Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of hsv glycoprotein-glycosaminoglycan interactions Dyer, Angela Patricia 1999

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

Item Metadata

Download

Media
831-ubc_2000-503046.pdf [ 9.53MB ]
Metadata
JSON: 831-1.0089674.json
JSON-LD: 831-1.0089674-ld.json
RDF/XML (Pretty): 831-1.0089674-rdf.xml
RDF/JSON: 831-1.0089674-rdf.json
Turtle: 831-1.0089674-turtle.txt
N-Triples: 831-1.0089674-rdf-ntriples.txt
Original Record: 831-1.0089674-source.json
Full Text
831-1.0089674-fulltext.txt
Citation
831-1.0089674.ris

Full Text

CHARACTERIZATION OF HSV GLYCOPROTEIN-GLYCOSAMINOGLYCAN INTERACTIONS by A N G E L A P A T R I C I A D Y E R B . Sc. (Biology), the University of Winnipeg, 1994 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Department of Microbiology and Immunology) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A November 29, 1999 © Angela Patricia Dyer, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. I Department of Mi Cfl b) O fogy £ ItVm\Af)0lojlu . The University of British Columbia Vancouver, Canada Date /VW . 2,0/^^ DE-6 (2/88) ABSTRACT The two herpes simplex virus (HSV) serotypes, HSV-1 and H S V - 2 , demonstrate different tissue tropisms in the human host. There is increasing evidence that the H S V serotypes may enter host cells differently, which may account, in part, for the different behaviours of these closely related viruses. Initial adsorption of H S V to the host cell is mediated by the interaction of viral glycoproteins with cell surface glycosaminoglycans such as heparan sulfate (HS) and chondroitin sulfate (CS). To investigate the contribution of various cell surface components in the infection pathway, we isolated a mutant cell line, sog9, which is unable to synthesize glycosaminoglycans (Banfield et al., 1995a). Although HSV-1 and H S V - 2 infection of sog9 cells is diminished, the cells are still infected at about 0.5% efficiency, which suggests that these cells normally express at least one nonglycosaminoglycan receptor. Sog9 cells were used to test whether glycosaminoglycan analogs, such as dextran sulfate (DS), could functionally substitute for cellular glycosaminoglycans to initiate H S V infection. High-molecular-weight DS added either prior to or during inoculation stimulated HSV-1 but not H S V - 2 infection by up to 35-fold. B y contrast, D S added after viral adsorption had no effect on infection efficiency. Moreover, DS stimulated HSV-1 infection at 4°C, indicating that this compound impinged on an early, energy-independent step in infection. Using radiolabeled virus, it was demonstrated that HSV-1 is more efficient than H S V - 2 in adsorbing to DS immobilized on microtiter wells. This raised the possibility that only HSV-1 could engage additional receptors to initiate infection in the presence of DS . To determine which viral component(s) facilitated D S stimulation, a panel of intertypic recombinants and deletion mutant viruses was investigated. These assays showed that DS stimulation of infection is mediated primarily by glycoprotein B (gB-1) and that this interaction was mediated by a domain other than the heparin-binding region in gB-1. Taken together, these results provide direct evidence that a principle role for cell surface glycosaminoglycans in H S V infection is to provide an efficient matrix for virus adsorption. To investigate further the interactions of H S V with cell surface glycosaminoglycans, a novel cell line, s o g 9 - E X T l , was used in this study. The ii expression of E X T 1, an enzyme in the heparan sulfate synthesis pathway, restores HS synthesis in sog9 cells (McCormick et al., 1998). Moreover, s o g 9 - E X T l cells are fully susceptible to H S V - 1 infection. Heparin and DS competition assays demonstrated that HSV-1 attachment to s o g 9 - E X T l cells is mediated by an interaction with heparan sulfate moieties on the cell surface. To determine which structural features of heparan sulfate were important for mediating attachment to sog9 -EXTl cells and control cell lines expressing H S , H S V - 1 infection in the presence of chemically modified heparin compounds was examined. It was found that 2-0-, 6-0- and N-sulfate groups of heparan sulfate are involved in viral attachment. Using H S V - 1 mutants deleted for the heparin-binding domains of gB and glycoprotein C (gC), it was shown that H S V - 1 attachment to glycosaminoglycans on sog9 -EXTl cells is primarily mediated by gC. The investigation of HSV-glycosaminoglycan interactions, and certainly of other aspects of the viral life cycle, is facilitated by the generation of H S V mutants. The construction of H S V recombinants has traditionally involved time-consuming purification procedures. To overcome this, attempts were made to develop i) an H S V - 2 cosmid set, which contains overlapping cosmids representing the entire H S V - 2 genome and ii) an H S V - 2 bacterial artificial chromosome ( B A C ) , which can be manipulated by bacterial genetics to generate the desired recombinant. This work demonstrates the potential advantages and disadvantages of using these different systems to construct H S V recombinants. i i i TABLE OF CONTENTS A B S T R A C T i i T A B L E OF C O N T E N T S iv L I S T O F T A B L E S ix LIST OF F I G U R E S x LIST OF A B B R E V I A T I O N S x i i i A C K N O W L E D G E M E N T S xvi i D E D I C A T I O N xv i i i C H A P T E R 1: L I T E R A T U R E R E V I E W 1 1.0 I N T R O D U C T I O N 1 1.1 C L A S S I F I C A T I O N 1 1.2 C L I N I C A L P A T H O L O G Y O F HSV-1 A N D H S V - 2 2 1.3 H S V S T R U C T U R E 3 1.3.1. The H S V Genome 3 1.3.2. The H S V Capsid 5 1.3.3. The H S V Tegument 5 1.3.4. The H S V Envelope 6 1.4 T H E H S V L I F E C Y C L E 6 1.4.1 The H S V Lytic Cycle 6 1.4.1.1. Vi ra l Replication 7 1.4.1.2. Vi ra l Egress 10 1.4.2 H S V Latency 12 1.5 T H E H S V E N T R Y P A T H W A Y 14 1.5.2 Vira l Binding 23 1.5.3 Stable Attachment of Virus to the Host Cel l Surface 28 iv 1.5.3.1. Hve A 30 1.5.3.2. HveB/Prr2 31 1.5.3.3. H v e C / P r r l 31 1.5.4. Vira l Penetration 35 1.5.5. Cell-to-Cell Spread and Cell-to-Cell Fusion 36 1.5.6 H S V Infection of Mouse Cel l lines Defective in G A G synthesis 39 1.6 G L Y C O S A M I N O G L Y C A N S 39 1.6.1 Glycosaminoglycan Structure 41 1.6.2. Chain Initiation 41 1.6.3. Biosynthesis of Heparin and HS 44 1.6.4. EXT I: A novel enzyme of the G A G synthesis pathway 47 1.6.5. Core Proteins 49 1.6.6 Dextran Sulfate: A Glycosaminoglycan Analog 50 1.7 V I R A L E N T R Y A N D T I S S U E T R O P I S M 52 1.7.1 H S V Serotype Differences in Glycosaminoglycan-Binding 53 1.7.2 Protein-Glycosaminoglycan Interactions 55 1.8 C O N S T R U C T I O N O F H S V M U T A N T S : T E C H N I Q U E S F O R M A N I P U L A T I N G T H E H S V G E N O M E 58 1.8.1 H S V Cosmids 59 1.8.2 H S V Bacterial Artificial Chromosomes 61 1.9 R A T I O N A L E O F S T U D Y 64 C H A P T E R 2: M A T E R I A L S A N D M E T H O D S 68 2.0 Materials 68 2.1 plasmids 69 2.2 Cel l Lines 72 2.3 Viruses A n d Vira l Stock Production 72 2.4 Virus Titer Determination 74 v 2.5 DS Stimulation Assays 74 2.6 Plaque Inhibition Assays 75 2.7 Vira l Penetration Assays 75 2.8 Preparation of radiolabelled virus 76 2.9 Binding of H S V to immobilized DS 76 2.10 Time Course Experiment 77 2.11 Preparation of Vira l D N A 77 2.12 Construction of H S V - 2 gB(tocZ) Vira l Mutant 78 2.13 Construction of H S V - 2 Vira l Cosmid Library 79 2.13.1 Preparation of viral D N A 79 2.13.2 Cosmid vector preparation 80 2.13.3 Ligation of H S V - 2 Viral D N A Fragments and P M S I Cosmid Vector 80 2.13.4 Packaging H S V - 2 Cosmids into Lambda Phage 81 2.13.5 Infection 81 2.13.6 Restriction mapping H S V - 2 Cosmid Clones 81 2.13.7 Cosmid Sequencing 82 2.14 Co-transfection of H S V cosmid clones 83 2.15 Construction of H S V 2 - B A C 84 2.15.1 Production of B A C Virus 84 2.15.2 Confirmation of p B A C - T K Sequences in H S V 2 - B A C 84 2.15.3 Plaque purification of ACV-resistant H S V 2 - B A C 85 2.15.4 Isolation of H S V - 2 recombinant as a B A C plasmid 85 C H A P T E R 3: D E X T R A N S U L F A T E C A N A C T A S A N A R T I F I C I A L R E C E P T O R T O M E D I A T E HSV-1 I N F E C T I O N 87 3.0 I N T R O D U C T I O N 87 3.1 R E S U L T S 88 3.1.1 Effect of sulfated polyanions on H S V infection of L and sog9 cells 88 vi 3.1.2 Characterization of the Interaction of Dextran Sulfate with sog9 Cells 90 3.1.3 Analysis of Vira l Attachment in the Presence of Dextran Sulfate 100 3.1.4 Effect of Dextran Sulfate on Vira l Penetration 105 3.1.5 Effect of Dextran Sulfate on H S V Infection of GAG-deficient sog8 Cel ls . . . . 107 3.2 D I S C U S S I O N 107 C H A P T E R 4: U S E O F N O V E L EXT 1-EXPRESSING C E L L L I N E S T O C H A R A C T E R I Z E H S V - G L Y C O S A M I N O G L Y C A N I N T E R A C T I O N S 112 4.0 I N T R O D U C T I O N 112 4.1 R E S U L T S 113 4.1.1 Effect of Sulfated Polyanions on HSV-1 Infection of Mouse EXT-1 Cells. . .113 4.1.2 Analysis of HSV-1 Infection of Mouse EXT1 Cells in the Presence of Modified Heparin Compounds 117 4.1.3 Relative Contributions of H S V gB and gC to Infection of Host Cells 124 4.2 D I S C U S S I O N 127 C H A P T E R 5: G L Y C O P R O T E I N B M E D I A T E S D E X T R A N S U L F A T E S T I M U L A T I O N O F HSV-1 I N F E C T I O N 132 5.0 I N T R O D U C T I O N 132 5.1 R E S U L T S 133 5.1.1 Mapping the DS Activation Site on the Virus 133 5.1.2 Contributions of gB-1 and gC-1 Heparin-Binding Domains to the Interaction of HSV-1 with Dextran Sulfate 139 5.1.3 Analysis of DS-mediated Infection Using HSV-1 gB Antigenic Variants 142 5.2 D I S C U S S I O N 144 C H A P T E R 6: S U M M A R Y 147 R E F E R E N C E S 152 A P P E N D I X I: C O S M T D A N D B A C T E C H N O L O G Y F O R G E N E R A T I N G H S V R E C O M B I N A N T S 177 vi i 7.0 I N T R O D U C T I O N 177 7.1 R E S U L T S 178 7.1.1 Construction of H S V - 2 gB2"(/acZ) virus 178 7.1.2 Cosmid Technology for Generating H S V Mutants: Construction and Characterization of an H S V - 2 Cosmid Library 180 7.1.3 Construction of H S V 2 - B A C 192 7.2 D I S C U S S I O N 195 vi i i LIST OF TABLES Table 1 Homology between glycoproteins encoded by alphaherpesviruses 18 Table 2 Relative infectivities of H S V virus on control and mutant cell lines 40 Table 3 Viruses used in this study 73 Table 4 Effect of dextran sulfate addition on HSV-1 infection 101 Table 5 Relative infectivity of H S V heparin-binding mutants on E X T 1 cell lines 126 Table 6 Effect of dextran sulfate on H S V infection 135 ix LIST OF FIGURES C H A P T E R 1 Figure 1.1 Electron micrograph of H S V particle 4 Figure 1.2 Schematic representation of the H S V reproductive cycle 8 Figure 1.3 Illustration of H S V latency 13 Figure 1.4 Functions of H S V glycoproteins 15 Figure 1.5 Organization of glycoprotein genes in the H S V - 1 genome 17 Figure 1.6 Structure of typical N-linked and O-linked oligosaccharides 20 Figure 1.7 Functional domains of HSV-1 glycoproteins gC-1, gB-1 and gD-1 25 Figure 1.8 Schematic representation of the molecular structures of HveB/Prr2, H v e C / P r r l , HIgR and P V R 32 Figure 1.9 Structure of heparin/heparan sulfate 42 Figure 1.10 Structure of chondroitin sulfate 43 Figure 1.11 Polymer modification reactions involved in the biosynthesis of heparin and heparan sulfate 45 Figure 1.12 Structure of the glycosaminoglycan analog dextran sulfate 51 Figure 1.13 Strategy for the generation of H S V recombinants using a cosmid-based system 60 Figure 1.14 The B A C system for construction of H S V recombinants 62 Figure 1.15 Rationale of study 66 C H A P T E R 2 Figure 2.1 Schematic diagram of the pHS208-lacZ plasmid, the P M S I cosmid vector, and the p B A C - t k plasmid 70 x C H A P T E R 3 Figure 3.1 Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9 Figure 3.10 Figure 3.11 Figure 3.12 C H A P T E R 4 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Effect of soluble heparin on HSV-1(F) plaque formation 89 Effect of soluble dextran sulfate on HSV-1 infection 91 Effect of soluble D S on herpes simplex virus infection 93 Effect of soluble DS on HSV-1(F) infection of sog9 cells 95 Effect of dextran sulfate on H S V - l ( F ) infection of sog9-EXT1...96 cells and L cells Effect of soluble D S on HSV-1(F) plaque formation 97 Size requirement for dextran sulfate 98 Effect of pretreating sog9 cells with DS on HSV-1(F) infection 99 Binding of HSV-1 and H S V - 2 to DS 103 Time course of HSV-1 infection of sog9 cells in the presence of D S 104 Rates of H S V - 1 penetration into sog9 cells in the presence and in the absence of DS 106 H S V - l A g C 2 - 3 infection of sog8 cells in the presence of DS 108 Effect of soluble native bovine heparin on HSV-1(F) infection of sog9, s o g 9 - E X T - l , L and L - E X T 1 cells 114 Effect of DS on HSV-1 infection of mouse E X T - 1 cell lines 115 Effects of O-desulfated heparin compounds on H S V - 1 infection of sog9, s o g 9 - E X T l , L and L - E X T 1 cells 118 Effect of N-desulfated.N-reacetylated heparin on HSV-1(F) infection of sog9, s o g 9 - E X T l , L and L - E X T 1 cells 121 Effects of N - / 0 - desulfated heparin compounds on H S V - 1 (F) plaque formation 122 xi CHAPTER 5 Figure 5.1 Effect of soluble DS on herpes simplex virus infection 136 Figure 5.2 Effect of DS on infection of sog9 cells by HSV-1 heparin-binding mutants 140 Figure 5.3 Effect of DS on infection of sog9 cells by HSV-1 gB mar mutants 143 C H A P T E R 6 Figure 6.1 Proposed model for dextran sulfate stimulation of HSV-1 infection of sog9cells 149 Appendix I Figure 7.1 Strategy for construction of a HSV-2 intertypic recombinant expressing gB-1 in place of gB-2 using a traditional plasmid-based strategy 179 Figure 7.2 Bg l II restriction profile of HSV-2 genomic D N A 182 Figure 7.3 Bg l II restriction analysis of HSV-2(G) cosmid clones 183 Figure 7.4 Bg l II restriction maps for all four isomeric forms of HSV-2(G) genomic D N A 184 Figure 7.5 Mapping the HSV-2 cosmids by sequence analysis 186 Figure 7.6 Impaired release of HSV-2 viral inserts from the cosmid backbone 190 Figure 7.7 Construction of HSV-2 B A C 193 Figure 7.8 Identification of B A C sequences in HSV-2 B A C 194 xi i LIST OF ABBREVIATIONS A adsorbance a.a amino acid A C V acyclovir B A C bacterial artificial chromosome B H V Bovine herpesvirus B M E beta-mercaptoethanol B S A bovine serum albumin C H O Chinese Hamster Ovary C M chloramphenicol C P E cytopathic effect cpm counts per minute CS chondroitin sulfate d H 2 0 distilled water D M E M Dulbecco's modified eagle media D N A deoxyribonucleic acid ddNTP dideoxyribonucleoside triphosphate dNTP deoxyribonucleoside triphosphate DS dextran sulfate D T T dithiothreitol E B V Epstein Barr virus E D T A ethylenediaminetetraacetic acid x i i i E R endoplasmic reticulum E t O H ethanol F B S fetal bovine serum Fuc fucose G A G glycosaminoglycan Gal galactose G a l N A c N-acetylgalactosamine Glc glucosamine G l c A glucuronic acid G l c N A c N-acetylglucosamine G 1 C N S O 3 N-sulfoglucosamine h hour(s) H B D heparin-binding domain H C M V human cytomegalovirus H I V human immunodeficiency virus H P L C high-pressure liquid chromatography HS heparan sulfate HSV-1 herpes simplex virus type 1 HSV-2 herpes simplex virus type 2 Hve herpesvirus entry mediator IdoA iduronic acid IgG immunoglobulin G I I inverted long segment xiv lis inverted long and short segments Is inverted short segment Kbp kilobase pairs kDa kilodaltons M 6 P mannose-6-phosphate M 6 P R mannose-6-phosphate receptor Man mannose min minute(s) M O I multiplicity of infection M W molecular weight N A N A N-acetylneuraminic acid NP-40 Nonidet P-40 P prototype position PBS phosphate buffered saline P F U plaque forming unit(s) P G proteoglycan Prr Poliovirus receptor-related protein PrV pseudorabies virus Pvr poliovirus receptor roe rate of entry mutation rpm revolutions per minute syn syncytial mutation T E T r i s - E D T A xv tk thymidine kinase T N F tumor necrosis factor V H S virion host shutoff protein vol volume V Z V Varicella Zoster virus U L unique long segment Us unique short segment xvi ACKNOWLEDGEMENTS There are many individuals whom I would like to thank for their help during the course of this work. I would first like to thank my supervisor Dr. Frank Tufaro, for giving me the opportunity to work in his laboratory and for providing a challenging, high caliber work environment -1 learned alot! Thanks Frank for all the conferences, financial assistance, computer lessons and innumerable lab get-togethers - great choice of wine!!! Additionally I would like to thank the members of my committee, Dr. Shirley Gil lam, Dr. Mike Gold and Dr. Rob McMaster, for their help and advice throughout this project and in the preparation of this thesis. I would also like to thank the Department of Microbiology and Immunology, especially the administrative staff who were always able to answer my questions. Thanks to Dr. Bernard Roizman at the University of Chicago for giving me the opportunity to work in his lab for a month, and to Chip V a n Sant for teaching me all there is to know and more about cosmids. I would also like to acknowledge the help of Dr. B . Horsburgh at Neurovir Inc., who provided reagents and guidance on the H S V B A C project. Thanks to N S E R C and U B C for providing funding over the past few years. I would like to thank all of the members of the Tufaro lab, both past and present -I could not have asked to work with a better group of people! Thanks especially to Lesley, Peter, Diane, Tina, Sonia, Craig, Sophie, Andrew and Yves for their friendship, humor, and help during my time in the lab. I w i l l always remember our all-you-can-eat sushi lunches and Peter's infamous feasts ( "What do you mean you can't eat anymore?")! A special thanks goes to Lesley Esford for her friendship, helpful scientific discussions, and for being a great travel buddy. To Ilo, Shamit, Zahara and Tina - sincere thanks for your warm friendship and for being there when I needed help the most. Words cannot fully express my gratitude to my family. M u m , Dad, Jacqueline and Nan - thank you for your constant love and unwavering belief in my ability. A n d to Ross.... your love, constant encouragement and sense of humor have kept me going with a laugh and smile. xvi i For Mum and Dad, J.J. and Nan, with love xvi i i Chapter 1: Literature Review CHAPTER 1: LITERATURE REVIEW "Nature is to be found in her entirety nowhere more than in her smallest creatures " Pliny, Roman writer 1.0 INTRODUCTION There are two herpes simplex virus (HSV) serotypes, H S V - 1 and H S V - 2 . It is now estimated that the worldwide prevalence of HSV-1 is 70-90% (Roizman, 1993). In the U.S. , the prevalence of H S V - 2 is approximately 20.8% (Aurelius, 1998). HSV-1 preferentially infects the oral mucosa and is associated with facial vesicular lesions, while H S V - 2 primarily infects the genital mucosa (Roizman and Sears, 1996). Why the viruses have predilections for different body sites remains unclear. Recent evidence, however, suggests that the H S V serotypes may enter host cells differently, which may account, in part, for the different behaviors of these viruses. This research examines the initial interactions of H S V - 1 and H S V - 2 with the host cell, providing further evidence that the association of virus with cell surface glycosaminoglycan moieties plays a pivotal role in H S V infection and is a possible determinant of H S V tissue tropism. 1.1 CLASSIFICATION HSV-1 and H S V - 2 belong to the family Herpesviridae, the sub-family alphaherpesvirinae and the genera simplexvirus. (Roizman and Sears, 1996). The alphaherpesviruses are characterized by their variable host range, short reproductive 1 Chapter 1: Literature Review cycles (e.g. 18 h for H S V ) and rapid development of cytopathic effect (CPE) in cell culture. They are also neurotropic and can establish latency in sensory ganglia. In addition to the two H S V serotypes, varicella zoster virus ( V Z V ) , pseudorabies virus (PrV) and bovine herpesvirus ( B H V ) also belong to the alphaherpesvirinae. The two other sub-families of the Herpesviridae are the betaherpesvirinae and the gammaherpesvirinae. The betaherpesviruses have a relatively long reproductive cycle and infected cells often become enlarged (cytomegalia) both in vivo and in vitro. Human cytomegalovirus ( H C M V ) is an example of a betaherpesvirus. Finally, the gammaherpesvirinae are lymphotropic viruses and specifically infect B and T lymphocytes. The prototype virus of this sub-family is Epstein-Barr virus ( E B V ) . 1.2 C L I N I C A L P A T H O L O G Y O F H S V - 1 A N D H S V - 2 H S V causes two types of infections in humans, lytic and latent. The virus infects and replicates at the site of entry (lytic infection) and is then transported to the sensory neurons that innervate it, becoming latent. The virus can reactivate from latency when exposed to the proper stimulus, becoming evident once more in the form of mucocutaneous lesions (Whitley, 1996). H S V infections range from minor nuisances to life-threatening disease. In the immunocompetent host, H S V infections can be asymptomatic, but painful recurrent mucocutaneous lesions are common. Typically, HSV-1 infects the mouth and lips, causing gingivostomatitis. H S V - 2 causes symptomatic lesions at genital sites and has also been recently described as causing hyperproliferative lesions (Aurelius, 1998). Severe clinical H S V problems, including keratoconjunctivitis, encephalitis and disseminated disease, are rare, but have high morbidity and mortality. Interestingly, H S V - 2 has also been recently identified as a D N A tumor virus (Aurelius, 1998). It encodes a protein kinase whose expression is required for neoplastic 2 Chapter 1: Literature Review transformation of human cells and tumour growth in animals. It is not yet known whether the virus is associated with malignancy in the human host. 1.3 H S V S T R U C T U R E The members o f the Herpesviridae family are morphologically similar. They consist of four components: (a) a core containing a linear double-stranded D N A genome, (b) an icosahedral capsid surrounding the core, (c) a tegument surrounding the capsid, and (d) an outer envelope containing glycoprotein spikes (Fig. 1.1). Here, the structure of H S V wi l l be summarized. 1.3.1. The H S V Genome Present understanding of H S V genome organization comes from studies of H S V - 1 . The genome exists as linear double-stranded D N A and is approximately 150kb in length (Roizman and Sears, 1996). The genome is composed of two unique regions, a 108 kb unique long sequence (U L ) and a 13 kb unique short sequence ( U s ) (Sarisky and Weber, 1994). The U L and Us components are flanked by repeat sequences a, b and c, which are 500 bp, 9 kb and 6.5 kb in length, respectively. The final structure of the genome thus appears as ab-U L -b ' a ' c ' -U s -ca . The a sequences can be present in one to many copies. A n interesting feature of the H S V genome is the ability of the U L and Us components to invert relative to one another. Thus, studies have shown that plaque-purified HSV-1 exists as an equimolar mixture of the four isomers (Delius and Clements, 1976; Hay ward et al, 1975). The isomers have been designated as P (prototype), I L (inverted U L sequence), I s (inverted U s sequence) and ISL (inversion of both U s and U L ) (Fig. 7.3, Appendix) (Roizman and Sears, 1996). The precise mechanism by which the H S V genome isomerizes remains unclear, although studies have indicated that homologous 3 Chapter 1: Literature Review Figure 1.1. Electron micrograph of H S V particle (X100,000). The H S V virion consists of a double stranded linear D N A genome contained within an icosahedral capsid (A). Surrounding the capsid is the amorphous tegument (B). The virion envelope (C) contains glycoprotein spikes which are indicated by the arrow. 4 Chapter 1: Literature Review recombination through any of the repeats of the HSV-1 genome (the a, b and c sequences) w i l l promote inversion of the U L and U s components (Sarisky and Weber, 1994; Martin and Weber, 1996). The function of the inversions in the viral life cycle is also enigmatic, since genomes restricted to one orientation, as a result of deletion of the internal repeats, can still yield infectious virus (Poffenberger et al., 1983). The HSV-1 genome contains three origins of D N A replication: oriL, which is situated in the U L segment of the genome, and two copies of oriS in the c sequences that flank the U s segment. HSV-1 is known to express at least 84 different proteins, 38 of which are essential for growth in cell culture (Aurelius, 1998). B y contrast, the ORFs that are essential for H S V - 2 replication have not been identified. Although the H S V - 2 genome is organized similar to that of H S V - 1 , the overall D N A sequence similarity between the two viruses is only 50% (Aurelius, 1998). 1.3.2. The H S V Caps id B y electron microscopy, the capsid of HSV-1 appears as an icosahedron containing 162 capsomers (Davison et ai, 1992). There are 150 hexameric capsomers and 12 pentameric capsomers. The capsid is approximately 100 nm in diameter and is assembled in the nuclei of virus-infected cells (Roizman, 1993). 1.3.3. The H S V Tegument The space between the capsid and the envelope is termed the tegument. The thickness of the tegument is variable and frequently distributed asymmetrically (Roizman, 1993). The tegument contains several important regulatory proteins, including the transactivator V P 16 and the virion host shut-off (VHS) protein, which together create an environment that promotes viral D N A replication (Ward and Roizman, 1994). 5 Chapter 1: Literature Review 1.3.4. The HSV Envelope The viral envelope is acquired as the D N A capsid buds from the inner nuclear membrane into the cytoplasm of the host cell (Roizman and Sears, 1996). The viral envelope is primarily composed of lipids and contains numerous glycoproteins, which appear as 8-24 nm long spikes when viewed by electron microscopy (Asher et al., 1969; Epstein et al., 1962; Stannard et al, 1987). In H S V - 1 , ten envelope glycoproteins have been identified to date: gB, gC, gD, gE, gG, gH, g i , gj, gL and g M . Several of these glycoproteins are important for mediating viral entry and egress, although only gB, gD, gH, gL are essential for virus replication (Little et al., 1981; Cai et ah, 1988; Desai et al., 1988; Ligas and Johnson, 1988; Herold et al, 1991; Fuller and Lee, 1992; Baines et al, 1993; Roop et al., 1993; Laquerre et al, 1998b). H S V synthesizes an additional glycoprotein, gK, although this glycoprotein does not appear to be incorporated into the envelope of infectious virus (Jayachandra et al, 1997). Synthesis of the H S V envelope glycoproteins is described later in Section 1.5.1. In addition to the glycoproteins, the HSV-1 envelope also contains several non-glycosylated membrane proteins, including U L 2 0 (Ward et al., 1994), U L 4 5 (Cockrell and Muggeridge, 1998) and U L 4 3 (MacLean et al, 1991). These proteins are not essential for viral growth in tissue culture, although U L 2 0 and U L 4 5 have been shown to function in viral egress and cell-cell fusion, respectively (Baines et al., 1991; Visa l l i and Brandt, 1993; Haanes etal., 1994). 1.4 THE HSV LIFE CYCLE 1.4.1 The HSV Lytic Cycle The H S V replication cycle is relatively short, lasting approximately 18 hours, and results in the production of infectious viral progeny and the destruction of the host cell (Ward and Roizman, 1994). In cell culture, virally infected cells become round in appearance and lose their adhesive properties. Other cytopathic effects include 6 Chapter 1: Literature Review degradation of cell nucleoli and chromatin. The lytic cycle of H S V can be divided into three stages: (a) viral entry, (b) viral replication, and (c) viral egress (Fig. 1.2). A s it is the subject of this thesis, viral entry wi l l be described later in detail. 1.4.1.1. Viral Replication Upon virus entry (see Section 1.5), the viral capsid is released into the cell cytoplasm and transported to the nuclear pores. The cellular cytoskeleton probably mediates the transport of HSV-1 capsids to nuclear pores, since drugs such as nocodazole, which disrupts microtubules, have been shown to inhibit neuritic transport of HSV-1 (Dales and Chardonnet, 1977; Kristensson et al., 1986). Once the virus reaches the nuclear pores, capsid uncoating occurs. The linear D N A is released into the cell nucleus and immediately circularizes (Roizman and Sears, 1996). The viral D N A is now ready to undergo replication. Two viral tegument proteins, V H S and V P 16, are released into the cell cytoplasm concomitant with viral entry. These proteins create an environment that is favourable for viral gene expression (Ward and Roizman, 1994). V H S mediates the degradation of host cell m R N A , and in this manner downregulates host protein synthesis (Kwong and Frenkel, 1989; Read and Frenkel, 1993). V P 16 acts in trans to initiate the transcription of viral genes (Batterson and Roizman, 1983; Campbell et al., 1984). H S V gene expression is coordinately regulated and occurs sequentially in a cascade fashion (Roizman and Sears, 1996). The three sets of genes expressed are the a , p* and y genes. The a genes are the first to be expressed, and their expression is maximal 2 to 4 h post-infection. The a genes primarily encode regulatory proteins, which in turn are required for the induction of the p" genes. The (3 group of polypeptides, whose rate of synthesis peaks at approximately 5 to 7 h post-infection, are involved in viral D N A synthesis. The (3 proteins also include the viral thymidine kinase (tk) and the viral D N A polymerase. The viral tk and D N A polymerase are targets of H S V antiviral agents, which 7 Chapter 1: Literature Review Figure 1.2. Schematic representation of the H S V reproductive cycle. (1) H S V attachment to cell surface receptors is followed by (2) fusion of the virion envelope with the cell membrane (penetration). (3) The nucleocapsid and two tegument proteins, V H S and V P 16, are released into the cell cytoplasm. (4) The nucleocapsid is transported to nuclear pores where the viral genome is released into the nucleus and circularizes. V P 16 initiates transcription of the a genes. Subsequently, the p* genes are transcribed, whose protein products are necessary for (5) D N A replication. Concomitant with D N A replication, is transcription of the y genes, which encode the capsid proteins and glycoproteins. (6) Concatamer length D N A is packaged into preformed capsids. (7) Nucleocapsids acquire an envelope by budding from glycoprotein-modified patches on the inner nuclear membrane. (8) In the model of egress shown here, enveloped viruses travel through the E R and Golgi apparatus, at which time the glycoproteins are modified. (9) Mature enveloped virions are transported to the cell membrane in vesicular structures and are then released (10) into the extracellular space. See text for details. 8 Chapter 1: Literature Review 9 Chapter 1: Literature Review include the guanosine analog acyclovir ( A C V ) . Unlike its cellular counterpart, H S V tk recognizes nucleoside analogs, such as A C V , as substrates and phosphorylates them. H S V D N A polymerase is less selective than cellular D N A polymerases and can incorporate the triphosphate form of A C V into viral D N A , at which point D N A synthesis is terminated. D N A synthesis begins following the production of the (3 proteins. It is detectable as early as 3 h and continues for up to 15 h post-infection. D N A synthesis is a requirement for the expression of some, but not all, of the y genes. Collectively, the y genes encode for structural components of the virion, including capsid proteins and envelope glycoproteins. Although some of the y genes (e.g. gC) are expressed late in infection, it should be noted that others, such as the genes specifying gB and gD, are expressed early in infection. Once synthesized in the rough endoplasmic reticulum (ER), the viral glycoproteins are transported to the inner nuclear membrane, Golgi membranes and the cell plasma membrane. Vi ra l D N A is synthesized by a rolling-circle mechanism in globular structures in the nucleus called replication compartments. This yields long head-to-tail concatamers (Jacob et al, 1979; Liptak et al, 1996; Boehmer and Lehman, 1997). Genome length D N A is then cleaved and packaged into preformed capsids. Associated with D N A replication, cleavage and packaging, is the isomerization of the D N A (Dutch et al, 1992). Mature capsids containing D N A appear in the nucleus within 6 h post-infection. 1.4.1.2. V i r a l Egress The egress of H S V from an infected host cell is believed to initiate when mature capsids attach to patches in the inner nuclear membrane that contain immature viral envelope glycoproteins. Capsids bud through these regions, acquiring an envelope (Darlington and Moss, 1968; Roizman and Sears, 1996). Subsequent routes taken by the particles as they travel to the extracellular space, and by which the envelope 10 Chapter 1: Literature Review glycoproteins are processed, remain unclear. One view is that the enveloped virions are transported in transport vesicles through the endoplasmic reticulum (ER) and Golg i complex, finally reaching the plasma membrane where they are released. The glycoproteins are believed to mature as the virus is transported along the secretory pathway (Campadelli-Fiume et al., 1991). Evidence for this model comes from a study by Johnson and Spear (1982) who demonstrated that monensin, an ionophore that disrupts the budding of vesicles from the Golgi , inhibits the transport of H S V particles to the cell surface. Treatment of H S V infected cells with monensin led to the accumulation of enveloped virus particles in what appeared to be Golgi-derived vacuoles. This supports the hypothesis that the viral glycoproteins are modified as the enveloped virus moves through the Golgi complex. A n alternative model for egress involves fusion of enveloped virus particles in the perinuclear space with the outer nuclear membrane, thereby releasing naked nucleocapsids into the cell cytoplasm (Jones and Grose, 1988; Whealy et al., 1991). The nucleocapsids are then re-enveloped in the trans Golgi . This would yield a virion envelope with fully matured glycoproteins. Evidence for this model comes from a study by Whealy et al. (1991), who demonstrated that treatment of PrV-infected cells with brefeldin A , an inhibitor that destroys the Golgi apparatus, resulted in the accumulation of envelope-free capsids in the cytoplasm. In controls, the capsids were associated with Golg i vesicles and by electron microscopy, appeared to acquire a membrane envelope from the trans Golgi . There is evidence that HSV-1 may use a similar egress pathway. Brefeldin A treatment of HSV-1 infected cells results in the accumulation of naked nucleocapsids in the cell cytoplasm (Cheung et al., 1991). Moreover, Browne and colleagues demonstrated that restriction of the expression of glycoprotein H to the ER-inner nuclear membrane by means of an ER-retention motif, resulted in the release of virus containing no detectable amounts of the glycoprotein (1996). These data suggest that the H S V - 1 virion acquires an envelope from a subcellular component other than the ER-inner nuclear membrane. Support for this model also comes from the general observation that naked nucleocapsids 11 Chapter 1: Literature Review are prevalent in HSV-infected cells and are found adjacent to membrane-bound vesicular structures (possibly Golgi-derived vesicles) (Roizman and Sears, 1996). Virus egress is dependent on Golgi-dependent maturation of viral glycoproteins. A s discussed later (Section 1.5.1), defects in glycoprotein processing block the transport of virus to the extracellular space. The process of virion translocation across the cytoplasm to the extracellular space is also regulated by various viral components, including the protein product encoded by the U L 2 0 gene. In several cell lines, the Golgi apparatus fragments as a result of infection, and in these cells, virions devoid of U i 2 0 get trapped in the perinuclear space (Baines et al, 1991; Campadelli-Fiume et al, 1993). U L 2 0 thus appears to facilitate viral exocytosis when the Golgi apparatus is not intact. In addition to U L 2 0 , g H (Browne et al, 1996; Desai et al, 1988), gD (Campadelli-Fiume et al, 1991), and g K (Hutchinson and Johnson, 1995; Jayachandra et al, 1997) are involved in the transport o f enveloped virus: virions accumulate within the cytoplasm when any one of these proteins is absent. 1.4.2 H S V Latency M u c h of the clinical disease arising from H S V infection is due to the ability of the virus to interact with neurons of the peripheral nervous system. This allows the virus to survive throughout the life of the infected individual in a latent state and evade the host immune response. Following primary infection and replication at peripheral sites, virus attaches to and enters sensory nerve terminals (Vahlne et al., 1978). H S V particles travel along axons to neuronal cell bodies within the sensory ganglia (McLennan and Darby, 1980). These are preferentially the trigeminal and dorsal root ganglia for HSV-1 and H S V - 2 , respectively (Efstathiou et al, 1986; Galloway et al, 1982). Latent virus is reactivated by hormonal imbalance, emotional stress or injury to tissues that are innervated by neurons which are latently infected with the virus. Figure 1.3 illustrates herpes simplex virus latent infection. H S V latency has recently been reviewed by Whitely et al. (1998) and Efstathiou et al. (1999). Despite extensive research, however, 12 Chapter 1: Literature Review Chapter 1: Literature Review the mechanisms by which the virus establishes and maintains a latent state or is reactivated, remain poorly understood. 1.5 THE HSV ENTRY PATHWAY In general, entry of enveloped viruses into cells requires the binding of virus to receptors on the cell surface, followed by either direct fusion of the virion envelope with the cell plasma membrane or endocytosis of the virion particle into the cell. Vi ra l entry by endocytosis requires a low p H to initiate membrane fusion, as weak bases that increase p H inhibit virus entry. Data from electron microscopy studies suggests that H S V can penetrate cells by either fusion at the plasma membrane and/or by endocytosis (Morgan et al., 1968; Dales and Silverberg, 1969). There is compelling evidence, however, that H S V entry occurs by a pH-independent fusion mechanism, rather than by endocytosis. Agents that inhibit endocytosis (e.g. chloroquine) do not block H S V entry (Fuller et al., 1989). Moreover, it appears that entry of H S V by endocytosis results in degradation of the virus and thus a non-productive infection (Campadelli-Fiume et al., 1988). Numerous lines of evidence indicate that H S V - 1 entry is comprised of a cascade of virus-cell interactions, involving several viral glycoproteins and at least two cell surface components. The process of HSV-1 entry into cells can be divided into three distinct phases: binding, stable attachment of the virion particle to the cell surface, and penetration. Before discussing details of the entry pathway, the synthesis of viral glycoproteins that mediate H S V entry into host cells wi l l be described in detail. 1.5.1 HSV Glycoproteins: Genetic Loci, Synthesis and Processing The H S V envelope glycoproteins are important determinants of H S V pathogenicity. The glycoproteins play critical roles in viral entry, egress, cell fusion and immune evasion. They also influence tissue tropism and host range. The functions of the glycoproteins in HSV-1 infection are illustrated in Figure 1.4. The organization of 14 Chapter 1: Literature Review Figure 1.4 Functions of HSV glycoproteins. 15 Chapter 1: Literature Review glycoprotein genes in the genome of HSV-1 is shown in Figure 1.5. Precise map locations of the glycoproteins have come from nucleotide sequence analyses. It is significant that several of the glycoproteins are conserved among members of the Alphaherpesvirinae (Table 1). Five families of glycoproteins, gB, gC, gD, gH and gL, participate in the entry into cells of H S V - 1 , H S V - 2 , P r V and B H V - 1 (Spear, 1993). gB and gH appear to be the most strongly conserved of the glycoprotein families with respect to amino acid sequence (Manservigi and Cassai, 1989). In fact all herpesviruses (alpha-, beta- and gamma-) encode members of gB and gH families. This likely reflects the important functions these glycoproteins have in viral infection. It is also interesting that glycoproteins not essential for viral replication are clustered in the U s region. O f all the H S V U s-encoded glycoproteins, only glycoprotein D is required for viral infection. However, in both P rV and B H V , gD has been shown to be dispensable for growth in cell culture, likewise to the other U s-encoded glycoproteins (Schmidt et al., 1997; Schroder et al., 1997). These observations support the hypothesis that this cluster of dispensable genes might have been aquired during the evolution of herpesviruses, perhaps having evolved from one dispensable precursor gene (McGeoch 1990; Schroder et al., 1997). The H S V glycoproteins have the characteristics of integral membrane proteins. Nucleotide sequence analyses of gB-1, gC-1 and gD-1, for example, have shown that (i) the N-terminal portions of these proteins are hydrophobic and act as signal sequences which are cleaved while (ii) the C-terminal regions possess a hydrophobic transmembrane anchoring region and a highly basic carboxy tail (Watson et ah, 1982; Frink et al, 1983; Bzick et al, 1984). In general, the biosynthesis of H S V glycoproteins is similar to that of eukaryotic cell glycosylated proteins (Campadelli-Fiume and Sefarini-Cessi, 1985). Non-glycosylated precursors of the viral glycoproteins are synthesized on polyribosomes bound to the rough endoplasmic reticulum. Concomitantly, the proteins are then transported into the lumen of the rough endoplasmic reticulum by the signal sequence. The unglycosylated protein precursors then undergo glycosylation using host machinery. Studies have shown that glycoproteins specified by HSV-1 contain both O-linked 16 Chapter 1: Literature Review u , s H H H r gL gM gH gB H t e : MTRc gC gK gG gJ gD gi gE Figure 1.5. Organization of glycoprotein genes in the HSV-1 genome. (TR) terminal repeat; (IR) internal repeat. 17 Chapter 1: Literature Review a o o G <D & x) CD T3 O O c o c '53 o 8* o, I—I I tool c <D ID CD >S w> o "3 S o 4) S H a C/3 c o 3 ^  '3 c ic u <D >> a ,o P <+H pj/ 1/1 a bo tu OH c/3 <D o <D G o oo > O v oo '3 1 0 e : >^ i ^ -f? • u % bo ^ j .S <u' 1 1 > 00 a bO ID •a CU (D O a o °a . 3 <D CJ G o is (D a ID CM •a a c/3 o ID > oo G o . 3 <D O O > p p > oo CD O CD 00 OH <D CJ CD<D O G o CD G CL, > oo 1/3 a bo CD a o, </3 CD O CD CJ P CD cj CD CJ C o c M G 3 G -o G ^ ! G 3 OH > 00 i a a, 1/5 CD O CD O IT) 1/3 o OH o z >> O0 CH H H H H O * < ° OH > l-l OH bO X ID O O * ,o Z 00 c S o o o o o o o > < Z Z Z Z Z Z Z bJO 3 £ bO bO K b0 H H O H - , bO bO bfi OH bo b0 bO bO bO G (D c/3 CD O G > oo 05 b0 'bo *W) lab ^b0 ' t i H H O bO bfl bfl 18 Chapter 1: Literature Review (Oloffson et al., 1981; Johnson and Spear, 1983) as well as N-linked oligosaccharides (Sefarini-Cessi and Campadelli-Fiume, 1981; Wenske et al., 1982). N-l inked glycosylation is initiated in the E R , when a preformed branched oligosaccharide [(glucose)3 -(mannose)9- (N-acetylglucosamine)2 ] linked to a dolichol phosphate l ipid carrier is transferred en bloc to asparagine residues on the nascent polypeptide. These asparagine residues are found within the tripeptide recognition sequences Asn-X-Ser or Asn-X-Thr (where X is any amino acid except proline) (Marshall, 1972). The asparagine-linked (N-linked) oligosaccharides are further modified during transit through the Golgi apparatus. This processing w i l l produce an oligosaccharide chain referred to as the high mannose oligosaccharide, which contains only mannose and N-acetylglucosamine (Fig. 1.6B). In some instances, N-linked glycans are not processed beyond this stage, such as the adenovirus E3 glycoprotein (Kornfeld and Wold, 1981). Complex N-linked glycans are produced when high mannose forms undergo further trimming in the Golgi . These oligosaccharides have the core structure [(mannose)3-(N-acetylglucosmamine)2 ] and several side chains containing N -acetylglucosamine, galactose and N-acetylneuraminic acid residues (Fig. l .6C). Fucose can also be added to the side chains. The mature forms of H S V glycoproteins possess complex glycans (Campadelli-Fiume and Sefarini-Cessi, 1985). HSV-1 gB is an exception to this, in that not all of its high mannose glycans undergo conversion. Thus, it carries both types of N-l inked glycans (Wenske et al., 1982; Johnson and Spear, 1983). B y contrast to N-glycosylation, O-glycosylation of H S V glycoproteins is a late post-translational event (Johnson and Spear, 1993). O-glycosylation initiates when N -acetylgalactosamine is linked to a serine or threonine residue in the polypeptide chain (Fig. 1.6A). For H S V glycoproteins, this appears to occur in the Golgi apparatus. The assembly of sugars added later (galactosamine, N-acetylgalactosamine, fucose and sialic acid) involves the sequential action of specific glycosyltransferases located in the Golgi complex. Significantly, O-glycosylation of viral proteins occurs less frequently than N -19 Chapter 1: Literature Review Figure 1.6. Structure of typical N-linked and O-linked oligosaccharides. (A) Structure of an O-linked oligosaccharide, here shown linked to a serine hydroxyl group. (B) Structure of a high-mannose N-linked oligosaccharide, linked to asparagine. (C) Structure of a complex N-linked oligosaccharide. Complex N-linked oligosaccharides are formed by modifications to the high mannose form in the Golgi complex. Asterisks indicate the five residues always found in N-linked oligosaccharides. ( N A N A ) , N-acetylneuraminic acid (sialic acid); (Gal), galactose; (GlcNAc) , N-acetylglucosamine; (GalNAc), N -acetylgalactosamine; (Man), mannose; (Fuc), fucose. 20 Chapter 1: Literature Review O-linked oligosaccharide O II -C [NANA] I [GJJO I [GaiNAcj o CH2 I NH C H -I NANA I o c NH-B Man Man Man I 1 1 |Man | |Man | [Man | I Man N-linked oligosaccharide (high mannose) pan~|*' I |GlcNAc |: iGlcNAcf p a n l * NH :0 O II -C NH-CH2 — CH asparagine o II c NH— | N A N A | | N A N A | |NANA | I 1 1 UD I 1 1 |GlcNAc| |GlcNAc| |GlcNAc| I |Man |* [Man f |Man p N-linked oligosaccharide (complex) |GlcNAc|* I |GlcNAc|*4-I NH [FuT O II C NH-O CH2 — CH C asparagine NH— 21 Chapter 1: Literature Review glycosylation (Olofsson et al., 1981; Johnson and Spear, 1983). It has been postulated that the conversion of high mannose forms to complex N-linked glycans, may interfere with accessibility of serine/threonine sites to the transferase (Sefarini-Cessi et al, 1983a). Structures of N and O-linked oligosaccharides vary immensely, and the glycan side chains of an individual glycoprotein can thus be quite heterogeneous. Johnson and Spear (1993) noted that O-linked oligosaccharides from gC and gD ranged in size from 215 daltons to as large as 2750 daltons. For both N - and O- glycosylation, the conformation of the protein may, in part, determine the extent of glycosylation and degree of heterogeneity (Roizman and Sears, 1996). The role of N-glycosylation with regard to H S V infectivity has been well characterized. Studies using the antibiotic tunicamycin, which inhibits the dolichol-dependent transfer of the mannose-rich backbone of the oligosaccharide chain to the protein, have demonstrated that N-glycosylation of viral glycoproteins is necessary for the production of infectious progeny (Pizer at el., 1980). A subsequent study demonstrated that in tunicamycin-treated cells, virion envelopment is defective (Peake et al., 1982). To test directly whether the conversion of high-mannose oligosacharrides to complex glycans is required for infectivity, Sefarini-Cessi and colleagues (1983) examined HSV-1 infection in cell lines defective in Golgi enzymes responsible for the addition of terminal sugars to N-linked glycans. In this cell line, therefore, the H S V glycoproteins were not fully processed. It was interesting then that this had no effect on the production of infectious progeny. The authors concluded that the presence of complex glycans is not a requirement for H S V infectivity and that high mannose forms are sufficient in this regard. The conversion of high mannose forms to complex N-linked glycans does, however, appear to facilitate virion egress and virus-induced cell-fusion (Sefarini-Cessi et al, 1983). Several other groups have also shown that viral egress is hindered when there are defects in the glycosylation pathway (Johnson and Spear, 1982; Peake et al, 1982). The role of O-glycosylation in H S V infectivity is not as well understood. A n interesting observation, however, was made by Tal-Singer and colleagues (1995) who 22 Chapter 1: Literature Review noted that in H S V - 1 gC, most of the O-linked saccharides were clustered in the N -terminus. In mucins, which are large mucous glycoproteins, clustering of O-linked glycans is believed to force the protein into an extended fibre-like structure (Jentoft, 1990). Indeed, gC-1 has a rod-like appearance when examined by electron microscopy, and it may be that this conformation facilitates the protein's interaction with cell surface glycosaminoglycans and thereby mediate the initial attachment of virus to cell surface glycosaminoglycans (Tal-Singer et al, 1995). 1.5.2 Viral Binding The current model for H S V entry predicts that there is an initial interaction of H S V particles with cell surface glycosaminoglycans (GAGs) which are abundant on most cell surfaces. Evidence for this interaction stems from observations that soluble heparan sulfate (HS) and heparin, which is structurally similar to H S , can reduce H S V infection by over 90% when present during inoculation (Gruenheid et al, 1993). Because soluble glycosaminoglycans do not appear to reduce the infectivity of the virion itself, it is likely that soluble G A G s act as competitive inhibitors to viral attachment. Removal of cell surface G A G s , by incubation of the host cell with heparin- and chondroitin-lyases, also reduces H S V infection, which suggests that HS and chondroitin sulfate (CS) have roles in H S V infection (WuDunn and Spear, 1989). Furthermore, animal cell mutants with defects in G A G synthesis show either complete (Shieh et al, 1992) or partial resistance to H S V infection (Gruenheid et al, 1993; Banfield et al, 1995a). H S V was the first virus shown to bind H S (WuDunn and Spear, 1989). Other herpesviruses have since been found to use HS as an initial receptor, including pseudorabies virus (Mettenleiter et al, 1990), cytomegalovirus (Neyts et al, 1992), bovine herpesvirus 1 (Okazaki et al, 1991) and human herpesvirus 7 (Secchiero et al, 1997). Recent work has also demonstrated that foot-and-mouth-disease virus (Jackson et al, 1996), respiratory syncytial virus (Krusat and Streckert, 1997), dengue virus (Chen et 23 Chapter 1: Literature Review al, 1997), adeno-associated virus type 2 (Summerford and Samulski, 1988) and human immunodeficiency virus type 1 (Patel et al., 1993), all use cell surface H S G A G s to attach to cells. Like H S V , many of the aforementioned viruses use additional cellular receptors besides H S . The functional role of H S in H S V infection appears to be to concentrate viral particles at the cell surface, thereby increasing the frequency of a subsequent interaction with its secondary receptor (Johnson and Ligas, 1988). Initial attachment of H S V virions to cell surface heparan sulfate and chondroitin sulfate G A G s is mediated by gC and/or gB. The gene encoding HSV-1 gC (gC-1) has been extensively studied. Sequence data predict a primary protein product of 511 amino acids with a predicted molecular weight of 55,000 (Swain et al., 1985). B y contrast, H S V - 2 gC-2 is 480 amino acids in length and has a predicted molecular weight of 51,600. Relative to the other H S V glycoproteins, gC-1 and gC-2 show the least degree of conservation, with limited areas of D N A and amino acid similarity. The functional arrangement, however, seems to be similar. Both glycoproteins have a hydrophobic signal sequence, a hydrophobic membrane-anchoring region, and a short acidic carboxy terminus. Both N -and O-linked oligosaccharides have been described for the two proteins (Wenske et ah, 1982; Olofsson etal, 1983; Swain etal, 1985; Johnson and Spear, 1983). The functional domains of gC-1 are depicted in Figure 1.7. The involvement of HSV-1 gC in virion attachment to host cells is supported by several lines of evidence. First, virions devoid of gC are impaired in their ability to bind to cells (Herold et al, 1991; Herold et al, 1994; Laquerre et al, 1998b). Second, neutralizing antibodies against gC-1 can block the binding of particles to host cells (Fuller and Spear, 1985). That gC-1 interacts with H S G A G s is demonstrated by the ability of soluble gC-1 to adsorb to heparin affinity columns (Herold et al, 1991). Moreover, heparitinase treatment of cells, which removes any cell surface H S G A G s , reduces the attachment of soluble gC-1 by 50% (Tal-Singer et al, 1995). Finally, removal of the heparin-binding domains of gC-1 significantly reduces the attachment efficiency of virus to cells (Trybala et al, 1994; Tal-Singer et al, 1995). A s discussed later (section 1.7.2), 24 Chapter 1: Literature Review gC-l HBD 143-150 HBD 33-123 _ 1 HBD 247 T COOH 511 a.a. gB-1 HBD 68-76 Neuroinvasive Determinant 523 Viral Penetration 241-441 Oligomerization 626-653 I I 1 _ syn 858 I COOH 903 a. gD-l Neuroinvasive Determinant 84 Penetration 1 231-244 HveA HveA/HveC Binding Binding 11-19 234-275 | COOH 369 a.a Figure 1.7 Functional domains of the HSV-1 glycoproteins gC-l, gB-1 and gD-l. The filled boxes represent signal sequences. Hatched boxes indicate transmembrane regions. The amino acids comprising the different glycoprotein domains are indicated. (HBD) Heparin-binding domain; (syn) syncytial; (Hve) Herpes virus entry mediator. 25 Chapter 1: Literature Review these heparin-binding regions have high positive charge densities, thereby promoting binding to negatively charged heparan sulfate moieties. It should be stressed, however, that the polycationic nature of heparan sulfate-binding domains (HBDs) is but one parameter that governs the interaction of proteins with G A G s . The involvement of g C - l in the initial adsorption of HSV-1 to host cells is made complicated by the observation that gC is not essential for viral infection (HeroId et al., 1991; Laquerre et al., 1998b). Virions devoid of g C - l are still infectious, although less efficient in attachment. This prompted Hero Id and co-workers (1994) to investigate whether there was gC-independent mode of binding. The ability of H S V - 1 gC-negative virus to bind to cells devoid of HS G A G s was found to be significantly reduced compared to wild-type virus, indicating that gC-independent binding of HSV-1 requires cell surface HS. Moreover, H S V - 1 particles devoid of both gB-1 and g C - l have a drastically reduced ability to bind to cells compared to gCl-deficient and wild-type virions. These data, along with the finding that gB has an affinity for heparin under physiological conditions, led to the conclusion that in the absence of g C - l , gB-1 mediates adsorption to H S G A G s (Herold etal, 1991). g C - l homologs in P rV and B H V - 1 also function in viral attachment to HS G A G s (Karger and Mettenleiter, 1993). It was originally thought that H S V - 2 gC would function to mediate H S V - 2 attachment to cellular HS moieties since soluble gC-2, like g C - l , binds to heparin affinity columns (Gerber et al, 1995). However, by contrast to H S V - 1 gC, H S V - 2 gC does not appear to play a predominant role in viral adsorption, because gC2-negative virions can bind to the cell surface as efficiently as wild-type H S V - 2 (Gerber et al, 1995). This raises the interesting possibility that gB-2, which is also a heparin-binding protein (Williams and Straus, 1997), may be the principle glycoprotein involved in mediating H S V - 2 viral attachment to G A G s . Glycoprotein B is required for productive infection of both H S V - 1 and H S V - 2 . gB-1 and gB-2 are highly conserved, having an overall nucleotide and amino acid sequence similarity of 86% (Stuve et al, 1987). For both proteins, the length of the entire coding 26 Chapter 1: Literature Review sequence is 904 amino acids, although the mature gB-2 is 7 amino acids longer than gB-1 due to its shorter signal peptide (the length of the signal peptides for gB-1 and gB-2 are 29 and 22 amino acids, respectively). These glycoproteins have 3 domains. For gB-1 and gB-2 respectively, they include an amino terminal hydrophilic extracellular domain of 696' and 700 amino acids, a transmembrane anchor region of 69 and 54 amino acids, and a charged carboxy terminus 109 and 106 amino acids, the latter two domains being highly conserved between the proteins (Pellet et al., 1985; Stuve et al., 1987). Marked divergence, however, is observed within the amino-terminal 85 amino acids of the protein (Stuve et al., 1987). HSV-1 gB has the consensus sequence for six N-glycosylation sites and contains 10 cysteine residues. These are conserved in gB-2. Figure 1.7 shows the functional domains of gB-1. The biologically active form of gB is a multimer (Claesson-Welsh et al., 1986; Cai et al., 1988; Highlander et al., 1991). It has been demonstrated that a region in the gB-1 ectodomain, comprising residues 626 to 653, is sufficient for oligomerization (Laquerre et al., 1996). gB monomers self-associate through a hydrophobic interaction. The formation of disulfide bridges involving cysteines on the monomer units is not required for dimerization, although this contributes to oligomer stability. Other viral glycoproteins, including those of human immunodeficiency virus (Earl et al., 1990) and Semliki Forest Virus (Kielian et al., 1990) form oligomers. Oligomerization of viral glycoproteins can impart new functions to the protein (Doms et al, 1993). Thus oligomeric forms can possess functions, such as receptor binding, that the monomer units would not. Recently, Laquerre and colleagues (1998b) identified the heparin-binding domain of gB-1 (see Section 1.7.2). While this domain is not required for HSV-1 infectivity, it is involved in viral adsorption. A mutant HSV-1 deficient in the heparin-binding domain of gB-1 was shown to bind to cells less efficiently than wild-type virus (20% reduction in binding). Significantly, this reduction in binding was not as great as that observed for a gC-deficient HSV-1 virus (65% reduced-binding). Having both mutations in a single virus (i.e. deletions in the H B D of gB-1 as well as the gC-coding sequence) was additive, and 27 Chapter 1: Literature Review reduced binding to cells by 80%. Thus, the relative contribution of gC-1 to HS-binding is greater than that of gB-1. It is also worthy to note that the H B D of gB-1 is located within the region most divergent between gB-1 and gB-2. Future studies w i l l need to determine whether this domain functions similarly in the two glycoproteins with respect to affinity and specificity for G A G s . 1.5.3 Stable Attachment of Virus to the Host Cell Surface Several lines of evidence support a model whereby H S V entry involves a second n o n - G A G receptor. For example, sog9 cells, despite being GAG-deficient, are still susceptible to H S V infection (Banfield et al., 1995a). Subramanian and colleagues (1994) isolated swine testis cells which were defective in HSV-1 entry, despite expressing a full complement of functional cell surface HS moieties. This indicated that these cells lacked a functional non-HS receptor required for efficient viral entry. Finally, Sears et al. (1991), have provided conclusive evidence that HSV-1 is able to bind at least two cell surface receptors. M D C K cells, a polarized cell line, express HS G A G s primarily on the basal cell surface. It was demonstrated that gC-deficient virions are only able to infect via the basal surface of M D C K cells whereas wild-type HSV-1 was able to infect at both the basal and apical surfaces. These data suggest that one receptor interacts with gC and is found on the apical cell surface, while a second receptor is found only at the basal cell surface. The accumulative work of several laboratories has shown that gD interacts with its own cellular receptor. Glycoprotein D is found in the envelope of H S V - 1 and H S V - 2 , and for both viruses, this protein is required for entry into mammalian cells (Highlander et al, 1987; Johnson and Ligas, 1988; Ligas and Johnson, 1988). The two proteins are structurally similar (Eisenberg et al, 1980; Lasky and Downbenko, 1984; Chiang et al, 1994) and appear to be functionally interchangeable, since gD-2 can be recombined into HSV-1 with no effect on virus infection (Muggeridge et al, 1992). The gene for HSV-1 gD encodes for a protein 394 amino acids in length which like gB and gC, has an N -28 Chapter 1: Literature Review terminal ectodomain and a hydrophobic transmembrane anchor sequence near the carboxy terminus (Fig. 1.7) (Matthews etal, 1983). Evidence that gD interacts with its own receptor came first from the observation that g D - l interferes with infection by HSV-1 (Spear, 1993). For example, cells which constitutively express g D - l are resistant to infection by HSV-1 (Campadelli-Fiume et al., 1988; Johnson and Spear, 1989). Presumably the gD produced by the cell sequesters the cellular receptors and prevents them from binding to gD in the virus. This resistance to infection is due to the failure of the virus to penetrate into gD-expressing cells, not the inability of the virus to bind to H S G A G s . gD-mediated interference is also observed with H S V - 2 , P r V and B H V - 1 (Spear, 1993). Moreover, it appears that in certain cell types, some of these viruses use a common gD receptor for entry, since cells expressing a member of the gD family can be resistant to infection by a heterologous virus (Spear, 1993 ). Johnson and colleagues (1990) have demonstrated that soluble forms of HSV-1 and H S V - 2 gD bind to a limited number of sites on the cell surface, approximately 4 x 10 5 to 5 x 10 5 per cell (Johnson et al, 1990). Binding of soluble g D - l can also be reduced by treating cells with proteases but is unaffected when cell surface HS is removed. Thus, the gD receptor is not an H S moiety. Brunetti et al. (1994) later proposed that H S V - 1 gD interacts with host cell mannose-6-phosphate receptors (MPRs) . M P R s are found on the surfaces o f cells, primarily in clathrin coated pits (Dahms et al, 1989). Soluble g D - l bound to the 275-kDa M P R and the 46-kDa M P R . Moreover, the authors observed that g D - l was modified by mannose-6-phosphate moieties. This group subsequently demonstrated that soluble forms of the M P R s could inhibit HSV-1 entry into monkey cells (Brunetti et al, 1995). However, mouse cells lacking the receptor are still susceptible to infection, suggesting that M P R s are not essential for entry in all cell types. 29 Chapter 1: Literature Review 1.5.3.1. Hve A Collectively, the above data provided strong evidence for at least one other receptor involved in H S V entry. To determine the identity of this receptor, Montgomery and colleagues (1996) transfected an expression library into C H O cells, which express G A G s but are resistant to entry of H S V - 1 , P rV and B H V - 1 (Shieh et al, 1992; Warner et al, 1998). C H O cells are partially susceptible to H S V - 2 infection. One clone, which encoded a member of the T N F receptor family, restored full susceptibility to H S V - 1 and H S V - 2 infection in the C H O cells and was shown to have herpesvirus entry activity. This cell surface protein has now been designated as herpesvirus entry protein A (HveA). The interaction of H S V - 1 and H S V - 2 with Hve A was later shown to be mediated by gD (Whitbeck et al, 1997; Nicola et al, 1998). Characteristic of the T N F receptor family, Hve A , also known as T R 2 (Kwon et al., 1997) and A T A R (another TRAF-associated receptor) (Hsu et al., 1997), is a type I membrane glycoprotein with cysteine-rich repeats in the ectodomain (Naismith and Sprang, 1998). Lymphotoxin a and L I G H T (homologous to /ymphotoxins, exhibits inducible expression, and competes with g D for Z/veA, a receptor expressed by T lymphocytes) are members of the T N F family and are ligands for Hve A (Mauri et al, 1998). The TNF-related cytokine-receptor systems feed into signalling pathways involved in either cell death or cell survival. This raises the possibility that H S V gD-1 may be able to modify HveA-signalling activities during viral entry which may, in turn, aid in virus survival and persistance. Although H v e A is expressed in many tissue types, including lung and liver, the receptor is expressed most abundantly in lymphoid organs and cells (Marsters et al, 1997). This corroborates the findings of Montgomery and colleagues (1996), who observed that anti-Hve A antibody prevented H S V - 1 infection of human T lymphocytes but not of other human cell types. This demonstrated that (i) H v e A is the principle co-receptor into human lymphoid cells and (ii) other receptors must exist for H S V entry, since H S V can also infect cells of neuronal and epithelial origin. 30 Chapter 1: Literature Review 1.5.3.2. HveB/Prr2 A subsequent study by Warner and colleagues (1998) demonstrated that poliovirus receptor-related protein 2 (Prr2) (Eberle et al., 1995) mediated entry of wi ld-type H S V - 2 strains, particular viable HSV-1 gD mutants and P r V into C H O cells. Prr2, designated as herpesvirus entry protein B (HveB), failed to support entry of wild-type H S V - 1 strains or B H V - 1 . The structure of Prr2/HveB is illustrated in Figure 1.8. Prr2/HveB is expressed in keratinocytes and neuronal cells. Prr2 is a member of the poliovirus receptor subfamily of the immunoglobulin (Ig) superfamily. The poliovirus receptor family also includes the poliovirus receptor (Pvr) and poliovirus receptor-related protein 1 (Prrl) (Fig. 1.8). B y contrast to Pvr, P r r l and Prr2 do not function in poliovirus entry and are only related to Pvr by amino acid sequence (Eberle et al., 1995). Recently, P r r l and Prr2 were renamed as nectin-1 and nectin-2a, respectively (Takahashi et al., 1999). The name nectin comes from the Latin "necto" meaning "to connect". Both nectin-1 and nectin-2a have been shown to function as cell-cell adhesion molecules (Aoki et al. 1997; Takahashi et al, 1999). 1.5.3.3. HveC/Prrl The ability of wild-type H S V - 2 , but not H S V - 1 , to use HveB as a receptor could account, in part, for serotype differences in tissue tropism and pathogenicity. Neither H v e A nor HveB, however, function as a co-receptor of both serotypes into epithelial cells, the mucosal epithelia being the site of initial replication. Moreover, neither Hve A nor Hve B can serve as a co-receptor for the other alphaherpesviruses, P r V and B H V - 1 . For these reasons, Geraghty and colleagues (1998) set out to isolate a common coreceptor for H S V - 1 , H S V - 2 , P r V and B H V - 1 . In their study, they identified poliovirus receptor-related protein l (P r r l ) protein (Lopez etal, 1995), designated as herpesvirus entry protein C (HveC), as being able to mediate entry of all these viruses into resistant C H O cells (Fig. 1.8). Additionally, they demonstrated that Pvr itself mediates B H V - 1 and PrV infection, but not H S V infection. Cocchi et al. (1998b) later identified an isoform of 31 Chapter 1: Literature Review Figure 1.8 . Schematic representation of the molecular structures of HveB/Prr2, H v e C / P r r l , HIgR, and P V R . These receptors are members of the l g superfamily. Shown here are the V and C domains formed by cysteine bonds and the number of amino acid residues in each region. From Warner et a/.(1996) and Cocchi et al. (1998). 32 Chapter 1: Literature Review HveB/Prr2 NH, Trans-membrane region 28 «- 122 COOH HveC/Prrl NH, 28 «- 139 COOH HIgR NH, COOH PVR NH, 33 Chapter 1: Literature Review HveC, designated the herpes Ig-like receptor (HIgR), as being able to mediate H S V - 1 , H S V - 2 and B H V - 1 infection (Fig. 1.8). HIgR appears to be a splice variant of HveC; the two share an identical ectodomain, comprising one variable (V) and two constant (C2) domains. The proteins differ in their cytoplasmic and transmembrane domains. Significantly, HveC and HIgR are expressed in human epithelial and neuronal cells and as such are strongly favoured as the receptors that allow both HSV-1 and H S V - 2 to infect epithelial mucosa and spread to the nervous system (Geraghty et al., 1998; Cocchi et al. 1998b). Likewise to HveA, HveC and HIgR interact with HSV-1 gD. Moreover, Cocchi and colleagues (1998a) demonstrated that the V domain of HIgR interacts with gD. The gD-HveA and gD-Hve C interactions have been studied in detail (Knimmenacher et al., 1998). Soluble truncated forms of Hve A and Hve C directly bind to H S V gD both in solution and at the surface of virions. Competition assays with neutralizing anti-gD monoclonal antibodies showed that the region between amino acids 234 and 275 in g D - l may contain a common domain which interacts with both H v e A and HveC. However, the gD-HveA and gD-HveC interactions also had a number of significant differences. Disruption of the N-terminal domain of gD by linker insertion at amino acid 34 significantly reduced binding of gD to Hve A but not Hve C. Moreover, competition assays with anti-gD monoclonal antibodies directed against N-terminal amino acids 11 to 19 blocked the interaction between H v e A and HSV-1 (KOS) but did not block the interaction of Hve C with virus. Collectively, these data indicate that the N-terminal region of gD is important for interactions with Hve A but not HveC (Krummenacher et al, 1998). The Ig superfamily members Pvr, Prr-1, Prr-2 and HIgR share a similar molecular structure defined by the six conserved cysteine residues (Fig. 1.8). Alphaherpesviruses may have evolved to recognize these highly conserved domains in the Ig superfamily. Clearly, however, different HSV-1 and H S V - 2 strains vary in their ability to utilize HveA, HveB, Hve C and HIgR for infection. This, and the fact that these receptors are 34 Chapter 1: Literature Review expressed in distinct cell types, w i l l , in part, determine the pattern of viral spread in an infected individual. 1.5.4. Viral Penetration The H S V glycoproteins gB, gD, gH and gL are required for viral penetration and are essential for productive H S V infection (Cai et al., 1988; Ligas and Johnson, 1988; Forrester et al., 1992; Hutchinson et al., 1992). Homologs of gB, gD, gH and gL in P rV and B H V - 1 have also been shown to be involved in penetration (Spear, 1993; Karger and Mettenleiter, 1993; van Drunen Littel-van den Hurk et al, 1996; Klupp et al, 1988). Unlike H S V gD, however, B H V and PrV gD homologs are not essential for this process (Schmidt et al, 1997; Schroder et al, 1997). H S V mutants deleted for either gB, gD and/or gH-gL are able to attach to cells but unable to penetrate. Polyethylene glycol, a membrane fusogen, can enable adsorbed mutant viruses to penetrate and initiate infection. Furthermore, anti-gB, gD and g H neutralizing antibodies permit attachment to cells but prevent penetration (Fuller and Spear, 1987; Fuller et al, 1989; Navarro et al, 1992). The activity of gH is dependent on gL (Hutchinson et al, 1992). g H and gL form a hetero-oligomer which is incorporated into virions. Co-expression of gL and g H is required for normal post-translational modifications, folding and intracellular transport of both glycoproteins (Gompels and Minson, 1989; Roop et al, 1993). Thus, Roop et al. (1993) observed that an HSV-1 mutant deleted in gL was unable to incorporate g H into the virion envelop and enter into cells, although it could adsorb efficiently to the cell surface. Domains critical for viral penetration have been mapped for HSV-1 gB. Anti-gB monoclonal antibodies that inhibit viral penetration map to amino acid residues 241 to 441 in the gB molecule, a region that is centrally located within the ectodomain (Fig. 1.7) (Highlander et al, 1988). With respect to gD-1, earlier studies identified a functional region comprising residues 231 to 244, which when altered significantly disrupted gD function (Feenstra et al, 1990; Muggeridge et al, 1990). It is interesting that this region overlaps with the region identified as interacting with H v e A and HveC. It w i l l be 35 Chapter 1: Literature Review important to establish how the interaction of gD with its cellular receptor facilitates the fusion of the virion envelope with the cell plasma membrane. Despite these findings, the molecular mechanisms underlying H S V penetration remain unclear. It is evident, however, that gB-1, g D - l , gH-1, gL-1 and also g C - l can associate with one another in the virion envelope (Handler et al., 1996a). Chemical cross-linking studies have shown that these glycoproteins can form homodimers and hetero-oligomers of gD linked to gB, gC linked to gB and gD, and gH-gL linked to gD and gC. The gH-gL complex does not appear to associate with gB. These findings were expanded by another study which examined cross-linking patterns of these glycoproteins during entry (Handler et al., 1996b). Changes in cross-linking profiles were not detected during viral attachment. However, complexes of gB and gD which were present during attachment disappeared as penetration proceeded. It is possible that during penetration, the glycoproteins undergo a conformational change or that the virion envelope undergoes a physical change (such as during the formation of a fusion pore). Together, these findings support the concept that the H S V viral glycoproteins form a complex during entry. This is not necessarily surprising, since all known virus fusion proteins are oligomers and many cellular fusion events are dependent on the formation of multi-subunit complexes (White, 1990). 1.5.5. Cell-to-Cell Spread and Cell-to-Cell Fusion Two processes related to, but distinct from viral entry, are cell-to-cell spread of virus and the fusion of cells infected with virus. During wild-type H S V infection in tissue culture, it is typical for cells to become rounded as single cells, with very little fusion of infected cells (Hutchinson et al., 1992b). Plaque formation results from virus spreading from an infected cell directly into adjacent uninfected cells. Cell-to-cell spread requires the glycoproteins gB, gD and gH-gL. If viruses deficient in these glycoproteins are grown on complementing cell lines (to provide the missing glycoprotein) the mutant viruses can 36 Chapter 1: Literature Review enter cells but cannot spread to adjacent cells and form plaques (Cai et al., 1988; Forrester et ai, 1992; Ligas and Johnson, 1988; Roop et al., 1993). Glycoproteins gE and g i , which form a complex, are also involved in cell-to-cell spread. These glycoproteins are not required for entry or for the production of infectious progeny. However, gE- and g l -negative H S V mutants yield significantly smaller plaques in vitro, and when injected into mice, spread poorly beyond the initial site of infection (Dingwell et al., 1994). Moreover, H S V particles deleted for either gE or g l fail to spread efficiently within the nervous system (Dingwell et al., 1995). A recent study by Dingwell and Johnson (1998) demonstrated that the gE-gl complex co-localizes with cell-junction components, supporting a model whereby gE-gl mediate transfer of virus to adjacent cells across cell junctions. This mode of transmission is particularly important to H S V pathogenesis, since the virus is therefore able to avoid neutralization by anti-HSV antibodies. Entry of H S V into cells involves fusion of the virion envelope with the cell plasma membrane. In a similar process, newly synthesized proteins expressed on the plasma membrane of an infected cell induce fusion with neighboring uninfected cells. A s mentioned previously, wild-type H S V infection causes limited amounts of cell fusion, although polykaryocytes have been observed in lesions of some infected individuals. This phenomenon is best observed when cells are infected with H S V syncytial mutants which yield large, multi-nucleated cells (Hutchinson et al., 1992). A syncytial (syri) mutation has been mapped to the cytoplasmic tail of gB; an Arg-to-His alteration at residue 858 accounts for the syncytial phenotype (DeLuca et al., 1982). H S V - 1 strains A N G and ANG-path possess this syn mutation in gB, along a second mutation at residue 553, termed a fast rate of entry (roe) determinant in gB (Saharkhiz-Langroodi and Holland, 1997). Together, these mutations allow for a unique type of cell fusion in tissue culture called fusion-fr om-without (FFWO). F F W O is the rapid induction of cell fusion that is observed at a high m.o.i. and occurs without viral protein synthesis. Additionally, gD and gH are involved in cell fusion, as antibodies directed against these glycoproteins w i l l inhibit syncytia formation (Gompels and Minson, 1986; 37 Chapter 1: Literature Review Highlander et al, 1987). Recently, Turner and colleagues (1998) demonstrated that transient expression of gB, gD, gH-gL of HSV-1 in cells was necessary and sufficient to induce cell fusion. Omitting any one of the glycoproteins reduced polykaryocyte formation. Finally, many syn mutations have been mapped to H S V - 1 g K (Dolter et al., 1994). g K has been detected in the nuclear and perinuclear space, but not in the viral envelope or in the plasma membrane (Hutchinson et ah, 1992a; Hutchinson et al., 1995b). Because it is not present in the plasma membrane, it is unlikely that g K could directly influence cell fusion. Recently, however, HSV-1 g K was shown to be required for efficient viral envelopment and egress of infectious virions from the cytoplasm to the extracellular space (Hutchinson and Johnson, 1995a; Jayachandra et al, 1997). Further studies are necessary to determine whether altered virion cell surface transport could influence cell-cell fusion. Cellular factors have also been shown to influence viral-induced cell fusion. Shieh and Spear (1994) demonstrated that herpesvirus-induced cell fusion was dependent on cell surface heparan sulfate. Infectious D N A from the HSV-1 syncytial mutant HSV-1 (MP) was transfected into heparan sulfate-deficient mutant C H O cells. The ability of H S V - 1 (MP) D N A to induce syncytia on the mutant cell line was reduced by 10-fold when compared to syncytia formation in cells expressing heparan sulfate. Soluble heparin was also shown to be able to substitute for cell surface H S in promoting fusion of transfected C H O cells. What is the role of H S in viral-induced cell fusion? The interaction of G A G with a GAG-binding protein can significantly alter the activity of the protein (Jackson et al. 1991). The role of HS/heparin in viral-induced cell fusion could, therefore, be to interact with a heparin-binding protein and thereby trigger fusogenic activity (Shieh and Spear, 1994). Given its role in viral penetration and cell-cell fusion, the heparin-binding glycoprotein gB is a possible candidate for H S activation of fusion. More recently, Terry-Allison and colleagues (1998) demonstrated that H v e A also mediates cell-cell fusion. Cells resistant to HSV-1-induced cell fusion became susceptible when they expressed HveA. 38 Chapter 1: Literature Review 1.5.6 HSV Infection of Mouse Cell lines Defective in GAG synthesis. The approach our lab has taken to investigate the H S V entry pathway has involved the isolation of mouse L cell mutants defective in G A G synthesis. Gruenheid and colleagues (1993) infected mouse cells with HSV-1 and subsequently harvested any cells that survived to form colonies. This led to the isolation of the mutant cell line gro2C, which synthesizes chondroitin sulfate but not heparan sulfate. Gro2C cells are 90% resistant to H S V - 1 infection relative to parental control L cells. The observation that gro2C cells were susceptible to HSV-1 infection, despite being deficient in H S G A G s , suggested that HSV-1 could interact with cell surface CS to mediate infection (Banfield et al, 1995b). This was supported by the observation that soluble CS inhibited H S V - 1 infection on gro2C cells. The sog9 cell line was derived from gro2C cells (Banfield et ah, 1995b) and exhibits a three-order of magnitude decrease in susceptibility to HSV-1 infection compared with control L cells. This observation can be explained by the fact that the sog9 cell line has additional defects in the G A G synthesis pathway and fails to synthesize any G A G s . Importantly, HSV-1 infection of sog9 cells is not significantly reduced by soluble H S , indicating that infection is G A G independent. Thus glycosaminoglycans do not appear to be essential for H S V infectivity. However, in the absence of G A G s , the ability of the virus to bind to the cell surface is significantly reduced. Table 2 shows the relative infectivities of HSV-1 and H S V - 2 on the L , gro2C and sog9 cell lines. 1.6 GLYCOSAMINOGLYCANS A glycosaminoglycan is a linear heteropolysaccharide consisting of specific repeating disaccharide units (Jackson et al, 1991). Usually one monosaccharide of the disaccharide repeat is a hexuronic acid, either glucuronic acid or iduronic acid, and the 39 Chapter 1: Literature Review Table 2. Relative infectivities of H S V virus on control and mutant cell lines Virus Relative Infectivity (%) L(control) gro2C sog9 H S V - l ( K O S ) 100 10 0.3 H S V - l ( F ) 100 20 0.8 HSV-2(G) 100 3.3 1 (From Banfield et al., 1995). 40 Chapter 1: Literature Review other sugar is a hexosamine, either N-acetylglucosamine or N-acetylgalactosamine. One or both of the sugars are variably N - and O-sulfated, which contributes to the diversity of these structures and their high negative charge density. The most common G A G s include heparin, heparan sulfate (HS), dermatan sulfate (DS) and chondroitin sulfate (CS). These can be linked through an O-glycosidic linkage to a serine residue in a protein core. A G A G covalently linked to a protein in this manner is termed a proteoglycan. Proteoglycans and their attached G A G s have a variety of roles in cell-cell interactions and serve as activators of growth and anti-coagulation factors. 1.6.1 Glycosaminoglycan Structure Heparan sulfate, which is produced in nearly all cell types studied, and heparin, which is synthesized by connective tissue-type mast cells, are both synthesized as proteoglycans (Lindero/ . , 1993). Biosynthesis of HS and heparin proteoglycans involves the initial formation of a G A G structure composed of alternating D-glucuronic (GlcA) and N-acetylglucosamine (GlcNAc) units (Fig. 1.9). A s discussed later, this initial structure is modified through a series of reactions, which leads to the formation of polymers with extensive structural diversity. B y comparison, the modifications of CS are much less extensive, resulting in a relatively homogeneous molecule. Chondroitin sulfate is synthesized from an initial polymerization product consisting of alternating G l c A and N -galactosamine residues (GalNAc) (Fig. 1.10). Ester-linked sulfate groups are added typically at the C-4 and/or C-6 positions on G a l N A c residues. In some instances, the G l c A residue epimerizes to iduronic acid (IdoA). In this case, the polysaccharide is called dermatan sulfate or chondroitin sulfate B . 1.6.2. Chain Initiation Synthesis of CS and H S begins after translation of the core protein and transfer of xylose to specific serine residues in the core protein (Kjellen and Lindahl, 1991). 41 Chapter 1: Literature Review A . CH 2 OH HNAc OH n B. CH 2 OS0 3 COO CH,OH HNAc OH NHSO, OS0 3 Figure 1.9 Structure of heparin/heparan sulfate. (A) The initial polymerization product consists of a repeating disaccharide unit containing D-glucuronic acid (GlcA) and N -acetylglucosamine (GlcNAc) . (B) Structure of heparin/HS following N-deacetylation/N-sulfation, epimerization of G l c A to IdoA, and O-sulfation. 42 Chapter 1: Literature Review CH 2 OH COO o y o a K OH HNAc O H n B HNAc c CH 2OH CH 2OH 0,SO i o,so OH OH HNAc HNAc Figure 1.10 Structure of chondroitin sulfate. (A) The initial polymerization product consists of a repeating disaccharide unit of N-galactosamine (GalNAc) and D-glucuronic acid. (B) Structure of chondroitin-4-sulfate (chondroitin sulfate A). (C) Structure of dermatan sulfate (chondroitin sulfate B) . 43 Chapter 1: Literature Review Xylosylation is catalyzed by xylosyltransferase using UDP-xylose as the sugar donor and there is some evidence that this begins in the E R and continues in the Golgi (Silbert and Sugumaran, 1995). In several proteins, xylosylation occurs on S - G - X - G tetrapeptide sequences (where X is variable). The existence of a consensus sequence for G A G attachment is, however, controversial. The formation of polysaccharide chains continues in the cis and medial Golgi by the addition of galactose (Gal) and glucuronic acid (GlcA) residues. This results in the formation of what is termed the tetrasaccharide linkage region (-glucuronic acid-galactose-galactose-xylose-Ser), which is common to CS , heparin and heparan sulfate proteoglycans. The pathway diverges after the formation of the linkage tetrasaccharide. A t this point, the addition of G a l N A c or G l c N A c commits the structure to be synthesized either as CS or HS/heparin, respectively. Kitagawa and colleagues (1999) recently demonstrated that the gene EXTL2 encodes an enzyme which transfers G a l N A c and G l c N A c to the common GAG-protein linkage region. Thus, E X T L 2 is likely the critical enzyme that determines whether the polysaccharide chain is to be heparan sulfate or chondroitin sulfate. 1.6.3. Biosynthesis of Heparin and HS H S G A G s have considerable structural diversity. A s is relevant to this thesis, it is important to consider the means by which the structural heterogeneity of H S arises. Considered here is the synthesis of HS/heparin, which is illustrated in Figure 1.11. Polymerization of these G A G s begins after the formation of the tetrasaccharide linkage (Silbert and Sugumaran, 1995). In the case of heparan sulfate, polymerization is catalyzed by heparan sulfate polymerases, including those encoded by EXT1 and EXT2 (Lind et al, 1998). The identification of E X T 1 as a component of the G A G synthesis pathway is described later in Section 1.6.4. The G A G chain is formed by the transfer of alternating G l c A and G l c N A c monosaccharides from UDP-sugar nucleotides. Coupled with polymerization is the first modification of the [ G l c A - G l c N A c ] n polymer, N -deacetylation, with subsequent N-sulfation of the G l c N A c units (Salmivirta et al., 1995). 44 Chapter 1: Literature Review Figure 1.11. Polymer modification reactions involved in the biosynthesis of heparin and heparan sulfate. N-deactylation/N-sulfation is the first modification. Regions that remain N-acetylated w i l l not be further modified. N-sulfated regions can be further modified by C-5 epimerization of Glc A residues to Ido A , which is often concomitant with 2-0-sulfation. The final modification shown here is 6-0-sulfation of G l c N S 0 3 residues. 45 Chapter 1: Literature Review C H 2 O H COO" O S O UDP — • / \ — UDP H N A c Polymerization C H 2 O H C O O O \ / O O H N A c | N-Deacetylation/ C H 2 O H • COO" N-Sulfation O N X O O H N S O 3 C-5 Epimerization C H 2 O H I + 2-0- Sulfation O N XI O o <coo H N S O ; OS03-6-O-SuIfation HNSO3- OSO3-46 Chapter 1: Literature Review This is followed by C-5 epimerization of G l c A units to IdoA, which is coupled with O-sulfation at C2. Finally, G l c N S 0 3 units are modified by O-sulfation at C-6. The stepwise nature of this process should be emphasized, as the product of any given reaction wi l l be the substrate for the next series of modifications (Salmivirta et al., 1995). This is significant since the enzymes catalyzing these reactions have substrate specificity. Therefore, the polymer product of one reaction may only contain a limited number of monosaccharide units that satisfy the substrate specificity for the next enzyme in the pathway. Throughout the modification process, therefore, the heterogeneity and complexity of the polysaccharide w i l l increase. For example, the initial N-deacetylation/ N-sulfation reaction w i l l determine which G l c A units w i l l epimerize to IdoA. N-sulfate groups are required for substrate recognition by enzymes catalyzing the C-5 epimerization and O-sulfation reactions. Regions that remain N-acetylated, therefore, are not further modified and w i l l not contain IdoA and O-sulfate residues. A t least 50% of the H S polymer consists of such unmodified domains, which alternate with heterogeneous N-sulfate domains. Heparin, by contrast, consists of long stretches of highly modified domains occasionally interspersed by short unmodified regions. The reasons why particular regions of a polysaccharide remain N-acetylated or why other regions are N -deacetylated/N-sulfated remain elusive. It w i l l be important to understand how polymer modification is regulated, since it is by this process that protein-binding domains in the HS chain are initially formed. 1.6.4. EXT1: A novel enzyme of the G A G synthesis pathway There are obviously numerous enzymes involved in proteoglycan biosynthesis, not all of which have been purified and cloned. A review of the accumulating information with regards to these enzymes is not within the scope of this thesis. Nonetheless, it relevant to the work conducted in this study to consider one of the enzymes involved in the G A G synthesis pathway, namely E X T - 1 . Our laboratory isolated E X T - 1 using a 47 Chapter 1: Literature Review functional assay based on the ability of HSV-1 to infect cells by attaching to cell surface H S (McCormick et al., 1998). A s previously mentioned, sog9 cells do not synthesize any glycosaminoglycans. To determine the defect in the sog9 cell line, sog9 cells were transfected with pools of HeLa cell c D N A s , followed by incubation with H S V - 1 (McCormick et al., 1998). Several rounds of screening resulted in the isolation of a clone encoding the gene EXT1, which restored wild-type levels of susceptibility to HSV-1 infection in sog9 cells. To determine whether HSV-1 attachment in s o g 9 - E X T l cells was mediated by an interaction with a cell surface G A G , HSV-1 infection assays on sog9 cells expressing E X T 1 ( sog9-EXTl) were performed in the presence of soluble heparin. HSV-1 infection of s o g 9 - E X T l cells was effectively inhibited by the presence of the polyanion, suggesting that sog9-EXTl cells resembled normal L cells with regard to HSV-1 infection and were decorated with H S G A G s . This was confirmed by anion-exchange H P L C analysis, which revealed that sog9-EXTl cells synthesized a single HS G A G peak. Recent work by Craig McCormick in our laboratory has demonstrated that EXT-1 encodes a heparan sulfate polymerase (McCormick et al., 1999). Sog9 cells harbour a 322 bp deletion in the EXT1 gene. This creates a premature stop codon which results in the production of a 335 amino acid protein, instead of the normal 746 amino acid protein product (Dr. F. Tufaro, personal communication) EXT1 is one of three genetic loci for hereditary multiple exostoses ( H M E ) . H M E is an autosomal dominant disorder characterized by the formation of cartilage-capped tumours (exostoses). This condition can lead to short stature, skeletal abnormalities and malignant transformation of exostoses to chondrosarcomas (Hennekam, 1991) and osteosarcomas (Luckert-Wicklund et al, 1995). The EXT gene family includes EXT1 on 8q24.2 (Ahn et al, 1995), EXT2 on 1 l p l 1-13 (Stickens et al, 1996) and EXT3 on 19p (Le Merrer et al, 1994). This family has recently been extended to include the E X T - l i k e genes EXTL1, EXTL2 and EXTL3, although none of these have been linked with H M E (Kitagawa, 1999). 48 Chapter 1: Literature Review It is becoming increasingly evident that E X T family members play pivotal roles in the G A G synthesis pathway. A s mentioned earlier, both EXT1 and EXT2 encode an H S polymerase (Lind et ah, 1998). EXTL2 encodes a glycosyltransferase that transfers G l c N A c / G a l N A c to the GAG-protein linkage region (Kitagawa, 1999). We previously demonstrated that E X T 1 is an ER-resident transmembrane glycoprotein that has a type II configuration (that is, it has a cytoplasmic amino terminus and a lumenal carboxy terminus) (McCormick et al., 1998). Recent work by Craig McCormick in our laboratory has demonstrated that E X T 1 forms a complex with E X T 2 and in doing so is transported to the Golgi complex from the E R (Craig McCormick, personal communication). Our hypothesis is that E X T 1 functions in the Golgi apparatus where it modifies proteoglycans transversing the secretory organelles en route to the cell surface. This hypothesis is now being tested. 1.6.5. Core Proteins Many distinct proteins serve as core proteins for proteoglycans. It is the combination of a core protein and specific G A G chain(s) that bestows a unique structure and function on any given proteoglycan. There are three general classes of proteoglycans. The first includes proteoglycans which are secreted and deposited into the extracellular matrix (Esko, 1991). These include PGs such as aggrecan and basement membrane proteoglycans. The second family comprises PGs which are anchored to the plasma membrane, such as syndecan. PGs can be anchored to the cell membrane either by a hydrophobic peptide region or by phosphatidylinositol (PI). The third group of PGs comprises those which are typically found intracellularly in secretory granules and include the P G serglycin. The primary sequences of known PGs range in molecular weight from 20,000 (serglycin) to greater than 200,000 (aggrecan) (Hascall et al, 1994). Some core proteins contain only one or few G A G chains, while others can contain as many as 100 or more. Moreover, core proteins can contain more than one type of glycosaminoglycan. 49 Chapter 1: Literature Review Which cell surface proteoglycans carry the HS moieties that H S V interacts with? Treatment of B H K cells with phosphatidylinositol-specific phospholipase C partially reduced HSV-1 binding to cells (Langeland and Moore, 1990). This suggests that PI-linked proteoglycans are among the PGs that can serve as receptors for the virus. Further studies are required to identify specific proteoglycans mediating viral attachment. 1.6.6 Dextran Sulfate: A Glycosaminoglycan Analog Dextrans are high molecular weight polymers of D-glucopyranose produced from sucrose by the bacterium Leuconostoc mesenteroides (Lindberg and Svenson, 1968). Dextran sucrase catalyzes the transfer of glucosyl groups from sucrose to the growing dextran chain, where they are linked by al—> 6 linkages. Branching of the chain arises from a l —>3 linkages. Dextran sulfate is a synthetic sulfated carbohydrate derived from the esterification of dextran. The structure of dextran sulfate is shown in Figure 1.12. Dextran sulfate is a highly sulfated glycosaminoglycan analog and as such is a useful reagent for studying GAG-l igand interactions. For example, dextran sulfate (DS) has been shown to be as active as heparin in binding fibronectin (Ruoslahti, 1988). This activity of dextran sulfate demonstrates that GAG-l igand interactions are not always specific and can be functions of charge density. Indeed, large glycosaminoglycan chains tend to interact more strongly with ligands than small ones (Klebe and Mock, 1982). DS is an effective inhibitor of enveloped virus infection (Baba et al, 1988). DS can inhibit viral infection by a number of mechanisms. For example, dextran sulfate causes certain strains of Coxsackie virus to aggregate, an effect which seems to inhibit plaque formation by repressing viral release (Totsuka et al, 1981). Ohki and colleagues (1992) demonstrated that DS inhibits the binding and subsequent fusion of Sendai virus with erythrocyte ghosts. This may be due to the steric hindrance of a large polyanionic macromolecule adhered to the virion surface. DS has also been shown to inhibit human 50 Chapter 1: Literature Review CH 2 0S0 3 Figure 1.12 Structure of the glycosaminoglycan analog dextran sulfate. 51 Chapter 1: Literature Review immunodeficiency virus type 1 (HIV-1) binding, replication and formation of syncytia (Baba et al., 1988a; Baba et al., 1988b; Lederman et al., 1989; Ida et al, 1994). Callahan et al. (1991) observed that DS interacted with one of the H I V glycoproteins, gpl20. This interaction inhibits binding of H I V to the cell surface. Finally, D S inhibits H S V attachment and infection of cells (Banfield et al, 1995a). The inhibitory action of D S on the virus can be attributed, in part, to a reduction in electrostatic binding. The positively charged sites on the proteins are likely sequestered by negative sulfate groups on dextran sulfate and are therefore be unavailable to interact with cell surface HS . In light of this, it was surprising that D S enhances HSV-1 infection of GAG-deficient sog9 cells (Dyer et al, 1997; this study). The work described in this thesis supports a model whereby in the absence of H S , HSV-1 can use D S as a surrogate receptor to mediate viral attachment and infection. 1.7 VIRAL ENTRY AND TISSUE TROPISM The initial interaction of a virus with its host cell is a key determinant of host range and tissue tropism. Broad host range viruses bind to widely distributed cell surface molecules. Influenza virus, for example, attaches to sialic acid moieties (Rossman, 1994). Narrow host range viruses, on the other hand, recognize very specific molecules, as exemplified by rhino viruses which bind to intercellular adhesion molecule-1 ( ICAM-1) . Structural features of viral envelope glycoproteins are also key determinants of viral tropism. The host range of avian leukosis virus ( A L V ) , for example, is defined by the env gene, which encodes two glycoproteins (Bova et al, 1986). Nucleotide differences in the env genes of different A L V subgroups permit the virus to use different host cell receptors. Similarly, the conformation of the HIV-1 envelope glycoprotein has been shown to play a role in the host range of this virus (Stamatatos and Cheng-Mayer, 1993). Cellular proteinases, which can activate particular viral fusion proteins involved in penetration, also play a pivotal role in tissue tropism (Nagai, 1995). The availability of a 52 Chapter 1: Literature Review particular proteinase in the host, as well as the cleavability of the viral fusion protein, can determine the tropism of some viruses, such as influenza A . For many years, it has been known that HSV-1 and H S V - 2 have type-selective cell surface receptors. Vahlne and colleagues (1979) demonstrated that UV-irradiated HSV-1 interfered with the subsequent adsorption of HSV-1 but not H S V - 2 particles to the cell surface. These findings correlated with those of Addison et al. (1984), who demonstrated that a temperature-sensitive HSV-1 mutant, which was defective in cell penetration at the non-permissive temperature, could block entry of wild-type H S V - 1 but not H S V - 2 . It has also been observed that HSV-1 adsorption to human synaptosomes and glial cells is higher than that of H S V - 2 . Binding of H S V - 1 , but not H S V - 2 to synaptosomes, is blocked by exposure to H S V - 1 , indicating the existence of type-selective receptors (Vahlne et al., 1980). The identification of HveA, HveB, Hve C and HIgR as receptors for H S V , as well as the observation that strains of HSV-1 and H S V - 2 display preferences for binding these receptors, has contributed significantly to our understanding of the different behaviours of the two serotypes with regard to tissue tropism (Geraghty et al., 1998). 1.7.1 HSV Serotype Differences in Glycosaminoglycan-Binding There is increasing evidence that the interaction of the H S V glycoproteins with G A G s may play an important role in the tissue tropism of the virus and that H S V - 1 and H S V - 2 use different surface glycoproteins to bind to G A G s and other receptors. The observation that HSV-1 gC-negative viruses are severely impaired in their ability to bind to HS G A G s , while H S V - 2 gC-negative virions can bind to cells as efficiently as wild-type H S V - 2 (Gerber et al, 1995), has lead to the speculation that gB-2 may be the principle H S V - 2 glycoprotein that interacts with G A G s . Several other studies have also provided indirect evidence that the GAG-binding functions of gB and gC may be different for the H S V serotypes. For example, neomycin, a cationic inositol lipid-complexing agent, 53 Chapter 1: Literature Review is able to inhibit binding of HSV-1 but not H S V - 2 to the cell surface (Langeland et al., 1987). The resistance of H S V - 2 to neomycin was later shown to map to the N-terminal portion of gC-2 (Oyan et al., 1993). Oyan and colleagues (1993) also demonstrated that in the presence of neomycin, inhibition of binding of gC-deficient H S V - 2 virions approached that of wild-type H S V - 1 . The authors suggested that H S V - 2 could use two adsorption pathways: one pathway is neomycin resistant and involves gC-2 binding to a specific receptor, while the second is neomycin sensitive and mediated by the H S V - 1 receptor. The extent to which this may determine the epidemiology of H S V is unclear, although the findings are intriguing and suggest that the domains of gC involved in binding may be type-specific. With regards to gB, our lab has demonstrated that the GAG-binding properties of gB-1 and gB-2 differ (Dyer et al, 1997; this study). Soluble dextran sulfate is able to rescue H S V - 1 , but not H S V - 2 infection of GAG-deficient sog9 cells. Using a panel of deletion mutants, the ability of HSV-1 to interact with dextran sulfate was shown to map to gB-1. The work presented in this thesis details the mechanism of D S stimulation, as well as attempts to identify the domains of gB-1 that interact with D S . Such domains may be important determinants of the interaction of H S V with cell surface G A G s . Recently, Herold and colleagues (1996) identified the specific features of H S that are important for binding HSV-1 and H S V - 2 by comparing the abilities of modified heparin compounds to inhibit plaque formation. Modified heparin compounds which structurally resembled the cell surface heparan sulfate receptor competitively inhibited viral attachment and infection. Key structural features for interactions of both H S V - 1 and H S V - 2 with cell surface G A G s included N-sulfate and carboxyl groups. Important determinants for H S V - 1 infection in particular were 6-0 and 2-,3-0 sulfations. Taken together, these results suggest that differences in the interactions of H S V - 1 and H S V - 2 with cell surface H S may influence tissue tropism. 54 Chapter 1: Literature Review 1.7.2 Protein-Glycosaminoglycan Interactions In order to understand H S V attachment and tissue tropism, it is important to address the factors that govern protein-GAG interactions. The analysis of protein-GAG interactions, however, is complicated due to the structural heterogeneity of G A G s and the diversity of proteins which bind them. Thus, factors that influence one type of protein-G A G interaction may not be the same for an interaction involving different GAG-protein components. In general, interactions between HS/heparin and proteins are ionic in nature and depend on the presence of sulfate groups (Kjellen and Lindahl 1991; Salmivirta et al., 1996). Negatively charged sulfate groups can interact with positively charged groups of amino acids on the protein. Frequently, therefore, binding can be of low specificity and affinity (electrostatic binding). In this instance, different polysaccharide sequences would be able to interact with a given protein. Conversely, binding can be of high specificity and affinity. For example, antithrombin binds with high affinity to a specific pentasaccharide sequence (Salmivirta et al., 1996). Thus, the arrangement of sulfate groups and carboxyl groups, as well as the sugar composition of the polysaccharide can also influence protein binding. The ability of proteins to bind to HS/heparin likely depends on the charge density, structure and conformation of defined heparin-binding domains ( H B D ) that are found in these proteins. Cardin and Weintraub (1989) analyzed various heparin-binding proteins and identified two consensus sequences for glycosaminoglycan recognition: [-X-B - B - X - B - X ] and [ - X - B - B - B - X - X - B - X ] , where B is a basic residue and X is a hydrophobic residue. Other proteins may, however, have different GAG-recognition motifs. Sobel et al. (1992) demonstrated that the heparin-binding domain of Human von Willebrand Factor (a plasma glycoprotein) consists of the sequence [ X - B - B - X - X - B - B - B -X - X - B - B - X ] . Some heparin-binding proteins do not contain any of these consensus motifs (Margalit et al., 1993). Thus, Caldwell et al. (1996) examined the importance of non-basic amino acids in HS/heparin binding. Glycine, serine and proline were found to contribute to HS-peptide interactions. Glycine and serine, due to their small side chains, 55 Chapter 1: Literature Review rninirnize steric hindrance and when located between basic residues, can facilitate electrostatic interactions with anionic sulfate groups. Proline, on the other hand, introduces " kinks" in the protein and in this way may help to expose the critical binding site. Margalit and co-workers (1993) concluded that the spatial arrangement of basic amino acids within an a-helix or P-sheet provided the critical secondary structure for the interaction with HS/heparin. The heparin-binding domains of both HSV-1 gB and gC have been identified (Trybala et al, 1994; Tal-Singer et al, 1995; Laquerre et al, 1998b). To study the heparin-binding properties of gB-1, Laquerre and colleagues (1998) constructed the H S V -1 mutant KgBpK", which carried a deletion in the lysine-rich heparin-binding domain of gB-1 (amino acids 68 to 76) (Fig. 1.7). This virus showed reduced adsorption to cells. Moreover, solubilized gB-1 p K _ protein had lower affinity for heparin-acrylic beads than wild-type gB-1. Taken together, these results demonstrated that the heparin-binding domain of gB-1 had been inactivated. The gB-1 heparin-binding domain has the sequence K P K K N K K P K , which is similar to the consensus motifs predicted for other heparin-binding proteins (Cardin and Weintraub, 1989). The heparin-binding region of gB-2 has not been identified although clusters of basic amino acids are present in the N-terminus and have been proposed as the heparin-binding sites (Williams arid Straus, 1997). The interaction of g C - l with HS has been extensively studied. A study by Trybala and colleagues (1994) identified two sites on the protein critical for gC-mediated attachment of H S V - 1 to HS . One site included residues Arg-143, Arg-145, Arg-147 and Thr-150. A second site included Gly-247 (Fig. 1.7). The two regions are separated by 100 amino acids, although in a model proposed by the authors, these regions would be in close to one another in the three-dimensional structure of the glycoprotein, possibly forming a single attachment site. Another significant conclusion of this study was that the binding of HS to these regions of g C - l was specific. Only synthetic peptides having the specified amino acid sequence bound irreversibly and with high affinity to HS . Synthetic peptides corresponding to other regions of g C - l or which contained basic 56 Chapter 1: Literature Review residues but in a different sequence, bound weakly to HS . This clearly demonstrates that the topographical arrangement of amino acids is an important determinant of the HS-gC interaction. Finally, a study by Tal-Singer (1995) and workers demonstrated that there was an additional H B D in the N-terminus of gC-1, this one mapping to amino acids 33 to 123. There is evidence that the heparin binding domains of HSV-1 gC and gB interact with different structural features of HS. Herold and colleagues (1995) compared the ability of modified heparin compounds to inhibit infection of wild-type H S V - 1 as well as gC- and gB-negative viruses. 2-3-0- desulfated heparin and carboxyl-reduced heparin inhibited attachment of gC-positive viruses (wild-type and gB-negative) to cells, but minimally inhibited attachment of gC-negative H S V - 1 . These results suggest that carboxyl-reduced and 2-,3-0- reduced heparin compete with structural features of cell surface HS that bind to gC but not to gB. Using plasmon resonance-based biosensor technology, Williams and Straus (1997) directly examined the specificity of gB-2 binding to heparin. Heparin, which is more sulfated than heparan sulfate, was a better competitive inhibitor of gB-2 binding to immobilized heparin. Moreover, desulfated, N -acetylated heparin was unable to compete for gB-2 binding. Taken together, these results demonstrate that the degree of sulfation is an important determinant for gB-2 binding. Significantly, iduronic acid was also important for recognition by gB-2, since heparin, H S and chondroitin sulfate B (dermatan sulfate), which all contain IdoA, could inhibit binding of gB-2 to heparin, whereas chondroitin sulfate A , which lacks IdoA, was a weak inhibitor. The structural requirements of H S for interaction with H S V - 1 gC have been studied in detail. Using gC-1 affinity chromatography, Feyzi and colleagues (1997) demonstrated that the shortest HS sequence that could bind gC-1 comprised 10-12 monosaccharide units and contained at least one 2-0- and one 6-O-sulfate group. The authors concluded that these groups have to be arranged in a specific order within the polysaccharide sequence, since H S fragments which did not bind to gC-affinity columns 57 Chapter 1: Literature Review were of similar in charge and composition to those fragments which bound to gC. The HS-binding fragment also contained N-sulfate groups, although these sulfate groups were considered to be less important than the 2-0- and 6-0-sulfations. This was based on the finding that N-desulfated heparin reduced the inhibitory effect of heparin by 10-fold in a virus infectivity assay whereas removal of 2-0- or 6-0- sulfate groups reduced the antiviral activity of heparin by approximately 100-fold. 1.8 CONSTRUCTION OF HSV MUTANTS: TECHNIQUES FOR MANIPULATING THE HSV GENOME It is evident that much of what is known about the biology of H S V has come from studies involving the construction of various deletion mutants or intertypic recombinant viruses. The present study is no exception, in that in order to understand better the interaction of HSV-1 with dextran sulfate, it was desirable to generate particular H S V recombinants (see Chapter 6). For this reason, it is worthwhile to review the various techniques that are used for the construction of viral mutants. Unt i l the complete D N A sequence of HSV-1 was published in 1988 by McGeoch and colleagues, HSV-1 recombinants were constructed with minimal knowledge of viral D N A sequences. Mutants could be selected on the basis of plaque morphology (Ejercito et al, 1968), drug resistance (Kit and Dubs, 1963), or temperature sensitivity (Schaffer et al., 1970). The drug resistant mutants contain lesions in genes encoding enzymes that are sensitive to these inhibitors. For example, H S V mutants containing lesions in the viral tk gene are sensitive to the drug A C V . Other mutants having defects in the envelope glycoproteins have been selected on the basis of resistance to monoclonal antibodies. These monoclonal antibody-resistant (mar) mutants have proven to be useful tools for identifying the functional domains of some of the H S V glycoproteins (Trybala et al., 1994). 58 Chapter 1: Literature Review Current methods for constructing H S V mutants include the following: (i) a plasmid-based method, (ii) the use of H S V cosmids and (iii) the use of H S V bacterial-artificial chromosomes ( H S V - B A C ) . The plasmid-based method relies on co-transfection of a plasmid bearing the mutated gene (which should also contain a marker) and the complete viral D N A genome. Markers can include the HSV-1 tk gene (Post et al., 1981) or a detectable marker such as E.coli lac Z(Goldstein and Weller, 1988). Recombination between homologous regions in the plasmid and viral D N A wi l l produce mutant progeny, which can be isolated from wild-type virus on the basis of expression of the marker gene. A limitation of this method, however, is the inability to isolate mutant progeny in the absence of wild-type virus. This problem is overcome using H S V cosmid and B A C technology to produce recombinants (Cunningham and Davison, 1993; Horsburgh and colleagues, 1999), both of which are described below. 1.8.1 H S V Cosmids Cosmid sets that contain the genomes of other herpesviruses, including varicella-zoster virus, cytomegalovirus and Epstein-Barr virus, have been created (Cohen and Seidel, 1993; Tompkinson et ai, 1993; Kemble et ai, 1996). The method is illustrated in Figure 1.13. Briefly, cosmids containing HSV-1 fragments are constructed. Cosmid sets containing 4 to 5 cosmids, whose viral inserts overlap and represent the entire genome, are then identified. The viral inserts are co-transfected into cells and subsequent recombination between overlapping viral sequences yields infectious progeny. The cosmid system has been successfully used to produce recombinant H S V - 1 particles (Cunningham and Davison, 1993; Fraefel et al., 1996). This entails manipulation of cosmid D N A such that wild-type cosmids are replaced with cosmids containing the desired mutation(s). Co-transfection of these cosmids wi l l produce the desired recombinant in the absence of any wild-type progeny. Despite its advantages, the cosmid-based system does have several drawbacks. 59 Chapter 1: Literature Review 20 40 60 80 100 120 140 T gB I kb Figure 1.13. Strategy for the generation of H S V recombinants using a cosmid -based system. The H S V genome is depicted as two unique regions (horizontal lines), the larger U L and the smaller Us, flanked by inverted repeats (rectangles). Overlapping cosmids are represented by open bars. The cosmid carrying a mutation in the gene of interest (gB) is co-transfected with the other cosmid clones. Recombination between overlapping cosmid sequences w i l l form a complete H S V genome which can undergo replication. 60 Chapter 1: Literature Review First, mutating viral sequences in cosmids is not necessarily a trivial task due to the large size of the cosmid inserts. In some instances, it may be necessary to perform a series of subcloning steps in order to obtain a manageable size of D N A before mutating the gene of interest. The cosmid then has to be reconstructed to contain the altered sequence. This can obviously be a tedious process. Finally, H S V cosmids are not stable in bacteria. H S V cosmids maintained in E.coli are prone to deletion, recombination and rearrangement and are, thus often heterogeneous in nature (Van Zig l et al, 1988; Cunningham and Davison, 1993; Horsburgh et al. in press). 1.8.2 HSV Bacterial Artificial Chromosomes Some of the difficulties experienced with H S V cosmids can be overcome using H S V - B A C s , in which the entire H S V genome is cloned into a single bacterial artificial chromosome (Stavropoulos and Strathdee, 1998; Horsburgh et al, 1999). H S V - B A C D N A is stable in bacteria. (Horsburgh et al., 1999). This may be because the copy number of B A C s in bacteria is only approximately one copy per cell which prevents recombination. Moreover, H S V - B A C technology exploits bacterial genetics for the construction of recombinant herpesviruses, allowing rapid and direct cloning of viral recombinants. The H S V - B A C mutagenesis procedure is illustrated in Figure 1.14. Two starting reagents are required, the first being an H S V - B A C . The H S V - B A C is constructed by integration of a B A C vector into viral D N A (Horsburgh et al., 1999). The B A C vector contains B A C sequences and a marker gene, such as chloramphenicol. The site of integration of the B A C vector into the viral genome needs to be at a locus not essential for viral replication. The viral tk gene is one such locus and also allows for selection of infectious H S V - B A C on the basis of A C V resistance. Recombinant H S V - B A C viruses are harvested and circular viral D N A is isolated and electroporated into E.coli. Colonies containing H S V - B A C are selected on the basis of resistance to chloramphenicol. 61 Chapter 1: Literature Review Figure 1.14. The B A C system for construction of H S V recombinants. (A) Vi ra l D N A and a vector carrying B A C sequences flanked by viral sequences are co-transfected into mammalian cells. The B A C D N A is integrated into the viral genome by homologous recombination to generate H S V - B A C . (B) To generate mutants, a vector carrying the altered allele (mut) is transformed into bacteria containing H S V - B A C . The mutant allele is integrated into the H S V - B A C genome by a two-step replacement strategy. (C) The mutant H S V - B A C plasmid is transfected into mammalian cells to produce infectious recombinant virus. 62 Chapter 1: Literature Review Eukaryotic cell 1 .transfection 2.recombination 3.isolation of BAC genome 63 Chapter 1: Literature Review The second reagent required for B A C mutagenesis is a gene replacement plasmid, which contains the desired mutated gene flanked by H S V sequences to target the replacement gene to a particular locus within the H S V - B A C . The gene replacement vector is transformed into E.coli that contain H S V - B A C . The mutated gene is transferred to the H S V - B A C using a 2-step homologous recombination strategy. Mutant H S V - B A C D N A is then isolated from positive clones and transfected into mammalian cells. H S V - B A C D N A is infectious in mammalian cells and cytopathic effects should be observed within 2 to 3 days post-transfection. Using this procedure, Horsburgh and colleagues (1999) constructed a recombinant HSV-1 virus in as few as seven days. Moreover, all mutant viruses are produced in the absence of wild-type virus and thus, there is no need to purify recombinant virus. Since, all herpesvirus genomes circularize in the nucleus of infected cells, it should be possible to clone them as B A C s and perform mutagenesis in bacteria. Murine C M V has also recently been cloned into a B A C (Messerle et al., 1997). The B A C mutagenesis system should prove to be a powerful tool for manipulating the H S V genome, thereby facilitating construction of H S V mutants. 1.9 RATIONALE OF STUDY A t the time this project was initiated, the H S V entry pathway, and its role in determining H S V tropism, was not well understood. The approach our laboratory took to investigating H S V entry involved the isolation of H S V resistant cell lines (Gruenheid et al., 1993; Banfield et al., 1995a). It was previously demonstrated that H S V infection of sog9 cells, which do not synthesize any cell surface G A G s , was reduced by nearly three orders of magnitude relative to that of control mouse L cells. Surprisingly, however, the addition of the G A G analog dextran sulfate could enhance HSV-1 infection of sog9 cells 64 Chapter 1: Literature Review (Banfield et al., 1995). Moreover, D S stimulation of infection was type-specific, in that only H S V - 1 , but not H S V - 2 infection, was enhanced in the presence of the G A G analog. The overall rationale for this study is depicted in Figure 1.15. One of the objectives of this work was to determine the mechanism of D S stimulation of H S V - 1 infection of sog9 cells. DS could act to tether the virus to the cell surface and thereby facilitate infection. Alternatively, DS could enhance infection by facilitating viral penetration and/or a post-entry process. Since D S stimulated H S V - 1 , but not H S V - 2 infection, another aim of this study was to map the viral components that facilitated DS type-specific enhancement of infection. The differences between HSV-1 and H S V - 2 in their interactions with DS might reflect differences in how the virions interact with H S G A G s . Mapping the DS activation site on the virus could potentially identify a virion component that may be a determinant of the different behaviours of H S V - 1 and H S V - 2 in the human host. The observation of differences in the ability of HSV-1 and H S V - 2 to use DS as a surrogate receptor on GAG-deficient sog9 cells, underscored the intricate nature of H S V glycoprotein interactions with glycosaminoglycans. The third objective of this work was to determine which viral and cellular determinants were important for interaction of H S V with its natural cell surface receptor, heparan sulfate (Chapter 4). One approach that was used was to test H S V - 1 infection on control and EXT1 -expressing cell lines in the presence of modified heparin compounds. This could help to identify the structural features of H S important for H S V infection of EXT1 -expressing cells and of control cell lines. B y testing HSV-1 heparin-binding mutants, the roles of g C - l and gB-1 in mediating HSV-1 attachment to EXT1 -expressing cell lines could also be assessed. Finally, as part of the effort to characterize virion components mediating DS stimulation of HSV-1 infection, it was desirable to construct viruses bearing deleted and chimeric forms of particular H S V glycoproteins. A s mentioned earlier, however, the construction of H S V recombinants is often a difficult and time-consuming process. Thus, the final objective of this work was to develop a system which would make the H S V 65 OBSERVATION Chapter 1: Literature Review DS stimulates HSV-1 but not HSV-2 infection of GAG-deficient sog9 cells Q: Can DS substitute for HS as a receptor for HSV-1 attachment? A: YES Q: Are HSV-GAG interactions specific or non-specific? examine HSV-1 infection of EXTl-cell lines; determine which HS moieties are important for infection A: Both Q: Which HSV glycoprotein confers DS stimulation? test intertypic recombinants A: HSV-1 gB i Q: Can DS stimulation of HSV-2 be restored by expressing gB-1 in place of gB-2? construct HSV-2 gB intertypic virus A: Unknown; could not construct virus using plasmid-based strategy Q: Does the gB-1 HBD mediate DS stimulation of HSV-1 infection? test HSV-1 gB HBD mutant A: No Q: Will cosmid and/or BAC system facilitate construction of HSV-2 mutants? construct HSV-2 cosmid set and HSV-2 BAC A; Unknown; had difficulties generating these reagents; each system for generating mutants has its own merits and drawbacks Figure 1.15. Rationale of study. 66 Chapter 1: Literature Review genome more amenable to genetic manipulation (Chapter 6). Specifically, I wanted to explore the possibility of using H S V - 2 cosmids and H S V - 2 B A C s for constructing H S V - 2 recombinants. Other research groups have used the cosmid and B A C technologies to successfully construct H S V - 1 mutants (Cunningham and Davison, 1993; Horsburgh et al, 1999). The development of similar systems for H S V - 2 would greatly facilitate investigations of this virus. 67 Chapter 2: Materials and Methods CHAPTER 2: MATERIALS AND METHODS 2.0 MATERIALS A l l tissue culture reagents (media, fetal bovine serum (FBS), penicillin-streptomycin and trypsin), tissue culture dishes, L ipo fec tAMINE, G418, T4 D N A ligase, electrophoresis grade agarose, acrylamide, Proteinase K , calf intestinal alkaline phosphatase, D N A z o l , D H 1 0 B E.coli, X-ga l , Taq D N A polymerase, sequencing primers and most restriction enzymes, were from Canadian Life Technologies (Burlington, ON). The X L I I Packaging System and the X L 1 Blue strain of E.coli for generation of the H S V -2 cosmid library were from Stratagene (La Jolla, C A ) . Gelase and Gelase buffer were purchased from Epicentre Technologies (Madison, WI) and Nucleobond A X 5 0 0 D N A purification kits were from the Nest Group (Southboro, M A ) . The T7 Sequenase v 2.0 7-deaza-dGTP sequencing kit and oc-[ 3 2P]-dATP were purchased from Amersham (Oakville, ON.) . [ 3 5S]-methionine was from Dupont-NEN (Missassauga, ON) . Bovine serum albumin (BSA) , ampicillin, RNAse A and acyclovir ( A C V ) were purchased from Sigma Chemical Co. (St. Louis, M O ) . Scintillation cocktail (Ready-Safe) and all ultracentrifuge rotors and tubes were from Beckman (Palo Alto , C A ) . Human IgG was purchased from I C N ( St. Laurent, PQ) while Qiagen D N A purification kits were from Qiagen (Missassauga, ON.) Pme I restriction enzyme was purchased from N e w England Biolabs (Missassauga, ON.) D S with a molecular weight ( M W ) of 500,000 ( M W 500,000 DS) was from Pharmacia (Baie d ' Urfe, PQ). M W 5000,15,000, and 50,000 DS were from Sigma, as was heparan sulfate. Modified heparin compounds were a generous gift from Dr. U . Lindhal, Uppsala University, Sweden. 68 Chapter 2: Materials and Methods 2.1 P L A S M I D S Standard molecular biology techniques were used in the course of this study. Small-scale plasmid D N A purification, restriction enzyme analysis and gel electrophoresis were performed according to "Methods in Molecular Cloning" (Sambrook et al. 1989). Large scale D N A purification was performed using either Nucleobond or Qiagen purification kits according to the manufacturer's specifications. The plasmid pHS208 contains the gene encoding for gB-2 in the plasmid vector pBR322 and was a gift from Chiron Inc. The plasmid pHS208-/acZ (Fig.2.1A)is derived from pHS208. In pHS208-/acZ, the gB-2 gene is disrupted by a lacZ gene cassette. To construct this plasmid, the plasmid pUC249, a gift from Neurovir Inc., was digested with Sail to yield a 4.6 kb fragment containing lacZ driven by the h C M V promoter. The h C M V - l a c Z cassette was then ligated into Sail digested pHS208 to produce pHS208-/acZ. The P M S I cosmid vector (Fig. 2.1 B) was a kind gift from Dr. Bernard Roizman (University of Chicago, IL.). This vector was derived from the Stratagene SuperCos 1 cosmid vector and was generated by replacing the cloning site of SuperCos 1 with one which was flanked by two Pmel sites. The plasmid p B A C - T K (Fig. 2.1C), which contains H S V tk sequences flanking bacterial artificial chromosome ( B A C ) elements and the chloramphenicol resistance gene, was a gift from Neurovir Inc. The B A C elements and the chloramphenicol marker gene in p B A C - t k were derived from p B e l o B A C (Shizuya etal, 1992). 69 Chapter 2: Materials and Methods Figure 2.1. Schematic diagram of the pHS208-lacZ plasmid, the P M S I cosmid vector, and the p B A C - t k plasmid. (A) The pHS208-lacZ plasmid. pHS208-lacZ contains the E.coli lac Z gene driven by the h C M V promoter. The h C M V / l a c Z cassette is flanked by H S V - 2 gB-2 sequences. pHS208-lacZ also contains an ampicillin resistance gene (amp). (B) The P M S I cosmid vector. P M S I was constructed by C. V a n Sant in the laboratory of Dr. B . Roizman (University of Chicago) and is derived from the Stratagene Super C O S 1 vector. P M S I has two cos sites that facilitate packaging of the vector into lambda phage heads. Two Pmel sites are also indicated in the illustration. Work presented in Chapter 6, however, demonstrates that one of these sites (indicated by the asterisk) is not functional. (C) The p B A C - t k plasmid. This plasmid was constructed by Neurovir Inc. and contains B A C sequences flanked by HSV-1 tk sequences. p B A C - t k also contains a chloramphenicol ( C M ) resistance gene. Relevant restriction sites for all the vectors are indicated. 70 Chapter 2: Materials and Methods Sal I (6963) Hind III (3375) Xbal(1172) 71 Chapter!: Materials and Methods 2.2 CELL LINES The parental L cell line used for all experiments was the clone I D line of LMtk" murine fibroblasts. The procedure for isolation of the mutant gro2C and sog9 cell lines was described previously (Gruenheid et al., 1993; Banfield et al., 1995a). L , gro2C and sog9 cells were grown at 37°C in Dulbecco Modified Eagle medium ( D M E M ) supplemented with 10% F B S in a 5% C 0 2 atmosphere. S o g 9 - E X T l and L - E X T 1 cells were generated by liposome-mediated transfection with the plasmid p E X T l , followed by selection in media supplemented with 700 ug/ml G418 (McCormick et al., 1998). Clonal versions of these cell lines were derived from individual transfection colonies. The parental C H O K 1 and mutant C H O cell lines were a gift from J.D. Esko and were grown in Ham's F12 media supplemented with 10% F B S (Esko et al., 1985). Vero cells were obtained from S. McKnight and grown in D M E M with 10% F B S . The D6 cell line, a Vero-derived cell line expressing HSV-1 gB-1, was a gift from S. Person (Cai etal., 1987). D6 cells were grown in D M E M , supplemented with 10% F B S and 500ug/ml G418. 2.3 VIRUSES AND VIRAL STOCK PRODUCTION Viruses used in this study, and the research labs in which these viruses were purified and/or constructed, are shown in Table 3. To produce viral stocks, Vero cells were grown overnight in T-flasks and infected with virus at a multiplicity of infection (m.o.i) of 0.1 After a 1 h adsorption period at 37° C, virus was removed and D M E M with 4% F B S was added. When general cytopathic effect was evident (approximately 96 h post-infection), cells were scraped off the dish and transferred, along with the supernatant, into a 50 ml conical tube. To release intracellular virus, harvested cells were subjected to two rounds of freeze-thawing. To pellet cellular debris, tubes were spun 72 Chapter 2: Materials and Methods Table 3. Viruses used in this study Virus Strain Description Source HSV-1 (F) Wild type Dr. B. Roizman (University of Chicago) HSV-1 (KOS) Wild type Dr. D. Coen (Harvard Medical School) HSV-2 (G) Wild type Dr. S. Sacks (UBC) HSV-2 (333) Wild type Dr. J. Smiley (University of Alberta) RSIG25 HSV-l;gC-2 Dr. B. Roizman R7015 HSV-l;gD-2,gE-2;gG-2 HSV-1 (MP) HSV-1; gC-deficient syncytial strain RHIG13 HSV-1; gB-2 UL10" HSV-1; gM deficient R126 HSV-1; gB mar mutant R233 HSV-1; gB mar mutant HSV-l(KOS) AgC2-3 HSV-1; gC-deficient Dr. B. Herold, (University of Chicago) HSV-2(G)gC2- HSV-2; gC-deficient KCZ HSV-l(KOS); gC-deficient Dr. J. Glorioso (University of Pittsburgh KgBpK" HSV-l(KOS); gB heparin-binding domain deleted n KgBpKgC" HSV-l(KOS); gB heparin-binding domain deleted and gC-deficient M HSV-1 (ANG) Wild type; gB syn mutation; rate of entry (roe) mutation in gB Dr. G. Cohen (University of Pennsylvania) HSV-1 (ANG-path) HSV-1 A N G derivative; mutation in a.a 84 of gD; roe and syn mutation in gB I I 73 Chapter 2: Materials and Methods at 3000 rpm for 5 min at 4° C. The supernatant was collected, aliquoted and stored at -80° C. 2.4 VIRUS TITER DETERMINATION Vero cells were plated in 6-well dishes one day prior to inoculation. Confluent monolayers were rinsed with P B S and inoculated with diluted virus preparations in lOOul of D M E M . After a 1 h adsorption period at 37°C, the viral inoculum was removed and the monolayer was rinsed with P B S . Cells were overlaid with D M E M containing 0.1% pooled human IgG which neutralized any extracellular virus. Plaques were visualized and counted after 3 days by fixation and staining of the cells for 10 min with 5% methylene blue in 70% methanol. 2.5 DS STIMULATION ASSAYS Assays for D S stimulation were performed on sog9 cells plated in 6-well dishes. Confluent sog9 cell monolayers were pre-treated with DS in D M E M for selected lengths of time prior to infection or inoculated with virus in D M E M previously mixed with DS . The cells pre-incubated with D S were rinsed three times with P B S prior to infection. After a 60 min viral adsorption period at 37°C, the inoculum was removed and the cells were washed with P B S . The cells were then overlaid with D M E M containing 0.1% pooled human IgG. Plaques were visualized and counted 96 h post-infection. A s a baseline measure of infectivity of HSV-1 on sog9 cells in the absence of DS , monolayers were infected with virus such that approximaly 30 to 40 plaques were consistently counted. 74 Chapter 2: Materials and Methods 2.6 PLAQUE INHIBITION ASSAYS Confluent monolayers of L , sog9, s o g 9 - E X T l , and L - E X T 1 cells were rinsed once with P B S and then inoculated with HSV-1 (F) previously mixed with bovine heparin or with one of the following chemically modified heparin compounds: 6-O-desulfated heparin; 2-O-desulfated heparin; N - 2 desulfated heparin, N-desulfated/N-reacetylated heparin. N-2,6- O-desulfated heparin; N-6-O-desulfated heparin; and 2,6-0 desulfated heparin. The production of these modified heparin compounds has been previously described (Feyzi et al., 1997). After a 60 min incubation at 37°C, the inoculum was removed and the cells were washed with P B S . A s before, the cells were overlaid with D M E M containing IgG to facilitate plaque formation and plaques were counted 3 days postinfection by fixation and staining of the cells for 10 min with 5% methylene blue in 70% methanol. 2.7 VIRAL PENETRATION ASSAYS Confluent monolayers of sog9 cells growing in 6-well dishes were rinsed with P B S and incubated with 1 ml of 100u.g/ml D S in D M E M at 4°C for 30 min. The cells were rinsed three times with P B S and inoculated with HSV-1 at 4°C for 30 min. The dishes were then shifted to 37°C to allow penetration to proceed. A t various times after the temperature shift, the wells were treated for 2 min with 2 ml of low-pH citrate buffer (40 m M sodium citrate, 10 m M K C I , 135 m M N a C l [pH 3.0]) to remove any bound virus that had not penetrated into cells. The monolayers were washed three times with P B S and overlaid with D M E M containing 4% F B S and 0.1% IgG. Alternatively cells were inoculated with virus in the presence of 10 ug/ml of DS at 37°C. A t various times, the virus was removed and the monolayers were washed with citrate buffer. Plaques were visualized and counted 96 h postinfection by fixing and staining the cells as described above. 75 Chapter 2: Materials and Methods 2.8 PREPARATION OF RADIOLABELED VIRUS Monolayers of L cells were infected with H S V at a M.O. I , o f 10. A t 2 h postinfection, the medium was changed to methionine-free labeling medium (methionine-free D M E M : 0.1 volume of D M E M / 1 0 % F B S - 4 % dialyzed FBS-500u€ i of [ 3 5S] methionine. When a generalized C P E was evident, the medium was removed from the infected-cell monolayers and subjected to low-speed centrifugation to pellet cell debris. The supernatant was sedimented through a 30% sucrose pad formed in 50 m M NaCl-10 m M Tris (pH 7.8) for 2 h at 39,000 rpm in a Beckman SW41 rotor. Following centrifugation, the sucrose was removed by aspiration, and the virus pellet was suspended in P B S at 4°C. For determination of viral titers, a sample of virus was diluted serially with medium and used to inoculate monolayers of Vero cells growing in 6-well dishes. Plaques were scored after 3 days. 2.9 BINDING OF HSV TO IMMOBILIZED DS A modified protocol by Leong et al. (1995) was used to examine binding of HSV-1 and H S V - 2 particles to immobilized DS . Nunc-Maxisorp 96-well microtiter dishes were coated with 50 ul of P B S containing 5 mg of DS. The dishes were incubated overnight at 4°C. The wells were rinsed twice with P B S and then blocked for 2 h in 3.5% bovine serum albumin ( B S A ) in 50 m M Tris (pH 7.5)-100 m M N a C l - 1 m M M g C l 2 - l m M C a C l 2 at 20°C. 3 5S-labeled H S V was diluted in adsorption buffer (PBS plus 0.1% glucose). The blocking solution was removed, and 50 ul of virus was added to each well . The microtiter dishes were centrifuged for 15 min at 1,100 x g at 4° C, and then rocked for 45 min at 20°C. The wells were washed three times with P B S , and the dishes were incubated for 10 min with 100 ul of lysis buffer ( l O m M T r i s - H C l [pH 7.4], 150mM N a C l , 1% Nonidet P -40 ,1% sodium deoxycholate). The dishes were rinsed once more with lysis buffer and all rinses were added to scintillation vials. The radioactivity 76 Chapter 2: Materials and Methods 2.10 TIME COURSE EXPERIMENT Sog9 cells grown in 35-mm dishes were inoculated with HSV-1 in D M E M containing 10 ug/ml of D S D M E M and incubated at 37°C. Control dishes were infected with virus in the absence of DS . A t different times postinoculation, virus was removed and 2 ml of D M E M with 4% F B S and 0.1% IgG was added to cells after the monolayers were rinsed with P B S . Plaques were counted 3 days postinfection. 2.11 PREPARATION OF VIRAL DNA H S V - 2 D N A for construction of the H S V - 2 cosmid library was prepared as follows. Two roller bottles of Vero cells were infected with H S V - 2 (G) at M.O. I = 1.0. Two days later, cells were collected from the roller bottles and centrifuged at 5000 rpm for 5 min. The cell pellet was washed with P B S and spun once more as before. Cells were resuspended in 20 ml of resuspension buffer (150 m M N a C l , 10 m M Tris [pH 7.4], 1.5 m M MgC12). NP-40 was added to a final concentration of 0.1%, vortexed briefly, and the nuclei pelleted at 5000 rpm for 5 min. The supernatant was collected and adjusted to have a final concentration of 0.2% SDS, 5 m M E D T A and 50 m M p - M E . The lysate was extracted twice with an equal volume of 1:1 phenohchloroform and each time centrifuged for 10,000 rpm for 10 min. Two volumes of ethanol were added and the lysate centrifuged again at 10,000 rpm for 10 min. The resulting D N A pellet was resuspended in 2 ml of T E and RNAse was added to a final concentration of 20 ug/ml. The D N A was allowed to dissolve at 37°C for 15min. The viral D N A solution was then loaded on top of a linear 5-20% potassium acetate gradient in T E (10 m M Tr i s -HCl [pH 8] and 5 m M E D T A ) which was centrifuged in an SW41 rotor at 27,000 rpm for 16 h at 4°C. The supernatant was gently removed and 3 ml of T E was added to the pellet. The D N A was ethanol precipitated once more and redissolved in 0.5 ml T E [pH 7.5]. 77 Chapter 2: Materials and Methods For all other experiments, viral D N A was prepared by infecting Vero cells grown in 10 cm dishes at a M.O. I , o f 0.1. When generalized C P E was evident, cells and medium were collected in a 15 ml conical tube. Cells were spun down for 5 min at 3,000 rpm , the supernatant was removed and the cells washed by resuspending in 5 ml P B S . The cells were pelleted again and the P B S removed. The cells were resuspended in 1 ml of D N A extraction buffer (100 m M N a C l , 10 m M Tr i s -HCl [pH 8], 25 m M E D T A [pH 8], 0.5% SDS) and proteinase K was added to a final concentration of 0.1 mg/ml. The cells were incubated overnight at 65°C, following which the lysate was transferred to a 1.5 m l vial . The D N A was extracted once with an equal volume of phenol and extracted once more with an equal volume of 1:1 phenol:chloroform and then again with an equal volume of chloroform. The viral D N A was precipitated with 1/2 volume of 7 .5M ammonium acetate and 2 volume of ethanol. The pellet was resuspended in 50ul of T E and the D N A later analyzed by restriction digest. 2.12 C O N S T R U C T I O N O F HSV-2 gB (LACZ) V I R A L M U T A N T Construction of the H S V - 2 gB" lacZ virus involved co-transfection of pHS208-lacZ with H S V - 2 D N A into gB-1 complementing D6 cells. The day prior to transfection, D6 cells were plated in 60 mm dishes at a density of 1 x 10 6 cells per dish. pHS208-/acZ was linearized by overnight digestion with X h o l . One microgram of linearized pHS208-lacZ was mixed with 4 ug of H S V - 2 genomic D N A . 100 ul of sterile d H 2 0 and 200 ul of serum-free, antibiotic-free D M E M were added to the D N A solution. In a separate tube, 10 ul L ipo fec tAMINE was diluted in 300 ul serum-free antibiotic-free D M E M . The D N A and L ipo fec tAMINE solutions were then mixed together and kept at room temperature for 15 min. In the meantime, the D6 cells were rinsed with serum-free, antibiotic-free D M E M . The transfection mixture was combined with 2.4 m l of serum-free, antibiotic-free D M E M and was then added to the cells. Following an incubation period of 18 h at 37°C, the transfection mixture was removed and regular D M E M / 1 0 % 78 Chapter 2: Materials and Methods F B S was added to the monolayers. When generalized C P E was evident, infected cells were scraped off the dishes and virus was harvested as previously described. To isolate H S V - 2 gB'lacZ virus, confluent monolayers of D6 cells in 6-well dishes were infected with serially diluted virus stock. After a 1 h incubation period at 37°C, virus was removed and monolayers were rinsed with P B S . Cells were then overlaid with 0.8% agarose in D M E M / 4 % F B S . When plaques were observed (3 days post-infection), cell monolayers were stained with 0.1% X-gal in X-gal buffer (10 m M potassium ferricyanide, 1 OmM potassium ferrocyanide, 1 m M M g C l 2 in a final volume of 200 ml in PBS) . Plaques expressing lacZ (those which stained blue) were picked and agar plugs containing virus were added to D6 cells growing in 96-well dishes. Infection was allowed to proceed for 3-6 days. If C P E was evident, cells were scraped and virus harvested. To confirm the gB27acZ genotype, isolated virus was once more used to infect D6 cells grown in 6-well dishes. Following a 1 h incubation at 37°C, virus was removed and D M E M containing 4% F B S and 0.1% pooled human IgG was added to the cell monolayers. Plaques were stained with X-gal to detect lacZ expression. 2.13 CONSTRUCTION OF HSV-2 VIRAL COSMID LIBRARY 2.13.1 Preparation of viral DNA H S V - 2 D N A was prepared as described above. Vi ra l D N A was partially digested with Sau3AI. One unit of Sau3AI was added to 40 ug of D N A at room temperature and samples were collected at 7, 9 and 11 min. The enzyme was inactivated by adding 50 m M E D T A and then heat inactivated for 20 min at 65°C. The samples were loaded onto a 0.6% low melt gel and run with P a d digested HSV-1 cosmid as a 40 kb control. The 40 kb H S V - 2 fragment was excised from the gel and purified using Gelase. For Gelase purification, the gel fragment was incubated in 1 ml Gelase buffer for 30 min on ice. After 30 min, the remaining buffer was removed and the gel melted at 65 °C for 20 min. The 79 Chapter 2: Materials and Methods sample was then incubated at 40°C for 20 min and 20 ul of Gelase was added to each sample. This was allowed to incubate at 40°C for 2 hours after which 0.1 volume of sodium acetate was added. The sample was chilled on ice for 15 min and then centrifuged at 12,000 rpm for 15 min at 4°C. The supernatant was transferred to a new tube to which 2 volumes of cold isopropanol was added. The sample was gently mixed and then centrifuged for 15 min at 10,000 rpm, 4°C. The pellet was allowed to dry in the fumehood and then resuspended in 15 ul dH 2 0. 2.13.2 Cosmid vector preparation 20 p.g of the cosmid vector P M S I was digested with X b a l at 37°C overnight. The enzyme was heat inactivated at 65 °C for 20 min and the sample dephosphorylated with l u l of calf alkaline phosphatase for 30 min at 37°C. The sample was heat inactivated again and then digested with BamHI at 37°C overnight. 2.13.3 Ligation of HSV-2 Viral DNA Fragments and PMSI Cosmid Vector A ligation mix of 20 | i l total, containing 10X T4 D N A ligase buffer, the entire volume of partially digested H S V - 2 viral D N A , 0.5 | i g of prepared P M S I vector and 1.0 unit of T4 D N A ligase, was used. The ligation was performed overnight at 2-8°C. Once ligated, 30 u l of water was added and the sample was gently mixed with 500|il o f 1-butanol to precipitate the D N A . The tube was spun for 10 min, 10,000 rpm at 4°C and the pellet resuspended in 5 | i l o f d H 2 0 . The sample was then ready to be packaged into lambda phage. 80 Chapter 2: Materials and Methods 2.13.4 Packaging HSV-2 Cosmids into Lambda Phage Stratagene's X L II packaging system was used to package ligated D N A . Packaging was done according to the manufacturer's specifications. Packaged cosmid D N A was stored in 500 ul of S M buffer (5.8 g N a C l , 2.0 g M g S 0 4 , 7 H 2 0 , 50 ml I M Tr i s -HCl [pH7.5], 5.0 m l 2%w/v gelatin per I L d H 2 0 ) at 2-8°C until infection. 2.13.5 Infection X L - 1 Blue M R strain E.coli (Stratagene) was streaked onto an L B plate the day before infection. One colony was picked and incubated for 5 h at 37°C in 50 ml of L B supplemented with 0.2% maltose and 10 m M M g S 0 4 . The culture was centrifuged at 4000 x g for 5 min after which the supernatant was removed. The pellet was re-suspended in 1 m l of cold M g S 0 4 and the adsorbance (A) of the culture was adjusted to A600=1.0 in 10 m M M g S 0 4 . For infection, 50 (al of bacteria was combined with 50 ul of packaged D N A and incubated at room temperature for 30 min. Following this incubation, 400 ul of L B was added to the tube and the infection allowed to continue for 60 min at 37°C. The tube was gently inverted every 15 min. Finally, the tube was centrifuged and the bacterial pellet resuspended in 50 ul of L B . This was followed by plating the sample onto L B plates containing 100|ig/ml ampicillin and then incubating overnight at 37°C. 2.13.6 Restriction mapping HSV-2 Cosmid Clones Using standard alkaline lysis, cosmid D N A was prepared from 200 individual colonies. Cosmids were digested with B g l II and where necessary, K p n I, to map the viral D N A fragments contained in each cosmid. Cosmid D N A was digested as follows: 2ul cosmid D N A in 2 p i enzyme buffer and 15.5 p i d H 2 0 , was digested with 0.5 u l restriction enzyme for 6-18 hours at 37°C. For controls, parental HSV-2(G) viral D N A was digested with B g l II and K p n I, respectively. For these digests, 1 ug viral D N A was 81 Chapter 2: Materials and Methods 0.5 ul of enzyme in a 50 ul reaction for 6-18 h at 37°C. Digested cosmid and control D N A was electrophoresed on a 0.5% agarose gel. 2.13.7 C o s m i d Sequencing To establish an overlapping set of cosmids that spanned the entire H S V - 2 genome, it was necessary to sequence the ends of selected viral inserts. Sequencing was done according to the Sanger dideoxy method using the T7 Sequenase v 2.0 7-deaza-dGTP sequencing kit. Cosmid D N A to be sequenced was prepared using Nucleobond 500X columns. Approximately 10 (xg of cosmid D N A was denatured in 0.1 volume of 2 M N a O H , 2 m M E D T A for 30 min at 37°C. Denatured D N A was precipitated by adding 0.1 volume sodium acetate and 2 volumes of ethanol and transferred to -30°C for a minimum o f 15 min. Precipitated D N A was washed two times with 70% ethanol and resuspended in 7 ul d H 2 0 , 2 ul T7 Sequenase Reaction buffer and 1 ul (lOng) of primer. The forward primer had the sequence 5 ' A A A G T G C C A C C T G A C G T C and initiated at position 7,841 in the P M S I cosmid vector. The reverse primer had the sequence 5 ' C C G C T T A T C A T C G A T A A G and initiated at position 80 in the cosmid vector. Primers were allowed to anneal for 2 min at 65°C, after which the reaction was allowed to cool for 30 min at room temperature. The samples were briefly centrifuged and stored on ice. In the meantime, 2.5 ul o f each chain terminator (ddG, ddA, ddT, ddC) was aliquoted to wells in a multi-well dish. The labelling mix (7.5 \iM 7-deaza-dGTP, 7.5 (0.M dCTP, 7.5 | i M dTTP) was diluted 1:5 in d H 2 0 and the Sequenase was diluted 1:8 in Sequenase dilution buffer. To annealed D N A , 1 ul 0 .1M D T T , 2 ul diluted labelling mix, 0.5 ul [a 3 2 P ] - d A T P and 2 ul of diluted Sequenase was added. D N A was labelled at room temperature for 5 min. To terminate the reaction, 3.5 ul of the labelling reaction was added to each termination mix in the multi-well dish, which had been pre-warmed to 37°C. The reactions were kept at room temperature for an additional 15 min. Finally, 4 ul o f stop solution (95% formamide, 20 m M E D T A , 0.05% bromophenol blue, 0.05% 82 Chapter 2: Materials and Methods xylene cyanol) was added to each reaction followed by an incubation at 75 °C for 2 min. The reactions were then run on a 6% polyacrylamide gel (5.7 g acrylamide, 0.3 g bis-acrylamide, 48 g urea, 10ml 10X T B E buffer, 40 ml dH 2 0, 10% ammonium persulfate, 50p,l T E M E D ) . The gel was transferred to Whatman 3 M M paper and dryed for 1 h on a gel dryer at 70°C. The gel was placed in an X-ray cassette with Kodak X - O M A T A R film and kept at room temperature overnight. The following day, the X-ray fi lm was removed and processed. 2.14 CO-TRANSFECTION OF HSV COSMID CLONES Vero cells were plated at a density of 1-1.2 x 10 6 cells per dish in 60 mm dishes the day prior to transfection. Cosmids were first linearized by overnight digestion with Pme I. For each transfection, 5 ul of cosmid mix (which consisted of 250 ng/pl of each linearized cosmid) was combined with 95 pi of antibiotic-free O p t i M E M . This was then mixed with 100 p i of L ipo fec tAMINE stock solution (12 pi of L ipo fec tAMINE plus 88 pi of O p t i M E M ) and incubated at room temperature for 45 minutes. Prior to the addition of the transfection mixture, the cells were washed with antibiotic and serum-free O p t i M E M to remove any F B S present. The transfection mix was mixed thoroughly with 1.8 ml of O p t i M E M and added to the dish. The cells were incubated at 37°C with 5% C 0 2 for 2-4 h. Cells were washed once with D M E M to remove all L ipo fec tAMINE and then overlaid with 3 ml of 5% F B S D M E M and placed at 34°C with 5% C 0 2 . Three to four days post-transfection, plates were examined for evidence of C P E . 83 Chapter 2: Materials and Methods 2.15 CONSTRUCTION OF HSV2-BAC 2.15.1 Production of BAC Virus The day prior to transfection, 60mm dishes were seeded with l x l O 6 Vero cells per dish. p B A C - T K was linearized by digestion with Hind III overnight. For transfection, l u g of linearized p B A C - T K and 4 ug of H S V - 2 viral D N A were mixed together. To this D N A mix was added 100 ul of d ^ O and 200 u l of serum-free, antibiotic-free Opti-M E M . In a separate tube, IOJLLI L ipo fec tAMINE and 300ul O p t i - M E M were combined. The D N A and L ipo fec tAMINE solutions were mixed together and allowed to sit at room temperature for 15 min. Following this incubation, 2.4 ml of O p t i - M E M was added to the mixture, which was then overlaid on cells which had been previously rinsed with O p t i - M E M . The cells were incubated with the transfection mixture for 18 h at 37°C, after which the mixture was removed and replaced by D M E M / 1 0 % F B S . When cells showed C P E , cells were scraped into sterile tubes and the mixture was frozen and thawed to release virus. The virus titer was determined on fresh Vero cell monolayers. T K -deficient virus were identified by infecting dishes of cells at an M.O. I , o f 0.01 and adding 100 m M A C V . A C V resistant plaques arising within several days are tk-deficient. To concentrate tk-deficient virus, viral infection was allowed to continue until the cell monolayer was completely infected. The virus was then harvested. 2.15.2 Confirmation of pBAC-TK Sequences in HSV2-BAC Vira l D N A was purified from the ACV-resistant viral stock. To confirm that ACV-resistance was due to incorporation of p B A C - T K into the H S V - 2 tk locus, P C R was performed to amplify cm sequences. The primers used were 5 ' A G G C C G G A T A G C T T G T G C and 5' C G G A A C A G A G A G C G T C A C A . P C R reactions were performed in a 50 | i l volume and contained the following: 5 (0.110X P C R 84 Chapter 2: Materials and Methods buffer, 1.5 u l l O m M dNTP mixture, 1.5 u l 50mM M g C l 2 , 2 u l primer mix ( l O u M each) 10 u l viral D N A , 5ul D M S O and 24.5 u l d H 2 0 . A s a control reaction, 1 ug p B A C - T K D N A was used as template D N A . The tubes were incubated in a thermocycler at 94°C for 5 minutes to completely denature the template D N A . Midway during this incubation, 0.5 j i l Taq D N A polymerase was added to the tubes. The P C R amplification was then allowed to proceed for 25 cycles which comprised the following steps: denaturation at 94°C for 1 min, primer annealing at 55°C for 1 min and extension at 72°C for 1 min. Tubes were incubated for an additional 10 min at 72°C and then maintained at 4°C. The amplification products were analyzed by gel electrophoresis. 2.15.3 Plaque purification of ACV-resistant HSV2-BAC To plaque purify ACV-resistant H S V 2 - B A C , the ACV-resistant H S V 2 - B A C stock was serially diluted and used to infect confluent monolayers of Vero cells grown in 6-well dishes. After a 1 h adsorption period at 37°C, the virus was removed and the monolayers were washed with P B S . To facilitate plaque formation, monolayers were overlaid with 0.8% agarose in D M E M / 4 % F B S containing 100 m M A C V . Approximately 3 days post-infection, plaques were picked and agarose plugs containing virus were used to infect Vero cells grown in 96-well dishes. Once again, the infection was performed in D M E M / 4 % F B S supplemented with lOOmM A C V . Vi ra l isolates were harvested when C P E was evident and analyzed by P C R for virus containing B A C sequences (Section 2.15.2). 2.15.4 Isolation of HSV-2 recombinant as a BAC plasmid To isolate the H S V - 2 recombinant as a B A C plasmid in E.coli, 10 cm dishes of confluent Vero cells were infected with H S V 2 - B A C isolates at an M.O . I , o f 3. Two hours later, the medium was removed and 1 ml of D N A z o l was added per 10 cm plate. The cells were lysed by gentle pipetting and the resulting D N A precipitated from the cell 85 Chapter 2: Materials and Methods lysate by adding 0.5 ml ethanol per ml of D N A z o l . Samples were gently mixed and centrifuged for 2 min at 12,000 x g. The supernatant was removed and the D N A pellet was washed with 70% ethanol. After centrifugation, the pellet was allowed to air dry and the D N A resuspended in 100 u l of T E (10 m M Tr is -Cl , 1 m M E D T A [pH8.0]). One ul of purified D N A was added to an equal volume of d H 2 0 and electroporated (2.5 k V , 200Q) into 40 u i electrocompetent D H 1 0 B E.coli. After electroporation, 1 ml L B media was added to the cells and the mixture was transferred to a conical tube. The tube was shaken at 200 rpm at 37°C for 1 h, after which the cells were plated onto L B plates containing 20ug/ml chloramphenicol. The plates were incubated overnight at 37°C to allow bacterial colonies to grow. 86 Chapter 3: Introduction CHAPTER 3: DEXTRAN SULFATE CAN ACT AS AN ARTIFICIAL RECEPTOR TO MEDIATE HSV-1 INFECTION 3.0 INTRODUCTION To investigate the H S V entry pathway, our laboratory isolated and characterized H S V resistant mouse cell lines (Chapter 1, Table 2). Parental L cells, which express both H S and C S G A G s on their cell surface, were infected with H S V - 1 at a low M O I (Gruenheid et al, 1993). Cells which survived to form colonies were harvested and plated. In this manner, the gro2C cell line was established. Gro2C cells are 90% resistant to infection compared to parental L cells. H P L C analysis demonstrated that gro2C cells synthesized chondroitin sulfate as efficiently as control L cells but generated little, i f any, heparan sulfate, a known receptor for H S V attachment. Despite the lack of HS moieties, gro2C cells are still partially susceptible to H S V infection, and further analysis indicated that CS can act as receptor for viral binding in the absence of HS . Sog9 cells were sequentially isolated from gro2C cells and exhibited a three-order-of-magnitude reduction in susceptibility to H S V - 1 compared to L cells. This resistance to infection is likely due to the fact that sog9 cells do not synthesize any cell surface HS or CS G A G s (Banfield et ai, 1995a). Sog9 cells however, retain some susceptibility to infection, providing evidence for a GAG-independent entry pathway. Taken together, these studies demonstrated that HSV-1 need not interact specifically with H S to gain entry into mouse fibroblasts, although the H S V - H S interaction significantly enhances infection efficiency. To characterize furhter the mutant cell lines, Banfield et al. (1995a) compared the inhibitory effects of sulfated polyanions on H S V - 1 infection. Parental L cells were infected with HSV-1 in the presence of soluble HS to test i f virus used cellular HS as a receptor for viral binding. Soluble H S inhibited L cell infection, demonstrating that it could act as a competitive inhibitor of virus attachment to cell surface H S . Due to its high negative charge density, soluble HS also inhibited infection of gro2C cells by competing 87 Chapter 3: Introduction with viral attachment to CS GAGs. Sog9 cells, however, do not express any GAG moieties, and thus, it was not surprising that HSV-1 infection of sog9 cells was resistant to inhibition by soluble HS. To assess whether the inhibitory effects of HS on L and gro2C cell infection were specific or ionic in nature, cells were infected with HSV-1 in the presence of the highly sulfate GAG analog dextran sulfate. DS inhibited infection of L cells, suggesting that some of the inhibitory effect of HS on L cell infection was, in part, due to electrostatic forces. Gro2C cell infection was neither enhanced nor inhibited in the presence of DS. Surprisingly, however, DS stimulated HSV-1 infection of sog9 cells by up to 25-fold. Moreover, this effect was type-specific, in that HSV-2 infection of sog9 cells was not enhanced by the addition of DS. I decided to investigate further the effects of dextran sulfate on sog9 cell infection for two reasons. First, understanding how DS could enhance viral infection on a GAG-deficient cell line might provide further insight into the initial interactions of the virus with cell surface molecules. Second, little is known of the differences between HSV-1 and HSV-2 and their interactions with cell surface GAGs. Differences in the interactions of the HSV serotypes with DS might reflect variations in the manner in which these viruses normally recognize and attach to GAGs on the host cell. This chapter summarizes experiments that were conducted to determine the mechanism by which DS enhances HSV-1 infection. 3.1 RESULTS 3.1.1 Effect of sulfated polyanions on HSV infection of L and sog9 cells To expand on the earlier findings of Banfield et al. (1995a), monolayers of L and sog9 cells were infected with HSV-1 in the presence of increasing concentrations of either soluble heparin or DS (0-300 ug/ml). HSV-1 infection of control L cells was sensitive to inhibition by soluble heparin, as 30 ug heparin/ml reduced L cell infection by 90% (Fig. 3.1). By contrast, HSV-1 infection of sog9 cells was only decreased by 28% at the highest 88 Chapter 3: Results Figure 3.1. Effect of soluble heparin on HSV-1(F) plaque formation. Monolayers of L and sog9 cells were inoculated with HSV-1 (F) diluted in various concentrations of heparin. After a 1 h adsorption period, the virus was removed and medium containing 0.1% pooled human IgG was added to the monolayer to allow for plaque formation. Data are expressed as percents of the number of plaques that form in the absence o f heparin. Values represent the average ± range from two independent experiments. 89 Chapter 3: Results concentration tested. Next, L and sog9 cells were infected in the presence of DS (Fig 3.2 A and B). DS was a strong inhibitor of HSV-1 infection of L cells, since as little as 3 ug/ml DS could reduce infection by 70%. By contrast, low concentrations of DS stimulated HSV-1 infection on sog9 cells by as much as 25-fold. This effect was dose-dependent as increasing DS concentration reduced the level to which infection was enhanced. This effect was likely caused by the propensity of soluble DS to bind virus in solution. Nevertheless, infection was still enhanced using 300 ug/ml DS, the highest concentration tested. It is worthwhile to note that in Figure 3.2, only the results from a single determination are shown for DS inhibition of L cell infection, despite the fact that this experiment was repeated three to four times with similar results. Other data throughout this thesis have been presented in a similar fashion. The reason for this is that different stocks of virus and/or polyanion were used in some of these experiments, introducing variation in the absolute number of plaques formed. In these instances, it was not possible nor appropriate to average the data from different experiments. Despite this variation in absolute plaque number, however, the overall results for different experiments were the same. Therefore, the experiments presented are representative experiments. To confirm that DS stimulation was type-specific, plaque assays were done with HSV-2 (G). DS did not significantly enhance HSV-2 infection (less than four-fold) (Fig. 3.3). These results are consistent with the earlier findings of Banfield et al. (1995a) and confirmed that DS promoted infection of sog9 cells in a manner that was dependent on both the virus strain and the cell phenotype. 3.1.2 Characterization of the Interaction of Dextran Sulfate with sog9 Cells There were several possibilities as to how dextran sulfate could enhance HSV-1 infection: (a) DS could act to tether the virus at the cell surface, (b) it could enhance viral penetration and/or (c) it might impinge on a post-entry step. It was reasoned that if (a), 9 0 Chapter 3: Results Figure 3.2. Effect of soluble dextran sulfate on HSV-1 infection. (A) Effect of soluble dextran sulfate on H S V - 1 (F) plaque formation on control L cells. (B) Effect of soluble dextran sulfate on infection of sog9 cells (and for comparison, control L cells). It should be noted that the scales for each graph are different. For both experiments, monolayers of cells were inoculated with HSV-1 (F) was diluted in various concentrations of dextran sulfate. After a 1 h adsorption period, the virus was removed and medium containing 0.1% pooled human IgG was added. Data are expressed as percents of the plaque formation that occurs in the absence of DS . Three to four independent experiments were performed with similar results. Where error bars are shown, the value represents the average ± range from two independent experiments. Where no error bars are shown, the value represents the results from a single experiment. 91 Chapter 3: Results 92 Chapter 3: Results Figure 3.3 Effect of soluble D S on herpes simplex virus infection. Fold stimulation of plaque formation observed when sog9 cell monolayers were inoculated with either HSV-1(F) or HSV-2(G) diluted in media containing various concentrations of D S . After a 1 h adsorption period at 37°C, the inoculum was removed and medium containing 0.1% pooled human IgG was added to the monolayers. Plaques were counted after three days. Each datum point represents the average o f two determinations which did not vary by more than 4-fold at any point. Results are ratios of plaque formation on DS-treated monolayers to plaques formed on untreated controls. Values represent the average ± range from two independent experiments. 93 Chapter 3: Results DS tethered the virus to the cell, then DS would likely interact at some level with the sog9 cell surface. To test this possibility, sog9 cell monolayers were incubated with D S for up to 60 min, rinsed, and incubated with HSV-1 for an additional 60 min. A s shown in Figure 3.4 , H S V - 1 infection was stimulated by more than 25-fold after only a few minutes of pre-incubation. Longer incubations up to 1 h did not significantly increase infection efficiency. D S could likely interact with sog9 cells, in part, because of their lack of cell surface G A G s . To test this, sog9-EXTl cells, which express cell surface H S G A G s , were pre-treated with 30 ug/ml DS for 10 min (Fig. 3.5). The monolayers were then rinsed and infected with virus. HSV-1 infection of sog9 -EXTl cells was clearly inhibited by the addition of DS and in a manner similar to that observed with parental L cells. To establish optimum conditions for enhancing HSV-1 infection of sog9 cells, confluent monolayers of cells were treated with various concentrations of D S either before or during inoculation with HSV-1 (Fig. 3.6). Maximum enhancement (35-fold stimulation) was achieved by pre-incubating cell monolayers with low concentrations of DS . A s expected, less stimulation was achieved with simultaneous DS treatment. Experiments also determined that there was a size dependency for D S enhancement of HSV-1 infection (Fig. 3.7). M W 5,000 DS and M W 15,000 DS were not active (less than two-fold) in stimulating H S V - 1 infection, whereas up to 14-fold stimulation was observed with M W 50,000 D S . M W 500,000 D S possessed the most stimulatory activity (up to 35-fold). It appears from these data that relatively long-chain DS is required for efficient stimulation of HSV-1 infection. To assess the stability of DS on the sog9 cell surface, cell monolayers were treated with two different concentrations of D S for 10 min, incubated in D M E M for various times (pre-treatment window), and then infected with HSV-1 (Fig. 3.8) With 10 ug/ml of D S , stimulation declined rapidly as the length of the pre-treatment window was increased, and by 2 h, stimulation was reduced to two-fold over that of controls. However, D S stimulation persisted for up to 60 min when 100 ug of DS/ml was used. A t this 94 Chapter 3: Results OH ' ' ' 1 ' 1 ' 1 • . • 1 • 1 0 10 20 30 40 50 60 70 DS Pre-incubation (min) Figure 3.4. Effect o f soluble D S on H S V - 1 (F) infection of sog9 cells. Sog9 cell monolayers were incubated with lOug /ml D S in D M E M at 37°C for various lengths of time prior to inoculation with virus. Following removal of DS , the cells were inoculated with H S V - 1 and incubated for 1 h at 37°C. Virus was removed, and D M E M containing 0.1% pooled human IgG was added to the monolayers. Plaques were counted after three days. Each datum point represents the average o f two determinations which did not vary by more than approximately 5-fold at any point. Results are ratios of plaque formation on DS-treated monolayers to that on control monolayers not exposed to D S . Two independent experiments were performed with similar results. The results from a single determination are shown. 95 Chapter 3: Results Figure 3.5. Effect of dextran sulfate on HSV-1(F) infection of s o g 9 - E X T l cells and L cells. Monolayers of L and s o g 9 - E X T l cells were incubated with various concentrations of D S for 10 min at 37°C. Following removal of DS , the cells were inoculated with H S V - l ( F ) for 1 h at 37°C. Virus was removed, and D M E M containing 0.1% pooled human IgG was added to the monolayers. Plaques were counted after three days. Two independent experiments were performed with similar results. The results from a single determination are shown. 96 Chapter 3: Results O (/) = Q 0) i O 3 - U . •I? U 3 simultaneous " • — pre-incubation 1 2 Log DS Added (ng/ml) Figure 3.6. Effect of soluble DS on HSV-1 plaque formation. sog9 cell monolayers were either (i) incubated with DS in D M E M for 10 min at 37°C prior to infection or (ii) inoculated with virus in the presence of various concentrations of D S . Equivalent concentrations of HSV-1 were used for both experiments. After a l h adsorption period at 37°C, the inoculum was removed and medium containing 0.1% pooled human IgG was added to the monolayers. Plaques were counted after three days. Results are the ratios of plaques formed on DS-treats monolayers to plaques formed on untreated controls. Values represent the average ± the range from two determinations. 97 Chapter 3: Results 20 OH • 1 • 1 . 1 1 2 3 Log DS Concentration (Mg/ml) Figure 3.7. Size requirement for dextran sulfate. D S of different molecular weights was diluted in D M E M and then added to monolayers of sog9 cells for 10 min at 37°C prior to infection. D S was removed and H S V - 1 (F) was added for l h at 37°C. The inoculum was removed and medium containing 0.1% pooled human IgG was added to the monolayers. The results from a single determination are shown. Data are ratios of plaque formation on DS-treated monolayers to that on control monolayers not exposed to D S . Two independent experiments were performed with similar results. The results from a single experiment are shown. 98 Chapter 3: Results Pretreatment Window (min prior to infection) Figure 3.8. Effect of pretreating sog9 cells with DS on HSV-1(F) plaque formation. Monolayers of sog9 cells were incubated with 10 or lOOpg of DS/ml in D M E M at 37°C for 10 min at different times prior to inoculation with HSV-1 (pretreatment window). The monolayers were rinsed three times with P B S after treatment with D S , following which the cells were either (i) incubated with D M E M at 37°C until infection or (ii) infected immediately after a lOmin pretreatment with DS in D M E M . The cells were rinsed once with P B S before a l h adsorption period at 37°C, and the inoculum was removed and replaced with D M E M containing 0.1% pooled human IgG. Plaque numbers were determined after 3 days. Data are ratios of plaque formation on DS-treated monolayers to that on control monolayers not exposed to DS . The results shown are averages± the ranges from two independent experiments. 99 Chapter 3: Results concentration, maximal stimulation was achieved when the cells were treated 30 min prior to infection. It was thus concluded that DS rapidly adsorbed to cells in active form and was then inactivated by either dissociation, degradation, or both. To determine whether DS was toxic to cells, which could account for the loss of D S stimulation when cells were treated hours in advance of infection, cells incubated with DS 3 h prior to infection were treated a second time just before inoculation. DS stimulation was normal in these cells, which showed that D S pretreatment did not reduce cell viability or susceptibility to infection (data not shown). 3.1.3 Analysis of Viral Attachment in the Presence of Dextran Sulfate On the basis of these results, it appeared that D S stimulated infection by binding stably to the host cell, where it acted as a matrix for subsequent virus adsorption and entry. This possibility was tested directly by experiments in which inoculation of cell monolayers with H S V - 1 preceded DS treatment (Table 4, experiments A , B , and C) or in which cells were incubated with DS before inoculation (Table 4, experiment D). These assays took advantage of the fact that viral adsorption can be experimentally separated from penetration simply by keeping the temperature of incubation at 4°C. The temperature can then be shifted to 37°C to allow adsorbed virus to penetrate into cells. These experiments demonstrated that DS did not affect virus already bound to the cell surface (Table 4, experiment C) , nor did it influence the internalization of bound virus (Table 4, experiment A ) . Experiment B (Table 4) showed that D S did not influence virus once it had been internalized, eliminating the possibility that DS impinged on a post-entry step. Only in experiment D (Table 4), in which cells were incubated with D S at 4°C prior to inoculation, was infection stimulated. From these results, it can be concluded that DS stimulates infection only when it is present prior to or during viral inoculation and that this activity is energy independent. 100 Chapter 3: Results d +1 Cd TJ CD e 0> + •4—» ed • O 3 O c 1 <D co C 2 TJ Q 3-a TJ TJ I t co G t' o d +1 00 m t u co G s o , p U 101 +1 2 5 0 o + ed 3 O c CD CO 0< • 3 3 Q tD CO G o CD O TJ CD en -t-t G U _ 2 c td a <D CD TJ CD CO 1/1 a CM X CD P 3 CO CD TJ 3 CJ CD CJ .a G TJ CD TJ CD O CD CM X <D *ed O G O o 3 TJ G O o a CD £ CO G O cd o 2 CD CJ O -4—» o CO TJ cd TJ CD .2 < Q ^ c <2 -a cd J 3 oo Q cd £> 3 O G •a "§ -2 1 3 £ CO cd fe > 00 U TJ U G CD s CD CJ Q .g E E £ CD CM X CD >-O fe 1 I •c * x ep CD OH 00 Q CD O c CD C/3 •9 CD J3 TJ <D C 'ed JO o TJ co fe w r= *=; .O <-! cd is .a « o . o TJ § CM £ o cj 00 Q * '1 ° Cd JD 3 O ,o ij -2 CD CJ G ej CO ii - H CL, CD JS TJ CD G •a •4—» £> O C 3 ed © c/3 ! J l> ed ed c/3 •§ .§ .a M-I a ° CD 73 ^ o 2 8 bfl G o C CD ' o CD G O cj O O CD JS 3 G CD CM X CD G CD TJ G CD CM CD TJ G bfi o •2 £ <D 3 a* ed CD M CM « ^ <>i >• J3 a +i CD bfi ed CD > ed CD JS G CD C/3 a CM a CO CD 3 >Chapter 3: Results These results are consistent with previous findings demonstrating that D S stimulated adsorption of radiolabelled HSV-1 to sog 9 cells at 4°C (Banfield et al., 1995a). Banfield and colleagues (1995a) showed that D S could enhance adsorption to sog9 cells by nearly 5-fold, but exerted little effect on virus adsorption to control L cells and gro2C cells. I wanted to demonstrate definitively that HSV-1 could bind D S . To do this, radiolabelled H S V was incubated with DS immobilized on microtiter wells (Fig.3.9). A s a control for adventitious binding, virus was incubated with wells coated with B S A . In these experiments, approximately 20% of input H S V - 1 bound to D S , which was substantially more than that observed for the B S A control wells. B y contrast, less than 5% of input H S V - 2 bound to DS . These data are consistent with a model in which H S V -1 adsorbs more efficiently than H S V - 2 to sog9 cells in the presence of D S on the cell surface. This could account in part for the results showing that only H S V - 1 infection is stimulated by D S . If D S was tethering HSV-1 to the cell surface, then it could possibly accelerate the interaction of virus with a secondary, saturable receptor. To test this, sog9 cell monolayers were inoculated with HSV-1 in the presence or absence of D S for up to 6 h (Fig. 3.10). In the absence of DS, virus infection continued to increase during the 6 h incubation, which suggested that saturation of cell surface receptors had not been achieved. To control for degradation and/or dissociation of DS during this incubation, an experiment was performed in which fresh virus mixed with D S was added every 2 h throughout the 6-h time course (data not shown). The results from the two experiments were indistinguishable. A n interpretation of these results is that in the absence of the primary cell surface gylcosaminoglycan receptor, HSV-1 could not efficiently engage a downstream receptor. In the presence of DS , however, saturation was reached by 1 h of incubation with virus. Moreover, by 6 h, there was only a 3 +/- 0.5-fold difference between DS-stimulated and control infections. These data are consistent with a model in which DS functions to tether HSV-1 to the cell surface, thereby facilitating its interaction with a downstream receptor. 102 Chapter 3: Results T3 -a T J < 3 T J C O CO w 3 "a (D "O T J < w 3 T J c 3 O CQ w 3 20 h Virus Added (CPMX10 3 ) 20 40 60 Virus Added (CPMX10 4 ) 80 Figure 3.9. Binding of HSV-1 (A) and H S V - 2 (B) to DS . HSV-1 and H S V - 2 were labelled with [ 3 5S]methionine and purified by centrifugation through a sucrose gradient. Maxisorp 96-well dishes were coated with 5mg/ml DS and incubated with virus for 2 h at room temperature. Unbound material was removed by several washes with P B S . Bound virus was harvested with lysis buffer and transferred to scintillation vials for quantitation of radioactivity by liquid scintillation spectroscopy. Two independent experiments were performed with similar results. The results from a single experiment are shown. Open circles, binding to D S ; closed circles, binding to B S A control. 103 Chapter 3: Results Figure 3.10. Time course of HSV-1 infection of sog9 cells in the presence of D S . Sog9 cells were inoculated with HSV-1 previously mixed with lOug/ml D S and incubated at 37°C. A t the indicated times post-infection , the inoculum was removed and D M E M with pooled human IgG was added. Values represent the average ± range from two independent experiments. 104 Chapter 3: Results 3.1.4 Effect of Dextran Sulfate on Viral Penetration To investigate whether DS exerted any effect on the rate of viral penetration, which might be suggestive of an alternative entry pathway, several penetration assays were performed. In one experiment, virus was adsorbed to the sog9 cell surface in the presence or absence of 10 pg/ml D S at 37°C for 1 h to allow for virus adsorption and entry (Fig. 3.11 A ) . A t various times, monolayers were washed with citrate buffer (pH 3.0) to inactivate extracellular virions, and the resulting plaques were counted after 3 days. The adsorption-penetration rates for DS-treated and control cells were similar. In a variation on this assay, cells were pretreated with D S for 30 min at 4°C, rinsed, incubated with virus for 30 min at 4°C, and then incubated at 37°C (Fig. 3.11 B) . Monolayers were then treated with citrate buffer. Once again, there was no significant difference in virus penetration between DS-treated and untreated controls. On the basis of these results, it appeared that HSV-1 engaged the normal entry pathway in the presence of DS . Dextran sulfate, however, is capable of adsorbing specifically to several cell surface receptors, including scavenger receptors (Krieger, 1992; Krieger et al., 1993). It has also been shown to stimulate endocytosis in certain cell types (Thiele and Steinbach, 1994) and to block endocytosis by competing with other molecules for the endocytic machinery (Tokuda et al., 1993; X u et al., 1993; Greenspan and Gutman, 1994). L o w temperature blocks the movement of ligands between compartments of the endocytic pathway. To inhibit steps in endocytosis, sog9 cells were treated with 100 (ig/ml of DS at 4°C for 30 min. The cells were rinsed and then incubated with H S V - 1 at 4°C for 30 min. The incubation temperature was then shifted to either 15°C, to inhibit steps in endocytosis, or 37°C (control) for 30 min. The results indicated that H S V - 1 infection was stimulated by 20-fold at both temperatures (data not shown). These results suggest that DS-mediated infection occurs via the normal entry pathway. 105 Chapter 3: Results 120 "I ol ' ' ' a E o Q. 20 0 -I 1 0 20 40 60 80 Time (min) Figure 3.11. Rates of H S V - 1 penetration into sog9 cells in the presence and in the absence of D S . (A) Confluent monolayers in 35-mm-diameter dishes were inoculated with H S V - 1 (F) in the presence of 10|0,g/ml DS at 37°C for 1 h. The monolayers were washed at various times with citrate wash buffer (pH 3.0), and the resulting plaques were counted. (B) Sog9 cell monolayers were incubated with lOOpg/ml D S / m l in D M E M for 30 min at 4°C. The monolayers were rinsed three times with P B S , and then the inoculum was added. Following a 30-min adsorption period at 4°C, the cells were transferred to a 37°C environment. A t different times after the temperature shift, the monolayers were treated with citrate wash buffer. Two independent experiments were performed with similar results. Results from single experiments are shown. The results are P F U surviving citrate treatment expressed as a percentage of the number of plaques obtained after 60 min of infection, which for each experiment was taken as 100%. 106 Chapter 3: Results 3.1.5 Effect of Dextran Sulfate on HSV Infection of GAG-deficient sog8 Cells I wanted to test the hypothesis that DS could enhance HSV-1 infection of other cell lines that are deficient in cell surface GAGs. One cell line that was tested was the sog8 cell line (Fig.3.12). Similar to sog9 cells, sog8 cells were isolated from gro2C cells and do not express any cell surface GAGs. Due to additional defects in this cell line (Esford, 1999), HSV-1 is unable to form plaques on sog8 cells. To quantify infection of sog8 cells in the presence of DS, I used a mutant HSV-1, HSV-lAgC2-3, in which part of gC is deleted and replaced with the lacZ gene. HSV-1 glycoprotein C is not required for DS stimulation of sog9 cell infection (see Chapter 5), and thus this particular virus allowed the measurement of sog8 cell infection as a function of the number of lacZ-expressing cells. As shown in Figure 3.12, DS enhanced infection of sog8 cells by up to 16-fold. This result supported a model whereby DS could replace cell surface HS as a matrix for viral attachment. 3.2 DISCUSSION In this study, I investigated a type-specific phenotype in which sog9 cell infection by HSV-1, but not HSV-2, could be partially rescued by the addition of soluble dextran sulfate. The simplest model to account for DS-mediated HSV-1 infection is that DS binds to sog9 cells in a saturable, reversible manner and tethers the virion to the cell surface. These data, showing that cells can be treated with DS prior to infection, that DS stimulates infection at 4°C and that DS interacts with virions, all lend support to this model. Moreover, there was a size requirement for DS. This makes sense, since it is more likely that virus would collide with and subsequently attach to long-chain DS than to low M W DS. This interaction with DS then allows HSV-1, but not HSV-2, to infect 107 Chapter 3 .'Discussion 20 0> LL OH • 1 . , , 1 1 2 3 Log Dextran Sulfate Added (ug/ml) Figure 3.12 H S V - l A g C 2 - 3 infection of sog8 cells in the presence of D S . sog8 cell monolayers were incubated with 30|J,g/ml D S at 37°C for 10 min prior to infection. The D S was removed and the monolayers rinsed three times with P B S . Cells were incubated with virus for l h at 37°C, after which the inoculum was removed and medium was added to the monolayers. 24 h post-infection, cells were stained with 0.1% X-ga l in X-ga l buffer and incubated for 6 h at 37°C. Blue (infected) cells were counted. Two independent experiments were performed with similar results. Results from a single experiment are shown. 108 Chapter 3 .-Discussion cells more efficiently. If this model is correct, D S essentially functions as a substitute for cell surface heparan sulfate, albeit one that confers type-specific infection. Heparan sulfate is a ubiquitous cellular G A G that promotes efficient H S V adsorption to the host cell (WuDunn and Spear, 1989). Heparan sulfate may also serve other functions as well , such as promoting efficient fusion of the virus with the host cell (Shieh and Spear, 1994). Because the structure of D S does not resemble that of the highly complex heparan sulfate molecule, it is unlikely that DS could provide more than a simple tethering function during infection. This may account, in part, for the inability of DS to completely restore wild-type levels of infection in the sog9 cells. Moreover D S is not covalently linked to the host cell surface, which is likely to severely compromise its ability to stabilize HSV-1 virions that collide with the cell surface. The finding that D S could actually stimulate HSV-1 infection is highly significant. Stimulation o f infection requires that specific host-virus interactions be maintained. Indeed, this data indicates that DS mediates HSV-1 infection via the normal entry pathway. Time course assays of H S V infection in the presence of DS suggest that D S is capable of binding to HSV-1 and then "handing-off' the particle to its co-receptor, such that saturation of the second receptor is quickly achieved. This investigation into the mechanism of D S stimulation thus corroborates the paradigm of H S V entry, in which cellular G A G s serve as a matrix for viral binding, concentrating viral particles at the host cell surface and facilitating attachment to a downstream receptor. It is well established that D S normally inhibits infection of cells by enveloped viruses. D S has been shown to prevent fusion of influenza virus with erythrocytes (Krumbiegel et al., 1992) and the formation of syncytia in human T-cell lymphotropic virus type 1- infected cells (Ida et al., 1994). Infection by human cytomegalovirus (Neyts et al., 1992) and Sendai virus is also inhibited by this polyanion (Ohki et al., 1992). It was therefore surprising that HSV-1 infection of sog9 cells was markedly enhanced. These results also indicate that this effect was mediated specifically by D S ; H S V - 1 infection was not affected by soluble heparin. Earlier findings demonstrated that heparan 109 Chapter 3 .Discussion sulfate and chondroitin sulfate neither enhanced or inhibited HSV-1 infection on sog9 cells (Banfield et al., 1995a). This is interesting, because heparan sulfate is the natural cell surface receptor for the virus. Several cell types, including murine macrophages and endothelial cells, as well as C H O cells, express scavenger receptors capable of binding various polyanionic ligands (Krieger, 1992; Krieger et al., 1993). In particular, scavenger receptors present on the surface of murine macrophages have broad ligand binding specificity and can bind D S and fucoidan but not heparin or chondroitin sulfate. Although the ligands for these receptors are all polyanions, not all polyanions can function as ligands. It is possible, therefore, that sog9 cells posses a receptor with a binding specificity similar to that of macrophage scavenger receptors, and this may account for the observation that cells pretreated with D S retain enhanced susceptibility to HSV-1 even after the soluble D S is removed. If this is true, then presumably sog9 -EXTl and L cells also express this receptor. However, pre-treatment of L cells and sog9 -EXTl cells with D S did not enhance viral infection (plaque formation was inhibited). The stimulatory effects of D S may not detectable in cells that display heparan sulfate. To demonstrate definitively that DS could act as an artificial co-receptor to initiate viral infection, D S stimulation assays were performed using another GAG-deficient cell line, sog8. HSV-1 infection of sog8 cells line was enhanced by soluble dextran sulfate. This was not necessarily surprising, since the sog8 and sog9 cell lines were both derived from gro2C cells. During the course of the present investigation, however, another project was initiated in our laboratory which allowed us to test the D S stimulation model in an unrelated GAG-deficient cell type. Mouse muscle fibers exhibit a maturation-dependent loss of susceptibility to infection by H S V - 1 , such that H S V can infect newborn muscle fibers in vitro, but not those from older animals (Acsadi et al., 1994; Feero et al., 1997). Work by Sonia Yeung in our laboratory has demonstrated that heparan sulfate biosynthesis is down regulated during skeletal muscle maturation (Yeung et al., 1999). Newborn muscle fibers express significant amounts of heparan sulfate and chondroitin sulfate. B y contrast, G A G expression in mature myofibers is reduced. 110 Chapter 3 .'Discussion Further experiments demonstrated that mature fibers expressed secondary viral receptors but that in the absence of heparan sulfate, HSV-1 was unable to effectively bind to these receptors. In an attempt to overcome this block to infection, DS was added to cultured mature myofibers either prior to or during infection. In both experiments, infection was enhanced by the addition of dextran sulfate. In another experiment, adult mice were injected with HSV-1 in the tibialis anterior muscle, along with dextran sulfate. Remarkably, H S V infectivity was enhanced by DS , consistent with the in vitro results. This provides sound support for a model in which DS functions as a surrogate receptor to stabilize the virus at the cell surface, thereby allowing H S V to engage a downstream receptor. I l l Chapter 4: Introduction CHAPTER 4: USE OF NOVEL £AT7-EXPRESSING C E L L LINES TO CHARACTERIZE HSV-GLYCOSAMINOGLYCAN INTERACTIONS 4.0 I N T R O D U C T I O N The initial attachment of H S V to cell surface G A G s is a critical component of the H S V entry pathway; the exact nature of this interaction, however, is poorly understood. Several lines of evidence indicate that H S V gB and gC are able to interact with cell surface G A G s and thus, play an important role in viral adsorption. This interaction would appear to be partly ionic in nature, as demonstrated in this study whereby D S , a polyanion structurally distinct from H S , could act as a surrogate H S V receptor on G A G -deficient cell lines. Nevertheless, recent studies have indicated that certain structural features of cell surface H S may also be significant in mediating attachment (Feyzi et al, 1997; Herold et al. 1995, Lycke et al, 1991; Herold et al, 1996). To examine further H S V attachment to G A G s , HSV-1 infection of novel mouse cell lines expressing E X T 1 was investigated. S o g 9 - E X T l cells were generated by transfecting sog9 cells with a single c D N A clone encoding the gene EXT1 (McCormick et al, 1998). EXT1 expression restored wild-type levels of susceptibility to HSV-1 infection in sog9 cells. The rescue of HSV-1 infection conferred by EXT1 expression is due to the fact that EXT1 encodes a heparan sulfate polymerase, thereby restoring HS synthesis in sog9 cells. Moreover, the type of HS that is synthesized in EXT1 -expressing cells appears to be distinct from that produced in wild-type L cells. H P L C analysis demonstrated that L - E X T 1 and s o g 9 - E X T l cells synthesize a heparan sulfate G A G that is eluted at lower salt concentration. This suggests that H S produced by EXT I-expressing cells is either undersulfated and/or of shorter chain length relative to HS produced in control L cells. One objective of the work presented in this chapter, therefore, was to identify the specific features of H S that are important for viral binding and infection of EXT1 and non-EXTl cell lines. To do this, 112 Chapter 4: Introduction HSV-1 plaque formation on L cells, sog9 cells, L - E X T 1 cells and sog9 -EXTl cells was examined in the presence of chemically modified heparin compounds. The hypothesis underlying these experiments was that H S V - 1 infection would be inhibited by modified heparin compounds which resembled viral receptors on the cell surface. Modified heparin compounds which were no longer similar to viral G A G receptors would not inhibit HSV-1 infection. It was also desirable to determine i f HSV-1 attachment to these novel EXT-1 expressing cell lines employed the heparin-binding glycoproteins gB and gC . For this, a panel of H S V mutants bearing deletions in the heparin-binding domains of gB and gC was tested. B y doing so, the ability of these glycoproteins to functionally recognize a unique HS moiety on EXT1 cells could be assessed. The results obtained suggest that H S V recognizes similar HS features in EXT-1 and non-£AT7 cell lines and that in both instances, this recognition is primarily mediated by the heparin-binding domain of gC. 4.1 RESULTS 4.1.1 Effect of Sulfated Polyanions on HSV-1 Infection of Mouse EXT-1 Cells In a previous study, EXT-1 expression was shown to restore H S V - 1 infection of sog9 cells to wild-type levels (McCormick et al, 1998). We hypothesized that this was due to the ability of the virus to attach to HS G A G s on s o g 9 - E X T l cells. To test this, HSV-1 infection assays were performed on sog9-EXTl cells in the presence of soluble heparin (Fig. 4.1). HSV-1 infection of L cells, L - E X T 1 cells and sog9 -EXTl cells was effectively inhibited, indicating that heparin could act as a competitor for H S V attachment to cellular G A G s . To substantiate these results, HSV-1 infection assays were performed once again, only this time using soluble DS (Fig. 4.2). Under these conditions, sog9-E X T 1 cells behaved like control L cells and L - E X T 1 cells in that D S inhibited H S V - 1 infection. A s observed previously, DS stimulated HSV-1 infection of sog9 cells. Taken together, these results indicate that sog9-EXTl cells resemble control L cells with regard 113 Chapter 4: Results Native Bovine Heparin Added (|jg/ml) Figure 4.1. Effect of soluble native bovine heparin on HSV-1(F) infection o f sog9, sog9-E X T - 1 , L and L - E X T 1 cells. Cel l monolayers were inoculated with H S V - 1 diluted in media containing various concentrations of unmodified heparin. After a 1 h adsorption period at 37°C, virus was removed and media containing 0.1% pooled human IgG was added to the cells. Plaques were counted after three days. Results are the number of plaques formed in the presence of heparin expressed as percentages of plaque formation in the absence of heparin. The value shown is the average ± range from two independent experiments. 114 Chapter 4: Results Figure 4.2. Effect of D S on HSV-1 infection of mouse E X T - 1 cell lines. Graphs A and B show experiment plotted two different ways in order to illustrate both the stimulatory (B) and inhibitory (A) effects of DS on HSV-1 infection. Cel l monolayers were inoculated with HSV-1 (F) in the presence of various concentrations of soluble DS . Following a 1-h adsorption period at 37°C, virus was removed and D M E M containing 0.1% pooled human IgG was added. Plaques were counted after three days. Two individual experiments were performed with similar results. Data shown are from single experiments and are expressed as percentages of the infection that occurs in the absence of dextran sulfate. In a variation of this experiment, cell monolayers were pretreated with DS prior to infection. The data obtained were similar to the results shown. 115 Chapter 4: Results Chapter 4: Results to H S V - 1 infection. Moreover, it would appear that EXT1 expression radically alters the cell surface architecture of sog9 cells, such that sog9-EXTl cells express cell surface G A G s to which H S V - 1 can functionally attach. 4.1.2 Analysis of HSV-1 Infection of Mouse EXT1 Cells in the Presence of Modified Heparin Compounds Both heparin and heparan sulfate display a number of chemical groups that may potentially contribute to viral attachment. Several lines of evidence indicate that both N -and O - sulfate groups are key determinants of viral binding to heparan sulfate (Lycke et al, 1991; Sheih et al, 1992; and Herold et al, 1994). In an attempt to define the importance of specific HS sulfate groups synthesized by control L cells and EXT1-expressing cell lines, H S V infection assays were performed on sog9, s o g 9 - E X T l , L - E X T 1 and L cells in the presence of chemically desulfated heparins. To determine the relative contributions of 6-0 and 2-0 sulfations to viral infection, the effects of 6-0 desulfated, 2 -0 desulfated, and 2, 6-0 desulfated heparin on the inhibition of viral plaque formation were compared (Fig. 4.3). None of the aforementioned modified heparin compounds had any significant effect on infection of sog9 cells. However, both 6-0-desulfated and 2 -0 desulfated heparin inhibited H S V - 1 (F) infection on L , L - E X T - 1 and sog9-EXTl cells, albeit less efficiently than unmodified heparin (Fig. 4.3 A and B) . A t the highest dose tested, plaque formation on s o g 9 - E X T l cells in the presence either 6-0 or 2-0 desulfated heparin was reduced by approximately 30%. Infection of L and L - E X T 1 cells was somewhat more sensitive to the antiviral effects of these compounds, as plaque formation was inhibited by up to 50% and 60%, respectively. Removal of both 2-0 and 6-0 sulfate groups, however, abrogated the ability of heparin to act as a competitor for viral attachment, since 2,6-0 desulfated heparin had little inhibitory effect on HSV-1 infection of either L , L E X T 1 or sog9 -EXTl cells (Fig. 4.3 C) . Taken together, these results suggest that the H S chains expressed on 117 Chapter 4: Results Figure 4.3. Effects of O-desulfated heparin on HSV-1 infection of sog9, s o g 9 - E X T l , L and L - E X T 1 cells. Cel l monolayers were infected with HSV-1 diluted in various concentrations of either (A)2-0-desulfated heparin, (B) 6-O-desulfated heparin or (C)2,6-O-desulfated heparin. Cell monolayers were inoculated with HSV-1 diluted in media containing various concentrations of the O-desulfated heparin compounds. After a 1-h adsorption period at 37°C, the inoculum was removed and media containing 0.1% pooled human IgG was added. Plaques were counted after 3 days. Results are plaques formed in the presence of O-desulfated heparin expressed as percentages of plaque formation in the absence of O-desulfated heparin. Values represent the average ± range from two independent experiments. 118 119 Chapter 4: Results L , L - E X T 1 and s o g 9 - E X T l are rich in O-sulfate groups and that these moieties are important for virus-cell interactions. I next wanted to test the importance of N-sulfation in mediating H S V adsorption and infection. For this, a modified heparin compound in which all N-sulfate groups were replaced by the charge neutral acetyl group (N-desulfated/N-acetylated heparin) was used. Interestingly, HSV-1 infection on s o g 9 - E X T l , L and L - E X T 1 cells was inhibited as effectively by N-desulfated/N-acetylated heparin as it was by unmodified heparin (Fig. 4.4). This could imply that N-sulfate groups are not important for H S recognition by H S V - 1 . These results, however, are difficult to interpret, since previous studies have observed that reacetylation of heparin at N-desulfated sites partially restores antiviral activity to the compound (Herold et al, 1995; Feyzi et al. 1997). I was, therefore, unable to establish the role of N-sulfate groups in HSV-1 attachment using this particular heparin preparation. To examine the combined contributions of N - and O- sulfated regions in virus-cell interactions, N - ; 6-O-desulfated, N-;2-0-desulfated and N - ; 2-/6-0-desulfated heparin compounds were tested in HSV-1 infection assays (Fig 4.5). None of the three N - ; 0 -desulfated heparin compounds inhibited HSV-1 infection of L , L - E X T 1 cells or sog9-E X T 1 cells. Moreover, N;6-0-desulfated and N;2-0-desulfated heparin were less effective than their 6-O-desulfated and 2-0 desulfated counterparts at inhibiting viral infection (Fig.4.3). This supports the notion that N-sulfate groups are involved in viral-cell interactions. Taken together, these results demonstrate that both O- and N-sulfated regions of H S contribute to viral attachment. 120 Chapter 4: Results 120 0 20 40 60 80 100 120 N-Desulfated/N-Acetylated Heparin Added (ug/ml) Figure 4.4. Effect of N-desulfated/N-acetylated heparin on HSV-1(F) infection of sog9, s o g 9 - E X T l , L and L - E X T 1 cells. Cel l monolayers were inoculated with HSV-1 diluted in media containing various concentrations of N-desulfated /N-acetylated heparin. After a l h adsorption period at 37°C, the inoculum was removed and media containing 0.1% pooled human IgG was added. Plaques were counted after three days. Results are plaques formed in the presence of N-desulfated/N-acetylated heparin expressed as percentages of plaque formation in the absence of the modified heparin compound. Values are the averages ± the range from two independent experiments. 121 Chapter 4: Results Figure 4.5. Effects of N-/0-desulfated compounds on HSV-1(F) plaque formation on sog9, s o g 9 - E X T l , L and L - E X T 1 cells. Ce l l monolayers were inoculated with H S V - 1 diluted in media containing various concentrations of either (A) N-;6-0-desulfated heparin, (B) N-;2-0-desulfated heparin, or (C)N-;2-/6-0-desulfated heparin. After a l h adsoprtion period at 37°C, the inoculum was removed and media containing 0.1% pooled human IgG was added. Plaques were counted after 3 days. Results are plaques formed in the presence of N-/0-desulfated heparin expressed as percentages of plaque formation in the absence of N-/0-desulfated heparin. Values represent the average ± the range from two independent experiments. 122 Chapter 4: Results c o u c o (0 E cr (0 0 20 40 60 80 100 120 N-;6-0-Desulfated Heparin Added (ng/ml) B 0 10 20 30 40 N-;2-0-Desulfated Heparin Added (ug/ml) c o u c o re E cr ro 120 N-;2-/6-0-Desulfated Heparin Added (pg/ml) 123 Chapter 4: Results 4.1.3 Relative Contributions of HSV gB and gC to Infection of Host Cells From these results, it appeared that HSV-1 interacted with similar structural features of H S on both EXT I cells and control L cells. It was reasoned that these H S moieties were likely being recognized by H S V gB and /or gC, two glycoproteins previously shown to bind heparin. To test this hypothesis, plaque formation on these cell lines was compared using a panel of H S V gB and gC mutants that have reduced heparin-binding capabilities. Results from these experiments are shown in Table 5. To discern any differences between infection of EXT1 and non-£AT7 cells with the H S V mutants, plaque formation on each cell line was calculated relative to infectivity observed on sog9 cells. A s a control, wild-type H S V - 1 (KOS) was used, since al l the mutants were originally derived from this strain. In agreement with an earlier study on HSV-1 infection of s o g 9 - E X T l cells (McCormick et al., 1998), infection of s o g 9 - E X T l cells by wild-type HSV-1 was at a level similar to that observed for control L cells. To determine i f gC played a role in the infection of these cell lines, the gC-deficient virus H S V - l ( K O S ) A g C 2 - 3 , was tested. The relative infectivity of HSV-1 (KOS)AgC2-3 on s o g 9 - E X T l , L , and L - E X T 1 cells was substantially less than that of wild-type H S V -l ( K O S ) . This was similar to the results obtained using another gC-deficient H S V - 1 , K C Z . The relative infectivity of K C Z on sog9 -EXTl cell lines was lower than that observed for H S V - l ( K O S ) A g C 2 - 3 . The reason for this is not clear. Collectively, however, these observations demonstrate that gC facilitates HSV-1 infection of both control L cell and EXT I -expressing cell lines. The H S V - 1 mutant K g B p K - , in which a putative HS binding lysine-rich (pK) domain of gB is deleted (Laquerre et al., 1998b), was tested next. Significantly, the relative infectivity of K g B p K / on sog9 -EXTl and L cells was comparable to that of wi ld-type H S V - 1 . This implied that the heparin-binding domain of gB was not essential for mediating H S V - 1 infection. To extend this observation further, infection of the cell lines 124 Chapter 4: Results Table 5. Relative infectivity of HSV-1 heparin-binding mutants on E X T 1 cell lines. a Data shown is H S V plaque formation on s o g 9 - E X T l , L and L - E X T 1 cells relative to that observed on sog9 cells (=1.0) for each strain tested. For each virus strain, cell lines were inoculated with equivalent amounts of virus for l h at 37°C. Virus was removed and media containing IgG was added to facilitate plaque formation. Plaques were counted after three days. The results shown are from a single determination, which were similar to data obtained from two other experiments. ( N A ) Not available. * See Table3 (Chapter 2) for information regarding virus strains used in this experiment. 125 Chapter 4: Results OS u bO a, PQ bO oo O OO O U at. a, PQ Ml co oo CN 00 CN 03 U 9 U bo PQ bo CN co CN 00 OH PQ bO oo N U O •si-CO ^ c*"> J CN > r j K < co 00 co CN CN O IT) 3 CS H >o ON 60 O 00 o co X w I ON bO O 00 CO CN ON ON in v O 126 Chapter 4: Results using the double-mutant virus K g B p K " g C and the two repaired viruses, K g B p K R g C and K g B p K ' g C R , was compared. The relative infectivity K g B p K ' g C R was comparable to that of wild-type H S V - l ( K O S ) . B y contrast, the relative infectivities of K g B p K ' g C and K g B p K R g C on the various cell lines were similar to that observed with K C Z . Taken together, these results support a model whereby HSV-1 gC is the primary mediator of attachment to H S G A G s presented on EXT1 cells and control L cells. Moreover, the p K sequence of gB is not essential for infection of these cell lines. 4.2 DISCUSSION Heparan sulfate recognition by H S V is an important determinant of cell and tissue tropism. The interaction of HSV-1 with H S was studied by examining the ability of modified heparin compounds to inhibit viral infection of novel EXT I -expressing cell lines. Previous analysis in our laboratory indicated that the HS moiety expressed on EXT1 cell lines is distinct from that presented on control L cells (McCormick et al., 1998). A n advantage of using these cell lines, therefore, was that they allowed the comparison of the ability o f virus to use different HS structures as attachment receptors and to assess which features of H S were important for viral binding. EXT\ expression rescues HSV-1 infection of sog9 cells to wild-type levels (McCormick et al., 1998). We hypothesized that the rescue of HSV-1 infection conferred by EXT I expression was the result of the enhanced expression of H S G A G s in sog9-E X T 1 cells. In this chapter, it was shown that HSV-1 infection of s o g 9 - E X T l cells is inhibited by the addition of soluble heparin and dextran sulfate. The ability of these polyanions to inhibit HSV-1 infection is a clear indicator that HSV-1 can engage the cell surface H S presented on sog9 -EXTl cells and thereby gain entry into host cells. Having determined that HSV-1 could use HS on sog9-EXTl cells to mediate infection, I wanted to identify structural features of the molecule that were important for viral recognition. Previous studies have shown that in particular, sulfate groups are 127 Chapter 4: Discussion important for viral attachment. For example, cells expressing undersulfated heparan sulfate due to genetic mutation bind H S V poorly (Shieh et al, 1992). To clarify which sulfate groups were present on the H S G A G s on EXT1 and won-EXTl cell lines, and whether or not they were important for viral infection, the effects of various desulfated heparin compounds on H S V - 1 infection of the EXT1 cell lines was tested. 2 -0 and 6-0 desulfation of heparin reduced the antiviral activity of the compound on s o g 9 - E X T l , L -E X T 1 and control L cells. Interestingly, however, these compounds were less effective at inhibiting viral infection of sog9-EXTl cells than they were on L and L - E X T 1 cells. One interpretation of this result is that sog9 -EXTl cells express HS that has relatively more 2-O and 6-0 sulfate groups than L and L - E X T 1 cells. Infection of all the cell lines was, however, insensitive to the addition of 2,6-0 desulfated heparin. It can be concluded, therefore, that the H S G A G s on L , s o g 9 - E X T l and L - E X T 1 possess both 2-0- and 6-0 sulfate groups and that these are important for viral infection. This is in agreement with results from another study which demonstrated that 6-0-desulfation of heparin markedly reduced the antiviral activity for HSV-1 on Vero cells (Herold et al., 1995). Work by Feyzi et al. (1997) further demonstrated that the minimal requirements for HSV-1 binding to H S consisted of 10-12 monosaccharide units containing at least one 2-0 and one 6-0-sulfate group. In addition to O-sulfate groups, N-sulfate groups have also been shown to be important for binding of HSV-1 to H S (Feyzi et al., 1997). Cel l lines mutant in N -sulfotransferase display a reduced susceptibility to H S V infection (Shieh et al, 1992). Moreover, N-desulfation of heparin has been shown to abolish its antiviral activity for HSV-1 (KOS) (Herold et al. ,1995). The contribution of N-sulfate groups to viral binding was examined by testing N-desulfated/N-acetylated heparin in the H S V infectivity assay. H S V infection of all EXT1 and L cells was inhibited as effectively by N-desulfated/N-acetylated heparin as it was by unmodified heparin. This is contrary to a report by Feyzi et al. (1997), who showed that N-acetylation of N-desulfated sites reduced the inhibitory effect of heparin in a virus infectivity assay by 10-fold. It is difficult to explain this 128 Chapter 4: Discussion discrepancy. It should be noted, however, that Herold and colleagues (1995) observed that re-acetylation o f heparin at N-desulfated sites partially restored the antiviral activity of heparin. N-desulfation yields a positive charge ( N H + 3 ) in the molecule, whereas, N -desulfation, N-reacetylation produces a neutral charge. A s proposed by Herold et al. (1995), it may be that it is the absence of a positive charge at that site, rather than the presence of a negatively charged sulfate group, that is important for antiviral activity. Nonetheless, the observation that N-desulfated/N-acetylated heparin inhibits HSV-1 infection of the mouse cell lines is significant. The presence of a neutral group clearly allows the modified compound to retain enough structural similarity to compete with cell surface HS for viral binding. This demonstrates that not only are the charged sulfate groups of H S determinants of viral binding, but importantly, so is the actual backbone of the molecule. This is in agreement with Lycke and colleagues (1991) who demonstrated that HSV-1 could bind to a heparan sulfate preparation which contained 1.5 sulfate groups/disaccharide unit but not with a dermatan sulfate preparation which had a similar negative charge density (1.3 sulfate groups/disaccharide unit). The ability of H S V to recognize G A G s is a function of the two heparin-binding glycoproteins, gC and gB. Unti l recently, the role of gB in the attachment of virus to HS G A G s has been difficult to establish. Laquerre and colleagues (1997) constructed a mutant H S V - 1 particle deleted in the heparin-binding region of gB (KgBpK") and demonstrated that binding of this virus to cells was reduced 20% compared to wild-type virus. The binding capacity of a gC-deficient HSV-1 ( K C Z ) was reduced by 65% and that of a double gB/gC mutant ( K g B p K ' g C ) by 80%. These data led the authors to conclude that the heparin-binding region of gB contributes less to virus binding to H S than gC. The infectivity of these heparin-binding mutants was tested using our EXT1 -expressing cell lines. These results support those of Laquerre and colleagues (1997), in that the relative infectivity of H S V - 1 AgC2-3, K C Z and K g B p K ' g C " on L , L - E X T 1 and s o g 9 - E X T l cells was reduced compared to that of wild-type H S V - 1 . The relative infectivity of K g B p K " , on the otherhand, was similar to that of wild-type virus. Thus, H S V - 1 gC is likely the 129 Chapter 4: Discussion principle virion component mediating HSV-1 attachment to H S G A G s on EXT1 cells. Finally, it was interesting that the two gC-deficient viruses used in this study, K C Z and HSV-1 AgC2-3, had different relative infectivities on the cell lines tested; the relative infectivity of K C Z was noticeably lower than that of HSV-1 AgC2-3. It is difficult to account for this result. One possibility is that one of these viruses carries a secondary mutation in another viral glycoprotein and that this affects a step in viral infection. Work is underway in our laboratory to investigate this possibility. The initial aim of this study was to compare HSV-1 infection of novel EXT1-expressing cells with control L cells. These results indicate that HSV-1 infection of sog9-EXT1 cells resembles that of normal L cells. Infection of both EXT-1 and non-EXTl cells was facilitated by HSV-1 gC. Moreover, HSV-1 appears to recognize similar H S features in both cell types. This does not imply, however, that the H S moieties on EXT1 and control cells are identical, since H P L C data clearly suggests otherwise (McCormick et al., 1998). Rather, H S V - 1 has likely evolved the ability to recognize particular polysaccharide sequences which may be present in different types of H S G A G s . Given that H S expressed on the surface of cells may differ with respect to chain length, fine structure and amount produced, it would be of benefit for HSV-1 to be somewhat versatile in its ability to bind HS . H o w could the virus adapt to different HS types? Trybala and colleagues (1998) recently addressed this question by analyzing the interaction of P r V gC heparin-binding domain mutants with modified heparin preparations. Their results indicated that different regions of the gC heparin-binding domain may promote P rV binding to different structural features of H S . It remains to be determined whether the same holds true for the heparin-binding region of HSV-1 gC. It does appear, however, that HSV-1 gC and H S V - 2 gC have evolved to recognize different HS features. Herold and colleagues (1996) demonstrated that H S V - 1 and H S V - 2 differed in their susceptibility to modified heparin compounds. O-sulfate groups were determinants primarily for HSV-1 infection. These differences in susceptibility to modified heparin compounds mapped to gC. This suggests that 130 Chapter 4: Discussion differences in the interactions of HSV-1 and H S V - 2 with cell surface HS may influence tissue tropism. The studies reported here confirm that the specific interaction o f H S V - 1 with HS is important for an efficient infection of cells by the virus. 131 Chapter 5: Introduction CHAPTER 5: GLYCOPROTEIN B MEDIATES DEXTRAN SULFATE STIMULATION OF HSV-1 INFECTION 5.0 I N T R O D U C T I O N Several lines of evidence show that for both HSV-1 and H S V - 2 , the initial interaction with cells is binding of virus to cell surface HS. This, however, would appear to contradict observations that the two serotypes display differences in epidemiology, cell tropism and susceptibility to inhibitors of viral binding. These results, however, can be rationalized by the observations that i) the relative contributions of viral glycoproteins to viral binding are different (Herold et al., 1994), ii) the glycoproteins of HSV-1 and H S V - 2 recognize different structural features of heparan sulfate (Gerber et al., 1995) and iii) HSV-1 and H S V - 2 may use distinct secondary receptors to gain entry into the host cell (Geraghty et al. 1998). A t the time this work was initiated very little was understood about the different entry pathways used by HSV-1 and H S V - 2 . Work presented earlier in this thesis on H S V infection of sog9 cells suggested that a particular component of the HSV-1 particle could interact with the G A G analogue dextran sulfate, since H S V - 1 , but not H S V - 2 , infection of sog9 cells could be enhanced by the addition of the polyanion. Differences in the ability of HSV-1 and H S V - 2 to interact with DS could reflect differences in the adsorption of these viruses to G A G moieties displayed on the host cell surface. The aim of the work presented in this chapter, therefore, was to identify the viral component(s) that could mediate an interaction of HSV-1 with dextran sulfate. 132 Chapter 5: Results 5.1 RESULTS 5.1.1 Mapping the DS Activation Site on the Virus In an attempt to map the viral components that facilitate DS stimulation, sog9 cells were treated with DS either before or during inoculation with a panel of intertypic recombinants and deletion mutants (Table 6). Sog9 cell infection with several o f the glycoprotein mutants, including, UL10" (HSV-1 gM"), HSV-1 macroplaque (gC-), HSV-1 A N G (gB syri) and HSV-1 ANG-path (gB syn; base change at amino acid 84 in gD), was enhanced 5- to 14-fold in the presence of D S , a level of stimulation that is less than that of HSV-1 (F), but greater than that observed for H S V - 2 (G). These results suggest that: i) gM-1 and gC-1 are not essential for DS stimulation and ii) the HSV-1 particle can functionally interact with D S regardless of syn mutations in gB and a base change in gD, the latter alteration conferring neuroinvasive properties to ANG-path . It cannot be concluded, however, that these glycoproteins do not contribute to DS-mediated infection since stimulation by DS is greater when these glycoproteins are intact. The most significant finding in this series of experiments was that D S was unable to stimulate (less than two-fold) infection of the H S V - 1 intertypic recombinant RH1G13 , in which gB-1 is replaced by gB-2 (Table 6, Figure 5.1). This indicated that D S stimulation is mediated, at least in part, by gB-1. B y contrast, infection with RS1G25, an HSV-1 strain in which gC-1 is replaced by gC-2, was stimulated by D S , although by only five- to sevenfold, significantly lower than the level observed using HSV-1 (F). This implied a role for gC-1 in DS-mediated infection, which was somewhat surprising because H S V - l ( K O S ) A g C 2 - 3 , a gC-deficient HSV-1 strain was stimulated by up to 18-fold (Figure 5.1, Table 6). This is similar to the stimulation observed with wild-type H S V -1(F). One way to account for these results was the possibility that gC-2 is an inhibitor of D S stimulation. I tested this, using a gC-2-deficient strain, HSV-2(G)gC2", and found that it was stimulated by up to eight-fold when sog9 cells were pre-treated with DS . This r 133 Chapter 5: Results a (-), stimulation of the prototype H S V - 2 strain G , which varied from zero- to fourfold above that of controls in different experiments; (++), fold stimulation of the prototype HSV-1 strain F, which varied from 15- to 35-fold above that of controls in different experiments; (+), stimulation which lies intermediate to those of the two prototype strains. *HSV-1 background containing gC-2 in place of g C - l . Other strains of virus are designated in a similar manner. c D a t a published by Banfield et al. 1995, J. V i r o l 69:3290-3298 134 Chapter 5: Results Table 6. Effect of dextran sulfate on H S V infection Virus Strain Phenotype Stimulation by D S " HSV-1 (F) wi ld type ++ HSV-1 ( K O S ) c wild type ++ H S V - 2 (G) wi ld type -RSIG25 H S V - 1 ; gC-2* + RHIG13 H S V - 1 ; gB-2 -H S V - l ( K O S ) A g C 2 - 3 H S V - 1 ; gC-l-deficient ++ HSV-2(G)gC2- H S V - 2 ; gC-2-deficient + K C Z H S V - l ( K O S ) ; gC-deficient ++ K g B p K " H S V - l ( K O S ) ; gB H B D mutant ++ K g B p K g C H S V - l ( K O S ) ; gB H B D mutant; gC-deficient ++ R7015 H S V - 1 ; gD-2, gE-2, gG-2 ++ U L 1 0 - H S V - 1 ; gM-deficient + H S V - l ( M P ) H S V - 1 ; gC-deficient syncytial strain + H S V - 1 ( A N G ) wi ld type; gB syn mutation; gB fast rate of entry determinant (roe) + HSV-1 (ANG-path) HSV-1 A N G derivative; gB syn and roe mutations; neuroinvasive determinant at residue 84 in gD + 135 Chapter 5: Results Figure 5.1. Effect of soluble D S on herpes simplex virus infection. ( A and B) Fold stimulation of H S V plaque formation observed when sog9 cell monolayers were treated with D S prior to infection. Sog9 cells grown in 6-well dishes were incubated at 37°C for 10 min with D S diluted in D M E M . The wells were then rinsed three times with P B S , and the viral inoculum was added. After a 1-h adsorption period at 37°C, the virus was removed and medium containing 0.1% pooled human IgG was added to facilitate plaque formation. (C and D) Fold stimulation of H S V plaque formation when sog9 cell monolayers were inoculated with virus in the presence of various concentrations of D S . Virus was incubated with cells for 1-h at 37° C, after which the inoculum was removed. Medium containing IgG was added to facilitate plaque formation. Data are fold stimulation of plaque formation that occurs in the presence of DS . Values represent the average ± the range of two independent experiments. 136 Chapter 5: Results I i I 1 2 Log DS Added (ng/ml) Log DS Added (\ig/m\) 137 Chapter 5: Results Log DS Added (ug/ml) HSV-1 (F) HSV-2 (G) HSV-1 (KOS)AgC2-3 HSV-2(G)gC-1 2 Log DS Added (ug/ml) 138 Chapter 5: Results level of infection was substantially better than that for H S V - 2 . On the basis of these results, it can be concluded that gB-1, in part, facilitates D S stimulation and that gC-2 may be a weak inhibitor o f DS-mediated stimulation of infection. This property might account, in part, for the poor DS stimulation observed with RS1G25 and control H S V - 2 . 5.1.2 Contributions of gB-1 and gC-l Heparin-Binding Domains to the Interaction of HSV-1 with Dextran Sulfate The above findings suggest that the interactions of gB-1 and gB-2 with DS are quite distinct from one another. I therefore wanted to identify the domains of gB-1 that mediate D S stimulation of HSV-1 infection, since these regions could be important for interactions of HSV-1 and H S V - 2 with host cell G A G s . One candidate domain was the heparan sulfate (HS) binding domain in gB-1 (amino acids 68 to 76) since i) DS could possibly interact with the positively charged lysine residues found in this domain and ii) the HS-binding domain lies within the most divergent region between gB-1 and gB-2. To determine whether or not the HS binding domain of gB-1 was involved in DS-mediated H S V - 1 infection, the D S stimulation assay was performed using the virus K g B p K " , an HSV-1 virus in which only the gB H S binding domain (pK) has been removed. Laquerre and colleagues (1998b) demonstrated that the p K sequence of gB-1 was solely responsible for the H S binding function of gB. A s shown in Figure 5.2, infection of sog9 cells by K g B p K " was stimulated by as much as 37-fold in the presence of D S . This clearly indicates that the HS-binding domain of gB is not required for DS-mediated stimulation of infection. One possibility to account for this observation is that in the absence of the gB H S - binding domain, HSV-1 uses gC to interact with DS . This hypothesis was tested directly using the double HSV-1 mutant, H S V - l g B p K " g C in which both the H S binding domain of gB and the entire coding region of gC have been deleted. A s a control for the gC deletion, the gC-deficient HSV-1 virus K C Z was also tested. Infection of sog9 cells by 139 Chapter 5: Results Figure 5.2 Effect of D S on infection of sog9 cells by H S V - 1 heparin-binding mutants. (A) sog9 cell monolayers were inoculated with virus in the presence of various concentrations of D S . (B) sog9 cell monolayers were incubated with D S in D M E M for 10 min at 37°C and washed three times with P B S prior to inoculation with virus. Equivalent concentrations of virus were used for both experiments. After a 1-h adsorption period at 37°C, the inoculum was removed and medium containing 0.1% pooled human IgG was added to the monolayers. Plaques were counted after three days. The results are the ratios of plaques formed on DS-treated monolayers to plaques formed on untreated controls. Two independent experiments were performed with similar results. The results from single experiments are shown. 140 Chapter 5: Results Log Dextran Sulfate Added (ug/ml) 141 Chapter 5: Results K C Z and K g B p K R g C was stimulated by 35-fold and 20-fold, respectively. Taken together these results provide evidence that gC is not essential for viral interactions with DS. 5.1.3 Analysis of DS-mediated Infection Using HSV-1 gB Antigenic Variants The interaction of gB-1 with DS might be determined by not only the primary sequence of the molecule, but also by its secondary and tertiary structure. To test whether or not DS-mediated HSV-1 infection was sensitive to changes in the secondary structure of gB-1, two H S V - 1 gB monoclonal antibody resistant (mar) mutants, R126 and R233 were studied. These H S V variants were isolated on the basis of their ability to escape neutralization by various gB-specific monoclonal antibodies (Pellet et al., 1985). Mutant R126 contains three substitutions, i.e., Thr for Ser+283, and G i n for Arg+305 and A s n for Ser+443. Mutant R233 contains a single substitution of Thr+285. These substitutions alter the secondary structure of the protein at both epitopic sites, which are in exposed domains of the gB-1 molecule. The R126 and R233 viruses were either added to cells pre-treated with D S or diluted in D M E M containing D S and then used to inoculate sog9 cells. A s shown in Figure 5.3, R126 infection was stimulated by D S although less than that of HSV-1(F) in both experiments. One interpretation of these results could be that the site comprising these residues is involved at some level in recognizing DS . B y contrast, DS stimulation of R233 infection was near wild-type levels when virus and D S were added to cells simultaneously. However, R233 infection was somewhat lower than that of HSV-1(F) when sog9 cells were pre-treated with DS . Likewise to R126, the mutation in gB of R233 did not block the ability of D S to stimulate HSV-1 infection. However, because a slightly lower infection efficiency was observed using R233 in one of the experiments, the possibility that this region of gB (comprising residue 285) is involved in mediating interactions with D S cannot be ruled out. 142 Chapter 5: Results 30 Log Dextran Sulfate Added (ug/ml) Figure 5.3. Effect of DS on infection of sog9 cells by HSV-1 gB mar mutants. (A) sog9 cell monolayers were inoculated with virus in the presence of various concentrations of DS. (B) sog9 cell monolayers were incubated with D S in D M E M for 10 min at 37°C and washed three times with P B S prior to inoculation with virus. Equivalent concentrations o f virus were used for both experiments. After a 1 h adsorption period at 37°C, the inoculum was removed and medium containing 0.1% pooled human IgG was added to the monolayers. Plaques were counted after three days. The results are the ratios o f plaques formed on DS-treated monolayers to plaques formed on untreated controls. Values represent the average ± range from two independent experiments. 143 Chapter 5: Discussion 5.2 D I S C U S S I O N It has been reported recently that HSV-1 and H S V - 2 display type-specific differences in their interactions with host cells. These differences include preference for binding to various cell types (Vahlne et al., 1979; Vahlne et al., 1980), binding to sulfated glycosaminoglycans (Herold et al., 1996) and interactions with specific cellular components, such as the C 3 B receptor and Pr r l (Eisenberg et al., 1987; Friedman et al., 1984; Friedman et al., 1986; Fries et al., 1986; Geraghty et al., 1998). In this study, an additional type-specific phenotype was investigated which involves the interaction of virions with host cell surfaces devoid of G A G s . Infection of sog9 cells by H S V - 1 , but not H S V - 2 , could be rescued by the addition of dextran sulfate. What could account for the failure of DS to stimulate H S V - 2 infection? There are several possibilities which must be considered. It is clear that D S stimulation of HSV-1 is mediated, at least in part, by gB-1 and that gB-2 is inactive in this capacity when it is present in an otherwise unperturbed HSV-1 virion (RH1G13; Table 6). This observation is extended by the work of others in our laboratory, who have recently shown that infection of GAG-deficient adult mouse muscle fibers by RH1G13 is unresponsive to the addition of D S (Yeung et al., 1999). The role of gB-1 could be to interact with a distinct cellular receptor accessible to the virion at all times. This interaction is promoted when the virion is held in place even transiently by cell surface DS . If RH1G13 , which contains gB-2 in an HSV-1 virion, can still interact with DS , then it is likely that access to the gB-2 receptor is not enhanced in this instance. gB-1 and gB-2 are highly conserved, with 86% amino acid similarity (Stuve et al., 1987). For the most part, regions that have been shown to have functional significance are conserved, including cysteine residues and predicted glycosylation sites in the external domain (Norton et ai, 1998; Stuve et al, 1987). Despite the overall similarity, however, there exist clustered regions of marked divergence between the two proteins. Could the structural differences between gB-1 and gB-2 alter the ability of virus to interact with DS? 144 Chapter 5: Discussion Interestingly, and perhaps most relevant, are clustered amino acid substitutions within the N-terminal 85 amino acids of the mature protein and a second region that includes amino acids 451 to 495. The N-terminal divergence includes domains of high positive charge density that could interact with G A G s , including the gB-1 H B D (Laquerre et al., 1998b). Studies of GAG-binding proteins such as gC of H S V - 1 , B H V - 1 and P rV , show that sulfated polyanions are recognized by protein domains that are rich in basic residues, particularly lysine (Flynn and Ryan, 1995; Langeland et al., 1988; Liang et al., 1993; Tal-Singer et al, 1995; Trybala et al, 1994). Moreover, the inhibitory effects of D S on HIV-1 infection appear to be mediated by specific binding of the polyanion to positively charged amino acids concentrated in the V 3 loop of the envelope glycoprotein gpl20 (Callahan, 1991). I was curious as to whether or not the H B D of gB was involved in mediating DS stimulation. The results show that infection by K g B p K - , an H S V - 1 particle deleted for the gB H B D , was stimulated as efficiently by DS as wild-type infection. It can therefore be concluded that the H B D of gB-1 is not required for DS-mediated infection. It could be that the binding of D S to proteins is dependent upon specific peptide domains containing suitably positioned cationic residues and in this respect, some site in gB, other than the H B D , may be able to mediate binding to DS . Using another approach to identify a possible domain of gB that could interact with DS , the effect of DS on sog9 cell infection by two gB-1 mar mutants, R126 and R233, was tested. HSV-1 mar mutants have been successfully used in past studies to help localize heparan sulfate binding sites in gC (Trybala et al., 1994) and to identify domains of gB that function in viral penetration (Highlander, 1989). Despite significant changes in the secondary structure of gB between N-terminal amino acids 283 to 305, DS was still able to stimulate infection of both R126 and R233, implying that this region of the molecule may not be critical for mediating an interaction of the virus with DS . Nevertheless, the level of DS stimulation of R126, and in one experiment with R233, was lower than that of wild-type H S V - 1 . Therefore, one cannot exclude the possibility that this region assists in mediating the interaction. Perhaps this region is only one of several 145 Chapter 5: Discussion peptide domains on gB that help stabilize the interaction with D S . It is interesting that for several other GAG-binding proteins, such as gC, the consensus sequence elements are not close together in the primary sequence and only by folding of the protein structure, do these regions come together to form a crevice into which the G A G binds (Cardin and Weintraub, 1989; Trybala etal., 1994). The molecular recognition of G A G s by proteins is clearly a complex process. Therefore, other approaches wi l l need to be used to elucidate the role of gB-1 in D S -mediated infection. Future experiments could include an assessment of the ability of soluble gB as well as truncated forms of the protein to bind to DS . The construction and characterization of viruses bearing gB-1 /gB-2 chimaeric proteins could also help identify key sites of the protein. Nevertheless, the results presented here support the concept that gB-1 and gB-2 may differ in their ability to adsorb to sulfated polyanions such as DS or to recognize a cellular receptor. Differences in gB-1 and gB-2 can also be inferred from studies showing that gC-deficient H S V - 2 exhibits no loss in specific binding activity, specific infectivity, or rate of viral penetration (Gerber et al., 1995). This is very different from studies on gC-deficient H S V - 1 , which showed a serious impairment in virus adsorption. It may be that gC-1 and gB-2 predominate in their respective viruses to regulate the early interactions that lead to a productive infection. It was also interesting to discover that gC-2 reduces DS-mediated infection. In this instance the presence of gC-2 may interfere with the binding of the virion to D S at the cell surface. Alternatively, gC-2 may directly impede the interaction o f gB-2 with DS . These experiments using a gC" virus support the possibility that gC-2 is responsible for at least part of the unresponsiveness of the H S V - 2 virion to DS stimulation. It is important to consider that the H S V - 2 virion does show some level of D S stimulation in the absence of gC-2, thereby indicating that H S V - 2 can engage the DS-mediated pathway. There is no evidence, however, that gB-2 mediates this process in H S V - 2 . 146 Chapter 6: Summary CHAPTER 6: SUMMARY The H S V entry pathway normally involves viral attachment to cell surface HS and CS G A G s (Sheih et al., 1992; Banfield et al., 1995a). This interaction is mediated by the heparin-binding proteins gB and gC in the virion envelope (Herold et al. 1991, Laquerre et al., 1998). These glycoproteins contain heparin-binding regions that recognize and bind to particular structural features of G A G moieties (Herold et al, 1995; Tal-Singer et al, 1995; Feyzi et al, 1997; Laquerre et al, 1998). The binding of virus to G A G receptors is a critical component of the H S V entry pathway as it i) concentrates the virus at the cell surface and ii) brings the virus in proximity with its co-receptor. H S V co-receptors have recently been identified as HveA, a member of the T N F receptor family, and HveB, HveC and HIgR, members of the Ig superfamily (Montgomery et al, 1996; Cocchi et al, 1998b; Geraghty et al, 1998). Significantly, different HSV-1 and H S V - 2 strains vary in their ability to use these co-receptors. This likely influences the spread of virus in the human host. Attachment of the H S V to its co-receptor primes the virus for penetration into the cell, which involves the fusion of the virion envelope with the cell plasma membrane and results in the release of the viral capsid into the cell cytoplasm. The process is not yet fully understood, although appears to be regulated by the combined activities of gB, gD, g H and gL (Handler et al. 1996b). In this study, I uncovered a functional difference between HSV-1 and H S V - 2 in their ability to use dextran sulfate as a surrogate receptor on glycosaminoglycan-deficient sog9 cells. The absence of glycosaminoglycans on the sog9 cell surface reduces the adsorption of H S V in a manner that allows for a more sensitive readout of the virus-host interactions that ensue during the infection process. The significance of this study is that DS stimulates HSV-1 infection of sog9 cells, thereby acting as an artificial receptor to initiate a productive infection. B y contrast with previous experiments in which HSV-1 and H S V - 2 were differentiated by their susceptibility to inhibitors, I have identified an 147 Chapter 6: Summary interaction that promotes infection. This distinction is significant; whereas inhibition of infection could be caused by relatively non-specific blocking, stimulation of infection requires that specific host-virus interactions be maintained. A model for DS stimulation of HSV-1 infection of sog9 cells is illustrated in Figure 6.1. In Chapter 3, it was demonstrated that sog9 cells can be treated with DS prior to infection, that DS interacts with virions and that DS stimulates infection at 4°C. This supports a model in which DS binds to sog9 cells and tethers the virion at the cell surface. Possible receptors for DS on sog9 cells include scavenger receptors, which are capable of binding a variety of polyanions (Krieger et al. 1993). Using intertypic recombinants, it was demonstrated that HSV-1 interacted with DS, in part, via glycoprotein B. This interaction of virus with DS likely serves to concentrate HSV-1 virions at the cell surface, much the same way as the virus' natural receptor, heparan sulfate, does. In this manner, HSV-1 is brought into proximity with a second, downstream cell surface receptor, such as HveC. Once the virus engages this receptor, viral penetration can ensue. In the absence of DS, however, the probability of HSV-1 engaging its co-receptor is reduced, which in turn decreases productive infection. The striking differences in the behaviour of HSV-1 and HSV-2 in the DS assay are most likely defined by differences in the propensity of principally gB to interact with sulfated polyanions, as well as perhaps additional cell surface receptors. It will be interesting to take advantage of the differences in susceptibility conferred by DS to identify and characterize the domains of gB that mediate this effect. These domains are likely to be important in conferring type-specific properties on the respective virions. Results presented in this thesis show that the heparin-binding region of gB-1 is not required for mediating an interaction of HSV-1 with DS. The interactions of proteins with glycosaminoglycans are defined by several parameters and are not simply electrostatic in nature. It seems likely then that there may be another domain of gB-1 comprising a specific sequence and spatial arrangement of amino acid residues with which DS can interact. 148 Chapter 6: Summary Figure 6.1 Proposed model for dextran sulfate stimulation of H S V - 1 infection of sog9 cells. (A) Normal infection (e.g. L cell infection) involves the binding of virus to cell surface H S and/or C S proteoglycans. This facilitates a downstream interaction with a co-receptor, such as HveC. Stable attachment of H S V with its co-receptor promotes penetration of the virus into cells. (B) H S V infection of sog9 cells. In the absence of cell surface G A G s , the ability o f H S V to engage its co-receptor is reduced, resulting in decreased infection efficiency. (C) HSV-1 infection of sog9 cells in the presence of soluble dextran sulfate. DS binds to the sog9 cells surface, possibly to scavenger receptors. D S functions to tether the virus at the cell surface. This interaction is mediated, in part, by HSV-1 gB HSV-1 and facilitates attachment of the virus with its co-receptor, resulting in a productive infection. 149 Chapter 6: Summary A Lcell C sog9 cell + dextran sulfate 150 Chapter 6: Summary B y contrast to DS-mediated infection, HSV-1 infection of cell lines expressing heparan sulfate, including the novel EXT I -expressing cell lines, is mediated principally by g C - l . Moreover, the H B D of gB-1 appears to play a minimal role in binding of virus to heparan sulfate G A G s . Using chemically modified heparin compounds, I determined that the virion recognizes particular structural features of HS, including, 2-0-, 6-0- and N -sulfate groups. These findings further suggest that the cell and tissue tropism of H S V may, in part, be due to the structural variations of HS. The recent advances in identifying factors governing H S V - G A G interactions, such as those described herein, have made a substantial contribution to our understanding of the H S V attachment process. This in turn has had a significant impact on other areas of H S V research. For example, Laquerre and colleagues (1998a) altered the tropism of HSV-1 by replacing the HS binding domain of g C - l with erythropoietin (EPO). This engineered virus, but not wild-type virus, could bind to cells expressing the E P O receptor, and as such is the first evidence that HSV-1 attachment can be targeted to non-HSV cell surface receptors. The ability to alter host range of the virus can be exploited for H S V -based gene therapy purposes. This is underscored by work in our laboratory demonstrating that in vivo, DS can mediate HSV-1 infection of adult mouse myofibres, cells which are normally refractory to infection due to their lack of HS G A G s (Yeung et al, 1999). Skeletal muscle is an ideal seeding site for the treatment of a variety of disorders, including muscular dystrophy. Therefore, DS-mediated infection represents an approach for expression of H S V vectors in muscle fibers. On a final note, this study highlights the complex, multi-faceted nature of the interaction of H S V with host cell glycosaminoglycans. Clearly, both viral and host cell components have their own unique parts to play in this exchange. B y continuing to identify the finer aspects of this interaction, the role of H S V attachment in determining H S V epidemiology can be fully defined. 151 References REFERENCES Acsadi, G., A. Jani, B. Massie, M. Simoneau, P. Holland, K. Blaschuk. 1994. A differential efficiency of adenovirus-mediated in vivo gene transfer into skeletal muscle cells of different maturity. Hum. M o l . Gen. 3:579-84. Addison, C , F.J. Rixon, J.W. Palfreyman, M. O'Hara, and V.G. Preston. 1984. Characterization of a herpes simplex virus type 1 mutant which has a temperature-sensitive defect in penetration of cells and assembly of capsids. Virology 138:246-259. Ahn, J., H. Ludecke, S. Lindow, W.A. Horton, B. Lee, M.J. Wagner, B. Horsthemke, and D.Wells. 1995. Cloning of the putative tumor suppressor gene for hereditary multiple exostoses (EXT1). Nat.Gen. 11:137-143. Aoki, J., S. Koike, H. Asou, I. Ise, H. Suwa, T. Tanaka, M. Miyasaka, and A. Nomoto. 1997. Mouse homolog of poliovirus receptor-related gene 2 product, mPRR2, mediates homophilic cell aggregation. Exp. Cell Res. 235:374-384. Asher, Y., M . Heller, and Y. Becker. 1969. Incorporation of lipids into herpes simplex virus particles. J. Gen. V i r o l . 4: 65-76. Aurelius, L. 1998. Herpes simplex virus type 2: unique biological properties include neoplastic potential mediated by the P K domain of the large subunit of ribonucleotide reductase. Front. Biosci . 3:237-249. Baba, M., R.Pauwels, J. Balzarini, J. Arnout, J. Desmyter, and E. de Clercq. 1988a. Mechanism of inhibitory action of dextran sulfate and heparin on replication of human immunodeficiency virus in vitro. Proc. Natl. Acad. Sci. U S A 85:6132-6136. Baba, M. , R. Snoeck, R. Pauwels, and E. de Clercq. 1988b. Sulfated polysaccharides are potent and selective inhibitors of various enveloped viruses, including herpes simplex virus, cytomegalovirus, vesicular stomatitis virus and human immunodeficiency virus. Microb. Agents Chemo. 32:1742-1745. Baines, J.D, and B. Roizman. 1993. The U L 1 0 gene of herpes simplex virus 1 encodes a novel viral glycoprotein, g M , which is present in the virion and plasma membrane of infected cells. J. V i ro l . 67:1441-1452. 152 References Baines, J.D., P.L. Ward, G. Campadelli-Fiume, and B. Roizman. 1991. The U L 2 0 gene of herpes simplex virus 1 encodes a function necessary for viral egress. J. V i ro l . 65:6414-6424. Banfield, B.W., Y. Leduc, L. Esford, K. Shubert, and F. Tufaro. 1995a. Sequential isolation of proteoglycan synthesis mutants by using herpes simplex virus as a selective agent: evidence for a proteoglycan-independent virus entry pathway. J. V i r o l . 69: 3290-3298. Banfield, B.W., Y. Leduc, L.Esford, R.J. Visalli, C.R. Brandt and F.Tufaro. 1995b. Evidence for an interaction of herpes simplex virus with chondroitin sulfate proteoglycans during infection. Virology 208:531-539. Batterson, W., and B. Roizman. 1983. Characterization of the herpes simplex virion-associated factor responsible for the induction of a genes. J. V i ro l . 46:371-377. Boehmer, P.E., and I.R. Lehman. 1997. Herpes simplex virus D N A replication. Annu. Rev. Biochem. 66:347-384. Bova, C.A., J.P. Manfredi, R. Swanstrom. 1986. Env genes of avian retroviruses: nucleotide sequence and molecular recombinants define host range determinants. Virology 152:343-354. Browne, H., S. Bell, T. Minson and D. Wilson. 1996. A n endoplasmic reticulum-retained herpes simplex virus glycoprotein H is absent from secreted virions: evidence for reenvelopment during egress. J. V i ro l . 70:4311-4316. Brunetti, C.R., R.L. Burke, B. Hoflack, T. Ludwig, K.S. Dingwell and D.C. Johnson. 1995. Role of mannose-6-phosphate receptors in herpes simplex virus entry into cells and cell-to-cell transmission. J. V i ro l . 69:3517-3528. Brunetti, C.R., R.L. Burke, S. Kornfeld, W. Gregory, F.R. Masiarz, K.S. Dingwell, and D.C. Johnson. 1994. Herpes simplex virus glycoprotein D acquires mannose 6-phosphate residues and binds to mannose 6-phosphate receptors. J. B i o l . Chem. 269:17067-17074. Bzik, D., B. Fox, N. DeLuca, and S. Person. 1984. Nucleotide sequence of the glycoprotein gene, gB, of herpes simplex virus type 1. Virology 133:301-314. Cai, W., C. Gu, and S. Person. 1988. Role of glycoprotein B of herpes simplex virus type 1 in viral entry and cell fusion. J. V i ro l . 62:2596-2604. 153 References Caldwell, E.E.O, V. D. Nadkarni, J. R. Fromm, R. J. Linhardt, and J.M. Weiler. 1996. Importance of specific amino acids in protein binding sites for heparin and heparan sulfate. Int. J. Biochem. Cel l . B i o l . 28:203-216. Callahan, L.N., M. Phelan, M.Mallinson, and M.A. Norcross. 1991. Dextran sulfate blocks antibody binding to the principle neutralizing domain of human immunodeficiency virus type 1 without interfering with gpl20-CD4 interaction. J. V i r o l . 65: 1543-1550. Campadelli-Fiume,G., M. Aresenakis, F. Farabegoli, and B. Roizman. 1988. Entry of herpes simplex virus in B J cells that constitutively express viral glycoprotein D is by endocytosis and results in degradation of the virus. J. Virology. 62:159-167. Campadelli-Fiume, G., R. Brandimarti, C. Dilazzaro, P.L. Ward, B. Roizman, and M.R. Torrisi. 1993. Fragmentation and dispersal of Golgi proteins and redistribution of glycoproteins and glycolipids processed through Golgi following infection with herpes simplex virus 1. Proc. Natl. Acad. Sci. U S A . 90:2798-2802. Campadelli-Fiume, G., and F. Sefarini-Cessi. 1985. Processing of the oligosaccharide chains of herpes simplex virus type 1 glycoproteins. In. Roizman, B , ed. The herpesviruses vol 3. New York. Plenum Press pp. 357-382. Campadelli-Fiume, G., D. Stirpe, A. Boscaro, E. Avitabile, L. Foa-Tomasi, D. Barker and B. Roizman. 1990. Glycoprotein C-dependent attachment of herpes simplex virus to susceptible cells leading to productive infection. Virology 178:213-222. Campbell, M.E.M., J.W. Palfreyman, and C M . Preston. 1984. Identification of the herpes simplex virus D N A sequences which encodes a trans-acting polypeptide responsible for the stimulation of the immediate-early transcription. J. M o l . B i o l . 180:1-19. Cardin, A., and H.J.R.Weintraub. 1989. Molecular modeling of protein-glycosaminoglycan interactions. Arteriosclerosis 9:21-32. Chang, H., G.H. Cohen, and R. J. Eisenberg. 1994. Identification of functional regions of herpes simplex virus glycoprotein D by using linker-insertion mutagenesis. J. V i ro l . 68: 2529-2543. Chang, Y . E , L. Menotti, F. Filatov, G. Campadelli-Fiume, and B. Roizman. 1998. U L 27.5 is a novel fl gene antisense to the herpes simplex virus 1 gene encoding glycoprotein B . J. V i r o l . 72:6056-6064. 154 References Chen, Y.T., Maguire, R. Hileman, J.R. Fromm, J.D. Esko, R.J. Linhardt, and R.M.Marks. 1997. Dengue virus infectivity depends on envelope protein binding to target cell heparan sulfate. Nat. Med. 3:866-871. Cheung, P., B. Banfield, and F. Tufaro. 1991. Brefeldin A arrests the maturation and egress of herpes simplex virus particles during infection. J. V i ro l . 65:1893-1904. Chiang, H.Y., G.H. Cohen, and R.J. Eisenberg. 1994. Identification of functional regions of herpes simplex virus glycoprotein gD by using linker-insertion mutagenesis. J .Virol . 68:2529-2543. Chung, C , J. Hsiao, Y. Chang, and W. Chang. 1998. A 2 7 L protein mediates vaccinia virus interaction with cell surface heparan sulfate. J. V i ro l . 72:1577-1585. Claesson-Welsh, L., and P.G. Spear. 1986. Oligomerization of herpes simplex virus glycoprotein B . J. V i r o l . 60: 803-806. Cocchi, F., M. Lopez, L. Menotti, M. Aoubala, P. Dubreuil, and G. Campadelli-Fiume. 1998a. The V domain of herpesvirus Ig-like receptor (HIgR) contains a major functional region in herpes simplex virus-1 entry into cells and interacts physically with the viral glycoprotein D . Proc. Natl. Acad. Sci. 95:15700-15705. Cocchi, F., L. Menotti, P. Mirandola, M. Lopez, and G. Campadelli-Fiume. 1998b. The ectodomain of a novel member of the immunoglobulin subfamily related to the poliovirus receptor has the attributes of a bona fide receptor for herpes simplex virus types 1 and 2 in human cells. J. V i ro l . 72:9992-10002. Cockrell, A.S., and M.L Muggeridge. 1998. Herpes simplex virus 2 U L 4 5 is a type II membrane protein. J. V i ro l . 72:4430-4435. Cohen, J.L, and K.E. Seidel. 1993. Generation of varicella-zoster virus ( V Z V ) and viral mutants from cosmid D N A s : V Z V thymidylate synthetase is not essential for replication in vitro. Proc. Natl . , Acad. Sci. U S A 90:7376-7380. Cunningham, C , and A. Davison. 1993. A cosmid-based system for constructing mutants of herpes simplex virus type 1. Virology. 197:116-124. Dahms, N.M., P. Lobel, and S. Kornfeld. 1989. Mannose-6-phosphate receptors and lysosomal enzyme targeting. J. B io l . Chem. 264:12115-12118. Dales, S. and Y. Chardonnet. 1977. Early events in the interaction of adenoviruses with HeLa cells. IV. Association with microtubules and the nuclear pore complex during vectorial movement of the inoculum. Virology 56:465-483. 155 References Dales, S., and H. Silverberg. 1969. Viropexis of herpes simplex virus by H e L a cells. Virology. 37:475-80. Darlington, R.W., and L.H. Moss III. 1968. Herpesvirus envelopment. J. V i r o l . 2:48-55. Davison, M.D., F.J. Rixon, and A.J.Davison. 1992. Identification of gene encoding two capsid proteins (VP24 and VP26) of herpes simplex virus type 1. J. Gen. V i r o l . 73:2709-2713. Delius, H., and J.B. Clements. 1976. A partial denaturation map of herpes simplex virus type 1 D N A : evidence for inversions of the unique D N A regions. J. Gen. V i r o l . 33:125-133. DeLuca, N, D.J. Bzik, V . C . Bond, S. Person, and W. Snipes. 1982. Nucleotide sequences of herpes simplex virus type 1 (HSV-1) affecting virus entry, cell fusion, and production of glycoprotein gB (VP7). Virology 122:411-423. Desai, P.J., P.A. Schaffer, and A.C. Minson. 1988. Excretion of non-infectious virus particles lacking glycoprotein H by a temperature-sensitive mutant of herpes simplex virus type 1: evidence that gH is essential for virion infectivity. J.Gen. V i r o l . 69:1147-1156. Dingwell, K.S., C.R. Brunetti, R.L. Hendricks, Q.Tang, M. Tang, A.J. Rainbow, and D.C. Johnson. 1994. Herpes simplex virus glycoproteins E and I facilitate cell-to-cell spread in vivo and across junctions of cultured cells. J. V i ro l . 68:834-845. Dingwell, K.S., L.C. Doering, and D.C. Johnson. 1995. Glycoproteins E and I facilitate cell-to cell spread in vivo and across junctions of cultured cells. J. V i ro l . 69:7087-7098. Dingwell, K.S. , and D.C. Johnson. 1998. The herpes simplex virus gE-gl complex facilitates cell-to-cell spread and binds to components of cell junctions. J. V i ro l . 72: 8933-8942. Dolter, K.E. , R. Ramaswamy, and T.C. Mettenleiter. 1994 Syncytial mutations in the herpes simplex virus type 1 g K (UL53) gene occur in two distinct domains. J. V i r o l . 68:8277-8281. Doms, R. W., R. Lamb, J.K. Rose, and A. Helenius. 1993. Folding and assembly of viral membrane proteins. Virology 193:545-562. 156 References Dyer, A.P., B.W. Banfield, D. Martindale, D. Spannier, and F. Tufaro. 1997. Dextran sulfate can act as an artificial receptor to mediate a type-specific herpes simplex virus infection via glycoprotein B . J. V i ro l . 71:191-198. Dutch, R.E., R. C. Bruckner, E.S. Mocarski, and I. R. Lehman. 1992. Herpes simplex virus type 1 recombination: role of D N A replication and viral a sequences. J. V i ro l . 66:277-285. Earl, P.L, R.W. Doms, and B. Moss. 1990. Oligomeric structure of the human immunodeficiency virus type 1 envelope glycoprotein. Proc. Natl . Acad. Sci. U S A 87:648-652. Eberle, F., P. Dubreuil, M. Mattei, E. Devilard, and M. Lopez. 1995. The human PRR2 gene, related to the human poliovirus receptor gene (PVR), is the true homolog of the murine M P H gene. Gene 159:267-272. Efstathiou, S., H.J. Field, P.D. Griffiths, E.R. Kern, S.L. Sacks, N.M. Sawtell, and L.R. Stanberry. 1999. Herpes simplex virus latency and nucleoside analogues. Antiviral Res. 41:85-100. Efstathiou, S., A.C. Minson, H.J. Field, J.R. Anderson, and P. Wildly. 1986. Detection of herpes simplex virus-specific D N A sequences in latently infected mice and humans. J. V i r o l . 57:446-455. . Eisenberg, R.J., M . Ponce de Leon, and G.H. Cohen. 1980. Comparative structural analysis of glycoprotein gD of herpes simplex virus types 1 and 2. J. V i r o l . 35:428-435. Eisenberg, R.J., M . Ponce de Leon, H.M. Friedman, L.F.Fries, M.M. Frank, J.C. Hastings, and G.H. Cohen. 1987. Complement component C3b binds directly to purified glycoprotein C of herpes simplex virus types 1 and 2. Microb. Pathog. 3:423-435. Ejercito, P.M., E.D. Kieff, and B. Roizman. 1968. Characterization of herpes simplex virus strains differing in their effects on social behaviour of infected cells. J. Gen. V i r o l . 2:357-364. Epstein, N.A. 1962. Observations on the mode of release of herpes virus from infected HeLa cells. J. Cel l . B io l . 12:589-597 Esford, L.E. 1999. The role of cell surface architecture in the HSV-1 life cycle and in H A binding ability of CD44. Ph.D. thesis. 157 References Esko, J.D. 1991. Genetic analysis of proteoglycan structure, function and metabolism. Curr. Opin. Cel l B i o l . 3:805-816. Esko, J.D., T . E . Stewart, W.H. Taylor. 1985. Animal cell mutants defective in glycosaminoglycan biosynthesis. Proc. Natl. Acad. Sci. U S A 82:3197-3201. Feenstra, V., M . Hodaie, and D.C. Johnson. 1990. Deletions in herpes simplex virus glycoprotein D define nonessential and essential domains. J. V i r o l . 64:2096-2102. Feero, W.G. , J.D. Rosenlatt, J . Huard, S.C. Watkins, M . Epperly, P.R. Clemens. 1997. Vi ra l gene delivery to skeletal muscle: insights on maturation-dependent loss of fiber infectivity for adenovirus and herpes simplexvirus type 1 viral vectors. Hum. Gene Ther. 8:371-80. Feyzi, E . , E . Trybala, T. Bergstrom, U . Lindahl, and D. Spillmann. 1997. Structural requirement of heparan sulfate for interaction with herpes simplex virus type 1 virions and isolated glycoprotein C. J. B io l . Chem. 272:24850-24857. Flynn, S.J., and P.Ryan. 1995. A heterologous heparin-binding domain can promote functional attachment of a pseudorabies virus gC mutant to cell surfaces. J. V i ro l . 69:834-839. Forrester, A., H . Farrell, G . Wilkinson, J . Kaye, N.Davis-Poynter, and T. Minson. 1992. Construction and properties of a mutant of herpes simplex virus type 1 with glycoprotein H coding sequences deleted. J. V i ro l . 66:341-348. Fraefel, C , S. Song, F. Lim, P. Lang, L . Yu, Y. Wang, P. Wild, and A.I. Geller. 1996. Helper virus-free transfer of herpes simplex virus type 1 plasmid vectors into neural cells. J. V i ro l . 70:7190-7197. Friedman, H . M . , G .H. Cohen, R.J. Eisenberg, C A . Seidel, and D.B. Cines. 1984. Glycoprotein C of herpes simplex virus 1 acts as a receptor for the C3b complement component on infected cells. Nature. 309:633-635. Friedman, H . M . , J .C. Glorioso, G.H. Cohen, J .C. Hastings, S.L. Harris, and R.J. Eisenberg. 1986. Binding of complement component C3b to glycoprotein C of herpes simplex type 1: mapping of gC-binding sites and demonstration of conserved C3b binding in low-passage clinical isolates. J. Vi ro l . 60: 470-475. Fries, L . F . , H . M . Friedman, G.H. Cohen, R.J. Eisenberg, C H . Hammer and M . M . Frank. 1986. Glycoprotein C of herpes simplex virus type 1 is an inhibitor of the complement cascade. J. Immunol. 137:1636-1641. 158 References Frink, R.J., R.J. Eisenberg, G.Cohen, and E.K. Wagner. 1983. Detailed analysis of the portion of the herpes simplex virus type 1 genome encoding glycoprotein C. J. V i r o l . 45:634-647. Fuller, A.O., and W.C. Lee. 1992. Herpes simplex type 1 entry through a cascade of virus-cell interactions required different roles of gD and gH in penetration. J. V i ro l . 66:5002-5012. Fuller, A.O., R.E. Santos, and P.G. Spear. 1989. Neutralizing antibodies specific for glycoprotein H of herpes simplex virus permit virus attachment by prevent penetration. J. V i ro l . 63:3435-3443. Fuller, A.O., and P.G. Spear. 1987. Anti-glycoprotein D antibodies that permit adsorption but block infection by herpes simplex virus 1 prevents virion-cell fusion at the cell surface. Proc. Natl. Acad. Sci. U S A 84:5454-5458. Fuller, A.O., and P.G. Spear. 1985. Specificities of monoclonal and polyclonal antibodies that inhibit adsorption of herpes simplex virus to cells and lack of inhibition by potent neutralizing antibodies. J. V i r o l . 55:475-482. Galloway, D.A, C M . Fenoglio, and J.K. McDougall. 1982. Limited transcription of the herpes simplex virus genome when latent in human sensory ganglia. J. V i ro l . 41:686-91. Geraghty, R. J., C. Krummernacher, G.H. Cohen, R.J. Eisenberg, and P.G. Spear. Entry of alphaherpesviruses mediated by poliovirus receptor-related protein 1 and poliovirus receptor. Science 280:1618-1620. Gerber, S.L, B.J. Belva, and B.C. Herold. 1995. Differences in the role of glycoprotein C of HSV-1 and H S V - 2 in viral binding may contribute to serotype differences in cell tropism. Virology. 214:29-39. Goldstein, D.J., and S.K. Weller. 1998. Herpes simplex virus type 1-induced ribonucleotide reductase activity is dispensable for virus growth and D N A synthesis: isolation and characterization of an ICP6 lacZ insertion mutant. J. V i ro l . 62:196-205. Gompels, U., and A . C Minson. 1989. Antigenic properties and cellular localization of herpes simplex virus glycoprotein H synthesized in a mammalian cell expression system. J. V i ro l . 63:4744-4755. Gompels, U., and A.C. Minson. 1986. The properties and sequence of glycoprotein H of herpes simplex virus type 1. Virology. 153:230-247. 159 References Gray, C.P., and R.H.C. Kaerner. 1984. Sequence of the putative origin of replication in the U L region of herpes simplex virus type 1 A N G D N A . J. Gen. V i r o l . 65:2109-2119. Greenspan, P., and R.L. Gutman. 1994. Endocytosis of sulfatides by macrophages: relationship to the cellular uptake up phosphatidylserine. J. Leukocyte B i o l . 55:99-104. Gruenberg, J. and K.E.Howell. 1989. Membrane traffic in endocytosis: insights from cell-free assays. Annu. Rev. Cel l B io l . 5:453-481. Gruenheid, S., L.Gatzke, H. Meadows, and F. Tufaro. 1993. Herpes simplex virus infection and propagation in a mouse L-cel l mutant lacking heparan sulfate proteoglycans. J. V i ro l . 67:93-100. Haanes, E.J., C M . Nelson, C L . Soule, and J.L. Goodman. 1994. The U145 gene product is required for herpes simplex virus type 1 glycoprotein B-induced cell fusion. J. Vi ro l . 68:5825-5834. Handler, C.G., G.H. Cohen, and R.J. Eisenberg. 1996. Cross-linking of glycoprotein oligomers during herpes simplex virus type 1 entry. J. V i ro l . 70:6076-6082. Handler, C G , R.J. Eisenberg, and G.H. Cohen. 1996. Oligomeric structure of glycoproteins in herpes simplex virus type 1. J. V i ro l . 70:6067-6075. Hascall, V . C , A . Calabro, R.J. Midura, and M. Yangishita. 1994. Isolation and characterization of proteoglycans. Methods. Enzymol. 230:390-417. Hayward, G.S., R.J. Jacob, S.C Wadsworth, and B. Roizman. 1975. Anatomy of herpes simplex virus D N A : evidence for four populations of molecules that differ in the relative orientations of their long and short segments. Proc. Natl. Acad. Sci. U S A 72:4243-4247. Hecht, J.T., D. Hogue, Y.Wang, S.H. Blanton, M.Wagner, L . C Strong, W. Raskind, M.F.Hansen, and D. Wells. 1997. Hereditary multiple exostoses ( E X T ) : mutational studies of familial E X T 1 cases and EXT-Associated malignancies. Am.J.Hum.Genet. 60:80-86. Hennekam, R . C 1991. Hereditary multiple exostoses. J. Med. Genet. 28:262-266. Herold, B.C., S.I. Gerber, B.I. Belval, A . M . Siston, and N. Shulman. 1996. Differences in the susceptibility of herpes simplex virus types 1 and 2 to modified heparin compounds suggest serotype differences in viral entry. J. V i r o l . 70:3461-3469. 160 References Herold, B.C., S. I. Gerber, T. Polonsky, B. J. Belval, P. N. Shaklee, and K. Holme. 1995. Identification of structural features of heparin required for inhibition of herpes simplex virus type I binding. Virology 206:1108-1116. Herold, B.C., R.J. Visalli, N. Sumarski, C.R. Brandt, and P.G. Spear. 1994. Glycoprotein C-independent binding of herpes simplex virus to cells requires cell surface heparan sulfate and glycoprotein B . J.Gen. V i ro l . 75:1211-1222. Herold, B.C., D. WuDunn, N . Soltys, and P.G. Spear. 1991. Glycoprotein C of herpes simplex virus type 1 plays a principle role in the adsorption of virus to cells and in infectivity. J. V i r o l . 65:1090-1098. Highlander, S.L., W. Cai, S. Person, M. Levine, and J.C. Glorioso. 1988. Monoclonal antibodies define a domain on herpes simplex virus glycoprotein B involved in virus penetration. J. V i r o l . 62:1881-1888. Highlander, S.L., W.F. Goins, S. Person, T.C.Holland, M. Levine, and J.C. Glorioso. 1991. Oligomer formation of the gB glycoprotein of herpes simplex virus type 1. J. V i ro l . 70:3461-3469. Highlander, S.L., S.L. Sutherland, PJ. Gage, D.C. Johnson, M. Levine, and J.C.Glorioso. 1987. Neutralizing monoclonal antibodies specific for herpes simplex virus glycoprotein D inhibit virus penetration. J. V i r o l . 61:3356-3364. Horsburgh, B.C., M.M. Hubinette, and F. Tufaro. 1999. Genetic manipulation of H S V using bacterial artificial chromosomes. Methods Enzymol. 306:337-352. Hsu, H., I. Solovyev, A . Colomero, R. Elliot, M. Kelly, and W.J. Boyle. 1997. A T A R , a novel tumor necrosis factor receptor family member, signals through T R A F 2 and T R A F 5 . J. B i o l . Chem. 272: 13471-13474, Hubbard, S.C. 1988. Regulation of glycosylation. J. B io l . Chem. 263:19303-19317. Hung, S.-L. C. Peng, I. Kostavasili, H.M. Friedman, J.D. Lambris, R.J. Eisenberg, and G.H. Cohen. 1994. The interaction of glycoprotein C of herpes simplex virus types 1 and 2 with the alternative complement pathway. Virology 203:299-312. Hutchinson, L., H. Browne, V. Wargent, N. Davis-Poynter, S. Primorac, K. Goldsmith, A . C . Minsom, and D.C. Johnson. 1992a. A novel herpes simplex virus glycoprotein, gL, forms a complex with glycoprotein H (gH) and affects normal folding and surface expression of gH. J. Vi ro l . 66:2240-2250. 161 References Hutchinson, L., K. Goldsmith, D. Snoddy, H. Ghosh, F. L. Graham, and D.C. Johnson. 1992b. Identification and characterization of a novel herpes simplex virus glycoprotein K , involved in cell fusion. J. V i ro l . 66:5603-5609. Hutchinson. L., and D. Johnson. 1995. Herpes simplex virus glycoprotein K promotes egress of virus particles. J. Vi ro l . 69: 5401-5413. Hutchinson, L., C. Roop-Beauchamp, and D. Johnson. 1995. Herpes simplex virus glycoprotein K is known to influence fusion of infected cells, yet is not on the cell surface. J. V i ro l . 69:4556-4563. Ida, H., A . Kurata, K. Eguchi, I. Yamashita, M. Nakashima, M. Sakai, Y. Kawabe, T. Nakamura, and S. Nagataki. 1994. Mechanism of inhibitory effect of dextran sulfate and heparin on human T-cell lymphotropic virus type 1 (HTLV-l ) - induced syncytium formation in vitro: role of cell-to-cell contact. Antiviral Res. 23:143-159. Jackson, R.L., S.J. Busch, and A.D.Cardin. 1991.Glycosaminoglycans: molecular properties, protein interactions, and role in physiological processes. Physiol. Rev. 71:481-539. Jackson, T., F. M. Ellard, R .A. Ghazaleh, S.M. Brookes, W.E. Blakemore, A . H . Corteyn, D.I. Stuart, J.W. Newman, and A .M.Q. King. 1996. Efficient infection of cells in culture by type O foot-and-mouth disease virus requires binding to cell surface heparan sulfate. J. V i ro l . 70:5282-5287. Jacob, R.J., L.S. Morse, and B. Roizman. 1979. Anatomy of herpes simplex virus D N A XIII. Accumulation of head to tail concatamers in nuclei of infected cells and their role in generation of the four isomeric arrangements of viral D N A . J. V i ro l . 9:448-457. Jayachandra, S., A . Baghian, and K. Kousoulas. 1997. Herpes simplex virus type 1 glycoprotein K is not essential for infectious virus production in actively replicating cells but is required for efficient envelopment and translocation of infectious virions from the cytoplasm to the extracellular space. J. V i ro l . 71:5012-5024. Jentoft, N. 1990. Why are proteins O-glycosylated? TIBS 15:291-294. Johnson, D., R. L. Burke, and T. Gregory. 1990. Soluble forms of herpes simplex virus glycoprotein D bind to a limited number of cell surface receptors and inhibit virus entry into cells. J. V i ro l . 64:2569-2576. Johnson, D . C , and M.W. Ligas.1988. Herpes simplex viruses lacking glycoprotein D are unable to inhibit virus penetration: quantitative evidence for virus-specific cell surface receptors. J. V i r o l . 62:4605-4612. 162 References Johnson, D. and P.G. Spear. 1982. Monensin inhibits the processing of herpes simplex virus glycoproteins, their transport to the cell surface, and the egress of virions from infected cells. J. Vi ro l . 43:1102-1112. Johnson, D., and P.G. Spear. 1983. O-linked oligosaccharides are acquired by herpes simplex virus glycoproteins in the Golgi Apparatus. Cel l 32:987-997. Johnson, D., and P.G. Spear. 1989. Herpes simplex virus glycoprotein D mediates interference with herpes simplex virus infection. 63:819-827. Jones, F., and C. Grose. 1988. Role of cytoplasmic vacuoles in varicella-zoster virus glycoprotein trafficking and virion envelopment. J. V i ro l . 62: 2701-2711. Jons, A., J. M . Dijkstra, and T.C. Mettenleiter. 1998. Glycoproteins M and N of pseudorabies virus form a disulfide-linked complex. J. Vi ro l . 72:550-557. Kaerner, H.C., C H . Shroder, A. Ott-Hartman, G. Kumel, and H. Kaerner. 1983. Genetic variability of herpes simplex virus: development of a pathogenic variant during passaging of a nonpathogenic herpes simples virus type 1 in mouse brain. J. V i r o l . 46:83-91. Karger, A., and T.C. Mettenleiter. 1993. Glycoproteins g i l l and gp50 play dominant roles in the biphasic attachment of pseudorabies virus. Virology 194:654-664. Kemble, G., G. Duke, R. Winter, and R. Spaete. 1996. Defined large-scale alterations of the human cytomegalovirus genome constructed by cotransfection of overlapping cosmids. J. V i ro l . 70:2044-2048. Kielian, M. , S. Jurgerwirth, K.U. Sayad, and S. DeCandido. 1990. Biosynthesis, maturation, and acid activation of the semliki forest virus fusion protein. J. V i r o l . 64:4614-4624. Kitagawa, H., H. Shimakawa, and K. Sugahara. 1999. The tumor suppressor E X T -like gene EXTL2 encodes an a l , 4-N-acetylhexoaminyltransferase that transfers N -acetylgalactosamine and N-acetylglucosamine to the common glycosaminoglycan-protein linkage region. J. B io l . Chem. 274:13933-13937. Klebe, R.J. and P.J. Mock. 1982. Effect of glycosaminoglycans on fibronectin-mediated cell attachment. J.Cell. Physiol. 112:5-9. 163 References Klupp, B.G., J. Baumeister, P. Dietz, H. Granzow, and T.C. Mettenleiter. 1998. Pseudorabies virus glycoprotein K is a virion structural component involved in virus release but is not required for entry. J. V i r o l . 72:1949-1958. Kit, S. and D.R. Dubs. 1963. Nonfunctional thymidine kinase cistron in bromodeoxyuridine resistant strains of herpes simplex virus. Biochem. Biophys. Res. Comm. 13:500-504. Kjellen, L. and U. Lindahl. 1991. Proteoglycans: structures and interactions. Annu.Rev. Biochem. 60:443-475. Kornfeld, R., and W.S.M. Wold. 1981. Structure of the oligosaccharides of the glycoprotein coded by early region E3 of adenovirus 2. J. V i ro l . 40:440-449. Krieger, M. 1992. Molecular flypaper and atherosclerosis: structure of the macrophage scavenger receptor. Trends Biochem. Sci. 17:141-146. Krieger, M. , S.Acton, J.Ashkenas, A.Pearson, M. Penman, and D. Resnich. 1993. Molecular flypaper, host defense and atherosclerosis. J. Biol .Chem. 268:4569-4572. Kristensson, K., E. Lycke, M. Roytta, B. Svennerholm, and A. Vahlne. 1986. Neuritic transport of herpes simplex virus in rat sensory neurons in vitro. Effects of substances interacting with microtubular function and axonal flow. J. Gen. V i ro l . 67:2023-2028. Krumbiegel, M. , D.S. Dimitov, A. Puri, and R. Blumenthal. 1992. Dextran sulfate inhibits fusion of influenza virus and cells expressing influenza hemagglutin with red blood cells. Biochim. Biophys. Acta. 110:158-164. Krummenacher, C , A. V. Nicola, J. C. Whitbeck, H. Lou, W. Hou, J.D. Lambris, R. Geraghty, P.G. Spear, G.H. Cohen and R. J. Eisenberg. 1998. Herpes simplex virus glycoprotein D can bind to poliovirus receptor-related protein 1 or herpes entry mediator, two structurally unrelated mediators of virus entry. J. V i ro l . 72:7064-7074. Krusat, T., and H.J. Streckert. 1997. Heparin-dependent attachment of respiratory syncytial virus (RSV) to host cells. Arch. Vi ro l . 142:1247-1254. Kwon, B., K.B.Tan, J. Ni, K. Oh, Z.H. Lee, K.K. Kim, Y. Kim, S. Wang, R. Gentz, G. Yu, J. Harrop, S.D. Lyn, C. Silverman, T.G. Porter, A. Truneh, and P.R. Young. 1997. A newly identified member of the tumor necrosis factor receptor superfamily with a wide tissue distribution and invelvement in lymphocyte activation. J. B i o l . Chem. 272: 14272-14276. 164 References Kwong, A.D., and N. Frenkel. 1987. Herpes simplex virus-infected cells contain a function(s) that destabilizes both host and viral m R N A s . Proc. Natl . Acad. Sci. U S A 84:1926-1930. Langeland, N., H. Holmsen, J.R. Lillehaug, and L.Haar. 1987. Evidence that neomycin inhibits binding of herpes simplex virus type 1 to the cellular receptor. J. V i r o l . 61:3388-3393. Langeland, N., and L.J. Moore. 1990. Reduction of HSV-1 binding to B H K cells after treatment with phophatidylinositol-specific phospholipase C. F E B S Lett 277:253-256. Langeland, N., L.J. Moore, H. Holmsen, and L.Haar. 1988. Interaction of polylysine with the cellular receptor for herpes simplex virus type 1. J. Gen V i r o l . 69:1137-1145. Langeland, N., A.M. Oyan, H.S. Marsden, A. Cross, J.C. Glorioso, L.J.Moore, and L.Haar. 1990. Localization on the herpes simplex virus type 1 genome of a region encoding proteins involved in adsorption to the cellular receptor. J. V i ro l . 64:1271-1277. Laquerre, S., D.B. Anderson, R. Arganin, and J.C. Glorioso. 1996. Herpes simplex virus type 1 glycoprotein B requires a cysteine residue at position 633 for folding, processing and incorporation into mature infectious virus particles. J. V i r o l . 72:4940-4949. Laquerre, S., D.B. Anderson, D.B. Stolz, and J.C. Glorioso. 1998a. Recombinant herpes simplex virus type 1 engineered for targeted binding to erythropoietin receptor-bearing cells. J. Vi ro l . 72:9683-9697. Laquerre.S., R. Argnani, D.B. Anderson, S. Zuccini, R. Manservigi, and J.C. Glorioso. 1998b. Heparan sulfate proteoglycan binding by herpes simplex virus type 1 glycoprotein B and C, which differ in their contributions to virus attachment, penetration, and cell-to-cell spread. J. Vi ro l . 72:6119-6130. Lasky, L.A., and D.J. Dowbenko. 1984. D N A sequence analysis of the type-common glycoprotein-D genes of herpes simplex virus types 1 and 2. D N A 3:23-29. LeMerrer, M. , L . Legeal-Mallet, P.M. Jeannin, B. Horsthemke, A. Schinzel, H. Plauchu, A. Toutain, F.Achard, A.Munnich, and P. Maroteaux. 1994. A gene for hereditary multiple exostoses maps to chromosome 19p. Hum. M o l . Genet. 3:717-722. Lederman, S., R. Gulick, and L. Chess. 1989. Dextran sulfate and heparin interact with CD4 molecules to inhibit the binding of coat protein (gpl20) of H I V . J. Immunol. 143:1149-1154. 165 References Leong, J.M., P.E. Morrissey, E. Ortega-Barria, M.E.A. Pereira, and J. Coburn. 1995. Hemagglutin and proteoglycan binding by the Lyme disease spirochete, Borrelia burgdorferi. Infect. Immun. 63:874-883. Liang, X., L.A. babiuk, and T.J. Zamb. 1993. Mapping of heparin-binding structures of bovine herpesvirus 1 and pseudorabies virus g i l l glycoproteins. Virology 194:233-243. Ligas, M.W., and D. Johnson. 1988. A herpes simplex virus mutant in which glycoprotein D sequences are replaced by P-galactosidase sequences binds to but is unable to penetrate into cells. J. V i ro l . 62:1486-1494. Lind, T., U. Lindahl, and K. Lidholt. 1993. Biosynthesis of heparin/heparan sulfate. J. of B io l . Chem. 268:20705-20708. Lind, T., F. Tufaro, C. McCormick, U. Lindahl, and Kerstin Lidholt. 1998. The putative tumor suppressors E X T 1 and E X T 2 are glycosyltransferases required for the biosynthesis of heparan sulfate. J. B i o l . Chem. 273:26265-26268. Lindberg, S., and S. Svenson. 1968. Structural studies on dextran from Leuconostoc mesenteroides N R R L B-512. Acta Chem. Scand. 22:1907-1912. Liptak, L.M. , S.L. Uprichard, and D.M. Knipe. 1996. Functional order of assembly of herpes simplex virus D N A replication proteins into prereplicative sites. J. V i r o l . 70:1759-67. Little, S.P., J.T. Jofre, R.J. Courtney, and P.A. Schaffer. 1981. A virion-associated glycoprotein essential for infectivity of herpes simplex virus type 1. Virology 115:149-158. Lopez, M., F. Eberle, M. Mattei, J. Gabert, F. Birg, F. Bardin, C. Maroc, and P. Dubreuil. 1995. Complementary D N A Characterization and chromosomal localization of a human gene related to the poliovirus receptor-encoding gene. 155:261-265. Luckert-Wicklund, C , R.M. Pauli, D. Johston, and J.T. Hecht. 1995. Natural history study of hereditary multiple exostoses. A m . J. Med. Genet. 55:43-46. Lycke, E., M. Johansson, B. Svennerholm, and U. Lindahl. 1991. Binding of herpes simplex virus to cellular heparan sulfate, an initial step in the adsorption process. J. Gen. Vi ro l . 72:1131-1137. MacLean, C.A., S. Efsetsthiou, M.L. Elliot, F.E. Jamieson, and D.J. McGeoch. 1991. Investigation of herpes simplex virus type 1 genes encoding multiply inserted membrane proteins. J. Gen. V i r o l . 72:897-906. 166 References MacLean, C.A., L .M. Robertson, and F.E. Jamieson. 1993. Characterization of the U L 1 0 gene product of herpes simplev virus type 1 and investigation of its role in vivo. J. Gen. Vi ro l . 74:975-983. McCormick, C , G. Duncan, K. T. Goutsos, T. Lindt, K. Lindholt, and F. Tufaro. 1999. The putative tumor suppressors E X T 1 and E X T 2 form a stable complex that accumulates in the Golgi and catalyzes the synthesis of heparan sulfate. P N A S (in press) McCormick C , Y. Leduc, D. Martindale, K. Mattison, L.E. Esford, A.P. Dyer and Frank Tufaro. 1998. The putative tumor suppressor E X T 1 alters the expression of cell-surface heparan sulfate. Nat. Gen. 19:158-161. McGeoch, D.J. 1990. Evolutionary relationships of virion glycoprotein genes in the S regions of alphaherpesvirus genomes. J. Gen. Vi ro l . 71:2361-2367. McGeoch, D.J., M.A. Dalrymple, A.J. Davison, A. Dolan, M.C. Frame, D. McNab, L.J. Perry, J.E. Scott, and P. Taylor. 1988. The complete D N A sequence of the long unique region in the genome of herpes simplex virus type 1. J. Gen. V i r o l . 69:1531-1574. McLennan, J.L., and G. Darby. 1980. Herpes simplex virus latency: the cellular location of virus in dorsal root ganglia and the fate of the infected cell following virus activation. J. Gen. Vi ro l . 51:233-243. Manservigi, R., and E. Cassai. 1991. The glycoproteins of the human herpesviruses. Comp. Immun. Microbiol , infect. Dis. 14:81-95. Margalit, H., N. Fischerr, and S. A. Ben-Sasson. 1993. Comparative analysis of structurally defined heparin binding sequences reveals a distinct spatial distribution of basic residues. J. B i o l . Chem. 268:19228-19231. Marshall, R. 1972. Glycoproteins. Annu. Rev. Biochem. 41:673-702. Martin. D.W., and P.C. Weber. 1996. The a sequence is dispensable for isomerization of herpes simplex virus type I genome. 70:8801-8812. Matthews, J.T., G.H. Cohen, and R.J. Eisenberg. 1983. Synthesis and processing of glycoprotein D of herpes simplex virus types 1 and 2 in an in vitro system. J. V i ro l . 48:521-533. 167 References Mauri, D. N., R. Ebner, R.I. Montgomery, K.D. Kochel, T. Cheung, G. Yu, S. Ruben, M . Murphy, R.J. Eisenberg, G.H. Cohen, P.G. Spear, C. F. Ware. 1998. L I G H T , a new member of the T N F superfamily, and lymphotoxin a are ligands for herpesvirus entry mediator. Immunity. 8:21-30. Messerle, M. , I. Crnkovic, W. Hammerschmidt, H. Ziegler, and U. Koszinowski. 1997. Cloning and mutagenesis of a herpesvirus genome as an infectious bacterial artificial chromosome. Proc. Natl. Acad. Sci U S A 94:14759-14763. Mettenleiter, T.C., L. Zsak, F. Zuckerman, N. Snugg, H. Kern, and T. Ben-Porat. 1990. Interaction of glycoprotein g i l l with cellular heparinlike substance mediates adsorption of pseudorabies virus. J. V i r o l . 64:278-286. Montgomery, R.I., M.S. Warner, B.J. Lum and P.G. Spear. 1996. Herpes simplex virus 1 entry into cells mediated by a novel member of the T N F / N G F receptor family. Cel l 87:427-436. Morgan, C , H.M. Rose, and B. Mednis. 1968. Electron microscopy of herpes simplex virus 1 entry. J. V i r o l . 2:507-516. Muggeridge, M.L, H.Y. Chaing, G.H. Cohen, and R.J. Eisenberg. 1992. Mapping of a functional site on herpes simplex virus glycoprotein D . J. Cel l Biochem. Suppl. 16C:132. Nagai, Y. 1995. Virus activation by host proteinases. A pivotal role in the spread of infection, tissue tropism and pathogenicity. 39:1-9. Naismith, J.H., and S. R. Sprang. 1998. Modularity in the TNF-receptor family. TIBS. 23:74-79. Navarro, D., P.Paz, and L. Pereira. 1992 Domains of herpes simplex virus 1 glycoprotein B that function in virus penetration, cell-to-cell spread, and cell fusion. Virology 186:99-112. Neyts, J. R. Snoeck, D. Schots, J. Balzarini, J.D. Esko, A. Van-Schepdael, and E. De-Cercq. 1992. Sulfate polymers inhibit the interaction of human cytomegalovirus with cell surface heparan sulfate. Virology 189:45-58. Nicola, A.V., M. Ponce de Leon, R. Xu, W. Hou, J.C. Whitbeck, C. Krummenacher, R. Montgomery, P.G. Spear, R. J. Eisenberg, and G.H. Cohen. 1998. Monoclonal antibodies to distinct sites on herpes simplex virus (HSV) glycoprotein D block H S V binding to H V E M . J. V i ro l . 72:3595-3601. 168 References Norton, D.D, D.S. Dwyer, and M.I. Muggeridge. 1998. Use of a neural network secondary structure prediction to define targets for mutagenesis of herpes simplex virus glycoprotein B . Virus Res. 55:37-48. Ohki, S. K. Arnold, N. Srinivasakumar, and T.D. Flanagan. 1992. Effect of anionic polymers on fusion of Sendai virus with human erythrocyte ghosts. Antiviral . Res. 18:163-177. Okazaki, K., T. Matsuzaki, Y. Sugahara, J. Okada, M. Hasebe, Y., Iwamura, M. Ohnishi, T. Kanno, M. Shimizu, E. Honda, and Y. Kono. 1991. B H V - 1 adsorption is mediated by the interaction of glycoprotein g i l l with a heparin-like moiety on the cell surface. Virology 181:666-670. Olofsson, S., J. Blomberg, and E. Lycke. 1981. O-glycosidic carbohydrate-peptide linkages of herpes simplex virus glycoproteins. Arch. Vi ro l . 70:321-329. Oyan, A.M., K.E. Dolter, N. Langeland, W.F. Goins, J.C. Glorioso, L.Haar, and CS. Crumpacker. 1993. Resistance of herpes simplex virus type 2 to neomycin maps to the N-terminal portion of glycoprotein C. J. V i ro l . 67:2434-2441. Patel, M., M. Yanagishita, G. Rodriquez, D.C. Bou-Habib, T. Oravecz, B.C. Hascall, and M.A. Norcross. 1993. Cell-surface heparan sulfate proteoglycan mediates HIV-1 infection of T-cell lines. A I D S Res. Hum. Retroviruses 9:167-174. Peake, M.L., P. Nystrom, and L. Pizer. 1982. Herpesvirus glycoprotein synthesis and insertion into plasma membranes. J. Vi ro l . 42:678-690. Pellett, P.E., Konstantin G. Kousoulas, L. Pereira, and B. Roizman. 1985. Anatomy of the herpes simplex virus 1 strain F glycoprotein B gene: primary sequence and predicted protein structure of the wi ld type and of monoclonal antibody-resistant mutants. J. V i r o l . 53:243-253. Phillippe, C. D.Porter, M.E. Emerton, D.E. Wells, A. Hamish, R.W. Simpson, and A. P. Monaco. 1997. Mutation screening of the E X T 1 and E X T 2 genes in patients with hereditary multiple exostoses. A m . J. Hu. Genet. 61:520-528. Pizer, L.I., G. Cohen, and R.J. Eisenberg. 1980. Effect of tunicamycin on herpes simplex virus glycoproteins and infectious virus production. J. V i ro l . 34:142-153. 169 References Poffenberger, K.L. , E. Tabares, and B. Roizman. 1983. Characterization of a viable, noninverting herpes simplex virus 1 genome derived by insertion and deletion of sequences at the junction of components L and S. Proc. Natl. Acad. Sci . U S A 80:2690-2694. Post, L.E. , S. Mackem, and B. Roizman. 1981. Regulation of a genes of herpes simplex virus: expression of chimeric genes produced by fusion of thymidine kinase with a gene promoters. Ce l l 24:555-565. Quinn, J.P., and D.J. McGeoch. 1985. D N A sequence of the region in the genome of herpes simplex virus type 1 containing the genes for D N A polymerase and the major D N A binding protein. Nucleic Acids Res. 13:8143-8163. Read, G.S., and Frenkel, N. 1983. Herpes simplex virus mutants defective in the virion-associated shutoff of host polypeptide synthesis and exhibiting abnormal synthesis of alpha (immediate-early) viral polypeptides. J. V i ro l . 68:498-512. Roizman, B. 1996. Herpesviridae. In: Fields' Virology, 3rd ed. B . N . Fields et al., Eds. Raven Press, N e w York, N . Y . , pp. 2221-2230. Roizman, B. and Sears, A.E. 1996. Herpes simplex viruses and their replication. In: Fields' Virology, 3rd ed. B . N . Fields et al., Eds., Raven Press, New York, N Y . , pp.2231-2295. Roizman, B. 1993. The family herpesviridae: a brief introduction. In:The human herpesviruses. B . Roizman et al., Eds., Raven Press, New York, N . Y . , pp. 1-9. Rossman, M.G. 1994. Vira l cell recognition and entry. Protein Sci. 3:1712-1725. Roop, C . L. Hutchinson, and D. C . Johnson. 1993. A mutant herpes simplex virus type 1 unable to express glycoprotein L cannot enter cells, and its particles lack glycoprotein H . J. V i r o l . 67:2285-2297. Ruoslahti, E. Structure and biology of proteoglycans. Annu. Rev. Cel l B i o l . 4:229-55. Saharkhiz-Langroodi, A., and T . C . Holland. 1997. Identification of the fusion-from without determinants of herpes simplex virus glycoprotein B . Virology 227:153-159. Salmivirta, M. , K. Lindholt, and U. Lindahl. 1996. Heparan sulfate: a piece of information. F A S E B . 10:1270-1279. 170 References Sambrook, J., E.F.Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N Y . Sarisky, R.T., and P. C. Weber. 1994. Requirement for double-strand breaks but not for specific D N A sequences in herpes simplex virus type 1 genome isomerization events. J. Vi ro l . 68:34-47. Schaffer, P., V. Vonka, R. Lewis, and M. Benyesh-Melnick. 1970. Temperature-sensitive mutants of herpes simplex virus. Virology 42:1144-1146. Schmidt, J., B. Klupp, A. Karger, and T.C. Mettenleiter. 1997. Adaptability in herpesviruses: glycoprotein D-independent infectivity of pesudorabies virus. J. V i ro l . 71:17-24. Schroder, C , and G.M. Keil. 1999. Bovine herpesvirus 1 requres glycoprotein H for infectivity and direct spreading and glycoproteins gH(W450) and gB for glycoprotein D -independent cell-to-cell spread. J. Gen. V i ro l . 80:57-61. Schroder, C. G. Linde, F. Fehler, and G.M. Keil. 1997. From essential to beneficial glycoprotein D loses importance for replication of bovine herpesvirus 1 in cell culture. J. V i ro l . 71: 25-33. Sears, A.E., B. S. McGwire, and B. Roizman. 1991. Infection of polarized M D C K cells with herpes simplex virus 1: two assymetrically distributed cell receptors interact with different viral proteins. Proc. Natl. Acad. Sci. U S A 88:5087-5091. Secchiero, P., D. Sun, A.L. de Vico, R. W. Crowley, M. S. Reitz, JR., G. Zauli, P. Lusso, and R.C. Gallo. 1997. Role of the extracellular domain of human herpesvirus 7 glycoprotein B in virus binding to cell surface heparin sulfate proteoglycans. J. V i ro l . 71:4571-4580. Sefarini-Cessi, F., and Campadelli-Fiume. 1981. Studies on benzhydrazone, a specific inhibitor of herpesvirus glycoprotein synthesis: size distribution of glycopeptides and endo-P-N-acetylglucosaminidase H treatment. Arch . V i ro l . 70:331-343. Sefarini-Cessi, F., F. Dall'Olio, M. Scannavini, F. Costanzo, and G. Campadelli-Fiume. 1983a. N-Acetylgalactosaminyltransferase activity involved in O-glycosylation of herpes simplex virus type 1 glycoproteins. J. V i ro l . 48: 325-329. Sefarini-Cessi, F., F. Dall'Olio, M . Scannavini, and G. Campadelli-Fiume. 1983b. Processing of herpes simplex virus 1 glycans in cells defective in glycosyl transferases of the Golgi system: relationship to cell fusion and virion egress. Virology 131:59-70. 171 References Shieh, M. , T D. WuDunn, RJ. Montgomery, J.D. Esko, and P.G. Spear. 1992. Cel l surface receptors for herpes simplex virus are heparan sulfate proteoglycans. J. Cel l . B io l . 116:1273-1281. Shieh, M.T., and P.G. Spear. 1994. Herpesvirus-induced cell fusion that is dependent on cell surface heparan sulfate or soluble heparin. J. Vi ro l . 68:1224-1228. Shizuya, H., B. Birren, U.J. Kim, V. Mancino, T. Slepak, Y. Tachiri, and M. Simon. 1992. Cloning and stable maintenance of human D N A in Escherichia coli using an F factor based vector. Proc. Natl. Acad. Sci. U S A 89:8794-8797. Silbert, J.E., and G. Sugumaran. 1995. Intracellular membranes in the synthesis, transport, and metabolism of proteoglycans. Biochim.Biophis. Acta. 1241:371-384 Smiley, M. L. and H.M. Friedman. 1985. Binding of complement component C3b to glycoprotein C is modulated by sialic acid on herpes simplex virus type 1-infected cells. J. V i ro l . 55:857-861. Sobel, M., D. F. Soler, J. Kermode, and R.B. Harris. 1992. Localization and characterization of a heparin binding domain peptide of Human von Willebrand factor. J. B io l . Chem. 267:8857-8862. Spear, P.G. 1993. Entry of alphaherpesviruses into cells. Sem. Vi ro l . 4:167-180. Spivack, J.G., and N .W. Fraser. 1987. Detection of herpes simplex virus type 1 transcripts during latent infection in mice. J. Vi ro l . 62:1479-1485. Stamatatos, L. , and C. Cheng-Mayer. 1993. Evidence that the structural conformation of envelope gpl20 affects human immunodeficiency virus type 1 infectivity, host range, and syncytium forming ability. 67:5635-5639. Stannard, L .M. , A . O. Fuller, and P.G. Spear. 1987. Herpes simplex virus glycoproteins associated with different morphological entities projecting from the virion envelope. J. Gen. V i r o l . 68:715-725. Stickens, D., G. Clines, D. Burbee, P. Ramos, S. Thomas, D. Hogue, J.T. Hecht, M. Lovett and G.A.Evans. 1996. The EXT2 multiple exostoses gene defines a family of putative tumour suppressor genes. Nat.Gen. 14:25-32. Stravropoulos, T., and C. A . Strathdee. 1998. A n enhanced packaging system for helper-dependent herpes simplex virus vectors. J. V i ro l . 72:7137-7143. 172 References Stuve, L.L. , S. Brown-Shimer, C. Pachl, R. Najarian, D. Dina, and R.L. Burke. 1987. Structure and expression of the herpes simplex virus type 2 glycoprotein gB gene. J. V i ro l . 61:326-335. Subramanian, G., D.S. McClain, A. Perez, and A.O. Fuller. 1994. Swine testis cells contain functional heparan sulfate but are defective in entry of herpes simple virus. J. Vi ro l . 68:5667-5676. Summerford, C., and R.J. Samulski. 1998. Membrane-associated heparan sulfate proteoglycan is a receptor for adeno-associated virus type 2 virions. J. V i r o l . 72:1438-1445. Sunstrum, J.C., C.E. Chrisp, M. Levine, and J.C. Glorioso. 1998. Pathogenicity of glycoprotein C negative mutants of herpes simplex virus type 1 for the mouse central nervous system. Virus. Res. 11:17-32. Swain, M.A., R.W. Peet, and D. A. Galloway. 1985. Characterization of the gene encoding herpes simplex virus type 2 glycoprotein C and comparison with the type I counterpart. 53:561-569. Swain, M.A. and D.A. Galloway. 1983. Nucleotide sequence of the herpes simplex virus type 2 thymidine kinase gene. J. V i ro l . 46:1045-1050. Takahashi, K., H. Nakanishi, M. Miyahara, K. Mandai, K. Satoh, A. Satoh, H. Nishioka, J. Aoki, A. Nomoto, A. Mizoguchi, and Y. Takai. 1999. Nectin/PRR: A n immunoglobulin-like cell adhesion molecule recruited to cadherin-based adherens junctions through interaction with afadin, a P D Z domain-containing protein. J. Cel l B i o l . 145:539-549. Tal-Singer, R., C. Peng, M. Ponce de Leon, W.R. Abrams, B.W. Banfield, F. Tufaro, G.H. Cohen, and R.J. Eisenberg. 1995. Interaction of herpes simplex virus glycoprotein gC with mammalian cell surface molecules. J. V i ro l . 69:4471-4483. Terry-Allison, T., R.I. Montgomery, J.C. Whitbeck, R. Xu, G.H. Cohen, R. J. Eisenberg, and P. Spear. 1998. H v e A (herpesvirus entry mediator A ) , a co-receptor for herpes simplex virus entry, also participates in virus-induced cell fusion. J. V i ro l . 72:5802-5810. Thiele, B., and F. Steinbach. 1994. Dextran sulfate induces a P K C and actin dependent internalisation. Immunol. Lett. 42:105-110. 173 References Tokuda, H., S. Masuda, Y. Takakura, H.Sezaki, and M. Hashida. 1993. Specific uptake of succinylated proteins via a scavenger receptor-mediated mechanism in cultured brain microvessel endothelial cells. Biochem. Biophys. Res. Commun. 196:18-24. Tompkinson, B., E. Robertson, R. Yalamanchili, R. Longnecker, and E. Kieff. 1993. Epstein-Barr virus recombinants from overlapping cosmid fragments. J. V i ro l . 67:7298-7306. Totsuka, A., A. Mukoyama, and I. Tagaya. 1981. Strain-specific aggregation of ,., enterovirus by dextran sulfate. Arch. V i ro l . 70:123-135. ' * f Trybala, E., T. Bergstrom, B. Svennerholni', S. Jeansson, jlc. Glorioso, and S. Olofsson. 1994. Localization of a functional site on herpes simplex virus type 1 glycoprotein C involved in binding to cell surface heparan sulfate. J. Gen. V i ro l . 75:743-752. Trybala, E, T. Bergstrom, D. Spillman, B. Svennerholm^ S- J- Flynn, and P. Ryan. 1998. Interaction between pseudorabies virus and heparin/heparan sulfate. J. B i o l . Chem. 273:5047-5052. Vahlne, A, B. Nystrom, M. Sandberg, A. Hamberger, and E. Lycke. 1978. Attachment of herpes simplex virus to neurons and glial cells. J. Gen. V i ro l . 40:359-71. Vahlne, A., B. Svennerholm, and E. Lycke. 1979. Evidence for herpes simplex virus type-selective receptors on cellular plasma membranes. J. Gen. V i ro l . 44:217-225. Vahlne, A. B. Svennerholm, M. Sandberg, A. Hamburger, and E. Lycke. 1980. Differences in attachment between herpes simplex type 1 and type 2 viruses to neurons and glial cells. Infect. Immun. 28:675-680. Van Drunen Little-van den Hurk, S., S. Khattar, S. K. Tikoo, L.A. Babiuk, E. Baranowski, D. Plainchamp, and E. Thiry. 1996. Glycoprotein H (gll /gpl08) and glycoprotein L form a functional complex which plays a role in penetration, but not in .attachment, of bovine herpesvirus 1. J. Gen. V i ro l . 77: 1515-1520. Van Zigl, M., W. Quint, J. Briare, T. de Rover, A. Gielkens, and A. Berns. 1988. Regeneration of herpesviruses from molecularly cloned subgenomic fragments. J. Vi ro l . 62:2191-2195. Visalli, R.J. and C .R. Brandt. 1993. The HSV-1 U L 4 5 18kDa gene product is a true late protein and a component of the virion. Virus Res. 29:167-178. 174 References Ward, P.L., G. Campadelli-Fiume, E. Avitabile, and B. Roizman. 1994. Localization and putative function of the U L 2 0 membrane protein in cells infected with herpes simplex virus 1. J. V i r o l . 68:7406-7417. Ward, P.L. and B. Roizman. 1994. Herpes simplex genes: the blueprint of a successful pathogen. T IG . 10:267-274. Warner, M. S., R. Geraghty, W. M. Martinez, R. I. Montgomery, J.C. Whitbeck, R. Xu. R. J. Eisenberg, G.H. Cohen, and P. G. Spear. 1998. A cell surface protein with herpesvirus entry activity (HveB) confers susceptibility to infection by mutants of herpes simplex virus type 1, herpes simplex virus type 2 and pseudorabies virus. Virology 246:179-189. Watson, R.J., J. H. Weis, J.S. Salstrom, and L.W. Enquist. 1982. Herpes simplex virus type 1 glycoprotein D gene: nucleotide sequence and expression in Escherichia coli. Science 218:381-384. Weller, S.K., A. Spadarp, J.E. Schaffer, A. Murray, A.M. Maxam, and P.A. Schaffer. 1985. Cloning, sequencing, and functional analysis of oriL, a herpes simplex type 1 origin of D N A synthesis. M o l . Cel l . B i o l . 5:930-942. Wenske, E.A., M.W. Bratton, and R.J. Courtney. 1982. Endo-[3-N-acetylglucosaminidase H sensitivity of precursors to herpes simplex virus type 1 glycoproteins gB and gC. J. Vi ro l . 44:241-248. Whealy, M.E., J.P. Card, R.P. Meade, A.K. Robbins, and L.W. Enquist. 1991. Effect of brefeldin A on alphaherpesvirus membrane protein glycosylation and virus egress. J. Vi ro l . 65: 1066-1081. Whitbeck, J . C , C. Peng, H. Lou, R. Xu, S. Willis, M. P. de Leon, T. Peng, A. V . Nicola, R. Montgomery, M. Warner, A. M. Soulika, L. A. Spruce, W. T. Moore, J. D. Lambris, P.G. Spear, G. H. Cohen, R. Eisenberg. 1997. Glycoprotein D of herpes simplex virus (HSV) binds directly to H V E M , a member of the tumor necrosis factor receptor superfamily and a mediator of H S V entry. J. V i ro l . 71:6083-6093. White, J. 1990. Vi ra l and cellular membrane fusion proteins. Annu.Rev. Physiol. 52:675-97. Whitley, R.J. 1996. Herpes simplex viruses. In: Fields' virology, 3 r d ed., B . N . Fields et al., Eds., Raven Press, N e w York, N . Y . pp.2297-2342 Whitley, R.J., D.W. Kimberlin, and B. Roizman. 1998. Herpes simplex viruses. Cl in . Infect. Dis. 26:541-555. 175 References Williams, R.K., and S.E. Straus. 1997. Specificity and affinity of binding of herpes simplex virus type 2 glycoprotein B to glycosaminoglycans. J. V i r o l . 71:1375-1380. Wittels, M. , and P.G. Spear. 1990. Penetration of cells by herpes simplex virus does not require a low pH-dependent endocyctic pathway. Virus Res. 18:271-290. WuDunn, D., and P.G. Spear. 1989. Initial interaction of herpes simplex virus with cells is binding to heparan sulfate. J . V i ro l . 63:52-58. Xu, X., H.G. Remold, and J.P. Caulfield. 1993. Potential role for scavenger receptors of human monocytes in the kil l ing of Schistosoma mansoni. A m . J. Pathol. 142:685-689. Yeung, S.N., K. Bockhold, and F. Tufaro. 1999. Efficient infection of mature skeletal muscle with herpes simplex virus vectors by using dextran sulfate as a co-receptor 6:1536-1544. Yuhasz, S.A., and J.G. Stevens. 1993. Glycoprotein B is a specific determinant of herpes simplex virus type 1 neuroinvasiveness. J. V i r o l . 67:5948-5954. Yoshitake, A . , X. Xuan, and H. Otuska. 1997. Identification and characterization of bovine herpesvirus-1 glycoproteins E and I. J. Gen. V i ro l . 78:1399-1403 176 Appendix I APPENDIX I: COSMID AND BAC TECHNOLOGY FOR GENERATING HSV RECOMBINANTS 7.0 INTRODUCTION Much of our understanding of the biology and molecular biology of the two H S V serotypes has come largely from studies of H S V - 1 . It is becoming increasingly evident, however, that HSV-1 and H S V - 2 are quite different from one another and that more research on H S V - 2 , in particular, is required. Studies on H S V - 2 are somewhat hindered, though, because of a general scarcity of H S V - 2 mutants and unfortunately, the process of constructing H S V recombinants is complex and often lengthy. The large viral genome of 152 kb is just one factor that complicates the manipulation of viral D N A . Conventional methods for construction of viral recombinants have involved co-transfection of cells with intact viral D N A and a plasmid which has been modified by insertion of a marker. Mutants arise by recombination and are isolated from wild-type virus by virtue of expression of the marker. Two new technologies have been developed which may expedite and simplify the construction of HSV-1 mutants. The first is a cosmid-based system, which is centered around a set of cosmids whose inserts overlap and represent the entire HSV-1 genome. When co-transfected into cells, viral plaques are produced via recombination between overlapping D N A fragments (Cunningham and Davison, 1993). Mutants are constructed by replacing a wild-type cosmid with a specifically mutated derivative. The second method relies on bacterial artificial chromosome ( B A C ) technology to clone the entire HSV-1 genome as a single molecule (Messerle et al, 1997; Horsburgh et al, 1999). Vira l recombinants are generated by using bacterial genetics to mutate the BAC-cloned virus genome. 177 Appendix I Given that the heparin-binding domain of gB was not required for D S stimulation, I wanted to determine which other domains of gB could be important for mediating this effect. This would require the generation of a variety of H S V gB mutants. One recombinant that I wanted to generate was an H S V - 2 particle expressing gB-1 in place of gB-2. This intertypic virus could then be used to test whether or not expression of gB-1 could restore D S stimulation of H S V - 2 . A s evident from the work presented in this chapter, it is often difficult to construct H S V gB mutants because gB is essential for viral replication. In light of this, we wanted to explore the possibility of developing a reagent, such as an H S V - 2 cosmid set or an H S V - 2 B A C , that would make the H S V - 2 genome more accessible to mutagenesis. For various reasons, construction of both the H S V - 2 cosmid set and B A C proved troublesome. The work described herein addresses particular problems that can arise when manipulating the H S V genome. A discussion of some of the advantages and disadvantages of the various systems used to construct H S V recombinants is included. 7.1 RESULTS 7.1.1 Construction of HSV-2 gB2\lacZ) virus To confirm further the role of gB-1 in mediating DS stimulation, I wanted to construct and later characterize an H S V - 2 intertypic recombinant expressing gB-1 in place of gB-2. Figure7.1 illustrates the plasmid-based strategy that was used for the construction of this mutant. The plasmid pHS208-/acZ, which encodes the E. coli lacZ gene flanked by gB-2 sequences, was linearized and co-transfected into gB-1 complementing D6 cells with H S V - 2 genomic D N A . Recombination between homologous sequences in pHS208-/acZand in the viral D N A would generate virus deficient in gB-2 and expressing lacZ. To select for recombinant virus, infected monolayers were stained with buffer containing X-gal . The appearance of blue-staining plaques indicated that the desired H S V - 2 gB knockout virus had likely been constructed. 178 Appendix I Co-transfect plasmid D N A and H S V - 2 viral D N A into gB complementing cell line Recombination of gB-2/'lac Z cassette with gB-2 sequences in viral genome J Isolate gB-2 knockout virus ilacZ-expressing viral plaques) Co-transfect H S V - 2 gB'QacZ) genomic D N A and linearized plasmid D N A into Vero cells Recombination between gB-2 sequences and insertion of gB-1 into H S V - 2 genome Isolation of H S V - 2 virus expressing gB-1 in place of gB-2 Figure 7.1. Strategy for construction of an H S V - 2 intertypic recombinant expressing gB-1 in place of gB-2 using a traditional plasmid-based strategy. 179 Appendix I This was also supported by the observation that /acZ-expressing plaques were not observed when non-complementing cell lines were infected with virus. The next step was to plaque purify the H S V - 2 gB" (lacZ) virus. Several attempts to do so, however, failed, simply because the size of the plaques produced by the mutant were too small to pick, i.e. after 3 days post-infection, the number of blue-staining infected cells in a given plaque ranged from 5-20. B y comparison, wild-type H S V can be expected to produce plaques containing approximately 100 - 200 infected cells. When plaques formed by the H S V - 2 gB2'(lacZ) virus were picked, wild-type H S V - 2 virus was always isolated instead. In an attempt to produce larger H S V - 2 gB" (lac Z) plaques, infection was allowed to proceed beyond the standard three days. This was to little avail, however, since wild-type H S V - 2 infection proceeded at a faster rate than the mutant, resulting in any uninfected cells being infected by wild-type H S V - 2 . I was unable to plaque purify the H S V - 2 gB'(lacZ) virus and could not, therefore, generate the H S V - 2 / g B - l intertypic virus according to the strategy originally proposed. The difficulties encountered using a plasmid-based method for constructing the H S V - 2 mutant underscored the need to develop a system that would facilitate the manipulation of the H S V - 2 genome. For this reason, attempts were made to develop an infectious H S V - 2 cosmid set. 7.1.2 Cosmid Technology for Generating H S V Mutants: Construction and Characterization of an HSV-2 Cosmid Library To develop a cosmid-based system for generating H S V - 2 mutants, an H S V - 2 cosmid library was first constructed. In characterizing the cosmid clones in the library, B g l II restriction analysis was used to approximate the locations of viral cosmid inserts with respect to the H S V - 2 genome. B g l II was chosen for restriction analysis because compared to other enzymes, B g l II digestion of wild-type H S V - 2 D N A yielded a banding pattern that was the most effective for mapping purposes. The B g l II restriction profile 180 Appendix I for HSV-2(G) is shown in Figure 7.2. The B g l II H , K , and M fragments are of lower intensity than the other fragments in the B g l II H S V - 2 restriction profile due to recombination in the repeat regions of the H S V - 2 genome. This recombination generates the four isomeric forms of the H S V - 2 genome, the Bg l II restriction maps of which are shown in Figure7.3. Approximately 200 cosmid clones were analyzed and as one might expect, viral fragments from the different isomeric forms of the genome were found. For simplicity, I studied cosmid clones in which viral inserts were of the prototype form. The results from B g l II restriction analysis are shown in Figure7.4. The generation of a functional cosmid set required cosmids to contain viral D N A that was all of the prototype form, with the exception of one cosmid which contained the B g l II D K fragment found the I L I S form of the genome. The latter cosmid was required in order to provide sequences that could overlap with D N A at the ends of the genome. To confirm that viral inserts overlapped with one another, the ends of selected cosmids were sequenced. The results from sequencing analysis are presented in Figure 7.5. These results are, however, obviously incomplete, in that for the majority of cosmids studied, only one end of the viral insert was sequenced. The reason for this is that by this point in the study, I was aware that there were inherent problems with the H S V - 2 cosmid library and the method by which I was characterizing it. First, it was extremely difficult by B g l II restriction analysis to distinguish between the sequences found in the repeat regions of the H S V - 2 genome and, moreover, to determine from what isomeric form of the genome they were derived. As shown in Figure 7.3, the H M (23 kb) and H K (24 kb) fragments, found in the prototype and Is (inverted short) isomers respectively, are of similar size. Likewise, it was difficult to distinguish between the D M (32 kb) and D K (33 kb) fragments found in the I L (inverted long) and I L I S isomeric forms, respectively. Thus, it was only when the ends of cosmids were later sequenced, that the viral inserts could be accurately mapped. Indeed, sequence analysis of cosmids pF2-35 and pF-98 (Fig. 7.5), demonstrated that the viral inserts did not contain the D K fragment, as 181 Appendix I O P Figure 7.2. Bgl II restriction profile of HSV-2 genomic DNA. HSV-2 genomic D N A was digested with Bgl II restriction enzyme overnight and the products were analyzed on a 0.5% agarose gel stained with ethidium bromide. Each viral D N A fragment is designated by a letter. For the corresponding HSV-2 Bgl II restriction maps, see Figure 6.3. 182 Appendix I HSV-2 (G) Viral Genome 20 40 60 80 100 1 i >0 140 D R P G u L J 0 l i C N I I I H u s j M Q L K ' I I I H 1 N c 1 1 0 J G i P R i 11 I D I I I I I M Q L K D RP II 1 G J 0 1 c I I I N I H I K L Q M H II 1 1 N c 0 J G P R D I K L Q M kb P Figure 7.3 B g l II restriction maps for all four isomeric forms of H S V - 2 (G) genomic D N A . The dashed line indicates the point of genomic inversion. (UL) unique long sequence, (Us) unique short sequence, (P) prototype form of the genome, (IL) inverted long sequence, (I s) inverted short sequence, (ILs) inverted long and short sequences. 183 Appendix I Figure 7.4 B g l II restriction analysis of HSV-2(G) cosmid clones. H S V - 2 cosmids were digested with B g l II restriction enzyme and the digestion products were analyzed by agarose gel electrophoresis and mapped accordingly. V i ra l cosmid inserts mapping to the prototype position of the viral genome are shown here. The different cosmid clones are desginated by a number. 184 Appendix I 2 0 4 0 6 0 8 0 100 120 140 I I I I I I I k b U L U s D R P G J O C N l H i M Q L K j i i_i i i i I 57 66 93 IZZZZ 1 8 19 ZZLZZ 2 8 29 69 40 42 86 88 59 80 60 297 65 105 77 87 23 24 33 81 51 56 52 5 16 39 62 82 39 47 61 64 97 20 46 100 P 25 30 8 9 98 2 55 I 1 7 " 2 6 54 48 185 Appendix I Figure 7.5 Mapping the H S V - 2 cosmids by sequence analysis. Each cosmid is designated by pF-2, followed by a number. Only the ends of viral inserts were sequenced. Bo ld crosses indicate that the sequence was obtained using the forward sequencing primer; regular crosses indicate that the sequence was obtained using the reverse primer. Asterisks indicate a discrepancy between the sequencing results and data obtained from B g l II restriction analysis. See text for details. 186 Appendix I 2 0 4 0 6 0 8 0 100 1 20 140 kb D R P G O N I H I MQL K 103,302 + 111,767 pF2-2 pF2-55 pF2-297* 151,500 pF2-30 - f " 141,678 + + 140,740 133,131 6,769 + 6,780 + pF2-82 pF2-62 pF2-61 67,451 + pF2-46 * 6,825 + pF2-77 49,519 + pF2-105 49,371 pF2-75 - \ ~ 41,230 + 85,472 + pF2-98 * 113,674 pF2-35 99,087 + 85,414 + pF2-20 76,513 + 187 Appendix I was suggested by restriction analysis (Fig.7.4). This particular problem was compounded by the fact that cosmids encoding the D K fragment would be under-represented in the library to begin with, simply due to the recombinogenic nature of the viral genome. Restriction analysis also proved inefficient for identifying cosmids encoding the B g l II D fragment (24 kb), since this fragment was also of similar size to fragments encoding the aforementioned repeat regions ( H M , H K , D M and D K ) . Thus, cosmid pF2-297, which by restriction analysis contained the B g l II D fragment (Fig. 7.4 ), was by sequence analysis, shown to have one end of its viral insert localized to the B g l K fragment (Fig. 7.5) In this instance, the B g l II D fragment may have been mistaken for either the Bg l II H K or D K fragment during restriction analysis. In addition to problems in identifying cosmids containing the B g l II D and D K fragments, difficulties were also encountered with cosmids mapping to B g l II fragments N , I and H . For example, restriction analysis clearly indicated that the cosmids pF2-46, pF2-62 and pF2-82, contained viral D N A mapping to Bg l II fragments N , I and/or H . (Fig.7.4) Sequencing the ends of these cosmids, however, suggested that these clones contained D N A mapping between, and partly including, the B g l II D and N fragments (Fig.7.5). This is a confusing result since the B g l II G , J, O and C, fragments were clearly not observed when these cosmids are analyzed by restriction digestion. Moreover, this observation implied that these cosmids possessed viral D N A inserts of up to 70kb! Vira l fragments of only 40 kb were inserted into the cosmid vector to begin with, and this is the upper limit of what lambda phage can package (C. van Sant, personal communication.) One way to account for this result is i f viral sequences in this region (Bgl fragments N , I, and H) are particularly recombinogenic. In this manner, the cosmids may have acquired part of the B g l II D fragment. This would likely not be detected by restriction analysis but would be detected when the cosmid ends were sequenced. In the cosmid system, a virus genome is reconstructed via homologous recombination between the overlapping sequences in the different clones. It is worthy to note, however, that the cosmid vector backbones are excised prior to co-transfection of 188 Appendix I the clones into cells. The presence of the cosmid backbone could interfere with recombination between overlapping viral sequences. For this reason, the cosmid cloning vector used in this study, P M S I , contained two Pme I sites which when cleaved, would release the viral D N A from the cosmid vector (Fig. 2.1). When H S V - 2 cosmids from the library were digested with P M S I , I expected to observe two bands by gel electrophoresis: a high molecular weight band representing the viral D N A insert and an 8 kb band representing the cosmid vector. The smaller band representing the cosmid backbone was not observed (Fig. 7.6A). One possibility to account for this was i f either one or both of the Pme I sites was not functioning. Sequence analysis using the 5' sequencing primer demonstrated that for one of the sites, the sequence was 5 ' - G T T T A A A G - 3 ' , instead of having the correct Pme I recognition sequence of 5 ' - G T T T A A A C - 3 ' (Fig.7.6B). I could show that the other Pme I site in the vector was functional by double digesting P M S I with Hind III (which cut at only one site) and Pme I. This yielded the two expected fragments of sizes 3.3 and 4.5 kb (Fig 7.6C). Since one Pme I site was functional, attempts were made to identify a second restriction site in the P M S I vector that could be used to release viral inserts from the vector. Unfortunately, there were no restriction enzyme sites in the vector which would also not digest the viral D N A insert (data not shown). Thus, viral inserts could not be released from the cosmid vector. For the reasons addressed above, I was unable to generate a set of H S V - 2 cosmids which, when co-transfected into cells, would yield infectious virus. This was a disappointing outcome because the development of such a system would have tremendous utility for generating mutant H S V - 2 viruses. A t this phase of the work, however, research by other investigators had demonstrated the feasibility to mutate the HSV-1 genome when maintained as a B A C in E. coli ( Messerle et al, 1997; Horsburgh et al, 1999). This prompted the construction of an H S V 2 - B A C . 189 Appendix I Figure 7.6. Impaired release of H S V - 2 viral inserts from the cosmid backbone. (A) H S V - 2 cosmid clones were digested with Pmel and analyzed by gel electrophoresis. The 8kb D N A fragment representing the P M S I cosmid vector is absent (indicated by the arrow). (B) Sequence analysis of an H S V - 2 cosmid clone using the forward sequencing primer. One of the Pme I sites in the P M S I cosmid vector is dysfunctional because of a single bp change, resulting in the sequence 5 ' - G T T T A A A G - 3 ' instead of the correct sequence of 5 ' - G T T T A A A C - 3 ' . (C) The second Pmel site in the P M S I cosmid vector is functional. P M S I was double digested with Pmel and Hind III and the products examined by gel electrophoresis. This yielded two expected fragments of 3.3 kb and 4.5 kb in length. 190 1 kb D N A ladder! P m e I digested H S V - 2 cosmids < * \ B Appendix I T C G A 191 Appendix I 7.1.3 Construct ion of HSV2-BAC The strategy for construction of the H S V 2 - B A C is outlined in Figure 7.7. The vector p B A C - T K was used to insert B A C sequences into the H S V - 2 genome at the thymidine kinase (tk) locus. The tk locus allows for identification of integration events by screening for acyclovir resistant plaques. It is worthy to note that the tk sequences in p B A C - T K are derived from HSV-1 (F). The HSV-1 and H S V - 2 tk genes are quite similar, varying by only 19% with regard to nucleotide sequence (Swain and Galloway, 1983). It was postulated that this degree of D N A sequence similarity was sufficient for recombination to occur between the HSV-1 tk sequences in p B A C - T K and the tk locus in H S V - 2 . p B A C - T K was linearized by digestion at the unique Hind III site and co-transfected with infectious H S V - 2 (G) D N A into Vero cells. Recombinant virus which was resistant to 50 u M A C V was harvested. The presence of B A C sequences in the viral genome was verified by P C R using primers which amplified chloramphenicol sequences in the B A C vector (Fig.7.8). A small amount of 600 bp product was produced, which was similar in size to that obtained when P C R was performed on the control p B A C - T K vector. In addition to the correct 600 bp product, P C R of control p B A C - t k also produced a smaller amount of lkb contaminating product. This 1 kb product was not observed in all P C R trials and its nature is not clear. To plaque purify a recombinant B A C virus, approximately 50 ACV-resistant plaques were picked and analyzed by P C R (data not shown). For some viral isolates, no P C R products were detected. For others, P C R of chloramphenicol sequences generated multiple products, instead of the expected 600 bp product. Nonetheless, a small amount of 600 bp product was observed for four isolates, in addition to unexpected larger molecular weight products. Due to the fact that non-specific sequences were being amplified in the P C R assay, I could not be certain that this 600bp band represented B A C sequences. To test directly i f these viral isolates encoded B A C sequences, circular D N A 192 Appendix I HSV-2 genomic DNA (tk+/ACV-sensitive) tk B A C vector u n i q u e site HSV-2 BAC (tk-/ACV-resistant) r i — B A C Figure 7.7 Construction of H S V - 2 B A C . A B A C vector containing B A C sequences flanked by HSV-1 tk sequences is linearized at a unique restriction site and co-transfected into cells with H S V - 2 genomic D N A . B A C sequences are integrated by homologous recombination into the tk locus in the viral genome. H S V - 2 B A C virus is selected on the basis of resistance to A C V . The B A C sequences also encode a chloramphenicol marker which is used for selection of H S V - 2 B A C . 193 Appendix I 1 kb D N A ladder 1 2 Figure 7.8. Identification of B A C sequences in HSV-2 B A C . The presence of the B A C vector in the HSV-2 genome was established using P C R primers which amplified chloramphenicol sequences. This figure shows the P C R products obtained from an ACV-resistant HSV-2 viral stock produced from co-transfection of HSV-2 genomic D N A and pBAC-tk. Lane 1 shows the PCR products when pBAC-tk is used as template. The 600 bp band represents cm sequences found in the B A C vector. Lane 2 shows results from P C R when D N A from the ACV-resistant HSV-2 B A C viral stock was used. A faint 600 bp band is visible, indicated by the arrow. 194 Appendix I from each of the four isolates was electroporated into E. coli and cells were plated onto agar plates containing chloramphenicol. Although this procedure was repeated several times, I was unable to isolate recombinant colonies resistant to chloramphenicol and, therefore, unable to isolate H S V 2 - B A C . 7.2 D I S C U S S I O N During the course of my investigation into the mechanism of DS-mediated H S V infection, it became desirable to construct an H S V - 2 intertypic virus in which gB-2 was replaced with gB-1. Initial attempts to construct this mutant were made using a plasmid-based method. This strategy relied on the generation of an H S V - 2 gB knockout virus as an intermediate. B y co-transfecting H S V - 2 D N A and a plasmid bearing the gB-2 gene modified by insertion of a lac Z marker, a virus that formed small, /acZ-expressing plaques on a gB-1 complementing cell line was generated. These plaques were likely formed by the H S V - 2 gB-(ZacZ) virus, although due to the small plaque size, the virus could not be purified. There are several possibilities to account for the difficulties I encountered when isolating the virus. The first, and most obvious, is that the high background of parental H S V - 2 plaques may have obscured the recombinant plaques, particularly since the mutant was disabled in its native gB and relied on gB-1 supplied in trans for infection. This in itself raises the question as to whether or not larger plaques would have been produced had the infection been carried out using a gB-2 complementing cell line, rather than the gB-1 complementing D6 cells. The observation that plaques expressing lacZ were only observed on complementing cells, and not on control Vero cells, does, however, suggest that gB-1 can complement the gB-2 defect. Nonetheless, there have been no studies to date which have examined the ability of gB-1 to functionally substitute for gB-2 and, therefore, it is not certain whether or not gB-1 can fully complement for gB-2 in this assay. 195 Appendix I Given the complexity of the viral genome, one must be wary when modifying viral sequences. The construction of plasmid pHS208-/acZ used in this investigation, involved the deletion of gB-2 sequences and then insertion of the lacZ cassette. B y doing so, the 2.7 kb open-reading frame (ORF) that runs antisense to the gB-2 O R F was also disrupted. A t the time of this investigation, it was not known that this O R F , now called U L 27.5, encodes a protein of 985 amino acids (575 amino acids for HSV-1) which accumulates in the cytoplasm of infected cells (Chang et al, 1998). The gB/U L 27.5 genes are the third set of genes in the H S V genome located antisense to one another. Although the function of U L 27.5 is not known, it is worthwhile to consider that the loss of this protein may have contributed to the small plaque phenotype of H S V - 2 gB" (lacZ) virus. The difficulties encountered in constructing H S V - 2 mutants using a plasmid-based method made the development of a cosmid-based system for H S V - 2 a particularly desirable objective. Cosmid libraries of V S V , H S V - 1 , and E B V have proven useful tools for the production of recombinant virus (Cohen and Seidel, 1993; Tompkinson et al, 1993; Kemble et al, 1996). The major advantage of the cosmid system is that it permits the isolation of mutant virus in the absence of wild-type virus, eliminating the need for multiple rounds of plaque purification. For several reasons, I was unable to generate an H S V - 2 cosmid set. First, and importantly, viral inserts could not be removed from the P M S I cosmid vector. This would obviously affect the reconstitution of an intact H S V - 2 genome in that during recombination between cosmids, the cosmid vector would also be incorporated. Had the P M S I vector contained the necessary restriction sites for excision of the viral insert, the generation of an H S V - 2 cosmid set would still have been hindered by the instability of particular viral sequences (Bgl II fragments N , I and H) in the cosmids. Other research groups have noted that some H S V cosmids maintained in E. coli are prone to deletion, recombination and rearrangement (Horsburgh et al, 1999). Particularly problematic is that the genome contains palindromic sequences that are often unstable in bacteria. The HSV-1 origin of replication, or i L , for example, consists of a large palindrome which is readily deleted when cloned in bacteria (Gray and Kaerner, 1984; 196 Appendix I Quinn and McGeoch, 1985; Weller et al, 1985). Hence, this region is unstable in the HSV-1 cosmids and the restriction pattern of serially passaged cosmid clones can alter significantly (Cunningham and Davison, 1993; Stravopoulos and Strathdee, 1998; Horsburgh etal, 1999) The heterogeneity of cosmid clones is a problem inherent to the cosmid system. Moreover, the reconstitution of viral genomes using cosmids relies on several recombination events in mammalian cells which are difficult to control. These properties are undesirable for the production of reproducible infectious viral D N A . B y contrast, herpesvirus sequences maintained as a B A C appear to be stable in bacteria (Messerle et al, 1997; Horsburgh et al, 1999). Horsburgh and colleagues (1999) propose that this may be due to the copy number of B A C s in bacteria: approximately one copy per cell. Another advantage of the BAC-system is that the starting material is homogeneous in nature, implying that the properties of H S V virion preparations that are produced from B A C D N A are consistent in their biological properties. In this study, attempts were made to construct an H S V - 2 B A C following the protocol of Horsburgh and colleagues (1999). A viral stock containing ACV-resistant virus which also encoded chloramphenicol sequences (derived from the B A C vector) was generated. This indicated that the H S V - 2 B A C virus had likely been produced. However, several attempts to plaque purify the recombinant B A C virus were unsuccessful, suggesting that perhaps the concentration of H S V - 2 B A C virus in the viral stock was rather low. A reduced concentration of H S V - 2 B A C virus could also explain why P C R of chloramphenicol sequences produced a minimal amount of product (Fig. 7.8A). These difficulties in producing H S V - 2 B A C virus may be due to the fact that the incorporation of B A C elements into H S V - 2 relied on recombination of HSV-1 tk sequences in the p B A C - T K plasmid with tk sequences in the H S V - 2 genome. Although the tk genes in HSV-1 and H S V - 2 have significant sequence similarity, the 19% nucleotide variation between them may be sufficient to reduce the efficiency of recombination of B A C sequences into the tk locus of the H S V - 2 genome. Another explanation for these results is 197 Appendix I that not all ACV-resistant (tk") virus contained B A C sequences. It is possible that during drug treatment, spontaneous ACV-resistant, tk" variants were generated. In the future, this problem could be avoided by integration of the B A C vector and a marker, such as lacZ, into another non-essential gene. Because the gene is not essential for replication in vitro, the insertion of B A C sequences does not interfere with analyses performed with recombinant viruses. In light of the success other labs have had using B A C technology to construct herpesvirus mutants, work is ongoing in our laboratory to develop an H S V - 2 B A C . This reagent w i l l be useful for the rapid generation of various H S V - 2 gB mutants which can then be used to help identify domains of gB involved in mediating DS stimulation of HSV-1 infection. The generation of viral recombinants requires that a vector containing the desired mutation be transformed into H S V - B A C containing bacteria. Following selection, the B A C D N A can be harvested from bacteria and transfected into mammalian cells. Recombinant virus can be harvested 2 days later. In this manner, Horsburgh and colleagues (1999) constructed an HSV-1 virus deleted for the viral packaging/cleavage sites within seven days. B y contrast, construction of herpesvirus mutants by the plasmid or cosmid-based methods can be quite challenging. In the cosmid system, fragments of the cosmid are subcloned until a manageable sized D N A fragment is obtained. A t this point, the appropriate mutation can be made and, the cosmid, in a series of cloning steps, can be reconstructed to contain the desired mutation. The cosmid and B A C systems represent cutting-edge technologies for the manipulation of H S V and other herpesviruses. These reagents clearly have advantages over the conventional plasmid-based method for constructing recombinants. The difficulties experienced in this study in constructing both an H S V - 2 cosmid set and an H S V - 2 B A C , however, demonstrate that these methodologies still have several shortcomings. Thus, it is perhaps best to view the different plasmid, cosmid and B A C systems as being complementary to one another, since each has different merits. For the construction of any given viral recombinant, the benefits and drawbacks of using each of 198 Appendix I these systems should be carefully considered. Choosing the most suitable method wi l l certainly facilitate the generation of the desired recombinant. 199 

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.831.1-0089674/manifest

Comment

Related Items