Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Prevention of viral infection via modification of virus or cells with methoxypoly (ethylene glycol) McCoy, Lori L. 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2005-995135.pdf [ 18.3MB ]
Metadata
JSON: 831-1.0092388.json
JSON-LD: 831-1.0092388-ld.json
RDF/XML (Pretty): 831-1.0092388-rdf.xml
RDF/JSON: 831-1.0092388-rdf.json
Turtle: 831-1.0092388-turtle.txt
N-Triples: 831-1.0092388-rdf-ntriples.txt
Original Record: 831-1.0092388-source.json
Full Text
831-1.0092388-fulltext.txt
Citation
831-1.0092388.ris

Full Text

Prevention of Viral Infection via Modification of Virus or with Methoxypoly(Ethylene Glycol) By Lori L. McCoy B. Sc., Georgia Institute of Technology, 1999 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Pathology and Laboratory Medicine) T H E UNIVERSITY OF BRITISH COLUMBIA January 2005 © Lori L. McCoy, 2005 Abstract Viral entry into cells is typically mediated by specific interactions between viral proteins and cell surface receptors. Recent pharmacological methods have attempted to exploit this specificity. For example, the interaction of rhinoviruses with ICAM-1 (its cell receptor) can be inhibited by either free zinc or soluble ICAM-1. However, this and other approaches are highly virus specific. Therefore a broad-spectrum method of preventing viral infections is needed. One potential nonspecific method of preventing viral infections is through the modification of host cells with a nontoxic physical barrier. It is my hypothesis that this may be accomplished by the covalent derivatization of cell membranes with methoxypoly(ethylene glycol) [mPEG]. Our previous research demonstrated that covalent modification of mammalian cells with activated mPEG produced a protective barrier that functioned, in part, as a molecular sieve. While small molecules (e.g., water and glucose) readily pass through, larger molecules (e.g., antibodies), particles (e.g., immune complexes) and cells were excluded from interacting with membrane components. My study further extended these findings to models of viral pathogenesis. Five viruses were employed: Simian virus 40 [SV40], Theiler's murine encephalomyelitis virus [TMEV], mouse adenovirus [MAV], rat coronavirus [RCV] and cytomegalovirus [CMV]. Importantly, these viruses varied in mode of entry, size and structure. As demonstrated in this thesis, modification of either the virus or target cell effectively blocked viral infection. For example, cells challenged with unmodified SV40 were 47% infected at 24 ii hours while < 4 and 0% of cells modified with 2.4 and 15 mM 5 kDa cyanuric chloride activated mPEG (CmPEG) were infected, respectively. The broad spectrum effects of mPEG grafting were demonstrated by the findings that modification of host cells with only 0.2 mM activated mPEG (5 kDa) resulted in a 95%, 78% and 47% reduction in plaque formation compared to control cells challenged with RCV, M A V and T M E V , respectively. Further studies were conducted to determine the effects of reaction time, temperature, polymer size and linker chemistry on the antiviral efficacy of mPEG gafting. In summary, these studies show that mPEG modification of viruses and/or host cells is a potent and broad spectrum method of preventing viral infection. iii Table of Contents Abstract....: ii-iii List of Tables viii-ix List of Figures x-xiii List of Abbreviations xiv-xv Glossary xvi Acknowledgments xvii Chapter 1- Introduction 1.0 Overview 1 1.1 Rationale and Objectives 1-6 1.2 Current Methods of Preventing Viral Infection 6-14 1.3 Surface Modification Methods 14-15 1.4 Pegylation Technology 16-32 1.4.1 Pegylation: Previous and Ongoing Work 16-22 1.4.2 Pegylation Theory 22-31 1.4.3 PEG Immunogenicity? 32-33 1.5 Viruses and Infection 33-37 1.5.1 Structure 33-35 1.5.2 How Viruses Infect Host Cells 35-37 1.6 Human Common Cold Infections 37-39 1.7 SV40 Background 40-42 1.8 Summary 42-43 iv Chapter 2- Materials and Methods 2.1 Choice of Model Viruses 44-45 2.2 Materials 46-47 2.2.1 Cell Lines and Viruses 46 2.2.2 Antibodies 47 2.2.3 mPEG Species and Other Reagents 47 2.3 mPEG Modification of Virus and Cells 48-51 2.3.1 mPEG Modification of Virus 48-49 2.3.2 mPEG Modification of Viral Target Cells 49-51 2.4 Measuring the Efficacy of Immunocamouflage 52-58 2.4.1 Immunostaining 52-53 2.4.2 Plaque Assay 54-55 2.4.3 Detection of SV40 Entry: Immunostaining for VP1 56-58 2.5 Cell Maintenance and Viral Stock (Lysate) Production 58-61 2.5.1 Cell Culture 58-59 2.5.2 Stocks (Lysate Production) 59-61 2.6 Statistical Analysis 61 Chapter 3- Pegylation of SV40 Viral Particles Prevents Infection 3.1 Modification of SV40 Particles with CmPEG, BTCmPEG or SPAmPEG 5 kDa Prevents Infection 62-65 v 3.2 SV40 Modification using Different Molecular Weight mPEGs Prevents Infection ...65-69 3.3 Combination Studies .;. 69-73 3.4 Discussion 73-76 Chapter 4- Pegylation of Host (CV-1) Cell Surfaces Prevents SV40 Infection 4.1 Non-covalent versus Covalent mPEG Modification of Target Cells 77 4.1.1 Non-Covalent (Unactivated) mPEG does not Inhibit SV40 Infection 77-80 4.1.2 Covalent Modification of CV-1 Cells with CmPEG 5 kDa Prevents SV40 Infection..... 81-83 4.2 Effect of Reaction Time, pH and Temperature on Pegylation of Cells 84-91 4.3 Effect of Reaction Time, pH and Temperature on mPEG Protection When Challenged with Increasing Viral Concentrations 91-99 4.4 mPEG Dosage Schedule 100-102 4.5 CmPEG Modification Prevents Infection of Cells Incubated with Mucin 102-106 4.6 mPEG Inhibits SV40 Entry 106-110 4.7 Discussion 110-113 Chapter 5- Comparison of the Effects of mPEG Linker Chemistry and Polymer Size on Viral Infection 5.1 Pegylation of Cells with SPAmPEG Prevents SV40 Infection 114-120 5.2 Pegylation of Cells with BTCmPEG Prevents SV40 Infection 121-127 vi 5.3 Combination Studies 127-131 5.4 Discussion 131-137 Chapter 6- Pegylation of Cells Prevents Infection by a Broad Range of Virus Types 6.0 Overview 138-140 6.1 Pegylation of Cells with BTCmPEG 5 kDa Prevents SV40 Infection 140-142 6.2 Pegylation of Cells with BTCmPEG Prevents Theiler's Murine Encephalomyelitis Virus Infection 142-146 6.3 Pegylation of Cells with BTCmPEG Prevents Mouse Adenovirus Infection 146-149 6.4 Pegylation of Cells with BTCmPEG Prevents Rat Coronavirus Infection 150-153 6.5 Pegylation of MRC-5 Cells with BTCmPEG Prevents Cytomegalovirus Infection 153-156 6.6 Discussion 156-157 Chapter 7- Summary 7.1 Discussion 158-175 7.2 Potential Effects of Pegylation on Normal Function 176-178 7.3 Future Directions 179-181 References 182-195 vii L i s t of Tables Table 1.1 Immunocamouflage of Cellular Epitopes 22 Table 1.2 Common Cold Viruses 35 Table 2.1 Comparison of Common Cold Viruses to Model Systems 45 Table 2.2 Description of Each Cell Line and the Corresponding Virus Table 46 Table 2.3 Protein Sequence of VP1 49 Table 2.4 Protein Sequences of Virus Receptors 51 Table 3.1 Comparison of Protection by 5 kDa mPEGs with Varied Linker Chemistries 64 Table 3.2 Comparison of SPAmPEG Inhibition of SV40 Infection at Various Molecular Weights 66 Table 3.3 Comparison of BTCmPEG Inhibition of SV40 Infection at Various Molecular Weights 68 Table 3.4 Comparison of BTCmPEG Inhibition of SV40 Infection 71-72 Table 4.1 Amino Acid Sequence of Mucin 105 Table 5.1 SPAmPEG 20 kDa More Effectively Prevents Progeny Virus Infection than 2 or 5 kDa 116 Table 5.2 BTCmPEG 20 kDa is Less Effective at 24 hours but More Effective at 72 hours than 3.4 or 5 kDa 122 viii Table 5.3 BTCmPEG Combinations More Effectively Prevent SV40 Infection than Single Molecular Weights 128-130 Table 5.4 Summary of 5 kDa Polymer Data- SPAmPEG is Less Effective than BTC and CmPEG 134 Table 5.5 Comparison of 20 kDa mPEGs 136 Table 6.1 Description of Each Virus Model and Corresponding Cell Line 139 Table 6.2 Plaque Assay Comparison of BTCmPEG Protection Against T M E V Infection 145 Table 6.3 Plaque Assay Comparison of BTCmPEG Protection Against M A V Infection 149 Table 6.4 Plaque Assay Comparison of BTCmPEG Protection Against R C V Infection 153 Table 6.5 IE72 Immunostaining Comparison of BTCmPEG Protection Against C M V Infection 155 Table 7.1 Well-Known Human Viruses Share Characteristics with Model Viruses 167 Table 7.2 Low mPEG Concentrations Inhibit Larger Viruses More Readily than Smaller Viruses 174 Table 7.3 Comparison of Immunocamouflage to Current Methods of Preventing Viral Infection 175 ix L i s t of Figures Figure 1.1 Hypothesis 3 Figure 1.2 Specific Aim 1 4 Figure 1.3 Specific Aim 2 5 Figure 1.4 Specific Aim 3 6 Figure 1.5 Rhinovirus Saturated with Soluble ICAM-1 12 Figure 1.6 Red Blood Cell Immunocamouflage 17 Figure 1.7 Pegylation of Red Blood Cells Obscures Surface Charge 19 Figure 1.8 T Cell Modification Prevents Stimulation by Non-Self Cells 21 Figure 1.9 Structure of Lysine 24 Figure 1.10 CmPEG-Lysine Reaction 25 Figure 1.11 BTCmPEG-Lysine Reaction 26 Figure 1.12 SPAmPEG-Lysine Reaction 27 Figure 1.13 Mushroom and Brush PEG Regimes 30 Figure 1.14 mPEG Zone of Exclusion 31 Figure 1.15 Virus Structure 34 Figure 1.16 Viral Receptor Mediated Endocytosis 36 Figure 1.17 Viral Fusion 37 Figure 1.18 Normal versus SV40 Infected Cell Cycle Progression 41 Figure 2.1 Cover Slip Experiment 53 Figure 2.2 Plaque Assay Experiment 55 x Figure 2.3 VP1 versus T antigen Staining 57 Figure 3.1 Modification ofthe SV40 Capsid is More Protective with the Combination of 3.4/20 than 5/20 76 Figure 4.1 Unactivated mPEG Does Not Prevent Infection 79 Figure 4.2 Microscopic Look at the Effects of Unactivated mPEG on SV40 Infection 80 Figure 4.3 CmPEG Modification of CV-1 Cells Prevents SV40 Infection 82 Figure 4.4 Microscopic Look at the Effects of CmPEG 5 kDa on SV40 Infection 83 Figure 4.5 A Reaction of 5 Minutes is Sufficient for Protection (pH 8.4, Room Temperature, 24 Hours Post SV40 Challenge) 85 Figure 4.6 A Reaction of 5 Minutes is Sufficient for Protection (pH 8.4, Room Temperature, 72 Hours Post SV40 Challenge) 86 Figure 4.7 A 5-Minute Reaction is Sufficient for mPEG Protection with Physiologic Reaction Conditions (24 hrs) 88 Figure 4.8 A 5-Minute Reaction is Sufficient for mPEG Protection with Physiologic Reaction Conditions (72 hrs) 89 Figure 4.9 CmPEG Modification of CV-1 Cells Prevents SV40 Infection Over a Wide Range of SV40 Concentrations (Four Logs), 24 Hours Post Challenge 93 Figure 4.10 CmPEG Modification of CV-1 Cells Prevents SV40 Infection Over a Wide Range of S V40 Concentrations (Four Logs), 72 Hours Post Challenge 95 xi Figure 4.11 A 5 Minute Modification of CV-1 Cells Prevents SV40 Infection Over a Wide Range of S V40 Concentrations (Four Logs), 24 Hours Post Challenge 96 Figure 4.12 A 5 Minute Modification of CV-1 Cells Prevents SV40 Infection Over a Wide Range of SV40 Concentrations (Four Logs), 72 Hours Post Challenge 98 Figure 4.13 CmPEG Modification of CV-1 Cells for 5 Minutes Prevents Infection with Increased Viral Titers, 24 Hours Post Infection 99 Figure 4.14 A 5 Minute mPEG Reaction at pH 7.8 (32 °C) Effectively Prevents SV40 Infection for 24 Hours 101 Figure 4.15 Pegylation of Cells Prevents Infection in the Presence of Mucin 104 Figure 4.16 CmPEG Modification of Cells Inhibits SV40 Entry 108 Figure 4.17 A Microscopic Look at SV40 Entry 109 Figure 4.18 Actin Staining of Modified and Unmodified Cells 110 Figure 5.1 Modification of CV-1 Cells with SPAmPEG 2, 5 or 20 kDa Prevents Viral Infection Over a Wide Range of Virus Titers, 24 Hours 118 Figure 5.2 Modification of CV-1 Cells with SPAmPEG 2, 5 or 20 kDa Prevents Viral Infection Over a Wide Range of Virus Titers, 72 Hours .119 Figure 5.3 Modification of CV-1 Cells with BTCmPEG 3.4, 5 or 20 kDa Prevents Viral Infection Over a Wide Range of Virus Titers, 24 Hours 124 Figure 5.4 Modification of CV-1 Cells with BTCmPEG 3.4, 5 or 20 kDa Prevents Viral Infection Over a Wide Range of Virus Titers, 72 Hours 126 xii Figure 6.1 BTCmPEG 5 kDa Modification of Cell Monolayers Inhibits Plaque Formation 141 Figure 6.2 Modification of Cells with BTCmPEG Prevents T M E V Infection 143 Figure 6.3 T M E V Plaque Assay Pictures 144 Figure 6.4 Modification of Cells with BTCmPEG Prevents M A V Infection 147 Figure 6.5 M A V Plaque Assay Pictures 148 Figure 6.6 Modification of Cells with BTCmPEG Prevents R C V Infection 151 Figure 6.7 R C V Plaque Assay Pictures : 152 Figure 7.1 Masking Surface Charge with mPEG Plays a Role in Preventing Viral Infection 160 Figure 7.2 Modification of Cells versus SV40 Capsids 163 Figure 7.3 Importance of Physiologic mPEG Reaction Conditions 164 Figure 7.4 mPEG Modification of Cell Surfaces Prevents Infection of a Broad-Spectrum of Viruses 168 Figure 7.5 Prevention of Progeny Virus Infection by 20 kDa mPEGs 171 Figure 7.6 C M V Progeny Virus Infection is More Inhibited than that of SV40 at Low Concentrations of mPEG 173 Figure 7.7 Prevention Time-Release Effect of mPEG Modified Virus In Vivo 178 xiii List of Abbreviations APN aminopeptidase N A T C C American Tissue Culture Collection BCS bovine calf serum BTCmPEG benzotriazole carbonate mPEG CAR coxsackie and adenovirus receptor CmPEG cyanuric chloride mPEG C M V cytomegalovirus D M E M Dulbecco's modified Eagle's media DNA deoxyribonucleic acid ds double stranded EGFR epidermal growth factor receptor F12K Kaigen's modification of Ham's F12 media FBS fetal bovine serum FDA Food and Drug Administration ICAM-1 intracellular adhesion molecule-1 ID infectious dose IE72 immediate early protein (72 kDa) M A V mouse adenovirus M E M minimal essential media MHC-1 major histocompatibility complex-1 xiv MOI multiplicity of infection mPEG methoxypoly(ethylene glycol) PBS phosphate buffered saline PEG poly(ethylene glycol) pfu plaque forming unit RBC red blood cell RCV rat coronavirus RNA ribonucleic acid RSV respiratory syncytial virus SARS severe acute respiratory syndrome sICAM soluble ICAM-1 SPAmPEG succinimidyl propionate mPEG ss single stranded SV40 simian virus 40 T Antigen transforming antigen T M E V Theiler's murine encephalomyelitis virus VP1 viral protein 1 xv Glossary APN coronavirus receptor Balb/3T3 mouse embryonic fibroblast cell line BHK-21 hamster kidney fibroblast cell line CAR adenovirus receptor CV-1 African green monkey kidney cell line EGFR C M V receptor ICAM-1 rhinovirus receptor IE72 C M V immediate early protein used to monitor infection L2 rat lung epithelial cell line MHC-1 SV40 receptor MRC-5 human lung fibroblast cell line T Antigen SV40 early protein used to monitor infection VP1 major coat protein of SV40 xvi Acknowledgements The work that went into this thesis could not have been completed without the support of a number of wonderful people. First, I would like to thank my supervisor, Mark, for reading countless drafts of everything I have written, for his sense of humor, and for giving in to my insistence that this would be my project. I also need to thank Dr. John Lehman for laying the foundation for my work by teaching me many of the virology techniques used in this thesis. I am also indebted to my thesis committee: Cedric Carter, Dana Devine and Ed Pryzdial for their guidance through my UBC experience. Additionally, Dana's laboratory for giving up some highly valued space while we waited for our laboratory to be found/renovated and for becoming my Canadian family. I may not miss Vancouver when I leave, but I will certainly miss all of you. I'd also like to thank the 'support staff that I left behind when I followed my project to Canada. First, Danielle for bringing some much needed life to Albany Med and being a friend I know will always have my back. I hope I was as helpful during your degree process as you have been during mine. To my family for your encouragement and love- thanks Mom, Dad, Jill, Aunt Mildred and Dolores. Finally, I'd like to dedicate this work to two important women in my life that I lost during this process- Gram and Dolores I will miss you both deeply. Thank you for being among my most vocal supporters, you are remembered fondly and with love. xvii Chapter 1: Introduction 1.0 Overview A crucial step in viral infection of permissive cells is entry. Cell entry is typically mediated via highly specific interactions between viral capsid or proteins and cell surface receptors (Flint, et al, 2000). Current preventative therapies attempt to exploit this specificity, making them inherently virus specific and therefore of limited usefulness towards other viral agents. This chapter will provide a brief description of the objectives of this thesis project. The drawbacks of several common antiviral strategies and preventative drug treatments will be discussed and the merits of covalent modification of viruses or cell surfaces with methoxy(polyethylene glycol) [mPEG] as a preventative therapy against viral infection will be illustrated. 1.1 Rationale and Objectives The general objective of this thesis was to determine whether modification of viruses or host cell surfaces with activated mPEG could prevent viral infection by a wide range of viruses. Previous work from Dr. Scott's laboratory had clearly demonstrated that pegylation of cells inhibits receptor-ligand interactions resulting in decreased red blood cell agglutination, T cell proliferation, and antibody recognition of surface antigens (Scott, et al, 1997; Scott and Murad, 1998; Murad, et al, 1999A and B; Bradley, et al, 2001; Chen and Scott, 2001; Bradley, et al, 2002; Chen and Scott 2003; Scott and Chen 2004). Considering the importance of receptor-1 ligand interactions to viral infection, I hypothesized that the covalent modification of host cells and/or virus particles with methoxypolyfethylene glycol) [mPEG] would inhibit viral infection. The hypothesis of this thesis is outlined in Figure 1.1. Using the nasal passage epithelium as an example, virus would be introduced into the local environment where upon it recognizes cellular receptors (1). For rhinovirus, the most common cold virus, this receptor is ICAM-1 (Staunton, et al, 1989; Greve, et al, 1989; Bella, et al 1998; Charles, et al 2003; Pitaranta, et al. 1998). Following binding to the receptor (1), virus would be internalized (2) where it undergoes multiple rounds of replication (3). The virus would then be packaged into the capsid proteins and shed (4) into the extracellular environment (5) where upon it would continue the disease process by infecting new, naive, epithelial cells (1). However, as hypothesized, an activated mPEG aerosol can be used to interrupt this infectious process. This is proposed to be accomplished via the two mechanisms shown. In mechanism 1, any free virus would be covalently modified by the mPEG. As hypothesized, the virus would no longer recognize or interact with the cell surface receptors, thus preventing entry and subsequent infection of the cell. In mechanism two, viral infection is inhibited through the direct covalent grafting of mPEG to the nasal epithelium itself. As shown diagrammatically, the derivatization of the epithelium by mPEG is proposed to give rise to a protective barrier that camouflages cell surface proteins including the viral receptors utilized during infection. As a consequence, the viral particle is unable to effectively initiate and/or propagate the infection. While the example depicted in Figure 1.1 is that of nasal epithelium, this methodology is thought to be applicable to all circumstances where viral infection is a possibility (e.g., vaginal and oral surfaces, etc.). 2 © ^S^> mPEG-Modified ... . . . . . " ' ~9 ..M^&M';-}.&£:M>V> Virus Particle § Normal Epithelial Cells Rhinovirus Influenza Virus Disease Pathogenesis Normal Nasopharyngeal Epithelial Cells L- . «• ^ 1 r - ' Immunocamouflaged 1 ICAM-1 or Of/?©/" V/rus Receptor Virus Particle (#-ii?9> mPEG-Modified ttiWW' Virus Particle Nasal Aerosol/Nebulizer mPEG mPEG •WWCtftS Hydration vK-'Sy Sphere Figure 1.1: Hypothesis Normal viral disease pathogenesis is noted in the left panel and includes: (1) Attachment (for rhinovirus this receptor is ICAM-1), (2) Entry, (3) Replication, (4) Release (5) Free virus. An aerosolized spray of activated mPEG (mPEG) may be used to interrupt this infectious process. As shown, modified virus and/or host cells can no longer recognize or interact with each other, thus preventing infection. As previously stated, current methods of prevention of viral infection target specialized interactions of a single strain of virus with its specific receptor (univalent in nature) while, in contrast, mPEG modification of cells is hypothesized to be non-specific by nature and capable of preventing infection by a wide range of viral agents (i.e. broad spectrum). The non-specific inhibition arises as a consequence of stochastically modifying lysine residues of any membrane-bound protein (e.g., ICAM-1, MHC-1, etc.) resulting in the global camouflage of receptors necessary for viral infection. Three specific aims were designed to test my hypothesis: 3 1. Determine whether the covalent modification of host cells or virus particles with mPEG abrogates viral invasion thereby preventing infection (Figure 1.2). 2. Determine the effects of polymer size and linker chemistry on prevention of viral infection (Figure 1.3). 3. Determine whether pegylation of target cells is universally effective in preventing viral infection (Figure 1.4). Specific Aim 1 Modification of Cells Unmodified Cells mPEG Modified Cells Modification of Virus H I , cbc5ocb Unmodified Virus mPEG Modified Virus Figure 1.2: Specific Aim 1 The first goal of specific aim 1 is to determine if the covalent modification of target cells prevents viral infection. As shown in panel A, unmodified receptors interact with viral particles and infection will result. However, if these cells are covalently modified with mPEG viral invasion is thwarted due to the gel-like nature of the PEG brush border (stippled area), which prevents large molecules (i.e. virus) from reaching the cell surface. The second goal of specific aim 1 is to determine if the covalent modification of virus particles prevents infection. Once again in panel B, unmodified virus is capable of interacting with its receptor and infecting the cell. However, by modifying the virus we produce a protective layer that prevents the virus from binding its cell surface receptor. 4 Specific Aim 2 Linker Chemistry CH3(-0-CH 2 -CH 2 ) n -CK - c l c m P E G Y N CH 3 0(CH 2 CH 2 0) n CON N B T C - P E G 0 C H 3 0 ( C H 2 C H 2 0 ) n C H 2 C H 2 C O N CO  S P A - P E G 0 Polymer Size Comparative Chain Length 2 kDa _ 3.4 kDa 5 kDa 20 kDa Comparative Area of Protection 2 kDa 3.4 kDa it %t "„". s". %t C*» V"A^  t^ •.»«.*» «*• ."• **• «»• •", 5 kDa «». «"»»"«« .*« «"•«" * .»*.»« «*; **• ^  - •*««*« -V «* • «*.• -V ? **«**« «•«.*« «* * «** «"«.» • • Figure 1.3: Specific Aim 2. The goal of specific aim 2 is to delineate the differences in protection between 5 kDa mPEGs with linker molecules: cyanuric chloride (cmPEG), benzotriazole carbonate (BTCmPEG), and succinimidyl propionate (SPAmPEG). Each type of mPEG is shown in the upper box. The second goal of aim 2 is to compare the protection afforded by various sized mPEG chain lengths. For example BTCmPEG will be compared at 3.4, 5 and 20 kDa molecular weights, while SPAmPEG will be compared at 2, 5 and 20 kDa. As shown in the lower box, there is a large variation in the area of protection afforded by each molecular weight mPEG. This difference is due to flexibility, as well as the high level of hydration of the PEG chain. The combined effect of these properties is formation of a barrier, which inhibits virus-receptor interactions. 5 Figure 1.4: Specific Aim 3 The objective of specific aim 3 is to determine the effectiveness of mPEG modification of target cells against viral infection of a wide range of viruses. The viruses are shown above from smallest to largest diameter. The viruses are Theiler's murine encephalomyelitis virus (TMEV; ~20 nm), simian virus 40 (SV40; 45 nm), mouse adenovirus (MAV; 70-90 nm), rat coronavirus (RCV; 80-160 nm), and human cytomegalovirus (CMV; -200 nm). 1.2 Current Methods of Preventing Viral Infection Current methods to prevent viral disease range from natural drug treatments (e.g., mega doses of vitamin C, oral zinc and Echinacea) to disinfecting all surfaces. These methods are not always practical and many have been proven ineffective in preventing viral disease (Block and Mead, 2003 and Yale and Liu, 2004). 6 To effectively prevent viral disease, one must target propagation of the virus (e.g., antiviral drugs), which is the most common approach, the virus itself (e.g., sanitation and vaccination), or initial virus-receptor interactions (e.g., Zicam), which have only more recently been explored. Once infected with a common cold virus or influenza, there is only a brief window of time (1-2 days) in which antiviral drug administration may be effective in preventing replication and propagation of the virus; thus attenuating the onset or severity of disease. Among the best described antiviral agent is Amantadine, which is used to prevent the uncoating of influenza A. However, Amantadine is only effective during the early stages of viral infection (McKinlay, et al. 1992; Flint, et al. 2000). Within cells, Amantadine specifically inhibits viral uncoating by blocking the activity of the proton channel of the influenza A M2 protein (Kandel and Hartshorn, 2001). With this channel blocked, the pH within the endosome containing influenza A cannot be altered sufficiently for the virus to uncoat. However, if Amantadine is administered after the virus has uncoated, the infection will proceed normally and the resulting progeny virus will propagate the infection. Amantadine is also highly specific to influenza A, as it is not effective against other strains of influenza much less different viral families. Additional drawbacks have been the emergence of drug resistant strains of influenza and as with many antiviral drugs, Amantadine therapy has serious toxic side effects. For example, Amantadine affects the nervous system, and approximately 10% of people using the drug experience nervousness, depression, anxiety, difficulty concentrating, and lightheadedness (Osondu-Alilonu and Gross 2003). Additional common complaints include nausea, vomiting, and indigestion. Rarely, Amantadine can cause hallucinations and seizures (Kandel and Hartshorn, 2001; Osondu-Alilonu and Gross 2003). 7 Other drugs targeting viral propagation include those commonly used to fight long term viral diseases (e.g., HIV infection) include nucleoside analogs and non-nucleoside reverse transcriptase inhibitors (McKinlay, etal. 1992; Flint, et al. 2000). Both classes of drugs attempt to prevent the replication of a virus following successful cellular entry. Nucleoside analogs are similar in structure to normal cellular nucleoside bases, varying from normal bases by only a minor modification in the ribose molecule, and compete with the physiological nucleosides for incorporation into the replicating viral genome (Flint, et, al, 2000). This modification causes DNA synthesis to end because phosphodiester bonds needed to elongate the strand can no longer be made. Unfortunately, nucleoside analogs are toxic and cause a wide spectrum of side effects including: destruction of bone marrow, accumulation of lactic acid throughout the body, polyneuropathy (non-inflammatory degeneration of nerves), and pancreatitis (Brinkman, et al. 1999). Additional frequent complaints include fatigue, headache and a variety of gastrointestinal problems such as abdominal discomfort, nausea, vomiting and diarrhea (Galli, et al. 2002). In contrast to nucleoside analogs, non-nucleoside reverse transcriptase inhibitors bind directly and non-competitively to the viral reverse transcriptase enzyme (Flint, et, al, 2000). The non-nucleoside reverse transcriptase inhibitor binding site is in close proximity to the substrate-binding site for nucleosides, and the resulting complex blocks the binding site ofthe reverse transcriptase, thereby significantly reducing the rate of polymerization (Flint, et al. 2000). As with the preceding compounds, toxicity and allergic reactions are noteworthy concerns 8 (Sulkowski, et al. 2002). Furthermore, the most significant problem with non-nucleoside reverse transcriptase inhibitors has been the development of resistant viral strains. Both nucleoside analogs and non-nucleoside reverse transcriptase inhibitors have been shown to decrease the overall viral load in patients (Conway, 2000; Squires, et al. 2003). However, a population of virus that is capable of propagating the infection almost invariably remains. For this reason, nucleoside analogs as well as non-nucleoside reverse transcriptase inhibitors need to be taken over a long period of time and often in cocktails of several different drugs to curtail the emergence of resistant populations (Hanna and D'Aquila, 1999; Johnson, 2002). In addition, due to the toxicity associated with these compounds, they are not suitable treatments for a majority of viral infections and are only appropriate for viruses characterized by high morbidity and mortality (e.g., HIV) (Preston, et al. 2003). Due to the drawbacks of anti-viral medications, the holy grail of antiviral medicine remains prevention. Unfortunately, there are few methods of thwarting viral infection currently in use. These prophylactic antivirals include personal hygiene, vaccination, and remarkably few over-the-counter preventative drug treatments. As shown throughout history, the most effective method of reducing transmissible diseases, such as common colds and bacterial infections, is good hygiene and sanitation. Physically removing germs through hand washing has been identified as a major factor in preventing both viral and bacterial infections in hospitals as well as at home (Lintz, et. al. 1976; Hall, et. al. 1978). 9 The more advanced method of preventing viral infection through vaccination is highly effective when the proper vaccines are available. There are currently several vaccines to prevent viral infections in humans, some of which include: chicken pox (varicella zoster), hepatitis A and B (hepatitis virus), poliomyelitis (poliovirus), measles (morbillivirus), mumps (mumps virus), rubella (rubivirus), rabies (rhabdovirus), and the flu (influenza). While vaccines generally provide excellent protection against infection, there are several drawbacks. In the case ofthe influenza, a new vaccine must be administered in the early part of each flu season because the infectious strains vary year to year (Belshe and Mendelman, 2003). The World Health Organization epidemiologically tracks which strains of influenza are causing disease throughout the world and recommend the combination of strains to be included in the annual vaccine. Unfortunately, the tracking system is not absolutely accurate and the potential to miss an infectious strain remains a real risk. In addition, the vaccines for measles, mumps, rubella, and polio are attenuated live virus preparations that have the potential to cause the very infection they are meant to prevent. Most often these unexpected infections occur in individuals with undiagnosed immunodeficiencies (McLaughlin, et al. 1988; Ion-Nedelcu, et al 1994; Weber and Rutala 2003; Edghill-Smith, et al. 2003; Amorosa and Isaacs, 2003; Hidalgo, et al. 2003). Fortunately, these problems rarely occur. Other uncommon, though potentially serious side effects of vaccines include: severe systemic reaction characterized by respiratory symptoms, fainting, itching, and hives known as anaphylaxis, febrile seizures, sub-acute sclerosing panencephalitis (central nervous system disease caused by infection of the brain by the morbillivirus and marked by intellectual deterioration, convulsions, and paralysis), and hypersensitivity reactions (Gupta and Bajpai, 1966; Aho, et al. 1967; Annamalai, et al. 1967; Heymann, 1967; Black, et al. 2004; CDC- no author listed, 2004). More importantly 10 however, the majority of viral diseases have no vaccines, consequently the development of preventative drug therapies is a major research focus. One of the few preventative drugs currently in use in the United States and Canada is Zicam. Zicam, which is sold over-the-counter, is a highly concentrated zinc gel (zincum gluconicum) containing Z n 2 + ions, which interact with rhinovirus capsid proteins. Crystallographic evidence suggests the surface of rhinovirus-14 contains binding sites for at least 360 Zn ions (Novick, et al. 1996). When bound, zinc ions physically block a portion of the binding pocket of the rhinovirus capsid thereby interfering with rhinovirus-receptor interactions (Novick, et al. 1996; Hirt, etal. 2000; Eby, etal. 1984; Godfrey, etal. 1992; Korant and Butterworth, 1976). The receptor for up 80-90% of rhinovirus strains is Intracellular Adhesion Molecule-1 (ICAM-1) (Marlin, et al., 1990). Nasal administration of the zinc preparation within 24-48 hours of symptom onset and has proven effective in shortening the length of illness (Hirt, et al. 2000), most likely due to the prevention of secondary infection by progeny virus. However, the zinc treatment is only effective for rhinovirus infections (not other common cold viruses such as Adenovirus and Coronavirus) due to its specificity in interfering with the ICAM-rhinovirus interaction. Additionally, there are at least two other experimental methods of preventing rhinoviral infection. The first is the application of soluble Intracellular Adhesion Molecule (sICAM), which competes with cell-associated ICAM-1 for binding sites on the rhinovirus capsid, thereby preventing viral entry (Marlin, et al. 1990; Huguenel, et al. 1997; Turner, et al. 1999). Therefore, as shown in Figure 1.5, excess sICAM present in nasal passages saturates 11 rhinoviral-binding sites before the virus comes into contact with a cell, preventing infection. In vitro studies conducted by Marlin, et al. (1990) showed that an sICAM dose of 10 u.g/mL inhibited 90% of the rhinovirus-54 cytopathic effects (resulting from viral infection) seen in control cells. Importantly, when sICAM was tested on herpes simplex virus type 1 (HSV-1), coxsackie virus B1 and poliovirus there was no significant inhibition of cytopathic effect compared to control cells (Marlin, et al. 1990). The lack of prevention of infection by non-rhinovirus agents was expected due to the specificity of sICAM to viruses that utilize the ICAM-1 receptor. Furthermore, sICAM was unable to prevent infection of rhinovirus-2, which is among the 10-20% of rhinovirus strains that do not use the ICAM-1 receptor. An additional randomized clinical trial was conducted by Turner, et al. (1999), in which sICAM was administered intranasally either in solution or powder form seven hours pre-inoculation with rhinovirus-39, or 12 hours post-inoculation. As with conventional treatments for the common Normal Rhinovirus Capsid Rhinovirus Saturated with ICAM Figure 1.5: Rhinovirus Saturated with Soluble ICAM-1 Above is an illustration of the rhinovirus capsid with ICAM-1 binding sites noted. Shown in B is the virus saturated with soluble ICAM. Inspired by the electron micrograph from http://www.virology.net/Big_Virology/BVRNApicorna.html 12 cold, sICAM was effective in reducing the symptoms of infection despite very weak prevention of initial rhinovirus infection, as demonstrated by 92% of the placebo-treated subjects and 85% . of the sICAM-treated subjects who became infected (Turner, et al. 1999). Clinical results were much less effective than the previous in vitro studies (85% of patients infected versus 10%o of cells). The disparity in results is likely due to the clearance of sICAM by mucociliary defenses of the nasal mucosa during the 7-hour delay between drug delivery and viral challenge. A second experimental method of preventing rhinoviral infections is the application of anti-ICAM-1 antibodies into the nasal passage prior to viral inoculation (Hayden, et al. 1988). In human trials by Hayden, et al. (1988), it was shown that administration of rhinovirus receptor murine monoclonal antibodies (RRMA) did not reduce the infection or illness rates. However, RRMA treatment was associated with a 1-2 day delay in the onset of viral shedding and common cold symptoms. The drawbacks of antibody therapy are, as with soluble ICAM, rapid clearance and, in many cases, immunogenicity, because of the non-human origin of the antibodies (Charles, et al. 2003; Luo, et al. 2003). Furthermore, as with the soluble ICAM treatment, anti-ICAM antibody therapy is only effective against the 80-90% of rhinovirus strains that utilize the ICAM-1 receptor. In contrast, the covalent modification of cells with mPEG [CH3(CH2CH20)n-CH2CH20H] to prevent viral infection as hypothesized by this thesis would overcome the problems of rapid clearance, immunogenicity and virus specificity of these aforementioned treatments. Perhaps the most significant advantage of mPEG grafting is its proposed broad-spectrum effect. As 13 stated, due to the specificity of action, Zicam, soluble ICAM, and anti-ICAM antibody treatments are incapable of preventing infection by common cold viruses other than rhinoviruses. It has been estimated that rhinoviruses account for only 30-35% of adult colds, while coronaviruses, adenoviruses and several other viruses are responsible for the remainder of colds (National Institutes of Health, 2003). For this reason, a more universal method of preventing viral infection is needed and mPEG grafting may give rise to this broad-spectrum antiviral prophylaxis. 1.3 Surface Modification Methods to Increase or Decrease Immune Recognition The cell membrane is a complex mixture of lipids, proteins and carbohydrates. These components function together during vital cell tasks such as cell signaling, endocytosis and immunorecognition. Due to the wide range of cellular activities involving membrane components, alteration of cell surfaces has been a major field of research. Applications for surface engineered cells include drug delivery, prevention of tissue rejection and nerve regeneration (Kellam et ai, 2003). Thus far, cells have been modified through glycosylphosphatidylinositol (GPI) anchored proteins and cholesterol tethered molecules both of which readily insert into the lipid bilayer of cell membranes (Sloan et al, 1998; McHugh et al., 1999; Premkumar et al., 2001; Pekari et al., 2003; Xue et al., 2003). The GPI-anchored or cholesterol-linked molecules of interest include cell surface receptors, enzymes and antigens. Typically antigens are added to cancer cells to initiate an immune response from natural killer and T cells (McHugh et al., 1999). Tumor cell 14 membranes have also been modified through metabolic engineering where cells are incubated with precursors of cell surface moieties that are taken up and metabolized, resulting in the incorporation of unnatural structures on the exterior of the cell (Herrler et al, 1992; Keppler et al., 1995; Oetke et al., 2003). Again, the metabolic engineering technique is often utilized to initiate an immune response to eliminate cancer cells. Conversely, cell membranes have also been modified to reduce or eliminate immune responses. Cell surfaces have been modified by enzymatic cleavage of surface antigens as well as through the covalent ligation of functional groups or polymers such as mPEG to prevent immunorecognition and tissue rejection (Lenny et al., 1991; Scott et al. 1997; Yarema et al., 1998; Scott and Murad, 1998; Lee et al, 1999; Panza et al, 2000; Scott et al. 2000; Bradley et al, 2001; Chen and Scott, 2001; Bradley et al, 2002; Scott and Chen, 2004; Doucet et al, 2004). For example, studies by Lenny, et al, (1991) utilized the enzyme a-galactosidase to convert red blood cells from group B to O by removing the immunodominant carbohydrates. One-unit transfusions of enzyme converted red blood cells were administered to either group A or O recipients. No increase in serum anti-B antibodies were seen at any time during the studies in patients receiving blood treated with 90 or 185 U/mL a-galactosidase for A and O recipients, respectively. Post-transfusion sera were unable to agglutinate or lyse the O-converted red blood cells and importantly, these red blood cells showed normal in vivo survival time. This technology has now reached the point of Phase I and II clinical trials (Kruskall et al, 2000). 15 1.4 Pegylation Technology 1.4.1 Pegylation: Previous and Ongoing Work Pegylation of cell membranes may provide a broad spectrum approach to preventing viral infection. Pegylation of proteins has been used for years to prevent aggregation during long term storage, increase systemic circulation of drug therapies, and reduce drug toxicity (Barrett andHanigan, 1975; Abuchowski, et al., 1977; Kawai, et al, 1980; Kinstler, et al, 1996; Goodson and Katre, 1990; Cantin, et al, 2002). Most importantly, pegylated proteins exhibit decreased in vivo immunogenicity following repeated administration (Ogris, et al, 1999) suggesting that mPEG is interfering with immunorecognition. The earliest attempts to modify cell surfaces with mPEG focused on pegylating red blood cells (RBCs) in order to mask surface antigens (Scott, et al. 1997; Chen and Scott 2001). As shown in Figure 1.6A human RBCs demonstrate a loss of antibody mediated aggregation of the major blood group antigen A and minor blood group antigen RhD (Scott, et al. 1997; Scott and Murad, 1998; Scott et al 2000; Bradley, et al, 2001; Chen and Scott, 2001; Bradley, et al, 2002; Scott and Chen, 2004). This loss of antibody mediated aggregation was due to the direct camouflaging of blood group antigens by the grafted mPEG. Additionally, human peripheral blood monocytic cells were unable to phagocytose the mPEG modified sheep RBCs (B) as cell-cell interactions were inhibited by mPEG. Consequently, due to mPEG inhibition receptor-ligand interactions, these modified sheep RBCs also displayed significantly reduced immunorecognition and antibody response in mice (C). Importantly, as depicted in Figure 1.6D, pegylated RBCs showed normal in vivo survival. These results arise consequent to the 16 covalent attachment of mPEG to cell surface proteins, which results in a physical and immunological barrier that prevents receptor-ligand interactions and immune recognition. Anti-Blood Group A Serum Control 0.2 mM 0.6 mM 1.2 mM Anti-RhD Monoclonal Antibody Control 0.6 mM 1.2 mM 2.4 mM • Control RBC • 1.2 mM CmPEG • 5.0 mM CmPEG X Xenogeneic Sheep RBC Allogeneic C57B1/6 RBC Syngeneic Balb/c RBC mM mPEG/ml Sheep RBC C 1100 (8 E 5 BE 0 75 m cc 1 50 | 25 0 GL D 0.8 mM ^ j T \ /0.4 mM i AflmM 2.0 mM 0 mM Normal Mouse RBC Survival Is 50 Days ™ 10 20 30 Days 40 50 Figure 1.6: Red Blood Cell Immunocamouflage Previous work in our laboratory showed that pegylation of RBCs demonstrate (A) loss of antibody binding/agglutination of A/B and RhD blood group antigens; (B) loss of antigenic recognition necessary for phagocytosis; (C) loss of immunogenicity; and (D) normal in vivo survival. Figure modified from Chen and Scott 2001. 17 In addition to the immunocamouflage effect seen in pegylated red blood cells, mPEG modification greatly reduces electrophoretic mobility. Electrophoretic mobility of RBCs is a function of membrane surface charge as well as the drag coefficient of the cell itself. The molecular origin of surface charge has not been fully determined; however, Thethi, et. al. (1997) suggest the electric charge can originate in three ways: 1) adsorption or desorption of ions from the media, 2) dissociation or ionization of surface groups, or 3) chemical reaction of surface groups with substances in the media. Figure 1.7 shows particle electrophoresis of red blood cells modified by increasing concentrations of two different activated mPEG species [benzotriazole carbonate mPEG (BTCmPEG 5 or 20 kDa) or succinimidyl propionate mPEG (SPAmPEG 2, 5 or 20 kDa)]. Under the conditions used by Bradley, et al. (2002), normal (unmodified) human RBCs have a mobility rate of approximately -1.18 pm'cm/volt'sec. As shown, mPEG-derivatization resulted in obscuring the surface charge of the RBC causing a decrease in electrophoretic mobility. It was found that the obscured charge due to pegylation was a function of both polymer density (mM) and polymer length (kDa). As the mPEG concentration and molecular weight increased, the mobility of modified cells shifted towards zero. Interestingly, complete neutralization of the electrophoretic mobility was noted at very low derivatization concentrations of the 20 kDa BTC- and SPAmPEGs (Bradley, et al, 2002). As hypothesized in this thesis, obscuring cellular or viral surface charge and accessibility would result in interference with receptor-ligand interactions and therefore viral attachment and subsequent infection. 18 0 2 4 6 8 0 2 4 6 B T C m P E G (mM) S P A m P E G (mM) Figure 1.7: Pegylation of RBCs Obscures Surface Charge. P v B C were modified as washed cells at a 12% hematocrit for 1 hour at room temperature. The electrophoretic mobilities of control and mPEG-modified cells were determined using measurements for ten RBC per experiment. Normal human RBC had a mobility rate of approximately -1.18 jim'cm/volt'sec compared to 0 pm'cm/volfsec for cells modified by 1.2 mM BTC- or SPAmPEG 20 kDa. Symbols show the mean of 2 independent experiments ± one SD. Figure modified from Bradley, et al, 2002. The immunocamouflage efficacy of mPEG modified cell surfaces has been further explored using lymphocytes and pancreatic islet cells (Murad, et al 1999A, Chen and Scott 2001). Transfusion or transplantation of T lymphocytes into an allogeneic recipient can provoke a potent immune response in immunocompromised patients known as graft-versus-host disease. In studies by Murad, et al. (1999A) and Chen and Scott (2001 and 2003), peripheral blood mononuclear cells (PBMCs) were isolated from disparate donors for use in mixed lymphocyte reactions (MLR). In control MLRs massive proliferation of the cells was noted by H-thymidine incorporation. In contrast, if one population of cells was covalently modified with 19 mPEG, allorecognition was blocked. For example, MLRs with human PBMCs modified with only 0.6 mM CmPEG showed a greater than seventy-five percent decrease in proliferation, while 1.2 mM CmPEG modification decreased proliferation by greater than ninety-five percent (Murad, et al. 1999A). Furthermore, in a murine model of transfusion-associated graft versus host disease (GVHD), pegylation of allogeneic donor cells prevented induction of GVHD (Chen and Scott 2001 and 2003). Mechanistically, the loss of allorecognition is due to the inhibition of receptor-ligand interactions by the grafted mPEG, which is highly relevant to virus-receptor interactions. For example, the normal function of human intercellular adhesion molecule-1 (ICAM-1) is to provide adhesion between endothelial cells and leukocytes after injury or stress; however, ICAM-1 is also used as a receptor by the major group of human rhinoviruses (Bella, et al., 1998). As shown diagrammatically in Figure 1.8, previous studies from our laboratory showed interactions between T cells and antigen presenting cells including: CD50-CD1 la, CD28-CD80, M H C II- TCR/CD3, CD58-CD2 were inhibited following mPEG grafting. Table 1.1 displays the decrease in antibody recognition of these cell surface epitopes as the concentration of mPEG used to modify the murine splenocytes increased. All of these epitopes play important roles in stimulatory, adhesion and allorecognition pathways, which are mediated by specific protein-protein interactions much like the virus-receptor interactions required for viral infection (Rudin, et al, 1996; Gudmundsdottir, et al, 1999; Abbas, 2000). Additionally, using a rat model of diabetes, pegylated pancreatic islet cells were found to retain normal structure and function. As shown by Chen and Scott, modified pancreatic islet cells were able to engraft and more importantly, were able to reestablish blood glucose homeostasis. Interestingly, pegylated islets appeared to function better than unmodified cells, likely due to 20 mPEG protection from immunologic responses resulting from damage during ex vivo processing (Chen and Scott, 2001). CelhCell Interaction Figure 1.8: T Cell Modification Prevents Stimulation by Non-Self Cells mPEG derivatization prevents both the initial adhesion (CD50:CD1 la/CD18) necessary for allorecognition of foreign tissues via cell- cell interaction as well as the stronger MHC- mediated recognition and costimulatory events. As shown by the stippled areas, multiple adhesion, recognition, and stimulatory pathways are blocked following pegylation. Modified from Chen and Scott, 2001. 21 Table 1.1: Immunocamouflage of Cellular Epitopes Epitope CmPEG Derivatization Concentration (mM per 5.5 x 107 Splenocytes) 0 mM 1.2 mM 2.4 mM CD3e 83% 42% (49%) 25% (70%) TCR 69% 29% (58%) 12% (83%) CDlla 96% •73% (24%) 61% (36%) CD2 77% 72% (6%) 67% (13%) MHC Class II 51% 32% (37%) 15% (71%) CD4 41% 35% (15%) 9% (78%) CD25 10% 6% (40%) 5% (50%) CD28 4% 3% (25%) 1% (75%) CD80 4% 3% (25%) 1% (75%) The values given represent the percent positive splenocytes detected via flow cytometry using epitope specific antibodies at the indicated cmPEG derivatization concentrations. The values inside the parentheses are the average percent decrease in epitope detection relative to the unmodified control splenocytes. The table values represent the average value for a minimum of three separate experiments. Modified from Chen and Scott, 2001. 1.4.2 Pegylation Theory The immunocamouflage effects discussed in section 1.4.1 are based on the physical and chemical properties of methoxypoly(ethylene glycol) [mPEG]. This compound is nontoxic, water-soluble and is approved for use in humans by the United States Food and Drug Administration (Herold, et al, 1989). The polyethylene glycol (PEG) chain itself is highly 22 hydroscopic and therefore in solution PEG is well hydrated (Hermans, 1982). Multiple linker chemistries exist (e.g., cyanuric chloride, benzotriazole carbonate, or succinimidyl propionate as used in this thesis) which are used to covalently graft the mPEG to primary amines such as the epsilon amino group of lysine residues as well as N-terminal amines within proteins. To produce an activated PEG (i.e., a polymer capable of covalently binding to a substrate) the chemical linker is attached to a terminal hydroxyl group (OH) of the polyethylene glycol chain Because the remaining hydroxyl group of PEG can confer some residual reactivity, it is replaced by a methyl group (CH3) to produce methoxypoly(ethylene glycol) [mPEG]. While theoretically it is possible for any appropriately charged amino group to be modified by the chemistries described, lysine residues are the primary site of mPEG binding, known as pegylation. This has been clearly shown in the extensive literature on protein pegylation as exemplified by the study of Bailon, et al. (2001). Despite the ability of mPEG to react with primary amines, when interferon was modified with a 40 kDa branched mPEG the N-terminal cysteine was not a pegylation site. According to amino acid sequencing and mass spectrometric analysis only lysines 31, 70, 83, 121, 131 and 134 were sites of modification (Bailon, et al, 2001). Due to the importance of the covalent bond formed between activated mPEG and lysine residues to the work in this thesis, the structure of lysine is depicted in Figure 1.9 and the reaction schemes for each linker moiety are shown in Figures 1.10-1.12. As shown in Figure 1.9, the a- and 8 - amino groups are positively charged under neutral or acidic conditions. The 23 pK a for the a - group is 9.2 and 10.5 for the e- group. Importantly, the amino group reacts with the activated mPEGs when it is in the basic, NH2 state (Jackson et al, 1986). Acidic/Neutral ^ ° + H 3 N — — C H H— — C H H— — C H H— — C H H— — C H 1 e" + N H 3 L y s i n e oc-amino group • pK a=9.2 O Basic H 2 N-H-H-pK a = 10.5 •H -H H C H H C H •H NH; Figure 1.9: Structure of Lysine. As illustrated above, under acidic or neutral conditions the a and e amino groups of the lysine residues exist in the ionized form (N l iV ) . Under basic conditions, like those used during the pegylation reaction, these groups remain in the NH2 state. Note that the a amino group has a pK a of 9.2 while that of the e group is 10.5 (Modified from Zubay, 1998). 24 Protein^ H2N C H H C H H-H-H-•C H •C H •C H NH 2 H 3C(0-CH 2-CH 2)n-0. Cyanuric Chloride activated mPEG H 3C(0-CH 2-CH 2) n-O s .N CI 3 J\ CI 6 5 NH H C H H C H H C H H C H H2N C H O Protein + HC1 H 3C(0-CH 2-CH 2) n-O s .N CI 6 5, Hydrolysis + H 2 O CI H 3C(0-CH 2-CH 2) n-0 N + HC1 .N. XI 6 5, OH Figure 1 . 1 0 : CmPEG-Lysine Reaction Scheme. The amino group acts as a nucleophile to displace the chlorine atom (CI) attached to the carbon in position 4 in CmPEG. Note the CI attached to carbon 6 may retain some reactivity under conditions used in my experiments. In addition, once the CmPEG is added to the buffer there is a competing hydrolysis reaction in which the chlorine is displaced by a hydroxyl group (OH). 25 P r o t e i n ^ H 2 N C H H C H H C H H C H H C H Benzotriazole Carbonate activated mPEG N H 2 H3co + H 3 C O /n In - N -H - C -H H - C -H H -c-H / \ - O -N H H H N H 2 < Protein H 3 C O O Hydrolysis O—J-C O N In N H 3 C O + H 2 0 O O - j - C OH In / \ HON N Figure 1.11: BTCmPEG-Lysine Reaction Scheme. The amino group acts as a nucleophile to displace the oxygen attached to the benzotriazole ring. Note that there is a competing hydrolysis reaction for BTCmPEG, which more readily occurs under neutral or acidic conditions than at basic pHs. 26 Protein. C ^ H2N C H H C H H- -H H C H H C H NH2 Succinimidyl Proprionate activated mPEG H H C o-+ H 3CO-^-C H H H H O -C C - c -H H -O N H H H H O H H H H H 3CO-^-C- -O-H H H H H H H H H H -C — NH? < Protein O Hydrolysis H H H H / I H 3CO-f-C-H -O-H H + H 2 0 H H H3CO-f^-C C O-H H H H 0 -OH H H Figure 1.12: SPAmPEG-Lysine Reaction Scheme. As with the BTC- and CmPEG reaction, the SPAmPEG reaction occurs between the amino group of lysine, which acts as a nucleophile to displace the single bonded oxygen of the carboxyl group. Note that there is also a competing hydrolysis reaction for SPAmPEG, which more readily occurs under neutral or acidic conditions than at basic pHs. 27 In the case of CmPEG (Figure 1.10), the amino group of the lysine residue displaces the chlorine of cyanuric chloride activated mPEG (CmPEG) through nucleophilic attack, resulting in an amine bond anchoring mPEG to the protein. The competing hydrolysis reaction occurs in approximately 10 minutes; therefore CmPEG is used in excess in order to compensate for the formation of inert hydrolysis products (Jackson et al., 1986). Unfortunately, Jackson et ah, (1986) did not specify the hydrolysis half-life of mPEG at a specific pH. The pH is important in hydrolysis reactions, as the rate of hydrolysis triples when the pH is lowered by one unit [e.g., from 8 to 7] (Shearwater, 2000). It should be noted that despite the unfortunate naming of this compound, cyanuric chloride mPEG does not contain cyanide, nor is cyanide created during the reaction with lysine as shown in Figure 1.10. Also note that CmPEG contains two chlorine atoms that may react with the amino group of lysine; however, Jackson, et al. (1986) clearly demonstrated that under highly basic conditions (pH 9.2), the residual chlorine was not sufficiently active for reaction with a second amino group or hydrolysis. A similar nucleophilic reaction occurs with benzotriazole carbonate mPEG (BTCmPEG). In this reaction, the single bonded oxygen of the ester linkage of BTCmPEG is attacked by the amine group of lysine to form a carbamate [0-C-(C=0)-N] bond as illustrated in Figure 1.11 (Roberts, et ai, 2002). Unlike the CmPEG reaction, the BTCmPEG-lysine reaction produces a sizable leaving group that was a potential toxicity concern. However, peripheral blood mononuclear cell (PBMC) studies to determine propidium iodide exclusion by flow cytometry showed there was no in vivo toxicity (Chen and Scott, 2001). As with CmPEG, there is a 28 competing hydrolysis reaction for BTCmPEG, which has a half-life of 13.5 minutes at pH 8.0 (Shearwater, 2000). Figure 1.12 depicts the succinimidyl propionate mPEG (SPAmPEG) reaction where the ester linkage is attacked by the nucleophilic amino group to form an amide bond. These bonds secure the polymer to the protein until acted upon by enzymatic degradation or protein turnover (Roberts, et al, 2002). As with BTCmPEG, the SPAmPEG reaction with lysine produces a leaving group, which was shown to have no toxicity in a PBMC model (Chen and Scott, 2001). As with the other two linker moieties, SPAmPEG also undergoes a competing hydrolysis reaction in solution. This reaction has a half-life of 16.5 minutes at pH 8.0 (Shearwater, 2000). It is important to note that the competing hydrolysis reactions that occur with each of the linker moieties, occur more readily at neutral or acidic pHs (Shearwater, 2000). Covalently bound mPEG functions in part as a physical barrier in addition to obscuring surface charge both of which block/impair receptor-ligand interactions (Scott, et al, 2000; Bradley et al 2002). Due to mPEG existing as a highly hydrated molecule with free rotation every 4-5 ethoxy units, a gel-like zone of exclusion is formed which inhibits large molecules from passing through (Scott et al 1997; Bradley et al 2001; Bradley et al. 2002). This mPEG barrier varies in density depending upon the surface protein topography of the cell. As shown in Figure 1.13, at low polymer densities, mPEG exists as separate coils known as mushrooms (de Gennes, 1980; Allen, etal. 2002). High density mPEG is said to exist in the brush regime, where the PEG chains extend away from the point of attachment (Carignano and Szleifer, 2000). The PEG in these high density or PEG-rich domains were shown by Satulovsky (2000) 29 A. Low Density Mushroom B. High Density Brush Border Figure 1.13: Mushroom and Brush P E G Regimes Low PEG densities result in a mushroom regime where each polymer exists as a separate coil (A). In contrast, high PEG density results in a stretching of the polymer into the brush regime where the PEG chains interact with each other (B). Due to the heterogeneity of the cell membrane, both regimes are likely to exist following the mPEG reaction. Modified from de Gennes, 1980. to interact via non-covalent attractive forces known as van der Waals forces and interchain hydrogen bonding resulting in entanglement and greater uniformity compared to PEG-poor domains. Due to the heterogeneity of cell surface topography, both regimes are likely to exist following the pegylation reaction. For example, large surface proteins extending well away from the membrane will be weakly immunocamouflaged. In contrast, proteins adjacent to the cell surface are significantly better camouflaged by the grafted mPEG as the density of these proteins is greater, therefore proteins that are not directly modified are likely to be protected by the modification of adjacent proteins (note the area with thick grey lines in Figure 1.14). The size exclusion gradient that is formed in these PEG-rich regions allows small molecules such as water and glucose to freely navigate through to the cell surface while larger molecules such as 30 immunoglobulins and viruses are not able to penetrate as deeply and are therefore blind to the surface antigens/proteins. Relative mPEG Density is also a Function of Cell Topography Figure 1.14: mPEG Zone of Exclusion The density of the mPEG zone of exclusion/ hydration (indicated by gray bars of different thickness) is variable. This zone acts as a molecular sieve allowing small molecules (e.g., glucose) to pass through unencumbered. Meanwhile larger molecules and particles (e.g., virus) and cells are limited in their ability to penetrate the gradient. Molecules of intermediate sizes (e.g., IgG) are able to penetrate further than large molecules, but cannot reach the cell surface. Note the thickness of the grey lines denotes mPEG density, which is maximal midway from the cell surface to the tip of the longest protein, though this will vary depending upon the surface topography. This figure does not adequately address the variation in mPEG density with non-uniform protein distribution. Figure modified from Bradley, et al, 2001. 31 1.4.3 PEG Immunogenicity? Despite significant research and clinical use which suggests that mPEG is non-immunogenic, there is a small body of literature that suggests otherwise (Richter and Akerblom 1983 and 1984; Leger, et al. 2001, Armstrong et al. 2003, Fisher et al. 2003, Garratty, 2004). In the early 80's the first reports of anti-PEG antibodies were published. Richter and Akerblom (1983) were able to raise antibodies in a small proportion of rabbits against pegylated superoxide dismutase and ragweed pollen extract. Mild human anti-PEG antibody responses were also observed following administration of PEG-ragweed extract and honeybee venom. Interestingly, this response was found to dwindle with increased exposure time and was weak enough to be of no clinical significance (Richter and Akerblom 1984). In addition, studies by Cheng, et al. (2000) have also shown that a mouse IgM antibody against PEG caused the rapid clearance of PEG conjugated (3-glucuronidase (|3G) from Balb/c mice without toxicity. This antibody, AGP3, was generated against succinimidyl propionate mPEG modified (3G and has been made commercially available (Tsai, et al. 2001). Their data suggested that the AGP3 antibody did not bind the linker moiety used for conjugation as a tresylated mPEG modified protein was also recognized by AGP3 (Cheng, et al. 1999). These studies do appear to contradict years of clinical and non-clinical use of PEG in humans. PEG has been used in cosmetics, foods and in the pharmaceutical industry for drug compounding for over 30 years. The first pegylated proteins appeared in the literature in 1977, when Abuchowski, et al. (1977) modified bovine catalase. They found that pegylated catalase was non-immunogenic/antigenic, while maintaining enzyme activity. Since this 32 groundbreaking work, additional proteins have been pegylated for therapeutic uses including: asparaginase, adenosine deaminase, bilirubin oxidase, interferon, interleukin-2, methionase, superoxide dismutase, and thrombin with no detrimental immune response to PEG reported (Park, et al, 1981; Kimura, et al, 1988; Goodson and Katre, 1990; Jackson, et al, 1990; Kita, et al, 1990; Nakagomi and Ajisaka, 1990; Bailon, et al, 2001; Avramis, et al, 2002; Soares, et al, 2002; Hoffman, 2003) - despite often decades of use in patient populations. Even if antibodies are produced, the data of Richter and Akerblom (1984) suggests this would be of little/no clinical concern. 1.5 Viruses and Infection According to the hypothesis of this thesis, the concept of immunocamouflage can be translated into the prevention of viral infection based on the similarities in virus-receptor interactions required for entry and protein-protein interactions required for immunorecognition. Because this thesis is geared toward preventing infection of the nasopharynx by respiratory viruses, the viruses that cause the common cold will be briefly discussed in terms of structure and cell entry. Importantly however, these viruses (rhinovirus, adenovirus, and coronavirus) share characteristics with nearly all viruses, therefore the protective effects of immunocamouflage may be extrapolated to most other virus families. 1.5.1 Structure The generic illness known as the common cold results from infection by one of several hundred viruses. The typical common cold viruses include members of the rhinovirus, adenovirus and 33 coronavirus viral groups. These viruses range in size from approximately 20 to 200 nanometers (nm) in diameter (Fields, 1998). The viral genomes consist of either RNA or DNA, which can be single or double stranded, circular or linear. As shown in Figure 1.15, the outer protective coat of these viruses can be either a naked capsid (also referred to as non-enveloped) composed entirely of viral proteins, or a host-cell derived lipid envelope surrounding the protein capsid (i.e. enveloped). Naked Virus Enveloped Virus Figure 1.15: Virus Structure A naked virus is shown on the left and consists of a protein capsid and the viral genome. Pictured on the more structurally complex enveloped virus, which is composed of a lipid membrane taken from the host cell, envelope proteins, viral capsid and genetic material. As shown in Table 1.2, the common cold viruses are representative ofthe wide range of sizes, structures and genomes in the virus world. The smallest of the common cold viruses are rhinoviruses, which are approximately 30 nm in diameter. The largest cold viruses are coronaviruses, which range in size from 80-200 nm (Fields, 1998; Flint, 2000). In addition to the differences in size, structure and genome, common cold viruses also differ in mode of entry into cells. Rhinoviruses and adenoviruses enter via receptor mediated endocytosis, while coronavirus utilizes receptor-mediated fusion for cell entry (Fields, 1998). Envelope Proteins Lipid Bilayer (Envelope) Genome (RNA or DNA) 34 . Common Cold Viruses -, S " J , - ' 4 ' ' ' ' ' ; ' r ^ ' i i l ' - i ' Virus and Family Receptor Mode of Entry Particle Size Virus Structure Genome Type Rhinovirus Picornaviridae Intercellular Adhesion Molecule-1 (ICAM-1) Receptor Mediated Endocytosis -30 nm Naked Icosahedral Capsid single stranded RNA (ssRNA) Adenovirus Adenoviridae Coxsackie and Adenovirus Receptor (CAR) Receptor Mediated Endocytosis 70-90 nm Naked Icosahedral Capsid Linear double stranded DNA (dsDNA) Coronavirus Coronaviridae Aminopeptidase N (APN) Fusion 80-160 nm Enveloped Helical Capsid ssRNA Table 1.2: Common Cold Viruses. Common cold viruses: rhinovirus, adenovirus and coronavirus are compared based on receptor utilized, mode of entry, diameter, structure and genome type. Note that there is a wide range in viral particle size and structure among common cold viruses. 7.5.2 How Viruses Infect Host Cells The two main modes of viral infection mentioned above are receptor mediated endocytosis, and fusion, both of which heavily rely upon receptor-ligand interactions. As shown in Figure 1.16, during receptor mediated endocytosis viral capsid proteins make highly specific interactions with cell surface receptors (e.g., ICAM-1 for Rhinovirus). These interactions signal the cell to engulf the virus within a vesicle, which in turn is brought into the cell. Once inside the permissive cell, the virus uncoats, early viral proteins are translated, and the genome is replicated. Late in the infection process, viral coat proteins are translated and the progeny virions assemble. 35 Attachment Endocytosis Viral Receptor i N i t f * Vesic le containing Virus Permissive Cells Naked Virus — Figure 1.16: Viral Receptor Mediated Endocytosis Depicted above are the first two stages of viral infection: Attachment, and Endocytosis. During attachment, the viral coat proteins make highly specific interactions with cell surface receptors. These interactions cause cell signaling resulting in endocytosis ofthe virus particle. As with receptor mediated endocytosis, the process of viral fusion is deeply reliant on receptor-ligand interactions. Figure 1.17 depicts the first stages of viral fusion. Virally encoded envelope proteins recognize and bind to cell surface receptors facilitating fusion of the two membranes and the internalization of the viral capsid. The outcome is a single, continuous lipid bilayer containing a segment of viral membrane displaying envelope proteins and an infected cell. The remaining events in infection of enveloped viruses are largely the same as those described for receptor mediated endocytosis. 3 6 At tachment Fus ion Vira l Envelope Prote ins Figure 1.17: Viral Fusion Depicted above are the first two stages of viral infection: Attachment, and Fusion. During attachment, the viral envelope proteins make highly specific interactions with cell surface receptors. These interactions cause cell signaling resulting in fusion of the viral and cell membrane, and entry of the virus particle. 1.6 Human Common Cold Infections Common colds are caused by a multitude of viruses, therefore a general discussion of the immunological response to viral infection is necessary. The innate immune response of interferon-beta and natural killer (NK) cells are the first response to infected cells. Interferon-beta is released from infected fibroblast cells resulting in an 'antiviral state' of nearby cells and activated NK cells. NK cells lyse infected cells through the release of porforin, which creates pores in the cell membrane and granzymes that induce apoptosis (Abbas, 2000). Complement proteins also recognize viruses and recruit additional complement proteins that assemble into protease complexes. The breakdown products generated by these proteases 37 covalently bind the virus and serve as opsonins to promote phagocytosis. Additionally, the final protease complex of the complement cascade cleaves C5 into a potent pro-inflammatory peptide C5a and C5b, which remains attached to the target cell. C5a stimulates an influx of neutrophils while C5b initiates the formation of a complex of proteins (C6-C9) which are assembled into a membrane pore that causes the lysis of infected cells. Adaptive immunity also plays an important role in viral clearance. B cells release antibodies, which block virus binding and entry into host cells. Additionally, cytotoxic T lymphocytes (CTLs) recognize and lyse infected cells. All of these immune responses, as well as the physical responses such as increased body temperature, coughing and sneezing work to resolve viral infections. For example, the upper respiratory tract provides optimal conditions for rhinovirus infection with temperatures ranging from 31-35°C regardless of ambient temperature (Hendley, 1999). Rhinoviruses have been shown to infect human respiratory epithelium however, there is no evidence that cells in the sub-mucosa of the respiratory tract become infected, nor has the virus been detected in the blood (Hendley, 1999; Halperin, et al., 1983). Human infection can occur through inhalation of virus particles or direct contact of virus-contaminated hands with nasal mucosa. Once the virus is present in the nasal cavity it can bind to the cell surface receptor, ICAM-1, enter the cell and begin replicating. Symptoms of the common cold generally appear within 8-10 hours and peak in 1-3 days (Kirkpatrick, 1996). Epithelial cells infected with rhinovirus secrete cytokines, which causes 38 the local vasculature to become leaky. The area of infection becomes inflamed due to the extravasation of serum proteins such as albumin, and polymorphonuclear cells that invade the mucosa (Hendley, 1999; Turner, et al, 1998; Kirkpatrick, 1996). When enough cells are infected, the combined effect from the release of cytokines produces the symptoms associated with the common cold: post-nasal drip, nasal discharge, sneezing, coughing, congestion and headache. The two cytokines released from infected cells, interferon-gamma and interleukin-8, stimulate a cytokine cascade that is responsible for amplifying the immune response. One of the important kinins involved in this cascade is bradykinin, which is responsible for dilating blood vessels in the area and stimulates glandular secretion. Additionally, bradykinin stimulates nerve fibers during the cytokine cascade resulting in sneezing or coughing as well as pain. Studies by Proud, et al. (1988 and 1995) showed a close association between symptoms of a cold and increased levels of bradykinin. In addition to cytokine release, infected cells are replicating and shedding progeny virus in order to propagate the infection. During the peak in cold symptoms, progeny viruses are diluted by secretions and are likely to be swallowed or expunged by coughing or sneezing. Rhinoviruses that are swallowed are unlikely to propagate the infection because of the increased temperature within the core of the body as rhinoviruses are extremely inefficient at replicating when temperatures rise above 33°C. In addition, the acidic pH encountered in the stomach causes the virus to irreversibly disassemble (Hendley, 1999). 39 1.7SV40 Background SV40 was the basis of a majority of the work presented in this thesis therefore it is vital to understand the biology of this particular virus. The most important early gene transcribed during SV40 infection is the large transforming antigen (T antigen). Once translated, T antigen is found in the nucleus and is responsible for nearly every aspect of SV40 replication. It has the ability to bind DNA polymerase alpha, both single and double stranded DNA and RNA, pRB (protein product of the retinoblastoma gene), and p53 among others (Fields, 1998). When concentrations of T antigen are low (e.g., the beginning stages of viral infection), it will bind to its own promoter causing an increase in T antigen transcription. By binding to the promoter, T antigen is able to auto-regulate the messenger RNA (mRNA) and as the concentration increases (e.g., as infection continues), T antigen forms a doughnut-shaped hexamer that binds at the origin of replication on the SV40 genome (Gai, et al, 2004; Jiao and Simmons, 2003; Alexandrov, et al, 2002; VanLoock, et al., 2002; Fields, 1998). It then acts as a helicase to bi-directionally unwind the DNA, allowing the replication of the viral genome. In addition, the ability of T antigen to bind hypophosphorylated pRb results in an alteration in the cell cycle. As shown in Figure 1.18, in an uninfected cell, synthesis phase (S phase) of the cell cycle occurs when cyclin E/cdk 2 phosphorylates pRb, thereby freeing the transcription factor, E2F. In contrast, an SV40 infected cell will enter S phase shortly after infection. Following translation T antigen binds pRb, preventing pRB from interacting with E2F. Therefore, the cell enters the S phase of the cell cycle (Fields, 1998). In addition, T antigen binds and sequesters p53, preventing the induction of p21, thereby preventing cell cycle arrest 40 Normal S-Phase Progression SV40 Infected Cells S-Phase Figure 1.18: Normal versus SV40-Infected Cell Cycle Progression. In an uninfected cell, S phase of the cell cycle occurs when cyclin E/cdk 2 phosphorylates pRb, thereby freeing the transcription factor, E 2 F . In contrast, an SV40 infected cell will enter S phase shortly after infection. Following T antigen translation it binds pRb, preventing pRB from interacting with E 2 F . Therefore, the cell enters the S phase of the cell cycle. 41 in G i or G 2 (Fields, 1998). This combination forces the cell to remain in S phase, allows the viral genome to be copied without restriction and maintains the hostage status of cellular machinery (Lehman, et al. 2000). As the infection continues, late genes encoding the viral coat proteins, viral proteins 1, 2, and 3 (VP1, VP2, VP3) are transcribed. VP1 is the major coat protein of SV40 and is readily detected in the cytoplasm of infected cells (O'Farrell and Goodman, 1976). In preparation for viral packaging, the coat proteins accumulate just outside the nuclear membrane. At the same time, viral DNA begins the process of replication. As the genome is reproduced, it is brought to the waiting coat proteins, and daughter virion particles are formed. These particles migrate to the cell membrane and are released en mass through vacuoles. In time, the vacuoles burst and the newly made SV40 particles are free to infect other cells. 1.8 Summary According to the hypothesis of this thesis, the specific interactions required for receptor mediated endocytosis and cell fusion can be inhibited by the covalent modification of either the virus or host cell with mPEG (Figure 1 . 1 ) . In contrast to existing antiviral prophylaxis, pegylation has multiple benefits. First, whereas the current preventative therapies described (Zicam, sICAM, and RRMA) are virus-specific, pegylation of cells will prevent infection from a broad spectrum of viruses. Secondly, unlike R P v M A (anti-ICAM antibody therapy), mPEG derivatization is non-immunogenic. Thirdly, while drug resistant strains have been described 42 for nearly all antivirals currently in use (e.g., nucleoside analogs, non-nucleoside reverse transcriptase inhibitors, Amantadine, etc.), due to the multivalent nature of pegylation, modification of cells and/or virus is unlikely to cause resistance. Finally, unlike many existing agents pegylation of cells is unlikely to exhibit any significant toxicity at either the site of application or systemically. 43 Chapter 2: Materials and Methods 2.1 Choice of Model Viruses As shown in Table 2.1, the virus models used in this thesis are representative of the wide range of viruses causing the common cold. The smallest of the model viruses is Theiler's murine encephalomyelitis virus (TMEV) which, like rhinovirus, is approximately 20-30 nm in diameter. The largest model viruses are rat coronavirus (RCV) and cytomegalovirus (CMV), which range in size from 80-200 nm, a similar size range to that of human coronaviruses (Fields, 1998). In addition to the differences in size, structure and genome, common cold viruses also differ in mode of entry into cells. Rhinoviruses and adenoviruses enter via receptor mediated endocytosis, while coronavirus utilizes fusion for cell entry. Again, the model viruses infect cells utilizing these same modes of entry. Simian virus 40 (SV40), mouse adenovirus (MAV), and T M E V enter via receptor mediated endocytosis, while RCV and C M V fuse with the cell to gain entry (Fields, 1998). It should be noted that the majority of the work for this thesis focused on SV40. SV40 was chosen as the initial model for human adenovirus infection based on similarities in size, genome and mode of entry. 44 Table 2.1: Comparison of Common Cold Viruses to Model Systems. Virus and Family Receptor Mode of Entry Particle Size Virus Structure Genome Type Common Cold Viruses Rhinovirus Picornaviridae Intercellular Adhesion Molecule-1 (ICAM-1) Receptor Mediated Endocytosis -30 nm Naked Icosahedral Capsid single stranded RNA (ssRNA) Adenovirus Adenoviridae Coxsackie and Adenovirus Receptor (CAR) Receptor Mediated Endocytosis 70-90 nm Naked Icosahedral Capsid Linear double stranded DNA (dsDNA) Coronavirus Coronaviridae Aminopeptidase N (APN) Fusion 80-160 nm Enveloped Helical Capsid ssRNA Viral Models SV40 Papovaviridae Major Histocompatibility Molecule-1 (MHC-1) Receptor Mediated Endocytosis -45 nm Naked Icosahedral Capsid Circular dsDNA MAV Adenoviridae murine homologue of Coxsackie and Adenovirus Receptor (mCAR) Receptor Mediated Endocytosis 70-90 nm Naked Icosahedral Capsid Linear dsDNA RCV Coronaviridae Not yet identified Fusion 80-160 nm Enveloped Helical Capsid ssRNA CMV Herpesviridae Epidermal Growth Factor Receptor (EGFR) Fusion -200 nm Enveloped Icosahedral Capsid Linear dsDNA TMEV Picornaviridae Not yet identified Receptor Mediated Endocytosis 20-30 nm Naked Icosahedral Capsid ssRNA Common cold viruses (Rhinovirus, Adenovirus and Coronavirus) are compared to each ofthe experimental viruses used. They are: SV40 (Simian Virus 40), M A V (Mouse Adenovirus), RCV (Rat Coronavirus), C M V (Cytomegalovirus), and T M E V (Theiler's Murine Encephalomyelitis Virus). Note that there is a wide range in viral particle size and structure for our experimental viruses as there are among common cold viruses. 45 2.2 Materials 2.2.1 Cell Lines and Viruses All cell lines (Balb/3T3, BHK-21, CV-1, L2, MRC-5, and pAb 101) were purchased from the American Tissue Culture Collection (Manassas, V A , USA). Each of these cell lines are either fibroblast or epithelial in nature and have been derived from several animal species. The specific cell lineages are described in Table 2.2. SV40 (SV40) samples were the gift of Dr. John Lehman (East Carolina University, NC, USA) and cytomegalovirus samples were the gift of Dr. Edward Pryzdial (Canadian Blood Services/ University of British Columbia, Vancouver, BC, Canada). Mouse adenovirus, rat coronavirus and Theiler's murine encephalomyelitis virus were obtained from American Tissue Culture Collection. Table 2.2: Description of each cell line and the corresponding virus. Target Cell Line Cell Lineage Cell Morphology Virus Virus Abbreviation CV-1 (ATCC CCL-70) Monkey Kidney Fibroblast/ Epithelial Simian Virus 40 SV40 Balb/3T3 (ATCC CCL-163) Mouse Embryo Fibroblast Mouse Adenovirus M A V L2 (ATCC CCL-149) Rat Lung Epithelial Rat Coronavirus RCV MRC-5 (ATCC CCL-171) Human Lung Fibroblast Cytomegalovirus C M V BHK-21 (ATCC CCL-10) Hamster Kidney Fibroblast Theiler's Murine Encephalomyelitis Virus T M E V 46 2.2.2 Antibodies Mouse anti-VP 1 antibodies used in immunostaining for viral coat proteins were the gift of Dr. John Lehman. Mouse anti-SV40 Large T antigen antibodies were collected from pAb 101 hybridoma cell line obtained from the A T C C and were used as the primary antibody in immunostaining to detect SV40 infection of cells. Alexa-488 goat anti-mouse antibodies used as the secondary antibody for all immunostains were obtained from Molecular Probes (Burlington, ON, Canada). Mouse anti-CMV IE72 antibodies used to detect C M V infection of cells were purchased from Accurate Scientific (Westbury, NY, USA). 2.2.3 mPEG Species and other Chemicals Three forms of activated methoxypoly(ethylene glycol) [mPEG] were used throughout this thesis. Cyanuric chloride activated methoxypoly(ethylene glycol) was obtained from Sigma-Aldrich (Oakville, ON, Canada). Benzotriazole carbonate- mPEG (BTCmPEG), and succinimidyl propionate- mPEG (SPAmPEG) were purchased from Nektar Therapeutics (formerly Shearwater Polymers; Huntsville, A L , USA). CmPEG was only commercially available in the 5 kDa molecular weight whereas BTCmPEG was available in 3.4, 5, and 20 kDa and SPAmPEG in 2, 5 and 20 kDa molecular weights. Matrigel basement membrane and bovine submaxillary mucin were acquired from BD Biosciences (Mississauga, ON, Canada). 47 2.3 mPEG Modification of Virus and Cells 2.3.1 mPEG Modification of Virus Viral stocks were combined with either activated or unactived (non-covalent) mPEG in PBS.at a pH of 8.0. Following a thorough mixing, the reaction was incubated at room temperature for 30 minutes. In order to then rid the sample of unmodified virus, an aqueous polymer two-phase system of PEG and Dextran was used (Mizouni 2000). PEG 8 kDa (43%, Sigma) was layered over 5%> Dextran T500 (Pharmacia) in 150 mM NaCl and 10 mM sodium phosphate buffer, and the viral sample was added. The layers were mixed and allowed to separate for 1 hour at room temperature. In this two-phase system, unmodified virus had a greater affinity for the Dextran phase and interface while modified virus separated into the PEG phase. This system was sensitive to changes in pH, salt concentration, and temperature but does not involve shearing forces or extreme pH, which could cause the virus coat proteins to disassemble. The only virus pegylated for this work was S V40, whose capsid is primarily composed of viral protein 1 (VP1). The capsid of SV40 is composed of 72 VP1 pentamers each of which contain 25 lysine residues (Fields, 1998). Therefore there are many potential sites on the virus particle for mPEG modification (Table 2.3). 48 Table 2.3: Protein Sequence of VP1 M K M A P T K R K G SCPGAAPKKP K E P V Q V P K L V IKGGIEVLGV KTGVDSFTEV ECFLNPQMGN PDEHQKGLSK SLAAEKQFTD DSPDKEQLPC YSVARIPLPN LNEDLTCGNI L M W E A V T V K T EVIGVTAMLN 1.11SGTQKTIII: NGAGKP1QGS NFHFFAVGGE P L E L Q G Y L A N Y R T K Y P A Q T V TPKNATVDSQ Q M N T D H K A V L D K D N A Y P V E C WVPDPSKNEN T R Y F G T Y T G G ENVPPVLFIIT N T A T T V L L D E Q G V G P L C K A D SLYVSAVDIC GLFTNTSGTQ QWKGLPRYFK ITLRKRSVKN PYIMSI I.I.SI) LI.NR.RTQRVD GQPMIGMSSQ V E E V R V Y E D T EELPGDPDMI RYIDEFGQTT TRMQ As noted in sequence above, the major viral coat protein contains 25 lysine (K) residues resulting in an abundance of potential sites for the covalent attachment of mPEG species utilized in this thesis. Protein ID: NP043126 2.3.2 mPEG Modification of Viral Target Cells mPEG derivatization of target cells (CV-1, BHK-21, L2, Balb/3T3, and MRC-5 cell lines) was performed as previously described, with slight modifications due to cell type (Scott, et al. 1997 Scott and Murad 1998; Murad, et al. 1999A and B). Cells were split into either 35 or 60 mm petri dishes and were grown until 75-100% confluent (approximately 3 x 104 cells per cover slip or 3 xlO 5 cells per 35 mm petri dish). The cells were briefly washed with phosphate buffered saline (PBS, pH 7.4), and then overlaid with 1 mL of the mPEG containing solution (PBS, pH 8.4), which was made fresh just prior to the reaction. The reaction was incubated at room temperature for 30 minutes. Following this incubation, the cells were washed with complete media to remove any excess mPEG prior to viral infection. Additional complexity was added to the mPEG reaction with host cells in order to better simulate an in vivo environment. To examine the effects of mucus on the derivatization 49 reaction, bovine sub-maxillary mucin was added to cells and incubated for 30 minutes prior to pegylation. The reaction was carried out in the manner described above. Further aspects of the mPEG reaction were also examined, including: pH, time, and temperature. The pH of the mPEG overlay was lowered from 8.4 to a physiological 7.8 to determine if the lower pH provided equivalent protection from viral infection. Due to the chemistry of the mPEG reactions discussed above, high pH is optimal, however in my initial studies cells had a tendency to peel off the cover slips when high pH values were used (e.g., 9.2). Therefore a balance between optimal reaction conditions and cell viability had to be achieved. Importantly, there was no significant difference in cell viability between pH 8.4 and 7.8, however it was thought that a pH of 7.8 would be less stressful for cells of the nasopharynx than 8.4. The mPEG reaction time was also altered in several studies to determine the optimal amount of time required for maximal antiviral protection as well as optimal human compliance (for a final nasal spray). These mPEG derivatization times ranged from 1-60 minutes. Additionally, the effects of temperature on the derivatization reaction were observed. The initial reactions were carried out at room temperature and were subsequently altered to 32°C, the mean temperature ofthe human nasal cavity regardless of ambient temperature. Each of these reaction parameters were altered in a stepwise fashion to determine individual as well as combined effects on the levels of antiviral protection following the mPEG reaction. Similarly to the SV40 major coat protein, the known cell surface receptors for each of the viruses used in this thesis have a number of lysine residues available for modification with mPEG, as shown in Table 2.4. It should be noted that mPEG does not specifically target these 50 receptors and will covalently bind any lysine residue of any protein. This non-specific protein binding results in the physical mPEG barrier described in Chapter 1. Table 2.4: Protein Sequences of Virus Receptors A. MHC-1 L T K T W A G S H S L K Y I H T S V S R PGRGEPRF1S V G Y Y D D T Q F V R F D S D A A S P R M Q P R A P W V E Q H G P E Y W D Q F . T R S A R D T A Q T F R V N L N T L R G Y Y N Q S E G G S H T L Q W M Y G C D I G P D G R F L R G Y E Q F A Y D G K D Y L T L N E D L R S W S A V D T A A Q I S E Q K S N D G S E A E H Q R A Y L E D T C V E W L R R Y L E N G K E T L Q R S E P P K T B. C A R M A R L L C F V L L C G 1 A D F T S G L S I T T P E Q R I E K A K G E T A Y L P C K F T L S P E D Q G P L D I E W L IS P S D N Q I V D Q V 1 I I Y S G D K I Y D N Y Y P D E K G R V H F T S N D V K S G D A S I N V T M L Q L S D 1 G T Y Q C K Y K K A P G V A N K K F L L T Y L V K P S G T R C F Y D G S E E I G N D F K L K C E P K E G S L P L Q I F W Q K L S D S Q T M P T P W L A E M T S P V I S V K N A S S E Y S G T Y S C T V Q N R V G S D Q C M L R L D V V P P S N R A G T I A G A V I G T L L A L V L I G A I L F C C H R K R R E E K Y E K E V H H D I R E D V P P P K S R T S T A R S Y I G S N H S S E G S M S P S N M E G Y S K T Q Y N Q V P S E D F E R A P Q S P T L A P A K F K Y A Y K T D G I T V V C. E G F R M R P S G T A R T T E I . V E L T A L C A A G G A E E E K K V C Q G T S N R L T Q L G T F E D H F L S L Q R M Y N N C E V V L G N L E 1 T Y V Q R N Y D L S F L K T I Q E V A G Y V L I A L N T V E R 1 P L E N L Q I I R G N A L Y E N T Y A L A I L S N Y G T N R T G L R E L P M R N L Q E I L 1 G A V R F S N N P I L C N M D T I Q W R D I V Q N V F M S N M S M D L Q S H P S S C P K C D P S C P N G S C W G G G E E N C Q K L T K 1 I C A Q Q C S H R C R G R S P S D C C H N Q C A A G C T G P R E S D C L V C Q K F Q D E A T C K D T C P P L M I . Y N P T T Y Q M D V N P E G K Y S F G A T C V K K C P R N Y V V T D H G S C V R A C G P D Y Y E V E E D G I R K C ' K K C D G P C R K V C N G I G I G E F K D T L S I N A T N ' I K H F K Y C T A I S G D I . H I L P V A F K G D S F T R T P P L D P R E L E 1 L K T V K E I T G F L L I Q A W P D N W T D L H A F E N L E I I R G R T K Q H G Q F S L A V V G L N I T S L G L R S L K E I S I ) G D V 1 1 S G N R N L C Y A N T I N W K K L F G T P N Q K T K 1 M N N R A E K D C K A V N H V C N P L C S S E G C W G P E P R D C V S C Q N V S R G R E C V E K C N 1 L E G E P R E F V E N S E C I Q C H P E C L P Q A M N 1 T C T G R G P D N C 1 Q C A H Y 1 D G P H C V K T C P A Q I M G E N N T L V W K Y A D A N N V C H L C H A N C T Y G C A G P G L Q G C E V W P S G Y V Q W Q W I L K T F W I Protein sequences for: (A) MHC-1 (Cercopithecus aethiops) [Protein I D : AAL34325], (B) Coxsackie and Mouse Adenovirus Receptor (Mus musculus) [Protein I D : AAH16457] and (C) Epidermal Growth Factor Receptor (Mus musculus) [Protein I D : AAH23729] are shown. Each viral receptor has several lysine residues ( K ) that can be modified by mPEG. However, it should be noted that other cell surface proteins are also modified during pegylation resulting in the mPEG barrier. 51 2.4 Measuring the Efficacy of Immunocamouflage Against Viral Infection 2.4.1 Immunostaining To determine the effects of mPEG modification of cells on viral infection, either the CV-1 or MRC-5 cells lines were used with SV40 or C M V , respectively, using the experimental scheme illustrated in Figure 2.1. Following the pegylation reaction described above, infectious dose 50 percent (ID50 at 24 hours; ~ 105 pfu/mL) of virus, was added to each chamber and the cover slips were incubated for 2 hours at 37°C with 5% CO2 to allow for viral attachment. After this time, the virus was removed and 1 mL minimal essential media with 5% fetal bovine serum (Invitrogen) was added. The cover slips were incubated at 37°C with 5% CO2 for 24 to 72 hours. At each 24 hour time point, the cover slips were removed from the media and were briefly washed in PBS (pH 7.4). The cells were then permeablized in methanol (-20° C) for 15 minutes. Next, the cover slips were immunostained for SV40 large T antigen or C M V IE72 in order to detect infected cells (Lehman, et al. 1993). After permeablization of the cells, the cover slips were allowed to dry before being washed three times with PBS (pH 7.4). Next, mouse anti-T antigen antibodies or mouse anti-IE72 antibodies were added. The cover slips were then incubated for 30 minutes at room temperature and were again washed three times with PBS (pH 7.4). The secondary antibody, Alexa488 goat anti-mouse antibody (35 uL of the following dilution: 250 uL antibody into 9.75 mL PBS, pH 7.5), was then added to each cover slip. Again, the cover slips were incubated at room temperature for 30 minutes. Following this final incubation, the cover slips were washed 52 three times with PBS (pH 7.4) and three times with distilled water to remove any residual salts. Once the cover slips were dry, they were mounted onto slides with glycerol. Cover Slip Experiment Figure 2.1: Cover Slip Experiments. Cells are grown on cover slips treated with Matrigel Basement Membrane until confluent (A). The appropriate concentration of mPEG is added to each dish containing two cover slips and incubated for 1-60 minutes at 4°C, 32°C, 37°C, or room temperature (B). Virus is then added and incubated for 1-2 hours (dependent upon the virus) before removal and addition of media. Modified cells are hypothesized to be protected from viral infection via the mPEG barrier, which physically blocks virus-receptor interactions (C). Control cells are treated with PBS or unactivated mPEG and are therefore vulnerable to viral infection (D). Because the anti-T antigen antibodies were harvested from a hybridoma cell line, each stock vial had to be titrated prior to use in experiments. Generally the anti-T antigen antibody stocks were diluted 1:2 to 1:12 in PBS (pH 7.4) and tested against the secondary antibody on both 53 SV40 infected and control CV-1 cells. The dilution offering minimal background fluorescence as observed on the control cells and maximal brightness on infected cells was chosen. Five random fields on each cover slip were counted at 20x magnification, for a total of ten counts per mPEG concentration and time point. Fields were first observed under phase contrast for a total cell count. Each field was then viewed under U V light, for the infected cell count. The first field was selected at the far left of the cover slip followed by the field in the center of the cover slip and the far right, followed by the upper and lower center ofthe cover slip forming a cross when the fields are connected. 2.4.2 Plaque Assay Because antibodies were not commercially available for several of the viruses, plaque assays were performed to determine the effects of pegylation on viral infection. Following pegylation, monolayers of CV-1, Balb/3T3, L2, or BHK-21 cells grown in 35 mm petri dishes were washed with serum free media (1 mL each). Viral stocks (SV40, M A V , R C V or T M E V respectively) were diluted 10"3-10"10 and 1 mL of the appropriate dilution was added to each plate. Dishes were incubated at 37°C and were rocked every 10-15 minutes for 1-2 hours depending upon the virus used. A mixture containing equal parts 0.6% melted Bacto-Agar cooled to 45°C and 2x M E M (without phenol red, Gibco) with 10% FBS was added and allowed to cool for 15-20 minutes in order for the agar to solidify. The plates were then incubated at 37°C for 3-10 days depending upon the virus, the agar was removed, cells were treated with ice cold methanol for 15 minutes and stained with neutral red (Tollefson, et al. 1998; Tremblay, et al. 2001; 54 Wentworth and French 1970; Chou and Scott 1988; Musiani, et al. 1988). The experimental design is illustrated in Figure 2.2. Plaq ue Assay Experiment B Virus is added • Figure 2.2: Plaque Assay Experiments. Cells are grown on petri dishes treated with Matrigel Basement Membrane until confluent (A). The appropriate concentration of mPEG is added to each dish and incubated for 30 minutes at room temperature (B). Virus is then added and incubated for 1-2 hours (dependent upon the virus) before removal and addition of media overlay (C). Plates are stained with neutral red and plaques are counted (D, F). Control cells are treated with PBS or unactivated mPEG and as hypothesized are vulnerable to viral entry/infection (E) leading to a larger number of plaques (F). 5 5 2.4.3 Detection of SV40 Entry: Immunostaining for VP1 In order to determine if viral particles were able to enter modified target cells during early stages of infection, VP1 staining was conducted. VP1 was the major capsid protein of SV40 and was not expressed until late stages of infection, therefore any VP1 observed within the first 12 hours of SV40 exposure would arise from the presence of intact viral particles subsequent to cell entry. As shown in Figure 2.3, cells challenged with SV40 showed no signs of VP1 production after 6 hours. However, by 24 hours post viral challenge VP1 was prevalent in the cytoplasm of infected cells. In contrast, at 6 hours cells stained positive for T antigen (an early protein), which persisted throughout the viability of the cells (beyond 96 hours). For this reason, T antigen staining was used to determine the number of infected cells for the majority of SV40 experiments, while VP1 staining was only used in the entry studies. During the short incubation periods (1-3 hours) there would not be sufficient time for an infected cell to produce new VP1 protein. As shown in Figure 2.3, no VP1 is detectable within 6 hours since this protein is only synthesized late in the infection. Therefore during short incubation periods post viral challenge, any VP1 detected is the result of recent SV40 entry. " 56 A. VP1 Staining 6 hours Post Challenge B. VP1 Staining 24 hours Post Challenge C. T Antigen Staining 6 hours Post D. T Antigen Staining 24 hours Post Challenge Challenge Figure 2.3: VP1 versus T antigen Staining. Cells challenged with SV40 showed no signs of VP1 production after 6 hours (A). However, by 24 hours post viral challenge (B) VP1 was prevalent in the cytoplasm of infected cells. In contrast, at 6 (C) and 24 (D) hours cells stained positive for T antigen. To measure viral entry, VP1 immunostaining was performed 1-3 hours post viral challenge. Confluent cover slips were pegylated with 0 or 5 mM CmPEG for 30 minutes at room temperature. SV40 was added to each dish (2x ID50, ID50, and 10"1 ID50) and incubated for 1-3 hours. At each time point, the petri dishes were removed from the incubator, virus was removed, cells were washed twice with MEM 5% FBS, and three times with PBS (pH 7.4). To ensure the VP1 detected was the result of viral infection rather than being washed in when the 5 7 cells were permeablized with methanol, the cover slips were transferred to clean petri dishes and washed three more times with PBS. Cells were then permeablized by addition of 1 mL of ice-cold methanol to each petri dish for 15 minutes. The cover slips were removed from the methanol and allowed to dry. They were then washed three times with PBS (pH 7.4) prior to addition of mouse anti- VP1 antibodies. The cover slips were incubated for 30 minutes at room temperature followed by washing with PBS (pH 7.4). The secondary antibody, Alexa488 goat anti-mouse antibody was added and incubated for 30 minutes at room temperature. Again the cover slips were washed in PBS (pH 7.4) followed by distilled water to remove salts. The cover slips were allowed to dry at room temperature before mounting onto slides with glycerol. 2.5 Cell Maintenance and Viral Stock (Lysate) Production 2.5.1 Cell Culture The CV-1, MRC-5 and BHK-21cell lines were maintained in minimal essential media (MEM, Gibco-Invitrogen, Burlington, ON, Canada) with 10% fetal bovine serum (FBS, Invitrogen) plus 200 U/mL penicillin and 200 pg/mL streptomycin (Gibco-Invitrogen). The Balb/3T3 cell line was similarly maintained in Dulbecco's modified Eagle's media (DMEM with bovine calf serum rather than FBS). L2 cells were cultured in Ham's F12K media with 10% FBS. All cultures were incubated at 37°C with 5% C O 2 . They were subcultured in a ratio from 1:2 to 1:10 using 0.25% (w/v) Trypsin-0.03% (w/v) EDTA (Gibco-Invitrogen) to re-suspend cells. Matrigel Basement Membrane (BD Biosciences) was used during experiments to provide cells with better anchorage on glass cover slips. It was diluted 1:11 in serum free medium, and 1 mL was added to each petri dish in order to coat the cover slips with a thin layer. The petri dishes 58 were incubated at 37°C for at least 30 minutes to solidify the gel, and any remaining liquid was removed prior to addition of cells. Anti-T antigen antibody producing hybridoma cell line, pAb 101 (ATCC TIB-117), were maintained in Dulbecco's modified Eagle's media (Gibco-Invitrogen) with 20% FBS plus 200 U/mL penicillin and 200 pg/mL streptomycin (Gibco). Cultures were incubated at 37°C with 5% CO2. Cells were subcultured when they reached a density of 5x 105 cells per mL. 2.5.2 Virus Preparation (Lysate Production) Cells were grown until 75-100% confluent (depending on the virus used) in 2 L roller bottles with a constant rolling motion or in 150 cm 2 petri dishes at 37°C. Viral stock was added with a multiplicity of infection (MOI) between 0.1 and 0.0001. The cells were incubated for 1-2 hours (virus specific) at 37°C with the viral solution, at which time, the solution was removed and fresh media containing 2% FBS was added. Cells were incubated for 4-14 days or until cytopathic effects develop to the point of complete destruction of the monolayer prior to viral harvest. For SV40 after ~10 days, the lysate was placed in conical tubes and frozen at -80°C followed by rapid thawing in a 37°C water bath. This freeze/ thaw cycle was repeated for a total of three times in order to disrupt any intact cellular membranes that remain, thereby freeing a large number of viruses, which were trapped in vesicles. The infected media was then centrifuged for 10 minutes at 4°C, 400 x g to remove cell debris. The supernatant was then layered over a 59 saturated potassium bromide gradient and centrifuged for 6 hours at 35,000 x g. The virus was collected and dialyzed against PBS overnight and stored at -80°C (Mizouni 2000). With M A V infection, following ~5 days the virus-rich media was placed in conical tubes and frozen at -80°C followed by rapid thawing in a 37°C water bath. This freeze/ thaw cycle was repeated for a total of three times in order to disrupt any intact cellular membranes that remain, thereby freeing a large number of viruses (Temple, etal. 1981; Larsen and Nathan 1977). Lysates were then sonicated for 30 seconds and centrifuged for 10 minutes at 800 x g to remove cellular debris. Thirty milliliters of this extract were then layered over a gradient composed of 3 mL CsCl 1.5 g/cm3 and 5 mL CsCl 1.2 g/cm3 (0.02 M Tris HC1, pH 8.7). This was centrifuged at 40,000 x g for 90 minutes at 4°C. The recovered virus particles were then dialyzed overnight against 1000 times the volume of buffer (10 mM Tris HC1, pH 8.0, 0.25 M NaCl, 1 mM EDTA, 10 mM MgCl 2) and stored at -80°C (Temple, et al. 1981). In the case of RCV, after 7-14 days virus was purified using a discontinuous 60% and 30% (w/w) sucrose in TNE buffer (50 mM Tris HC1, pH 7.5; 100 mM NaCl; 1 mM EDTA) and centrifuged at 110,000 x g for 2 hours at 4°C. R C V was recovered from the interface and stored at -80°C (Gagneten, et al. 1996; Kunita, et al. 1993). C M V infected cell culture media was collected and the cells scraped from the roller bottle or petri dishes. This crude lysate was centrifuged at 1,500 x g at 4°C for 10 minutes. The cell free media was layered onto potassium-tartrate-glycerol gradient prepared in 0.05 M Tris HC1 pH 60 7.4, 0.10 M NaCl (TN buffer) and was centrifuged for 15 minutes at 90,000 x g. The banded virus was collected and stored at -80°C (Irmiere and Gibson, 1983; Talbot and Almeida 1977). Following ~4 days of incubation with T M E V , the lysate was placed in conical tubes and frozen at -80°C followed by rapid thawing in a 37°C water bath. This freeze/thaw cycle was repeated for a total of three times in order to disrupt any intact cellular membranes that remained, thereby maximizing the release of viruses. T M E V was purified by centrifuging the crude lysate for 10 minutes at 400 x g to remove the cellular debris. The lysate was then layered over a 20- 70% continuous sucrose gradient and was centrifuged at 210,000 x g for 3 hours at 4°C. The purified virus was stored in the -80°C freezer (Rhozon, et al. 1983). 2.6 Statistical Analysis The standard deviation for all experiments was calculated using the pooled estimate of the 2 2 2 variance where s= square root [(ni-l)si + (n2-l)s2 +( n3-l)s3 ]/(nj + n 2 + n3-3)] (Rosner, 1995). One or two-way analysis of variance (ANOVA) were performed using SPSS version 11 for Macintosh. Data found to be statistically significant (p < 0.05) were then subjected to a Tukey analysis. Analysis of data from three independent experiments was performed to determine whether the mPEG treatments resulted in significantly lower rates of infection than controls. A final analysis was conducted comparing the effectiveness of the three linker chemistries (CmPEG 5 kDa versus BTCmPEG 5 kDa and SPAmPEG) or molecular weight mPEGs of the same species (i.e. SPAmPEG 2 kDa versus SPAmPEG 5 kDa). Tukey's method was chosen because sample sizes from each population were equal and pair-wise comparisons of the mean infection rates were the primary interest (Kleinbaum and Kupper, 1978). 61 Chapter 3: Pegylation of SV40 Viral Particles Prevents Infection :7 ''I Chapter Overview ' '''^':\^':^v-.' • • Modification of the: S V40 Virus by CmPEG, BTCmPEG, of SP^PEG^j^ai-Whiai^...'' effect does linker chemistry have on preventing infection? ; & . • SV40. Modification by Different Molecular. Weight mPEGs: What effect does polymer size have? • Combinatorial Studies: Can additional protection against viral infection be gained by using two different molecular weight mPEGs? 3.1 Modification ofSV40 Particles with CmPEG, BTCmPEG or SPAmPEG 5 kDa Prevents Infection As described in Chapter 2, Simian Virus 40 (SV40) is a relatively small (45 nm diameter) DNA virus with a capsid composed primarily of viral protein 1 (VP1). Previous work on pegylation of SV40 demonstrated that it was possible to modify the virus with CmPEG 5 kDa, which resulted in decreased infection of CV-1 cells (Mizouni, 1998, 1999 and 2000). This work was replicated in order to provide a direct comparison to my additional work with other forms of mPEG. In the current studies, virus was modified at a pH of 8.0 for 30 minutes with concentrations of CmPEG 5 kDa ranging from 0- 15 mM. Cells were challenged with modified 62 virus and infection was monitored for 72 hours through SV40 T antigen staining. Table 3.1 shows the effects of mPEG modification with three different linker chemistries on SV40 infection. As noted previously, these linker moieties demonstrate comparable efficacies at the 5 kDa molecular weight. As shown, at 24 hours post viral challenge cells exposed to unmodified virus stained positive for T antigen at a rate of 49% and decreased in a dose dependent manner, with 0%> of cells becoming infected when challenged with 15 mM CmPEG modified SV40 (p<0.001). At 48 hours, cells challenged with unmodified virus were 76%o positive, while only 0.05% of cells challenged with modified (15 mM) virus were T antigen positive (p<0.001). Even by 72 hours post challenge, when controls were 99%» T antigen positive, cells challenged with SV40 modified by CmPEG showed a dose dependent decrease in infection from 85% at 0.2 mM to 0.3% infected at 15 mM (pO.OOl). The 72 hour time point demonstrates the extremely potent effect of direct viral pegylation since any progeny virus from successful entry would greatly magnify the infection rate as the progeny virus would not be pegylated. Clearly, the modification of viral particles with CmPEG 5 kDa provided a potent means of preventing viral infection. In order to compare the efficacy of protection from benzotriazole carbonate activated mPEG (BTCmPEG) and succinimidyl propionate activated mPEG (SPAmPEG) to CmPEG similar SV40 studies were performed. As with the CmPEG experiments, virus was modified at a pH of 8.0 for 30 minutes with concentrations of 5 kDa BTCmPEG ranging from 0- 15 mM. Cells were challenged with control and BTCmPEG modified virus and monitored for a total of 72 hours via T antigen staining. As shown in Table 3.1, there are similar dose dependent reductions in the percent cells infected following challenge with B T C - or SPAmPEG at all 63 Table 3.1: Comparison of Protection by 5 kDa mPEGs with Varied Linker Chemistries. 24 Hours 5 kDa 0 mM 0.2mM 0.6mM 1.2mM 2.4mM 5 mM lOmM 15mM CmPEG 49.3 ±1.9 21.43 ±1.0 14.033 ±0.6 10.533 ±1.1 4.733 ±1.5 l . l 3 3 ±1.0 O.I33 ±0.2 O33 ±0 BTCmPEG 51.4 ±2.4 44.3 ±4.8 ** 40.03 ±5.2 ** 32.433 ±4.5 ** 21.433 ±2.0 ** 9.93 3 ±1.6 l.O 3 3 ±0.3 0.233 ±0.1 SPAmPEG 50.6 ±0.1 44.333 ±0.3 ** 41. I 3 3 ±0.6 ** 38.533 ±1.6 ** 31.333 ±0.8 ** 15.633 ±0.6 2.633 ±0.5 0.533 ±0.4 48 1 Hours 5 kDa 0 mM 0.2mM 0.6mM 1.2mM 2.4mM 5 mM lOmM 15mM CmPEG 75.8 ±2.7 45.933 ±2.4 25.033 ±0.8 22.733 ±1.5 11.633 ±1.4 2.633 ±1.6 0.633 ±0.3 0.133 ±0.1 BTCmPEG 78.4 ±2.3 63.2 ±9.7 59.9 ±10.6 ** 54.53 ±12.8 ** 43.23 ±10.0 ** 24.433 ±6.5 8.233 ±3.2 0.733 ±0.3 SPAmPEG 75.3 ±6.0 75.6 ±5.8 ** 65.33 ±1 ** 59.833 ±1.7 ** 55.233 ±1.9 ** 20.933 ±0.4 5.333 ±0.5 1.733 ±0.4 72 Hours 5 kDa 0 mM 0.2mM 0.6m VI 1.2mM 2.4mM 5 mM lOmM 15mM CmPEG 99.1 ±0.2 84.533 ±5.7 36.233 ±0.8 34.933 ±0.4 22.133 ±3.6 6.333 ±3.8 2.533 ±2.2 0.333 ±0.3 BTCmPEG 99.5 ±0.4 80.03 ±3.1 67.333 ±6.3 ** 63.333 ±7.6 * 48.733 ±4.2 * 36.333 ±3.5 * 13.333 ±2.8 2.033 ±1.1 SPAmPEG 99.4 ±0.2 96.9 ±0.9 81.633 ±1.5 ** 70.633 ±1.4 ** 63.033 ±1.6 ** 50.233 ±0.9 ** 10.433 ±0.8 3.833 ±0.5 Data presented in this table represents the average percent cells infected of n=3 independent experiments ± standard deviation (SD). * Significantly different from the equivalent concentration of CmPEG 5 kDa (** p <0.001; * p < 0.05). t Significant differences between BTC and SPAmPEG 5 kDa at equivalent concentrations (f f p < 0.001; f p < 0.05).3 Significantly different from the control for that mPEG (0 mM) at that particular time point (33 p< 0.001,3 p< 0.05). Modification of SV40 with each mPEG species at 5 kDa prevented infection of unmodified cells. BTC- and SPAmPEG were significantly different from each other at any mPEG concentration. However, CmPEG 5 kDa was found to be significantly more effective at preventing infection of modified virus at low mPEG concentrations (0.2-2.4 mM at 24 hours and 0.6-5 mM at 72 hours). 64 three time points. Hence, using three different linker chemistries, comparable antiviral activity was noted with the 5 kDa polymer. 3.2 SV40 Modification Using Different Molecular Weight mPEGs Prevents Infection of Cells As previously reported by Bradley, et al. (2002) polymer size is an important factor in the ability of mPEG to effectively immunocamouflage red blood cells. Their work demonstrated that a larger polymer size was more effective in preventing allorecognition of surface proteins in addition to better obscuring surface charge. To further delineate the ability of mPEG modification of viruses to prevent infection, the effect of both decreased (2 kDa) and increased (20 kDa) polymer size was explored. As with the experiments performed in the previous section, SV40 was modified at a pH of 8.0 for 30 minutes with SPAmPEG (2 or 20 kDa), or BTCmPEG (3.4 or 5 kDa) at concentrations ranging from 0-15 mM. CV-1 target cells were challenged with control or mPEG modified virus and infection was monitored over 72 hours through T antigen immunostaining. Results for each molecular weight SPAmPEG modified SV40 are shown in Table 3.2. As shown in the previous section, there were dose-dependent reductions in the percent cells infected at each time point for all three molecular weight SPAmPEGs, with few significant differences between the molecular weights at equimolar mPEG concentrations. 65 Table 3.2: Comparison of SPAmPEG Inhibition of SV40 Infection at Various Molecular Weights. 24 Hours SPAmPEG 0 mM 0.2mM 0.6mM 1.2mM 2.4m M 5mM lOmM 15mM 2 kDa 49.8 ±1.5 39.1" ±2.0 37.333 ±2.4 32.033 ±1.5 27.033 ±1.5 t t 11.533 ±1.2 1.233 ±0.7 0.433 ±0.2 5 kDa 50.6 ±0.6 44.333 ±0.3 41.133 ±0.6 38.533 ±1.6 31.333 ±0.8 15.633 ±0.6 2.633 ±0.5 0.533 ±0.4 20 kDa 50.3 ±0.4 42.8 ±2.6 36.833 ±2.5 28.233 ±4.1 12.333 ±5.1 ** t t 2.53 3 ±2.2 * 0.833 ±1.3 0 3 3 ±0.1 4! 3 hours SPAmPEG 0 mM 0.2mM 0.6mM 1.2mM 2.4mM 5 mM lOmM 15mM 2 kDa 78.6 ±2.1 61.733 ±2.5 58.233 ±2.3 50.133 ±1.8 34.833 ±1.3 19.633 ±1.6 3.433 ±1.5 1.133 ±0.2 5 kDa 75.3 ±6.0 75.6 ±5.8 65.33 ±1.0 59.833 ±1.7 55.233 ±1.9 20.933 ±0.4 5.333 ±0.5 1.733 ±0.4 20 kDa 73.3 ±3.3 63.5 ±15.7. 52.3 ±13.3 43.13 ±9.0 27.733 ±13.8 * 8.333 ±6.9 2.23 3 ±2.8 0.133 ±0.1 72 Hours SPAmPEG 0 mM 0.2 mM 0.6mM 1.2mM 2.4 mM 5 mM lOmM 15mM 2 kDa 99.4 ±0.3 93.1 ±3.1 87.233 ±2.7 62.333 ±1.2 50.133 ±1.1 42.733 ±4.3 10.133 ±1.6 3.53 3 ±0.6 5 kDa 99.4 ±0.2 96.9 ±0.9 81.633 ±1.5 70.633 ±1.4 63.033 ±1.6 50.233 ±0.9 10.433 ±0.8 3.833 ±0.5 20 kDa 99.3 ±0.2 93.7 ±1.7 73.8 ±8.1 67.63 ±7.8 47.733 ±15.3 19.433 ±16.5 ** 5.033 ±6.1 0.733 ±0.8 Data presented in this table represents the average percent cells infected of n=3 independent experiments ± standard deviation. * Significantly different from SPAmPEG 5 kDa at equivalent concentrations (** p <0.001; * p < 0.05). t Significantly differences between SPAmPEG 2 and 20 kDa at equivalent concentrations (tt p < 0.001; t p < 0.05). 3 Significantly different from the control for that mPEG (0 mM) at that particular time point (33p< 0.001,3 p< 0.05). Modification of SV40 with each SPAmPEG molecular weight prevented infection of unmodified cells. SV40 modification with SPAmPEG 20 kDa was more effective than either 2 or 5 kDa at 2.4 mM, 24 hours post challenge. 66 These results surprisingly demonstrated that the 2 kDa polymer was only less effective than the 5 kDa polymer at a concentration of 2.4 mM (p<0.001). In contrast, SPAmPEG 2 kDa was equally effective as the 5 kDa polymer at low and high derivatization concentrations (0.2-1.2mM and 10-15 mM). Based on the work by Bradley, et al. (2002), which demonstrated that high molecular weight mPEGs were better able to shield surface charge and inhibit immunorecognition, the low molecular weight SPAmPEG (2 kDa) appeared equally effective as the 5 kDa polymer. The single significant difference seen at 2.4 mM is likely biologically irrelevant despite the mathematical significance. Interestingly, the 20 kDa polymer was no more effective than the 5 kDa polymer at low derivatization concentrations (0.2-0.6 mM; 24 hours). At moderate mPEG concentrations (2.4 mM), the 20 kDa polymer was statistically more effective than the smaller (2 and 5 kDa) polymers at 24 hours post challenge; however, these differences may not exist in an in vivo model. Additional experiments were conducted with BTCmPEGs of 3.4 and 20 kDa molecular weights using the reaction conditions previously described. The results of these experiments are shown in Table 3.3 and compared to those obtained from BTCmPEG 5 kDa studies previously described. As with the SPAmPEG studies, all three molecular weight BTCmPEGs show a dose dependent decrease in infection rates through 72 hours. There were few significant differences in infection rates between the cells challenged with 3.4, 5, or 20 kDa molecular weight BTCmPEG. Interestingly, at moderate mPEG derivatization concentrations (0.6-2.4 mM) the 3.4 kDa polymer was more effective than the slightly larger 5 kDa polymer. In contrast, results obtained at low (0.2 mM) and high (5-15 mM) mPEG concentrations were indistinguishable 67 Table 3.3: Comparison of BTCmPEG Inhibition of SV40 Infection at Various Molecular Weights. 24 Hours BTCmPEG OmM 0.2mM 0.6mM 1.2 mM 2.4mM 5 mM 10 mM 15 mM 3.4 kDa 49.4 ±2.4 36.7 ±4.7 28.03 ±6.0 * 18.733 ±7.1 * 7.633 ±0.9 * 2.03 3 ±2.2 0.733 ±1.0 0.133 ±0.1 5 kDa 51.4 ±2.4 44.3 ±4.8 40.03 ±5.2 32.433 ±4.5 21.433 ±2.0 9.93 3 ±1.6 l .O 3 3 . ±0.3 0.233 ±0.1 20 kDa 50.9 ±0.8 43.0 ±1.0 37.03 ±3.5 28.633 ±5.4 12.833 ±5.5 2.73 3 ±0.9 0.533 ±0.8 O 3 3 ±0.1 4. 8 Hours BTCmPEG OmM 0.2mM 0.6 mM 1.2mM 2.4 m M 5 mM 10 mM 15 mM 3.4 kDa 73.0 ±11.4 49.33 ±8.9 39.93 ±10.6 27.633 ±10.5 14.233 ±2.8 4.03 3 ±3.5 1.833 ±2.5 0.533 ±0.6 5 kDa 78.4 ±2.3 63.3 ±9.7 59.9 ±10.6 54.53 ±12.8 43.3 3 ±10.0 24.433 ±6.5 8.233 ±3.2 0.733 ±0.3 20 kDa 75.0 ±5.1 59.8 ±15.7 52.0 ±11.6 45.7 ±6.8 29.63 ±17.4 13.933 ±15.2 1.433 ±2.4 0.133 ±0.2 72 Hours BTCmPEG OmM 0.2 mM 0.6mM 1.2 mM 2.4mM 5 mM 10 mM 15 mM 3.4 kDa 97.5 ±3.8 84.4 ±12.2 65.0 ±14.1 55.63 ±25.9 * 27.3 3 3 ±7.1 ** 7.033 ±4.6 * 4.33 3 ±4.3 1.633 ±1.6 5 kDa 99.5 ±0.4 80.09 ±3.1 67.333 ±6.3 63.333 ±7.6 48.733 ±4.2 36.333 ±3.5 13.333 ±2.8 2.033 ±1.1 20 kDa 99.4 ±0.3 91.2 ±4.8 72.9 ±7.0 63.43 ±6.2 46.73 ±13.9 26.033 ±26.4 4.433 ±5.1 1.333 ±1.7 Data presented in this table represents the average percent cells infected of n=3 independent experiments ± standard deviation (SD). * Significantly different from BTCmPEG 5 kDa at equivalent concentrations (** p <0.001; * p < 0.05). f Significant differences between BTCmPEG 3.4 and 20 kDa at equivalent concentrations (ft P < 0.001; t p < 0.05).3 Significantly different from the control for that mPEG (0 mM) at that particular time point (33 p< 0.001,3 p< 0.05). Modification of SV40 with each BTCmPEG molecular weight prevented infection of unmodified cells. SV40 modification with BTCmPEG 3.4 kDa was more effective than 5 kDa at 0.6, 1.2 and 2.4 mM, 24 hours post challenge, and 5 mM at 72 hours. 68 from those observed with the 5 kDa BTCmPEG. Surprisingly, no significant variance was noted between the results obtained from the 20 kDa BTCmPEG and those obtained from the 5 kDa polymer. In sum, both the SPAmPEG and BTCmPEG studies suggest that polymer size is not a crucial factor in viral derivatization efficacy at very high concentrations of mPEG. This may be due to the small size of the virus particles versus the large size of mammalian cells studied by Bradley, et al. (2002). Additional research has shown that less mPEG is required for small proteins to be completely immunocamouflaged than large proteins, therefore the even larger cells with heterogeneous proteins would likely display differences based on polymer size more readily than the smaller, homogeneous viral capsid (Bowen, et al., 1999; Kim and Park, 2001). 3.3 Combination Studies Having deduced that modification of SV40 with various molecular weight mPEGs prevented infection of CV-1 cells, it was hypothesized that additional protection might be gained by combining mPEGs of different molecular weights through the creation of a more dense brush border surrounding the viral capsid. This hypothesis was tested using polymer combinations BTCmPEG. The three equimolar combinations used were: 3.4/5kDa, 3.4/20 kDa, and 5/20 kDa BTCmPEG. As with the previous BTCmPEG experiments, virus was modified at a pH 8.0 for 30 minutes with 0-15 mM of the polymer combinations. Cells were challenged with modified virus and infection was monitored over 72 hours via T antigen staining. 69 Data obtained from the BTCmPEG 3.4/5, 3.4/20 and 5/20 kDa combinations are shown in Table 3.4. Similarly to the results presented in the previous sections, all three time points show a dose-dependent reduction in the percentage of cells infected. For example, at 24 hours post challenge, CV-1 cells exposed to unmodified virus were 51% positive for T antigen while infection rates for cells inoculated with modified SV40 decreased from 35% to 0% (0.2-15 mM) as the concentration BTCmPEG 3.4/5 increased (p<0.001). As noted in Table 3.4, the combination of 3.4/5 kDa significantly improved protection over that of 5 kDa BTCmPEG alone at low to moderate concentrations over the course of 72 hours (0.6 and 2.4 mM at 24 hours; 1.2-5 mM at 48 hours; 2.4-5 mM at 72 hours). No differences were observed between the 3.4/5 kDa combination and 5 kDa alone at concentrations greater than 5 mM. Next, the combination of 3.4/20 kDa was used to modify SV40. As presented in Table 3.4, at 24 hours post challenge cells exposed to unmodified virus were 51% positive for T antigen, while rates of infection for cells inoculated with modified SV40 decreased in a dose dependent fashion ranging from a high of only 6% to a low of 0% (0.2-15 mM) with increasing concentrations of BTCmPEG 3.4/20 kDa (pO.OOl). The comparison of 3.4/20 kDa to 5 kDa BTCmPEG (Table 3.4) uncovered significant differences at low (0.2-1.2 mM) derivatization concentrations. At 24 and 72 hours post challenge with the combination of 3.4/20 kDa modification proved more effective in preventing infection than the 5 kDa alone at derivatization concentrations of 0.2-1.2 mM. In addition, at 48 hours this combination was also more effective at the 0.2-5 mM derivatization concentrations. 70 Table 3.4: Comparison of B T C m P E G Inhibition of SV40 Infection. * Significantly different from BTCmPEG 3.4 & 5 kDa (** p < 0.001; * p < 0.05) t Significantly different from BTCmPEG 3.4 & 20 kDa (tf p < 0.001; f p < 0.05) X Significantly different from BTCmPEG 5 & 20 kDa (U p < 0.001; J p < 0.05) § Significantly different from BTCmPEG 5 kDa (§§ p < 0.001; § p < 0.05) ¥ Significantly different from CmPEG 5 kDa (¥¥ p < 0.001; ¥ p < 0.05) 3 Significantly different from the control for that mPEG (0 mM) at that particular time point (33 p< 0.001,3 p< 0.05). Data presented in this table represents the average percent cells infected of n=3 independent experiments ± standard deviation (SD), and significance between the BTCmPEG combinations shown are between equivalent concentrations of total mPEG (i.e. 5 mM 3.4/5 kDa versus 5 mM 5/20 kDa). Modification of SV40 with each combination of the BTCmPEG molecular weights prevented infection of unmodified cells. SV40 modification with BTCmPEG 3.4 and 20 kDa was more effective than the other combinations at low concentrations (0.2-1.2 mM). 24 Hours B T C m P E G OmM 0.2 m M 0.6mM 1.2 m M 2.4mM 5 m M 10 m M 15 m M 3.4 & 5 51.3 34.63 30.03 15.633 1.833 0.733 0.133 O 3 3 ±0.8 ±10.8 ±12.0 ±10.2 ±1.6 ±1.0 ±0.1 ±0 ft tt f t §§ §§ ¥ ¥¥ 3.4 & 20 50.6 5.933 2.833 2.03 3 0.833 0.13 3 o3 3 o3 3 ±1.3 ±1.3 ±1.4 ±1.2 ±0.2 ±0.1 ±0 ±0 ** ** * tt tt tt §§ §§ §§ §§ ¥¥ ¥ 5 & 2 0 50.3 41.333 32.533 27.833 7.333 O.I 3 3 o3 3 0 3 3 ±2.1 ±2.5 ±3.0 ±2.5 * ±2.0 ±0.1 ±0 ±0 tt tf tt ¥¥ ¥¥ ¥¥ 5 kDa 51.4 443 40.0 32.433 21.4* 9.93 3 f :?' 'i-o'^ ±2.4 ±4.8 ±5.2 . ±4.5 ±2.0 1 ±1.6. ±0.3 , ;±o.r C m P E G 5 49.3 . '• 21.4 , J 14.033 10.533 • 4.73 3 " 1.133 < 0.T33,.;,. ;,±1;S:\: 2 ±0.6 ±1.1 ±1.5 ±1.0 . i&>0-2.V-71 48 Hours BTCmPEG OmM 0.2 mM 0.6mM 1.2mM 2.4mM 5 mM 10 mM 15 mM 3.4 & 5 73.1 54.73 49.03 25.93 6.333 1.733 0.733 0.233 ±4.7 ±10.7 ±9.3 ±2.3 ±2.7 ±1.2 ±0.3 ±0.1 t t tt ¥ §§ §§ § 3.4 & 20 75.2 9.733 7.633 6.333 3.63 3 0.333 O.I 3 3 O 3 3 ±1.9 ±1.6 ** ±2.0 ** ±2.1 ±1.4 ±0.1 ±0.1 ±0 tt §§ ¥¥ tt §§ §§ §§ § 5&20 71.3 44.533 36.933 11.333 1.8 3 3 . 0.0233 O 3 3 O 3 3 ±3.7 ±3.0 ±3.6 ±6.4 ±4.0 ±1.9 ±0.1 ±0.1 t t §§ tt §§ §§ §§ § ;0';5;kDa 78.4 63.3 59.9 54.5'3 43.3 3 : 24.433 ; 8.2 3" 0".733 ••• - ±2.3 ±9.7 '• ±10.6. ±12.8 ±10.0 : ±6.5 ±3.2 ' " ±0.3 ; CmPEG5 75.8 45.93 3 25.033 22.733 11.633 • 2.6" 0.6" 0.1 " • _ ±2.7':; ; •:±'2/4r'....; ••• ±1-5 •i±'i.4^ f,:.±o.3%. 72 Hours BTCmPEG OmM 0.2mM 0.6 mM 1.2mM 2.4mM 5 mM 10 mM 15 mM 3.4 & 5 98.4 75.43 68.13 50.133 14.3 3 3 4.53 3 2.33 3 0.633 ±0.6 ±13.8 t t ±13.1 tt ¥¥ ±11.2 ±6.4 §§ ±1.5 §§ ±1.2 ±0.2 3.4 & 20 99.1 39.533 32.133 27.333 15.433 4.63 3 1.933 0.233 ±0.4 ±3.5 ±2.9 ±3.1 ±2.1 ±1.8 ±1.0 ±0.1 ** % §§ ¥¥ ** §§ §§ 5&20 99.0 69.133 53.3 3 3 32.233 7.933 1.233 O.I 3 3 O.I 3 3 ±0.6 ±3.2 t t ±3.6 ±6.4 §§ ±4.0 §§ ±1.9 §§ ±0.1 ±0.1 5 kDa 99.5 80.03 67.3.33 . 63,333 48.733 ,36.3 3 3 ' 13.333 X O 3 3 , ' ±0.4 ±3.1 ±6.3 ' ±7.6 ' ±4.2' ±3.5 ±'2.8 ±1.1 CmPEG 5 99.1 84.533 36.233 • 34.9,". ,22.133 6.33 3 2.5 3 3'- •0.333 • i V V 1 ' 1 ' ' ' , ±0.2 . ±5.7>„ ±0.8 ±0.4 i ±3.6 ±3:8 „ ±2.2 ±0.3 72 Finally, the combination of 5/20 kDa BTCmPEG was used to modify SV40 following the protocol previously outlined. Shown in Table 3.4, at 24 hours post challenge cells exposed to unmodified virus were 50% positive for T antigen and cells challenged with BTCmPEG 5/20 kDa decreased from 41% to 0%> (0.2-15 mM) T antigen positive in a dose dependent manner (p<0.001). Similar dose response curves were seen at 48 and 72 hours post challenge. Interestingly, at 24 hours post infection there were no differences observed between 5/20 kDa and 5 kDa BTCmPEG. However, at 48 hours the combination of 5/20 kDa was more effective at low and moderate derivatization concentrations (0.2-5 mM) and at moderate concentrations (1.2-5 mM) 72 hours post viral challenge. 3.4 Discussion My results further confirmed the findings of Mizouni (2000), who utilized only the 5 kDa CmPEG when modifying the SV40 capsid. In addition, my studies extended Mizouni's work by exploring different linker moieties (cyanuric chloride, benzotriazole carbonate and succinimidyl propionate), as well as additionalmolecular weights of mPEG and combination studies using two different molecular weight BTCmPEGs. Results presented in this chapter demonstrated that modification of SV40 with CmPEG, SPAmPEG and BTCmPEG prevented infection of unmodified CV-1 cells. These data corroborate previous research on mPEG derivatized red blood cells conducted in Dr. Scott's laboratory, which found that SPAmPEG and BTCmPEG modified cells had similar biophysical 73 and immunocamouflaging properties as CmPEG (Scott et al, 2000; Bradley et al., 2002). This study demonstrated that the only differences thought to exist between linker molecules were the rates of the derivatization reaction and hydrolysis. It was also shown that 20 kDa mPEG more effectively camouflaged red blood cells compared to the 3.4 and 5 kDa mPEGs. Comparable antiviral effects were obtained following modification of SV40 using the different linker chemistries. Modification with CmPEG 5 kDa was more effective in preventing SV40 infection than BTCmPEG or SPAmPEG at low concentrations (0.2-2.4 mM), 24 hours post exposure. Because CmPEG reacts faster, a greater percentage of mPEG molecules are utilized during the 30 minute reaction (Murad, et al. 1999A). However, there were no significant differences at 10-15 mM between any of the mPEGs (combinations included). With higher concentrations (10 and 15 mM) the reaction was saturated to the extent that unmodified lysine residues were blocked by previously modified proteins adjacent to them. My work showed that individual polymer sizes varied little in the protection against SV40 infection at high derivatization concentrations (10-15mM) and very few differences existed at concentrations lower than 5 mM. The most significant protection at low concentrations was afforded by BTCmPEG 3.4 kDa. It may be that the surface topography of SV40's capsid is more optimally coated with mPEG by the 3.4 kDa polymer. As mentioned in Chapter 2, the capsid is composed of 360 copies of viral protein 1 (VP1) that are organized into 72 pentamers (Flint, et al, 2000). Hence there are a set number of lysine residues that can be modified (the VP1 amino acid sequence is shown in Table 2.3), which are always positioned a certain distance from each other based on the icosahedral symmetry of the capsid (Fields, 1998), therefore an optimal polymer size or combination should exist (even if only theoretically). 74 With the concept of optimal capsid pegylation in mind, the combinations of BTCmPEG molecular weights were compared. It was found that modification of SV40 with the combination of BTCmPEG 3.4/20 kDa was more effective at preventing infection than the other combinations or the single molecular weight polymers alone. In addition, 3.4/20 kDa provided significantly more protection against viral infection than the 3.4 kDa alone at low concentrations (0.2-1.2 mM; pO.OOl). In contrast, the 5/20 kDa combination offered no additional protection when compared to the 5 kDa alone at 24 hours. The 5 and 20 kDa polymers may occlude unmodified lysine residues in a manner that results in a less dense mPEG brush border than the 3.4/20 kDa combination. As illustrated in Figure 3.1, the 20 kDa polymer may sterically hinder mPEG modification of adjacent unmodified lysine residues by 5 kDa or 20 kDa polymers, whereas the 3.4 kDa polymer is not inhibited from reacting. Throughout this chapter it was clearly demonstrated that covalent modification of SV40 with mPEG was a powerful means of preventing viral infection. This method may be good for viral inactivation in acellular products (e.g., plasma), however, it would be highly ineffective for the prevention of common cold infections in the real world as only in a lab is virus contained in clearly labeled vials perfect for pegylation. To this end, we hypothesized that an activated mPEG solution can be used to prevent common cold infections in other ways such as through the modification of viral target cells within the nasopharyngeal cavity. 75 VP1 Figure 3.1: Modification of the SV40 Capsid is M o r e Protective with the Combination of 3.4/20 kDa than 5/20 kDa . Illustrated above is a portion of the SV40 capsid composed of discrete VP1 molecules with lysine residues (K) at a constant distance (DI or D2) apart. The combination of 3.4 and 20 kDa (A) may be more effective in preventing SV40 infection because the two polymer sizes do not sterically interfere with the others binding to unmodified lysine residues. As illustrated in B , the 20 kDa polymer may sterically hinder the 5 kDa polymer from reacting with the unmodified lysine residues (or vice versa), resulting in gaps in the mPEG brush border. 76 Chapter 4: Pegylation of Host (CV-1) Cell Surfaces Prevents SV40 Infection Chapter Over\ iew • ITfects of Non-Cox alcnt xersus Coxalenl mPI.G Modification of Target Cells: Is covalcnt grafting of m P E G to host cells essential for antiviral efficacy? • Effect of Reaction Time, pi I and Temperature on Pegv lalion Induced Protection from ? Viral Infection: Can immunocamouflage occur under ph>siological conditions? • Effect of Reaction Time, pi 1 and Temperature on mPFG Protection from Viral Election «hen Challenged uiih Increasing Viral Titers: How effective is immunocamouflage against increasing viral exposure? • lined ofDelayed Viral Challenge Following Pegylation ofCells: Is protection • i:iTcei of Mucin on Pegylation: Will nasal secretions interfere uilh pollution? • Effect of mPF.G on Viral Entry: Does mPEG present viral entry? 4.1 Non-covalent versus Covalent mPEG Modification of Target Cells 4.1.1 Non-Covalent (Unactivated) mPEG does not Inhibit SV40 Infection According to the hypothesis of this thesis, covalent grafting of the mPEG is essential for the antiviral effect, as only the covalently bound mPEG creates a bio-physical barrier, which inhibits virus-receptor interactions. In contrast, unactivated (i.e. mPEG lacking the chemical linker moiety) polyethylene glycol, which is not covalently bound, would not camouflage the viral receptors and should not prevent viral entry and infection. 77 However, in order to ensure that the unactivated mPEG was not interfering with the infection process (e.g., cell entry), cells were treated with soluble mPEG (4 kDa) for 30 minutes prior to infection. As shown in Figures 4.1 and 4.2, increasing concentrations of unactivated mPEG had no effect on SV40 infection as measured by T antigen expression. At 24 hours post SV40 challenge control cells were approximately 50% positive for large T antigen. Similarly, cells exposed to increasing concentrations of unactivated mPEG (0.2-15 mM) were also 50% positive at 24 hours. By 72 hours post SV40 exposure, all concentrations of non-covalent mPEG (0.2- 15 mM) were approximately 100% infected, as were the control cells. These data clearly demonstrated that unactivated mPEG was not capable of interfering with SV40 infection of CV-1 cells. 78 100 T3 -t—> o 75 t/2 U +-» fi <D O VH 24 Hours 48 Hours -o- -®- -©-50J T - H -25 72 Hours 1 1 r— 0 5 10 15 Concentration Unactivated PEG 4kDa (mM) Figure 4.1: Unactivated mPEG does not Prevent Infection Confluent CV-1 cells were exposed to unactivated mPEG at concentrations ranging from 0-raM. At 24 hours cells exposed to all concentrations of mPEG were approximately 50% infected. By 72 hours the number of infected cells rose to nearly 100% for all mPEG concentrations. n= 3, standard deviation is shown. ** p < 0.001; * p < 0.05 significantly different from time matched controls (0 mM unactivated mPEG). 79 A. 0 mM mPEG 4 kDa Figure 4.2: Microscopic Look at the Effects of Inactivated mPEG on SV40 Infection Cells were fixed and stained for SV40 T antigen at 24 hour intervals for 72 hours. The phase contrast pictures are shown on the left, while the same field is shown under U V fluorescence on the right. Note that the nucleus of infected cells will appear green under fluorescence. At 72 hours, both control cells ( A ) and those exposed to 15 mM unactivated mPEG ( B ) were 92% infected. Shown at 20x magnification. 80 4.1.2 Covalent Modification of CV-1 Cells with CmPEG 5 kDa Prevents SV40 Infection To experimentally determine whether covalently grafted mPEG would prevent viral infection, CV-1 cells were modified with increasing concentrations of cyanuric chloride activated mPEG (CmPEG 5 kDa). As described in the methods, confluent CV-1 cells were incubated with CmPEG at room temperature for 30 minutes prior to SV40 challenge. At 24 hour intervals, cells were stained for SV40 large T antigen. As. shown in Figure 4.3, at 24 hours post challenge 50% of unmodified control cells were infected, while only 9%> of cells modified by 2.4 mM CmPEG were T antigen positive (pO.OOl). Higher concentrations of CmPEG further reduced the percentage of infected cells (e.g., 5% at 15 mM). By 72 hours (Figure 4.3), control cells were nearly 100% infected while only 24%> of cells modified by 15 mM CmPEG were infected (pO.OOl) as illustrated in photomicrographs (Figure 4.4). These data clearly demonstrated that covalent modification of CV-1 cells with CmPEG could drastically reduce viral entry and subsequent infection as measured by T antigen immunostaining. 81 Concentration cmPEG 5kDa (mM) Figure 4.3: C m P E G Modification of CV-1 Cells Prevents SV40 Infection. CV-1 cells were modified with CmPEG 5 kDa at room temperature for 30 minutes prior to SV40 challenge. Infection was monitored through T antigen staining at 24 hour intervals for 72 hours. As shown, there is a significant decrease in the percent of modified cells infected compared to unmodified controls. The standard deviation of n=3 independent experiments is shown. ** p < 0.001, which represents significant differences between cells modified by CmPEG versus unmodified control cells at the same time point. Importantly, there is also a significant difference between 0.2 mM CmPEG and 15 mM at each time point. 82 A. 0 mM CmPEG 5 kDa B. 1 5 mM CmPEG 5 kDa Figure 4.4: Microscopic Look at the Effects of C m P E G 5 kDa on SV40 Infection. Cells were fixed and stained for SV40 T antigen at 24 hour intervals for 72 hours. The phase contrast pictures are shown on the left, while the same field is shown under U V fluorescence on the right. Note that the nucleus of infected cells will appear green under fluorescence. As seen in A, control cells challenged with SV40 were nearly 100% infected at 72 hours. The second picture is that of cells modified by 15 mM CmPEG prior to infection; note the dramatic reduction in the number of infected cells. All pictures were taken at 20x magnification with Olympus Magnafire camera and software. 8 3 4.2 Effect of Reaction Time, pH and Temperature on Pegylation of Cells Based on our previously published studies on red blood cells, lymphocytes and pancreatic islets, the initial mPEG reactions were completed at a pH of 8.4 for 30 minutes at room temperature. However, these conditions would not be optimal for prophylactic use in humans, due to the alkaline pH (hence potentially irritating), temperature differential (intranasal temperature averages 32 °C) and the 30-minute application time. In order to optimize the pegylation reaction for human use, experiments were performed varying the reaction time, pH and temperature. Once the protective effects of CmPEG modification of target cells was confirmed, the next question addressed was how long a reaction time was required for maximal CmPEG protection to occur. To answer this question, CV-1 cells were modified for 1-60 minutes at a pH of 8.4 and room temperature. As depicted in Figure 4.5, control cells were 50% infected at 24 hours. Surprisingly, even cells modified for 1 and 2 minutes with 15 mM CmPEG were only 16%o and 13%o infected, while those modified for 5, 15, 30 and 60 minutes were 10%o, 7%, 6% and 5% infected, respectively (p<0.001). Interestingly, the longer mPEG reaction times (15, 30 and 60 minutes) provided only slightly better protection from infection than the 5 minute exposure. It was thought that longer reaction times would result in a more dense brush border and therefore greater antiviral efficacy when other reaction conditions were altered (such as pH), however my results showed that near maximal protection was achieved within 5 minutes. Furthermore, as demonstrated in Figure 4.6, by 72 hours post challenge control cells were 99%> infected, whereas cells modified for 1 and 2 minutes with 15 mM CmPEG were only 24% and 26%> ' 84 infected. Further reductions were seen in cells modified for 5, 15, 30 and 60 minutes with infection rates of 20%, 16%, 15% and 20%, respectively (pO.OOl). Based on these findings, as well as keeping in mind human nature, it was hypothesized that a minimum application time of 3-5 minutes w i l l be sufficient to give near maximal protection. 50 A -<-> U t/3 l u C 25 <u a S-H m— 1 M i n Rxn o — 2 M in Rxn O — 5 M i n Rxn * — 15 M in Rxn O — 30 M i n Rxn * — 60 M in Rxn —r— 10 —r-15 Concentration CmPEG 5 kDa (mM) Figure 4.5: A Reaction of 5 Minutes is Sufficient for Protection (pH 8.4, Room Temperature, 24 Hours Post SV40 Challenge) CV-1 cells were modified for 1, 2, 5, 15, 30 and 60 minutes at a pH of 8.4 and room temperature. A l l control cells were 5 0% infected at 24 hours. Cells modified for 1 and 2 minutes with 15 m M C m P E G were 16% and 13%infected, while those modified for 5, 15, 30 and 60 minutes were 10%, 7%, 6%, and 5% infected respectively. The standard deviation of n=3 independent experiments is shown. * p < 0.05 Significantly different from the 15 minute reaction at the equivalent concentration, t p < 0.05 Significantly different from the 60 minute reaction. Importantly, there is also a significant difference between 0.2 m M CmPEG and 15 m M for each reaction time. 85 i o o A 1 7 5 C I—I U c <u o OH 50 -J 25 4 - H — 72 Hr 1 Min Rxn - o — 72 Hr 2 Min Rxn - G — 72 Hr 5 Min Rxn - A — 72 Hr 15 Min Rxn - • — 72 Hr 30 Min Rxn - * — 72 Hr 60 Min Rxn 0 5 10 15 Concentration C m P E G 5 kDa (mM) Figure 4.6: A Reaction of 5 Minutes is Sufficient for Protection (pH 8.4, Room Temperature, 72 Hours Post SV40 Challenge) CV-1 cells were modified for 1, 2, 5, 15, 30 and 60 minutes at a pH of 8.4 and room temperature. All control cells were 99% infected at 72 hours. Cells modified for 1 and 2 minutes with 15 mM CmPEG were 24% and 26% infected, while those modified for 5, 15, 30 and 60 minutes were 20%, 16%, 15%, and 20% infected respectively. The standard deviation of n=3 independent experiments is shown. ** p < 0.001; * p < 0.05 Significantly different from the 15 Minute Reaction at the equivalent concentration. f t P< 0.001; t p < 0.05 Significantly different from the 60 Minute Reaction. Importantly, there is also a significant difference between 0.2 mM CmPEG and 15 mM for each reaction time. 86 While it was apparent that the reaction time could be dramatically decreased, the previous experiments were completed at a pH of 8.4, and room temperature. Theoretically, a high pH is ideal for the pegylation reaction as it favors the deprotonation of the lysine residues (NH3+ to N H 2 ) , with pH 9.2 being optimal (Zubay 1998). Despite the foregoing, pegylation of cells would ideally occur at a pH closer to physiological conditions to avoid nasal irritation and burning. Consequently, experiments were conducted at pH 7.8 in order to both maintain a slightly basic environment and be closer to physiological pH. Furthermore, temperature is an important factor in reaction efficacy, with 4°C being optimal (Scott, et. ai, 1997). However, because the temperature of the nasal cavity, site of infection for common colds, is between 31 and 34 °C regardless of ambient temperature, further experiments were conducted at a pH of 7.8 and a temperature of 32 °C. To determine if less alkaline conditions could be utilized, CV-1 cells were pegylated for 1- 60 minutes at pH 7.8 and room temperature. As seen in Figure 4 . 7 A at 24 hours, control cells were approximately 50% infected at all reaction times. Cells modified for 1 and 2 minutes with 15 mM CmPEG were 8% and 6% infected, while those modified for 5, 15 and 60 minutes were 5%, 3% and 2% infected respectively. Shown in Figure 4 . 8 A , by 72 hours post infection all control cells were 99% infected. In contrast, cells modified for periods of time as short as 1 and 2 minutes with 15 mM CmPEG were only 21% and 17% infected, while those modified for 5 and 15 minutes were 17% and 12% infected respectively. As previously noted, a 3-5 minute derivatization time again provided near maximal protection against viral infection. As before, grafting times longer than 15 minutes had no additional benefits (p>0.05). 87 A. pH 7.8, RT C. pH 8.4, RT 50 T 3 <U u e a 25 OH 24 Hr 1 Min Rxn 24 Hr2 Min Rxn 24 Hr 5 Min Rxn 24 Hr 15 Min Rxn 24 Hr 30 Min Rxn 24 Hr 60 Min Rxn 10 15 50 T3 U c 25 o Concentration C m P E G 5 kDa (mM) 0 5 10 15 Concentration C m P E G 5 kDa (mM) B. pH 7.8, 32°C 100 u U , - 50-1 <u o cu 25 72 Hr 1 Min Rxn 72 Hr 2 Min Rxn 72 Hr 5 Min Rxn 72 Hr 15 Min Rxn 72 Hr 30 Min Rxn 72 Hr 60 Min Rxn 0 5 10 15 Concentration C m P E G 5 kDa (mM) Figure 4.7: A 5-Minute Reaction is Sufficient for mPEG Protection with Physiologic Reaction Conditions (24 hrs) As shown in A (pH 7.8, RT), at 24 hours, unmodified cells were 50% infected at all reaction times. Cells modified for 1 minute with 15 mM CmPEG were 8% infected, while those modified for 5 minutes were 5% infected. Similarly, B shows that following the reaction at pH 7.8, 32°C, 24 hours post SV40 challenge control cells were approximately 50% infected at all reaction times. Cells modified for 1 minute with 15 mM CmPEG were 18% infected, while those modified for 5 minutes were 14%, infected. As shown in C (pH 8.4, RT), all three of these reaction conditions yield similar response curves. The standard deviation of n=3 independent experiments was shown in each graph. * p < 0.05, Significantly different from the 15 minute reaction time, t p < 0.05; t t p < 0.001, Significantly different from the 60 minute reaction time. 88 A . pH7.8,RT C. pH8.4,RT Concentration C m P E G 5 kDa (mM) Concentration C m P E G 5 kDa (mM) B. pH 7.8, 32°C 100 £ ? 5 c U - 5 0 u 25 72 H r 1 M i n R x n 72 H r 2 M i n R x n 72 H r 5 M i n Rxn 72 H r 15 M i n R x n 72 Hr 30 M i n R x n 72 Hr 60 M i n Rxn 10 Concentration C m P E G 5 kDa (mM) Figure 4.8: A 5-Minute Reaction is Sufficient for mPEG Protection with Physiologic Reaction Conditions (72 hrs) A s shown in A (pH 7.8, RT) , at 72 hours, unmodified cells were 99% infected at all reaction times. Cells modified for 1 minute with 15 m M C m P E G were 21% infected, while those modified for 5 minutes were 17%o infected. Similarly, B shows that following the reaction at p H 7.8, 32°C, 72 hours post SV40 challenge control cells were approximately 100% infected at all reaction times. Cells modified for 1 minute with 15 m M C m P E G were 40% infected, while those modified for 5 minutes were 36%>, infected. A s shown in C (pH 8.4, RT) , all three of these reaction conditions yield similar response curves. The standard deviation of n=3 independent experiments was shown in each graph. * p < 0.05, Significantly different from the 15 minute reaction time, t p < 0.05; f f p < 0.001, Significantly different from the 60 minute reaction time. 89 More surprisingly, covalent derivatization of cells at pH 7.8 proved superior to pH 8.4 in preventing viral infection. Statistical comparison of the data from reactions at pH 8.4 (Figures 4.7C and 4.8C) and 7.8 demonstrated significant benefits at the 5 minute reaction times at the 2.4, 5, 10 and 15 mM concentrations of CmPEG at 24 hours. The 5 minute reaction at pH 8.4 resulted in 15%, 14%, 12% and 10% infected, while the reaction at pH 7.8 yielded 7%, 6%, 5% and 5% infected cells (2.4, 5, 10 and 15 mM respectively). It is possible that a pH 7.8 may have prevented alkaline cell injury during the pegylation reaction thereby maintaining the monolayer integrity by preventing cell contraction which would result in the exposure of unmodified portions of the cell membrane to viral attachment. In sum, these results demonstrate that mPEG grafting at pH 7.8 versus 8.4 (room temperature) provided superior . prophylaxis against SV40 infection. As before, a five minute reaction was sufficient for near maximal mPEG protection against SV40 infection. Because the nasopharyngeal cavity has a mean temperature of 32 °C, regardless of ambient temperature, additional experiments using a pH of 7.8 were conducted at 32 °C. CV-1 cells were pegylated for 1- 60 minutes at pH 7.8 and 32 °C prior to SV40 challenge. At 24 hours (Figure 4.7B), control cells were approximately 50% infected at all reaction times. Cells modified for 1 and 2 minutes with 15 mM CmPEG were 18% and 16% infected, while those modified for 5, 15 and 60 minutes were 14%, 13%, and 12% infected respectively. It should be noted that the difference in infection rates between 15 and 60 minutes was not statistically significant. Shown in Figure 4.8B, by 72 hours post challenge all control cells were 99% infected. Cells modified for 1 and 2 minutes with 15 mM CmPEG were 40% and 39% infected, while those modified for 5, 15 and 60 minutes were 36%, 28%, and 37% infected 90 respectively. As with previous results, these data showed that a 5 minute reaction time offered near maximal protection against viral infection 24 hours post SV40 challenge. Statistical comparison of cell derivatization at room temperature versus 32 °C does suggest that room temperature is superior in conferring viral protection. However, even at 32 °C, membrane pegylation provides a potent prophylaxis against viral infection. 4.3 Effect of Reaction Time, pH and Temperature on mPEG Protection When Challenged with Increasing Viral Concentrations As clearly demonstrated in the preceding sections, cell surface pegylation provides a significant prophylactic effect against viral infection under more physiological conditions (pH 7.8, 32 °C following a 5-minute derivatization time). The next question we chose to address was how effective pegylation was against increasing viral titers. Importantly, the viral titer used in the previous experiments was quite large at 109 pfu/mL, but to ensure mPEG was effective against varying levels of exposure (from inhaling a droplet of a sneeze to a 50 mL conical tube of virus), serial (log scale) dilutions of viral stocks were used. CV-1 cells were modified for 30 minutes at room temperature, pH 8.4, with CmPEG 5 kDa prior to viral challenge. The lowest concentration of virus used was 106 pfu/mL and the highest concentration used was 2x1010 pfu/mL. Interestingly, virus concentrations of 106 pfu/mL resulted in only 2% of unmodified control cells becoming infected while 0% of modified cells (regardless of CmPEG concentration) were infected at 24 hours post challenge (p<0.001). It should be noted that the 91 discrepancy of pfu/mL versus percent infection at 24 hours results from the titer calculation being determined 7-10 days post infection through plaque assays versus 24 hours via T antigen staining. In contrast at 24 hours, control cells challenged with the highest concentration of virus were 70% infected (Figure 4.9A), while those modified by 15 mM CmPEG were only 9% infected (pO.OOl). More importantly, at levels of viral exposure more closely approximating those seen in the real world (107 pfu/mL), mPEG modification was highly effective. Control cells challenged with 107pfu/mL were 6% T antigen positive at 24 hours compared to 0% positive following treatment with 10 and 15 mM CmPEG (pO.OOl). By 72 hours (Figure 4.9B), control cells challenged with 107 pfu/mL were 43% infected, while only 3% of those modified by 15 mM CmPEG were infected (pO.OOl). Even when challenged with 108, 109 and 1010 pfu/mL, mPEG grafting provided highly significant protection against viral infection with only 3%, 6% and 7% of cells infected at 24 hours following modification with 15 mM CmPEG (pO.OOl). These data clearly demonstrated that mPEG modification of host cells prevented infection when challenged with viral loads ranging from real world conditions to ridiculously large amounts of virus. 92 A. 24 Hours I I V V 1 1 r 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) B. 72 Hours Figure 4.9: C m P E G Modification of CV-1 Cells Prevents SV40 Infection Over a Wide Range of SV40 Concentrations (Four Logs) Cells were modified at pH 8.4, room temperature, for 30 minutes by 0-15 mM CmPEG prior to infection with SV40 ranging in concentration from 106- 1010'3 pfu/mL. A: Control cells exposed to the highest concentration of virus were 70% infected at 24 hours, while those modified by 15 mM CmPEG were only 9% infected. Unmodified cells inoculated with 107 pfu/mL were 6% infected at 24 hours compared to 0% cells treated with 10 and 15 mM CmPEG. B: Control cells challenged with 107 pfu/mL were 43 % infected, while those modified by 15 mM CmPEG were 3% infected. n=3, Standard deviation is shown. **p< 0.001 significantly different from the 0 mM control for that particular viral dilution. 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) 93 Based on the results from the reaction time experiments in which it was determined that a 5 minute reaction was sufficient for near maximal protection against viral infection, the previous study was repeated using a 5 minute reaction time. As shown in Figure 4.10, at 24 hours control cells challenged with the highest concentration of virus (2xl0 1 0pfu/mL) were 79% infected, while those modified by 15 mM CmPEG were only 15% infected (pO.OOl) versus the 9% observed using the 30 minute derivatization time. Control cells challenged with 107 pfu/mL were 7% infected at 24 hours compared to 0% positive following prophylactic treatment with 5 and 15 mM CmPEG (pO.OOl). Not surprisingly, challenging control cells with 2xl0 1 0 pfu/mL of SV40 resulted in nearly 100% cells being T antigen positive at 72 hours while those modified by 15 mM CmPEG were only 33% T antigen positive. More importantly, by 72 hours post challenge with lower viral titers (107 pfu/mL), control cells were 39% T antigen positive (Figure 4.11), while those modified by 15 mM CmPEG were only 0.1% positive (pO.OOl). Statistical analysis of the 5 and 30 minute reactions times (Figures 4.10B and 4.11B), verified that there were no significant differences between the two. Thus, these data confirmed that a 5 minute reaction was sufficient for producing a highly effective protection against viral infection of both small and large viral titers. 94 A. 5 Minute Reaction at pH 8.4, RT 100 T3 U 8 75 4) U « 50 C <u o u 25 0 0 mM 0.2 mM 1.2 mM 5 mM 15 mM 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) B. 30 Minute Reaction at pH 8.4, RT Figure 4.10: A 5 Minute Modification of CV-1 Cells Prevents SV40 Infection Over a Wide Range of SV40 Concentrations (Four Logs), 24 Hours Post Challenge As shown in A, cells were modified for 5 minutes at pH 8.4, RT, by 0-15 mM CmPEG prior to inoculation with SV40 ranging in concentration from 106- 2xl0 ! 0pfu/mL. Unmodified CV-1 (control) cells challenged with 107 pfu/mL were 7% infected, while those modified by 15 mM CmPEG were 0% infected. Challenge of control cells with 2x 1010 pfu/mL were nearly 79% infected at 24 hours while those modified by 15 mM CmPEG were 15% infected. The standard deviation of n=3 independent experiments was shown. For comparison, Figure 4.9A has been included in panel B, note the similar shape of both curves. ** p < 0.001; * p < 0.05 significantly different from the 0 mM control for that particular viral dilution. 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 SV40 Concentration (pfu/mL) E+ll 95 A. 5 Minute Reaction at pH 8.4, RT 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) B. 30 Minute Reaction at pH 8.4, RT Figure 4.11: A 5 Minute Modification of CV-1 Cells Prevents SV40 Infection Over a Wide Range ofSV40 Concentrations (Four Logs), 72 Hours Post Challenge As shown in A, cells were modified for 5 minutes at pH 8.4, RT, by 0-15 mM CmPEG prior to inoculation with SV40 ranging in concentration from 106- 2x10'0 pfu/mL. Control cells challenged with 107 pfu/mL were 39% infected, while those modified by 15 mM CmPEG were 0.1% infected. Challenge of control cells with 2xl0 1 0 pfu/mL were nearly 100% infected at 72 hours while those modified by 15 mM CmPEG were 33% infected. For comparison, Figure 4.9B has been included in panel B, note the similar shape of both curves. ** p < 0.001; * p < 0.05 significantly different from the 0 mM control for that particular viral dilution. 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) 96 While the previous experiments utilized a pH of 8.4, this is less than optimal for intranasal application as previously discussed. To determine if pH 7.8 provided similar protection, CV-1 cells were modified at room temperature for 5 minutes at pH 7.8 prior to challenge with viral loads ranging from 106- 2xl0 1 0 pfu/mL. As shown in Figure 4.12A, at 24 hours control cells challenged with the highest concentration of virus were 87% infected, while those modified by 15 mM CmPEG were only 25% infected (p<0.001) versus the 15%> following derivatization at pH 8.4. Under a more lower viral concentration model of infection, control cells challenged with 107 pfu/mL were 9%> infected at 24 hours compared to only 0.03%> T antigen positive following treatment with 15 mM CmPEG (pO.OOl) following 5-minute derivatization at pH 7.8. By 72 hours (Figure 4.13A), control cells inoculated with 107 pfu/mL were 32% infected, while those modified by 15 mM CmPEG were only 0.8% infected (pO.OOl). Again, these data confirm that a 5 minute reaction time at pH 7.8 was sufficient for protection against SV40 infection over a wide range of viral concentrations. Finally, in order to further test the efficacy of the mPEG prophylaxis under full physiological conditions, log scale experiments were conducted following derivatization at 32 °C (pH 7.8, 5 minutes) and then challenged with viral loads ranging from 106- 2xl0 1 0 pfu/mL. As shown in Figure 4.12B at 24 hours, control cells challenged with the highest concentration of virus were 82%o infected, while those modified by 15 mM CmPEG were 36% infected (pO.OOl). Using the lower viral titer of 107 pfu/mL control cells were 7% infected at 24 hours whereas cells treated with 5 and 15 mM CmPEG compared were negative for viral infection (pO.OOl). By 72 hours (Figure 4.13B), control cells challenged with 107 pfu/mL were 30% infected, while those modified by 15 mM CmPEG were only 0.4%> infected (pO.OOl). Again, these results 97 confirmed significant mPEG protection against SV40 infection subsequent to a reaction of five minute reaction at pH 7.8 and 32 °C. A. 5 Minute Reaction: pH 7.8, RT 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) B. 5 Minute Reaction: pH 7.8, 32"C 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 1E+12 SV40 Concentration (pfu/mL) Figure 4.12: CmPEG Modification of CV-1 Cells for 5 Minutes Prevented Infection with Increased Viral Titers, 24 Hours Post Infection Cells modified for 5 minutes at pH 7.8 (RT) 24 hours post challenge (A), control cells incubated with the highest concentration of virus were 87% infected, while those modified by 15 mM CmPEG were only 25%o infected. Control cells challenged with 107 pfu/mL were 9% infected at 24 hours compared to 0.03%o positive following treatment with 15 mM CmPEG. B shows the reaction performed at pH7.8, 32 °C. At 24 hours, control cells challenged with the highest concentration of virus were 82% infected, while those modified by 15 mM CmPEG were only 36%o infected. Control cells infected with 107 pfu/mL were 7% infected at 24 hours compared to 0%> positive following treatment with 5 and 15 mM CmPEG. Note the similar shape of curves generated under different reaction conditions. For comparison of reaction conditions, the 5 and 30 minute reactions at pH 8.4 can be found in Figure 4.10. The standard deviation of n=3 independent experiments is shown. **p< 0.001; *p<0.05 significantly different from the 0 mM control for that particular viral dilution. 98 A. 5 Minute Reaction: pH 7.8, RT 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 SV40 Concentration (pfu/mL) B. 5 Minute Reaction: pH 7.8, 32°C E+l l Figure 4.13: CmPEG Modification of CV-1 Cells for 5 Minutes Prevented Infection with Increased Viral Titers, 72 Hours Post Infection A depicts the 5 minute reaction at pH 7.8 (RT). At 72 hours, control cells inoculated with 107 pfu/mL were 32% infected, while those modified by 15 mM CmPEG were 0.8% infected. Inoculation of control cells with 2x1010 pfu/mL resulted in nearly 100% infected at 72 hours while those modified by 15 mM CmPEG were 55% infected. B shows the results when the reaction was performed at pH 7.8, 32 °C. At 72 hours, control cells challenged with 107 pfu/mL were 14% infected, while those modified by 15 mM CmPEG were 0.4% infected. Infection of control cells with 2xl0 1 0 pfu/mL were nearly 100% infected at 72 hours while those modified by 15 mM CmPEG were 48% infected. Note the similar shape of curves generated under different reaction conditions. For comparison of reaction conditions, the 5 and 30 minute reactions at pH 8.4 can be found in Figure 4.11. The standard deviation of n=3 independent experiments is shown. **p<0.001;*p<0.05 significantly different from the 0 mM control for that particular viral dilution. 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 1E+12 SV40 Concentration (pfu/mL) 99 4.4 mPEG Dosage Schedule While viral infection immediately following mPEG derivatization is clearly prevented, an important question that needed to be addressed was the time frame of the prophylactic effect as this will govern both application scheduling as well as the practicality of this prophylactic approach. Due to protein turnover and possible enzymatic reactions that could destroy the protective mPEG barrier, it was hypothesized that dosing would need to occur every 12- 48 hours. Additionally, a study conducted by Carson, et al. (1981), suggests that nasal epithelial cells have a rate of turnover similar to that of intestinal epithelia, approximately 3 days, suggesting a maximal delay of 72 hours prior to re-application of mPEG. To test the longevity and efficacy of mPEG prophylaxis, delayed infection experiments were conducted. Cells were modified with CmPEG for 5 minutes at pH 7.8 (32 °C) while the viral challenge was delayed for up to 72 hours following mPEG derivatization. Cells were monitored for infection 24 hours post viral challenge only. As shown in Figure 4.14, unmodified cells infected with SV40 without delay were 50% infected compared to 12%) of cells modified by 10 mM CmPEG (pO.OOl). Following a delay of 12 hours, unmodified cells were 49%> infected while those modified by 10 mM were 20%> infected (pO.OOl). Importantly, after a delay of 24 hours only 18% of the 10 mM CmPEG modified cells were infected compared to 50%> of the control cells (pO.OOl). Interestingly, even following a 60 hour delay, 10 mM modified cells were 34%> infected compared to 50%> of the unmodified control cells (pO.OOl). Thus, while a loss in prophylaxis was noted over 24 hours and even more so after 60 hours, the mPEG antiviral effect remained highly significant (pO.OOl). 100 5 Minute Reaction: pH 7.8, 32 °C Figure 4.14: A 5 Minute mPEG Reaction at pH 7.8 (32 °C) Effectively Prevents SV40 Infection for 24 Hours. As shown, cells were modified for 5 minutes at pH 7.8 (32 °C). Unmodified cells challenged with SV40 without delay were 50% infected compared to 18% of cells modified by 10 mM CmPEG. A 24 hour delay resulted in 50% of control cells infected while 26% of cells modified by 10 mM CmPEG were infected. Note, similar curves were obtained for all reaction conditions (data not shown). The standard deviation of n=3 experiments is shown. ** p < 0.001; * p < 0.05 significantly different from time matched controls (0 mM CmPEG). 101 As demonstrated in Figure 4.14, mPEG derivatization of target cells effectively prevented viral infection following a delay of up to 60 hours prior to SV40 challenge. These findings were consistent despite changes in pH, reaction time and temperature (data not shown). Based on these studies, we hypothesized that a suitable intranasal product would have a pH of 7.8, should be in place for 3-5 minutes and would effectively modify the cells at 32 °C (the mean nasal temperature). As indicated by the data, the more time that elapsed prior to SV40 challenge, the less effective the prophylaxis provided by the mPEG barrier. This loss of mPEG brush border efficacy was not the result of cell death as control cell counts were similar to those of modified cells at all time points. Based on these findings, it is anticipated that daily (24 hour intervals) application of mPEG solution would provide near maximal protection. Furthermore, should an application be missed, significant protection would remain for 48 or even 60 hours. 4.5 CmPEG Modification Prevents Infection of Cells Incubated with Mucin An additional measure taken to test the effectiveness of mPEG modification in an in v/vo-like model was to treat cells with mucin prior to pegylation. Since the ultimate goal of this project was to optimize mPEG modification for intranasal administration, factors such as mucus secretion have the potential to alter the efficacy of pegylation. In order to test the effectiveness of grafting in the presence of mucus, cells were incubated with bovine sub-maxillary mucin for 30 minutes prior to pegylation. The mucin utilized in these experiments was at pH 7.4, as normal nasal mucus secretions have been shown to have a pH range between 5.5 and 7.56 102 (Buhrmester, 1933; Fabricant, 1941; Gatto, 1981). As shown in Figure 4.15, at 24 hours post infection, 50% of control cells (with and without mucin) were infected. Mucin-treated cells modified with 2.4 mM CmPEG were 16%> infected, while those treated with 15 mM CmPEG were 6% infected. Prolonged incubation to 72 hours resulted in 99%> of control cells infected compared to 41% modified by 2.4 mM and 33% modified by 15 mM CmPEG. Interestingly, statistical analysis of these data in comparison to previous identical experiments without mucin (Figure 4.3) completed at room temperature showed no significant differences between the two. At first glance, the lack of difference between these experiments is surprising because mucin is a secreted glycoprotein with a molecular weight upwards of 1 MDa, which should theoretically compete with cell surface proteins for mPEG, however there are several plausible explanations for these results. First, when choosing a mucin concentration for these experiments I discovered that mucin alone prevented SV40 infection at high concentrations (0.01-1 mg/mL); therefore a low concentration of mucin (0.001 mg/mL) was used. This low mucin concentration likely had little effect on the pegylation efficiency of cell surface proteins. Additionally, the protein sequence of bovine sub-maxillary mucin (BSM) shown in Table 4.1 103 < S — 24 Hours •o— 48 Hours u - l 1 1 1 0 5 10 15 Concentration cmPEG 5 kDa (mM) Figure 4.15: Pegylation of Cells Prevents Infection in the Presence of Mucin Cells were incubated with mucin for 30 minutes prior to pegylation (30 minute reaction at pH 8.4, RT). At 24 hours post challenge, 50% of control cells were infected. Cells modified with 2.4 mM CmPEG were 16% infected, while those treated with 15 mM CmPEG were 6% infected. At 72 hours 99% of control cells were infected compared to 41% modified by 2.4 mM and 33% modified by 15 mM CmPEG. ** p < 0.001; * p < 0.05 104 Table 4.1: Amino Acid Sequence of Mucin BSM 1: 1589 Amino Acid Sequence (AAC39250) SESPTLSPCiVTRTTALRGSETRVPSfGVSGLPGS1"(XKiSAATGCiSGA(iS(iP'I"APVSGET RTSVISCTfNVPVSGAPVTPGSSAGSSCiAPG'IXiGiKiSETASPLSG A A GTSATG SRTS IPP SGAPVTPEPPLISTGASAGPPASSESTVLPGATGTDVLRSGTSLPVSGGAVTPA SSPCi G SSAnAGPAVCiSQ ' I 'TVQVSGAFATHVKASNTNGSSAFJSETrcATAGETLTSE'IS l V S S ATR A P S S A V T R A P V T YDDVSGISF1SSSGRS RT1VIGSPSSVSSAE Q I A P S L S T D G L E G JTKISDVDARTIRPSYGALGA 1 G S S I G E I G ITS TSPEFTETSSFSVGLRTTRPSSGETGT TLIESSTSASSSEESGTTGSIAGLRRTNRISLIRSGTT RPSSGETETTVIESRVS GSSDEG IXJT1GSTAGIA4RTTRISVVVSGTTGPSSGHTGSAVSK FRTSGSLGKGSET TVST PGLAR MTRISFCXiSRTTRQSSGETGTTVIESRT^^ SRETETTVTESRNNGSLGEGSGTTGAIAGl.TRrTRlSVVGSG'IIRPSSGETRTTVIESITRR TSAEGSQTTGSAVGLITATRISSADLQTLGPLSGETRTTV1GSGTSGKSGEVSGLTRSPAE RTI I TRISHVASGTSAPSSGMTRTTVTSGVA SRTSGLSSGEKGTSVT ETR TSGSS1EG SQTTRTADRLTMTTRTSVWSGTDAPSSGTSGTIRSSVDLTGJTKVSVIGEGTIEPSTVEL WTTEPRDLGSSTTVFSAGA1GTTRPGTSGASRPSVVGSETAGPLSAKTETTVIRSGSSGS SLEGRGTSGSTDGLTGTTTISFVGLGTTGPSARGSRPTGKGDIRSSTTVSSVDATGNIR SGGSGTTGPSIVGSETVGPSSGEAGTTVTGSGTSGKSAERLGTTVSTDRLRRTTRISLVSL GTTGPSSGVMRTTQTSIVGLETTRSSTGVLVTTSTSAESLRTTGPSPGGLWTTGTSVEGS ETTGSSTG K I T G A R R T T WHSG SHVATYEG TSG K F SKA A IS GSSHT EATTLIVSNSTS GTGLRPEDNTAVAGGQATGRVTGTTKVIPGTTVAPGSSNTESTTSLGESRTRIGRITGAT TGTSERSSPGSKTGNTGAISGTTVAPRSSNTGATTSLGSGETSQGGIKIVTMGVTTGTTI APGSSNTKAJTPTEVRTTTEVRTATETTTSRHSSDATGSGIQTGITGTGSGTTSSPGGFN AEAJTl'Kf:HVRTfi;TRII.SG'I'S"rGVGRO'fSTAVVSGRVTGVSi;SSSPGTJK^^^^ G I S T T G S T S K S N R I T T S S R I P Y P E T T V V A T G E O E T E T K T C J C T T S L P P P P A C Y G P L G E K K S P G D I W T A N C H K C T C T D A E T V D C K L K E C P S P P T C K P E E R L V K F K D N D T C C E I A Y C E P R T C LFNNNDYEVGASFADPNNPClSYSCHBT( iFVAVV0DCPK(7rWCAEEDRVYDSTKCCY TC , KPYCRSSSVNVTVNYNGCKKKVEMARCAGl*X 1 KKTIKYr)YIMI"Q)LKN^ NYEYREIDLI)CPDGGTIPYRYRHIITCSCFDICQQSMTSTVS As shown above, there are only 38 lysine (K) residues in the mucin protein sequence. Many are in close proximity to serine (S) or threonine (T) amino acids, which are commonly linked to polysaccharides. These polysaccharides may sterically hinder mPEG from reacting with nearby lysine residues. Serine and threonine residues within 3 amino acids of a lysine are underlined (e.g., ST). 105 provides valuable information on the availability of lysine residues for mPEG modification. Note, that within the sequence for BSM 1, there are only 38 lysine residues of 1589 total amino acids (-2%). These lysines are generally found in close proximity to serine and threonine residues that are points of attachment for O-linked saccharides, which constitute between 50 and 80% of the molecular weight of mucin (Bhargava, et. al, 1990; Jiang, et. al, 1998). There are hundreds of saccharides attached to the core mucin protein that can vary in length from disaccharides to large oligosaccharides of approximately 20 monosaccharides (Jiang, etal, 1998), and these sugars may sterically hinder mPEG binding to adjacent lysine residues resulting in levels of protection against viral infection similar to those seen in the absence of mucin. 4.6 mPEG Inhibits SV40 Entry To determine if pegylation of the host cells prevented viral binding and entry, immunostaining for VP1 (Viral protein 1) was performed within 3 hours of viral challenge. VP1 is the major capsid protein of SV40 and is not expressed until late (18-20 hours post infection) in the infectious process (Fields, 1998). As such, any VP1 detected within the first 3 hours of infection is the result of SV40 binding to and subsequently entering cells. For this experiment, cells were infected with SV40 and incubated for up to 3 hours. At each time point, cells were thoroughly washed prior to methanol fixation to prevent exogenous virus from entering the cell during permeablization as described in the methods. 106 As shown in Figure 4.16, cells modified with 5 mM CmPEG and infected with 1010 pfu/mL were 0.2%, 0.2% and 0.4% VP1 positive at 1, 2 and 3 hours respectively. In sharp contrast, unmodified cells were 21%, 18% and 13% infected at 1, 2 and 3 hours. The decrease in VP1 positivity noted in the control cells likely resulted from the uncoating of the SV40 capsid subsequent to cell entry. As shown in Chapter 2, VP1 is not detectable in cells at 6 hours post SV40 challenge due to the rapid degradation upon uncoating. The disparity in the ability of SV40 to enter modified versus unmodified cells was readily observed microscopically (Figure 4.17). Note the uninfected control cells showed no fluorescence at any time point, while unmodified infected cells were observed to have small round positively stained vesicles resulting from the endocytosis of SV40. Cells modified with 5 mM CmPEG had significantly fewer cells with SV40 filled endosomes. A second set of controls of both modified and unmodified cells were stained for actin to demonstrate the permeability of both cell populations (control and pegylated). As noted in Figure 4.18, no difference in the actin staining intensity between the two populations was observed. 107 1E+09 1E+10 1E+11 V i r a l D o s e ( p f u / m L ) Figure 4.16: C m P E G Modification of Cells Inhibits SV40 Entry VP1 immunostaining was completed at 1, 2 and 3 hours following SV40 infection. Cells modified with 5 mM CmPEG and infected with 1010 pfu/mL were 0.2%, 0.2% and 0.4% VP1 positive at 1, 2 and 3 hours respectively. In contrast, unmodified cells were 21%, 18% and 13% positive at 1, 2 and 3 hours. The standard deviation of n=3 independent experiments is shown. * p < 0.05; ** p < 0.001 significantly different from time matched controls (0 mM CmPEG). 108 Unmodified Cells 2 Hours Post Infection Phase Contrast 5 mM Modified Cells 2 Hours Post Infection Phase Contrast Unmodified Cells 2 Hours Post Infection 5 mM Modified Cells 2 Hours Post Infection UV Fluorescence UV Fluorescence Figure 4.17: A Microscopic Look at SV40 Entry The picture at top-left was representative of unmodified CV-1 cells 2 hours post infection with SV40 shown under phase contrast. The arrowheads point out endosomes stained positive for VP1, which was shown in the bottom-left picture. None of these vesicles were found in the pictures of cells modified by 5 mM CmPEG prior to infection top and bottom-right. 109 0 mM CmPEG 5 kDa 5 mM CmPEG 5 kDa Figure 4.18: Actin Staining of Modified and Unmodified Cells Cells were PEGylated with 0 or 5 mM CmPEG 5 kDa and infected with SV40. Cells were stained using a protocol similar to VP1 staining using anti-actin antibodies rather than anti-VPl antibodies. As shown above, actin is prevalent in the cytoplasm of both unmodified and mPEG modified cells. Therefore the antibodies were not inhibited from entering PEGylated cells following methanol permeablization. 4.7 Discussion The pegylation of cells is a novel prophylactic approach in the prevention of viral infections. It was hypothesized that mPEG modification of virus target cells would prevent viral infection based on our laboratory's earlier work on red blood cells, lymphocytes and ongoing work with pancreatic islet cells where pegylation of these cells was shown to prevent protein-protein and cell-cell interactions. Because viral infection of cells is highly dependent upon receptor-ligand interactions for entry, mPEG derivatization of host cells should similarly prevent these interactions. 110 Indeed, as shown throughout this chapter, modification of target cells prevents viral infection. Significantly, the derivatization reaction which was initially 30 minutes would be difficult for humans to comply with. Importantly, it was found that a 3-5 minute reaction time provided near maximal protection against viral infection resulting in improved compliance over that achievable with a 30-minute application. Furthermore, a decrease in the pH of the derivatization reaction from 8.4 to 7.8 had little effect on the levels of mPEG protection afforded. In fact, a reaction pH of 7.8 appeared to better prevent infection suggesting the higher pH caused slight cell injury (e.g., cell contraction). Even the smallest change in cell shape or structure following pegylation could result in a gap in the protective barrier allowing viral entry and subsequent infection. Additionally, increasing the mPEG reaction temperature to that of the nasopharyngeal cavity (32 °C) resulted in similar levels of protection to reactions completed at room temperature. Another important factor in compliance with a final mPEG nasal spray is the dosage schedule. It was originally hypothesized that application of mPEG would need to occur every 12-48 hours in order to maintain the protective barrier. This supposition was supported by the data presented in this chapter, indicating that a once daily (24 hour) dosage schedule is appropriate for optimal maintenance of a highly protective mPEG barrier against viral infection. Interestingly, significant protection was seen even following a 48-60 hour delay prior to viral challenge. The protection afforded following this lengthy delay suggests that if a dose is missed, significant protection against viral infection would remain. I l l The once-daily dosage schedule based on our in vitro model is likely to translate well into human nasopharynx because the epithelial cells of the upper respiratory tract are renewed every 3 days (Carson et al., 1981). However, our model is much less complex than an animal or human model beginning with the cells used. The CV-1 cell line contains both epithelial and fibroblast cells while the upper respiratory tract would also include secreting cells, nerve fibers, and ciliated epithelium. It is possible that pegylating these cells may somewhat diminish the sense of smell during treatment, but it would return to normal within a few days as the pegylated proteins turnover. It is worth noting that the sense of smell is tremendously impacted during a typical common cold infection as well. Additional complexity was added to the initial in vitro experiments in order to more accurately reproduce nasal cavity conditions. Because the nasal cavity contains complex secretions such as mucus, cells were incubated with mucin prior to pegylation to test efficacy in this more protein-rich environment. As shown in this chapter, the presence of mucin did not alter mPEG protection following a 30 minute reaction at room temperature suggesting the presence of mucus may have minimal adverse effects on the mPEG reaction in vitro. However, these experiments did not account for continuous nasal discharge during pegylation, which may hinder modification ofthe nasopharynx. Additionally, there are other components of nasal discharge that remain unknown factors that may reduce the efficacy of pegylation. Should this be the case, it is possible that simply increasing the concentration of mPEG used would overcome this loss of pegylation efficacy. 112 Finally, it is evident from my studies that the covalent attachment of mPEG to cells prevented SV40 entry. There were questions as to whether the virus was being trapped within the mPEG brush border or interfering with virus uncoating or transport to the nucleus once inside the mPEG modified cell. However, VP1 immunostaining demonstrated that while virus entered unmodified cells, entry into the pegylated cells was dramatically reduced. Hence, the modification of cells appeared to prevent virus-receptor interactions and/or membrane interactions required for endocytosis of the virus to occur. All of the data presented in this chapter supports a daily nasal application of an mPEG solution for 3-5 minutes as a means of preventing viral infection. Such an application would optimally protect the nasal epithelium from infection by a wide range of virus concentrations for 24 hours. Additionally, should a daily application be missed, protection would remain for 48 to 72 hours. As discussed in the next chapter, enhancement of mPEG protection against viral infection may be possible by altering the linker moiety and/or polymer size. 113 Chapter 5: Comparison of the Effects of m P E G Linker Chemistry and Polymer Size on Viral Infection Chapter Overview •: Modification Of Host Cells with SPAmPEG or BTCmPEG Prevents SV40 Infection: Do changes in linker chemistry and polymer size affect mPEG protection against viral infection? '.. . . • Combination Studies: Can better protection be achieved using mixtures of polymer molecular weights? 5.1 Pegylation of Cells with SPAmPEG Prevents SV40 Infection As shown in the previous chapter, mPEG modification of cell surface proteins significantly inhibited SV40 infection. The next step in the optimization of pegylation was to determine if other linker chemistries or polymer sizes could provide better protection against viral infection. The first linker moiety tested was succinimidyl propionate (SPAmPEG 2, 5, and 20 kDa). As with the early experiments performed using CmPEG 5 kDa, the first infection assays performed for SPAmPEG 2 kDa involved a 30-minute reaction at pH 8.4 conducted at room temperature on confluent CV-1 cells. These reaction conditions were used in two sets of experiments. The first was performed with only one concentration of virus and the second with a 4-log range in titers (106- 2x1010). 114 Table 5.1 displays the data from the SPAmPEG 2, 5, and 20 kDa cell derivatization studies. Each SPAmPEG molecular weight gives rise to a dose dependent reduction in the percent cells infected similar to the results seen in Chapter 4 with CmPEG. At 24 hours post viral challenge unmodified cells were 51% positive for T antigen, while 24% and 6% of cells modified by 2.4 and 15 mM SPAmPEG 2 kDa were infected, respectively (pO.OOl). At 48 hours, unmodified cells were 73% T antigen positive while only 44% and 36% of 2.4 and 15 mM SPAmPEG 2 kDa modified cells were positive (pO.OOl). By 72 hours post exposure, unmodified cells were 99% infected, while 61% and 55% of cells modified by 2.4 and 15 mM were infected (pO.OOl). In comparison, derivatization with CmPEG 5 kDa resulted in infection rates of 9% and 5% at 24 hours (2.4 and 15 mM) and 28% and 24% at 72 hours post challenge. These results demonstrated that cells modified by SPAmPEG 2 kDa were significantly protected from viral infection. However, the degree of protection appeared to be significantly (pO.01) less than that afforded by the 5 kDa CmPEG. Similar results were obtained with SPAmPEG 5 and 20 kDa as shown in Table 5.1. 115 Table 5.1: SPAmPEG 20 kDa More Effectively Prevents Progeny Virus Infection than 2 or 5 kDa 24 Hours SPAmPEG 0 m M 0.2mM 0.6 m M 1.2 m M 2.4 m M 5 m M 10 mM 15 m M 2 kDa 50.7 41.3 3 3 36.7 3 3 31.7 9 3 23.6 3 3 9.5 3 3 9.6 3 3 6.2 3 3 ±1.8 ±8.5 ±7.1 ±10.9 ±10.1 ±4.6 ±4.4 ±3.9 * * , ** 50.1 ' ?33.r 3 i ' ""20.8 3 3 v 13.43 3 1-2.73" 10.23 3 ±1.6 . i .\V±3*:0,{.:i /:;.±2v6;': i ;-^±2.0; ' ±2.7;;' • ±3.2.\>:>; ±3.0 20 kDa 50.4 40.3 3 3 24.1 3 3 24.3 3 3 14.43 3 8.93 3 7.1 3 3 2.6 3 3 ±2.2 ±3.5 ** ±3.4 ±3.3 ±3.3 ±3.0 ±2.5 ±1.4 t t t t t 48 Hours SPAmPEG 0 mM 0.2 mM 0.6 mM 1.2 mM 2.4 mM 5 mM 10 mM 15 mM 2 kDa 72.8 61.7 57.1 3 3 46.2 3 3 42.3 3 3 46.8 3 3 63.4 3 3 35.7 3 3 ±2.3 ±3.4 ±11.9 ±5.5 ±5.5 ±25.3 * ±12.9 ** ±18.6 t t t t t t t t "'5td>a V'V-: * •..•75:9- 70.2 3 51.1 3 3 > 43.8 3 3 42.6" ;*-36:933 -. :' '35;o:33'':i;' 27.2" : ; ' - . . ' • " • • ±2.8 • ,-. '±3.5 , ±3.5 .... -•-4.8 . ±8,1, • ' ' ±5'.4 . •. ±6.5 ' . ±7.5 20 kDa 75.3 66.9 3 3 51.6 3 3 45.0 3 3 31.9 3 3 20.8 3 3 V 13.633 6.7 3 3 ±3.8 ±2.9 ±3.1 ±3.4 ±5.8 ±2.7 ±3.4 ±2.5 * ** ** ** t t f t t t t t 72 Hours SPAmPEG 0 mM 0.2mM 0.6 mM 1.2 mM 2.4 mM 5 mM 10 mM 15 mM 2 kDa 99.2 76.0 3 3 62.333 64.1 3 3 61.1 3 3 58.7 3 3 65.0 3 3 55.5 3 3 ±0.5 ±12.8 ±9.0 ±11.9 ±11.7 ±4.3 ±6.4 ±12.6 ** * ** ** ** ** t t t t t t 5 kDa 99.3 89.3 3 3 " 70.4 3 3 1 ' 63.4 3 3 . .50.8'33 47.7 3 3 . 49.6 3 3 41.3 s91'' ±0.6, ' ±1\8'" • ±3.2 ±2.7 ±3.4 ' ±3.4 ±13.2 ±5.4 20 kDa 99.3 65.9 3 3 64.4 3 3 59.9 3 3 49.0 3 3 43.6 3 3 23.0 3 3 18.033 ±0.5 ±14.3 ±4.2 ±2.3 ±3.0 ±3.4 ±3.8 ±3.2 ** ** ** ** ** ** t t t t f t The average percent of infected cells is shown for n=3 independent experiments, ± the standard deviation. Note at 24 hours the only significant differences between the three molecular weights occur at low mPEG concentrations (0.2-2.4 mM). At 72 hours post challenge, SPAmPEG 20 kDa modified cells (0.2, 10 and 15 mM) were significantly less infected than either the 2 or 5 kDa polymer. * Significantly different from SPAmPEG 5 kDa at equimolar concentrations (* p < 0.001; * p < 0.05). f Significant differences between SPAmPEG 2 and 20 kDa at equimolar concentrations (f f p < 0.001; f p < 0.05). 3 Significantly different from the control for that mPEG (0 mM) at that particular time point (33p< 0.001,3p< 0.05). 116 Once it was shown that SPAmPEG modification of cells prevented SV40 infection, the efficacy of protection of each molecular weight was tested over a 4-log scale viral dose range (106-2xl0 1 0 pfu/mL). As shown in Figure 5.1A at 24 hours, control cells challenged with the highest concentration of virus (2xl0 1 0pfu/mL) were 78% infected, while those modified by 1.2 and 15 mM SPAmPEG 2kDa were 45% and 38% infected (pO.OOl). Experimental rhinovirus infections of humans utilize between 3 and 3000 times the tissue culture infectious dose 50 (TCID50) to ensure infection of the subjects, which in my experiments is equivalent to 109-2xl0 1 0 pfu/mL (Gwaltney, et al, 1997 and 2002). These viral concentrations are thought to be much higher than real world conditions; therefore more lower viral loads (106-107 pfu/mL) were also tested in our SV40 tissue culture model. Control cells challenged with 10 pfu/mL were 2% T antigen positive at 24 hours compared to 0.02% and 0%> positive following treatment with 1.2 and 15 mM SPAmPEG 2 kDa (pO.OOl). By 72 hours (Figure 5.2A), control cells challenged with 107 pfu/mL were 30%> infected, while those modified by 1.2 and 15 mM SPAmPEG were 4% and 1% infected (pO.OOl). Similar results were obtained from experiments utilizing the 5 kDa or 20 kDa SPAmPEG and are displayed in Figure 5.1B and C. At 24 hours control cells challenged with the highest concentration of virus were 75% infected, while those modified by 1.2 and 15 mM SPAmPEG 5 kDa were 36% and 23% infected (pO.OOl; Figure 5.1B). In comparison, 61% and 18% of cells modified by 1.2 and 15 mM SPAmPEG 20 kDa and challenged with 2xl0 1 0 pfu/mL of virus were infected at 24 hours, while control cells were 76% infected (pO.OOl; Figure 5.1C). Using lower viral concentrations (107 pfu/mL) control cells were 4% T antigen positive at 24 hours compared to 0.1%> and 0% positive following treatment with 1.2 and 15 mM 117 A. Modification of Cells with SPAmPEG 2 kDa 100 75 u c 50 o 0* 25 4 m— OmM o— 0.2 mM a — 1.2 mM «— 5mM T — 15 mM 0 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) B. Modification of Cells with SPAmPEG 5 kDa C. Modification of Cells with SPAmPEG 20 kDa 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) D. Modification of Cells with CmPEG 5 kDa 100 75 u U C 50 I) o 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 25 04 — • — 0 mM —o— 0.2 mM —A— 1.2 mM — » — 5 mM —V— 15 mM 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) S V 4 0 Concentration (pfu/mL) Figure 5.1: Modification of CV-1 Cells with SPAmPEG 2, 5 or 20 kDa Prevents Viral Infection over a Wide Range of Virus Titers, 24 hours A: Control cells infected with 107 pfu/mL were 2% T antigen positive at 24 hours compared to 0.02% and 0% positive following treatment with 1.2 and 15 mM SPAmPEG 2 kDa. B: Control cells inoculated with 107 pfu/mL were 4% T antigen positive at 24 hours compared to 0.1 % and 0%> positive following treatment with 1.2 and 15 mM SPAmPEG 5 kDa. C: Control cells infected with 107 pfu/mL were 5% T antigen positive at 24 hours compared to 0.03% and 0% positive following treatment with 1.2 and 15 mM SPAmPEG 20 kDa. D: Data from CmPEG 5 kDa modification of CV-1 cells from Chapter 4 was shown here for comparison purposes. The standard deviation of n=3 independent experiments was shown. ** p < 0.001; * p < 0.05 compared to viral dilution controls exposed to 0 mM mPEG. 118 A. Modification of Cells with SPAmPEG 2 kDa C. Modification of Cells with SPAmPEG 20 kDa 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) B. Modification of Cells with SPAmPEG 5 kDa 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) D. Modification of Cells with CmPEG 5 kDa 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 1 E + 0 3 IE+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) SV40 Concentration (pfu/mL) Figure 5.2: Modification of CV-1 Cells with SPAmPEG 2, 5 or 20 kDa Prevents Viral Infection over a Wide Range of Virus Titers, 72 hours A: Control cells inoculated with 107 pfu/mL were 30% infected, while those modified by 1.2 and 15 mM SPAmPEG 2 kDa were 4% and 1% infected. B: Control cells infected with 107 pfu/mL were 28% infected, while those modified by 1.2 and 15 mM SPAmPEG 5 kDa were 6% and 0.5% infected. C: Control cells infected with 107 pfu/mL were 3% infected, while those modified by 1.2 and 15 mM SPAmPEG were 2% and 0.04% infected. D: Data CmPEG 5 kDa modification of CV-1 cells from Chapter 4 was shown here for comparison purposes. The standard deviation of n=3 independent experiments was shown. ** p < 0.001; * p < 0.05 compared to viral dilution controls exposed to 0 mM mPEG. 119 SPAmPEG 5 kDa (pO.OOl). By 72 hours (Figure 5.2B), control cells challenged with 107 pfu/mL were 28% infected, while those modified by 1.2 and 15 mM SPAmPEG were 6% and 0.5% infected (pO.OOl). In contrast, cells modified by 20 kDa SPAmPEG (1.2 and 15 mM) were 0.03% and 0% T antigen positive 24 hours post challenge with 107 pfu/mL while control cells were 5% positive. Seventy-two hours post challenge (Figure 5.2C), control cells exposed to 107 pfu/mL were 23% infected, while 2% and 0.04% of those modified by 1.2 and 15 mM SPAmPEG 20 kDa were infected (pO.001). Results obtained from the experiments with SPAmPEG showed that modification of cells with any of the three molecular weights prevented infection. As shown in Table 5.1, at 24 hours the only significant differences between the three molecular weights occurred at low mPEG concentrations (0.2-2.4 mM). Interestingly, in contrast to the red cell immunocamouflage studies of Bradley, et al. (2002), small polymer size (2 kDa) did not exhibit any consistent effect on antiviral efficacy at low (<1.2mM) moderate (1.2-2.4 mM) or high (>5 mM) SPAmPEG derivatization concentrations. In contrast, SPAmPEG 20 kDa modification of cells was more effective at preventing progeny virus infection. For example as shown in Table 5.1, at 72 hours post viral challenge, cells modified with high concentrations (10-15mM) of SPAmPEG 20 kDa were significantly less infected than either of the smaller (2 or 5 kDa) polymers suggesting progeny virus is unable to propagate the infection. It is thought that these viruses became trapped within mPEG-coated vesicles or just below the mPEG brush border. 120 5.2 Pegylation of Cells with BTCmPEG Prevents SV40 Infection As shown throughout both this chapter as well as Chapter 4, mPEG modification of cell surface proteins inhibits SV40 infection. To further optimize pegylation of cells to inhibit viral infection, benzotriazole carbonate mPEG (BTCmPEG) 3.4, 5, and 20 kDa were tested. As with the experiments described above, the infection assays performed for BTCmPEG involved a 30-minute reaction at room temperature (pH 8.4) on confluent CV-1 cells. Table 5.2 presents the data obtained from the BTCmPEG 3.4, 5, and 20 kDa experiments. Note that as with the CmPEG experiments from Chapter 4, and the SPAmPEG experiments described above, there was a dose dependent reduction in cells modified by each molecular weight BTCmPEG. For example, at 24 hours post challenge, unmodified cells were 51% positive for T antigen, while 7% and 0.5%> cells modified by 2.4 and 15 mM BTCmPEG 3.4 kDa were infected respectively (p<0.001). At 48 hours, unmodified cells were 67%) positive while only 12% and 2%o of 2.4 and 15 mM BTCmPEG 3.4 kDa modified cells were positive (pO.OOl). By 72 hours post SV40 challenge, unmodified cells were 99%) T antigen positive, while only 28% and 11% of cells modified by 2.4 and 15 mM were infected (pO.OOl). Interestingly, when compared to 5 kDa BTCmPEG, there was a trend toward increased protection against viral infection with the 3.4 kDa polymer. For example, as shown in Table 5.2, at 24 hours post SV40 challenge unmodified cells were 49%> positive for T antigen, while 14%o and 7% of cells modified by 2.4 and 15 mM BTCmPEG 5 kDa were infected, respectively (pO.OOl). At 48 hours, unmodified cells were 76% infected while only 32% and 15% of 2.4 and 15 mM BTCmPEG 5 kDa modified cells were positive (pO.OOl). By 72 hours post 121 Table 5.2: BTCmPEG 20 kDa is Less Effective at 24 hours but More Effective at 72 hours than 3.4 or 5 kDa 24 Hours BTCmPEG 0 mM 0.2mM 0.6 mM 1.2 mM 2.4 mM 5 mM 10 mM 15 mM 3.4 kDa 50.7 19.5 3 3 12.0 3 3 10.2 3 3 7.4 3 3 4.4 3 3 1.833 0.5 3 3 ±2.0 ±3.6 ±3.7 ±2.6 ±2.0 ±1.4 ±1.2 ±0.4 ** ** t t t t t t t t 5kl)a 49.2 25 .7 3 3 ' 16.7 " 16.8 3 3 14.2 3 3 17.6 3 3 8.0 3 3 7 .1 3 3 . . -l... 1F' ±13.0 ±6.4 • • '±6 :0 . y'a (6.1 ±1.3 ' ±1.6 • ±0.8 ±2.6 . 20 kDa 52.2 39 .0 3 3 34 .1 3 3 30 .1 3 3 19.6 3 3 17.3 3 3 15.2 3 3 14.1 3 3 ±10.1 ±2.8 ±5.3 ±2.8 ±12.3 ±11.5 ±10.4 ±9.9 ** ** ** ** ** ** ** t t t t t t f t 4 I Hours BTCmPEG 0 mM 0.2mM 0.6 mM 1.2 mM 2.4 mM 5 mM 10 mM 15 mM 3.4 kDa 67.3 32.9 3 3 19.8 3 3 16.2 3 3 12.2 3 3 9 .1 3 3 5.5 3 3 2 .5 3 3 ±4.2 ±5.2 ±4.4 ±5.5 ±3.5 • ±2.1 ±1.0 ±1.4 ** ** ** ** ** * ** t t t t t t t t t t t f f t 5 kDa > 75'.7 44.4 3 3 32.6 3 3 33.2 3 3 32.2 3 3 25.9 3 3 15.6 3 3 • 14.6 3 3 ':"±i'2i8 ±11.4 ±14.0 ±11.1 ±'5.7 ' ±5.3 ±5.3 ' '±2.6 20 kDa 75.8 54 .5 3 3 45.6 3 3 37 .1 3 3 27.7 3 3 25.6 3 3 20.6 3 3 18.3 3 3 ±3.5. ±4.5 # ±3.1 ** ±2.4 ±2.1 ±2.6 ±1.2 ±1.4 t f t t t t t t t t t t . t t 72 Hours BTCmPEG 0 mM 0.2 mM 0.6 mM 1.2 mM 2.4 mM 5 mM 10 mM 15 mM 3.4 kDa 98.7 50.4 3 3 42 .1 3 3 42 .3 3 3 27.8 3 3 26 .1 3 3 14.3 3 3 10.8 3 3 ±1.0 ±3.8 ±6.9 ±5.2 ±3.9 ±3.2 ±3.0 ±2.2 ** ** * ** ** t t t t 5 kDa 98.9 , 62 .8 3 3 . . 43 .6" 37.8* , 41 .8 3 3 3 4 : 3 3 3 25 .6 3 3 ' 21 .0 3 3 . ±0.4 ±15.1 \ ±3:8 '.\ ±7 .9 . ' - ±9.4 ±6.1 : ±2.7 • 20 kDa 99.5 66 .9 3 3 58 .1 3 3 44.8 3 3 34 .1 3 3 26.0 3 3 17.5 3 3 12.5 3 3 ±0.5 ±2.6 ±3.2 ±2.7 ±3.0 ±2.1 ±1.4 ±1.5 ** * ** * * t t t t The average percent of infected cells is shown for n=3 independent experiments, ± the standard deviation. Note at 24 hours BTCmPEG 20 kDa was significantly less effective in preventing initial SV40 infection. However, at 72 hours post challenge, BTCmPEG 20 kDa modified cells (0.2, 10 and 15 mM) were significantly less infected than those modified by the 5 kDa molecular weight and approximately equal to those modified by 3.4 kDa BTCmPEG. * Significantly different from BTCmPEG 5 kDa at equimolar concentrations (** p < 0.001; * p < 0.05). f Significant difference between BTCmPEG 3.4 and 20 kDa at equimolar concentrations (tf p < 0.001; f p < 0.05). "Significantly different from the control for that mPEG (0 mM) at that particular time point (33 p< 0.001, 3p< 0.05). 122 exposure, unmodified cells were 99% T antigen infected, while 42% and 21%> of cells modified by 2.4 and 15 mM were infected (pO.OOl). Next, to determine if a larger polymer would provide superior protection against viral infection similar experiments were conducted with the BTCmPEG 20 kDa. As shown in Table 5.2, at 24 hours post viral challenge 20 kDa modified cells were infected at rates comparable to those modified by the lower molecular weight BTCmPEGs. Interestingly, at 72 hours the data suggests a failure of progeny virus to infect other cells within the monolayer as 14% of cells modified by 15 mM were infected at 24 hours compared to 13%> at 72 hours. Once it was determined that BTCmPEG 3.4, 5, and 20 kDa modification of cells prevented SV40 infection, the efficacy of the mPEG protection was challenged with a wide range of viral loads. As shown in Figure 5.3A, at 24 hours, control cells challenged with the highest concentration of virus (2xl0 1 0pfu/mL) were 76%> infected, while those modified by moderate (1.2mM) and high (15 mM) concentrations of BTCmPEG 3.4 kDa were 30% and 19% infected respectively (pO.OOl). Figure 5.3B depicts BTCmPEG 5 kDa modified cells at 24 hours, with control cells challenged with the highest concentration of virus 76% infected, while those modified by 1.2 and 15 mM BTCmPEG 5 kDa were 32% and 18% infected (pO.OOl). Similar results are illustrated in Figure 5.3C when BTCmPEG 20 kDa was utilized; at 24 hours, control cells challenged with 2xl0 1 0 pfu/mL were 77% infected, while those modified by 1.2 and 15 mM BTCmPEG 20 kDa were 34% and 15% infected (pO.OOl). 123 A. Modification of Cells with BTCmPEG 3.4 C. Modification of Cells with BTCmPEG 20 kDa Prevents Infection kDa Prevents Infection 100 75 1 50-1 25 04 — O m M - o — 0.2 m M -A— 1.2 m M _ • — 5 m M - * — 15 m M 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) B. Modification of Cells with BTCmPEG 5 kDa Prevents Infection 100 £ 75 U 50 4 25 0 — • — O m M — o — 0.2 m M —A— 1.2 m M — • — 5 m M — T — 15 m M 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) 100 75 4 O n 50 25 4 - B — O m M - o — 0.2 m M ~A— 1.2 m M - » — 5 m M - » — 15 m M 0 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) D. Modification of Cells with CmPEG 5 kDa Prevents Infection 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) Figure 5.3: Modification of CV-1 Cells with B T C m P E G 3.4, 5 or 20 kDa Prevents Viral Infection over a Wide Range of Virus Titers, 24 hours. A: Control cells infected with 107 pfu/mL were 6% T antigen positive at 24 hours compared to 0.18% and 0% positive following treatment with 1.2 and 15 mM BTCmPEG 3.4 kDa. B: 24 hours, control cells infected with the highest concentration of virus were 76% infected, compared to were 32% and 18% (1.2 and 15 mM BTCmPEG 5 kDa). C: Control cells infected with 107 pfu/mL were 7% infected compared to 0.04% and 0% positive following treatment with 1.2 and 15 mM BTCmPEG 20 kDa. D: Data CmPEG 5 kDa modification of CV-1 cells from Chapter 4 was shown here for comparison purposes. The standard deviation of n=3 independent experiments was shown. ** p < 0.001; * p < 0.05 compared to viral dilution controls exposed to 0 mM mPEG. 124 At more lower viral concentrations (107 pfu/mL), control cells were 6% T antigen positive at 24 hours compared to 0.18% and 0% positive following treatment with 1.2 and 15 mM BTCmPEG 3.4 kDa (p<0.001). By 72 hours (Figure 5.4A), control cells inoculated with 107 pfu/mL were 27% infected, while those modified by 1.2 and 15 mM BTCmPEG 3.4 kDa were 6% and 0.6%> infected (p<0.001). Similar results were obtained with BTCmPEG 5 kDa; at 24 hours control cells challenged with 107 pfu/mL were 4%> T antigen positive compared to 0%> antigen positive following treatment with either 1.2 or 15 mM BTCmPEG 5 kDa (pO.OOl). By 72 hours (Figure 5.4B), control cells challenged with 107 pfu/mL SV40 were 27% infected, while those modified by 1.2 and 15 mM BTCmPEG 5 kDa were only 2% and 0.05% infected (pO.OOl). Additionally, comparable results were acquired following derivatization of cells with BTCmPEG 20 kDa. Further demonstrating the potency of mPEG grafting, control cells challenged with 107 pfu/mL were 7%> T antigen positive at 24 hours compared to only 0.04% and 0%> positive following treatment with 1.2 and 15 mM BTCmPEG 20 kDa (pO.OOl). Interestingly, the 20 kDa polymer shows almost no increase in infection rates between 24 and 72 hours (Figure 5.4C), with cells modified by 1.2 and 15 mM BTCmPEG 20 kDa were only 2%o and 0.04%> infected, while control cells challenged with 107 pfu/mL were 27%> infected (pO.OOl). 125 A . Modification of Ceils with B T C m P E G 3.4 kDa Prevents Infection C . Modification of Cells with B T C m P E G 20 kDa Prevents Infection 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) B. Modification of Cells with B T C m P E G 5 kDa Prevents Infection 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) D . Modification of Cells with C m P E G 5 kDa Prevents Infection 100 4 1E+03 1E+04 1E+05 1E+06 1E+07 1E+08 1E+09 1E+10 1E+11 SV40 Concentration (pfu/mL) Figure 5.4: Modification of CV-1 Cells with B T C m P E G 3.4, 5 or 20 kDa Prevents Viral Infection over a Wide Range of Virus Titers, 72 hours. A: Control cells infected with 107 pfu/mL were 27% infected, while those modified by 1.2 and 15 mM BTCmPEG 3.4 kDa were 6% and 0.6% infected. B: Control cells infected with 107 pfu/mL compared to 2% and 0.05% 27% (1.2 and 15 mM BTCmPEG 5 kDa were infected). C: Control cells infected with 107 pfu/mL were 27% infected, compared to 2% and 0.05% (1.2 and 15 mM BTCmPEG 20 kDa). D: Data CmPEG 5 kDa modification of CV-1 cells from Chapter 4 was shown here for comparison purposes. The standard deviation of n=3 independent experiments was shown, p < 0.001; *.p < 0.05 compared to viral dilution controls exposed to 0 mM mPEG. 126 As shown in Table 5.2, BTCmPEG 20 kDa was significantly less effective in preventing initial SV40 infection (24 hour time point). As expected, cells modified by 3.4 or 5 kDa BTCmPEG exhibited higher rates of infection after the 72 hour incubation than they had at 24 hours due to the production of progeny virus, suggesting the infection proceeded as normal in these cells. In contrast, infection rates of cells modified by BTCmPEG 20 kDa remained static over the entire period of study following derivatization with concentrations greater than 5 mM as clearly documented in Table 5.2, indicating the progeny virus were unable to propagate the infection. The lack of progeny virus infection suggested that virus was either trapped in mPEG modified vesicles or became trapped by the mPEG brush border while exiting the cells. 5.3 Combination Studies As in Chapter 3, it was hypothesized that combining mPEGs of different molecular weights would increase the level of protection against viral infection. The enhanced protection would arise as a consequence of an improved brush border. The combinations tested consisted of equimolar combinations (3.4/5 kDa, 3.4 /20 kDa, and 5 /20 kDa) of BTCmPEG. As shown in Table 5.3, the data from each of the three combinations is compared to the data obtained form BTCmPEG 5 kDa. First, the BTCmPEG 3.4/5 kDa combination was examined. As presented in Table 5.3, at 24 hours post infection unmodified cells were 50% positive for T antigen, while 12%> and 0.2%> of cells modified by 2.4 and 15 mM BTCmPEG 3.415 kDa mixture were infected, respectively (pO.OOl). Using the combination of BTCmPEG 3.4 120 kDa yielded similar results at 24 127 Table 5.3: BTCmPEG Combinations More Effectively Prevent SV40 Infection than Single Molecular Weights The average percent of infected cells is shown for n=3 independent experiments, ± the standard deviation. Note there are few differences between the BTCmPEG concentrations 24 hours post challenge, though the combination of 5 and 20 kDa is slightly more effective. However the combinations are more effective at both 24 and 48 hours than individual molecular weights. Once again at 72 hours the combination of 5 and 20 kDa appears slightly more effective than the other combinations. * Significantly different from BTCmPEG 3.4/5 kDa at equimolar concentrations (** p<0.001; *p<0.05) t Significantly different from BTCmPEG 3.4/20 kDa at equimolar concentrations (ft p < 0.001; fp<0.05) % Significantly different from BTCmPEG 5/20 kDa at equimolar concentrations (ft p< 0.001; $p< 0.05) § Significantly different from BTCmPEG 5 kDa at equimolar concentrations (§§ p< 0.001; § p < 0 . 0 5 ) 3 Significantly different from the control for that mPEG (0 mM) at that particular time point (33p<0.001,3p< 0.05). (TABLE ON FOLLOWING PAGES) 128 Table 5.3: B T C m P E G Combinations More Effectively Prevent SV40 Infection than Single Molecular Weights 24 Hours B T C m P E G 0 m M 0.2mM 0.6 m M 1.2 m M 2.4 m M 5 mM 10 m M 15 m M 3.4/5 50.4 19.2 99 15.1 33 12.2 33 5.5 33 2.6 33 0.6 33 0.2 33 ±2.0 ±3.6 ±2.8 ±2.0 ±1.9 ±1.5 ±0.6 ±0.3 §§ §§ § § 3.4 / 20 46.9 25.7 33 19.0 33 15.8 33 11.9 33 5.9 99 3.4 99 1.9 99 ±6.6 ±13.9 ±14.7 ±11.3 ±6.9 ±2.9 ±3.0 ±1.2 tt t §§ 5/20 42.5 15.1 33 11.8 33 10.0 33 8.4 33 7.6 99 5.0 99 2.0 99 ±6.3 ±5.6 ±5.5 ±5.8 ±3.9 ±2.9 ±1.9 ±1.3 t t t § §§ 5 kDa ••T' 49.2 • 25.7 33• 16.8 33 14.2 39 17.6 33 8.0 33 \ 7.1 33 -13.0 .;, ±6 ,4 : +6.0 " ±6.1 '. • x l .3 t 1.6 • ± 0 . 8 . ± 1 . 0 : ' 4 1 Hours B T C m P E G 0 m M 0.2mM 0.6 m M 1.2 m M 2.4 m M 5 m M 10 mM 15 m M 3.4/5 72.3 28.8 33 24.5 33 20.1 33 14.8 93 6.4 33 2.9 33 1.3 33 ±3.5 ±3.4 ±3.7 ±3.0 ±2.7 ±2.7 ±1.5 ±0.9 ' i t t §§ §§ §§ §§ §§ §§ 3.4 / 20 67.2 34.5 33 31.3 33 27.3 33 24.0 33 15.0 33 8.3 39 4.7 99 ±7.8 ±14.7 ±13.0 ±14.0 ±12.8 * ±3.2 * ±2.3 ±2.2 tt tt tt tt § §§ § 5/20 63.7 18.0 33 16.2 33 15.8 33 11.4 33 11.6 99 8.8 99 7.5 99 ±9.7 ±6.9 * ±5.6 ±5.5 ±4.4 ±3.8 ±3.8 ±3.3 t t t t t t t t §§ §§ §§ §§ §§ 5 kDa 75'.7 l 44.439 32.6.39'. 33.2 33 , 32.2 33 ff 25.9 3 3 \ ; 15.6 33 14.6 99 ±12:8 ±11.4 ±14.0 ±11.1 ±5.7 • ±5.3 ' ' ±5.3 ±2.6 129 72 Hours BTCmPEG 0 mM 0.2mM 0.6 mM 1.2 mM 2.4 mM 5mM 10 mM 15mM 3.4/5 96.9 ±1.9 43.2 33 ±3.8 XX §§ 34.5 33 ±3.1 X §§ 31.1 33 ±2.2 §§ 23.0 33 ±2.4 §§ 13.2 33 ±2.8 §§ 7.3 33 ±2.6 t t X §§ 4.2 33 ±2.1 §§ 3.4 / 20 92.0 ±10.2 40.3 33 ±13.9 36.2 33 ±14.4 31.3 33 ±15.7 28.3 33 ±13.6 16.6 33 ±7.2 10.6 33 ±4.3 ** 5.3 33 ±1.4 tt §§ tt §§ §§ §§ §§ 5/20 .87.2 ±10.1 26.5 33 ±8.1 ** 26.5 33 ±7.9 * 28.0 33 ±7.2 24.5 33 ±5.6 20.1 33 ±5.9 15.6 33 ±4.8 * 11.5 33 ±5.5 t t §§ t t §§ §§ §§ §§ §§ §§ 5 kDa . 98.9 ±0.4 62.8 33 ±15.1/ 43.6 33 ±3.8' 37.8 33 • ±2.6 41.8 33, ' - ±7:9 ~ 34.3 33 < • ±9.4 25.6 33 '.±6.1 21.0 33 ±2.7 130 hours post SV40 challenge with unmodified cells being 47% positive for T antigen, while significantly fewer cells modified by 2.4 and 15 mM BTCmPEG 3.4 /20 kDa, 12% and 2%, were infected respectively (p<0.001). A final combination of 5/20 kDa BTCmPEG was then examined. As depicted in Table 5.3, at 24 hours post infection unmodified cells were 42% positive for T antigen, while 8% and 2% of cells modified by 2.4 and 15 mM BTCmPEG 5 /20 kDa were infected respectively (pO.OOl). As shown in Table 5.3, similar results were observed for all three combinations examined at 48 and 72 hours post SV40 challenge. Analysis of these combination studies suggests that modification of cells with the BTCmPEG combinations offered improved protection when compared to the individual molecular weights (Table 5.3). For example, cells modified by 5 mM BTCmPEG 5 kDa were significantly (pO.OOl) more infected at 24 hours than the 5/20 kDa combination (18% versus 8% T antigen positive). As shown in Table 5.3, there were few differences between the BTCmPEG combinations 24 hours post challenge, though the combinations of 3.4/5 and 5/20 kDa were slightly more effective in preventing SV40 infection than the 3.4/20 kDa combination. 5.9 Discussion As seen throughout this chapter, modification of viral target cells with various forms of mPEG was shown to prevent SV40 infection. Based on the work by Bradley, et al. (2002), where grafting high molecular weight mPEGs better protected red blood cells from immunorecognition and obscured surface charge, it was hypothesized that by increasing the mPEG polymer length, modified cells would be further protected from viral infection. The hypothesis was supported by data from Dr. Scott's laboratory in which it was found that murine 131 Major Histocompatibility Complex-1 (MHC-1) was readily camouflaged with mPEG (Chen and Scott, 2001). As discussed in Chapter 2, the Cercopithicus aethiops (monkey) homologue of MHC-1 is the receptor for SV40 found on the CV-1 cells used throughout this thesis. Due to the slight size differential between the murine MHC-1 and the monkey homologue, 348 (murine MHC-1) compared to 193 amino acids in length (Velloso, et al. 2004; Alvarez, et al. 1997), it was hypothesized that all molecular weight mPEGs examined would prevent infection, with high molecular weight mPEG (20 kDa) better shielding MHC-1 as well as other surface proteins. Interestingly, when comparing SPAmPEGs of varying molecular weight, at 24 hours post viral challenge, there was no single molecular weight that was significantly more effective than the other two at low derivatization concentrations (0.2-2.4 mM). For example, SPAmPEG 5 kDa (0.6 mM) was significantly more effective than 2 kDa in preventing SV40 infection, but equally effective as SPAmPEG 20 kDa (Table 5.1). At higher derivatization concentrations (5-15 mM SPAmPEG), all polymer sizes afforded significant protection against viral entry. However, no consistent molecular weight effect was noted at 24 hours. Furthermore, the 20 kDa SPAmPEG polymer was able to alter the course of SV40 infection. In contrast to the 2 and 5 kDa SPAmPEGs, infection rates of cells modified by the 20 kDa polymer remained fairly static over the 72 hour period. The data suggest that SPAmPEG 20 kDa inhibits propagation of the infection. It is theorized that the progeny viruses become trapped in either mPEG coated vesicles or below the mPEG brush border and are therefore inhibited from efficiently propagating the infection. In order to test the hypothesis that progeny virus are produced in 20 132 kDa mPEG modified cells, progeny virus production could be measured by 3H-thymidine incorporation 72 hours post viral challenge. Similar results were obtained with BTCmPEG. Interestingly, modification of cells with BTCmPEG 20 kDa was significantly less effective in preventing SV40 infection than 3.4 or 5 kDa at 24 hours post infection. However, similarly to the SPAmPEG data discussed above, at 72 hours post infection, BTCmPEG 20 kDa was more effective in preventing progeny virus infection than the 3.4 or 5 kDa mPEGs. Again, the low rates of infection at 72 hours are likely the result of mPEG trapping of the progeny virus. As shown in Table 5.4, there were few significant differences between the three linker moieties at 24 hours; however, SPAmPEG 5 kDa was less effective at low concentrations (0.2-2.4 mM) than CmPEG 5 kDa. In contrast, at high concentrations (10-15 mM) there were no significant differences between the linker chemistries. These results may be due to the relative rates of hydrolysis of the linker molecules (Roberts, et al., 2002) with SPAmPEG more readily hydrolyzing than CmPEG resulting in a fewer SPAmPEG molecules capable of reacting with lysine residues. Additionally, according to the Shearwater catalogue (2000), SPAmPEG undergoes hydrolysis in vivo due to the ester linkage in the backbone of SPAmPEG (structure shown in Figure 2.1). Thus, a less dense mPEG brush border would result in a greater likelihood of viral infection. This difference in infection rates is also likely due to the reactivity of CmPEG toward nucleophilic functional groups other than amines such as cysteinyl and tyrosyl residues, which would increase the number of sites available for modification (Zalipsky and Lee, 1992). By 72 hours post challenge, it became obvious that cells modified by 133 SPAmPEG 5 kDa had significantly higher rates of infection than cells modified by either CmPEG or BTCmPEG at every concentration tested. The higher percentage of infection could be the result of increased SPAmPEG hydrolysis or higher rates of turnover of the SPAmPEG Table 5.4: Summary of 5 kDa Polymer Data- SPAmPEG is Less Effective than BTC or CmPEG 24 Hours Linker 0 mM 0.2 m M 0.6 mM 1.2 mM 2.4 mM 5 mM 10 mM 15 mM CmPEG 46.6 21333 17.099 12.699 9.099 7.599 6.199 5.199 ±7.5 ±9.1 ±7.1 ±7.3 ±8.4 ±5.4 ±3.9 ±5.3 t t BTCmPEG 49.2 25.799 16.799 16.899 14.299 17.699 8.099 7.199 ±13.0 ±6.4 ±6.0 ±6.1 ±1.3 ±1.6 ** ±0.8 ±1.0 SPAmPEG 50.1 33.199 28.999 22.799 20.899 13.499 12.799 10.299 ±1.6 ±3.0 ±2.6 ±2.0 ±2.7 ±1.8 ±3.2 ±3.0 ** ** ** ** 72 Hours Linker 0 mM 0.2mM 0.6 mM 1.2 mM 2.4 mM 5 mM 10 mM 15 mM CmPEG 98.5 29.599 29.899 29.699 27.699 26.399 22.599 23.599 ±0.7 ±12.8 ±12.4 ±15.6 ±6.3 ±10.7 ±2.9 ±7.9 f t t t t t BTCmPEG 98.9 62.899 43.699 37.899 41.899 34.399 25.699 21.099 ±0.4 ±15.1 ±3.8 ±2.6 ±7.9 ±9.4 ±6.1 ±2.7 ** ** ** SPAmPEG 99.3 89.399 70.499 63.499 50.899 47.799 49.699 41.399 ±0.6 ±1.8 ±3.2 ±2.7 ±3.4 ±3.4 ±13.2 ±5.4 ** ** ** ** ** ** ** t t t t t t t . t t f t t t The average percent of infected cells is shown for n=3 independent experiments, ± the standard deviation. Data for SPAmPEG and BTCmPEG 5 kDa are also shown in Tables 5.1 and 5.2. * Significantly different from CmPEG 5 kDa at equimolar concentrations (** p < 0.001; * p < 0.05) f Significantly different from BTCmPEG 5 kDa at equimolar concentrations (ff p < 0.001; f p < 0.05) 9 Significantly different from the control for that mPEG (0 mM) at that particular time point (33p< 0.001,3 p< 0.05). 134 modified proteins compared to those of BTC- or CmPEG. According to Table 5.4, it also appears that cells modified by low concentrations (0.2-2.4 mM) of BTCmPEG are significantly more prone to infection than those modified by CmPEG at 72 hours; however high BTC- and CmPEG concentrations (10-15 mM) were equally infected 72 hours post virus challenge. This dramatic difference in relative rates of infection between BTC- and CmPEG from 24 to 72 hours suggests that the less dense (0.2-0.6 mM) BTCmPEG brush border degrades more readily than that of CmPEG. The more-dense (5-15 mM), and therefore more protective, BTCmPEG brush border maintains equivalent infection rates to CmPEG from 24 to 72 hours post SV40 challenge (Table 5.4). To test this hypothesis, radio-labeled B T C - and CmPEG could be used to determine both the initial level of membrane derivatization and if any differential loss between CmPEG and BTCmPEG occurs over 72 hours. A comparison of the 20 kDa mPEGs is shown in Table 5.5, and confirms previous findings. BTCmPEG 20 kDa was less protective from initial SV40 infection than SPAmPEG (24 hour time point) suggesting that differences in the rates of hydrolysis are responsible. According to the Shearwater catalogue (2000), the hydrolysis half-life of BTCmPEG is 13.5 minutes at pH 8, 25 °C compared to 16.5 for SPAmPEG, suggesting there would have been somewhat fewer intact BTCmPEG molecules. However, my data shows SPAmPEG was less effective in preventing viral infection than BTCmPEG. This may be due to the slightly different reaction chemistries and polarities of the linker moieties, which may require somewhat different local environments for reactions to occur. Together, these results suggest that the CmPEG chemistry is best for modification of cells followed by BTCmPEG, with SPAmPEG a distant third. 135 Table 5.5: Comparison of 20 kDa mPEGs 24 Hours Linker 0 mM 0.2mM 0.6mM 1.2mM 2.4 m M 5 mM 10 mM 15 mM BTCmPEG 52.2 ±10.1 39.0" ±2.8 34.1" ±5.3 30.1" ±2.8 19.6" ±12.3 17.3" ±11.5 15.2" ±10.4 14.1" ±9.9 SPAmPEG 50.4 ±2.2 40.3" ±3.5 24.1" ±3.4 ** 24.3" ±3.3 14.4" ±3.3 8.9" ±3.0 ** 7.1" ±2.5 ** 2.6" ±1.4 ** 72 Hours Linker 0 mM 0.2mM 0.6mM 1.2mM 2.4 mM 5 mM 10 mM 15 mM BTCmPEG 99.5 ±0.5 66.9" ±2.6 58.1" ±3.2 44.8" ±2.7 34.1" ±3.0 26.0" ±2.1 17.5" ±1.4 12.5" ±1.5 SPAmPEG 99.3 ±0.6 65.9" ±14.3 64.4" ±4.2 59.9" ±2.3 ** 49.0" ±3.0 ** 43.6" ±3.4 ** 23.0" ±3.8 18.0" ±3.2 The average percent of infected cells is shown for n=3 independent experiments, ± the standard deviation. * Significantly different from BTCmPEG 5 kDa (** p < 0.001; * p < 0.05) at equimolar concentrations. 3 Significantly different from the control for that mPEG (0 mM) at that particular time point ("p< 0.001,3 p< 0.05). As previously discussed, combining BTCmPEGs of different molecular weights increased the protective effect against viral infection. While no combination examined was perfect, the data suggest that equimolar concentrations of 5/20 kDa and 3.4/5 kDa BTCmPEG were more effective than 3.4/20 kDa. Interestingly, at very low concentrations (0.2-1.2 mM), the 5/20 kDa combination was most effective; however at high concentrations (10-15 mM) the 3.4/5 kDa combination was most effective. This differential effect may be due to the self-limiting nature of the mPEG molecule. Because the 20 kDa polymers are very large, they are more likely to sterically hinder reactions with nearby lysines than the 3.4 or 5 kDa polymers resulting in fewer available residues. In other words, when high concentrations of the 20 kDa mPEG is used in 136 combination, the number of available, non-camouflaged, lysines are the limiting factor in the reaction consequently, fewer 20 kDa mPEG polymers are capable of covalently binding than when the two smaller molecular weights are used. Conversely, at low molar concentrations the limiting factor in the reaction is the mPEG. Under these conditions the 3.4 and 5 kDa polymers are less effective in combination than the 5/20 kDa because they are less capable of producing an efficient hydration barrier due to the smaller radius of gyration. These combination studies suggest that by combining all three molecular weight BTCmPEGs or in future studies, altering the concentrations of each molecular weight (e.g. 8 mM BTCmPEG 3.4 kDa and 2 mM 20 kDa) or combining different linker chemistries (e.g., CmPEG 5 kDa and BTCmPEG 20 kDa), additional protection may be available. In sum, modification of cells with mPEG was a powerful means of preventing SV40 infection. By combining multiple mPEG polymer lengths, the level of protection was enhanced and can be further titrated to meet the specific conditions of the human nasal cavity. 137 Chapter 6: Pegylation of Cells Prevents Infection by a Broad Range of Virus Types Chapter Overview ••} Prevention of Polyomavims Infections: Does mPEG modification prevent SV40 infection as assessed by plaque assays? • Prevention of Picornavirus Infections: Does mPEG modification prevent Theiler's . Murine Encephalomyelitis Virus infection? • Prevention of Adenovirus Infections: Does mPEG modification prevent Mouse Adenovirus infection? •'• Prevention of Coronavirus, Infections: Does mPEG modification prevent Rat Coronavirus infection? • Prevention of Herpesvirus Infections: Does mPEG modification prevent Cytomegalovirus infection? 6.0 Overview As shown in the previous chapters, mPEG modification of cell surface proteins inhibited SV40 infection. The next goal of this project was to determine whether pegylation of cells provided a broad spectrum antiviral effect. To examine this question, four additional viruses were used (Table 2.1 and 6.1). These viruses, which are representative of the common cold viruses were Theiler's murine encephalomyelitis virus (TMEV), mouse adenovirus (MAV) and rat coronavirus (RCV). In addition, human cytomegalovirus (CMV), a virus of importance to 138 transfusion and transplantation medicine was also examined. The virus family members share common traits such as viral structure, mode of entry, type of genome and in many cases size. As discussed in the Methods section, T M E V is a member of the Picornaviridae family and as such shares many traits with Rhinovirus, Mouse Adenovirus is a member of the Adenoviridae family as is human adenovirus, and RCV as well as human coronaviruses are members of the Coronaviridae family. In addition, C M V is a member of the Herpesviridae family and while Herpesviruses are not a cause of the common cold, C M V is an important blood born pathogen and represents an additional test of the broad spectrum antiviral activity of membrane pegylation. Table 6.1: Description of each virus model and corresponding cell line. Virus Virus Abbreviation Target Cell Line Cell Lineage Cell Morphology Simian Virus 40 SV40 CV-1 (ATCC CCL-70) Monkey Kidney Fibroblast/ Epithelial Mouse Adenovirus M A V Balb/3T3 (ATCC CCL-163) Mouse Embryo Fibroblast Rat Coronavirus R C V L2 (ATCC CCL-149) Rat Lung Epithelial Cytomegalovirus C M V MRC-5 (ATCC CCL-171) Human Lung Fibroblast Theiler's Murine Encephalomyelitis Virus T M E V BHK-21 (ATCC CCL-10) Hamster Kidney Fibroblast Throughout this chapter results were presented from the PEGylation of each cell line infected with the corresponding virus. This table lists the species and tissue of origin as well as the morphology of each cell line. 139 Based on my SV40 data as well as studies on red blood cells, T lymphocytes and pancreatic islets, BTCmPEG was chosen as the primary candidate polymer chemistry. BTCmPEG demonstrated comparable antiviral efficacy to CmPEG and in RBC studies demonstrated superior immunocamouflage efficacy and bioavailability (Scott, et al, 2000; Bradley, et al, 2002). 6.1 Pegylation of Cells with BTCmPEG 5 kDa Prevents SV40 Infection In Chapters 4 and 5, SV40 infection was monitored via T antigen staining within 72 hours post viral challenge. Unfortunately, there are no comparable commercially available antibodies to early viral proteins for T M E V , M A V and RCV. Therefore in order to detect infection in these viruses, plaque assays were performed. In order to compare the plaque assay results with these viruses to the SV40 findings, SV40 plaque assays were conducted on BTCmPEG 5 kDa modified CV-1 cells. As described in Chapter 2, all plaque assay experiments involved a 30-minute mPEG derivatization reaction at room temperature (pH 8.4), followed by viral challenge with 10"3-10"8 dilutions of SV40 (105-1 pfu/mL) and cells were incubated for up to 9 days following viral challenge in order for the plaques to fully develop. Results from BTCmPEG 5 kDa modification of CV-1 cells are depicted in Figure 6.1. Similarly to the results presented in Chapters 4 and 5, the plaque assay demonstrated a dose dependent reduction in SV40 infection of cells. For instance, control cells exposed to the 10*3 dilution (105 pfu/mL) of SV40 stock had an average of 88 plaques while those cells modified by 0.2, 1.2 and 15 mM BTCmPEG 5 kDa averaged only 4, 1 and 0 plaques respectively (>90% 140 1E-08 1E-07 1E-06 1E-05 0.0001 0.001 0.01 Viral Dilution Figure 6.1: B T C m P E G 5 kDa Modification of Cell Monolayers Inhibits Plaque Formation. Unmodified control cells exposed to the 10"3 SV40 dilution averaged 88.3 plaques while those cells modified by 0.2 and 15 mM BTCmPEG 5 kDa averaged only 3.7 and 0 plaques respectively. The standard deviation of n=3 independent experiments is shown. ** p < 0.001 compared to unmodified controls at equivalent viral dilutions. reduction in cytopathic effects for all BTCmPEG concentrations; pO.OOl). Likewise, an average of 12 plaques were observed on cells challenged with the 10"4 dilution (104 pfu/mL), compared to 2 (>75% reduction), 0 and 0 plaques (pO.OOl) found on the monolayers modified by 0.2, 1.2 and 15 mM BTCmPEG 5 kDa. It is worth noting that the SV40 viral doses were 141 104-109 fold less than that used for T antigen staining, as those viral concentrations would have resulted in a complete destruction of the unmodified monolayers. 6.2 Pegylation of Cells with BTCmPEG Prevents Theiler's Murine Encephalomyelitis Virus Infection The antiviral efficacy of BTCmPEG grafting to cell surface proteins was examined in response to TMEV. As previously stated T M E V is a member of the Picornaviridae family and is therefore a close relative of Rhinoviruses. The target cell line for T M E V is BHK-21, hamster derived kidney fibroblasts (Table 6.1). The BHK-21 cells were modified as described in Chapter 2 with BTCmPEG. As shown in Figure 6.2, significantly fewer plaques (pO.OOl) were observed on cells modified by all BTCmPEG species when compared to unmodified cells 5 days post challenge. As hypothesized, mPEG grafting resulted in a dose dependent decrease in plaque formation as evidenced by the dose response curves and photomicrographs (Figure 6.2 and 6.3). For example, at the 10"3 dilution (105 pfu/mL) an average of 244 plaques were counted in the unmodified control cells. Cells modified by 0.2 mM BTCmPEG 5 kDa had an average of 131 plaques (-50% reduction) while those modified by 15 mM BTCmPEG 5 kDa had an average of only 12 plaques (>90% reduction in cytopathic effects; pO.OOl). Control cells exposed to smaller viral loads (tO"4 dilution) had an average of 52 plaques while an average of only 9 and 1 plaque(s) were observed on cells modified by BTCmPEG 5 kDa derivatization concentrations as low as 0.2 mM (>75% reduction) or as high as 15 mM (>90% reduction), respectively. 142 A. BTCmPEG 5 kDa B. BTCmPEG 20 kDa 350 325-1 300 275 -250 -225 200-1 175 150-1 125 100 75 50 25 O m M 0.2 m M 0.6 m M 1.2 m M — « -2.4 m M —f-5 m M 10 m M 15 m M 1E-08 1E-07 1E-06 1E-05 0.0001 Viral Dilution BTCmPEG 5/20 kDa 0.001 0.01 500 450 400 4 O m M — • — 2.4 m M 0.2 m M — o — 5 m M 0.6 m M —#— 10 m M 1.2 m M — p — 15 m M 1E-07 1E-06 1E-05 0.0001 Viral Dilution 0.001 0.01 1E-08 1E-07 1E-06 1E-05 0.0001 0.001 0.01 Viral Dilution Figure 6.2: Modification of Cells with BTCmPEG Prevents TMEV Plaque Formation. BHK-21 cells were modified with BTCmPEG 5 kDa and plaques were counted 5 days post infection. On the 10"3 dilution plate there were 244 plaques counted on control cells. Cells modified by 0.2 mM BTCmPEG 5 kDa had an average of 131 plaques and those modified by 15 mM BTCmPEG 5 kDa had an average of 12 plaques. Similar curves were seen with BTCmPEG 20 kDa (B), and the combination of 5/20 kDa molecular weights (C). The standard deviation of n=3 independent experiments was shown. ** p < 0.001; * p 0.05 compared to unmodified controls at equivalent viral dilutions. < 143 A. B. Figure 6.3: TMEV Plaque Assay Pictures (A) Unmodified T M E V cells 5 days post T M E V challenge stained with neutral red. The plaque appears as a knot of cells (the large darkly stained mass) with a clear area to one side. (B) BHK-21 cells modified by 15 mM BTCmPEG 5 kDa were protected from M A V infection and display no plaques. Both pictures are magnified lOx. The larger BTCmPEG polymer (20 kDa) provided similar levels of protection. As shown in Figure 6.2B, the 10" dilution gave rise to an average of 216 plaques on unmodified cells. In contrast, BHK-21 cells modified by 0.2 mM BTCmPEG 20 kDa had an average of 95 plaques (>50% reduction) while those modified by 15 mM BTCmPEG 20 kDa had an average of only 3 plaques (>90% reduction in cytopathic effects; pO.OOl for all concentrations). Similarly, the 10"4 dilution plates averaged 87 plaques on the unmodified cells while 64 (>25% reduction) and 0 plaques (>90% reduction) were observed on those modified by 0.2 and 15 mM BTCmPEG 20 kDa, respectively. For comparison purposes, a subset of the data from the BTCmPEG 5 kDa, 20 kDa and 5/20 kDa is presented in Table 6.2. As with the SV40 data presented in Chapter 5, comparison of the efficacy of the 5 and 20 kDa polymers suggested that when using a single molecular weight mPEG, polymer size was not a significant factor (Table 5.2 and 6.2). 144 Table 6.2: Plaque Assay Comparison of B T C m P E G Protection Against T M E V Infection. B T C 0 m M 0.2 m M 0.6mM 1.2mM 2.4m M 5 m M 10 m M 15 m M m P E G 5 kDa 243.7 131 93.3 86 68 40.3 27 11.7 ±61.9 ±46.5 ±10.6 ±9.2 ±8.5 ±4.5 ±5.6 ±2.5 20 kDa 216 95.3 69.7 63.3 29.3 14.7 4.7 3.3 ±40.3 ±7.0 ±4.7 ±6.5 ±6.2 ±9.1 ±5.5 ±2.5 5 & 2 0 316 56 25.7 15.3 5.7 2.7 1.3 1.7 ±169.9 ±26.1 ±6.1 ±5.5 ±3.5 ±2.1 ±1.5 ±2.1 * Significantly different from BTCmPEG 5 kDa ** p < 0.001; * p < 0.05 at equimolar concentrations. f Significantly different from BTCmPEG 20 kDa ft P < 0.001; f p < 0.05 at equimolar concentrations. The average number of plaques for n=3 independent experiments from the 10~3 T M E V dilution are shown above. Full data is shown in Figure 6.2. To determine if the anti-TMEV efficacy of BTCmPEG could be enhanced, equimolar concentrations of BTCmPEG 5/20 kDa were examined. As shown in Figure 6.2C, significantly fewer plaques were observed on cells modified by BTCmPEG 5/20 kDa than unmodified cells 5 days post infection. For example, on the 10"3 dilution plate there were 316 plaques counted on unmodified cells. In contrast, cells modified with as little as 0.2 mM BTCmPEG 5/20 kDa exhibited an average of 56 plaques (>75% reduction in cytopathic effects) while those modified by 15 mM BTCmPEG 5/20 kDa had an average of 2 plaques (>90% reduction; p<0.001 for all mPEG concentrations). The 10"4 dilution plates averaged 40 plaques for unmodified cells while 18 (>50% reduction) and 0 plaques (>90% reduction) were observed on those modified by 0.2 and 15 mM BTCmPEG 5/20 kDa, respectively. Thus, despite statistical analysis that showed no significant differences between the levels of protection when mPEG polymer size was increased (Table 6.2), these data demonstrated that 145 the polymer combination of BTCmPEG 5/20 kDa provided superior protection against T M E V infection. 6.3 Pegylation of Cells with BTCmPEG Prevents Mouse Adenovirus Infection As previously discussed, M A V is a close relative of human adenovirus (Chapter 2) and was chosen to represent this important cause of human respiratory infections. Plaque assays were conducted on control and BTCmPEG modified Balb/3T3 cells, a mouse embryonic fibroblast cell line (Table 6.1). As shown in Figures 6.4A and 6.5, BTCmPEG provided potent protection against M A V infection 7 days post viral challenge. For example, cells challenged with the 10~3 M A V dilution averaged 83 plaques on unmodified cells, while cells modified by as little as 0.2 mM BTCmPEG 5 kDa averaged only 19 plaques (p<0.001). Furthermore, cells modified by 15 mM BTCmPEG 5 kDa averaged only 1 plaque (p<0.001). Cells challenged with the 10"4 dilution averaged 14 plaques for unmodified cells while those modified by 0.2 and 15 mM BTCmPEG 5 kDa averaged 4 and 0 plaques, respectively. Interestingly, even at the lowest mPEG concentration (0.2 mM) both the 10"3 and 10"4 dilutions result in an approximately 75% reduction in cytopathic effects and greater than 90% at 15 mM BTCmPEG. 146 A. BTCmPEG 5 kDa B. BTCmPEG 20 kDa 150 1E-08 1E-07 1E-06 1E-05 0.0001 0.001 0.01 Viral Dilution C. BTCmPEG 5/20 kDa OmM — • — 2.4 mM 0.2 mM — o — 5 mM 0.6 mM — 1 0 mM 1.2 mM —•?— 15 mM 1E-08 1E-07 1E-06 1E-05 0.0001 0.001 0.01 Viral Dilution OmM — * — 1.2 mM —t,— 10 mM 0.2 mM — • — 2.4 mM — 7 — 15 mM 0.6 mM — » — 5mM 1E-08 1E-07 1E-06 1E-05 0.0001 0.001 0.01 Viral Dilution Figure 6.4: Modification of Cells with BTCmPEG Prevents MAV Infection. Balb/3T3 cells were modified with BTCmPEG 5 kDa for 30 minutes prior to infection with M A V and plaques were counted 7 days post infection. On the 10"3 dilution plate there were 83 plaques counted on control cells. Cells modified by 0.2 mM BTCmPEG 5 kDa had an average of 19 plaques and those modified by 15 mM BTCmPEG 5 kDa had an average of 1 plaque. Similar results were obtained with BTCmPEG 20 kDa and the combination of 5/20 kDa. The standard deviation of n=3 independent experiments was shown. ** p < 0.001; * p < 0.05 compared to unmodified controls at, equivalent viral dilutions. 147 A. B. Figure 6.5: M A V Plaque Assay Pictures (A) Unmodified Balb/3T3 cells 7 days post M A V challenge. Plaques appear as clear areas where the cells have died as a result of infection. (B) Balb/3T3 cells modified by 15 mM BTCmPEG 5 kDa were protected from M A V infection and display no plaques. BTCmPEG 20 kDa provided similar protection. As shown in Figure 6.4B, fewer plaques were observed on cells modified by BTCmPEG 20 kDa than unmodified cells 7 days post infection. For example, unmodified cells exposed to the 10"3 dilution averaged 75 plaques, while cells modified by 0.2 mM BTCmPEG 20 kDa averaged 13 plaques (>75% reduction) and those modified by 15 mM BTCmPEG 20 kDa averaged only 1 plaque (>90% reduction). Similarly, cells exposed to the 10"4 M A V dilution averaged 10 plaques for unmodified cells while 1 and 0 plaques were observed on those modified by 0.2 and 15 mM BTCmPEG 20 kDa, respectively. As previously noted with both SV40 and T M E V , increasing the polymer size did not have additional protective effects against viral infection. However, in order to determine if mPEG protection could be enhanced by combining polymers, derivatization with BTCmPEG 5/20 kDa 148 was examined. Figure 6.4C, displays the dose dependent reduction in plaque formation seen in cells modified by the BTCmPEG 5/20 kDa mixture. Unmodified cells exposed to the 10"3 dilution averaged 79 plaques, while cells modified by 0.2 or 15 mM BTCmPEG 5/20 kDa averaged 12 (>75% reduction in cytopathic effects) and 1 plaque (>90% reduction), respectively. Similarly, control cells exposed to the 10"4 dilution averaged 11 plaques compared to 2 and 0 plaques on monolayers modified by 0.2 and 15 mM the BTCmPEG combination respectively. Unlike the T M E V experiments previously described, the combination of BTCmPEG 5/20 offered no significant benefit compared to the single molecular weight polymers (Table 6.3), with all three resulting in a greater than 75% reduction in cytopathic effects with as little as 0.2 mM BTCmPEG and greater than 90% reduction for cells modified with 15 mM. Table 6.3: Plaque Assay Comparison of BTCmPEG Protection Against MAV Infection. BTC 0 mM 0.2 mM 0.6mM 1.2mM 2.4mM 5 mM 10 mM 15 mM mPEG 5 kDa 83 19 13.3 11 8 6.7 3 1.3 ±19.0 ±12.5 ±10.8 ±8.9 ±7.6 ±6.5 ±2.6 ±1.2 20 kDa 75.3 13.3 8.3 3.3 2 1.3 0.3 0.3 ±9.1 ±1.5 ±2.5 ±1.5 ±2.6 ±1.5 ±0.6 ±0.6 5&20 79 12.3 8.3 5.3 1.7 0.7 0.3 0.3 ±21.4 ±7.6 ±4.2 ±3.2 ±1.6 ±1.2 ±0.6 ±0.5 * Significantly different from BTCmPEG 5 kDa ** p < 0.001; * p < 0.05 at equimolar concentrations. t Significantly different from BTCmPEG 20 kDa ft p < 0.001; t p < 0.05 at equimolar concentrations. The average number of plaques for n=3 independent experiments from the 10"3 M A V dilution are shown above. Complete data is presented in Figure 6.4. 149 6.4 Pegylation of Cells with BTCmPEG Prevents Rat Coronavirus Infection Rat Coronavirus was examined as a representative of human Coronaviruses which include common cold viruses as well as the Severe Acute Respiratory Syndrome (SARS) virus. The L2 rat lung epithelial cell line (Table 6.1) was modified as discussed in Chapter 2. As shown in Figures 6.6A and 6.7, significantly fewer plaques were observed on cells modified by BTCmPEG 5 kDa than in the unmodified cells 7 days post infection. For example, unmodified L2 cells exposed to the 10"3 dilution averaged 156 plaques, while those modified even at the lowest concentration, 0.2 mM averaged only 9 plaques (>90% reduction in cytopathic effect) and 0 plaques were observed within cells modified with 15 mM BTCmPEG 5 kDa (pO.OOl). A similar decrease in plaques was observed in cells challenged with the 10"4 R C V dilution. Control cells averaged 54 plaques while only 4 and 0 plaques, a greater than 90% reduction at both concentrations, were observed on those modified by 0.2 and 15 mM BTCmPEG 5 kDa, respectively (pO.OOl). The efficacy of large molecular weight mPEG (20 kDa) was also examined. As shown in Figure 6.6B, there was a dose dependent decrease in plaque formation (pO.OOl for all concentrations at the 10"3 dilution) for cells modified by BTCmPEG 20 kDa, but as with SV40, T M E V and M A V , no significant effect of polymer size on antiviral efficacy was noted. To determine if polymer combinations would have additional antiviral effects, BTCmPEG 5/20 kDa was used (Figure 6.6C). Control cells challenged with the 10" R C V dilution averaged 133 plaques, in comparison, cells modified with only 0.2 mM BTCmPEG 5/20 kDa had an average of only 9 plaques (>90% reduction in cytopathic effects), while those modified by 15 150 A. B T C m P E G 5 kDa B. B T C m P E G 20 kDa 0.01 Viral Dilution Viral Dilution C. B T C m P E G 5/20 kDa Figure 6.6: Modification of Cells with B T C m P E G Prevents R C V Infection. L2 cells were modified with BTCmPEG 5 kDa for 30 minutes prior to infection with RCV and plaques were counted 7 days post infection. On the 10"3 dilution plate there were 156 plaques counted on control cells. Cells modified by 15 mM BTCmPEG 5 kDa had an average of 0 plaques. Similar results were obtained with BTCmPEG 20 kDa and the combination of 5/20 kDa. The standard deviation of n=3 independent experiments was shown. ** p < 0.001; * p < 0.05 compared to unmodified controls at equivalent viral dilutions. o.oi Viral Dilution 151 A. B. ': S . . •» .; •. * •'.' .'. • • ' ... * - * ' ' Figure 6.7: R C V Plaque Assay Pictures (A) Unmodified L2 cells 7 days post RCV challenge. The edge of a plaque is shown and appears as a clear area at the top of the picture where the cells have died as a result of infection. (B) L2 cells modified by 15 mM BTCmPEG 5 kDa were protected from M A V infection and display no plaques. Cells were stained with neutral red and pictures taken with a lOx magnification. mM BTCmPEG 5 and 20 kDa had an average of 0 plaques (p<0.001). A similar decrease in plaque formation was observed with cells exposed to the 10"4 dilution. As displayed in Table 6.4, statistical analysis of the 10" R C V dilution data demonstrated no further enhancement of the already potent antiviral effects of the 5 kDa and 20 kDa polymers. 152 Table 6.4: Plaque Assay Comparison of B T C m P E G Protection Against R C V Infection. B T C mPEG 0 m M 0.2mM 0.6 m M 1.2mM 2.4mM 5 m M 10 mM 15 mM 5 kDa 156 8.7 3 3 1.3 0 0 0 ±11.5 ±6.0 ±3.6 ±4.4 ±1.2 ±0 ±0 ±0 20 kDa 171.3 11.3 3.3 0.7 0.3 0 0 0 ±48.6 ±4.5 ±1.2 ±0.6 ±0.6 ±0 ±0 ±0 5 & 2 0 133 9 2.7 0.3 0.3 0 0 0 ±29.1 ±3.6 ±1.2 ±0.6 ±0.5 ±0 ±0 ±0 * Significantly different from BTCmPEG 5 kDa ** p < 0.001; * p < 0.05 at equimolar concentrations. f Significantly different from BTCmPEG 20 kDa f t P < 0.001; f P < 0.05 at equimolar concentrations. The average number of plaques for n=3 independent experiments from the 10"3 RCV dilution are shown above. Full data is shown in Figure 6.6. 6.5 Pegylation of Cells with BTCmPEG Prevents Cytomegalovirus Infection In addition to the model viruses representing human respiratory pathogens, cytomegalovirus infection was examined. As previously mentioned C M V is a member of the herpesvirus family and is an important blood and tissue borne pathogen. As with SV40 T antigen staining, C M V infection was monitored via immunostaining for the 72 kDa immediate early protein (IE72) at 24 hour intervals. The cell line used in these studies were the MRC-5, human lung fibroblasts (Table 6.1). At 24 hours post C M V challenge, 49% of unmodified cells stained positive for IE72, while cells modified by BTCmPEG 5 kDa demonstrated a dose-dependent decrease in IE72 153 expression. At the 2.4 and 15 mM concentrations cells were 14% and only 4% infected, respectively. At 48 hours, 66% of unmodified MRC-5 cells were positive while only 19% and 7% of 2.4 and 15 mM BTCmPEG 5 kDa modified cells were positive. By 72 hours post exposure, unmodified cells were 93% IE72 positive, with only 25% and 13% of cells modified by 2.4 and 15 mM were infected with CMV. As can be noted by these findings, BTCmPEG provided potent protection against viral entry (24 hours) and propagation (48-72 hours). This data is presented in Table 6.5 along with data from the BTCmPEG 20 kDa and 5/20 kDa experiments for comparison. In order to determine if a larger polymer size increased the antiviral efficacy of mPEG, BTCmPEG 20 kDa was examined. As shown in Table 6.5, at 24 hours post C M V challenge, 50% of unmodified cells stained positive for IE72, compared to 16% and 6% of cells modified by 2.4 and 15 mM BTCmPEG 20 kDa. Forty-eight hours post challenge, 68% of unmodified cells were positive while only 18% and 7% of 2.4 and 15 mM BTCmPEG 20 kDa modified cells stained IE72 positive. At 72 hours post C M V exposure, unmodified cells were 92% IE72 positive, while 21%> and 7% of cells modified by 2.4 and 15 mM were positive. Thus as with all viruses tested, the larger polymer size (20 kDa) did not increase the already potent antiviral efficacy of the 5 kDa polymer. To assess if increased efficacy of mPEG protection could be obtained through a combination of molecular weights, an equimolar mixture of BTCmPEG 5/20 kDa was studied (Table 6.5). Combinatorial studies demonstrate no biologically important benefits to the 5/20 kDa polymer mixture over the single molecular weights. However, at very low derivatization concentrations 154 Table 6.5: IE72 Immunostaining Comparison of BTCmPEG Protection Against CMV Infection. 24 Hours BTC mPEG 0 mM 0.2mM 0.6mM 1.2mM 2.4 mM 5mM 10 mM 15 mM 5 kDa 49.1 ±1.5 28.5" ±2.2 24.7" ±3.1 20.3" ±2.1 14.4" ±2.0 9.1" ±1.8 6.0" ±1.4 4.4" ±1.1 20 kDa 49.6 ±2.7 27.9" ±1.5 25.0" ±1.8 19.5" ±1.5 15.9" ±1.2 12.3" ±1.8 8.3" ±1.0 5.9" ±1.2 5/20 48.3 ±2.1 21.2" ±1.8 * 18.3" ±1.4 * 15.5" ±1.7 13.4" ±1.5 10.6" ±0.5 7.3" ±1.1 4.2" ±1.4 48 Hours BTC mPEG 0 mM 0.2mM 0.6 m IV1 1.2mM 2.4mM 5mM 10 mM 15 mM 5 kDa 66.0 ±2.7 35.3" ±1.8 29.6" ±1.1 24.2" ±1.5 18.8" ±1.8 12.9" ±1.4 9.9" ±1.9 7.5" ±1.6 20 kDa 68.4 ±2.6 30.6" ±2.6 28.1" ±2.5 22.0" ±1.1 18.5" ±1.5 14.0" ±1.3 10.4" ±1.2 7.2" ±1.6 5/20 70.1 ±3.0 25.4" ±1.9 21.3" ±2.0 18.0" ±1.3 16.4" ±2.4 13.3" ±1.2 9.7" ±1.7 6.2" ±1.7 72 Hours BTC mPEG 0 mM 0.2mM 0.6 m M 1.2mM 2.4m M 5 mM 10 mM 15 mM 5 kDa 92.8 ±3.0 41.7" ±2.0 + 35.9" ±1.9 30.8" ±2.9 25.1" ±2.2 19.9" ±2.2 15.9" ±2.0 12.7" ±2.1 20 kDa 92.2 ±1.6 . 34.0" ±3.4 * 31.1" ±1.5 24.2" ±1.8 20.7" ±2.2 14.9" ±1.7 10.9" ±1.6 7.3" ±1.5 5/20 90.4 ±3.8 28.3" ±2.1 ** 24.6" ±1.4 ** 21.0" ±2.3 ** 19.1" ±1.3 15.0" ±1.4 11.7" ±0.9 6.8" ±1.8 The average percent of infected cells is shown for n=3 independent experiments, ± the standard deviation. * Significantly different from BTCmPEG 5 kDa ** p < 0.001; * p < 0.05 at equimolar concentrations. f Significantly different from BTCmPEG 20 kDa ff p < 0.001; t p < 0.05 at equimolar concentrations. 3 Significantly different from the control for that mPEG (0 mM) at that particular time point (" p< 0.001,3 p< 0.05). 155 (0.2-1.2 mM) a slight, but statistically significant benefit was noted for the 5/20 kDa combination (Table 6.5). This effect was probably not biologically significant because the differences were so slight. 6.6 Discussion As shown throughout this chapter, mPEG modification of a variety of cells prevents infection by a variety of viruses representing a broad spectrum of viral families. Notably, viruses within the same families as common cold viruses were significantly inhibited from infecting cells modified by BTCmPEG compared to unmodified cells. Also of importance was the significant protection against C M V , a representative of the Herpesvirus family, infection of modified MRC-5 cells. C M V is not a common cold virus; however it is an important blood and tissue borne pathogen which causes several important human diseases, and can be passed through close personal contact or blood products. As clearly demonstrated in this chapter, the immunocamouflage effect induced by mPEG grafting results in a potent broad-spectrum antiviral effect. Importantly, this antiviral effect is successful in preventing infection by viruses that enter cells via both receptor mediated endocytosis and membrane fusion. The viruses examined in this chapter also differed in structure, genome type and size (Table 2.1). The size difference between the viruses is quite large, with T M E V only 20 nm in diameter while R C V and C M V are as large as 200 nm, hence these data demonstrate that the mPEG brush border is sufficient to block even very small viruses. This size differential covers most human viruses and was necessary to show mPEG efficacy against very small and large 156 viruses. Taken collectively, the data presented in this chapter showed that mPEG modification was indeed a broad-spectrum means of preventing viral infection as modified cells were significantly less infected with SV40, T M E V , M A V , RCV or C M V . Because of the broad spectrum efficacy of mPEG grafting, several applications for this approach exist. One such application is a nasal spray or inhaler containing activated mPEG for common cold viruses or respiratory viruses such as respiratory syncytial virus (RSV) and SARS. Based on the data presented in Chapters 3-6, mPEG modification of nasal or upper respiratory epithelial cells would prevent viral infection, shorten the duration of illness and decrease severity of symptoms in a manner similar to Zicam (though on a broader spectrum of viruses). For non-respiratory viruses such as HIV and herpesvirus, a vaginal application of activated mPEG may also prevent infection as their modes of entry are fusion, the same as that of RCV and C M V . Additionally, pegylation of cellular blood products can prevent HIV infection of naive T cells as well as the escape of progeny virus from modified cells. Furthermore, studies by Mizouni (2000) demonstrate that mPEG may be used in acellular blood products (e.g., plasma) to inactivate blood borne viruses (e.g., CMV). 157 Chapter 7: Summary 7.1 Discussion To date, prophylactic antiviral drug therapies approved for human use are limited to vaccines and very few preventative agents such as the zinc nasal gel, Zicam. As discussed in Chapter 1, there are many drawbacks to vaccines including allergic reactions, infection and more importantly virus specificity. Virus specificity is also the primary drawback of Zicam. Zicam is a novel over-the-counter treatment that specifically inhibits rhinovirus infection and has been shown to reduce the length of illness (Godfrey, et al., 1992; Novick, et al, 1996; Hirt, et al., 2000). This improved rate of convalescence is attributed to the zinc ions of Zicam associating with progeny rhinovirus capsids thereby preventing binding of the cell surface receptor, ICAM-1 (Eby, et al., 1984). Unfortunately due to this specificity, Zicam only inhibits infections of rhinovirus strains utilizing the ICAM-1 receptor (80-90% of strains) and has no effect on adenovirus or coronavirus infections. Therefore, a broad-spectrum method of preventing viral infections is needed. Based on the immunocamouflage of red blood cells, T lymphocytes and pancreatic islet cells performed in Dr. Scott's laboratory, I hypothesized that the modification of viruses and/or cell surfaces with activated mPEG would prevent infection by a broad-spectrum of viruses (Figure 1 . 1 ) . It is important to note that several characteristics of mPEG have made it useful in these and other studies, including: water solubility, immunocamouflage capacity, lack of toxicity and 158 immunogenicity, and efficient clearance from the body via the kidney (Kozlowski, et al., 2001). As described in Chapter 1, when mPEG is covalently bound to proteins (either viruses or cell surfaces), a physical barrier is created that prevents protein-protein, cell-cell and virus-receptor interactions. In addition to the physical nature of the barrier, mPEG also masks surface charge (Scott, et al., 1998; Bradley, et al., 2002). Under normal circumstances, virus-receptor interactions are in part governed by electrostatic interactions (Conti, et al., 1991; Mastromarino, etal, 1991; Supteri, etal., 1993; Callahan, 1994) and as hypothesized in this thesis, pegylation of host cells of viruses will interfere with these interactions. When a virus approaches its cell surface receptor these electrostatic charges act magnetically to pull the two proteins together in the proper alignment. However, as illustrated in Figure 7.1, modification of the viral capsid with mPEG shields the surface charge of the virus, which prevents the electrostatic attraction. Similarly, as shown by Bradley, et. al. (2002) and discussed in Chapter 1, modification of target cells shields the charge of the entire cell surface thereby inhibiting electrostatic interactions with the unmodified virus. This masking of charge is similar to placing a non-conducting material such as a block of wood between two magnets; the charge on one side is not recognized by the opposing charge and attraction (i.e. binding) does not occur. Therefore modified viruses will deflect off cell surface receptors and entry/infection will not occur. Furthermore, unmodified virus will be inhibited from properly binding the masked cell surface proteins due to the physical barrier and electrostatically neutral nature of the mPEG barrier. 159 A t t r a c tion r r + No Attraction $1 NC NC mPEG NC = No / Obscured Charge 3 Pegylated Virus D Pegylated C e l l Surface Proteins x y (V aru s Figure 7.1: Masking Surface Charge with mPEG Plays a Role in Preventing Viral Infection. As shown in A, magnetic forces of opposite charge are attracted to each other when placed in close proximity. Similarly, electrostatic reactions are in part responsible for virus-receptor interactions (B). By modifying surfaces (viral coat proteins or cell surface proteins) with mPEG (C and D), these electrostatic forces are nullified resulting in the virus deflecting off the receptor (i.e. preventing virus binding). 160 Through in vitro studies, it was shown that mPEG derivatization of viruses was an extremely potent means of preventing infection of unmodified target cells. In Chapter 3,1 explored the effects of covalent modification of virus particles with activated mPEG species of molecular weights ranging from 2-20 kDa, as well as three different linker chemistries (cyanuric chloride, benzotriazole carbonate and succinimidyl propionate). Interestingly, there were no significant differences in the levels of protection among the individual mPEG molecular weights. This lack of differences is likely the result of the uniform nature of the SV40 capsid as well as the small size of the virus particle in relation to the mPEG polymer. According to the data presented in Chapter 3, the equimolar mixture of BTCmPEG 3.4 /20 kDa is best for modifying the capsid of SV40. It seems likely that this combination provides the most dense mPEG barrier, which is optimal for preventing interactions with the SV40 receptor, MHC-1. Additional mPEG modification of SV40 performed by Mizouni (2000) yielded very interesting results, which may improve vaccine development. When modified virus was injected intraperitoneally into Balb/c mice an enhanced immune response developed than in mice challenged with unmodified SV40. These results were surprising given that all previous work in our laboratory showed pegylation of foreign cells resulted in a drastic reduction in the immune response (Scott, et al, 1997; Scott and Murad, 1998; Murad, et al, 1999A and B; Bradley, et al, 2001; Chen and Scott, 2001; Bradley, et al, 2002; Chen and Scott 2003; Scott and Chen 2004). The explanation for this unexpected increase in antibody production in mice exposed to modified SV40 is proposed to be two-fold. First, because SV40 is a naked virus (i.e. protein capsid rather than enveloped), the capsid will naturally break up over time and any 161 unmodified portions of the protein clusters will become visible to the immune system ultimately resulting in antibody production. Secondly, as many of these proteins remain bound to mPEG, which increases the circulation time throughout the body, partially modified VP1 clusters will continue to disassemble resulting in a time-release effect that enhances the immune response (Torchilin 1994; Shorr, et al, 1999; Bandas, et al, 2003). Despite the convincing ability of the pegylation of virus particles to prevent infection presented in Chapter 3, this method would be ineffective in preventing viral respiratory infections in the real world where viruses go unseen rather than in neatly labeled vials. Therefore, to determine if healthy cells could be prophylactically protected, viral target cells were modified with activated mPEG to test the efficacy of the antiviral protection. Unlike SV40, which has a capsid composed primarily of 360 viral protein 1 (VP1) molecules (Flint, et al, 2000), cells vary in their surface topography based on the irregular sizes of external protein domains. As shown in Figure 7.2, cell surfaces may be sparsely or densely populated with proteins of varying size, which may affect the mPEG barrier. In addition to irregularity of the cell surface, pegylation of cells is also affected by protein turnover, which will deplete the mPEG brush border over time. It also bears noting that cells are much larger than SV40, which is only 450 angstroms in diameter. The mPEG brush border is thought to extend approximately 100 angstroms from the point of protein attachment for a 5 kDa linear PEG (de Gennes, 1980; Allen, et al, 2002); therefore the mPEG brush border would be visible at the magnification of the SV40 picture in Figure 7.2. 162 B. Sparsely Populated Densely Populated Figure 7.2: Modification of Cells versus SV40 Capsids. As seen in A, cell surfaces vary in the size of external protein domains as well as protein density leading to potential differences in the mPEG brush border. Conversely, the capsid of SV40 is composed of 360 VP1 proteins resulting in the uniform pegylation of viruses in solution (B). The mPEG is depicted as black (A) or grey (B) projections emanating from the cells or virus respectively. It should be noted that the diameter of SV40 is 450 angstroms (45 nm) while the mPEG brush border is approximately 100 angstroms therefore the mPEG would be visible at the magnification shown. However cells are much larger and are not drawn to scale in order to show the large proteins and brush border. B was modified from Liddington, et al, 1991. CmPEG modification of host cells was discussed at length in Chapters 4-6. Pegylation of cells was clearly shown to prevent viral infection over a wide range of reaction conditions including changes in reaction time, pH and temperature. However, to accomplish this antiviral effect, it was crucial that we also maintain more physiological conditions during host cell derivatization to ensure a viable monolayer remained. The importance of these mPEG reaction conditions is depicted in Figure 7.3. Note that cell shrinkage or loss of cells (through death or disassociation from the petri dish) due to adverse pH exposes unmodified proteins to which viruses can attach and gain entry into the cell. Notably, the physiologic reaction conditions (5 minute reaction, pH 7.8, 32°C) provided near maximal protection against viral infection. These results are 163 Physiological Conditions Pegylated m P E G Modif ied Proteins Control Cel l Sur face Proteins - * — i Harsh Conditions ! Unmodif ied Proteins are Exposed Extreme Conditions Floater Unmodif ied Proteins are Exposed Figure 7.3: Importance of Physiologic mPEG Reaction Conditions. As illustrated above, adverse mPEG reaction conditions initially used in my studies (pH 9.0, data not shown) such as alkaline pH can cause cell shrinkage or in extreme cases cell loss resulting in exposure of unmodified proteins. These unmodified proteins are potential sites for viral entry. 164 significant for formulating a final mPEG nasal spray that would be well tolerated (i.e. little nasal irritation due to high pH) and includes a reaction time that is likely to be complied with (3-5 minutes versus 30). Another significant factor to be considered for an mPEG nasal spray is the dosing schedule. Importantly, as shown in Chapter 4, near maximal mPEG protection remains for 24 hours in vitro. Furthermore, significant protection remains for up to 60 hours post-modification. These findings indicate that a once-daily dosing of mPEG would be optimal in providing maximal protection, however if a dosage is missed, some protection against viral infection would remain for up to 60 hours. Because the ultimate goal of this project will be a human nasal spray, mPEG modification of cells was also tested in the presence of mucin to model the presence of human mucus. Notably, the presence of mucin during pegylation had no significant effect on the efficacy of mPEG protection suggesting mucus will not interfere in the modification of nasal epithelium; however other secreted proteins may reduce the density of cell derivatization by reacting with the activated mPEG. This may be a problem when the work is translated into the more complex animal models. One way of determining whether mucus secretions inhibit in vivo pegylation would be to compare infection rates in animals where the nasal passage is rinsed the with saline prior to modification versus animals that do not undergo the saline rinse. Studies were also conducted utilizing additional viruses with a range of characteristics to demonstrate mPEG modification of cells yields a broad-spectrum antiviral effect. Three 165 viruses which enter cells by receptor mediated endocytosis (SV40, Theiler's murine encephalomyelitis virus and mouse adenovirus) were shown to be significantly inhibited from infecting modified cells as illustrated in plaque assay experiments. Furthermore, two viruses that utilize fusion to enter cells (rat coronavirus and cytomegalovirus) were also significantly inhibited from infecting the mPEG-modified cells, as described in Chapter 6. These two methods of entry are utilized by a majority of viruses including those of human significance, as shown in Table 7.1. The viruses tested in my studies vary in size from 20-200 nm in diameter, have genomes of either RNA or DNA and are representative of five different virus families. Due to the considerable differences between these viruses, it is obvious that mPEG modification of target cells in vitro prevents infection from a very broad-spectrum of viruses (Figure 7.4), unlike the other therapies described in Chapter 1 (Zicam, sICAM and RRMA). In aggregate, the above data are compelling evidence that viral infection is prevented by mPEG modification of cells, but did not directly confirm our claim that the mPEG barrier prevented virus-receptor interactions thereby preventing entry. To delineate the mechanism by which the pegylation of cells prevented viral infection, viral entry studies were conducted using the SV40/CV-1 model system. Because VP1 is only produced late in the SV40 infection, any VP1 detected within the first three hours of infection is a result of viral entry into the cells. Significantly, pegylated cells were shown to contain far fewer endocytosed VP1 positive vesicles. This lack of VP1 within cells 1-3 hours post viral challenge is strong evidence that the virus is inhibited from entering mPEG modified cells rather then the mPEG interfering in the infectious cycle after virus entry. Indeed, based on the T antigen staining studies, it appears that mPEG does not interfere with the production of progeny virus, as the number of infected 166 Table 7.1: Well-Known Human Viruses Share Characteristics with Model Viruses. Virus Family Mode of Entry Diameter Genome HIV Fusion 80-100 nm ssRNA Retroviridae Influenza Fusion 90-120 nm ssRNA Orthomyxoviridae Poliovirus Receptor Mediated -30 nm ssRNA Picornaviridae Endocytosis Respiratory Syncytial Virus Paramyoviridae Fusion 150-300 nm ssRNA Rotavirus Receptor Mediated 60-80 nm dsRNA Reoviridae Endocytosis SARS Virus Fusion -200 nm RNA Coronaviridae Variola (Smallpox) Fusion 200 x 200 x 250 nm dsDNA Poxviridae West Nile Virus Fusion 40-60 nm ssRNA Flaviviridae Model Viruses SV40 Papovaviridae Receptor Mediated Endocytosis -45 nm dsDNA MAV Adenoviridae Receptor Mediated Endocytosis 70-90 nm dsDNA RCV Coronaviridae Fusion 80-160 nm ssRNA CMV Herpesviridae Fusion -200 nm dsDNA TMEV Picornaviridae Receptor Mediated Endocytosis -20 nm ssRNA As shown, many human viruses enter cells via receptor mediated endocytosis or fusion similarly to the model viruses used in this thesis. In addition, the model viruses vary in size as much as the human viruses listed above. 167 I II Biiiiiiis l i i i i p i i V i r u s K e y Rotavirus Adenovirus West Nile Virus -JfSk- Variola TMEV Poliovirus SV40 Figure 7.4: mPEG Modification of Cell Surfaces Prevents Infection of a Broad-Spectrum of Viruses. As shown above, the model viruses (TMEV, SV40, Adenovirus, Coronavirus and CMV) are prevented from infecting cells. Due to the similar characteristics shown in Table 7.1, additional human viruses (Poliovirus, Rotavirus, West Nile Virus and Variola) should also be inhibited from infecting mPEG modified cells. cells increased from 24 to 48 and 72 hours. While the production of progeny virus was not directly measured, stringent washing steps following the initial viral challenge make it highly unlikely that sufficient residual virus remained to account for the increase in infection rates noted over 72 hours. Importantly, modified cells remain viable following pegylation as determined by trypan blue exclusion and cell counts for modified versus control cells were equivalent over the 72 hour period, therefore the increase in infection rates was not a result of cell death. 168 Crucial to the antiviral efficacy of mPEG grafting is the linker chemistry used to graft the polymer to the cell. In addition to the cyanuric chloride mPEG experiments conducted, benzotriazole carbonate and succinimidyl propionate mPEGs were also used to modify cells. Data from these experiments suggests that concentrations of CmPEG 5 kDa lower than 5 mM are more effective at preventing viral infection 24 hours post virus challenge than equivalent concentrations of all molecular weights of the mPEGs tested. However, there were no significant differences at moderate to high concentrations (5-15 mM) of mPEG, which are likely to be used in an mPEG nasal spray due. Indeed, higher mPEG concentrations (10-15 mM) are most desirable for nasal application due to the greater viscosity as well as the increased protective effect seen in the in vitro studies. In addition, high concentrations would likely combat the potential loss of mPEG due to reactions with secreted proteins in the nasopharynx. Several differences were noted between the mPEG linker moieties. As discussed by Bradley, et al. (2002), the BTCmPEG and SPAmPEG reactions occur more slowly than the CmPEG reaction with lysine residues. The slower reaction times and increased rates of hydrolysis were noted in both viral modification and cell surface derivatization with cells significantly less infected when either the cells or virus was modified by CmPEG than those modified by BTCmPEG or SPAmPEG (5 kDa molecular weight). In addition, due to the hydrolysis reactions noted for all three linker moieties, a nasal spray will need to be designed to ensure the active form of mPEG is applied. This may be accomplished through the use of a powder application, or single use aerosolizers in which the mPEG is solubilized just prior to use. In 169 addition to the effect of hydrolysis and reaction rates of the various linker chemistries on the antiviral efficacy of mPEG, the linker moieties may modify different lysine residues. Interestingly, despite the somewhat less effective prevention of initial viral infection (24 hours post challenge) observed in cells modified with BTCmPEG, the 20 kDa polymer proved to be much more effective in preventing progeny virus infection (48 and 72 hours) than the other mPEGs tested (Tables 5.2 and 5.5). As depicted in Figure 7.5, the protection against progeny virus infection is thought to result from virus becoming trapped within mPEG-coated vesicles or by the mPEG layer while exiting the cells. These vesicles would be inhibited from endocytic or fusion events due to the mPEG barrier, thereby preventing progeny virus infection. Furthermore, as long as the cell membrane remains intact, the viruses being shed from infected cells will be inhibited from passing back through the mPEG layer, which would inhibit their ability to infect cells. Therefore, the mPEG brush border may act to contain progeny virus, inhibiting propagation of the infection. Protection against progeny virus infection is important in a clinical setting, as the common cold will resolve more rapidly if the infection cannot be propagated. Indeed this is the primary mechanism by which Zicam functions. In addition, the self-limiting grafting effects of preformed polymers may be important as large polymers tend to occlude additional binding sites while small polymers allow a higher degree of grafting. This is clearly shown by the study of Bailon, et. al. (2001), in which 40 kDa branched mPEG polymers modified interferon at fewer sites than a 5 kDa linear mPEG (Monkarch, et. al. 1997). The 5 kDa mPEG modified all 11 lysine residues, but not the N-170 terminal cysteine, while the 40 kDa polymer was only able to modify 6 of the lysine residues. A similar phenomenon likely occurs with the modification of cells as well as virus particles, as the 2 or 3.4 kDa polymers would be less sterically hindered from reacting with surface lysine residues than the 20 kDa mPEGs. Formation of m P E G Coated Vesicles Containing Virus Entrapment of Virus Particles within the m P E G Brush Border m P E G Coated Vesicle • Virus m P E G Figure 7.5: Prevention of Progeny Virus Infection by 20 kDa mPEGs. The protection against progeny virus infection by 20 kDa mPEGs, described in Chapter 4, is thought to be the result of virus becoming trapped within mPEG coated vesicles or just below the mPEG layer while exiting the cells. Additionally, when comparing the naked virus, SV40, with the enveloped virus, CMV, it appears CMV progeny were more inhibited from infecting cells than SV40. As illustrated in Figure 7.6, the average difference in the percent cells infected with SV40 or CMV at 72 hours versus 24 hours is significant at low (0.2 and 0.6 mM) BTCmPEG 20 kDa concentrations. Cells are less infected with progeny CMV than with progeny SV40. This inhibition may be the result of CMV acquiring a portion of the pegylated cell surface proteins as the virus buds off 171 from the cell. The inadvertent pegylation of C M V further inhibits the virus from interacting with receptors for attachment/fusion, thereby preventing infection. Alternatively, these results could reflect the simple matter of virus size, where the mPEG barrier is more effective against larger viruses such as C M V (200 nm) compared with the smaller SV40 (45 nm). Another interesting difference was observed between T M E V , M A V and RCV infections of BTCmPEG-modified cells. M A V , the largest ofthe protein capsid viruses (70-90 nm) was more inhibited from infecting cells modified with low to moderate concentrations of mPEG (0.2-2.4 mM) than T M E V , the smallest virus utilized (20-30 nm). Additionally, RCV (comparatively large enveloped virus; approximately ten times the size of TMEV) was more inhibited than the smaller viruses at low to moderate concentrations. The R C V inhibition is likely two-fold. First, the size of the virus is much larger than T M E V or M A V and therefore should be greatly inhibited from passing through the mPEG brush border. Secondly, because RCV is an enveloped virus any initial infection may result in pegylated progeny virus as described above for C M V . Unfortunately, the effect of size on enveloped virus infection of pegylated cells could not be determined from the RCV and C M V studies based on similarity in size (160 vs. 200 nm) as well as the different experimental designs. Table 7.2 displays the percent inhibition of cytopathic effect for BTCmPEG 5 kDa modified cells infected with T M E V , M A V or RCV. Note that high concentrations showed there are relatively minor differences between the three viruses, suggesting that the more dense brush border is equally capable of preventing infection from very small, medium and very large viruses. 172 SV40 72 Hours C M V 72 Hours 15 Concentration B T C m P E G 20 kDa (mM) Figure 7.6 C M V Progeny Virus Infection is More Inhibited than that of SV40 at Low Concentrations of mPEG. The difference in rates of infection were calculated as percent cells infected at 72 hours minus percent cells infected at 24 hours for each of n=3 independent experiments. Note at low BTCmPEG 20 kDa concentrations (0.2 and 0.6 mM) cells are significantly more infected with progeny SV40 than C M V (* p < 0.05). The standard deviation is shown. This may be due to either the size differential (SV40- 45 nm vs. C M V - 200 nm) or the difference in virus structure (e.g., naked vs. enveloped) 173 Table 7.2: Low mPEG Concentrations Inhibit Larger Viruses More Readily than Smaller Viruses Virus 0.2mM 0.6 mM 1.2 mM 2.4 mM 5 mM 10 mM 15 mM RCV 94.6 98.2 98.2 99.0 100 100 100 BTC 5 kDa ±3.4 ±2.1 ±2.6 ±0.6 ±0 ±0 ±0 MAV 78.2 85.1 87.7 89.5 92.5 96.6 98.5 BTC 5 kDa ±13.5 ±12.1 ±10.1 ±8.9 ±7.9 ±3.6 ±1.3 TMEV 47.0 60.8 63.8 66.5 82.9 88.8 95.1 BTC 5 kDa ±5.5 ±5.4 ±5.2 ±4.7 ±3.7 ±0.9 ±1.1 The percent inhibition of cytopathic effects for RCV, M A V and T M E V are shown above and was determined by dividing the number of plaques from modified cells by the number of plaques from control cells, subtracting that number from 1 and multiplying by 100. Note the smaller virus (TMEV) is much less inhibited at low to moderate mPEG concentrations (0.2-2.4) than the larger viruses (MAV and RCV). In contrast, high concentrations of mPEG (15 mM) are nearly identical in the percent inhibition of infection. In sum, unlike other current methods of preventing viral infection, such as anti-ICAM antibody therapy (RRMA), soluble ICAM (sICAM), Zicam, and vaccination, pegylation of cell surfaces or virus particles is a multivalent prophylactic antiviral treatment. As shown in Table 7.3 mPEG modification is the only preventative drug therapy that is effective against a broad spectrum of viruses. As previously discussed, RRMA, sICAM and Zicam are only effective against rhinoviruses that utilize the ICAM-1 receptor, and are ineffective in preventing infections by other viruses. While vaccination is an important method of preventing disease over several years, no single inoculation can prevent infection/illness by more than a few viruses (e.g., M M R vaccine) or strains of virus (e.g., flu shots). Additionally, these current methods of preventing viral infection have important and potentially serious drawbacks including being highly specific, potentially immunogenic/allergenic, being rapidly cleared from 174 the body or site of infection, and development of resistant strains (or existence of inherently resistant viral strains). Immunocamouflage stands out as a broad-spectrum antiviral with few drawbacks and is therefore a potent means of preventing viral infections. Table 7.3: Comparison of Immunocamouflage to Current Methods of Preventing Viral Infection Effective Against mPEG Modification RRMA (anti-ICAM Ab) sICAM Zicam Vaccination Adenoviridae + * ND * ND ND* -Coronaviridae + * ND * ND ND* -Herpesviridae + ND* - * ND + Papovaviridae + * ND -* ND -Picornaviridae + + + + + Antiviral Traits Non-Specific + - - - -Non-Immunogenic + - + + -Slow Drug Clearance + + Lack of Inherent Resistance + Longevity of Protection + +/+++ ND Not Determined. While these treatments were not tested/reported for the all the viral families shown, there is no evidence suggesting that they would be effective due to the reported specificity of the drug on a particular receptor-ligand model - often restricted to a small number of viruses within the above families. As shown above, mPEG modification of cells and/or viruses is effective against a broad spectrum of viral families and has few drawbacks unlike the other methods of preventing infection. 175 7.2 Potential Effects of Pegylation on Normal Function Based on the hypothesis of this thesis, pegylation of the nasal epithelium would inhibit virus-receptor interactions in the upper respiratory tract. In so doing, fewer cells should initially become infected. The remaining virus should be washed into the lower respiratory tract or stomach where many common cold viruses cannot effectively replicate. For those cells that do become infected, it is thought that the infection will proceed as normal, with progeny virus produced. However, these viruses would be less likely to infect the pegylated cells of the nasopharynx based on two factors: 1) my studies suggest progeny virus may become trapped by the mPEG brush border, and 2) the existing mPEG barrier on uninfected cells prophylactically prevents further viral infection. The result is thought to be fewer cells releasing cytokines and therefore less cellular infiltrate and secretions leading to an absence or decrease in the severity of symptoms. One concern may be that NK cell, macrophage or T cell mediated killing of infected pegylated epithelia may be impaired due to the inhibition of cell-cell interactions. While this may be the case, cells infected with virus often express a different complement of proteins on their cell surfaces (Abbas, 2000). The turnover or alteration in cell surface proteins would reduce the density of the mPEG brush border following infection. This reduction in the amount of mPEG derivatization of the cell surface proteins would likely allow the cell-cell interactions required for immune effector cells to function properly. On a related note, pegylation of the nasal cavity may reduce the sense of smell. However, this reduction would also be experienced during a cold due to the large volume of secretions in the nasopharynx that block smell receptors. Due to the rapid turnover of cells in the nasal cavity 176 discussed in Chapter 4, any loss in the sense of smell would be short-lived once use ofthe mPEG spray is ceased (Carson, et al., 1981). Some questions might also be raised as to the inadvertent production of pegylated super viruses that are capable of infecting cells in other areas of the body by escaping immune surveillance. This is highly unlikely to occur for a number of reasons. First, if the virus is sufficiently pegylated so that it is incapable of infecting the upper respiratory tract, it will remain pegylated and therefore unable to infect other cells throughout the body. Secondly, if this pegylated virus encounters adverse conditions such as pH or temperature, which increase the rate of capsid disassembly, the virus will be further inactivated. Thirdly, should this pegylated virus particle enter the blood stream, the result may in fact be an enhanced immune response like that seen by Mizouni (2000). As shown in Figure 7.7, the mPEG modification may in fact act as a time-release capsule slowly releasing portions of the viral capsid as it breaks apart and exposing these proteins to the immune system. 177 mPEG Modified Virus in Circulation Unmodified V P 1 Clusters Unmodified V P 1 X6b <9? aw igylated V P 1 Clusters W h e n the c a p s i d deg rades , s o m e proteins lack m P E G and are recogn ized by the immune sys tem Pegylated VP1 Further degradat ion of the c a p s i d results in more protein b e c o m i n g a c c e s s i b l e enhanc ing the immune response Figure 7.7: Time-Release Effect of mPEG Modified Virus In Vivo. In the mouse model explored by Mizouni (2000), an enhanced immune response was observed following intraperitoneal administration of mPEG modified SV40. This enhanced response is thought to be due to a time-release effect of mPEG. The virus capsid naturally breaks apart over time and in doing so, some of the proteins that are shed will lack the immunocamouflage of mPEG. Therefore, the immune system will recognize and respond to the foreign proteins as they are slowly released from the degrading capsid. Modified from Mizouni, 2000. 178 7.3 Future Directions Further examination of the broad-spectrum antiviral effect of mPEG grafting should be expanded to include several of the viruses listed in Table 7.1 including RSV, SARS, Influenza and HIV. There are currently plans for RSV and.HIV work to be completed in Mark Scott's laboratory or through collaborations. Preliminary results with RSV have further demonstrated the broad-spectrum antiviral effect of mPEG derivatization of either virus particles or their host cells. Therefore, pegylation as a method of preventing viral infection may have profound effects on outbreaks of viral diseases caused by respiratory syncytial virus, influenza and common cold viruses. When illness is recognized, mPEG may be administered to the symptomatic patient as well as anyone that person has come into contact with. Administration to symptomatic patients would result in the modification of both progeny viruses and cells causing an attenuation of infection, similar to the attenuation of rhinovirus infection seen with Zicam treatment. Furthermore, mPEG administration to hospital workers and family members would protect them from viral infection through the modification of cell surfaces. Additionally, because mPEG treatments are easily administered and react quickly, they may be a vital defense against bioterrorist use of the smallpox virus, Variola, against which few in the western world have been immunized since it was declared eradicated on May 8th 1980 (Pennington, 2003). In fact, vaccination was ceased in the United States in 1972, resulting in nearly half the present population vulnerable (Bray, 2003). Variola used for bioterrorism would likely be an aerosolized form in which infection will occur through either the respiratory tract 179 or breaks in the skin, therefore an mPEG nasal spray and/or inhaler could be used to prevent infection for the millions of people not currently immunized. Moreover, this thesis has clearly shown that mPEG modification of cells prevents a broad-spectrum of viral infections. This potential universality of cellular pegylation can be further tested with additional viruses including HIV, respiratory syncytial virus, and even bacteriophage or plant viruses such as T4 or tobacco mosaic virus. Conversely, these viruses can also be pegylated to test the universality of virus modification in preventing infection of unmodified target cells. In addition, structural modeling of viruses may be used to predict which combinations of the various molecular weight mPEGs would be optimal for mPEG modification of viruses to prevent infection. Results from the work presented in this thesis have created exciting possibilities for additional future studies. The most obvious future work based on this thesis is an animal model of nasal cavity pegylation to prevent viral infection. To this end, a respiratory syncytial virus (RSV) guinea pig model is currently being explored in our laboratory. Ultimately, clinical trials of an mPEG nasal spray would need to be conducted if this prophylactic treatment is ever to reach a consumer market. Additional in vitro studies combining different mPEG linker chemistries and molecular weights (e.g., CmPEG 5 kDa with BTCmPEG 20 kDa) would be beneficial in further optimizing the mPEG nasal spray. In addition, the results from Mizouni (2000), which indicated an increased immune response to modified virus discussed above, opens further avenues of study. Based on these studies, it is 180 thought that heavily derivatized virus may be useful as a vaccine. The modified and disassembled viral capsids of poliovirus, rotavirus or adenoviruses can be examined for their ability to generate similar immunologic responses to those seen with SV40. Enveloped viruses such as RSV and West Nile virus may be tested as well, though results will vary depending upon the amount of disruption of the lipid bilayer following pegylation. Thus far modification of cells has not resulted in the shearing of the lipid membrane; therefore pegylation of enveloped viruses may not result in the time-release effects seen in viral capsid pegylation. Nevertheless, it is a worthwhile avenue of study. 181 References Abbas AK, Lichtman A H , Pober JS, ed. 2000. Cellular and Molecular ImmunoloRV. WB Sanders Company, New York. 253-355. Abuchowski A, McCoy JR, Palczuk NC, van Es T, Davis FF. 1977. Effect of covalent attachment of polyethylene glycol on immunogenicity and circulating life of bovine liver catalase. JBiol Chem.\0;252(\l):3582-3586. Aho K, Somer T, Salo OP. 1967. Rheumatoid factor and immuno-conglutinin responses following various vaccinations. Proc Soc Exp Biol Med. 124(1): 229-233. Alexandrov AI, Botchan MR, Cozzarelli NR. 2002. Characterization of simian virus 40 T-antigen double hexamers bound to a replication fork. The active form of the helicase. JBiol Chem. 211 (Al): 44886-44897. Allen C, Dos Santos N, Gallagher R, Chiu GNC, Shiu Y, Li M , Johnstone SA, Janoff AS, Mayer LD, Webb MS, Bally MB. 2002. Controlling the physical behavior and biological performance of liposome formulations through use of surface grafted poly(ethylene glycol). Bioscience Reports. 22(2): 225-250. Alvarez M , Martinez-Laso J, Varela P, Diaz-Campos N, Gomez-Casado E, Vargas-Alarcon G, Garcia-Torre C and Arnaiz-Villena A. 1997. High polymorphism of Mhc-E locus in non-human primates: alleles with identical exon 2 and 3 are found in two different species. Tissue Antigens. 49(2): 160-167. Amorosa V K , Isaacs SN. 2003. Separate worlds set to collide: smallpox, vaccinia virus vaccination, and human immunodeficiency virus and aquired immunodeficiency syndrome. Clin Infect Dis. 37(3): 426-432. Annamalai A, Shreekumar S, Manoharan V. 1967. Neurological complications following antirabies vaccination. Trans R Soc Trop MedHyg. 61 (6): 839-841. Armstrong JK, Leger R, Wenby RB, et. al. 2003. Occurance of an antibody to poly(ethylene) glycol. Blood. 102: 556a (abstr). 182 Avramis VI, Sencer S, Periclou AP, Sather H, Bostrom BC, Cohen LJ, Ettinger A G , Ettinger LJ, Franklin J, Gaynon PS, Hilden JM, Lange B, Majlessipour F, Mathew P, Needle M , Neglia J, Reaman G, Holcenberg JS, Stork L. 2002. A randomized comparison of native Escherichia coli asparaginase and polyethylene glycol conjugated asparaginase for treatment of children with newly diagnosed standard-risk acute lymphoblastic leukemia: a Children's Cancer Group study. Blood. 99(6): 1986-1994. Bailon P, Palleroni A, Schaffer CA, Spence CL, Fung WJ, Porter JE, Ehrlich GK, Pan W, Xu ZX, Modi MW, Farid A, and Berthold W. 2001. Rational Design of a Potent, Long-Lasting Form of Interferon: A 40 kDa Branched Polyethylene Glycol-Conjugated Interferon r-2a for the Treatment of Hepatitis C. Bioconjugate Chem. 12: 195-202. Barrett WE, Hanigan JJ. 1975. The bioavailability of griseofulvin PEG ultramicrosize (Gris-PEG) tablets in man under steady-state conditions. Curr Ther Res Clin Exp. 18(3):491 -500. Bella J, Kolatkar PR, Marlor CW, Greve JM, Rossman M G . 1998. The structure of the two amino-terminal domains of human ICAM-1 suggests how it functions as a rhinovirus receptor and as an LFA-1 integrin ligand. Pro. Natl. Acad. Sci. USA 95: 4140-4145. Belshe RB, Mendelman PM. 2003. Safety and Efficacy of Live Attenuated, Cold-Adapted, Influenza Vaccine-Trivalent. Immunol. Allergy Clin. North Am. 23(4): 745-67. Bendas G, Rothe U, Scherphof GL and Kamps JA. 2003. The influence of repeated injections on pharmacokinetics and biodistribution of different types of sterically stabilized immunoliposomes. Biochim. Biophys. Acta. 1609(1): 63-70. Bhargava A K , Woitach JT, Davidson EA and Bhavanandan VP. 1990. Cloning and cDNA sequence of a bovine submaxillary gland mucin-like protein containing two distinct domains. Proc. Natl. Acad. Sci. USA. 87:6798-6802. Black S, Shinefield H, Hansen J, Lewis E, Su L, Coplan P. 2004. A post-licensure evaluation of the safety of inactivated hepatitis A vaccine (VAQTA, Merk) in children and adults. Vaccine. 22(5-6): 766-772. Block KI and Mead M N . 2003. Immune system effects of echinacea, ginseng and astragalus: a review. Integr. Cancer Ther. 2(3): 247-267. Bradley AJ, Murad K L , Regan K L , Scott MD. 2002. Biophysical consequences of linker chemistry and polymer size on stealth erythrocytes: Size does matter. Biochimica et Biophysica Acta, 1561(2): 147-58. Bradley AJ, Test TT, Murad K L , Mitsuyoshi J, Scott MD. 2001. Complement interactions with methoxypoly(ethylene glycol)-modified human erythrocytes. Transfusion, 41: 1225-1233. 183 Bowen S, Tare N, Inoue T, Yamasaki M , Okabe M , Horii I, Eliason JF. 1999. Relationship between molecular mass and duration of activity of polyethylene glycol conjugated granulocyte colony-stimulating factor mutein. Exp Hematol. 27(3):425-432. Bray M . 2003. Pethogenesis and potential antiviral therapy of complications of smallpox vaccination. Antiviral Res. 58(2): 101-114. Brinkman K, Smeitink JA, Romijn JA, Reiss P. 1999. Mitochondrial toxicity induced by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-therapy-related lipodystrophy. Lancet 354(9184): 1112-5. Buhrmester CC. 1933. A study of the hydrogen-ion concentration, nitrogen content, and viscosity of nasal secretions. Ann. Otolo. Rhinol. Laryngol. 42: 1041-1057. Callahan L. 1994. HIV-1 virion-cell interactions: an electrostatic model of pathogenicity and syncytium formation. AIDS Res Hum Retroviruses. 10(3):231-233. Cantin A M , Woods DE, Cloutier D, Dufour EK, Leduc R. 2002. Polyethylene glycol conjugation at Cys232 prolongs the half-life of alphal proteinase inhibitor. Am J Respir Cell Mol Biol. 27(6): 659-65. Carignano M A , Szleifer 11. 2000. Prevention of protein adsorption by flexible and rigid chain molecules. Colloids Surf B Biointerfaces. 18(3-4): 169-182. Carson JL, Collier A M , Knowles MR, Boucher RC and Rose JR. 1981. Morphometric aspects of ciliary distribution and ciliogenesis in human nasal epithelium. Proc. Natl. Acad. Sci. USA. 78(11 ):6996-6999. Centers for Disease Control and Prevention (No Author Listed). 2004. Update: adverse events following civilian smallpox vaccination- United States, 2003. MMWR Morb Mortal WklyRep. 53(5): 106-107. Charles CH, Luo GX, Kohlstaedt LA, Morantte IG, Gorfain E, Cao L, Williams JH, Fang F. 2003. Prevention of human rhinovirus infection by multivalent Fab molecules directed against ICAM-1. Antimicrobial Agents and Chemotherapy. 47: 1503-1508. Charles CH, Luo GX, Kohlstaedt L A , Morantte IG, Gorfain E, Cao L, Williams JH, Fang F. 2003. Prevention of human rhinovirus infection by multivalent fab molecules directed against ICAM-1. Antimicrob. Agents Chemotherapy. 47(5): 1503-8. Chen A M and Scott MD. 2003. Immunocamouflage: Prevention of transfusion-induced graft-versus-host disease via polymer grafting of donor cells. J Biomed Mater Res. 67A(2):626-36. Chen A M , Scott MD. 2001. Current and future applications of immunologic attenuation via PEGylation of cells and tissues. BioDrugs. 15(12): 833-847. 184 Cheng T L , Wu PY, Wu MF, Chern JW and Roffler SR. 1999. Accelerated clearance of polyethylene glycol-modified proteins by anti-polyethylene glycol IgM. Bioconjugate Chem. 10:520-528. Cheng TL, Chen B M , Chern JW, Wu MF and Roffler SR. 2000. Efficient clearance of poly(ethylene glycol)-modified immunoenzyme with anti-PEG monoclonal antibody for prodrug cancer therapy. Bioconjugate Chem. 11:258-266. Chou S, Scott K M . 1988. Rapid quantitation of cytomegalovirus and assay of neutralizing antibody using monoclonal antibody to the major immediate early viral protein. J. Clin. Microbiology 26: 504-507. Conti C, Mastromarino P, Riccioli A, Orsi N. 1991. Electrostatic interactions in the early events of V S V infection. Res Virol. 142(1): 17-24. Conway B. 2000. Initial Therapy with Protease Inhibitor-Sparing Regimens: Evaluation of Nevirapine and Delavirdine. Clin Infect. Dis. Supplement 2: SI30-4. de Gennes PG. 1980. Conformation of polymers attached to an interface. Macromolecules. 13: 1069-1075. Doucet J, Gao ZH, MacLaren L A , McAlister V C . 2004. Modification of xenoantigens on porcine erythrocytes for xenotransfusion. Surgery. 135(2): 178-86. Eby GA, Davis DR, Halcomb WW. 1984. Reduction in duration of common colds by zinc gluconate lozenges in a double blind study.. Antimicrob. Agents Chemother. 25: 20-24 Edghill-Smith Y, Venzon D, Karpova T, McNally J, Nacsa J, Tsai WP, Tryniszewska E, Moniuszko M , Manischewitz J, King LR, Snodgrass SJ, Parrish J, Markham P, Sowers M , Martin D, Lewis M G , Berzofsky JA, Belyakov IM, Moss B, Tartaglia J, Bray M , Hirsch V, Golding H, Franchini G. 2003. Modeling a safer smallpox vaccination regimen, for human immunodeficiency virus type 1 -infected patients, in immunocompromised macques. J Infect Dis. 188(8): 1181-1191. Fabricant, ND. 1941. Significance of the pH of nasal secretions. Arch. Otolaryngol. 34:150-163. Fields BN, Knipe PM, Howley, ed. 1998. Papovaviridae. Fundamental Virology, 2 n d Edition. Lippincott- Raven Publishers, Philadelphia, pgs. 917-945. Fisher TC, Armstrong JK, Wenby RB, et. al. 2003. Isolation and identification of a human antibody to poly(ethylene glycol). Blood. 102:559a (abstr). Flint SJ, Enquist LW, Krug R M , Racaniello VR, Skalka A M . 2000. Prevention and control of viral diseases. Principles of Virology: Molecular biology, pathogenesis, and control, A S M Publishing, New York, pgs. 662-712. 185 Fu J, Fiegel J, Krauland E, Hanes J. 2001. New polymeric carriers for controlled drug delivery following inhalation or injection. Biomaterials 23(22):4425-4433. Gaertner DJ, Winograd SR, Compton FX, Paturzo A L , Smith A L . 1993. Development and optimization of plaque assays for rat coronaviruses. J. Virol. Methods 43: 53-64. Gagneten S, Scanga CA, Dveksler GS, Beauchemin N, Percy D, Holmes K V . 1996. Attachment glycoproteins and receptor specificity of rat coronaviruses. Lab. Anim. Sci. 46: 159-166. Galli M , Ridolfo A L , Adorni F, Gervasoni C, Ravasio L, Corsico L, Gianelli E, Piazza M , Vaccarezza M , d'Arminio Monforte A, Moroni M . 2002. Body habitus changes and metabolic alterations in protease inhibitor-naive HIV-1-infected patients treated with two nucleoside reverse transcriptase inhibitors. J Acquir Immune Defic Syndr. 29(1):21-31. Garratty, G. 2004. Progress in modulating the RBC membrane to produce transfusable universal/stealth donor RBCs. Trans. Med. Reviews. 18(4): 245-256. Gatto LA. 1981. pH of mucus in rat trachea. J. Appl. Physiol. Respirat. Environ. Exercise Physiol. 50(6): 1224-1226. Godfrey JC, Conant Sloane B, Smith DS, Turco JH, Mercer N, Godfrey NJ. 1992. Zinc gluconate and the common cold: a controlled study. JInt Med Res. 20:234-246. Goodson RJ, Katre NV. 1990. Site-directed PEGylation of recombinant interleukin-2 at its glycosylation site. Biotechnology (N Y). 8(4): 343-6. Greve JM, Davis G, Meyer A M , Forte CP, Yost SC, Marlor CW, Kamarck M E and McClelland A. 1989. The major human rhinovirus receptor is ICAM-1. Cell. 56(5):839-847. Gudmundsdottir H, Turka L A . 1999. T cell Costimulatory Blockade: New Therapies for Transplant Rejection. J Am Soc Nephro. 10:1356-1365. Gupta JP, Bajpai HS. 1966. Neuro-paralytic complications following antirabic vaccination. Indian J Med Sci. 20(4): 276-279. Gwantley JM and Druce H M . 1997. Efficacy of brompheniramine maleate for the treatment of rhinovirus colds. Clin Infect. Dis. 25:1188-1194. Gwantley JM, Winther B, Patrie JT and Hendley JO. 2002. Combined antiviral-antimediator treatment for the common cold. J. Infect. Dis. 186:147-154. 186 Hall CB, Geiman JM, Douglas RG Jr, Meagher MP. 1978. Control of nosocomial respiratory syncytial viral infections. Pediatrics. 62(5):728-732. Halperin SA, Eggleston PA, Hendley JO, Suratt PM, Groschel DH, Gwaltney JM Jr. 1983. Pathogenesis of lower respiratory tract symptoms in experimental rhinovirus infection. Am. Rev. Respir. Dis. 128: 17-20. Hanna GJ, D'Aquila RT. 1999. Antiretroviral Drug Resistance in HIV-1. Current Infectious Disease Report. 1(3): 289-297. Hayden FG, Gwaltney JM, Colono RJ. 1988. Modification of experimental rhinovirus colds by receptor blockade. Antiviral Res. 9(4): 233-47. Hendley JO. 1999. Clinical virology of rhinoviruses. Ad. Virus Res. 54: 435-466. Hermans J. 1982. Excluded volume theory of polymer-protein interactions based on polymer chain statistics. J. Chem. Phys. 77: 2193-2203. Herold, D. A.; Keil, K.; Bruns, D. E. 1989. Oxidation of polyethylene glycols by alcohol dehydrogenase. Biochem. Pharmacol. 38: 73-76. Herrier G, Gross HJ, Imhof A, Brossmer R, Milks G, Paulson JC. 1992. A synthetic sialic acid analogue is recognized by influenza C virus as a receptor determinant but is resistant to the receptor-destroying enzyme. J Biol Chem. 267(18): 12501-12505. Heymann K. 1967. Herpes zoster following vaccination. Practitioner. 198(184): 280-282. Hidalgo S, Garcia Erro M , Cisterna D, Freire M C . 2003. Paralytic poliomyelitis caused by a vaccine-derived polio virus in an antibody-deficient Argentinian child. Pediatr Infect DisJ. 22(6): 570-572. Hirt M , Nobel S, Barron E. 2000. Zinc nasal gel for the treatment of common cold symptoms: a double-blind placebo-controlled trial. Ear Nose & Throat Journal 79(10): 778-780. Hoffman, R M . 2003. PEG-methioninase. Adv Exp Med Biol. 519:69-79. Ion-Nedelcu N, Dobrescu A, Strebel PM, Sutter RW. 1994. Vaccine-associated paralytic poliomyelitis and HIV infection. Lancet 343(8888): 51-52. Irmiere A, Gibson W. 1983. Isolation and characterization of noninfectious virion-like particles released from cells infected with human strains of cytomegalovirus. Virology 130:118-133. Jackson CJ, Charlton JL, Kuzminski K, Lang G M , Sehon A H . 1987. Synthesis, isolation, and characterization of conjugates of ovalbumin with monomethoxypolyethylene glycol using cyanuric chloride as the coupling agent. Anal Biochem. 165(1): 114-27. 187 Jackson RM, Veal CF, Beckman JS, Brannen A L . 1990. Polyethylene glycol-conjugated superoxide dismutase in unilateral lung injury due to re-expansion (re-oxygenation). Am J Med Sci. 300(1 ):22-28. Jiang W, Woitach JT, Keil RL and Bhavanandan VP. 1998. Bovine submaxillary contains multiple domains and tandemLy repeated non-identicle sequences. Biochem. J. 331:193-199. Jiao J, Simmons DT. 2003. Nonspecific double-stranded DNA binding activity of simian virus 40 large T antigen is involved in melting and unwinding of the origin. J Virol. 77(23): 12720-8. Johnson D. 2002. Therapeutic Management of HIV. Oral Dis. Supplement 2: 17-20. Kandel R, Hartshorn K L . 2001. Prophylaxis and treatment of influenza virus infection. Biodrugs. 15(5): 303-23. Kawai F, Kimura T, Tani Y, Yamada H, Kurachi M . 1980. Purification and characterization of polyethylene glycol dehydrogenase involved in the bacterial metabolism of polyethylene glycol. Appl Environ Microbiol. 40(4):701-5. Kellam B, De Bank PA, Shakesheff K M . 2003. Chemical modification of mammalian cell surfaces. Chem Soc Rev. 32(6):327-337. Keppler OT, Stehling P, Herrmann M , Kayser H, Grunow D, Reutter W, Pawlita M . 1995. Biosynthetic modulation of sialic acid-dependent virus-receptor interactions of two primate polyoma viruses. J Biol Chem. 270(3): 1308-1314. Kim JJ, Park K. 2001. Glucose-binding property of pegylated concanavalin A. Pharm Res. 18(6):794-799. Kimura M , Matsumura Y, Miyauchi Y, Maeda H. 1988. A new tactic for the treatment of jaundice: an injectable polymer-conjugated Bilirubin oxidase. Proc Soc Exp Biol Med. 188(3):364-9. Kinstler OB, Brems DN, Lauren SL, Paige A G , Hamburger JB, Treuheit MJ. 1996. Characterization and stability of N-terminally PEGylated rhG-CSF. Pharm Res. 13(7): 996-1002. Kirkpatrick GL. 1996. The common cold. Prim Care. 23(4):657-75. Kita Y, Rohde MF, Arakawa T, Fagin KD, Fish EN, Banerjee K. 1990. Characterization of a polyethylene glycol conjugate of recombinant human interferon-gamma. Drug Des Deliv. 6(3):157-167. 188 Kleinbaum DG, Kupper L L . 1978. One-way analysis of variance. Applied Regression Analysis and other Multivariable Methods. Duxbury Press, North Scituate Massachusetts, pp. 244-277. Korant BD, Butterworth BE. 1976. Inhibition by zinc of rhinovirus protein cleavage: Interaction of zinc with capsid polypeptides. J. Virology. 18: 298-306. Kozlowski A, Charles SA, Harris JM. 2001. Development of pegylated interferons for the treatment of chronic Hepatitis C. BioDrugs. 15(7): 419-429. Kunita S, Mori M , Terada E. 1993. Sequence analysis of the nucleocapsid protein gene of rat Coronavirus SDAV-681. Virology 193: 520-523. Kuskall MS, AuBuchon JP, Anthony K Y , Herschel L, Pickard C, Biehl R, Horowitz M , Brambilla DJ and Popovsky MA. 2000. Transfusion to blood group A and O patients of group B RBCs that have been enzymatically converted to group O. Transfusion. 40:1290-1298. Larsen SH, Nathans D. 1977. Mouse adenovirus: growth of plaque purified FL virus in cell lines and characterization of viral DNA. Virology'82: 182-195. Lee JH, Baker TJ, Mahal LK, Zabner J, Bertozzi CR, Wiemer DF, Welsh MJ. 1999. Engineering novel cell surface receptors for virus-mediated gene transfer. J Biol Chem. 274(31):21878-21884. Leger RM, Arndt P, Garratty G. 2001. Normal donor sera can contain antibodies to polyethylene glycol. Transfusion. 41:29S (abstr). Lehman JM, Laffin J, Friedrich TD. 1993. Quantification of simian virus 40 T-antigen correlated with the cell cycle of permissive and non-permissive cells. Cytometry. 14: 401-410. Lehman JM, Laffin J, Friedrich TD. 2000. Simian virus 40 induces multiple S phases with the majority of viral DNA replication in the G2 and second S phase in CV-1 cells Exp Cell Res. 258(1): 215-22. Lenny L L , Hurst R, Goldstein J, Benjamin LJ, Jones RL. 1991. Single-unit transfusions of RBC enzymatically converted from group B to group O to A and O normal volunteers. Blood. 77(6): 1383-1388. Liddington RC, Yan Y, Moulai J, Sahli R, Benjamin T L , Harrison SC. 1991. Structure of simian virus 40 at 3.8-A resolution. Nature 354(6351):278-284. Lintz D, Kapila R, Pilgrim E, Tecson F, Dorn R, Louria D. 1976. Nosocomial Salmonella epidemic. Arch Intern Med. 136(9):968-973. 189 Luo GX, Kohlstaedt L A , Charles CH, Gorfain E, Morantte I, Williams JH, Fang F. 2003. Humanization of an anti-ICAM-1 antibody with over 50-fold affinity and functional improvement. J Immunol Methods. 275(1-2): 31-40. Marlin SD, Staunton DE, Springer TA, Stratowa C, Sommergruber W, Merluzzi VJ. 1990. A soluble form of intercellular adhesion molecule-1 inhibits rhinovirus infection. Nature. 344(6261):70-2. Mastromarino P, Conti C, Petruzziello R, Lapadula R, Orsi N. 1991. Effect of polyions on the early events of Sindbis virus infection of Vero cells. Arch Virol. 121(1-4): 19-27. McHugh RS, Nagarajan S, Wang Y C , Sell KW, Selvaraj P. 1999. Protein transfer of glycosyl-phosphatidylinositol-B7-l into tumor cell membranes: a novel approach to tumor immunotherapy. Cancer Res. 59(10):2433-2437. McKinlay M A , Pevear DC. 1992. Treatment of the picornavirus common cold by inhibitors of viral uncoating and attachment. Annu. Rev. Microbiol. 46:635-654. McLaughlin M , Thomas P, Ontorato I, Rubinstein A, Oleske J, Nicholas S, Krasinski K, Guigli P, Orenstein W. 1988. Live virus vaccines in human immunodeficiency virus-infected children: a retrospective survey. Pediatrics. 82(2): 229-233. Mizouni, S.K., Lehman, J.M., Cohen, B. and Scott, M.D. 1998. Viral modification with methoxypoly(ethylene glycol): Implications for gene therapy and viral inactivation. Blood. (Suppl. 1), 4627. Mizouni, S.K., Bradley, A.J., and Scott, M.D. 1999. Use of a two-phase partitioning system to purify an immunologically attenuated viral vector. Blood. (Suppl. 1):5081 (415b). Mizouni, S.K. Viral modification with methoxypoly(ethylene glycol): Implication for viral inactivation and gene therapy. Masters Thesis, Albany Medical College, Accepted May, 2000. Murad K L , Gosselin EJ, Eaton JW, Scott MD. 1999A. Stealth cells: prevention of Major histocompatibility complex class 11-mediated T-cell activation by cell surface modification. Transfusion Medicine. 94(6): 2135-2141. Murad K L , Mahany K L , Kuypers FA, Brugnara C, Eaton JW, Scott MD. 1999B. Structural and functional consequences of antigenic modulation of red blood cells with methoxypoly(ethylene glycol). Blood., 93:2121-2127. Musiani M , Zerbini M , Gentilomi G, LaPlaca M . 1988. Rapid quantitative assay of cytomegalovirus infectivity. J. Virol. Methods 20: 333-340. 190 Nakagomi K, Ajisaka K. 1990. Chemical modification of alpha-thrombin and in vitro characterization of its anticoagulant activity. Biochem Int. 22(l):75-84. National Institutes of Health. 2003. The common cold. http://www.niaid.nih.gov/factsheets/cold.htm Novick SG, Godfrey JC, Godfrey NJ, Wilder HR. 1996. How does zinc modify the common cold? Clinical observations and implications regarding mechanisms of action. Med Hypotheses. 46(3):295-302. Oetke C, Brossmer R, Mantey LR, Hinderlich S, Isecke R, Reutter W, Keppler OT, Pawlita M . 2002. Versatile biosynthetic engineering of sialic acid in living cells using synthetic sialic acid analogues. J Biol Chem. 277(8):6688-6695. O'Farrell PZ, Goodman H M . 1976, Resolution of simian virus 40 proteins in whole cell extracts by two-dimensional electrophoresis: heterogeneity of the major capsid protein. Cell. 9(2):289-298. Ogris M , Brunner S, Schuller S, Kircheis R, Wagner E. 1999. PEGylated DNA/transferrin-PEI complexes: reduced interaction with blood components, extended circulation in blood and potential for systemic gene delivery. Gene Ther. 6(4): 595-605. Osondu-Alilonu N, Gross PA. 2003. Influenza: the recurring, re-emerging infection. Current Treatment Optionsin Infectious Diseases. 5: 129-136. Panza JL, Wagner WR, Rilo HL, Rao RH, Beckman EJ, Russell AJ. 2000. Treatment of rat pancreatic islets with reactive PEG. Biomaterials. 21(11): 1155-64. Park Y K , Abuchowski A, Davis S, Davis F. 1981. Pharmacology of Escherichia coli-L-asparaginase polyethylene glycol adduct. Anticancer Res. l(6):373-376. Pekari K, Schmidt RR. 2003. A variable concept for the preparation of branched glycosyl phosphatidyl inositol anchors. J Org Chem. 68(4): 1295-1308. Pennington H. 2003. Smallpox and bioterrorism. Bull. World Health Organ. 81(\0):762-767. Pinheiro JP, Lanvers C, Wurthwein G, Boos J. 2002. Pharmacology of PEG-asparaginase in childhood acute lymphoblastic leukemia (ALL). Blood. 100(5): 1923-1924. Pitaranta A, Hayden FG. 1998. Rhinoviruses: important respiratory pathogens. Ann. Med. 30: 529-537. Premkumar DR, Fukuoka Y, Sevlever D, Brunschwig E, Rosenberry T L , Tykocinski M L , Medof M E . 2001. Properties of exogenously added GPI-anchored proteins following their incorporation into cells. J Cell Biochem. 82(2):234-45. 191 Proud D, Reynolds CJ, Lacapra S, Kagey-Sobotka A, Lichtenstein L M , Naclerio RM. 1988. Nasal provocation with bradykinin induces symptoms of rhinitis and a sore throat. Am Rev Respir Dis. 137(3):613-6. Proud D, Bathon JM, Togias A G , Naclerio RM. 1995. Inhibition of the response to nasal provocation with bradykinin by HOE-140: efficacy and duration of action. Can J Physiol Pharmacol. 73(7):820-6. Preston SL, Piliero PJ, Drusano GL. 2003. Pharmacodynamics and clinical use of anti-HIV drugs. Infectious Disease Clinics of North America. 17: 651-74. Richter AW and Akerblom E. 1983. Antibodies against polyethylene glycol produced in animals by immunization with monomethoxy polyethylene glycol modified proteins. Int Arch Allergy Appl Immunol. 70(2): 124-131. Richter A W and 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(l):36-39. Roberts MJ, Bently MD, Harris JM. 2002. Chemistry for protein PEGylation. Advanced Drug Delivery Reviews, 54: 459-476. RosnerB. 1995. Fundamentals of Biostatistics 4 th Edition. Wadsworth Publishing Company, Bemnont, CA. pgs. 299-344. Rozhon EJ, Kratochvil JD, Lipton HL. 1983. Analysis of genetic variation in Theiler's virus during persistent infection in the mouse central nervous system. Virology 128: 16-32. Rudin C M , Van Dongen J, Thompson CB. 1996. Apoptotic signaling lymphocytes. Curr Opin Hematol. 3:35-40. Soares A L , Guimaraes G M , Polakiewicz B, de Moraes Pitombo RN, Abrahao-Neto J. 2002. Effects of polyethylene glycol attachment on physicochemical and biological stability of E. coli L-asparaginase. Int J Pharm. 237(1-2): 163-170. Satulovsky J, Carignano M A , Szleifer I. 2000. Kinetic and thermodynamic control of protein adsorption. Proc Natl Acad Sci USA. 97(16):9037-9041. Scott M D and Chen A M . 2004. Beyond the Red Cell: PEGylation of other blood cells and tissues. Transfus Clin Biol. 11(1): 40-6. Scott MD, Bradley AJ, Murad K L . 2000. Camouflaged blood cells: Low-technology bioengineering for transfusion medicine. Transfusion Medicine Reviews. 14(1): 53-63. 192 Scott MD, Murad K L , Eaton JW. 1997. The other blood substitute: Antigenically inert erythrocytes. In Advances in Blood Substitutes: Industrial Opportunities and Medical Challenges (Winslow RM, Vandegriff KD, Intaglietta M eds) Birkhauser, Boston. Pgs. 133-150. Scott MD, Murad K L . 1998. Cellular camouflage: Fooling the immune system with polymers. Current Pharmaceutical Design. 4: 423-438. Shearwater Polymers Inc. 2000. Functionalized biocompatible polymers for research and pharmaceuticals: Polyethylene glycol and derivatives. Huntsville, A L pp. 1-50. Shao N, Xue J, Guo Z. 2003. Chemical synthesis of CD52 glycopeptides containing the acid-labile fucosyl linkage. J Org Chem. 68(23):9003-9011. Shorr RG, Kwong S, Gilbert C and Benesch RE. 1999. Changes in the functional properties of bovine hemoglobin induced by covalent modification with polyethylene glycol. Artif. Cells Blood Substit. Immobil. Biotechnol. 27(3): 185-202. Sloand E M , Maciejewski JP, Dunn D, Moss J, Brewer B, Kirby M , Young NS. 1998. Correction of the PNH defect by GPI-anchored protein transfer. Blood. 92(11): 4439-4445. Squires K, Pozniak A L , Pierone G Jr, Steinhart CR, Berger D, Bellos NC, Becker SL, Wulfsohn M , Miller MD, Toole JJ, Coakley DF, Cheng A, Study 907 Team. 2003. Tenofovir Disoproxil Fumarate in Nucleoside-Resistant HIV-1 Infection: a randomized trial. Ann Internal Medicine. 139 (5 Pt 1): 313-20. Stauton DE, Merluzzi VJ, Rothlein R, Barton R, Marlin SD and Springer TA. 1989. A cell adhesion molecule, ICAM-1, is the major surface receptor for rhinoviruses. Cell. 56(5):849-853. Sulkowski MS, Thomas DL, Mehta SH, Chaisson RE, Moore RD. 2002. Hepatotoxicity associated with nevirapine or efavirenz-containing antiretroviral therapy: role of hepatitis C and B infections. Hepatology. 35(1): 182-9. Superti F, Marziano M L , Tinari A, Donelli G. 1993. Effect of polyions on the infectivity of SA-11 rotavirus in LCC-MK2 cells. Comp Immunol Microbiol Infect Dis. 16(l):55-62. Talbot P, Almeida JD. 1977. Human cytomegalovirus: purification of enveloped virions and dense bodies. J. Gen. Virology 36: 345-349. Temple M , Antoine G, Delius H, Stahl S, Winnacker EL. 1981. Replication of mouse adenovirus strain FL DNA. Virology 109: 1-12. 193 Tollefson A E , Hermiston TW, Wold WSM. 1998. Preparation and titration of CsCl-banded adenovirus stock. Methods in Molecular Medicine, Vol. 21: Adenovirus Methods and Protocols Humana Press, Totowa, NJ. TorchilinVP. 1994. Immunoliposomes and pegylated immunoliposomes: possible use for targeted delivery of imaging agents. Immunomethods. 4(3):244-258. Tremblay JD, Sachsenmeier KF, Pipas JM. 2001. Propagation of wild-type and mutant SV40. Methods in Molecular Biology, Vol. 165: SV40 Protocols Humana Press, Totowa, NJ. Tsai N, Cheng T L , and Roffler SR. 2001. Sensitive measurement of polyethylene glycol-modified proteins. BioTechniques 30:396-402. Turner RB, Weingand KW, Yeh CH, Leedy DW. 1998. Association between interleukin-8 concentration in nasal secretions and severity of symptoms of experimental rhinovirus colds. Clin. Infect. Dis. 26(4):840-6. VanLoock MS, Alexandrov A, Yu X, Cozzarelli NR, Egelman EH. 2002. SV40 large T antigen hexamer structure: domain organization and DNA-induced conformational changes. Curr Biol. 12(6): 472-476. van Regenmortel M H V , Fauquent C M , Bishop DHL, Carstens EB, Estes M K , Lemon SM, Maniloff J, Mayo M A , McGeoch DJ, Pringle CR, Wickner RB. 2000. Virus Taxonomy: Classification and Nomenclature of Viruses 7th ICTV Report. Academic Press, NY. pgs. 137-157, Velloso L M , Michaelsson J, Ljunggren HG, Schneider G and Achour A. 2004. Determination of structural principles underlying three different modes of lymphocytic choriomeningitis virus escape from C T L recognition. J. Immunol. 172(9):5504-5511. Waldmann TA. 2003. Immunotherapy: past, present and future. Nature Medicine. 9(3): 269-277. Weber DJ, Ratula WA. 2003. Immunization of immunocompromised persons. Immunol Allergy Clin North Am. 23(4): 605-634. Wentworth BB, French L. 1970. Plaque assay of cytomegalovirus strains of human origin. Proc. Soc. Exp. Biol. Med. 135: 253-258. Yale SH, Liu K. 2004. Echinacea purpurea therapy for the treatment of the common cold: a randomized, double-blind, placebo-controlled clinical trial. Arch Intern Med. 164(11):1237-1241. Yarema KJ, Mahal L K , Bruehl RE, Rodriguez EC, Bertozzi CR. 1998. Metabolic delivery of ketone groups to sialic acid residues. Application to cell surface glycoform engineering. J Biol Chem. 273(47):31168-31179. 194 Zalipsky S and Lee C. 1992. Use of functionalized poly(ethylene glycol)s for modification of polypeptides. Poly(ethylene Glycol) Chemistry: Biotechnical and biomedical applications. (JM Harris ed) Plenum Press, NY; p 347-367. Zubay GL. 1998. Building blocks ofproteins: amino acids, peptides and polypeptides. Biochemistry 4th Edition. Wm. C. Brown Publishers; p 64. 195 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092388/manifest

Comment

Related Items