CHARACTERIZING and MANIPULATING BIOLOGICAL INTERACTIONS of VIRUSES by NEETU MEHEK GULATI Submitted in Partial Fulfillment Of

Home , Negative stain

CHARACTERIZING AND MANIPULATING BIOLOGICAL INTERACTIONS OF

VIRUSES

NEETU MEHEK GULATI

Submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy

Dissertation Advisors:

Dr. Phoebe L. Stewart and Dr. Nicole F. Steinmetz

Department of Pharmacology

CASE WESTERN RESERVE UNIVERSITY

January 2018 CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

Neetu Mehek Gulati candidate for the Doctor of Philosophy degree*.

(signed) Vera Moiseenkova-Bell (chair of the committee)

Phoebe L. Stewart

Nicole F. Steinmetz

Derek J. Taylor

Jun Qin

Vivien C. Yee

(date) September 8, 2017

*We also certify that written approval has been obtained for any proprietary material contained therein.

Table of Contents

List of Tables ...... vii

List of Figures ...... viii

Acknowledgements ...... xi

List of Abbreviations ...... xiv

Abstract ...... xxi

Chapter 1: Introduction ...... 1

1.1 Viruses – for good and bad ...... 1

1.2 Bioinspired shielding strategies for nanotechnology applications . 2

1.2.1 Abstract ...... 2

1.2.2 Introduction ...... 3

1.2.3 PEG and other synthetic polymeric shielding strategies ...... 4

1.2.4 Bio-inspired shielding strategies ...... 9

1.2.5 Beyond shielding – Immune editing ...... 21

1.2.6 Conclusions ...... 22

1.3 Tobacco mosaic virus as a nanotechnology platform ...... 25

1.3.1 Tobacco mosaic virus infection and production ...... 25

1.3.2 Tobacco mosaic virus structure ...... 25

1.3.3 Tobacco mosaic virus nanoparticles ...... 28

1.4 Human papillomavirus infection and neutralization ...... 28

1.4.1 Human papillomavirus and cancer ...... 28

1.4.2 Human papillomavirus vaccines ...... 29

1.4.3 Human papillomavirus structure ...... 30

1.4.4 Human papillomavirus cell entry mechanism and infection pathway ...... 35

1.4.5 Human papillomavirus and human alpha-defensin 5 ...... 37

1.5 Aims of dissertation ...... 40

Chapter 2: Structural characterization of SA-TMV ...... 41

2.1 Abstract ...... 41

2.2 Introduction ...... 42

2.3 Materials and Methods ...... 45

2.3.1 Virus propagation and purification ...... 45

2.3.2 TMV sCy5 labeling ...... 45

2.3.3 TMV conjugation ...... 46

2.3.4 SDS-PAGE analysis ...... 47

2.3.5 Western blot analysis ...... 47

2.3.6 Immuno-dot blots ...... 48

2.3.7 Negative stain transmission electron microscopy ...... 48

2.3.8 Cryo-electron microscopy and tomography ...... 48

2.3.9 Subtomogram averaging of SA ...... 49

2.3.10 Subtomogram averaging of SA-TMV segment ...... 50

2.3.11 Analysis of volume coverage ...... 50

2.3.12 Analysis of surface coverage ...... 51

2.4 Results and Discussion ...... 51

2.4.1 Particle preparation ...... 51

2.4.2 Imaging and processing of naked and SA-coated particles ...... 57

2.4.3 Quantification of TMV coverage ...... 66

2.5 Conclusions ...... 69

2.6 Acknowledgements ...... 70

2.7 Supplementary information ...... 71

Chapter 3: Immune recognition of SA-TMV ...... 72

3.1 Abstract ...... 72

3.2 Introduction ...... 73

3.3 Materials and Methods ...... 76

3.3.1 Virus propagation and purification ...... 76

3.3.2 TMV sCy5 labeling ...... 77

3.3.3 TMV external conjugation ...... 77

3.3.4 SDS-PAGE analysis ...... 78

3.3.5 Western blot analysis ...... 79

3.3.6 Negative stain transmission electron microscopy ...... 79

3.3.7 Dot blots ...... 80

3.3.8 In vivo studies and ELISA assays ...... 80

3.3.9 Confocal microscopy ...... 81

3.3.10 Lysosomal extraction experiments ...... 82

3.4 Results and Discussion ...... 83

3.4.1 Synthesis and characterization of SA-TMV constructs ...... 83

3.4.2 Effects of SA coverage and linker length on shielding TMV ...... 92

iii

3.4.3 Antibody response to multiple administrations of SA-TMV constructs ...... 97

3.4.4 Trafficking of SA-TMV nanoparticles upon cell uptake ...... 103

3.4.5 SA-TMV cleavage and degradation in the lysosome ...... 109

3.5 Conclusions ...... 112

3.6 Acknowledgements ...... 113

Chapter 4: cryoEM of HPV PsV with HD5 ...... 114

4.1 Introduction ...... 114

4.2 Materials and Methods ...... 119

4.2.1 Protein disorder prediction and modeling ...... 119

4.2.2 HPV16 pseudovirus production ...... 119

4.2.3 CryoEM grid preparation ...... 119

4.2.4 CryoEM imaging and data collection ...... 120

4.2.5 Particle picking and CTF correction ...... 120

4.2.6 3D structure determination and filtering ...... 120

4.3 Results and Discussion ...... 121

4.3.1 Sub-nanometer resolution structures of HPV16 and HPV16+HD5 by RELION ...... 121

4.3.2 Gaussian filtered HPV16 and HPV16+HD5 maps reveal additional density in the HD5 structure ...... 126

4.4 Conclusions ...... 135

Chapter 5: Conclusions and Future Directions ...... 137

5.1 Scope of work ...... 137

5.2 Viruses in nanomedicine: evading immune recognition of TMV nanoparticles by stealth-camouflage with SA ...... 138

5.2.1 Tobacco mosaic virus as a nanocarrier for drug delivery applications ...... 140

5.2.2 Novel SA self-camouflage platform ...... 143

5.2.3 Considerations beyond SA-TMV ...... 149

5.3. Viruses in human disease: characterization of HD5 neutralization of HPV infection ...... 152

5.3.1 Beyond the current vaccinations strategies and the need for more information ...... 154

5.3.2 HD5-based therapies for treatment of human papillomavirus .. 157

5.3.3 Beyond papillomavirus – broad neutralization of infectious pathogens ...... 157

5.4. Conclusions ...... 158

Appendix 1: Models for Adenovirus Neutralization by HD5 ...... 160

A1.1 Abstract ...... 160

A1.2. Acknowledgements ...... 165

Appendix 2: Shielding of SA on TMV nanoparticle ...... 166

Appendix 3: TEM of targeted PVX ...... 170

Appendix 4: TEM of silica-coated TMV rods and spheres ...... 172

Appendix 5: TEM of short TMV rods ...... 174

Appendix 6: TEM of serum-aggregated PVX nanoparticles ...... 175

Appendix 7: C shell script for preparing and submitting Frealign 7.07 scripts ...... 177

A3.1 Script for making Frealign submission scripts ...... 178

A3.2 Example template script ...... 189

Bibliography ...... 191

List of Tables

Table 3.1. Physical characteristics of SA coverage and PEG linkers of SA-TMV

constructs...... 90

vii

List of Figures

Figure 1.1. Uricase activity in plasma over 21 days in subjects with and without

PEG antibodies ...... 7

Figure 1.2. Antitumor effects of heparin nanoparticle (NP) formulation,

doxorubicin, and doxorubicin-loaded heparin-NP in mice model of

subcutaneous squamous cell carcinoma ...... 12

Figure 1.3. RBC-membranes for use as camouflage for nanoparticles ...... 15

Figure 1.4. Repeated administration of SA-coated TMV NPs does not produce an

immune response ...... 19

Figure 1.5. Map of polymorphic aggregates of TMV CPs based on varying pH

and ionic strength ...... 27

Figure 1.6. Known domains of the L2 minor capsid protein ...... 32

Figure 1.7. Identification of L2 density regions by comparison with a simulated L1

capsid ...... 34

Figure 1.8. Cell entry pathway of HPV after furin cleavage ...... 36

Figure 2.1. Production and characterization of SA-TMV ...... 55

Figure 2.2. Transmission electron microscopy of TMV and SA-TMV ...... 58

Figure 2.3. Cryo-electron tomography analysis of SA-TMV nanoparticles ...... 61

Figure 2.4. Subtomogram averaging analysis of SA orientation on the TMV

surface...... 63 viii

Figure 2.5. Subtomogram averaging analysis of SA-TMV segments ...... 65

Figure 2.6. Analysis of SA volume coverage of TMV rod ...... 67

Figure 2.S1. Cryo-electron tomography analysis of naked TMV nanoparticles .. 71

Figure 3.1. Cartoon depiction of production of four PEG-TMV and four SA-TMV

constructs, with varying PEG linker length and SA coverage quantity ...... 85

Figure 3.2. Characterization of SA-TMV constructs ...... 87

Figure 3.3. Antibody recognition of SA-TMV and PEG-TMV constructs ...... 94

Figure 3.4. Quantification of antibodies generated after repeated administration

of SA-TMV constructs ...... 98

Figure 3.5. Analysis of SA-TMV recognition by antibodies in plasma from

immunized mice ...... 102

Figure 3.6. Confocal microscopy of SA-TMV uptake by RAW 264.7 murine

macrophage cells ...... 104

Figure 3.7. Stability of Cy5-labeled SA-TMV in lysosomal extract (LE) over time

...... 111

Figure 4.1. Disorder profile of L2 protein of HPV16 by amino acid sequence .. 118

Figure 4.2. Representative cryo-electron micrographs of HPV16 (left) and

HPV16+HD5 (right) ...... 122

Figure 4.3. CryoEM reconstructions of HPV16 and HPV16+HD5 ...... 124

Figure 4.4. Plot showing the gold standard Fourier shell correlation (FSC) vs.

spatial frequency of the icosahedrally averaged reconstructions for HPV16

and HPV16+HD5 ...... 125

Figure 4.5. Gaussian filtered maps of HPV16 and HPV16+HD5 reconstructions

...... 127

Figure 4.6. Pores in the HPV16 capsid ...... 130

Figure 4.7. Central slabs of the HPV16 (light blue) and HPV16+HD5 (red) filtered

maps compared to the L1 pentamer atomic structure (green) ...... 132

Figure 5.1. Schematic of human serum albumin (HSA) transport in endothelial

cells showing FcRn receptor binding in endosomes at pH 6.0 leading to

recycling and release, or trafficking to a lysosome for degradation ...... 145

Figure 5.2. Schematic of the uptake of albumin-bound paclitaxel into tumors,

mediated by gp60 for transcytosis through the endothelium and the albumin-

binding protein SPARC within the interstitial tumor space ...... 148

Figure A1.1. Proposed binding mode for HD5 with the HAdV5 penton of the

defensin-sensitive chimera ...... 163

Figure A1.2. Proposed alternative binding mode for HD5 with HAdV12 penton 164

Figure A2.1. Sections of cryo-electron micrographs of A) bare TMV nanoparticles

and B) TMV nanoparticles conjugated to SA via a short (8-mer) PEG linker

...... 169

Acknowledgements

First and foremost, I would like to thank my advisors, Dr. Phoebe Stewart

and Dr. Nicole Steinmetz, for their guidance, encouragement, and support

throughout my graduate career. Phoebe welcomed me into her lab as a naïve

first year graduate student and introduced me to the world of cryo-electron

microscopy. Nicole welcomed me with open arms midway through my PhD and

expanded my horizons into new techniques and skills. Their time, mentorship,

and training have shaped my graduate career and made me the scientist I am

today.

I would also like to thank the members of my Thesis Committee for their

support encouragement during these past few years: Dr. Vera Moiseenkova-Bell,

Dr. Jun Qin, Dr. Vivien Yee, and Dr. Derek Taylor. Their breadth and depth of

knowledge has been invaluable during the progress of my Ph.D.

To my lab members, past and present, I am thankful for the help,

comradery, and shared experiences. I would like to thank Dr. Justin Flatt, Dr.

Seth Villareal, and Dr. Tara Fox for welcoming me when I first joined Phoebe’s

lab and helping me navigate research questions as well as departmental

requirements and providing me with a lab environment that encouraged and fostered my growth. I would also like to thank the current members of Phoebe’s lab for continuing to make the lab a place for learning and friendships throughout the ups and downs of graduate school: Sahil Gulati, Corey Emerson, and Linda

Thomas. I would like to thank the many members of Nicole’s lab over the years

for welcoming me and embracing me as one of their own: Dr. Sourabh Shukla,

Dr. Mike Bruckman, Dr. Andrzej Pitek, Dr. Patricia Lam, Dr. Brylee Tiu, Dr. Duc

Le, Dr. Andy Hu, Dr. Ravi Patel, Dr. Chao Wang, Dr. Hui Cai, Dr. Amy Wen, Dr.

Karin Lee, Dr. Anna Czapar, Abner Murray, Paul Chariou, Richard Lin, Brendan

Barton, Mike Jandzinski, Frank Veliz, Dan Kernan, Christina Franke, Bindi Patel, many visiting scholars, and a multitude of wonderful undergraduate students. In particular, I would like to thank Andrzej for supporting my research and being a wonderful collaborator throughout our shared time in the lab. I am also extremely grateful to Heather Holdaway and Dr. Sudheer Molugu at the Cleveland Center for Membrane Structural Biology for advising me on best practices with the ever- temperamental transmission electron microscopes I used during my graduate career.

I would also like to thank the Pharmacology Department as a whole, and in particular the administrative staff and the students, for supporting me and believing in me during graduate school. I would also like to thank Dr. Paul

MacDonald in the Graduate Education Office for encouraging me to follow my own path in graduate school and Dr. John Mieyal for being a sounding board whenever I met challenges in navigating through the program. I am grateful for the Pharmacology training grant funded by the National Institutes of Health, which supported me for two years of my research.

I would like to thank Jacqueline Wallat for her help with confocal microscopy, the people of the Small Animal Imaging Center for taking care of the animals used in my studies, and the electron microscopy consortium at Florida

State University for the use of their microscope. I would also like to thank my

xii

collaborators Dr. Jason Smith and Dr. Mayim Wiens for their expertise, samples,

and for inviting me to their laboratory to learn more about their research.

I am extremely grateful to the students of the various graduate student

governments I have been a part of, and to the administrators that supported us.

These organizations kept me sane by providing me with another outlet for my

energy and for teaching me that anything is possible with the right resources and

enough perseverance.

I would also like to thank my friends, near and far, for believing in me and

being there for me through this long and eventful journey. I appreciate the

encouragement in all my successes, and for the motivation to push through challenges. Though there are too many friends to name, I am eternally grateful for all they have done. I would especially like to thank Ken Farabaugh for his patience throughout the stresses of research, writing, and planning my future after graduate school.

Last but certainly not least, I want to thank my family for always believing in me throughout this journey. The demands of graduate school have not just affected me, but also the people I care about. Despite this, they have stood by my side and supported me, even when I did not know I needed it. They have taught me how important a good education is, and ensured I prioritized school above passing distractions. They have been my rock throughout this entire experience, and I would have never made it without them.

xiii

List of Abbreviations

2D two-dimensional

3D three-dimensional

A alanine

AAV adeno-associated virus

ABC accelerated blood clearance

AdV adenovirus

AIDS acquired immunodeficiency syndrome

AMG aminoguanidine

AP alkaline phosphatase

AsC ascorbic acid

BCIP 5-bromo-4-chloro-3-indolyl phosphate

BSA bovine serum albumin

CDC Centers for Disease Control and Prevention

CP coat protein

CPMV cowpea mosaic virus cryoEM cryo-electron microscopy cryo-EM cryo-electron microscopy cryoET cryo-electron tomography cryo-ET cryo-electron tomography

CTF contrast transfer function

CWRU Case Western Reserve University

D aspartic acid

xiv

DAPI 4',6-diamidino-2-phenylindole

DE Direct Electron

DMEM Dulbecco's Modified Eagle Medium

DMSO dimethyl sulfoxide

DNA deoxyribonucleic acid

DOTA 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid

DPBS Dulbecco’s phosphate buffered saline

E glutamic acid

EDC N-ethyl-N’-(3-dimethylaminopropyl)carbodiimide hydrochloride

EDTA ethylenediaminetetraacetic acid

EGFR epidermal growth factor receptor

ELISA enzyme-linked immunosorbent assay

ELP elastin-like peptide

EMAN Electron Micrograph Analysis

EMD Electron Microscopy Databank

Fab fragment antibody-binding

FcRn neonatal Fc receptor

FDA United States Food and Drug Administration

FP false positive

FSC fourier shell correlation fstep fine step

G glycine

GAG glycosaminoglycan

GCTF GPU-accelerated contrast transfer function program

Gd gadolinium

Glu glutamic acid

Gly glycine gp glycoprotein

H histidine

HA hyaluronic acid

HAdV human adenovirus

HAdV12 human adenovirus type 12

HAdV35 human adenovirus type 35

HAdV5 human adenovirus type 5

HD5 human alpha-defensin 5

HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid hipr high phase residual

HIV human immunodeficiency virus

HPC High Performance Computing Cluster

HPV human papillomavirus

HPV11 human papillomavirus type 11

HPV16 human papillomavirus type 16

HPV18 human papillomavirus type 18

HPV31 human papillomavirus type 31

HPV33 human papillomavirus type 33

xvi

HPV45 human papillomavirus type 45

HPV52 human papillomavirus type 52

HPV58 human papillomavirus type 58

HPV6 human papillomavirus type 6

HSA human serum albumin

HSPG heparan sulfate proteoglycan receptor

I isoleucine

IgG immunoglobulin G

IgM immunoglobulin M

K lysine

LAMP-1 lysosome-associated membrane glycoprotein-1

LE lysosomal extract

LLV leukolike vectors lopr low phase residual

LPS lipopolysaccharide

Lys lysine

MAL maleimide

MPS mononuclear phagocytic system

MRI magnetic resonance imaging

NAMD Nanoscale Molecular Dynamics

ND10 nuclear domain 10

NHS N-hydroxysuccinimide

NIH National Institutes of Health

xvii

NMR nuclear magnetic resonance

NP nanoparticle

P proline

PAGE polyacrylamide gel electrophoresis

PAMP pathogen-associated molecular pattern

PBS phosphate buffered saline

PDB Protein Databank

PEG polyethylene glycol

PNB polynorbornene

PNPP p-nitrophenyl phosphate

POX polyoxazoline

POZ polyoxazoline

Pro proline

PsV pseudovirus

PVX potato virus X

Q glutamine

R arginine

RBC red blood cell

RELION Regularized Likelihood Optimization

RF flory dimension

RGB red/green/blue

RGD arginylglycylaspartic acid

RNA ribonucleic acid

xviii

RT room temperature

RT-PCR reverse transcription polymerase chain reaction

S serine

SA serum albumin

SAT N-succinimidyl-S-acetylthioacetate sCy5 sulfo-Cy5

SDS sodium dodecyl sulfate

Si silica

SNP spherical nanoparticle

SPARC secreted protein acidic and rich in cysteine

T threonine

TBS tris-buffered saline

TEM transmission electron microscopy

TLR toll-like receptor

TMV tobacco mosaic virus

TMV-lys T158K tobacco mosaic virus

UCSF University of California San Francisco

V valine v/v volume by volume

Val valine

VLP virus-like particle

VNP viral nanoparticle vs versus

xix

W tryptophan w/v weight by volume

WB western blot

Y tyrosine

Characterizing and Manipulating Biological Interactions of Viruses

Abstract

NEETU M. GULATI

Viruses are both pathogenic and can be repurposed for novel roles in nanotechnology. Tobacco mosaic virus (TMV) plant viral nanoparticles are promising drug delivery agents. In this work TMV nanoparticles are stealth- coated with serum albumin (SA), the most abundant protein in human plasma.

SA-camouflaged nanoparticles successfully evade immune recognition in vitro.

Cryo-electron tomography of SA-conjugated TMV reveals that SA provides steric hindrance for TMV-specific antibodies and that SA adopts random orientations relative to the viral rod. Random orientations could be advantageous for preventing an immune response based on pathogen-associated molecular patterns (PAMPs). Studies on SA-TMV constructs with different SA coverage levels and linker lengths indicate that constructs with high SA coverage with short linkers are best for immune evasion. In vivo antibody production experiments reveal that administration of SA-TMV constructs results in the production of TMV- specific antibodies but not SA-specific antibodies. Nevertheless, intact SA- shielded nanoparticles are protected from recognition by these antibodies.

Confocal microscopy results suggest that after SA-TMV is taken up by

xxi

macrophages, the TMV and stealth-coating components are processed through different pathways: TMV is degraded in the lysosome while SA is recycled to the cell surface. These results highlight the potential for tuning the immune evasion properties of a viral-based nanoparticle. Numerous viruses can infect humans and cause disease and there are only a handful of anti-viral drugs currently available. Human papillomavirus (HPV) 16, together with HPV18, are the two most oncogenic HPV types and are thought to cause 70% of cervical cancer cases worldwide. HPV16 can be neutralized by human alpha-defensin 5 (HD5), an innate immune peptide. Cryo-electron microscopy structural studies reveal that HD5 stabilizes the HPV16 capsid and strengthens the interaction between the capsid and viral genome, thus presumably blocking separation of the capsid and the genome necessary for productive infection. Characterizing the neutralization mechanism of HD5 is significant for developing HD5-based HPV16 treatments, as no therapies currently exist for the treatment of HPV infections.

Overall this body of work reveals the importance of characterizing and manipulating the intricate interactions between viruses and biological factors encountered in their natural hosts and in nanotechnology applications.

xxii

Chapter 1: Introduction

1.1 Viruses – for good and bad

Viruses are pathogenic agents that cannot replicate on their own. Instead, they infect living cells of organisms and require cellular machinery for replication.

Viruses are diverse pathogens, with many shapes, sizes, and host organisms.

Many cause human disease, for example human immunodeficiency virus causes

AIDS, human papillomavirus (HPV) causes cervical cancer, and influenza virus causes the flu. Other viruses cannot replicate in humans, instead causing disease in plants or bacteria or other organisms. For example, tobacco mosaic virus (TMV), the first characterized virus in 1898, causes disease in tobacco plants, but is non-toxic and non-infectious in humans.

Because of the range of viruses, different viruses play different roles in biomedical research. As viruses have been designed by nature to deliver a specific cargo (their genome) to cells, they can also be utilized to deliver other cargo, such as drugs or imaging agents. However it is important to choose viruses for this purpose that will not cause disease on their own. To this end,

TMV and other plant viruses are ideal candidates for viral-based nanoparticle carriers. Nevertheless, plant viruses and other nanoparticles are foreign to the human body, and are rapidly cleared unless shielded from immune recognition.

Section 1.2 of the Introduction will cover shielding strategies for nanoparticles,

including TMV nanoparticles. Section 1.3 of the Introduction will focus on the use

of TMV for nanoparticle development.

While some viruses can be used for good, to treat diseases such as

cancer and cardiovascular disease, other viruses are disease-causing on their

own. In these situations, it is important to study viruses to understand the

mechanism of infection, and how to prevent it. Section 1.4 of the Introduction will

focus on one such disease-causing virus, HPV.

Structural biology techniques can be used to study virus structures, alone

or in the presence of ligands or other protein molecules. Two such techniques,

cryo-electron microscopy (cryoEM) and cryo-electron tomography (cryoET) techniques will be discussed in Section 1.5 of the Introduction.

1.2 Bioinspired shielding strategies for nanotechnology applications

The material in Section 1.2 is adapted from a review manuscript in

preparation for Biomaterials (expected submission September 2017): Gulati NM,

Stewart PL, Steinmetz NF. Bioinspired shielding strategies for nanotechnology

applications.

1.2.1 Abstract

Nanoparticle delivery systems offer advantages over free drugs, in that

they can deliver a high payload of molecules while limiting off-target side effects.

Modifying nanoparticle surfaces is often required to reduce clearance and

thereby increase circulation times. Many nanoparticles are coated with

polyethylene glycol (PEG) to prevent immune recognition and clearance.

However the prevalence of PEG-specific antibodies limit its utility in nanomedicine. This review highlights alternative bio-inspired nanoparticle shielding strategies, which may be more beneficial moving forward than PEG and other synthetic polymer coatings.

1.2.2 Introduction

Nanoparticles have become an exciting research topic in recent years, due to their potential for delivering large payloads while limiting off-target effects. They offer hope for new therapies and provide opportunities to revitalize drugs taken off the market due to toxicity. Nanoparticles come in many different shapes and sizes, and can be loaded with drugs or contrast agents as well as targeting ligands. One of the many advantages that nanoparticle engineering provides is the possibility of tuning the carrier to a specific biological problem. Nanoparticle research has rapidly grown in recent years, with over 20,000 papers published in

2016 alone. Despite this, many nanoparticle formulations fail to translate to the clinic.

Engineered nanoparticles face numerous biological barriers upon administration (Blanco et al., 2015). One of the first challenges is avoiding recognition and clearance by the immune system. Nanoparticles associate with naturally occurring proteins and other molecules (termed opsonization) leading to formation of a ‘protein corona’ which alters their in vivo properties (Kokkinopoulou et al., 2017; Lee et al., 2015b). This corona is more prominent for synthetic nanoparticles than proteinaceous viral-based nanocarriers, but in both cases the

corona can include immune proteins such as immunoglobulins and complement proteins (Docter et al., 2014; Pitek et al., 2016a). These proteins tag nanoparticles for clearance by phagocytic cells of the mononuclear phagocyte system (MPS), thereby preventing nanocarriers from ever reaching their target sites. Once enveloped with a protein corona, nanoparticles are sequestered in

MPS organs, including the liver, spleen, and kidney (Pallardy et al., 2017).

Nanoparticles can also trigger the adaptive immune system, leading to an increase in production of neutralizing antibodies, which can be especially challenging for nanotherapeutic or imaging strategies that require repeat administrations. Antibody production and the adaptive immune response lead to accelerated blood clearance (ABC) and increased accumulation in organs such as the liver, reducing the efficacy of nanoparticle formulations (Abu Lila et al.,

2013).

1.2.3 PEG and other synthetic polymeric shielding strategies

Nanoparticles are frequently PEGylated to avoid immune recognition and to provide better pharmacokinetic profiles. PEG is a flexible hydrophilic polymer that was originally considered to have little to no immunogenicity and few biological interactions, and therefore was expected to be advantageous as a biologic shielding agent (Abuchowski et al., 1977). By grafting PEG to the surface of a nanoparticle formulation, protein adsorption and antibody binding are decreased, as is uptake by phagocytotic cells. Reduced uptake occurs because the PEG layer forms a hydrophilic barrier on the surface of nanoparticles and blocks receptor interactions via steric hindrance (Drobek et al., 2005). The

effectiveness of PEG to increase circulation times has been demonstrated on

nanoparticles with a range of shapes, sizes, and composition (Bruckman et al.,

2014a; Hatakeyama et al., 2013; Jokerst et al., 2011; Lee et al., 2015a; Perry et

al., 2012). In fact, several clinically approved nanoparticle therapies such as

Doxil® include PEGylation. During the design of Doxil, adding PEG to

doxorubicin-containing liposomes was found to increase the circulating half-life

from approximately 10 min to over 40 hours (Northfelt et al., 1996). Nevertheless,

choosing the correct PEG formulation can be tricky. PEG polymers can have

different physical properties by varying characteristics such as chain length and

number of branch arms. The density of the PEG coating can alter its

conformation as well, allowing it to adopt either a mushroom-like globular conformation at low density or a more extended brush-like conformation at high

density. The effectiveness of PEG as a shielding agent is dependent on the

chosen physical characteristics (Jokerst et al., 2011; Lee et al., 2015a; Perry et

al., 2012), implying that optimization of the PEG shield is usually necessary. In practice nanoparticles have been optimized with various PEG formulations, for example 10 kDa molecular weight PEG best reduced clearance for chitosan/siRNA nanoparticles delivered intravenously, while 1 kDa molecular weight PEG was best for oral administration of prodrug-based micelles (Li et al.,

2015; Yang et al., 2017).

Despite the benefits of PEG shielding, there have been recent challenges.

While initial reports on PEG suggested low immunogenicity, more recent reports indicate a significant level of immunogenicity as PEG-specific antibodies have

been found in the general population. A recent study found that up to 25% of the

population have anti-PEG antibodies in 2012, up from 0.2% in 1984 (Garay et al.,

2012; Richter and Akerblom, 1984). This is likely due to the prevalence of PEG in

everyday products. Not only is PEG used as an anti-fouling agent in biomedical

applications, but it is also found in cosmetics and food products (Garay et al.,

2012). Development of anti-PEG antibodies can be especially detrimental for

nanoparticle formulations that require repeat administration. Newer nanoparticle

formulations that rely on PEG shielding have struggled in clinical trials mainly

because of the immunogenicity of PEG.

A Phase 1 clinical trial published in 2006 found that treatment of

hyperuricemia in gout patients by PEGylated uricase led to induction of PEG-

specific IgM and IgG antibodies in 5 of 13 patients (Ganson et al., 2006). PEG-

uricase was cleared in these subjects by 10 days post injection, whereas the 8

patients who did not show measurable antibodies against PEG had circulating

PEG-uricase 21 days post-injection (Figure 1.1). In 2007, Armstrong et al. found

that antibodies against PEG affected the efficacy of a PEGylated asparaginase

therapy for treatment of acute lymphoblastic leukemia patients (Armstrong et al.,

2007). Rapid clearance of PEG-asparaginase was observed for 15 patients, of

whom 12 had measurable antibodies against PEG. Since serum asparaginase

levels need to remain elevated for 21 days to achieve a response in acute

lymphoblastic leukemia, the authors concluded that rapid clearance by PEG-

specific antibodies could render the treatment ineffective. Other studies have also found that PEGylated nanoparticles are rapidly cleared upon repeat

administrations because of a PEG-specific immune response (Dams et al., 2000;

Hamad et al., 2008; Ishida et al., 2006). Therefore, in recent years, alternative strategies for nanoparticle shielding have been explored.

Figure 1.1. Uricase activity in plasma over 21 days in subjects with and without PEG antibodies. A) Subjects without PEG antibodies had “long circulating” uricase after a single injection as indicated by plasma uricase activity detectable for 21 days. B) Uricase activity among subjects with anti-PEG antibodies had “early elimination” of uricase. Reproduced with permission from

(Ganson et al., 2006).

While PEG is the most well characterized shielding polymer, other

hydrophilic synthetic polymers have been studied. These alternative polymers

are less commercially available and not used in as many products. Therefore the

risk of antibodies against them in the general population is small. A diverse group

of polymers with anti-fouling properties exists, each with their own advantages.

For example, polynorbornene (PNB) shielding has been shown to reduce

antibody recognition more successfully than PEGylation on icosahedral

nanoparticles (Lee et al., 2017). This reduced antibody recognition could be

because the PNB coating occupies a significantly smaller hydrodynamic volume

than a comparison PEG coating, as determined by small-angle neutron scattering. Cryo-electron microscopy was used in this study to reveal the nearly complete coverage of the viral surface with a compact PNB layer.

Another class of hydrophilic polymers, polyoxazolines (POX/POZ), have been shown to reduce antibody recognition and macrophage uptake of viral nanoparticles better than PEGylated nanoparticles while maintaining a similar pharmacokinetic profile (Bludau et al.). Other polymers provide benefits such as being environmentally responsive or biodegradable (Chen et al., 2015; Muller et al., 2017). Many other polymers exist that provide anti-fouling and shielding effects (Knop et al., 2010). However, the risk of antibody recognition remains with repeated use, as has been observed for PEG. Therefore, a promising alternative approach has been to develop shielding agents based on biologically relevant molecules, including lipids, carbohydrates, and proteins.

1.2.4 Bio-inspired shielding strategies

The in vivo environment is diverse, with a plethora of cells, proteins, and small molecules circulating at any given moment. Bio-inspired shielding strategies harness the body’s complexity, by cloaking nanoparticles in biodegradable natural polymers normally found in circulation. Thus, nanoparticles are not just hidden from clearance mechanisms through ‘passive shielding,’ but are camouflaged within their environment through ‘active stealthing.’ In this section, various bio-inspired shielding strategies are discussed, including those based on carbohydrates, lipids, and proteins.

Carbohydrate-based shielding strategies. While carbohydrates can be found on foreign pathogens, such as lipopolysaccharides (LPS) in gram-negative bacteria, they are also critical in the human body. They are used as energy sources and are found in extracellular matrices and on cell surfaces. For example, eukaryotic cells are naturally coated with glycosaminoglycans (GAGs), a class of negatively charged polysaccharides which are frequently conjugated to proteins on the cell surface or within the extracellular matrix. Hydrophilic polysaccharides including GAGs and other carbohydrates, can provide shielding effects for nanoparticles by mimicking naturally occurring GAGs. Carbohydrate- based shields have limited toxicity and immunogenicity. Furthermore, carbohydrate shielding mechanisms can also benefit from the natural functions of carbohydrates in the body by targeting certain carbohydrate-binding receptors.

One carbohydrate that has been found to provide anti-fouling effects is chitosan, a derivative of chitin. Because of its ability to electrostatically interact

with cells in the tumor environment (Amoozgar et al., 2012; Gerweck and

Seetharaman, 1996), it has been of interest as a pH-sensitive stealth coating.

Low molecular weight chitosan reduced phagocytic uptake and reduced protein

adsorption similarly to PEG (Amoozgar et al., 2012). However, chitosan coating

led to IgG antibody binding in the protein corona, much like PEGylated

nanoparticles, so it may have incomplete shielding or could lead to accelerated

blood clearance.

Dextran, a branched polysaccharide, has been well-characterized for its

anti-fouling effects in polymeric films (Ferrer et al., 2010; Kozak et al., 2011;

Perrino et al., 2008). Recently studies have investigated the shielding properties

and reduction of protein adsorption when dextran was used to coat

nanoparticles. Dextran coating on porous silica nanoparticles resulted in reduced

adsorption of high molecular weight plasma proteins and immune proteins.

However, the coating could not shield lower molecular weight plasma protein

binding, and also induced complement activation (Wang et al., 2015). These

results could be due to the fact that the shielding properties of dextran are

dependent on chain length, where differently coated nanoparticles have different

protein adsorption patterns, as has been seen previously for synthetic polymer

coatings (Labarre et al., 2005).

Heparin is a GAG known for its anticoagulant properties. It can also

provide shielding properties when used to coat numerous types of nanoparticles

(Bellido et al., 2015; Jaulin et al., 2000; Socha et al., 2008; Socha et al., 2009;

Wuang et al., 2006). Bellido et al. found that low-molecular weight heparin

coating improved the biological properties of mesoporous nanoparticles by preventing recognition and uptake by macrophages (Bellido et al., 2015). Heparin has also been shown to prevent angiogenesis and metastasis, making it an especially beneficial shielding agent for nanoparticles in cancer therapy (Lundin et al., 2000; Ono et al., 2002). Park et al. showed that heparin-coated nanoparticles loaded with the chemotherapy doxorubicin suppressed tumor growth in a mouse model of squamous cell carcinoma (Park et al., 2006). These nanoparticles showed reduced tumor growth compared to free doxorubicin or heparin-coated nanoparticles alone (Figure 1.2). The authors concluded this was due to the extended circulation times, along with a combination of the cytotoxicity from doxorubicin and the anti-proliferative effects of heparin.

Figure 1.2. Antitumor effects of heparin nanoparticle (NP) formulation,

doxorubicin, and doxorubicin-loaded heparin-NP in mice model of subcutaneous squamous cell carcinoma. Reproduced with permission from

(Park et al., 2006).

Another GAG, hyaluronic acid (HA), is ubiquitous within the body and found in high levels within the extracellular matrix (Toole, 2004). Stealth coatings based on HA have been shown to prevent cell interactions (Peer et al., 2008;

Upadhyay et al., 2010). Zhang et al. found that HA-coated liposomes did not

result in accelerated blood clearance and hypersensitivity upon repeat

administrations. PEG-coated liposomes, on the other hand, had increased

accumulation in the liver after repeat administrations due to the ABC

phenomenon (Zhang et al., 2016). Others have used HA not only to shield

nanoparticles but also as a targeting ligand, as HA interacts with CD44 receptors,

which are upregulated in some tissue types (Ebbesen et al., 2015; Negi et al.,

2015).

Lipid membrane shielding strategies. Because lipid membranes coat all

cells, nanoparticles enveloped with cell membrane-like shields can pass through

circulation incognito. A significant portion of research into lipid shielding

strategies has focused on membranes of blood particles, such as red blood cells,

and platelets. However, studies have also investigated the use of membranes

from other cell types, synthesized lipid coatings such as artificial membrane

‘wraps,’ and lipopeptide formulations.

A 2011 study first investigated the use of red blood cell (RBC) membranes

as an alternative shielding strategy to polyethylene glycol. RBC-vesicles were

derived from natural red blood cells harvested from mice, and used to coat

polymeric nanoparticles (Figure 1.3A). By doing so, the RBC-coated particles

retained the lipid bilayer structure and membrane proteins from the original cells.

RBC-coated particles exhibit reduced uptake into MPS organs and have longer circulation times compared to bare and PEGylated particles (Figure 1.3B, C) (Hu et al., 2011a; Rao et al., 2016). The biomimetic coating decreases macrophage uptake of nanoparticles and does not result in toxicity or accelerated blood clearance upon repeat administrations in vivo (Gao et al., 2013; Rao et al.,

2016). RBC-coating has also been shown to result in increased blood retention times and enhanced tumor uptake of nanoparticles (Piao et al., 2014; Ren et al.,

2016). Based on the success of RBC coatings, many other nanoparticle formulations have begun to explore membrane-based shielding strategies.

Figure 1.3. RBC-membranes for use as camouflage for nanoparticles. A)

Schematic of extraction of membranes from RBCs and method for coating nanoparticle surfaces. Modified with permission from (Hu et al., 2011a). B)

Biodistribution of Fe nanoparticles coated with PEG or RBCs in mice 24 hrs post- injection. Modified with permission from (Rao et al., 2016). C) Pharmacokinetics of Fe nanoparticles coated with PEG or RBCs and blood retention at 24 hrs post- injection in mice. Modified with permission from (Rao et al., 2016).

Along the same vein, white blood cell membranes have also been harnessed for stealth coatings. Leukolike vectors (LLVs) are coated with cellular

membranes harvested from mouse macrophages or human monocytes. In doing

so, LLVs avoid opsonization and have reduced uptake by the phagocytic immune

cells from which they were derived (Parodi et al., 2013). LLVs can activate

lymphocyte-receptor mediated pathways in vitro and have been shown to have

enhanced vascular permeability in vivo (Palomba et al., 2016). Macrophage cell

membrane-coated nanoparticles have also been found to have increased

circulation times and reduced retention in the organs of the MPS (Xuan et al.,

2015). Coating nanoparticles with cytotoxic T-lymphocyte membranes not only

serves to provide shielding properties such as reduced macrophage uptake, but

also enhances localization to gastric tumor tissue in vivo (Zhang et al., 2017).

This is likely because cytotoxic T-cells localize in gastric tissue, suggesting that the nanoparticle’s lipid-based coating provides targeting as well as shielding.

Platelets have the ability to marginate to the vascular wall and interact with

injury sites in the vasculature, thus nanoparticles stealth-coated with platelet

membranes could be beneficial in the treatment of diseases such as

atherosclerosis and thrombosis-related diseases. In one study nanoparticles coated with platelet membranes reduced particle uptake and complement activation, which the authors attributed to the platelet membrane-bound complement regulator proteins (Hu et al., 2015). The authors also found that the platelet-cloaked particles benefitted from platelet-like properties such as adhesion to damaged vasculature and binding to platelet-adhering pathogens.

Other cellular membranes have been explored for cloaking nanoparticles as well. Cancer cells, stem cells, and others have been used to provide membranes for nanoparticle cloaking for biomedical applications (Fang et al.,

2014; Gao et al., 2016; Zhang et al., 2015). Researchers have also investigated the possibility of designing synthetic lipid coatings that emulate biological membranes. For example, membrane ‘wraps’ are an artificial platform for shielding nanoparticles that can be tailored to target specific host organs (Xu et al., 2016). Lipopeptides can be inserted into liposomal nanoparticles to create a more biomimetic exterior, which helps limit serum protein adsorption (Ranalli et al., 2017).

Protein and polypeptide shielding strategies. Proteins are essential for numerous functions in the cell, serving as transporters, enzymes, signaling molecules, scaffolds, and many other important roles. Because they are so diverse and are found universally throughout the body, protein-based strategies hold great promise for stealth coating. Some of these strategies are based on naturally occurring proteins, while others involve synthetically derived polypeptides. Much of the work in protein stealth-coating strategies focuses on extending the half-life of protein-based therapies, but this approach holds promise for other nanoscale material applications as well.

CD47 is a cell membrane glycoprotein that has been reported to be a

‘marker of self,’ preventing clearance signaling through CD172a, a phagocytic cell receptor (Brown and Frazier, 2001; Oldenborg et al., 2000). When used to coat nanoparticles, peptides derived from CD47 have been shown to mimic the

CD47-CD172a pathway, thus preventing macrophage-mediated clearance

(Rodriguez et al., 2013). CD47 coating preferentially reduces nanoparticle uptake

by M1 macrophages, as compared to PEG which decreases clearance by all

macrophage phenotypes (Qie et al., 2016). This selective evasion may allow for

a more rationalized approach for nanoparticle shielding control.

Serum albumin (SA) is the most abundant protein in blood plasma. As such, it is a good candidate for camouflaging nanoparticles in the vascular environment. When compared with PEGylated particles, tobacco mosaic virus

(TMV) nanoparticles coated with SA had reduced antibody recognition and increased circulation times (Pitek et al., 2016b). Structural studies of SA- conjugated particles revealed that SA camouflage is due to steric hindrance as the SA molecules preventing antibodies from reaching the nanoparticle surface

(Gulati et al., 2017). Immune studies of these particles reveal that antibodies to the coating are not produced after repeat injections (Figure 1.4), which demonstrates the benefits of this coating over PEGylation (Gulati et al., in submission).

Figure 1.4. Repeated administration of SA-coated TMV NPs does not produce an immune response. A) SA-coated TMV NP administration schedule and assay protocol. B) Production of antibodies generated against the NP and the SA coating after repeat administrations for four SA-TMV constructs. Modified from (Gulati et al., in submission).

Elastin-like peptides (ELPs) are a class of genetically engineered protein- based polymers. They are based on a pentapeptide repeat of Val-Pro-Gly-X-Gly found in human elastin, where X can be any amino acid except proline. ELPs are thermally-responsive. Below their transition temperature they are soluble in aqueous solutions. Above their transition temperature, ELPs transition into an insoluble colloid-rich “coacervate” phase (Meyer and Chilkoti, 2004; Urry and

Pattanaik, 1997). ELPs have long circulation half-lives, which can be controlled based on their composition and chain length (Liu et al., 2006a; Liu et al., 2006b).

Furthermore, ELP-protein conjugates in clinical trials did not induce a significant immune response in most subjects (Gilroy et al., 2016). However, the use of

ELPs for coating nanoparticle formulations rather than protein-conjugates has yet to be explored.

PASylation is a synthetic peptide-based alternative to PEGylation that uses small amino acids to create a hydrophilic uncharged polypeptide chain with properties similar to PEG, and provides prolonged pharmacokinetics and other advantageous in vivo properties (Schlapschy et al., 2013). These chains are comprised of prolines, alanines, and serines and can be genetically encoded into protein-based therapies. Recombinant proteins are typically cleared via the kidney due to their small size (Ali et al., 2016). PAS-fusion proteins had increased hydrodynamic radius, thus prolonging circulation times in vivo

(Hedayati et al., 2017; Kuhn et al., 2016; Schlapschy et al., 2013). For example, intravital injection studies with a PASylated recombinant form of erythropoietin, a hormone that regulates the production of red blood cells, led to extended

circulation times compared to non-shielded erythropoietin. However, it was noted that the half-life of the PASylated form was shorter than that of the PEGylated form of the protein (Hedayati et al., 2017). Nevertheless, PASylated proteins have been observed to have more activity than their PEGylated counterparts

(Kuhn et al., 2016). It remains to be seen if PASylation can provide similar benefits for larger nanoparticles that are primarily cleared through the liver and spleen.

A zwitterionic synthetic peptide with a sequence of Glu-Lys repeats has been developed to prevent nonspecific protein adsorption and mimic naturally occurring protein surfaces, as the amino acids used are the two most prevalent amino acids on protein surfaces (Chen et al., 2009; White et al., 2012). Using this peptide-based strategy also allows for easy addition of targeting sequences to the end of the stealth peptide (Nowinski et al., 2014). The zwitterionic peptide coating was shown to prevent nonspecific cell uptake of gold nanoparticles in both macrophage and endothelial cell lines (Nowinski et al., 2014). However, the shielding effects of the zwitterionic peptide remain to be demonstrated in vivo.

1.2.5 Beyond shielding – Immune editing

For proteinaceous nanoparticles, an alternative to coating with shielding agents is to make the particles themselves less immunogenic. The antigenic epitopes of adeno-associated virus (AAV) gene delivery vectors have been mapped by a variety of techniques, including but not limited to directed evolution, peptide scanning, mutagenesis studies, and cryo-electron microscopy (cryoEM)

(Tseng and Agbandje-McKenna, 2014). Mapping of antigenic regions can then

be used to develop a new generation of vectors that are antigenically distinct.

CryoEM guided antigenic footprinting was used to guide directed evolution and

led to the production of AAV vectors with unique capsid antigenic motifs (CAMs)

that evade anti-AAV antibodies in sera from mice, nonhuman primates, and

humans without additional shielding molecules (Tse et al., 2017). The strategy of

developing new non-antigenic nanoparticles has drawbacks, however. Editing of

immunogenic epitopes can only be implemented for proteinaceous nanoparticles.

Modifying the vectors to be antigenically distinct may alter the ability of the virus to deliver as efficiently or alter which cells it can infect. Also, while second- generation vectors could evade pre-existing antibodies, there is a risk that newly- engineered epitopes could be produced, limiting their use in repeated administrations. Nevertheless, this strategy shows promise for specific applications.

1.2.6 Conclusions

Nanoparticle formulations offer great opportunities in drug delivery.

However, after administration immune recognition and clearance create challenges for nanoparticles remaining in circulation long enough to reach the site of disease. This can be especially detrimental upon repeat administration, when blood clearance can be accelerated. To address this issue, shielding strategies are frequently employed to coat nanoparticle formulations and to help them evade immune recognition. Typically, nanoparticles are coated with synthetic polymers, most commonly PEG. Polymeric coatings can effectively reduce nanoparticle-protein interactions. However, the widespread use of PEG

has limited its effectiveness because of the prevalence of PEG-specific

antibodies that allow for recognition and clearance of PEGylated nanoparticles.

Other polymeric coatings could suffer the same fate with increasing use.

Biopolymers, namely carbohydrates, lipids, and proteins, can provide similar shielding effects to PEG, while also camouflaging nanoparticles as ‘self.’

Bio-inspired shielding strategies have general advantages over their synthetic counterparts: they are biocompatible, biodegradable, and may be chosen so that they have low immunogenicity. Furthermore, because they are

derived from natural components of the body, they provide not just passive

evasion of immune recognition but an active mechanism of camouflage. In other

words, by appearing as if they belong carbohydrate-based shielding strategies

can mimic cell-surface polysaccharides, lipid-based strategies camouflage

nanoparticles as cell-like entities, and protein-based nanoparticle coatings blend

in with the proteins found universally within the body.

As the shielding requirements of different nanoparticle formulations will

vary depending on their application, several bio-inspired strategies have been

developed. These shielding strategies offer immune evasion while also providing

the ability of the shielding molecule to interact productively with other biological

macromolecules. Carbohydrate-, lipid-, and protein-based coatings are

biodegradable and do not provide as great a risk for repeat administrations as do

synthetic shielding strategies. Alternative approaches also exist, such as making

the nanoparticle itself less immunogenic, and these may have advantages in

specific circumstances.

In moving towards translation for nanoparticle therapies, it will be important to consider the balance between immune evasion and nanoparticle targeting. Since they are derived from natural components that sometimes have

built-in targeting ability, bio-inspired coatings may have the ability to deliver

nanoparticles to specific sites in the body. However, redirecting bio-shielded

nanoparticles may be an issue. For example, targeting platelet membrane-coated

nanoparticles away from vascular injury could prove difficult. The appropriate

coating will need to be chosen based on the requirements of the nanoparticle

application. Separate nanoparticle modifications may be needed for shielding

and for targeting. It will be important to investigate whether the effects of

targeting interfere with the stealth coating mechanism. It will also be important to

consider the function of the nanoparticle cargo. For drug delivery, long-circulating

nanoparticles are often the goal. However, when delivering contrast agents, it is not beneficial to have excessively long circulation times as imaging can be done within a matter of hours. Thus, having a way to control the half-life of a nanoparticle in the body by altering the coating would be advantageous, which could potentially be accomplished with a bio-inspired approach to shielding.

Because they offer immune evasion with many other inherent advantages, bio- inspired shielding strategies will likely play a significant role in future nanoparticle drug delivery applications.

1.3 Tobacco mosaic virus as a nanotechnology platform

1.3.1 Tobacco mosaic virus infection and production

TMV infects many species of plants, though it is most well-known for

infecting tobacco plants. The virus was initially discovered because it caused infections in tobacco crops in the 1800s, and was the first non-bacterial infectious

agent discovered (Scholthof et al., 2000). Since then, TMV has been the focus of

significant research in virology, and more recently in nanotechnology. Viruses can

be prepared in large quantities for use as nanoparticles by harnessing their viral

replication pathways in plants. Plants can be infected with purified virus via carborundum abrasion and leaves can be collected post-infection, from which

virus can be purified. Up to 4.5 mg TMV per gram of infected leaf material can be

obtained from Nicotiana benthamiana (tobacco) plants (Bruckman and

Steinmetz, 2014). This production can be scaled up to high yields for potential

nanotechnology applications.

1.3.2 Tobacco mosaic virus structure

TMV forms a 300 nm x 18 nm rigid rod structure, with a hollow 4 nm

interior channel. The virus self-assembles around its single-stranded RNA

genome, forming a helical assembly in which 2130 identical 158 amino-acid long

coat proteins (CPs) enclose the TMV genome in a right-handed helix (Creager

and Morgan, 2008). TMV can also form other structures by varying buffer

conditions (Figure 1.5) (Butler, 1984). TMV variants can be reassembled around

alternative RNA sequences, as long as the RNA contains an origin of assembly

sequence, or spherical TMV CP assemblies can be prepared using continuous

flow methods (Bruckman et al., 2015a; Eber et al., 2015; Lam et al., 2016).

Because the structure of TMV is known to atomic resolution, rational

manipulations to the virus are possible. Amino acids within the interior channel

and on the viral surface are available for chemical modification (Bruckman and

Steinmetz, 2014). Specific amino acid point-mutations can be made to provide

more residues available for conjugation, as well (Geiger et al., 2013). These

amino acid residues can be conjugated with imaging agents, shielding

molecules, targeting ligands, and therapeutics. Molecules can also be loaded into

the interior channel of TMV via electrostatic interactions.

Figure 1.5. Map of polymorphic aggregates of TMV CPs based on varying pH and ionic strength. Reproduced with permission from (Butler, 1984).

1.3.3 Tobacco mosaic virus nanoparticles

There are many advantages in using TMV as a nanoparticle. Unlike synthetic nanomaterials, TMV is genetically encoded to be monodisperse. The virus is non-infectious in mammals, biocompatible and has low immunogenicity.

Because the TMV structure is so well-defined, it is amenable to chemical and genetic engineering. Furthermore, because of its elongated shape, TMV has enhanced vascular targeting and tumor homing (Lee et al., 2013; Shukla et al.,

2015). These properties make TMV an excellent candidate for nanoparticle delivery platforms. The virus has been used for delivery of MRI contrast agents, chemotherapeutics, and photodynamic therapies (Bruckman et al., 2016;

Bruckman et al., 2014b; Czapar et al., 2016; Finbloom et al., 2016; Lee et al.,

2016). TMV can also be used for other nanotechnology applications such as immunotherapies, but those are beyond the focus of this dissertation.

1.4 Human papillomavirus infection and neutralization

1.4.1 Human papillomavirus and cancer

HPV is the most common sexually transmitted infection and nearly all sexually active people will get HPV at some point in their lives according to the

CDC. While most HPV infections go away on their own without symptoms, some

HPV infections can lead to health problems such as warts and cancer. Persistent

HPV infections, however, may to lead to precancerous lesions which may progress to cancerous outcomes (Bosch et al., 2002; Walboomers et al., 1999).

Depending on the route of infection and HPV strain, HPV can cause warts on the genital area as well as skin warts on the hands, feet, and other locations

(de Koning et al., 2015). These warts are very common with HPV infection, and

all HPV types are thought to be capable of creating warts. Only some types are

considered to be ‘high-risk’ types because they have been linked to cancers.

Nearly all cases of cervical cancer can be linked to a prior HPV infection. One

study found that, compared to uninfected women, the risk for developing cervical squamous cell carcinoma was 400 times higher following infection with HPV type

16 (HPV16) and 250 times higher following infection with HPV type 18 (HPV18)

(de Sanjose et al., 2010). Infections by HPV16 and HPV18 account for nearly

70% of all cervical cancer cases worldwide. HPV infection has also been linked to a majority of anogenital, head, and neck cancers (Doorbar et al., 2012). In fact,

HPV infection was attributed to 1.9 million new cases of cancer in 2008, accounting for approximately 15% of all new cases worldwide in that year (de

Martel et al., 2012).

1.4.2 Human papillomavirus vaccines

While there is no virus-specific treatment for HPV infection, there are two vaccines currently available for prevention of HPV-related diseases: Gardasil® and Cervarix®. Both are prophylactic vaccines directed against oncogenic HPV serotypes, using virus-like particles (VLPs) made up of the L1 major capsid protein of HPV. Because L1 differs between types, the vaccines are produced by combining multiple L1 VLPs specific to different HPV types. Cervarix® is a bivalent vaccine, protecting against HPV16 and HPV18, the two most common

HPV types in cancer, accounting for nearly 70% of cervical cancers worldwide.

Gardasil® was originally designed as a quadrivalent vaccine against HPV16,

HPV18, HPV6, and HPV11, which protect against the two most common HPV types in cancer and the two most common HPV types for causing genital warts.

In 2014, the FDA approved a nonavalent form of Gardasil®, to protect against the

HPV types included in the quadrivalent vaccine along with five additional HPV serotypes (HPV31, HPV33, HPV45, HPV52, and HPV58) that account for approximately 20% of cervical cancer cases (Harper and DeMars, 2017).

The HPV vaccines are meant to be administered before first exposure to

HPV. People aged 9-26 are eligible to be vaccinated but the CDC recommends vaccination of boys and girls at age 11-12 in order to administer the vaccine before sexual activity begins. Studies have shown HPV vaccines are efficacious for 10 years, but afterwards antibody titers may decrease (Kjaer et al., 2009;

Naud et al., 2014). Nevertheless, people can still be exposed to HPV after this

time so a risk remains.

1.4.3 Human papillomavirus structure

HPV is a double-stranded DNA virus, approximately 60 nm in size, that

infects basal epithelial cells, usually after cellular damage (Roberts et al., 2007;

Roden et al., 1994; Stanley, 2008). Over 120 different types of papillomaviruses

infect humans (Bernard et al., 2010). The genome of L2 is primarily hidden from

the surface of native HPV (Buck and Trus, 2012; Wang and Roden, 2013).

Regions of L2 have been identified as binding domains for L1 and DNA, as well

as transmembrane-like regions or nuclear localization signals (Figure 1.6)

(Becker et al., 2003; Bergant and Banks, 2013; Bienkowska-Haba et al., 2012;

Bossis et al., 2005; Bronnimann et al., 2013; Darshan et al., 2004; Fay et al.,

2004; Finnen et al., 2003; Florin et al., 2006; Kamper et al., 2006; Laniosz et al.,

2007; Mamoor et al., 2012; Richards et al., 2006; Woodham et al., 2012; Yang et al., 2003a; Yang et al., 2003b).

Figure 1.6. Known domains of the L2 minor capsid protein. Modified with

permission from (Wang and Roden, 2013). Prepared using data from (Becker et

al., 2003; Bergant and Banks, 2013; Bienkowska-Haba et al., 2012; Bossis et al.,

2005; Bronnimann et al., 2013; Christensen et al., 1991; Darshan et al., 2004;

Fay et al., 2004; Finnen et al., 2003; Florin et al., 2006; Kamper et al., 2006;

Laniosz et al., 2007; Mamoor et al., 2012; Richards et al., 2006; Rubio et al.,

2011; Woodham et al., 2012; Yang et al., 2003a; Yang et al., 2003b).

Two recent single particle cryo-electron microscopy (cryoEM) studies have revealed the structure of HPV PsVs from HPV type 16 to 9 Å and 4.3 Å resolution

(Cardone et al., 2014; Guan et al., 2017). Density for the L2 protein was not visualized in the 9 Å resolution structure (Cardone et al., 2014). In the 4.3 Å resolution structure, L2 was not clearly defined but by comparing the HPV PsV structure (containing L1 and L2) with a simulated structure of a capsid containing

L1 alone, differences suggested locations in which L2 may interact on the hexavalent capsomers (Figure 1.7) (Guan et al., 2017). This suggests that L2 may be disordered or lack 60-fold symmetry used in refining the HPV capsid structures.

Figure 1.7. Identification of L2 density regions by comparison with a simulated L1 capsid. A) Six copies of L1 were built into the HPV map to make an asymmetric unit. Hexavalent capsomers are indicated by a black hexagon.

Pentavalent capsomers are indicated by a black pentagon. B) Difference density

(red) attributed to L2 calculated by subtracting the simulated L1-only capsid map

from the experimental map. L2 appears to have exposed density on both

pentavalent and hexavalent capsomers, although it is more predominantly on the

hexavalent capsomers. C) Cross-section view to show apparent L2 density within

the capsomers. Reproduced with permission from (Guan et al., 2017).

1.4.4 Human papillomavirus cell entry mechanism and infection pathway

HPV Is thought to have a complex host cell entry pathway, with a half-time

of up to 12 hours (Schelhaas et al., 2012). The virus can initially only infect basal

epithelial cells, from which it propagates to other epithelial cells, and the viral life

cycle is tied closely to epithelial differentiation (Stanley, 2008). HPV initially binds

to heparan sulfate proteoglycan receptors (HSPGs) on the basement membrane

(Johnson et al., 2009; Kines et al., 2009). Upon binding to the HSPGs, the HPV

capsid undergoes a conformational change in which the N-terminus of L2

becomes surface-exposed (Bienkowska-Haba et al., 2012; Day et al., 2008). L2

is then cleaved by furin, which allows release from the HSPG receptors and a

previously-unexposed region of L1 binds an unidentified secondary receptor on basal epithelial cells (Figure 1.8A) which facilitates internalization (Figure 1.8B)

(Richards et al., 2006; Schelhaas et al., 2012; Selinka et al., 2007). Upon

internalization, the virus traffics to the endosome, leading to viral uncoating

(Figure 1.8C, D) (Campos et al., 2012; Richards et al., 2006). L2 and the viral

genome traffic along microtubules to the nucleus (Figure 1.8E, F) (Bergant

Marusic et al., 2012; Kamper et al., 2006). The L2-genome complexes localize

within the nucleus in ND10 bodies, which promotes transcription of the genome

(Day et al., 2004; Florin et al., 2002).

Figure 1.8. Cell entry pathway of HPV after furin cleavage. A) HPV binds to

an unidentified receptor which mediates cell entry. B) HPV enters the cell by an

endocytosis. C) By 4 hours, HPV localizes in the early endosome. D) By 12

hours, the virus uncoats and the genome is coupled with L2. E) The genome and

L2 travel via microtubules through the cytoplasm and enter the nucleus by 24

hours. F) The L2-genome complex co-localizes with ND10 and transcription of

the viral genome begins. Reproduced with permission from (Schiller et al., 2010).

1.4.5 Human papillomavirus and human alpha-defensin 5

Human alpha-defensin 5 (HD5) is a small, secreted immune protein with

broad antimicrobial activity against bacteria and viruses (Lehrer and Lu, 2012). It

is thought to disrupt lipid membranes in its activity against bacteria and

enveloped viruses, though the exact mechanism for this remains unclear. For

non-enveloped viruses, a different neutralization mechanism is required. HD5 is

known to prevent infection of multiple non-enveloped viruses, including

adenovirus (AdV), JC polyomaviruses, and HPV (Buck et al., 2006; Flatt et al.,

2013; Gounder et al., 2012; Smith et al., 2010; Zins et al., 2014). HD5 neutralizes

AdV by binding long flexible regions of the penton base viral capsid protein at the

interface between two capsid proteins, stabilizing the capsid and preventing it

from uncoating within the endosome after cell entry (Flatt et al., 2013; Nguyen et

al., 2010; Smith et al., 2010). JC polyomavirus capsids are also stabilized by

HD5, leading to altered intracellular trafficking (Zins et al., 2014). The mechanism

of HD5 neutralization of HPV also seems to be through stabilization of the viral

capsid, although how this occurs and which capsid proteins are involved remain

unclear (Wiens and Smith, 2017). It is known, however, that HD5 neutralizes

multiple types of HPV (Buck et al., 2006). HD5 prevents furin cleavage of the

HPV L2 protein, which is required for cell entry (Wiens and Smith, 2015).

However, HPV can be neutralized by HD5 even after the furin cleavage step

occurs (Wiens and Smith, 2017). HD5 neutralization involves prevention of the

HPV genome from escaping the endosome, and redirection of the virus towards

the lysosome (Buck et al., 2006). Recent data suggests this is because HD5

prevents dissociation of the L2-genome complex from L1 within the endosome, thereby redirecting the virus to the lysosome (Wiens and Smith, 2017). However, the mechanism of HD5 binding and its interactions with the HPV capsid are unknown. Understanding this mechanism could lead to the development of HD5-

based therapies for the treatment of HPV.

1.5 Cryo-microscopy techniques to study viral structures

Many structural biology techniques allow for studying virus structures, both alone or in the presence of ligands or binding proteins. Of these, cryo-electron

microscopy techniques offer the advantage of studying the virus in a native-like

state without dehydration effects by freezing samples in glass-like vitreous ice. In

recent years, cryo-electron microscopy techniques have gained momentum due

to a “resolution revolution,” in which structures solved by this technique have

attained near-atomic resolution (Kuhlbrandt, 2014). This has occurred largely due

to developments of new camera technologies with increased sensitivity in data

collection (Kuhlbrandt, 2014). Resolution of these structures is determined by

measuring the Fourier Shell Correlation (FSC), in which two independent

reconstructions of half-datasets are calculated and the consistency between the

two maps is determined in Fourier space (van Heel and Schatz, 2005). A FSC

threshold of 0.143 has been proposed as a realistic value for estimating the

resolution of a structure (Rosenthal and Henderson, 2003).

Reconstructions using single particle cryo-electron microscopy (cryoEM)

have frequently achieved resolutions of sub-4 Å since the “revolution in 38

resolution” (Orlov et al., 2017). Single particle cryoEM reconstructions are based

on the assumption that the input micrographs contain multiple projection images of the same object that is structurally and conformationally homogeneous.

Particles with different orientational views are extracted from micrographs and averaged together to form a three-dimensional structure. Sorting methods can be used to separate particles into different populations or classes which can be reconstructed into multiple different structures, but a single reconstruction will contain images from multiple different particles that are assumed to be of an identical population. Because of this, any areas of heterogeneity between particles will limit the attainable resolution and may not be visible in high- resolution reconstructions.

For heterogeneous samples, cryo-electron tomography (cryoET) is a useful alternative. Rather than determining a structure based on different particles, multiple images of the same particle at different angles are obtained.

These images can be built into a tilt series and aligned to reconstruct a tomogram without averaging different particles and assuming homogeneity.

However, cryoET does not provide as high resolution structures as single particle cryoEM. Nevertheless, if there are sub-structures that are homogeneous in nature, these can be combined using sub-tomogram averaging, yielding higher resolution of those sub-structures compared to cryoET alone (Wan and Briggs,

2016).

1.6 Aims of dissertation

It is evident that viruses are important macromolecules in biomedical research, both to develop new delivery systems and also to understand how to treat and prevent viral-based diseases. The purpose of this dissertation is to investigate mechanisms by which viruses interact with the immune system, and

how to manipulate them for biomedical applications. This theme will be

addressed throughout the dissertation. In Chapter 2, serum albumin

camouflaged tobacco mosaic virus nanoparticles are structurally characterized to

understand how a bio-inspired shielding strategy prevents immune recognition. In

Chapter 3, the immune recognition of serum albumin-coated tobacco mosaic

virus nanoparticles is investigated by studying macrophage uptake and antibody

production. In Chapter 4, the interactions between an immune peptide, HD5, and

the pathogenic human papillomavirus are structurally characterized to give

insights into neutralization of the virus. Finally, Chapter 5 will provide conclusions

and future studies related to the work presented here.

Chapter 2: Structural characterization of SA-TMV

The material in this chapter was adapted from Gulati NM, Pitek AS, Steinmetz

NF, Stewart PL, 2017. Cryo-electron tomography investigation of serum albumin- camouflaged tobacco mosaic virus nanoparticles. Nanoscale 9 (10), 3408-3415 with permission from the Royal Society of Chemistry (www.rsc.org).

2.1 Abstract

Nanoparticles offer great potential in drug delivery and imaging, but shielding strategies are necessary to increase circulation time and performance.

Structure–function studies are required to define the design rules to achieve effective shielding. With several formulations reaching clinical testing and approval, the ability to assess and detail nanoparticle formulations at the single

particle level is becoming increasingly important. To address this need, we use cryo-electron tomography (cryo-ET) to investigate stealth-coated nanoparticles.

As a model system, we studied the soft matter nanotubes formed by tobacco

mosaic virus (TMV) coated with human serum albumin (SA) stealth proteins.

Cryo-ET and subtomogram averaging allow for visualization of individual SA

molecules and determination of their orientations relative to the TMV surface,

and also for measurement of the surface coverage provided by added stealth

proteins. This information fills a critical gap in the understanding of the structural morphology of stealth-coated nanoparticles, and therefore cryo-ET may play an

important role in guiding the development of future nanoparticle-based

therapeutics.

2.2 Introduction

Nanoparticles (NPs) offer promise as drug delivery vehicles to target

drugs to diseased tissue, and they can be combined with contrast agents to

integrate imaging to assess therapy success and disease progression. While the

opportunities are wide ranging, challenges exist. When introduced into an

organism, nanoparticles face biological hurdles such as immune recognition

leading to clearance in non-target tissues. ‘Naked’ NPs can be recognized and

tagged by antibodies, complement proteins, and other components of the innate

immune system, leading to clearance by the mononuclear phagocyte system

(Lucas et al., 2016; Pitek et al., 2016a). To circumnavigate this, stealth coatings

have been developed to prevent carrier recognition and clearance, thereby

enhancing the pharmacokinetic profiles and performance.

Two principal methods have been developed: stealthing with hydrophilic

polymers and camouflage as ‘self’. Coating NPs with polyethylene glycol (PEG),

termed PEGylation, is a popular strategy. The hydrophilic polymer reduces

nanoparticle–protein interactions and therefore clearance (Owens and Peppas,

2006). However, PEG does not completely prevent immune recognition and its

efficacy is dependent on the physical characteristics of the specific polymer

selected (Lee et al., 2015a; Perry et al., 2012). Furthermore, recent data

suggests that PEG-specific antibodies are becoming more common in the human

population (up to 25% of the population in 2012 compared to 0.2% of the

population in 1984) (Garay et al., 2012; Richter and Akerblom, 1984). Recent

clinical trials have highlighted that the prevalence of anti-PEG antibodies can

impair the functionality and safety of PEGylated therapies (Ganson et al., 2016b;

Hershfield et al., 2014). Alternative polymer coatings and strategies are urgently

needed to overcome this technological hurdle. Thus, more recent approaches

make use of ‘self’ coatings, using proteins or peptides to mark the nanocarrier as

self therefore overcoming immune clearance (Pitek et al., 2016b; Rodriguez et

al., 2013; Sosale et al., 2015). Data indicate that these self-coatings may be

superior to contemporary PEG coatings as demonstrated by reduced immune

recognition along with enhanced circulation. In the present work, we set out to

structurally characterize a nanocarrier coated with the ‘self’ protein serum

albumin (SA), the most abundant plasma protein in humans.

While there are numerous methods to investigate NP stealth coatings,

challenges still remain to understand the precise mechanism of enhancing the in

vivo performance of nanocarriers. It is unclear why some stealth coatings

perform better than others, including why different formulations of the same

shielding strategy have varied effects. Thus, the need arises to characterize the

physical and morphological properties of various stealthing strategies when

applied to NP platforms. Multiple characterization techniques are usually

combined to characterize stealth coatings. For example, the loading density of stealth polymers, proteins, or peptides can be determined through NMR, surface plasmon resonance, gel electrophoresis, and fluorescent modeling (Demers et al., 2000; Garcia-Fuentes et al., 2004; Jokerst et al., 2011; Steinmetz et al.,

2009). Estimation of the size of coated nanoparticles can be obtained through

techniques such as dynamic light scattering and chromatography techniques

(Cheng et al., 2012; Li and Huang, 2009; Seehuber et al., 2012; Zanetti-Ramos

et al., 2009). Nevertheless, these techniques typically provide averaged

information on a population of nanoparticles, rather than data on individual NPs.

Techniques such as field emission scanning electron microscopy and negative

stain transmission electron microscopy (TEM) can be used to visualize individual

particles. But these techniques involve dried samples, whereas in the

bloodstream the NPs would be in a hydrated state (Duncanson et al., 2007; Gref

et al., 2001).

To overcome these technological challenges, this work utilizes cryo-ET

(Stewart, 2016) to visualize stealth coatings on the surface of a nanoparticle in a

hydrated state. As proof-of-concept, we use tobacco mosaic virus (TMV) as a

model NP platform coated with the stealth protein SA. TMV is a 300 nm by 18 nm

rigid soft matter nanotube which has served as a model organism in structural

biology studies, making it an ideal candidate platform for this study. It is a plant

virus that we have engineered as a drug delivery vehicle and MRI contrast agent

(Bruckman et al., 2014b; Czapar et al., 2016). However, as with other biologics,

TMV has a short circulating half-life as a result of immune recognition. To

overcome these properties, we recently developed SA-‘camouflaged’ TMV NPs,

which demonstrate reduced recognition by TMV-specific antibodies with no

change in macrophage uptake compared to ‘naked’ and PEG-coated NPs (Pitek

et al., 2016b). SA coating also extended the circulating half-life time more

effectively than PEG-coatings. Here, we investigate the physical characteristics

of SA-coated TMV NPs to understand the structural morphology of these

camouflaged NPs and how the camouflage may influence their function.

2.3 Materials and Methods

2.3.1 Virus propagation and purification

The following method was performed by Dr. Andrzej S. Pitek. T158K mutant of TMV (Geiger et al., 2013) (TMV-lys) was propagated in Nicotiana benthamiana plants through mechanical inoculation using 5-10 µg of virus per leaf. Viruses were isolated and purified using established methods to yield approximately 1 mg of virus per gram of infected leaf material (Bruckman and

Steinmetz, 2014).

2.3.2 TMV sCy5 labeling

The following method was performed by Dr. Andrzej S. Pitek. TMVs were labeled with the fluorescent dye Cy5 at glutamic acid residues lining the central channel of TMV. Glutamic acids were first modified with alkynes by EDC coupling for 24 hours using 1 mg/ml TMV, 100 equivalents of propargylamine (Sigma

Aldrich) per capsid protein with 50 equivalents of EDC (25 equivalents added at 0

and 18 hours) in 100 mM HEPES buffer, pH 7.4. This was followed by an alkyne- azide click reaction for 30 minutes by adding 1 equivalent of sCy5-azide

(Lumiprobe) per coat protein in the presence of 2 mM AMG (Fisher), 2 mM AsC

(Fisher), 1 mM CuSO4 (Fisher) in 10 mM potassium phosphate buffer, pH 7.4 on

ice. sCy5-labled TMV was purified by ultracentrifugation at 42,000 rpm for 3

hours on a 40% w/v sucrose cushion.

2.3.3 TMV conjugation

The following method was performed primarily by Dr. Andrzej S. Pitek.

Human serum albumin (SA; Sigma Aldrich) was conjugated to the solvent-

exposed exterior surface of TMV-lys through a PEG linker. SA was first conjugated using NHS-PEG4-SAT (ThermoFisher) at a 1-to-1 ratio in 10 mM

sodium phosphate buffer 125 mM saline, pH 7.4 containing 10% v/v DMSO

overnight at RT. De-acetylation solution (0.5 M hydroxylamine, 25 mM EDTA in

PBS, pH 7.2-7.5) was added to the reaction at a final concentration of 10% v/v to

de-protect the thiol group and this de-protection reaction was carried for 2 hours

at RT. TMV-lys was conjugated using maleimide-PEG4-NHS (ThermoFisher) at

10 equivalents PEG per TMV-lys coat protein in 10 mM potassium phosphate buffer, pH 7.4 containing 10% v/v DMSO for 2 hours at RT. The resulting SA-

PEG4-SH conjugates and MAL-PEG4-TMV conjugates were purified through a

PD MiniTrap G-25 desalting columns (GE). Purified SA-PEG4-SH was then

reacted with purified MAL-PEG4-TMV at a ratio of 6 equivalents of SA-PEG4-SH

per TMV coat protein overnight at room temperature. The reaction was quenched

for 1 hour at RT by addition of excess glycine and l-cysteine. SA-PEG8-TMV was

then purified by ultracentrifugation at 55,000 rpm for 3 hours on a 40% w/v sucrose cushion. Conjugation was verified and quantified using SDS-PAGE analysis.

2.3.4 SDS-PAGE analysis

TMV samples were denatured by boiling at 100°C for 5 minutes in gel loading buffer (62.5 mM Tris–HCl pH 6.8, 2% w/v SDS, 10% v/v glycerol, 0.01% w/v bromophenol blue, 10% v/v 2-mercaptoethanol). 20 µg TMV were loaded on

4-12% NuPAGE gels in MOPS running buffer and separated at 200 V for 45 min.

Gels were stained with Coomassie Blue and visualized using an AlphaImager imaging system (Biosciences). Gels were analyzed by densitometry using

ImageJ (Schneider et al., 2012).

2.3.5 Western blot analysis

Samples separated by SDS-PAGE (with 20 µg of loaded TMV) were transferred to nitrocellulose membranes at 100 V for 1 hour. Membranes were incubated in 5% w/v milk in TBS-Tween at RT for 1 hour, followed by incubation with either 0.5 µg/mL rabbit anti-TMV antibody (Pacific Immunology) or 0.5

µg/mL rabbit polyclonal anti-albumin antibody (Novus Biologicals) in 5% w/v milk in TBS-Tween for overnight at 4˚C. Samples were then washed three times for

10 min each in TBS-Tween and incubated in 1 µg/mL alkaline phosphatase goat anti-rabbit antibody in 5% w/v milk in TBS-Tween at RT for 1 hour, followed by three washes for 10 min each in TBS-Tween and 1 min wash in Millipore water.

Antibody binding was visualized using Novex AP Chromogenic Substrate

(BCIP/NBT; Invitrogen). 47

2.3.6 Immuno-dot blots

Dot blots were prepared by applying 1 µL spots of 150 µg/mL anti-TMV

and 150 µg/mL anti-CPMV control antibodies (Pacific Immunology) on a nitrocellulose membrane after the membrane has been equilibrated in 10 mM

PBS. Blots were incubated in 5% w/v milk in PBS at RT for 1 hour, washed three times for 5 min each in 10 mM PBS. The blots were then incubated in 40 µg/mL

Cy5-labeled VNPs for 2.5 hours at RT, followed by three washes for 5 min each in PBS. Blots were then dried and imaged for fluorescence using a Maestro

imaging system. Blots were analyzed for integrated density using ImageJ.

2.3.7 Negative stain transmission electron microscopy

3 µL SA-TMV or control naked TMV NPs at a concentration of 0.5 mg/mL were applied to glow-discharged carbon coated 200 mesh copper grids. After 1 min, excess sample was removed and grids were washed two times in dH2O for

1 min each. Grids were then stained two times for 30 sec each with 2% w/v

uranyl acetate, followed by blotting until dry with Whatman 1 blotting paper. Grids

were imaged in a FEI Tecnai G² Spirit (120kV) transmission electron microscope

with a 4k x 4k Gatan US4000 CCD camera.

2.3.8 Cryo-electron microscopy and tomography

3 µL SA-TMV or control naked TMV NPs at a concentration of 0.5 mg/mL

were mixed with 0.5 µL 10nm fiducial nanogold (Aurion). The solution was then applied to glow-discharged Quantifoil 2x2 200mesh holey carbon grids. Grids

were blotted until nearly dry (approximately 2.5 sec blot) and rapidly plunged into

liquid ethane cooled with liquid nitrogen using a manual plunger. Grids were 48

imaged in a JEOL 2200FS transmission electron microscope equipped with an energy filter using a Tietz TVIPS 4k x 4k CMOS camera (for cryo-electron microscopy) or a Direct Electron DE20 direct detector (for cryo-electron tomography). For tomography, SA-TMV or TMV grids were imaged as 0.5s long movies with four frames each, which were then motion-corrected using custom

EMAN2 (Tang et al., 2007) scripts to result in single images. Data was collected within a -70° to +70° tilt angle range or until a grid bar prevented further imaging.

Tilt series were processed with the IMOD processing package (Kremer et al.,

1996) to generate tomograms, which were visualized with UCSF Chimera

(Pettersen et al., 2004).

2.3.9 Subtomogram averaging of SA

Selected individual densities attributed to SA were cropped from the tomogram density using UCSF Chimera. These densities were analyzed with the

Jsubtomo software package (Huiskonen et al., 2014) using a 20 Å filtered map of the SA crystal structure (Sugio et al., 1999) (PDB: 1AO6) as a template map.

Refined SA positions were selected for which Jsubtomo found dimer related orientations of the SA template in the cropped cryo-EM density regions. The cryo-EM density from 10 refined SA positions, corresponding to five SA dimers, was averaged to generate an average SA density map. This average SA density map was positioned back into the tomogram using the refined SA positions to reveal the orientation of SA with respect to the TMV surface at multiple locations.

2.3.10 Subtomogram averaging of SA-TMV segment

The tomogram of a whole SA-TMV particle was cropped into 21 segments, each corresponding to the length of six turns of the TMV helix, using

UCSF Chimera. These density segments were analyzed with the Jsubtomo

software package (Huiskonen et al., 2014). An SA-TMV template map was

generated with an 8.8 nm diameter sphere, representing the same volume as

SA, positioned 2.4 nm away from the surface of a 20 Å filtered map of six helical

turns of TMV (PDB: 4UDV) (Fromm et al., 2015). Note that the average diameter

of the PEG linker is estimated to be 2.4 nm. Refined SA-TMV segment positions

were selected based on the highest cross-correlation score to the template map.

The aligned cryo-EM density from all segments was averaged to generate an

average SA-TMV segment map.

2.3.11 Analysis of volume coverage

Individual SA-TMV or TMV rods were cropped from the SA-TMV

tomogram or TMV tomogram, respectively, using UCSF Chimera. Rods were

oriented along the Z-axis and clipped to one-pixel wide segments in UCSF

Chimera. The integrated density of an oval containing only the TMV region of

each segment was determined using the Measure tool in ImageJ. An oval was

chosen because the density was skewed due to the missing wedge artifact of

tomography data collection. The integrated density was also measured for a

circle extending well beyond the TMV region to allow for SA. The TMV-only area

was subtracted from the larger SA and TMV area to calculate the area filled by

SA. This area was calculated for each one-pixel wide segment and plotted. The

integration of this plot results in the total volume coverage around the TMV. By

calculating the volume of a single SA in a similar fashion, the total number of SAs

on an individual rod were determined.

2.3.12 Analysis of surface coverage

Individual SA-TMV rods were cropped from the SA-TMV tomogram using

UCSF Chimera. Rods were oriented along the Y-axis and pseudo-colored red,

with a 300 nm by 18 nm blue cylinder also displayed using UCSF Chimera. At

one degree rotations, images were saved over a total of 360 degrees. For each

degree, the saved picture was opened in ImageJ and the amount of red and blue

shown in a one-pixel wide line at the center of the SA-TMV was determined using

the RGB Measure tool. The total amount of red in all 360 degrees divided by the

total amount of red + blue yields a percent coverage for a single SA-TMV rod.

2.4 Results and Discussion

2.4.1 Particle preparation

To prepare SA-coated NPs, TMV containing a T158K mutation (termed

TMV-lys) (Geiger et al., 2013) was incubated with MAL-PEG4-NHS yielding maleimide-functional MAL-PEG4-TMV. In a separate reaction, SA was reacted

with NHS-PEG4-SH to produce thiol-activated SA-PEG4-SH. These two products

were then reacted together via Michael addition to form SA-PEG8-TMV (in short

SA-TMV, Figure 2.1A). Conjugation and structural integrity of the SA-TMV

formulation was confirmed first by SDS-polyacrylamide gel electrophoresis (SDS-

PAGE; Figure 2.1B). TMV coat proteins (CPs) have a molecular weight of 17

kDa, SA has a molecular weight of 66.5 kDa. SA-TMV shows the expected band

pattern: free CP is detectable as a ~17 kDa band while conjugated SA-CP is

detectable as high molecular weight bands (Figure 2.1B). Western blotting

against SA (Figure 2.1C, red) and against TMV (Figure 2.1C, green) indicate the

higher order bands are in fact SA-conjugated CPs. According to densitometry

analysis, ~0.25-0.35 mg SA were conjugated per 1 mg TMV particles yielding

approximately 140-200 SAs per TMV.

TMV is comprised of 2,130 copies of an identical CP with the T158K

mutation available for conjugation on the solvent-exposed exterior TMV surface.

Each CP has an exposed exterior surface area of about 8 nm2. With the addition

of the PEG linker, extending on average 2.4 nm from the TMV surface as

calculated by the Flory dimension (Lee et al., 2015a), this generates an

expanded surface area of ~10 nm2 per PEG-TMV CP. Based on the SA crystal structure, the footprint of the SA dimer would be 28-47 nm2, depending on its

orientation. Since the footprint of SA is larger than the exposed surface area of a

single PEG-TMV CP, SA dimers would be unable to bind to each PEG-TMV CP

subunit. Steric constraints would limit SA dimers to binding to every 3-5 PEG-

TMV CPs on average. The theoretical maximum coverage of TMV, if each SA

dimer was oriented with its smallest footprint facing the TMV surface and with all

PEG linkers fully extended (2.8 nm), is approximately 800 SA dimers per TMV.

The practical maximum, however, would be much lower due to steric hindrance

between SA molecules, random orientational preference on the TMV surface, and PEG linker flexibility.

To estimate the surface coverage based on the densitometry analysis, both the number of SAs per TMV and the orientation of each SA relative to the

TMV rod must be considered. Assuming the exposed surface area of a TMV CP is approximately 8 nm2 and the footprint of SA is 28-47 nm2 depending on its

orientation, the theoretical surface coverage of TMV is in the range of 20-55%.

This underlines the importance to develop more quantitative experimental

methods to determine coverage.

To confirm that SA-TMV maintains stealth properties, we investigated

immune recognition from TMV-specific antibodies. Viral vectors are primarily

eliminated from the body through antibody clearance, and TMV-specific

antibodies have been previously found in blood collected from both smokers and

non-smokers, presumably due to the prevalence of the virus in crops and

tobacco cigarettes (Balique et al., 2012; da Silva et al., 2008; Hu et al., 2011b;

Wetter, 1975). Repeat administration of TMV-based therapeutics would also lead

to development of TMV-specific antibodies, so it is critical to test the ability of

stealth-coated particles to prevent immune recognition. Immuno dot blots with

anti-TMV antibodies and fluorescent TMV particles were performed to test the

shielding properties of the SA-TMV particles (Figures 2.1D and 2.1E). To

prepare the fluorescently labeled particles, Cy5 dyes were conjugated to the

interior channel of TMV using click chemistry. Dot blots were performed by

spotting polyclonal anti-TMV or control polyclonal anti-cowpea mosaic virus 53

(CPMV) antibodies on a nitrocellulose membrane, which was then incubated with

Cy5-labeled NPs. Recognition was measured using a Maestro Imaging system.

As seen previously, SA-TMV particles have reduced antibody recognition

compared to bare particles (Pitek et al., 2016b).

Figure 2.1. Production and characterization of SA-TMV. (A) Schematic representation of conjugation of TMV with SA. SA and TMV-lys are first separately conjugated with PEG linkers through NHS chemistry, followed by conjugation of SA-PEG4-SH and maleimide-PEG4-TMV to produce SA-PEG8-

TMV. Analysis of particles before and after conjugation by (B) SDS-PAGE and

(C) Western Blot (WB). Free SA was used as a reference for WB analysis. SA conjugation is indicated by the presence of multiple bands with MW > 64 kDa

(MW of single TMV-lys CP is ~17 kDa; MW of single SA is ~67 kDa). WB immune recognition by anti-TMV antibodies shown in green, anti-SA antibodies 55

shown in red. The data in panel C was provided by Dr. Andrzej S. Pitek. (D) Dot blot analysis of immune recognition of TMV and SA-TMV by anti-TMV and anti- cowpea mosaic virus (anti-CPMV, negative control). PBS is an additional negative control. (E) Quantitative densitometric analysis of the dot blots.

2.4.2 Imaging and processing of naked and SA-coated particles

Visualization in 2D. The physical characteristics of the ‘naked’ TMV and

SA-coated particles were obtained with a combination of negative stain and cryo-

electron microscopy (cryo-EM), followed by a detailed analysis using cryo-ET.

While cryo-samples are more technically challenging to prepare than negative

stain samples, they offer the advantage of visualization of the sample in a frozen

hydrated native-like state, albeit at lower contrast than observed with negative

stain.

As expected, TMV displays much better contrast by negative stain (Figure

2.2A) than by cryo-EM (Figure 2.2B), while both methods reveal smooth rod-

shaped particles. As we previously reported, the SA coating is detectable by

negative stain TEM (Pitek et al., 2016b). SA-TMV exhibits a rough surface

indicating presence of SA (Figure 2.2C). Nevertheless, individual SAs cannot be

identified in negative stain TEM images. In contrast, cryo-EM images of SA-TMV

do reveal individual SA molecules as dots on the surface of the rod-shaped TMV

(Figure 2.2D). From a single two-dimensional cryo-micrograph of SA-TMV, it is

apparent that there are numerous SAs per TMV nanoparticle and that the surface

of TMV is fairly well covered.

Figure 2.2. Transmission electron microscopy of TMV and SA-TMV. Bare

TMV-lys visualized by negative stain (A) and cryo (B) TEM. (C) Negative stain

images of SA-conjugated particles show a rougher, uneven surface. (D) Cryo-EM

images of SA-TMV samples reveal individual SA proteins on the TMV surface.

Electron dense 10 nm gold fiducials are visible in the cryo micrographs. (Scale bars = 100 nm).Visualization in 3D by cryo-electron tomography. To investigate the three-dimensional architecture of SA on the surface of TMV, cryo- 58

ET was performed. Data was collected using a JEOL 2200FS microscope

equipped with an energy filter and a DE20 direct detector device. Images were

motion corrected and processed using the IMOD software package (Kremer et

al., 1996), resulting in an aligned tilt series (see Movie 2.S1) and reconstructed

tomogram (see Movie 2.S2). A single slice from the reconstructed tomogram of

SA-TMV is shown in Figure 2.3A. As in the cryo-EM micrograph of SA-TMV

(Figure 2.2D), individual SA molecules are observed to dot the surface of the

TMV rods. When the entire tomogram is displayed as a three-dimensional

surface (Figure 2.3B), it becomes apparent that SA molecules surround each

TMV rod, leading to a relatively even coating of the NPs. An enlarged view of a

single SA-TMV rod in Figure 2.3C (left) shows a rough exterior with many

protrusions. By displaying a cylinder with an 18 nm diameter (blue) centered on

the SA-TMV density (red), it becomes possible to identify individual SA

molecules positions relative to the TMV rod (Figure 2.3C, right).

The radius of TMV is 9 nm and the site of attachment of the PEG linker

(Lys-158) is on the external surface of TMV. The average diameter of the PEG

linker is estimated to be 2.4 nm, as calculated from the Flory dimension of an 8-

mer linear PEG polymer (Lee et al., 2015a). Based on this information, on

average most SA dimers are expected to be attached 2.4 nm from the surface of

TMV or 11.4 nm from the central axis of TMV. Since the PEG linker is highly

flexible, it is possible that the SA dimers could begin close to the TMV surface, or

9 nm from the central axis. Alternatively if a PEG linker were fully extended to its

maximum length of 2.8 nm, then the distance between the SA dimer and TMV

would be 2.8 nm, or 11.8 nm from the central axis. The dimensions of the SA

dimer as calculated from the crystal structure are 6 nm x 6 nm x 10 nm (Sugio et

al., 1999). Taken together this means that the SA density in the SA-TMV NPs can range from 9 nm to 21.8 nm from the central axis of TMV. This agrees with

the experimental data, in which SA density extends from the surface of the TMV

NP in all directions for approximately 12 nm from the TMV surface (Figure 2.3C,

right).

Figure 2.3. Cryo-electron tomography analysis of SA-TMV nanoparticles.

(A) Central slice through processed tomogram of SA-TMV. A few electron dense

fiducial markers are visible in this slice. (Scale bar = 100 nm). (B) Three- dimensional representation of SA-TMV tomogram. (Scale bar = 100 nm). (C)

Single SA-TMV rod in blue, left, and in red with 300 nm x 18 nm cylinder shown

in blue, right. (Scale bar = 25 nm).

Subtomogram averaging of SA. To determine the orientations of SA

relative to the TMV surface, selected protrusions were cropped and aligned with

the SA crystal structure, and averaged using the subtomogram averaging

software Jsubtomo (Huiskonen et al., 2014; Sugio et al., 1999). The resulting

average SA density was then placed into the refined SA positions within the

tomogram to visualize how SA is oriented on the TMV surface. Figure 2.4A

shows the average SA density and docked SA crystal structure positioned with

respect to the TMV rod at multiple sites. These results indicate that there is not

one specific preferred orientation of SA relative to TMV. There are 120 lysine

residues in the SA dimer, of which 79 are solvent-exposed (Figure 2.4B). It is

possible that a number of these lysine side chains are reactive and serve as

anchors for the TMV linkage sites. Indeed, multiple lysine residues throughout

the SA structure, including Lys 137, Lys 190, Lys 351, Lys 505, and Lys 525,

have been found to be reactive groups in previous studies (Anraku et al., 2015;

Krantz et al., 2014; Meng et al., 2015; Shuck et al., 2014). The numerous

possible SA attachment points, along with the flexibility of the 8-mer PEG linker

would explain the heterogeneous display of SA observed for SA-TMV.

Heterogeneous display may be advantageous for in vivo applications of these

particles, to avoid innate immune recognition through pathogen-associated

molecular patterns (PAMPs).

Figure 2.4. Subtomogram averaging analysis of SA orientation on the TMV surface. (A) Averaged SA density (transparent gray) and the human SA crystal structure (Sugio et al., 1999) (PDB: 1AO6, gray ribbon) positioned with respect to

the TMV cylinder (blue) at five locations along a single SA-TMV nanoparticle.

These locations are shown both with (top) and without (bottom) the cryo-ET

density (red). (B) SA crystal structure (PDB: 1AO6) with the C-alpha atoms of

solvent-exposed lysine residues depicted as blue spheres.

Subtomogram averaging of SA-TMV segment. To identify any effect a

single SA might have on the positioning of its neighboring SA molecules, subtomogram averaging was performed on SA-TMV segments. The density map

for SA-TMV was divided into 21 segments, each containing ~6 helical turns of

TMV. Each segment was aligned to a generated template map of SA-TMV with a

sphere representing SA positioned in the middle of the segment at the expected

average distance from the TMV surface (Figure 2.5A). A sphere with the same

volume as the SA dimer was used in the template map to prevent bias in

searching for a specific orientation of SA relative to the TMV central axis. The

aligned segments were averaged using Jsubtomo (Figure 2.5B). Through this

analysis, any features that are repeated in all segments would show density in

the average map and differences would be averaged away. In comparison to the

template map, the average SA density is smaller than the sphere used in the

template map and also smaller than expected for an SA dimer (Figure 2.5C).

This suggests that the subtomogram averaged segments may be aligned on one

SA subunit, and that due to the multiple orientations of SA dimers relative to the

TMV surface the other SA subunit in each dimer is effectively averaged away.

Furthermore, there seems to be no pattern of neighboring SA molecules relative to the central SA dimer, as no density is apparent for additional SA dimers in the average SA-TMV segment map. Taken together, these results support the conclusion that although there is an average distance for SA dimers from the

TMV surface (~2.4 nm), the SA orientations are random and there is no interaction between neighboring SA dimers.

Figure 2.5. Subtomogram averaging analysis of SA-TMV segments. (A)

Template map used for subtomogram averaging, with six turns of TMV density

(PDB: 4UDV) shown in blue and a sphere representing SA shown in red. The sphere denoting SA is positioned 2.4 nm away from the TMV surface, as this distance is the average calculated length of an 8-mer PEG. The total possible zone for SA extends from the surface of TMV to 12.8 nm beyond the surface, or

21.8 nm from the central axis of TMV. (B) Results of averaging 21 aligned

segments of SA-TMV with TMV density in blue and SA density in red. (C)

Overlay of the template in gray mesh and the averaged density in solid blue

(TMV) and red (SA).

2.4.3 Quantification of TMV coverage

To calculate the coverage of the volume surrounding TMV by the SA

coating, three complete TMV rods were cut out from the three-dimensional

tomogram, and each TMV was sliced into multiple one-pixel wide segments

(Figure 2.6A). At each segment, the integrated density of the image was

measured to determine the area of the TMV portion alone, as well as the entire

SA-TMV area (Figure 2.6B). The TMV area was subtracted from the SA-TMV area to obtain a value for the SA area in each segment. These values are plotted for each TMV rod in Figure 2.6C. The same was done for a single naked TMV

rod extracted from a separate cryo tomogram of unmodified TMV (see Figure

2.S1). The coverage plot shows that SA surrounds TMV in a relatively even

fashion. The total integrated area under each SA curve can be quantified by

comparing it to the area calculated for a single simulated SA molecule. This

analysis indicates that each SA-TMV has a different number of conjugated SA

molecules, ranging from approximately 180 to 210 SAs per TMV. This is in good

agreement with the number calculated by the PAGE densitometry analysis (140-

200 SAs per TMV, see Figure 2.1). These results demonstrate that each SA-

TMV has a different level of camouflage coating.

Figure 2.6. Analysis of SA volume coverage of TMV rod. (A) Schematic representation of coverage area analysis. Each TMV rod is broken up into one- pixel wide segments, and the areas filled with density within the larger TMV+SA circle (red) as well as within the smaller TMV-only oval (cyan) are measured.

Subtracting the TMV-only area from TMV+SA area gives a measurement of the

SA area at each segment along the TMV rod. (B) Experimental example of a one-pixel wide segment from an SA-TMV tomogram with the measurement circle and oval shown. Note that an oval is used to measure the TMV-only area since the tomograms are not corrected for the missing wedge, leading to apparent elongation of TMV along one axis of its cross section. (C) Plot of the coverage for three SA-TMV rods (red) compared to a naked control TMV rod (blue).

The cryo-tomogram also facilitates a rough calculation of the percent

coverage of the TMV surface. The extent of uncoated and coated TMV surface can be quantified, analogous to the amounts of blue and red surface, respectively, as seen in Figure 2.3C, right. Three SA-TMV rods were analyzed and found to have 30-45% surface coverage. These numbers are within the range calculated from the dot blot densitometric analysis (20-55%), further suggesting that the SAs tend to orient in a random manner.

To understand the biological relevance of this coverage, atomic models were built for SA-TMV segments and modeled with both intact IgG and Fab fragments. The structural modeling indicates that the spacing and density of SA on the TMV surface would likely prevent antibody-antigen interactions with the

TMV surface – due to steric hindrance: i) SA is too densely packed to allow a

Fab arm to easily find an opening large enough to access the TMV surface, and ii) the Fab arm is too short to penetrate between SAs at their average distance from the TMV surface to reach TMV epitopes. This is assuming an average Fab length of 6.5 nm (Stanfield and Wilson, 2014). In rare cases, where the SA dimer is positioned next to the TMV surface and oriented such that its thinnest dimension (6 nm) is perpendicular to the TMV surface, then a Fab arm could reach an exposed TMV epitope. Mathematical calculations considering the average amount of exposed TMV surface area and the average antibody-antigen interaction patch size (Stanfield and Wilson, 2014) led to similar conclusions regarding SA shielding. The structural and mathematical modeling results correlate well with the experimental antibody binding data (Figure 2.1D),

indicating little antibody recognition for SA-TMV compared to strong recognition

for TMV.

2.5 Conclusions

This work demonstrates the utility of cryo-ET as a characterization tool for

stealth protein camouflaged NPs. When combined with subtomogram averaging,

cryo-ET provides a way to visualize individual SA molecules heterogeneously

displayed on the surface of TMV. While standard negative stain TEM can reveal

the presence of a stealth protein coating on the surface of TMV, individual SAs

are not resolved. Two dimensional cryo-TEM micrographs provide greater detail

and individual SA molecules are observed as dots. Three-dimensional cryo-ET

provides the possibility of quantitating surface coverage in a manner that is

complementary to gel densitometric analysis. Furthermore, subtomogram

averaging allows for analysis of the orientation of SA molecules relative to the

TMV surface and potentially for the detection of interactions between neighboring

SA molecules. With the rapid development of shielded nanoparticles for

biomedical applications, this study serves as a proof-of-concept for using cryo-ET

for visualization and quantitative structural analysis. It also opens the door for the

development of further structure-function studies through correlation of structural

morphology with in vivo efficacy.

2.6 Acknowledgements

We thank Dr. Sudheer Molugu for maintaining the TEM facility at the

Cleveland Center for Membrane and Structural Biology, as well as the CWRU

High Performance Computing staff for their assistance with campus computing

resources. We thank Christina Wege and her team (University of Stuttgart) for

providing TMV-lys. This work is supported by funding from the National Institutes

of Health (R01-CA202814 to NFS and T32 GM008803 to NMG).

2.7 Supplementary information

Figure 2.S1. Cryo-electron tomography analysis of naked TMV nanoparticles. (A) 0° tilt image of TMV-lys. (Scale bar = 100 nm). (B) Three-

dimensional representation of TMV tomogram. (Scale bar = 100 nm). (C) Single

TMV rod (red) with added 300 nm x 18 nm cylinder (blue) indicating that not

much density surrounds the bare TMV rod. (Scale bar = 25 nm).

MOVIE LEGENDS

Movie 2.S1. Aligned tilt series of SA-TMV processed using IMOD. Electron

dense 10 nm gold fiducials are visible. Movie is available at

http://www.rsc.org/suppdata/c6/nr/c6nr06948g/c6nr06948g1.mpg.

Movie 2.S2. Z-slices through reconstructed tomogram of SA-TMV processed using IMOD. Electron dense 10 nm gold fiducials are visible. Movie is available at http://www.rsc.org/suppdata/c6/nr/c6nr06948g/c6nr06948g2.mpg

Chapter 3: Immune recognition of SA-TMV

The material in this chapter is adapted from a manuscript submitted to

Advanced Healthcare Materials: Gulati NM*, Pitek AS*, Czapar AE, Stewart PL,

Steinmetz NF. The in vivo fates of plant viral nanoparticles camouflaged using self-proteins: overcoming immune recognition. If accepted, the material is copyright Wiley-VCH.

*Both authors contributed equally to this work.

As co-first author on this manuscript, I performed the cellular uptake and degradation experiments and wrote the manuscript. Dr. Andrzej S. Pitek prepared and characterized the SA-TMV constructs. We did the antibody production and recognition experiments together.

3.1 Abstract

Nanoparticles offer a promising avenue for targeted delivery of therapies.

To slow clearance, nanoparticles are frequently stealth-coated to prevent opsonization and immune recognition. Serum albumin (SA) has been used as a bio-inspired stealth coating. To develop this shielding strategy for clinical applications, it is critical to understand the interactions between the immune system and SA-camouflaged nanoparticles. This work investigates the in vivo

processing of SA-coated nanoparticles using tobacco mosaic virus (TMV) as a

model system. In comparing four different SA-formulations, the particles with high

SA coverage conjugated to TMV via a short linker performed the best at

preventing antibody recognition. Irrelevant of the coating chemistry, all

formulations led to similar levels of TMV-specific antibodies after repeat

administration in mice; importantly though, SA-specific antibodies were not

detected and the TMV-specific antibodies were unable to recognize shielded SA-

coated TMV. Upon uptake in macrophages, the shielding agent and nanoparticle

separate, where TMV trafficked to the lysosome and SA appears to recycle. The

distinct intracellular fates of the TMV carrier and SA shielding agent explain why

anti-TMV but not SA-specific antibodies are generated. This work characterizes

the outcomes of SA-camouflaged TMV after immune recognition, and highlights

the effectiveness of SA as a nanoparticle shielding agent.

3.2 Introduction

Nanoparticle-based delivery platforms have shown success in delivery of

therapeutic molecules to disease sites while limiting off-target side effects. Yet

despite the success of Doxil® (PEGylated liposomal doxorubicin) and other

clinically-approved nanoparticle formulations, barriers remain in the development

of these therapies. For a nanoparticle to successfully reach target tissue and

deliver its payload, it first must encounter and overcome the host’s defenses: the

immune system (Blanco et al., 2015). Nanoparticles can be recognized by

complement or other innate immune proteins as well as neutralizing antibodies,

which can adsorb to the surface of the nanoparticle to create a protein ‘corona’ and direct them for clearance through the mononuclear phagocyte system,

preventing them from ever reaching target tissue (Kokkinopoulou et al., 2017;

Lee et al., 2015b; Walczyk et al., 2010). This clearance can be reduced by

coating nanoparticles with shielding agents to create ‘stealth’ or ‘camouflage’

effects and thus overcome immune clearance.

The most common approach is to ‘stealth’ nanoparticles through

polyethylene glycol (PEG) coatings. This hydrophilic coating reduces

nanoparticle-protein interactions and thus reduces protein corona formation,

immune recognition, and premature clearance. However, PEGs come in different

shapes and sizes, and the effectiveness of shielding is dependent on the specific

PEG polymer chosen – there is no ‘one-fits-all’ solution and each PEGylated

nanoparticle system needs to be carefully optimized (Gref et al., 2000; Lee et al.,

2015a; Perry et al., 2012). Furthermore, with the prevalence of PEG in

commercial products, there has been an increase in PEG-specific antibodies

found in the human population (Ganson et al., 2016b; Garay et al., 2012). These

antibodies limit the effectiveness of the polymer shield, especially upon repeat administrations of a treatment (Ganson et al., 2006; Hershfield et al., 2014).

Newer strategies include alternate polymers as well as the use of ‘self’ coatings as stealth agents.

One avenue is to camouflage nanoparticles by coating them with ‘self’

molecules, including proteins and lipids, among others. It has been shown that

using a ‘self’ minimal peptide of CD47 prevented nanoparticle clearance by the

mononuclear phagocyte system by acting as an ‘eat me not’ signal (Rodriguez et

al., 2013). Nanoparticles formulated with self-coatings have been shown to

exhibit longer circulation times and decreased immune recognition compared to

PEGylated nanoparticles (Mariam et al., 2016; Pitek et al., 2016b; Rodriguez et

al., 2013). This highlights the potential of these ‘self’ coatings as nanoparticle

shielding agents.

Recently, our lab published results demonstrating the benefits of using

serum albumin (SA) as a ‘self’ protein camouflage (Pitek et al., 2016b). SA,

which functions to transport hydrophobic molecules in the blood, is the most

abundant protein in plasma, making it an ideal candidate for ‘self’ camouflage. As

a nanocarrier platform, we used tobacco mosaic virus (TMV). The nucleoprotein

components of TMV form a 300 nm x 18 nm rigid rod made up of 2130 identical

coat proteins. The TMV platform nanotechnology has been studied for drug

delivery and imaging applications (Bruckman et al., 2015b; Bruckman et al.,

2014b; Czapar et al., 2016). Tobacco mosaic virus and other plant viral nanoparticles are noninfectious in humans, making it an innovative candidate for therapeutic applications. However, like other proteinaceous nanoparticles, naked

TMV has a short half-life (3.5 minutes) in naïve mice (Bruckman et al., 2014a).

Moreover, anti-TMV antibodies can be found in human serum of both smokers

and nonsmokers due to the presence of TMV in tobacco and food products

(Asselin, 1979; Balique et al., 2012; Cai et al., 2003; Gottula and Fuchs, 2009;

Liu et al., 2013; Pinnow et al., 1981; Sakata et al., 1997). Therefore, it is

imperative that an effective stealth technology is applied for TMV-based contrast agents and therapeutic drug delivery.

We have shown that compared to their PEGylated counterparts, SA-TMV

nanoparticles had reduced antibody recognition and longer circulating half-lives

(Pitek et al., 2016b). Structural characterization of SA-camouflaged TMV indicated dense coverage of SA on the TMV surface and random orientations of

SA with respect to the TMV coat proteins (CPs) (Gulati et al., 2017). Here, we set out to characterize the SA-coated TMV nanoparticle system further and to gain insight into the structure-function relationship of SA coatings, displayed at distinct density and conjugated via short vs. long PEG linkers. We elucidate immune recognition and intracellular processing pathways of SA-coated TMV nanoparticles using a combination of tissue culture, in vivo and ex vivo assays.

3.3 Materials and Methods

3.3.1 Virus propagation and purification

The following method was performed by Dr. Andrzej S. Pitek. Tobacco mosaic virus or its T158K mutant (TMV-lys) (Geiger et al., 2013) was propagated by mechanical inoculation using 5-10 µg virus per leaf. Viruses were isolated and purified as previously described to yield approximately 1 mg virus / 1 g infected leaf material (Bruckman and Steinmetz, 2014).

3.3.2 TMV sCy5 labeling

The following method was performed by Dr. Andrzej S. Pitek. TMV particles were internally labeled at glutamic acid residues with sCy5 fluorescent dyes. First, glutamic acids were modified using EDC coupling with alkynes using

1 mg/mL TMV, 100 equivalents of propargylamine (Sigma-Aldrich) per capsid protein, and 50 equivalents EDC (25 equivalents added at 0 and 16 hours) in 100 mM HEPES buffer, pH 7.4 reacted for 20 hours at RT. Alkyne-azide click chemistry was then performed using 1.5 equivalent of sCy5-azide (Lumiprobe) per coat protein with 2 mg/mL TMV in the presence of 1 mM CuSO4 (Fisher), 2

mM AMG (Fisher), 2 mM AsC (Fisher) in 10 mM potassium phosphate buffer, pH

7.5 on ice for 30 min. TMV was purified by ultracentrifugation at 42,000 rpm for 3

hours on a 40% (w/v) sucrose cushion in 10 mM potassium phosphate buffer, pH

7.5.

3.3.3 TMV external conjugation

The following method was performed by Dr. Andrzej S. Pitek. Human or

mouse SA (Sigma-Aldrich or Bioworld respectively) was externally conjugated to

TMV-lys using combination of intermediate PEG linkers in multistep reaction.

First, SA was conjugated with NHS-PEG4-SAT (ThermoFisher) at 1:1 ratio in

phosphate buffered saline (PBS; 0.01 M Na2HPO4, 0.0018 M KH2PO4, 0.0027 M

KCl and 0.137 M NaCl, pH 7.4) containing 10% (v/v) DMSO overnight at RT. To de-protect the thiol group, 0.5 M hydroxylamine, 25 mM EDTA in PBS, pH 7.2-

7.5 was added at a final concentration of 10% (v/v) and incubated for 2 hours at

RT. Separately, TMV-lys was conjugated with maleimide-PEG4-NHS

(ThermoFisher) or maleimide-PEG24-NHS (ThermoFisher) at 10 (for high

coverage particles) or 3 (for low coverage particles) equivalents of PEG per coat

protein in 10 mM potassium phosphate buffer, pH 7.4 containing 10% (v/v)

DMSO for 2 hours at RT. The resulting maleimide-PEG-TMV conjugates were

purified twice through PD MiniTrap G-25 desalting columns (GE) and divided into

two aliquots: a) control PEG-TMV nanoparticles used in dot blot experiments,

and b) PEG-TMV nanoparticles subsequently combined with previously prepared

SA-PEG-SH at 6 (for high coverage particles) or 2 (for low coverage particles)

equivalents per CP, and reacted overnight at RT to yield SA-PEG8/28-TMV particles of variable SA coverage. Both a) and b) were quenched by addition of excess glycine and L-cysteine for 1 hour at RT. The resulting constructs were then purified by ultracentrifugation at 55,000 rpm for 3 hours on a 40% (w/v) sucrose cushion.

3.3.4 SDS-PAGE analysis

Samples were denatured by boiling at 100ºC for 5 minutes in gel loading buffer (62.5 mM Tris HCl pH 6.8, 2% w/v SDS, 10% v/v glycerol, 0.01 % w/v bromophenol blue, 10% v/v 2-β-mercaptoethanol). 20 µg TMV was loaded on 4-

12% Bis/Tris NuPAGE gels (ThermoFisher Scientific) and separated at 200 V for

35 minutes in MOPS running buffer. When fluorescent particles were used, the sCy5-labelled TMV CP bands were detected using a Maestro fluorescence imaging system. Subsequently, the gels were stained using Coomassie Blue, visualized using an AlphaImager imaging system (Biosciences) and underwent densitometry analysis using ImageJ software (Schneider et al., 2012).

3.3.5 Western blot analysis

The following method was performed by Dr. Andrzej S. Pitek. Samples

prepared by SDS-PAGE as described above were transferred to nitrocellulose

membranes at 30 V for 1 hour. Membranes were blocked with 5% (w/v) milk in

TBS-Tween (150 mM NaCl, 10 mM Tris HCl, 0.1% (v/v) Tween-20, pH 7.5) for 1

hour at RT. Membranes were then incubated with 0.5 µg/mL rabbit polyclonal

anti-TMV antibody (custom-made, Pacific Immunology) or rabbit polyclonal anti-

albumin antibody (NBP1-32458, Novus Biologicals) in 5% (w/v) milk in TBS-

Tween overnight at 4ºC. Membranes were then washed three times for 10 min

each in TBS-Tween, followed by incubation with 1 µg/mL alkaline phosphatase-

conjugated goat anti-rabbit antibody (G21079, Thermo Fisher Scientific) in 5%

(w/v) milk in TBS-Tween for 1 hour at RT, followed by three washes for 10 min

each in TBS-Tween and one 5 min wash in dH2O. Antibody binding was

visualized using Novex AP Chromogenic Substrate (BCIP/NBT; Invitrogen).

3.3.6 Negative stain transmission electron microscopy

3 µL sample at a concentration of 0.5 mg/mL was applied to glow-

discharged carbon coated 200 mesh copper grids. After 1 minute, excess sample

was removed. Grids were then washed two times in dH2O for 1 minute each,

then stained two times for 30 seconds each with 2% (w/v) uranyl acetate. Grids were then blotted until dry with Whatman 1 blotting paper. All grids were imaged on a JEOL 2200FS 200 kV transmission electron microscope equipped with an energy filter using a Tietz TVIPS 4k × 4k CMOS camera.

3.3.7 Dot blots

Blots were prepared by spotting 1 µL of 150 µg/mL rabbit polyclonal anti-

TMV antibody (custom-made, Pacific Immunology), rabbit polyclonal anti-CPMV

antibody (custom-made, Pacific Immunology), rabbit monoclonal anti-PEG

antibody (AB133471, Abcam), or 1:200 diluted mouse plasma on a nitrocellulose

membrane, after the nitrocellulose membrane was equilibrated in PBS. Blots

were incubated in 5% (w/v) milk in PBS at RT for 1 hour, then washed three

times for 5 minutes each in PBS. Blots were then incubated with fluorescently-

labeled particles for 2.5 hours at RT, then washed three times for 5 minutes each

in PBS. Blots were dried and imaged using a Maestro imaging system. Blots

were analyzed using ImageJ.

3.3.8 In vivo studies and ELISA assays

Animals were housed, bred, and handled in the Case Western Reserve

University Animal Resource Center in accordance with approved Institutional

Animal Care and Use Committee experimental protocols. Balb/C mice were

injected via tail vein with 200 µg SA-TMV constructs or TMV or SA + TMV on

days 0, 7, 14, 21, and 28. Blood was collected retro-orbitally in heparin-coated

tubes (ThermoFisher Scientific) on days 0, 14, 28 prior to injection, and on day

56. Plasma was separated by centrifugation at 2000 rcf for 10 min, then stored at

-20°C until analysis.

TMV-specific and SA-specific plasma IgG antibodies were determined by

ELISA analysis. 96-well Nunc Polysorb Immuno Plates (ThermoFisher Scientific)

were coated with 1 µg SA or TMV/well in coating buffer (0.05 M Na2CO3, 0.05 M

NaHCO3, 0.015 M NaN3 in dH2O, pH 9.6) and incubated overnight at 4°C. After

coating, wells were blocked with blocking buffer (2.5% (w/v) milk, 25% FBS in

PBS, pH 7.4) at 37°C for 1 hr, followed by four washes with washing buffer (0.1%

Tween-2 in PBS, pH 7.4). 100 µL of serially diluted plasma (1:200, 1:500, 1:1000,

or 1:5000) in blocking buffer was then added to each well and incubated at 37°C

for 2 hr. Wells were then washed five times with washing buffer, followed by the

addition of 100 µL alkaline phosphate-labeled goat anti-mouse IgG (Life

Technologies at 1:3000 in blocking buffer and incubated at 37°C for 1 hr. Wells

were then washed five times with washing buffer, then developed by adding 100

µL one-step PNPP substrate (ThermoFisher Scientific) for 10 min at 4°C. The

reactions were stopped by addition of 100 µL 2 M NaOH and absorbance was

measured at 405 nm using a Tecan Infinity 200 microplate reader.

3.3.9 Confocal microscopy

Confluent RAW 264.7 murine macrophage cells were cultured in DMEM

(Corning) containing 10% (v/v) fetal bovine serum and 1% (v/v)

penicillin/streptomycin at 37°C and 5% CO2. Confluent cells were removed using

0.05% (w/v) trypsin-EDTA (Corning) and cultured to glass coverslips in 24-well

untreated plates at 20,000 cells/well overnight. Cells were incubated for 6, 12, 24

hr, or 12 hr followed by wash and incubation with media for 12 hr. Following

incubation, cells were fixed in DPBS containing 5% (v/v) paraformaldehyde and

0.3% (v/v) gluteraldehyde for 10 min at room temperature. After fixation, cells

were permeabilized with 0.2% Triton-X 100 for 2 min followed by blocking in 5%

(v/v) goat serum, 0.5% (v/v) Fc block in DPBS for 45 min at room temperature.

Cells were then stained with 1:200 diluted primary antibodies in 5% (v/v) goat

serum, 0.5% (v/v) Fc block in DPBS for 1 hr at room temperature and 1:1000

diluted secondary antibodies in 5% (v/v) goat serum, 0.5% (v/v) Fc block in

DPBS for 1 hr at room temperature. Lysosomes were stained using goat anti-

mouse LAMP-1 conjugated to AlexaFluor 488 (328609; BioLegend). TMV was

stained using rabbit anti-TMV primary antibody and AlexaFluor 647-conjugated

goat anti-rabbit secondary antibody (A-21244; ThermoFisher Scientific). SA was

stained using chicken anti-SA primary antibody (AB106582, Abcam) and

AlexaFluor 555-conjugated goat anti-chicken secondary antibody (AB150174,

Abcam). Cells were washed three times with DPBS (Corning) between steps.

Coverslips were mounted onto slides using Fluoroshield with DAPI (Sigma-

Aldrich). Slides were imaged using a Leica TCS SPE confocal laser scanning

microscope and the data were processed with ImageJ.

3.3.10 Lysosomal extraction experiments

Animals were housed, bred, and handled in the Case Western Reserve

University Animal Resource Center in accordance with approved Institutional

Animal Care and Use Committee experimental protocols. Female FVB mice were

starved overnight and euthanized using carbon dioxide inhalation. Livers were

removed and lysosomes were extracted from them using the Lysosome Isolation

Kit (Sigma-Aldrich), based on the protocol provided with the kit, however no

protease inhibitors were used. Following lysosomal extraction, the presence of

lysosomal enzymes was determined using the Acid Phosphatase Kit (Sigma-

Aldrich). Lysosomal extract was then incubated with bovine serum albumin (BSA)

as an internal control to determine enzymatic activity of lysosomal proteases, as

lysosomal extract has been shown previously to degrade BSA (Wen et al., 2015).

After confirming enzymatic activity, Cy5-labeled high coverage, short linker SA-

TMV particles (2) were incubated in lysosomal extract at pH 5 or PBS at pH 5

(hydrochloric acid was used to adjust the pH) at a concentration of 1 mg/mL.

Samples were incubated at 37°C under gentle agitation and aliquots were taken at 6, 24, and 48 hrs and characterized by SDS-PAGE.

3.4 Results and Discussion

3.4.1 Synthesis and characterization of SA-TMV constructs

TMV T158K mutant (TMV-lys) (Geiger et al., 2013) was propagated and purified from Nicotiana benthamiana plants. A multistep protocol was used to produce a set of SA-TMV formulations (Figure 3.1). TMV-lys was incubated with maleimide (MAL) and N-hydroxysuccinimide (NHS) dually-functionalized PEG4 or

PEG24 linkers (MAL-PEG4/24-NHS). For high coverage particles, 10 equivalents of

MAL-PEG4/24-NHS were reacted per TMV-lys CP. For low coverage particles, 3

equivalents of MAL-PEG4/24-NHS were reacted per TMV-lys CP. Four resulting

PEG-TMVs were produced: low coverage MAL-PEG4-TMV (1’), high coverage

MAL-PEG4-TMV (2’), low coverage MAL-PEG24-TMV (3’), and high coverage

MAL-PEG24-TMV (4’). SA was reacted with NHS-PEG4-SH separately to produce

thiol-activated SA-PEG4-SH. The products from these two reactions were then

combined to react via Michael addition to form SA-PEG8-TMV or SA-PEG28-TMV.

To produce low coverage particles, 2 equivalents of SA-PEG4-SH were reacted

with 1’ or 2’. To produce high coverage particles, 6 equivalents of SA-PEG4-SH were reacted with 3’ or 4’. Four SA-TMV constructs were produced: particles with low SA coverage and a short 8-mer PEG linker (1), particles with high SA coverage and short linker (2), particles with low SA coverage and a long 28-mer

PEG linker (3), and particles with high SA coverage and a long PEG linker (4), as depicted in Figure 3.1.

Figure 3.1. Cartoon depiction of production of four PEG-TMV and four SA- TMV constructs, with varying PEG linker length and SA coverage quantity.

TMV shown in gray, PEG linker shown in cyan, and serum albumin shown in red.

Conjugation of PEG4 or PEG24 to TMV results in four PEG-TMV constructs: 1’

(short linker, low coverage), 2’ (short linker, high coverage), 3’ (long linker, low coverage), and 4’ (long linker, high coverage). PEG-TMV constructs are reacted with PEG4-conjugated SA to yield four SA-TMV constructs: 1 (short linker, low

coverage) as shown in the orange box, 2 (short linker, high coverage) in the blue

box, 3 (long linker, low coverage) in the purple box, and 4 (long linker, high

coverage) in the green box.

For each construct, the fully extended and average PEG linker length, as determined by Flory dimension (RF), were calculated and are reported in Table

3.1 (Lee et al., 2015a). RF is an estimation of the hydrodynamic radius of the

PEG linker, which was doubled to estimate the average PEG linker length.

Conjugation and structural integrity of the SA-TMV formulations were confirmed

by SDS-polyacrylamide gel electrophoresis (SDS-PAGE; Figure 3.2A), western

blot analysis (WB; Figure 3.2B), and negative stain transmission electron microscopy (TEM; Figure 3.2C). The SDS-PAGE gel confirms the production of

SA-TMV-lys coat protein (CP) conjugates of estimated molecular weight 83.5

kDa and above, corresponding to linked TMV-lys CP (17 kDa) and SA (66.5 kDa)

(Figure 3.2A). As expected, these higher molecular weight bands are not

present in the TMV-lys and SA controls. The multiple band pattern of the SA-

TMV conjugate may be explained by the fact that SA has multiple surface-

exposed lysines available for conjugation with PEG; thus, individual SAs could

display multiple PEG linkers. This could lead to polydisperse samples in which

one SA may be conjugated to more than one neighboring TMV-lys CP

simultaneously. Alternatively, modified CPs could engage through non-covalent

interactions, as has been seen previously for other modified TMV formulations

(Altintoprak et al., 2015; Bruckman et al., 2015a). The SDS-PAGE results are

supported by WB probed with antibodies against TMV (Figure 3.2B, pseudo-

colored green) and against SA (Figure 3.2B, pseudo-colored red). Overlaying

the two WBs together confirms that TMV and SA are present in the same high

molecular weight (SA-conjugated TMV CP) bands.

Figure 3.2. Characterization of SA-TMV constructs. A) Representative SDS-

PAGE gel. B) Overlaid WBs against TMV (green) and SA (red). C) Negative stain

TEM micrographs showing intact SA-TMV particles. The four SA-TMV constructs 87

are: 1 (short linker, low coverage), 2 (short linker, high coverage), 3 (long linker,

low coverage). and 4 (long linker, high coverage). Scale bars = 100 nm. The data

in A and B was provided by Dr. Andrzej S. Pitek.

As expected, the SA-conjugated TMV CP bands on the SDS-PAGE gel

are more intense for the high coverage particles (2 and 4) as compared to low

coverage particles (1 and 3), indicating a higher degree of coverage. Using

ImageJ for densitometry analysis of three SDS-PAGE gels corresponding to

three distinct batches of SA-TMV particles, the average amount of SA per TMV

was calculated for each formulation, as reported in Table 3.1. In preparing the particles, the ratios of SA to PEG-TMV were kept the same between high and low constructs, regardless of linker length (8x SA per TMV CP and 2x SA per

TMV CP, respectively). However, there is approximately seven times more SA conjugation for 2 than 1 (the two short linker constructs), but only four times more

SA conjugation for 4 than 3 (the two long linker constructs). Thus, the excess of reactive molecules is not the only limiting factor during conjugation. Furthermore, repeated conjugation experiments reveal that high coverage particles prepared using the long linker show reduced SA conjugation, as compared to those prepared with the short linker. Specifically, 2, which has approximately 300 SAs per TMV, has almost three times as much conjugated SA as 4, which has approximately 110 SAs per TMV. This could be because the long linker is more flexible and can adopt a tangled mushroom conformation, therefore its reactive end may not be surface exposed. For low coverage particles, 1 and 3 have

similar quantities of SA, with approximately 40 and 30 SAs per TMV,

respectively.

Table 3.1. Physical characteristics of SA coverage and PEG linkers of SA- TMV constructs.

SA-TMV Number of PEG Average Fully SA # SAs TMV construct PEG molecular PEG extended concentration per surface monomers weight diameter PEG (µg SA / mg TMVd) coverage (Da) (nm)a) length TMV)c) by SAe) (nm)b)

1 8 363 2.44 2.8 78.4 ± 23.1 44 ±13 7-12%

2 8 363 2.44 2.8 534.9 ± 146.8 301 ± 83 49-83%

3 28 1275 5.17 9.8 53.0 ± 20.3 30 ± 12 5-8%

4 28 1275 5.17 9.8 202.6 ±108.5 114 ± 61 19-31%

a)Determined by doubling RF of each PEG construct to get diameter length.;

b)Calculated by multiplying the number of PEG monomers by the length of a

single monomer (0.35 nm); c)Estimated by densitometry analysis of bands with

molecular weight >64 kDa from three SDS-PAGE gels of different batches of SA-

TMV constructs; d)Calculated using the SA concentration based on three SDS-

PAGE gels and the molecular weights of SA and TMV; e)Calculated using the

number of SAs per TMV and the cross-sectional area of SA (28-47 nm2) compared to the approximate surface area of TMV (17,040 nm2).

In both the SDS-PAGE gel and WB, unconjugated TMV CPs and SA are present in the SA-TMV constructs (Figure 3.2A, B). Free SA was present for all

SA-TMV formulations, and has been reported previously for SA-TMV (Pitek et al.,

2016b). This is likely due to non-covalently attached SA that could be adsorbed to TMV or forming multimers with covalently attached SA. While we cannot rule out that some amount of the non-conjugated SA is free SA that is not associated with TMV, this is unlikely given the extensive washing steps and ultracentrifugation procedures employed during the purification steps. The amount of free SA relative to the amount of TMV for each nanoparticle construct is as follows: 52.2 + 14.4 µg/mg (1), 151.9 + 44.8 µg/mg (2), 45.3 + 22.5 µg/mg

(3), and 94.4 + 18.5 µg/mg (4). The amount of free SA for low coverage particles

is similar to the amount of covalently-bound SA, supporting the possibility of SA

dimerization. For the high coverage particles, the amount of free SA is less than

the amount of conjugated SA, so it could be that not all SAs dimerize on the TMV surface or that the high coverage allows TMV-bound SAs to dimerize with one another. Regardless, these additional SA molecules could enhance camouflaging by sealing any gaps between covalently attached SAs on the TMV surface. For the long linker constructs, there is also an additional band of PEG24-conjugated

TMV CPs, with approximately 28.2 + 0.7% of TMV CPs being PEGylated in 3

and 44.0 + 5.5% of TMV CPs being PEGylated for 4. This indicates that not all

TMV CPs that are PEGylated are also conjugated with SA. Based on the PEG

coverage and the RF values reported in Table 3.1, we have determined that

these PEGs likely adopt a mushroom conformation on the TMV surface.

TEM of the nanoparticles shows that all SA-TMV constructs remain intact after conjugation, and that the characteristic 300 nm by 18 nm rigid rod structure of TMV is preserved. Additionally, the high coverage particles 2 and 4 show the characteristic rough TMV surface (Figure 3.2C); this is consistent with our previous TEM and structural studies (Gulati et al., 2017). It is apparent across the entire surface of 2, while 4 appears rough in patches. The low coverage

particles 1 and 3 do not have a rough TMV surface by TEM, likely because the

SA coverage is not dense enough to markedly affect the appearance of the

negative stain (Figure 3.2C).

3.4.2 Effects of SA coverage and linker length on shielding TMV

To determine the effectiveness of different SA coating strategies on

immune recognition, dot blots using fluorescently-labeled TMV particles and

TMV-specific antibodies were performed (Figure 3.3A). To do so, nitrocellulose

membranes were spotted in triplicate with 1 µL anti-TMV and 1 µL anti-cowpea

mosaic virus (CPMV) antibody at 150 µg/mL. CPMV is an icosahedral plant virus

and its antibodies act as a negative control. There are no structural similarities

between CPMV and TMV and there is no cross-reactivity between the antibodies.

After application of antibodies, membranes were blocked with 5% (w/v) milk, and

then incubated with fluorescent Cy5-labeled nanoparticles. SA-TMV constructs

(1-4) as well as the PEG-TMVs from which they were produced (1’-4’) were tested. After incubation with fluorescent SA-TMV and PEG-TMV particles, membranes were washed and dried, then imaged for fluorescence. If the

particles interact with the antibodies, there would be a strong fluorescent signal where antibody-binding occurred.

Figure 3.3. Antibody recognition of SA-TMV and PEG-TMV constructs. A)

Dot blots of anti-TMV or anti-CPMV recognition of fluorescent SA-TMV constructs (1-4) and their corresponding PEG-TMV control constructs (1’-4’). B)

Densitometry quantification of fluorescent TMV signal corresponding to antibody recognition by anti-TMV antibodies from A. C) Dot blots of anti-TMV or anti-PEG recognition of fluorescent high coverage SA-TMV constructs (2 and 4) and their corresponding PEG-TMV control constructs (2’ and 4’). D) Densitometry quantification of fluorescent TMV signal corresponding to antibody recognition by anti-TMV and anti-PEG antibodies from C.

Quantification of the dot blots indicates that for all cases, SA-TMV has reduced antibody recognition compared to its correlated PEG-TMV construct, as has been reported previously (Figure 3.3B) (Gulati et al., 2017; Pitek et al.,

2016b). We have also previously shown that both SA-TMV and PEG-TMV exhibit shielding properties and have significantly reduced binding compared to ‘naked’

TMV controls. Naked TMV samples were omitted in the data shown and we refer the reader to our previous paper (Pitek et al., 2016b). Here, the SA-coated TMV particles display shielding effects from best to worst as follows: 2, 4 >>>> 3 > 1

>>>> 4’ >>> 2’ > 3’ >> 1’. In all cases, when comparing particles with similar linker lengths high coverage was better, i.e. high coverage provides more efficient shielding and lower antibody recognition, compared to low coverage.

Previous structural studies of SA-TMV indicated that SA shielding of TMV is, at least in part, due to steric hindrance preventing TMV-specific antibodies from reaching beyond the SA shield and also preventing tumbling of antibodies along the surface.(Gulati et al., 2017) Therefore it seems logical that by lowering the number of shielding molecules on the TMV surface, there is more accessible area for epitope recognition, as well as antibody tumbling and binding.

Between PEG-TMV constructs, a longer PEG linker better shielded antibody recognition than a shorter linker in both cases, as has been reported previously for other plant viral nanoparticles (Lee et al., 2015a). For the SA-TMV constructs, the low coverage particles indicate that for similar SA coverage, a longer linker length is better (3 > 1). It is difficult to compare the high coverage particles, as 2 has approximately twice as much SA conjugated as 4, but there is

no significant difference in antibody recognition among high coverage particles.

This suggests that perhaps there is a threshold for the amount of necessary shielding agents on the TMV surface.

To investigate the effects of the PEG linker on shielding, it is important to understand whether or not the PEG linker is surface-exposed, which could further prevent access to the TMV surface. To do so, dot blots were performed to assay the ability of PEG-specific antibodies to recognize the SA-TMV and PEG-

TMV constructs (Figure 3.3C, D). High coverage particles were chosen for this experiment, as previous dot blots indicated that the high coverage particles showed the best reduction in antibody recognition, regardless of linker length.

While both the SA coatings on 2 and 4 provide protection from TMV-specific antibodies, only the coating on 2 protects against PEG-specific antibodies. Thus, due to the exposed PEG linker on 4, the particles must be shielded by both SA and PEG. Using the surface area of a TMV CP (8 nm2) and the cross-section of

SA (28-47 nm2), it is estimated that 49-83% of the TMV nanoparticle surface is coated by SA for 2 (Gulati et al., 2017). In comparison, 4 has only 19-31%

coverage. However, densitometry analysis indicates that 44% of TMV CPs are

PEGylated in 4. Using RF to calculate the size of PEG in a mushroom conformation, we determined that the additional PEG could fully cover the remaining TMV surface for 4. As both show successful protection against anti-

TMV antibodies, this suggests that the 49-83% coverage provided by SA in 2 is

sufficient for functional TMV camouflaging. Since 4 can be recognized by anti-

PEG antibodies this limits use of this construct to clinical applications in which

patients do not have pre-existing PEG antibodies. Overall, the in vitro antibody recognition data suggests that 2 would be the most clinically applicable construct.

3.4.3 Antibody response to multiple administrations of SA-TMV constructs

To address the immune response of differently coated SA-TMV nanoparticles in vivo, the antibody production of Balb/C mice in response to

these particles was assessed. Mice (n = 3) were intravenously injected with

either SA-TMV nanoparticles once per week for four weeks, mimicking a clinical setting in which patients may receive a weekly dosing with nanoparticles delivering chemotherapeutic regimen (Figure 3.4A). Blood was drawn via eye

bleed at days 0, 14, 28, and 56 to determine TMV-specific and SA-specific IgG

antibody levels resulting from repeat administration. ELISA measurements using

TMV or SA coated plates indicate that for all SA-TMV constructs, repeat

administrations produce a robust antibody response to TMV but not to SA

(Figure 3.4B). This is critical for the clinical applicability of these particles moving

forward, as an SA-specific antibody response would not only respond to the

stealth coating on SA-TMV particles but could also respond to naturally occurring

SA in the plasma.

Figure 3.4. Quantification of antibodies generated after repeated administration of SA-TMV constructs. A) Schematic of multiple administration and bleeding schedules (left); preparation of blood plasma and analysis by

ELISA (right). B) ELISA results showing production of TMV-specific but not SA-

specific antibodies. C) ELISA results showing increasing levels of TMV-specific antibodies after multiple administrations.

To investigate the differences between TMV-specific IgG antibody titers in

response to the different SA-TMV constructs, serial dilutions of plasma were

analyzed by ELISA on TMV-coated plates (Figure 3.4C). As expected, mice

showed no preexisting TMV-specific IgG antibodies prior to the first treatment

(day 0). At day 14 after two weeks of treatment, TMV and all SA-TMV constructs

showed a similar TMV-reactive IgG response. More significant IgG levels were

detected one week after the full treatment regimen at day 28, with all nanoparticles eliciting similar TMV-specific antibody responses. After the injections were stopped, mice retained circulating TMV-reactive IgGs, as seen on day 56. Taken together, the results from the ELISA experiments indicate that repeated administration of SA-camouflaged TMV nanoparticles leads to a robust anti-TMV immune response and this TMV-reactive IgG response occurs regardless of SA-coating formulation. Nevertheless, antibodies against the ‘self’ protein SA were not produced in the process, which indicates distinct processing of the TMV carrier vs. the SA coat. Given the prevalence of TMV antibodies in

the human population, in a clinical setting one would have to anticipate that a patient may already have TMV antibodies. Therefore, the key question is not

whether anti-TMV antibodies are produced, but whether or not these anti-TMV antibodies recognize the SA-TMV nanoparticle formulation. Accordingly, we investigated whether anti-TMV IgGs produced in response to repeat administration of SA-TMV would recognize the SA-TMV constructs. To do so, dot blots were performed with the diluted mouse plasma (1:200 dilution, as was used for ELISA experiments in Figure 3.4B) matched to the particles administered, i.e.

100

plasma from the mice administered 1 would be tested against 1, etc. Unlike bare

TMV particles, all SA-TMV constructs showed little antibody recognition (Figure

3.5A). As shown for purified TMV-specific antibodies, mouse plasma recognition was significantly reduced by SA coating (Figure 4B).

101

Figure 3.5. Analysis of SA-TMV recognition by antibodies in plasma from immunized mice. A) Dot blots indicating recognition of SA-TMV constructs by

antibodies in 1:200 diluted plasma from immunized mice. B) Densitometry quantification of fluorescent TMV signal from A corresponding to recognition by

anti-TMV antibodies in diluted plasma.

102

3.4.4 Trafficking of SA-TMV nanoparticles upon cell uptake

Confocal microscopy was used to understand immune cell processing of the SA-TMV nanoparticles, after cell uptake in macrophages. Since the antibody production experiments indicated that all formulations lead to the production of circulating anti-TMV antibodies but not anti-SA antibodies, we hypothesized that after nanoparticles are phagocytosed, SA and TMV undergo different fates. To test this, RAW 264.7 murine macrophage cells were incubated with 2, as this construct was identified as the most effective at ‘stealthing’ TMV. A non- conjugated mixture of SA and TMV served as a control. A time course experiment was performed, in which particles were incubated with cells for 0, 6,

12, 24 hours, or 12 hours followed by washing and an additional 12 hours of incubation (Figure 3.6A). After incubation, cells were fixed and stained with DAPI

(blue), AlexaFluor 488-conjugated LAMP-1 to stain for lysosomes (red), or antibodies against TMV (green) and SA (cyan) with fluorescent secondary antibodies. The results show that SA-TMV enters macrophage cells and that

TMV and the SA coating undergo different fates as was hypothesized. TMV appears within the lysosome by 12 hours and remains there until at least 24 hours after incubation (later time points were not tested). In contrast, SA appears to localize on the cell surface. For the control sample of SA mixed with TMV,

TMV was detected within the lysosome by 6 hours and remained there for 24 hours. SA from the mixed sample is only weakly visualized in macrophages.

103

Figure 3.6. Confocal microscopy of SA-TMV uptake by RAW 264.7 murine macrophage cells. A) Uptake time course of SA-conjugated TMV or mixed SA 104

and TMV (SA+TMV). SA labeled in cyan, TMV in green, lysosome LAMP-1 marker in red, and nucleus (DAPI) in blue. B) Colocalization analysis of the intracellular distribution of SA-TMV in RAW 264.7 cells at 6 hour time point. C) Z- stack cross-section of RAW 264.7 cell exposed to SA-TMV at 6 hour time point.

D) Colocalization analysis of the intracellular distribution of SA-TMV in RAW

264.7 cells at 12 hour time point. E) Z-stack cross-section of RAW 264.7 cell exposed to SA-TMV at 12 hour time point. Scale bars = 5 µm.

105

The discrepancy in the time before TMV is detected within the lysosome comparing SA-TMV vs. SA mixed with TMV is likely due to the shielding provided by the SA conjugation of the SA-TMV construct. Presumably the SA conjugation prevents TMV antibodies from binding and staining TMV until the 12 hour time point. By this point, the data indicates that the SA and TMV components of the construct have separated, thus allowing for staining of TMV in the lysosome. By

12 hours, SA is visualized at the periphery of the cell and remains there until at least 24 hours. As TMV is localized to the lysosome and SA is observed at the cell periphery there is no apparent colocalization of SA and TMV for the SA-TMV construct. This suggests that SA is cleaved and removed from the TMV carrier within the endolysosomal compartment and SA is recycled to the cell surface.

While SA remains conjugated to TMV, the virus is shielded and not visualized as colocalized with SA. Once SA is cleaved from SA-TMV, SA is likely trafficked away from the endolysosomal compartment.

To further investigate the fate of SA within macrophages, we compared the confocal results of the 12 hour incubation time point with that of 12 hours incubation followed by washing and an additional 12 hours of incubation (12 with

12 wash). The washing step is designed to remove any SA-TMV that has not already been taken up by the macrophages. For SA-TMV the results of the 12 with 12 wash condition indicate that the amount of SA is reduced compared to the 12 hour time point, suggesting either breakdown or cell export of SA during the 12 hours after the wash. In contrast, for SA mixed with TMV, there is never much intracellular signal for SA, suggesting that unconjugated SA is not as

106

readily taken up by macrophages as the SA-TMV construct. This observation correlates with previous findings that SA is taken up more readily when in a multimeric form (Schmidtke and Unanue, 1971).

The data presented in Figure 5A suggests that SA may be cleaved from

TMV between the 6 and 12 hour time points. To clarify the processing of the SA-

TMV construct 2, we investigated the colocalization of the SA, TMV, and the lysosome in more detail for time points of 6 hours (Figure 3.6B, C) and 12 hours

(Figure 3.6D, E). Again, at the 6 hour time point little to no TMV is detected, but

SA is detected, suggesting SA is still bound to the TMV surface and providing shielding. By measuring colocalization using the Mander’s colocalization coefficient test of five separate images, it appears that what little TMV is detectable is within the lysosome (M = 0.62 ± 0.12), but does not colocalize with

SA (M = 0.25 ± 0.12). The Mander’s colocalization coefficient also suggests that

SA is not in the lysosome (M = 0.37 ± 0.07). Therefore, SA could be trafficked from a different compartment, or could be cleaved from TMV before reaching the lysosome. Analyzing a z-stack cross-sectional view of a macrophage cell at 6 hours, puncta of SA are visualized within the cell but not colocalized with the lysosome (Figure 3.6C).

By 12 hours, the TMV signal increased as described for Figure 3.6A, suggesting SA has been cleaved by this time point (Figure 3.6D). Mander’s colocalization coefficient indicates again for the 12 hour time point that TMV is colocalized with the lysosome (M = 0.80 ± 0.08) but less so with SA (M = 0.49 ±

0.05). SA appears to be diffuse at the edges of the cell and could be at the cell

107

membrane, as has been reported previously (Schmidtke and Unanue, 1971). SA exhibits a low degree of lysosomal colocalization (M = 0.50 ± 0.03). It appears that the colocalization between SA and TMV or SA and lysosomes only occurs when TMV and lysosomes are at the edges of the cell where SA is found, and it may not be true colocalization. This is supported further by a z-stack cross- sectional view of a macrophage cell at 12 hours, where TMV clearly colocalizes with the lysosome but SA remains on the edges of the cell, possibly at the cell membrane (Figure 3.6E).

These results suggest that the conjugated nanoparticles are split into SA and TMV components after cell entry and traffic through different pathways. TMV travels to the lysosome, where it is likely degraded and processed, leading to presentation of TMV epitopes by antigen presenting cells and thereby stimulatin4g the production of antibodies (Watts, 2012). The results of the confocal microscopy experiments support the findings from the ELISA experiments (Figure 3.4C), where all SA-TMV constructs produced similar antibody responses. Presumably once the nanoparticles are taken up by immune cells, the cells process TMV similarly regardless of the construct. SA, on the other hand, appears to be exported from the cell, as it is found at the edges of the cell by 12 hours (Figure 3.6D) and the amount of SA within the cell appears to be reduced after washing (Figure 3.6A, 12 with 12 wash time point).

Furthermore, the confocal microscopy results never indicate strong colocalization of SA with the lysosome, possibly because SA is trafficked out of the cell by SA receptors such as FcRn, gp60, gp18, and SPARC after entering cells but before

108

reaching lysosomes (Anderson et al., 2006; Chaudhury et al., 2003; Merlot et al.,

2014; Schnitzer, 1992; Schnitzer and Oh, 1992; Schnitzer and Bravo, 1993;

Schnitzer and Oh, 1994). FcRn has been previously reported to rescue SA from lysosomal degradation for recycling back to circulation, supporting this hypothesis (Kim et al., 2006; Sarav et al., 2009).

3.4.5 SA-TMV cleavage and degradation in the lysosome

Endolysosomal proteasomes could be responsible for cleaving SA from

TMV, allowing for SA recycling and TMV degradation, as seen by confocal microscopy. To determine if SA could be cleaved from TMV within the cell, Cy5- labeled SA-TMV construct 2 was incubated with lysosomal extract (LE) isolated from mouse livers at pH 5. As a control, nanoparticles were separately incubated at pH 5 in phosphate buffered saline (PBS) buffer. At 6 hours, 24 hours, and 48 hours, nanoparticles were extracted and run on SDS-PAGE. Gels were imaged for fluorescence (from the Cy5 label), then stained with Coomassie blue (Figure

3.7A). By 48 hours, SA-TMV bands are absent from the gel but TMV CP monomer and dimer bands remain when 2 was incubated with LE (quantified in

Figure 3.7B). In contrast, when 2 was incubated in PBS at pH 5, all bands remain constant over time. This demonstrates that within the endolysosomal environment, SA can be cleaved from the TMV surface. While there is no increase in SA or TMV bands, it is likely that in the ex vivo experiments SA and

TMV fragments may be further degraded. However in vivo, while TMV would remain in the lysosome, SA could be recycled by its receptors before degradation could occur.

109

110

Figure 3.7. Stability of Cy5-labeled SA-TMV in lysosomal extract (LE) over time. A) SDS-PAGE results over time, imaged first by fluorescence (left), then stained with Coomassie blue and imaged by white light (right). B) Densitometry

quantification of fluorescent band intensity for TMV CP monomer (left), TMV CP

dimer (middle), and sum of SA-TMV bands (right).

111

3.5 Conclusions

This work highlights the utility of a bio-inspired camouflage, SA, for overcoming immune recognition of biologic-based nanoparticle delivery platforms. Our studies indicate that the effectiveness of SA shielding is dependent on the coverage characteristics, as has been reported previously for

PEG shielding mechanisms (Lee et al., 2015a; Perry et al., 2012). SA camouflage prevents antibody recognition, with higher coverage leading to more effective shielding. Linker length also contributes to stealth properties, with shorter linkers being more effective; likely due to providing a more efficient linker system for SA conjugation. Furthermore, to avoid antibody recognition by anti-

PEG antibodies, the best particles to develop for clinical translation are therefore the particles with high SA coverage and short PEG linkers (construct 2). While administration of SA-TMV particles in vivo results in production of TMV-specific antibodies, these anti-TMV antibodies do not recognize the SA-coated TMV nanoparticles. SA-specific antibodies are not produced, likely because TMV and

SA follow distinct intracellular pathways, i.e. endolysosomal degradation of TMV and recycling of SA. Previous findings show that small chemical modifiers are cleaved rapidly from TMV particles when exposed to endolysosomal conditions

(Wen et al., 2015). Here we show that even large cargoes, such as SA, are cleaved from the nanoparticles. This indicates a potential application for cargo delivery, where release in the endolysosome would be desired (e.g. cell permeable drugs or drugs acting on the endolysosome). TMV nanoparticles are already being explored as carriers for various drugs loaded into the central

112

channel of the virus (Chariou and Steinmetz, 2017; Czapar et al., 2016; Lee et al., 2016). Future applications could involve loading hydrophobic drugs onto the

SA component of an SA-TMV nanoparticle for a dual-delivery approach.

3.6 Acknowledgements

This work was supported in parts by a grant from NIH, R01-CA202814 (to

N.F.S.). N.M.G. was supported in part by NIH T32 GMS008803. We thank

Christina Wege and her team (University of Stuttgart) for providing TMV-lys.

113

Chapter 4: cryoEM of HPV PsV with HD5

The material in this chapter, along with work from others, is in preparation as a manuscript authored by Gulati NM, Thomas LD, Wiens ME, Smith JG,

Stewart PL entitled “Cryo-electron microscopy characterization of human alpha- defensin 5 neutralization of human papillomavirus 16.” This chapter will focus solely on the work that I performed for this paper in determining 3D reconstructions of HPV in the absence and presence of HD5 and analysis of the reconstructions. The simulation experiments that accompany the structural data will not be discussed as they were not performed by me.

4.1 Introduction

Human papillomavirus (HPV) infections are a major health burden. HPV is the most common sexually transmitted infection, and causes a range of conditions, including genital warts. While most cases of HPV do not cause major health concerns, persistent genital HPV infection is linked to nearly all cases of cervical cancer, the second most common cancer in women worldwide, according to the World Health Organization (World Health Organization, 2017).

Moreover, HPV infections can also cause other types cancers in men and women. HPV-related cancers are estimated to have caused 266,000 deaths worldwide in women (de Martel et al., 2017; World Health Organization, 2017).

114

Vaccines against certain strains of HPV are clinically available, but they are

optional and while the majority of adolescent females (48.7%) received at least one dose of the HPV vaccine in 2010, the number was significantly lower for adolescent males (1.4%) (Centers for Disease and Prevention, 2011). The vaccine must be given to adolescents before they become sexually active to be most effective, as HPV is typically transmitted through sexual contact between

genital skin surfaces (World Health Organization, 2017). However, studies have

shown that the vaccine only provides protection against HPVs for approximately

10 years (Naud et al., 2014).

The HPV vaccines on the market, Cervarix® and Gardasil®, protect against some oncogenic strains of the virus. Both contain viral-like particles, or

VLPs, of each strain they protect against. VLPs are formed by the major capsid protein L1 assembled into an icosahedral structure similar to native HPV.

Because the L1 proteins are type-specific, the vaccines are made up of multiple different VLP constructs. Therefore, the vaccines are expensive to manufacture and do not protect against every oncogenic HPV strain. There is a need for the cross-protective therapies for the prevention and treatment of HPV.

Human alpha-defensin 5 (HD5), a peptide of the innate immune system, has been shown to block infection of multiple serotypes of HPV, including HPV16

(Buck et al., 2006). There has been strong interest in defensins as potential pharmaceutical compounds because of their potent antimicrobial activity, but commercial realization has proved challenging. It is important to fully understand

the mechanism of action of these peptides to be able to translate them to the

115

clinic, yet how HD5 acts against HPV remains unclear. Defensins such as HD5 are thought to act against bacteria and enveloped viruses through membrane disruption, but this does not explain neutralization of non-enveloped viruses

(Lehrer and Lu, 2012).

Distinct mechanisms have been found for HD5 neutralization of non-enveloped viruses. HD5 prevents human adenovirus (AdV) infection by interacting with long intrinsically disordered regions on the viral capsid and by bridging and stabilizing the interaction between two capsid proteins (Flatt et al., 2013; Smith et al., 2010).

Consequently, upon cell entry the virus is unable to uncoat within the endosome

(Nguyen et al., 2010). JC polyomaviruses are also neutralized by HD5-induced viral stability, which leads to altered intracellular trafficking (Zins et al., 2014). For

HPV, the mechanism of HD5 neutralization is more complicated. HD5 has been shown to prevent furin cleavage of the HPV minor capsid protein L2, a step required for the cell entry pathway (Wiens and Smith, 2015). However, furin- cleaved HPV16 infection is still blocked by HD5 (Wiens and Smith, 2017). HD5 has also been shown to prevent the HPV genome from reaching the nucleus

(Buck et al., 2006). As with AdV, HD5 neutralizes HPV16 by preventing endosomal escape, leading to viral degradation in the lysosome (Wiens and

Smith, 2017). However, the critical binding sites for HD5 on the HPV capsid are unknown. The minor capsid protein L2 does have flexible, intrinsically disordered regions as determined by the protein disorder software package PrDOS (Figure

4.1), and it is partially surface-exposed for mature virions (Ishida and Kinoshita,

2007; Wang and Roden, 2013). We hypothesize that HD5 neutralization of

116

HPV16 might involve binding to the intrinsically disordered regions of L2, in analogy to AdV. To clarify the structural mechanism of HD5 neutralization of HPV, in this chapter, cryo-electron microscopy (cryoEM) is used to investigate the structural changes to the HPV16 capsid in the presence of HD5.

117

Figure 4.1. Disorder profile of L2 protein of HPV16 by amino acid sequence.

Disorder probability calculated by Protein Disorder prediction System (PrDOS)

(Ishida and Kinoshita, 2007). Predicted disordered regions are above the threshold (red) based on a false positive (FP) rate of 5.0%.

118

4.2 Materials and Methods

4.2.1 Protein disorder prediction and modeling

HPV16 L2 protein sequence was submitted to PrDOS for disorder prediction (Ishida and Kinoshita, 2007). PrDOS predicts disorder based on local amino acid sequence and sequences of homologous proteins for which structural information is available. The N- and C-terminal tails of an HPV16 L1 pentamer missing from the atomic model (PDB: 5KEP) were built using UCSF Chimera and minimized using NAMD (Guan et al., 2017; Pettersen et al., 2004; Phillips et al.,

2005).

4.2.2 HPV16 pseudovirus production

The following method was performed by Dr. Mayim Wiens. HPV16 pseudovirus (HPV PsV) containing the major capsid protein L1 and minor capsid protein L2 were prepared as previously described (Cardone et al., 2014; Wiens and Smith, 2015). Briefly, HPV16 PsVs were made via transfection of 293TT cells with plasmids encoding codon-optimized HPV16 L1 and L2. PsVs were matured in the lysate of these cells and purified by ultracentrifugation.

4.2.3 CryoEM grid preparation

Purified HPV16 PsVs were combined with HD5 (10 µM) and incubated on ice for 45 min. 3 µL HPV16 PsV alone or HPV16 PsV with HD5 was then applied to glow-discharged Quantifoil 2x2 400 mesh holey carbon grids. Grids were blotted until nearly dry and rapidly frozen in liquid ethane using a manual plunger.

119

4.2.4 CryoEM imaging and data collection

CryoEM micrographs were collected on an FEI Titan Krios 300 kV

transmission electron microscope with a Direct Electron DE20 direct detector for a total electron dose of 60 electrons/Å2. Micrographs were collected at 29,000X

magnification with a defocus range of 0.8 µm to 3.5 µm. The HPV16 PsV alone

dataset included 3304 micrographs and the HPV16 PsV with HD5 dataset

included 3868 micrographs. Frame alignment and radiation dose damage

compensation of the micrographs was performed with the DE_process_frames

software (Direct Electron).

4.2.5 Particle picking and CTF correction

Individual particles were selected in a semi-automated fashion using E2

Boxer in the EMAN2 software package (Tang et al., 2007). Estimations of the

defocus values for the micrographs were made using GCTF (Zhang, 2016).

4.2.6 3D structure determination and filtering

RELION 2.0 was used to extract individual particles from micrographs

(1.26 Å/pixel or binned data at 2.52 Å/pixel) and for subsequent steps in

refinement. A cryoEM structure of the HPV16 PsV (EMD: 5932) filtered to 60 Å

resolution was used as the initial model for 3D reconstruction (Cardone et al.,

2014). This model was used to determine the orientations of each particle. A

subset of particles was chosen through 2D classification and 3D classification for

each structure. Binned data (2.52 Å/pixel) was used for 2D classification and

unbinned data (1.26 Å/pixel) was used for 3D classification and refinement. Final

refinement of the maps used 2900 particles for HPV16 PsV alone and 1742

120

particles for HPV16 PsV with HD5 using RELION 2.1. After refinement and postprocessing, the final resolutions of the HPV16 alone and HPV16+HD5 maps were 5.3 Å and 4.9 Å, respectively, as measured at the 0.143 threshold of gold standard Fourier Shell Correlation plot. UCSF Chimera was used to filter the

HPV maps with the Gaussian filter tool and a 4.0 voxel width (Pettersen et al.,

2004).

4.3 Results and Discussion

4.3.1 Sub-nanometer resolution structures of HPV16 and HPV16+HD5 by

RELION

Samples of HPV16 PsV and HPV16 PsV with HD5 were vitrified and cryoEM data collection was performed on a Titan Krios with a Direct Electron

DE20 direct detector. Drift-corrected and dose-compensated micrographs of both samples showed similar morphology, with spherical capsids and prominent protrusions for the capsomers (Figure 4.2). HPV16 PsVs have diameters of ~60 nm, consistent with previous HPV16 structures (Cardone et al., 2014; Guan et al., 2017). However, the viral capsids in the cryo-electron micrographs are not completely uniform, and there is some heterogeneity in size and shape (Figure

4.2, arrows).

121

Figure 4.2. Representative cryo-electron micrographs of HPV16 (left) and

HPV16+HD5 (right). Arrows point to capsids with heterogeneous size or shape.

122

REgularized LIkelihood OptimizatioN (RELION) was used to sort the populations of HPV16 and HPV16+HD5 particles to include only those with spherical shape and a uniform diameter to produce a more homogenous dataset for reconstruction. Relion was then used to calculate 3D reconstructions from the homogeneous datasets (Figure 4.3A). The reconstruction of HPV16 resulted in a 5.3 Å resolution structure, while HPV16+HD5 resulted in a 4.9 Å resolution structure, as determined by the gold standard FSC curve at 0.143 Å (Figure 4.4).

Both structures were solved using particles with a 1.26 Å/pixel, although some initial processing was performed with binned images. At this resolution, HPV16 and HPV16+HD5 have similar capsid structures and there is no obvious density for HD5 (Figure 4.3B). This is unsurprising, however, as HD5 has been found to bind to intrinsically disordered regions on the AdV capsid (Flatt et al., 2013).

Therefore, HD5 could be binding to flexible regions on the HPV16 capsid. If this is the case, it would be difficult to observe density for HD5 in a high resolution cryoEM structure without filtering.

123

Figure 4.3. CryoEM reconstructions of HPV16 and HPV16+HD5. A) Radially colored representations of HPV16 at 5.3 Å resolution (left) and HPV16+HD5 at

4.9 Å resolution (right). B) Overlay and cross-section view of the two HPV16 maps. HPV16 is in light blue, HPV16+HD5 is in red. Scale bar = 100 Å.

124

Figure 4.4. Plot showing the gold standard Fourier shell correlation (FSC) vs. spatial frequency of the icosahedrally averaged reconstructions for

HPV16 and HPV16+HD5. The resolution of the reconstructions is assessed where the FSC curve crosses a correlation value of 0.143 (dashed line).

125

4.3.2 Gaussian filtered HPV16 and HPV16+HD5 maps reveal additional density in the HD5 structure

In order to visualize possible regions of HD5 binding, both the HPV16 and

HPV16+HD5 structures were filtered with UCSF Chimera’s Gaussian filter with a

4 voxel width (Pettersen et al., 2004). When filtered, the resolution of the structures is lowered but the signal-to-noise ratio is increased for regions with flexibility (Figure 4.5). After filtering additional density is revealed on the surface of the hexavalent capsomers, but not the pentavalent capsomers of the

HPV16+HD5 structure (Figure 4.5A). This density can be attributed to HD5 or to

HD5 in complex with flexible regions of the L2 capsid protein. Regions of L2 have been proposed to interact with the exposed surfaces of the hexavalent capsomers more prominently than pentavalent capsomers (Figure 1.7) (Guan et al., 2017). The hexavalent capsomers are those L1 pentamers that have six neighboring pentamers. Similarly, the pentavalent capsomers are those L1 pentamers that have five neighboring pentamers and sit on the icosahedral 5-fold symmetry axes. The reason for L2 binding preferentially to hexavalent compared to pentavalent capsomers is unclear.

126

Figure 4.5. Gaussian filtered maps of HPV16 and HPV16+HD5 reconstructions. A) An overlay of the two structures revealing additional density in on the outer capsid surface of the HPV16+HD5 complex on mainly the hexavalent capsomers. HPV16 alone is in light blue, HPV16+HD5 is in red. B) An overlay of 20 Å thick central slabs of the two structures reveals additional density within the core just below the capsid in the HPV16+HD5 structure. C) 20 Å thick central slab of the HPV16 structure. D) 20 Å thick central slab of the HPV16+HD5 structure. Scale bar = 100 Å.

127

If HD5 does interact with regions of the L2 protein on the capsid surface,

this could explain the ability of HD5 to prevent furin cleavage of L2 (Wiens and

Smith, 2015). Antibody binding experiments suggest that the L2 capsid protein is

only minimally exposed on the surface of mature virions (Buck and Trus, 2012;

Day et al., 2008; Rubio et al., 2011; Wang and Roden, 2013; Yang et al., 2003b).

At least portions of L2 are known to emerge from the virion during the cell entry process (Kines et al., 2009). Therefore we hypothesize that regions of L2 transiently pass through pores in the L1 capsid (Figure 4.6) and that HD5 stabilizes L2 bound on the exposed surfaces of the hexavalent capsomers.

Internally, the filtered structure of HPV16+HD5 has more prominent density within the core than HPV16 (Figure 4.5B-D). In general the core of

HPV16 does not follow icosahedral symmetry, but icosahedral symmetry has been applied during the reconstruction process. This leads to the most of the core density being averaged away during the reconstruction process. The finding of more prominent core density in the HPV16+HD5 structure indicates that the presence of HD5 leads to a tighter association between the HPV16 capsid and the core components. The core includes portions of L2 and the genome, or in the case of PsV, randomly incorporated DNA. Specific regions of L2 have been identified as interacting with DNA (Figure 1.6). It is unclear whether the internal core density is stabilized by interactions of HD5 on the exterior surface of the capsid or whether HD5 molecules are able to enter through pores in the capsid

(Figure 4.6) and stabilize the capsid/core interaction from within. Nevertheless, this structural result indicates an increased interaction between the HPV16

128

capsid and core in the presence of HD5, which is consistent with cell culture studies indicating that HD5 inhibits the dissociation between the capsid and genome (Wiens and Smith, 2017).

129

Figure 4.6. Pores in the HPV16 capsid. A) HPV16 capsid shown radially

colored with a black outline around a hexavalent capsomer and its six

neighboring capsomers. Each capsomer in HPV16 is an L1 pentamer. B) HPV16 capsid (transparent gray) with docked atomic structures of seven L1 pentamers

(PDB: 5KEP). C) A hexavalent capsomer and its six neighboring capsomers shown in ribbon representation. Red arrows indicate inter-capsomer pores and the black arrow indicates an example pore within one capsomer.

130

There is a gap between the internal core density and the viral capsid in the filtered HPV16+HD5 structure (Figure 4.7). This distance is approximately 25 Å, which is the approximate length of the flexible C-terminal tails of the L1 protein.

The C-terminal tails of L1 (aa 486 to 505) are not resolved in the HPV16 structures (Cardone et al., 2014; Guan et al., 2017). However, we have built extended atomic models for the N- and C-terminal tails of an L1 pentamer in order to get a more complete picture of possible interactions involving these tails with the virion. Several residues at the extreme C-terminal end of L1 are positively charged (K499, R500, K501, K502, R503, K504). These positively charged L1 residues could be interacting with the phosphate backbone of DNA or with negatively charged residues on HD5. There is one negatively charged residue of HD5 (E21) that has been shown to be critical for anti-viral activity against HPV16 (Tenge et al., 2014). If HD5 is able to enter through pores in the

HPV16 capsid, then it is possible that the L1 C-terminal tails are involved in interactions with both HD5 and the genome of a mature virion. This three-way interaction could lead to direct stabilization of the capsid/core interaction.

Docking of HD5 in the structure of the L1 capsid (PDB: 5KEP) suggests that the pores between the L1 capsomers are large enough for HD5 to pass through.

131

Figure 4.7. Central slabs of the HPV16 (light blue) and HPV16+HD5 (red) filtered maps compared to the L1 pentamer atomic structure (green). The distance between the outer capsid and the inner density observed in the

HPV16+HD5 structure is approximately 25 Å (gray bar), similar to the extended length of the flexible L1 C-terminal tails (gray bar). Scale bar (black) = 100 Å.

132

There are multiple pores within the HPV16 capsid through which L2 may extend or HD5 might enter. Pores exist within the center of each L1 pentameric capsomer (Figure 4.6C, black arrow), as well as between the different L1 capsomers (Figure 4.6C, red arrows), although the inter-capsomer pores are larger. It is likely that in a mature virion L2 is predominantly located with the viral genome in the core of the virion and that portions of L2 extend through capsid pores and interact with the exterior surface of the L1 hexavalent capsomers. This would be consistent with the known interaction data for L2 and with previous

HPV16 structural studies (Figure 1.7B, C) (Buck et al., 2008; Guan et al., 2017).

It is also likely that the interactions L2 has with the outer surface of the L1 capsid are transient in nature, since density for L2 was not well resolved in the high resolution HPV16 structures (Cardone et al., 2014; Guan et al., 2017). Based on our cryoEM structure of HPV16+HD5, we hypothesis that HD5 stabilizes a transient interaction of L2 with the outer surface of the L1 capsid. This stabilization on the outer capsid surface could lead indirectly to a tighter association between the HPV capsid and core. In the absence of HD5, the

L2/genome complex may be only loosely associated with the icosahedral capsid formed by L1 pentamers. Therefore, during cell entry the L1 capsid and the

L2/genome complex are easily separated by the appropriate cellular triggers. In the presence of HD5, flexible regions of L2 may be so tightly associated with the outer surface of the L1 capsid that separation of the L1 capsid and the

L2/genome complex is impeded. As has been shown, the hyper-stabilized

133

interaction of L1 with the L2/genome complex leads to viral neutralization (Wiens and Smith, 2017).

HD5 is known to stabilize the capsid of AdV, preventing viral escape from the endosome (Flatt et al., 2013; Nguyen et al., 2010; Smith et al., 2010). In the case of AdV, HD5 does not bind specifically to one capsid protein, but rather bridges the interaction between two capsid proteins. CryoEM structural studies combined with viral chimeras have indicated that HD5 binds at an interface between a mostly structured AdV capsid protein (fiber) and intrinsically disordered, highly flexible loops of another AdV capsid protein (penton base)

(Flatt et al., 2013; Smith et al., 2010). We propose a similar mechanism for HD5 stabilization of the HPV capsid: the immune peptide could stabilize transient interactions between the structured L1 capsomers and flexible, intrinsically disordered regions of L2. The involvement of intrinsically disorder regions of L2 in binding HD5 would explain the lack of observed structural differences in the high resolution structures of HPV16 and HPV16+HD5 (Figure 4.3).

Intrinsically disordered proteins and protein domains have important functions and are found in many eukaryotic organisms (Dyson and Wright, 2005).

Many of these sequences undergo transitions to become more structured in the presence of a binding ligand, as may be the case for L2 of HPV (Dyson and

Wright, 2005). Intrinsically disordered proteins are difficult to study using X-ray crystallography or cryoEM since a single ordered structural state does not exist.

Nevertheless, the presence of flexible regions can sometimes be identified by filtering cryoEM electron density. Other tools, such as molecular dynamics

134

simulations or spectroscopy techniques like nuclear magnetic resonance (NMR), can provide information on the structure and dynamics of flexible proteins or domains (Felli and Pierattelli, 2012; Salvi et al., 2016; Smith et al., 2012).

However, NMR and molecular dynamics simulations are challenging for large macromolecular assemblies such as viruses. Often intrinsically disordered proteins have highly conserved sequences, as is the case for the L2 protein between different HPV serotypes (Wang and Roden, 2013). This suggests that the intrinsically disordered regions play important functional roles. Therefore, even moderate resolution structural information is helpful in understanding the biological role played by proteins such as the HPV L2 capsid protein.

4.4 Conclusions

The cryoEM results presented here indicate that HD5 does not bind in a regular way to icosahedrally ordered sites on the HPV16 capsid. Rather,

Gaussian filtering indicates that HD5 binds to flexible regions on the virion. After filtering additional density is observed on the surface of the hexavalent L1 capsomers, and also within the core of the HPV16+HD5 structure. HD5 likely interacts with or stabilizes L2 on the surface of the hexavalent L1 capsomers.

HD5 also appears to directly or indirectly stabilize the association between the

HPV16 capsid and core components. Future work will be needed to determine specifically which regions of L2 are involved in binding HD5. A greater understanding of the mechanism of HD5 neutralization is important for the development of HD5-based therapies for treatment of HPV and other viral

135

infections. This structural study represents the first examination of where HD5 binds to the HPV16 virion and provides insight into the mechanism of HD5 neutralization.

136

Chapter 5: Conclusions and Future Directions

5.1 Scope of work

Viruses come in all shapes and sizes, and can infect all life forms, from plants to mammals to microorganisms. Because of the diversity of viruses, biomedical research of these pathogens is expansive, and not all viruses or materials derived thereof may be regarded as pathogen, but rather as a tool for bio- and nanotechnology. Based on their natural ability to deliver a cargo (their genome) to cells of their host organisms, viruses have been explored as nanoparticles in biomedical applications; this generally applies to plant viruses and bacteriophages as well as disarmed, non-infectious mammalian viruses. On the other hand, viruses that infect humans can cause devastating disease, and therefore significant research has focused on understanding the infectious mechanisms of pathogenic viruses and how to prevent or limit them. The work presented in this dissertation explores both sides of viral research. My studies led to new discoveries and understanding on how viruses – either virus-based nanotechnology or pathogen - interact with their biological environment.

137

5.2 Viruses in nanomedicine: evading immune recognition of TMV nanoparticles by stealth-camouflage with SA

To gain a deeper understanding of viral nanoparticle applications, it is first important to consider what happens to a nanoparticle upon entering the biological environment. After injection into the bloodstream, viral nanoparticles undergo opsonization and are rapidly cleared by the mononuclear phagocytic system (Kokkinopoulou et al., 2017; Lee et al., 2015b). Viruses and other nanoparticles can evade clearance using shielding strategies that involve coating the surface of the nanoparticle in an anti-fouling agent. In this work, I characterized serum albumin (SA) as a bioinspired shielding strategy using on a novel high-aspect ratio plant viral nanoparticle: tobacco mosaic virus (TMV).

Initially, I investigated the structure of stealth-coated viral nanoparticles to better understand the design rules of effective shielding strategies. As stealth- coated nanoparticle formulations become more relevant in the clinical setting, it is now more important than ever to assess and detail nanoparticle formulations at the molecular level. To tackle this, I studied the structural morphology of SA- camouflaged TMV nanoparticles using cryo-electron microscopy (cryoEM), cryo- electron tomography (cryoET), and subtomogram averaging (Chapter 2).

Previous data on SA coating demonstrated the effectiveness of SA as a bioinspired nanoparticle shielding mechanism, which, when used to coat TMV demonstrated better prevention of antibody recognition than coating the surface with polyethylene glycol (PEG), termed PEGylation (Pitek et al., 2016b).

However, the mechanism of shielding was not yet explored. Using cryoET and

138

subtomogram averaging, I was able to visualize individual SA molecules on the

TMV surface and determine that SAs adopt random orientations relative to the

TMV rod, likely due to the flexible linker between the two and also because of the

numerous possible attachment points on SA. This heterogeneous display could

be advantageous for the clinical translation of SA-coated TMV nanoparticles in

preventing innate immune pathogen-associated molecular patterns (PAMPs)

from recognizing the camouflage coating. With atomic modeling, I concluded that

SA likely prevents antibody-antigen interactions with the TMV surface at least

partially by steric hindrance, where SA packing effectively blocks the TMV

surface and prevents a Fab arm from accessing and recognizing necessary

epitopes on the nanoparticle surface. Overall, these studies serve not only as a

proof-of-concept of using cryoET and subtomogram averaging for characterizing

stealth-coated nanoparticles at the molecular level, but also highlight the

advantages of SA as a bio-inspired nanoparticle shielding mechanism.

To further understand the utility of SA-camouflaged TMV nanoparticles, I

sought to understand the immune processing of TMV with different formulations

of SA coating. Unlike PEGylation which results in the production of anti-PEG

antibodies, especially after repeat administrations (Armstrong et al., 2007;

Ganson et al., 2016a; Ganson et al., 2016b; Garay et al., 2012; Hershfield et al.,

2014; Ishida et al., 2006; Richter and Akerblom, 1984), SA is a naturally

occurring protein in the body and is in fact the most common protein in blood

plasma. By characterizing differently conjugated SA-TMV formulations for

antibody production and recognition in vivo and investigating the processing of

139

SA-TMV by phagocytic immune cells, I concluded that high coverage particles

prevent TMV-specific antibody recognition better than low coverage particles,

presumably because the amount of accessible TMV surface is reduced. While

high coverage particles with both linker lengths had similar shielding effects, the

formulation with a longer PEG linker was recognized by PEG-specific antibodies.

Furthermore, I found that all particle formulations led to the production of TMV- specific but not SA-specific antibodies. However, SA-coating reduced the ability of TMV-specific antibodies produced in vivo from recognizing TMV-based

nanoparticles, with higher coverage particles having less recognition. The

difference in antibody production is likely because SA and TMV undergo different

fates upon uptake by macrophage cells: TMV traffics to the lysosome, while SA

initially enters the cell but later traffics towards the edges of the cell possibly by

receptors that recycle SA. Understanding the immune processing for self-coated nanoparticles using SA, the most abundant plasma protein is critical for their

future applications in nanotechnology.

Some considerations for the continuation of the work on viruses in

nanomedicine presented in Chapters 2 and 3 and a discussion of future

implementation of these nanoparticles into the clinic is presented in the

remainder of Section 5.2.

5.2.1 Tobacco mosaic virus as a nanocarrier for drug delivery applications

Nanoparticle research typically focuses on spherical nanoparticles of

synthetic origin. High aspect-ratio viruses such as TMV have many advantages

as naturally-evolved nanomedical platforms. They can serve as vehicles to

140

deliver drugs, as imaging agents, or in gene therapies to deliver molecular cargo.

The exterior of the virus can be modulated with immune-shielding agents and targeting ligands. Plant viruses in particular have advantages in biomedical applications, as they are biocompatible, biodegradable, and non-infectious in mammals. Because of its elongated shape, TMV has improved characteristics in vivo compared to spherical particles, such as margination properties and homing to tumor tissue (Fratila et al., 2015; Hao et al., 2014; Lee et al., 2013; Petros and

DeSimone, 2010). TMV is also well-characterized and is amenable to chemical modifications and genetic engineering. The potential of TMV as a delivery vehicle for drugs and imaging agents has been demonstrated (Bruckman et al., 2016;

Bruckman et al., 2014b; Czapar et al., 2016; Finbloom et al., 2016; Kernan et al.,

2017; Lee et al., 2016; Niehl et al., 2015; Wen et al., 2012).

While TMV shows promise as a nanocarrier for translational applications, it requires shielding to evade immune recognition and clearance in order to be used successfully in the clinic. Without shielding, TMV has been shown to be rapidly cleared from circulation in naïve mice (Bruckman et al., 2014a). Towards this end, my work in Chapters 2 and 3 has focused on shielding TMV using a bioinspired ‘self’ coating with SA. However, other considerations for the translation of TMV must also be considered.

While TMV is noninfectious in humans and other mammals, it does infect multiple crops, including tobacco, tomatoes, and peppers (Asselin, 1979; Cai et al., 2003; Gottula and Fuchs, 2009; Pinnow et al., 1981; Sakata et al., 1997).

Before being used in a clinical setting, the environmental impact of TMV must be

141

studied. Research into the safety and outcomes of TMV and its effect on the

environment is required before clinical translation. If problems arise, non-

infectious TMV particles (containing non-genomic RNA sequences) could be

employed. RNA-free versions of TMV could also be produced; beyond avoiding

possible environmental safety concerns, such formulations would also have

reduced immunogenicity and prevent activation by toll-like receptors (TLR) (Gao

and Li, 2017). However the use of RNA does provide some advantages for TMV

in nanotechnology. RNA-containing TMVs can be easily tailored to a specific

length (and aspect ratio) which can help to control the nanoparticle’s in vivo

properties (Shukla et al., 2015). Furthermore, TMV could be used for delivery of

RNA cargo, thus necessitating inclusion of RNA. TMV is clearly a versatile

platform and there are many considerations into its design for specific

applications in nanotechnology.

The success of TMV as a nanocarrier has been primarily focused on in

vitro studies and preclinical animal models. Before translation, like any other

therapy, studies must be done to investigate the safety and efficacy of TMV in

non-human primates. It is known that TMV-specific antibodies are prevalent in

the human population, in both smokers and non-smokers (Liu et al., 2013). Thus

the use of TMV as a drug delivery vehicle in humans is dependent on the ability

to evade recognition by those antibodies. Specific formulations of TMV will have their own challenges and the safety of any TMV formulation must be investigated before undergoing clinical trials.

142

5.2.2 Novel SA self-camouflage platform

Shielding of TMV nanoparticles is critical for drug delivery applications in

preclinical models, as bare TMV is rapidly recognized and cleared by the immune

system. Initial studies of TMV as a nanocarrier, as well as many other nanoparticles, relied on PEGylation for stealth coating (Bruckman et al., 2014a;

Jokerst et al., 2011; Lee et al., 2015a; Perry et al., 2012; Veronese and Pasut,

2005). PEG has also been used in nanoparticle formulations in the clinic, such as

Doxil®. However, the increase in PEG-specific antibodies detected in humans

limits its use in nanoparticle formulations. I used structure-function studies to

show the benefits of a novel SA camouflage strategy (Chapters 2, 3). For the

translation of this strategy, multiple avenues of research must still be considered.

The data presented in Chapter 3 indicate that after SA-camouflaged TMV

nanoparticles are taken up by cells, the SA and TMV components undergo

different fates. While TMV co-localized with the lysosome by 12 hours, SA did not

significantly co-localize with the lysosome at any time point studied. Instead, SA

appeared to be diffusely located at the periphery of cells. It could be that SA is

recycled by scavenger receptors, such as the neonatal Fc receptor (FcRn), which

is known to bind and recycle SA (Figure 5.1) (Anderson et al., 2006; Chaudhury

et al., 2003; Kim et al., 2006; Sarav et al., 2009). Proteins that bind SA include

gp60, gp30, gp18, FcRn, and secreted protein acidic and rich in protein (SPARC)

(Anderson et al., 2006; Merlot et al., 2014; Milici et al., 1987; Schnitzer, 1992;

Schnitzer and Oh, 1992; Schnitzer and Bravo, 1993; Schnitzer and Oh, 1994).

They could be responsible for trafficking SA and preventing its degradation. To

143

clarify this, experiments could be done to inhibit or knockdown SA receptors in cell culture and repeat the confocal microscopy experiments to see if the localization of SA is affected. To check if SA degradation is increased when SA-

TMV is administered, experiments could be done by radioiodinating SA before conjugating it to TMV. Then, upon cell uptake the iodine content of within the cell and of degraded protein products could be measured, as has been previously done in studying albumin digestion by macrophages (Ehrenreich and Cohn,

1967). It is important to understand the fate of SA after cleavage from TMV, because deregulation of SA could lead to hyperalbumina (if too much SA is recycled) or hypoalbuminia (if degradation of SA from the nanoparticle formulation promotes clearance of free SA in the body as well).

144

Figure 5.1. Schematic of human serum albumin (HSA) transport in endothelial cells showing FcRn receptor binding in endosomes at pH 6.0 leading to recycling and release, or trafficking to a lysosome for degradation. HSA is shown in green. Reproduced with permission from (Elsadek and Kratz, 2012).

145

SA-TMV nanoparticles have been developed as a platform for drug

delivery applications. For further development of these particles, in vivo studies

are needed to study how the SA-coating affects the biodistribution of TMV. For

applications in cancer therapy, it is important to study the ability of these particles

to home to tumor tissue in comparison to deposition in organs of the mononuclear phagocytic system (MPS) including the liver. Some shielded nanoparticles have reduced clearance but also reduced uptake by target cells, so an important future consideration is whether these nanoparticles can actually reach the site of disease for drug delivery applications (Du et al., 1997;

Hatakeyama et al., 2013).

It is possible that SA-binding proteins could be involved in the

accumulation of SA-camouflaged nanoparticles in tumors. Abraxane®, a clinically

approved nanoparticle formulation of albumin-bound paclitaxel, has been thought

to accumulate in tumors, mediated by the albumin transport receptor gp60 across

endothelial cells (Schnitzer, 1992). These nanoparticles bind the receptor and

translocate to the tumor interstitium, where it may interact with SPARC, an

albumin-binding protein that is overexpressed in some tumors (Chang et al.,

2017; Porte et al., 1995; Prenzel et al., 2006; Shi et al., 2016; Zhu et al., 2016)

(Figure 5.2). Clinical studies have shown that SPARC expression correlates to

tumor response in patients with head and neck cancer treated with Abraxane®

(Desai et al., 2009). These albumin-binding proteins could also play a role in the

accumulation of SA-TMV formulations. If this were the case, then SA would not

only provide shielding for TMV nanoparticles but may also help with targeting to

146

tumor tissue. Even if this were not the case, future studies should investigate the use of targeting ligands in combination with SA coating for more specific delivery of TMV nanoparticles to target tissue.

147

Figure 5.2. Schematic of the uptake of albumin-bound paclitaxel into tumors, mediated by gp60 for transcytosis through the endothelium and the albumin-binding protein SPARC within the interstitial tumor space.

Reproduced with permission from (Elsadek and Kratz, 2012)

148

Abraxane® highlights another benefit of SA in nanoparticle formulations: it can be used to load drug molecules. In the body, SA serves as a transporter for a variety of molecules, including hormones, fatty acids, and some drugs (Evans,

2002). To this end, the SA molecules used as shielding agents on TMV could also be loaded with therapeutic molecules. Thus, SA-TMV could be a dual- loaded therapy for co-delivery of multiple synergetic drugs, or a combination of drugs and imaging agents towards a theranostic nanoparticle. Of course, future studies would need to be done to confirm that the loading of SA does not alter conjugation efficacy or the ability of SA to shield TMV.

5.2.3 Considerations beyond SA-TMV

The nanoparticles discussed in this dissertation were formulated by conjugating SA and TMV through a PEG linker. However, as discussed above,

PEG has its own set of challenges in nanoparticle formulations. Ideally, in moving this construct towards the clinic, the PEG linker should be modified. Alternate linkers could include other biocompatible polymers such as poly(phospoesters) or peptide-based polymers such as oligoethylamine, poly-aspartate, poly-lysine

(OEAL) (Chen et al., 2015; Muller et al., 2017). The proteins could also be ligated in vitro via sortase-mediated peptide fusion (Levary et al., 2011). Genetic engineering could be explored to develop a modified TMV coat protein that is fused with SA so that when the virus assembles SA is exposed on the surface of the virus. Genetic engineering or protein ligation would make more uniform particles; i.e. SA would be displayed symmetrically with a defined orientation.

Genetic engineering would be relatively inexpensive to manufacture compared to

149

in vitro conjugation strategies, as modified-TMVs could be produced in plants.

However, doing so may risk SA being exposed on the TMV surface in a repeated manner, which could trigger recognition of the SA-coated particles due to the development of pathogen-associated molecular patterns (PAMPs). This would then limit the utility of SA, and thus design considerations must be considered moving these particles towards the clinic. To avoid this, an additional long flexible peptide could be added between the sequence for the TMV coat protein (CP) and SA sequence, however much work would be needed to investigate the feasibility of this strategy and to confirm that any additional sequences do not result in a specific protein fold. If SA-camouflaging reduces uptake of TMV to tumors or to target cells, a cleavable linker could be developed, in which SA separates from TMV in acidic pH such as the tumor microenvironment.

SA is abundant in the human body, and also widely prevalent in mammals. In the studies presented in this work, the SA component of the SA-

TMV nanoparticles was tailored to the specific experiment. The majority of the in vitro experiments used human SA with the long-term goal of clinical application, while the cell culture and in vivo studies used particles containing mouse SA. As discussed earlier, SA may help target camouflaged TMV nanoparticles to tumors in cancer therapy, but for the development of drug delivery vehicles for other diseases (such as cardiovascular disease) alternative stealthing and targeting proteins could be investigated, such as blood proteins (fibrinogen, clotting factors, etc.) or antibodies (Pitek et al., in review). Soluble domains of protein receptors could also be used, as has been done for CD47 (Rodriguez et al.,

150

2013). Different proteins may provide different benefits based on their natural

abilities, however care must be considered in the choice of protein to ensure that

they provided stealthing effects. A protein not commonly found in the body, i.e. a

‘non-self’ protein, could result in an immune response rather than preventing one.

In developing any alternative stealth coatings, the ability to prevent antibody

recognition and macrophage uptake would need to be investigated. The

nanocarrier itself could be tailored as well to be less immunogenic, as has been

investigated for adeno-associated virus (AAV) (Tseng and Agbandje-McKenna,

2014). While my work has focused on using TMV, the work described here could be applicable to other viral nanoparticles or any other biologic as well.

The idea behind shielding nanoparticles from recognition arises because nanoparticles produce an immune response and are cleared upon entering the biological environment. My work has focused on evading the immune response that develops when TMV is recognized as ‘non-self.’ However, an alternate approach could be to make TMV less immunogenic. TMV variants could be developed, in which the antigenic epitopes of TMV are altered, so that TMV- specific antibodies cannot recognize TMV. The development of antigenically distinct virus variants is not new: studies in the field of AAV for gene therapy have explored developing AAV variants that evade immune recognition without

shielding (Tseng and Agbandje-McKenna, 2014)). Careful study would be

needed to ensure disruption of epitopes on the TMV surface, rather than altering

epitopes or creating new antigenic regions.

151

The work in Chapters 2 and 3 expands the knowledgebase regarding using SA-shielded TMV as a novel nanocarrier towards clinical applications. By characterizing the structure of these particles and their immune processing, I have laid the groundwork for future in vivo studies to test SA-TMV as a drug delivery vehicle and for the overall development of novel stealth-coated plant viral nanoparticles.

5.3. Viruses in human disease: characterization of HD5 neutralization of

HPV infection

Unlike tobacco mosaic virus, which is non-infectious in humans, many viruses can replicate in human cells. Unchecked, these viruses can go on to cause human diseases, many of which have no cure. Therefore, research into understanding how viruses infect human cells, and how to prevent or neutralize those infections, are important avenues of biomedical research. Human papillomavirus (HPV) infection is often asymptomatic. However, for patients who show symptoms of HPV infection, the consequences can be severe, including skin lesions and cancer. In the work presented in this dissertation, I studied the neutralization of HPV by an innate immune peptide, human alpha-defensin 5

(HD5) using single particle cryo-electron microscopy.

HD5 is known to neutralize HPV infection by preventing genome escape from the endosome (Buck et al., 2006). Previous data indicated that HD5 also prevents cleavage of the L2 minor capsid protein of HPV (Wiens and Smith,

2015). Nevertheless, viruses with pre-cleaved L2 were neutralized in the

152

presence of HD5 through the same endosomal entrapment mechanism (Wiens

and Smith, 2017). Recently, in vitro cell culture experiments have shown that

HPV prevents L1 and the L2-genome complex from dissociating in the endosome, which leads to viral trafficking to the lysosome and blocking of the infection process (Wiens and Smith, 2017). My work sought to identify details as

to how HD5 interacts with the HPV capsid to stabilize it and prevent infection

(Chapter 4).

Using cryo-electron microscopy (cryoEM), I solved the structures of a pseudovirus of HPV16 containing the L1 and L2 structural proteins in the absence and presence of HD5 and concluded that there is extra density on the

hexavalent capsomers in the HPV+HD5 structure that is not present in the

absence of HD5. As the L1 protein is not different between the pentavalent and hexavalent capsomers, this could be due to HD5 binding to the L2 protein, which is thought to be exposed on the external surfaces of the hexavalent capsomers

(Guan et al., 2017). In this manner, HD5 could stabilize the association between the L1 and L2 capsid proteins of the HPV capsid and interfere with their separation during cell entry. Additional density was also observed within the HPV capsid in the HD5-containing structure, providing evidence for HD5 stabilization of the viral capsid/genome interaction. It is unclear whether HD5 is internalized, or if the binding of HD5 on the external capsid surface allosterically affects the structure of the viral core. Nevertheless, this study provides the first structural evidence of how HD5 interacts with the HPV capsid and supports the idea that

153

the mechanism of neutralization involves stabilization of the capsid/genome interaction.

In the remainder of the section, some considerations for the continuation of the work presented on characterizing and neutralizing HPV infection in

Chapter 4 and its clinical implications are explored.

5.3.1 Beyond the current vaccinations strategies and the need for more information

While most HPV infections cause few symptoms, those that do can have devastating consequences. HPV infections that are persistent are more likely to lead to precancerous lesions (Bosch et al., 2002; Walboomers et al., 1999).

Thus, there is a need for HPV treatments and vaccines. The vaccines on the market (Gardasil® and Cervarix®) prevent HPV infection for certain HPV types, but they leave a lot to be desired. Firstly, they are not cross-protective, as they are based on the HPV L1 protein which differs between HPV types. Secondly, they are thought to protect for 10 years but may not protect beyond that (Kjaer et al., 2009; Naud et al., 2014). Thus, if a person is vaccinated as an adolescent as recommended, then by his or her mid-twenties the vaccine may no longer be effective. This leaves adults at risk of getting the virus, which is often transmitted through sexual contact. Thirdly, the vaccines do not treat HPV infections, but rather are meant to be given to people before first exposure to the virus.

Additionally, compliance with the vaccination schedule can sometimes be an issue: many people who get the first dose of the vaccine do not get second dose

(Centers for Disease and Prevention, 2011). There are also concerns within the

154

general population about the safety and necessity of vaccines in general, including the HPV vaccine, though there is no scientific basis for this speculation

(Velan, 2011). Clearly, while the available vaccines are useful in preventing disease, there is a need for HPV treatment options.

There are currently no therapies on the market for treating HPV infection.

New drugs to neutralize pre-existing infections would be beneficial in the current medical landscape. Those that could protect against many HPV types at once would be especially advantageous. Understanding how HPV is neutralized in biological systems, like with the immune peptide HD5, could lead to drugs that mimic those mechanisms. My work in Chapter 4 to characterize how the immune peptide HD5 neutralizes HPV infection works towards this goal. However, additional considerations must be made in continuing to characterize the structural basis of HPV neutralization.

One possible confounding factor in my work and the work of others is that this research has not been done with the infectious virus. Most researchers use pseudoviruses (PsVs) or virus-like particles (VLPs) because infectious HPV is difficult to purify in large yields in cell culture, and the HPV infection cycle is quite complicated (Schiller et al., 2010). These modified virions contain one or both of the ‘late’ structural proteins of HPV, but lack the ‘early’ proteins and the infectious

HPV genome (alternative DNA sequences are often used instead). The ‘early’ protein E4, however, is thought to have an expression cycle more similar to the late proteins (Doorbar, 2013). It could be that the missing proteins could be

155

somehow involved in the HD5 inhibition of HPV infection in vivo that is lacking in the in vitro studies.

This brings up another matter: HD5 neutralization in vitro could be different than the mechanism in vivo. In vitro studies are simplified systems.

There could be important proteins or other biomacromolecules missing in my cryo-electron microscopy studies and in the cell culture studies performed by others. Nevertheless, as HPV has been shown to be neutralized by HD5 in vitro, these studies are a good starting point.

While my studies reveal extra density on the external surface of the capsid that could be attributed to L2 and/or HD5, I was not able to identify a specific

HD5 binding site required for neutralization. This is not surprising, as HD5 is known to bind to intrinsically disordered regions for neutralization of adenovirus.

Our results indicate that this is likely what happens in the case of HPV as well.

Furthermore, as the atomic resolution structure of L2 and its precise location within the capsid are unknown, it will be challenging to elucidate specific interactions between HD5 and L2. To address this, cryoEM-guided simulations could be used to identify regions of the capsid with which HD5 may interact.

Rough models for L2 could possibly be built based on the available sequence information for L2 and low resolution cryoEM density for L2. The L2 models could be built traversing through the L1 capsid as indicated by prior cryoEM studies (Guan et al., 2017) and included in the simulations. It would also be interesting to repeat the cryoEM studies described here using HPV that contains

L2 that has been pre-cleaved with furin. If the density on the hexavalent

156

capsomers were to differ in the pre-cleaved structure, it could provide insights into HD5 neutralization (independent of furin cleavage) but also could provide further insight into the location of the L2 protein within the capsid. Finally, the structurally derived information could be confirmed through mutation studies, in which specific HPV residues are mutated to see if HD5 neutralization efficiency is diminished.

5.3.2 HD5-based therapies for treatment of human papillomavirus

HD5 is known to neutralize HPV infection in vitro. The structural studies presented in Chapter 4 indicate that HD5 stabilizes the interactions between the capsid and the core. This result is in alignment with previous studies that show

L1 does not dissociate from the genome in the presence of HD5, and that L2 and the genome are unable to escape the endosome. HD5 in the body is not enough to prevent infection of HPV, however. Understanding the interactions between

HD5 and the HPV capsid, opens the potential to develop new therapies that mimic HD5-HPV interactions. This may be challenging, as it may not be as simple as identifying a specific binding site but rather understanding how HD5 stabilizes flexible protein sequences, however the reward would be great if an

HPV treatment could be developed.

5.3.3 Beyond papillomavirus – broad neutralization of infectious pathogens

The mechanisms of action for the antimicrobial activities of HD5 are not fully clear. Because HD5 can act against a broad range of pathogens with different properties, HD5-based therapies could be beneficial not only in protecting against HPV infection but also in the case of other pathogens. HD5

157

acts against bacteria, enveloped viruses, and non-enveloped viruses (Lehrer and

Lu, 2012). The mechanism of action against bacteria and enveloped viruses is

thought to be due to membrane disruption, though research is ongoing in this area. HD5 stabilizes the virion and thus prevents endosomal escape for HPV as

well as AdV (Buck et al., 2006; Gounder et al., 2012; Nguyen et al., 2010; Wiens

and Smith, 2017). It is unknown whether this may hold true for other non-

enveloped viruses as well. If that were the case, it may be that by understanding

the mechanism of action for HD5, viral therapies may be developed that are

cross-protective against many viral pathogens.

The work in Chapter 4 provides insights the mechanism of HPV

neutralization by HD5. The structural insights suggest stabilization of the L2

protein on the exterior of the virus, and also stabilization within the core. More

information is clearly needed, but understanding the HD5-HPV interactions may

be a first step in the development of a new broad antiviral therapy against HD5

and possibly other non-enveloped viruses.

5.4. Conclusions

In this work, I have presented two sides of viral research for biomedical

applications. On the one hand, my structural characterization of the HPV16-HD5

interaction provides novel insights into the underlying mechanism of

neutralization of HPV16 by HD5 and thus presents an opportunity towards the

development of antiviral therapies. On the other hand, structural studies on the

SA-TMV complex reveals inights into the design principles for successful

158

nanotechnologies effctively shielded from the immune system. This work into the biological interactions of viruses provides insights into what may occur in vivo and how to control those interactions for the development of new therapies.

Further understanding of the interplay between viruses and their environments may lead to critical breakthroughs for pharmacology.

159

Appendix 1: Models for Adenovirus Neutralization by HD5

The material in this appendix was adapted from a published meeting abstract,

Gulati NM, Smith JG, Nemerow GR, Stewart PL, 2014. CryoEM Based Models for Adenovirus Neutralization by Human Alpha-Defensin 5. Microscopy and

Microanalysis 20 (S3), 1406-1407 with permission from Cambridge University

Press (www.cambridge.org). I presented this work orally at the Microscopy and

Microanalysis Meeting in Hartford, CT on Wednesday, August 6, 2014.

A1.1 Abstract

Defensins are peptides of the innate immune system with potent antimicrobial activity. Despite strong interest in defensins as potential pharmaceutical compounds, commercial development of these peptides has been challenging. New generations of modified antimicrobial peptides are being pursued to improve their potential as pharmaceuticals. Strategies include peptide mimetics, hybrid peptides, peptide congeners, stabilized peptides, peptide conjugates and immobilized peptides (Brogden and Brogden, 2011). Our laboratories have been working on understanding the molecular mechanisms of defensins against adenovirus to characterize key interaction sites. There are six human alpha-defensins, including HD5, and multiple beta-defensins. Defensin action against enveloped viruses includes membrane disruption, as well as interference with viral membrane fusion. Alpha-defensins also neutralize viruses

160

that lack envelopes including human adenovirus (HAdV), human papillomavirus

(HPV), and polyomaviruses.

We have used cryo-electron microscopy (cryoEM) to characterize the

critical binding site of HD5 on a neutralization-sensitive HAdV vector (Smith et

al., 2010). In our working structural model HD5 stabilizes the vertex region of the

capsid by bridging two capsid proteins, penton base and fiber. This is thought to

block further capsid uncoating steps required for infectivity. Viral infectivity

studies with virus chimeras comprised of capsid proteins from sensitive and

resistant serotypes supported this model (Smith et al., 2010). To further describe

the critical binding site, we determined subnanometer resolution cryoEM

structures of HD5 complexed with both neutralization-sensitive and –resistant

HAdV chimeras (Flatt et al., 2013). CryoEM guided molecular dynamics flexible

fitting (MDFF) was used to model the interactions between HD5 and the two

chimeras. We found that the long penton base loops of the neutralization- sensitive chimera enveloped HD5 and provided significant stabilization to the vertex region (Figure A1.1). It has been shown that HD5 dimerization is important for anti-viral activity against HAdV5 (Gounder et al., 2012).

In this study we analyzed the known sensitivity of multiple HAdV types to

HD5 (Smith et al., 2010) in terms of the emerging structural model for HD5 neutralization of adenovirus. The characteristics of this model are a negatively charged sequence in fiber, represented by 18-EDES-21 in the HAdV35 fiber, and an RGD-containing penton base loop of 39 or more residues with a predicted intrinsically disordered region. The RGD loop length was determined by 161

sequence alignment to the HAdV2 RGD loop (K297-Q374), which is missing due

to disorder in the crystal structure (PDB:1X9T). These two characteristics are

conserved in multiple adenovirus types shown to be sensitive to neutralization by

HD5: adenovirus species B1, B2, C, and E. Adenovirus types resistant to HD5

neutralization differ in their fiber sequence (most have the more hydrophobic

GYAR sequence) and have shorter RGD loops (20 to 35 residues). Interestingly,

one adenovirus type, species A (type 12), is HD5-sensitive and yet has a short

RGD loop of only 15 residues. Modeling of the HAdV12 vertex suggests an alternative binding mode for HD5 (Figure A1.2).

We propose that for HAdV12 the short intrinsically disordered RGD loop of

penton base is complemented by a longer fiber N-terminal region, which also has

predicted intrinsic disorder. Modeling indicates that these two intrinsically

disordered regions could bracket an HD5 monomer or dimer on two sides. In our

alternative model for defensin neutralization of HAdV, HD5 binds to intrinsically

disordered regions in two capsid proteins thereby stabilizing the vertex complex

of penton base and fiber, and blocking further capsid uncoating steps required for

infectivity.

162

Figure A1.1. Proposed binding mode for HD5 with the HAdV5 penton of the defensin-sensitive chimera. This model is based on hybrid cryoEM and MDFF results (Flatt et al., 2013; Smith et al., 2010). A) Atomic model of the penton base

(brown with RGD loops in yellow) and fiber (green) of the chimera with docked

HD5 monomers (red). The critical fiber sequence 18-EDES-21 is indicated by an asterisk. B) Enlarged view of one HD5 binding site. HD5 binds above the critical fiber sequence and is enveloped by the RGD loops. HD5 monomers are shown for simplicity, but dimers can be accommodated. Modified from (Flatt et al.,

2013).

163

Figure A1.2. Proposed alternative binding mode for HD5 with HAdV12 penton. A) Atomic model of HAdV12 penton base and fiber with one docked HD5 monomer, colored as in Figure A1.1 with the intrinsically disordered region at the

N-terminus of the fiber shown in blue. B) Enlarged view of one HD5 binding site.

HD5 is bracketed by intrinsically disordered peptide regions (wide tubes) in penton base and the fiber N-terminal region. An HD5 monomer is shown for simplicity, but a dimer can be accommodated.

164

A1.2. Acknowledgements

The authors acknowledge funding from the NIH National Institute of Allergy and

Infectious Diseases (AI042929) to PLS. We thank Dr. JW Flatt for many useful discussions.

165

Appendix 2: Shielding of SA on TMV nanoparticle

The material in this appendix was adapted from a published meeting abstract,

Gulati NM, Pitek AS, Steinmetz NF, Stewart PL, 2016. Characterization of the

Shielding Properties of Serum Albumin on a Plant Viral Nanoparticle. Microscopy

and Microanalysis 22 (S3), 1084-1085 with permission from Cambridge

University Press (www.cambridge.org). I presented this work orally at the

Microscopy and Microanalysis Meeting in Columbus, OH on Tuesday, July 26,

2016.

Abstract

Nanoparticle development is an important avenue of research for drug

delivery in the field of cancer, in order to deliver a high payload of therapeutic

molecules to a tumor site while reducing off-target effects elsewhere in the body.

However, a critical barrier for the success of nanoparticle platforms is the rapid

elimination of nanoparticles from circulation. To shield nanoparticles from

recognition and clearance by the immune system, drug delivery platforms are often coated in polyethylene glycol (PEG), which induces a ‘stealth effect’.

PEGylation reduces serum protein adsorption on the surface of the nanoparticle and therefore reduces immune detection; however PEGylation also has disadvantages. The effects of PEGylation are dependent on the physical characteristics of the PEG polymer used, including chain length and architecture 166

(Lee et al., 2015a), which may be due to the polymer’s flexible and hydrophilic nature. Also, a recent study has shown that as much as 25% of the population has PEG-specific antibodies, which would allow for enhanced clearance of

PEGylated nanoparticles (Garay et al., 2012). This is a drastic increase from the less than 1% of the population measured in a study performed in 1984 (Richter and Akerblom, 1984). Furthermore, a nanoparticle’s behavior in vivo can be altered by the proteins adsorbed to its surface during circulation, a phenomenon known as formation of a protein corona (Pitek et al., 2016a; Walczyk et al.,

2010). We have recently developed a nanoparticle platform with a manufactured protein corona that allows for shielding from the immune system. Our platform consists of a plant viral nanoparticle, tobacco mosaic virus (TMV), conjugated to serum albumin (SA) through a short PEG linker. SA is the most abundant plasma protein and therefore acts as a camouflage. Our preliminary results have shown that SA shielding reduces antibody recognition of the nanoparticle better than

TMV coated with a short PEG linker alone (data not shown). We predict the SA molecules are evenly distributed on the surface of our SA-shielded TMV nanoparticles. Furthermore, we expect that the short PEG linker attaching SA to the TMV surface does not allow much flexibility of the SA molecules. Therefore, we hypothesize that antibodies are unable to penetrate the SA layer in order to interact with the TMV nanoparticle surface. In this study, the physical properties of SA on the surface of TMV are characterized to better understand the mechanism of immune shielding.

167

TMV nanoparticles are produced through infection of Nicotiana benthamiana plants as described previously (Bruckman and Steinmetz, 2014).

Purified nanoparticles are conjugated to SA with a short PEG linker through NHS and maleimide chemistries. Conjugation to TMV is confirmed via SDS-PAGE analysis. Purified particles are characterized for antibody recognition through immunogold labeling and negative stain electron microscopy. Further characterization of SA on the surface of TMV is done through cryo-electron microscopy (cryoEM) and tomography (cryoET) to visualize coverage of TMV and the distribution of the shielding protein on the surface of TMV. CryoEM grids are imaged on a JEOL 2200FS 200kV (FEG, energy filter) transmission cryo- electron microscope with a DE20 direct electron detector.

Our results indicate that SA effectively reduces antibody recognition of

TMV nanoparticles as compared to bare TMV or TMV-PEG. This is presumably due to well-distributed SA proteins on the surface of TMV as shown by cryoEM images (Figure A2.1). CryoET studies will provide more detailed 3D information on SA coverage of the TMV surface. Further studies will investigate alternative

SA attachments to TMV, including longer PEG linkers and varying coverage levels.

168

A B

Figure A2.1. Sections of cryo-electron micrographs of A) bare TMV nanoparticles and B) TMV nanoparticles conjugated to SA via a short (8- mer) PEG linker. As compared to the bare TMV nanoparticles, TMV-SA particles show protrusions all along the length of the rod shaped TMV, indicative of well- distributed SA conjugated to the surface of the nanoparticles. Scale bar is 25 nm.

Black circles are 10nm gold fiducials.

169

Appendix 3: TEM of targeted PVX

Reprinted with permission from Chariou PL, Lee KL, Wen AM, Gulati NM,

Stewart PL, Steinmetz NF, 2015. Detection and imaging of aggressive cancer cells using an epidermal growth factor receptor (EGFR)-targeted filamentous plant virus-based nanoparticle. Bioconjugate Chemistry 26 (2), 262-269.

My role as co-author to this work was to perform TEM of modified PVX nanoparticles.

Abstract

Molecular imaging approaches and targeted drug delivery hold promise for earlier detection of diseases and treatment with higher efficacy while reducing side effects, therefore increasing survival rates and quality of life. Virus-based nanoparticles are a promising platform because their scaffold can be manipulated both genetically and chemically to simultaneously display targeting ligands while carrying payloads for diagnosis or therapeutic intervention. Here, we displayed a 12-amino-acid peptide ligand, GE11 (YHWYGYTPQNVI), on nanoscale filaments formed by the plant virus potato virus X (PVX).

Bioconjugation was used to produce fluorescently labeled PVX-GE11 filaments targeted toward the epidermal growth factor receptor (EGFR). Cell detection and 170

imaging was demonstrated using human skin epidermoid carcinoma, colorectal adenocarcinoma, and triple negative breast cancer cell lines (A-431, HT-29,

MDA-MB-231), all of which upregulate EGFR to various degrees. Nonspecific uptake in ductal breast carcinoma (BT-474) cells was not observed. Furthermore, co-culture experiments with EGFR(+) cancer cells and macrophages indicate successful targeting and partitioning toward the cancer cells. This study lays a foundation for the development of EGFR-targeted filaments delivering contrast agents for imaging and diagnosis, and/or toxic payloads for targeted drug delivery.

171

Appendix 4: TEM of silica-coated TMV rods and spheres

Reprinted with permission from Bruckman MA, Randolph LN, Gulati NM, Stewart

PL, Steinmetz NF, 2015. Silica-coated Gd(DOTA)-loaded protein nanoparticles

enable magnetic resonance imaging of macrophages. Journal of Materials

(www.acs.org).

My role as co-author to this work was to perform TEM of silica-coated TMV rods and spheres and immunogold-stained silica coated TMV rods and spheres.

Abstract

The molecular imaging of in vivo targets allows non-invasive disease

diagnosis. Nanoparticles offer a promising platform for molecular imaging

because they can deliver large payloads of imaging reagents to the site of

disease. Magnetic resonance imaging (MRI) is often preferred for clinical

diagnosis because it uses non-ionizing radiation and offers both high spatial

resolution and excellent penetration. We have explored the use of plant viruses

as the basis of for MRI contrast reagents, specifically Tobacco mosaic

virus (TMV), which can assemble to form either stiff rods or spheres. We loaded

TMV particles with paramagnetic Gd ions, increasing the ionic relaxivity

compared to free Gd ions. The loaded TMV particles were then coated with silica 172

maintaining high relaxivities. Interestingly, we found that when Gd(DOTA) was loaded into the interior channel of TMV and the exterior was coated with silica,

-1 -1 -1 -1 the T1 relaxivities increased by three-fold from 10.9 mM s to 29.7 mM s at 60

MHz compared to uncoated Gd-loaded TMV. To test the performance of the contrast agents in a biological setting, we focused on interactions with macrophages because the active or passive targeting of immune cells is a popular strategy to investigate the cellular components involved in disease progression associated with inflammation. In vitro assays and phantom MRI experiments indicate efficient targeting and imaging of macrophages, enhanced contrast-to-noise ratio was observed by shape-engineering (SNP > TMV) and silica-coating (Si-TMV/SNP > TMV/SNP). Because plant viruses are in the food chain, antibodies may be prevalent in the population. Therefore we investigated whether the silica-coating could prevent antibody recognition; indeed our data indicate that mineralization can be used as a stealth coating option to reduce clearance. Therefore, we conclude that the silica-coated protein-based contrast agent may provide an interesting candidate material for further investigation for in vivo delineation of disease through macrophage imaging.

173

Appendix 5: TEM of short TMV rods

Reprinted from Lam PL, Gulati NM, Stewart PL, Keri RA, Steinmetz NF, 2016.

Bioengineering of Tobacco Mosaic Virus to Create a Non-Infectious Positive

Control for Ebola Diagnostic Assays. Scientific Reports. 6, 23803 with permission

from the Nature Publishing Group (www.nature.com).

My role as co-author to this work was to perform TEM of short TMV rods loaded

with Ebola RNA sequences.

Abstract

The 2014 Ebola epidemic is the largest to date. There is no cure or

treatment for this deadly disease; therefore there is an urgent need to develop

new diagnostics to accurately detect Ebola. Current RT-PCR assays lack

sensitive and reliable positive controls. To address this critical need, we devised

a bio-inspired positive control for use in RT-PCR diagnostics: we encapsulated

scrambled Ebola RNA sequences inside of tobacco mosaic virus to create a

biomimicry that is non-infectious, but stable, and could therefore serve as a

positive control in Ebola diagnostic assays. Here, we report the bioengineering

and validation of this probe.

174

Appendix 6: TEM of serum-aggregated PVX nanoparticles

Reprinted with permission from Shukla S, Dorand RD, Myers JT, Woods SE,

Gulati NM, Stewart PL, Commandeur U, Huang AY, Steinmetz NF, 2016.

Multiple Administrations of Viral Nanoparticles Alter in Vivo Behavior-Insights from Intravital Microscopy. ACS Biomaterials Science and Engineering. 2 (5),

My role as co-author to this work was to perform TEM of PVX nanoparticles aggregated by serum.

Abstract

Multiple administrations of nanoparticle-based formulations are often a clinical requirement for drug delivery and diagnostic imaging applications. Steady pharmacokinetics of nanoparticles is desirable to achieve efficient therapeutic or diagnostic outcomes over such repeat administrations. While clearance through mononuclear phagocytic system is a key determinant of nanoparticle persistence in vivo, multiple administrations could potentially result in altered pharmacokinetics by evoking innate or adaptive immune responses. Plant viral nanoparticles (VNPs) represent an emerging class of programmable nanoparticle platform technologies that offer a highly organized proteinaceous architecture and multivalency for delivery of large payloads of drugs and molecular contrast 175

agents. These very structural features also render them susceptible to immune recognition and subsequent accelerated systemic clearance that could potentially affect overall efficiency. While the biodistribution and pharmacokinetics of VNPs have been reported, the biological response following repeat administrations remains an understudied area of investigation. Here, we demonstrate that weekly administration of filamentous plant viruses results in the generation of increasing levels of circulating, carrier-specific IgM and IgG antibodies. Furthermore, PVX specific immunoglobulins from the serum of immunized animals quickly form aggregates when incubated with PVX in vitro. Such aggregates of VNP-immune complexes are also observed in the mouse vasculature in vivo following repeat injections when imaged in real time using intravital two-photon laser scanning microscopy (2P-LSM). The size of aggregates diminishes at later time points, coinciding with antibody class switching from IgM to IgG. Together, our results highlight the need for careful in vivo assessment of (viral) nanoparticle-based platform technologies, especially in studying their performance after repeat administration. We also demonstrate the utility of intravital microscopy to aid in this evaluation.

176

Appendix 7: C shell script for preparing and submitting Frealign

7.07 scripts

The material in this appendix is reproduced from a script that I wrote

(/home/nmg42/frealign/make_frealign_submit-and-sbatch.csh) to prepare

multiple submission scripts on the HPC Cluster at CWRU for Frealign 7.07.

Frealign 7.07 is a program created by Niko Grigorieff for three-dimensional (3D) reconstructions of single-particle cryo-electron microscopy data (Grigorieff,

2007). It can be used for search and refinement of particle parameters, contrast transfer function (CTF) correction, and 3D reconstruction. The Stewart Lab has adapted Frealign 7.07 into three different modes. The first mode (basic) runs

Frealign in its original form. The second mode (hipr_var) was written by Dr.

Phoebe L. Stewart to rescue particles that have high phase residuals “hipr” by varying particle centers during the angular search. The third mode (lopr_fstep) was also written by Dr. Phoebe L. Stewart to refine only particles with low phase residuals “lopr” with a specified step size for refinement of angles (fstep).

The following script prepares submission scripts which divide the dataset into smaller groups for parallelized processing. If using the hipr_var mode of

Frealign 7.07, it creates scripts that specify the variations in particle centers specified by the user. All scripts prepared are based on a user-provided template script. Each script will be in a created folder, named based on the partition of the

177

dataset, and the centers used if using hipr_var. After creation, all scripts will be submitted to the queue for data processing.

A3.1 Script for making Frealign submission scripts

#!/bin/csh -f

############################################################

####

# This script will produce a folder with submit scripts for

# Frealign 7.07 based on an input template script. It will divide

# into many parts based on your number of particles. It will

# also add in the correct centers for hipr_var jobs.

# It will then submit the scripts to the CWRU HPC Slurm cluster.

# Lopr_fstep specific scripts may come soon...

# A version of this script (without sbatch) is also available

# in the same folder. It is recommended to first try making

# scripts without submitting them to the queue

# so the user can look over them and make sure everything

# is correct.

# Made by Neetu Gulati, April 17, 2017

####

############################################################

178

###Define Parameters

echo "This script is designed to produce submit scripts for Frealign 7.07” echo “based on a given template script." echo "It will divide your data into multiple parts, create individual folders for them,” echo “and write executable scripts in those folders." echo "If you are varying centers with Frealign hipr_var,” echo “it will also create folders for different centers." echo "This script will change particle numbers, center positions, and para file names.” echo “Everything else must be correct in your template script." echo " " echo " " echo "What is your input template script?" set input_script = $< echo "Input script is "$input_script echo " " echo "Which kind of Frealign are you using?" echo "Please enter either basic, hipr_var, or lopr_fstep"

###For now, this script will prepare files for hipr_var or otherwise.

###Lopr_fstep customization will be built in later. set frealign_flavor = $<

179

echo "Using "$frealign_flavor" frealign..." echo " " echo "What type of symmetry?" echo "Please make sure this symmetry matches your template script" set symm = $< echo "Symmetry line says: "$symm echo " " echo "What is your first particle number?" set init_part = $< echo "First particle at "$init_part echo "What is your last particle number?" set last_part = $< echo "Last particle is "$last_part echo "How many particles per group?"

@ divisor = $< echo "Each script will have "$divisor" particles" echo " " echo "What is the base of your output para/parash file name (no extention)?" set para_name = $< echo "Your para and parash files will be start with "$para_name

180

###This part of the script identifies how long the template script is and where the

###particle numbers will go (one line before the symmetry line)

set line_symm = `grep -n $symm $input_script | cut -d ':' -f 1` set line_count = `wc -l $input_script | cut -d ' ' -f 1`

@ lines_before = $line_symm - 2

@ line = $line_symm - 1 echo " "

###This part of the script says checks if you are using hipr_var.

###If so it asks about centers, otherwise it goes to the else. if ( $frealign_flavor == hipr_var ) then

echo "How many centers will you need?"

@ num_centers = $<

echo "Using "$num_centers" centers total"

@ c = 1

###For each center used, determine positions and make a folder

while ( $c <= $num_centers )

echo " "

echo "For center "$c

echo "What is the position in X?"

181

echo "If using a whole number, please do not use the decimal place" echo "(i.e. enter '10' instead of '10.0')" set x = $< echo $x set tmp_x_p1 = `echo $x | cut -d '.' -f 1` set tmp_x_p2 = `echo $x | cut -d '.' -f 2 -s` if ( "$tmp_x_p2" == "" ) then

set x_period = $tmp_x_p1

set has_decimal_x = false else

set x_period = $tmp_x_p1"p"$tmp_x_p2

set has_decimal_x = true endif if ($tmp_x_p1 < 0) then

set tmp_x = `echo $x_period | cut -c 2-`

set new_x = "m"$tmp_x

echo $new_x else

set new_x = $x_period

echo $new_x endif

182

echo "What is the position in Y?"

echo "If using a whole number, please do not use the decimal place"

echo "(i.e. enter '10' instead of '10.0')"

set y = $<

echo $y

set tmp_y_p1 = `echo $y | cut -d '.' -f 1`

echo $tmp_y_p1

set tmp_y_p2 = `echo $y | cut -d '.' -f 2 -s`

if ( "$tmp_y_p2" == "" ) then

set y_period = $tmp_y_p1

set has_decimal_y = false

else

set y_period = $tmp_y_p1"p"$tmp_y_p2

set has_decimal_y = true

endif

echo $y_period

if ($tmp_y_p1 < 0) then

set tmp_y = `echo $y_period | cut -c 2-`

set new_y = "m"$tmp_y

echo $new_y else

set new_y = $y_period

echo $new_y

183

endif

echo "Center at "$x","$y

set cent_folder = "cent_"$new_x"_"$new_y

mkdir $cent_folder

@ i = $init_part

@ j = $i + $divisor - 1

@ k = 1

###Within that folder, make a folder for each set of particles and write a script

###within it, based on the template script. while ( $i <= $last_part )

set p_value = "p"$k

echo "In the folder "$p_value

mkdir $cent_folder/$p_value

set new_script = $input_script"-"$cent_folder"-"$p_value

echo "Script will be called "$new_script

head -n $lines_before $input_script > $cent_folder/$p_value/$new_script

if ( $j <= $last_part ) then

echo "It will contain particles "$i" through "$j

echo $i" "$j >> $cent_folder/$p_value/$new_script

else

184

echo "It will contain particles "$i" through "$last_part

echo $i" "$last_part >> $cent_folder/$p_value/$new_script

endif

grep -A 5 $symm $input_script >> $cent_folder/$p_value/$new_script

if ( "$has_decimal_x" == "true" ) then

if ("$has_decimal_y" == "true" ) then

echo $x" "$y >> $cent_folder/$p_value/$new_script

else

echo $x" "$y".0" >> $cent_folder/$p_value/$new_script

endif

else

if ("$has_decimal_y" == "true" ) then

echo $x".0 "$y >> $cent_folder/$p_value/$new_script

else

echo $x".0 "$y".0" >> $cent_folder/$p_value/$new_script

endif

echo $para_name"-"$cent_folder"-"$p_value".para" >>

$cent_folder/$p_value/$new_script

echo $para_name"-"$cent_folder"-"$p_value".parash" >>

$cent_folder/$p_value/$new_script

185

@ lines_after = $line_count - $line_symm - 8

tail -n $lines_after $input_script >> $cent_folder/$p_value/$new_script

chmod +x $cent_folder/$p_value/$new_script

###Go into the folder and submit the script. USE WITH CAUTION!

###Comment out if unsure.

cd $cent_folder/$p_value

sbatch $new_script

cd ../..

@ i = $i + $divisor

@ j = $j + $divisor

@ k++

end

@ c = $c + 1

end

###If you are not varying centers, go straight to making particle division

###folders and scripts within them.

186

else

echo "moving to particle numbers..."

@ i = $init_part

@ j = $i + $divisor - 1

@ k = 1

while ( $i <= $last_part )

set p_value = "p"$k

echo "In the folder "$p_value

mkdir $p_value

set new_script = $input_script"-"$p_value

echo "The script will be "$new_script

head -n $lines_before $input_script > $p_value/$new_script

if ( $j <= $last_part ) then

echo "It will contain particles "$i" through "$j

echo $i" "$j >> $p_value/$new_script

else

echo "It will contain particles "$i" through "$last_part

echo $i" "$last_part >> $p_value/$new_script

endif

grep -A 5 $symm $input_script >> $p_value/$new_script

echo $para_name"-"$p_value".para" >> $p_value/$new_script

187

echo $para_name"-"$p_value".parash" >> $p_value/$new_script

@ lines_after = $line_count - $line_symm - 7

tail -n $lines_after $input_script >> $p_value/$new_script

chmod +x $p_value/$new_script

###Go into the folder and submit the script. USE WITH CAUTION!

###Comment out if unsure.

cd $p_value

sbatch $new_script

cd ..

@ i = $i + $divisor

@ j = $j + $divisor

@ k++

end endif end

188

A3.2 Example template script

#!/bin/tcsh

#SBATCH --nodes=1

#SBATCH --cpus-per-task=1

#SBATCH --time=100:00:00

#SBATCH --mem=4gb

/home/pls47/programs/frealign707/frealign/bin/frealign707_400lm_hipr_var_pad4

00.exe << END1

M 3 F F F F 0 F F T T

315.0 210.0 2.52 0.07 1.0 100.0 60.0 1.0 20 20

1 1 1 1 1

1 800

1.0 12.8 10.0 72.0 2.7 300.0 0.0 0.0

5.0 100.0 8.5 0.0

../../../stack_hpv-hd5_apix-2p52_black.mrc proj.mrc

../../hpv-hd5-400-rnd3.para

0.0 10.0 hpv-hd5-400-rnd3-cent_0_10-p1.para hpv-hd5-400-rnd3-cent_0_10-p1.parash

-100.0 12.8 10.0 72.0 2.7 300.0 0.0 0.0

../../../HPV-HD5-postprocess-8p54237.mrc

189

fweigh.mrc fqf.mrc famp.mrc fpha.mrc fpoi.mrc

END1

190

Bibliography

Abu Lila, A.S., Kiwada, H., Ishida, T., 2013. The accelerated blood clearance

(ABC) phenomenon: clinical challenge and approaches to manage. J

Control Release 172, 38-47.

Abuchowski, A., McCoy, J.R., Palczuk, N.C., van Es, T., Davis, F.F., 1977. Effect

of covalent attachment of polyethylene glycol on immunogenicity and

circulating life of bovine liver catalase. J Biol Chem 252, 3582-3586.

Ali, A., Suhail, M., Mathew, S., Shah, M.A., Harakeh, S.M., Ahmad, S., Kazmi, Z.,

Alhamdan, M.A., Chaudhary, A., Damanhouri, G.A., Qadri, I., 2016.

Nanomaterial Induced Immune Responses and Cytotoxicity. J Nanosci

Nanotechnol 16, 40-57.

Altintoprak, K., Seidenstucker, A., Welle, A., Eiben, S., Atanasova, P., Stitz, N.,

Plettl, A., Bill, J., Gliemann, H., Jeske, H., Rothenstein, D., Geiger, F.,

Wege, C., 2015. Peptide-equipped tobacco mosaic virus templates for

selective and controllable biomineral deposition. Beilstein J Nanotechnol

6, 1399-1412.

Amoozgar, Z., Park, J., Lin, Q., Yeo, Y., 2012. Low molecular-weight chitosan as

a pH-sensitive stealth coating for tumor-specific drug delivery. Mol Pharm

9, 1262-1270.

191

Anderson, C.L., Chaudhury, C., Kim, J., Bronson, C.L., Wani, M.A., Mohanty, S.,

2006. Perspective-- FcRn transports albumin: relevance to immunology

and medicine. Trends Immunol 27, 343-348.

Anraku, M., Shintomo, R., Taguchi, K., Kragh-Hansen, U., Kai, T., Maruyama, T.,

Otagiri, M., 2015. Amino acids of importance for the antioxidant activity of

human serum albumin as revealed by recombinant mutants and genetic

variants. Life Sci 134, 36-41.

Armstrong, J.K., Hempel, G., Koling, S., Chan, L.S., Fisher, T., Meiselman, H.J.,

Garratty, G., 2007. Antibody against poly(ethylene glycol) adversely affects

PEG-asparaginase therapy in acute lymphoblastic leukemia patients.

Cancer 110, 103-111.

Asselin, A., 1979. Presence of Tmv and Pvx in Various Tomato Cultivars.

Phytoprotection 60, 170-171.

Balique, F., Colson, P., Raoult, D., 2012. Tobacco mosaic virus in cigarettes and

saliva of smokers. J Clin Virol 55, 374-376.

Becker, K.A., Florin, L., Sapp, C., Sapp, M., 2003. Dissection of human

papillomavirus type 33 L2 domains involved in nuclear domains (ND) 10

homing and reorganization. Virology 314, 161-167.

Bellido, E., Hidalgo, T., Lozano, M.V., Guillevic, M., Simon-Vazquez, R.,

Santander-Ortega, M.J., Gonzalez-Fernandez, A., Serre, C., Alonso, M.J.,

Horcajada, P., 2015. Heparin-engineered mesoporous iron metal-organic

192

framework nanoparticles: toward stealth drug nanocarriers. Adv Healthc

Mater 4, 1246-1257.

Bergant, M., Banks, L., 2013. SNX17 facilitates infection with diverse

papillomavirus types. J Virol 87, 1270-1273.

Bergant Marusic, M., Ozbun, M.A., Campos, S.K., Myers, M.P., Banks, L., 2012.

Human papillomavirus L2 facilitates viral escape from late endosomes via

sorting nexin 17. Traffic 13, 455-467.

Bernard, H.U., Burk, R.D., Chen, Z., van Doorslaer, K., zur Hausen, H., de

Villiers, E.M., 2010. Classification of papillomaviruses (PVs) based on 189

PV types and proposal of taxonomic amendments. Virology 401, 70-79.

Bienkowska-Haba, M., Williams, C., Kim, S.M., Garcea, R.L., Sapp, M., 2012.

Cyclophilins facilitate dissociation of the human papillomavirus type 16

capsid protein L1 from the L2/DNA complex following virus entry. J Virol

86, 9875-9887.

Blanco, E., Shen, H., Ferrari, M., 2015. Principles of nanoparticle design for

overcoming biological barriers to drug delivery. Nat Biotechnol 33, 941-

951.

Bludau, H., Czapar, A.E., Pitek, A.S., Shukla, S., Jordan, R., Steinmetz, N.F.,

POxylation as an alternative stealth coating for biomedical applications.

European Polymer Journal.

193

Bodily, J., Laimins, L.A., 2011. Persistence of human papillomavirus infection:

keys to malignant progression. Trends Microbiol 19, 33-39.

Bosch, F.X., Lorincz, A., Munoz, N., Meijer, C.J., Shah, K.V., 2002. The causal

relation between human papillomavirus and cervical cancer. J Clin Pathol

55, 244-265.

Bossis, I., Roden, R.B., Gambhira, R., Yang, R., Tagaya, M., Howley, P.M.,

Meneses, P.I., 2005. Interaction of tSNARE syntaxin 18 with the

papillomavirus minor capsid protein mediates infection. J Virol 79, 6723-

6731.

Brogden, N.K., Brogden, K.A., 2011. Will new generations of modified

antimicrobial peptides improve their potential as pharmaceuticals? Int J

Antimicrob Agents 38, 217-225.

Bronnimann, M.P., Chapman, J.A., Park, C.K., Campos, S.K., 2013. A

transmembrane domain and GxxxG motifs within L2 are essential for

papillomavirus infection. J Virol 87, 464-473.

Brown, E.J., Frazier, W.A., 2001. Integrin-associated protein (CD47) and its

ligands. Trends Cell Biol 11, 130-135.

Bruckman, M.A., Steinmetz, N.F., 2014. Chemical modification of the inner and

outer surfaces of Tobacco Mosaic Virus (TMV). Methods Mol Biol 1108,

173-185.

194

Bruckman, M.A., VanMeter, A., Steinmetz, N.F., 2015a. Nanomanufacturing of

Tobacco Mosaic Virus-Based Spherical Biomaterials Using a Continuous

Flow Method. ACS Biomater Sci Eng 1, 13-18.

Bruckman, M.A., Randolph, L.N., Gulati, N.M., Stewart, P.L., Steinmetz, N.F.,

2015b. Silica-coated Gd(DOTA)-loaded protein nanoparticles enable

magnetic resonance imaging of macrophages. J Mater Chem B Mater Biol

Med 3, 7503-7510.

Bruckman, M.A., Czapar, A.E., VanMeter, A., Randolph, L.N., Steinmetz, N.F.,

2016. Tobacco mosaic virus-based protein nanoparticles and nanorods for

chemotherapy delivery targeting breast cancer. J Control Release 231,

103-113.

Bruckman, M.A., Randolph, L.N., VanMeter, A., Hern, S., Shoffstall, A.J., Taurog,

R.E., Steinmetz, N.F., 2014a. Biodistribution, pharmacokinetics, and blood

compatibility of native and PEGylated tobacco mosaic virus nano-rods and

-spheres in mice. Virology 449, 163-173.

Bruckman, M.A., Jiang, K., Simpson, E.J., Randolph, L.N., Luyt, L.G., Yu, X.,

Steinmetz, N.F., 2014b. Dual-modal magnetic resonance and fluorescence

imaging of atherosclerotic plaques in vivo using VCAM-1 targeted tobacco

mosaic virus. Nano Lett 14, 1551-1558.

Buck, C.B., Trus, B.L., 2012. The papillomavirus virion: a machine built to hide

molecular Achilles' heels. Adv Exp Med Biol 726, 403-422.

195

Buck, C.B., Day, P.M., Trus, B.L., 2013. The papillomavirus major capsid protein

L1. Virology 445, 169-174.

Buck, C.B., Day, P.M., Thompson, C.D., Lubkowski, J., Lu, W., Lowy, D.R.,

Schiller, J.T., 2006. Human alpha-defensins block papillomavirus infection.

Proc Natl Acad Sci U S A 103, 1516-1521.

Buck, C.B., Cheng, N., Thompson, C.D., Lowy, D.R., Steven, A.C., Schiller, J.T.,

Trus, B.L., 2008. Arrangement of L2 within the papillomavirus capsid. J

Virol 82, 5190-5197.

Butler, P.J., 1984. The current picture of the structure and assembly of tobacco

mosaic virus. J Gen Virol 65 ( Pt 2), 253-279.

Cai, W.Q., Fang, R.X., Shang, H.S., Wang, X., Zhang, F.L., Li, Y.R., Zhang, J.C.,

Cheng, X.Y., Wang, G.L., Mang, K.Q., 2003. Development of CMV- and

TMV-resistant transgenic chili pepper: field performance and biosafety

assessment. Mol Breeding 11, 25-35.

Campos, S.K., Chapman, J.A., Deymier, M.J., Bronnimann, M.P., Ozbun, M.A.,

2012. Opposing effects of bacitracin on human papillomavirus type 16

infection: enhancement of binding and entry and inhibition of endosomal

penetration. J Virol 86, 4169-4181.

Cardone, G., Moyer, A.L., Cheng, N., Thompson, C.D., Dvoretzky, I., Lowy, D.R.,

Schiller, J.T., Steven, A.C., Buck, C.B., Trus, B.L., 2014. Maturation of the

human papillomavirus 16 capsid. MBio 5, e01104-01114.

196

Centers for Disease, C., Prevention, 2011. National and state vaccination

coverage among adolescents aged 13 through 17 years--United States,

2010. MMWR Morb Mortal Wkly Rep 60, 1117-1123.

Chang, C.H., Yen, M.C., Liao, S.H., Hsu, Y.L., Lai, C.S., Chang, K.P., Hsu, Y.L.,

2017. Secreted Protein Acidic and Rich in Cysteine (SPARC) Enhances

Cell Proliferation, Migration, and Epithelial Mesenchymal Transition, and

SPARC Expression is Associated with Tumor Grade in Head and Neck

Cancer. Int J Mol Sci 18.

Chariou, P.L., Steinmetz, N.F., 2017. Delivery of Pesticides to Plant Parasitic

Nematodes Using Tobacco Mild Green Mosaic Virus as a Nanocarrier.

ACS Nano 11, 4719-4730.

Chaudhury, C., Mehnaz, S., Robinson, J.M., Hayton, W.L., Pearl, D.K.,

Roopenian, D.C., Anderson, C.L., 2003. The major histocompatibility

complex-related Fc receptor for IgG (FcRn) binds albumin and prolongs its

lifespan. J Exp Med 197, 315-322.

Chen, J., Dong, X., Feng, T., Lin, L., Guo, Z., Xia, J., Tian, H., Chen, X., 2015.

Charge-conversional zwitterionic copolymer as pH-sensitive shielding

system for effective tumor treatment. Acta Biomater 26, 45-53.

Chen, S., Cao, Z., Jiang, S., 2009. Ultra-low fouling peptide surfaces derived

from natural amino acids. Biomaterials 30, 5892-5896.

197

Cheng, T.L., Chuang, K.H., Chen, B.M., Roffler, S.R., 2012. Analytical

measurement of PEGylated molecules. Bioconjug Chem 23, 881-899.

Christensen, N.D., Kreider, J.W., Kan, N.C., DiAngelo, S.L., 1991. The open

reading frame L2 of cottontail rabbit papillomavirus contains antibody-

inducing neutralizing epitopes. Virology 181, 572-579.

Creager, A.N., Morgan, G.J., 2008. After the double helix: Rosalind Franklin's

research on Tobacco mosaic virus. Isis 99, 239-272.

Czapar, A.E., Zheng, Y.R., Riddell, I.A., Shukla, S., Awuah, S.G., Lippard, S.J.,

Steinmetz, N.F., 2016. Tobacco Mosaic Virus Delivery of Phenanthriplatin

for Cancer therapy. ACS Nano 10, 4119-4126. da Silva, R.M., de Souto, E.R., Pedroso, J.C., Arakava, R., Almeida, A.M.R.,

Barboza, A.A.L., Vida, J.B., 2008. Detection and Identification of TMV

Infecting Tomato Under Protected Cultivation in Parana State. Braz Arch

Biol Techn 51, 903-909.

Dams, E.T., Laverman, P., Oyen, W.J., Storm, G., Scherphof, G.L., van Der Meer,

J.W., Corstens, F.H., Boerman, O.C., 2000. Accelerated blood clearance

and altered biodistribution of repeated injections of sterically stabilized

liposomes. J Pharmacol Exp Ther 292, 1071-1079.

Darshan, M.S., Lucchi, J., Harding, E., Moroianu, J., 2004. The l2 minor capsid

protein of human papillomavirus type 16 interacts with a network of

nuclear import receptors. J Virol 78, 12179-12188.

198

Day, P.M., Baker, C.C., Lowy, D.R., Schiller, J.T., 2004. Establishment of

papillomavirus infection is enhanced by promyelocytic leukemia protein

(PML) expression. Proc Natl Acad Sci U S A 101, 14252-14257.

Day, P.M., Gambhira, R., Roden, R.B., Lowy, D.R., Schiller, J.T., 2008.

Mechanisms of human papillomavirus type 16 neutralization by l2 cross-

neutralizing and l1 type-specific antibodies. J Virol 82, 4638-4646. de Koning, M.N., Quint, K.D., Bruggink, S.C., Gussekloo, J., Bouwes Bavinck,

J.N., Feltkamp, M.C., Quint, W.G., Eekhof, J.A., 2015. High prevalence of

cutaneous warts in elementary school children and the ubiquitous

presence of wart-associated human papillomavirus on clinically normal

skin. Br J Dermatol 172, 196-201. de Martel, C., Plummer, M., Vignat, J., Franceschi, S., 2017. Worldwide burden

of cancer attributable to HPV by site, country and HPV type. Int J Cancer

141, 664-670. de Martel, C., Ferlay, J., Franceschi, S., Vignat, J., Bray, F., Forman, D.,

Plummer, M., 2012. Global burden of cancers attributable to infections in

2008: a review and synthetic analysis. Lancet Oncol 13, 607-615. de Sanjose, S., Quint, W.G., Alemany, L., Geraets, D.T., Klaustermeier, J.E.,

Lloveras, B., Tous, S., Felix, A., Bravo, L.E., Shin, H.R., Vallejos, C.S., de

Ruiz, P.A., Lima, M.A., Guimera, N., Clavero, O., Alejo, M., Llombart-

Bosch, A., Cheng-Yang, C., Tatti, S.A., Kasamatsu, E., Iljazovic, E., Odida,

199

M., Prado, R., Seoud, M., Grce, M., Usubutun, A., Jain, A., Suarez, G.A.,

Lombardi, L.E., Banjo, A., Menendez, C., Domingo, E.J., Velasco, J.,

Nessa, A., Chichareon, S.C., Qiao, Y.L., Lerma, E., Garland, S.M.,

Sasagawa, T., Ferrera, A., Hammouda, D., Mariani, L., Pelayo, A., Steiner,

I., Oliva, E., Meijer, C.J., Al-Jassar, W.F., Cruz, E., Wright, T.C., Puras, A.,

Llave, C.L., Tzardi, M., Agorastos, T., Garcia-Barriola, V., Clavel, C., Ordi,

J., Andujar, M., Castellsague, X., Sanchez, G.I., Nowakowski, A.M.,

Bornstein, J., Munoz, N., Bosch, F.X., Retrospective International, S.,

Group, H.P.V.T.T.S., 2010. Human papillomavirus genotype attribution in

invasive cervical cancer: a retrospective cross-sectional worldwide study.

Lancet Oncol 11, 1048-1056.

Demers, L.M., Mirkin, C.A., Mucic, R.C., Reynolds, R.A., 3rd, Letsinger, R.L.,

Elghanian, R., Viswanadham, G., 2000. A fluorescence-based method for

determining the surface coverage and hybridization efficiency of thiol-

capped oligonucleotides bound to gold thin films and nanoparticles. Anal

Chem 72, 5535-5541.

Desai, N., Trieu, V., Damascelli, B., Soon-Shiong, P., 2009. SPARC Expression

Correlates with Tumor Response to Albumin-Bound Paclitaxel in Head and

Neck Cancer Patients. Transl Oncol 2, 59-64.

Docter, D., Distler, U., Storck, W., Kuharev, J., Wunsch, D., Hahlbrock, A.,

Knauer, S.K., Tenzer, S., Stauber, R.H., 2014. Quantitative profiling of the

protein coronas that form around nanoparticles. Nat Protoc 9, 2030-2044.

200

Doorbar, J., 2006. Molecular biology of human papillomavirus infection and

cervical cancer. Clin Sci (Lond) 110, 525-541.

Doorbar, J., 2013. The E4 protein; structure, function and patterns of expression.

Virology 445, 80-98.

Doorbar, J., Ely, S., Sterling, J., McLean, C., Crawford, L., 1991. Specific

interaction between HPV-16 E1-E4 and cytokeratins results in collapse of

the epithelial cell intermediate filament network. Nature 352, 824-827.

Doorbar, J., Quint, W., Banks, L., Bravo, I.G., Stoler, M., Broker, T.R., Stanley,

M.A., 2012. The biology and life-cycle of human papillomaviruses. Vaccine

30 Suppl 5, F55-70.

Drobek, T., Spencer, N.D., Heuberger, M., 2005. Compressing PEG brushes.

Macromolecules 38, 5254-5259.

Du, H., Chandaroy, P., Hui, S.W., 1997. Grafted poly-(ethylene glycol) on lipid

surfaces inhibits protein adsorption and cell adhesion. Biochim Biophys

Acta 1326, 236-248.

Duncanson, W.J., Figa, M.A., Hallock, K., Zalipsky, S., Hamilton, J.A., Wong,

J.Y., 2007. Targeted binding of PLA microparticles with lipid-PEG-tethered

ligands. Biomaterials 28, 4991-4999.

Dyson, H.J., Wright, P.E., 2005. Intrinsically unstructured proteins and their

functions. Nat Rev Mol Cell Biol 6, 197-208.

201

Ebbesen, M.F., Olesen, M.T., Gjelstrup, M.C., Pakula, M.M., Larsen, E.K.,

Hansen, I.M., Hansen, P.L., Mollenhauer, J., Malle, B.M., Howard, K.A.,

2015. Tunable CD44-specific cellular retargeting with hyaluronic acid

nanoshells. Pharm Res 32, 1462-1474.

Eber, F.J., Eiben, S., Jeske, H., Wege, C., 2015. RNA-controlled assembly of

tobacco mosaic virus-derived complex structures: from nanoboomerangs

to tetrapods. Nanoscale 7, 344-355.

Ehrenreich, B.A., Cohn, Z.A., 1967. The uptake and digestion of iodinated human

serum albumin by macrophages in vitro. J Exp Med 126, 941-958.

Elsadek, B., Kratz, F., 2012. Impact of albumin on drug delivery--new applications

on the horizon. J Control Release 157, 4-28.

Evans, T.W., 2002. Review article: albumin as a drug--biological effects of

albumin unrelated to oncotic pressure. Aliment Pharmacol Ther 16 Suppl

5, 6-11.

Fang, R.H., Hu, C.M., Luk, B.T., Gao, W., Copp, J.A., Tai, Y., O'Connor, D.E.,

Zhang, L., 2014. Cancer cell membrane-coated nanoparticles for

anticancer vaccination and drug delivery. Nano Lett 14, 2181-2188.

Fay, A., Yutzy, W.H.t., Roden, R.B., Moroianu, J., 2004. The positively charged

termini of L2 minor capsid protein required for bovine papillomavirus

infection function separately in nuclear import and DNA binding. J Virol 78,

13447-13454.

202

Felli, I.C., Pierattelli, R., 2012. Recent progress in NMR spectroscopy: toward the

study of intrinsically disordered proteins of increasing size and complexity.

IUBMB Life 64, 473-481.

Ferrer, M.C., Yang, S., Eckmann, D.M., Composto, R.J., 2010. Creating

biomimetic polymeric surfaces by photochemical attachment and

patterning of dextran. Langmuir 26, 14126-14134.

Finbloom, J.A., Han, K., Aanei, I.L., Hartman, E.C., Finley, D.T., Dedeo, M.T.,

Fishman, M., Downing, K.H., Francis, M.B., 2016. Stable Disk Assemblies

of a Tobacco Mosaic Virus Mutant as Nanoscale Scaffolds for Applications

in Drug Delivery. Bioconjug Chem.

Finnen, R.L., Erickson, K.D., Chen, X.S., Garcea, R.L., 2003. Interactions

between papillomavirus L1 and L2 capsid proteins. J Virol 77, 4818-4826.

Flatt, J.W., Kim, R., Smith, J.G., Nemerow, G.R., Stewart, P.L., 2013. An

intrinsically disordered region of the adenovirus capsid is implicated in

neutralization by human alpha defensin 5. PLoS One 8, e61571.

Florin, L., Schafer, F., Sotlar, K., Streeck, R.E., Sapp, M., 2002. Reorganization of

nuclear domain 10 induced by papillomavirus capsid protein l2. Virology

295, 97-107.

Florin, L., Becker, K.A., Lambert, C., Nowak, T., Sapp, C., Strand, D., Streeck,

R.E., Sapp, M., 2006. Identification of a dynein interacting domain in the

papillomavirus minor capsid protein l2. J Virol 80, 6691-6696.

203

Fratila, R.M., Rivera-Fernandez, S., de la Fuente, J.M., 2015. Shape matters:

synthesis and biomedical applications of high aspect ratio magnetic

nanomaterials. Nanoscale 7, 8233-8260.

Fromm, S.A., Bharat, T.A., Jakobi, A.J., Hagen, W.J., Sachse, C., 2015. Seeing

tobacco mosaic virus through direct electron detectors. J Struct Biol 189,

87-97.

Ganson, N.J., Kelly, S.J., Scarlett, E., Sundy, J.S., Hershfield, M.S., 2006.

Control of hyperuricemia in subjects with refractory gout, and induction of

antibody against poly(ethylene glycol) (PEG), in a phase I trial of

subcutaneous PEGylated urate oxidase. Arthritis Res Ther 8, R12.

Ganson, N.J., Povsic, T.J., Sullenger, B.A., Alexander, J.H., Zelenkofske, S.L.,

Sailstad, J.M., Rusconi, C.P., Hershfield, M.S., 2016a. Pre-existing anti-

polyethylene glycol antibody linked to first-exposure allergic reactions to

pegnivacogin, a PEGylated RNA aptamer. J Allergy Clin Immunol 137,

1610-1613 e1617.

Ganson, N.J., Povsic, T.J., Sullenger, B.A., Alexander, J.H., Zelenkofske, S.L.,

Sailstad, J.M., Rusconi, C.P., Hershfield, M.S., 2016b. Pre-existing anti-

polyethylene glycol antibody linked to first-exposure allergic reactions to

pegnivacogin, a PEGylated RNA aptamer. J Allergy Clin Immunol 137,

1610-1613.e1617.

204

Gao, C., Lin, Z., Jurado-Sanchez, B., Lin, X., Wu, Z., He, Q., 2016. Stem Cell

Membrane-Coated Nanogels for Highly Efficient In Vivo Tumor Targeted

Drug Delivery. Small 12, 4056-4062.

Gao, D., Li, W., 2017. Structures and recognition modes of toll-like receptors.

Proteins 85, 3-9.

Gao, W., Hu, C.M., Fang, R.H., Luk, B.T., Su, J., Zhang, L., 2013. Surface

functionalization of gold nanoparticles with red blood cell membranes. Adv

Mater 25, 3549-3553.

Garay, R.P., El-Gewely, R., Armstrong, J.K., Garratty, G., Richette, P., 2012.

Antibodies against polyethylene glycol in healthy subjects and in patients

treated with PEG-conjugated agents. Expert Opin Drug Deliv 9, 1319-

1323.

Garcia-Fuentes, M., Torres, D., Martin-Pastor, M., Alonso, M.J., 2004. Application

of NMR spectroscopy to the characterization of PEG-stabilized lipid

nanoparticles. Langmuir 20, 8839-8845.

Geiger, F.C., Eber, F.J., Eiben, S., Mueller, A., Jeske, H., Spatz, J.P., Wege, C.,

2013. TMV nanorods with programmed longitudinal domains of differently

addressable coat proteins. Nanoscale 5, 3808-3816.

Gerweck, L.E., Seetharaman, K., 1996. Cellular pH gradient in tumor versus

normal tissue: potential exploitation for the treatment of cancer. Cancer

Res 56, 1194-1198.

205

Gilroy, C.A., Luginbuhl, K.M., Chilkoti, A., 2016. Controlled release of biologics for

the treatment of type 2 diabetes. J Control Release 240, 151-164.

Gottula, J., Fuchs, M., 2009. Toward a Quarter Century of Pathogen-Derived

Resistance and Practical Approaches to Plant Virus Disease Control. Adv

Virus Res 75, 161-183.

Gounder, A.P., Wiens, M.E., Wilson, S.S., Lu, W., Smith, J.G., 2012. Critical

determinants of human alpha-defensin 5 activity against non-enveloped

viruses. J Biol Chem 287, 24554-24562.

Gref, R., Quellec, P., Sanchez, A., Calvo, P., Dellacherie, E., Alonso, M.J., 2001.

Development and characterization of CyA-loaded poly(lactic acid)-

poly(ethylene glycol)PEG micro- and nanoparticles. Comparison with

conventional PLA particulate carriers. Eur J Pharm Biopharm 51, 111-118.

Gref, R., Luck, M., Quellec, P., Marchand, M., Dellacherie, E., Harnisch, S.,

Blunk, T., Muller, R.H., 2000. 'Stealth' corona-core nanoparticles surface

modified by polyethylene glycol (PEG): influences of the corona (PEG

chain length and surface density) and of the core composition on

phagocytic uptake and plasma protein adsorption. Colloids Surf B

Biointerfaces 18, 301-313.

Grigorieff, N., 2007. FREALIGN: high-resolution refinement of single particle

structures. J Struct Biol 157, 117-125.

206

Guan, J., Bywaters, S.M., Brendle, S.A., Ashley, R.E., Makhov, A.M., Conway,

J.F., Christensen, N.D., Hafenstein, S., 2017. Cryoelectron Microscopy

Maps of Human Papillomavirus 16 Reveal L2 Densities and Heparin

Binding Site. Structure 25, 253-263.

Gulati, N.M., Pitek, A.S., Steinmetz, N.F., Stewart, P.L., 2017. Cryo-electron

tomography investigation of serum albumin-camouflaged tobacco mosaic

virus nanoparticles. Nanoscale.

Gulati, N.M., Pitek, A.S., Czapar, A.E., Stewart, P.L., Steinmetz, N.F., in

submission. The in vivo fates of plant viral nanoparticles camouflaged

using self-proteins: overcoming immune recognition. Adv Healthc Mater.

Hamad, I., Hunter, A.C., Szebeni, J., Moghimi, S.M., 2008. Poly(ethylene glycol)s

generate complement activation products in human serum through

increased alternative pathway turnover and a MASP-2-dependent

process. Mol Immunol 46, 225-232.

Hao, N., Li, L., Tang, F., 2014. Shape-mediated biological effects of mesoporous

silica nanoparticles. J Biomed Nanotechnol 10, 2508-2538.

Harper, D.M., DeMars, L.R., 2017. HPV vaccines - A review of the first decade.

Gynecol Oncol 146, 196-204.

Hatakeyama, H., Akita, H., Harashima, H., 2013. The polyethyleneglycol

dilemma: advantage and disadvantage of PEGylation of liposomes for

207

systemic genes and nucleic acids delivery to tumors. Biol Pharm Bull 36,

892-899.

Hedayati, M.H., Norouzian, D., Aminian, M., Teimourian, S., Ahangari Cohan, R.,

Sardari, S., Khorramizadeh, M.R., 2017. Molecular Design, Expression

and Evaluation of PASylated Human Recombinant Erythropoietin with

Enhanced Functional Properties. Protein J 36, 36-48.

Hershfield, M.S., Ganson, N.J., Kelly, S.J., Scarlett, E.L., Jaggers, D.A., Sundy,

J.S., 2014. Induced and pre-existing anti-polyethylene glycol antibody in a

trial of every 3-week dosing of pegloticase for refractory gout, including in

organ transplant recipients. Arthritis Res Ther 16, R63.

Hu, C.M., Zhang, L., Aryal, S., Cheung, C., Fang, R.H., Zhang, L., 2011a.

Erythrocyte membrane-camouflaged polymeric nanoparticles as a

biomimetic delivery platform. Proc Natl Acad Sci U S A 108, 10980-10985.

Hu, C.M., Fang, R.H., Wang, K.C., Luk, B.T., Thamphiwatana, S., Dehaini, D.,

Nguyen, P., Angsantikul, P., Wen, C.H., Kroll, A.V., Carpenter, C., Ramesh,

M., Qu, V., Patel, S.H., Zhu, J., Shi, W., Hofman, F.M., Chen, T.C., Gao,

W., Zhang, K., Chien, S., Zhang, L., 2015. Nanoparticle biointerfacing by

platelet membrane cloaking. Nature 526, 118-121.

Hu, Q., Niu, Y., Zhang, K., Liu, Y., Zhou, X., 2011b. Virus-derived transgenes

expressing hairpin RNA give immunity to Tobacco mosaic virus and

Cucumber mosaic virus. Virol J 8, 41.

208

Huiskonen, J.T., Parsy, M.L., Li, S., Bitto, D., Renner, M., Bowden, T.A., 2014.

Averaging of viral envelope glycoprotein spikes from electron

cryotomography reconstructions using Jsubtomo. J Vis Exp, e51714.

Ishida, T., Kinoshita, K., 2007. PrDOS: prediction of disordered protein regions

from amino acid sequence. Nucleic Acids Res 35, W460-464.

Ishida, T., Ichihara, M., Wang, X., Yamamoto, K., Kimura, J., Majima, E., Kiwada,

H., 2006. Injection of PEGylated liposomes in rats elicits PEG-specific

IgM, which is responsible for rapid elimination of a second dose of

PEGylated liposomes. J Control Release 112, 15-25.

Jaulin, N., Appel, M., Passirani, C., Barratt, G., Labarre, D., 2000. Reduction of

the uptake by a macrophagic cell line of nanoparticles bearing heparin or

dextran covalently bound to poly(methyl methacrylate). J Drug Target 8,

165-172.

Johnson, K.M., Kines, R.C., Roberts, J.N., Lowy, D.R., Schiller, J.T., Day, P.M.,

2009. Role of heparan sulfate in attachment to and infection of the murine

female genital tract by human papillomavirus. J Virol 83, 2067-2074.

Jokerst, J.V., Lobovkina, T., Zare, R.N., Gambhir, S.S., 2011. Nanoparticle

PEGylation for imaging and therapy. Nanomedicine (Lond) 6, 715-728.

Kabsch, K., Mossadegh, N., Kohl, A., Komposch, G., Schenkel, J., Alonso, A.,

Tomakidi, P., 2004. The HPV-16 E5 protein inhibits TRAIL- and FasL-

209

mediated apoptosis in human keratinocyte raft cultures. Intervirology 47,

48-56.

Kamper, N., Day, P.M., Nowak, T., Selinka, H.C., Florin, L., Bolscher, J., Hilbig, L.,

Schiller, J.T., Sapp, M., 2006. A membrane-destabilizing peptide in capsid

protein L2 is required for egress of papillomavirus genomes from

endosomes. J Virol 80, 759-768.

Kernan, D.L., Wen, A.M., Pitek, A.S., Steinmetz, N.F., 2017. Delivery of

chemotherapeutic vcMMAE using tobacco mosaic virus nanoparticles.

Exp Biol Med (Maywood), 1535370217719222.

Kim, J., Bronson, C.L., Hayton, W.L., Radmacher, M.D., Roopenian, D.C.,

Robinson, J.M., Anderson, C.L., 2006. Albumin turnover: FcRn-mediated

recycling saves as much albumin from degradation as the liver produces.

Am J Physiol Gastrointest Liver Physiol 290, G352-360.

Kines, R.C., Thompson, C.D., Lowy, D.R., Schiller, J.T., Day, P.M., 2009. The

initial steps leading to papillomavirus infection occur on the basement

membrane prior to cell surface binding. Proc Natl Acad Sci U S A 106,

20458-20463.

Kjaer, S.K., Sigurdsson, K., Iversen, O.E., Hernandez-Avila, M., Wheeler, C.M.,

Perez, G., Brown, D.R., Koutsky, L.A., Tay, E.H., Garcia, P., Ault, K.A.,

Garland, S.M., Leodolter, S., Olsson, S.E., Tang, G.W., Ferris, D.G.,

Paavonen, J., Lehtinen, M., Steben, M., Bosch, F.X., Dillner, J., Joura,

210

E.A., Majewski, S., Munoz, N., Myers, E.R., Villa, L.L., Taddeo, F.J.,

Roberts, C., Tadesse, A., Bryan, J., Maansson, R., Lu, S., Vuocolo, S.,

Hesley, T.M., Saah, A., Barr, E., Haupt, R.M., 2009. A pooled analysis of

continued prophylactic efficacy of quadrivalent human papillomavirus

(Types 6/11/16/18) vaccine against high-grade cervical and external

genital lesions. Cancer Prev Res (Phila) 2, 868-878.

Knop, K., Hoogenboom, R., Fischer, D., Schubert, U.S., 2010. Poly(ethylene

glycol) in Drug Delivery: Pros and Cons as Well as Potential Alternatives.

Angew Chem Int Edit 49, 6288-6308.

Kokkinopoulou, M., Simon, J., Landfester, K., Mailander, V., Lieberwirth, I., 2017.

Visualization of the protein corona: towards a biomolecular understanding

of nanoparticle-cell-interactions. Nanoscale 9, 8858-8870.

Kozak, D., Chen, A., Bax, J., Trau, M., 2011. Protein resistance of dextran and

dextran-poly(ethylene glycol) copolymer films. Biofouling 27, 497-503.

Krantz, A., Hanel, A.M., Strug, I., Wilczynski, A., Wolff, J.J., Huang, W., Huang,

L.H., Settineri, T., Holmes, D.L., Hardy, M.C., Bridon, D.P., 2014. Site-

specific Labeling of a Protein Lysine Residue By Novel Kinetic Labeling

Combinatorial Libraries. Comput Struct Biotechnol J 9, e201403001.

Kremer, J.R., Mastronarde, D.N., McIntosh, J.R., 1996. Computer visualization of

three-dimensional image data using IMOD. J Struct Biol 116, 71-76.

211

Kuhlbrandt, W., 2014. Biochemistry. The resolution revolution. Science 343,

1443-1444.

Kuhn, N., Schmidt, C.Q., Schlapschy, M., Skerra, A., 2016. PASylated Coversin,

a C5-Specific Complement Inhibitor with Extended Pharmacokinetics,

Shows Enhanced Anti-Hemolytic Activity in Vitro. Bioconjug Chem 27,

2359-2371.

Labarre, D., Vauthier, C., Chauvierre, C., Petri, B., Muller, R., Chehimi, M.M.,

2005. Interactions of blood proteins with poly(isobutylcyanoacrylate)

nanoparticles decorated with a polysaccharidic brush. Biomaterials 26,

5075-5084.

Lam, P., Gulati, N.M., Stewart, P.L., Keri, R.A., Steinmetz, N.F., 2016.

Bioengineering of Tobacco Mosaic Virus to Create a Non-Infectious

Positive Control for Ebola Diagnostic Assays. Sci Rep 6, 23803.

Laniosz, V., Nguyen, K.C., Meneses, P.I., 2007. Bovine papillomavirus type 1

infection is mediated by SNARE syntaxin 18. J Virol 81, 7435-7448.

Lee, K.L., Carpenter, B.L., Wen, A.M., Ghiladi, R.A., Steinmetz, N.F., 2016. High

Aspect Ratio Nanotubes Formed by Tobacco Mosaic Virus for Delivery of

Photodynamic Agents Targeting Melanoma. ACS Biomater Sci Eng 2, 838-

844.

212

Lee, K.L., Hubbard, L.C., Hern, S., Yildiz, I., Gratzl, M., Steinmetz, N.F., 2013.

Shape matters: the diffusion rates of TMV rods and CPMV icosahedrons

in a spheroid model of extracellular matrix are distinct. Biomater Sci 1.

Lee, K.L., Shukla, S., Wu, M., Ayat, N.R., El Sanadi, C.E., Wen, A.M., Edelbrock,

J.F., Pokorski, J.K., Commandeur, U., Dubyak, G.R., Steinmetz, N.F.,

2015a. Stealth filaments: Polymer chain length and conformation affect

the in vivo fate of PEGylated potato virus X. Acta Biomater 19, 166-179.

Lee, P.W., Isarov, S.A., Wallat, J.D., Molugu, S.K., Shukla, S., Sun, J.E., Zhang,

J., Zheng, Y., Lucius Dougherty, M., Konkolewicz, D., Stewart, P.L.,

Steinmetz, N.F., Hore, M.J., Pokorski, J.K., 2017. Polymer Structure and

Conformation Alter the Antigenicity of Virus-like Particle-Polymer

Conjugates. J Am Chem Soc 139, 3312-3315.

Lee, Y.K., Choi, E.J., Webster, T.J., Kim, S.H., Khang, D., 2015b. Effect of the

protein corona on nanoparticles for modulating cytotoxicity and

immunotoxicity. Int J Nanomedicine 10, 97-113.

Lehrer, R.I., Lu, W., 2012. alpha-Defensins in human innate immunity. Immunol

Rev 245, 84-112.

Levary, D.A., Parthasarathy, R., Boder, E.T., Ackerman, M.E., 2011. Protein-

protein fusion catalyzed by sortase A. PLoS One 6, e18342.

Li, S.D., Huang, L., 2009. Nanoparticles evading the reticuloendothelial system:

role of the supported bilayer. Biochim Biophys Acta 1788, 2259-2266.

213

Li, Z.B., Han, X.P., Zhai, Y.L., Lian, H., Zhang, D., Zhang, W.J., Wang, Y.J., He,

Z.G., Liu, Z., Sun, J., 2015. Critical determinant of intestinal permeability

and oral bioavailability of pegylated all trans-retinoic acid prodrug-based

nanomicelles: Chain length of poly (ethylene glycol) corona. Colloid

Surface B 130, 133-140.

Liu, R., Vaishnav, R.A., Roberts, A.M., Friedland, R.P., 2013. Humans have

antibodies against a plant virus: evidence from tobacco mosaic virus.

PLoS One 8, e60621.

Liu, W., Dreher, M.R., Chow, D.C., Zalutsky, M.R., Chilkoti, A., 2006a. Tracking

the in vivo fate of recombinant polypeptides by isotopic labeling. J Control

Release 114, 184-192.

Liu, W., Dreher, M.R., Furgeson, D.Y., Peixoto, K.V., Yuan, H., Zalutsky, M.R.,

Chilkoti, A., 2006b. Tumor accumulation, degradation and

pharmacokinetics of elastin-like polypeptides in nude mice. J Control

Release 116, 170-178.

Lucas, A.T., Madden, A.J., Zamboni, W.C., 2016. Challenges in preclinical to

clinical translation for anticancer carrier-mediated agents. Wiley Interdiscip

Rev Nanomed Nanobiotechnol 8, 642-653.

Lundin, L., Larsson, H., Kreuger, J., Kanda, S., Lindahl, U., Salmivirta, M.,

Claesson-Welsh, L., 2000. Selectively desulfated heparin inhibits fibroblast

214

growth factor-induced mitogenicity and angiogenesis. J Biol Chem 275,

24653-24660.

Mamoor, S., Onder, Z., Karanam, B., Kwak, K., Bordeaux, J., Crosby, L., Roden,

R.B., Moroianu, J., 2012. The high risk HPV16 L2 minor capsid protein

has multiple transport signals that mediate its nucleocytoplasmic traffic.

Virology 422, 413-424.

Mariam, J., Sivakami, S., Dongre, P.M., 2016. Albumin corona on nanoparticles -

a strategic approach in drug delivery. Drug Deliv 23, 2668-2676.

McIntosh, P.B., Martin, S.R., Jackson, D.J., Khan, J., Isaacson, E.R., Calder, L.,

Raj, K., Griffin, H.M., Wang, Q., Laskey, P., Eccleston, J.F., Doorbar, J.,

2008. Structural analysis reveals an amyloid form of the human

papillomavirus type 16 E1--E4 protein and provides a molecular basis for

its accumulation. J Virol 82, 8196-8203.

Meng, X., Maggs, J.L., Usui, T., Whitaker, P., French, N.S., Naisbitt, D.J., Park,

B.K., 2015. Auto-oxidation of Isoniazid Leads to Isonicotinic-Lysine

Adducts on Human Serum Albumin. Chem Res Toxicol 28, 51-58.

Merlot, A.M., Kalinowski, D.S., Richardson, D.R., 2014. Unraveling the mysteries

of serum albumin-more than just a serum protein. Front Physiol 5, 299.

Meyer, D.E., Chilkoti, A., 2004. Quantification of the effects of chain length and

concentration on the thermal behavior of elastin-like polypeptides.

Biomacromolecules 5, 846-851.

215

Milici, A.J., Watrous, N.E., Stukenbrok, H., Palade, G.E., 1987. Transcytosis of

albumin in capillary endothelium. J Cell Biol 105, 2603-2612.

Muller, J., Bauer, K.N., Prozeller, D., Simon, J., Mailander, V., Wurm, F.R.,

Winzen, S., Landfester, K., 2017. Coating nanoparticles with tunable

surfactants facilitates control over the protein corona. Biomaterials 115, 1-

Naud, P.S., Roteli-Martins, C.M., De Carvalho, N.S., Teixeira, J.C., de Borba,

P.C., Sanchez, N., Zahaf, T., Catteau, G., Geeraerts, B., Descamps, D.,

2014. Sustained efficacy, immunogenicity, and safety of the HPV-16/18

AS04-adjuvanted vaccine: final analysis of a long-term follow-up study up

to 9.4 years post-vaccination. Hum Vaccin Immunother 10, 2147-2162.

Negi, L.M., Jaggi, M., Joshi, V., Ronodip, K., Talegaonkar, S., 2015. Hyaluronic

acid decorated lipid nanocarrier for MDR modulation and CD-44 targeting

in colon adenocarcinoma. Int J Biol Macromol 72, 569-574.

Nguyen, E.K., Nemerow, G.R., Smith, J.G., 2010. Direct evidence from single-

cell analysis that human {alpha}-defensins block adenovirus uncoating to

neutralize infection. J Virol 84, 4041-4049.

Niehl, A., Appaix, F., Bosca, S., van der Sanden, B., Nicoud, J.F., Bolze, F.,

Heinlein, M., 2015. Fluorescent Tobacco mosaic virus-Derived Bio-

Nanoparticles for Intravital Two-Photon Imaging. Front Plant Sci 6, 1244.

216

Northfelt, D.W., Martin, F.J., Working, P., Volberding, P.A., Russell, J., Newman,

M., Amantea, M.A., Kaplan, L.D., 1996. Doxorubicin encapsulated in

liposomes containing surface-bound polyethylene glycol:

pharmacokinetics, tumor localization, and safety in patients with AIDS-

related Kaposi's sarcoma. J Clin Pharmacol 36, 55-63.

Nowinski, A.K., White, A.D., Keefe, A.J., Jiang, S., 2014. Biologically inspired

stealth peptide-capped gold nanoparticles. Langmuir 30, 1864-1870.

Oldenborg, P.A., Zheleznyak, A., Fang, Y.F., Lagenaur, C.F., Gresham, H.D.,

Lindberg, F.P., 2000. Role of CD47 as a marker of self on red blood cells.

Science 288, 2051-2054.

Ono, K., Ishihara, M., Ishikawa, K., Ozeki, Y., Deguchi, H., Sato, M., Hashimoto,

H., Saito, Y., Yura, H., Kurita, A., Maehara, T., 2002. Periodate-treated,

non-anticoagulant heparin-carrying polystyrene (NAC-HCPS) affects

angiogenesis and inhibits subcutaneous induced tumour growth and

metastasis to the lung. Br J Cancer 86, 1803-1812.

Orlov, I., Myasnikov, A.G., Andronov, L., Natchiar, S.K., Khatter, H., Beinsteiner,

B., Menetret, J.F., Hazemann, I., Mohideen, K., Tazibt, K., Tabaroni, R.,

Kratzat, H., Djabeur, N., Bruxelles, T., Raivoniaina, F., Pompeo, L.D.,

Torchy, M., Billas, I., Urzhumtsev, A., Klaholz, B.P., 2017. The integrative

role of cryo electron microscopy in molecular and cellular structural

biology. Biol Cell 109, 81-93.

217

Owens, D.E., 3rd, Peppas, N.A., 2006. Opsonization, biodistribution, and

pharmacokinetics of polymeric nanoparticles. Int J Pharm 307, 93-102.

Pallardy, M.J., Turbica, I., Biola-Vidamment, A., 2017. Why the Immune System

Should Be Concerned by Nanomaterials? Front Immunol 8, 544.

Palomba, R., Parodi, A., Evangelopoulos, M., Acciardo, S., Corbo, C., de Rosa,

E., Yazdi, I.K., Scaria, S., Molinaro, R., Furman, N.E., You, J., Ferrari, M.,

Salvatore, F., Tasciotti, E., 2016. Biomimetic carriers mimicking leukocyte

plasma membrane to increase tumor vasculature permeability. Sci Rep 6,

34422.

Park, K., Lee, G.Y., Kim, Y.S., Yu, M., Park, R.W., Kim, I.S., Kim, S.Y., Byun, Y.,

2006. Heparin-deoxycholic acid chemical conjugate as an anticancer drug

carrier and its antitumor activity. J Control Release 114, 300-306.

Parodi, A., Quattrocchi, N., van de Ven, A.L., Chiappini, C., Evangelopoulos, M.,

Martinez, J.O., Brown, B.S., Khaled, S.Z., Yazdi, I.K., Enzo, M.V., Isenhart,

L., Ferrari, M., Tasciotti, E., 2013. Synthetic nanoparticles functionalized

with biomimetic leukocyte membranes possess cell-like functions. Nat

Nanotechnol 8, 61-68.

Peer, D., Park, E.J., Morishita, Y., Carman, C.V., Shimaoka, M., 2008. Systemic

leukocyte-directed siRNA delivery revealing cyclin D1 as an anti-

inflammatory target. Science 319, 627-630.

218

Perrino, C., Lee, S., Choi, S.W., Maruyama, A., Spencer, N.D., 2008. A

biomimetic alternative to poly(ethylene glycol) as an antifouling coating:

resistance to nonspecific protein adsorption of poly(L-lysine)-graft-dextran.

Langmuir 24, 8850-8856.

Perry, J.L., Reuter, K.G., Kai, M.P., Herlihy, K.P., Jones, S.W., Luft, J.C., Napier,

M., Bear, J.E., DeSimone, J.M., 2012. PEGylated PRINT nanoparticles:

the impact of PEG density on protein binding, macrophage association,

biodistribution, and pharmacokinetics. Nano Lett 12, 5304-5310.

Petros, R.A., DeSimone, J.M., 2010. Strategies in the design of nanoparticles for

therapeutic applications. Nat Rev Drug Discov 9, 615-627.

Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M.,

Meng, E.C., Ferrin, T.E., 2004. UCSF Chimera--a visualization system for

exploratory research and analysis. J Comput Chem 25, 1605-1612.

Phillips, J.C., Braun, R., Wang, W., Gumbart, J., Tajkhorshid, E., Villa, E., Chipot,

C., Skeel, R.D., Kale, L., Schulten, K., 2005. Scalable molecular dynamics

with NAMD. J Comput Chem 26, 1781-1802.

Piao, J.G., Wang, L., Gao, F., You, Y.Z., Xiong, Y., Yang, L., 2014. Erythrocyte

membrane is an alternative coating to polyethylene glycol for prolonging

the circulation lifetime of gold nanocages for photothermal therapy. ACS

Nano 8, 10414-10425.

219

Pinnow, D.L., Demski, J.W., Sterling, J., 1981. Tracking Tmv - the Virus That

Preys on Pimento Pepper Plants. Georgia Agr Res 22, 6-6.

Pitek, A.S., Wen, A.M., Shukla, S., Steinmetz, N.F., 2016a. The Protein Corona of

Plant Virus Nanoparticles Influences their Dispersion Properties, Cellular

Interactions, and In Vivo Fates. Small 12, 1758-1769.

Pitek, A.S., Jameson, S.A., Veliz, F.A., Shukla, S., Steinmetz, N.F., 2016b. Serum

albumin 'camouflage' of plant virus based nanoparticles prevents their

antibody recognition and enhances pharmacokinetics. Biomaterials 89,

89-97.

Pitek, A.S., Wang, Y., Gulati, S., Stewart, P.L., Simon, D., Steinmetz, N.F., in

review. Elongated plant virus-based nanoparticles for enhanced delivery of

thrombolytic therapies. Mol Pharm.

Porte, H., Chastre, E., Prevot, S., Nordlinger, B., Empereur, S., Basset, P.,

Chambon, P., Gespach, C., 1995. Neoplastic progression of human

colorectal cancer is associated with overexpression of the stromelysin-3

and BM-40/SPARC genes. Int J Cancer 64, 70-75.

Prenzel, K.L., Warnecke-Eberz, U., Xi, H., Brabender, J., Baldus, S.E.,

Bollschweiler, E., Gutschow, C.A., Holscher, A.H., Schneider, P.M., 2006.

Significant overexpression of SPARC/osteonectin mRNA in pancreatic

cancer compared to cancer of the papilla of Vater. Oncol Rep 15, 1397-

1401.

220

Qie, Y., Yuan, H., von Roemeling, C.A., Chen, Y., Liu, X., Shih, K.D., Knight, J.A.,

Tun, H.W., Wharen, R.E., Jiang, W., Kim, B.Y., 2016. Surface modification

of nanoparticles enables selective evasion of phagocytic clearance by

distinct macrophage phenotypes. Sci Rep 6, 26269.

Ranalli, A., Santi, M., Capriotti, L., Voliani, V., Porciani, D., Beltram, F., Signore,

G., 2017. Peptide-Based Stealth Nanoparticles for Targeted and pH-

Triggered Delivery. Bioconjug Chem 28, 627-635.

Rao, L., Xu, J.H., Cai, B., Liu, H., Li, M., Jia, Y., Xiao, L., Guo, S.S., Liu, W.,

Zhao, X.Z., 2016. Synthetic nanoparticles camouflaged with biomimetic

erythrocyte membranes for reduced reticuloendothelial system uptake.

Nanotechnology 27, 085106.

Ren, X., Zheng, R., Fang, X., Wang, X., Zhang, X., Yang, W., Sha, X., 2016. Red

blood cell membrane camouflaged magnetic nanoclusters for imaging-

guided photothermal therapy. Biomaterials 92, 13-24.

Richards, R.M., Lowy, D.R., Schiller, J.T., Day, P.M., 2006. Cleavage of the

papillomavirus minor capsid protein, L2, at a furin consensus site is

necessary for infection. Proc Natl Acad Sci U S A 103, 1522-1527.

Richter, A.W., Akerblom, E., 1984. Polyethylene glycol reactive antibodies in

man: titer distribution in allergic patients treated with monomethoxy

polyethylene glycol modified allergens or placebo, and in healthy blood

donors. Int Arch Allergy Appl Immunol 74, 36-39.

221

Rippe, R.A., Meinke, W.J., 1989. Identification and characterization of the BPV-2

L2 protein. Virology 171, 298-301.

Roberts, J.N., Buck, C.B., Thompson, C.D., Kines, R., Bernardo, M., Choyke,

P.L., Lowy, D.R., Schiller, J.T., 2007. Genital transmission of HPV in a

mouse model is potentiated by nonoxynol-9 and inhibited by carrageenan.

Nat Med 13, 857-861.

Roden, R.B.S., Kirnbauer, R., Jenson, A.B., Lowy, D.R., Schiller, J.T., 1994.

Interaction of Papillomaviruses with the Cell-Surface. Journal of Virology

68, 7260-7266.

Rodriguez, P.L., Harada, T., Christian, D.A., Pantano, D.A., Tsai, R.K., Discher,

D.E., 2013. Minimal "Self" peptides that inhibit phagocytic clearance and

enhance delivery of nanoparticles. Science 339, 971-975.

Rosenthal, P.B., Henderson, R., 2003. Optimal determination of particle

orientation, absolute hand, and contrast loss in single-particle electron

cryomicroscopy. J Mol Biol 333, 721-745.

Rubio, I., Seitz, H., Canali, E., Sehr, P., Bolchi, A., Tommasino, M., Ottonello, S.,

Muller, M., 2011. The N-terminal region of the human papillomavirus L2

protein contains overlapping binding sites for neutralizing, cross-

neutralizing and non-neutralizing antibodies. Virology 409, 348-359.

222

Sakata, Y., Chen, Y.Y., Chen, H.R., Zhao, J.H., Xu, Q.X., 1997. Identification of

strains of tobacco mosaic virus (TMV) isolated from Capsicums in

southcentral China. J Jpn Soc Hortic Sci 66, 99-104.

Salvi, N., Abyzov, A., Blackledge, M., 2016. Multi-Timescale Dynamics in

Intrinsically Disordered Proteins from NMR Relaxation and Molecular

Simulation. J Phys Chem Lett 7, 2483-2489.

Sarav, M., Wang, Y., Hack, B.K., Chang, A., Jensen, M., Bao, L., Quigg, R.J.,

2009. Renal FcRn reclaims albumin but facilitates elimination of IgG. J Am

Soc Nephrol 20, 1941-1952.

Schelhaas, M., Shah, B., Holzer, M., Blattmann, P., Kuhling, L., Day, P.M.,

Schiller, J.T., Helenius, A., 2012. Entry of human papillomavirus type 16 by

actin-dependent, clathrin- and lipid raft-independent endocytosis. PLoS

Pathog 8, e1002657.

Schiller, J.T., Lowy, D.R., 2012. Understanding and learning from the success of

prophylactic human papillomavirus vaccines. Nat Rev Microbiol 10, 681-

692.

Schiller, J.T., Day, P.M., Kines, R.C., 2010. Current understanding of the

mechanism of HPV infection. Gynecol Oncol 118, S12-17.

Schlapschy, M., Binder, U., Borger, C., Theobald, I., Wachinger, K., Kisling, S.,

Haller, D., Skerra, A., 2013. PASylation: a biological alternative to

223

PEGylation for extending the plasma half-life of pharmaceutically active

proteins. Protein Eng Des Sel 26, 489-501.

Schmidtke, J.R., Unanue, E.R., 1971. Macrophage-antigen interaction: uptake,

metabolism and immunogenicity of foreign albumin. J Immunol 107, 331-

338.

Schneider, C.A., Rasband, W.S., Eliceiri, K.W., 2012. NIH Image to ImageJ: 25

years of image analysis. Nat Methods 9, 671-675.

Schnitzer, J.E., 1992. gp60 is an albumin-binding glycoprotein expressed by

continuous endothelium involved in albumin transcytosis. Am J Physiol

262, H246-254.

Schnitzer, J.E., Oh, P., 1992. Antibodies to SPARC inhibit albumin binding to

SPARC, gp60, and microvascular endothelium. Am J Physiol 263, H1872-

1879.

Schnitzer, J.E., Bravo, J., 1993. High affinity binding, endocytosis, and

degradation of conformationally modified albumins. Potential role of gp30

and gp18 as novel scavenger receptors. J Biol Chem 268, 7562-7570.

Schnitzer, J.E., Oh, P., 1994. Albondin-mediated capillary permeability to

albumin. Differential role of receptors in endothelial transcytosis and

endocytosis of native and modified albumins. J Biol Chem 269, 6072-

6082.

224

Scholthof, K.-B.G., Shaw, J.G., Zaitlin, M., 2000. Tobacco Mosaic Virus: One

Hundred Years of Contributions to Virology. Journal of the History of

Biology 33, 604-606.

Seehuber, A., Schmidt, D., Dahint, R., 2012. Poly(acrylic acid)-poly(ethylene

glycol) layers on positively charged surface coatings: molecular structure,

protein resistance, and application to single protein deposition. Langmuir

28, 8700-8710.

Selinka, H.C., Florin, L., Patel, H.D., Freitag, K., Schmidtke, M., Makarov, V.A.,

Sapp, M., 2007. Inhibition of transfer to secondary receptors by heparan

sulfate-binding drug or antibody induces noninfectious uptake of human

papillomavirus. J Virol 81, 10970-10980.

Shi, D., Jiang, K., Fu, Y., Fang, R., Liu, X.I., Chen, J., 2016. Overexpression of

SPARC correlates with poor prognosis in patients with cervical carcinoma

and regulates cancer cell epithelial-mesenchymal transition. Oncol Lett 11,

3251-3258.

Shuck, S.C., Wauchope, O.R., Rose, K.L., Kingsley, P.J., Rouzer, C.A., Shell,

S.M., Sugitani, N., Chazin, W.J., Zagol-Ikapitte, I., Boutaud, O., Oates,

J.A., Galligan, J.J., Beavers, W.N., Marnett, L.J., 2014. Protein

modification by adenine propenal. Chem Res Toxicol 27, 1732-1742.

Shukla, S., Eber, F.J., Nagarajan, A.S., DiFranco, N.A., Schmidt, N., Wen, A.M.,

Eiben, S., Twyman, R.M., Wege, C., Steinmetz, N.F., 2015. The Impact of

225

Aspect Ratio on the Biodistribution and Tumor Homing of Rigid Soft-Matter

Nanorods. Adv Healthc Mater 4, 874-882.

Smith, J.G., Silvestry, M., Lindert, S., Lu, W., Nemerow, G.R., Stewart, P.L., 2010.

Insight into the mechanisms of adenovirus capsid disassembly from

studies of defensin neutralization. PLoS Pathog 6, e1000959.

Smith, W.W., Schreck, C.F., Hashem, N., Soltani, S., Nath, A., Rhoades, E.,

O'Hern, C.S., 2012. Molecular simulations of the fluctuating

conformational dynamics of intrinsically disordered proteins. Phys Rev E

Stat Nonlin Soft Matter Phys 86, 041910.

Socha, M., Lamprecht, A., El Ghazouani, F., Emond, E., Maincent, P., Barre, J.,

Hoffman, M., Ubrich, N., 2008. Increase in the vascular residence time of

propranolol-loaded nanoparticles coated with heparin. J Nanosci

Nanotechnol 8, 2369-2376.

Socha, M., Bartecki, P., Passirani, C., Sapin, A., Damge, C., Lecompte, T., Barre,

J., El Ghazouani, F., Maincent, P., 2009. Stealth nanoparticles coated with

heparin as peptide or protein carriers. J Drug Target 17, 575-585.

Sosale, N.G., Spinler, K.R., Alvey, C., Discher, D.E., 2015. Macrophage

engulfment of a cell or nanoparticle is regulated by unavoidable

opsonization, a species-specific 'Marker of Self' CD47, and target physical

properties. Curr Opin Immunol 35, 107-112.

Stanfield, R.L., Wilson, I.A., 2014. Antibody Structure. Microbiol Spectr 2.

226

Stanley, M., 2008. Immunobiology of HPV and HPV vaccines. Gynecol Oncol

109, S15-21.

Steinmetz, N.F., Hong, V., Spoerke, E.D., Lu, P., Breitenkamp, K., Finn, M.G.,

Manchester, M., 2009. Buckyballs meet viral nanoparticles: candidates for

biomedicine. J Am Chem Soc 131, 17093-17095.

Stewart, P.L., 2016. Cryo-electron microscopy and cryo-electron tomography of

nanoparticles. Wiley Interdiscip Rev Nanomed Nanobiotechnol, DOI:

10.1002/wnan.1417.

Sugio, S., Kashima, A., Mochizuki, S., Noda, M., Kobayashi, K., 1999. Crystal

structure of human serum albumin at 2.5 A resolution. Protein Eng 12,

439-446.

Tang, G., Peng, L., Baldwin, P.R., Mann, D.S., Jiang, W., Rees, I., Ludtke, S.J.,

2007. EMAN2: an extensible image processing suite for electron

microscopy. J Struct Biol 157, 38-46.

Tenge, V.R., Gounder, A.P., Wiens, M.E., Lu, W., Smith, J.G., 2014. Delineation

of interfaces on human alpha-defensins critical for human adenovirus and

human papillomavirus inhibition. PLoS Pathog 10, e1004360.

Toole, B.P., 2004. Hyaluronan: from extracellular glue to pericellular cue. Nat Rev

Cancer 4, 528-539.

227

Trus, B.L., Roden, R.B., Greenstone, H.L., Vrhel, M., Schiller, J.T., Booy, F.P.,

1997. Novel structural features of bovine papillomavirus capsid revealed

by a three-dimensional reconstruction to 9 A resolution. Nat Struct Biol 4,

413-420.

Tse, L.V., Klinc, K.A., Madigan, V.J., Rivera, R.M.C., Wells, L.F., Havlik, L.P.,

Smith, J.K., Agbandje-McKenna, M., Asokan, A., 2017. Structure-Guided

Iterative Evolution of Antigenically Advanced AAV Variants for Therapeutic

Gene Transfer. Mol Ther 25, 232-232.

Tseng, Y.S., Agbandje-McKenna, M., 2014. Mapping the AAV Capsid Host

Antibody Response toward the Development of Second Generation Gene

Delivery Vectors. Front Immunol 5, 9.

Upadhyay, K.K., Bhatt, A.N., Castro, E., Mishra, A.K., Chuttani, K., Dwarakanath,

B.S., Schatz, C., Le Meins, J.F., Misra, A., Lecommandoux, S., 2010. In

vitro and in vivo evaluation of docetaxel loaded biodegradable

polymersomes. Macromol Biosci 10, 503-512.

Urry, D.W., Pattanaik, A., 1997. Elastic protein-based materials in tissue

reconstruction. Ann N Y Acad Sci 831, 32-46. van Heel, M., Schatz, M., 2005. Fourier shell correlation threshold criteria. J

Struct Biol 151, 250-262.

Velan, B., 2011. Acceptance on the move: public reaction to shifting vaccination

realities. Hum Vaccin 7, 1261-1270.

228

Veronese, F.M., Pasut, G., 2005. PEGylation, successful approach to drug

delivery. Drug Discov Today 10, 1451-1458.

Walboomers, J.M., Jacobs, M.V., Manos, M.M., Bosch, F.X., Kummer, J.A., Shah,

K.V., Snijders, P.J., Peto, J., Meijer, C.J., Munoz, N., 1999. Human

papillomavirus is a necessary cause of invasive cervical cancer worldwide.

J Pathol 189, 12-19.

Walczyk, D., Bombelli, F.B., Monopoli, M.P., Lynch, I., Dawson, K.A., 2010. What

the cell "sees" in bionanoscience. J Am Chem Soc 132, 5761-5768.

Wan, W., Briggs, J.A., 2016. Cryo-Electron Tomography and Subtomogram

Averaging. Methods Enzymol 579, 329-367.

Wang, C.F., Makila, E.M., Bonduelle, C., Rytkonen, J., Raula, J., Almeida, S.,

Narvanen, A., Salonen, J.J., Lecommandoux, S., Hirvonen, J.T., Santos,

H.A., 2015. Functionalization of alkyne-terminated thermally

hydrocarbonized porous silicon nanoparticles with targeting peptides and

antifouling polymers: effect on the human plasma protein adsorption. ACS

Appl Mater Interfaces 7, 2006-2015.

Wang, J.W., Roden, R.B., 2013. L2, the minor capsid protein of papillomavirus.

Virology 445, 175-186.

Watts, C., 2012. The endosome-lysosome pathway and information generation in

the immune system. Biochim Biophys Acta 1824, 14-21.

229

Wen, A.M., Lee, K.L., Yildiz, I., Bruckman, M.A., Shukla, S., Steinmetz, N.F.,

2012. Viral nanoparticles for in vivo tumor imaging. J Vis Exp, e4352.

Wen, A.M., Infusino, M., De Luca, A., Kernan, D.L., Czapar, A.E., Strangi, G.,

Steinmetz, N.F., 2015. Interface of physics and biology: engineering virus-

based nanoparticles for biophotonics. Bioconjug Chem 26, 51-62.

Wetter, C., 1975. Tobacco mosaic virus and para-tobacco mosaic virus in

cigarettes. Naturwissenschaften 62, 533.

White, A.D., Nowinski, A.K., Huang, W.J., Keefe, A.J., Sun, F., Jiang, S.Y., 2012.

Decoding nonspecific interactions from nature. Chem Sci 3, 3488-3494.

Wiens, M.E., Smith, J.G., 2015. Alpha-defensin HD5 inhibits furin cleavage of

human papillomavirus 16 L2 to block infection. J Virol 89, 2866-2874.

Wiens, M.E., Smith, J.G., 2017. alpha-Defensin HD5 Inhibits Human

Papillomavirus 16 Infection via Capsid Stabilization and Redirection to the

Lysosome. MBio 8.

Wilson, R., Ryan, G.B., Knight, G.L., Laimins, L.A., Roberts, S., 2007. The full-

length E1E4 protein of human papillomavirus type 18 modulates

differentiation-dependent viral DNA amplification and late gene

expression. Virology 362, 453-460.

Woodham, A.W., Da Silva, D.M., Skeate, J.G., Raff, A.B., Ambroso, M.R., Brand,

H.E., Isas, J.M., Langen, R., Kast, W.M., 2012. The S100A10 subunit of

230

the annexin A2 heterotetramer facilitates L2-mediated human

papillomavirus infection. PLoS One 7, e43519.

World Health Organization, 2017. Human papillomavirus vaccines: WHO position

paper, May 2017-Recommendations. Vaccine.

Wuang, S.C., Neoh, K.G., Kang, E.T., Pack, D.W., Leckband, D.E., 2006.

Heparinized magnetic nanoparticles: In-vitro assessment for biomedical

applications. Adv Funct Mater 16, 1723-1730.

Xu, F., Reiser, M., Yu, X., Gummuluru, S., Wetzler, L., Reinhard, B.M., 2016.

Lipid-Mediated Targeting with Membrane-Wrapped Nanoparticles in the

Presence of Corona Formation. ACS Nano 10, 1189-1200.

Xuan, M., Shao, J., Dai, L., He, Q., Li, J., 2015. Macrophage Cell Membrane

Camouflaged Mesoporous Silica Nanocapsules for In Vivo Cancer

Therapy. Adv Healthc Mater 4, 1645-1652.

Yang, C., Gao, S., Dagnaes-Hansen, F., Jakobsen, M., Kjems, J., 2017. Impact

of PEG Chain Length on the Physical Properties and Bioactivity of

PEGylated Chitosan/siRNA Nanoparticles in Vitro and in Vivo. ACS Appl

Mater Interfaces 9, 12203-12216.

Yang, R., Yutzy, W.H.t., Viscidi, R.P., Roden, R.B., 2003a. Interaction of L2 with

beta-actin directs intracellular transport of papillomavirus and infection. J

Biol Chem 278, 12546-12553.

231

Yang, R., Day, P.M., Yutzy, W.H.t., Lin, K.Y., Hung, C.F., Roden, R.B., 2003b. Cell

surface-binding motifs of L2 that facilitate papillomavirus infection. J Virol

77, 3531-3541.

Zanetti-Ramos, B.G., Fritzen-Garcia, M.B., Cristian, S.D.O.B., Andr, A.P.B., Soldi,

V., Borsali, R., Creczynski-Pasa, T.B., 2009. Dynamic light scattering and

atomic force microscopy techniques for size determination of polyurethane

nanoparticles. Mat Sci Eng C-Bio S 29, 638-640.

Zhang, K., 2016. Gctf: Real-time CTF determination and correction. J Struct Biol

193, 1-12.

Zhang, L., Li, R., Chen, H., Wei, J., Qian, H., Su, S., Shao, J., Wang, L., Qian, X.,

Liu, B., 2017. Human cytotoxic T-lymphocyte membrane-camouflaged

nanoparticles combined with low-dose irradiation: a new approach to

enhance drug targeting in gastric cancer. Int J Nanomedicine 12, 2129-

2142.

Zhang, P., Chen, Y., Zeng, Y., Shen, C., Li, R., Guo, Z., Li, S., Zheng, Q., Chu, C.,

Wang, Z., Zheng, Z., Tian, R., Ge, S., Zhang, X., Xia, N.S., Liu, G., Chen,

X., 2015. Virus-mimetic nanovesicles as a versatile antigen-delivery

system. Proc Natl Acad Sci U S A 112, E6129-6138.

Zhang, Q., Deng, C., Fu, Y., Sun, X., Gong, T., Zhang, Z., 2016. Repeated

Administration of Hyaluronic Acid Coated Liposomes with Improved

232

Pharmacokinetics and Reduced Immune Response. Mol Pharm 13, 1800-

1808.

Zhu, A., Yuan, P., Du, F., Hong, R., Ding, X., Shi, X., Fan, Y., Wang, J., Luo, Y.,

Ma, F., Zhang, P., Li, Q., Xu, B., 2016. SPARC overexpression in primary

tumors correlates with disease recurrence and overall survival in patients

with triple negative breast cancer. Oncotarget 7, 76628-76634.

Zins, S.R., Nelson, C.D., Maginnis, M.S., Banerjee, R., O'Hara, B.A., Atwood,

W.J., 2014. The human alpha defensin HD5 neutralizes JC polyomavirus

infection by reducing endoplasmic reticulum traffic and stabilizing the viral

capsid. J Virol 88, 948-960.

233