Eiki Sekine Herpesvirus Transport and Genome Presentation During The

Herpesvirus Transport and Genome

Presentation during the Initiation of Infection

Eiki Sekine

Section of Virology

Faculty of Medicine

Imperial College London

THESIS SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

2019

Declaration of Originality

I hereby declare that all work presented in this thesis has been produced by me, and the use of other material are appropriately referenced and acknowledged.

Eiki Sekine

2019

The copyright of this thesis rests with the author. Unless otherwise indicated, its contents are licensed under a Creative Commons Attribution-Non Commercial 4.0 International Licence (CC BY-NC).

Under this licence, you may copy and redistribute the material in any medium or format. You may also create and distribute modified versions of the work. This is on the condition that: you credit the author and do not use it, or any derivative works, for a commercial purpose.

When reusing or sharing this work, ensure you make the licence terms clear to others by naming the licence and linking to the licence text. Where a work has been adapted, you should indicate that the work has been changed and describe those changes.

Please seek permission from the copyright holder for uses of this work that are not included in this licence or permitted under UK Copyright Law.

Acknowledgements

I would like to thank my supervisor and mentor Professor Peter O’Hare for his guidance and support during the course of the project. It has been a great pleasure working under his wing and I have gained invaluable experience during my PhD training. I also thank past and current O’Hare lab members (Nora,

Remi, Sonia, Magda, Hanan, Alberto, and Jonathan) who made the lab an enjoyable place to work in.

I am especially grateful to Catherine Teo who showed me the ropes during my early years in the lab and has been a great colleague.

I extend my thanks to Dr Vanessa Sancho-Shimizu as my co-supervisor, as well as Dr John Tregoning and Professor Wendy Barclay for their advice as my progress review panel throughout my PhD.

I would also like to thank Dr David Gaboriau and Dr Andreas Bruckbauer for teaching me how to use the super-resolution and widefield microscopes at the Facility for Imaging by Light Microscopy (FILM)*.

David kindly carried out the processing of images obtained from the widefield Nikon Ti-Eclipse microscope.

Finally, I would like to thank the Wellcome Trust for their generous grant as part of the Molecular and

Cellular Basis of Infection 4-year PhD Studentship that allowed me to carry out this project.

*The Facility for Imaging by Light Microscopy (FILM) at Imperial College London is part-supported by funding from the Wellcome Trust (grant 104931/Z/14/Z) and BBSRC (grant BB/L015129/1).

Abstract

Events controlling herpesvirus nuclear genome uncoating, nuclear transport, and early transcription remain poorly understood. I have developed methods to quantitatively examine these processes within individual cells and at the level of single molecules for both the genome and the mRNA transcripts produced from it. Using bioorthogonal nucleoside precursors for genome labelling in HSV-

1 infected cells and quantitative single particle analysis, I demonstrate extremely efficient precursor incorporation resulting in quantitative genome detection on a single particle basis in the population of progeny virus. I report a comprehensive analysis in infected cells of genome dynamics during capsid exit and nuclear import in which I, (i) demonstrate qualitative transitions in genome condensation state linked to transcription and replication, (ii) reveal novel processes in genome congregation dependent upon transcription and (iii) show the temporal switching in regulatory protein recruitment

(represented by ICP4) to distinct genome compartments. Furthermore, I have developed the simultaneous visualisation of HSV-1 genomes and their product transcripts. This has enabled me to conduct spatiotemporal analysis of immediate early (IE) gene transcription kinetics in relation to the template genomes. I demonstrate a biphasic nature of ICP0 transcription, heterogeneity of genome transcription bursting, and differences in IE gene transcript behaviour with multiplex analysis, all of which have laid the foundations for further multivariate quantitative analysis. Finally, I have started to decipher the relationship of infectious genomes with host cell antiviral factors, demonstrating the association of PML and γH2AX with genomes, as well as increased viral activity in the absence of antiviral DNA sensors IFI16 and cGAS, and potentially a new mechanism for general suppression of incoming viral genomes. Altogether my results reveal novel aspects of the spatiotemporal dynamics of HSV-1 genome uncoating and transport, together with their relationship with nascent transcript that can be integrated with previous biochemically based analysis and provide a framework for future investigation of the virus genome and its relationship with the host cell.

Table of Contents

Declaration of Originality ...... 2 Copyright Declaration ...... 3 Acknowledgements ...... 4 Abstract ...... 5 Table of Contents ...... 6 List of Figures ...... 9 List of Tables ...... 11 List of Abbreviations ...... 12 1. Introduction ...... 16 1.1 Virus nuclear genome transport ...... 16 1.2 Herpes Simplex Virus ...... 19 1.2.1 Transmission and disease ...... 20 1.3 HSV-1 virion structure ...... 20 1.4 Life cycle, gene expression, and replication of HSV-1 ...... 24 1.4.1 Attachment and entry ...... 24 1.4.2 HSV-1 transcription, RNA processing, and gene expression ...... 25 1.4.3 Latency ...... 27 1.4.4 Genome replication, capsid assembly, and egress ...... 28 1.5 Genome nuclear entry...... 31 1.6 Viral genome control by chromatin ...... 33 1.7 Immediate-early (IE) proteins ...... 34 1.8 Intrinsic and innate host antiviral responses ...... 37 1.8.1 Viral DNA sensing ...... 37 1.8.2 PML nuclear bodies ...... 39 1.8.3 DNA damage response proteins ...... 40 1.9 Bioorthogonal precursors and click chemistry ...... 41 1.10 Transcription analysis at the single cell level ...... 45 1.11 Objectives ...... 48 2. Materials & Methods ...... 51 2.1. Cell Culture ...... 51 2.1.1 Cell lines and culture media ...... 51 2.1.2 Cultivation and propagation of cell lines ...... 51 2.2 Virus Culture ...... 52 2.2.1 Infection of cell lines ...... 52 2.2.2 Preparation of virus stock ...... 53

2.2.3 Production of virus labelled with ethynyl deoxycytidine (EdC)...... 53 2.2.4 Titration of virus stocks ...... 53 2.2.5 Plaque assays in the presence of EdC ...... 54 2.2.6 Measuring viral yield in the presence of EdC...... 54 2.3 HSV-1EdC on coverslips ...... 54 2.3.1 Adsorption and visualisation of HSV-1EdC on coverslips ...... 54 2.3.2 Quantitative counts of HSV-1EdC on coverslips ...... 55 2.4 Drug treatments ...... 55 2.5 EdC pulse labelling of host cell DNA and HSV-1 replication compartments ...... 55 2.6 Immunofluorescence studies ...... 56 2.7 HSV-1 capsid detection by western blot ...... 57 2.8 Click chemistry for microscopy imaging ...... 57 2.9 RNA detection ...... 58 2.9.1 Single molecule RNA detection by RNAscope...... 58 2.9.2 DNase treatment of samples before RNA detection...... 59 2.10 Widefield microscopy ...... 59 2.11 Confocal microscopy ...... 60 2.12 Structural Illumination super-resolution microscopy ...... 60 2.13 Quantitative analysis of intracellular genome uncoating ...... 61 2.14 Quantitative analysis of genome clustering ...... 61 2.15 Semi-quantitative analysis of RNA and protein abundance...... 62 3. Development and validation of EdC-labelled HSV-1 (HSV-1EdC) ...... 65 3.1 EdC affects neither RPE-1 cell growth nor HSV-1 infectivity and replication ...... 65 3.2 EdC is incorporated into replicating host cellular DNA and viral DNA ...... 68 3.3 Validation of EdC incorporation into the HSV-1EdC genome ...... 72 3.4 HSV-1EdC and HSV-1 wt have similar capsid to pfu ratios ...... 76 3.5 HSV-1EdC genomes are only detected after capsid release during cell infection ...... 78 4. Spatiotemporal Resolution of nuclear genomes during early infection ...... 84 4.1 Quantitative analysis of the relationship between MOI and HSV-1EdC genome entry ...... 84 4.2 HSV-1EdC infection reveals temporal alterations in genome compaction state ...... 87 4.3 Spatiotemporal relationship of genome compaction state and ICP4 expression ...... 90 4.4 Transcription and translation coupled transitions in infecting genome compaction state ... 93 4.5 HSV-1 genome transport is independent of proteasome, protease, and nuclear export activity 104 4.6 Visualisation of single genome infections with HSV-1EdC ...... 112

4.7 Spatial analysis of infecting genomes versus RNA Polymerase II ...... 116 5. Simultaneous analysis of HSV-1 genomes and single molecule transcripts ...... 120 5.1 Validation of single molecule RNA detection by RNAscope ...... 120 5.2 RNAscope and click chemistry can be combined to simultaneously detect HSV-1 genomes and transcript ...... 122 5.3 Kinetics of genome nuclear entry and compaction state in relation to ICP0 transcript abundance and localisation ...... 126 5.4 Analysis of IE gene transcription in the absence of translation and viral DNA replication .. 131 5.5 Simultaneous ICP0 and ICP4 transcript analysis suggests different spatial distributions and template genome usage ...... 140 5.6 Spatiotemporal analysis of early/late genes in relation to input genomes ...... 142 5.7 Detection of multiplexed transcripts from single genome infections ...... 146 6. Host antiviral responses to infecting viral genomes ...... 149 6.1 Genomes colocalise with PML bodies and with their association continuing in the absence of transcription ...... 149 6.2 Genomes do not show preferential association with IFI16 ...... 154 6.3 The absence of DNA sensing antiviral components in keratinocytes results in increased viral activity 157 6.4 Association of DNA damage sensors with genomes ...... 162 7. CHAPTER VII: Discussion ...... 167 7.1 Production and validation of HSV-1EdC ...... 167 7.2 HSV-1 genome nuclear entry ...... 170 7.3 Genome compaction state transitions during infection ...... 175 7.4 ICP4 association with infecting HSV-1 genomes ...... 177 7.5 Spatiotemporal analysis of HSV-1 genomes and IE gene transcript ...... 179 7.6 Multiplexed spatial analysis of HSV-1 genomes with ICP0 and ICP4 transcript ...... 185 7.7 Transcriptional bursting ...... 186 7.8 Future directions ...... 188 7.9 Concluding remarks ...... 189 8. References ...... 191 9. Appendix ...... 226

List of Figures

Figure 1.1 Schematic representation of the HSV-1 virion, genome, and lytic transcriptional programme ...... 23 Figure 1.2 Schematic representation of the HSV-1 life cycle ...... 30 Figure 1.3 Structures of bioorthogonal analogues used in click chemistry ...... 44 Figure 1.4 Schematic of the RNAscope assay procedure...... 47 Figure 3.1: EdC has no effect on RPE-1 cell growth ...... 66 Figure 3.2: EdC does not significantly affect HSV-1 infectivity, replication, or yield ...... 67 Figure 3.3: EdC is incorporated into replicating cellular DNA ...... 69 Figure 3.4: EdC is incorporated into HSV-1 replication compartments ...... 71 Figure 3.5: Visualisation of HSV-1EdC on coverslips ...... 74 Figure 3.6 HSV-1EdC and HSV-1 wt show similar capsid to pfu ratios ...... 77 Figure 3.7: Visualisation of HSV-1EdC during cell infection ...... 79 Figure 3.8 Control experiments for cellular HSV-1EdC infection ...... 81 Figure 4.1 Quantitative analysis of genome uncoating at 0.5 hpi ...... 86 Figure 4.2 3D-SIM analysis of genome decompaction ...... 89 Figure 4.3 Spatiotemporal relationship of genome decompaction and ICP4 expression ...... 92 Figure 4.4 Control experiments on effects of drug treatments ...... 93 Figure 4.5 Quantitative analysis of the effect of transcription inhibition on genome decompaction ... 96 Figure 4.6 Effects of inhibition of transcription, translation, and virus DNA replication on transitions in genome decompaction ...... 99 Figure 4.7 Infecting genomes in CHX treated cells show a strong tendency for clustering compared to genomes in Act D treated cells ...... 103 Figure 4.8 Effects of drug treatment on nuclear entry ...... 107 Figure 4.9 Investigating the effect of TPCK on HSV-1 nuclear genome entry and protein expression 111 Figure 4.10 Visualisation of single genome infections with HSV-1EdC ...... 115 Figure 4.11 Spatiotemporal relationship of genome decompaction and host RNA Pol II...... 118 Figure 5.1 Validation of ICP0 transcript detection by RNAscope ...... 121 Figure 5.2 Simultaneous visualisation of HSV-1 genomes and its transcript ...... 125 Figure 5.3 Spatiotemporal relationship between genomes and ICP0 transcript ...... 129 Figure 5.4 Transcriptional bursts are observed colocalising with nuclear genomes ...... 130

Figure 5.5 Assessment of inhibiting translation and viral replication on ICP0 transcription ...... 134 Figure 5.6 Assessment of inhibiting translation and viral replication on ICP4 transcription ...... 137 Figure 5.7 Assessment of the contribution of ICP4 to early IE gene transcription inhibition ...... 139 Figure 5.8 Simultaneous visualisation of HSV-1 genomes and ICP0 and ICP4 transcript show different intranuclear transcript distribution and heterogeneous genome template usage ...... 141 Figure 5.9 The spatiotemporal relationship between genomes and VP16 transcript ...... 145 Figure 5.10 Simultaneous visualisation of HSV-1 genomes and ICP0 and ICP4 transcript in single genome infections ...... 147 Figure 6.1 PML associates with a subpopulation of nuclear genomes ...... 153 Figure 6.2 Genomes do not show preferential colocalisation with IFI16 during early infection ...... 156 Figure 6.3 The absence of DNA sensing antiviral components in keratinocytes results in increased viral activity ...... 161 Figure 6.4 γH2AX localises proximal to HSV-1 genomes during early infection ...... 163 Figure 7.1 Model for HSV-1 genome dynamics in nuclear entry, compaction, and ICP4 association .. 174 Figure 7.2 Model for IE gene HSV-1 transcription dynamics in relation to genomes proposing a new mechanism of general viral transcription from incoming genomes ...... 184

List of Tables

Table 2.1 List of cell lines ...... 51 Table 2.2 List of viruses ...... 52 Table 2.3 List of drugs ...... 55 Table 2.4 List of primary antibodies used in immunofluorescence assays ...... 56 Table 2.5 List of secondary antibodies used in immunofluorescence assays ...... 57 Table 2.6 List of RNAscope probes ...... 59

List of Abbreviations

3D-SIM 3D Structural Illumination Microscopy Act D Actinomycin D ACV Acyclovir AdV Adenoviruses AMP Adenosine Monophosphate bDNA branched DNA Cdk1 Cyclin dependent kinase 1 cGAMP cyclic GMP-AMP cGAS cyclic GMP-AMP Synthase CHO Chinese Hamster Ovary CHX Cycloheximide CTD C-Terminal Domain DDR DNA Damage Response DMEM Dulbecco’s Modified Eagle’s Medium E Early EdC 5-ethynyl-2’-deoxycytidine EdU 5-ethynyl-2’-deoxyuridine ER Endoplasmic Reticulum FBS Foetal Calf Serum FISH Fluorescence in situ hybridisation gB glycoprotein B GMP Guanosine Monophosphate HaCaT Human Epithelial Keratinocytes HAT Histone Acetyl Transferase HBV Hepatitis B Virus HCF-1 Host Cell Factor 1 HDAC Histone Deacetylase HeLa Henrietta Lax HHV-1 Human Herpesvirus type 1 HIV-1 Human Immunodeficiency Virus type-1 Hsp90 Heat shock protein 90

HSV-1 Herpes Simplex Virus type 1 HVEM Herpes Virus Entry Mediator ICP Infected Cell Protein IE Immediate Early IF Immunofluorescence IFI16 Interferon γ-Inducible Protein 16 IFN Interferon Ig Immunoglobulin IN Integrase IRF3 Interferon Regulatory Factor 3 ISG Interferon Stimulated Genes ISH in situ hybridisation KS Kolmogorov-Smirnoff L Late LAT Latency Associated Transcript MA Matrix Mal MyD88 adaptor-like protein MeOH Methanol MHC-I Major Histocompatibility Complex Class I MN Micrococcal Nuclease MOI Multiplicity of Infection MyD88 Myeloid Differentiation Factor 88 NCS Newborn Calf Serum NE Nuclear Envelope NHEJ Non-Homologous End Joining NLS Nuclear Localisation Sequence NP Nucleoprotein NPC Nuclear Pore Complex ns not significant Nup Nucleoporin ORF Open Reading Frame PAA Phosphonoacetic Acid PAMP Pathogen-Associated Molecular Pattern PBS Phosphate Buffered Saline

PFA Paraformaldehyde PIC Pre-Integration Complex PML Promyelocytic Leukaemia Protein PML-NB Promyelocytic Leukaemia Protein-Nuclear Body Pol II RNA Polymerase II PP1B Protein Phosphatase 1B PRR Pattern Recognition Receptor P-TEFb Positive Transcription Elongation Factor b PTM Post Translational Modification PY Pyrin RC Replication Compartment RPE-1 Retinal Pigment Epithelial cells SDI Spatial Distribution Index STING Stimulator of Interferon Genes SV40 Simian Virus 40 Tap Transporter associated with antigen processing TBK-1 TANK-Binding Kinase 1 TBP TATA-Binding Protein TF Transcription Factor TPCK l-(tosylamido-2-phenyl) ethyl chloromethyl ketone ts temperature sensitive UL Unique Long US Unique Short Vero African Green Monkey Kidney Fibroblasts vhs virion host shutoff VP Virion Polypeptide Vpr Viral protein R vRNP viral Ribonucleoprotein γH2AX gamma-Histone 2 AX

CHAPTER ONE Introduction

1. Introduction

1.1 Virus nuclear genome transport

All viruses require a host cell for their replication, and many carry out this process in the nucleus. The environment in the nucleus provides abundant materials and cellular machinery that are required for DNA replication, transcription, and RNA processing [1], [2]. The nucleus also enables the virus to interfere and subvert host cell gene expression for its own benefit, while also offering ways to establish latency and integration of genetic material into host chromosomes. DNA viruses in particular have evolved wide ranging strategies to exploit, for example, nuclear DNA and RNA polymerases, transcription factors, and splicing factors [3]–[5]. RNA viruses do not generally replicate in the nucleus, but several do require nuclear events such as RNA splicing and genome integration [6], [7].

Despite the advantages of viruses using the host cell nucleus, it also presents a significant hurdle. The nuclear envelope (NE) presents a major barrier to infection and trafficking of viral components. Infecting viruses have evolved methods to deliver the genome and other necessary viral proteins into the nucleus in order to utilise the nucleus as a site of replication. This involves selective targeting, intracellular transport, and translocation across the NE of key viral components [8]–[10].

The mechanisms that viruses employ to enter the nucleus differ considerably between virus families.

The size of capsids and nucleoprotein complexes are the main factors that determines this, due to the size limit of cargo transport through the nuclear pore complexes (NPCs) [4].

All DNA viruses require their genomes to be deposited in the nucleus for replication to occur

(with one exception of poxviruses), and display several variations of NPC/NE engagement with genome translocation depending on class and size. DNA viruses with large capsids such as herpesviruses, adenoviruses, and hepatitis B virus (HBV) target nucleocapsids for transport to the NPC.

Here the nucleocapsids undergo partial or full disassembly, and genome uncoating takes place together with its translocation into the nucleoplasm. Herpesviruses are among the largest and most

16 complex viruses that replicate in the nucleus [11]. They are enveloped viruses that contain an isometric nucleocapsid of 120 nm in diameter. Herpes simplex virus type-1 (HSV-1), the prototypical member of the family, is the main focus of my investigations. Details of HSV-1 nuclear genome entry are described in Section 1.5. Briefly, after cytoplasmic entry by membrane fusion, the nucleocapsid is targeted to the NPC via microtubules. Here, a number of events trigger the portal of the capsid to open and the double-stranded DNA genome is threaded through the NPC into the nucleus.

Adenoviruses are large non-enveloped viruses between 60 to 90 nm in diameter and a linear double-stranded DNA genome. They are composed of up to 15 different structural proteins, and are complex and functionally sophisticated [12]. They have been extensively studied and provide one of the best understood models of virus nuclear entry [5], [13]–[15]. Adenovirus 2 capsids are partially dissociated during entry into the cytoplasm via endosomes, and subsequently recruit motor proteins dynein and kinesin-1. This allows movement of the capsid towards the NPC [16], [17] where it simultaneously binds with the NPC (via Nup214/Nup 358) and kinesin. Kinesin-1 then attempts to move away from the NPC while still attached to the capsid. The capsid in turn is attached to the NPC, and so results in the disintegration of the capsid by force and genome uncoating. This action also dislocates the Nup214/Nup358 complex from the central NPC channel [16]. The exposed adenovirus

DNA and is bound by several adenovirus proteins such as protein VII which is thought to have a histone-like DNA-condensing function and aids in the threading through of the genome through the

NPC as a nucleoprotein. Interaction with host proteins are also required to facilitate genome translocation, such as histone H1 which, upon binding to capsids stably docked at the NPC, recruits import factors importin 7 and importin β [18].

Hepatitis B virus (HBV), a prototypical member of the hepadnaviridae, is another DNA virus that replicates in the nucleus. It is an enveloped virus with an icosahedral capsid of 32 nm in diameter.

The protein shell is composed of 240 copies of a single protein known as the core protein [19], [20].

HBV, in the same theme as adenoviruses, use their own structural proteins to engage soluble nuclear

17 import receptors that mediate genome import into the nucleus. With HBV, the core protein is responsible for this and binds NPCs specifically [21]. However, in contrast to larger herpesviruses and adenoviruses, HBV is thought to be physically small enough to cross the NPC intact [22]. Current observations show that HBV capsids appear to traverse the NPC as whole particles until they encounter the NPC ‘basket’, where they disassemble prior to genome release into the nucleus [23].

Smaller DNA viruses such as parvoviruses and polyoma viruses possess capsids that are small enough to physically fit through the NPC. However, these viruses enter the nucleus by disrupting the

NE architecture, promoting transport of capsids across the disrupted membrane [24]. Recent evidence from studying parvovirus H1 indicates that binding to the NPC is necessary for NE disruption [25]. The capsid interacts with NPC proteins (including Nup358 and Nup153) which triggers a conformational change that exposes VP1, a viral structural protein. VP1 penetrates the nearby pore membrane, releasing Ca2+ ions from the NE lumen which in turn results in lamin phosphorylation, depolymerisation, and dissociation from the nuclear membrane [25]. These combined activities disrupt the NE and promote capsids transport into the nucleus. Polyoma viruses such as SV40 enter the cell by using endocytosis involving small flask-shaped invaginations known as caveolae [26]–[29].

The exact pathway is poorly understood, although it is thought that capsids move from endosomal compartments to the endoplasmic reticulum (ER) [27], [30]–[33]. During this process, capsids are partially disassembled exposing capsid proteins VP2 and VP3 which perforate the ER membrane and target the capsids to the nucleus via nuclear localisation signals, with subsequent transport of capsids through the NPC [34]–[36]. However, more recent evidence suggests that SV40 may enter the nucleus directly from the ER, bypassing the cytoplasm, by forming holes in the NE [37].

As mentioned previously, some families of RNA viruses require the nucleus for replication.

Human immunodeficiency virus type-1 (HIV-1) is a member of the lentivirus family of retroviruses and requires nuclear import of its genome for integration into the host DNA. Having entered the cell, genomic HIV-1 RNA is reverse transcribed into linear double-stranded DNA, which forms a key

18 component of a nucleoprotein complex known as the pre-integration complex (PIC). Three other components of this complex, namely integrase (IN), the matrix protein (MA), and the accessory protein viral protein R (Vpr) are implicated in PIC nuclear import [7]. The exact mechanism of this process is controversial, although it appears to be mediated by nuclear localisation sequence (NLS)- like sequences in the IN and MA proteins, and also the importin-β like properties of Vpr [9], [38]–[43].

Influenza viruses are another RNA virus family that relies on nuclear replication and transcription. They are enveloped animal viruses with a segmented, negative-sense RNA genome. There are eight distinct

RNA molecules that comprise the genome that are packaged into viral ribonucleoprotein complexes

(vRNPs) that contain multiple copies of nucleoprotein (NP) and a single copy of a heterotrimeric polymerase [6]. The size of the rod-shaped particles (up to 80 nm in length and 20 nm in diameter) is small enough for active passage of intact vRNPs through the NPC [44], [45]. All protein components of the vRNPs contain NLS type sequences and interact with importin proteins, although the available evidence suggests that sequences of the NP protein are most likely responsible for nuclear import of the viral genome [46]–[49].

Different virus families have evolved divergent strategies to overcome host barriers, and simultaneously exploit host cell machinery to achieve the ultimate objective of their own replication in the nucleus. It is an important step during the initiation of infection, and greater understanding of the exact mechanisms behind it are crucial in developing novel therapeutics to prevent viral infection.

1.2 Herpes Simplex Virus

The Herpesviruses are a family of double-stranded DNA viruses. They are classified into three subfamilies depending upon their genomic similarities; the Alpha-, Beta-, and Gammaherpesviruses.

They are among the largest and most complex viruses that replicate in the nucleus, and encode a wide repertoire of proteins that are involved in various metabolic processes. Replicated genomes are assembled into capsids in the nucleus and virions mature as they exit the cell through the cytoplasm.

Lytic infection results in the destruction of the host cell, although herpesviruses are also able to

19 establish life-long latency. Alphaherpesviruses are neurotropic with a relatively short replication cycle and rapid cell-to-cell spread resulting in cell lysis. Beta- and gammaherpesviruses on the other hand tend to have longer replicative life cycles with a slower progression of infection [11], [50], [51]. In addition, alphaherpesviruses infect a variety of host organisms and cell types, are able to establish latent infection in sensory nerve ganglia [52]. In contrast, betaherpesviruses can only infect a limited range of hosts and can establish latency in secretory glands, lymphocytes, and hepatocytes.

Gammaherpesviruses exhibit a restricted host range that include B cells and T cells [53]. HSV-1 is an alphaherpesviruses, and is also a human herpesvirus, a collective term for the 9 herpesviruses that infect humans [54]. HSV-1 is also known as human herpesvirus type-1 (HHV-1). HSV-1 is a prototypic model virus of the herpesvirus family, and is well characterised with respect to its host cell responses and infectious behaviour. Thus, it was selected as the main focus of my investigation.

1.2.1 Transmission and disease

The natural host cells of HSV-1 are epithelial cells, where primary lytic infection occurs, and sensory neurons where the virus can establish latency that can result in lifelong infections [55]. HSV-

1 infection is transmitted via physical contact and commonly causes localised mucocutaneous lesions

[56]. Oral and ocular lesions are the most common, although HSV-1 induced genital diseases are becoming increasingly more prevalent [57]. HSV-1 can cause more severe outcomes such as blindness, meningitis, and encephalitis [58], [59], with neonates and immunocompromised patients being particularly susceptible. The virus is a leading cause of fatal encephalitis and viral corneal blindness in more developed countries [60], [61].

1.3 HSV-1 virion structure

The HSV-1 virion is one of the largest and most complex viruses that has been structurally studied. It is comprised of four distinct structural layers; a glycoprotein containing envelope, a proteinaceous tegument layer, a capsid, and the DNA core [62] (Fig 1.1a). The virion is enveloped in a

20 host-cell derived lipid membrane and forms spherical particle of 186 nm in diameter. When the glycoprotein spikes are accounted for the full diameter is 225 nm [63]. There are 13 different glycoproteins that are identified in the membrane envelope and these are involved in various aspects of the viral life cycle, including entry, egress, and acquisition of the final envelope [64], [65].

The whole HSV-1 virion is composed of 40 proteins, both from viral and cellular origin, and of these 26 are associated with the tegument. Many of these proteins are critical for the virus, and one of their essential roles is to rapidly modulate the host cell environment to the advantage of the virus.

Examples include VP16 (transactivator that promotes viral gene transcription [66]), virion host shutoff protein (vhs, degrades mRNA and suppresses host cell protein synthesis [67]), and VP1/2 (mediates capsid transport to the nuclear pore and viral DNA entry into the nucleus [68]). There are also RNA- binding proteins (e.g. US11, UL47, and UL49) within the tegument that are thought to bind viral and cellular transcripts that are packaged into the virion [69].

The icosahedral capsid has a diameter of 120 nm and contains four principal constituent proteins. VP5 is the major capsid protein that forms the hexon and penton subunits which contain 5 and 6 copies of VP5 respectively. The capsid is composed of 162 capsomeres, which are in turn made up of 150 hexons and 12 pentons [70]. VP26 is an accessory protein located on the outer surfaces of the hexons, while VP23 and VP19C form a triplex that binds surrounding capsomeres for stabilisation

[71]. The capsid also contains the UL6 protein which forms the portal on one of the vertices on the capsid. It is through here the viral genome is packaged into the capsid during virion maturation, and also released during infection [72].

The HSV-1 genome contains approximately 94 open reading frames (ORFs) which in turn encodes at least 89 unique proteins, 10 long non-coding RNAs and several microRNAs [73]–[77]. The exact number of ORFs is still under debate and there is much speculation that there could be many more hidden ORFs. The genome is composed of a unique long (UL) and a unique short (US) region, and each are flanked by inverted repeats [78], [79] (Fig 1.1b). Of the currently identified genes, UL contains

56, while US only contains 12 [75]. Transcription of HSV-1 genes is enabled by the host cell RNA polymerase II (Pol II). HSV-1 protein nomenclature is based on one of three categories; genomic location of the encoding ORF (e.g., UL49), relative size on a polyacrylamide gel (e.g. virion polypeptide,

VP16), or infected cell protein (e.g. ICP4). Viral glycoproteins are designated by letters (e.g. gB) [50].

Figure 1.1 Schematic representation of the HSV-1 virion, genome, and lytic transcriptional programme (a) HSV-1 virion structure (reproduced from [80]). (b) Schematic of the HSV-1 genome which consists of long and short unique regions (UL and US) each bounded by terminal (T) and internal (I) repeat regions (RL and RS). The transcript map of the five IE genes is shown. There are two copies of ICP0 and ICP4, one in each repeat region. (c) Schematic of representation of the HSV-1 lytic transcriptional programme. The immediate early genes are transcribed first, with their products driving early gene transcription, which in turn facilitate viral DNA replication. Late genes are mostly transcribed from the replicated genome template (described in detail in Section 1.4.2). Figure modified from Gruffat et al. (2016) [81].

1.4 Life cycle, gene expression, and replication of HSV-1

1.4.1 Attachment and entry

The HSV-1 life-cycle can be divided into the following five steps; host cell entry, viral gene expression, DNA replication, virion assembly, and mature virion egress (Fig 1.2). The duration of the whole life-cycle lasts approximately 18 hours in permissive cell lines, although this varies with the host cell. Currently there are two proposed HSV-1 entry mechanisms. The first is the fusion of the viral envelope with the plasma membrane mediated by the interaction of viral glycoproteins with specific cell surface receptors. The second entry pathway is endocytosis of the enveloped virion which subsequently fuses with intracellular vesicles, and is utilised by the virus as an alternative pathway

[82]. Which pathway is used is cell type dependent [83]. In Vero cells, HSV-1 enters via fusion and delivers the capsid directly into the cytoplasm [84], [85]. In other cells such as Chinese hamster ovary

(CHO), HeLa cells and keratinocytes, HSV-1 enters via the endocytic pathway [86]–[88]. The attachment of the virion to the plasma membrane is mediated by the interaction of glycoproteins gC and gB with cell surface glycosaminoglycans, in particular heparan sulphate [89]. In the fusion pathway, the interaction between gD, gB, and the heterodimer gH/gL facilitates the fusion process [90], [91], where gD binds to one of its cell surface receptors. These are one of nectin-1, nectin-2, herpes virus entry mediator (HVEM), and 3-O-sulphated heparan sulphate.

Following the entry of the nucleocapsid, dissociation of the tegument occurs in the cytoplasm, but some tegument proteins such as VP1-2 and VP16 remain bound. These recruit the cellular microtubule motor dynein and its cofactor dynactin which facilitates microtubule-dependent transport of the capsid to the nuclear pore [84], [92]–[95]. VP1-2 contains an NLS [96] and interacts with importin-β to anchor the capsid to the NPC [97]–[99]. The minor capsid protein pUL25 associates with NPC protein complexes Nup358/RanBP2 and Nup214/CAN, and this interaction facilitates the capsid binding to the NPC [100]–[103]. Through a series of triggering events, the HSV-1 genome is

24 translocated into the nucleus (described in Section 1.5), where the genome is transcribed and replicated.

1.4.2 HSV-1 transcription, RNA processing, and gene expression

Transcription of the HSV-1 genome is carried out using the host cell RNA Polymerase II (Pol II) transcriptional machinery. The transcription cycle consists of discrete stages transcription initiation, followed by subsequent elongation and termination [104]–[108]. Pol II transcription typically starts with the binding of gene-specific regulatory factors near the site of transcription initiation. These factors interact with the transcription machinery, resulting in the recruitment of the transcription machinery to a core promoter [109], [110]. Pol II and at least five general transcription factors, namely

TFIIB, TFIID, TFIIE, TFIIF, and TFIIH, and assemble to form a pre-initiation complex before every round of transcription at this site [111]. The pre-initiation complex is initially in a “closed” state and is not in an active conformation to begin transcription. Subsequently, the pre-initiation complex undergoes dramatic conformational changes into to an “open” promoter complex, where around 11 to 15 base pairs of the promoter DNA are unwound to form a “transcription bubble” [112]. Multiple short RNAs of 3 to 10 base pairs may be produced, known as abortive products, before Pol II productively initiates synthesis of full length RNAs [113], [114]. After synthesis of around 30 base pairs of RNA, Pol II releases its contacts with the core promoter and the rest of the transcriptional machinery and enter the stage of transcription elongation [115], [116]. Phosphorylation of the Pol II C-terminal domain (CTD) is required for productive transcription elongation, and this is mediated by positive transcription elongation factor b (P-TEFb) which is composed of the catalytic subunit cyclin-dependent kinase 9

(cdk9) and its regulatory subunit, cyclin T [117]–[119]. Phosphorylation of Serine-2 in the heptapeptide repeats of Pol II CTD is recognised as a marker for productive elongation [120], [121]. Interactions between the promoter and enhancer proteins have been shown to fine-tune the elongation process

[122], [123]. Furthermore, topoisomerases and PARP1 ADP-ribosylate enzymes have also been implicated in regulating elongation [115].

The final step of the transcription cycle is termination, which results in the release of Pol II and

RNA from the DNA template [124]. The main proposed pathways of transcription termination in mammalian cells is poly(A)-dependent termination, where termination is coupled to RNA maturation.

Transcription of the poly(A) site of the gene results in pausing of Pol II and endoribonucleolytic cleavage of the nascent transcript, together with polyadenylation of the upstream cleavage product

[125], [126]. More than 14 proteins are thought to coordinate the termination process [127], [128], and while the exact mechanism is as yet unknown, CPSF and CstD are thought to be key Pol II termination factors that assemble on the Pol II CTD [124]. Once the nascent transcript is produced, it is processed into pre-mRNA, and as mentioned above, this maturation process occurs co- transcriptionally [129]. Other maturation processes include the addition of a 7-methyl G5’ppp5’N cap to the 5’ end of all Pol II transcripts when the nascent RNA is about 25 base pairs long, and is facilitated by guanylyltransferase and 7-methyltransferase [130], [131]. Splicing can also be carried out co- transcriptionally; it has been shown that introns can be removed by the spliceosome within 30 seconds of the transcription of the 3’ splice site [132], although post-transcriptional splicing has also been documented [133], [134]. The spliceosome consists of five uridine-rich snRNPs and many other proteins to form a large, dynamic complex that associates with Pol II via P-TEFb [135].

The HSV-1 gene expression programme is temporally coordinated and are categorised into three groups which are sequentially synthesised [136]; the α genes (immediate early, IE), β genes

(early, E), and γ genes (late, L) (Fig 1.1c). IE genes are the first set of viral genes transcribed during infection and they encode proteins which regulate the expression of E and L genes. The fastest rate of synthesis of IE proteins occurs approximately 2 to 4 hours after infection. Expression of IE genes is directly regulated by virion transactivator protein VP16 and occurs in the absence of de novo protein synthesis [137]. VP16 is transported into the nucleus independently of the capsid [138], [139] and activates IE gene expression by forming a three-component protein complex with host cell factor 1

(HCF-1) and transcription factor Oct-1 [139]–[142]. There are five genes categorised as true HSV-1 IE genes (ICP0, ICP4, ICP22, ICP27, and ICP47). All IE proteins except ICP47 activate transcription of the

26 subsequent group of E genes. E genes encode enzymes involved in nucleotide metabolism include thymidine kinase (ICP36) and ribonucleotide reductase (ICP6) and proteins that are essential for viral

DNA replication including DNA polymerase and the major single stranded DNA binding protein (ICP8).

E genes are further categorised into β1 and β2 subclasses according to their kinetic behaviour. The β1 proteins (e.g. ICP6 and ICP8) are synthesised first while β2 proteins (e.g. ICP36 and DNA polymerase) are synthesised later. Certain glycoproteins are dependent upon β gene expression (e.g. gD, gE, gI, gG)

[50]. The final group of L genes is also categorised into two further subclasses; the γ1 and γ2 genes.

The earlier expressed γ1 genes (e.g. VP5) are independent of DNA synthesis inhibitors, while the later expressed γ2 genes (e.g. gC, vhs) are strictly dependent upon viral DNA replication [81], [136], [143].

The production of L gene proteins, particularly structural proteins involved in virion structure and assembly, are a characteristic feature of a productive viral infection.

1.4.3 Latency

HSV-1 establishes latency in sensory neurons at the oral or genital mucosa. The process occurs during the delivery of the viral particle to the neuronal cell body via retrograde microtubule-associated transport. The establishment of latency relies on specific features of neurons that are important for the repression of lytic gene expression, in particular IE gene expression [144], [145]. The infecting virions travel relatively long distances through the neuronal axons, and so VP16 in the virion tegument is not efficiently delivered to the ganglion [146]. Thus, the formation of the VP16-induced trimeric protein complex is restricted. This is also due to limited HCF-1 and Oct-1 [147], [148]. In addition, the expression of latency-associated transcripts (LATs) plays a regulatory role in the establishment of latency. LATs are the only transcripts that are produced during latency [149], and a LAT-encoded miRNA is proposed to prevent neuronal apoptosis during latency establishment and/or reactivation

[150]. Additionally, it has been proposed that LATs inhibit IE gene expression, possibly by inhibiting

ICP0 synthesis by miRNA inhibition, together with the proposal that LATs downregulate IE mRNA levels via a transacting mechanism [151]. It is thought that LATs modify the chromatin on the HSV-1 genome

27 to promote stable, but ultimately reversible silencing [152]. The circularised HSV-1 genome is maintained as a circular episome during latency [153], [154], and the latent viral genome becomes rapidly associated with histones and is assembled into nucleosomes [155]. Reactivation from latency in neurons is facilitated by specific production of the HSV-1 regulatory proteins VP16 or ICP0 by cellular factors. Latency is reversed by a two-stage reactivation programme initiated by a pre-existing epigenetic switch [156], [157]. The first stage consists of a generalised burst of HSV-1 gene transcription that is independent of VP16 [157]. The second stage of reactivation closely resembles the cascade of viral gene expression that is observed during de novo infection, and involves chromatin remodelling and histone lysine modifications [158].

1.4.4 Genome replication, capsid assembly, and egress

The linear viral genome is circularised and the replicated by a rolling-circle mechanism [159] which produces concatemeric molecules that are cleaved during nucleocapsid assembly. Rolling-circle replication is facilitated by seven viral proteins (UL9, ICP8, UL5, UL8, UL52, UL30, and UL42) [160],

[161]. A small number of cellular proteins are also involved [162] including DNA ligase, topoisomerase

II, cellular chaperone heat shock protein 90 (Hsp90), and various components of the DNA repair and homologous recombination systems [160], [163] (also further described in Section 1.9). Virus-encoded enzymes such as thymidine kinase (UL23), ribonucleotide reductase (UL39, UL49), deoxyuridine triphosphatase (UL50), uracil N-glycosylase (UL2), and alkaline nuclease (UL12) play a role in nucleotide metabolism. These are necessary for viral DNA synthesis and repair due to the expression of the counterpart host cell enzyme being suppressed.

The viral DNA concatemers are cleaved and subsequently packaged into capsids that are formed autocatalytically. This is carried out with the aid of viral scaffolding proteins [164]–[166].

Intranuclear capsids bud through the inner nuclear membrane, facilitated by UL31, UL34, and US3, and the capsids acquire a primary envelope during this nuclear egress process [71], [167]–[169]. The nucleocapsids enter the cytoplasm via de-envelopment of primary enveloped particles at the outer

28 nuclear membrane [170], [171]. The final incorporation of tegument proteins and secondary envelopment subsequently takes place in cytoplasmic compartments. This results in a complex network of protein-protein interactions, where tegument proteins interact with both the capsids on one side and the cytoplasmic tails of envelope glycoproteins on the other. These interactions link the structural components for the final envelopment process and secure the integrity of the virion [172].

UL36 and UL37 proteins are required for active transport prior to secondary envelopment [173]. This occurs by budding of the inner tegument-coated capsids through membrane-associated outer tegument-glycoprotein structures [170], [171]. The egress of progeny virions is associated with microtubule-based transport which is facilitated by the interaction of UL37 with dystonin [174].

Mature virions exit the cell via the secretory pathway [175].

Figure 1.2 Schematic representation of the HSV-1 life cycle Reproduced from [176]. The HSV-1 virion attaches and fuses with the host cell membrane via a glycoprotein mediated process and releases the nucleocapsid into the cytoplasm. The nucleocapsid is then transported to the nuclear pore via microtubules where the genome is released into the nucleoplasm. The genome is transcribed and replicated according to the viral programme, resulting in the production of progeny nucleocapsids. These nucleocapsids then undergo envelopment and exit the cell through a series of budding and fusion events.

1.5 Genome nuclear entry

As previously described in Section 1.1, DNA viruses such as HSV-1 must deliver their genome to the nucleoplasm for its transcription and replication to occur. Once the virion has entered the cell, the nucleocapsid is transported to the NPC where it docks, and then the genome must be translocated through the NPC into the nucleoplasm. The exact mechanism for this in herpesviruses is poorly understood, and studies have been limited by difficulties in specifically detecting the input genomes.

Certain viral proteins have been implicated in this translocation process that include VP1-2, pUL25, and pUL6.

VP1-2 is an inner tegument protein encoded by the UL36 gene in HSV-1 that is conserved across the herpesvirus family. It is a multifunctional protein and while it has been well known that it plays many roles in the virus life cycle, including entry, transport, and assembly [177]–[182]. Early analysis of the HSV-1 temperature sensitive (ts) mutant tsB7 provided the first indication of an essential role for VP1-2 in HSV-1 replication [97], [138], [182], which has been supported by results showing the failure of VP1-2 negative capsids to spread infection between nuclei within syncytia [102].

Temperature-sensitive mutants generally exhibit a significant reduction in protein function when at its restrictive temperature, or non-permissive temperature. With HSV tsB7, infection carried out at the non-permissive temperature, viral gene expression does not take place and capsids were observed to accumulate at the nuclear pore without any genome release. Further analysis of this mutant definitively showed that this phenotype of temperature sensitive entry was in the UL36 gene, and was mapped to a single nucleotide substitution from tyrosine to histidine at position 1453 of the VP1-2 protein [183]. It has also been reported that VP1-2 is proteolytically cleaved when the nucleocapsid docks at the nuclear pore, and this process is necessary for viral genome translocation into the nucleus

[68]. Infection of cells with HSV-1 tsB7 in the presence of serine-cysteine protease inhibitor l-

(tosylamido-2-phenyl) ethyl chloromethyl ketone (TPCK) demonstrated that viral gene expression was dramatically decreased in the presence of the inhibitor. Observations of infection with the tsB7

31 mutant at the non-permissive temperature not resulting in any cleavage product and viral gene expression have led to proposals that the tsB7 phenotype during early infection is caused by the inability of the mutant VP1-2 to be cleaved [68], [183].

The minor capsid protein pUL25 (encoded by the UL25 gene) is another protein implicated to be involved in genome nuclear entry. Its main function is packaging the viral DNA into the capsid during nucleocapsid assembly and retaining it within the capsid [184]–[186]. It binds as a heterodimer with another minor capsid protein pUL17 around the capsid vertices [187] and also interacts with VP1-

2 [188]. In a similar fashion to the aforementioned tsB7 mutant, a ts mutant of the UL25 gene known as ts1249 showed a phenotype where nucleocapsids could enter the cytoplasm but could not subsequently release genomes into the nucleus at the non-permissive temperature [189]. It has been shown that pUL25 interacts with the nucleoporin hCG1 and CAN/Nup214, the latter of which is thought to act as a nuclear receptor for herpesvirus capsids [100]. VP1-2 and pUL6 is also bound by pUL25 when the nucleocapsid is bound at the NPC, thus it is thought to act as an interface between the two elements, and is proposed to be involved in the triggering process for viral DNA release into the nucleus. However, there are caveats in that these studies assessing nuclear genome release with

VP1-2 and pUL25 use surrogate measures such as viral gene expression for nuclear genome translocation, and do not use direct methods of genome detection.

pUL6 is the portal protein of the HSV-1 nucleocapsid. It exists at a unique site and forms a channel through which viral DNA can pass as it enters, and possibly exits, the capsid [184], [190]–[192].

Experiments in cells where nucleocapsids were treated with trypsin showed that pUL6 cleavage correlated with DNA release, with the proposal that portal protein cleavage is one of the processes required for nuclear genome release [193]. Potential viral proteins candidates for this cleavage include

VP24 (UL26 gene product) and VP1-2 itself.

1.6 Viral genome control by chromatin

Host nuclear DNA is organised in chromatin, where it wraps around nucleosomes that are composed of an octamer core of histones. Chromatin is essential to compact the long eukaryotic genome to fit in the nucleus, but it also regulates accessibility of proteins to the DNA, and consequently all processes requiring access to nuclear DNA. Regulation is largely mediated by post- translational modifications (PTMs) of histones (known as the ‘histone code’) which provide docking sites for proteins that regulate transcription and nucleosome stability [194]–[200]. Current evidence suggests that HSV-1 gene expression is also regulated by chromatin and histone PTMs. Using micrococcal nuclease (MN) digestion, it was shown that HSV-1 DNA in latently infected cells were inaccessible to a similar degree to inactive cellular genes, and HSV-1 DNA in lytic infections was highly accessible [155], [201]–[210]. Latent HSV-1 genomes are associated with mostly with PTMs associated with gene silencing with the notable expression of the LAT gene. In contrast, in lytic infections, highly transcribed genes are associated with PTMs that are distinctive of gene activation while non- transcribed genes had PTMs associated with gene silencing [211]–[213]. Specific histone variants that are associated with gene expression were consistent with viral transcription status where histone H3 variant H3.3 associated preferentially with transcribed viral genes [214]. The most recent model of the viral genome regulation by chromatin that reconciles the current evidence is that HSV-1 DNA is associated with unstable nucleosomes during lytic infection [204], [205], [215]. Furthermore, it is thought that infected cells mobilise histones from cellular loci on to incoming viral genomes and attempt to assemble silencing chromatin on them as part of an antiviral response [215]–[218].

However, viruses have evolved proteins that counteract this process to allow viral gene expression and genome replication, and modulate the PTMs to its advantage.

All three major HSV-1 transcription activators, VP16, ICP0, and ICP4, have been shown to be capable of indirectly inducing chromatin modifications and their stability. Histones are reversibly acetylated by histone acetyl-transferases (HATs) and histone deacetylases (HDACs). Their acetylation

33 neutralises charge-dependent interactions with DNA and adjacent nucleosomes, and is a marker of active gene transcription due to increased DNA accessibility. VP16 and ICP0 recruit HATs (e.g.

CBP/p300, PCAF, GCN5) to viral chromatin while disrupting HDAC activity [211], [219]–[227]. VP16 also recruits ATP-dependent chromatin remodellers BRG and BRM (part of the SWI/SNF family) to IE gene promoters resulting in increased gene expression [211], [220], [228]. Ubiquitination of histone H2A is associated with gene repression, and is carried out by cellular ubiquitin ligases RNF8 and RNF168. ICP0 is thought to degrade these ligases via the proteasome to prevent viral gene silencing by the host

[229], [230]. ICP4 also functionally interacts with two chromatin architectural proteins, HMGA and B

[231], [232], and disrupts the silencing of cellular heterochromatin together with ICP0, although the biological significance of this is still unclear [233], [234].

1.7 Immediate-early (IE) proteins

IE proteins play a multifunctional role that includes modulating host cell functions and subversion of host cell processes for the benefit of the virus. Four of the five IE proteins (ICP4, ICP27,

ICP0, and ICP22) play crucial roles in the regulation of HSV-1 gene expression. ICP4 and ICP27 are essential regulatory proteins in all experiments systems, both in vitro and in vivo. ICP0 and ICP22 are also important in regulating the viral gene expression, although they have been shown to be redundant in some models.

ICP4 is the major transcriptional transactivator of E and L gene expression [235], [236], but can also repress its own transcription in an autoregulatory manner as well as the other IE genes [237]–

[239]. It performs these functions by binding to a specific cognate DNA sequence [240]–[242] and modulating the activity of host cellular RNA Polymerase II (Pol II) on viral genes [243]–[245]. ICP4 is essential for the transition from IE gene transcription to the later stages of viral transcription [236],

[246]–[248]. The binding of ICP4 to the DNA is necessary [244] but insufficient for the activation of viral gene expression [249], [250]. ICP4 activates transcription through interactions with the basic RNA

Pol II transcription machinery [231], [250]. ICP4 forms a three-component complex on the DNA with

TATA-binding protein (TBP) and transcription factor IIB (TFIIB) [251], and promotes the formation of transcription pre-initiation complexes [252] by facilitating the binding of TFIID to the TATA box in the presence of TFIIB [253]. The current evidence is that the mechanism of gene expression activation by

ICP4 occurs at the very earliest stages of transcription. Various mutants have been generated to study the effects of depleting the ICP4 protein, and one of those mutants is the dl1403 mutant that has a 2 kb deletion in both copies of the RS1 gene encoding ICP4 [254].

In contrast to ICP4, ICP0 is not considered a classic transcriptional activator since it transactivates a wide range of cellular promoters without binding to DNA or cellular transcription factor recruitment [255]–[257]. Nevertheless, ICP0 can enhance the expression of any viral gene that exhibits at least a basal level of transcription [258]. Thus, ICP0 can further enhance the transcription of IE and E genes as these are transcribed at basal levels by the host transcription machinery.

Additionally, ICP4, ICP27, and ICP0 have been shown to functionally interact with each other, facilitating cooperative activation of viral gene expression [259]–[261]. ICP0 is not required for viral replication [254], [262], but it is necessary for productive infection in low-multiplicity infections [254],

[255], [263]. ICP0 also interacts with several host cell proteins, including cyclin D3 [264], transcription factor BMAL1 [265], translation elongation factor EF-1δ [266], ubiquitin-specific protease 7 (also known as herpes-associated ubiquitin specific protease) [267], and promyelocytic leukaemia nuclear bodies (PML-NBs) [268]–[270]. The RING-finger domain of ICP0, which confers E3 ubiquitin ligase activity to the protein, is implicated in the disaggregation of PML-NBs and the proteasome-dependent degradation of its constituents [271], [272]. These include PML and Sp100, especially their

SUMOylated isoforms [273]. In addition to PML-NB components, ICP0 has been shown to degrade several host proteins in a RING-finger dependent manner such as the catalytic subunit of DNA- dependent protein kinase and centromeric proteins C and A [274]–[278]. ICP0 has also been shown to be important in preventing the host cell interferon antiviral response from repressing the HSV-1 genome by impairing the activation of the interferon regulatory mediators that include interferon γ-

35 inducible protein 16 (IFI16), myeloid differentiation factor 88 (MyD88), and MyD88 adaptor-like protein (Mal) [279], [280].

ICP27 stimulates the transcription of E and L genes in a promoter-independent manner [281],

[282]. It is required for efficient DNA replication, and enables this by aiding the accumulation of E gene products. It also required for the transition from E to L gene expression by activating L genes while repressing IE and E genes [283], [284]. ICP27 is primarily functional at the post-transcriptional level, and is involved at all stages of mRNA biogenesis from transcription, RNA processing and export, through to translation. During early infection, it inhibits cellular pre-mRNA splicing [285], [286] by interacting with splicing factors, and this contributes towards the shutoff of host gene expression

[287], [288]. ICP27 also recruits RNA Pol II to viral transcription sites by interacting with its C-terminal domain (CTD) [289], [290]. Another property of ICP27 is that it shuttles between the nucleus and cytoplasm [291], [292], and it binds specifically to intronless mRNAs in both cellular compartments and promotes the nuclear export of viral intronless mRNAs [291]. This process is also mediated by the interaction of ICP27 with mRNA export adaptor proteins Aly/REF and SRp20, as well as cellular mRNA export receptor TAP/NXF1, which then translocates through the NPC [293]–[296]. ICP27 recruits cellular translation initiation factors upon shuttling to the cytoplasm, and thus facilitates translation of certain viral mRNAs [297]–[299]. ICP27 has also been involved in nuclear protein quality control

[300]–[302], cell cycle control [303], [304], activation of stress signalling pathways [305], [306], and inhibition of apoptosis [307]–[309].

ICP47 is an IE protein that binds with high affinity to transporter associated with antigen processing (TAP-1/TAP-2). This binding prevents TAP-1/-2 from transporting antigenic peptides into the ER for antigen presentation on the cell surface [310]–[312]. This aids the virus in evading host cell immunity by effectively blocking the major histocompatibility complex class I (MHC-I) antigen presentation pathway [313].

ICP22 is the final IE protein and its function is less clear than the other four members. It has been shown to modify RNA Pol II during productive infection, and induces the loss of serine-2 phosphorylation of the RNA Pol II C-terminal domain [314], a mark that is associated with productive elongation [120], [121]. It is proposed that ICP22 directly interacts with cyclin-dependent kinase 9 to inhibit RNA Pol II transcription elongation, thus inhibiting host gene expression during infection [315].

It has also been suggested that ICP22 enhances L gene expression by altering the activity of cyclin dependent kinase 1 (Cdk1) [316]–[319], and can also repress transcription of IE genes [320].

1.8 Intrinsic and innate host antiviral responses

1.8.1 Viral DNA sensing

Intrinsic, innate, and adaptive arms of the host immune system cooperatively suppress the replication and spread of infecting viral pathogens. Intrinsic immunity is the first line of intracellular defence against infection, and is conferred by constitutively expressed host-cell restriction factors

[321]–[323]. In contrast, innate immune effectors are upregulated following the detection of key pathogen-associated molecular patterns (PAMPs) by pattern recognition receptors (PRRs). These

PAMPs are usually molecules unique to pathogens including viral nucleic acids. PRR activation induces expression of host antiviral genes, principally cytokines (including interferons) and interferon stimulated gene (ISG) products [324]–[326]. This induced innate immune response confers an antiviral state in the cells that limits viral propagation and stimulates adaptive immune responses. As a result, many viruses have evolved methods to counteract host cell immune defences to successfully propagate to new host cells.

One of the first compartments the PRRs are able to detect viral DNA is the cytoplasm. Despite the HSV-1 genome being protected inside the capsid, digestion and damage of the capsid can occur, exposing the genome to cytoplasmic viral DNA sensors. For example, in certain cells types such as macrophages, proteolytic digestion of the capsid results in abundant viral DNA being released into the

37 cytoplasm available for detection [327]. One of the main cytoplasmic DNA sensors is cyclic GMP-AMP synthase (cGAS) which detects double-stranded DNA [328]–[331]. Upon binding to its target, cGAS is activated and uses ATP and GTP to produce cyclic GMP-AMP (cGAMP) through its enzymatic activity

[332]–[334]. cGAMP acts as a signalling molecule and in turn activates the downstream stimulator of interferon genes (STING). STING activation recruits the kinase TANK-binding kinase 1 (TBK-1) [335] which phosphorylates and activates the transcription factor interferon regulatory factor 3 (IRF3)

[336]–[338], inducing the production of interferon-β (IFN-β) [328], [329], [339]. However, cGAS DNA sensing has been shown to not be exclusive to the cytoplasm, and there is evidence for its role in nuclear DNA sensing where it interacts synergistically with interferon gamma inducible protein 16

(IFI16) [340]. HSV-1 has been shown to counteract this cGAS/STING-mediated interferon activation via the vhs protein which selectively degrades cGAS mRNA [341]. Tegument protein VP22 also abrogates the same pathway by inhibiting cGAS enzymatic activity [342].

Previously it was thought that most intracellular surveillance of viral DNA occurred in the cytoplasm and endosomal compartments in order to prevent false recognition of host cell DNA as foreign. However, recent studies have established the existence of cellular DNA sensors that detect viral DNA within the nucleus to trigger immune signalling. One of the key nuclear DNA sensors that has been identified is IFI16. It was initially discovered as a cytosolic DNA sensor [327], [343], [344], but has since been found to detect nuclear viral DNA as well [345]–[348]. It is member of the interferon- inducible human PYHIN protein family [349], [350], and has two C-terminal HIN200 domains that bind to DNA in a sequence-independent manner [351], together with an N-terminal pyrin (PY) domain that mediates assembly of IFI16 oligomers [348], [352]. IFI16 directly binds and detects incoming viral DNA, and signals to activate STING by an as yet unknown mechanism, resulting in the production of IFN-β via IRF-3 signalling [343], [353]–[359]. IFI16 has also been proposed to induce innate signalling via an

IFI16 containing oligomeric signalling platform termed as an ‘inflammasome’ [345], [360], [361]. It has also been reported that IFI16 can suppress viral gene expression more directly by introducing repressive chromatin structures onto the HSV-1 genome [354], [362], although it is still unclear if this

38 is directly induced by IFI16 or a consequence of the downstream IRF-3 signalling. There is much controversy surrounding the spatiotemporal dynamics of IFI16 detection of viral DNA. It has been reported that IFI16 localisation on viral genomes is transient and occurs in multiple phases upon initial nuclear entry [363]–[365]. Others have reported that IFI16 does not preferentially localise to viral genomes during the very earliest stages of infection, and requires viral gene expression or genome replication for IFI16-genome association [366]. ICP0 has been shown to counteract the effects of IFI16 genome detection, and is thought to inhibit IFI16 recruitment onto viral genomes [367], as well as contributing to IFI16 degradation via the proteasome [347], [368].

1.8.2 PML nuclear bodies

PML nuclear bodies (PML-NBs), also known as ND10 bodies, are proteinaceous subnuclear organelles that have been implicated to be prominent factors in targeting incoming nuclear genomes for silencing. They range in size from 0.1 to 1 µm in diameter, numbering approximately 7 to 40 bodies per nucleus [369]. PML protein serves as a scaffold protein which recruits over 60 other proteins that interact and localise with PML proteins, resulting in the assembly of PML-NBs [370], [371]. Major components include Sp100, human death associated protein (hDaxx), ATRX, and small ubiquitin-like modifier (SUMO) proteins. SUMOylation of PML and other components are essential for PML-NB formation and function [372]. PML bodies regulate several cellular processes including proliferation, senescence, apoptosis, DNA damage [373]–[376], and intrinsic antiviral activity that target both DNA and RNA viruses [377], [378]. PML bodies are thought to also function as early sites for viral transcription and replication. It has been demonstrated that PML bodies constitute a preferential site for infecting HSV-1 genome localisation and subsequent replication compartment (RC) development

[379]–[381]. Incoming HSV-1 genomes associate with ICP4 which is recruited into foci adjacent to PML bodies [382], which in turns nucleates the formation of PML-NB-like bodies [383]. Association of HSV-

1 genomes with PML bodies requires ICP4 and ICP27 and is mediated by the interaction of these proteins with hDaxx [384]. It is thought that both ICP4 and ICP27 and their association with PML-NBs

39 contributes to active viral gene expression. Newly synthesised ICP0 colocalises with PML bodies and promotes degradation of SUMOylated forms of PML and Sp100 proteins [272], [273], [385], [386], resulting in the dispersal of PML-NBs. This disruption correlates with the initiation of HSV-1 lytic gene expression [387]. The targeting of PML for degradation is part of the virus overcoming host antiviral defences, with viral gene expression and yield being significantly increased in PML depleted cells in the presence of IFN production [388]. Thus, ICP0 is essential for resistance of HSV-1 to IFN-α/β [279].

There is also evidence to suggest that IFI16 also contributes to PML-NB mediated restriction of HSV-1

[367].

1.8.3 DNA damage response proteins

The cellular DNA damage response (DDR) is a complex network of signalling pathways that safeguards cellular DNA to maintain genomic integrity. This can potentially be compromised during replication, as well as endogenous damage and exogenous agents. DDR activation is triggered by damaged DNA sites such as DNA double-stranded breaks (detected by host cell kinases ATM and DNA-

PK), accumulation of single-stranded DNA, and stalled replication forks (detected by host cell kinase

ATR) [389]. Recognition of DNA damage results in the recruitment of cellular factors to repair the damaged sites, and is coordinated with arrest of the cell cycle [390]–[393]. Recent work has shown that host DNA repair proteins are able to recognise viral DNA [394], and HSV-1 infection is influenced by the DDR [395]. During lytic infection, HSV-1 recruits several cellular DNA repair and recombination proteins into viral RCs, and examples such as Mre11, ATM, ATR/ATRIP, and WRN have been shown to enhance viral replication [395]–[398]. However, other components are thought to have putative roles in the host antiviral response. For example, DNA-PKcs and Ku70, which are proteins involved in non- homologous end joining (NHEJ), have been shown to be detrimental to HSV-1 replication [228], [277].

Ubiquitin ligases involved in DNA repair, RNF8 and RNF168, are thought to be part of intrinsic antiviral defence. They ubiquitinate histone H2A in viral chromatin which results in less accessibility to the viral

40 genes, and hence suppression of viral gene expression. ICP0 has been shown to degrade these ligases via the proteasome as a countermeasure [229], [230], [399].

One of the DDR components involved in the complex virus-host interactions is histone variant

H2AX. It constitutes approximately 10% of the total histone H2A of chromatin of a typical cell. The exact functions of H2AX are still poorly understood, however, H2AX is phosphorylated by ATM and

ATR in response to DNA damage at sites up to several kilobase pairs proximal to the damaged site

[400]. During HSV-1 infection, H2AX is phosphorylated during viral IE gene expression, and the phosphorylated form, known as γH2AX, is observed at sites associated with incoming viral genomes during early infection [399]. The amount of H2AX increases throughout the HSV-1 transcriptional programme [401]–[403]. It is not clear what the trigger is for H2AX phosphorylation, although the possibilities include the free DNA ends from the infecting linear viral genome, or replication and recombination of the viral genome that will cause the appearance of single-strand nicks and gaps [404].

The exact mechanism and function of H2AX phosphorylation in the context of viral infection is yet to be elucidated, although it is proposed H2AX is required for efficient viral replication [399], [405], while

H2AX phosphorylation is suggested to be a host response to HSV-1 infection that has little impact on the progression of viral infection [404].

1.9 Bioorthogonal precursors and click chemistry

Quantitative data at the single particle level on localisation, uncoating, and transport of the infecting genome is critical for fully understanding virus infection and pathogenesis. Factors which have limited the quantitative spatiotemporal analysis of genome transport and presentation are the complexity and lack of sensitivity of methods to directly visualise and measure viral genomes. In particular, the inability to differentiate genomes that are encapsidated from those that have dissociated, the inability to readily differentiate incoming from replicated genomes, and the incompatibility of certain detection methods with immunohistochemistry for simultaneous detection of host and viral proteins. One of the most frequently used techniques, fluorescence in situ

41 hybridisation (FISH), has provided insights, but still presents several limitations [406]. FISH can neither discriminate input from replicated genomes nor distinguish between encapsidated genomes versus released genomes due to the harsh treatment conditions. Other methods such as the incorporation of multimerised binding sites for fluorescent DNA binding proteins such as YFP-TetR [407] offer the potential for live cell imaging, but require specialised recombinant viruses or cells lines and can still be limited in detecting infecting genomes. These limitations and issues of complexity and sensitivity have meant that many studies on aspects of virus infection have almost invariably relied on indirect, surrogate measures of viral genome detection. Previous investigations, for example, on the effect of inhibitors on DNA entry or analysis of DNA sensing, have often inferred effects on genome localisation from DNA binding protein localisation. Genomes that are not bound by these surrogate markers will not be detected for a multitude of reasons, including unknown but specific differences between genomes, occlusion by other factors, spatial segregation dictating protein association, repression by chromatin, non-nuclear localisations where surrogate markers may not colocalise. Such assays also disclose little information on other aspects of genome uncoating and the morphological states of infecting genomes. These and other considerations limit the current understanding of genome entry, uncoating, nuclear translocation, and physical transitions, all of which are necessary to elucidate the exact mechanism and processes governing viral infection and host responses.

In this regard, the development of bioorthogonal metabolic precursors combined with cycloaddition to corresponding capture reagents is increasingly being exploited in various approaches to biological processes, including mechanisms in infection and immunity [226], [406], [408]–[410].

Specifically, alkyne-derivatised nucleosides that are incorporated into host cell macromolecules are covalently attached to azide-coupled fluorochromes, making use of copper(I)-catalysed Huisgen 1,3- dipolar cycloadditions [411], [412], and these reactions are commonly known as ‘click chemistry’ [413],

[414]. This Huisgen cycloaddition involves the reaction of a 1,3-dipole (e.g. azides, nitriloxides, or diazoalkanes) with a dipolarophile (e.g. alkenes, alkynes, or carbonyls) via a concerted mechanism to form a five-membered heterocycle [415], [416]. Within click chemistry, the cycloaddition between an

42 the azide-labelled fluorochrome and an alkyne-labelled macromolecule produces a 1,2,3-triazole ring, covalently attaching both molecules together (Fig 1.3).

Analysis of DNA synthesis by click chemistry has been evaluated in several systems [226],

[417]–[419] and has been used in spatial, biochemical, and systems approaches to investigate DNA replication and the cell cycle. Alkyne-derivatised nucleosides used in these studies include the deoxythymidine analogue 5-ethynyl-2’-deoxyuridine (EdU) [418], [420] and deoxycytidine analogue 5- ethynyl-2’-deoxycytidine (EdC) [226], [421]. Other optimised nucleoside analogues are available, such as (2’S)-2’-deoxy-2’fluoro-5-ethynyluridine (F-ara-EdU) [422]. Click chemistry and bioorthogonal labelling techniques have recently been exploited for analysis of adenovirus (Adv) infection [406] using viruses incorporating EdU and EdC in their genomes for the spatiotemporal investigation of genome trafficking at the single particle level. De novo HSV DNA synthesis has also been analysed using EdC incorporation [423].

Figure 1.3 Structures of bioorthogonal analogues used in click chemistry The schematic illustrates the in vivo incorporation of EdC into genomes (solid black dots within the polynucleotide), and the subsequent in vitro cycloaddition reaction to covalently crosslink an azide fluorochrome-coupled capture reagent (green stars) to the alkyne-modified analogues. The structure of the unmodified deoxycytidine is shown for reference. The structures of alternative nucleotide analogues to EdC (EdU and F-ara-EdU) are also shown.

1.10 Transcription analysis at the single cell level

Many cellular processes, including viral infection and the host antiviral response, are the result of the changes in the overall rate of transcription in genes. Thus, the ability to measure gene transcriptional activity allows the investigation of changes in gene output and how they influence and carry out these cellular processes. Transcription of RNA from genes, both host and viral, is a highly regulated process, occurring at multiple levels. Promoter accessibility, chromatin architecture, and transcription factor (TF) recruitment being some of the many variables that are involved in transcriptional regulation [424]–[431]. Traditional methods of studying RNA transcriptional output from genes were carried out on a population level, such as northern blots and reverse-transcription

PCR (RT-PCR). Although useful, this masks any individual cell variation and does not give any information on cell specific responses, for example in tissues where there are multiple cell types. Thus, gene transcription was thought to occur at a continuous and smooth rate, resulting in a Poisson-like distribution of transcript per cell [432]. Early single-cell studies on gene output relied on protein output and could only measure the combined transcriptional and translational output [433], [434].

More recently, studies have used single cell and single molecule techniques, such as fluorescence in situ hybridisation (FISH) of RNAs, to analyse gene transcripts and their kinetics. These studies have all pointed towards a model where genes fluctuate between a transcriptionally active ‘on’ state and a refractory ‘off’ state, resulting in stochastic production of RNA in both eukaryotes and prokaryotes.

This process has been termed either transcriptional bursting or the random telegraph model of gene expression [432], [435]–[445].

Transcriptional bursting has been widely observed across the human genome independently of proximal sequences and is not confined to specific gene loci [443]. The exact mechanism behind transcriptional bursting is still unknown, although at least in E. coli it has been proposed that bursting occurs as a results of DNA topology, where supercoiling of the DNA after transcription causes refractory periods [446]. Regulation and kinetics of transcriptional bursting is currently intensively

45 studied, and the burst size and frequency are two factors that can be influenced to regulate gene transcription [447]. It has been reported that interaction of cis-elements such as enhancer-promoter chromatin looping can contribute to regulation by controlling bursting frequency [436], while gene length is inversely correlated with burst size (but not burst frequency) [448]. Histone acetylation at the gene promoter can enhance transcription by increasing the burst frequency [449], demonstrating that chromatin modulation can influence bursting as well. Regulation from trans factors such as TFs have been shown, for example, a system measuring c-Fos transcription has shown that increased TF binding on the promoter increases burst size, while TF concentration correlates with burst frequency

[444]. More recent studies have attempted to reconcile the transcriptional bursting model with the paradigm of transcriptional control by RNA polymerase recruitment and promoter-proximal pausing and release [450]. In this study it was found that burst initiation is required before polymerase recruitment, and biological stimuli changed only burst frequency and polymerase pause release rates.

Other studies have proposed that transcriptional bursting activity of genes occurs across a spectrum of activity rather than genes having a binary on or off state [451].

As previously mentioned, in order to obtain data of cell transcriptional heterogeneity, single cell and single molecule analysis of transcripts is required. Traditionally, RNA FISH has been used for such purposes. A recent development of this in situ hybridisation (ISH) technique has been the detection of RNA molecules with amplification of the target probes with branched DNA (bDNA), known as bDNA ISH for RNA, and marketed as RNAscope [452], [453]. A novel probe design strategy is used where, in addition to the 18 to 25-base sequence complementary to the target RNA, each probe also contains a spacer sequence and a 14-base tail sequence (conceptualised as a Z-shape). A pair of target probes (double Z), each containing a different tail sequence, hybridise continuously to a target region of approximately 50 bases long. The two tail sequences together form a 28-base hybridisation site for the preamplifier. This preamplifier contains 20 binding sites for the amplifier, which in turn contains 20 binding sites for the label probe. The double-Z design of the probes ensures high specificity and low background since it is highly unlikely that a nonspecific hybridisation event

46 would juxtapose a pair of target probes along an off-target RNA molecule to form the 28-base hybridisation site for the preamplifier. The large amplification step also ensures high sensitivity, and this enables single molecule sensitivity and specificity for visualising RNA [452]. This bDNA ISH methodology for detecting RNA has been successfully utilised to carry out sensitive studies of single

RNA molecules [454]–[456].

Figure 1.4 Schematic of the RNAscope assay procedure Reproduced from [452]. Step 1- Cells or tissues are fixed and permeabilised to allow target probe access. Step 2- Target RNA-specific oligonucleotide proves (Z) are hybridised in pairs (ZZ) to multiple RNA targets. Step 3- Multiple signal amplification molecules are hybridised, each recognising a specific target probe, and each unique label probe is conjugated to a different fluorophore. Step 4- signals are detected using fluorescent microscopes.

1.11 Objectives

Specific and direct visualisation of the initial infecting genomes has not been possible with traditional methods. Thus, the exact mechanism of genome release into the nucleus, as well as the spatiotemporal relationship of the HSV-1 genome and its association with host and viral factors is currently not well understood. These are important topics since further understanding of them allows for identification of potential key steps and interactions in the viral lifecycle that could be targeted to stop the progression of infection for therapeutic use. Other questions include what is the behaviour and fate of the input genome during the progression of infection, and what is the spatiotemporal relationship between the viral genomes and product transcript. Again, answering these are crucial in understanding how the virus proceeds with infection in host cells. My investigation addressed the above questions using the following framework:

i. Objective: Develop and produce an assay for genome uncoating by producing HSV-1 with its genome labelled with EdC (HSV-1EdC) for direct genome visualisation.

Result: Efficient nucleotide analogue incorporation into viral DNA resulted in establishment of an assay that is able to quantitatively visualise single infecting genomes.

ii. Objective: Quantitative spatiotemporal analysis of HSV-1EdC genomes uncoated into the host nucleoplasm using wide-field and super-resolution microscopy, and investigate factors affecting nuclear genome entry.

Hypothesis: Input genomes during infection maintain a constant morphology during the progression of infection

Result: The hypothesis was rejected; input genomes demonstrated qualitative transitions in genome condensation state linked to transcription and replication, genomes showed a novel process of genome congregation dependent upon transcription, and there was temporal switching in viral regulatory protein recruitment to distinct viral genome compartments.

iii. Objective: Quantitative spatiotemporal analysis of the transcriptional output of HSV-1 genomes and their relationship with genomes.

Hypothesis: There is heterogeneity in genome template usage and production of transcript from different genome foci.

Result: The hypothesis was accepted; multiple genome infections showed heterogeneity in transcript production between different genomes, as well as on the same genome with

different transcripts. Transcription bursting was observed, as well as a biphasic nature of ICP0 transcription during the progression of infection. Investigations with drug inhibitors have implicated potential new mechanism for general suppression for input viral genomes. My investigations have laid down the foundations for further multivariate quantitative analysis of what factors influence transcriptional output from infecting genomes.

iv. Objective: Investigate the interactions and relationship between HSV-1 genomes and components implicated in the host antiviral response.

Result: Intrinsic immunity component PML and DNA damage pathway component γH2AX associate with viral genomes upon entry, and increased productive viral activity is observed in the absence of antiviral DNA sensors IFI16 and cGAS.

CHAPTER TWO Materials & Methods

2. Materials & Methods

2.1. Cell Culture

2.1.1 Cell lines and culture media

Human epithelial keratinocytes (HaCaT) cells (both wild type and knock-out lines, detailed in table 2.1) and African green monkey kidney fibroblasts (Vero) cells were grown in Dulbecco’s Modified

Eagle’s Medium (DMEM; Gibco) with 10% foetal bovine serum (FBS; Gibco). RPE-1 cells (human telomerase immortalised retinal pigment epithelial cells) were grown in DMEM/F12 medium (Sigma) supplemented with 200 mM glutamine and 10% newborn calf serum (NCS; Gibco). All cell growth media were supplemented with 100 units/ml penicillin and 100 µg/ml streptomycin (Gibco). Cell cultures were incubated in a 37 °C humidified incubator supplemented with 5% CO2.

Table 2.1 List of cell lines

Cell line Growth Medium Serum Origin RPE-1 DMEM/F12 10% NCS ATCC© CRL-4000TM Vero ATCC© CRL-1587DTM HaCaT HaCaT cGAS-/- DMEM 10% FBS Dr Leonie Unterholzner [457] HaCaT IFI16-/- HaCaT STING-/-

2.1.2 Cultivation and propagation of cell lines

Cells were sub-cultured from a confluent monolayer at a ratio of approximately 1:8 for a 175 cm2 flask. The culture medium was aspirated and cells were washed with 5 ml of 1x PBS (Sigma), followed by incubation at 37 °C in 7.5 ml of 0.05% trypsin/EDTA (Invitrogen) until cell detachment. 15 ml of culture medium was added to neutralise the effect of trypsin and the cell aggregates were dispersed by pipetting. The suspended cells were then seeded in new flasks for passaging or on

51 coverslips for further experiments. Cells counts were carried out with a Countess Automated Cell

Counter (Invitrogen). The cell suspension was mixed with trypan blue and a 10 µl sample was loaded on to a haemocytometer for automatic counting. Cells were plated on sterile 16 mm diameter glass coverslips (VWR) for cell imaging experiments.

2.2 Virus Culture

2.2.1 Infection of cell lines

Cell monolayers were infected with different viruses (described in table 2.2) at different multiplicities of infection (MOI) as indicated in the text. Unless otherwise specified, for standard infections, cells were chilled in the fridge for 20 min and then washed once in ice-cold PBS. The concentrated virus was diluted as appropriate in ice-cold serum-free culture medium, and 0.5 ml of this inoculum was then applied to the cells. This was incubated for 45 min at 4 °C on a rotating shaker for virus adsorption. The cells were then transferred directly to an incubator at 37 °C and 5% CO2, and this time point was determined as the start of infection (i.e. 0 hours post infection, hpi). Cells were then incubated for the appropriate time. At 1 hpi, the virus inoculum was removed and replaced with culture medium with 2% serum. Mock-infected controls were prepared as above but with no virus present in the inoculum.

Table 2.2 List of viruses

Virus Description Source HSV-1[17] Wild-type HSV-1 strain 17 O’Hare Lab HSV-1[17]EdC EdC-labelled HSV-1[17] O’Hare Lab HSV-1[17].dl1403 HSV-1[17] ICP0 deletion mutant Dr Mike Nicoll HSV-1[KOS] Wild-type HSV-1 strain KOS O’Hare Lab HSV-1[KOS].n12 HSV-1[KOS] ICP4 deletion mutant Dr Neal DeLuca

2.2.2 Preparation of virus stock

Herpes simplex virus type-1, strain 17 (HSV-1 [17]), was produced by infection of confluent

RPE-1 cells plated in roller bottles (850 cm2). RPE-1 cells were infected at a multiplicity of infection

(MOI) of 0.025 in serum-free media at 37 °C. At 2 hpi, an equal volume of medium with 4% NCS was added to give a final concentration of 2% NCS. The HSV-1 infected cells were harvested at approx. 48 hpi when cell cytopathic effects were evident. The cell culture suspension was spun down at 4,500 rpm, 15 min, 4 °C to remove the cell debris and the supernatant was spun in a Sorvall centrifuge

(Thermo Scientific, SS34 rotor) at 19,000 rpm, 90 min, 4 °C. The pellets were resuspended in PBS and were further spun at 4,500 rpm, 10 min, 4 °C to remove any remaining debris. Concentrated virus stock was stored at -80 °C, and was termed HSV-1EdC. Titres were determined by plaque assays (see

Section 2.2.4). The ‘Mock EdC’ inoculum used for control experiments (Fig 3.8) was prepared by pulsing uninfected RPE-1 cells for 48 h with 5 µM EdC, harvesting the supernatant and concentrating exactly as if preparing virus from infected cells.

2.2.3 Production of virus labelled with ethynyl deoxycytidine (EdC)

HSV-1EdC was produced as above, except 5 µM ethynyl deoxycytidine (EdC, Sigma) in DMSO was added to the culture medium at 6 and 24 hpi. Before infection of HaCaT cells, the concentrated

HSV-1EdC virus stock was dialysed with PBS using Slide-A-Lyzer MINI Dialysis Device (20K MWCO,

Thermo Fisher) for 5 h at 4 °C.

2.2.4 Titration of virus stocks

RPE-1 cells were seeded in 12-well plates (2 x 105 cell/well). They were infected with three dilutions of virus stock (10-6, 10-7, and 10-8) in serum-free medium and placed on an orbital shaker in the incubator (37 °C, 5% CO2) for 1.5 h. The inoculum was then replaced by medium containing 2% serum and 2% pooled neutralising human serum (Sigma). At 48 hpi, cells were fixed with 3.7% formaldehyde and stained with crystal violet. Plaques were counted and the virus titre was calculated.

2.2.5 Plaque assays in the presence of EdC

RPE-1 cells were seeded in 6-well plates (106 cell/well) and infected with approx. 50 pfu per well in medium containing EdC as denoted in the text. Cells were infected, fixed, and stained as above

(Section 2.2.4). Plaque numbers were counted and plaque areas were measured with Image Pro Plus

7.0 software.

2.2.6 Measuring viral yield in the presence of EdC

RPE-1 cells seeded on 75 cm2 flasks were infected with HSV-1[17] at MOI 0.025 in serum-free medium. At 2 hpi, an equal volume of medium with 4% FCS was added to give a final concentration of

2% NCS. 5 µM EdC or control DMSO was added at 4 hpi and the infected cells were harvested at approx.

48 hpi when cell cytopathic effects were evident. The supernatant and cells were separated by spinning at 4,600 rpm, 10 min, 4 °C. The cell pellet was resuspended in 1 ml serum-free medium and disrupted with three cycles of freeze-thawing in ethanol on dry ice. Remaining cell debris was removed by spinning at 4,600 rpm, 10 min, 4 °C to obtain cell-associated virus. Supernatant and cell-associated virus titres were measured by plaque assays (see Section 2.2.4).

2.3 HSV-1EdC on coverslips

2.3.1 Adsorption and visualisation of HSV-1EdC on coverslips

Glass (borosilicate) coverslips were coated in poly-lysine (100 µg/ml) for 1 h and then washed three times in PBS. HSV-1[17] or HSV-1EdC was diluted in PBS and 5 µl of stock containing 6 x 105 pfu was placed on the coverslip for 15 min. Viruses were then fixed in 4% PFA in PBS for 15 min and processed for immunofluorescence and click chemistry (see Sections 2.5 and 2.7).

2.3.2 Quantitative counts of HSV-1EdC on coverslips

Counts of HSV-1EdC fixed on glass coverslips was carried out using the ‘find maxima’ function in ImageJ. The total counts from the merged image (both red and green channels) were calculated, and then the counts from each channel was calculated. The number of foci which were red or green only was calculated by subtracting the number of foci recorded in the green and red channels respectively from the total counts. The counts of foci both positive for red and green were counted by subtracting the red and green only foci from the total counts.

2.4 Drug treatments

Drugs used are listed in the table below. Unless otherwise stated, drugs were added to the medium at the beginning of the 45 min virus adsorption period (described in 2.2.1).

Table 2.3 List of drugs

Drug Function Medium Concentration Source Actinomycin D Inhibits RNA polymerases I and II DMSO 5 µg/ml Sigma Acyclovir Inhibits HSV-1 replication DMSO 0.5 mM Cayman Cycloheximide Inhibits translation Cell culture 100 µg/ml Sigma medium Leptomycin B Inhibits nuclear export by CRM MeOH 20 nM Sigma MG132 Inhibits proteasome activity DMSO 10 µM Calbiochem Nocodazole Disassembles microtubules DMSO 2 µM Sigma Phosphonoacetic Inhibits HSV-1 replication Cell culture 400 µg/ml Sigma acid medium TPCK Inhibits cysteine proteases DMSO 10 or 20 µM Sigma

2.5 EdC pulse labelling of host cell DNA and HSV-1 replication compartments

RPE-1 cells were seeded on glass coverslips and were infected with HSV-1[17] wild-type at

MOI 5 in serum-free medium. The inoculum was replaced by 2% NCS medium at 1.5 hpi. 5 µM EdC

55 was added at 1 hpi. At 5 hpi, cells were fixed with 4% PFA in PBS for 15 min and processed for immunofluorescence and click chemistry (see Sections 2.6 and 2.7).

2.6 Immunofluorescence studies

Coverslips with viruses or cells attached were fixed with 4% paraformaldehyde (PFA) in PBS

(15 min), and then quenched in 0.1 M glycine in PBS (5 min). Coverslips were permeabilised with 0.5%

Triton X-100 (Sigma) in PBS (15 min), washed twice, and were incubated in 10% FBS in PBS (15 min) for blocking. Coverslips were incubated with primary antibody (Table 2.3) diluted in 10% FBS in PBS for 45 min and then washed 3 times in PBS. They were subsequently incubated in secondary antibody

(Table 2.4) together with DAPI (100 µg/ml) for 45 min. Coverslips were washed 3 times in PBS and mounted on glass slides with ProLong® Gold Antifade Mountant (Molecular Probes). All processing was done at room temperature.

Table 2.4 List of primary antibodies used in immunofluorescence assays

Antibody target Type Animal of origin Dilution Source α-Tubulin Monoclonal Mouse 1:1000 Sigma γH2AX (Phospho-S319) Polyclonal Rabbit 1:250 Abcam Cyclin B1 Monoclonal Mouse 1:100 Thermo Fisher ICP0 Monoclonal Mouse 1:300 Virusys ICP4 Monoclonal Mouse 1:400 Virusys ICP8 Monoclonal Mouse 1:100 Abcam ICP27 Monoclonal Mouse 1:100 Virusys IFI16 Monoclonal Mouse 1:200 Santa Cruz PML Polyclonal Rabbit 1:300 Regal Group RNA Pol II CTD Monoclonal Mouse 1:150 Santa Cruz VP5 Monoclonal Mouse 1:400 Virusys

Table 2.5 List of secondary antibodies used in immunofluorescence assays

Antibody target Type Animal of origin Dilution Source Mouse IgG (Alexa Fluor 594) Secondary Goat 1:750 Invitrogen Mouse IgG (Alexa Fluor 488) Secondary Goat 1:750 Invitrogen Rabbit IgG (Alexa Fluor 594) Secondary Goat 1:750 Invitrogen Rabbit IgG (Alexa Fluor 488) Secondary Goat 1:750 Invitrogen

2.7 HSV-1 capsid detection by western blot

Concentrated HSV-1 wt and HSV-1EdC (107 pfu each) was mixed with Laemmli buffer and heated for 5 min at 95 °C. Samples were loaded on a 10% polyacrylamide gel. Proteins resolved by electrophoresis were transferred to nitrocellulose membranes which were blocked with Odyssey© blocking solution (LI-COR). After blocking, membranes were incubated overnight at 4 °C with primary antibody (anti-VP5 mouse monoclonal antibody, 1:500 dilution, from Virusys) diluted in blocking solution. Membranes were washed three times with 0.05% Tween in PBS and incubated with secondary antibody (goat anti-mouse IgG Dylight 680, 1:10,000 dilution, from Pierce Biotechnology) in blocking solution for 1 h at RT in the dark. Visualisation of protein bands was performed using the

LI-COR Odyssey infrared imaging system.

2.8 Click chemistry for microscopy imaging

Coverslips with viruses or cells attached were fixed with 4% paraformaldehyde (PFA) in PBS

(15 min), and then quenched in 0.1 M glycine in PBS (5 min). Coverslips were permeabilised with 0.5%

Triton X-100 (Sigma) in PBS (15 min), washed twice, and were incubated with a reaction mix for 2 h:

10 μM Alexa Fluor 488-azide (Invitrogen); 1 mM CuSO4; 10 mM sodium ascorbate; 10 mM amino- guanidine and 1 mM Tris(3-hydroxypropyltriazolylmethyl)-amine (Sigma-Aldrich) in PBS pH 7.4.

Coverslips were washed three times with PBS and mounted on glass slides with ProLong® Gold

Antifade Mountant. If immunofluorescence studies were to be carried out on the same sample, the

57 click chemistry was carried out first. The sample was then blocked in 10% FBS in PBS for 15 min and then processed for primary and secondary antibodies as described in Section 2.5.

2.9 RNA detection

2.9.1 Single molecule RNA detection by RNAscope

Single molecule RNA fluorescent in situ hybridisation (smRNA-FISH) was performed using the

RNAscope Multiplex Fluorescent Kit v2 (Advanced Cell Diagnostics, ACD Bio, Italy) according to the manufacturer’s instructions. Probes were designed by ACD Bio (Table 2.6). Briefly, ZZ probe pairs targeting HSV-1 RNAs as indicated and a positive control probe to the cellular mRNA for PP1B

(cyclophilin B), were designed and synthesised by Advanced Cell Diagnostics. Mock-infected or HSV infected cells, at different times or under different experimental regimes as indicated, were washed in PBS, fixed for 30 min with 4% PFA in PBS, treated sequentially with hydrogen peroxide and then with RNAscope Protease III (ACD Bio). The samples were then incubated with the smRNA-FISH probes

(see table below). Incubation and hybridization steps were performed at 40 °C for 2 hrs in a HybEZ™ hybridisation oven (ACD Bio), which provides a sealed, temperature-controlled humidifying environmental chamber. After washing, the ISH signal was amplified using the ADC Pre-Amplifier, and the distinct channel signals (C1, C2 or C3) developed with TSA Plus Fluorescein, TSA Plus Cyanine 3, or

TSA Plus Cyanine 5 (Perkin Elmer, USA). Cells were counterstained with DAPI, washed in PBS. Where simultaneous analysis of transcripts and genomes were performed samples were first processed for smRNA-FISH, then processed for click chemistry (see Section 2.7) and mounted onto coverslips with

ProLong Gold.

Table 2.6 List of RNAscope probes

Protein Gene Species Probe Name ACD Bio Catalogue Number ICP0 RL2 HSV-1 V-HHV1-RL2-C2 479931-C2 ICP4 RS1 HSV-1 V-HSV-1-RS1 533741 VP16 UL48 HSV-1 V-HSV1-UL48-C2 423471-C2 PP1B PP1B Human 3-plex Positive Control Probe 320681 Hs (Channel 2)

2.9.2 DNase treatment of samples before RNA detection.

After fixation of cells with PFA and treatment with hydrogen peroxide, cells were incubated in 10 units of DNase I (amplification grade, Thermo Scientific) in DNase I Reaction Buffer for 30 min at

37 °C. One unit is defined as the increase in absorbance of a high molecular weight DNA solution at a rate of 0.001 A260 units/min/ml of a reaction mixture at 25 °C. After the incubation, the digestion was stopped by treating cells with 2.5 mM EDTA in dH2O for 10 min at room temperature. Cells were then subsequently processed for RNAscope as described in Section 2.9.1, resuming with treatment

RNAscope Protease III.

2.10 Widefield microscopy

Images were acquired with Zeiss Axiovert 135 TV microscope using Zeiss 63x lens (Plan-

APOCHROMAT, 1.4 numerical aperture), 40x (Plan-Neofluar, 0.75 numerical aperture), 20x (Plan-

Neofluar, 0.5 numerical aperture) or 10x (EC Plan-Neofluar, 0.3 numerical aperture) objective lens and

Retiga 2000R camera with Image Pro Plus 7.0 software.

For imaging of multiplexed genomes and RNA, images were acquired on a widefield Nikon Ti-

Eclipse microscope, with a Plan Apo λ 60x oil immersion objective 1.4NA, a 1.5x zoom, a CoolLED pE-

4000 light source and a Hamamatsu Flash 4.0 camera (2048 by 2048 pixels, xy pixel size of 72nm). 3D stacks were acquired with a z-step of 0.1µm. Files were deconvolved in NIS-Elements version 4.00

(Nikon, UK) (Blind mode, 5 iterations) and processed with the open-source bioimaging software Icy version 2.0.2.0 [458]. Files were processed as maximum intensity projections and analysed with custom-made protocols in Icy developed by Dr David Gaboriau, Facility for Imaging by Light

Microscopy (FILM), Imperial College London.

2.11 Confocal microscopy

For confocal microscopy, images were acquired with Zeiss 510 Laser Scanning Confocal

Microscope using argon lasers at 405 nm, 488 nm and 543 nm with Zeiss LSM 5 software. Each channel was collected separately, with images at 512 x 512 or 1024 x 1024 pixels, with 4x averaging and with or without a variable zoom factor. Single confocal sections were acquired or multiple z-sections at 1

µm intervals which were then compiled for maximum projection display.

2.12 Structural Illumination super-resolution microscopy

Super-resolution imaging was performed on an Elyra PS1 system (Carl Zeiss) with an

Apochromat 63x 1.4 NA oil objective lens, 488nm and 561nm excitation lasers and images were captured on a sCMOS PCO Edge camera. The camera pixel size is 6.5 µm and with 63x objective and additional 1.6x tube lens, corresponding to 64 nm in the object plane. For analysis of infected cells, image stacks (2 µm) were acquired in Frame Fast mode (single multiband cube) with a z-step of 110 nm and 15 raw images (five phases, three angles) per plane. Raw data was then computationally reconstructed using the ZEN software to obtain a super-resolution 3D image stack with a pixel size of

32 nm in xy and 105 nm in z. The SIMCheck ImageJ/Fiji plugin [459] was used to perform quality control on both raw and reconstructed data and to estimate lateral (x-y) resolution (approximately 120 nm) and axial (z) resolution (approximately 300 nm). Images from the different colour channels were registered in ZEN with alignment parameters obtained from calibration measurements with either virus capsids simultaneously labelled in both red and green channels or with TetraSpeck calibration beads of 0.1 μm diameter (Thermofisher).

Volume analysis was performed the object analyser module of the Huygens image processing suite version 4.0 (SVI, The Netherlands). The image is segmented into defined objects by the seed- threshold level adjustment, and connection process. Introduction of a watershed increases segmentation reliability further. Detected objects are automatically labelled and submitted to a continuous Iso Surface renderer. The segmented image is shown as a coloured iso-surface image.

Object statistics are reported for each object, including geometrical data and spatial location. A simulated fluorescence process (SFP) computing algorithm allows visualization of the 3D data and production of the rendered image as an animation. Using this method, the volume and sphericity of genomic foci was estimated on coverslips and after nuclear entry.

2.13 Quantitative analysis of intracellular genome uncoating

Images were acquired by standard wide-field microscopy (see Section 2.10). Images were then processed using Image J denoise plugin and corrected for background. For quantitative evaluation of genome foci, images were imported into Image Pro Plus and then subject to thresholding and segmentation modules to define object masks for genomes which were quantified for various parameters. Using the DAPI outlines and the population analysis tool, the number of genomes within multiple nuclei (approximately numbers are shown in the text) was calculated for each condition under study which include different MOIs, different times after infection, and various drug treatments.

For cytoplasmic genome counts (Fig 4.1), the total number of genomes detected inside nuclei was subtracted from the total number of genomes detected.

2.14 Quantitative analysis of genome clustering

Spatial clustering analysis of EdC-labelled genomes was carried out using the Spatial Statistics

2D/3D ImageJ plugin [460]. The plugin analyses the overall distribution of inter-point distances including any local clusters and calculates whether there is evidence for a non-random distribution in the population of cells. A binary mask of each nucleus is generated together with a mask of the

61 genomes using the ‘Find Maxima’ function. The plugin calculates for every nucleus, the distances between every point and its nearest neighbour and generates a cumulative distribution function (CDF) of those distances (the G-function). To compute this function, first the average CDF of a completely random distribution is estimated over a set (500 iterations) of randomly generated point patterns, specific to each reference structure (nucleus) and the number of points (genomes) in that structure.

Second, the expected variation of CDFs around their average is estimated using a second set of randomly generated binomial point patterns. The relative position of the observed CDF for the actual test set within this range of variation is used to assign a p-value to the observed pattern, termed the

‘Spatial Distribution Index’ (SDI). Point patterns that tend to clustering have an SDI closer to 0 while patterns tending to even spacing have an SDI close to 1. The CDFs of SDIs of two different populations are compared using the Kolmogorov-Smirnoff (KS) test, which is non-parametric and distribution free.

A p-value for the difference between the two populations is calculated, as well as the D statistic which is the largest deviation between the two CDFs. The average CDF corresponding to a completely random distribution, (specific to each reference structure (nucleus) and the number of points) is first estimated over a set (500 iterations) of randomly generated point patterns. The expected variation of

CDFs around their average is then estimated using a second set of randomly generated binomial point patterns. The relative position of the actual observed CDF for the test set within this range of variation is used to assign a p-value to the observed pattern, termed the ‘Spatial Distribution Index’ (SDI).

2.15 Semi-quantitative analysis of RNA and protein abundance

For analysis of RNA abundance (Fig 5.3), the total RNAscope signal in each field was calculated using the ‘measure’ function in ImageJ. The nuclei were masked using the DAPI image and the mask was subtracted from the total image to obtain the cytoplasmic signal. The signal within the nuclear mask was defined as the nuclear signal. The calculated total signal was divided by the total number of nuclei to obtain the relative abundance of RNA per cell. For analysis of ICP4 abundance (Fig 6.3), the total protein immunofluorescence signal in each field was calculated using the ‘measure’ function in

ImageJ and was divided by the number of total nuclei to obtain the ICP4 abundance per cell. In both experiments approximately 100 cells were analysed for each condition.

CHAPTER THREE

Development and validation of EdC-labelled HSV-1 (HSV-1EdC)

Data from Figures 3.2, 3.4, and 3.7 are previously published [461].

3. Development and validation of EdC-labelled HSV-1 (HSV-1EdC)

3.1 EdC affects neither RPE-1 cell growth nor HSV-1 infectivity and replication

In order to validate that EdC is a suitable cytosine analogue, its effect on the host cell and HSV-

1 growth was assessed. Cells were plated at subconfluency and incubated in the presence of varying concentrations of EdC for 48 h. On inspection with brightfield microscopy and crystal violet staining, cells had no morphological aberrations and there was no difference in cell density (Fig 3.1a). There was also no significant difference in the viable cell counts between cells incubated in the presence or absence of EdC (Fig 3.1b). Thus, EdC does not appear to have any major detrimental effects on RPE-1 growth.

Plaque assays were conducted to investigate if EdC had any detrimental effects on HSV-1 infectivity and replication using key measurements of plaque numbers and plaque sizes respectively.

RPE-1 cells were infected in the presence of varying concentrations of EdC. The results demonstrated that there was no difference in either the number of plaques formed (Fig 3.2a and b) or their size (Fig

3.2a and c), indicating that EdC does not significantly affect HSV-1 infectivity or replication. The effects of EdC were also examined on multi-step viral production. To this end, RPE-1 cells were infected at a low multiplicity of infection (MOI) of 0.025 in the absence or presence of EdC (5 µM) and virus was harvested at 48 hpi. At this concentration the supernatant virus titre was 4-fold lower, while the cell- associated titre was 5-fold lower which is a comparatively modest effect relative to the high virus titres in the order of 107-108 (Fig 3.2d). Therefore 5 µM was chosen as the EdC concentration when producing HSV-1EdC and also labelling up HSV-1 replication compartments (RCs) in infected host cells.

Figure 3.1: EdC has no effect on RPE-1 cell growth (a) RPE-1 cells were incubated in EdC at the concentrations denoted for 48 h. Phase images of cells were obtained at 10x magnification on a brightfield microscope (top row), and cells were also stained with crystal violet (bottom row), scale bar 100 µm. (b) RPE-1 cells were plated at 3 x 105 cell per well and incubated with the denoted concentration of EdC for 48 h. Cells were treated with 0.05% trypsin/EDTA and the number of cells were counted by using a haemocytometer and an automated cell counter. A one-way analysis of variance test was used to compare the four cell counts (ns = not significant, p = 0.89)

Figure 3.2: EdC does not significantly affect HSV-1 infectivity, replication, or yield (a) RPE-1 cells were infected with approximately 50 pfu of HSV-1 wt per well and EdC was added at the denoted concentrations. Cells were fixed after 48 h and stained with crystal violet. Plaque numbers were counted (b) and plaque area was calculated (c) using Image Pro Plus software. Values of the controls (0 µM EdC) were set at 100%. A one-way analysis of variance test was used to compare the four cell counts (ns = not significant, (b) p = 0.70, (c) p = 0.26). (d) Multi step growth yield assay; RPE- 1 cells were infected with HSV-1 wt at MOI 0.025 and EdC was added at 4 hpi. The virus was harvested at 48 hpi. Produced virus was titrated on RPE-1 cells in the absence of EdC. Images obtained with brightfield microscope, scale bar 1 mm. A one-way analysis of variance test was used to compare the four cell counts (ns = not significant, p = 0.89). Figure previously published [461].

3.2 EdC is incorporated into replicating host cellular DNA and viral DNA

To undertake spatial analysis of EdC incorporation into DNA, RPE-1 cells were incubated with

EdC for 4 h and visualised by click chemistry by covalently coupling the EdC to a fluorescent capture reagent (Fig 3.3). Cells that had been in S-phase at some point during the pulse showed EdC signal.

The pattern of EdC was heterogeneous and in some cells EdC was observed in distinct puncta (Fig 3.3, panels I and II), while in others it was distributed throughout the whole nucleus with intense foci (Fig

3.3, panel III), or in a more homogeneous pattern (Fig 3.3, panel IV). In some cells, intense foci of EdC colocalised with intense DAPI staining (Fig 3.3, white arrowheads). Cell counts showed that approximately 30% of cells were positive for EdC (Fig 3.4b ‘Mock’ data).

Next, cells infected with wild-type HSV-1 (HSV-1 wt) were incubated with EdC to verify whether it can be incorporated into newly synthesised viral DNA. RPE-1 cells were infected with HSV-

1 wt at MOI 5 and were pulsed with EdC from 1-5 hpi. EdC was visualised by click chemistry and cells were costained with ICP8, a viral single-stranded DNA binding protein which acts as a marker for HSV-

1 replication compartments (RC) [462]. Mock infected cells did not show any nuclear ICP8 staining (Fig

3.4a, panel I), while 30% of cells were positive for EdC from cellular DNA synthesis (Fig 3.4b). Cells infected with HSV-1 were ICP8 positive to different degrees. In infected cells where infection was still in the relatively early stages, ICP8 could be observed in discrete foci which colocalised with weak EdC foci (Fig 3.4a, panel II white arrowheads), each focus likely originating from a single genome or multiple genomes. In cells where infection had progressed further, ICP8 staining was more intense and RCs had fused together to occupy most of the nucleus, again colocalising with EdC (Fig 3.4a, panel

III). These cells show clear incorporation of EdC into HSV-1 replication compartments, since there is little colocalisation of host cell DNA (DAPI) with ICP8 and EdC. This is in agreement with previous results which suggest that host DNA may be excluded from replication compartments [463], [464].

Figure 3.3: EdC is incorporated into replicating cellular DNA RPE-1 cells were incubated with 5 µM EdC for 2 h. Cells were then fixed and the EdC was visualised by coupling to a fluorescent capture reagent, and costained with DAPI. White arrowheads indicate examples of areas of strong EdC and DAPI overlapping signal. Images were acquired on a fluorescence microscope, scale bar 10 µm.

Figure 3.4: EdC is incorporated into HSV-1 replication compartments (a) RPE-1 cells were infected with HSV-1 wt at MOI 5 and pulsed with 5 µM EdC from 1-5 hpi. Cells were fixed at 5 hpi and EdC was visualised by coupling to a fluorescent capture reagent. Cells were costained with ICP8 antibody and DAPI. White arrowheads denote early replication compartments with clear ICP8 and EdC colocalisation, while white vertical arrows denote ICP8 foci that do not appear to colocalise with EdC. Images were obtained on a fluorescence microscope, scale bar 10 µm. (b) Cell counts of EdC and ICP8 positive cells during the timecourse. The total number of cells were identified by DAPI staining and cells were counted as positive if the nuclear signal exceeded the background value. Counts were done using the Image Pro Plus population analysis tool. Data from (a) previously published [461].

In other infected ICP8 positive cells, there appeared to be evidence of host cell DNA replication

(Fig 3.4a, panels IV and V). The EdC signal in these cells lacked colocalisation with ICP8 suggesting that it predominantly represented host cell DNA synthesis. This in turn suggests that these cells were infected during S-phase since HSV-1 infection blocks G1-S transition [465]. With a similar pattern to infected cells without host DNA synthesis, some cells at a relatively early stage in infection had discrete

ICP8 foci (Fig 3.4a, panel IV white arrows). Others with more progressed infections showed larger more developed replication compartments (Fig 3.4a, panel V). However, there was no obvious colocalisation of ICP8 and EdC which suggests that in cells infected in S-phase, there is relatively little EdC incorporated into viral RCs and more into replicating host cell DNA. Quantitative analysis of the infected cells showed that when pulsed with EdC at 1-5 hpi, approximately 55% of the cells were positive for EdC, and 60% were positive for ICP8 (Fig 3.4b). The number of cells positive for EdC and

ICP8 increased to 85% and 95% respectively later on in infection when cells were pulsed with EdC at

5-8 hpi. This is as expected since more cells will have had the opportunity to become infected and move further on into the replicative stage of infection.

3.3 Validation of EdC incorporation into the HSV-1EdC genome

Having established that EdC has little effect on either virus replication or cell viability and growth, 5 µM EdC was used for large scale production of HSV-1 (see materials and methods 2.2.3). In order to confirm what proportion of viral genomes had been labelled with EdC, an in vitro assay was used to examine purified capsids on coverslips. Previously it has been reported [193], [466] that when

HSV capsids were adsorbed on to solid surfaces, they underwent some form of structural rearrangement such that with increased temperatures, genome expulsion from the capsid could be observed. Although the exact mechanism is not known, these observations indicate that attachment of the capsid to a solid surface induces a conformational change facilitating DNA release. This process was exploited, and the HSV-1EdC was adsorbed to borosilicate coverslips to examine copper-catalysed cycloaddition of the genome to a fluorescent capture reagent. The adsorbed virions were processed for click chemistry and co-immunostained with VP5, the HSV-1 major capsid protein (Fig 3.5a(i)). Using the ‘Find Maxima’ function in ImageJ, local maxima above the background threshold were identified as particles positive for VP5 and EdC. Approximately 12,500 particles were identified from multiple fields, of which 93% were positive for both EdC and VP5, appearing as yellow particles (Fig 3.5, white arrowheads; Fig 3.5b). Approximately 5% of particles were positive for VP5 only (Fig 3.5a(i), horizontal white arrows), suggesting the presence of capsids containing a non EdC-labelled genome or (no genome at all). There was a rare population (2% of total particles) that were positive for EdC only (Fig

3.5a(i), vertical white arrows), which may indicate uncoated genomes which have separated from damaged capsid. EdC signal above background was not observed from the control HSV-1 wt (Fig

3.5a(ii); Fig 3.5b).

To ensure the EdC signal detected with HSV-1EdC was not from any EdC-labelled DNA contamination in the virus sample, the HSV-1EdC was clicked in solution, and then adsorbed to the coverslip (Fig 3.5(iii)). No EdC signal above background was detected in this sample, confirming the signal observed in HSV-1EdC is from bona fide EdC-labelled genomes, showing the extremely high

72 specificity and sensitivity of this method. From my analysis, the percentage of HSV-1EdC particles with detectable genomes was comparable, if not slightly greater than that for similarly labelled adenovirus

(using a combination of EdA and EdC), where approximately 90% of capsids applied to coverslips contained detectable genomes [406]. This information is important since if incorporation efficiency was low, e.g., only 10% of particles were detected or the efficiency was unknown, then subsequent studies examining genome localisation would be inaccurate.

In the original observations using purified capsids, Newcomb and colleagues observed that

DNA detected as elongated strands with progressively increasing ejection at elevated temperatures

[193], [466]. Although there was considerable heterogeneity between particles, release could be substantially prevented if the capsids were first cross-linked with PFA. In my experiments, genomes were detected genomes within virions as punctate foci, but elongated genome release was not observed due to analysis being carried out on extracellular virions and not capsids. To examine the detection and possible ejection of HSV-1EdC genomes further, virions were adsorbed onto coverslips and then subjected to elevated heat treatment. This resulted in numerous elongated filamentous stands ejected from virion (Fig 3.5c). This was accompanied by an increase in the numbers of capsids in which genomes were not detected as well as an increase in the numbers of punctate genome foci that did not colocalise with capsids. This is expected, since heat treatment is disruptive and would denature the VP5 epitopes for detection by immunofluorescence detection. The results are consistent with previous data obtained by electron microscopy and demonstrate that HSV-1EdC genomes when released from heat-disrupted virions could readily be detected on coverslips by the cycloaddition reaction. Furthermore, they indicate that when absorbed onto solid supports at lower temperatures, the genome is accessible to the fluorescent capture reagent, presumably due to some conformational perturbation of the capsid, but maintained in the confines of the capsid or virion (discussed in Section

7.1).

Figure 3.5: Visualisation of HSV-1EdC on coverslips (a) (i) and (ii): Glass coverslips were coated in poly-lysine (100 µg/ml) and 6 x 105 pfu of HSV-1EdC or HSV-1 wt was applied. Virions were fixed and permeabilised, and EdC was visualised by coupling to a fluorescent capture reagent. Virions were co-immunostained with VP5 (major capsid protein). White arrowheads denote dots where VP5 and EdC colocalise (93%). White horizontal arrows denote dots where only VP5 is present (5%), while white vertical arrows denote dots where only EdC is present (2%). Images were obtained with a fluorescent microscope, scale bar 10 µm. (iii) HSV-1EdC was also clicked in solution, and then applied to the glass coverslip for subsequent fixation, permeabilisation and VP5 detection. (b) Quantitative counts for HSV-1EdC and HSV-1 wt when clicked after adsorption to the coverslip. Capsids and genomes were identified by the ImageJ ‘Find Maxima’ function using an appropriate threshold. Percentages were calculated from a total population of 12,500. (c) Glass coverslips were coated in poly-lysine (100 µg/ml) and 6 x 105 pfu of HSV-1EdC was applied. The sample was heated at 70 °C for 2 min, and then fixed and processed for genome and VP5 detection. Scale bar 10 µm.

3.4 HSV-1EdC and HSV-1 wt have similar capsid to pfu ratios

HSV-1EdC was further characterised by comparing the capsid to pfu ratio with wild type HSV-1, since drastically different ratios can signify significantly altered characteristics. For example, a ten-fold increase in capsid to pfu ratio may indicate extremely inefficient production of viable virions. 104 pfu of HSV-1EdC and HSV-1 wt was adsorbed onto glass coverslips and processed for VP5 immunofluorescence. The density of VP5 in representative fields were relatively similar between the two viruses, albeit the coverslip was slightly more densely populated with VP5 with HSV-1EdC (Fig 3.6a).

Quantitative counts confirm this, where counting the total number of VP5 particles from 104 pfu was, on average, 105 for HSV-1EdC, and 5 x 104 for HSV-1 wt (Fig 3.6b). This indicates a capsid to pfu ratio of approximately 10:1 and 5:1 respectively, with a two-fold difference in ratios being observed between the two virus samples which is not significant in the context of HSV-1 virus production. The quantity of VP5 protein in the same number of pfu was compared between the two viruses by western blot (Fig

3.6c), which showed no significant difference. Thus HSV-1EdC, despite having a marginally increased capsid to pfu ratio, maintains similar characteristics to HSV-1 wt, demonstrating that EdC-labelling of the genome has had little effect on the virus.

Figure 3.6 HSV-1EdC and HSV-1 wt show similar capsid to pfu ratios 104 pfu of HSV-1EdC and HSV-1 wt was adsorbed to glass coverslips, fixed and membrane permeabilised, and detected for VP5 immunofluorescence. (a) shows four representative fields from each virus imaged by a fluorescence microscope, while (b) shows the total number of VP5 particles detected from 104 pfu. Counts were calculated from three independent experiments. VP5 particles were counted by the ‘Find Maxima’ function in ImageJ. (c) VP5 was detected from a Western blot of 107 pfu of HSV-1EdC and HSV-1 wt. The molecular weight of VP5 is 149 kDa.

3.5 HSV-1EdC genomes are only detected after capsid release during cell infection

HSV-1EdC was then used to infect RPE-1 cells to see if EdC-labelled genomes could be visualised in infected host cells. Viral entry was synchronised by infecting cells at 4 °C for 45 min where virions attach to the surface of the host cell membrane but do not fuse. Infected cells were then either fixed for processing or raised to 37 °C which allows viral entry and initiation of infection. At 4 °C the virions were distributed randomly on the surface of the cell as denoted by VP5 immunostaining. In contrast to adsorption onto coverslips, there was no significant genome signal from HSV-1EdC virions adsorbed to cells (Fig 3.7a, HSV-1EdC 4 °C). This is consistent with the lack of genome detection in physiological buffer [193], [466], and supports the proposal for a conformational change of the capsid upon adsorption to solid supports.

After raising the temperature to 37 °C, the distribution of VP5 becomes asymmetrical and the signal was often concentrated to one side of the nucleus. This indicates that the virions have fused with the host cell membrane and the capsids have been transported to the nuclear periphery via microtubules and the microtubule organising centre [95]. After shifting to 37 °C for 2 h, numerous genome foci were observed, the majority of which were detected in the nucleus (Fig 3.7a, HSV-1EdC

4 °C -> 37 °C). Frequently, small genome foci were observed in close proximity (but not colocalising) with VP5 (Fig 3.6b, white horizontal arrows) which may genomes that have recently uncoated from their parent capsid. Larger more diffuse nuclear genomes were also observed (Fig 3.7b, white arrowheads) which suggests the genome is no longer spatially confined. There was also a minority population of genome foci which were detectable in the cytoplasm (Fig 3.7b, white vertical arrows).

Infection with HSV-1 wt shows the same change in VP5 pattern upon shifting from 4 to 37 °C, however, there is no EdC signal detected as expected (Fig 3.7a, HSV-1 wt). These results show the EdC- labelled genome is not accessible to the fluorescent capture reagent when adsorbed to the cell surface, and is only exposed after virion fusion with the cell membrane and subsequent uncoating.

Figure 3.7: Visualisation of HSV-1EdC during cell infection (a) RPE-1 cells were infected with HSV-1 wt or HSV-1EdC at MOI 10 at 4 °C. Cells were then either fixed at 4 °C or after 2 h of incubation at 37 °C. They were then permeabilised, EdC was visualised by coupling to a fluorescent capture reagent, and co-immunostaining with a VP5 antibody and DAPI counterstain was carried out. Image denoted with an asterisk (＊) is expanded in (b). White arrowheads denote larger isolated genomes, white horizontal arrows denote genomes that are proximal but not colocalising with VP5 capsids, while white vertical arrows denote genomes detected in the cytoplasm. Images were obtained with a fluorescent microscope, scale bar 10 µm. Figure previously published [461].

Control experiments modifying various conditions were carried out to further verify these observations. Pre-treatment of the HSV-1EdC virus before infection with DNase had no effect on the ability to observe genomes in the nucleus (Fig 3.8(i)), showing that the observed EdC signal does not originate from any contaminating EdC-labelled DNA. The ability to see genomes was prevented by treatment with neutralising antibody against HSV-1 during infection, demonstrating the dependence of nuclear genome observation on the active infectious process and virion cell entry (Fig 3.8(ii)). The ability to see genomes was also dependent on the copper-catalysed cycloaddition reaction (Fig 3.8(iii)), since removing the key component of Cu(I) from the reaction mixture prevented their detection.

Finally, the nuclear EdC signal was not due to any contaminating EdC in solution in the HSV-1EdC inoculum (Fig 3.8(iv)).

Figure 3.8 Control experiments for cellular HSV-1EdC infection RPE-1 cells were infected with HSV-1EdC at MOI 10 and fixed at 2 hpi for detection by cycloaddition and immunofluorescence for VP5 (scale bar 10 µm). Experimental variations from the normal process were as follows: (i) the inoculum was treated with DNase I (500 U/ml) for 0.5 h at 10 °C prior to infection; (ii) the virus inoculum was treated with clinical grade neutralizing antibody IVIg (100 mg/ml) for 0.5 hr at room temperature prior to infection; (iii) During the cycloaddition, Cu(I) was omitted from the reaction mixture; (iv) cells were infected with a ‘mock’ inoculum which consisted of concentrated supernatant from uninfected RPE-1 cells pulsed and prepared as for an infected HSV-1EdC stock.

Chapter 3 Summary

In this chapter, I aimed to develop and produce an assay for genome uncoating by labelling the HSV-1 genome with the nucleotide analogue EdC for direct genome visualisation. EdC incorporation into the genome was efficient and with little to no side effects, and an assay for specifically visualising uncoated input genomes was established.

CHAPTER FOUR

Spatiotemporal Resolution of nuclear genomes during early infection

Data from Figures 4.1, 4.2, 4.6, 4.7, and 4.8 are previously published [461].

4. Spatiotemporal Resolution of nuclear genomes during early infection

4.1 Quantitative analysis of the relationship between MOI and HSV-1EdC genome entry

The first part of the initial HSV-1EdC characterisation involved a quantitative study of the relationship between MOI and the number of genomes that are released into the nucleus. Thirty minutes after shifting to 37 °C (0.5 hpi) was the time point used to quantitate this relationship since genome foci became more difficult to quantitate accurately at later timepoints because of the changing genome morphology after 1 hpi (see sections 4.2 and 4.3). An MOI of 10 was used since the numbers of nuclear genomes observed under this condition was optimal for showing features of the nuclear genome population and their decompaction. Figure 4.1a shows a representative maximum projection of a cell infected with HSV-1EdC at 0.5 hpi, illustrating the frequently observed proximal positioning of genomes and capsids. This most likely represents the uncoating event where the genome has just exited the capsid but has not yet physically moved far away from its parent capsid that it has been released from (Fig 4.1a inset).

At this early stage in infection, HSV-1EdC nuclear viral genome foci were homogeneous in size and shape and were spherical. The average numbers of genomes observed at different MOIs

(calculated from approximately 200 nuclei for each MOI) are shown in Figure 4.1b (raw data shown in

Fig 4.1c), while frequency distributions are presented in Figure 4.1d. There was a clear relationship with increasing genomes with increasing MOI, although this was not directly proportionate. For example, the median number of foci at MOI 5 was approximately 3, while at MOI 10 and MOI 50, the median was 6 and 10 foci per nucleus respectively. Furthermore, there were some nuclei, constituting a minority population, where extremely high numbers of nuclear foci could be observed, with maximum numbers for MOI 10, 20, and 50 being 21, 27, and 34 foci respectively (Fig 4.1b and c).

Capsid-free cytoplasmic foci (examples illustrated in Fig 4.1e, left panel) were infrequently observed,

84 although at higher MOIs these could represent 7-8% of total foci. Despite these being a minor population, the presence of these cytoplasmic genomes could be relevant in host pathogen detection and antiviral responses.

Figure 4.1 Quantitative analysis of genome uncoating at 0.5 hpi (a) Representative high magnification image of an individual cell infected with HSV-1EdC (MOI 10) at 0.5 hpi. Infection was as described in Fig 3.7. The expanded inset shows juxtaposition of uncoated compact genomes (green) and parent capsid (red), scale bar 10 µm. (b) Box and whisker plot for the number of genomes observed in cell nuclei at 0.5 hpi at increasing MOI. Box limits represent 2nd and 3rd quartiles with the horizontal bar in the middle showing the median and whiskers showing up the 5-95% range of the total population. Exceptional outliers (less than 5% of population) are shown as individual dots. The mean value is indicated by a ‘+’. Raw data for this summary is shown in the table (c) below. (d) Histograms for the number of genomes observed in cell nuclei at 0.5 hpi at each MOI. Bin width was set at 1 genome and approximately 200 nuclei for each MOI were analysed for (b) and (c). (e) Distribution of total labelled foci seen in the cytoplasm versus the nucleus at each MOI. Examples of cytoplasmic genomes are shown in the image denoted by a white arrow, scale bar 10 µm. Figure previously published [461].

4.2 HSV-1EdC infection reveals temporal alterations in genome compaction state

Having investigated genome transport by standard microscopy, I wished to study the genomes in more detail. Standard wide field microscopy is diffraction limited with other limitations in optical imaging and capture. Therefore to pursue higher resolution analysis of genome entry I next turned to using 3D structured illumination microscopy (3D-SIM) [467].

Initially, genome presentation at the earliest times of infection at 0.5 hpi was studied. Cells were imaged by 3D-SIM, an example of which is shown in Fig 4.2a(i), and subsequently the raw data was processed using the ‘Object analyser’ module of Huygens image processing software. After 3D segmentation, this allows the calculation of geometrical and spatial localisation data for individual objects. Figure 4.2a(ii) shows the same cell in Fig 4.2a(i) which has been 3D-rendered, and Figure

4.2a(iii) shows a zoomed section of the cell periphery which has been tilted to show the Z-dimension.

This field shows capsids in red, genomes in green, and DAPI nuclear stain in blue with transparencies applied. Pairs of genomes proximal to their putative parent capsids can be observed as previously identified (Fig 4.1a), with very few genomes detected in the cytoplasm. Capsids appear slightly oblong due to the lower optical resolution in the Z-dimension than in the X- and Y-dimensions. This is because of the inherent properties of point spread functions. This optical property applies to genome foci as well, but does not affect the main conclusions of the subsequent analysis.

Genomes inside capsids adsorbed onto coverslips (Fig 4.2b(i)) were compared with infected cell nuclear genome foci at 0.5 and 2 hpi (Fig 4.2b(ii) and (iii) respectively). For clarity the nuclear DAPI signal of the infected cell is omitted. The 3D genome objects were quantified for volume (Fig 4.2c) and shape (Fig 4.2d). This latter measurement reflects sphericity, where a measurement of 1 indicates the shape of perfect sphere while lower scores indicate progressively irregular shapes. The results show a substantial increase (approximately 3-fold) in median volume of the nuclear genomic foci compared to those still packaged inside capsids (Fig 4.2b (i) and (ii); Fig 4.2c). Nuclear foci at 0.5 hpi were comparatively homogeneous and with only marginal differences in sphericity (Fig 4.2c and 4.2d). The

87 linear length of the 150 kilobase long HSV-1 genome is approximately 50 µm and would stretch across a typical cell nucleus several times. The results demonstrate that while HSV-1 genomes expand, they are initially constrained and condensed to a relatively consistent and compact spherical volume. By 2 hpi, the median volume of the foci had increased further by another 2 to 4-fold, but this was also accompanied by increased heterogeneity in size and a marked increase in the irregularity of its shape as the foci became more decondensed and dissipated. As the infection progressed beyond 2 h, the genome signal was difficult to quantitate since it dissipated into aggregates of low intensity, but frequently with residual smaller condensed foci (see section 4.3).

Figure 4.2 3D-SIM analysis of genome decompaction (a) 3D-SIM data of a cell infected with HSV-1EdC (MOI 10) at 0.5 hpi. Panel (i) shows a maximum Z- projection of the cell while panel (ii) shows a visualisation of the raw data by iso-rendering in Huygens analysis software as described in materials and methods, scale bar 5 µm. Panel (iii) shows a zoomed section of panel (ii) which has been tilted to show the Z-dimension, scale bar 1 µm. Red objects denote capsids detected by VP5 immunofluorescence, green objects denote EdC-labelled genomes, while the blue object is nuclear DAPI staining. (b) 3D-SIM data of HSV-1EdC genomes on coverslips (i) compared to inside infected nuclei at 0.5 and 2 hpi ((ii) and (iii) respectively) visualised after 3D-SIM by iso- rendering in Huygens analysis software, scale 1 µm. Quantitative analysis of genome volume (c) and sphericity (d) is shown as box and whisker plots. Boxes show 2nd and 3rd quartiles with a horizontal bar in the middle showing the median. Whiskers show up to 5-95% of the total population. 50 genomes were analysed for each category. One-way analysis of variance tests were used to compare the three populations of genome volume and sphericity (*** = p < 0.0001). Data previously published [461].

4.3 Spatiotemporal relationship of genome compaction state and ICP4 expression

To examine the genome decompaction process further, I extended the analysis with increased temporal resolution. The analysis was also extended to investigate the relationship of genomes with

ICP4, which is the major HSV-1 IE regulator of transcription. Representative fields (three examples at each timepoint) illustrating several features are shown in Figure 4.3a. As shown previously (Fig 4.2b), within 0.5 h after shifting to 37 °C, intense punctate genome foci were observed (green) which could be observed prior to ICP4 accumulation (Fig 4.3a panels i-iii). By 1 hpi, ICP4 became detectable which was accompanied by the progressive increase in genome size. There were multiple patterns of localisation which included cells that had ICP4 recruited to a subset of genomes with several foci still having undetectable levels (panel iv), and cells which had accumulated ICP4 on most, if not all genomes (panels v and vi). At this time there was comparatively little diffuse ICP4 with the majority co-located with genome foci. The trend of genome decompaction accompanied by preferential ICP4 genome association continued on to 2 hpi (panels vii-ix).

By 3 hpi, a qualitative change was observed in the patterns and relationship between ICP4 and genomes. The infecting genome foci appeared as increasingly heterogeneous dispersed aggregates of lower intensity (as shown in section 4.2), together with residual punctate foci (denoted by white arrows) frequently on the periphery of the dispersed pattern (Fig 4.3a panels x-xii). However, while

ICP4 exhibited quite precise co-localisation with the genome foci at 1 hpi, by 3 hpi ICP4 was observed more within the dispersed areas of genome labelling, lacking any distinct enrichment in the remaining punctate foci. This was a distinct qualitative feature where foci remaining at later times on the periphery of the diffuse areas either completely lacked or were not enriched in ICP4, while at earlier times foci selectively recruited ICP4. This distinction of ICP4 localisation on earlier condensed genomes

(Fig 4.3a, panel v) against later more decondensed foci (panel x) is shown in expanded panels (Fig 4.3b).

A separate analysis by 3D-SIM is shown in Figure 4.3c which shows a more condensed genome signal

(white arrows) on the edge of an extended decondensed area at 3 hpi, with ICP4 localised

90 preferentially within the decondensed material and mostly excluded from the remaining punctate focus.

Figure 4.3 Spatiotemporal relationship of genome decompaction and ICP4 expression (a) Representative images of cells infected with HSV-1EdC (MOI 10) and fixed at 0.5 hpi (i-iii), 1 hpi (iv- vi), 2 hpi (vii-ix), or 3 hpi (x-xii) with subsequent detection by cycloaddition and immunofluorescence for ICP4, scale bar 10 µm. Insets from panels v and x are shown magnified in (b) to illustrate a shift from preferential ICP4 association with genome foci immediately after infection, but reduced or absence of association on foci remaining at later timepoints. White arrows denote residual genome foci on the edge of ICP4 bodies at 3 hpi. (c) 3D-SIM data of a cell nucleus infected with HSV-1EdC and fixed at 3 hpi showing residual EdC-labelled infecting genomes remaining as tighter foci on the periphery of replication compartments, marked by ICP4 (red) and the absence of significant ICP4 recruitment to those remaining genome foci. Figure previously published [461].

4.4 Transcription and translation coupled transitions in infecting genome compaction state

To examine the effects of basic cellular processes acting on viral genome localisation and morphological condensation state, inhibitors of transcription, translation, and viral DNA synthesis, were used. Cells were infected at MOI 10 either untreated or in the presence of a series of inhibitors, each added 1 h prior to raising the temperature to 37 °C as follows; actinomycin D (Act D, 5 μg/ml) to inhibit transcription, cycloheximide (CHX, 100 μg/ml) to inhibit translation, or acyclovir (ACV, 500 μM) or phosphonoacetic acid (PAA, 400 μg/ml) to inhibit viral DNA replication. Control experiments confirmed the expected activity of the drugs. In cells infected with HSV-1EdC, ICP4 protein synthesis was completely blocked in the presence of Act D and CHX but not by PAA or ACV (Fig 4.4).

Figure 4.4 Control experiments on effects of drug treatments Representative images of the localisation of ICP4 in cells infected with HSV-1EdC (MOI 10) and either untreated or incubated in the presence of Actinomycin D (Act D; 5 µg/ml), acyclovir (ACV; 500 µM), phosphonoacetic acid (PAA; 400 µg/ml), cycloheximide (CHX; 100 µg/ml), or no treatment as indicated. Cells were fixed (3 hr post infection) and processed for immunofluorescence of ICP4, scale bar 10 μm.

Initially, the effect of inhibiting transcription was studied. Representative results for genome uncoating and morphology with or without Act D treatment at each of four early timepoints are shown in Figure 4.5a. In the absence of drug treatment, genome foci were observed in the nucleus by 0.5 hpi together with the temporal increase in volume, irregular morphology (1, 2 hpi), progressive decondensation, and dissipation (3 hpi) as described previously (Section 4.3). Two significant conclusions could be made from results of inhibition of transcription.

Firstly, it was clear that inhibition of transcription had a significant effect on the progressive increase in genome volume and eventual decompaction, which were almost completely inhibited (Fig

4.5a). Genome morphology did not change with time and individual genomes remained homogeneous punctate foci. This was observed at extended times well beyond the relatively early time frames up to

3 hpi that had been studied so far. In fact, genome maintenance as discrete punctate foci was observed at least up until 8 hpi by which time in untreated cells, genomes largely dissipated and had become virtually undetectable. This observation allowed quantitation of individual nuclear genome foci at relatively later times in infection, albeit in an artificial setting where both host cell and viral transcription is absent. The results showed that the majority of nuclear genome entry occurred during the early window up to 0.5 hpi, although a proportion enters the nucleus between 0.5 to 1 hpi (Fig

4.5b). By 1 hpi, the vast majority of infecting genomes had entered the nucleus, with very few entering after this timepoint, suggesting that at this MOI, 1 hour is the time frame for nuclear genome entry of infecting virus. Secondly, in the presence of actinomycin D, there was no significant effect on either the average initial numbers or morphology of uncoated nuclear genome foci at 0.5 hpi (Fig 4.5b), indicating that the transcription process itself did not play any significant role in genome uncoating and transport into the nucleus, for example, by a transcription-coupled ratcheting process.

Unfortunately, direct comparisons at later timepoints was not feasible since the genomes in untreated infections were difficult to quantitate.

This difference in decompaction due to inhibition of transcription was again quantitated using

3D-SIM super-resolution microscopy. In a similar process to Figure 4.2, nuclear genomes at 0.5 and 2 hpi in the presence of Act D or with no treatment (Fig 4.5c, nuclear DAPI signal omitted for clarity).

The 3D genome objects for each condition and timepoint were quantified for volume (Fig 4.5d) and shape (Fig 4.5e). The results show that at both times the genomes were observed, inhibition of transcription significantly decreases the genome volume and maintains them as punctate foci. This effect was evident previously at the later timepoints after 2 hpi from observing fluorescence microscope images (Fig 4.5a). However, even as early as 0.5 hpi there is a significant difference in volume where untreated genomes have uniformly expanded slightly more than their counterparts in the presence of Act D. A small but significant increase in the genome volume was observed from 0.5 to 2 hpi in the absence of transcription, and this may represent the rate at which genomes passively decondense without any active processes (Fig 4.5d). There was no difference in the shape of genomes at 0.5 hpi with or without Act D. Without transcription, the genomes maintained their homogeneous spherical shape and showed a large difference with the untreated genomes that had expanded to become more diffuse and heterogeneous (Fig 4.5e).

Figure 4.5 Quantitative analysis of the effect of transcription inhibition on genome decompaction (a) Representative images of cells infected with HSV-1EdC (MOI 10) and incubated in the presence of actinomycin D (Act D, 5 μg/ml) 1 hour before infection or no treatment. Cells were fixed at the denoted timepoints for processing for click chemistry, scale bar 10 µm. (b) Box and whisker plots for the number of genomes observed in cell nuclei at MOI 10. Cells were infected with HSV-1EdC and were untreated (left graph) or treated with Act D (right graph). Genome counts were not possible with the untreated sample after 0.5 hpi due to changes in genome morphology. Box limits represent 2nd and 3rd quartiles with the horizontal bar in the middle showing the median and whiskers showing up the 5-95% range of the total population. The mean value is indicated by a ‘+’. Approximately 200 nuclei were analysed for each timepoint. (c) 3D-SIM data of HSV-1EdC genomes inside infected nuclei at 0.5 and 2 hpi with Act D or no treatment. Genomes were visualised after 3D-SIM by iso-rendering in Huygens analysis software, scale 1 µm. Quantitative analysis of the volume (d) and sphericity (e) of genomes on coverslips, at 0.5, and 2 hpi with or without Act D treatment. The data is shown as box and whisker plots. Boxes show 2nd and 3rd quartiles with a horizontal bar in the middle showing the median. Whiskers show up to 5-95% of the total population. 50 genomes were analysed for each category. Unpaired two-tailed t-tests were used for comparison of two populations, while one-way analysis of variance tests were used for comparison of three populations (* = p<0.05, *** = p<0.0001, ns = not significant).

These results that were obtained in the absence of transcription were distinct from those obtained after inhibition of virus DNA synthesis. Similar numbers of uncoated infecting genome foci were initially observed, not unexpectedly. In this case however, a progressive increase in foci volume was observed, with a clear difference especially by 3 hpi for the ACV/PAA treated cells versus the Act

D treated cells (Fig 4.6, small arrows vs white arrowheads). On the other hand, inhibition of DNA replication clearly had an effect, preventing the later dissipation and decrease in intensity seen in untreated cells (Fig 4.6, 3 hpi, untreated vs +ACV/PAA), indicating that these latter events were likely related to genome replication or other coupled events. The smaller genome foci size and tighter morphology in the presence of Act D compared to that observed in the presence of ACV or PAA also indicates that events linked to or downstream of transcription per se, but not requiring DNA replication, are reflected in increasing volume and irregularity and decreased compaction of the uncoated genomic foci.

We also examined genome localisation in the presence of cycloheximide. Previous data has shown that in the presence of CHX transcription does occur, unlike in the presence of Act D, but in the absence of HSV-1 protein synthesis, transcription is limited to IE loci with little or no E or L gene transcription nor DNA replication [468]. CHX did not block uncoating and the numbers of nuclear genome foci at 0.5 hpi were similar in CHX treated to untreated and to each of the other inhibitors

(Fig 4.6, +CHX). However, a distinct feature in genome morphology was observed at subsequent times with CHX treatment. Whereas inhibition of transcription resulted in the maintenance of tighter more condensed and largely singular foci (Fig 4.5a and Fig 4.6, +Act D), genome localisation in the presence of CHX resulted in a distinct congregation and clustering of genomic foci (Fig 4.6, +CHX, circled clusters).

While individual singular foci could still be observed in CHX treated cells, this congregation was a noticeable qualitative change in genome presentation especially at later times (see Fig 4.7a).

Figure 4.6 Effects of inhibition of transcription, translation, and virus DNA replication on transitions in genome decompaction Representative images of cells infected with HSV-1EdC (MOI 10) and incubated in the presence of Act D (5 µg/ml), CHX (100 µg/ml), PAA (400 µg/ml), Acyclovir (500 µM), or no treatment. Cells were fixed at the denoted timepoints indicated for processing for click chemistry. Arrows and circles indicate qualitative features of genome localisation under each condition as discussed in the text, scale bar 10 µm. Figure previously published [461].

To examine this difference in an unbiased quantitative manner, I used the spatial statistics and focus clustering algorithm in ImageJ [460]. This algorithm (see Section 2.11) examines the positions of objects within a reference structure (nuclei in this case) and assesses clustering using a normalized measure of the difference between the observed distribution of inter-point nearest neighbour distances and a completely random one. This difference is termed the Spatial Distribution

Index (SDI). Figure 4.7 shows a schematic illustration of the analysis of the SDI. Fig 4.7b illustrates theoretical nuclei with random, clustered or dispersed patterns. The algorithm compares the cumulative inter-point distance frequencies (CDFs) for a theoretical completely random distribution

(Fig 4.7b, thick black line in each graph; 95% confidence limits in thin grey lines) with the actual cumulative distribution obtained for foci in each example pattern (Fig 4.7b, red lines in each graph). A nucleus with an actual random distribution of foci (left cell) will have a distribution that could be anywhere with respect to the theoretical CDF. Nuclei with clustered foci (middle cell) or evenly spaced out foci (right cell) will have CDFs that deviate significantly to the left or right of the theoretical CDF respectively (Fig 4.7b, red and black CDFs in each panel). SDIs are then calculated as a probability index with an SDI close to 0 indicating a more clustered pattern, while SDIs closer to 1 indicate a pattern that is more evenly spread out. The theoretical overall distributions of SDIs for populations of cells is then calculated (Fig 4.7c). A population of nuclei with a truly random pattern will show an approximately even distribution of SDI values between 0 and 1 while nuclei with clustered foci will show an SDI distribution with a distinct leftward shift towards lower values, while those with evenly spaced foci show a shift towards 1. I validated this approach using the completely random pattern of spots when viruses were applied to coverslips (Fig 4.7d). The CDF function of 50 fields (used as surrogate ‘nuclei’) of approximately 50 capsid foci were calculated and an SDI distribution was derived from those values. The resulting SDI distribution was compared to an expected SDI distribution for a randomly distributed population using the Kolmogorov-Smirnov test, and was not found to be significantly different (p-value = 0.508; D statistic = 0.16), validating this method of analysis.

100

We applied this nearest neighbour analysis to virus genomes in infected cells for individual nuclei (approximately 50 nuclei, 3 hpi) in the presence of Act D or CHX. SDIs were calculated for the foci in each nucleus and compiled into a distribution of SDIs across the cells for each condition.

Consistent with the indication from visual inspection (Figs 4.6 and 4.7a), the results indicate a clear difference in the population distribution of inter-foci distances in Act D versus CHX treated cells (Fig

4.7e). While the Act D pattern deviates from random to some degree across the population, this was bordering on statistical significance (p-value = 0.056; D statistic = 0.26). What was clear was the distinctly different trends for CHX versus Act D, indicating a highly significant change in relative localisation and clustering of genomes in the presence of CHX versus Act D (p-value < 0.0001; D statistic = 0.52). Altogether from the spatial analysis of the images and clustering analysis, the results strongly support the conclusions for a distinct difference between Act D and CHX treated cells and the proposal that transcription is recognised in the host cell and results in distinct events that I have termed genome congregation.

101

102

Figure 4.7 Infecting genomes in CHX treated cells show a strong tendency for clustering compared to genomes in Act D treated cells (a) Comparison of genome localisation at 3 hpi in Act D treated vs CHX treated cells. Circles indicate the feature of genome clustering seen in CHX treated cells as opposed to the more typical individual foci (arrowed) for Act D. (b) Schematic illustration of theoretical random, clustered or evenly spaced patterns of foci in an individual nucleus. The spatial distribution analysis examines the overall distribution of inter-point distances, including any local clusters, and calculates whether there is any evidence for a non-random distribution in the population of cells. The distances between each point and its single nearest neighbour are generated as a cumulative distribution function (CDF) of those distances (see Section 2.11). A theoretical CDF for a truly random distribution (thick black line; 95% confidence limits, thin grey lines) is generated (specific for each nucleus and number of foci) and then compared with the actual CDF obtained for the test set (red line). For an actual random pattern, the red line will be located randomly, while for clustered and evenly spaced the red line will deviate left or right respectively. This difference between the observed and theoretical random distribution is transformed into a spatial distribution index (SDI) which has a value assigned between 0 and 1. Point patterns that tend to be clustered have an SDI closer to 0 while evenly spaced patterns have an SDI closer to 1. (c) The SDIs from each cell in the population is compiled into a distribution of SDIs. A population of nuclei with randomly distributed foci will have an SDI distribution that is uniform, while a population of nuclei with clustered and evenly spaced foci will have SDI distributions that are skewed towards 0 and 1 respectively. The SDI distributions of different populations are compared using the Kolmogorov-Smirnoff (KS) test, which is non-parametric and distribution free. A p-value for the difference between the two populations is calculated, as well as the D statistic which is the largest deviation between the two CDFs. (d) Validation of spatial distribution index using a random sample. Capsids applied on coverslips were fixed, stained with anti-VP5. Images were split into 50 fields of approximately 50 capsids and processed to calculate the SDI distribution of the population which is expected to be random. (e) The SDI distributions among populations of Act D and CHX treated cells were calculated. The difference in SDI distributions was highly significant with that of CHX reflecting a clear trend to clustering. Histograms of the SDIs were calculated from 50 nuclei from each group and the difference in the two distributions was calculated using the Kolmogorov-Smirnoff test (p < 0.0001, D = 0.52). Data previously published [461].

103

4.5 HSV-1 genome transport is independent of proteasome, protease, and nuclear export activity

In a previous study it was concluded that proteasome inhibition prevented HSV-1 genome transport to the nucleus [469]. This conclusion came from an indirect surrogate measure of β- galactosidase enzyme activity from a recombinant virus as a measure of genome import, and the results showed that treatment by MG132 suppressed β-galactosidase activity. MG132 is a reversible proteasome inhibitor that functions by mimicking peptide transition states that are formed during cleavage and inhibiting the major peptidase sites in the proteasome 20S core particle [470]. My results established a direct assay to visualise the HSV-1 genome, and I therefore next addressed directly whether proteasome inhibition had any detectable effect for genome uncoating and nuclear import.

MG132 (10 μM) was added to cells 1 h prior to infection with HSV-1EdC at MOI 10, and genome localisation assessed at 0.5 hpi compared to untreated cells. The distribution of numbers of nuclear genome foci in individual cells was assessed for at least 200 cells in each condition (Fig 4.8a, insets show representative individual nuclei). Overall there was no significant effect of MG132 on either total numbers or distribution of HSV genomes transported to the nucleus (Fig 4.8b).

Inhibition of CRM1-dependent nuclear export by Leptomycin B treatment has previously been shown to inhibit Adv genome nuclear entry [406]. The effect of Leptomycin B was also examined (as used in the previous studies). In contrast to the effects on Adv, no significant effect on HSV-1EdC genome uncoating and import was observed. By comparison, when nocodazole treatment was examined, a drug which depolymerises microtubules and has been previously shown in many studies to inhibit HSV infection and capsid transport [84], [471], [472], a very striking inhibition of the appearance of nuclear genomes was observed. Control experiments for the activities of MG132 and

Leptomycin B using known targets and effects confirmed their action (Fig 4.8c). Inhibition of proteasome activity with MG132 has been shown to cause increases in the number of PML bodies

[473], [474], Leptomycin B is known to inhibit export of cyclins [475] and treatment with nocodazole

104 depolymerises microtubules [476], [477]. These effects of the drugs were all observed as expected.

Taken together with the positive control for the suppression of genome nuclear entry by nocodazole, the results indicate neither proteasome function nor nuclear export are required for the initial stages of HSV-1 nuclear transport, uncoating, and nuclear import.

105

106

Figure 4.8 Effects of drug treatment on nuclear entry Cells were mock-treated or treated with MG132 (10 μM), Leptomycin B (20 nM), or nocodazole (2 μM) as indicated. Inhibitors were added to cells for 1 h prior to infection with HSV-1EdC (MOI 10). Cells were analysed at 0.5 hpi for the localisation of EdC-labelled genomes as described for other figures. (a) Each panel shows a representative image at high magnification (x63 objective) together with histograms of quantitative evaluation of the frequency of numbers of genomes per nucleus observed for each condition (at least 200 nuclei for each). (b) Box and whisker plots for data in (a). Box shows 2nd and 3rd quartiles with a horizontal bar in the middle showing the median, while whiskers show up to 5–95% of the total population. ‘+’ denotes the mean value. Unpaired two-tailed t-tests were used for comparison of two populations (ns = not statistically significant, *** = p<0.0001). In this experiment infection even in the untreated sample was somewhat less efficient than standard, but there was no significant difference with either MG132 or Leptomycin B at 0.5 hpi. In contrast, Nocodazole treatment resulted in a substantial and significant reduction in accumulation of uncoated nuclear genomes. (c) Uninfected cells were treated with MG132 (10 μM), Leptomycin B (20 nM), or nocodazole (2 μM), for 1.5 h, as during analysis of infection. Cells were then fixed and processed for immunofluorescence for PML, α-tubulin, or cyclin B1 respectively (red channel). Cells were counterstained with DAPI. Panels (i) and (ii) show MG132 treatment increases PML number and size per nuclei; panels (iii) and (iv) demonstrate microtubule depolymerisation upon nocodazole treatment; panels (v) and (vi) show inhibition of nuclear export and nuclear accumulation of cyclin B1. Scale bar for MG132 and nocodazole panels, 10 μm; scale bar for Leptomycin treatment, 100 μm. Data previously published [461].

107

In addition to proteasome activity, protease activity has also been implicated to be important in early HSV-1 nuclear genome entry. One study has concluded that specific proteolytic cleavage of

VP1-2, a large tegument protein, is required for genome transport into the nucleus. This conclusion was based on the observation that treatment with serine-cysteine protease inhibitor L-(tosylamido-2- phenyl) ethyl chloromethyl ketone (TPCK) during infection could block this process [68]. However, this was again using indirect methods of observing ICP4 protein production and bulk HSV-1 genome nuclease resistance.

The proposal that proteolytic cleavage is required for nuclear genome entry was assessed directly by examining nuclear entry of individual HSV-1EdC genomes in the presence of TPCK. Treatment with TPCK concentrations of up to 20 µM did not prevent nuclear genome entry (Fig 4.9a, Panels (iii) and (v), white horizontal arrows). It also appeared that at later timepoints approaching 3 hpi, TPCK seemed to inhibit the genome decompaction process and ICP4 production in a dose dependent manner. Decondensed genomes with preferential ICP4 colocalisation were observed as expected in the untreated cells (Fig 4.9a, Panel (ii), white vertical arrows), and were occasionally observed in the

10 µM TPCK treated cells (Fig 4.9a, Panel (iv), white vertical arrow). However, at 20 µM TPCK, ICP4 was not observed and the genomes were maintained as punctate foci (Fig 4.9a, Panel (vi), horizontal white arrow), similar to when cells are infected in the presence of Act D (Fig 4.9a, Panels (vii) and (viii)).

The nuclear genome numbers were again quantitated and showed while 10 µM TPCK treatment did not prevent nuclear genome transport, it resulted in approximately a 1.5-fold decrease (Fig 4.9b).

Unfortunately, 20 µM TPCK treatment was affecting the viability of the cells, and sufficient numbers of cells could not be analysed for this data point.

Another observation that I made was that TPCK treatment seemed to affect the intracellular localisation of the ICP4 protein. Whereas in the untreated cells the ICP4 was entirely nuclear at the observed times (Fig 4.9a, Panels (i) and (ii)), with TPCK treatment cytoplasmic ICP4 was being observed

(Fig 4.9a, Panel (iv), white asterisk). The question of whether TPCK is interfering with the nuclear

108 import of viral proteins, two other IE proteins, ICP0 and ICP27 were assessed. All three IE proteins studied were impaired in their nuclear import and appeared in the cytoplasm (Fig 4.9c).

Thus, my current data suggests that proteases inhibited by TPCK are not involved in any of the processes up to and including genome transport to the nuclear pore and genome release into the nucleus, despite some issues with cell viability at higher concentrations of TPCK (see Section 7.2).

However, there may be a previously unidentified role for protease function that is involved in regulating the genome decondensation process, as well as efficient nuclear import of IE proteins.

109

110

Figure 4.9 Investigating the effect of TPCK on HSV-1 nuclear genome entry and protein expression (a) Vero cells were treated with TPCK (10 or 20 μM), Act D (5 µg/ml), or no treatment as indicated. Inhibitors were added to cells for 1 h prior to infection with HSV-1EdC (MOI 10). Cells were fixed at 1 and 3 hpi for processing for click chemistry and immunofluorescence for ICP4. White arrows and asterisks denote features described in the text, scale bar 10 µm. (b) Box and whisker plots for data in (a). Box shows 2nd and 3rd quartiles with a horizontal bar in the middle showing the median, while whiskers show up to 5–95% of the total population. ‘+’ denotes the mean value. At least 70 nuclei were analysed for each category. 10 µm TPCK treatment did not completely inhibit the appearance of uncoated nuclear genomes. Unpaired two-tailed t-tests were used for comparisons of two populations (ns = not statistically significant, ** = p<0.005). (c) Experiment investigating the effect of TPCK treatment on intracellular HSV-1 protein localisation during infection. Cells were infected with HSV- 1EdC (MOI 10) for 3 h with or without 10 µM TPCK and fixed for immunofluorescence for the denoted HSV-1 protein, scale bar 10 µm.

111

4.6 Visualisation of single genome infections with HSV-1EdC

My results to date demonstrated the ability to directly examine genomes transport with novel results on condensation state and mechanism of uncoating and the spatial relationship within the cell to key transcription factors. However, they also revealed the complexity of genome analysis and with the aim of simplifying the system for further quantitative spatial analysis and exploring the limits of sensitivity, I next wished to address whether genome detection could be addressed under circumstances where a single HSV-1 genome infected the cell. To this end, the infection was carried out at an extremely low MOI of 0.05, such that probabilistically it was unlikely that a cell would be infected with more than one virion capable of initiating a productive infection.

The results show that the technique was indeed sensitive and specific enough to image single genome infections (Fig 4.10). During the infection timecourse under that condition, there were several features of genome entry and the relationship with ICP4 that were observed. During the earliest stages at 1 hpi, the genome was observed in its punctate form prior to ICP4 expression as previously observed

(Fig 4.10, Cell A), but there was a subsequent stage where ICP4 protein was observed but not preferentially localised to the infectious genome (Fig 4.10, Cell B, white horizontal arrows). As the single particle infection progressed, ICP4 become associated with the genome around 2-3 hpi (Fig 4.10,

Cells C and D, white horizontal arrows), however, the vast majority of ICP4 was localised diffusely in the cytoplasm. This was in contrast to infections with multiple genomes where most of the ICP4 present specifically localised to genomes (Fig 4.3a). Another observation was that ICP4 foci were observed that did not localise to genomes (Fig 4.10, Cells C and D, white vertical arrows). ICP4 binds to HSV-1 gene promoter sequences through specific binding sites [240], [478], and is associated with transcriptional activity of viral genomes. Accumulation of ICP4 in the nucleus away from genomes may be the result of excess ICP4 localising in host transcription factories due to its affinity and interactions with host transcriptional machinery such as TATA-binding protein and TFIIB [251], [253]. During the later stages of infection around 3-4 hpi, ICP4 levels increased inside the nucleus (Fig 4.10 Cell E), and

112 eventually the genome became extremely dissipated and difficult to detect as genome transcription and replication caused its decondensation (Fig 4.10 Cells F and G). Larger irregular shaped bodies of genomes that were observed in multiple genome infections were not observed (Fig 4.3a, Panels (vii)-

(ix)).

These results set the foundations for the potential use of single genome infections in further studies, including studying the output transcript from a single infectious genome (see Section 5.7).

113

114

Figure 4.10 Visualisation of single genome infections with HSV-1EdC Cells were infected at an extremely low MOI of 0.05 such that each infected cell only contained one genome, and were fixed at various timepoints for processing for click chemistry and ICP4 immunofluorescence. Cells A to G are representative examples of qualitatively different cells observed during the timecourse and are arranged temporally, although different states were observed at multiple timepoints due to the heterogeneity of the progression of infection. White horizontal arrows denote the location of the genome, while white vertical arrows denote ICP4 foci which do not colocalise with the genome, scale bar 10 µm.

115

4.7 Spatial analysis of infecting genomes versus RNA Polymerase II

Having established the close spatiotemporal relationship of ICP4 localisation with genomes during infection (Section 4.3), the relationship of input genomes and the host transcriptional machinery was investigated. HSV-1 relies on the host RNA Pol II to transcribe itself [479], and studies directly imaging the genome and polymerase together have not been attempted before.

Uninfected cells were fixed and processed for Pol II immunofluorescence. Although all of the

Pol II signal was observed in the nucleus, it was not homogeneous and there were foci where Pol II was more concentrated than others (Fig 4.11a, white arrowheads). This could be the visualisation of nuclear transcription factories, where components required for transcription are pooled into hotspots of subnuclear foci to efficiently carry out transcription [480], [481]. Cells were then infected with HSV-

1EdC (MOI 10) and fixed at various points after infection, and were processed for click chemistry and

RNA Pol II immunofluorescence. Immediately after nuclear genome entry, the Pol II distribution did not change and Pol II foci were still present which were not localising with genomes (Fig 4.11b, 0.5 hpi, white arrowheads).

However, after viral protein production had begun after 1 hpi, there was a significant alteration in the intranuclear Pol II localisation where genomes started to recruit Pol II (Fig 4.11b, 1 hpi, white vertical arrows) in a manner similar to ICP4 (Fig 4.3a), although with Pol II only a proportion of the total nuclear protein is recruited. This association continues as the genomes decompact (Fig

4.11b, 2 hpi), and eventually is recruited to replication compartments (Fig 4.11b, 3 hpi) with residual punctate genome foci being observed on the periphery, again similar to what was observed with ICP4

(Fig 4.11b, 3hpi, white horizontal arrows). These results show that after Pol II is strongly recruited to viral genomes, far more efficiently than any host genome locus which presumably enables the highly efficient process of viral genome transcription during lytic infection. This process could also be simultaneously aided by the displacement of Pol II from host cell genes. It has been shown that ICP22 is responsible for inhibiting transcription elongation of host cell genes by interacting with cdk9 [315],

116 and it is possible that this increases the free nuclear Pol II pool available for viral gene transcription.

Serine-2 phosphorylation of the Pol II CTD is a marker of productive elongation [120], [121], and observing the spatial relationship of Ser2 phosphorylated Pol II and genomes using specific antibodies would be able to provide further information on which genomes are currently undergoing transcriptional elongation, together with how much of the pool of actively elongating Pol II are associated with genomes.

Chapter 4 Summary

In this part of my investigation I quantitatively and spatiotemporally analysed the HSV-1EdC genome uncoating process using the newly established assay. This was observed using wide-field and super-resolution microscopy. The input genomes demonstrated qualitative transitions in genome condensation state linked to transcription and replication, as well as a novel process of genome congregation that was dependent upon transcription. The recruitment of viral regulatory protein ICP4 was also investigated and showed temporal switching between genome foci and distinct viral genome compartments. Furthermore, proteasome activity and cysteine-protease activity were found to be dispensable for nuclear genome entry.

117

Figure 4.11 Spatiotemporal relationship of genome decompaction and host RNA Pol II (a) Representative images of mock infected cells fixed and visualised for RNA Pol II immunofluorescence. White arrowheads denote Pol II foci, scale bar 10 µm. (b) Representative images of cells infected with HSV-1EdC (MOI 10) and fixed at the denoted timepoints with subsequent detection by click chemistry and RNA Pol II immunofluorescence. White arrowheads denote areas of concentrated Pol II that do not colocalise with genomes, vertical white arrows denote bodies where RNA Pol II preferentially colocalises to genomes, while horizontal white arrows denote putative residual genome foci on the periphery of supposed replication compartments.

118

CHAPTER FIVE

Simultaneous analysis of HSV-1 genomes and single molecule transcripts

119

5. Simultaneous analysis of HSV-1 genomes and single molecule

transcripts

5.1 Validation of single molecule RNA detection by RNAscope

Having validated and quantitatively analysed early genome entry and nuclear presentation with HSV-1EdC, the visualisation of genomes was applied to study the output from genomes. This was carried out by studying transcription, and the differences in possible outputs depending on potential interactions with viral and host factors. This required the simultaneous visualisation of genomes with click chemistry, together with viral RNA transcripts. Previous comparative analysis of RNA transcript detection has indicated that analysis using serially arrayed branched Z-form probes offers the most sensitive method of transcript detection while also allowing multiplexed detection of several genes simultaneously [454], [455]. Moreover, because of the requirement for adjacent hybridisation of multiple probe pairs to generate the template for amplification, this method has a negligible background signal. However, the most powerful trait of this is that it allows not only single cell measurements, but spatiotemporal analysis at the single transcript resolution. This is in contrast to population-based techniques such as Northern blots which conflates single cell variability. This method of RNA detection will subsequently be referred to as detection by RNAscope.

In order to assess transcriptional output from genomes, I first designed a probe for HSV-1 ICP0

(at its 3’ exon) as a representative immediate-early (IE) gene. This was used to examine ICP0 transcripts in mock-infected or HSV-1 infected cells at MOI 10 (Fig 5.1a). The results show ICP0 transcript exclusively within HSV-1 infected cells (Fig 5.1a(ii)), with no significant background in mock- infected ones (Fig 5.1a(i)). The transcripts appeared mainly in the nucleus in large intense bodies of transcript, while the cytoplasmic transcript, for the most part, appeared as well resolved dots at this timepoint. Detection of ICP0 transcript required de novo transcription since infecting in the presence of Actinomycin D completely abolished detection (Fig 5.1a(iii)). While not unexpected, this result

120 indicates that HSV-1 virions do not contain significant levels of transcript in the input inoculum. As further confirmation that this hybridisation signal represented bona fide ICP0 transcript, no signal was detected in cells infected with HSV-1 dl1403, an HSV-1 ICP0 deletion mutant [254] (Fig 5.1a(iv)). The infectivity of this mutant was confirmed by combining immunofluorescence staining for ICP4 protein, as another representative IE gene product, together with RNAscope detection for ICP0 transcript (Fig

5.1b). While cells infected with the wild type virus showed signal for both (Fig 5.1b(ii)), only ICP4 was detected in cells infected with HSV-1 dl1403 (Fig 5.1b(iii)).

Figure 5.1 Validation of ICP0 transcript detection by RNAscope (a) Representative images of cells that were mock infected or infected with HSV-1 (wt or dl1403 as denoted, MOI 10) and fixed at 1 hpi. Cells were subsequently processed for RNAscope with an ICP0 probe as described in materials and methods (section 2.9). For panel (iii), cells were pre-treated with Act D (5 µg/ml) 1 h before infection. (b) Cells were mock infected or infected with HSV-1 (wt or dl1403 as denoted, MOI 10) and fixed at 1 hpi. Simultaneous visualisation of RNA and protein was carried out with cells being processed first for RNAscope to detect ICP0 transcript and then immunofluorescence to detect ICP4 protein.

121

5.2 RNAscope and click chemistry can be combined to simultaneously detect HSV-1 genomes and transcript

In order to assess the compatibility of RNAscope and click chemistry, preliminary experiments were carried out. The incorporation of EdC into uninfected DNA previously mentioned in Section 3.2 was used as a measure of the click chemistry reaction, while a probe that hybridises with protein phosphatase 1B (PP1B) RNA was used as a measure of the RNAscope protocol. During optimisation of simultaneous visualisation of click chemistry and RNAscope, the order of processing after fixation of the cells was found to be crucial (Fig 5.2a). Cells were pulsed with 5 µM EdC for 4 h and then fixed for processing. Incorporated EdC and PP1B RNA were detected without problem separately, however, when the cells were processed for click chemistry first and subsequently for RNAscope, the latter signal was no longer detected. This was in contrast to when RNAscope processing was carried out first before click chemistry, and with this order both RNA and EdC could be visualised simultaneously

(further discussed in Section 7.5).

This assay was next extended to try to simultaneously detect EdC-labelled genomes and its transcript product using ICP0 RNA as a prototypical example of an IE gene. Cells were infected with

HSV-1EdC and fixed after 1 h. They were then processed for RNAscope first and then subsequently for click chemistry. Both were successfully simultaneously visualised, and at 1 hpi genomes were observed in the nucleus in a relatively compact state as previously observed (Fig 4.3a), although in addition there were multiple compact genomes appearing in the cytoplasm. ICP0 transcript were observed both in the nucleus and cytoplasm with many transcripts being resolved as single dots. However, there was a distinctive feature in the nuclear transcript population where large bodies of transcript were specifically localised around genomes (this feature is examined further in Section 5.3, Fig 5.4).

The probable reason for previously unseen cytoplasmic genomes appearing is due to the protease digestion step in the processing for RNAscope. This is required to weakly digest proteins and

122 tissue so that the probe can hybridise with its target RNA, however, an unintended consequence of this appears to be partial breakdown of the VP5 capsid of nucleocapsids in the cytoplasm of the infected cell. This then allows the genomes inside the cytoplasmic capsids to be accessible to the click chemistry reaction and are visualised which does not occur when the RNAscope processing is not carried out (Fig 3.7). The main focus of this part of the investigation is on the relationship between the actively transcribing nuclear genomes and the RNAs that they produce, therefore in subsequent images the cytoplasmic genomes have been excluded for clarity. An example is shown in Fig 5.2(c) which shows an infected cell from the same experiment as Fig 5.2(b) where only the nuclear genomes are shown in the masked image.

123

124

Figure 5.2 Simultaneous visualisation of HSV-1 genomes and its transcript (a) Representative fields of cells were pulsed with 5 µM EdC for 4 h and fixed. They were then processed for (i) click chemistry only, (ii) RNAscope only, (iii) click chemistry first and RNAscope second, or (iv) RNAscope first and click chemistry second. A PP1B probe was used as a positive control for RNAscope detection. (b) Cells were infected with HSV-1EdC at MOI 10 and fixed at 1 hpi, and then processed for RNAscope first and click chemistry second. (c) A representative cell from the experiment in (b) is shown masked for only the genomes inside the nucleus. Images of cells in subsequent experiments are shown in this manner for clarity.

125

5.3 Kinetics of genome nuclear entry and compaction state in relation to ICP0 transcript abundance and localisation

Having successfully developed simultaneous visualisation of EdC-labelled genomes and its product RNAs, the spatiotemporal relationship between genomes and ICP0 transcript, chosen as a prototypical IE gene, was investigated.

Cells were infected with HSV-1EdC at MOI 5 and fixed at various times after infection for processing for RNAscope and click chemistry (Fig 5.3a). In this case, based on initial experiments (data not shown), the MOI was adjusted from 10 to 5 in order to aid spatial resolution of the product transcripts. Results on the morphological changes of the genomes are consistent with those observed previously (see Fig 4.2). During the very earliest stages of infection at 0.5 hpi, most ICP0 RNA is concentrated in large bodies around the compact genomes (Fig 5.3a, 0.5 hpi). At this earliest stage of infection, there was a clear heterogeneity in the association between individual genomes and transcript output. Thus, within each individual nucleus large single transcript foci, precisely colocalised with and surrounded single genomes (Fig 5.3a, white arrows). Other genomes within the same nuclei had no transcript association (Fig 5.3a, 0.5 hpi, white arrowheads). By 1 hpi, the ICP0 transcript distribution is still mainly nuclear. Large transcriptional bodies colocalising with genomes are still observed, although the transcripts start to become dispersed in the nucleus and are also observed in the cytoplasm, indicating the start of RNA cytoplasmic export (Fig 5.3a, 1 hpi). As infection progresses, the ICP0 transcript distribution altered significantly. Most of the transcripts are now localised in the cytoplasm while the amount of intranuclear transcript has decreased as a result of continued export and perhaps together with a slowing down or shut down of ICP0 transcription from the genomes (Fig

5.3a, 2 hpi). By 3 hpi, nuclear transcription appears to resume, with transcripts localising in increasing abundance with genomes which are becoming more diffuse (Fig 5.3a, 3 hpi). This trend continues so that by 4 hpi there is abundant nuclear transcription, with large transcript compartments appearing

126 to localise within viral replication compartments suggesting high levels of transcription from the replicating genomes (further discussed in Section 7.5).

The intracellular distribution of the ICP0 transcript was semi-quantitatively analysed by measuring the total signal intensity of the RNA and the ratio of nuclear and cytoplasmic transcript (Fig

5.3b). Consistent with visual inspection, almost all transcript at the earliest time analysed (0.5 hpi) was nuclear, although by 1 hpi there is a slight increase in the cytoplasmic population. At 2 hpi this distribution shifted significantly, where the majority (approximately 70%) of the transcript is now localised outside the nucleus, most likely being exported for subsequent translation. However, from this point onwards nuclear transcription resumes, and the relative preponderance in the nucleus increases as infection progresses. The total abundance of RNA in the infected cells increased throughout the infection (Fig 5.3c), with the largest increase in total RNA between 2 and 3 hpi when transcription from replicated genomes was initiated.

127

128

Figure 5.3 Spatiotemporal relationship between genomes and ICP0 transcript (a) Representative images of cells that were infected with HSV-1EdC at MOI 5 and were fixed at the denoted times. Cells were processed for click chemistry and RNAscope for ICP0 transcript. White arrows denote genomes colocalising with transcript foci, white arrowheads denote genomes not associated with any transcript, scale bar 10 µm. Cells are masked for nuclear genomes. (b) Ratios (given as percentages) of nuclear versus cytoplasmic transcript calculated from the signal intensity using ImageJ. Total signal was calculated and the nuclear signal was subtracted to obtain the cytoplasmic signal. Data was obtained from approximately 100 cells. (c) The same data is expressed as the mean intensity of signal per cell, which was obtained by calculating the total signal and dividing by the number of nuclei. The mean signal per cell at 0.5 hpi is assigned an arbitrary value of 1.

129

One key observation was the presence of larger bodies of transcript that colocalised with genomes. This feature is illustrated in Fig 5.4 (white arrows). Their defining characteristics are their larger size and more intense signal than individual transcripts (examples of which are marked out by white arrowheads), and their preferential association with genomes. One hypothesis is that these are transcriptional ‘bursts’ from genomes. Current evidence from analysis of transcription dynamics from gene loci indicate that transcription is non-continuous, with genes being transcribed in alternating ‘on’ and ‘off’ states, and the ‘on’ state results in multiple transcripts accumulating around the site of transcription [432], [450], [482]. The key feature of IE transcription bursts is expanded upon in subsequent experiments carried out in Section 5.6 (also discussed in Section 7.7).

Figure 5.4 Transcriptional bursts are observed colocalising with nuclear genomes Representative image of a cell infected with HSV-1EdC (MOI 5) and fixed at 0.5 hpi. Cells were processed for click chemistry and RNAscope for ICP0 transcript, scale bar 10 µm. Cells are masked for nuclear genomes. White arrows denote transcription bursts colocalising with genomes, white arrowheads denote single ICP0 transcripts.

130

5.4 Analysis of IE gene transcription in the absence of translation and viral DNA replication

The investigation of the relationship between genomes and emergent IE transcripts was extended to investigate how ongoing protein translation and DNA replication would influence viral genome transcriptional output. Cells were either pre-treated with 5 µg/ml CHX for 1 h before infection, or untreated and infected with HSV-1EdC (MOI 5), and fixed for processing for ICP0 RNA detection and click chemistry at various times (Fig 5.5a). In the absence of translation, there was an immediate effect from the earliest time after infection (0.5 hpi) where increased numbers of transcriptional bursts were observed with a higher proportion of genomes associating with them (Fig 5.5a, 0.5 hpi Untreated, white arrows). At 2 hpi, as previously described, in the untreated cells nuclear transcription decreases and the distribution of transcript is mostly cytoplasmic. However, in the presence of CHX the transcriptional bursts around the genome become even larger and more intense. This trend of elevated transcriptional activity in the presence of CHX was sustained throughout the infection, and by 4 hpi, there were extremely high levels of transcript present in the cytoplasm likely due to continued export together with ongoing superinduced transcript output in the nucleus. The extremely intense cytoplasmic transcript signal was saturated at the normal detection settings used for the experiment. A representative image obtained at a reduced exposure times is shown (Fig 5.5a, panel with asterisk). The nuclear transcript is still preferentially located around genomes, although even at this later stage in infection there appear to be genomes that are not transcriptionally active for ICP0 and are not specifically associated with ICP0 transcript.

The abundance of transcript was again analysed semi-quantitatively at the population level

(Fig 5.5c). Even at 0.5 hpi, CHX treated cells produce approximately 3 times as much transcript. This fold increase becomes larger during the progression of infection with the amount of transcript detected in the CHX treated cells approximately 16 and 8 times higher than the equivalent untreated cells at 2 and 4 hpi respectively. The difference in cellular distribution at 4 hpi is also apparent with

131 the cytoplasmic transcript in the CHX treated cells accounting for almost 85% of the total transcript, while in the untreated cells this only amounts to 35%. The results indicate that blocking translation leads to transcriptional hyperinduction of IE genes, possibly by preventing the synthesis of a protein that suppresses and controlling ICP0 gene transcription. The most likely candidate for this is a viral protein ICP4, and is discussed later in the section (Fig 5.7).

The same experimental procedure was carried out to examine the effect of blocking viral replication, in this case using 400 µg/ml PAA (Fig 5.5b). Here, no major significant difference was observed at the earlier times, although at 2 hpi, there appeared to be slightly less transcript produced overall in the presence of PAA. The most significant difference was observed later in the infection at

4 hpi, where in the presence of PAA, there were much reduced levels of transcript, with virtually none found on and around the genomes (Fig 5.5b, 4 hpi +PAA, white arrowheads). This suggests that after initial competence for transcription, the infecting genomes are subsequently suppressed and that later viral transcription, including IE genes, preferentially takes place on replicated genomes (further discussed in Section 7.5). The quantification confirmed the observations (Fig 5.5c), where there was around a 40% decrease in transcript abundance at 2 hpi. However, the difference was most striking at

4 hpi with a near 50-fold change in the mean transcript abundance between the untreated and PAA treated cells.

During viral replication, large numbers of single stranded genomes will be produced which may potentially act as a non-specific target that the ICP0 probe may bind to. In order to validate the hypothesis that the transcript signal that was observed in the replication compartments after 3 hpi was bona fide ICP0 transcript, DNase treatment was carried out in a control experiment to remove any residual DNA templates. Cells were infected with HSV-1 wt, fixed at 1 and 4 hpi and were processed for both ICP0 and ICP4 transcript (Fig 5.5d). In the untreated infection, both IE transcripts showed strong signal in the replication compartments at 4 hpi (Fig 5.5d(i)). When the infection was carried out in the presence of PAA, this intense transcript signal was absent, showing that it is

132 dependent on viral replication occurring (Fig 5.5d(iii)). When the cells were treated with DNase after fixation in an untreated infection, the DAPI signal was lost, demonstrating the activity of the DNase

(Fig 5.5d(ii)). However, there was no loss in signal from the ICP0 and ICP4 transcripts, confirming that the probe was specifically targeting RNA templates and that this was not due to any cross reaction to the DNA.

133

Figure 5.5 Assessment of inhibiting translation and viral replication on ICP0 transcription (a) Representative images of cells infected with HSV-1EdC (MOI 5) and incubated in the presence of 100 µg/ml CHX 1 hour before infection or no treatment. Cells were fixed at the denoted times, and processed for RNAscope for ICP0 transcript detection and click chemistry. (*) An image of the same field with reduced sensitivity is shown for the +CHX 4 hpi panel. White arrows denote genomes localising with ICP0 transcript, white arrowheads denote genomes not localising with any transcript. Scale bar 10 µm, cells are masked for nuclear genomes. (b) Same experimental procedure as in (a), except cells were treated with 400 µg/ml PAA. White arrowheads denote genomes not localising with transcript. Scale bar 10 µm, cells are masked for nuclear genomes. (c) ICP0 transcript abundance is quantified as mean ICP0 signal per cell by measuring the total intensity and dividing by the number of nuclei. The mean signal per cell at 0.5 hpi (untreated) is assigned an arbitrary value of 1. Data was obtained from approximately 100 cells. Data for CHX treatment is shown on a different axis for clarity. (d) Representative images of cells infected with HSV-1 wt (MOI 5) and were either (i) untreated, (ii) treated with DNase (10 units) for 1 hour at 37 °C after fixation, or (iii) incubated in the presence of 400 µg/ml PAA 1 hour before infection. Cells were fixed at 1 or 4 hpi and processed for simultaneous ICP0 and ICP4 RNAscope detection. Scale bar 10 µm.

134

135

These observations with ICP0 transcription begin to reveal novel features of viral genome transcript dynamics during early infection. In order to assess whether these are unique to ICP0 or if they are indicative of the general behaviour of IE genes, the same type of analysis was also carried out with the ICP4 gene (Fig 5.6). Somewhat surprisingly, the distribution of the ICP4 gene at earlier times

(0.5 hpi) was qualitatively different to that of ICP0, ICP4 transcript appeared to be distributed more disperse inside the nucleus with more individual transcripts, while ICP0 was found to be mostly localised in bursts around the genome (Fig 5.6(i), further analysed in Section 5.6 and discussed in

Section 7.6). However, the qualitative results upon treatment with CHX and PAA were highly similar with ICP4. With CHX treatment, transcriptional bursts were more frequent and more intense, and transcriptional activity was elevated throughout the timecourse with high levels of transcript being produced at later times (Fig 5.6(ii)). With PAA treatment, viral replication dependent transcription at later times was suppressed and there was very little ICP4 transcript present at 4 hpi (Fig 5.6(iii)). This demonstrates that ICP4 transcriptional regulation is similar to that of ICP0, although interestingly there were definite qualitative differences in intranuclear distribution where ICP4 appeared to be more mobile and dispersed (addressed later in Section 5.6).

136

Figure 5.6 Assessment of inhibiting translation and viral replication on ICP4 transcription Representative images of cells infected with HSV-1EdC (MOI 5) and were either (i) untreated, (ii) incubated in the presence of 100 µg/ml CHX 1 hour before infection, or (iii) incubated in the presence of 400 µg/ml PAA 1 hour before infection. Cells were fixed at the denoted times and processed for RNAscope detection for ICP4 RNA and click chemistry. Scale bar 10 µm, cells are masked for nuclear genomes.

137

The two main results of the above experiments (i) translation inhibition causing uncontrolled transcription and (ii) prevention of new template synthesis resulting in shutdown of transcription, indicate that a protein that is newly synthesised during the early stages of infection is suppressing IE gene transcription. The main candidate for this is the ICP4 protein, and its role in inhibiting the transcription of IE genes during early infection is well documented [246], [483], [484]. In order to assess the contribution of ICP4 to the suppression of IE genes, a comparison was carried out using

HSV-1 n12 which is a mutant that has a partial deletion in the ICP4 gene, resulting in no functional

ICP4 protein being produced. The HSV-1 n12 mutant (made in complementing cell lines) is infectious

(Fig 5.7a), and was used to infect cells in parallel with the wild type virus in the presence of CHX (Fig

5.7b). If the ICP4 protein is solely responsible for the transcriptional suppression of IE genes, the expectation was that cells infected with the n12 mutant should produce as much transcript as when treated with CHX. Both ICP0 and ICP4 transcript were used as readouts for the IE genes (HSV-1 n12 still produces transcript from the ICP4 gene locus [254] which can be detected by the ICP4 RNA probes used). When cells infected with HSV-1 wt are treated with CHX, the transcriptional activity increases as previously observed (comparison of panels (i) and (iii)). However, when infected with HSV-1 n12, although the transcript increases this was at a much-reduced level compared to CHX treatment of the wild type virus (panels (ii) and (iii)). There was no difference in the transcript abundance when cells were infected with either HSV-1 wild type or n12 in the presence of CHX which was expected. While

ICP4 does contribute to the suppression of IE genes, the fact that global translation inhibition further increases IE gene transcription indicates the presence of another protein, potentially an unknown cellular repressor, that is responsible for the suppression of viral transcription (discussed in Section

7.5, Fig 7.2).

138

Figure 5.7 Assessment of the contribution of ICP4 to early IE gene transcription inhibition (a) Representative fields of cells infected with either HSV-1[KOS] wt or HSV-1[KOS] n12 (MOI 2), fixed at 2 hpi and processed for ICP0 immunofluorescence. Scale bar 10 µm. (b) Representative images of cells infected with HSV-1[KOS] wt or n12 (MOI 2) and incubated in the presence of 100 µg/ml CHX 1 hour before infection or no treatment. Cells were fixed at 3 hpi, and processed for RNAscope for simultaneous ICP0 and ICP4 transcript detection. Scale bar 10 µm, images from this experiment were obtained from a confocal microscope.

139

5.5 Simultaneous ICP0 and ICP4 transcript analysis suggests different spatial distributions and template genome usage

As previously mentioned in the previous section, when comparing the intranuclear distributions of ICP0 and ICP4 transcript, there appeared to be qualitative differences. In order to further investigate this observation, simultaneous visualisation of genomes and both ICP0 and ICP4 transcript was carried out on an advanced widefield microscope system. Cells were infected with HSV-

1EdC (MOI 5) and fixed at 0.5 hpi before processing for detection of genomes, and subsequently ICP0 and ICP4 transcript. Representative examples of cells are shown in Fig 5.8. When comparing the intranuclear distribution of transcript, ICP4 appeared to be more widely and evenly distributed around the nucleus with fewer bursts and more single transcripts. Contrastingly, ICP0 transcript appeared to be confined to a smaller area in the nucleus and was more often concentrated in transcription bursts.

This difference in spatial distribution of transcript could be due to differences in RNA processing since mRNA is spliced unlike that of ICP4 (further discussed in Section 7.6).

There also appears to be differential genome template usage for transcription of the two genes. The majority of genomes that are active are associated with both transcripts (Fig 5.8, arrowheads). However, there is a minority population that appears to be active for only either of the genes. There are genomes that are only associated with ICP0 (Fig 5.8, horizontal arrows), and those that are only associated with ICP4 (Fig 5.8, vertical arrows). This suggests that even within the same transcript class of IE genes, there is differential template usage in transcript production during lytic infection (further discussed in Section 7.6).

140

Figure 5.8 Simultaneous visualisation of HSV-1 genomes and ICP0 and ICP4 transcript show different intranuclear transcript distribution and heterogeneous genome template usage Representative images of cells infected with HSV-1EdC (MOI 5). Cells were fixed at 0.5 hpi and processed for ICP4 and ICP0 RNA detection and click chemistry. Arrowheads denote genomes associated with both transcripts, horizontal arrows denote genomes associated with only ICP0 transcript, while vertical arrows denote genomes associated with only ICP4 transcript. Scale bar 10 µm, cells are masked for nuclear genomes. Nuclei are marked out with white dotted lines. Images acquired with a NIKON Ti-Eclipse widefield microscope and processing of the raw image files were carried out by David Gaboriau.

141

5.6 Spatiotemporal analysis of early/late genes in relation to input genomes

Having initially focused on IE genes, the investigation was extended by comparing the transcriptional outputs from different kinetic classes of viral genes and their spatiotemporal relationship with viral genomes. Early genes were compared with IE genes, since they are the second group of genes that are transcribed during HSV-1 lytic infection and are under IE gene control. VP16 was selected as an example of an important HSV-1 gene that is transcribed in the latter stages of infection which acts as a key activator of lytic infection and a putative regulator of the lytic-latent program [156]. VP16 is transcribed late in the early gene programme and is often referred to as a delayed-early (E) or late (L) gene.

Cells were infected with HSV-1EdC and simultaneously visualised for infectious genomes and

VP16 RNA using RNAscope (Fig 5.9a). VP16 transcript was not produced until 1 hpi, where in a similar manner to the IE genes, transcriptional bursts could be observed specifically colocalising with a subset of genomes. Nuclear transcript increased as the infection progressed, and by 4 hpi, transcription from replicated genome templates was highly active with intense transcript signal being observed in replication compartments. Unlike ICP0, there was much less transcript in the cytoplasm which may indicate different export kinetics or a shorter transcript half-life in the cytoplasm. It is unlikely to be a result of lack of detection of cytoplasmic RNA molecules, since there was abundant detection of nuclear transcript which is more difficult for the target probes to access. A previous study which visualised VP16 transcript during HSV-1 infection in HeLa cells reported a largely cytoplasmic distribution during early infection [485]. The differences in VP16 distribution may arise from the different cell types having different efficiencies of exporting the mRNA.

Infections in the presence of CHX and PAA were also undertaken for direct comparisons (Fig

5.9b). No VP16 transcript was observed with CHX treatment (Fig 5.9b(ii)). This can be explained by delayed-E and L genes requiring ICP4 to drive their transcription [81], [486]. Thus, in the absence of protein translation, the VP16 gene locus is not transcribed. The nuclear genomes become clustered as

142 the infection progresses which is consistent with previous observations (Fig 4.7). In the presence of

PAA, early VP16 transcription is not affected with similar levels and intranuclear distribution of transcript at 2 hpi (Fig 5.9b(iii)). However, at 4 hpi the transcript levels in the PAA treated cells decrease dramatically, and what little transcript is present is localised in the cytoplasm, suggesting transcriptional shutoff and cytoplasmic export of the remaining nuclear transcript. Similar observations at 4 hpi with IE genes in presence of PAA could be explained by auto-regulation by ICP4 and possible other factors (Fig 5.7). However, with delayed early and late genes it would be expected that their transcription should continue with ICP4 protein being present with only viral replication being inhibited. One explanation of this could be due to the ICP4 being initially synthesised, but then

ICP4 protein levels gradually decaying as a result of ICP4 autorepression. This would result in initial

VP16 transcription that would fall off as ICP4 levels decreased. When viral replication is permissive, multiple new genome templates are produced from which ICP4 is transcribed (Fig 5.3a), thus driving the further transcription of VP16. Multiplexing genome detection with VP16 transcript and ICP4 immunofluorescence to investigate the relationship between ICP4 protein abundance and VP16 transcript abundance would provide further evidence for this proposal. Another explanation could be that the unknown cellular repressor, discussed in Section 5.4 (Fig 5.7), that is known to act on IE gene expression acts more generally and also represses HSV-1 early and delayed genes. Based on my results,

I propose a model of HSV-1 transcription dynamics (further discussed in Section 7.5, Fig 7.2), where this cellular repressor suppresses all viral transcription when viral replication is inhibited.

These results show the different kinetics and behaviours of later transcribed viral genes compared to IE genes, and further investigation by multiplexing HSV-1 gene transcripts from the different classes will allow information to be gained regarding the relative transcription kinetics between them and how the genomes transcribe them.

143

144

Figure 5.9 The spatiotemporal relationship between genomes and VP16 transcript (a) Representative fields and images of cells infected with HSV-1EdC (MOI 5). Cells were fixed at the denoted times and processed for VP16 RNA detection and click chemistry. Scale bars 10 µm, cells are masked for nuclear genomes. (b) Representative images of cells infected with HSV-1EdC (MOI 5) and were either (i) untreated, (ii) incubated in the presence of 100 µg/ml CHX 1 hour before infection, or (iii) incubated in the presence of 400 µg/ml PAA 1 hour before infection. Cells were fixed at the denoted times and processed for RNAscope detection for VP16 RNA and click chemistry. Scale bar 10 µm, cells are masked for nuclear genomes.

145

5.7 Detection of multiplexed transcripts from single genome infections

The previous results described in Section 5.5 of multiplexed IE gene transcripts with genomes have been insightful into the relationship between infecting HSV-1 genomes and their product transcripts. However, quantitative analysis of infections where there are numerous transcripts originating from multiple genomes is extremely complex. In order to simplify the system, single genome infections were utilised. In this setup, it can be clearly be deduced that the transcripts observed must have originated from the single infecting genome in the nucleus. Cells were infected with a low MOI of 0.005 and fixed at 1 hpi for detection of genomes and transcripts (Fig 5.10). A heterogeneous population of cells were observed where a subset of cells had infectious genomes that are active for transcription which colocalised with transcription bursts for either or both ICP0 and ICP4

(Fig 5.10(i)). In another subset of cells, the genomes were not closely associated with ICP0 or ICP4 mRNA (Fig 5.10(ii)), indicating that these were not currently active for transcription for either transcript. A final subpopulation showed infected nuclei with genomes where transcript was not detectable, perhaps indicating some form of host antiviral suppression of viral transcription (Fig

5.10(iii)). This successful visualisation of single genome infections and their transcript products lays the groundwork for their multivariate quantitative analysis (discussed in Section 7.7).

Chapter 5 Summary

In this part of my investigation I quantitatively and spatiotemporally analysed the transcriptional output in relation to the HSV-1 genomes. A biphasic nature of ICP0 transcript production was demonstrated, and transcriptional bursting was observed on the input genomes.

These transcriptional bursts were heterogeneous between genomes for the same transcript, as well as being heterogeneous between transcripts on the same genome, showing previously unseen dynamics, all of which have laid the foundations to further multivariate quantitative analysis. Finally,

146 studies using inhibitors have implicated a potential new mechanism for general suppression of the input viral genome.

Figure 5.10 Simultaneous visualisation of HSV-1 genomes and ICP0 and ICP4 transcript in single genome infections Cells were infected with HSV-1EdC (MOI 0.005). Cells were fixed at 1 hpi and processed for ICP4 and ICP0 RNA detection and click chemistry. Arrowheads denote genomes. (i) Image of representative cell with genome colocalising with bursts of both ICP0 and ICP4. (ii) Image of representative cell with genome proximal to ICP0 and ICP4 transcript. (iii) Image of representative cell with genome not associated with either transcript. Scale bar 10 µm, cells are masked for nuclear genomes. Nuclei are marked out with white dotted lines. Images acquired with a NIKON Ti-Eclipse widefield microscope and processing of the raw image files were carried out by David Gaboriau.

147

CHAPTER SIX

Host antiviral responses to infecting viral genomes

148

6. Host antiviral responses to infecting viral genomes

6.1 Genomes colocalise with PML bodies and with their association continuing in the absence of transcription

The applications of specifically visualising uncoated viral genomes is not limited to studying early entry mechanisms and viral transcriptional outputs. The single cell quantitative measurements that this technique can obtain, unlike traditional assays, can be used to study multiple other aspects of infection. One of those aspects is the study of host antiviral responses to infecting viral genomes.

While the host cell has multiple layers of antiviral defence, one of the first mechanisms of resistance that the virus encounters is intrinsic immunity, which involves the constitutively expressed cellular antiviral proteins [322], [353]. These factors include core constituent proteins of Promyelocytic

Leukaemia Nuclear Bodies (PML-NBs), of which most notable is PML itself. PML was selected for investigation since it has been well documented to associate with incoming viral genomes in the nucleus, and is an important factor in cellular antiviral function. My investigation is the first one that directly visualises genomes with PML (although one other lab has subsequently conducted experiments using this technique under different conditions [366]).

Cells were infected with HSV-1EdC at MOI 10 and fixed at various times for processing for click chemistry and PML immunofluorescence (Fig 6.1a). Even as early as 0.5 hpi, PML can be seen colocalising with nuclear genomes (Fig 6.1a). Many genomes appear to be still associated with PML up to 1 hpi, although after this time, lytic infection progresses and the PML protein is degraded by

ICP0. This results in loss of PML signal from 2 hpi onwards. Extending the infection timecourse to 8 hpi, the initial input genomes become too dissipated to detect (Fig 6.1a, 8 hpi). Higher resolution images obtained by 3D-SIM give further details (Fig 6.1b). During early infection when PML is still intact, while many genomes colocalise with PML (Fig 6.1b, 0.5 hpi, white arrows), some are not associated at all

(Fig 6.1b, 0.5 hpi, white arrowhead).

149

PML is degraded by ICP0 protein that is newly synthesised from the infecting genomes.

Therefore, Act D was used to inhibit transcription in order to study the relationship of genomes and

PML in the absence of any targeted degradation (Fig 6.1c). Act D inhibiting transcription caused the genomes to remain in tight punctate foci for the duration of the infection as previously observed (Fig

4.5). PML was seen to colocalise with a subset of genomes during early infection. However, in the absence of any ICP0, this association continued throughout the timecourse, even up to 8 hpi (although by this time some of the PML bodies had become more diffuse and disaggregated presumably due to the global inhibition of transcription preventing new PML production to replenish them). The higher resolution 3D-SIM images showed that at 2 hpi with Act D, almost all of the genomes that had entered the nucleus colocalised with PML (Fig 6.1d, 2 hpi, white arrows), but there was still a minority subpopulation of genomes that seemed to not associate with PML (Fig 6.1b, 2 hpi, white arrowhead).

This heterogeneity of genome-PML association may be important in the context of infection progression. It could be hypothesised that genomes that are not associated with PML have successfully evaded being locked down by PML-NBs (or PML-NBs have failed to accumulate around these genomes). Thus, it would follow that only this subpopulation ends up being actively transcribed and produces ICP0 to initiate lytic infection by releasing other genomes from PML-NB inhibition.

Combining genome detection and PML immunofluorescence with RNAscope detection for IE genes would shed further light on whether genomes with PML are truly silenced and whether the non- associating genomes are active. One other study has directly visualised genome association with PML

[366], although in this case cells were infected with single genomes of HSV-1 ICP0 deletion mutants, which do not degrade PML. It is interesting to note that preventing ICP0 production by transcription inhibition does not always result in all genomes associating with PML. It could be proposed that these subsets of genomes managed to evade initial detection by host cell DNA sensors and PAMPs, although how these genomes differ would be challenging to study. Heterogeneity between cells is also possible, with perhaps a subset of infecting cells not progressing through infection due to all genomes being inhibited by PML, while others have non-repressed genomes which allows progression of infection.

150

Further quantitative analysis with productive infection markers such as ICP4 protein would be required.

151

152

Figure 6.1 PML associates with a subpopulation of nuclear genomes (a) Representative images of cells infected with HSV-1EdC (MOI 10). Cells were fixed at the denoted timepoints for processing for click chemistry and PML immunofluorescence. (b) Cells were infected as described in (a), and analysed with 3D-SIM. Horizontal arrows indicate colocalising genomes and PML, while the arrowhead indicates a genome that is not associated with PML. (c) Representative images of cells infected with HSV-1EdC (MOI 10) and incubated in the presence of actinomycin D (Act D, 5 µg/ml) 1 hour before infection. Cells were fixed at the denoted timepoints for processing for click chemistry and PML immunofluorescence. (d) Cells were infected as described in (c), and analysed with 3D-SIM. Horizontal arrows indicate colocalising genomes and PML, while the arrowhead indicates a genome that is not associated with PML. All scale bars 5 µm.

153

6.2 Genomes do not show preferential association with IFI16

While intrinsic immunity acts as one of the first lines of host antiviral defence, the innate immune system is the other mechanism for mounting a rapid, non-specific response to pathogens.

One protein that is part of host cell innate immunity is IFI16 which recognises foreign DNA. It has been found to promote the assembly of heterochromatin on Herpesvirus DNA, thus silencing the genome and inhibiting viral IE gene expression and replication [354]. IFI16 is also found to trigger downstream

IRF-3 signalling, resulting in a broader antiviral response [325], [353]. Thus, preliminary experiments were carried out into the spatiotemporal relationship between HSV-1 genomes and this important innate immune factor to establish the groundwork for further investigations.

IFI16 was visualised by standard immunofluorescence, however, the intranuclear distribution changed dramatically with the method of fixation used (Fig 6.2a). When cells were crosslinked with

4% PFA, which was the method used consistently throughout all experiments, IFI16 showed a nuclear distribution which was depleted from the nucleolus (Fig 6.2a (i)). This is not in agreement with the general consensus that cellular IFI16 is concentrated in the nucleolus with no stimulus. This was indeed the case when cells were dehydrated with methanol (MeOH) (Fig 6.2a (ii)), suggesting that PFA fixation was not successfully cross-linking the protein in the nucleolus. When cells were infected with HSV-1EdC, there was no specific association of IFI16 observed at 1 hpi when fixed with PFA (Fig 6.2b, PFA).

Fixation with MeOH, which appeared to better preserve the true localisation of IFI16, did not show any clear colocalisation either, although the nucleolar IFI16 was now dissipated so that the intranuclear distribution was largely homogeneous. There were also multiple cytoplasmic genomes detected when fixing with MeOH, suggesting that it was allowing genomes still inside nucleocapsids to be accessible to the click chemistry reaction.

These preliminary results show that there is no specific accumulation of IFI16 around infectious genomes during early infection using two different fixation methods. Previous investigations have suggested that IFI16 undergoes multiphasic redistribution into punctate foci inside

154 the nucleus on to viral genomes [363], although there has been little direct evidence to show the direct association of these foci with genomes. More recent studies employing direct visualisation of genomes with bioorthogonal chemistry similar to that used in my investigation have shown colocalisation of IFI16 and nuclear genomes, however, there appear to be several conflicting observations. Results from a study by Alandijany et al. (2018) are consistent with my observations, and they show IFI16 dispersed around the nucleus and not colocalising with viral genomes during early infection (albeit using an ICP0-null mutant virus and infecting cells with single genomes) [366].

Interestingly, they showed highly specific IFI16 enrichment on to genomes in cells at the edge of plaques that are infected with large numbers of genomes, suggesting MOI or cell priming by interferon may be important factors. Other studies such as that of Cabral et al. [364] report contrasting results, where selective IFI16 colocalisation is observed with HSV-1 wt genomes early during single genome infections. The disparity in results with previous investigations could be explained by differences in experimental setup, timescale, and antibody usage, although further experiments will be required to draw any firm conclusions on the relationship between genomes and IFI16. It would be interesting to repeat infections and observe IFI16 and genomes at plaque edges, as well as conducting single genome infections to see if I can replicate the results of other groups.

155

Figure 6.2 Genomes do not show preferential colocalisation with IFI16 during early infection (a) Cells were fixed with 4% PFA in PBS for 10 min at room temperature or 100% methanol (MeOH) for 10 min at -20 °C, and then processed for IFI16 immunofluorescence, scale bar 10 µm. (b) Representative images of cells infected with HSV-1EdC (MOI 10). Cells were fixed with either PFA or MeOH as in (a) and processed for click chemistry and IFI16 immunofluorescence.

156

6.3 The absence of DNA sensing antiviral components in keratinocytes results in increased viral activity

In addition to IFI16, two other components of the innate immune system that have been implicated to be important in detecting DNA viruses are cGAS and STING. cGAS is one of the main cytoplasmic double-stranded DNA sensors, and detection of viral DNA by cGAS results in the production of cGAMP [329]. This then triggers an antiviral response that includes the production of interferon-β and upregulation of interferon stimulated genes that have antiviral functions [334]. This can both be an autocrine as well as a paracrine effect with infected cells potentially priming neighbouring uninfected cells for subsequent infection. STING is a downstream effector of cGAS, and is activated by cGAMP signalling. STING also acts as a node in the innate immune signalling pathway and is activated by other components such as IFI16 signalling [353]. Thus, the current model of innate immune DNA sensor activation results in a cellular transcriptional response including ISG upregulation, but in the case of DNA acting as the PAMP, direct cell intrinsic suppression of viral transcriptional output is thought to be another mechanism. This is illustrated with the previously mentioned putative role of IFI16 in mobilising silencing heterochromatin onto viral genomes [354], [362]. However, there is further complexity regarding more direct mechanisms of action of DNA sensors that do not rely on

ISG upregulation and paracrine signalling, where a recent study has suggested that IFI16 binds and sequesters the transcription factor Sp1 during HIV-1 infection to suppress viral gene expression [487].

Sp1 is also involved in the activation of HSV-1 IE and E gene transcription [488], implying that IFI16 could be acting via this route in HSV-1 infection as well. In order to evaluate these earlier response pathways and the contribution of DNA sensing components to suppressing viral transcriptional activity, keratinocyte cell lines that have individual components knocked out were used in preliminary infection experiments at early time points. Any phenotypic differences in the infection outcome in the absence of each component would provide valuable insight into their role in host antiviral immune responses.

157

Wild type HaCaT cells, together with cell lines in which each of cGAS, IFI16, and STING were knocked-out were infected with HSV-1EdC (MOI 1). Cells were fixed at various times during the infection and genomes were detected together with ICP4, which was selected as an indicative measure of infectious viral activity in the cells (Fig 6.3a). Infection of the wild type cells (Fig 6.3a wt) resulted in punctate genomes being first detected in the nuclei at 0.5 hpi. ICP4 was subsequently detected at 1 hpi in the subset of cells that were infected with active genomes, mainly in the nucleus. By 2 hpi, the nuclear genomes were still relatively punctate and ICP4 production was more pronounced in cells that were infected with more genomes (cells marked with arrowheads), with ICP4 being observed in both the nucleus and cytoplasm. In contrast, cells with fewer genomes tended to have no or little ICP4 present (cells marked with white arrows). A similar progression of infection was observed in the STING-

/- cells (Fig 6.3a STING-/-). Infection of the cGAS-/- and IFI16-/- cells resulted in increased ICP4 production

(Fig 6.3a cGAS-/- and IFI16-/-). Even at the earlier times of 1 hpi, in both knock-out cells, ICP4 was observed abundantly in both the nucleus and cytoplasm. At the later time of 2 hpi, it was observed that the cells expressing the highest levels of ICP4 were more intense than that of any wild type cell, and an increased proportion of cells showed high ICP4 production.

Semi-quantitative analysis of the intensity of ICP4 signal at 2 hpi confirmed these initial observations (Fig 6.3b). The total intensity of ICP4 was measured across multiple fields and divided by the number of nuclei to obtain the mean ICP4 signal intensity per cell. The values were normalised so that the wild type value is 1. The STING-/- cells only showed a minor increase (1.2-fold), while the cGAS-

/- and IFI16-/- showed a significantly larger increase in ICP4 production (2.8- and 4.6-fold increases respectively). It can be concluded that the absence of these two viral DNA sensors results in increased viral activity during infection. This is most likely due to reduced viral DNA surveillance by the host cell resulting in a weaker antiviral response (Fig 6.3c (ii) and (iii)). The more potent increase in ICP4 production in the IFI16 deletion mutant compared to the cGAS mutant suggests that the larger effect of knocking out IFI16 may suggest nuclear DNA sensing plays a larger role in suppressing early viral transcription than cytoplasmic DNA sensing. One explanation could be that the viral genomes are

158 much more readily detected earlier in the nucleus rather than in the cytoplasm where most genomes will be encapsidated. However, cGAS and IFI16 have both been shown to detect DNA in the nucleus and cytoplasm respectively [340], [345]–[348], and so further investigation would be required to substantiate this proposal. Alternatively, IFI16 itself could have a more significant role in cell intrinsic viral gene suppression than cGAS. This is consistent with reports of IFI16 directly interacting with and suppressing viral gene transcription by chromatin modulation [354], [362], and also with the more recent reports of IFI16 binding and sequestering the transcription factor Sp1 [487], [488].

Knocking out STING did not have as large an effect as knocking out cGAS or IFI16. This may be somewhat surprising, since it could be expected that knocking out a downstream effector that is also a signalling node for both DNA sensors would disrupt the antiviral signalling more and result in increased viral activity. One explanation for the negligible effect of STING depletion may be that at this early time point, the intrinsic antiviral activity of sensors such as IFI16 speculated above is the only significant factor that suppresses viral activity. Many ISGs that are activated by STING pathway signalling act by inhibiting viral replication or by inducing antiviral responses in other cells by paracrine signalling [489], [490], which occurs at a much later stage in infection than that observed in these experiments. Another explanation may be that cGAS and IFI16 have redundancy in their downstream effectors and signal via STING independent pathway, and in fact depleting one of the DNA sensors has more of an effect on derepressing viral transcription (Fig 6.3c (iv)). The data obtained was only semi- quantitative, and further analysis of ICP4 production per cell, as well as genome counts per cell would be highly informative since it would show whether these innate immune components had any influence on nuclear genome entry.

159

160

Figure 6.3 The absence of DNA sensing antiviral components in keratinocytes results in increased viral activity (a) Representative fields of HaCaT wt, STING-/-, cGAS-/-, and IFI16-/- cells infected with HSV-1EdC at MOI 1. Cells were fixed at the denoted times and processed for click chemistry and ICP4 immunofluorescence. Arrowheads denote cells with high numbers of genomes expressing higher levels of ICP4, white arrows denote cells with few or no genomes expressing little or no ICP4. Scale bar 10 µm. (b) Semi-quantitative analysis of ICP4 signal intensity at 2 hpi. The total intensity of ICP4 was measured across multiple fields of cells and divided by the number of nuclei in each field to obtain the mean ICP4 signal intensity per cell per field. The data shows the mean ICP4 signal intensity per cell from five fields (approximately 100 cells in total). The values were normalised so that the mean wild type value is 1. Error bars show standard deviations. Unpaired two-tailed t-tests were used for comparison of two populations (ns = not significant, ** = p<0.005, *** = P<0.0001). Data is from one biological replicate (n = 1). (c) Proposed model for the observations made showing infection of (i) wild type, (ii) cGAS-/-, (iii) IFI16-/-, and (iv) STING-/- HaCaT cells. Diagrams show transcription of IE genes from the viral genome (blue) being suppressed by the action of IFI16 and ISGs. Red ‘+’ represent the relative IE gene transcriptional output.

161

6.4 Association of DNA damage sensors with genomes

Infecting HSV-1 not only interacts with host antiviral defence systems, but in addition they also interfere with other cellular systems. One of these is the DNA damage response, and while its exact relationship with HSV-1 infection is extremely complex, it has been indicated that the virus can activate and exploit the response in a manner that aids viral replication [395]. One of the components in the DNA damage response pathway that has been implicated to be responsible for this is γH2AX which is the phosphorylated form of histone H2AX [405]. To this end, some preliminary experiments were conducted to study the spatiotemporal relationship of γH2AX and infecting genomes.

In mock infected cells, γH2AX was largely distributed diffusely in the nucleus, although there were concentrated bodies of the protein observed in the nucleoplasm (Fig 6.3a, white arrowheads).

There was heterogeneity with regard to the number of these aggregate bodies, being more numerous in a subpopulation of cells. When cells were infected with HSV-1EdC, no clear association was observed at 0.5 hpi (Fig 6.3a, 0.5 hpi). However, as the infection progressed to 1 hpi, a distinct spatial relationship between genomes and γH2AX was observed. Aggregate bodies of γH2AX localised adjacent to the infecting genomes, with the γH2AX body often forming an arc in a way that surrounds them (Fig 6.3b, white boxes). Moreover, the genomes and γH2AX was never observed to specifically colocalise unlike any previous proteins that have been studied in this investigation.

This is possibly the first time that this association of γH2AX was observed by directly visualising genomes, although a previous investigation observed similar effects with a surrogate output for genomes using ICP4 protein [399]. The results indicate that through some unknown mechanism, infection recruits γH2AX in close proximity to the genomes but not together. This may be linked to its role during infection that has been suggested to be to enhance HSV-1 replication, although further investigation would be required to elucidate the exact purpose of this recruitment.

162

Figure 6.4 γH2AX localises proximal to HSV-1 genomes during early infection (a) Representative images of cells (4 per timepoint) infected with HSV-1EdC (MOI 10). Cells were fixed at the denoted times and processed for click chemistry and immunofluorescence for γH2AX. White arrowheads in the mock infected cells denote aggregate bodies of γH2AX, scale bar 10 µm. (b) Enlarged images of cells at 1 hpi from (a). White boxes highlight bodies of γH2AX localising adjacent to genomes in an arc-like fashion.

163

164

Chapter 6 Summary

In this final part of my investigation, I focused on the interactions and relationship between

HSV-1 genomes and components implicated in the host antiviral response. I have demonstrated for the first time direct colocalisation of PML-NBs with HSV-1 genomes, as well as the spatial association of genomes with γH2AX. I have also made inroads into studying the effects of the absence of key innate immune pathway components, including IFI16 and cGAS, and have observed increased viral activity in their absence.

165

CHAPTER SEVEN

Discussion

166

Figure 7.4 has been previously published in modified form [461].

7. CHAPTER VII: Discussion

7.1 Production and validation of HSV-1EdC

The premise of producing viruses with labelled DNA during cell infection relies on the analogue being physiologically similar enough to the original biomolecule, in this case deoxycytidine. It is required that the analogue does not induce any adverse effects on viral production in order that the progeny viruses are as close to the parent wild type as possible. My results show that at the concentration necessary for labelling viral genomes of 5 µM EdC, there were no significant effects on

RPE-1 cell growth and morphology (Fig 3.1). EdC was selected over the main alternative ethynyl- labelled nucleoside of EdU due to the HSV-1 genome being approximately 68% GC content [491], and also the latter’s increased toxicity in cells relative to EdC [492]. Both EdC and EdU have been shown to have cell growth inhibitory effects on cells due to their inhibition of thymidylate synthase [493], [494] that has been shown to be time dependent. Assessments with A549 cells have shown significant cytotoxicity after 48 h with 10 µM EdC [226], however, the RPE-1 cells used in my investigations were not affected under similar conditions. During virus production in the presence of EdC, almost all cells will undergo cytopathic effects and cell lysis by 48 h, and so there are no adverse effects on the cells for the duration of virus infection and replication.

The next issue was whether EdC has any influence on the virus, and my results showed that it did not have any significant effects on infectivity, replication, or yield (Fig 3.2). Previous work where

Vero cells were infected with HSV-1 in the presence of lower EdC concentrations showed little difference in titre [406]. This is consistent with the 4 to 5-fold decrease in titre in the multi-step growth assay conducted in RPE-1 cells which is proportionally a modest change from the initial supernatant titre of 5 x 108 pfu/ml. In addition, the EdC-labelled virus is only used in the very earliest window of

167 infection during a very limited phase of the lifecycle (up to 4 hpi) to visualised the infecting genomes, and so is unlikely to be influenced by any detrimental effects of the nucleoside analogues.

Efficient incorporation of EdC into replicating host DNA and replicating viral DNA has been demonstrated (Fig 3.3 and 3.4 respectively). EdC-containing viral RCs were shown to colocalise with the major virus DNA binding protein ICP8 in cells that were infected outside of S-phase as expected, since HSV-1 infection blocks G1-S transition [465]. On the other hand, cells infected during S-phase appeared to incorporate EdC preferentially into replicating host DNA rather than viral RCs. In the latter case, it is likely the case that EdC is still incorporated at low levels in replicating viral DNA, but is extremely low and difficult to detect when compared to the signal from the highly active replicating host DNA (Fig 3.4a panels IV and V). These features of virus DNA replication in relation to the cell cycle are entirely consistent with previous work [462], [464], [495].

However, results demonstrating the incorporation of EdC into RCs and its minimal effect on virus yields do not necessarily mean that it would be incorporated into mature infectious virions.

Information on the efficiency and proportion of particles that contain detectable EdC is necessary for subsequent analyses and I exploited an in vitro assay described by Newcomb et al. (2007). They showed that HSV-1 genomes are ejected from the capsid upon attachment to solid supports due to undefined structural perturbations [193], [466]. Consistent with their findings, my results show that adsorption on borosilicate glass coverslips results in a structural change in the capsid that permits access to the catalytic molecules involved in the cycloaddition reaction including the azide-tagged fluorescent coupling reagent. Quantitative counts showed that vast majority (93%) were virions which had their genomes successfully labelled with EdC and were detectable (Fig 3.5). This is in contrast to previous attempts where EdC was only poorly incorporated into HSV-1 DNA [423]. The experimental conditions that I used and also the host cells being RPE-1 cells (as opposed to MRC5 cells) could be an important factor in the successful introduction of EdC into the nucleotide synthesis pathway. A small minority of particles detected were only positive for either VP5 (2%) or EdC (5%). The former can be

168 explained by the documented presence of capsids that do not contain genomes that are produced during infection. During capsid maturation in the host cell, three types of capsid are produced, namely

A, B, and C, representing empty capsids, scaffold containing capsids, and viral DNA containing capsids

[496]. A and B capsids contained in the supernatant during HSV-1EdC production will be purified together with virions containing C capsids, resulting in VP5 only particles. The identity of the EdC only particles is potentially more challenging to explain, and possibilities include residual host DNA from lysed cells, or mature capsids where the VP5 was present but could not be detected. In either case, they form an extremely small minority population and do not present an issue when quantitatively analysing genomes during cellular infection.

In my experimental setup, upon adsorption the genomes were retained inside the particle (Fig

3.5a), while purified capsids released their genomes [466]. It is possible that while there is some structural change in the virion capsid allowing the coupling reaction to the DNA, the genome is nevertheless retained in the confines of the particle due to the surrounding viscous tegument and enveloping membrane. It was also previously demonstrated that when the genome was released from purified capsids, it was ejected in a polarised manner likely from the portal, and that the proposed structural alteration may impact directly or indirectly on portal integrity [466]. A portal-specific alteration is possible, but not necessary to explain my observations, and it could be that some more global perturbations around the capsid shell could allow access to the components of the cycloaddition reaction, the largest being the azide-tagged fluorescent capture reagent (molecular weight 861). Indeed, there could be distinct perturbations with one type allowing access to small compounds and another involving changes including at the portal, promoting genome release. The ability to identify the genome within the capsid might be exploited for other types of analysis, such as in vitro biophysical analysis of genome transitions [497], or the identification of specific host components that may promote release. One explanation for how structural perturbations of the virion could arise is the ionic interactions with the borosilicate glass. Silicate glass surfaces immersed in water are known to acquire a negative surface charge through ionic dissociation [498], and this will occur

169 when the virion sample diluted in dH2O is applied to the coverslip. The electrostatic charges between the solid support and virion may be enough to induce these structural perturbations. Fixation of virions to neutral supports may shed further light on this, especially since experiments by Newcomb et al. were conducted by adsorbing viruses on to a hydrophilic Formvar-carbon coated electron microscope grid.

Similar analysis has been conducted with adenovirus (Adv) on solid supports [406]. With Adv,

EdC-labelled genomes were not detected upon initial adsorption of the virus to coverslips but were observed only after heat disruption. Heat treatment revealed internal protein VII by immunofluorescence and allowed cycloaddition labelling of the genome which also remained tightly associated with the capsid [406]. The ejection of the HSV-1 genome (from capsids or heated virions) compared to the maintained association of the Adv genome (from heated capsids) most likely reflects differences in the pressurisation state of genomes within capsids [499] and the lack of DNA packaging proteins within HSV-1 compared to Adv where the genome is associated with several core proteins, in particular protein VII [500]. These differences in internal pressure and protein-genome association are likely reflected in differences in mechanism in nuclear pore engagement and genome import.

7.2 HSV-1 genome nuclear entry

Uncoated nuclear genomes could be detected within 30 min of infection at 37 °C with HSV-

1EdC (Fig 3.7). Using similar methods to examine Adv infection, genomes could not be detected in the nucleus at 30 min and quantitative analysis on nuclear entry was performed at 2.5 h. This is a comparatively late point in my analyses, when genomes were already uncoated, decondensed, and in many cases beginning to replicate. This does not necessarily indicate that HSV-1 genome import is more rapid than Adv which would require a direct comparison with similarly labelled viruses with an identical experimental setup. Nonetheless for the majority of Adv capsids, their genomes become rapidly accessible to detection by click chemistry in the cytoplasm. This reflects the initial stages of

Adv cytoplasmic genome uncoating, where the capsid is disintegrated by the physical force of motor

170 proteins attached it [406], [500]–[502]. Complete Adv genome uncoating from the capsid is inhibited by leptomycin B which blocks the attachment of cytosolic adenoviruses to the NPC [16]. Given differences in entry processes, it is not unsurprising that unlike Adv, HSV-1 genome entry is not sensitive to Leptomycin B inhibition of nuclear export (Fig 4.8).

For HSV-1EdC, a minor subpopulation of genomes could be detected that had uncoated in the cytoplasm, present in a subpopulation of cells (Fig 3.7). Although unlikely to contribute to the nuclear genome pool, these cytoplasmic genomes may play a distinct role in the cell population as a whole.

This includes host responses from subsets of cells that have detected HSV-1 infection with cytosolic antiviral DNA sensors that might elicit paracrine effectors to other cells via interferon signalling. I conclude that for HSV-1, most capsids do not undergo structural transitions that perturb access, at least those that are able to be resolved by cycloaddition labelling with azide-tagged fluorescent coupling reagents. Structural transitions allowing genome release are likely to be tightly coupled with engagement with the nuclear pore. However, other factors including cell type could influence capsid integrity and genome accessibility, for example, differences in entry by fusion at the plasma membrane when compared to endocytosis. The ability to positively identify genomes associated with perturbed capsids will be useful in future studies including investigation of capsid integrity as a function of cell type and the influence of prior immune stimulation by different pathways.

There was no significant qualitative or quantitative difference in the presence of MG132 (Fig

4.8), and so it can be concluded that proteasome function is not required for capsid transport or genome nuclear import. This is in contrast to a previous publication which indicated that proteasome function was required after cell entry for efficient delivery of incoming HSV-1 capsids to the nucleus and subsequent gene expression [469]. While there could be several possible explanations for the differences in observations, in my analysis I directly measure genome import. These previously published results were based on measurements of gene expression, using a readout of β-galactosidase at 6 hpi and measuring of capsid localisation of a fluorescent virus at 2.5 hpi. Proteasome inhibition

171 could have inhibited a number of processes involved in surrogate read-out later in infection, or it could have inhibited bulk capsid dynamics that were not important for early genome delivery.

Another inhibitor of proteolytic cleavage that was assessed was TPCK. In previous findings it has been reported that TPCK prevents the release of viral DNA from the capsid by inhibiting VP1-2 cleavage [68]. Jovasevic et al. reported this result using ICP4 protein expression and nuclease resistance of viral DNA inside the capsid as proxy readouts in the presence of TPCK. My findings using direct visualisation of genomes were in disagreement (despite both using Vero cells as hosts) and demonstrated that TPCK did not prevent nuclear genome entry during infection of HSV-1EdC, and is not involved in any processes up to and including genome transport to the nuclear pore and genome release into the nucleus (Fig 4.9). This apparent discrepancy would imply that the genomes that had entered into the nucleus at 2 hpi in the presence of TPCK were somehow nuclease resistant, since they are clearly visible at similar time frames (Fig 4.9a). It could be argued from the decreasing numbers of genomes that appear in the nuclei in the presence of TPCK (Fig 4.9b) that its inhibitory effects are dose dependent with higher concentrations of TPCK completely preventing nuclear genome entry. However, this could not be confirmed due to the highly toxic effects of TPCK that were being exhibited on the Vero cells, where any concentrations above 20 µM would kill the Vero cells

(Jovasevic et al. were consistently using 30 µM TPCK). My results suggest that any proteases inhibited by TPCK are not involved in nuclear genome entry, although due to the pleotropic effects of TPCK (Fig

4.9c), assessment with a more specific protease inhibitor would lead to a better understanding for the necessary triggers for the genome release process.

As summarised (Fig 7.1), at the earliest detectable time (stage 1, within 0.5 hpi), uncoated nuclear genomes were in a comparatively homogeneous, roughly spherical form that had expanded to approximately 3-fold of the volume within virions (Fig 4.2). Thus, while there is a distinct genome decompaction after nuclear import, this is constrained in a relatively regular manner. The distribution of the numbers of genomes appearing in the nucleus and the relationship to MOI bear similarity to

172 those from physical analysis of adenovirus genome entry [406], and are relevant also to conclusions from other studies on the numbers of herpesvirus nuclear genomes that participate in transcription and replication [503]–[505]. I observed that at a standard MOI of 10 pfu/cell, equating to approximately 100 capsids (based on results from Fig 3.6), the mean number of nuclear genome foci was approximately 5, and around 90% of cells had fewer than 10 (Fig 4.1), although a small subset could contain relatively high numbers of foci. Increasing the MOI by 5-fold did not increase nuclear genome numbers 5-fold, indicating the operation of some form of limit on infection and a decreasing efficiency of nuclear import at higher MOIs. This could be explained due to a bottleneck(s) in the pathway of viral genome entry, for example, the saturation of cell surface receptors from the sheer number of particles (both infectious and non-infectious). Further experiments where cells were infected with increasing MOI in the presence of Act D would give further insight into how many genomes would ultimately reach the nucleus, given a generous amount of time for all of the viruses to be transported through potential bottlenecks.

Similar conclusions of increasing MOI not being proportional to nuclear genome numbers were made for Adv nuclear entry [406]. The fate of capsids which do not uncoat at the nuclear pore cannot be discriminated, while for Adv, genomes appear to be lost from the capsid. Based on the mathematical modelling of simultaneous infections with strains of pseudorabies virus expressing individual fluorophores, it has also been estimated that an average of approximately 5 infecting genomes are expressed per cell at an MOI of 10, and that even at MOI 100, the mean is no more than

7 to 8 genomes [503]. It is clear that at the more extreme ends of distributions from my analysis that high numbers of genomes can enter the nucleus (Fig 4.1b). However, at a standard MOI of 10, the mean numbers of physical nuclear genomes are of the same approximation as the numbers of genomes that have been proposed to actively express their genes or replicate [503]. One implication from this is that once imported into the nucleus, the efficiency of transcription may be relatively high.

Many genomes were indeed associated with ICP4 (see Section 7.4), although this does not necessarily

173 show that these genomes are transcribed. Further analysis of viral genomes in relation to their transcriptional output is discussed in Section 7.5.

Figure 7.1 Model for HSV-1 genome dynamics in nuclear entry, compaction, and ICP4 association The proposed model for spatiotemporal dynamics of the infecting HSV-1 genome. The genome is indicated in blue. Progressive phases reflecting observations on certain qualitative features of genome organisation (which will naturally not occur completely synchronously), are demarked as phases 1 to 4 with approximate timings post infection. For clarity and ease of discussion, the inner part of the circle indicates only genomes, while the outer part indicates the association of genomes with the regulatory protein ICP4 (indicated in red). The bottom sections in shaded background indicate features delineated in the presence of inhibitors. Replicated progeny genomes are indicated in phases 3 and 4 in black. Details of the model are as discussed in the text. Figure modified from previously published paper [461].

174

7.3 Genome compaction state transitions during infection

After nuclear import, the HSV-1 genome initially expands and continues to decondense in a series of transitions that could be discriminated into distinct phases by using chemical inhibitors. In the absence of transcription, genomes remained relatively compact and were maintained in that form in the nucleus for at least 8 h (Fig 4.5). De novo virus protein synthesis is therefore not required for the initial compact state of the genome. Transcription in itself contributes little, if at all, to genome decompaction, with genomes maintained as punctate foci, although volumetric analysis with super- resolution microscopy may show minor changes. Previous work has demonstrated histone and nucleosome deposition on viral genomes [208], [506], although more recent work suggests that HSV-

1 genomes are associated with associated with unstable nucleosomes during lytic infection [204],

[205], [215]. Whether histones are responsible for the compact state of the early infecting genomes is unknown, but what is certain is that a factor is required for this and the decondensation process is an active one. Allowing transcription but in the absence of de novo translation revealed a distinct feature that is not observed when transcription was blocked. Genomes move in close proximity to each other while maintaining their compact morphology in a process that I have termed genome congregation (Fig 4.7).

Several mechanisms could contribute to this process. One explanation could be that the transcribed genomes are transported to nearby transcription factories. These are substructures in the nucleus, composed of an average of 8 molecules of RNA polymerase (ranging from 4 to 30 molecules), that are specialised in efficient transcription and mRNA processing [480], [507]. There is evidence to suggest that the RNA polymerase molecules localised in these transcription factories are attached to the nuclear matrix and are stationary, reeling in and displacing the gene that is being transcribed [508],

[509]. The genomes could be reeled in to transcription factories in a similar manner, resulting in genome congregation, and this would not be observed in the absence of transcription. Putative transcription factories can be observed in the cell, marked by RNA Pol II foci (Fig 4.11a), although it is

175 evident that free RNA Pol II is recruited to transcribing genomes since the number of foci increases, as well as their intensity (Fig 4.11b). It would be interesting to observe the localisation of genomes and

RNA Pol II in the presence of CHX. If the above proposal is true, then genome clusters would be expected to localise with pre-existing putative transcription factory RNA Pol II foci. It could also be the case that genome congregation is a host response to infection (with no viral proteins yet made), specifically recognising the process of transcription and sequestering genomes as a result. Alternative explanations are possible, but understanding the mechanism of genome congregation will be important for any full understanding of genome dynamics, competency for transcription, and host responses to infection.

When translation was allowed to occur, the genomes were found localised in irregular shaped foci that were larger than those found in the absence of transcription (Fig 4.6, 3 hpi, white arrows).

These foci were similar in size and frequency to the congregation of genomes that were observed when only transcription was allowed, which is likely to be the genomes decompacting and merging into a larger body. The fact that this only occurs when translation is permissive signifies that a newly synthesised protein is responsible for this process. The potential candidates are the seven viral E gene products that are essential for viral replication since. These are UL9 (origin binding protein), ICP8

(single stranded binding protein), UL8/UL5/UL52 complex (three-unit helicase/primase complex),

UL30 (viral polymerase), and UL42 (polymerase accessory factor) [160], [161]. These could potentially interact with the DNA and alter the macrostructure, which may result in the apparent merging of the previously individual genomes. The possibility of host cell proteins cannot be ruled out, although most proteins known to interact with the viral genome such as DNA ligase and topoisomerase II would be expressed constitutively, thus cellular protein levels not being affected much by a short period of translation inhibition.

During uninhibited infection, genomes underwent further progressive decondensation, eventually becoming difficult to discriminate and dissipating within viral RCs (Fig 4.3a). I frequently

176 observed longer lived residual condensed foci, usually at the periphery of RCs (Fig 4.3c, Fig 7.1 stages

2-4). The intermediate stages of these genome transitions did not require DNA replication per se, with the enlarged and decondensed morphology of infecting genomes in the presence of DNA replication inhibitors being distinct from that observed in the presence of Act D or CHX (Fig 4.6, Fig 12 summary schematic, shaded sectors). I conclude that recruitment of regulatory and/or replication factors combined with the more extensive early transcription, results in further changes and decondensation of the genome. Meanwhile, downstream DNA replication and associated processes e.g. recombination, branching, and extensive late transcription [160], [161] results in more complete decondensation of input genomes at later times. It is possible that the longer-lived condensed foci remaining on the periphery of replication compartments represent either replicated parental strands that remain as foci, or potentially a subset of parental genomes that were not acted upon by either replication or transcription. Rolling circle replication [159], [160], [510] acting on HSV-1EdC would result in one labelled parental strand remaining at the replication fork, while the other parental strand would progressively move away as replication and unlabelled progeny DNA accumulates. It is not currently technically possible to discriminate between these possibilities and other explanations are also possible, but in this regard the pattern of genome association with ICP4 warrants discussion.

7.4 ICP4 association with infecting HSV-1 genomes

While there was heterogeneity at an individual genome level with some genomes not accumulating ICP4, the majority of condensed genomes recruited and were enriched for ICP4 by 1 to

2 hpi (as discussed above, this does not necessarily imply productive transcription from all genomes).

However, as infection progressed there was a clear distinction in this association. Those genomes that remained as condensed foci, found mainly on the periphery of RCs, were selectively depleted for ICP4 and frequently devoid of the protein altogether (Fig 4.3). One possible explanation is that these foci never accumulated ICP4. This would imply significant selectivity, since ICP4 would clearly have been initially recruited to certain other foci, and ICP4 was also present in adjacent decondensed replication

177 centres in the same nuclei. Alternatively, it could have been the case that ICP4 was recruited to many foci but different downstream pathways dictate either maintenance of a more condensed state coupled with displacement of ICP4 or progressive decondensation and association with ICP4. For example, IFI16, a nuclear host antiviral DNA sensor, has been shown to bind viral DNA, resulting in repressive chromatin structures being formed on the genome [354], [362], although under the conditions I used in my investigation, I could not observe any specific localisation between genomes and IFI16. Future work developing methods for the simultaneous visualisation of infecting genome localisation and condensation, active transcription and protein localisation will help address the nature of these relationships revealed here.

My results on spatial analyses are relevant to the interpretation of the many previous biochemical analyses on the nature of the infecting HSV-1 genome. Micrococcal nuclease digestion experiments of the bulk virus genome population strongly indicate that the considerable majority of infecting genomes released from the capsid are randomly digested and not assembled into any conventional nucleosomal organisation [155], [204]–[207], [209], [210]. On the other hand, chromatin immunoprecipitation analyses, which usually address a minor fraction of the total DNA, suggests that histones in some form are associated with at least a population of genomes [202], [208], [211], [511],

[512]. One model attempting to integrate results from different approaches proposes that infecting genomes associate with some form of nucleoprotein complex that includes histones but in a nonconventional highly distributive, rapidly associating/dissociating organisation [204], [215]. I show that after capsid exit and nuclear import, genomes expand but in a constrained and distinct state. They then further decondense in a discernible fashion prior to replication, and potentially associated transcription results in further extensive dissipation within the nucleus. In addition to heterogeneity arising from overlapping temporal transitions, heterogeneity arises from subpopulations of genomes that may not associate with e.g. ICP4, or which at later times remain in a more condensed configuration. Thus, certain proteins may be selectively associated with specific subpopulations of these genomes (c.f. longer lived condensed foci), and thus antibodies to such proteins sample only

178 those genome populations. Future work combining bioorthogonal chemistry for spatial analyses of genomes and ongoing transcription and replication together with immunofluorescence analysis to localise viral and host cell proteins will be necessary to resolve these questions.

7.5 Spatiotemporal analysis of HSV-1 genomes and IE gene transcript

In order to enable simultaneous visualisation of viral genomes and its transcripts, bioorthogonal chemistry was successfully combined with RNAscope, although there were certain experimental procedures that were required. My optimisation experiments found that the RNAscope processing must precede genome detection by click chemistry, otherwise all of the RNA signal would be abrogated (Fig 5.2a). This could be due to RNase contamination of the click chemistry reagents degrading the target RNA, especially since there is no RNase inhibitor in any of the click chemistry buffers. Alternatively, the click chemistry reaction itself could be contributing to the degradation of

RNA, or altering it so that the probes can no longer hybridise with the target RNA, while a simpler explanation would be that the relatively unstable RNA self-catalyses during the click chemistry processing period. In any case, processing for RNAscope first would result in the accumulation of fluorescent dye where the RNA is localised. This accumulated dye is not affected by the click chemistry reaction, and even if the target RNA is damaged or broken down subsequently, the signal for the RNA would still be in place, enabling simultaneous visualisation with genomes. One unintended effect of the RNAscope processing was the visualisation of cytoplasmic genomes (Fig 5.3b). Under normal circumstances, the genomes would only be accessible to the azide-tagged fluorescent coupling reagent once the genome had been uncoated from the capsid (usually at the nuclear pore). However, it appears that the protease digestion step in the RNAscope procedure that is required for accessibility to the target RNAs is also partially digesting the VP5 capsid. Visualisation of cytoplasmic genomes, both coated and uncoated, is not an issue in my current investigation of nuclear transcribing genomes and their transcripts, although it may present an issue when attempting to study cytoplasmic uncoated genomes in a similar context.

179

The total ICP0 transcript abundance increases as the infection progresses (Fig 5.3c), and this kinetic correlates with a previous study where HSV-1 IE gene transcript accumulation was measured with RNA-seq [513]. Although the exact times will differ due to different host cells (MRC5 cells and trigeminal neurons) and MOI, the overall trends are very similar. However, the transcriptional activity of the ICP0 gene appeared to exhibit a biphasic nature (Fig 5.3, summarised in Fig 7.2). During the earliest stages when genomes are punctate foci, a subpopulation are tightly colocalised with transcript as it is produced in the proximity of the template genome. I have defined these colocalising bodies proximal to genomes that are larger than single transcripts as transcription bursts (discussed further in Section 7.7). After initial ICP0 transcription is activated from the infectious genomes, the rate of production from the nucleus appears to decrease for a short window before being reactivated later on in infection. This decrease of ICP0 transcription was mostly due to ICP4 transcriptional suppression, although subsequent experiments with ICP4 mutant virus indicated the presence of a putative cellular repressor (discussed below, Fig 7.2). ICP4 inhibits IE gene transcription such as ICP0, as well as itself in an autoregulatory manner [237]–[239], [514], and from my previous results ICP4 protein specifically associates with genomes at this stage in infection (Fig 4.3a vii-ix). This proposal is also supported by my data where infection is carried out in the presence of CHX. Under these conditions, transcriptional activity is constantly elevated throughout the duration of the infection due to the inhibition of translation, preventing ICP4 protein from suppressing ICP0 transcription (Fig 5.5a, Fig 7.2b). This observation is consistent with previous data where production of IE gene RNA was increased in the presence of CHX [468], where Northern blots were used to observe RNA from whole cell populations.

It is perhaps worth noting that these conclusions assume a relatively constant rate of ICP0 transcript export, and with the current data, a possible explanation of the decreased nuclear transcript in the middle of the two phases (with no drug) being due to accelerated nuclear export cannot be completely ruled out as a contributing factor. However, ICP4 protein inhibition of the ICP0 gene is the most plausible explanation, especially considering the supporting data with drug inhibition.

180

After the initial suppression, the second phase of ICP0 transcription resumes from newly replicated genomes. This is supported by the fact that blocking viral transcription with PAA results in complete shutdown of ICP0 transcription due to ICP4 suppression (Fig 5.5b), indicating that the second transcription phase of ICP0 is replication dependent. The distribution of bulk transcript that is produced late in the timecourse is also morphologically very similar to viral replication compartments

(c.f. Fig 3.4a III and 4.3a). It would be expected that this bulk transcript would colocalise with EdC if it were pulse-chased during the infection, and also ICP8. These results are consistent with a previous study where HFFF cells were infected with HSV-1, where ICP0 transcript detected by RNAscope appeared in RC-like bodies in the nucleus at 6 to 8 hpi with very intense signal [515]. In a similar observation to that made when analysing the relationship of ICP4 protein and genomes (see Section

7.4), there were again residual punctate genomes observed on the periphery of the large bodies of

RNA where the viral RCs are thought to be (Fig 5.3a). These peripheral genomes were selectively depleted for ICP0, suggesting they are not transcriptionally active despite the extremely high levels of viral transcription occurring elsewhere in the nucleus. This provides further cause to believe they are a subset of genomes that successfully entered the nucleus but did not activate their transcriptional program. HSV-1 genome replication results in many single-stranded templates in the nucleoplasm

[160], and it was initially suspected that the signal observed was from the RNAscope probes cross reacting to the single-stranded DNA. This was confirmed not to be the case with DNase treatment not affecting the signal at all (Fig 5.5d), and further evidence that the signal was from bona fide ICP0 RNA would be provided if RNase treatment abrogated all signal.

ICP4 is one of the regulators of HSV-1 transcriptional activity and contributes to the biphasic nature of ICP0 transcription kinetics. However, my results show that it is not the only factor that is involved in early IE gene regulation. Infection with an ICP4 deletion mutant (HSV-1 n12) virus resulted in derepressed IE genes and thus increased their transcriptional activity as expected compared to wt virus. However, infection with the mutant n12 virus in the presence of CHX further increased IE gene activity, indicating that CHX was preventing the synthesis of another protein that was contributing

181 towards viral gene repression. The identity of this is either another viral protein, or a host cell protein that is (a) constitutively synthesised and has a rapid turnover cycle, or (b) rapidly upregulated in response to viral infection, since CHX treatment only starts approximately 1 h before the virus first enters the cell. One putative candidate could be ICP22, since it has been shown in previous investigations that it could repress transcription from viral genes, including genes with IE promoters

[320]. Indeed ICP22 has been shown to inhibit host cell gene expression by preventing phosphorylation of Serine-2 of the RNA Pol II CTD by directly interacting with the kinase responsible, namely cdk9 [315]. However, the evidence for inhibition of viral gene transcription by ICP22 is less convincing since the effect of ICP22 repression has only been shown in transient reporter assays using

IE gene promoters in vitro [320], [516]. Also experiments using ICP22 knock-out viruses showed no increase in IE gene transcription [517], and it has also been shown that expression of UL13, a viral serine-threonine kinase, during the infection results in abrogation of the supposed ICP22 mediated transcriptional repression [320], [518]. Thus, the identity of the unidentified repressor is more likely to be a host cell protein and not a viral one, since ICP4 is the only HSV-1 gene product that has been definitively shown to repress IE genes [246], [483], [484]. The presence of a cellular repressor that suppresses all viral transcription in the absence of viral replication would be entirely consistent with my results and would be a putative new mechanism for general suppression on incoming DNA.

Thus, collating my results I have proposed a model for HSV-1 transcription dynamics in relation to genomes (Fig 7.2). Under normal circumstances, the cellular repressor is transcribed and translated, and is subsequently degraded at a certain rate. Upon infection and nuclear genome entry, the cellular repressor attempts to inhibit all viral transcription, but IE gene transcription is still permissible and produces IE transcript (Fig 7.2a). ICP4 protein is produced and suppresses IE gene transcription, but simultaneously promotes the transcription of early (E) and late (L) genes. The cellular repressor is again attempting to inhibit viral transcription but is unsuccessful when viral replication is permissive

(Fig 7.2b). Once the incoming genome has replicated and there are multiple templates, transcription of IE genes, as well as E and L genes is extremely active from replicated genomes with the cellular

182 repressor being overwhelmed and ineffective (Fig 7.2c). In the absence of any translation, the cellular repressor is overturned and degraded while not being replenished. ICP4 is transcribed but the protein is not produced, resulting in uninhibited IE gene transcription (Fig 7.2d). Only in the absence of viral replication can the cellular repressor fully inhibit the incoming genome even with the attempted activation of E and L genes by ICP4 (Fig 7.2e). One interesting experiment to conduct would be to infect cells in the presence of PAA and observe genomes and an E or L gene transcript, but remove the drug at differing times to reverse its effect. This would give insight into whether the inhibiting effect of the cellular repressor on the incoming genome is dependent upon the complete absence of viral replication, or whether the crucial factor is the time in which the replication process is completed.

Possible candidates for this cellular repressor include components of PML-NBs which are well- documented host antiviral factors involved in genome silencing and transcription inhibition [377],

[378]. PML-NB components such as PML and Sp100 have been shown to be upregulated via the interferon response during HSV-1 infection [519]–[521]. Although ICP0 produced by HSV-1 targets

PML-NB components for degradation, my data shows that PML protein is present and associates with genomes until at least 1 hpi (Fig 6.1a), which may be enough time for PML-NBs to elicit some form of transcriptional suppression. While PML protein in PML-NBs appear to be stable for at least 3 h without any new protein translation (Fig 6.1c), it is possible that CHX treatment disrupts PML-NB stability or prevents the synthesis of other components that have a higher turnover. IFI16 could be another candidate for the unknown host factor due to its previous observed functions as a nuclear DNA sensor that restricts viral transcription, although further investigation would be required on its turnover in the cell and effect on genomes, especially since my preliminary data did not show their specific association with IFI16 (Fig 6.2b). An advanced method of identifying the repressor would be via isolation of the genome with click chemistry and analysing the proteins that are pulled down together

(detailed in Section 7.8). An alternative method is CRISPR Affinity Purification in situ of Regulatory

Elements, commonly known as CAPTURE. This involves a biotinylated dCas9 with a guide RNA to a genomic sequence of choice, and once the dCas9 has targeted the genomic site of interest,

183 formaldehyde crosslinking and subsequent pulldown by streptavidin can isolate the interacting proteins for mass spectrometry analysis [522]. This could be applied to investigate the proteins that are interacting with the IE gene loci in the viral genome.

mRNA

Figure 7.2 Model for IE gene HSV-1 transcription dynamics in relation to genomes proposing a new mechanism of general viral transcription from incoming genomes The proposed model for IE gene transcription during infection as described in the text. The blue object represents the incoming viral genome, while the black object in (c) indicates replicated genomes. ‘IE’ denotes the IE gene loci within the genome; ‘E/L’ denotes the early and late gene loci within the genome; ‘CR’ denotes the putative cellular repressor protein (green), while ‘ICP4’ denotes the ICP4 protein (red). Only one incoming genome is shown for clarity.

184

7.6 Multiplexed spatial analysis of HSV-1 genomes with ICP0 and ICP4 transcript

One intriguing observation was the difference in nuclear distribution between ICP0 and ICP4 when they were multiplexed with each other (Fig 5.8). This was somewhat unexpected, since ICP4 transcript showed extremely similar kinetics to ICP0 transcript during infection, and also showed near- identical responses to drug treatment (Fig 5.6), both being classified in the same IE gene class. An explanation for this difference in mobility of transcript could be due to the fact that ICP4 mRNA is not spliced and is produced from monocistronic transcript [523]. On the other hand, ICP0 mRNA is polycistronic, and is produced by splicing out two introns [523]. ICP0 mRNA is one of the few HSV-1 transcripts to undergo splicing, the others being LAT, ICP22, and ICP47 [74]. Current evidence shows that splicing of pre-mRNA takes place in close proximity to the transcription site, and in some cases splicing and pre-mRNA maturation occur co-transcriptionally [524]–[526]. It is possible that the requirement of ICP0 pre-mRNA splicing delays the release of ICP0 transcript, causing accumulation around the transcription site that is genomes, while ICP4 transcript is released quicker from the transcription site and is transported away faster. Further quantitative analysis would likely confirm the observations made above by measuring parameters such as the number of transcript bursts in each cell for each transcript, a measure of the spread of transcript using clustering analysis, and distances of individual transcripts from the nearest genome. It would also be of interest to try and detect ICP0 mRNA intronic sequences with RNAscope (the current probes only detect ICP0 mRNA exons), and see if they localise to genomes with ICP0 transcription bursts. Whether this difference in transcript behaviour has any biological significance such as implications on their function remains to be seen.

Furthermore, there appears to be differential genome template usage for ICP0 and ICP4 transcription between different genomes in the same nucleus (Fig 5.8). This is an observation that has previously not been made due to the inability to observe individual genomes and transcripts, and

185 analysis by population-based methods masking the behaviour of single cells and genomes.

Heterogeneity in genome IE gene transcription may have implications in understanding how the IE transcriptional programme is regulated, and how viral gene transcription is regulated by host antiviral suppression at the single molecule level.

7.7 Transcriptional bursting

One prominent feature that was observed in my RNAscope experiments were transcription bursts from viral genomes. I have defined these as being transcript bodies that are larger and more intense than single transcripts that are fully or partially colocalised with EdC-labelled genomes (Fig

5.4). There is currently much evidence in the literature for the transcription bursting paradigm which proposes a model where gene transcription stochastically fluctuates between active and inactive states, resulting in bursts of RNA production [435], [438], [439], [450]. Although much of the research has been conducted with eukaryotic cells [432], [437], [439], [443], transcription bursting has been observed with viral genes such as the HIV-1 genome [527], although this may not be surprising with viruses that use host cell machinery for transcription of their genes. Previous investigations have visualised transcription bursts that are morphologically similar to those identified in my experiments

[439], [482], although in these studies the gene locus was inferred by indirect methods, for example, the use of intron probes of the target mRNA [439]. In contrast, the direct visualisation of the HSV-1 genome in my experiments allows me to conclusively confirm the close spatial relationship of bursts and the template that it is transcribed from. It cannot be ruled out that these large transcriptional bodies proximal to genomes are a result of slow transcript translation away from the transcription site, and that the actual rate of transcription is constant. However, the evidence overwhelmingly suggests that they are indeed a product of transcriptional bursting.

Heterogeneity is observed with respect to the presence of transcription busts within a single cell, and only a subpopulation appears to be transcriptionally active for the transcript being observed at any one time. Combining this technique with ICP4 immunofluorescence would give quantitative

186 information about the relationship between genomes that accumulate ICP4 and those that are actively transcribing (see Section 7.4). From the current data it cannot be deciphered whether the genomes that are not associated with transcript are permanently inactivated by host cell antiviral responses, or whether they are in the refractory phase of the burst kinetic. Combining this with immunofluorescence of host cell antiviral factors such as PML would further the understanding of any relationship between genome association of host cell factors and their transcriptional state. For example, it may be that genomes that evade PML association are the most transcriptionally active and are the key to the virus counteracting host antiviral responses. Thus, further quantitative analysis would yield much desired information about the genome transcriptional output and the state of the cell. Another interesting observation was that in the absence of translation, IE gene expression was highly active, and upon visual inspection there appeared to be a higher proportion of genomes active and associated with transcription bursts (Fig 5.5). Confirmation of this with quantitative analysis, together with information on the state of the cell would aid in understanding what factors are most important in dictating the transcriptional output, and hence the progression of infection, by observing the cells and genomes under these altered conditions,

Although multi-genome infections are informative, they are extremely complex to analyse with multiple transcripts being produced from multiple genomes. Thus, single genome infection systems provide a simpler system with fewer variables for analysis where all transcripts originate from a single genome (Fig 5.10). This allows quantitative measurement of parameters such as genome location inside the nucleus and association with host antiviral factors and chromatin assembly factors.

This will then be useful in determining how they influence the transcription state of the genome such as the frequency of transcription bursts on the genomes, as well as the nuclear and cytoplasmic distribution of nascent transcripts. The data obtained would enable further understanding of the relationship between genomes and their transcriptional output.

187

7.8 Future directions

Having established the methodology of labelling viral genomes with EdC for direct genome visualisation, it would be of great interest to assess the effects of key viral mutations on nuclear genome entry in order to elucidate the exact mechanism behind the process. Mutant candidates include HSV-1 VP1-2ΔNLS which lacks the nuclear localisation signal of VP1-2, and HSV-1 ΔUL37 which lacks the pUL25 protein, both of which are implicated to contribute towards nuclear genome entry

[102], [528]. Growing these deletion mutants requires the use of complementary cell lines instead of standard RPE-1 cells, and so generating EdC-labelled mutants would involve further optimisation experiments, such as confirming non-toxicity of EdC in the complementary cells and the labelling efficiency of the EdC-incorporated mutant virus genomes. Another approach would be to use siRNA- induced mRNA knockdown of the gene product of interest in wild type HSV-1 infections to assess their absence.

A different direction of investigation would be to use bioorthogonal chemistry for biochemical analysis. My experiments so far have involved covalently coupling the EdC-labelled viral genome to azide-tagged fluorescent capture reagents for spatial analysis. However, the EdC can be coupled to a biotinylated capture reagent for pulldown and isolation. By isolating HSV-1 genomes at different times during the infection, temporal information of the abundance and types of proteins associated with them can be obtained. This would be especially useful in identifying known and unknown host cell antiviral factors that may influence genome activity, as well as the newly synthesised protein that is responsible for the apparent merging of congregating genomes. Another method would be to use the aforementioned CAPTURE purification system. Similar analysis has been conducted on replicating

HSV-1 genomes [423], but this would be the first analysis to be done on input genomes. Pulldown by bioorthogonal chemistry and protein composition analysis by mass spectrometry has successfully been carried out with bioorthogonally labelled protein synthesised during HSV-1 infection [529].

However, the amount of DNA in input genomes is considerably less than the amount of newly

188 synthesised protein, and so large-scale infection and optimisation would be required for this biochemical analysis to be successful.

Finally, another extension of using click chemistry would be to generate HSV-1 viruses that have incorporated non-canonical amino acids that can be coupled to a capture reagent. These viruses have their proteins bio-orthogonally tagged and can be visualised by fluorescent capture reagents to spatially analyse viral protein localisation after entry into the cell [530] . Viruses could also be isolating with biotinylated capture reagents, and the associated host cell proteins analysed by mass spectrometry. This would provide insight into the spatiotemporal fate of input viral proteins during infection as well as temporal information and identification of antiviral host cell factors that act on viral proteins.

7.9 Concluding remarks

Using compatible biorthogonal nucleoside precursors for genome labelling in HSV-1 infected cells and quantitative individual particle analysis, I demonstrate efficient precursor incorporation resulting in quantitative detection on an individual particle basis in the population of progeny virus. I report a comprehensive analysis in infected cells of genome dynamics during capsid exit and nuclear import in which I (i) demonstrate qualitative transitions in genome condensation state linked to transcription and replication (ii) reveal novel processes in genome congregation dependent upon transcription and (iii) show the temporal switching in regulatory protein recruitment (represented by

ICP4) to distinct genome compartments. Furthermore, I have developed simultaneous visualisation of

HSV-1 genomes and their product transcripts. This has enabled me to conduct spatiotemporal analysis of IE gene transcription kinetics in relation to the template genomes. I demonstrate a biphasic nature of ICP0 transcription, heterogeneity of genome transcription bursting, and differences in IE gene transcript behaviour with multiplex analysis, all of which have laid the foundations for further multivariate quantitative analysis. Finally, I have started to decipher the relationship of infectious genomes with host cell antiviral factors, demonstrating the association of PML and γH2AX with

189 genomes, as well as increased viral activity in the absence of antiviral DNA sensors IFI16 and cGAS, and potentially a new mechanism for general suppression of incoming viral genomes.

Altogether my results reveal novel aspects of the spatiotemporal dynamics of HSV-1 genome uncoating and transport, together with their relationship with mRNA transcript that can be integrated with previous biochemically based analysis and provide a framework for future investigation of the virus genome and its relationship with the host cell.

190

8. References

[1] A. Whittaker, Gary R.; Kann, Michael; Helenius, G. R. Whittaker, M. Kann, and A. Helenius, “Viral entry into the nucleus,” Annu Rev Cell Dev Biol, vol. 16, pp. 627–651, 2000.

[2] N. Fay and N. Panté, “Nuclear entry of DNA viruses.,” Front. Microbiol., vol. 6, p. 467, 2015.

[3] G. R. Whittaker and A. Helenius, “Nuclear import and export of viruses and virus genomes,” Virology, vol. 246, no. 1, pp. 1–23, 1998.

[4] T. Hennig and P. O’Hare, “Viruses and the nuclear envelope,” Curr. Opin. Cell Biol., vol. 34, pp. 113–121, 2015.

[5] U. F. Greber and M. Fornerod, “Nuclear import in viral infections,” Curr Top Microbiol Immunol, vol. 285, pp. 109–138, 2005.

[6] G. Whittaker, “The role of nuclear import and export in influenza virus infection,” Trends Cell Biol., vol. 6, no. 2, pp. 67–71, Feb. 1996.

[7] M. Stevenson, “Portals of entry:uncovering HIV nuclear transport pathways,” Trends Cell Biol., vol. 6, no. 1, pp. 9–15, Jan. 1996.

[8] O. Kobiler, N. Drayman, V. Butin-Israeli, and A. Oppenheim, “Virus strategies for passing the nuclear envelope barrier,” Nucl. (United States), vol. 3, no. 6, 2012.

[9] S. Cohen, S. Au, N. Panté, and N. Pante, “How viruses access the nucleus,” Biochim Biophys Acta, vol. 1813, no. 9, pp. 1634–1645, Sep. 2011.

[10] G. R. Whittaker, “Virus nuclear import,” Adv. Drug Deliv. Rev., vol. 55, no. 6, pp. 733–747, Jun. 2003.

[11] B. Roizman and A. Sears, “Herpes simplex viruses and their replication,” in Fields Virology, B. Fields, D. Knipe, and P. Howley, Eds. Philadelphia: Lippincott-Raven, 1996, pp. 2:2231–95.

[12] T. Shenk, “Adenoviridae: The viruses and their replication,” in Fields Virology, B. Fields, D. Knipe, and P. Howley, Eds. Philadelphia: Lippincott-Raven, 1996, pp. 2:2111–48.

[13] D. Puntener and U. F. Greber, “DNA-tumor virus entry—From plasma membrane to the nucleus,” Semin. Cell Dev. Biol., vol. 20, no. 5, pp. 631–642, Jul. 2009.

[14] G. R. Nemerow, D. A. Cheresh, and T. J. Wickham, “Adenovirus entry into host cells: a role for αv integrins,” Trends Cell Biol., vol. 4, no. 2, pp. 52–55, Feb. 1994.

[15] U. F. Greber, M. Suomalainen, R. P. Stidwill, K. Boucke, M. W. Ebersold, and A. Helenius, “The role of the nuclear pore complex in adenovirus DNA entry,” EMBO J., vol. 16, no. 19, pp. 5998– 6007, Oct. 1997.

[16] S. Strunze et al., “Kinesin-1-Mediated Capsid Disassembly and Disruption of the Nuclear Pore Complex Promote Virus Infection,” Cell Host Microbe, vol. 10, no. 3, pp. 210–223, Sep. 2011.

[17] M. Suomalainen, M. Y. Nakano, S. Keller, K. Boucke, R. P. Stidwill, and U. F. Greber, “Microtubule-dependent plus- and minus end-directed motilities are competing processes for

191

nuclear targeting of adenovirus.,” J. Cell Biol., vol. 144, no. 4, pp. 657–72, Feb. 1999.

[18] P. K. Chatterjee, M. E. Vayda, S. J. Flint, T. Guan, G. Nemerow, and L. Gerace, “Interactions among the three adenovirus core proteins.,” J. Virol., vol. 55, no. 2, pp. 379–86, Aug. 1985.

[19] J. M. Kenney, C.-H. von Bonsdorff, M. Nassal, and S. D. Fuller, “Evolutionary conservation in the hepatitis B virus core structure: comparison of human and duck cores,” Structure, vol. 3, no. 10, pp. 1009–1019, Oct. 1995.

[20] R. A. Crowther et al., “Three-dimensional structure of hepatitis B virus core particles determined by electron cryomicroscopy,” Cell, vol. 77, no. 6, pp. 943–950, Jun. 1994.

[21] M. Kann, B. Sodeik, A. Vlachou, W. H. Gerlich, and A. Helenius, “Phosphorylation-dependent binding of hepatitis B virus core particles to the nuclear pore complex.,” J. Cell Biol., vol. 145, no. 1, pp. 45–55, Apr. 1999.

[22] N. Panté and M. Kann, “Nuclear Pore Complex Is Able to Transport Macromolecules with Diameters of ∼39 nm,” Mol. Biol. Cell, vol. 13, no. 2, pp. 425–434, Feb. 2002.

[23] B. Rabe et al., “Nuclear Entry of Hepatitis B Virus Capsids Involves Disintegration to Protein Dimers followed by Nuclear Reassociation to Capsids,” PLoS Pathog., vol. 5, no. 8, p. e1000563, Aug. 2009.

[24] S. Cohen, A. R. Behzad, J. B. Carroll, and N. Pante, “Parvoviral nuclear import: bypassing the host nuclear-transport machinery,” J. Gen. Virol., vol. 87, no. 11, pp. 3209–3213, Nov. 2006.

[25] M. Porwal et al., “Parvoviruses Cause Nuclear Envelope Breakdown by Activating Key Enzymes of Mitosis,” PLoS Pathog., vol. 9, no. 10, p. e1003671, Oct. 2013.

[26] S. Roy et al., “Dominant-negative caveolin inhibits H-Ras function by disrupting cholesterol-rich plasma membrane domains,” Nat. Cell Biol., vol. 1, no. 2, pp. 98–105, Jun. 1999.

[27] J. Kartenbeck, H. Stukenbrok, and A. Helenius, “Endocytosis of simian virus 40 into the endoplasmic reticulum.,” J. Cell Biol., vol. 109, no. 6, pp. 2721–2729, Dec. 1989.

[28] H. Ewers et al., “GM1 structure determines SV40-induced membrane invagination and infection,” Nat. Cell Biol., vol. 12, no. 1, pp. 11–18, Jan. 2010.

[29] B. Tsai, J. M. Gilbert, T. Stehle, W. Lencer, T. L. Benjamin, and T. A. Rapoport, “Gangliosides are receptors for murine polyoma virus and SV40,” EMBO J., vol. 22, no. 17, pp. 4346–4355, Sep. 2003.

[30] R. Geiger et al., “BAP31 and BiP are essential for dislocation of SV40 from the endoplasmic reticulum to the cytosol,” Nat. Cell Biol., vol. 13, no. 11, pp. 1305–1314, Nov. 2011.

[31] K. M. Giorda and D. N. Hebert, “Viroporins Customize Host Cells for Efficient Viral Propagation,” DNA Cell Biol., vol. 32, no. 10, pp. 557–564, Oct. 2013.

[32] R. Daniels, N. M. Rusan, P. Wadsworth, and D. N. Hebert, “SV40 VP2 and VP3 Insertion into ER Membranes Is Controlled by the Capsid Protein VP1: Implications for DNA Translocation out of the ER,” Mol. Cell, vol. 24, no. 6, pp. 955–966, Dec. 2006.

[33] G. S. Baer, T. S. Dermody, and B. Tsai, “Mutations in reovirus outer-capsid protein sigma3

192

selected during persistent infections of L cells confer resistance to protease inhibitor E64.,” J. Virol., vol. 71, no. 7, pp. 4921–8, Jul. 1997.

[34] A. Nakanishi, D. Shum, H. Morioka, E. Otsuka, and H. Kasamatsu, “Interaction of the Vp3 Nuclear Localization Signal with the Importin 2/ Heterodimer Directs Nuclear Entry of Infecting Simian Virus 40,” J. Virol., vol. 76, no. 18, pp. 9368–9377, 2002.

[35] A. Nakanishi, J. Clever, M. Yamada, P. P. Li, and H. Kasamatsu, “Association with capsid proteins promotes nuclear targeting of simian virus 40 DNA,” Proc. Natl. Acad. Sci., vol. 93, no. 1, pp. 96–100, Jan. 1996.

[36] A. Nakanishi, P. P. Li, Q. Qu, Q. H. Jafri, and H. Kasamatsu, “Molecular dissection of nuclear entry-competent SV40 during infection,” Virus Res., vol. 124, no. 1–2, pp. 226–230, Mar. 2007.

[37] V. Butin-Israeli et al., “Simian virus 40 induces lamin A/C fluctuations and nuclear envelope deformation during cell entry,” Nucleus, vol. 2, no. 4, pp. 320–330, Jul. 2011.

[38] M. I. Bukrinsky et al., “A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells,” Nature, vol. 365, no. 6447, pp. 666–669, Oct. 1993.

[39] U. von Schwedler, R. S. Kornbluth, and D. Trono, “The nuclear localization signal of the matrix protein of human immunodeficiency virus type 1 allows the establishment of infection in macrophages and quiescent T lymphocytes.,” Proc. Natl. Acad. Sci. U. S. A., vol. 91, no. 15, pp. 6992–6, Jul. 1994.

[40] Y. Jenkins, M. McEntee, K. Weis, and W. C. Greene, “Characterization of HIV-1 vpr nuclear import: analysis of signals and pathways.,” J. Cell Biol., vol. 143, no. 4, pp. 875–85, Nov. 1998.

[41] C. Depienne et al., “Characterization of the Nuclear Import Pathway for HIV-1 Integrase,” J. Biol. Chem., vol. 276, no. 21, pp. 18102–18107, May 2001.

[42] E. Le Rouzic et al., “Docking of HIV-1 Vpr to the nuclear envelope is mediated by the interaction with the nucleoporin hCG1.,” J. Biol. Chem., vol. 277, no. 47, pp. 45091–8, Nov. 2002.

[43] V. Zennou, C. Petit, D. Guetard, U. Nerhbass, L. Montagnier, and P. Charneau, “HIV-1 Genome Nuclear Import Is Mediated by a Central DNA Flap,” Cell, vol. 101, no. 2, pp. 173–185, Apr. 2000.

[44] I. Kemler, G. Whittaker, and A. Helenius, “Nuclear Import of Microinjected Influenza Virus Ribonucleoproteins,” Virology, vol. 202, no. 2, pp. 1028–1033, Aug. 1994.

[45] K. Martin and A. Helenius, “Transport of incoming influenza virus nucleocapsids into the nucleus,” Trends Cell Biol., vol. 2, no. 1, p. 8, 2003.

[46] P. Wang, P. Palese, and R. E. O’Neill, “The NPI-1/NPI-3 (karyopherin alpha) binding site on the influenza a virus nucleoprotein NP is a nonconventional nuclear localization signal.,” J. Virol., vol. 71, no. 3, pp. 1850–6, Mar. 1997.

[47] R. E. O’Neill and P. Palese, “NPI-1, the human homolog of SRP-1, Interacts with influenza virus nucleoprotein,” Virology, vol. 206, no. 1, pp. 116–125, Jan. 1995.

[48] W. W. Wu, Y.-H. B. Sun, and N. Pante, “Nuclear import of influenza A viral ribonucleoprotein complexes is mediated by two nuclear localization sequences on viral nucleoprotein,” Virol. J., vol. 4, no. 1, p. 49, Jun. 2007.

193

[49] J. F. Cros, A. García-Sastre, and P. Palese, “An Unconventional NLS is Critical for the Nuclear Import of the Influenza A Virus Nucleoprotein and Ribonucleoprotein,” Traffic, vol. 6, no. 3, pp. 205–213, Mar. 2005.

[50] D. Knipe et al., Fields Virology Vol.2, 5th ed. Lippincott Williams & Wilkins, Philadelphia, 2007.

[51] B. Roizman et al., “Herpesviridae,” Intervirology, vol. 16, no. 4, pp. 201–217, 1981.

[52] G. A. Karasneh and D. Shukla, “Herpes simplex virus infects most cell types in vitro: clues to its success.,” Virol. J., vol. 8, no. 1, p. 481, Oct. 2011.

[53] D. J. McGeoch, D. Gatherer, and A. Dolan, “On phylogenetic relationships among major lineages of the Gammaherpesvirinae,” J. Gen. Virol., vol. 86, no. 2, pp. 307–316, Feb. 2005.

[54] B. Roizman and P. Pellett, “The family herpesviridae: a brief introduction,” in Fields Virology, D. Knipe, P. Howley, and D. Griffin, Eds. Lippincott Williams & Wilkins, PA, USA, 2007, pp. 2479– 2500.

[55] E. E. Heldwein and C. Krummenacher, “Entry of herpesviruses into mammalian cells,” Cell. Mol. Life Sci., vol. 65, no. 11, pp. 1653–1668, Jun. 2008.

[56] J. Akhtar and D. Shukla, “Viral entry mechanisms: cellular and viral mediators of herpes simplex virus entry,” FEBS J., vol. 276, no. 24, pp. 7228–7236, Dec. 2009.

[57] W. E. Lafferty, L. Downey, C. Celum, and A. Wald, “Herpes Simplex Virus Type 1 as a Cause of Genital Herpes: Impact on Surveillance and Prevention,” J. Infect. Dis., vol. 181, no. 4, pp. 1454– 1457, Apr. 2000.

[58] S. A. Connolly, J. O. Jackson, T. S. Jardetzky, and R. Longnecker, “Fusing structure and function: a structural view of the herpesvirus entry machinery,” Nat. Rev. Microbiol., vol. 9, no. 5, pp. 369–381, May 2011.

[59] V. Sancho-Shimizu et al., “Genetic susceptibility to herpes simplex virus 1 encephalitis in mice and humans,” Curr. Opin. Allergy Clin. Immunol., vol. 7, no. 6, pp. 495–505, Dec. 2007.

[60] H. Shoji, K. Azuma, Y. Nishimura, H. Fujimoto, Y. Sugita, and Y. Eizuru, “Acute viral encephalitis: The recent progress,” Intern. Med., vol. 41, no. 6, pp. 420–428, 2002.

[61] K. R. Wilhelmus et al., “Acyclovir for the Prevention of Recurrent Herpes Simplex Virus Eye Disease,” N. Engl. J. Med., vol. 339, no. 5, pp. 300–306, Jul. 1998.

[62] A. Steven and P. Spear, “Herpesvirus capsid assembly and envelopment,” in Structural Biology of Viruses, W. Chiu, R. Burnett, and R. Garcea, Eds. Oxford University Press, New York, NY, 1997, pp. 312–351.

[63] K. Grünewald et al., “Three-Dimensional Structure of Herpes Simplex Virus from Cryo-Electron Tomography,” Science (80-. )., vol. 302, no. 5649, pp. 1396–1398, Nov. 2003.

[64] J. Zhang, C.-H. Nagel, B. Sodeik, and R. Lippé, “Early, active, and specific localization of herpes simplex virus type 1 gM to nuclear membranes.,” J. Virol., vol. 83, no. 24, pp. 12984–97, Dec. 2009.

[65] S. Loret, G. Guay, R. Lippe, and R. Lippé, “Comprehensive characterization of extracellular

194

herpes simplex virus type 1 virions,” J Virol, vol. 82, no. 17, pp. 8605–8618, Sep. 2008.

[66] C. I. Ace, T. A. McKee, J. M. Ryan, J. M. Cameron, and C. M. Preston, “Construction and characterization of a herpes simplex virus type 1 mutant unable to transinduce immediate- early gene expression.,” J. Virol., vol. 63, no. 5, pp. 2260–9, May 1989.

[67] A. Barzilai, I. Zivony-Elbom, R. Sarid, E. Noah, and N. Frenkel, “The herpes simplex virus type 1 vhs-UL41 gene secures viral replication by temporarily evading apoptotic cellular response to infection: Vhs-UL41 activity might require interactions with elements of cellular mRNA degradation machinery.,” J. Virol., vol. 80, no. 1, pp. 505–13, Jan. 2006.

[68] V. Jovasevic, L. Liang, and B. Roizman, “Proteolytic cleavage of VP1-2 is required for release of herpes simplex virus 1 DNA into the nucleus,” J Virol, vol. 82, no. 7, pp. 3311–3319, 2008.

[69] M. Donnelly, J. Verhagen, and G. Elliott, “RNA Binding by the Herpes Simplex Virus Type 1 Nucleocytoplasmic Shuttling Protein UL47 Is Mediated by an N-Terminal Arginine-Rich Domain That Also Functions as Its Nuclear Localization Signal,” J. Virol., vol. 81, no. 5, pp. 2283–2296, Mar. 2007.

[70] B. R. Bowman, M. L. Baker, F. J. Rixon, W. Chiu, and F. A. Quiocho, “Structure of the herpesvirus major capsid protein,” EMBO J., vol. 22, no. 4, p. 757, 2003.

[71] T. C. Mettenleiter, B. G. Klupp, and H. Granzow, “Herpesvirus assembly: a tale of two membranes,” Curr. Opin. Microbiol., vol. 9, no. 4, pp. 423–429, Aug. 2006.

[72] J. C. Brown and W. W. Newcomb, “Herpesvirus capsid assembly: insights from structural analysis,” Curr. Opin. Virol., vol. 1, no. 2, pp. 142–149, Aug. 2011.

[73] D. Tombácz et al., “Long-Read Isoform Sequencing Reveals a Hidden Complexity of the Transcriptional Landscape of Herpes Simplex Virus Type 1,” Front. Microbiol., vol. 8, p. 1079, Jun. 2017.

[74] J. Rajčáni, V. Andrea, and R. Ingeborg, “Peculiarities of Herpes Simplex Virus (HSV) Transcription: An overview,” Virus Genes, vol. 28, no. 3, pp. 293–310, Apr. 2004.

[75] D. J. McGeoch, F. J. Rixon, and A. J. Davison, “Topics in herpesvirus genomics and evolution,” Virus Res., vol. 117, no. 1, pp. 90–104, Apr. 2006.

[76] S. J. Macdonald, H. H. Mostafa, L. A. Morrison, and D. J. Davido, “Genome sequence of herpes simplex virus 1 strain KOS.,” J. Virol., vol. 86, no. 11, pp. 6371–2, Jun. 2012.

[77] T. Du, Z. Han, G. Zhou, G. Zhou, and B. Roizman, “Patterns of accumulation of miRNAs encoded by herpes simplex virus during productive infection, latency, and on reactivation.,” Proc. Natl. Acad. Sci. U. S. A., vol. 112, no. 1, pp. E49-55, Jan. 2015.

[78] E. D. Kieff, S. L. Bachenheimer, and B. Roizman, “Size, composition, and structure of the deoxyribonucleic acid of herpes simplex virus subtypes 1 and 2.,” J. Virol., vol. 8, no. 2, pp. 125– 32, Aug. 1971.

[79] S. Wadsworth, R. J. Jacob, and B. Roizman, “Anatomy of herpes simplex virus DNA. II. Size, composition, and arrangement of inverted terminal repetitions.,” J. Virol., vol. 15, no. 6, pp. 1487–97, Jun. 1975.

195

[80] Viralzone, “Alphaherpesvirinae virion,” 2017. [Online]. Available: https://viralzone.expasy.org/15?outline=all_by_species.

[81] H. Gruffat, R. Marchione, and E. Manet, “Herpesvirus Late Gene Expression: A Viral-Specific Pre-initiation Complex Is Key.,” Front. Microbiol., vol. 7, p. 869, 2016.

[82] J. Arii et al., “Entry of herpes simplex virus 1 and other alphaherpesviruses via the paired immunoglobulin-like type 2 receptor alpha.,” J. Virol., vol. 83, no. 9, pp. 4520–7, May 2009.

[83] A. V. Nicola and S. E. Straus, “Cellular and Viral Requirements for Rapid Endocytic Entry of Herpes Simplex Virus,” J. Virol., vol. 78, no. 14, pp. 7508–7517, Jul. 2004.

[84] B. Sodeik, M. W. Ebersold, and A. Helenius, “Microtubule-mediated transport of incoming herpes simplex virus 1 capsids to the nucleus,” J Cell Biol, vol. 136, no. 5, pp. 1007–1021, Mar. 1997.

[85] P. G. Spear and R. Longnecker, “Herpesvirus Entry: an Update,” J. Virol., vol. 77, no. 19, pp. 10179–10185, Oct. 2003.

[86] A. V Nicola, A. M. McEvoy, and S. E. Straus, “Roles for endocytosis and low pH in herpes simplex virus entry into HeLa and Chinese hamster ovary cells.,” J. Virol., vol. 77, no. 9, pp. 5324–32, May 2003.

[87] A. V Nicola, J. Hou, E. O. Major, and S. E. Straus, “Herpes simplex virus type 1 enters human epidermal keratinocytes, but not neurons, via a pH-dependent endocytic pathway,” J Virol, vol. 79, no. 12, pp. 7609–7616, 2005.

[88] E. Rahn, P. Petermann, M.-J. Hsu, F. J. Rixon, and D. Knebel-Mörsdorf, “Entry Pathways of Herpes Simplex Virus Type 1 into Human Keratinocytes Are Dynamin- and Cholesterol- Dependent,” PLoS One, vol. 6, no. 10, p. e25464, Oct. 2011.

[89] B. C. Herold, R. J. Visalli, N. Susmarski, C. R. Brandt, and P. G. Spear, “Glycoprotein C- independent binding of herpes simplex virus to cells requires cell surface heparan sulphate and glycoprotein B,” J. Gen. Virol., vol. 75, no. 6, pp. 1211–1222, Jun. 1994.

[90] T. Gianni, M. Amasio, and G. Campadelli-Fiume, “Herpes simplex virus gD forms distinct complexes with fusion executors gB and gH/gL in part through the C-terminal profusion domain.,” J. Biol. Chem., vol. 284, no. 26, pp. 17370–82, Jun. 2009.

[91] E. Avitabile, C. Forghieri, and G. Campadelli-Fiume, “Cross talk among the glycoproteins involved in herpes simplex virus entry and fusion: the interaction between gB and gH/gL does not necessarily require gD.,” J. Virol., vol. 83, no. 20, pp. 10752–60, Oct. 2009.

[92] K. Döhner et al., “Function of Dynein and Dynactin in Herpes Simplex Virus Capsid Transport,” Mol. Biol. Cell, vol. 13, no. 8, pp. 2795–2809, Aug. 2002.

[93] B. J. Kelly, C. Fraefel, A. L. Cunningham, and R. J. Diefenbach, “Functional roles of the tegument proteins of herpes simplex virus type 1,” Virus Res., vol. 145, no. 2, pp. 173–186, Nov. 2009.

[94] H. Mabit et al., “Intact microtubules support adenovirus and herpes simplex virus infections.,” J. Virol., vol. 76, no. 19, pp. 9962–71, Oct. 2002.

[95] K. Radtke et al., “Plus- and minus-end directed microtubule motors bind simultaneously to

196

herpes simplex virus capsids using different inner tegument structures,” PLoS Pathog, vol. 6, no. 7, p. e1000991, Jul. 2010.

[96] F. Abaitua and P. O’Hare, “Identification of a Highly Conserved, Functional Nuclear Localization Signal within the N-Terminal Region of Herpes Simplex Virus Type 1 VP1-2 Tegument Protein,” J. Virol., vol. 82, no. 11, pp. 5234–5244, Jun. 2008.

[97] W. Batterson, D. Furlong, and B. Roizman, “Molecular genetics of herpes simplex virus. VIII. further characterization of a temperature-sensitive mutant defective in release of viral DNA and in other stages of the viral reproductive cycle.,” J. Virol., vol. 45, no. 1, pp. 397–407, Jan. 1983.

[98] P. M. Ojala, B. Sodeik, M. W. Ebersold, U. Kutay, and A. Helenius, “Herpes simplex virus type 1 entry into host cells: reconstitution of capsid binding and uncoating at the nuclear pore complex in vitro.,” Mol. Cell. Biol., vol. 20, no. 13, pp. 4922–31, Jul. 2000.

[99] Y. Yamauchi, K. Kiriyama, N. Kubota, H. Kimura, J. Usukura, and Y. Nishiyama, “The UL14 tegument protein of herpes simplex virus type 1 is required for efficient nuclear transport of the alpha transinducing factor VP16 and viral capsids.,” J. Virol., vol. 82, no. 3, pp. 1094–106, Feb. 2008.

[100] D. Pasdeloup, D. Blondel, A. L. Isidro, and F. J. Rixon, “Herpesvirus capsid association with the nuclear pore complex and viral DNA release involve the nucleoporin CAN/Nup214 and the capsid protein pUL25,” J Virol, vol. 83, no. 13, pp. 6610–6623, Jul. 2009.

[101] A. M. Copeland, W. W. Newcomb, and J. C. Brown, “Herpes simplex virus replication: roles of viral proteins and nucleoporins in capsid-nucleus attachment.,” J. Virol., vol. 83, no. 4, pp. 1660–8, Feb. 2009.

[102] A. P. E. Roberts, F. Abaitua, P. O’Hare, D. McNab, F. J. Rixon, and D. Pasdeloup, “Differing roles of inner tegument proteins pUL36 and pUL37 during entry of herpes simplex virus type 1,” J Virol, vol. 83, no. 1, pp. 105–116, Jan. 2009.

[103] K. Rode et al., “Uncoupling uncoating of herpes simplex virus genomes from their nuclear import and gene expression.,” J. Virol., vol. 85, no. 9, pp. 4271–83, May 2011.

[104] E. Solomon, L. Berg, and D. Martin, Biology 8th Edition. Brooks/Cole Publishing Co, 2008.

[105] X. Liu, D. A. Bushnell, and R. D. Kornberg, “RNA Polymerase II Transcription: Structure and Mechanism,” Biochim. Biophys. Acta, vol. 1829, no. 1, p. 2, Jan. 2013.

[106] S. Hahn, “Structure and mechanism of the RNA Polymerase II transcription machinery,” Nat. Struct. Mol. Biol., vol. 11, no. 5, p. 394, 2004.

[107] S. Sainsbury, C. Bernecky, and P. Cramer, “Structural basis of transcription initiation by RNA polymerase II,” Nat. Rev. Mol. Cell Biol., vol. 16, no. 3, pp. 129–143, Mar. 2015.

[108] P. Cramer, “Organization and regulation of gene transcription,” Nature, vol. 573, no. 7772, pp. 45–54, Sep. 2019.

[109] M. Ptashne and A. Gann, Genes and Signals. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 2002.

197

[110] M. P. Cosma, “Ordered recruitment: gene-specific mechanism of transcription activation.,” Mol. Cell, vol. 10, no. 2, pp. 227–36, Aug. 2002.

[111] G. Orphanides, T. Lagrange, and D. Reinberg, “The general transcription factors of RNA polymerase II.,” Genes Dev., vol. 10, no. 21, pp. 2657–2683, Nov. 1996.

[112] W. Wang, M. Carey, and J. Gralla, “Polymerase II promoter activation: closed complex formation and ATP-driven start site opening,” Science (80-. )., vol. 255, no. 5043, pp. 450–453, Jan. 1992.

[113] D. S. Luse and G. A. Jacob, “Abortive initiation by RNA polymerase II in vitro at the adenovirus 2 major late promoter.,” J. Biol. Chem., vol. 262, no. 31, pp. 14990–7, Nov. 1987.

[114] F. C. P. Holstege, U. Fiedler, and H. T. Timmers, “Three transitions in the RNA polymerase II transcription complex during initiation,” EMBO J., vol. 16, no. 24, pp. 7468–7480, Dec. 1997.

[115] F. X. Chen, E. R. Smith, and A. Shilatifard, “Born to run: control of transcription elongation by RNA polymerase II,” Nat. Rev. Mol. Cell Biol., vol. 19, no. 7, pp. 464–478, Jul. 2018.

[116] R. J. Sims, R. Belotserkovskaya, and D. Reinberg, “Elongation by RNA polymerase II: the short and long of it.,” Genes Dev., vol. 18, no. 20, pp. 2437–68, Oct. 2004.

[117] Z. Ni et al., “P-TEFb Is Critical for the Maturation of RNA Polymerase II into Productive Elongation In Vivo,” Mol. Cell. Biol., vol. 28, no. 3, pp. 1161–1170, Feb. 2008.

[118] J. Peng, N. F. Marshall, and D. H. Price, “Identification of a Cyclin Subunit Required for the Function of Drosophila P-TEFb,” J. Biol. Chem., vol. 273, no. 22, pp. 13855–13860, May 1998.

[119] I. Jonkers, H. Kwak, and J. T. Lis, “Genome-wide dynamics of Pol II elongation and its interplay with promoter proximal pausing, chromatin, and exons,” Elife, vol. 3, p. e02407, Apr. 2014.

[120] S. Egloff and S. Murphy, “Cracking the RNA polymerase II CTD code,” Trends Genet., vol. 24, no. 6, pp. 280–288, Jun. 2008.

[121] S. Egloff, M. Dienstbier, and S. Murphy, “Updating the RNA polymerase CTD code: adding gene- specific layers.,” Trends Genet., vol. 28, no. 7, pp. 333–41, Jul. 2012.

[122] A. Krumm, L. B. Hickey, and M. Groudine, “Promoter-proximal pausing of RNA polymerase II defines a general rate-limiting step after transcription initiation.,” Genes Dev., vol. 9, no. 5, pp. 559–572, Mar. 1995.

[123] K. Yankulov, J. Blau, T. Purton, S. Roberts, and D. L. Bentley, “Transcriptional elongation by RNA polymerase II is stimulated by transactivators,” Cell, vol. 77, no. 5, pp. 749–759, Jun. 1994.

[124] J. N. Kuehner, E. L. Pearson, and C. Moore, “Unravelling the means to an end: RNA polymerase II transcription termination,” Nat. Rev. Mol. Cell Biol., vol. 12, no. 5, pp. 283–294, May 2011.

[125] J. Logan, E. Falck-Pedersen, J. E. Darnell, and T. Shenk, “A poly(A) addition site and a downstream termination region are required for efficient cessation of transcription by RNA polymerase II in the mouse beta maj-globin gene.,” Proc. Natl. Acad. Sci., vol. 84, no. 23, pp. 8306–8310, Dec. 1987.

[126] E. Whitelaw and N. Proudfoot, “Alpha-thalassaemia caused by a poly(A) site mutation reveals

198

that transcriptional termination is linked to 3’ end processing in the human alpha 2 globin gene.,” EMBO J., vol. 5, no. 11, pp. 2915–22, Nov. 1986.

[127] S. Millevoi and S. Vagner, “Molecular mechanisms of eukaryotic pre-mRNA 3′ end processing regulation,” Nucleic Acids Res., vol. 38, no. 9, pp. 2757–2774, May 2010.

[128] C. R. Mandel, Y. Bai, and L. Tong, “Protein factors in pre-mRNA 3′-end processing,” Cell. Mol. Life Sci., vol. 65, no. 7–8, pp. 1099–1122, Apr. 2008.

[129] D. Bentley, “The mRNA assembly line: transcription and processing machines in the same factory,” Curr. Opin. Cell Biol., vol. 14, no. 3, pp. 336–342, Jun. 2002.

[130] S. Shuman, “Origins of mRNA identity: capping enzymes bind to the phosphorylated C-terminal domain of RNA polymerase II.,” Proc. Natl. Acad. Sci. U. S. A., vol. 94, no. 24, pp. 12758–60, Nov. 1997.

[131] S. McCracken et al., “5’-Capping enzymes are targeted to pre-mRNA by binding to the phosphorylated carboxy-terminal domain of RNA polymerase II.,” Genes Dev., vol. 11, no. 24, pp. 3306–18, Dec. 1997.

[132] I. Wetterberg, J. Zhao, S. Masich, L. Wieslander, and U. Skoglund, “In situ transcription and splicing in the Balbiani ring 3 gene,” EMBO J., vol. 20, no. 10, pp. 2564–2574, May 2001.

[133] P. J. Lopez and B. Séraphin, “Uncoupling yeast intron recognition from transcription with recursive splicing,” EMBO Rep., vol. 1, no. 4, pp. 334–339, Oct. 2000.

[134] I. Wetterberg, G. Baurén, and L. Wieslander, “The intranuclear site of excision of each intron in Balbiani ring 3 pre-mRNA is influenced by the time remaining to transcription termination and different excision efficiencies for the various introns.,” RNA, vol. 2, no. 7, pp. 641–51, Jul. 1996.

[135] Y. W. Fong and Q. Zhou, “Stimulatory effect of splicing factors on transcriptional elongation,” Nature, vol. 414, no. 6866, pp. 929–933, Dec. 2001.

[136] R. W. Honess and B. Roizman, “Regulation of herpesvirus macromolecular synthesis. I. Cascade regulation of the synthesis of three groups of viral proteins.,” J. Virol., vol. 14, no. 1, pp. 8–19, Jul. 1974.

[137] P. O’Hare, “The virion transactivator of herpes simplex virus,” Semin. Virol., vol. 4, no. 3, pp. 145–155, Jun. 1993.

[138] W. Batterson and B. Roizman, “Characterization of the herpes simplex virion-associated factor responsible for the induction of alpha genes.,” J. Virol., vol. 46, no. 2, pp. 371–7, May 1983.

[139] L. E. Post, S. Mackem, and B. Roizman, “Regulation of α genes of herpes simplex virus: Expression of chimeric genes produced by fusion of thymidine kinase with α gene promoters,” Cell, vol. 24, no. 2, pp. 555–565, May 1981.

[140] M. E. M. Campbell, J. W. Palfreyman, and C. M. Preston, “Identification of herpes simplex virus DNA sequences which encode a trans-acting polypeptide responsible for stimulation of immediate early transcription,” J. Mol. Biol., vol. 180, no. 1, pp. 1–19, Nov. 1984.

[141] T. Gerster and R. G. Roeder, “A herpesvirus trans-activating protein interacts with transcription factor OTF-1 and other cellular proteins.,” Proc. Natl. Acad. Sci. U. S. A., vol. 85, no. 17, pp.

199

6347–51, Sep. 1988.

[142] S. Stern, M. Tanaka, and W. Herr, “The Oct-1 homoeodomain directs formation of a multiprotein-DNA complex with the HSV transactivator VP16,” Nature, vol. 341, no. 6243, pp. 624–630, Oct. 1989.

[143] P. Mavromara-Nazos and B. Roizman, “Activation of herpes simplex virus 1 γ2 genes by viral DNA replication,” Virology, vol. 161, no. 2, pp. 593–598, Dec. 1987.

[144] C. M. Preston, “Repression of viral transcription during herpes simplex virus latency,” J. Gen. Virol., vol. 81, no. 1, pp. 1–19, Jan. 2000.

[145] S. Efstathiou and C. M. Preston, “Towards an understanding of the molecular basis of herpes simplex virus latency,” Virus Res., vol. 111, no. 2, pp. 108–119, Aug. 2005.

[146] T. M. Kristie and B. Roizman, “Differentiation and DNA contact points of host proteins binding at the cis site for virion-mediated induction of alpha genes of herpes simplex virus 1.,” J. Virol., vol. 62, no. 4, pp. 1145–57, Apr. 1988.

[147] G. Kolb and T. M. Kristie, “Association of the cellular coactivator HCF-1 with the Golgi apparatus in sensory neurons.,” J. Virol., vol. 82, no. 19, pp. 9555–63, Oct. 2008.

[148] N. D. Lakin et al., “Down regulation of the octamer binding protein Oct-1 during growth arrest and differentiation of a neuronal cell line,” Mol. Brain Res., vol. 28, no. 1, pp. 47–54, Jan. 1995.

[149] E. K. Wagner and D. C. Bloom, “Experimental investigation of herpes simplex virus latency.,” Clin. Microbiol. Rev., vol. 10, no. 3, pp. 419–443, Jul. 1997.

[150] A. Gupta, J. J. Gartner, P. Sethupathy, A. G. Hatzigeorgiou, and N. W. Fraser, “Anti-apoptotic function of a microRNA encoded by the HSV-1 latency-associated transcript,” Nature, vol. 442, no. 7098, pp. 82–85, Jul. 2006.

[151] N. Mador, D. Goldenberg, O. Cohen, A. Panet, and I. Steiner, “Herpes simplex virus type 1 latency-associated transcripts suppress viral replication and reduce immediate-early gene mRNA levels in a neuronal cell line.,” J. Virol., vol. 72, no. 6, pp. 5067–75, Jun. 1998.

[152] M. P. Nicoll et al., “The HSV-1 Latency-Associated Transcript Functions to Repress Latent Phase Lytic Gene Expression and Suppress Virus Reactivation from Latently Infected Neurons,” PLOS Pathog., vol. 12, no. 4, p. e1005539, Apr. 2016.

[153] D. L. Rock and N. W. Fraser, “Detection of HSV-1 genome in central nervous system of latently infected mice,” Nature, vol. 302, no. 5908, pp. 523–525, Apr. 1983.

[154] S. Efstathiou, A. C. Minson, H. J. Field, J. R. Anderson, and P. Wildy, “Detection of herpes simplex virus-specific DNA sequences in latently infected mice and in humans.,” J. Virol., vol. 57, no. 2, pp. 446–455, Feb. 1986.

[155] S. L. Deshmane and N. W. Fraser, “During latency, herpes simplex virus type 1 DNA is associated with nucleosomes in a chromatin structure.,” J. Virol., vol. 63, no. 2, pp. 943–947, Feb. 1989.

[156] A. R. Cliffe et al., “Neuronal Stress Pathway Mediating a Histone Methyl/Phospho Switch Is Required for Herpes Simplex Virus Reactivation,” Cell Host Microbe, vol. 18, no. 6, pp. 649–658, Dec. 2015.

200

[157] J. Y. Kim, A. Mandarino, M. V. Chao, I. Mohr, and A. C. Wilson, “Transient Reversal of Episome Silencing Precedes VP16-Dependent Transcription during Reactivation of Latent HSV-1 in Neurons,” PLoS Pathog., vol. 8, no. 2, p. e1002540, Feb. 2012.

[158] J. M. Hill et al., “Inhibition of LSD1 reduces herpesvirus infection, shedding, and recurrence by promoting epigenetic suppression of viral genomes,” Sci. Transl. Med., vol. 6, no. 265, pp. 265ra169-265ra169, Dec. 2014.

[159] R. Skaliter and I. R. Lehman, “Rolling circle DNA replication in vitro by a complex of herpes simplex virus type 1-encoded enzymes,” Proc Natl Acad Sci U S A, vol. 91, no. 22, pp. 10665– 10669, Oct. 1994.

[160] S. K. Weller and D. M. Coen, “Herpes simplex viruses: mechanisms of DNA replication,” Cold Spring Harb Perspect Biol, vol. 4, no. 9, p. a013011, Sep. 2012.

[161] D. Wilkinson and S. Weller, “The Role of DNA Recombination in Herpes Simplex Virus DNA Replication,” IUBMB Life (International Union Biochem. Mol. Biol. Life), vol. 55, no. 8, pp. 451– 458, Aug. 2003.

[162] H. J. Zuccola, D. J. Filman, D. M. Coen, and J. M. Hogle, “The Crystal Structure of an Unusual Processivity Factor, Herpes Simplex Virus UL42, Bound to the C Terminus of Its Cognate Polymerase,” Mol. Cell, vol. 5, no. 2, pp. 267–278, Feb. 2000.

[163] A. D. Burch and S. K. Weller, “Herpes Simplex Virus Type 1 DNA Polymerase Requires the Mammalian Chaperone Hsp90 for Proper Localization to the Nucleus,” J. Virol., vol. 79, no. 16, pp. 10740–10749, Aug. 2005.

[164] J. D. Tatman, V. G. Preston, P. Nicholson, R. M. Elliott, and F. J. Rixon, “Assembly of Herpes Simplex virus Type 1 Capsids Using a Panel of Recombinant Baculoviruses,” J. Gen. Virol., vol. 75, no. 5, pp. 1101–1113, May 1994.

[165] D. R. Thomsen, W. W. Newcomb, J. C. Brown, and F. L. Homa, “Assembly of the herpes simplex virus capsid: requirement for the carboxyl-terminal twenty-five amino acids of the proteins encoded by the UL26 and UL26.5 genes.,” J. Virol., vol. 69, no. 6, pp. 3690–3703, Jun. 1995.

[166] W. W. Newcomb et al., “Assembly of the Herpes Simplex Virus Capsid: Characterization of Intermediates Observed During Cell-free Capsid Formation,” J. Mol. Biol., vol. 263, no. 3, pp. 432–446, Nov. 1996.

[167] T. C. Mettenleiter, “Intriguing interplay between viral proteins during herpesvirus assembly or: The herpesvirus assembly puzzle,” Vet. Microbiol., vol. 113, no. 3–4, pp. 163–169, Mar. 2006.

[168] A. E. Reynolds, B. J. Ryckman, J. D. Baines, Y. Zhou, L. Liang, and R. J. Roller, “U(L)31 and U(L)34 proteins of herpes simplex virus type 1 form a complex that accumulates at the nuclear rim and is required for envelopment of nucleocapsids.,” J. Virol., vol. 75, no. 18, pp. 8803–17, Sep. 2001.

[169] A. E. Reynolds, E. G. Wills, R. J. Roller, B. J. Ryckman, and J. D. Baines, “Ultrastructural localization of the herpes simplex virus type 1 UL31, UL34, and US3 proteins suggests specific roles in primary envelopment and egress of nucleocapsids.,” J. Virol., vol. 76, no. 17, pp. 8939– 52, Sep. 2002.

[170] D. C. Johnson and J. D. Baines, “Herpesviruses remodel host membranes for virus egress,” Nat.

201

Rev. Microbiol., vol. 9, no. 5, pp. 382–394, May 2011.

[171] J. N. Skepper, A. Whiteley, H. Browne, and A. Minson, “Herpes Simplex Virus Nucleocapsids Mature to Progeny Virions by an Envelopment Deenvelopment Reenvelopment Pathway,” J. Virol., vol. 75, no. 12, pp. 5697–5702, Jun. 2001.

[172] T. C. Mettenleiter, “Herpesvirus assembly and egress.,” J. Virol., vol. 76, no. 4, pp. 1537–47, Feb. 2002.

[173] M. Sandbaumhüter et al., “Cytosolic herpes simplex virus capsids not only require binding inner tegument protein pUL36 but also pUL37 for active transport prior to secondary envelopment,” Cell. Microbiol., vol. 15, no. 2, pp. 248–269, Feb. 2013.

[174] D. Pasdeloup, M. McElwee, F. Beilstein, M. Labetoulle, and F. J. Rixon, “Herpesvirus Tegument Protein pUL37 Interacts with Dystonin/BPAG1 To Promote Capsid Transport on Microtubules during Egress,” J. Virol., vol. 87, no. 5, pp. 2857–2867, Mar. 2013.

[175] D. C. Johnson and P. G. Spear, “Monensin inhibits the processing of herpes simplex virus glycoproteins, their transport to the cell surface, and the egress of virions from infected cells.,” J. Virol., vol. 43, no. 3, pp. 1102–1112, Sep. 1982.

[176] F. J. Ibáñez et al., “Experimental Dissection of the Lytic Replication Cycles of Herpes Simplex Viruses in vitro,” Front. Microbiol., vol. 9, p. 2406, Oct. 2018.

[177] P. J. Desai, “A null mutation in the UL36 gene of herpes simplex virus type 1 results in accumulation of unenveloped DNA-filled capsids in the cytoplasm of infected cells.,” J. Virol., vol. 74, no. 24, pp. 11608–18, Dec. 2000.

[178] W. Fuchs, B. G. Klupp, H. Granzow, and T. C. Mettenleiter, “Essential Function of the Pseudorabies Virus UL36 Gene Product Is Independent of Its Interaction with the UL37 Protein,” J. Virol., vol. 78, no. 21, pp. 11879–11889, Nov. 2004.

[179] J. I.-H. Lee, G. W. G. Luxton, and G. A. Smith, “Identification of an essential domain in the herpesvirus VP1/2 tegument protein: the carboxy terminus directs incorporation into capsid assemblons,” J Virol, vol. 80, no. 24, pp. 12086–12094, Dec. 2006.

[180] G. W. G. Luxton, J. I.-H. Lee, S. Haverlock-Moyns, J. M. Schober, and G. A. Smith, “The pseudorabies virus VP1/2 tegument protein is required for intracellular capsid transport.,” J. Virol., vol. 80, no. 1, pp. 201–9, Jan. 2006.

[181] S. Böttcher, H. Granzow, C. Maresch, B. Möhl, B. G. Klupp, and T. C. Mettenleiter, “Identification of functional domains within the essential large tegument protein pUL36 of pseudorabies virus.,” J. Virol., vol. 81, no. 24, pp. 13403–11, Dec. 2007.

[182] D. M. Knipe, W. Batterson, C. Nosal, B. Roizman, and A. Buchan, “Molecular genetics of herpes simplex virus. VI. Characterization of a temperature-sensitive mutant defective in the expression of all early viral gene products.,” J. Virol., vol. 38, no. 2, pp. 539–47, May 1981.

[183] F. Abaitua, T. Daikoku, C. M. Crump, M. Bolstad, and P. O’Hare, “A single mutation responsible for temperature-sensitive entry and assembly defects in the VP1-2 protein of herpes simplex virus,” J Virol, vol. 85, no. 5, pp. 2024–2036, 2011.

[184] A. K. Sheaffer et al., “Herpes Simplex Virus DNA Cleavage and Packaging Proteins Associate

202

with the Procapsid prior to Its Maturation,” J. Virol., vol. 75, no. 2, pp. 687–698, Jan. 2001.

[185] A. R. McNab et al., “The product of the herpes simplex virus type 1 UL25 gene is required for encapsidation but not for cleavage of replicated viral DNA,” J Virol, vol. 72, no. 2, pp. 1060– 1070, Feb. 1998.

[186] N. D. Stow, “Packaging of genomic and amplicon DNA by the herpes simplex virus type 1 UL25- null mutant KUL25NS,” J Virol, vol. 75, no. 22, pp. 10755–10765, Nov. 2001.

[187] B. L. Trus et al., “Allosteric Signaling and a Nuclear Exit Strategy: Binding of UL25/UL17 Heterodimers to DNA-Filled HSV-1 Capsids,” Mol. Cell, vol. 26, no. 4, pp. 479–489, May 2007.

[188] K. E. Coller, J. I.-H. Lee, A. Ueda, and G. A. Smith, “The Capsid and Tegument of the Alphaherpesviruses Are Linked by an Interaction between the UL25 and VP1/2 Proteins,” J. Virol., vol. 81, no. 21, pp. 11790–11797, Nov. 2007.

[189] V. G. Preston, J. Murray, C. M. Preston, I. M. McDougall, and N. D. Stow, “The UL25 gene product of herpes simplex virus type 1 is involved in uncoating of the viral genome.,” J. Virol., vol. 82, no. 13, pp. 6654–66, Jul. 2008.

[190] W. W. Newcomb et al., “The UL6 gene product forms the portal for entry of DNA into the herpes simplex virus capsid.,” J. Virol., vol. 75, no. 22, pp. 10923–32, Nov. 2001.

[191] A. H. Patel and J. B. MacLean, “The product of the UL6 gene of herpes simplex virus type 1 is associated with virus capsids.,” Virology, vol. 206, no. 1, pp. 465–78, Jan. 1995.

[192] Homa and Brown, “Capsid assembly and DNA packaging in herpes simplex virus.,” Rev. Med. Virol., vol. 7, no. 2, pp. 107–122, Jul. 1997.

[193] W. W. Newcomb, F. P. Booy, and J. C. Brown, “Uncoating the herpes simplex virus genome,” J Mol Biol, vol. 370, no. 4, pp. 633–642, 2007.

[194] B. D. Strahl and C. D. Allis, “The language of covalent histone modifications,” Nature, vol. 403, no. 6765, pp. 41–45, Jan. 2000.

[195] T. Jenuwein and C. D. Allis, “Translating the histone code.,” Science, vol. 293, no. 5532, pp. 1074–80, Aug. 2001.

[196] R. N. Dutnall, “Cracking the Histone Code One, Two, Three Methyls, You’re Out!,” Mol. Cell, vol. 12, no. 1, pp. 3–4, Jul. 2003.

[197] B. M. Turner, “Histone acetylation and an epigenetic code,” BioEssays, vol. 22, no. 9, pp. 836– 845, Aug. 2000.

[198] S. L. Berger, “Histone modifications in transcriptional regulation,” Curr. Opin. Genet. Dev., vol. 12, no. 2, pp. 142–148, Apr. 2002.

[199] M. S. Cosgrove and C. Wolberger, “How does the histone code work?,” Biochem. Cell Biol., vol. 83, no. 4, pp. 468–476, Aug. 2005.

[200] M. S. Cosgrove, J. D. Boeke, and C. Wolberger, “Regulated nucleosome mobility and the histone code,” Nat. Struct. Mol. Biol., vol. 11, no. 11, pp. 1037–1043, Nov. 2004.

203

[201] M. W. Ferenczy and N. A. DeLuca, “Epigenetic Modulation of Gene Expression from Quiescent Herpes Simplex Virus Genomes,” J. Virol., vol. 83, no. 17, pp. 8514–8524, Sep. 2009.

[202] D. M. Knipe and A. Cliffe, “Chromatin control of herpes simplex virus lytic and latent infection,” Nat. Rev. Microbiol., vol. 6, no. 3, pp. 211–221, Mar. 2008.

[203] M. W. Ferenczy and N. A. DeLuca, “Reversal of heterochromatic silencing of quiescent herpes simplex virus type 1 by ICP0.,” J. Virol., vol. 85, no. 7, pp. 3424–35, Apr. 2011.

[204] J. J. Lacasse and L. M. Schang, “Herpes simplex virus 1 DNA is in unstable nucleosomes throughout the lytic infection cycle, and the instability of the nucleosomes is independent of DNA replication.,” J. Virol., vol. 86, no. 20, pp. 11287–300, Oct. 2012.

[205] J. J. Lacasse and L. M. Schang, “During Lytic Infections, Herpes Simplex Virus Type 1 DNA Is in Complexes with the Properties of Unstable Nucleosomes,” J. Virol., vol. 84, no. 4, pp. 1920– 1933, Feb. 2010.

[206] S. S. Leinbach and W. C. Summers, “The Structure of Herpes Simplex Virus Type 1 DNA as Probed by Micrococcal Nuclease Digestion,” J. Gen. Virol., vol. 51, no. 1, pp. 45–59, Nov. 1980.

[207] A. F. Lentine and S. L. Bachenheimer, “Intracellular organization of herpes simplex virus type 1 DNA assayed by staphylococcal nuclease sensitivity,” Virus Res., vol. 16, no. 3, pp. 275–292, Jul. 1990.

[208] J. R. Kent, P.-Y. Zeng, D. Atanasiu, J. Gardner, N. W. Fraser, and S. L. Berger, “During Lytic Infection Herpes Simplex Virus Type 1 Is Associated with Histones Bearing Modifications That Correlate with Active Transcription,” J. Virol., vol. 78, no. 18, pp. 10178–10186, Sep. 2004.

[209] M. I. Muggeridge and N. W. Fraser, “Chromosomal Organization of the Herpes Simplex Virus Genome during Acute Infection of the Mouse Central Nervous System,” J Virol, vol. 59, no. 3, pp. 764–767, Sep. 1986.

[210] M. E. Mouttet, D. Guétard, and J. M. Béchet, “Random cleavage of intranuclear herpes simplex virus DNA by micrococcal nuclease.,” FEBS Lett., vol. 100, no. 1, pp. 107–9, Apr. 1979.

[211] F. J. Herrera and S. J. Triezenberg, “VP16-Dependent Association of Chromatin-Modifying Coactivators and Underrepresentation of Histones at Immediate-Early Gene Promoters during Herpes Simplex Virus Infection,” J. Virol., vol. 78, no. 18, pp. 9689–9696, Sep. 2004.

[212] J. Oh and N. W. Fraser, “Temporal Association of the Herpes Simplex Virus Genome with Histone Proteins during a Lytic Infection,” J. Virol., vol. 82, no. 7, pp. 3530–3537, Apr. 2008.

[213] J. Huang et al., “Trimethylation of Histone H3 Lysine 4 by Set1 in the Lytic Infection of Human Herpes Simplex Virus 1,” J. Virol., vol. 80, no. 12, pp. 5740–5746, Jun. 2006.

[214] B. J. Placek et al., “The histone variant H3.3 regulates gene expression during lytic infection with herpes simplex virus type 1.,” J. Virol., vol. 83, no. 3, pp. 1416–21, Feb. 2009.

[215] K. Conn, L. Schang, K. L. Conn, and L. M. Schang, “Chromatin Dynamics during Lytic Infection with Herpes Simplex Virus 1,” Viruses, vol. 5, no. 7, pp. 1758–1786, Jul. 2013.

[216] K. L. Conn, M. J. Hendzel, and L. M. Schang, “Linker histones are mobilized during infection with herpes simplex virus type 1.,” J. Virol., vol. 82, no. 17, pp. 8629–46, Sep. 2008.

204

[217] C. Boutell and R. D. Everett, “Regulation of alphaherpesvirus infections by the ICP0 family of proteins,” J. Gen. Virol., vol. 94, no. Pt_3, pp. 465–481, Mar. 2013.

[218] K. L. Conn, M. J. Hendzel, and L. M. Schang, “Core histones H2B and H4 are mobilized during infection with herpes simplex virus 1.,” J. Virol., vol. 85, no. 24, pp. 13234–52, Dec. 2011.

[219] A. S. Belmont, S. Dietzel, A. C. Nye, Y. G. Strukov, and T. Tumbar, “Large-scale chromatin structure and function,” Curr. Opin. Cell Biol., vol. 11, no. 3, pp. 307–311, Jun. 1999.

[220] S. Memedula and A. S. Belmont, “Sequential Recruitment of HAT and SWI/SNF Components to Condensed Chromatin by VP16,” Curr. Biol., vol. 13, no. 3, pp. 241–246, Feb. 2003.

[221] T. Tumbar, G. Sudlow, and A. S. Belmont, “Large-scale chromatin unfolding and remodeling induced by VP16 acidic activation domain.,” J. Cell Biol., vol. 145, no. 7, pp. 1341–54, Jun. 1999.

[222] H. Gu and B. Roizman, “Herpes simplex virus-infected cell protein 0 blocks the silencing of viral DNA by dissociating histone deacetylases from the CoREST REST complex,” Proc. Natl. Acad. Sci., vol. 104, no. 43, pp. 17134–17139, Oct. 2007.

[223] G. T. Melroe, L. Silva, P. A. Schaffer, and D. M. Knipe, “Recruitment of activated IRF-3 and CBP/p300 to herpes simplex virus ICP0 nuclear foci: Potential role in blocking IFN-β induction,” Virology, vol. 360, no. 2, pp. 305–321, Apr. 2007.

[224] M. Kalamvoki and B. Roizman, “Circadian CLOCK histone acetyl transferase localizes at ND10 nuclear bodies and enables herpes simplex virus gene expression,” Proc. Natl. Acad. Sci., vol. 107, no. 41, pp. 17721–17726, Oct. 2010.

[225] P. Lomonte, J. Thomas, P. Texier, C. Caron, S. Khochbin, and A. L. Epstein, “Functional Interaction between Class II Histone Deacetylases and ICP0 of Herpes Simplex Virus Type 1,” J. Virol., vol. 78, no. 13, pp. 6744–6757, Jul. 2004.

[226] D. Qu et al., “5-Ethynyl-2’-deoxycytidine as a new agent for DNA labeling: detection of proliferating cells,” Anal Biochem, vol. 417, no. 1, pp. 112–121, Oct. 2011.

[227] H. Gu, Y. Liang, G. Mandel, and B. Roizman, “Components of the REST/CoREST/histone deacetylase repressor complex are disrupted, modified, and translocated in HSV-1-infected cells,” Proc. Natl. Acad. Sci., vol. 102, no. 21, pp. 7571–7576, May 2005.

[228] T. J. Taylor and D. M. Knipe, “Proteomics of herpes simplex virus replication compartments: association of cellular DNA replication, repair, recombination, and chromatin remodeling proteins with ICP8.,” J. Virol., vol. 78, no. 11, pp. 5856–66, Jun. 2004.

[229] F. Mattiroli et al., “RNF168 Ubiquitinates K13-15 on H2A/H2AX to Drive DNA Damage Signaling,” Cell, vol. 150, no. 6, pp. 1182–1195, Sep. 2012.

[230] C. E. Lilley et al., “A viral E3 ligase targets RNF8 and RNF168 to control histone ubiquitination and DNA damage responses,” EMBO J., vol. 29, no. 5, pp. 943–955, Mar. 2010.

[231] M. J. Carrozza and N. DeLuca, “The High Mobility Group Protein 1 Is a Coactivator of Herpes Simplex Virus ICP4 In Vitro,” J. Virol., vol. 72, no. 8, pp. 6752–6757, Aug. 1998.

[232] C. A. Panagiotidis and S. J. Silverstein, “The Host-Cell Architectural Protein HMG I(Y) Modulates Binding of Herpes Simplex Virus Type 1 ICP4 to Its Cognate Promoter,” Virology, vol. 256, no.

205

1, pp. 64–74, Mar. 1999.

[233] C. A. Smibert and J. R. Smiley, “Differential regulation of endogenous and transduced beta- globin genes during infection of erythroid cells with a herpes simplex virus type 1 recombinant.,” J. Virol., vol. 64, no. 8, pp. 3882–94, Aug. 1990.

[234] J. R. Smiley, D. C. Johnson, L. I. Pizer, and R. D. Everett, “The ICP4 binding sites in the herpes simplex virus type 1 glycoprotein D (gD) promoter are not essential for efficient gD transcription during virus infection.,” J. Virol., vol. 66, no. 2, pp. 623–31, Feb. 1992.

[235] P. J. Godowski and D. M. Knipe, “Transcriptional control of herpesvirus gene expression: gene functions required for positive and negative regulation.,” Proc. Natl. Acad. Sci. U. S. A., vol. 83, no. 2, pp. 256–60, Jan. 1986.

[236] R. J. Watson and J. B. Clements, “A herpes simplex virus type 1 function continuously required for early and late virus RNA synthesis,” Nature, vol. 285, no. 5763, pp. 329–330, May 1980.

[237] N. A. DeLuca and P. A. Schaffer, “Activation of immediate-early, early, and late promoters by temperature-sensitive and wild-type forms of herpes simplex virus type 1 protein ICP4.,” Mol. Cell. Biol., vol. 5, no. 8, pp. 1997–2008, Aug. 1985.

[238] E. K. Lium, C. A. Panagiotidis, X. Wen, and S. Silverstein, “Repression of the alpha0 gene by ICP4 during a productive herpes simplex virus infection.,” J. Virol., vol. 70, no. 6, pp. 3488–96, Jun. 1996.

[239] M. S. Roberts, A. Boundy, P. O’Hare, M. C. Pizzorno, D. M. Ciufo, and G. S. Hayward, “Direct correlation between a negative autoregulatory response element at the cap site of the herpes simplex virus type 1 IE175 (alpha 4) promoter and a specific binding site for the IE175 (ICP4) protein.,” J. Virol., vol. 62, no. 11, pp. 4307–20, Nov. 1988.

[240] S. W. Faber and K. W. Wilcox, “Association of the herpes simplex virus regulatory protein ICP4 with specific nucleotide sequences in DNA.,” Nucleic Acids Res., vol. 14, no. 15, pp. 6067–83, Aug. 1986.

[241] M. T. Muller, “Binding of the herpes simplex virus immediate-early gene product ICP4 to its own transcription start site.,” J. Virol., vol. 61, no. 3, pp. 858–865, Mar. 1987.

[242] T. M. Kristie and B. Roizman, “Alpha 4, the major regulatory protein of herpes simplex virus type 1, is stably and specifically associated with promoter-regulatory domains of alpha genes and of selected other viral genes,” Proc. Natl. Acad. Sci., vol. 83, no. 10, pp. 3218–3222, May 1986.

[243] K. E. Allen and R. D. Everett, “Mutations which alter the DNA binding properties of the herpes simplex virus type 1 transactivating protein Vmw175 also affect its ability to support virus replication.,” J. Gen. Virol., vol. 78, no. 11, pp. 2913–2922, Nov. 1997.

[244] T. Paterson and R. D. Everett, “The regions of the herpes simplex virus type 1 immediate early protein Vmwl75 required for site specific DNA binding closely correspond to those involved in transcriptional regulation,” Nucleic Acids Res., vol. 16, no. 23, pp. 11005–11025, Dec. 1988.

[245] T. Paterson and R. D. Everett, “Mutational dissection of the HSV-1 immediate-early protein Vmw175 involved in transcriptional transactivation and repression,” Virology, vol. 166, no. 1, pp. 186–196, Sep. 1988.

206

[246] N. A. DeLuca, A. M. McCarthy, and P. A. Schaffer, “Isolation and characterization of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate-early regulatory protein ICP4.,” J. Virol., vol. 56, no. 2, pp. 558–570, Nov. 1985.

[247] R. A. Dixon and P. A. Schaffer, “Fine-structure mapping and functional analysis of temperature- sensitive mutants in the gene encoding the herpes simplex virus type 1 immediate early protein VP175.,” J. Virol., vol. 36, no. 1, pp. 189–203, Oct. 1980.

[248] C. M. Preston, “Control of herpes simplex virus type 1 mRNA synthesis in cells infected with wild-type virus or the temperature-sensitive mutant tsK.,” J. Virol., vol. 29, no. 1, pp. 275–84, Jan. 1979.

[249] A. A. Shepard, A. N. Imbalzano, and N. A. DeLuca, “Separation of primary structural components conferring autoregulation, transactivation, and DNA-binding properties to the herpes simplex virus transcriptional regulatory protein ICP4.,” J. Virol., vol. 63, no. 9, pp. 3714– 28, Sep. 1989.

[250] M. J. Carrozza and N. A. DeLuca, “Interaction of the viral activator protein ICP4 with TFIID through TAF250.,” Mol. Cell. Biol., vol. 16, no. 6, pp. 3085–93, Jun. 1996.

[251] C. A. Smith, P. Bates, R. Rivera-Gonzalez, B. Gu, and N. A. DeLuca, “ICP4, the major transcriptional regulatory protein of herpes simplex virus type 1, forms a tripartite complex with TATA-binding protein and TFIIB.,” J. Virol., vol. 67, no. 8, pp. 4676–87, Aug. 1993.

[252] B. Gu and N. DeLuca, “Requirements for activation of the herpes simplex virus glycoprotein C promoter in vitro by the viral regulatory protein ICP4.,” J. Virol., vol. 68, no. 12, pp. 7953–65, Dec. 1994.

[253] B. Grondin and N. DeLuca, “Herpes simplex virus type 1 ICP4 promotes transcription preinitiation complex formation by enhancing the binding of TFIID to DNA.,” J. Virol., vol. 74, no. 24, pp. 11504–10, Dec. 2000.

[254] N. D. Stow and E. C. Stow, “Isolation and characterization of a herpes simplex virus type 1 mutant containing a deletion within the gene encoding the immediate early polypeptide Vmw110,” J Gen Virol, vol. 67 ( Pt 12, no. 12, pp. 2571–2585, Dec. 1986.

[255] W. Cai and P. A. Schaffer, “Herpes simplex virus type 1 ICP0 regulates expression of immediate- early, early, and late genes in productively infected cells.,” J. Virol., vol. 66, no. 5, p. 2904, 1992.

[256] R. Jordan and P. A. Schaffer, “Activation of gene expression by herpes simplex virus type 1 ICP0 occurs at the level of mRNA synthesis.,” J. Virol., vol. 71, no. 9, pp. 6850–62, Sep. 1997.

[257] M. Kalamvoki and B. Roizman, “Role of Herpes Simplex Virus ICP0 in the Transactivation of Genes Introduced by Infection or Transfection: a Reappraisal,” J. Virol., vol. 84, no. 9, pp. 4222– 4228, May 2010.

[258] G. Nabel, S. Rice, D. Knipe, and D. Baltimore, “Alternative mechanisms for activation of human immunodeficiency virus enhancer in T cells,” Science (80-. )., vol. 239, no. 4845, pp. 1299–1302, 2006.

[259] F. Yao and P. A. Schaffer, “Physical interaction between the herpes simplex virus type 1 immediate-early regulatory proteins ICP0 and ICP4.,” J. Virol., vol. 68, no. 12, pp. 8158–68, Dec. 1994.

207

[260] P. O’Hare, C. R. Goding, and A. Haigh, “Direct combinatorial interaction between a herpes simplex virus regulatory protein and a cellular octamer-binding factor mediates specific induction of virus immediate-early gene expression.,” EMBO J., vol. 7, no. 13, pp. 4231–4238, Dec. 1988.

[261] Z. Zhu, W. Cai, and P. A. Schaffer, “Cooperativity among herpes simplex virus type 1 immediate- early regulatory proteins: ICP4 and ICP27 affect the intracellular localization of ICP0.,” J. Virol., vol. 68, no. 5, pp. 3027–40, May 1994.

[262] W. R. Sacks and P. A. Schaffer, “Deletion mutants in the gene encoding the herpes simplex virus type 1 immediate-early protein ICP0 exhibit impaired growth in cell culture.,” J. Virol., vol. 61, no. 3, pp. 829–39, Mar. 1987.

[263] R. D. Everett, “Construction and Characterization of Herpes Simplex Virus Type 1 Mutants with Defined Lesions in Immediate Early Gene 1,” J. Gen. Virol., vol. 70, no. 5, pp. 1185–1202, May 1989.

[264] Y. Kawaguchi, C. Van Sant, and B. Roizman, “Herpes simplex virus 1 alpha regulatory protein ICP0 interacts with and stabilizes the cell cycle regulator cyclin D3.,” J. Virol., vol. 71, no. 10, pp. 7328–36, Oct. 1997.

[265] Y. Kawaguchi et al., “Herpes simplex virus 1 alpha regulatory protein ICP0 functionally interacts with cellular transcription factor BMAL1,” Proc. Natl. Acad. Sci., vol. 98, no. 4, pp. 1877–1882, Feb. 2001.

[266] Y. Kawaguchi, R. Bruni, and B. Roizman, “Interaction of herpes simplex virus 1 alpha regulatory protein ICP0 with elongation factor 1delta: ICP0 affects translational machinery.,” J. Virol., vol. 71, no. 2, pp. 1019–24, Feb. 1997.

[267] R. D. Everett, M. Meredith, and A. Orr, “The ability of herpes simplex virus type 1 immediate- early protein Vmw110 to bind to a ubiquitin-specific protease contributes to its roles in the activation of gene expression and stimulation of virus replication.,” J. Virol., vol. 73, no. 1, pp. 417–26, Jan. 1999.

[268] G. G. Maul and R. D. Everett, “The nuclear location of PML, a cellular member of the C3HC4 zinc-binding domain protein family, is rearranged during herpes simplex virus infection by the C3HC4 viral protein ICP0,” J. Gen. Virol., vol. 75, no. 6, pp. 1223–1233, Jun. 1994.

[269] R. D. Everett and G. G. Maul, “HSV-1 IE protein Vmw110 causes redistribution of PML.,” EMBO J., vol. 13, no. 21, pp. 5062–5069, Nov. 1994.

[270] G. G. Maul, H. H. Guldner, and J. G. Spivack, “Modification of discrete nuclear domains induced by herpes simplex virus type 1 immediate early gene 1 product (ICP0),” J. Gen. Virol., vol. 74, no. 12, pp. 2679–2690, Dec. 1993.

[271] R. Everett, P. O’Hare, D. O’Rourke, P. Barlow, and A. Orr, “Point mutations in the herpes simplex virus type 1 Vmw110 RING finger helix affect activation of gene expression, viral growth, and interaction with PML-containing nuclear structures.,” J. Virol., vol. 69, no. 11, pp. 7339–44, Nov. 1995.

[272] R. D. Everett et al., “The Disruption of ND10 during Herpes Simplex Virus Infection Correlates with the Vmw110- and Proteasome-Dependent Loss of Several PML Isoforms,” J. Virol., vol. 72, no. 8, pp. 6581–6591, Aug. 1998.

208

[273] M. K. Chelbi-Alix and H. de Thé, “Herpes virus induced proteasome-dependent degradation of the nuclear bodies-associated PML and Sp100 proteins,” Oncogene, vol. 18, no. 4, pp. 935–941, Jan. 1999.

[274] C. Boutell et al., “A Viral Ubiquitin Ligase Has Substrate Preferential SUMO Targeted Ubiquitin Ligase Activity that Counteracts Intrinsic Antiviral Defence,” PLoS Pathog., vol. 7, no. 9, p. e1002245, Sep. 2011.

[275] K. M. Eidson, W. E. Hobbs, B. J. Manning, P. Carlson, and N. A. DeLuca, “Expression of herpes simplex virus ICP0 inhibits the induction of interferon-stimulated genes by viral infection.,” J. Virol., vol. 76, no. 5, pp. 2180–91, Mar. 2002.

[276] S. P. Lees-Miller, M. C. Long, M. A. Kilvert, V. Lam, S. A. Rice, and C. A. Spencer, “Attenuation of DNA-dependent protein kinase activity and its catalytic subunit by the herpes simplex virus type 1 transactivator ICP0.,” J. Virol., vol. 70, no. 11, pp. 7471–7, Nov. 1996.

[277] J. Parkinson, S. P. Lees-Miller, and R. D. Everett, “Herpes Simplex Virus Type 1 Immediate-Early Protein Vmw110 Induces the Proteasome-Dependent Degradation of the Catalytic Subunit of DNA-Dependent Protein Kinase,” J. Virol., vol. 73, no. 1, pp. 650–657, Jan. 1999.

[278] R. D. Everett, W. C. Earnshaw, J. Findlay, and P. Lomonte, “Specific destruction of kinetochore protein CENP-C and disruption of cell division by herpes simplex virus immediate-early protein Vmw110,” EMBO J., vol. 18, no. 6, pp. 1526–1538, Mar. 1999.

[279] P. Härle, B. Sainz, D. J. . Carr, and W. P. Halford, “The Immediate-Early Protein, ICP0, Is Essential for the Resistance of Herpes Simplex Virus to Interferon-α/β,” Virology, vol. 293, no. 2, pp. 295– 304, Feb. 2002.

[280] M. Perusina Lanfranca, H. Mostafa, D. Davido, M. Perusina Lanfranca, H. H. Mostafa, and D. J. Davido, “HSV-1 ICP0: An E3 Ubiquitin Ligase That Counteracts Host Intrinsic and Innate Immunity,” Cells, vol. 3, no. 2, pp. 438–454, May 2014.

[281] R. M. Sandri-Goldin and G. E. Mendoza, “A herpesvirus regulatory protein appears to act post- transcriptionally by affecting mRNA processing.,” Genes Dev., vol. 6, no. 5, pp. 848–863, May 1992.

[282] K. D. Perkins, J. Gregonis, S. Borge, and S. A. Rice, “Transactivation of a Viral Target Gene by Herpes Simplex Virus ICP27 Is Posttranscriptional and Does Not Require the Endogenous Promoter or Polyadenylation Site,” J. Virol., vol. 77, no. 18, pp. 9872–9884, Sep. 2003.

[283] S. Jean, K. M. LeVan, B. Song, M. Levine, and D. M. Knipe, “Herpes Simplex Virus 1 ICP27 Is Required for Transcription of Two Viral Late (γ2) Genes in Infected Cells,” Virology, vol. 283, no. 2, pp. 273–284, May 2001.

[284] S. A. Rice and D. M. Knipe, “Genetic evidence for two distinct transactivation functions of the herpes simplex virus alpha protein ICP27.,” J. Virol., vol. 64, no. 4, pp. 1704–15, Apr. 1990.

[285] M. A. Hardwicke and R. M. Sandri-Goldin, “The herpes simplex virus regulatory protein ICP27 contributes to the decrease in cellular mRNA levels during infection.,” J. Virol., vol. 68, no. 8, pp. 4797–4810, Aug. 1994.

[286] W. R. Hardy and R. M. Sandri-Goldin, “Herpes simplex virus inhibits host cell splicing, and regulatory protein ICP27 is required for this effect.,” J. Virol., vol. 68, no. 12, pp. 7790–7799,

209

Dec. 1994.

[287] A. Lindberg and J.-P. Kreivi, “Splicing Inhibition at the Level of Spliceosome Assembly in the Presence of Herpes Simplex Virus Protein ICP27,” Virology, vol. 294, no. 1, pp. 189–198, Mar. 2002.

[288] K. S. Sciabica, Q. J. Dai, and R. M. Sandri-Goldin, “ICP27 interacts with SRPK1 to mediate HSV splicing inhibition by altering SR protein phosphorylation,” EMBO J., vol. 22, no. 7, pp. 1608– 1619, Apr. 2003.

[289] J. Q. Dai-Ju, L. Li, L. A. Johnson, and R. M. Sandri-Goldin, “ICP27 interacts with the C-terminal domain of RNA polymerase II and facilitates its recruitment to herpes simplex virus 1 transcription sites, where it undergoes proteasomal degradation during infection.,” J. Virol., vol. 80, no. 7, pp. 3567–81, Apr. 2006.

[290] C. Zhou and D. M. Knipe, “Association of herpes simplex virus type 1 ICP8 and ICP27 proteins with cellular RNA polymerase II holoenzyme.,” J. Virol., vol. 76, no. 12, pp. 5893–904, Jun. 2002.

[291] R. M. Sandri-Goldin, “ICP27 mediates HSV RNA export by shuttling through a leucine-rich nuclear export signal and binding viral intronless RNAs through an RGG motif.,” Genes Dev., vol. 12, no. 6, pp. 868–79, Mar. 1998.

[292] M. D. Koffa et al., “Herpes simplex virus ICP27 protein provides viral mRNAs with access to the cellular mRNA export pathway,” EMBO J., vol. 20, no. 20, pp. 5769–5778, Oct. 2001.

[293] I.-H. B. Chen, L. Li, L. Silva, and R. M. Sandri-Goldin, “ICP27 Recruits Aly/REF but Not TAP/NXF1 to Herpes Simplex Virus Type 1 Transcription Sites although TAP/NXF1 Is Required for ICP27 Export,” J. Virol., vol. 79, no. 7, pp. 3949–3961, Apr. 2005.

[294] I.-H. B. Chen, K. S. Sciabica, and R. M. Sandri-Goldin, “ICP27 interacts with the RNA export factor Aly/REF to direct herpes simplex virus type 1 intronless mRNAs to the TAP export pathway.,” J. Virol., vol. 76, no. 24, pp. 12877–89, Dec. 2002.

[295] F. P. Hernandez and R. M. Sandri-Goldin, “Head-to-tail intramolecular interaction of herpes simplex virus type 1 regulatory protein ICP27 is important for its interaction with cellular mRNA export receptor TAP/NXF1.,” MBio, vol. 1, no. 5, Nov. 2010.

[296] L. A. Johnson, L. Li, and R. M. Sandri-Goldin, “The Cellular RNA Export Receptor TAP/NXF1 Is Required for ICP27-Mediated Export of Herpes Simplex Virus 1 RNA, but the TREX Complex Adaptor Protein Aly/REF Appears To Be Dispensable,” J. Virol., vol. 83, no. 13, pp. 6335–6346, Jul. 2009.

[297] E. C. Fontaine-Rodriguez and D. M. Knipe, “Herpes simplex virus ICP27 increases translation of a subset of viral late mRNAs.,” J. Virol., vol. 82, no. 7, pp. 3538–45, Apr. 2008.

[298] E. C. Fontaine-Rodriguez, T. J. Taylor, M. Olesky, and D. M. Knipe, “Proteomics of herpes simplex virus infected cell protein 27: association with translation initiation factors,” Virology, vol. 330, no. 2, pp. 487–492, Dec. 2004.

[299] O. Larralde, R. W. P. Smith, G. S. Wilkie, P. Malik, N. K. Gray, and J. B. Clements, “Direct stimulation of translation by the multifunctional herpesvirus ICP27 protein.,” J. Virol., vol. 80, no. 3, pp. 1588–91, Feb. 2006.

210

[300] L. Li, L. A. Johnson, J. Q. Dai-Ju, and R. M. Sandri-Goldin, “Hsc70 Focus Formation at the Periphery of HSV-1 Transcription Sites Requires ICP27,” PLoS One, vol. 3, no. 1, p. e1491, Jan. 2008.

[301] A. D. Burch and S. K. Weller, “Nuclear sequestration of cellular chaperone and proteasomal machinery during herpes simplex virus type 1 infection.,” J. Virol., vol. 78, no. 13, pp. 7175–85, Jul. 2004.

[302] C. M. Livingston, M. F. Ifrim, A. E. Cowan, and S. K. Weller, “Virus-Induced Chaperone-Enriched (VICE) Domains Function as Nuclear Protein Quality Control Centers during HSV-1 Infection,” PLoS Pathog., vol. 5, no. 10, p. e1000619, Oct. 2009.

[303] B. Song, J. J. Liu, K.-C. Yeh, and D. M. Knipe, “Herpes Simplex Virus Infection Blocks Events in the G1 Phase of the Cell Cycle,” Virology, vol. 267, no. 2, pp. 326–334, Feb. 2000.

[304] B. Song, K.-C. Yeh, J. Liu, and D. M. Knipe, “Herpes Simplex Virus Gene Products Required for Viral Inhibition of Expression of G1-Phase Functions,” Virology, vol. 290, no. 2, pp. 320–328, Nov. 2001.

[305] D. Hargett, T. McLean, and S. L. Bachenheimer, “Herpes simplex virus ICP27 activation of stress kinases JNK and p38.,” J. Virol., vol. 79, no. 13, pp. 8348–60, Jul. 2005.

[306] P. A. Gillis, L. H. Okagaki, and S. A. Rice, “Herpes Simplex Virus Type 1 ICP27 Induces p38 Mitogen-Activated Protein Kinase Signaling and Apoptosis in HeLa Cells,” J. Virol., vol. 83, no. 4, pp. 1767–1777, Feb. 2009.

[307] G. Zachos, M. Koffa, C. M. Preston, J. B. Clements, and J. Conner, “Herpes Simplex Virus Type 1 Blocks the Apoptotic Host Cell Defense Mechanisms That Target Bcl-2 and Manipulates Activation of p38 Mitogen-Activated Protein Kinase To Improve Viral Replication,” J. Virol., vol. 75, no. 6, pp. 2710–2728, Mar. 2001.

[308] M. Aubert, S. A. Rice, and J. A. Blaho, “Accumulation of Herpes Simplex Virus Type 1 Early and Leaky-Late Proteins Correlates with Apoptosis Prevention in Infected Human HEp-2 Cells,” J. Virol., vol. 75, no. 2, pp. 1013–1030, Jan. 2001.

[309] J. A. Blaho, C. Mitchell, and B. Roizman, “An amino acid sequence shared by the herpes simplex virus 1 α regulatory proteins 0, 4, 22, and 27 predicts the nucleotidylylation of the U(L)21, U(L)31, U(L)47, and U(L)49 gene products,” J. Biol. Chem., vol. 269, no. 26, pp. 17401–17410, 1994.

[310] K. Früh et al., “A viral inhibitor of peptide transporters for antigen presentation,” Nature, vol. 375, no. 6530, pp. 415–418, Jun. 1995.

[311] A. Hill et al., “Herpes simplex virus turns off the TAP to evade host immunity,” Nature, vol. 375, no. 6530, pp. 411–415, Jun. 1995.

[312] I. A. York, C. Roop, D. W. Andrews, S. R. Riddell, F. L. Graham, and D. C. Johnson, “A cytosolic herpes simplex virus protein inhibits antigen presentation to CD8+ T lymphocytes,” Cell, vol. 77, no. 4, pp. 525–535, May 1994.

[313] R. Tomazin et al., “Herpes simplex virus type 2 ICP47 inhibits human TAP but not mouse TAP.,” J. Virol., vol. 72, no. 3, pp. 2560–3, Mar. 1998.

211

[314] K. A. Fraser and S. A. Rice, “Herpes Simplex Virus Immediate-Early Protein ICP22 Triggers Loss of Serine 2-Phosphorylated RNA Polymerase II,” J. Virol., vol. 81, no. 10, pp. 5091–5101, May 2007.

[315] J. Zaborowska et al., “Herpes Simplex Virus 1 (HSV-1) ICP22 Protein Directly Interacts with Cyclin-Dependent Kinase (CDK)9 to Inhibit RNA Polymerase II Transcription Elongation,” PLoS One, vol. 9, no. 9, p. e107654, Sep. 2014.

[316] R. Bruni and B. Roizman, “Herpes simplex virus 1 regulatory protein ICP22 interacts with a new cell cycle-regulated factor and accumulates in a cell cycle-dependent fashion in infected cells.,” J. Virol., vol. 72, no. 11, pp. 8525–31, Nov. 1998.

[317] R. Bruni, B. Fineschi, W. O. Ogle, and B. Roizman, “A novel cellular protein, p60, interacting with both herpes simplex virus 1 regulatory proteins ICP22 and ICP0 is modified in a cell-type- specific manner and Is recruited to the nucleus after infection.,” J. Virol., vol. 73, no. 5, pp. 3810–7, May 1999.

[318] A. Satyanarayana and P. Kaldis, “Mammalian cell-cycle regulation: several Cdks, numerous cyclins and diverse compensatory mechanisms,” Oncogene, vol. 28, no. 33, pp. 2925–2939, Aug. 2009.

[319] S. J. Advani, R. R. Weichselbaum, and B. Roizman, “The role of cdc2 in the expression of herpes simplex virus genes.,” Proc. Natl. Acad. Sci. U. S. A., vol. 97, no. 20, pp. 10996–1001, Sep. 2000.

[320] C. Prod’hon, I. Machuca, H. Berthomme, A. Epstein, and B. Jacquemont, “Characterization of Regulatory Functions of the HSV-1 Immediate-Early Protein ICP22,” Virology, vol. 226, no. 2, pp. 393–402, Dec. 1996.

[321] T. Komatsu, K. Nagata, H. Wodrich, T. Komatsu, K. Nagata, and H. Wodrich, “The Role of Nuclear Antiviral Factors against Invading DNA Viruses: The Immediate Fate of Incoming Viral Genomes,” Viruses, vol. 8, no. 10, p. 290, Oct. 2016.

[322] N. Yan and Z. J. Chen, “Intrinsic antiviral immunity,” Nat. Immunol., vol. 13, no. 3, pp. 214–222, Mar. 2012.

[323] D. Blanco-Melo, S. Venkatesh, and P. D. Bieniasz, “Intrinsic Cellular Defenses against Human Immunodeficiency Viruses,” Immunity, vol. 37, no. 3, pp. 399–411, Sep. 2012.

[324] B. A. Diner, K. K. Lum, and I. M. Cristea, “The emerging role of nuclear viral DNA sensors.,” J. Biol. Chem., vol. 290, no. 44, pp. 26412–21, Oct. 2015.

[325] D. M. Knipe, “Nuclear sensing of viral DNA, epigenetic regulation of herpes simplex virus infection, and innate immunity,” Virology, vol. 479–480, pp. 153–159, May 2015.

[326] A. Dempsey and A. G. Bowie, “Innate immune recognition of DNA: A recent history,” Virology, vol. 479–480, pp. 146–152, May 2015.

[327] K. A. Horan et al., “Proteasomal Degradation of Herpes Simplex Virus Capsids in Macrophages Releases DNA to the Cytosol for Recognition by DNA Sensors,” J. Immunol., vol. 190, no. 5, pp. 2311–2319, Mar. 2013.

[328] J. Wu et al., “Cyclic GMP-AMP Is an Endogenous Second Messenger in Innate Immune Signaling by Cytosolic DNA,” Science (80-. )., vol. 339, no. 6121, pp. 826–830, Feb. 2013.

212

[329] L. Sun, J. Wu, F. Du, X. Chen, and Z. J. Chen, “Cyclic GMP-AMP synthase is a cytosolic DNA sensor that activates the type I interferon pathway.,” Science, vol. 339, no. 6121, pp. 786–91, Feb. 2013.

[330] X. Cai, Y.-H. Chiu, and Z. J. Chen, “The cGAS-cGAMP-STING Pathway of Cytosolic DNA Sensing and Signaling,” Mol. Cell, vol. 54, no. 2, pp. 289–296, Apr. 2014.

[331] P. Gao et al., “Cyclic [G(2′,5′)pA(3′,5′)p] Is the Metazoan Second Messenger Produced by DNA- Activated Cyclic GMP-AMP Synthase,” Cell, vol. 153, no. 5, pp. 1094–1107, May 2013.

[332] X. Zhang et al., “Cyclic GMP-AMP Containing Mixed Phosphodiester Linkages Is An Endogenous High-Affinity Ligand for STING,” Mol. Cell, vol. 51, no. 2, pp. 226–235, Jul. 2013.

[333] A. Ablasser et al., “cGAS produces a 2′-5′-linked cyclic dinucleotide second messenger that activates STING,” Nature, vol. 498, no. 7454, pp. 380–384, Jun. 2013.

[334] E. J. Diner et al., “The Innate Immune DNA Sensor cGAS Produces a Noncanonical Cyclic Dinucleotide that Activates Human STING,” Cell Rep., vol. 3, no. 5, pp. 1355–1361, May 2013.

[335] J. L. Pomerantz and D. Baltimore, “NF-kappa B activation by a signaling complex containing TRAF2, TANK and TBK1, a novel IKK-related kinase,” EMBO J., vol. 18, no. 23, pp. 6694–6704, Dec. 1999.

[336] M. Yoneyama, W. Suhara, Y. Fukuhara, M. Fukuda, E. Nishida, and T. Fujita, “Direct triggering of the type I interferon system by virus infection: activation of a transcription factor complex containing IRF-3 and CBP/p300,” EMBO J., vol. 17, no. 4, pp. 1087–1095, Feb. 1998.

[337] M. Sato, N. Tanaka, N. Hata, E. Oda, and T. Taniguchi, “Involvement of the IRF family transcription factor IRF-3 in virus-induced activation of the IFN-β gene,” FEBS Lett., vol. 425, no. 1, pp. 112–116, Mar. 1998.

[338] R. Lin, C. Heylbroeck, P. M. Pitha, and J. Hiscott, “Virus-dependent phosphorylation of the IRF- 3 transcription factor regulates nuclear translocation, transactivation potential, and proteasome-mediated degradation.,” Mol. Cell. Biol., vol. 18, no. 5, pp. 2986–96, May 1998.

[339] H. Ishikawa and G. N. Barber, “The STING pathway and regulation of innate immune signaling in response to DNA pathogens,” Cell. Mol. Life Sci., vol. 68, no. 7, pp. 1157–1165, Apr. 2011.

[340] M. H. Orzalli et al., “cGAS-mediated stabilization of IFI16 promotes innate signaling during herpes simplex virus infection,” Proc. Natl. Acad. Sci., vol. 112, no. 14, pp. E1773–E1781, Apr. 2015.

[341] C. Su and C. Zheng, “Herpes Simplex Virus 1 Abrogates the cGAS/STING-Mediated Cytosolic DNA-Sensing Pathway via Its Virion Host Shutoff Protein, UL41,” J Virol, vol. 91, no. 6, pp. e02414-16, Mar. 2017.

[342] J. Huang, H. You, C. Su, Y. Li, S. Chen, and C. Zheng, “Herpes Simplex Virus 1 Tegument Protein VP22 Abrogates cGAS/STING-Mediated Antiviral Innate Immunity.,” J. Virol., vol. 92, no. 15, pp. e00841-18, Aug. 2018.

[343] L. Unterholzner et al., “IFI16 is an innate immune sensor for intracellular DNA,” Nat Immunol, vol. 11, no. 11, pp. 997–1004, 2010.

213

[344] M. R. Jakobsen et al., “IFI16 senses DNA forms of the lentiviral replication cycle and controls HIV-1 replication.,” Proc. Natl. Acad. Sci. U. S. A., vol. 110, no. 48, pp. E4571-80, Nov. 2013.

[345] N. Kerur et al., “IFI16 Acts as a Nuclear Pathogen Sensor to Induce the Inflammasome in Response to Kaposi Sarcoma-Associated Herpesvirus Infection,” Cell Host Microbe, vol. 9, no. 5, pp. 363–375, May 2011.

[346] T. Li, B. A. Diner, J. Chen, and I. M. Cristea, “Acetylation modulates cellular distribution and DNA sensing ability of interferon-inducible protein IFI16.,” Proc. Natl. Acad. Sci. U. S. A., vol. 109, no. 26, pp. 10558–63, Jun. 2012.

[347] M. H. Orzalli, N. A. DeLuca, and D. M. Knipe, “Nuclear IFI16 induction of IRF-3 signaling during herpesviral infection and degradation of IFI16 by the viral ICP0 protein,” Proc Natl Acad Sci U S A, vol. 109, no. 44, pp. E3008-17, Oct. 2012.

[348] T. Li, J. Chen, and I. M. Cristea, “Human Cytomegalovirus Tegument Protein pUL83 Inhibits IFI16-Mediated DNA Sensing for Immune Evasion,” Cell Host Microbe, vol. 14, no. 5, pp. 591– 599, Nov. 2013.

[349] S. A. Schattgen and K. A. Fitzgerald, “The PYHIN protein family as mediators of host defenses,” Immunol. Rev., vol. 243, no. 1, pp. 109–118, Sep. 2011.

[350] B. A. Diner et al., “The functional interactome of PYHIN immune regulators reveals IFIX is a sensor of viral DNA,” Mol. Syst. Biol., vol. 11, no. 2, p. 787, Feb. 2015.

[351] T. Jin et al., “Structures of the HIN Domain:DNA Complexes Reveal Ligand Binding and Activation Mechanisms of the AIM2 Inflammasome and IFI16 Receptor,” Immunity, vol. 36, no. 4, pp. 561–571, Apr. 2012.

[352] S. R. Morrone, T. Wang, L. M. Constantoulakis, R. M. Hooy, M. J. Delannoy, and J. Sohn, “Cooperative assembly of IFI16 filaments on dsDNA provides insights into host defense strategy.,” Proc. Natl. Acad. Sci. U. S. A., vol. 111, no. 1, pp. E62-71, Jan. 2014.

[353] M. H. Orzalli and D. M. Knipe, “Cellular Sensing of Viral DNA and Viral Evasion Mechanisms,” Annu. Rev. Microbiol., vol. 68, no. 1, pp. 477–492, Sep. 2014.

[354] J. A. DeCaprio, D. M. Knipe, C. Berrios, S. E. Conwell, and M. H. Orzalli, “Nuclear interferon- inducible protein 16 promotes silencing of herpesviral and transfected DNA,” Proc. Natl. Acad. Sci., vol. 110, no. 47, pp. E4492–E4501, 2013.

[355] Y. Tanaka and Z. J. Chen, “STING specifies IRF3 phosphorylation by TBK1 in the cytosolic DNA signaling pathway.,” Sci. Signal., vol. 5, no. 214, p. ra20, Mar. 2012.

[356] D. B. Stetson and R. Medzhitov, “Recognition of Cytosolic DNA Activates an IRF3-Dependent Innate Immune Response,” Immunity, vol. 24, no. 1, pp. 93–103, Jan. 2006.

[357] B. Zhong et al., “The Adaptor Protein MITA Links Virus-Sensing Receptors to IRF3 Transcription Factor Activation,” Immunity, vol. 29, no. 4, pp. 538–550, Oct. 2008.

[358] J. V. Falvo, B. S. Parekh, C. H. Lin, E. Fraenkel, and T. Maniatis, “Assembly of a Functional Beta Interferon Enhanceosome Is Dependent on ATF-2–c-jun Heterodimer Orientation,” Mol. Cell. Biol., vol. 20, no. 13, pp. 4814–4825, Jul. 2000.

214

[359] D. Thanos and T. Maniatis, “Identification of the rel family members required for virus induction of the human beta interferon gene.,” Mol. Cell. Biol., vol. 15, no. 1, pp. 152–164, Jan. 1995.

[360] M. A. Ansari et al., “Constitutive interferon-inducible protein 16-inflammasome activation during Epstein-Barr virus latency I, II, and III in B and epithelial cells.,” J. Virol., vol. 87, no. 15, pp. 8606–23, Aug. 2013.

[361] K. E. Johnson, L. Chikoti, and B. Chandran, “Herpes simplex virus 1 infection induces activation and subsequent inhibition of the IFI16 and NLRP3 inflammasomes,” J Virol, vol. 87, no. 9, pp. 5005–5018, May 2013.

[362] K. E. Johnson, V. Bottero, S. Flaherty, S. Dutta, V. V. Singh, and B. Chandran, “IFI16 Restricts HSV-1 Replication by Accumulating on the HSV-1 Genome, Repressing HSV-1 Gene Expression, and Directly or Indirectly Modulating Histone Modifications,” PLoS Pathog., vol. 10, no. 11, 2014.

[363] B. A. Diner, K. K. Lum, J. E. Toettcher, and I. M. Cristea, “Viral DNA Sensors IFI16 and Cyclic GMP-AMP Synthase Possess Distinct Functions in Regulating Viral Gene Expression, Immune Defenses, and Apoptotic Responses during Herpesvirus Infection,” MBio, vol. 7, no. 6, 2016.

[364] J. M. Cabral, H. S. Oh, and D. M. Knipe, “ATRX promotes maintenance of herpes simplex virus heterochromatin during chromatin stress,” Elife, vol. 7, pp. 1–32, Nov. 2018.

[365] R. D. Everett, “Dynamic Response of IFI16 and Promyelocytic Leukemia Nuclear Body Components to Herpes Simplex Virus 1 Infection,” J Virol, vol. 90, no. 1, pp. 167–179, Jan. 2016.

[366] T. Alandijany et al., “Distinct temporal roles for the promyelocytic leukaemia (PML) protein in the sequential regulation of intracellular host immunity to HSV-1 infection,” PLoS Pathog, vol. 14, no. 1, p. e1006769, 2018.

[367] D. Cuchet-Lourenço, G. Anderson, E. Sloan, A. Orr, and R. D. Everett, “The Viral Ubiquitin Ligase ICP0 Is neither Sufficient nor Necessary for Degradation of the Cellular DNA Sensor IFI16 during Herpes Simplex Virus 1 Infection,” J. Virol., vol. 87, no. 24, pp. 13422–13432, Dec. 2013.

[368] B. . L. Diner KK.; Javitt, A.; Cristea, IM., “Interactions of the Antiviral Factor Interferon Gamma- Inducible Protein 16 (IFI16) Mediate Immune Signaling and Herpes Simplex Virus-1 Immunosuppression,” Mol Cell Proteomics, vol. 14, no. 9, pp. 2341–2356, 2015.

[369] C. A. Ascoli and G. G. Maul, “Identification of a novel nuclear domain.,” J. Cell Biol., vol. 112, no. 5, pp. 785–95, Mar. 1991.

[370] Z. G. Wang et al., “Role of PML in cell growth and the retinoic acid pathway.,” Science, vol. 279, no. 5356, pp. 1547–51, Mar. 1998.

[371] A. Rabellino and P. P. Scaglioni, “PML Degradation: Multiple Ways to Eliminate PML,” Front. Oncol., vol. 3, p. 60, Mar. 2013.

[372] A. Flotho and F. Melchior, “Sumoylation: A Regulatory Protein Modification in Health and Disease,” Annu. Rev. Biochem., vol. 82, no. 1, pp. 357–385, Jun. 2013.

[373] M. Pearson and P. G. Pelicci, “PML interaction with p53 and its role in apoptosis and replicative senescence,” Oncogene, vol. 20, no. 49, pp. 7250–7256, Oct. 2001.

215

[374] G. Dellaire and D. P. Bazett-Jones, “PML nuclear bodies: dynamic sensors of DNA damage and cellular stress,” BioEssays, vol. 26, no. 9, pp. 963–977, Sep. 2004.

[375] R. Bernardi, A. Papa, and P. P. Pandolfi, “Regulation of apoptosis by PML and the PML-NBs,” Oncogene, vol. 27, no. 48, pp. 6299–6312, Oct. 2008.

[376] E. Krieghoff-Henning and T. G. Hofmann, “Role of nuclear bodies in apoptosis signalling,” Biochim. Biophys. Acta - Mol. Cell Res., vol. 1783, no. 11, pp. 2185–2194, Nov. 2008.

[377] T. Regad, G. Maul, V. Lallemand-Breitenbach, P. P. Pandolfi, H. de Thé, and M. K. Chelbi-Alix, “PML mediates the interferon-induced antiviral state against a complex retrovirus via its association with the viral transactivator,” EMBO J., vol. 20, no. 13, pp. 3495–3505, Jul. 2001.

[378] R. D. Everett and M. K. Chelbi-Alix, “PML and PML nuclear bodies: Implications in antiviral defence,” Biochimie, vol. 89, no. 6–7, pp. 819–830, Jun. 2007.

[379] A. M. Ishov and G. G. Maul, “The periphery of nuclear domain 10 (ND10) as site of DNA virus deposition.,” J. Cell Biol., vol. 134, no. 4, pp. 815–26, Aug. 1996.

[380] G. G. MAUL, A. M. ISHOV, and R. D. EVERETT, “Nuclear Domain 10 as Preexisting Potential Replication Start Sites of Herpes Simplex Virus Type-1,” Virology, vol. 217, no. 1, pp. 67–75, Mar. 1996.

[381] S. L. Uprichard and D. M. Knipe, “Assembly of Herpes Simplex Virus Replication Proteins at Two Distinct Intranuclear Sites,” Virology, vol. 229, no. 1, pp. 113–125, Mar. 1997.

[382] R. D. Everett, G. Sourvinos, and A. Orr, “Recruitment of herpes simplex virus type 1 transcriptional regulatory protein ICP4 into foci juxtaposed to ND10 in live, infected cells.,” J. Virol., vol. 77, no. 6, pp. 3680–9, Mar. 2003.

[383] R. D. Everett and J. Murray, “ND10 components relocate to sites associated with herpes simplex virus type 1 nucleoprotein complexes during virus infection.,” J. Virol., vol. 79, no. 8, pp. 5078–89, Apr. 2005.

[384] Q. Tang, L. Li, A. M. Ishov, V. Revol, A. L. Epstein, and G. G. Maul, “Determination of minimum herpes simplex virus type 1 components necessary to localize transcriptionally active DNA to ND10.,” J. Virol., vol. 77, no. 10, pp. 5821–8, May 2003.

[385] R. D. Everett, C. Boutell, and B. G. Hale, “Interplay between viruses and host sumoylation pathways,” Nat. Rev. Microbiol., vol. 11, no. 6, pp. 400–411, Jun. 2013.

[386] S. Müller and A. Dejean, “Viral immediate-early proteins abrogate the modification by SUMO- 1 of PML and Sp100 proteins, correlating with nuclear body disruption.,” J. Virol., vol. 73, no. 6, pp. 5137–43, Jun. 1999.

[387] R. D. Everett, “ICP0, a regulator of herpes simplex virus during lytic and latent infection,” BioEssays, vol. 22, no. 8, pp. 761–770, Jul. 2000.

[388] A. V Chee, P. Lopez, P. P. Pandolfi, and B. Roizman, “Promyelocytic leukemia protein mediates interferon-based anti-herpes simplex virus 1 effects.,” J. Virol., vol. 77, no. 12, pp. 7101–5, Jun. 2003.

[389] A. Ciccia and S. J. Elledge, “The DNA damage response: making it safe to play with knives.,” Mol.

216

Cell, vol. 40, no. 2, pp. 179–204, Oct. 2010.

[390] N. Hustedt and D. Durocher, “The control of DNA repair by the cell cycle,” Nat. Cell Biol., vol. 19, no. 1, pp. 1–9, Jan. 2017.

[391] A. Maréchal and L. Zou, “DNA damage sensing by the ATM and ATR kinases.,” Cold Spring Harb. Perspect. Biol., vol. 5, no. 9, p. a012716, Sep. 2013.

[392] J. C. Saldivar, D. Cortez, and K. A. Cimprich, “The essential kinase ATR: ensuring faithful duplication of a challenging genome,” Nat. Rev. Mol. Cell Biol., vol. 18, no. 10, pp. 622–636, Aug. 2017.

[393] S. P. Jackson and J. Bartek, “The DNA-damage response in human biology and disease,” Nature, vol. 461, no. 7267, pp. 1071–1078, Oct. 2009.

[394] M. D. Weitzman, C. E. Lilley, and M. S. Chaurushiya, “Genomes in Conflict: Maintaining Genome Integrity During Virus Infection,” Annu. Rev. Microbiol., vol. 64, no. 1, pp. 61–81, Oct. 2010.

[395] C. E. Lilley et al., “DNA repair proteins affect the lifecycle of herpes simplex virus 1,” Proc. Natl. Acad. Sci., vol. 102, no. 16, pp. 5844–5849, Apr. 2005.

[396] K. N. Mohni, C. M. Livingston, D. Cortez, and S. K. Weller, “ATR and ATRIP Are Recruited to Herpes Simplex Virus Type 1 Replication Compartments Even though ATR Signaling Is Disabled,” J. Virol., vol. 84, no. 23, pp. 12152–12164, Dec. 2010.

[397] D. E. Wilkinson and S. K. Weller, “Recruitment of Cellular Recombination and Repair Proteins to Sites of Herpes Simplex Virus Type 1 DNA Replication Is Dependent on the Composition of Viral Proteins within Prereplicative Sites and Correlates with the Induction of the DNA Damage Response,” J. Virol., vol. 78, no. 9, pp. 4783–4796, May 2004.

[398] T. J. Taylor and D. M. Knipe, “Proteomics of Herpes Simplex Virus Replication Compartments: Association of Cellular DNA Replication, Repair, Recombination, and Chromatin Remodeling Proteins with ICP8,” J. Virol., vol. 78, no. 11, pp. 5856–5866, Jun. 2004.

[399] C. E. Lilley, M. S. Chaurushiya, C. Boutell, R. D. Everett, and M. D. Weitzman, “The intrinsic antiviral defense to incoming HSV-1 genomes includes specific DNA repair proteins and is counteracted by the viral protein ICP0,” PLoS Pathog, vol. 7, no. 6, p. e1002084, Jun. 2011.

[400] J. S. Iacovoni et al., “High-resolution profiling of gammaH2AX around DNA double strand breaks in the mammalian genome.,” EMBO J., vol. 29, no. 8, pp. 1446–57, Apr. 2010.

[401] H. Li et al., “Chk2 is required for HSV-1 ICP0-mediated G2/M arrest and enhancement of virus growth.,” Virology, vol. 375, no. 1, pp. 13–23, May 2008.

[402] D. E. Wilkinson and S. K. Weller, “Herpes simplex virus type I disrupts the ATR-dependent DNA- damage response during lytic infection.,” J. Cell Sci., vol. 119, no. Pt 13, pp. 2695–703, Jul. 2006.

[403] S. Smith and S. K. Weller, “HSV-I and the cellular DNA damage response,” Future Virol., vol. 10, no. 4, pp. 383–397, Apr. 2016.

[404] C. Botting, X. Lu, and S. J. Triezenberg, “H2AX phosphorylation and DNA damage kinase activity are dispensable for herpes simplex virus replication,” Virol J, vol. 13, no. 1, p. 15, Dec. 2016.

217

[405] B. J. Placek and S. L. Berger, “Chromatin dynamics during herpes simplex virus-1 lytic infection,” Biochim. Biophys. Acta - Gene Regul. Mech., vol. 1799, no. 3–4, pp. 223–227, 2010.

[406] I. H. Wang et al., “Tracking viral genomes in host cells at single-molecule resolution,” Cell Host Microbe, vol. 14, no. 4, pp. 468–480, 2013.

[407] G. Sourvinos and R. D. Everett, “Visualization of parental HSV-1 genomes and replication compartments in association with ND10 in live infected cells.,” EMBO J., vol. 21, no. 18, pp. 4989–97, Sep. 2002.

[408] K. E. Beatty et al., “Live-Cell Imaging of Cellular Proteins by a Strain-Promoted Azide-Alkyne Cycloaddition,” ChemBioChem, vol. 11, no. 15, pp. 2092–2095, Oct. 2010.

[409] S. P. Somasekharan et al., “Identification and quantification of newly synthesized proteins translationally regulated by YB-1 using a novel Click-SILAC approach,” J Proteomics, vol. 77, pp. e1-10, Dec. 2012.

[410] M. C. Hagemeijer, A. M. Vonk, I. Monastyrska, P. J. M. Rottier, and C. A. M. de Haan, “Visualizing Coronavirus RNA Synthesis in Time by Using Click Chemistry,” J. Virol., vol. 86, no. 10, pp. 5808– 5816, May 2012.

[411] C. W. Tornøe, C. Christensen, and M. Meldal, “Peptidotriazoles on Solid Phase: [1,2,3]- Triazoles by Regiospecific Copper(I)-Catalyzed 1,3-Dipolar Cycloadditions of Terminal Alkynes to Azides,” J. Org. Chem., vol. 67, no. 9, pp. 3057–3064, May 2002.

[412] S. I. Presolski, V. P. Hong, and M. G. Finn, “Copper-Catalyzed Azide-Alkyne Click Chemistry for Bioconjugation.,” Curr. Protoc. Chem. Biol., vol. 3, no. 4, pp. 153–162, 2011.

[413] H. C. Kolb, M. G. Finn, and K. B. Sharpless, “Click Chemistry: Diverse Chemical Function from a Few Good Reactions,” Angew. Chemie Int. Ed., vol. 40, no. 11, pp. 2004–2021, Jun. 2001.

[414] M. Meldal and C. W. Tornøe, “Cu-catalyzed azide-alkyne cycloaddition.,” Chem. Rev., vol. 108, no. 8, pp. 2952–3015, Aug. 2008.

[415] R. Huisgen, “Kinetics and Mechanism of 1,3-Dipolr Cycloadditions,” Angew. Chemie Int. Ed. English, vol. 2, no. 11, pp. 633–645, Nov. 1963.

[416] R. Huisgen, “1,3-Dipolar Cycloadditions. Past and Future,” Angew. Chemie Int. Ed. English, vol. 2, no. 10, pp. 565–598, Oct. 1963.

[417] Z. Cseresnyes, U. Schwarz, and C. M. Green, “Analysis of replication factories in human cells by super-resolution light microscopy,” BMC Cell Biol., vol. 10, no. 1, p. 88, Dec. 2009.

[418] A. Salic and T. J. Mitchison, “A chemical method for fast and sensitive detection of DNA synthesis in vivo,” Proc. Natl. Acad. Sci., vol. 105, no. 7, pp. 2415–2420, Feb. 2008.

[419] N. Schmidt, T. Hennig, R. A. Serwa, M. Marchetti, and P. O’Hare, “Remote Activation of Host Cell DNA Synthesis in Uninfected Cells Signaled by Infected Cells in Advance of Virus Transmission.,” J. Virol., vol. 89, no. 21, pp. 11107–15, Nov. 2015.

[420] Johannes Gierlich, Glenn A. Burley, Philipp M. E. Gramlich, and David M. Hammond, and T. Carell*, “Click Chemistry as a Reliable Method for the High-Density Postsynthetic Functionalization of Alkyne-Modified DNA,” 2006.

218

[421] L. Guan, G. W. van der Heijden, A. Bortvin, and M. M. Greenberg, “Intracellular Detection of Cytosine Incorporation in Genomic DNA by Using 5-Ethynyl-2′-Deoxycytidine,” ChemBioChem, vol. 12, no. 14, pp. 2184–2190, Sep. 2011.

[422] A. B. Neef and N. W. Luedtke, “Dynamic metabolic labeling of DNA in vivo with arabinosyl nucleosides.,” Proc. Natl. Acad. Sci. U. S. A., vol. 108, no. 51, pp. 20404–9, Dec. 2011.

[423] J. A. Dembowski and N. A. DeLuca, “Selective recruitment of nuclear factors to productively replicating herpes simplex virus genomes,” PLoS Pathog, vol. 11, no. 5, p. e1004939, 2015.

[424] G. Orphanides and D. Reinberg, “A unified theory of gene expression.,” Cell, vol. 108, no. 4, pp. 439–51, Feb. 2002.

[425] M. C. Thomas and C.-M. Chiang, “The General Transcription Machinery and General Cofactors,” Crit. Rev. Biochem. Mol. Biol., vol. 41, no. 3, pp. 105–178, Jan. 2006.

[426] G. Dieci and A. Sentenac, “Detours and shortcuts to transcription reinitiation,” Trends Biochem. Sci., vol. 28, no. 4, pp. 202–209, Apr. 2003.

[427] B. Li, M. Carey, and J. L. Workman, “The Role of Chromatin during Transcription,” Cell, vol. 128, no. 4, pp. 707–719, Feb. 2007.

[428] A. Saunders, L. J. Core, and J. T. Lis, “Breaking barriers to transcription elongation,” Nat. Rev. Mol. Cell Biol., vol. 7, no. 8, pp. 557–567, Aug. 2006.

[429] K. Struhl, “Chromatin structure and RNA polymerase II connection: implications for transcription.,” Cell, vol. 84, no. 2, pp. 179–82, Jan. 1996.

[430] X. Darzacq and R. H. Singer, “The Dynamic Range of Transcription,” Mol. Cell, vol. 30, no. 5, pp. 545–546, Jun. 2008.

[431] D. Yean and J. Gralla, “Transcription reinitiation rate: a special role for the TATA box.,” Mol. Cell. Biol., vol. 17, no. 7, pp. 3809–3816, Jul. 1997.

[432] A. Raj, C. S. Peskin, D. Tranchina, D. Y. Vargas, and S. Tyagi, “Stochastic mRNA synthesis in mammalian cells,” PLoS Biol., vol. 4, no. 10, pp. 1707–1719, 2006.

[433] M. B. Elowitz, “Stochastic Gene Expression in a Single Cell,” Science (80-. )., vol. 297, no. 5584, pp. 1183–1186, Aug. 2002.

[434] B. B. Kaufmann and A. van Oudenaarden, “Stochastic gene expression: from single molecules to the proteome,” Curr. Opin. Genet. Dev., vol. 17, no. 2, pp. 107–112, Apr. 2007.

[435] D. M. Suter, N. Molina, D. Gatfield, K. Schneider, U. Schibler, and F. Naef, “Mammalian Genes Are Transcribed with Widely Different Bursting Kinetics,” Science (80-. )., vol. 332, no. 6028, pp. 472–474, Apr. 2011.

[436] C. R. Bartman, S. C. Hsu, C. C.-S. Hsiung, A. Raj, and G. A. Blobel, “Enhancer Regulation of Transcriptional Bursting Parameters Revealed by Forced Chromatin Looping,” Mol. Cell, vol. 62, no. 2, pp. 237–247, Apr. 2016.

[437] D. Zenklusen, D. R. Larson, and R. H. Singer, “Single-RNA counting reveals alternative modes of gene expression in yeast,” Nat. Struct. Mol. Biol., vol. 15, no. 12, pp. 1263–1271, Dec. 2008.

219

[438] J. R. Chubb, T. Trcek, S. M. Shenoy, and R. H. Singer, “Transcriptional Pulsing of a Developmental Gene,” Curr. Biol., vol. 16, no. 10, pp. 1018–1025, May 2006.

[439] K. Bahar Halpern et al., “Bursty Gene Expression in the Intact Mammalian Liver,” Mol. Cell, vol. 58, no. 1, pp. 147–156, Apr. 2015.

[440] T. Lionnet et al., “A transgenic mouse for in vivo detection of endogenous labeled mRNA,” Nat. Methods, vol. 8, no. 2, pp. 165–170, Feb. 2011.

[441] R. A. Coleman, Z. Liu, X. Darzacq, R. Tjian, R. H. Singer, and T. Lionnet, “Imaging Transcription: Past, Present, and Future.,” Cold Spring Harb. Symp. Quant. Biol., vol. 80, pp. 1–8, Jan. 2015.

[442] I. Golding, J. Paulsson, S. M. Zawilski, and E. C. Cox, “Real-Time Kinetics of Gene Activity in Individual Bacteria,” Cell, vol. 123, no. 6, pp. 1025–1036, Dec. 2005.

[443] R. D. Dar et al., “Transcriptional burst frequency and burst size are equally modulated across the human genome.,” Proc. Natl. Acad. Sci. U. S. A., vol. 109, no. 43, pp. 17454–9, Oct. 2012.

[444] A. Senecal et al., “Transcription Factors Modulate c-Fos Transcriptional Bursts,” Cell Rep., vol. 8, no. 1, pp. 75–83, Jul. 2014.

[445] A. Kalo et al., “Cellular Levels of Signaling Factors Are Sensed by β-actin Alleles to Modulate Transcriptional Pulse Intensity,” Cell Rep., vol. 13, no. 6, pp. 1284–1285, Nov. 2015.

[446] D. Levens and D. R. Larson, “A new twist on transcriptional bursting.,” Cell, vol. 158, no. 2, pp. 241–242, Jul. 2014.

[447] D. Nicolas, N. E. Phillips, and F. Naef, “What shapes eukaryotic transcriptional bursting?,” Mol. Biosyst., vol. 13, no. 7, pp. 1280–1290, Jun. 2017.

[448] G. Otto, “Enhancers and promoters regulate burst kinetics,” Nat. Rev. Mol. Cell Biol., vol. 20, no. 3, pp. 134–135, Mar. 2019.

[449] D. Nicolas, B. Zoller, D. M. Suter, and F. Naef, “Modulation of transcriptional burst frequency by histone acetylation.,” Proc. Natl. Acad. Sci. U. S. A., vol. 115, no. 27, pp. 7153–7158, Jul. 2018.

[450] C. R. Bartman et al., “Transcriptional Burst Initiation and Polymerase Pause Release Are Key Control Points of Transcriptional Regulation,” Mol. Cell, vol. 73, no. 3, pp. 519-532.e4, 2019.

[451] A. M. Corrigan, E. Tunnacliffe, D. Cannon, and J. R. Chubb, “A continuum model of transcriptional bursting,” Elife, vol. 5, Feb. 2016.

[452] F. Wang et al., “RNAscope: a novel in situ RNA analysis platform for formalin-fixed, paraffin- embedded tissues.,” J. Mol. Diagn., vol. 14, no. 1, pp. 22–9, Jan. 2012.

[453] A. N. Player, L.-P. Shen, D. Kenny, V. P. Antao, and J. A. Kolberg, “Single-copy Gene Detection Using Branched DNA (bDNA) In Situ Hybridization,” J. Histochem. Cytochem., vol. 49, no. 5, pp. 603–611, May 2001.

[454] N. Battich, T. Stoeger, and L. Pelkmans, “Image-based transcriptomics in thousands of single human cells at single-molecule resolution,” Nat Methods, vol. 10, no. 11, pp. 1127–1133, 2013.

[455] N. Battich, T. Stoeger, and L. Pelkmans, “Control of Transcript Variability in Single Mammalian

220

Cells,” Cell, vol. 163, no. 7, pp. 1596–1610, 2015.

[456] T. Stoeger, N. Battich, M. D. Herrmann, Y. Yakimovich, and L. Pelkmans, “Computer vision for image-based transcriptomics,” Methods, vol. 85, pp. 44–53, Sep. 2015.

[457] J. F. Almine et al., “IFI16 and cGAS cooperate in the activation of STING during DNA sensing in human keratinocytes.,” Nat. Commun., vol. 8, p. 14392, 2017.

[458] F. de Chaumont et al., “Icy: an open bioimage informatics platform for extended reproducible research,” Nat. Methods, vol. 9, no. 7, pp. 690–696, Jul. 2012.

[459] G. Ball, J. Demmerle, R. Kaufmann, I. Davis, I. M. Dobbie, and L. Schermelleh, “SIMcheck: a Toolbox for Successful Super-resolution Structured Illumination Microscopy,” Sci. Rep., vol. 5, no. 1, p. 15915, Dec. 2015.

[460] P. Andrey et al., “Statistical analysis of 3D images detects regular spatial distributions of centromeres and chromocenters in animal and plant nuclei,” PLoS Comput Biol, vol. 6, no. 7, p. e1000853, 2010.

[461] E. Sekine, N. Schmidt, D. Gaboriau, and P. O’Hare, “Spatiotemporal dynamics of HSV genome nuclear entry and compaction state transitions using bioorthogonal chemistry and super- resolution microscopy,” PLoS Pathog., vol. 13, no. 11, 2017.

[462] A. de Bruyn Kops and D. M. Knipe, “Formation of DNA replication structures in herpes virus- infected cells requires a viral DNA binding protein,” Cell, vol. 55, no. 5, pp. 857–868, 1988.

[463] F. Puvion-Dutilleul and S. Besse, “Induction of complete segregation of cellular DNA and non- encapsidated viral genomes in herpes simplex virus type 1 infected HeLa cells as revealed by in situ hybridization,” Chromosoma, vol. 103, no. 2, pp. 104–110, 1994.

[464] K. Monier, J. C. G. Armas, S. Etteldorf, P. Ghazal, and K. F. Sullivan, “Annexation of the interchromosomal space during viral infection,” Nat. Cell Biol., vol. 2, no. 9, pp. 661–665, Sep. 2000.

[465] G. L. Ehmann, T. I. McLean, and S. L. Bachenheimer, “Herpes simplex virus type 1 infection imposes a G(1)/S block in asynchronously growing cells and prevents G(1) entry in quiescent cells,” Virology, vol. 267, no. 2, pp. 335–349, 2000.

[466] W. W. Newcomb, S. K. Cockrell, F. L. Homa, and J. C. Brown, “Polarized DNA ejection from the herpesvirus capsid,” J Mol Biol, vol. 392, no. 4, pp. 885–894, 2009.

[467] M. G. L. Gustafsson et al., “Three-Dimensional Resolution Doubling in Wide-Field Fluorescence Microscopy by Structured Illumination,” vol. 94, no. June, 2008.

[468] C. M. Preston, M. J. Nicholl, and A. Rinaldi, “Herpes simplex virus type 1 immediate early gene expression is stimulated by inhibition of protein synthesis.,” J. Gen. Virol., vol. 79, no. 1, pp. 117–124, Jan. 1998.

[469] M. G. Delboy, D. G. Roller, and A. V Nicola, “Cellular proteasome activity facilitates herpes simplex virus entry at a postpenetration step,” J Virol, vol. 82, no. 7, pp. 3381–3390, 2008.

[470] A. L. Goldberg, “Development of proteasome inhibitors as research tools and cancer drugs.,” J. Cell Biol., vol. 199, no. 4, pp. 583–8, Nov. 2012.

221

[471] B. Sodeik, “Mechanisms of viral transport in the cytoplasm,” Trends Microbiol, vol. 8, no. 10, pp. 465–472, 2000.

[472] K. Kristensson, E. Lycke, M. Röyttä, B. Svennerholm, and A. Vahlne, “Neuritic transport of herpes simplex virus in rat sensory neurons in vitro. Effects of substances interacting with microtubular function and axonal flow [nocodazole, taxol and erythro-9-3-(2- hydroxynonyl)adenine].,” J. Gen. Virol., vol. 67 ( Pt 9), no. 9, pp. 2023–8, Sep. 1986.

[473] R. D. Everett et al., “A dynamic connection between centromeres and ND10 proteins.,” J. Cell Sci., vol. 112 ( Pt 20), pp. 3443–54, Oct. 1999.

[474] V. Lallemand-Breitenbach et al., “Role of promyelocytic leukemia (PML) sumolation in nuclear body formation, 11S proteasome recruitment, and As2O3-induced PML or PML/retinoic acid receptor alpha degradation.,” J. Exp. Med., vol. 193, no. 12, pp. 1361–71, Jun. 2001.

[475] F. Toyoshima, T. Moriguchi, A. Wada, M. Fukuda, and E. Nishida, “Nuclear export of cyclin B1 and its possible role in the DNA damage-induced G2 checkpoint.,” EMBO J., vol. 17, no. 10, pp. 2728–35, May 1998.

[476] M. A. Jordan and L. Wilson, “Microtubules and actin filaments: dynamic targets for cancer chemotherapy.,” Curr. Opin. Cell Biol., vol. 10, no. 1, pp. 123–30, Feb. 1998.

[477] K. H. Downing, “Structural Basis for the Interaction of Tubulin with Proteins and Drugs that Affect Microtubule Dynamics,” Annu. Rev. Cell Dev. Biol., vol. 16, no. 1, pp. 89–111, Nov. 2000.

[478] J. A. DiDonato, J. R. Spitzner, and M. T. Muller, “A predictive model for DNA recognition by the herpes simplex virus protein ICP4.,” J. Mol. Biol., vol. 219, no. 3, pp. 451–70, Jun. 1991.

[479] J. P. Weir, “Regulation of herpes simplex virus gene expression,” Gene, vol. 271, no. 2, pp. 117– 130, Jun. 2001.

[480] D. Rieder, Z. Trajanoski, and J. G. McNally, “Transcription factories.,” Front. Genet., vol. 3, p. 221, 2012.

[481] H. Sutherland and W. A. Bickmore, “Transcription factories: gene expression in unions?,” Nat. Rev. Genet., vol. 10, no. 7, pp. 457–466, Jul. 2009.

[482] K. Tantale et al., “A single-molecule view of transcription reveals convoys of RNA polymerases and multi-scale bursting,” Nat. Commun., vol. 7, 2016.

[483] P. O’Hare and G. S. Hayward, “Evidence for a direct role for both the 175,000- and 110,000- molecular-weight immediate-early proteins of herpes simplex virus in the transactivation of delayed-early promoters.,” J. Virol., vol. 53, no. 3, pp. 751–760, Mar. 1985.

[484] B. Gu, R. Rivera-Gonzalez, C. A. Smith, and N. A. DeLuca, “Herpes simplex virus infected cell polypeptide 4 preferentially represses Sp1-activated over basal transcription from its own promoter,” Proc. Natl. Acad. Sci., vol. 90, no. 20, pp. 9528–9532, Oct. 1993.

[485] G. Elliott, K. Pheasant, K. Ebert-Keel, J. Stylianou, A. Franklyn, and J. Jones, “Multiple Posttranscriptional Strategies To Regulate the Herpes Simplex Virus 1 vhs Endoribonuclease,” J. Virol., vol. 92, no. 17, pp. e00818-18, Sep. 2018.

[486] L. M. Wagner, A. Bayer, and N. A. Deluca, “Requirement of the N-terminal activation domain

222

of herpes simplex virus ICP4 for viral gene expression.,” J. Virol., vol. 87, no. 2, pp. 1010–8, Jan. 2013.

[487] D. Hotter et al., “IFI16 Targets the Transcription Factor Sp1 to Suppress HIV-1 Transcription and Latency Reactivation,” Cell Host Microbe, vol. 25, no. 6, pp. 858-872.e13, Jun. 2019.

[488] D.-B. Kim and N. A. DeLuca, “Phosphorylation of transcription factor Sp1 during herpes simplex virus type 1 infection.,” J. Virol., vol. 76, no. 13, pp. 6473–9, Jul. 2002.

[489] C. E. Samuel, “Antiviral actions of interferons.,” Clin. Microbiol. Rev., vol. 14, no. 4, pp. 778– 809, table of contents, Oct. 2001.

[490] A. García-Sastre and C. A. Biron, “Type 1 interferons and the virus-host relationship: a lesson in détente .,” Science, vol. 312, no. 5775, pp. 879–82, May 2006.

[491] J. C. Brown, “High G+C Content of Herpes Simplex Virus DNA: Proposed Role in Protection Against Retrotransposon Insertion.,” Open Biochem. J., vol. 1, pp. 33–42, Dec. 2007.

[492] E. De Clercq et al., “Antiviral, antimetabolic, and cytotoxic activities of 5-substituted 2’- deoxycytidines.,” Mol. Pharmacol., vol. 21, no. 1, 1982.

[493] P. J. Barr, P. A. Nolan, D. V Santi, and M. J. Robins, “Inhibition of thymidylate synthetase by 5- alkynyl-2’-deoxyuridylates.,” J. Med. Chem., vol. 24, no. 12, pp. 1385–8, Dec. 1981.

[494] E. DE CLERCQ et al., “Thymidylate Synthetase as Target Enzyme for the Inhibitory Activity of 5- Substituted 2’-Deoxyuridines on Mouse Leukemia L1210 Cell Growth,” Mol. Pharmacol., vol. 19, no. 2, 1981.

[495] C. J. Lukonis, J. Burkham, and S. K. Weller, “Herpes Simplex Virus Type 1 Prereplicative Sites Are a Heterogeneous Population: Only a Subset Are Likely To Be Precursors to Replication Compartments,” J Virol, vol. 71, no. 6, pp. 4771–4781, 1997.

[496] R. Tandon, E. S. Mocarski, and J. F. Conway, “The A, B, Cs of herpesvirus capsids.,” Viruses, vol. 7, no. 3, pp. 899–914, Feb. 2015.

[497] U. Sae-Ueng et al., “Solid-to-fluid DNA transition inside HSV-1 capsid close to the temperature of infection,” Nat. Chem. Biol., vol. 10, no. 10, pp. 861–867, Oct. 2014.

[498] S. H. Behrens and D. G. Grier, “The charge of glass and silica surfaces,” J. Chem. Phys., vol. 115, no. 14, pp. 6716–6721, Oct. 2001.

[499] D. W. Bauer et al., “Exploring the Balance between DNA Pressure and Capsid Stability in Herpesviruses and Phages,” J. Virol., vol. 89, no. 18, pp. 9288–9298, Sep. 2015.

[500] W. C. Russell, “Adenoviruses: update on structure and function,” J. Gen. Virol., vol. 90, no. 1, pp. 1–20, Jan. 2009.

[501] Y. Yamauchi and U. F. Greber, “Principles of Virus Uncoating: Cues and the Snooker Ball,” Traffic, vol. 17, no. 6, pp. 569–592, Jun. 2016.

[502] U. F. Greber, M. Willetts, P. Webster, and A. Helenius, “Stepwise dismantling of adenovirus 2 during entry into cells.,” Cell, vol. 75, no. 3, pp. 477–86, Nov. 1993.

223

[503] O. Kobiler, Y. Lipman, K. Therkelsen, I. Daubechies, and L. W. Enquist, “Herpesviruses carrying a Brainbow cassette reveal replication and expression of limited numbers of incoming genomes,” Nat. Commun., vol. 1, no. 1, p. 146, Dec. 2010.

[504] M. P. Taylor, O. Kobiler, and L. W. Enquist, “Alphaherpesvirus axon-to-cell spread involves limited virion transmission,” Proc. Natl. Acad. Sci., vol. 109, no. 42, pp. 17046–17051, Oct. 2012.

[505] E. M. Cohen and O. Kobiler, “Gene Expression Correlates with the Number of Herpes Viral Genomes Initiating Infection in Single Cells,” PLOS Pathog., vol. 12, no. 12, p. e1006082, Dec. 2016.

[506] J. Oh et al., “Genome wide nucleosome mapping for HSV-1 shows nucleosomes are deposited at preferred positions during lytic infection.,” PLoS One, vol. 10, no. 2, p. e0117471, 2015.

[507] D. A. Jackson, F. J. Iborra, E. M. Manders, and P. R. Cook, “Numbers and organization of RNA polymerases, nascent transcripts, and transcription units in HeLa nuclei.,” Mol. Biol. Cell, vol. 9, no. 6, pp. 1523–36, Jun. 1998.

[508] A. Papantonis et al., “Active RNA polymerases: mobile or immobile molecular machines?,” PLoS Biol., vol. 8, no. 7, p. e1000419, Jul. 2010.

[509] A. A. Gavrilov, I. S. Zukher, E. S. Philonenko, S. V. Razin, and O. V. Iarovaia, “Mapping of the nuclear matrix-bound chromatin hubs by a new M3C experimental procedure,” Nucleic Acids Res., vol. 38, no. 22, p. 8051, 2010.

[510] M. D. Challberg and T. J. Kelly, “Animal Virus DNA Replication,” Annu. Rev. Biochem., vol. 58, no. 1, pp. 671–713, Jun. 1989.

[511] A. R. Cliffe and D. M. Knipe, “Herpes Simplex Virus ICP0 Promotes both Histone Removal and Acetylation on Viral DNA during Lytic Infection,” J. Virol., vol. 82, no. 24, pp. 12030–12038, Dec. 2008.

[512] S. B. Kutluay and S. J. Triezenberg, “Role of chromatin during herpesvirus infections,” Biochim. Biophys. Acta - Gen. Subj., vol. 1790, no. 6, pp. 456–466, Jun. 2009.

[513] J. M. Harkness, M. Kader, and N. A. DeLuca, “Transcription of the Herpes Simplex Virus 1 Genome during Productive and Quiescent Infection of Neuronal and Nonneuronal Cells,” J. Virol., vol. 88, no. 12, pp. 6847–6861, Jun. 2014.

[514] N. Michael and B. Roizman, “Repression of the herpes simplex virus 1 alpha 4 gene by its gene product occurs within the context of the viral genome and is associated with all three identified cognate sites.,” Proc. Natl. Acad. Sci. U. S. A., vol. 90, no. 6, pp. 2286–90, Mar. 1993.

[515] K. Pheasant, C. S. Möller-Levet, J. Jones, D. Depledge, J. Breuer, and G. Elliott, “Nuclear- cytoplasmic compartmentalization of the herpes simplex virus 1 infected cell transcriptome is co-ordinated by the viral endoribonuclease vhs and cofactors to facilitate the translation of late proteins,” PLOS Pathog., vol. 14, no. 11, p. e1007331, Nov. 2018.

[516] W. Cun, M. Hong, L.-D. Liu, C.-H. Dong, J. Luo, and Q.-H. Li, “Structural and Functional Characterization of Herpes Simplex Virus 1 Immediate-Early Protein Infected-Cell Protein 22,” J. Biochem., vol. 140, no. 1, pp. 67–73, Jul. 2006.

[517] F. C. Purves, W. O. Ogle, B. Roizman, and B. Roizman, “Processing of the herpes simplex virus

224

regulatory protein alpha 22 mediated by the UL13 protein kinase determines the accumulation of a subset of alpha and gamma mRNAs and proteins in infected cells.,” vol. 90, no. 14, Jul. 1993.

[518] M. C. Long, V. Leong, P. A. Schaffer, C. A. Spencer, and S. A. Rice, “ICP22 and the UL13 protein kinase are both required for herpes simplex virus-induced modification of the large subunit of RNA polymerase II.,” J. Virol., vol. 73, no. 7, pp. 5593–604, Jul. 1999.

[519] D. G. Negorev, O. V. Vladimirova, and G. G. Maul, “Differential Functions of Interferon- Upregulated Sp100 Isoforms: Herpes Simplex Virus Type 1 Promoter-Based Immediate-Early Gene Suppression and PML Protection from ICP0-Mediated Degradation,” J. Virol., vol. 83, no. 10, pp. 5168–5180, May 2009.

[520] C. Lavau et al., “The acute promyelocytic leukaemia-associated PML gene is induced by interferon.,” Oncogene, vol. 11, no. 5, pp. 871–6, Sep. 1995.

[521] M. K. Chelbi-Alix et al., “Induction of the PML protein by interferons in normal and APL cells.,” Leukemia, vol. 9, no. 12, pp. 2027–33, Dec. 1995.

[522] C. Tsui et al., “dCas9-targeted locus-specific protein isolation method identifies histone gene regulators.,” Proc. Natl. Acad. Sci. U. S. A., vol. 115, no. 12, pp. E2734–E2741, Mar. 2018.

[523] J. R. Brister, D. Ako-Adjei, Y. Bao, and O. Blinkova, “NCBI viral genomes resource.,” Nucleic Acids Res., vol. 43, no. Database issue, pp. D571-7, Jan. 2015.

[524] D. L. Bentley, “Coupling mRNA processing with transcription in time and space.,” Nat. Rev. Genet., vol. 15, no. 3, pp. 163–75, 2014.

[525] L. Herzel, D. S. M. Ottoz, T. Alpert, and K. M. Neugebauer, “Splicing and transcription touch base: co-transcriptional spliceosome assembly and function,” Nat. Rev. Mol. Cell Biol., vol. 18, no. 10, pp. 637–650, Oct. 2017.

[526] J. Han, J. Xiong, D. Wang, and X.-D. Fu, “Pre-mRNA splicing: where and when in the nucleus,” Trends Cell Biol., vol. 21, no. 6, pp. 336–343, Jun. 2011.

[527] A. Singh, B. Razooky, C. D. Cox, M. L. Simpson, and L. S. Weinberger, “Transcriptional bursting from the HIV-1 promoter is a significant source of stochastic noise in HIV-1 gene expression.,” Biophys. J., vol. 98, no. 8, pp. L32-4, Apr. 2010.

[528] F. Abaitua, M. Hollinshead, M. Bolstad, C. M. Crump, and P. O’Hare, “A Nuclear localization signal in herpesvirus protein VP1-2 is essential for infection via capsid routing to the nuclear pore,” J Virol, vol. 86, no. 17, pp. 8998–9014, 2012.

[529] C. Su Hui Teo, R. A. Serwa, and P. O’Hare, “Spatial and Temporal Resolution of Global Protein Synthesis during HSV Infection Using Bioorthogonal Precursors and Click Chemistry,” PLos Pathog., vol. 12, no. 10, p. e1005927, 2016.

[530] R. A. Serwa, E. Sekine, J. Brown, S. H. C. Teo, E. W. Tate, and P. O’Hare, “Analysis of a fully infectious bio-orthogonally modified human virus reveals novel features of virus cell entry,” PLOS Pathog., vol. 15, no. 10, p. e1007956, Oct. 2019.

225

9. Appendix

PLoS Pathogens. Published.

Spatiotemporal Dynamics of HSV DNA Genome Nuclear Entry and

Compaction State Using Bioorthogonal Chemistry and Super-

resolution Microscopy

Eiki Sekine1, Nora Schmidt1, David Gaboriau2 and Peter O’Hare1*

1 Section of Virology, Faculty of Medicine, Imperial College London, St Mary’s Medical School, Norfolk Place, London W2 1PG, United Kingdom

2 Department of Medicine, Facility for Imaging by Light Microscopy, National Heart and Lung Institute, Imperial College London Exhibition Road, London SW7 2AZ, United Kingdom

*Corresponding author

Email: [email protected]

Short Title: Transport and Nuclear Entry of the HSV Genome

226