over-noise of surface-enhanced ramen spectroscopy-based diagnostics. In

general, one could aim to make use of emergent nanomaterial and molecular properties

122

that are available from assembled systems and not individual components, and where

temporal and/or spatial control of the property is advantageous.

Although I have described several possible uses of in vivo nano device

assembly, I hope that this concept becomes established as a “third option” for tumour

targeting applications, and that others will envision and describe many additional

applications.

6.2.3.1 In vivo assembly study 1: Testing alternative biomolecular assembly systems.

Hypothesis: Assembly of anti-mPEG IgG antibody fragments (Fab) onto mPEG-AuNP

will allow renal clearance of the probe and improved signal-over-noise.

Approach:

1. Process the anti-mPEG IgG from Epitomics using a Pierce antibody fragmention

kit. This will produce fragments of approximately 50 kDa, small enough to allow

renal clearance in mice.

2. Label the Fab with a near-infrared fluorescent dye, for example Alexa-fluor 750.

3. Characterize binding of the labeled Fab to mPEG-AuNP in vitro using a gel shift

assay, and quantify binding kinetics using either an ELISA-based protocol or

Biacore analysis service.

4. Synthesize particles for in vivo anchor component, aiming for 60-80 nm final

hydrodynamic diameter and using mPEG-SH 5 kDa.

5. Apply the anchor component to a tumour model, collecting tissue at 8 and 24

hours post-injection. Quantify particle content by ICP-MS.

6. Based on pharmacokinetic data determined from the biodistribution data,

determine an appropriate time to inject the mPEG-Fab-A750 probe. This should

be at a time point when little to no anchor component is still in circulation, but a

123

high dose remains in the tumour. Inject the probe and use the whole animal

optical imaging platform to measure kinetics and tumour accumulation. Measure

the rate of clearance and the improvement in background signal of the probe

versus streptavidin-Alexa fluor 750.

7. Compare the tumour accumulation kinetics resulting from in vivo assembly to that

measured for controls, including I) an Fab control probe that does not bind to

mPEG or any human antigen, II) a passively targeting fluorescent nanoparticle

control, III) other appropriate controls.

Expected Results: The time required for systemic clearance of the Fab cannot be

estimated, but ideally the probe will clear within 24 hours. This would provide a

much steeper decrease in background signal than streptavidin-Alexa fluor 750. If

the probe is able to spatially co-localize and bind the mPEG-AuNP, there should be

quite a significant improvement in signal-over-noise versus the system described in

Chapter 5.

124

6.2.3.2 In vivo assembly study 2: Optimizing contrast agent dosage to antigen availability.

Hypothesis: Higher levels of nanoparticle-bound antigens residing in the tumour, and

a lower ratio of contrast agent (probe) to antigen will greatly improve diagnostic signal-

over-noise.

Approach:

1. Synthesize 3 designs of biotin-AuNP using a consistent particle core size but

altering the number of biotin-PEG-SH on the surface. For example, synthesize a

stock of biotin-AuNP having 1000 biotins per particles, one with 500 and one with

100 biotins per particle. Also, synthesize mPEG-AuNP for use as a negative

assembly control.

2. Characterize streptavidin binding capacity of each biotin-AuNP design, as well as

fluorescence intensity of saturated, assembled constructs.

3. Apply the various biotin-AuNP designs to CD1 Nude mice bearing MDA-MB-435

tumours. Inject a consistent quantity of particles in each treatment. For each

design, 3 different quantities of strept-A750 should be injected. This produces a

total of 3 designs x 3 probe quantities = 9 treatments, plus controls. Use the

whole animal optical imaging platform to image accumulation of the probe over a

24-hour period. After completion of the 24 hours of imaging, harvest tissues and

image organs for probe biodistribution. Weigh all tissues after imaging, and save

the tumour sample.

4. Measure particle content of all tumour samples using ICP-MS. This will allow a

plot to be produced of fluorescence versus particle number (or antigen number).

5. Analyze imaging data and samples to correlate rate of accumulation, and signal-

over-noise, with particle content and antigen availability, as well as the ratio of

probe to antigen.

125

Expected Results: The signal-to-noise generated by in vivo assembly of strept-

A750 onto biotin-AuNP will be strongly dependent on the number of antigens

present, and the applied ratio of probe to antigen. It’s anticipated that increasing the

number of biotins per particle, or at least the number of strept-A750 that can be

bound without impacting fluorescence, should increase accumulation and signal-

over-noise. Increasing the number of biotins per particle could be detrimental to

blood pharmacokinetics and tumour accumulation at some threshold. Decreasing

the proportion of probe injected and thereby increasing the relative ratio of antigen to

probe should decrease background, and increase the proportion of probe captured

in the tumour versus accumulating non-specifically in other organs, and so should

increase diagnostic signal-over-noise. This study is expected to produce guidelines

on how I) in vivo assembly can be optimized to increase diagnostic sensitivity, and

II) the importance of probe quantity injected versus available antigen can be

optimized in an in vivo assembly or conventional active targeting strategy to improve

sensitivity.

6.2.3.3 In vivo assembly study 3: particle-to-particle assembly and control of geometry.

Hypothesis: Aggregates of metal nanoparticles gain interesting properties over

segregated particles, such as an enhancement of surface-enhanced ramon

spectroscopy (SERS) signal139 and changes in optical and electronic properties9, 140.

Controlling nanoparticle cluster or aggregate size and geometry is therefore an

important goal in many applications. Assembly of nanoparticles within tumours may

offer similar advantages, and the ability to control geometry of assembly would be very

powerful. A two component system of nanoparticles can be designed to undergo

molecular assembly both in vitro and within tumour interstitium. Precise engineering of

the molecular ligands controlling assembly should allow assembled geometry to be

controlled.

Approach:

126

1. Synthesize a two particle system. The first particle component should have

moderate to slow capacity to permeate the tumour interstitium, and should

contain either a biotin or streptavidin moiety on the surface, with the number of

sites being carefully controlled in the pegylation process. The second component

should have the complimentary moiety, and should be a much smaller particle to

allow faster permeation of the tumour extracellular matrix, allowing the smaller

particle to catch up to the larger particle when injected at a later time-point.

2. Apply the system of particles to CD1 Nude mice bearing MDA-MB-435 tumours.

Inject component 1, the larger component, and allow it to accumulate in tumour

tissue over ~ 24 hours. At this time, inject the second component. Allow the

second component to circulate, accumulate in tumours and assemble with the

larger component for 4-24 hours.

3. Euthanize the animal and collect tumour tissue for silver staining microscopy as

well as TEM.

4. Examine the distribution of particles throughout the tumour mass using silver

enhancement of histology samples and brightfield microscopy. If accumulation is

relatively good, carefully look through samples with TEM, finding particles and

examples of two-component assembly. Characterize the nature of assembly (ie

only 1 small and 1 large component, multiple small components per large

component, etc).

5. Try to replicate the geometry observed in vivo by setting up ratios of the two

components in solution reactions in vitro, applying these onto a TEM grid and

measuring geometry by TEM.

Expected Results: Geometry of assembly of two particle components will be

controllable in vitro. Specific ratios of component 2 to component 1 can be

determined by regulating the number of binding sites during pegylation. The major

challenge is to prove that assembly can be achieved in vivo, and that geometry can

be controlled in any way. This requires spatial co-localization of the two

components, which might be possible through careful design. Achieving specific

127

ratios of components in vivo may be much more difficult. If successful, this project

could provide guidelines for a SERS based detection strategy, whereby assembly of

particles in vivo results in enormous enhancement of SERS signal.

128

Appendix A

Curriculum Vitae of Steven D. Perrault

129

A. Personal Information

Name: Steven David Perrault

ISI ResearcherID: A-9457-2009

Mailing Address: 50 Lombard Street

Unit 1201

Toronto, Ontario

M5C 2X4

(647) 885 – 1346

Laboratory: 160 College Street

Terrence Donnelly Center for Cellular

and Biomolecular Research, Room 450

(416) 946-0015

Fax: Dept. (416) 978-4317

Email: [email protected]

Date & Place of Birth: September 6, 1978. St. Catharines, Ontario, Canada

Citizenship: Canadian

130

B. Education

Degree Institution Department Thesis Field Year

Ph.D. University of Toronto Institute of Biomaterials &

Biomedical Engineering

Nanostructures for cancer

diagnostics and therapeutics 2010

M.Sc. University of Guelph Biomedical Sciences Telomerase as a regulator of

eukaryotic cells.

2005

B.Sc.

University of Guelph Molecular Biology &

Genetics Telomere Length Analysis in

Cloned Goats and Their

Offspring

2002

C. Employment and Training History

Dates Employer Position

May 2005 – June 2006

Roche Diagnostics Canada: Clinical and Applied Science Division

Technical Support Rep.

February 2005 – Present Molecular Research Consulting

Proprietor

February 2005 – May 2005

University of Guelph

Research Technician

May 2002 – August 2002 Biomedical Sciences, University of Guelph

Student Researcher

May 2001 – August 2001 Biomedical Sciences, University of Guelph

Student Researcher

May 2000 – August 2000

Ontario Ministry of Agriculture, Vineland Research Station

Student Researcher

May 1999 – August 19999

Ontario Ministry of Agriculture, Vineland Research Station

Student Researcher

May 1998 – August 1998 Agriculture Canada, Vineland Research Station

131

D. Certifications

Dates Organization Certificate / Qualification

2005 – In Progress University of Guelph Certificate of Environmental Citizenship

2006 PADI Open Water Certification

2004 – 2005 LSAC Law School Admissions Test: 97th Percentile

2003 Guelph Grotto Lead Rock Climbing Certification

E. Community Involvement & Volunteering

Dates Organization Position

May 2006 – August 2006 HIV Projekt Belize Volunteer

September 2003 – December 2004

Graduate Students Association, University of Guelph

Elected Departmental Representative

September 2003 – August 2004

University of Guelph Senate Elected College

Representative

F. Scholarships, Fellowships and Awards

Dates Agency Value ($) Scholarship / Award

09/10 – 08/2012

09/2009 – 08/2010

09/2009 – 08/2010

NSERC

University of Toronto

Government of Ontario

80, 000

200

15, 000

Post-Doctoral Research Fellowship

Institutional ‘Best Paper” Prize.

Ontario Graduate Scholarship

132

Dates Agency Value ($) Scholarship / Award

09/2006 – 08/2009

NSERC 63, 000 NSERC PGS D

09/2006 – 08/2007 University of Toronto 5000 Graduate Research Scholarship

09/2005 – 08/2006

Ontario Graduate

Scholarship 15, 000 Ontario Graduate Scholarship

(declined)

09/2003 – 08/2004

Ontario Veterinary

College, University of Guelph

N/A Recognition of Contribution to the University Community

09/2003 – 08/2004

Ontario Veterinary


500 Excellence in Research Communication

09/2002 – 08/2003

Ontario Veterinary


2000 Graduate Research Scholarship

05/2002 – 08/2002 NSERC 6000 Undergraduate Student Research

Award

05/2001 – 08/2001

NSERC 6000 Undergraduate Student Research Award

G. Publications 16. Perrault SD, Chan WCW. In Vivo Assembly of Nanoparticle Components for Improved Tumour

Targeting. Submitted, Proceedings of the National Academy of Sciences, USA. January 2009.

15. Perrault SD, Chan WCW. Synthesis and Surface Modification of Highly Monodispersed, Spherical Gold Nanoparticles of 50-200 nm. J Am Chem Soc. 2009; 131(47):17042-3.

14. Perrault SD, Walkey C, Jennings T, Fischer HC, Chan WCW. Mediating Tumour Targeting Efficiency of Nanoparticles Through Design. Nano Letters. 2009; 9(5):1909-1915.

13. Chou, LYT, Fischer, HC, Perrault SD, Chan WCW. Visualizing Quantum Dots in Biological Samples Using Silver Staining. Analytical Chemistry, 2009; 81(11):4560-4565.

12. Wilcox JT, Semple E, Gartley C, Brisson BA, Perrault SD, Villagomez DA, Tayade C, Becker S, Lanza R, Betts DH. Characterization of Canine Embryonic Stem Cell Lines Derived from Different Niche Microenvironments. Stem Cell Dev. 2009 (Epub ahead of print).

133

11. Perrault SD, Hajek J, Zhong K, Graham C, Owino SO, Sichangi M, Smith G, Shi YP, Moore JM, Kain KC. HIV Co-Infection Increases Placental Parasite Burden and Antenatal Malaria Transmission Risk in Western Kenya. American Journal of Tropical Medicine and Hygiene, Jan 2009; 80(1):119-25.

10. Klostranec JM, Xiang Q, Farcas GA, Lee JA, Rhee A, Lafferty EI, Perrault SD, Kain KC, Chan WC. Convergence of quantum dot barcodes with microfluidics and signal processing for multiplexed high-throughput infectious disease diagnostics. Nano Letters. 2007; 7(9): 2812-8.

9. Jeon BG, Coppola G, Perrault SD, Rho GJ, Betts DH, King WA. S-adenosylhomocysteine treatment of adult female fibroblasts alters X-chromosome inactivation and improves in vitro embryo development after somatic cell nuclear transfer. Reproduction 2008; 135(6): 815-28.

8. Betts DH, Perrault SD, King WA. Low oxygen delays fibroblast senescence despite shorter telomeres. Biogerontology, 2008; 9(1): 19-31.

7. Alexander B, Coppola G, Perrault SD, Peura TT, Betts DH, King WA. Telomere length status of somatic cell sheep clones and their offspring. Mol Reprod Dev. 2007; 74(12): 1525-37.

6. Ortegon H, Betts DH, Lin L, Coppola G, Perrault SD, Blondin P, King WA. Genomic stability and physiological assessments of live offspring sired by a bull clone, Starbuck II. Theriogenology. 2007; 67(1):116-26.

5. Mastromonaco GF, Perrault SD, Betts DH, King WA. Role of chromosome stability and telomere length in the production of viable cell lines for somatic cell nuclear transfer.BMC Dev Biol. 2006; 6:41.

4. Betts DH, Perrault SD, Harrington L, King WA. Quantitative analysis of telomerase activity and telomere length in domestic animal clones. Methods Mol Biol. 2006;325:149-80.

3. King WA, Coppola G, Alexander B, Mastromonaco G, Perrault SD, Nino MI, Pinton A, Joudrey EM, Betts DH. The Impact of Chromosomal Alteration on Embryo Development. Theriogenology 2006; 65, 166 - 177.

2. Perrault SD Betts DH Global gene expression response to telomerase in bovine adrenal cortical cells. Biochemical and Biophysical Research Communications. 2005; 335(3): 925-936.

1. Betts, DH, Perrault SD, Petrik J, Favetta LA, Keefer CL, and King WA. Telomere length analysis in goat clones and their offspring. Molecular Reproduction and Development. 2005, 72(4): 461-470.

H. Technical Reports

2. Perrault SD, Baltzer HL (2006). Program Progress: Preliminary Findings & Summary Report on

Condom Vending Machines in Belize. Report prepared for Belizean Ministry of Health, Pan

American Social Marketing Organization, and HIV Projekt-Belize (volunteer position).

134

1. Betts DH, Perrault SD, Lin L, Kroetsch T, Bousquet D (2003). Normal telomere lengths of

spermatozoa from bull clone Starbuck II. Technical report prepared for L'Alliance Boviteq inc.

I. Book Chapters

1. Betts DH, Perrault SD, Harrington L, King WA (2004). Quantitative Analysis of Telomerase Activity

and Telomere Length in Domestic Animal Clones, in “Methods in Molecular Medicine: Nuclear

Reprogramming”, edited by Steve Pells, Humana Press, USA.

J. Conference Presentations, Posters & Proceedings

13. Perrault SD, Chan WCW (2010). A Systematic Mapping of Nanoparticle Design and Tumour

Accumulation for Improved Targeting. Oral Presentation, American Chemical Society National

Meeting, San Francisco.

12. Perrault SD, Chan WCW (2010). Nanotechnology and Medicine: Big Solutions From Small

Devices. Invited Speaker, Department of Physiology and Pharmacology, University of Western

Ontario.

11. Etame AB, Perrault SD, Smith C, Chan WCW, Rutka JT (2009). Nanoparticle Transport Into the

Central Nervous System. Neuro-Oncology 11(5): 578.

10. Perrault SD, Walkey C, Travis J, Fischer HC, Chan WCW (2009). Mapping the Tumour Targeting

Behaviour of Sub-100 nm Nanoparticles. Oral presentation, Biomedical Engineering Society

Annual Meeting, Pittsburg.

9. Perrault SD, Chan WCW (2009). Optimized Tumour Targeting of Nanoparticle-Mediated

Fluorescent Contrast Agents. Poster, World Molecular Imaging Conference, Montreal.

8. Perrault SD, Semple E, Betts DH (2005). Relative Quantification on the LightCycler: Quantification

of Oct4 in a Putative Canine Embryonic Stem Cell Line. Invited presentation: Roche Applied

Molecular Sciences Lightcycler User General Meeting, Montreal, Quebec.

7. Jeon B, Perrault SD, Rho G, Betts DH, King WA (2006). Developmental potential and

reprogramming efficiency of bovine embryos cloned with adult fibroblasts treated by a

135

demethylating agent, S-adenosyl-homocysteine. International Embryo Transfer Society. Orlando,

Florida.

6. Ortegon H, Lin L, Coppola G, Perrault SD, Kroetsch T, Betts DH, King WA, Bousquet D, Blondin

P(2006). Assessment of live offspring sired by a somatic cell nuclear transfer (SCNT) bull,

Starbuck II. International Embryo Transfer Society. Orlanda, Florida.

5. Perrault SD, King WA, Betts DH (2005). Real-Time Assays for Assessment of Telomere Length

and Telomerase Activity From Single Samples On the LightCycler. Poster, Canadian Federation of

Biological Societies: Second Northern Lights Conference. Guelph, Ontario.

4. Semple E, Baqir S., Gartley C, Brisson B, Perrault, SD, Cotton A, Beauchamp C, Betts DH (2005).

Isolation, Culture and Characterization of Putative Canine Embryonic Stem Cells. International

Society for Stem Cell Research. San Francisco, CA.

3. Perrault, SD, Hornsby PJ., Betts DH. (2004). Global gene expression response to telomerase in

bovine adrenal cortical cells. Poster, The Role of Telomeres and Telomerase in Cancer:

American Association of Cancer Research. San Francisco, California.

2. Perrault, SD. and Betts DH. (2002). Assessment of telomere length in cloned goats and their

offspring. Oral Presenation, Southern Ontario Reproductive Biology Conference. Toronto, Ontario.

1. Perrault, SD, Betts DH. (2002). Assessment of telomere length in cloned goats and their offspring.

Oral Presentation, 3rd Canadian Telomere Workshop. Vancouver, British Columbia

K. Teaching Experience

Dates Course & Responsibilities Total Hours

2002

Introductory Genetics, University of Guelph Leading bi-weekly teaching seminars

60 hours

2002-2004

Developmental Biology, University of Guelph Lab course instructor

20 hours

2002

Physiology of Aging, University of Guelph Overseeing students projects and presentations

40 hours

136

Dates Course & Responsibilities Total Hours

2006 - 2008

Biomedical Engineering Instrumentation & Technology Course administration

40 hours

137

Bibliography

1. Statistics Canada. Leading Causes of Death in Canada, 2005. 84-215-x2009000

(2009).

2. Heron, M. P. et al. Deaths: Final data for 2006. National Vital Statistics Reports 57

(2009).

3. Jain, R. K. Delivery of molecular and cellular medicine to solid tumors. Adv. Drug

Deliv. Rev. 46, 149-168 (2000).

4. Davis, M. E., Chen, Z. (. & Shin, D. M. Nanoparticle therapeutics: an emerging

treatment modality for cancer. Nature Reviews Drug Discovery 7, 771-782 (2008).

5. Heath, J. R. & Davis, M. E. Nanotechnology and cancer. Annu. Rev. Med. 59, 251-

265 (2008).

6. Weissleder, R. & Pittet, M. J. Imaging in the era of molecular oncology. Nature 452,

580-589 (2008).

7. Sanhai, W. R., Sakamoto, J. H., Canady, R. & Ferrari, M. Seven challenges for

nanomedicine. Nature Nanotechnology 3, 242-244 (2008).

8. Niemeyer, C. M. & Mirkin, C. A. Nanobiotechnology. (2005).

9. Lee, K. -. & El-Sayed, M. A. Gold and silver nanoparticles in sensing and imaging:

Sensitivity of plasmon response to size, shape, and metal composition. J Phys Chem B 110, 19220-19225 (2006).

10. West, J. L. & Halas, N. J. Engineered nanomaterials for biophotonics applications:

improving sensing, imaging, and therapeutics. Annu. Rev. Biomed. Eng. 5, 285-292

(2003).

138

11. Chan, W. C. W. & Nie, S. M. Quantum dot bioconjugates for ultrasensitive

nonisotopic detection. Science 281, 2016-2018 (1998).

12. Narayanan, R. & El-Sayed, M. A. Catalysis with transition metal nanoparticles in

colloidal solution: Nanoparticle shape dependence and stability. J Phys Chem B 109,

12663-12676 (2005).

13. Hermanson, G. T. Bioconjugate Techniques. (2008).

14. Lee, K. -., Kim, E. -., Mirkin, C. A. & Wolinsky, S. M. The use of nanoarrays for

highly sensitive and selective detection of human immunodeficiency virus type 1 in

plasma. Nano Lett. 4, 1869-1872 (2004).

15. Tasciotti, E. et al. Mesoporous silicon particles as a multistage delivery system for

imaging and therapeutic applications. Nature Nanotechnology 3, 151-157 (2008).

16. Niemeyer, C. M. Self-assembled nanostructures based on DNA: towards the

development of nanobiotechnology. Curr. Opin. Chem. Biol. 4, 609-618 (2000).

17. Aldaye, F. A., Palmer, A. L. & Sleiman, H. F. Assembling materials with DNA as the

guide. Science 321, 1795-1799 (2008).

18. Aldaye, F. A. et al. Modular construction of DNA nanotubes of tunable geometry and

single- or double-stranded character. Nature Nanotechnology 4, 349-352 (2009).

19. Walkey, C., Sykes, E. A. & Chan, W. C. Application of semiconductor and metal

nanostructures in biology and medicine. Hematology Am Soc Hematol Educ Program,

701-707 (2009).

20. Gentleman, D. J. & Chan, W. C. W. A systematic nomenclature for codifying

engineered nanostructures. Small 5, 426-431 (2009).

21. Paciotti, G. F. et al. Colloidal gold: A novel nanoparticle vector for tumor directed

drug delivery. Drug Deliv. 11, 169-183 (2004).

139

22. Qian, X. M. et al. In vivo tumor targeting and spectroscopic detection with surface-

enhanced Raman nanoparticle tags. Nat. Biotechnol. 26, 83-90 (2008).

23. Chithrani, B. D., Ghazani, A. A. & Chan, W. C. W. Determining the size and shape

dependence of gold nanoparticle uptake into mammalian cells. Nano Lett. 6, 662-668

(2006).

24. Chithrani, B. D. & Chan, W. C. W. Elucidating the mechanism of cellular uptake and

removal of protein-coated gold nanoparticles of different sizes and shapes. Nano Lett. 7, 1542-1550 (2007).

25. Jiang, W., Kim, B. Y. S., Rutka, J. T. & Chan, W. C. W. Nanoparticle-mediated

cellular response is size-dependent. Nature Nanotechnology 3, 145-150 (2008).

26. Hauck, T. S., Jennings, T. L., Yatsenko, T., Kumaradas, J. C. & Chan, W. C. W.

Enhancing the toxicity of cancer chemotherapeutics with gold nanorod hyperthermia.

Adv Mater 20, 3832-3838 (2008).

27. Panagi, Z. et al. Effect of dose on the biodistribution and pharmacokinetics of PLGA

and PLGA-mPEG nanoparticles. Int. J. Pharm. 221, 143-152 (2001).

28. Farokhzad, O. C. et al. Targeted nanoparticle-aptamer bioconjugates for cancer

chemotherapy in vivo. Proc. Natl. Acad. Sci. U. S. A. 103, 6315-6320 (2006).

29. Yu, H., Gibbons, P. C., Kelton, K. F. & Buhro, W. E. Heterogeneous seeded growth:

A potentially general synthesis of monodisperse metallic nanoparticles. J. Am. Chem.

Soc. 123, 9198-9199 (2001).

30. Jana, N. R., Gearheart, L. & Murphy, C. J. Wet chemical synthesis of high aspect

ratio cylindrical gold nanorods. J Phys Chem B 105, 4065-4067 (2001).

31. Jana, N. R., Gearheart, L. & Murphy, C. J. Evidence for seed-mediated nucleation in

the chemical reduction of gold salts to gold nanoparticles. Chemistry of Materials 13,

2313-2322 (2001).

140

32. Ji, X. et al. Size control of gold nanocrystals in citrate reduction: The third role of

citrate. J. Am. Chem. Soc. 129, 13939-13948 (2007).

33. Dhar, S., Gu, F. X., Langer, R., Farokhzad, O. C. & Lippard, S. J. Targeted delivery

of cisplatin to prostate cancer cells by aptamer functionalized Pt(IV) prodrug-PLGA-PEG

nanoparticles. Proc. Natl. Acad. Sci. U. S. A. 105, 17356-17361 (2008).

34. Park, J. W. et al. Anti-HER2 immunoliposomes: Enhanced efficacy attributable to

targeted delivery. Clinical Cancer Research 8, 1172-1181 (2002).

35. Gao, X. H., Cui, Y. Y., Levenson, R. M., Chung, L. W. K. & Nie, S. M. In vivo cancer

targeting and imaging with semiconductor quantum dots. Nat. Biotechnol. 22, 969-976

(2004).

36. Fang, C. et al. In vivo tumor targeting of tumor necrosis factor-alpha-loaded stealth

nanoparticles: Effect of MePEG molecular weight and particle size. European Journal of

Pharmaceutical Sciences 27, 27-36 (2006).

37. Avgoustakis, K. et al. Effect of copolymer composition on the physicochemical

characteristics, in vitro stability, and biodistribution of PLGA-mPEG nanoparticles. Int. J.

Pharm. 259, 115-127 (2003).

38. Elghanian, R., Storhoff, J. J., Mucic, R. C., Letsinger, R. L. & Mirkin, C. A. Selective

colorimetric detection of polynucleotides based on the distance-dependent optical

properties of gold nanoparticles. Science 277, 1078-1081 (1997).

39. 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, 55-75 (1951).

40. Frens, G. Controlled Nucleation for Regulation of Particle-Size in Monodisperse

Gold Suspensions. Nature-Physical Science 241, 20-22 (1973).

41. Frens, G. Particle size and sol stability in metal colloids. Kolloid-Z. u. Z. Polymere 250, 736-741 (1972).

141

42. Sau, T. K. & Murphy, C. J. Room temperature, high-yield synthesis of multiple

shapes of gold nanoparticles in aqueous solution. J. Am. Chem. Soc. 126, 8648-8649

(2004).

43. 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).

44. Storhoff, J. J. & Mirkin, C. A. Programmed materials synthesis with DNA. Chem.

Rev. 99, 1849-1862 (1999).

45. Dubertret, B., Calame, M. & Libchaber, A. J. Single-mismatch detection using gold-

quenched fluorescent oligonucleotides. Nat. Biotechnol. 19, 365-370 (2001).

46. Jin, R. C., Wu, G. S., Li, Z., Mirkin, C. A. & Schatz, G. C. What controls the melting

properties of DNA-linked gold nanoparticle assemblies? J. Am. Chem. Soc. 125, 1643-

1654 (2003).

47. Rosi, N. L. et al. Oligonucleotide-modified gold nanoparticles for intracellular gene

regulation. Science 312, 1027-1030 (2006).

48. Massich, M. D. et al. Regulating immune response using polyvalent nucleic acid-

gold nanoparticle conjugates. Molecular Pharmaceutics 6, 1934-1940 (2009).

49. Prigodich, A. E. et al. Nano-flares for mRNA Regulation and Detection. ACS Nano 3, 2147-2152 (2009).

50. Zheng, D., Seferos, D. S., Giljohann, D. A., Patel, P. C. & Mirkin, C. A. Aptamer

nano-flares for molecular detection in living cells. Nano Letters 9, 3258-3261 (2009).

51. Haes, A. J. & Van Duyne, R. P. A nanoscale optical blosensor: Sensitivity and

selectivity of an approach based on the localized surface plasmon resonance

spectroscopy of triangular silver nanoparticles. J. Am. Chem. Soc. 124, 10596-10604

(2002).

142

52. Cao, Y. C., Jin, R. & Mirkin, C. A. Nanoparticles with Raman spectroscopic

fingerprints for DNA and RNA detection. Science 297, 1536-1540 (2002).

53. Nie, S. M. & Emery, S. R. Probing single molecules and single nanoparticles by

surface-enhanced Raman scattering. Science 275, 1102-1106 (1997).

54. Nikoobakht, B. & El-Sayed, M. A. Preparation and growth mechanism of gold

nanorods (NRs) using seed-mediated growth method. Chemistry of Materials 15, 1957-

1962 (2003).

55. Loo, C., Lowery, A., Halas, N. J., West, J. & Drezek, R. Immunotargeted nanoshells

for integrated cancer imaging and therapy. Nano Letters 5, 709-711 (2005).

56. Park, J. -. et al. Cooperative nanomaterial system to sensitize, target, and treat

tumors. Proc. Natl. Acad. Sci. U. S. A. 107, 981-986 (2010).

57. Park, J. -. et al. Cooperative nanoparticles for tumor detection and photothermally

triggered drug delivery. Adv Mater 22, 880-885 (2010).

58. Hirsch, L. R. et al. Nanoshell-mediated near-infrared thermal therapy of tumors

under magnetic resonance guidance. Proc. Natl. Acad. Sci. U. S. A. 100, 13549-13554

(2003).

59. Melancon, M. P. et al. In vitro and in vivo targeting of hollow gold nanoshells

directed at epidermal growth factor receptor for photothermal ablation therapy.

Molecular Cancer Therapeutics 7, 1730-1739 (2008).

60. Ghazani, A. A. et al. High throughput quantification of protein expression of cancer

antigens in tissue microarray using quantum dot nanocrystals. Nano Lett. 6, 2881-2886

(2006).

61. Klostranec, J. M. et al. Convergence of quantum dot barcodes with microfluidics and

signal processing for multiplexed high-throughput infectious disease diagnostics. Nano

Lett. 7, 2812-2818 (2007).

143

62. Gao, X. H. et al. In vivo molecular and cellular imaging with quantum dots. Curr.

Opin. Biotechnol. 16, 63-72 (2005).

63. Huang, X. H., El-Sayed, I. H., Qian, W. & El-Sayed, M. A. Cancer cell imaging and

photothermal therapy in the near-infrared region by using gold nanorods. J. Am. Chem.

Soc. 128, 2115-2120 (2006).

64. Hatakeyama, H. et al. Tumor targeting of doxorubicin by anti-MT1-MMP antibody-

modified PEG liposomes. Int. J. Pharm. 342, 194-200 (2007).

65. Li, X., Ding, L., Xu, Y., Wang, Y. & Ping, Q. Targeted delivery of doxorubicin using

stealth liposomes modified with transferrin. Int. J. Pharm. 373, 116-123 (2009).

66. WHO. Cancer Control: Knowledge into Action. WHO Guide for Effective

Programmes. Early Detection. (2007).

67. Jain, R. K. Transport of molecules, particles, and cells in solid tumors. Annu. Rev.

Biomed. Eng. 1, 241-263 (1999).

68. Devalapally, H., Duan, Z. F., Seiden, M. V. & Amiji, M. M. Modulation of drug

resistance in ovarian adenocarcinoma by enhancing intracellular ceramide using

tamoxifen-loaded biodegradable polymeric nanoparticles. Clinical Cancer Research 14,

3193-3203 (2008).

69. Li, J. et al. The Enhancement Effect of Gold Nanoparticles in Drug Delivery and as

Biomarkers of Drug-Resistant Cancer Cells. ChemMedChem 2, 374-378 (2007).

70. Feldheim, D. L. & Foss, C. a. Metal Nanoparticles: Synthesis, Characterization, and

Application. (2002).

71. Faraday, M. Experimental Relations of Gold (and other metals) to Light.

Philosophical Transactions of the Royal Society 147, 145 (1857).

72. Brust, M., Walker, M., Bethell, D., Schiffrin, D. J. & Whyman, R. Synthesis of thiol-

derivatised gold nanoparticles in a two-phase liquid-liquid system. Journal of the

Chemical Society, Chemical Communications, 801-802 (1994).

144

73. McFarland, A. D. & Van Duyne, R. P. Single silver nanoparticles as real-time optical

sensors with zeptomole sensitivity. Nano Letters 3, 1057-1062 (2003).

74. Goel, R., Shah, N., Visaria, R., Paciotti, G. F. & Bischof, J. C. Biodistribution of TNF-

alpha-coated gold nanoparticles in an in vivo model system. Nanomed 4, 401-410

(2009).

75. Liu, Y. L. et al. Synthesis, stability, and cellular internalization of gold nanoparticles

containing mixed peptide-poly(ethylene glycol) monolayers. Anal. Chem. 79, 2221-2229

(2007).

76. Cederquist, K. B. & Keating, C. D. Curvature effects in DNA:Au nanoparticle

conjugates. ACS Nano 3, 256-260 (2009).

77. Xie, H. et al. Critical flocculation concentrations, binding isotherms, and ligand

exchange properties of peptide-modified gold nanoparticles studied by UV-visible,

fluorescence, and time-correlated single photon counting spectroscopies. Anal. Chem. 75, 5797-5805 (2003).

78. Brewer, S. H., Glomm, W. R., Johnson, M. C., Knag, M. K. & Franzen, S. Probing

BSA binding to citrate-coated gold nanoparticles and surfaces. Langmuir 21, 9303-9307

(2005).

79. Cerruti, M. et al. Poly(ethylene glycol) monolayer formation and stability on gold and

silicon nitride substrates. Langmuir 24, 10646-10653 (2008).

80. Klibanov, A. L., Maruyama, K., Torchilin, V. P. & Huang, L. Amphipathic

polyethyleneglycols effectively prolong the circulation time of liposomes. FEBS Lett. 268, 235-237 (1990).

81. Dunn, S. E. et al. In vitro cell interaction and in vivo biodistribution of poly(lactide-co-

glycolide) nanospheres surface modified by poloxamer and poloxamine copolymers. J.

Controlled Release 44, 65-76 (1997).

145

82. Peracchia, M. T., Vauthier, C., Passirani, C., Couvreur, P. & Labarre, D.

Complement consumption by poly(ethylene glycol) in different conformations chemically

coupled to poly(isobutyl 2-cyanoacrylate) nanoparticles. Life Sci. 61, 749-761 (1997).

83. Mosqueira, V. C. et al. Biodistribution of long-circulating PEG-grafted nanocapsules

in mice: effects of PEG chain length and density. Pharm. Res. 18, 1411-1419 (2001).

84. Hurst, S. J., Lytton-Jean, A. K. R. & Mirkin, C. A. Maximizing DNA loading on a

range of gold nanoparticle sizes. Anal. Chem. 78, 8313-8318 (2006).

85. Xu, S. et al. In situ studies of thiol self-assembly on gold from solution using atomic

force microscopy. J. Chem. Phys. 108, 5002-5012 (1998).

86. Chechik, P. & Stirling, C. J. M. in 551-640, 2003).

87. Dunn, S. E., Brindley, A., Davis, S. S., Davies, M. C. & Illum, L. Polystyrene-

Poly(ethylene Glycol) (Ps-Peg2000) Particles as Model Systems for Site-Specific Drug-

Delivery .2. the Effect of Peg Surface-Density on the In-Vitro Cell-Interaction and In-Vivo

Biodistribution. Pharm. Res. 11, 1016-1022 (1994).

88. Lundqvist, M. et al. Nanoparticle size and surface properties determine the protein

corona with possible implications for biological impacts. Proc. Natl. Acad. Sci. U. S. A. 105, 14265-14270 (2008).

89. Hobbs, S. K. et al. Regulation of transport pathways in tumor vessels: Role of tumor

type and microenvironment. Proc. Natl. Acad. Sci. U. S. A. 95, 4607-4612 (1998).

90. Beesley, J. E. in Methods in Molecular Biology (ed Manson, M.) 163 (The Humana

Press, Inc, Totowa, NJ, 1992).

91. Henglein, A. Formation and absorption spectrum of copper nanoparticles from the

radiolytic reduction of Cu(CN)2-. J Phys Chem B 104, 1206-1211 (2000).

92. Henglein, A. & Giersig, M. Formation of colloidal silver nanoparticles: Capping

action of citrate. J Phys Chem B 103, 9533-9539 (1999).

146

93. Shirtcliffe, N., Nickel, U. & Schneider, S. Reproducible preparation of silver sols with

small particle size using borohydride reduction: For use as nuclei for preparation of

larger particles. J. Colloid Interface Sci. 211, 122-129 (1999).

94. Henglein, A. & Meisel, D. Radiolytic control of the size of colloidal gold

nanoparticles. Langmuir 14, 7392-7396 (1998).

95. Teranishi, T. & Miyake, M. Size Control of Palladium Nanoparticles and Their

Crystal Structures. Chem. Mater. 10, 594-600 (1998).

96. Teranishi, T., Hosoe, M., Tanaka, T. & Miyake, M. Size control of monodispersed Pt

nanoparticles and their 2D organization by electrophoretic deposition. J Phys Chem B 103, 3818-3827 (1999).

97. Brown, K. R., Walter, D. G. & Natan, M. J. Seeding of colloidal Au nanoparticle

solutions. 2. Improved control of particle size and shape. Chemistry of Materials 12,

306-313 (2000).

98. Drucker, C. et al. Literatur. Fresenius, Zeitschrift f. anal. Chemie 67, 398-401

(1925).

99. Niu, J., Zhu, T. & Liu, Z. One-step seed-mediated growth of 30-150 nm

quasispherical gold nanoparticles with 2-mercaptosuccinic acid as a new reducing

agent. Nanotechnology 18 (2007).

100. Gentry, S. T., Fredericks, S. J. & Krchnavek, R. Controlled Particle Growth of

Silver Sols through the Use of Hydroquinone as a Selective Reducing Agent. Langmuir 25, 2613-2621 (2009).

101. Mostafavi, M., Marignier, J. L., Amblard, J. & Belloni, J. Nucleation Dynamics of

Silver Aggregates Simulation of Photographic Development Processes. Radiat. Phys.

Chem. 34, 605-617 (1989).

102. Linnert, T., Mulvaney, P., Henglein, A. & Weller, H. Long-Lived Nonmetallic Silver

Clusters in Aqueous-Solution - Preparation and Photolysis. J. Am. Chem. Soc. 112,

4657-4664 (1990).

147

103. Wang, S., Qian, K., Bi, X. & Huang, W. Influence of Speciation of Aqueous HAuCl4

on the Synthesis, Structure, and Property of Au Colloids. Journal of Physical Chemistry

C 113, 6505-6510 (2009).

104. Gachard, E. et al. Radiation-induced and chemical formation of gold clusters. New

Journal of Chemistry 22, 1257-1265 (1998).

105. Akerman, M. E., Chan, W. C. W., Laakkonen, P., Bhatia, S. N. & Ruoslahti, E.

Nanocrystal targeting in vivo. Proc. Natl. Acad. Sci. U. S. A. 99, 12617-12621 (2002).

106. Babincova, M., Sourivong, P., Leszczynska, D. & Babinec, P. Photodynamic

therapy of pigmented melanoma B16 using sterically stabilized fullerenosomes. Laser

Physics Letters 1, 476-478 (2004).

107. Matsumura, Y. & Maeda, H. A New Concept for Macromolecular Therapeutics in

Cancer-Chemotherapy - Mechanism of Tumoritropic Accumulation of Proteins and the

Antitumor Agent Smancs. Cancer Res. 46, 6387-6392 (1986).

108. Greish, K. Enhanced permeability and retention of macromolecular drugs in solid

tumors: a royal gate for targeted anticancer nanomedicines. J. Drug Target. 15, 457-464

(2007).

109. Maeda, H., Fang, J., Inutsuka, T. & Kitamoto, Y. Vascular permeability

enhancement in solid tumor: various factors, mechanisms involved and its implications.

Int. Immunopharmacol. 3, 319-328 (2003).

110. Peer, D. et al. Nanocarriers as an emerging platform for cancer therapy. Nat.

Nanotechnol 2, 751-760 (2007).

111. Schluep, T. et al. Pharmacokinetics and tumor dynamics of the nanoparticle IT-101

from PET imaging and tumor histological measurements. Proc. Natl. Acad. Sci. U. S. A. 106, 11394-11399 (2009).

112. Chung, T. H. et al. The effect of surface charge on the uptake and biological

function of mesoporous silica nanoparticles 3T3-L1 cells and human mesenchymal

stem cells. Biomaterials 28, 2959-2966 (2007).

148

113. Kang, B., Mackey, M. A. & El-Sayed, M. A. Nuclear targeting of gold nanoparticles

in cancer cells induces DNA damage, causing cytokinesis arrest and apoptosis. J. Am.

Chem. Soc. 132, 1517-1519 (2010).

114. Ding, L. H. et al. Molecular characterization of the cytotoxic mechanism of multiwall

carbon nanotubes and nano-onions on human skin fibroblast. Nano Letters 5, 2448-

2464 (2005).

115. Fujiwara, K. et al. Size-dependent toxicity of silica nano-particles to Chlorella

kessleri. Journal of Environmental Science and Health Part A-Toxic/hazardous

Substances & Environmental Engineering 43, 1167-1173 (2008).

116. Geng, Y. et al. Shape effects of filaments versus spherical particles in flow and

drug delivery. Nat. Nanotechnol 2, 249-255 (2007).

117. Shukla, R. et al. Biocompatibility of gold nanoparticles and their endocytotic fate

inside the cellular compartment: A microscopic overview. Langmuir 21, 10644-10654

(2005).

118. Chou, L. Y. T., Fischer, H. C., Perrault, S. D. & Chan, W. C. W. Visualizing

Quantum Dots in Biological Samples Using Silver Staining. Anal. Chem. 81, 4560-4565

(2009).

119. Jackson, H. et al. Quantum dots are phagocytized by macrophages and co-localize

with experimental glioma. J. Neurooncol. 87, 243-243 (2008).

120. Soo Choi, H. et al. Renal clearance of quantum dots. Nat. Biotechnol. 25, 1165-

1170 (2007).

121. Yuan, F. et al. Vascular-Permeability in a Human Tumor Xenograft - Molecular-

Size Dependence and Cutoff Size. Cancer Res. 55, 3752-3756 (1995).

122. Pluen, A., Netti, P. A., Jain, R. K. & Berk, D. A. Diffusion of macromolecules in

agarose gels: Comparison of linear and globular configurations. Biophys. J. 77, 542-552

(1999).

149

123. Dreher, M. R. et al. Tumor vascular permeability, accumulation, and penetration of

macromolecular drug carriers. J. Natl. Cancer Inst. 98, 335-344 (2006).

124. Schmidt, G. P., Reiser, M. F. & Baur-Melnyk, A. Whole-body MRI for the staging

and follow-up of patients with metastasis. Eur. J. Radiol. 70, 393-400 (2009).

125. Perrault, S. D., Walkey, C., Jennings, T., Fischer, H. C. & Chan, W. C. W.

Mediating Tumor Targeting Efficiency of Nanoparticles Through Design. Nano Letters 9,

1909-1915 (2009).

126. Larsen, E. K. et al. Size-Dependent Accumulation of PEGylated Silane-Coated

Magnetic Iron Oxide Nanoparticles in Murine Tumors. ACS Nano (2009).

127. Ayyagari, A. L. et al. Long-circulating liposomal contrast agents for magnetic

resonance imaging. Magn. Reson. Med. 55, 1023-1029 (2006).

128. Chen, W. et al. RGD peptide functionalized and reconstituted high-density

lipoprotein nanoparticles as a versatile and multimodal tumor targeting molecular

imaging probe. FASEB J. (2010).

129. Decuzzi, P., Pasqualini, R., Arap, W. & Ferrari, M. Intravascular Delivery of

Particulate Systems: Does Geometry Really Matter? Pharm. Res. 26, 235-243 (2009).

130. Tam, F., Goodrich, G. P., Johnson, B. R. & Halas, N. J. Plasmonic enhancement of

molecular fluorescence. Nano Letters 7, 496-501 (2007).

131. Cheng, D. & Xu, Q. Separation distance dependent fluorescence enhancement of

fluorescein isothiocyanate by silver nanoparticles. Chemical Communications, 248-250

(2007).

132. Pimm, M. V., Fells, H. F., Perkins, A. C. & Baldwin, R. W. I-131 and In-111 Labeled

Avidin and Streptavidin for Pre-Targeted Immunoscintigraphy with Biotinylated Anti-

Tumor Monoclonal-Antibody. Nucl. Med. Commun. 9, 931-941 (1988).

133. Vanosdol, W. W., Sung, C., Dedrick, R. L. & Weinstein, J. N. A Distributed

Pharmacokinetic Model of 2-Step Imaging and Treatment Protocols - Application to

150

Streptavidin-Conjugated Monoclonal-Antibodies and Radiolabeled Biotin. Journal of

Nuclear Medicine 34, 1552-1564 (1993).

134. Saga, T. et al. 2-Step Targeting of Experimental Lung Metastases with Biotinylated

Antibody and Radiolabeled Streptavidin. Cancer Res. 54, 2160-2165 (1994).

135. AlvarezDiez, T. M., Polihronis, J. & Reilly, R. M. Pretargeted tumour imaging with

streptavidin immunoconjugates of monoclonal antibody CC49 and In-111-DTPA-

biocytin. Nucl. Med. Biol. 23, 459-466 (1996).

136. Paganelli, G. et al. Antibody-guided three-step therapy for high grade glioma with

yttrium-90 biotin. Eur. J. Nucl. Med. 26, 348-357 (1999).

137. Sakahara, H. & Saga, T. Avidin-biotin system for delivery of diagnostic agents.

Adv. Drug Deliv. Rev. 37, 89-101 (1999).

138. Ntziachristos, V., Ripoll, J., Wang, L. H. V. & Weissleder, R. Looking and listening

to light: the evolution of whole-body photonic imaging. Nat. Biotechnol. 23, 313-320

(2005).

139. Gopinath, A. et al. Plasmonic nanogalaxies: Multiscale aperiodic arrays for

surface-enhanced Raman sensing. Nano Lett. 9, 3922-3929 (2009).

140. Rosi, N. L. & Mirkin, C. A. Nanostructures in biodiagnostics. Chem. Rev. 105,

1547-1562 (2005).

Documents

Gold Nanoparticles for Efficient Tumour Targeting ... · Representative TEM images, absorption spectra and DLS data of five gold nanoparticle sizes synthesized for testing blood pharmacokinetics