Utilization of a mouse model of infection to investigate Kaposi’s sarcoma-associated herpesvirus LANA function in vivo

Aura Daniela Gonçalves Franco

Thesis to obtain the Master of Science Degree in

Microbiology

Supervisors: Professor Doctor João Pedro Monteiro e Louro Machado de Simas Professor Doctor Arsénio do Carmo Sales Mendes Fialho

Examination Committee

Chairperson: Professor Doctor Jorge Humberto Gomes Leitão Supervisor: Professor Doctor João Pedro Monteiro e Louro Machado de Simas Member of the Committee: Doctor Sofia Isabel Arriaga Mimoso Cerqueira

October 2017 Acknowledgments

Em primeiro lugar, gostaria de agradecer ao meu orientador Doutor Pedro Simas por me ter dado a oportunidade de realizar a minha tese de mestrado no seu laboratório, pela orientação cientifica e por me ter proporcionado uma formação de qualidade durante estes meses. Um especial agradecimento à Doutora Marta Miranda pelo apoio, pela orientação científica e pela paciência que demonstrou para me ensinar e me aconselhar.

Quero também agradecer à Doutora Isabel Sá Correia, a coordenadora do Mestrado em Microbiologia, e ao Doutor Arsénio Fialho, o meu orientador interno, pela simpatia e disponibilidade que demonstraram para esclarecer todas as minhas questões.

Obrigada à Katia, a minha colega de laboratório, por todos os nossos serões no IMM, pela prontidão que sempre demonstraste para me ajudar e pela cumplicidade que desenvolvemos ao longo destes últimos meses.

A todos os meus amigos que acompanharam o meu percurso e com os quais eu posso sempre contar. Um obrigada especial à Jéssica, ao Tiago, à Inês e à Bá. Obrigada por todos os nossos momentos, por toda a força e conselhos que me deram e, principalmente, obrigada por terem acreditado em mim quando eu deixei de acreditar.

Aos meus familiares, ao meu pai, à minha mãe, à minha irmã, mas especialmente à minha tia, por ser o meu pilar, o meu braço direito, por ter aturado todas as crises existenciais que tive ao longo destes meses, por estar sempre presente quando eu precisei, por me ter sempre apoiado e por ter sempre uma palavra amiga para me dar.

E por fim, mas não menos importante, aos meus avós, especialmente ao meu avô Agostinho que tem sido uma fonte de inspiração nestes últimos anos. Obrigada do fundo do coração. Dedico-te esta tese, porque sem ti nada disto teria sido possível!

Um beijinho para todos!

2 Abstract

Kaposi’s sarcoma-associated herpesvirus is a herpesvirus that can infect humans and they have the ability to establish latency, causing lifelong infections in the host. The protein responsible for establishment of latency is the latency-associated nuclear antigen (LANA). Since this protein is of great importance, it is important to understand the mechanisms underlying the establishment of latency, to control gammaherpesviruses infections. One of the biggest problems to study these herpesviruses is their narrow host tropism, so the identification of murid herpesvirus 68 (MHV-68), a mouse equivalent virus which also encodes a LANA (mLANA), enabled the possibility of development of a mouse model to study gammaherpesviruses pathogenesis. Previous studies showed that the replacement of mLANA by kLANA in MHV-68 background was a good model to study KSHV pathogenesis in a mouse model of infection. The aim of this work was to test the influence of the aspartate and glutamate (DE) repeat region, in a mouse model of infection, by generating a recombinant virus: v-kLANA Δ331-495, in which the DE region was deleted. Two other recombinant viruses were added to this study, in order to complement the results: v-kLANA Δ465-929 and v-kLANA Δ332-929. In vivo results showed that v- kLANA Δ331-495 establish very low levels of latency, compared with v-kLANA, v-kLANA Δ332-929 failed to establish latency and v-kLANA Δ465-929 presented levels of latency similar to v-kLANA. Altogether, these data indicate that within internal region of LANA, the DE region exert a more important role in establishment of latency in the spleen, during an infection.

Keywords: KSHV; MHV-68; LANA; latency; internal repeat region; in vivo.

3 Resumo

O herpesvírus associado ao sarcoma de Kaposi é um herpesvírus que infeta humanos e têm a capacidade de estabelecer latência, causando infeções crónicas no hospedeiro. A proteína responsável pelo estabelecimento da latência é o antigénio nuclear associado à latência (LANA) e por isso é necessário compreender os mecanismos no estabelecimento da latência, para controlar estas infeções. Um dos maiores problemas para o estudo destes herpesvírus, é o facto de possuírem uma pequena gama de hospedeiros, por isso a identificação do herpesvírus murídeo 68 (MHV-68), um vírus equivalente ao KSHV que também codifica uma LANA (mLANA), possibilitou o desenvolvimento de um modelo animal para estudar a patogénese destes herpesvírus. Estudos realizados mostraram que a substituição da mLANA pela kLANA no genoma do MHV-68 é um bom modelo para estudar a patogenicidade do KSHV num modelo animal. O objetivo deste trabalho foi testar a influência da região aspartato e glutamato (DE), num modelo animal, gerando, para isso, um vírus recombinante: v-kLANA Δ331-495, na qual a região DE foi excluída. Dois outros vírus recombinantes foram adicionados a este estudo, para complementar os resultados: v-kLANA Δ465-929 e v-kLANA Δ332-929. Os resultados das experiências mostraram que v-kLANA Δ331-495 estabelece níveis inferiores de latência comparado ao v-kLANA, v-kLANA Δ332-929 não conseguiu estabelecer latência e v-kLANA Δ465-929 apresentou níveis de latência semelhantes aos da v-kLANA. De modo geral, todos os dados indicam que dentro da região interna da LANA, a região DE parece exercer um papel mais importante no estabelecimento de latência no baço, durante uma infeção.

Palavras-chave: KSHV; MHV-68; LANA; região interna repetitiva; in vivo.

4 Table of Contents

Acknowledgments ...... 2

Abstract ...... 3

Resumo ...... 4

Table of Contents ...... 5

Index of Figures ...... 7

Index of Tables ...... 8

Abbreviations...... 9

Chapter 1. Introduction ...... 12 1.1. Herpesviridae ...... 12 1.2. The Gammaherpesvirinae subfamily ...... 14 1.2.1. Kaposi’s sarcoma-associated herpesvirus (KSHV) ...... 15 1.2.1.1. KSHV latency-associated nuclear antigen (kLANA) ...... 18 1.2.1.2. kLANA internal repeat region ...... 19 1.2.2. Murine herpesvirus 68 (MHV-68) ...... 20 1.2.2.1. MHV-68 latency-associated nuclear antigen (mLANA) ...... 21

Chapter 2. Aim of the project ...... 23

Chapter 3. Materials and methods ...... 24 3.1. Plasmids ...... 24 3.2. Bacterial strains ...... 24 3.3. Generation of MHV-68 recombinant viruses ...... 25 3.3.1. Cloning procedures to engineering the virus kLANA ∆331-495 ...... 25 3.3.2. DNA plasmid isolation ...... 26 3.3.3. Mutagenesis ...... 27 3.4. Cell lines ...... 29 3.5. Viruses ...... 29 3.6. Reconstitution of MHV-68 virus ...... 30 3.6.1. Reconstitution of MHV-68 on BHK-21 ...... 30 3.6.2. Passage through NIH3T3-Cre cells to remove BAC sequences ...... 30 3.7. Production of working viral stocks ...... 30

5 3.8. Virus titration using suspension method (plaque assay) ...... 30 3.9. Protein expression analysis ...... 31 3.9.1. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) .... 31 3.9.2. Transfer of proteins to nitrocellulose membranes ...... 32 3.9.3. Western blot ...... 32 3.10. Multi-step growth curve ...... 33 3.11. Mice ...... 34 3.12. In vivo assays ...... 34 3.12.1. Ethics statement ...... 34 3.12.2. Infection of mice ...... 34 3.12.3. Single cell suspensions ...... 34 3.12.4. Infectious center assay ...... 34 3.12.5. Fluorescence-activated cell sorting (FACS) ...... 35 3.12.6. Limiting dilution analysis and real-time PCR of viral DNA positive cells ...... 35 3.13. Statistical analysis ...... 36

Chapter 4. Results ...... 37 4.1. Generation and characterization of a MHV-68 recombinant virus harboring a mutation on kLANA ...... 37 4.2. Viral titrations ...... 40 4.3. Expression of kLANA mutant proteins (Western blot) ...... 41 4.4. In vitro growth of the recombinant viruses ...... 43 4.5. In vivo assays ...... 44 4.5.1. Infectious center assay ...... 44 4.5.2. Fluorescence-activated cell sorting (FACS) ...... 45 4.5.3. Quantification of the frequency of viral DNA-positive total splenocytes ...... 47

Chapter 5. Discussion and future perspectives ...... 49

Chapter 6. References ...... 52

Chapter 7. Supplementary data ...... 57

6 Index of Figures

Figure 1 - Schematic representation of the multilayer organization of a herpesvirus...... 13

Figure 2 - Schematic representation of linearized KSHV genome...... 16

Figure 3 - Schematic representation of KSHV life cycle in infected cells and the different set

of genes expressed in each phase...... 17

Figure 4 - Schematic representation of kLANA protein...... 18

Figure 5 - Schematic diagram of KSHV LANA (kLANA) and MHV-68 LANA (mLANA)...... 22

Figure 6 - Schematic diagram of KSHV LANA and LANA mutant protein...... 23

Figure 7 - Schematic diagram of KSHV LANA and deletion mutants...... 38

Figure 8 - Restriction profiles of v-WT, v-kLANA and the recombinant genomes, in a non-yfp

and yfp background, respectively...... 39

Figure 9 - Detection of the expression of viral proteins in v-WT and chimeric viruses...... 42

Figure 10 - Growth curves of v-WT and chimeric viruses in both backgrounds...... 43

Figure 11 - Quantification of latent infection in spleen, 14 d.p.i., by ex vivo reactivation assay.

...... 45

Figure 12 - Flow cytometry analyses...... 47

Figure 13 - Quantification of viral DNA-positive cell in total splenocytes...... 48

Figure S1 - Schematic representation of the construction of a recombinant shuttle plasmid 59

7 Index of Tables

Table 1 – List of human herpesviruses...... 12

Table 2 - Primers used to amplify the desired fragment, with PstI restriction site in red and

codon for aa 330 of kLANA underlined...... 25

Table 3 - Composition of electrophoresis gels...... 32

Table 4 - Primary antibodies used in western blot...... 33

Table 5 - Secondary antibodies used in western blot...... 33

Table 6 - Viruses that were used to perform the in vitro and in vivo experiments...... 40

Table 7 - Viral stocks titers...... 41

Table 8 - Frequencies of MHV-68 latent infection in total splenocytes...... 47

8 Abbreviations

%GC Percentage of guanine and cytosine

AIDS Acquired immunodeficiency syndrome

BAC Bacterial artificial chromosome

BHK Baby hamster kidney

CPE Cytopathic effect

CWS Cell working stock

D.p.i. Days post-infection

DBD DNA binding domain

DMEM Dulbecco’s Modified Eagle Medium

DNA Deoxyribonucleic acid

EBNA-1 Epstein-Barr nuclear antigen-1

EBV Epstein-Barr virus

EDTA Ethylenediamine tetraacetic acid

FACS Fluorescence-activated cell sorting

GC Germinal center

GFP Green fluorescent protein

GMEM Glasgow Minimum Essential Medium

HCMV Human cytomegalovirus

HHV-6 Human herpesvirus 6

HHV-7 Human herpesvirus 7

HHV-8 Human herpesvirus 8

HIV Human Immunodeficiency Virus

HSV-1 Herpes simplex virus type 1

HSV-2 Herpes simplex virus type 2

ICTV International Committee on Taxonomy of Viruses

IE Immediate early

kLANA KSHV latency-associated nuclear antigen

9 KS Kaposi’s sarcoma

KSHV Kaposi’s sarcoma-associated herpesvirus

LANA Latency-associated nuclear antigen

LB Luria Bertani

LUR Long unique region

LZ Leucine-zipper

MCD Multicentric Castleman’s disease

MHV-68 Murine herpesvirus 68

mLANA MHV-68 latency-associated nuclear antigen

MOI Multiplicity of infection

miRNA MicroRNA

ORF Open reading frame

PBS Phosphate buffered saline

PBS-T PBS-Tween 20

PCR Polymerase chain reaction

PEL Primary effusion lymphoma

PFU Plaque forming unit

RNA Ribonucleic acid

rpm Revolutions per minute

RT Room temperature

RTA Replication and transcription activator

SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

SOB Super Optimal Broth

SOC SOB plus glucose

TEMED Tetramethylethylenediamine

TGS Tris-glycine-SDS

TR Terminal repeat

Viral Fas-associated death domain-like interleukin-1β-converting enzyme- v-FLIP inhibitory protein

10 VZV Varicella-zoster virus

WSM Working stock media

WT Wild type

11 Chapter 1. Introduction

According to the International Committee on Taxonomy of Viruses (ICTV), this family belongs to the order Herpesvirales together with the families Alloherpesviridae and Malacoherpesviridae. Herpesviridae is an ancient family of ubiquitous viruses that are widely spread across the nature (Davison, 2002) and it is the largest family of its order, containing over 120 herpesviruses. It includes reptile, bird and mammalian herpesviruses, where 8 of the 120 herpesviruses have been isolated from humans (Table 1) (Davison et al., 2009). Alloherpesviridae family includes the amphibian and bony fish herpesviruses, while Malacoherpesviridae family is only constituted by a single invertebrate virus (Davison et al., 2009).

Table 1 – List of human herpesviruses. This table lists the human herpesviruses, including some of their characteristics, like their common name, specie name according to ICTV, their genome size and subfamily (adapted from Davison, 2007).

ICTVa Genome size Abbreviation Subfamily Virus species name (kb)

Human Herpes simplex virus HSV-1 α 152 alphaherpesvirus 1 type 1

Human Herpes simplex vírus HSV-2 α 154 alphaherpesvirus 2 type 2 Human b VZV α ~125 Varicella-zoster virus alphaherpesvirus 3

Human Human HCMV β 229/236 betaherpesvirus 5 cytomegalovirusb Human b HHV-6 β 159/162 Human herpesvirus 6 betaherpesvirus 6 Human b HHV-7 β 145/153 Human herpesvirus 7 betaherpesvirus 7 Human b EBV γ 172 Epstein-Barr virus gammaherpesvirus 4

Human Kaposi’s sarcoma- KSHV γ ~140 gammaherpesvirus 8 associated herpesvirusb aInternational Committee on Taxonomy of Viruses. bMore than one strain has been sequenced.

12 Herpesviruses have co-evolved with their hosts over long periods, being well adapted to them (Davison, 2002). Molecular phylogenetic studies and host-specific occurrence suggest that herpesviruses speciation and host speciation occurred approximately at the same time. As result of this adaptation, it seems that each herpesvirus has a very narrow host range, usually limited to a single species. When there is an equilibrium between the virus and immune system of the host, in the majority of the cases, there is an absence of pathogenicity. Only when this equilibrium is disturbed, by factors like stress or diseases, the pathogenicity of the virus increases, and therefore can cause a disease in the host (Davison, 2002; Barton et al., 2011).

Herpesviridae is a very heterogeneous family and the viruses from this family can be identified among other viruses by their virion structure (Figure 1). This family is constituted by enveloped viruses that contain large linear double-stranded DNA genomes (ranging from 120 to 250 kb) with vast coding capacity (Davison, 2002; Barton et al, 2011). The viral genome is encased in a highly ordered icosahedral-shape capsid (125-130 nm in diameter), which is surrounded by an amorphous structure with variable size containing several viral proteins, called the tegument (Figure 1). The virion is enclosed by a lipid bi-layer envelope, derived from the host cell and contain a large number of viral glycoproteins on its surface, which are responsible for viral attachment and entry in host cells (Arvin et al., 2007; Hughes et al., 2002).

Figure 1 - Schematic representation of the multilayer organization of a herpesvirus. Virion structure of these viruses is constituted by: a core containing the double-stranded DNA, which is enclosed by an icosahedral capsid. The capsid is surrounded by the tegument, which has several viral proteins, and it is enveloped by a lipid bi-layer with glycoproteins on its surface (adapted from Hughes et al., 2002).

Herpesviruses are also characterized by having a life cycle with two distinct phases: a lytic phase and a latent phase, and both contribute to the pathogenicity of these viruses. One of the most remarkable biological characteristic of the herpesviruses is their ability to establish latency, causing lifelong infections in the host. When the virus is transmitted to a new host, in the site of infection (usually the epithelium of a mucosal surface) the virus starts to establish an acute infection (lytic phase)

13 characterized by an intense replication of the viral genome, leading to the production of infectious virions and, consequently, to the death of the infected cell. This first infection is rapidly resolved by host immune system (Stoopler, 2005; Wu et al., 2010).

The lytic infection is followed by a latent infection, which is maintained in specific types of cells. During latency, the viral gene expression is limited and there is no virion production. Periodically, in response to specific stimulus, the latent virus can reactivate, starting to express the viral genes associated with replication, leading to the production of new viruses that can be transmitted to new hosts (Whitley, 1996; Wu et al., 2010; Barton et al., 2011). The majority of herpesviruses infections are asymptomatic, however if the host has an immunocompromised system, the effects can be devastating (Barton et al., 2011; Whitley, 1996; Wu et al., 2010).

According to the ICTV this family is divided in three subfamilies: Alphaherpesvirinae, Betaherpesvirinae and Gammaherpesvirinae (Table 1). Among these subfamilies, there is a difference in the place where the virus establish latency. In Alphaherpesvirinae, the viruses establish latency mainly in neurons, in Betaherpesvirinae, they are found latent mainly in myeloid cell lineage and in Gammaherpesvirinae latency is establish predominantly in lymphocytes (Wu et al., 2010; King et al., 2011). The viruses from the three subfamilies can infect humans and cause infections (Stoopler, 2005; Wu et al., 2010).

Members of the Gammaherpesvirinae subfamily are widespread in nature. They are divided into four genera, according to ICTV, based on their DNA homology and genomic organization: Lymphocryptovirus (also known as gamma-1-herpesviruses), Macavirus, Percavirus and Rhadinovirus (also known as gamma-2-herpesviruses). From Gammaherpesvirinae, two viruses have been identified in humans: Epstein-Barr virus (EBV), which is a Lymphocryptovirus, and Kaposi’s sarcoma-associated herpesvirus (KSHV), which is a Rhadinovirus (Barton et al., 2011).

The study of human gammaherpesviruses EBV and KSHV is limited, because they have a very narrow host range. For this reason, the development of an animal model system in which the pathogenicity of the human gammaherpesviruses infection can be studied it is of considerable interest.

EBV and KSHV are lymphotropic viruses that are associated with the development of human cancers. These viruses establish latency preferentially in B-lymphocytes (Damania, 2004; Barton et al., 2011) and have the ability to drive the proliferation of latently infected lymphocytes, in order to establish and maintain a persistent viral reservoir of latent genomes, to colonize the host (Stevenson, 2004).

14

Kaposi’s sarcoma-associated herpesvirus (or human herpesvirus 8) was first identified in 1994 by Yuan Chang and his colleagues. They found unique sequences of herpesvirus in Kaposi sarcoma (KS) biopsies from AIDS patients (Chang et al., 1994). KSHV is the etiological agent of Kaposi’s sarcoma (KS) and consists on a proliferation of KSHV infected endothelial cells (Coscoy, 2007). These tumors are frequently found in the dermis, forming lesions that have red, brown or purple color, but can also occur in lungs, liver and intestines (Moore and Chang, 2003).

KSHV is also the etiological agent of primary effusion lymphoma (PEL) and multicentric Castleman’s disease (MCD) (Wu et al., 2010), a B-cell lymphoma and a B-cell lymphoproliferative disorder, respectively (Moore and Chang, 2003). The majority of primary KSHV infections are asymptomatic and result in a lifelong latency in the host (Dittmer and Damania, 2016). Recently, KSHV was associated with a new clinical condition, KSHV-inflammatory cytokine syndrome (KICS) (Polizzotto et al., 2016; Uldrick et al., 2010).

KSHV is one of the most common malignancies seen in patients with acquired immunodeficiency syndrome (AIDS). This type of malignancy caused by a virus develop with more frequency in individuals whose immune system is compromised, such as by infection with human immunodeficiency virus (HIV) or by using immunosuppressive drugs, for example, after an organ transplant (Wu et al., 2010).

Among human population, the prevalence of KSHV infection is less than 10%, in northern Europe, Asia and America, but in the Mediterranean regions, this prevalence increases to 30% and between 30-60% in most of sub-Saharan Africa (Moore and Chang, 2003; Uppal et al., 2014). The transmission modes of KSHV are still not completely understood, but it seems that KSHV can be transmitted by both sexual and non-sexual modes. Between children, it seems that the transmission mediated by saliva is the most common mode of KSHV transmission, while between adults it seems that KSHV can be transmitted through organ transplant, blood and sexual fluids (Bagni and Whitby, 2009; Uppal et al., 2014).

The KSHV genome consists of a long unique region (LUR) with approximately 145 kb (Figure 2). LUR is flanked by a region of 20-35 kb of non-coding terminal repeats (TRs), which comprise around 20% of the genome. These viral terminal repeats are constituted by 801 bp of repeated sequences that have a high G+C content (approximately of 80%) (Juillard et al., 2016; Moore & Chang, 2003), although the number of terminal repeats vary among the KSHV isolates (Wen and Damania, 2010). More than 90 genes have been identified in the LUR (Cai et al., 2010), but nearly 25 of these genes encode new proteins that were not found in other human herpesviruses (Ablashi et al., 2002).

15

Figure 2 - Schematic representation of linearized KSHV genome. KSHV genome consists in a long unique coding region with 145 kb. The coding regions contains over 90 ORFs, which are flanked with terminal repeat units. The protein encoded by each gene is labeled in the bottom, from left to right. The prefix K means that the ORF is unique to the KSHV. Arrows indicate transcription direction (adapted from Cai et al., 2010).

KSHV exhibits a life cycle with two distinct phases of infection, like the other gammaherpesviruses: lytic replication and latency, both with different patterns of gene expression and both contribute to the pathogenicity of the virus (Figure 3) (Uppal et al., 2014; Wu et al., 2010). Both latent and lytic phases are important for the long-term persistence of the virus in host cells (Purushothaman et al., 2016).

After infection, KSHV genome enters in the nucleus of the host cell, where it circularizes by fusing at the TR elements. In the nucleus, viruses can either enter in a latent phase, that allow them to persist in an immunological silent mode or enter in a lytic phase, which enables the production and transmission of virions to new hosts (Coscoy, 2007; Purushothaman et al., 2016). During latency, the limited number of genes expressed by KSHV facilitate the establishment of lifelong latency and its survival against the mechanisms of the host immune system (Purushothaman et al., 2016). This phase represents the main strategy used by this virus to escape host immune system, while maintaining its genome as a non-integrated mini-chromosome, with a circular form (episome) in the nucleus of infected cells (Wen and Damania, 2010; Wu et al., 2010). During the lytic phase, the viral episome gradually relaxes its chromatin structures, linearizing. Then, it starts to replicate its genome and viral genes are expressed in a regulated cascade mode. First, occurs the expression of the immediate early (IE) genes, which triggers the expression of early (E) genes, needed for viral DNA replication followed by the expression of late (L) genes, leading to virus production. This viral production consequently leads to cell death and transmission of the viruses (Delgado et al., 2012; Purushothaman et al., 2016; Uppal et al., 2014). IE genes are important, because they regulate the succeeding cascade events. The replication and transcription activator (RTA) is an IE lytic gene, encoded by ORF 50, that has been shown to be required and sufficient for initiating the lytic replication cycle, being the first gene to be expressed during the cascade events. On the other hand, it is thought that in order to establish latency, the expression of

16 RTA must be repressed, because the expression of this gene was found restricted in latently infected cells. Both early and late genes facilitate the replication of viral genome, viral assembly and the virus exit from the cells (Purushothaman et al., 2016; Wen and Damania, 2010).

Figure 3 - Schematic representation of KSHV life cycle in infected cells and the different set of genes expressed in each phase. KSHV life cycle is divided into two distinct phases of infection: a short lytic phase and a latent phase, both being important for the long-term persistence of the virus in the host (adapted from Purushothaman et al., 2016).

The major KSHV latency locus include viral genes, such as: the latency-associated nuclear antigen (LANA), encoded by ORF73; the viral cyclin D homolog (v-cyclin), encoded by ORF72; the viral Fas-associated death domain-like interleukin-1ß-converting enzyme-inhibitory protein (v-FLIP), encoded by ORFK13; Kaposin family (Kaposin A, B and C), encoded by ORFK12; and twelve microRNAs (miRNAs) (Figure 3) (Moore and Chang, 2003; Sin and Dittmer, 2013). LANA, v-cyclin and v-FLIP genes are also known as KSHV oncogenic cluster, because all these three regulatory proteins can interfere with several pathways that regulate cell growth and apoptosis, and they also can inhibit tumor suppressor (Dittmer, 1998; Moore and Chang, 2003). This inhibition is necessary, since the viral episome is replicated at the same time as the host cell. For this reason, the G1/S checkpoint arrest and apoptotic programs have to be inhibited so the cell cannot sense the latent viral infection (Moore and Chang, 2003).

It was discovered that KSHV and other gammaherpesviruses also encode several miRNAs. Although their exact function has not been completely determined, it is known that they do not encode proteins (Boss et al., 2009; Coscoy, 2007). These miRNAs are clustered together in the KSHV latency

17 locus, and some of them are expressed in latently infected cells (Habison et al., 2017). As they do not encode proteins, they cannot be target by the effectors of the host immune system. Therefore, the use of these miRNAs might represent a new and smart mechanism to evade the host immune system used by KSHV and other gammaherpesviruses (Coscoy, 2007).

Latency-associated nuclear antigen (LANA) is a multifunctional protein with 1162 amino acids which is encoded by ORF 73 and it is the major latent protein expressed through the viral life cycle (Purushothaman et al., 2016). This protein is constituted by several parts: the N-terminal domain that contains a proline-rich region (P), a set of internal repeat sequences - in which there is an aspartate- and glutamate-rich region (DE), a glutamine-rich region (Q), a predicted leucine zipper region (LZ), a glutamate- and glutamine-rich region (EQE) - and the C-terminal DNA binding domain, which associates with various host cell proteins and chromosomes (Figure 4).

Figure 4 - Schematic representation of kLANA protein. The different parts of the protein are indicated, in bold, namely proline-rich region (P) and the repetitive regions (DE, Q, L, EQE). Amino acids 5-13 mediate chromosome association and amino acids 996-1139 comprise TR DNA binding. The internal region of kLANA comprising amino acids 262-320 is required for episomal replication (adapted from Ponnusamy et al., 2015).

There are two key components to episome persistence in proliferating cells and kLANA is responsible for both: the episomal DNA needs to replicate with each cell division and episomes have to segregate to daughter cells nuclei following mitosis to avoid their destruction. This episome persistence occurs, because kLANA mediates the replication and tethers KSHV episome to the host mitotic chromosomes during the cell division, to ensure the correct segregation of the viral genome (Juillard et al., 2016). Both the N- and C-terminal regions of kLANA are important to maintain the KSHV genome as an episome, however they are not sufficient for an efficient episome maintenance (De Leon Vazquez and Kaye, 2011). Previous studies where the N- and C- terminal regions were fused showed that mutants retained the ability to bind to mitotic chromosomes, however these mutants were highly

18 deficient in episome maintenance, showing that the internal region of LANA has an important role for episome maintenance (De Leon Vazquez and Kaye, 2011; De Leon Vazquez et al., 2013).

The internal acidic repeat region of this protein still remains uncharacterized, although some authors defend that the internal repeats give stability to the kLANA, inhibit proteosomal degradation and retard protein synthesis (Kwun et al., 2007). Previous studies, also showed that the deletion of all internal kLANA sequences reduced the efficiency of DNA replication (De Leon Vazquez et al., 2013).

As described above, kLANA can also modulate host cell gene expression by interacting with cellular and viral proteins, such as tumor suppressors - p53, transcription factors, proteins that regulate the chromatin and histones, in order to establish a successful state of latency, repressing KSHV reactivation and escape to host immune surveillance. kLANA can also associates with some viral promoters to regulate transcription of viral genes. For instance, kLANA can negatively regulates the transcription of RTA and thus repress KSHV reactivation. kLANA inhibits RTA expression, by repressing the transcriptional activity of RTA/ORF50 promoter (Barbera, 2006; Purushothaman et al., 2016). kLANA can also regulates its expression by inducing transcription of kLANA promoter (Jeong et al., 2004). It has been suggested that these interactions affect the viral persistence, cellular transcription and growth transformation.

In summary, kLANA is a multifunctional protein with an important role in establish KSHV latency, viral episome replication and maintenance, regulation of host and viral transcription, and control of cell growth and proliferation. To accomplish all of these functions, kLANA interacts with a series of host cellular proteins, which are involved in different cellular processes.

The less characterized part of kLANA is the internal repeat region, maybe due to the difficulties in cloning this highly repetitious internal repeat region. This region is constituted by three segments: an aspartate and glutamate (DE) region, from amino acid 330 to 430, a glutamine (Q) region, from amino acid 440 to 756 and a leucine (L), valine (V), glutamate and glutamine (EQE) spaced in a leucine-zipper- like pattern (LZ), from amino acids 760 to 931 (Figure 4) (Alkharsah and Schulz, 2012). Several studies have been made in the last years, in order to better understand the role of this internal region. In 2012, Alkharsah and Schulz constructed a recombinant virus in which all the internal repeat region was deleted and found that this leaded to a loss of persistence of viral DNA, in vitro, although its ability to replicate viral DNA was maintained, meaning that this region has a crucial role in supporting the persistence of latent viral genomes (Alkharsah and Schulz, 2012). A similar result was showed by De Leon Vasquez and Kaye in 2011, where they fused N- and C- terminal of LANA deleting all the internal repeat region and observed that the mutants were highly deficient in episome maintenance (De Leon Vasquez and Kaye, 2011).

19 In 2007, Kwun and colleagues suggested that the central region of LANA, particularly, Q and EQE regions play an important role in inhibiting proteosomal degradation and retarding protein synthesis, having a similar function as the glycine-alanine repeats in Epstein-Barr nuclear antigen (EBNA-1) (Kwun et al., 2007). In 2013, De Leon Vasquez and colleagues engineered a series of LANA mutants with different deletions. They showed that when the amino acids residues 465 to 929 were deleted the levels of DNA replication capacity were 56.7% of the WT kLANA and when the amino acids residues 332 to 929 were deleted the levels of DNA replication capacity were only 36.7% of the WT kLANA. These results suggest that the internal repeat region between amino acids 331 to 465 may have an important contribution to episome replication in addition to the ones already identified (De Leon Vasquez et al., 2013).

There is a need of information regarding the internal repeat encompassing amino acids 330 to 465 (DE internal repeat). In literature, there is almost no information about this region and studies in vitro or in vivo using mutants lacking this region need to be done to better understand the exact function of this region.

A major limitation to the study of the human gammaherpesviruses is the absence of a small animal model, which allows the investigation of basic aspects of viral pathogenesis, due to the narrow host tropism exhibited by KSHV (Barton et al., 2001; Simas and Efstathiou, 1998). The identification of murine herpesvirus 68 (MHV-68), enabled the possibility of development of a mouse model to study gammaherpesvirus pathogenesis (Simas and Efstathiou, 1998).

MHV-68 was originally isolated from the bank vole (Myodes glareolus) and yellow-necked field mice (Apodemus flavicollis) in Slovakia (Blaskovic et al., 1980) and it is also classified as a gamma-2- herpesvirus, like KSHV. Its viral genome consists of 118 kb of unique DNA sequence with 46% of G+C content, flanked by sequences of terminal repeats (TR) (Efstathiou et al., 1990). It is estimated that the viral genome has approximately 80 proteins encoded by open reading frames (ORFs), where at least 63 are homologous to KSHV (Simas and Efstathiou, 1998). MHV-68 genome also has a number of genes that are unique to this virus, located at the left region of its genome.

MHV-68 also infects laboratory mice (Mus musculus), where it establishes a lifelong infection. Mice that are infected by intranasal inoculation develop a lytic infection, characterized by a severe virus replication in the lungs (Sunil-Chandra et al., 1992; Habison et al., 2017). This lytic infection is resolved by the immune system of the host between 10 to 12 days post-infection. Then, the viruses enter in the lymphoid tissue, in which they are transported until they reach the spleen, where they will establish latency (Frederico et al., 2014). In the spleen, viruses drive the proliferation of germinal center (GC) B cells (Habison et al., 2017) and gain access to memory B cell, the principal long-term reservoir of this

20 viruses (Flãno et al., 2002). The peak of latency occurs approximately 14 days post-infection and it is characterized by the maximal numbers of splenocytes that are latently infected. After this peak, the numbers of latently infected cells start to decrease until it reaches a steady-state level (Simas and Efstathiou, 1998).

Like KSHV, MHV-68 encodes a latency-associated nuclear antigen (mLANA), which is encoded by ORF73. mLANA is homologous in sequence and function to kLANA, but has only 314 amino acids in length, which is considerably smaller than kLANA (Figure 5). This difference in size is, mostly due to the absence of the internal acidic repeat and glutamine-rich sequence in mLANA. Analogously to kLANA, mLANA N-terminal region contains a proline-rich region, which interacts with some cellular proteins of the host. The C-terminal region of mLANA has amino acid homology with kLANA DNA binding domain (DBD) (Correia et al., 2013; Grundhoff and Ganem, 2003).

Like kLANA, mLANA also acts on viral TR to mediate episome persistence (Habinson et al., 2017), by tethering them to host mitotic chromossomes, allowing the segregation of episome to the daughter cells (Habinson et al., 2012).

The MHV-68 genome, like the other gammaherpesviruses, persists as a multi-copy, circularized, non-integrated episome in latently infected cells. Unlike kLANA, the functions of mLANA can be directly studied in a mouse model of infection. Previous studies demonstrated that MHV-68 virus that do not express mLANA or have mutation in DNA-binding domain fail in establish viable latency, in vivo, thereby demonstrating that mLANA has a critical role in the establishment of latency and in the persistence of the virus in the host (Correia et al., 2013).

In summary, mLANA exhibits functional similarity to kLANA, being essential to viral episome persistence and regulating cellular transcription. However, the viral genome manipulation and consequent study in MHV-68 is easier. Thereby, and given the similar characteristic between mLANA and kLANA, Habison et al. engineered a chimeric virus in which mLANA was replaced by kLANA and its 5’ untranslated region (UTR) in the MHV-68 genome. They constructed this chimeric virus (v-kLANA) in order to study kLANA pathogenesis in an in vivo infection. They showed that the levels of latency in vivo of v-kLANA were lower compared to the v-WT, but both v-WT and v-kLANA infected cells had similar genome copy numbers. Thus, kLANA rescued mLANA deficient MHV-68, allowing the chimeric virus to establish a latent infection in vivo (Habison et al., 2017). This finding shows that this is a viable model to study KSHV pathogenesis in vivo by infect small animal models with v-kLANA.

21

Figure 5 - Schematic diagram of KSHV LANA (kLANA) and MHV-68 LANA (mLANA). The proline-rich region (P) and C-terminal region harboring the DNA binding domain (DBD) are shared by both proteins. DBD also mediates dimerization. The internal region of kLANA, which comprises aspartate-glutamate (DE), glutamine (Q) and glutamine-glutamate (EQE) and also a putative leucine zipper (LZ), is absent in mLANA (adapted from Ponnusamy et al., 2015).

Due to the oncogenic potential associated with KSHV, there is a strong need for a vaccine development to prevent or reduce the establishment of persistent infection, and thus lower the risk of developing tumors. The similarity of MHV-68 with KSHV, and the establishment of MHV-68 as an in vivo model of gammaherpesvirus infection has allowed the exploration of the principle vaccination strategies. Based on previous studies with MHV-68, one of the strategies for the development of the vaccine could be the use of attenuated KSHV viruses unable to establish latency (Wu et al., 2010).

22 Chapter 2. Aim of the project

Kaposi´s sarcoma-associated herpesvirus (KSHV) is a gamma-2 herpesvirus which is capable to establish latency. During latent phase, one of the proteins expressed is the latency-associated nuclear antigen (LANA). Since KSHV malignancies are associated with latent infection, the study of the mechanisms underlying LANA functions is essential. The researches on KSHV are limited, because it has a narrow host range, so an animal model of infection is crucial for this investigation. The murid herpesvirus 68 (MHV-68) is a natural rodent pathogen and it is genetically related to KSHV. MHV-68 also encodes a LANA (mLANA) which is homologous in function to KSHV LANA (kLANA). Recently, a chimeric virus in which mLANA was replaced by kLANA in MHV-68 genome was generated and it was demonstrated that kLANA is able to support MHV-68 latency, after an infection.

The aim of this project was to assess the impact of kLANA DE internal repeat in the establishment of latency after an infection, using an animal mouse model. To do this assessment, a deletion between amino acids 331-495 of kLANA (v-kLANA∆331-495) was performed (Figure 6). The mutant kLANA was introduced in MHV-68 genome, replacing its original protein (mLANA) and thus creating a chimeric virus where it was possible to better understand the importance of kLANA DE internal repeat in vivo.

Ultimately, this study will provide more information about an essential KSHV latency protein, specially about its internal repeat region. Thus, the results of this study may help in the investigation of vaccine development, by using the information of this chimeric virus to try to fight KSHV infection.

Figure 6 - Schematic diagram of KSHV LANA and LANA mutant protein. The different parts of the protein are indicated, in bold, namely proline-rich region (P), the aspartate and glutamate (DE), the glutamine (Q), the glutamate and glutamine (EQE), the leucine zipper (LZ) regions and the DNA binding domain (DBD). Amino acid residues are indicated below the different domains. The deleted protein constructed is represented by kLANA ∆331-495 (adapted from De Leon Vazquez et al., 2013).

23 Chapter 3. Materials and methods

pSP72_PCR1_4 was constructed by Dr Marta Miranda (Habison et al., 2017). It contains DNA encoding the full length kLANA (KSHV genome coordinates 123808-127886, U75698) and its 5’ untranslated region (UTR) flanked at the left side by the MHV-68 upper flank (genome coordinates 104710-105092, U97553) (Figure S1A, Supplementary data). pSP72_PCR1_5 was constructed by Dr Marta Miranda (Habison et al., 2017). It contains DNA encoding the full length kLANA and its 5’ untranslated region (KSHV genome coordinates 123808- 127886, U75698), flanked by MHV-68 sequences. PCR5 fragment or MHV-68 lower flank contains the restriction site of BglII. pSP72 kLANA Δ331-495 was generated in this project. It contains DNA encoding kLANA with amino acid residues 331-495 deleted and flanked by MHV-68 sequences. Deleted region encompasses the DE internal repeat (Figure S1B, Supplementary data). The BamHI-G shuttle plasmid was constructed by Dr Sofia Marques. It contains the MHV-68 genomic BamHI-G fragment (genomic coordinates 101654-106903, U97553) cloned into the pST76K- SR plasmid (shuttle vector). The BamHI-G fragment contains the mLANA coding sequence (genome coordinates 103927-104868). kLANA Δ331-495 BamHI-G recombinant shuttle was generated in this project, where mLANA was replaced by kLANA coding sequence in the MHV-68 bacterial artificial chromosome (BAC). It also contains the 5’UTR region and the MHV-68 flanks (upper and lower flanks) (Figure S1C, Supplementary data).

E. coli XL10-Gold Ultracompetent Cells (Agilent Technologies) were used to transform the ligation reactions between inserts and vectors, during the cloning procedures. E. coli DH10B harboring MHV-68 cloned in a BAC were kindly provided by Dr Heiko Adler and Dr Ulrich Koszinowski and were used to produce recombinant viruses, during the mutagenesis procedure (Adler et al., 2000). E. coli DH10B containing yfp MHV-68 BAC were kindly provided by Dr Samuel Speck and were used for mutagenesis of yfp MHV-68 recombinant viruses (Collins et al., 2009).

24

3.3.1.1. Polymerase Chain Reaction (PCR) to introduce the deletion into pSP72_PCR1_4 The fragment between the PstI restriction site in the 5’ UTR of kLANA and codon for amino acid 330 of kLANA was amplified by PCR, using pSP72_PCR1_4 as template and primers depicted in Table 2. The primers designed were specific for this procedure, because both had a PstI restriction site. The generated fragment had 1139 bp. PCR was performed using a high fidelity Pfu DNA Polymerase (Promega). PCR reactions were prepared in a total volume of 100µl (made up in MilliQ sterile water) and consisted of 30µM of each primer, Pfu buffer (Promega), 200 µM of each deoxynucleotide (dNTP), 6 U of Pfu DNA polymerase (Promega) and 20ng of DNA template. DNA was amplified on a MyCicler thermal cycler (BioRad), under the following condition parameters: an initial melting step of 95ºC for 2 min followed by 25 cycles of amplification, starting with a denaturation step at 94ºC for 45 sec, an annealing step at 63ºC for 45 sec and extension at 72ºC for 2 min and 20 sec. A final step of strand extension was performed at 72ºC for 5 min. PCR product was analyzed by gel electrophoresis to compare with the expected size.

Table 2 - Primers used to amplify the desired fragment, with PstI restriction site in red and codon for aa 330 of kLANA underlined.

Oligonucleotides Sequence (5’-3’) Upper primer (IMMAF3) GTACTGCAGCCTGCTACTGTG Lower primer (IMMAF2) AAACTGCAGATCCTTATTGTCATTGTCATC

3.3.1.2. Cloning the PCR product into construct plasmid PCR product obtained in section 3.3.1.1 was cut from the gel, purified from agarose using QIAquick gel extraction kit (Qiagen), according to manufacturer’s instructions and digested with PstI restriction endonuclease. pSP72_PCR1_4 (construct plasmid) was also digested with PstI restriction endonuclease and analyzed by gel electrophoresis to assess if the plasmid had the expected size. The desired fragment was cut from the gel and purified from agarose using the same extraction kit as above. Both insert and vector were ligated as described in section 3.3.1.5., and the resulting plasmid was pSP72_PCR1_4 kLANA ∆331-495. The ligations were transformed into XL10-Gold Ultracompetent cells (section 3.3.1.6). Colony PCR was performed to verified the orientation of the insert in the plasmid. DNA was isolated from colonies with the insert in the correct orientation, by plasmid miniprep (section 3.3.2.). PCR desired region was sequenced at GATC Biotech according to the Sanger method. DNA sequences were analyzed and compared with pSP72_PCR1_4 sequence.

25 3.3.1.3. Subcloning PCR5 into construct plasmid pSP72_PCR1_5 was digested with XhoI and HindIII restriction endonucleases to release the fragment between them. This purified fragment was subcloned into pSP72_PCR1_4 kLANA ∆331-495 obtained in section 3.3.1.2, that had been also digested with the same two restriction endonucleases. The resulting plasmid was pSP72_kLANA ∆331-495. The ligations were transformed as explained above, DNA was isolated from colonies by plasmid maxiprep to increase the DNA concentration and plasmid structure was screened by restriction analysis with the appropriate endonucleases.

3.3.1.4. Subcloning of insert into BamHI-G shuttle vector pSP72_kLANA ∆331-495 obtained in section 3.3.1.3. and BamHI-G shuttle were both digested with BglII restriction endonuclease, analyzed by gel electrophoresis and the correspondent bands were cut from the gel. Gel purified insert and vector were ligated and the ligations were transformed as explained above. The resulting plasmid was kLANA ∆331-495 BamHI-G recombinant shuttle. DNA was isolated from colonies by plasmid miniprep and plasmid structure was screened by restriction analysis with the appropriate endonucleases. After the screen analysis, glycerol stocks were made and stored at -80ºC.

3.3.1.5. DNA ligation Purified inserts and vectors were ligated using T4 DNA ligase (Thermo Scientific). Approximately 50 ng of vector DNA were ligated with 1-3 fold molar excess of insert in 20µl reactions (made up in sterile MilliQ water), containing ligase buffer and 2 U of T4 ligase. Sticky-end ligations were performed at 22ºC during 1h and 14ºC during 8h.

3.3.1.6. Transformation of ultracompetent cells E. coli XL10-Gold ultracompetent cells were transformed by the heat shock method. Cells were gently thawed on ice and 45µl were distributed into 15mL falcons. 2µl of ß-Mercaptoethanol provided with the kit were added to each falcon and cells were incubated on ice for 10 min. 1.5µl of each ligation reaction were added and cells were incubated for 30 min on ice. Cells were heat shocked for 30 sec at 42ºC and subsequently chilled on ice for 2 min. 500µl of Super Optimal Broth (SOC) or Luria Bertani (LB) medium were added and cells were incubated for 1h at 37ºC or 30ºC for shuttle vector, with agitation of 230 rpm. Cells were then spread on LB agar plates containing the appropriated antibiotic and incubate overnight at 37ºC or 30ªC.

Plasmid DNA was isolated from plasmid-containing E. coli strain grown in LB broth containing the appropriate antibiotic(s), using an alkaline lysis method. This method was modified according to the scale of the preparation and the copy number of the plasmid. The antibiotics used were: ampicillin at 100µg/mL, kanamycin at 30µg/mL and/or chloramphenicol at 15-30µg/mL.

26 Small scale plasmid preparation For small scale plasmid preparations or plasmid minipreps, 3-10 mL of LB broth cultures containing the appropriate antibiotic were inoculated with a single bacterial colony and incubated overnight with shaking at 37ºC or 30ºC for shuttle plasmid. Cultures were centrifuged for 10 min at 4000 rpm. Plasmid DNA was obtained using the Wizard Plus SV Minipreps DNA Purification System (Promega), by column purification of DNA prepared by alkaline lysis, according to manufacturer’s instructions. DNA was eluted in 70µl of nuclease free water and stored at -20ºC.

Large scale plasmid preparation For large scale plasmid preparations or plasmid maxipreps, cultures of plasmid-containing bacteria were prepared by inoculation of 200 mL of LB broth cultures containing the appropriate antibiotic from single colonies and incubated overnight with shaking at 37ºC. Bacterial cultures were pelleted by centrifugation at 6000 rpm for 10 min at 4ºC and DNA was purified using the Plasmid Purification MAXI Kit (JETSTAR), following manufacturer’s instructions. DNA was resuspended in 100- 400µl of MilliQ and stored at -20ºC.

3.3.2.1. Quantification of nucleic acids DNA from maxipreps and minipreps was quantified by ultraviolet spectrophotometry at 260nm by a Nanodrop ND-1000 spectrophotometer.

3.3.2.2. Restriction endonuclease digestion Restriction endonuclease digestion of plasmids or gel purified PCR products was used either to prepare linear or insert DNA. To screen DNA for a desired digestion profile or for the presence of the inserts cloned in vectors. Restriction enzyme assays were performed using the appropriated restriction endonuclease and correspondent reaction buffer, according to manufacturer’s instructions. Digestion of plasmid or PCR products for subsequent ligation and cloning was performed overnight at 37ºC, with 1-5 µg of DNA and 1-20 U of restriction endonuclease in a reaction volume of 50-60µl. Multiple digestions of the same DNA were performed when possible, using the same buffer.

In order to generate MHV-68 recombinant viruses, we used a mutagenesis procedure in E. coli DH10B, according to the Two-Step-Replacement Strategy, described by O’ Connor et al.. First, the BamHI-G shuttle containing the kLANA with the desired mutations was transformed into E. coli DH10B competent bacteria harboring a BAC with the entire WT MHV-68 genome cloned, using the heat shock method. Bacteria were spread in LB plates containing the selection marker of shuttle plasmid (kanamycin) and the selection marker of BAC (chloramphenicol) and incubated 1-2 days at 30ºC. In this step a recombination event between homologous MHV-68 sequences in the BamHI-G recombinant shuttle and MHV-68 BAC, mediated by protein RecA interaction. This protein is encoded

27 by the shuttle plasmid and it is responsible for the complete integration of BamHI-G shuttle plasmid into BAC genome, forming the co-integrates. To select these co-integrates, the bacteria colonies were subsequently spread in LB plates containing kanamycin/chloramphenicol and incubated overnight at 43ºC. This step was repeated to ensure that the co-integrates were picked. Afterwards, the largest colonies were picked and plated only on LB agar with chloramphenicol and incubated 1-2 days at 30ºC. In this step, the co-integrates can resolve themselves by homologous recombination to the expected BAC mutant state. In order to isolate the resolved clones with BAC mutants, bacteria were spread into LB agar with chloramphenicol and 5% of sucrose, which is a counter selection against SacB gene encoded by BamHI-G shuttle plasmid, and incubated 1-2 days at 30ºC. After, colonies were picked once and plated in parallel first in kanamycin and then in chloramphenicol LB agar, and incubated overnight at 37ºC. Sucrose-resistance and kanamycin-sensitivity are two independent indications that the co- integrates are well resolved. To identify MHV-68 BAC recombinants containing kLANA coding sequences, a colony PCR was performed using primers that hybridize to the 3’end of kLANA, encoding the DBD domain: upper primer - 1019 F (5’-GTCCCTTACAGACAGATAGATGATTG-3’) and lower primer – 1150 R (5’-AAAAAGCTTTTATTCCCCTGGCTGGGTTAATG-3’). PCR positive clones were grown in LB medium with chloramphenicol and minipreps were prepared and characterized by restriction profile analysis using EcoRI and BamHI to assess the integrity of the genome and confirm kLANA desired deletions. Clones with the expected profile were grown and BAC maxipreps were prepared.

3.3.3.1. BAC DNA preps The colonies that were positive in the colony PCR were inoculated in 10mL of LB broth with chloramphenicol and grown overnight at 37ºC with shaking, in order to obtain BAC DNA miniprep. In the next day, bacterial cells were centrifuged at 4000 rpm for 10 min, at 4ºC. Pellet was resuspended in 200μL of buffer S1 (50mM Tris-HCl, 10mM EDTA, 100μg/mL RNase A, pH 8.0) and then, 200μL of Buffer S2 (200mM NaOH, 1% SDS) were added to lysate the cells. The cell suspension was mixed by inversion of the tubes carefully and left 5 min at RT. Next, 200μL of pre-cooled buffer S3 (2.8M KAc, pH 5.1) was added, the suspension was mixed by inversion and incubated on ice for 15 min. Then, the suspension was centrifuged at 13000 rpm for 15 min, at 4°C, to separate the cell debris from the BAC DNA. The supernatant was collected, transfer to a new tube where an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1, Applichem A0944, 0250) was added and mixed by inversion and incubated for 5 min at RT. The mixture was centrifuged at 13000 rpm for 10 min at 4ªC and the aqueous phase was transferred to a new tube. 0.7 volumes of isopropanol were added to each tube, mixed by inversion and the mixtures were incubated for 5 min at RT. Then, the tubes were centrifuged at 13000 rpm for 20 min at 4ºC, to pellet the BAC DNA and the supernatant was discarded. Pellet was washed with 200μL of 70% ETOH and centrifuged at 13000 rpm for 10 min at 4ºC. Finally, the pellet was dried and resuspended in 60μL of nuclease-free water and stored at 4ºC. For transfections, a large amount of pure BAC DNA was necessary, so DNA maxipreps were produced. In this procedure, colonies were inoculated in 250mL of LB broth with the appropriate antibiotic and BAC DNA was purified using with the NucleoBond® BAC 100 kit (Macherey-Nagel), according to manufacturer’s instructions.

28

Baby hamster kidney (BHK-21) fibroblast cells were used for growing and titrating viral stocks, in vitro growth curves, ex vivo reactivation and plaque assays. They were maintained in Glasgow Minimum Essential Medium (GMEM) supplemented with 10% fetal bovine serum (FBS), 10% tryptose phosphate broth (TPB), 100 U/mL penicillin-streptomycin and 2mM L-glutamine. Mouse embryonic fibroblasts (NIH-3T3) expressing the CRE recombinase were used to remove the bacterial artificial chromosome (BAC) cassette, during the construction of MHV-68 recombinant viruses. They were maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented as above, except TPB. This cell line was established by transduction of NIH-3T3 cells with a retrovirus derived from Phoenix-ecotropic cells transfected with pMSCV-NEO (cre of bacteriophage P1 cloned into the EcoRI-XhoI sites of pMSCV-NEO) and selection with 1mg/mL G418 (Stevenson et al., 2002) BHK-21 and NIH-3T3-Cre cell cultures were grown in a humidified tissue culture incubator at

37ºC under 5%CO2.

Wild type MHV-68 was originally isolated by Prof. Dr Blaskovic (Blaskovic et al., 1980). The wild type virus (v-WT) used in this work was derived from a genomic BAC and was kindly offered by Dr Heiko Adler and Dr Ulrich Koszinowski. This virus is essentially a MHV-68 clone G2.4, that was isolated from virus grown in BHK-21 cells, but contains a single loxP site (Efstathiou et al., 1990; Adler et al., 2000; Adler et al., 2001). It was used in all the in vitro and in vivo experiments as a control virus. MHV-68 recombinant virus expressing yfp was derived from a genomic BAC and was provided by Dr Samuel Speck. This virus expresses the yellow fluorescent protein (yfp), driven by the human cytomegalovirus (HCMV) and cloned in the intergenic region between open reading frames 27 and 29b (Collins et al., 2009). This virus allows direct detection of infected cells based on yfp expression and was used in all the yfp in vitro and in vivo experiments as a control virus. Chimeric kLANA MHV-68 virus, in which the mLANA was replaced by the full length kLANA and 5’ UTR (v-kLANA) was constructed by Dr Marta Miranda and it was also used in all the in vitro and in vivo experiments as a control virus. A chimeric kLANA MHV-68 expressing yfp was constructed in Psimas Lab (Habison et al., 2017) and it was also used in all the yfp in vitro and in vivo experiments as a control virus. v-kLANA ∆332-929 was constructed by André Seixas (unpublished data, Seixas, Miranda e Simas). It is a recombinant virus that has a deletion between amino acids 332 to 929, having the totality of the internal repeat region of kLANA deleted. v-kLANA ∆465-929 was constructed by André Seixas (unpublished data, Seixas, Miranda e Simas). It is also a recombinant virus that has a deletion encompassing amino acids 465 to 929, which removes part of the glutamine region (Q), all the leucine zipper (LZ) and the glutamate and glutamine (EQE) regions.

29

The reconstitution of MHV-68 virus was performed on BHK-21 cells. One day before the experiment, the cells were split in order to obtain 106 cells/6cm2dish. 1µg of BAC DNA maxiprep was diluted into 500µL of unsupplemented GMEM (with a cut tip) in a eppendorf. Then, 2µL of X-tremeGENE HP DNA Transfection Reagent (Roche Applied Science) was added directly into the media and mixed gently. The mixture was left incubating at room temperature for 20 min. The transfection complexes were added to the dishes and incubated at 37ºC. After some days, when it was possible to observe a lot of cytopathic effect (CPE), but still some intact cells, the cells were scrapped onto media, harvested and stored at -80ºC (BAC+ viral stocks).

In order to remove completely the BAC sequences, BAC+ viral stocks were passed through NIH3T3-CRE cells. On the day before infection, 1.75x105 cells/well were seeded in 6-well plates and the BAC cassette was removed by cell passage coupled with limiting dilutions. Therefore, in each passage the virus gradually loses the GFP signal stemming from the GFP expression cassette present in the BAC sequences. After the incubation period (usually 3 days) on the last passage, the cells were scrapped and stored at -80ºC (BAC- viral stocks). These viral stocks were used to produce working stocks.

Viral stocks were produced by infecting 5x106 BHK-21 cells with a MOI of 0.002 plaque forming 2 units (PFU)/cell, in 175 cm flasks. After the infection, cells were incubated for 4 days at 37ºC, 5% CO2 and after the incubation period, they were scrapped into supplemented GMEM and centrifuged for 5 min and 4ºC, at 1500 rpm. The supernatant was transferred to 30 mL bottles and ultracentrifuged at 12000 rpm, during 2h at 4ºC to pellet virus. The supernatant was discarded and to the pellet was added 2mL of fresh media (softened for 1h). The pellet was resuspended in that same volume, distributed in 75µL aliquots and stored at -80ºC. These stocks are the working stock media (WSM) and were used for all the infectivity assays. The cell pellet was resuspended in 2mL of fresh supplemented GMEM, aliquoted and stored at -80ºC. These are the cell working stocks (CWS) and can be used to produce subsequent viral stocks.

To perform virus titration, first 100µL of virus were added to 15 mL falcon tubes containing 900µL of supplemented GMEM, in order to dilute the virus 10 times. Then another 10-fold serial dilutions were made (10-3 to 10-8). 1 mL of 2.5x105 BHK-21 cells were added to the falcons and incubated in a rotating

30 table, during 1h at room temperature and 30 rpm. After the incubation period, 2mL of fresh supplemented GMEM were added to the tubes, mixed by inversion of the tubes and cells/virus mixture were added to 6-well plates. Plates were incubated for 4 days at 37ºC. Cells were fixed with 4% formaldehyde in PBS and stained with 0.1% toluidine blue. Viral plaques were counted using a magnifying glass. Titer was given using the formula:

1 1 Titer = nr of plaques x x dilution inoculum

To confirm protein expression of WT and kLANA protein mutants, on BHK-21 cells and in a yfp and non-yfp background, 2x105 cells/well in a 12-well plate were infected with 3 PFU/cell in a total volume of 500µL. As a control, a well with non-infected cells (mock) was analyzed in parallel. The inoculum was added to the wells and cells were incubated during 2h, at 37ºC with a gentle rock every 30 min. After the incubation period, the inoculum was removed and cells were washed gently with 1mL of pre-warmed media and 1mL of fresh supplemented media was added to each well. Cells were incubated again during 4h at 37ºC. After that time, cells were washed twice with ice cold PBS. 110µL of TM lysis buffer (150mM NaCl, 10mM Tris-HCl pH 7.4, 1mM Na3VO4, 1mM NaF, 1% Triton X-100, complete protease inhibitors (Roche), MilliQ H2O) were added to each well, cells were harvested into eppendorfs, left for 10 min on ice and and frozen at -20ºC. The samples were centrifuged (13000rpm during 10 min on a refrigerated centrifuge, to remove cell debri) and the supernatant (approximately 100µL) was transferred to a new tube. 100µL of 2x Laemmli’s Buffer (100mM Tris-HCl pH 6.8, 20% glycerol, 4% SDS, 10% b-mercaptoethanol, 0.1% bromophenol blue, MilliQ H2O) were added to the cell lysates and these were heated at 95ºC for 3 min, in a heat block, before loading into gel.

The technique of polyacrylamide gel electrophoresis under denaturing conditions consists in the separation of proteins of a sample based on their molecular weight. With the application of an electric field proteins migrate through the gel depending on its size. The lower molecular weight proteins migrate faster than the heavier proteins, separating easily. The acrylamide gels consist of two phases (Table 3). The stacking is a constant phase and is used to compact samples, allowing a simultaneously progress of the samples. The gel used had a percentage of 5% of acrylamide. The resolving phase is where the separation of proteins occurs and its acrylamide percent can vary between 5%-15%, depending of the molecular weight of the interest proteins. The higher the percentage of acrylamide, the smaller the mesh size of the pore, allowing a

31 better separation of lower molecular weight proteins. In this case, the resolving gel had a percentage of 10% of acrylamide.

Table 3 - Composition of electrophoresis gels.

Volume for 3mL of stacking gel Volume for 20mL of (pH 6.8) resolving gel (pH 8.8) H2O 2.1 7.9 30% Acrylamide mix 0.5 6.7 Tris-Cl (1.0M, pH6.8) 0.38 - Tris-Cl (1.5M, pH 8.8) - 5.0 SDS (10%) 0.03 0.2 10% Ammonium persulfate 0.03 0.2 TEMED 0.003 0.008

Once the gels have polymerized, 50µL of each sample (corresponding to lysates of about 1.25x105 cells) and a molecular weight marker (Dual color, BioRad) were loaded into the gels and the step of electrophoresis was performed. This step was performed in a Bio-Rad tray with TGS buffer 1x (25mM Tris-base, 192mM glycine and 0.1% SDS, pH 8.3) at 180 V constant, for 1h.

After the electrophoresis, proteins are separated along the acrylamide gel according to their molecular weight. These proteins were transferred from gels to nitrocellulose membranes (one membrane to each gel), by the wet-transfer method. The transfer was carried out in transfer buffer

(700mL MilliQ H2O, 100mL 10x Tris-Glycine (1x TG: 25mM Tris pH 8.3 and 192mM Glycine), 200mL methanol, 3.7mL 10% SDS), overnight at 100mA, at 4ºC. After the transfer, the nitrocellulose membranes were incubated with Ponceau dye for 2 min to visualize proteins in the membrane.

The membranes were blocked using a blocking solution (PBS+0.05% Tween-20 (PBS-T) + 2.5g of non-fat powder milk) in a rotating table for 30 min, 10 rpm at RT. Then, the membranes were washed twice with PBS-T and incubated with blocking solution containing each of the primary antibodies (Table 4) in sealed bags, in a rotating table for 2h, 10 rpm at RT. After the incubating period, the membranes were washed 3x with PBS-T, during 5 min each and incubated with blocking solution containing the secondary antibodies (Table 5) specific to the species of the primary antibodies, during 30 min, 10 rpm at RT. The membranes were washed again, 3x with PBS-T for 5 min.

32 Table 4 - Primary antibodies used in western blot.

Primary Molecular Species Dilution Supplier antibody weight Advanced kLANA LN53 Rat 250 kDa** 1:1000 Biotech mLANA mAb 6A3 Mouse 50 kDa 1:10 Psimas Lab ** M3 Rabbit 43 kDa 1:2000 Psimas Lab *** EGFP* Mouse 27 kDa 1:2000 Sigma Actin Rabbit 42 kDa 1:2000 Sigma * This antibody was only used to detect protein expression of the viruses with a yfp background. ** (Pires de Miranda et al., 2012). *** (Jensen et al., 2003).

Table 5 - Secondary antibodies used in western blot.

Secondary Immunized Dilution Supplier antibody species Anti-mouse-HRP* Goat 1:5000 Jackson Anti-rat-HRP Goat 1:5000 Immunoresearch Anti-rabbit-HRP Donkey 1:5000 GE Healthcare * HRP, horse radish peroxidase.

To detect bound antibodies, the membranes were covered with the same quantity of hydrogen peroxide and luminol from SuperSignalÒ West Pico Chemiluminescence Substrate (Thermo Scientific) for 2 min and then exposed to the X-ray film. Films were revealed in Curix developer.

To perform this assay, in the day before infection, 5x104 BHK-21 cells/well were seeded in 24- well plates. On the day of infection, the viruses were diluted in media to 0.01 PFU/cell in 200µL of complete GMEM. The media was removed and cells were infected with control (v-WT and v-kLANA) or mutant viruses. Cells were incubated during 1h at 37ºC. After the incubation period, the inoculum was removed and the cells were washed twice with 500µL of PBS. Then, 1mL of fresh media was added in each well and cells were incubated at 37ºC. Cells and media corresponding of time-point zero were harvested right after adding the 1mL of media and stored at -80ºC until titration. This assay consists in recovering the cells and media overtime (0, 24, 48, 72, 96 and 120 hours post-infection (h.p.i.)), freeze and determine virus titers by plaque assay in duplicate.

33

6-8 week-old female C57BL/6 J used in this work were from Charles Rivers Laboratories. Animals were housed and subjected to experimental procedures in specific pathogen-free conditions, at Instituto de Medicina Molecular animal facility, Lisbon, Portugal.

This study was in accordance with the Federation of European Laboratory Animal Science Associations guidelines on laboratory animal welfare.

To infect mice, the viruses were diluted to a concentration of 104 PFU in 20µL of PBS. 7 week- old mice were anaesthetized with isoflurane and inoculated intranasally with the viruses (five mice for each virus). At 14 days post-infection, mice were sacrificed by CO2 inhalation and the spleens were removed to a falcon containing 5mL of PBS+2% FBS.

Spleens were mechanically disrupted and filtered through a 100µm cell strainer, in order to remove some debris. Cells were centrifuged at 1200 rpm, for 10 min at 4ºC, and the pellet was resuspended in 1mL of Red Blood Cell Lysis Buffer (RBL) (154mM ammonium chloride, 14mM sodium hydrogen carbonate, 1mM EDTA pH 7.3). Cells were incubated on ice for 5 min and then, washed with 10mL of PBS+2% FBS. Cells were centrifuged at 1200 rpm, for 5 min at 4ºC. Pellet was resuspended in 1mL of PBS+2% FBS and each splenocyte suspension was divided in two falcons (500µL each): one falcon was used to assess viral latency by infectious center assay/reactivation assay (ICA) and the other falcon was used to flow cytometry analysis.

The reactivation assay allows the determination of the titers of latent virus. The co-culture of single cell suspension of splenocytes with BHK-21, leads to reactivation of latent virus and consequently to the formations of viral plaques. The falcon intended to assess viral latency was centrifuged at 1200 rpm, for 5 min at 4ºC. The pellet was resuspended in 1mL of complete GMEM and then, 1.5mL of complete GMEM was added to perform a total volume of 2.5mL, which correspond to half of a spleen. 10-fold serial dilutions were prepared and plated, in duplicate, in 6 cm2 plates containing 4.5x105 BHK-21 cells. Plates were incubated for 5 days at 37ºC. Another assay was performed with the spleens to search for the presence of pre-formed infectious viruses. In this assay, the splenocytes suspensions were freeze-thawed, plated and incubated for 4 days, at 37ºC. In both assays, after the incubation period, cells were fixed with 1% formaldehyde

34 in PBS and stained with 0.1% toluidine blue. Viral plaques were counted in a magnifying glass and infectious centers (PFU/spleen) were determined.

After the splenocytes suspensions were prepared, the cells were filtered through a 40µL cell strainer and incubated with purified Rat anti-mouse CD16/CD32 (1:100 in PBS+2% FBS) for 15 min on ice, to block Fc receptors. After washing with PBS+2% FBS and centrifugation, splenocytes were surfaced stained by incubation for 25 min on ice, in the dark, with the appropriated antibodies diluted in PBS+2% FBS: APC-H7 Rat anti-mouse CD19 (1:400) (BD Biosciences), Anti-Human/Mouse GL7 eFluorÒ 660 (1:200) (eBiosciences) and PE conjugated hamster anti-mouse CD95 (1:800) (BD Biosciences). Later, cells were washed twice with PBS+2% FBS, to remove the unbound antibodies and centrifugation. Cells were resuspended in PBS+2% FBS and transferred to FACS tubes. For each infection group, 1x106 GC B cells (CD19+CD95hiGL7hi) were purified by a BD FACSAria (BD Biosciences) cell sorters. The total splenocytes were analyzed on a LSR Fortessa (BD Biosciences), using FACSDiva software (BD Biosciences) for acquisition and FlowJo (Tree Star, Inc.) for analysis

A pool for each infectious group was prepared (5 mice per pool), from the splenocyte suspensions. The splenocytes from each pool were counted and diluted to 2x106 cells in 100µL PBS+2% FBS. 2-fold serial dilutions were prepared and eight replicates of each dilution added to PCR tubes containing 10µL of lysis buffer (10mM Tris-HCl pH8.3, 3mM MgCl2, 50mM KCl, 0.45% NP-40, 0.45% Tween-20, 0.5mg/mL proteinase K). PCR tubes were left overnight at 37ºC and in the next day, proteinase K was inactivated at 95ºC for 5 min in a thermocycler. The samples were analyzed by real- time PCR, on a Rotor Gene 6000 thermocycler (Corbett Life Science) using a fluorescent Taqman probe and primers specific for the MHV-68 M9 gene: M9-F (upper primer): 5’- GCCACGGTGGCCCTCTA-3’, M9-R (lower primer): 5’-CAGGCCTCCCTCCCTTTG-3’ and M9-T probe: 5’- 6-FAM - CTT CTG TTG ATC TTC C – MGB-3’. PCR reactions was prepared in a total volume of 25µL, containing 2.5µL of cell suspension lysate, 200µM of each primer, 300µM of probe, 1x Platinum Quantitative PCR SuperMix-

UDG (Invitrogen), 5mM MgCl2 and nuclease free water. The cycling program consisted of an initial melting step of 95ºC for 10 min followed by 40 cycles of amplification, starting at 95ºC for 15 sec and then 60ºC for 1 min. In all PCR runs, a positive control was added. This positive control consisted of a series of dilutions of pGBT9-M9, a plasmid that contains the M9 gene. Real-time PCR results were analyzed on the Rotor Gene 6000 software. In all dilutions tested each replicate was scored positive or negative based on comparison with the negative (water sample) or the positive (plasmid containing M9 gene) controls.

35

All the statistical analysis was performed with GraphPad Prism Software. To calculate the frequency of cells with viral DNA, for limiting dilution analysis, 95% confidence intervals were determined as described (Marques et al., 2003).

36 Chapter 4. Results

As described above, mLANA and kLANA share a lot of similar characteristics. Habison et al. replaced the endogenous ORF73 (mLANA) of MHV-68 genome by KSHV ORF73 (kLANA) and 5’UTR, but not the kLANA promoter, to assess the viability of this chimeric virus to study kLANA in vivo. The resulting chimeric virus was named v-kLANA. They showed that the levels of latency in vivo of v-kLANA were lower compared to the v-WT, but both v-WT and v-kLANA infected cells had similar genome copy numbers. Thus, kLANA rescued mLANA deficient MHV-68, allowing the chimeric virus to establish a latent infection in vivo (Habison et al., 2017). This finding shows that this is a viable model to study KSHV pathogenesis in vivo by infect small animal models with v-kLANA.

As the internal region of kLANA is constituted by a series of internal repeats and their functions are not fully known, some in vitro studies are being done to address the relevance of the different kLANA internal repeats. The aim of this work was to evaluate the importance of kLANA DE internal repeats in an in vivo infection, so a recombinant virus containing a deletion encompassing amino acids 331 to 495 of kLANA, which corresponds to the DE internal repeat region, was engineered (v-kLANA Δ331-495) in backgrounds of wild type MHV-68 (non-yfp) or yellow fluorescent protein (yfp). However, two other recombinant viruses from previous work in our laboratory were added to the in vitro and in vivo experiments, in order to complement the results and compare the contributions of the DE region with the other internal repeat regions of kLANA (unpublished data, Seixas, Miranda e Simas): v-kLANA Δ332- 929, which has a deletion in amino acids 332 to 929, having the totality of the internal repeat region of the protein deleted and v-kLANA Δ465-929, which has a deletion encompassing amino acids 465 to 929, which removes part of the glutamine region (Q), all the leucine zipper (LZ) and the glutamate and glutamine (EQE) regions (Figure 7).

37

Figure 7 - Schematic diagram of KSHV LANA and deletion mutants. The different domains of the protein are indicated, in bold, namely proline-rich region (P), the aspartate and glutamate (DE), the glutamine (Q), the glutamate and glutamine (EQE), the leucine zipper (LZ) regions and the DNA binding domain (DBD). Amino acid residues are indicated below the different domains. The fold deficiency in episome maintenance in vitro was determined by comparison with kLANA WT and an up arrow means that mutant showed a bigger deficiency in episome maintenance than kLANA (adapted from De Leon Vazquez et al., 2013).

The MHV-68 recombinant virus was generated by a mutagenesis procedure in E. coli DH10 according to a Two-Step-Replacement-Strategy (O’Connor et al., 2009), and using a BAC with the entire MHV-68 genome cloned, as described in detail in Materials and Methods, section 3.3.3. This technique allows the maintenance of viral genome as a BAC in E. coli, where a homologous recombination occurs between the shuttle plasmid and the BAC (Adler et al., 2003), allowing the introduction of the desired mutations into MHV-68 genome. After the mutagenesis procedure and in order to identify which clones had the BAC’s with the kLANA inserted, a colony PCR with specific primers to C-terminal region of kLANA was performed. PCR positive clones were then analyzed by restriction digestion using two different restriction enzymes: EcoRI and BamHI (Figure 8A and Figure 8B, respectively), in order to assess the integrity of the viral genome and to verify if the desired mutation in kLANA was inserted. v- WT and v-kLANA were also digested to compare the restriction profile with the ones from the recombinant viruses. In Figure 8, some bands have colorful circles or boxes that show specific bands from the digestion of the viral genome with the enzymes. These specific bands allow the assessment of integrity of viral genome and verification of the introduction of the desired mutation by comparison of the sizes of bands.

38 A

B

Figure 8 - Restriction profiles of v-WT, v-kLANA and the recombinant genomes, in a non-yfp and yfp background, respectively. (A) Restriction profile analysis using EcoRI. Red circle corresponds to full length kLANA band. (B) Restriction profile analysis using BamHI. Orange circles correspond to specific bands of v-WT and green circles correspond to specific bands of v-kLANA and v-kLANA Δ331-495. Blue boxes show specific band that correspond to the inserted mutation on v-kLANA Δ331-495.

39 Recombinant viruses which the integrity of the genome was intact and mutation was well introduced, were then transfected into BHK-21 cells to start the reconstitution of the viruses. The BAC sequences were completely removed by transfecting the reconstituted viruses in NIH-3T3 fibroblasts expressing CRE recombinase.

In order to start in vitro experiments, viral stocks were produced, as described in Materials and Methods, section 3.7. In this step, the other viruses: v-kLANA Δ465-929 and v-kLANA Δ332-929, were added to the work to give a better comprehension of the importance of each internal repeat of kLANA, since they have different internal repeats deleted. Viruses used in the in vitro and in vivo experiments are showed in Table 6.

Table 6 - Viruses that were used to perform in vitro and in vivo experiments. The “#” corresponds to the number of the clone.

non- Viruses yfp Observations yfp v-kLANA ∆331-495 X DE internal repeat deleted #27 v-kLANA ∆331-495 X DE internal repeat deleted #36 v-kLANA ∆465-929 1 X Q, LZ and EQE internal repeats deleted #32 v-kLANA ∆331-495 X DE internal repeat deleted #15 v-kLANA ∆465-929 X Q, LZ and EQE internal repeats deleted #161 v-kLANA ∆332-929 X All the internal repeat region deleted #21 1 Virus from previous work in our laboratory that were used in this work (unpublished data, Seixas, Miranda e Simas).

Following the viral stocks production, they were titrated to determine the concentration in each viral stock. This is an important step, because in order to perform all the in vitro and in vivo experiments, a certain concentration of virus must be added and it is only possible to add the correct concentration by determining the exact amount of virus in each viral stock. The viral titration method is described in detail in Materials and Methods, section 3.8. The concentrations of the different viral stocks are showed in Table 7. Although the concentration of some viral stocks is lower, it is within the range of values of expected titers.

40 Table 7 - Viral stock titers.

Titers Titers yfp viruses non-yfp viruses (PFU/mL) (PFU/mL)

v-WT.yfp 4x107 v-WT 1.3x107

v-kLANA.yfp 5x106 v-kLANA 7.7x106

v-kLANA ∆331-495 v-kLANA ∆331-495 7 3x107 1.3x10 #27.yfp #15 v-kLANA ∆331-495 v-kLANA ∆465-929 6 1.6x107 1.2x10 #36.yfp #16 v-kLANA ∆465-929 v-kLANA ∆332-929 7 5x106 2.4x10 #32.yfp #2

To investigate if all recombinant viruses expressed correctly the different kLANA mutant proteins, BHK-21 cells were infected with a MOI of 3 PFU/cell during 6h and a western blot was performed using antibodies against kLANA and other cellular proteins. The antibodies used are in Table 4, in the Section 3.9.3, in Materials and Methods. kLANA LN53 is a commercial antibody that recognizes the repetitive glutamic motifs EQEQE found in the glutamate and glutamine repeat region (EQE) of kLANA. This antibody detected full length kLANA with a molecular weight of approximately 250 kDa and mutant kLANA proteins with a lower molecular weight, as expected. The immunoblotting worked very well, however, recombinant virus v-kLANA ∆465-929 #32 lacks the EQE internal repeat of the protein, so the detection in this mutant did not work. mLANA was detected only in the WT sample, as expected and the blot showed a protein with a molecular weight of approximately 50 kDa. Detection of M3 (a chemokine binding protein expressed in lytic infection) showed that the levels of infection were similar between v- WT and chimeric viruses, indicating comparable infection levels, in yfp background (Figure 9A), but in the viruses in a non-yfp background, the levels in kLANA and in mutant kLANA were higher than in the WT (Figure 9B). Actin protein was also detected and worked as a control for cell infection. Here, the levels of detection between all samples were similar and the blot showed a protein with a molecular weight of approximately 42 kDa. GFP protein was also detected to verify if all the samples expressed this protein. The levels of detection were similar between samples, except for the mutant v-kLANA ∆331- 495 #27 where the detection level was lower.

41

yfp yfp A yfp

#27. 495 495 #36. 495 #32. 929 - - -

331 331 465 yfp

△ △ △ yfp

WT. kLANA. kLANA kLANA kLANA - - - - - Mock v v v v v

250 kDa - α - kLANA

50 kDa - α - mLANA

50 kDa - α – M3

37 kDa -

50 kDa -

α – Actin

37 kDa - 37 kDa - α – EGFP 25 kDa -

495 B -

331

△

kLANA kLANA WT - - - Mock v v #15 v

α - kLANA 250 kDa -

50 kDa - α - mLANA

50 kDa - α – M3 37 kDa -

50 kDa - α – Actin 37 kDa -

Figure 9 - Detection of the expression of viral proteins in v-WT and chimeric viruses. BHK-21 were infected with 3 PFU/cell during a period of 6h. The proteins were detected with the antibodies indicated on the right and the molecular weight protein markers are indicated on the left. (A) Viruses in yfp background. (B) Viruses in a non-yfp background.

42

Although it is known that ORF73 from MHV-68 is not essential for its growth in vitro (Fowler et al., 2003), it was important to verify if the growth of recombinant viruses was different from v-WT or v- kLANA. To assess their growth, a multi-step growth curve was performed. In this procedure, BHK-21 cells were infected with a MOI of 0.01 PFU/cell and every day, at the same hour, during 5 days, time points were collected. Later, the time points were titrated by freeze-thawed, titers were determinate by plaque assay and a growth curve was constructed (Figure 10). As in other experiments, both v-WT and v-kLANA were used as control viruses.

8

v -W T .y fp A 6

L 8 v -k L A N A .y fp m /

U v -k L A N A D 3 3 1 -4 9 5 # 2 7 .y fp F 4 P v --kWL AT .NyAfp D 3 3 1 -4 9 5 # 3 6 .y fp g 6 o L L v -k L A NNAA .Dy f3p3 2 -9 2 9 # 1 3 .y fp m / 2 U v -k L A N A D 43 63 51 -94 29 95 # 32 27 .y fp F 4 P v -k L A N A D 3 3 1 -4 9 5 # 3 6 .y fp g

o 0 L v -k L A N A D 3 3 2 -9 2 9 # 1 3 .y fp 2 0 2 4 4 8 7 2 9 6 1 2 0 T im e a f t e r in f e c t io n ( h ) v -k L A N A D 4 6 5 -9 2 9 # 3 2 .y fp

0 0 2 4 4 8 7 2 9 6 1 2 0 T im e a f t e r in f e c t io n ( h )

B 8

v -W T 6 L 8 v -k L A N A m /

U v -k L A N A D 3 3 1 -4 9 5 # 1 5 F 4

P v -kWLTA N A 3 3 2 -9 2 9 # 2 g 6 D o

L L v -k L A N A D 4 6 5 -9 2 9 # 1 6 m / 2 U v -k L A N A D 3 3 1 -4 9 5 # 1 5 F 4 P v -k L A N A 3 3 2 -9 2 9 # 2 g D

o 0 L v -k L A N A D 4 6 5 -9 2 9 # 1 6 2 0 2 4 4 8 7 2 9 6 1 2 0 T im e a f t e r in f e c t io n ( h )

0 Figure 100 - Growth2 4curves of4 8 v-WT and7 2 chimeric9 6 viruses1 2 in0 both backgrounds. To assess growth of the viruses in vitro, BHK-21T i mweree a finfectedt e r in f e withc t io 0.01n ( h )PFU/cell of the indicated viruses. At the indicated times post- infection, the samples were harvested and titrated by plaque assay of frozen-thawed samples. There was no significant difference between infectious groups. (A) Growth curve of the yfp viruses. (B) Growth curve of the non-yfp viruses.

43 In yfp background (Figure 10A), viruses had a similar growth between them and similar to the growth of other v-kLANA viruses previously observed in our laboratory (Habison et al., 2017). In non- yfp background (Figure 10B), v-kLANA D331-495 #15 had a bigger titer at 0h (first time point) in comparison with the others, but its growth was similar to that observed in other viruses. This difference may be due to the fact that a bigger amount of inoculum was introduced , during infection of BHK-21 with v-kLANA D331-495 #15. In addition to that, there were no significant differences between the growth of viruses. This means that the deletions in the kLANA regions do not affect the lytic replication of the viruses. Since the growth of v-kLANA was also similar to the others, this indicates that the substitution of mLANA for kLANA also does not affect the lytic replication, confirming that the ORF73 is not essential for the growth in vitro.

It is also known that ORF73 in both MHV-68 and KSHV is essential to the establishment of latency in the spleen and that chimeric v-kLANA can rescue mLANA deficient MHV-68, enabling v-kLANA to establish latency in vivo (Habison et al., 2017). To assess if the recombinant viruses with the lacking parts of the internal repeat region could also establish latency in spleens, C57BL/6 J mice were intranasally inoculated with 104 PFU of v-WT or with the recombinant viruses, described in section 3.5, in Materials and Methods. Latent load in spleen (Figure 11, closed circles) was determined by quantification of ex vivo reactivation of competent viruses in total splenocytes by an infectious center assay at 14 days post-infection (d.p.i.). When the latent viruses in the splenocytes are co-cultured with fibroblasts, they form viral plaques within cell monolayer that can be quantified. To verify if no lytic infectious viruses were present in the splenocytes, the splenocyte suspensions were freeze-thawed and co-cultured with fibroblasts (Figure 11, open circles). It was performed two assays: one with the recombinant viruses in a yfp background and another with recombinant viruses in a non-yfp background.

In the yfp assay (Figure 11A), as expected, v-kLANA.yfp was able to establish latency in the spleen with lower levels (~2 log) than those of v-WT.yfp (Habison et al., 2017). The levels of latency in v-kLANA D465-929 #32.yfp were similar to the levels of v-kLANA.yfp, as observed in previous work in our laboratory. In v-kLANA D331-495 #27.yfp the ability to establish latency was abrogated, this means that its levels were below the limit of detection of this assay and in v-kLANA D331-495 #36.yfp the levels of latency were low. These results indicate that the DE internal repeat is important for the establishment of latency in the spleen.

In the non-yfp assay (Figure 11B) the levels of latency of v-kLANA were lower (~2 log) than those of v-WT, as in yfp assay. The latency levels of v-kLANA D332-929 #2 were below the detection limit of this assay and latency levels of v-kLANA D465-929 #16 were ~1 log lower compared with the ones of

44 v-kLANA D465-929 #32, in yfp assay. Since results from v-kLANA Δ331-495 #27.yfp, v-kLANA Δ331- 495 #36.yfp and v-kLANA Δ331-495 #15, which are independent viruses, were similar, this is likely to be its phenotype.

A B

Figure 11 - Quantification of latent infection in spleen, 14 d.p.i., by ex vivo reactivation assay. Mice were intranasally inoculated with 104 PFU of the indicated viruses and 14 d.p.i. latent viruses in the spleen were titrated by infectious center assay (closed circles). Titers of pre-formed infectious viruses were determined in freeze- thawed splenocytes suspensions (open circles). Each circle corresponds to a titer of an individual mouse. Horizontal bars show arithmetic means. The dashed line represents the limit of detection of the assay. (A) Assay with yfp viruses. (B) Assay with non-yfp viruses.

Simultaneous to infection center assay, the spleens infected with v-WT or recombinant yfp viruses were analyzed by flow cytometry. Splenocyte suspensions were incubated with appropriate antibodies, as described in section 3.12.5., in Materials and Methods.

Analyzing the graphics, it can be observed that the percentage of B cells that were GC B cells varied between ~1-7% in the infection groups (Figure 12A). The amount of GC B cells was higher in the spleens of mice infected with v-WT.yfp when compared with v-kLANA.yfp. However, it was significantly lower when compared with the v-kLANA D331-495 #27 and v-kLANA D331-495 #36 mutants. This result was unexpected, in light of results in Figure 11A, since the amount of GC B cells is related with the latent viral load (Habison et al., 2017). v-kLANA.yfp and v-kLANA D465-929 #32.yfp had a similar amount of GC B cells, as expected.

Using the yfp expression, it was possible to determine the frequency of GC B cells that were infected with virus. Here, the percentage of infection varied between ~0-4%, where the higher percentage of yfp+ GC B cells was in v-WT.yfp infected mice (Figure 12B). The percentage of infected

45 cells was almost null in v-kLANA D331-495 #27.yfp, which is in agreement with the previous results from infectious center assay. The percentage of infected cells was unexpectedly too low in v-kLANA D465- 929 #32.yfp, since in Figure 11A its titer was similar to v-kLANA.

Approximately 70% of the cells infected with v-WT.yfp were GC B cells and only ~50% of v- kLANA.yfp infected cells were GC B cells (Figure 12C). Among infected cells with v-kLANA D331-495 #27.yfp or v-kLANA D331-495 #36.yfp, about 8% and 32%, respectively, were GC B cells. Among the infected cells with v-kLANA D465-929 #32.yfp, 25% were GC B cells.

A

B

46 C

Figure 12 - Flow cytometry analyses. In the left panel are the FACS plots from individual mice. In the bottom are the quantification graphs, where each point corresponds to an individual mouse. Bars correspond to mean values. (A) Identification of the total number of GC B cells. (B) Identification of infected GC B cells, using YFP expression. (C) Identification of YFP positive cells that were GC B cells.

4.5.3. Quantification of the frequency of viral DNA-positive total splenocytes

To complement the results from infectious center assay, it was also investigated the frequency of infected cells by limiting dilution coupled with real-time PCR in total splenocytes. This experiment allows to directly indicate the numbers of latently infected cells at 14 d.p.i., in the spleen of mice intranasally infected with v-WT or recombinant viruses. Total splenocytes were subjected to 2-fold dilutions, with 8 replicates and lysed. Cell lysates were analyzed by real-time PCR, using primers and probe specific for MHV-68 M9 gene (as described in Materials and Methods, section 3.12.6.). The experiment with the v- WT and recombinant viruses in a yfp background did not work, so the graphic represented is only from the experiment with viruses in a non-yfp background.

v-WT had the expected frequency of latently infected cells in total splenocytes population (Figure 13 and Table 8) and the frequency of viral DNA-positive cells for v-kLANA was ~1 log lower than v-WT (Habison et al., 2017). There were no differences between v-kLANA and v-kLANA D465-929 #16, as expected due to previous work in our laboratory and v-kLANA D331-495 #15 had a frequency of latently

1 infected cells 2 to ~2 2 log lower than those of v-WT. The v-kLANA D332-929 #2 had a frequency of viral DNA-positive cells lower in total splenocytes compared with the other viruses, however, this result is in agreement with results from the infectious center assay.

47 These results are in agreement with the ones from infectious center assay, except for recombinant virus v-kLANA D465-929 #16 that in this experiment had a frequency of latently infected cells similar to v-kLANA, but in infectious center assay had a titer much lower than v-kLANA.

Figure 13 - Quantification of viral DNA-positive cell in total splenocytes. Mice were intranasally inoculated with 104 PFU of the indicated viruses and 14 d.p.i. the frequencies of viral infection in total splenocytes were determined by limiting dilution coupled with real-time PCR. Data were obtained from pools of five spleens per infection group. Bars represent the frequency of viral DNA-positive cells with 95% confidence intervals.

Table 8 - Frequencies of MHV-68 latent infection in total splenocytes.

Cell Reciprocal frequencyb of D.p.i. Virus Subpopulation viral DNA positive cells v-WT 98 (63-219) v-kLANA 1100 (689-2733)

a 13727 (8221-41568) Total Splenocytes 14 v-kLANA D331-495 #15 v-kLANA D465-929 #16 1565 (956-4286)

v-kLANA D332-929 #2 27865 (17805-64059) a Data were obtained from a pool of 5 spleens. b Frequencies were calculated by limiting dilution coupled to real-time PCR, with 95% confidence intervals (numbers in parentheses).

Together, these data shows that among the internal repeats of kLANA, the aspartate and glutamate (DE) repeat seems to have a much more important role for an efficient establishment of latency in the spleen, during an in vivo infection.

48 Chapter 5. Discussion and future perspectives

This work led to the construction of a new v-kLANA MHV-68 recombinant virus, v-kLANA Δ331- 495, containing a deletion corresponding to amino acids 331 to 495 of kLANA which encompasses all of the DE internal repeat region. This recombinant virus was constructed to assess the impact of the kLANA DE internal repeat deletion in an in vivo model of infection, in order to better understand the function of this specific region. Two other recombinant viruses from previous work (unpublished data, Seixas, Miranda e Simas) were added to this study to complement the results and compare the contributions of the DE region with the other internal repeat regions: v-kLANA Δ465-929, which has part of the Q, the LZ and the EQE internal repeat region of kLANA deleted and v-kLANA Δ332-929, which has the entire internal repeat region deleted. Results indicate that from all the internal repeats of kLANA, the DE region is essential for an efficient establishment of latency in the spleen.

Starting with in vitro experiments, our results showed that despite the mutation on kLANA, recombinant virus v-kLANA Δ331-495 expressed correctly mutant kLANA with a lower molecular weight than original kLANA, as well as other cellular proteins tested (Figure 9). This result means that the deletion introduced did not affect expression of kLANA. This recombinant virus has an advantage over v-kLANA Δ465-929 and v-kLANA Δ332-929, which is the fact that expression of kLANA can be detected using a very sensitive commercial antibody (LN53) that detects several epitopes in the EQE repeat region. Since v-kLANA Δ465-929 lacks this internal repeat, expression of kLANA could not be detected (Figure 9A). Previous work from our laboratory showed that expression of kLANA from v-kLANA Δ332- 929 could not be detected due to the same reason. Thus, any phenotype observed in this last virus, could be due to low level or no expression of the protein (although in previous work it was observed that mRNA levels were good). v-kLANA Δ331-495 had a normal in vitro growth, similar to v-WT and v-kLANA recombinant viruses, in both non-yfp and yfp backgrounds (Figure 10). This means that neither the replacement of mLANA with kLANA or deletions of internal repeats of kLANA affected lytic replication in vitro. These results were not unexpected, since it is known that ORF73 from MHV-68 is not essential for growth of the virus in vitro (Fowler et al., 2003; Habison et al., 2017).

Our in vivo results from the infectious center assay showed for the first time that at 14 d.p.i v- kLANA Δ331-495 was able to establish latency in spleen with values near to the limit of detection of the assay. This recombinant virus exhibited ~1 log of deficiency compared with v-kLANA and ~3 log of deficiency compared to v-WT (Figure 11). v-kLANA Δ465-929 had a peak of latency similar to v-kLANA (Figure 11A), which is in agreement with results from previous work in our laboratory. Consistent with previous results v-kLANA Δ332-929, failed to establish latency, presenting levels below the limit of detection of this assay (Figure 11B). These results showed that the capacity to establish latency in v- kLANA Δ331-495 is markedly less than v-kLANA, but it is higher than v-kLANA Δ332-929. Together, these data indicate that DE internal repeat is required for the establishment of latency in the spleen, and

49 this internal repeat must be present in kLANA in order to have a correct establishment of latency during an in vivo infection. The results from FACS showed that among the population of GC B cells only a small percentage were yfp positive in both v-kLANA Δ331-495 independent viruses. This percentage was lower than that observed for v-kLANA.yfp. Between the yfp positive B cells only 8% and 32% (in v-kLANA Δ331-495 #27 and v-kLANA Δ331-495 #36, respectively) were GC cells. Altogether these results are in agreement with results from infectious center assay (Figure 11A) and results from frequency of viral DNA-positive cells (Figure 13), since a lower percentage of GC B cells infected means a lower load of latent viruses in the spleen. However, data from FACS regarding v-kLANA Δ 465-929 were not in agreement with the remaining results, since the percentage of yfp positive GC B cells was lower that observed with the v- kLANA and the levels of reactivating virus (infectious center assay) were the same for both viruses (Figure 11A). The infectious center assay results are corroborated by the results of the frequency of viral DNA- positive cells, where the frequencies of v-kLANA and v-kLANA Δ 465-929 were similar (Figure 13 and Table 8) and v-kLANA Δ332-929 presented a frequency lower than the other viruses. The frequency of viral DNA-positive cells in total splenocytes in v-kLANA Δ331-495 was between frequencies of v-kLANA, v-kLANA Δ 465-929 and v-kLANA Δ332-929. Despite the deletions, v-kLANA Δ331-495 and v-kLANA Δ465-929 were able to establish latency in the spleen, indicating that they were able to maintain the episome persistence, thus showing that deletions did not affect the kLANA DNA binding domains. It is known that the internal region of kLANA has an important role for episome maintenance (De Leon Vazquez and Kaye, 2011; De Leon Vasquez and Kaye, 2013) and as in v-kLANA Δ331-495 this episome persistence was reduced compared with v-kLANA Δ465-929, this indicated that within the internal region of LANA, the deleted region must exert a more important role in episome maintenance. Altogether, the results from these three assays were in agreement with each other and that validated the phenotype of each recombinant virus.

Results of v-kLANA Δ465-929 #32.yfp and v-kLANA Δ465-929 #16 in the two in vivo experiments were not concordant with each other, since the experiments showed different results for the same clone. So, to standardize the phenotype of this virus, experiments need to be repeated.

All results from v-kLANA Δ332-929 were consistent and showed a recombinant virus that is incapable to establish latency in the spleen as shown by infectious center assay and supported by the frequency of viral DNA-positive cells in total splenocytes. These data are in line with previous work of our laboratory and with in vitro studies made by others (De Leon Vasquez and Kaye, 2013).

In summary, this work, showed that v-kLANA Δ331-495 establish very low levels of latency in the spleen, with values remarkably lower than v-kLANA. Future work still needs to be done in order to better understand the functions of the DE internal repeat. First, using this virus in mouse model of infection experiments should be repeated to increase the number of animals to achieve statistical

50 significance. Then, other times after infection may be analyzed (for example day 11 and day 21) to assess if the reduced latency at day 14 is in fact just lower magnitude of infection and not altered kinetics of infection. Lung lytic replication should also be assessed to rule out a defect that could affect latency establishment after intranasal inoculation.

The molecular functions of the DE repeat region are unknown. However, it is known that many proteins interact with different parts of LANA (De Leon Vasquez and Kaye, 2011), so it would be interesting to perform more protein-protein interaction studies with the aim of identifying more proteins that interact with this specific internal region or within all the internal repeat region. Then, map the localization of those interactions to clarify and unveil the functions of LANA internal repeat region.

This work showed that the DE internal repeat region is required for normal levels of v-kLANA latency. Thus, in the future, the recombinant v-kLANA can be used to test in vitro and then in vivo drugs that target and inhibit the DE region functions in order to try to fight KSHV infection.

51 Chapter 6. References

Ablashi, D., Chatlynne, L., Whitman, J. and Cesarman, E. (2002). Spectrum of Kaposi's Sarcoma- Associated Herpesvirus, or Human Herpesvirus 8, Diseases. Clinical microbiology reviews, 15:439-464.

Adler, H., Messerle, M., Wagner, M. and Koszinowski, U. (2000). Cloning and mutagenesis of the murine gammaherpesvirus 68 genome as an infectious bacterial artificial chromosome. Journal of virology, 74:6964– 74.

Adler, H., Messerle, M. and Koszinowski, U. (2001). Virus reconstituted from infectious bacterial artificial chromosome (BAC)-cloned murine gammaherpesvirus 68 acquires wild-type properties in vivo only after excision of BAC vector sequences. Journal of virology, 75(12):5692-5696.

Adler, H., Messerle, M. and Koszinowski, U.H. (2003). Cloning of herpesviral genomes as bacterial artificial chromosomes. Medical virology, 13(2):111– 121.

Alkharsah, K. and Schilz, T. (2012). A Role for the Internal Repeat of the Kaposi's Sarcoma-Associated Herpesvirus Latent Nuclear Antigen in the Persistence of an Episomal Viral Genome. Journal of virology, 86(3):1883-1887.

Arvin, A., Campadelli-Fiume, G., Mocarski, E., Moore, P., Roizman, B., Whitley, R. and Yamanishi, K. (2007). Human herpesviruses: Biology, Therapy and Immunoprophylaxia. 1st edition. Cambridge: Cambridge university press.

Bagni, R. and Whitby, D. (2009). Kaposiʼs sarcoma-associated herpesvirus transmission and primary infection. Current opinion in HIV and AIDS, 4:22-26.

Barbera, A. (2006). The Nucleosomal Surface as a Docking Station for Kaposi's Sarcoma Herpesvirus LANA. Science, 311:856-861.

Barton, E., Mandal, P. and Speck, S. (2011). Pathogenesis and Host Control of Gammaherpesviruses: Lessons from the Mouse. Annual review of immunology, 29:351-397.

Blaskovic D., Stancekova M., Svobodova J. and Mistrikova J. (1980). Isolation of five strains of herpesviruses from two species of free living small rodents. Acta virology. 24:468.

Boss, I., Plaisance, K. and Renne, R. (2009). Role of virus-encoded microRNAs in herpesvirus biology. Trends in microbiology, 17:544-553.

52

Cai, Q., Verma, S., Lu, J. and Robertson, E. (2010). Molecular Biology of Kaposi's Sarcoma-associated Herpesvirus and Related Oncogenesis. Advances in virus research, 87-142.

Chang, Y., Cesarman, E., Pessin, M., Lee, F., Culpepper, J., Knowles, D. and Moore, P. (1994). Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma. Science, 266:1865-1869.

Collins, C.M., J.M. Boss, and S.H. Speck. (2009). Identification of infected B-cell populations by using a recombinant murine gammaherpesvirus 68 expressing a fluorescent protein. Journal of virology, 83:6484-6493.

Correia, B., Cerqueira, S., Beauchemin, C., Pires de Miranda, M., Li, S., Ponnusamy, R., Rodrigues, L., Schneider, T., Carrondo, M., Kaye, K., Simas, J. and McVey, C. (2013). Crystal Structure of the Gamma-2 Herpesvirus LANA DNA Binding Domain Identifies Charged Surface Residues Which Impact Viral Latency. PLoS pathogens, 9.

Coscoy, L. (2007). Immune evasion by Kaposi's sarcoma-associated herpesvirus. Nature reviews immunology, 7:391-401.

Damania, B. (2004). Oncogenic γ-herpesviruses: comparison of viral proteins involved in tumorigenesis. Nature reviews microbiology, 2:656-668.

Davison, A. (2002). Evolution of the herpesviruses. Veterinary microbiology, 86:69-88.

Davison, A., Eberle, R., Ehlers, B., Hayward, G., McGeoch, D., Minson, A., Pellett, P., Roizman, B., Studdert, M. and Thiry, E. (2009). The order Herpesvirales. Archives of virology, 154:171-177.

Davison AJ. Comparative analysis of the genomes. In: Arvin A, Campadelli-Fiume G, Mocarski E. (2007). Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge: Cambridge university press. Chapter 2.

De Leon Vazquez, E. and Kaye, K. (2011). The Internal Kaposi's Sarcoma-Associated Herpesvirus LANA Regions Exert a Critical Role on Episome Persistence. Journal of virology, 85:7622-7633.

De Leon Vazquez, E., Carey, V. and Kaye, K. (2013). Identification of Kaposi's Sarcoma-Associated Herpesvirus LANA Regions Important for Episome Segregation, Replication, and Persistence. Journal of virology, 87:12270-12283.

53 Delgado, T., Sanchez, E., Camarda, R. and Lagunoff, M. (2012). Global Metabolic Profiling of Infection by an Oncogenic Virus: KSHV Induces and Requires Lipogenesis for Survival of Latent Infection. PLoS pathogens.

Dittmer, D., Lagunoff, M., Renne, R., Staskus, K., Haase, A., and Ganem, D. (1998). A Cluster of Latently Expressed Genes in Kaposi’s Sarcoma-Associated Herpesvirus. Journal of virology, 72:8309–8315.

Dittmer, D. and Damania, B. (2016). Kaposi sarcoma–associated herpesvirus: immunobiology, oncogenesis, and therapy. Journal of clinical Investigation, 126:3165-3175.

Efstathiou, S., Ho, Y. and Minson, A. (1990). Cloning and molecular characterization of the murine herpesvirus 68 genome. Journal of general virology, 71:1355-1364.

Flaño, E., Husain, M., Sample, T., Woodland, L. and Blackman, A. (2000). Latent murine gamma- herpesvirus infection is established in activated B-cells, dendritic cells, and macrophages. Journal of immunology, 165(2):1074– 81.

Fowler, P., Marques, S., Simas, JP. and Efstathiou, S. (2003). ORF73 of murine herpesvirus-68 is critical for the establishment and maintenance of latency. Journal of general virology, 84(12):3405– 3416.

Frederico, B., Chao, B., May, J., Belz, G. and Stevenson, P. (2014). A Murid Gamma-Herpesviruses Exploits Normal Splenic Immune Communication Routes for Systemic Spread. Cell host & microbe, 15:457-470.

Grundhoff, A. and Ganem, D. (2003). The Latency-Associated Nuclear Antigen of Kaposi's Sarcoma- Associated Herpesvirus Permits Replication of Terminal Repeat-Containing Plasmids. Journal of virology, 77:2779-2783.

Habison, A., de Miranda, M., Beauchemin, C., Tan, M., Cerqueira, S., Correia, B., Ponnusamy, R., Usherwood, E., McVey, C., Simas, J. and Kaye, K. (2017). Cross-species conservation of episome maintenance provides a basis for in vivo investigation of Kaposi's sarcoma herpesvirus LANA. PLOS pathogens, 13(9).

Hughes, M., Morgan, H. and Rixon, F. (2002). Measuring the dielectric properties of herpes simplex virus type 1 virions with dielectrophoresis. Biochimica et Biophysica Acta (BBA) - General subjects, 1571:1-8.

54 Jensen, K., Chen, C., Hipkin, W., Wiekowski, M. and Schwarz, A. (2003) Disruption of CCL21-induced chemotaxis in vitro and in vivo by M3, a chemokine-binding protein encoded by murine gammaherpesvirus 68. Journal of virology, 77: 624-630.

Jeong, J.H., Orvis, J., Kim, J.W., McMurtrey, C.P., Renne, R. and Dittmer, D.P. (2004). Regulation and autoregulation of the promoter for the latency-associated nuclear antigen of Kaposi's sarcoma- associated herpesvirus. Journal of biological chemistry, 279:16822-16831.

Juillard, F., Tan, M., Li, S. and Kaye, K. (2016). Kaposi’s Sarcoma Herpesvirus Genome Persistence. Frontiers in microbiology, Chapter 7.

King, A., Lefkowitz, E., Adams, M. and Carstens, E. (2011). Virus taxonomy: Ninth report of the International Committee on Taxonomy of Viruses. 1st ed. Oxford: Elsevier, 111-122

Kwun, H., da Silva, S., Shah, I., Blake, N., Moore, P. and Chang, Y. (2007). Kaposi's Sarcoma- Associated Herpesvirus Latency-Associated Nuclear Antigen 1 Mimics Epstein-Barr Virus EBNA1 Immune Evasion through Central Repeat Domain Effects on Protein Processing. Journal of virology, 81:8225-8235.

Marques, S., Efstathiou, S., Smith, K., Haury, M. and Simas, J. (2003). Selective Gene Expression of Latent Murine Gammaherpesvirus 68 in B Lymphocytes. Journal of virology, 77:7308-7318.

Moore, P. and Chang, Y. (2003). Kaposi's Sarcoma–Associated Herpesvirus Immunoevasion and Tumorigenesis: Two Sides of the Same Coin?. Annual review of microbiology, 57(1):609-639.

Pires de Miranda, M., Lopes, F., McVey, C., Bustelo, X. and Simas, J. (2012). Role of Src Homology Domain Binding in Signaling Complexes Assembled by the Murid γ-Herpesvirus M2 Protein. Journal of biological chemistry, 288(6):3858-3870.

Polizzotto, M., Uldrick, T., Wyvill, K., Aleman, K., Marshall, V., Wang, V., Whitby, D., Pittaluga, S., Jaffe, E., Millo, C., Tosato, G., Little, R., Steinberg, S., Sereti, I. and Yarchoan, R. (2016). Clinical Features and Outcomes of Patients with Symptomatic Kaposi Sarcoma Herpesvirus (KSHV)- associated Inflammation: Prospective Characterization of KSHV Inflammatory Cytokine Syndrome (KICS). Clinical infectious diseases, 62:730-738.

Ponnusamy, R., Petoukhov, M., Correia, B., Custodio, T., Juillard, F., Tan, M., Pires de Miranda, M., Carrondo, M., Simas, J., Kaye, K., Svergun, D. and McVey, C. (2015). KSHV but not MHV-68 LANA induces a strong bend upon binding to terminal repeat viral DNA. Nucleic acids research, p.987.

55 Purushothaman, P., Dabral, P., Gupta, N., Sarkar, R. and Verma, S. (2016). KSHV Genome Replication and Maintenance. Frontiers in microbiology, 7.

Simas, J. and Efstathiou, S. (1998). Murine gammaherpesvirus 68: a model for the study of gammaherpesvirus pathogenesis. Trends in microbiology, 6:276-282.

Sin, S. and Dittmer, D. (2013). Viral latency locus augments B-cell response in vivo to induce chronic marginal zone enlargement, plasma cell hyperplasia, and lymphoma. Blood, 121:2952-2963.

Stevenson, P. (2004). Immune evasion by gamma-herpesviruses. Current opinion in immunology, 16:456-462.

Stevenson, P., May, J., Smith, X., Marques, S., Adler, H., Koszinowski, U., Simas, J. and Efstathiou, S. (2002). K3-mediated evasion of CD8+ T cells aids amplification of a latent γ-herpesvirus. Nature immunology.

Stoopler, E. (2005). Oral herpetic infections (HSV 1–8). Dental clinics of North America, 49:15-29.

Sunil-Chandra NP., Efstathiou S., Arno J. and Nash AA. (1992). Virological and pathological features of mice infected with murine γ-herpesvirus 68. Journal of general virology. 73:2347–56

Uldrick, T., Wang, V., O’Mahony, D., Aleman, K., Wyvill, K., Marshall, V., Steinberg, S., Pittaluga, S., Maric, I., Whitby, D., Tosato, G., Little, R. and Yarchoan, R. (2010). An Interleukin-6–Related Systemic Inflammatory Syndrome in Patients Co-Infected with Kaposi Sarcoma–Associated Herpesvirus and HIV but without Multicentric Castleman Disease. Clinical infectious diseases, 51:350-358.

Uppal, T., Banerjee, S., Sun, Z., Verma, S. and Robertson, E. (2014). KSHV LANA-The Master Regulator of KSHV Latency. Viruses, 6:4961-4998.

Wen, K. and Damania, B. (2010). Kaposi sarcoma-associated herpesvirus (KSHV): Molecular biology and oncogenesis. Cancer letters, 289:140-150.

Whitley RJ. Herpesviruses. In: Baron S, editor. Medical microbiology. 4th edition. Galveston (TX): University of Texas Medical Branch at Galveston; (1996). Chapter 68.

Wu, T., Blackman, M. and Sun, R. (2010). Prospects of a novel vaccination strategy for human gamma- herpesviruses. Immunologic research, 48:122-146.

56 Chapter 7. Supplementary data

A

PCR to introduce the delection

PstI PstI

kLANA 1-495 Digestion with PstI

Cloning into pSP72_PCR1_4

57

B

Digestion with XhoI and HindIII

XhoI HindIII

MHV 68 lower flank Fragment released

102,728

Subcloning into pSP72_PCR1_4

D331-495

105,090 104,710

58

C

BglII BglII 102,728 105,090

Lower flank kLANA △331-495 5’UTR Upper flank Digestion with BglII

Subcloning into shuttle vector

Figure S1 - Schematic representation of the construction of a recombinant shuttle plasmid. (A) The deletion was introduced by PCR, using the primer IMMAF2. The amplified fragment was digested with PstI and was cloned into pSP72_PCR1_4. (B) The pSP72_PCR1/2/5 was digested with XhoI and HindIII to release the lower flank of MHV-68 and then this fragment was subcloned into pSP72_PCR1_4. (C) pSP72_kLANA ∆331-495 was digested with BglII and the fragment released was subcloned into the BamHI-G shuttle vector.

59