Medizinische Hochschule Hannover Detection of Murine Herpesvirus 68

Medizinische Hochschule Hannover

Institut für Virologie/ Helmholtz Zentrum für Infektionsforschung

Detection of murine herpesvirus 68 by the innate immune system and studies on Mus musculus rhadinovirus 1 in its natural host

INAUGURALDISSERTATION zur Erlangung des Grades einer Doktorin der Naturwissenschaften - Doctor rerum naturalium - (Dr. rer. nat.)

vorgelegt von

Sripriya Murthy

aus Bangalore

Hannover 2016

Angenommen vom Senat der Medizinischen Hochschule Hannover am 16.09.2016

Gedruckt mit Genehmigung der Medizinischen Hochschule Hannover

Präsident: Prof. Dr. med. Christopher Baum

Betreuer: Prof. Dr. rer. nat. Melanie M. Brinkmann

Kobetreuer: Prof. André Bleich, PhD

1. Gutachter: Prof. Dr. rer. nat. Melanie M. Brinkmann

2. Gutachter: Prof. André Bleich, PhD

3. Gutachter: Prof. Dr. rer. nat. Beate Sodeik

Tag der mündlichen Prüfung vor der Prüfungskomission: 16.09.2016

Prof. Dr. rer. nat. Christine Falk

Prof. Dr. rer. nat. Melanie M. Brinkmann

Prof. André Bleich, PhD

Prof. Dr. rer. nat. Beate Sodeik

Table of contents

1 Introduction ...... 1 1.1 Herpesviridae ...... 1 1.1.1 Betaherpesvirinae ...... 2 1.1.2 Gammaherpesvirinae ...... 4 1.2 Host defense ...... 9 1.2.1 Pattern recognition receptors ...... 12 1.2.2 UNC93B and TLR trafficking ...... 20 1.2.3 TLR-mediated detection and control of MHV68 ...... 24 1.3 Aims of the study ...... 28

2 Materials and Methods ...... 29 2.1 Buffers and solutions ...... 29 2.2 Commercially available kits ...... 30 2.3 Cloning of plasmid controls ...... 30 2.3.1 Bacteria ...... 31 2.3.2 Determination of copy number to assess PCR sensitivity ...... 33 2.4 Extraction of DNA from tissue samples ...... 33 2.5 PCR analysis for the detection of MmusRHV1, MCMV1, MCMV2 and MHV68 ...... 34 2.5.1 Nested PCR to detect the presence of MmusRHV1, MCMV1 and MCMV2 ...... 34 2.5.2 Non-invasive blood PCR to screen for the presence of MmusRHV1 ...... 34 2.5.3 Limiting dilution nested PCR for the detection of MHV68 ORF50 ...... 37 2.6 Cell culture ...... 38 2.6.1 Primary cells – preparation of dendritic cells with Flt3 ligand and GM-CSF ...... 38 2.7 Viruses ...... 39 2.8 Cell sorting ...... 40 2.8.1 Fluorescence-activated cell sorting (FACS) ...... 40 2.8.2 Magnetic-activated cell sorting (MACS) ...... 40 2.9 Stimulation of MmusRHV1 reactivation ...... 42

2.10 Stimulation of TLR with viruses and TLR ligands for TNFα and IFNα measurement by enzyme-linked immunosorbent assay (ELISA) ...... 42 2.11 ELISA ...... 43 2.11.1 TNFα ELISA ...... 44 2.11.2 IFNα ELISA ...... 45 2.12 Mice ...... 45 2.13 Genotyping of mouse strains used in the study of MHV68 infection both in vitro and in vivo ...... 49 2.14 In vivo infection ...... 51 2.15 Perfusion ...... 51 2.16 Determination of viral titers in organs by tissue culture

infective dose (TCID50) ...... 52 2.17 Ex vivo reactivation assay...... 53 2.18 Statistical analyses ...... 54

3 Results ...... 55 3.1 Attempt to establish a novel autologous mouse model for herpesvirus infection ...... 55 3.1.1 Colony screen verified presence of MmusRHV1 and MCMV2 in wild mouse colonies ...... 56 3.1.2 Efforts to isolate MmusRHV1 in an in vitro setting ...... 57 3.1.3 Maintenance of wild mouse colony as a resource of MmusRHV1 at the HZI ...... 59 3.1.4 Establishment of a non-invasive blood PCR screen for MmusRHV1 ...... 63 3.1.5 Co-housing strategy and design to study natural transmission of MmusRHV1 ...... 65 3.2 The role of Toll-like receptors in the detection of MHV68 ...... 69 3.2.1 Identification of cell types responsible for the production of type I interferons upon MHV68 infection in vitro ...... 69 3.2.2 Characterization of the role of TLR expressed by pDC for sensing MHV68 in vitro ...... 72 3.3 Characterization of the role of TLR in sensing MHV68 infection in vivo...... 78 3.3.1 Role of TLR in the detection of MHV68 after intranasal infection ...... 80 3.3.2 Role of TLR in the detection of MHV68 after intraperitoneal infection ...... 88 3.3.3 Role of TLR in the detection of MHV68 after intravenous infection ...... 93

3.4 Role of TLR in the establishment of latency and reactivation of MHV68 after intravenous infection ...... 103 3.4.1 The role of TLR in the establishment of latency ...... 105 3.4.2 The role of TLR in reactivation of MHV68 ...... 108

4 Discussion ...... 114 4.1 Establishment of an autologous model to study gammaherpesvirus infections ...... 114 4.1.1 Isolation of MmusRHV1...... 114 4.1.2 Characterization of the natural transmission of MmusRHV1 in Mus musculus ...... 116 4.2 Characterization of co-infections and its effect on herpesvirus latency in naturally infected mouse models ...... 120 4.3 Role of UNC93B and specifically TLR7 and 9 in the detection of MHV68 in vitro ...... 124 4.4 Role of TLR during MHV68 infection in vivo upon different routes of infection ...... 127 4.4.1 Intranasal infection ...... 127 4.4.2 Intraperitoneal infection ...... 131 4.4.3 Intravenous infection ...... 132 4.5 Role of TLR and gender for establishment of MHV68 latency ...... 138 4.6 Influence of gender and TLR on reactivation of MHV68 ...... 140

List of abbreviations ...... 146

List of tables ...... 150

List of figures ...... 151

List of references ...... 154

Appendix ...... 174

1. Introduction

1 Introduction 1.1 Herpesviridae

The Herpesviridae is a family of large double-stranded deoxyribonucleic acid (dsDNA) viruses that share a common morphology. A typical herpes virion consists of a core containing a linear dsDNA genome ranging from 124-295 kb in length which is packaged into an icosahedral capsid that is approximately 125 nm in diameter. The complex of the core and its capsid constitutes the nucleocapsid. An amorphous tegument layer which consists of proteins that are required for viral entry, replication and egress surrounds the nucleocapsid (Guo et al., 2010). The tegument is enclosed by an envelope containing glycoprotein spikes on its surface. These glycoproteins play an important role in viral attachment and entry via cell surface receptors (Grunewald et al., 2003; Heldwein and Krummenacher, 2008) (Figure 1).

Figure 1: Herpesviridae virion structure. Herpesviral dsDNA is enclosed by an icosahedral capsid. A protein layer comprises the tegument located between the capsid and envelope which consists of glycoproteins necessary for attachment and entry.

The herpesvirus lifecycle can be divided into four stages: primary infection, lytic replication, latency and reactivation. Herpesviruses employ multiple cellular receptors for entry, and the interaction with cellular receptors is mediated by viral glycoproteins embedded in the lipid envelope of the virion (Fields Virology, 6th edition). Herpesviruses enter cells by either fusion of the virion envelope with the plasma membrane, or endocytosis (Nicola et al., 2003). Lytic replication results in the de novo production of virions that can infect other cells and hosts. This phase of the herpesvirus life cycle involves regulated viral gene expression, replication of the viral genome, assembly of virions, egress and transmission. A defining property of members of the herpesvirus family is their ability to enter a dormant state called latency (Fields Virology, 6th edition). In cells harboring latent virus, the viral genomes form a closed circular molecule called

1. Introduction

the episome, and only a small subset of viral genes is expressed. The pool of latently infected cells serves as a reservoir of viral persistence, and no infectious progeny are released unless the virus reactivates in latently infected cells under specific physiological conditions like stress or illness (Virgin HW and Speck SH, 1999).

Herpesviruses are divided into three subfamilies: Alphaherpesvirinae, Betaherpesvirinae and Gammaherpesvirinae. Gammaherpesviruses are further sub-divided into rhadinoviruses and lymphocryptoviruses (Pellet and Roizman, 2007). To date, 8 human herpesviruses (HHV) that naturally infect humans have been identified: herpes simplex virus type 1 (HSV-1) also known as HHV-1, herpes simplex virus type 2 (HSV-2) also known as HHV-2, varicella-zoster virus (VZV) also known as HHV-3, Epstein-Barr virus (EBV) also known as HHV-4, human cytomegalovirus (HCMV) also known as HHV-5, human herpesvirus 6 (variants A and B) and 7 (HHV-6A, HHV-6B, HHV-7) and Kaposi’s sarcoma-associated herpesvirus (KSHV) also known as HHV-8 (Fields Virology, 6th edition). Over 200 non-human vertebrate herpesviruses have been identified (Ehlers et al., 2008).

Members of the alphaherpesvirinae have a wide host range, a relatively short reproductive cycle and the ability to establish latent infections primarily in sensory ganglia (Fields Virology, 6th edition). Betaherpesvirinae on the other hand have a restricted host range and long reproductive cycles. Infected cells become enlarged (cytomegaly) and the virus can establish latency in secretory glands, lymphoreticular cells, kidneys and other tissues (Sinzger et al., 2008). Similar to betaherpesvirinae, members of the gammaherpesvirinae family exhibit a limited host range and latency is established in the lymphoid tissue (Fields Virology, 6th edition).

1.1.1 Betaherpesvirinae

Cytomegaloviruses belong to the betaherpesvirus family and can establish a mostly asymptomatic, lifelong infection in immunocompetent individuals. HCMV is a betaherpesvirus that has a worldwide distribution. In developing countries, the seroprevalence reaches 70-90%, while lower rates of 30-40% found in developed countries (Ramanan and Rezonable, 2013; Pembrey et al., 2013). The virus can be transmitted by direct contact with body fluids such as saliva, urine and breastmilk, by sexual contact, via blood transfusion or organ transplantation (Fields Virology, 6th edition). HCMV can infect a wide range of cells including fibroblasts, endothelial and epithelial cells, stromal cells and macrophages (Sinzger et al., 2008). HCMV- related diseases can manifest in immunocompromised individuals after solid organ transplantation and hematopoietic cell allografts, in individuals undergoing immunosuppressive

1. Introduction

therapies, and in people with acquired or genetic immunodeficiency (Fields Virology, 6th edition). Lytic infection of HCMV in immunocompromised individuals can result in clinical syndromes including pneumonitis, encephalitis, colitis and retinitis (Hutter et al., 1989). During pregnancy, an active CMV infection can be transmitted to the fetus (Kenneson and Cannon, 2007) and infection of the fetus during the first and second trimester can result in permanent damage such as microcephaly, hearing loss, hepatitis, damage to the central nervous system and - in very few cases - death (Manicklal et al., 2013; Cannon et al., 2011; Swanson and Schleiss, 2013).

Cytomegaloviruses are extremely species specific (Kim and Carp, 1971). In order to better understand HCMV pathogenesis and the role of the immune system in its detection, it is important to study CMV infection in animal models. CMV infection has been studied in mouse, rat, guinea pig and non-human primate models. Among these models, murine cytomegalovirus (MCMV) is by far the best studied.

Murine cytomegalovirus

Murine cytomegalovirus (MCMV) is used for infection studies in its natural host the house mouse (Mus musculus), and is known to cause acute and latent infections (Hudson et al., 1979). MCMV has a similar viral life cycle and cell tropism as HCMV (Tsutsui et al., 2005; Reddehase et al., 2002). Two MCMV strains are currently studied in detail: the less virulent, laboratory-passaged Smith strain, which was isolated in the laboratory of Margaret Smith in 1954 (Smith et al., 1954) and the more virulent K181 strain, which was isolated by the group of June Osborne in the 1970s (Hudson et al., 1988). Both strains have different phenotypes during infection (Smith et al., 2008). The passage histories of the Smith and K181 strain of MCMV are ambiguous, and the mouse strain from which they were derived has not been clearly recorded, making it difficult to track changes to the genome (Smith et al., 2008). However, it is noteworthy that the outbred strain of Swiss mice used during that period was of the Mus musculus background (Frazer et al., 2007; Yang et al., 2007).

Sequencing of wild MCMV strains (WP15B and G4) isolated from mice captured in the wild and comparison to the K181 laboratory strain sequence have led to the identification of several sequence variations in the m157 gene that can directly ligate the natural killer (NK) cell-activating receptor, Ly49H (Brown et al., 2001; Smith et al., 2002; Scalzo et al., 1990). This difference in sequence is presumably responsible for better replication of the WP15B and G4 strains in C57BL/6 spleens and livers when compared to passaged laboratory strains (Smith et al., 2008).

1. Introduction

Mus musculus cytomegalovirus 2

With the help of a consensus PCR with degenerate primers, Teterina and colleagues detected a novel betaherpesvirus in organ samples from inbred Mus musculus captured in the wild (Teterina et al., 2009). Consensus PCR on lung samples that were co-cultured with L929 cells revealed the presence of DNA polymerase (DPOL) and glycoprotein B (gB) sequences of an unknown murine betaherpesvirus. Since murine cytomegalovirus was already designated as MCMV, the novel virus was tentatively named Mus musculus cytomegalovirus 2 (MmusCMV-2). Based on the available sequence (3421 bp), MmusCMV-2 is closely related to the English isolate of rat cytomegalovirus (RCMV-E) (Teterina et al., 2009) (Figure 2).

Figure 2: Phylogenetic analysis of MCMV2. A phylogenetic tree was constructed using nucleic acid sequences encoded by the gB-DPOL segments of MCMV2 and those of known mouse, rat and human cytomegaloviruses. A multiple alignment of 3.4 kb was analyzed by neighbor-joining method. Results of the bootstrap analysis are indicated at the nodes of the tree. The original study and construction of the phylogenetic tree was performed by Teterina and colleagues (Teterina et al., 2009).This figure only serves as an illustrative example of the published phylogenetic tree. HHV-6a: Human herpesvirus 6a; RCMV-E: rat cytomegalovirus English isolate; MCMV1: murine cytomegalovirus 1; MCMV: murine cytomegalovirus 2; RCMV-M: rat cytomegalovirus Maastricht isolate; HCMV: human cytomegalovirus.

1.1.2 Gammaherpesvirinae

Gammaherpesviruses are lymphotropic viruses and are associated with the development of lymphoproliferative diseases and lymphomas (reviewed by Barton et al., 2011). Gammaherpesviruses are divided into two genera – lymphocryptoviruses and rhadinoviruses (Lacoste et al., 2010).

1. Introduction

Epstein-Barr virus

The genus lymphocryptovirus include HHV-4, also known as Epstein-Barr virus (EBV). EBV is one of the most common human viruses and distributed worldwide. EBV was the first gammaherpesvirus identified, and was discovered in cultured Burkitt’s lymphoma cells from an African patient in 1964 by Anthony Epstein and Yvonne Barr (Epstein et al., 1964). Over 90% of the world’s adult population is seropositive for EBV. Infection with EBV occurs mainly through contact with body fluids, primarily saliva. It can also spread during sexual contact, blood transfusions or organ transplantation (reviewed by Hjalgrim et al., 2007). Many people become infected with EBV during childhood, and usually EBV infections cause only very mild symptoms, or no clinical symptoms at all. However, during adolescence or later, primary infection of EBV can cause infectious mononucleosis. People with a compromised immune system can develop severe illness caused by EBV infection. Infection with EBV can cause lymphoproliferative disorders represented by a heterogeneous spectrum of diseases including Burkitt's lymphoma, Hodgkin's lymphoma and nasopharyngeal carcinoma (Ahmed and Baiocchi, 2016). These diseases share a degree of complexity involving lytic and latent viral expression and host factors like immunity, signal transduction and epigenetics. Therefore, it is essential to evaluate the pathogenesis of EBV and examine various experimental therapeutic strategies by using in vivo models of the disease.

Kaposi’s sarcoma-associated herpesvirus

The genus rhadinovirus (RHV) includes HHV-8 also known as Kaposi’s sarcoma-associated herpesvirus (KSHV). Classic Kaposi’s sarcoma (KS) was first described by Moritz Kaposi in 1872 as a skin cancer affecting elderly men of Ashkenazi origin in Vienna, Austria. In 1981, physicians in the United States of America (USA) observed an epidemic of KS among young men who have sex with other men (CDC, 1981). In 1994, DNA sequences of a new gammaherpesvirus were discovered in KS lesions (Chang et al., 1994). KS can be categorized into four epidemiological forms: (i) classic KS affects elderly men of eastern European or Mediterranean ancestry, (ii) endemic KS exists in parts of Africa, (iii) iatrogenic KS develops in immunosuppressed individuals and (iv) epidemic acquired immune deficiency syndrome (AIDS)- KS is a major AIDS-defining malignancy (reviewed by Mesri et al., 2010). KS incidence is 1 in 100,000 in the general population, while in human immunodeficiency virus (HIV)-infected individuals it is as high as 1 in 20 (Gallo et al., 1998). KSHV infection has also been linked to two lymphoproliferative disorders malignancies called primary effusion lymphoma (PEL) and

1. Introduction

multicentric Castleman’s disease (MCD), (Cesarman et al., 1995; Soulier et al., 1995; Moore and Chang, 2010).

The presence of anti-KSHV antibodies in infants suggests that transmission of KSHV from mother to child is likely (Wilkinson et al., 1999). A study of mothers and their children (<10 years old) found that about 30% of the children of KSHV-seropositive mothers were themselves HHV- 8 seropositive, whereas none of the children of KSHV-seronegative mothers were HHV-8 seropositive (Bourboulia et al., 1998). The steady increase in the prevalence of HHV-8 throughout childhood suggests that transmission of the virus from person to person, via nonsexual routes, may occur (Wilkinson et al., 1999; Mayama et al., 1998). In classic KS, saliva is the most likely route of transmission (He et al., 1998). In addition to possible sexual transmission of the virus, saliva is also thought to be the transmission route of KSHV in HIV- infected homosexual men (Pauk et al., 2000; Martro et al., 2007).

Due to strict species specificity animal models are used to study gammaherpesvirus infection. Rhesus Rhadinovirus (RRV) (Orzechowska et al., 2008) is used as a primate model to study gammaherpesvirus infection. RRV is closely related to KSHV, and in Rhesus macaques co- infection with simian immunodeficiency virus (SIV) is associated with the development of B-cell hyperplasia and lymphadenopathy, which resembles multicentric Castleman’s disease. Orzechowska and colleagues noted that immunocompetent animals infected with RRV exhibit viremia 2 weeks post infection, followed by a period where the virus is undetectable until the animals are rendered immunodeficient by infection with SIV. The advantage of a nonhuman primate model is that a genetic system exists to investigate the roles of specific viral open reading frames (ORFs) in different aspects of infection, for example, latency, persistence, and disease progression (Orzechowska et al., 2008).

Murid Herpesvirus 4 type 68 (MHV68)

Murid herpesvirus 4 type 68 (MuHV-4 type 68) also known and designated as MHV68 throughout this study, is a rhadinovirus used in small mammal studies. MHV68 is used extensively to study gammaherpesvirus infection and was first isolated from the bank vole (Myodes glareolus) in Slovakia. Its natural host was later found to be the wood mouse (Apodemus sylvaticus) (Blasdell et al., 2003; Hughes et al., 2010). Consistent with the classification of MHV68 as a gammaherpesvirus, it has been associated with lymphoproliferative disease in MHV68-infected immunocompetent mice, and immunosuppression by administration of cyclosporine A has been shown to enhance frequency of disease (Sunil-Chandra et al., 1994). Comparison of genomic organization of MHV68 with

1. Introduction

other gammaherpesviruses revealed that MHV68 has virus specific ORFs interspersed with co- linearly arranged genes. It is estimated that MHV68 encodes at least 80 gene products, of which 63 are co-linear and homologous to KSHV (reviewed by Simas and Efstathiou, 1998).

Upon intranasal infection with MHV68, the primary site of virus replication is the lung, though the virus can be detected in spleen, thymus, mesenteric lymph nodes, whole blood, heart, kidney and adrenal glands (Sunil-Chandra et al., 1992). Acute infection in the lung clears within 12 days post infection and is followed by the establishment of latent infection in the lymphoid tissue. B cells constitute the major latent reservoir of MHV68 (Sunil-Chandra et al., 1992). The establishment of latency in lymphoid tissue is characterized by splenomegaly (Bowden et al., 1997). Upon maturation of the B-cell response, germinal centers regress and there is a decrease in number of latently infected cells (Nash et al., 1994).

During 30 years of research, transmission of MHV68 had not been observed in captivity (Barton et al., 2011; Nash et al., 2001) until 2013 when François and colleagues reported sexual transmission of MHV68 between laboratory mice (François et al., 2013). The study used luciferase-expressing MHV68 and whole body imaging to investigate potential excretion sites of MHV68 in infected laboratory mice. The study identified genital excretion of MHV68 21 days after intranasal (i.n.) infection in female mice. The excretion occurred at the external border of the vagina and was dependent on the presence of estrogen. However, the study did not observe an association between MHV68 vaginal excretion and vertical transmission of MHV68 to the litter or horizontal transmission to other female mice. In contrast, efficient virus transmission to naïve males after sexual contact was observed. In vivo imaging also revealed that MHV68 first replicated in the penile epithelium and corpus cavernosum before spreading to draining lymph nodes and the spleen (François et al., 2013).

Mus musculus rhadinovirus 1

The fact that the natural host of MHV68, the wood mouse, and M. musculus are separated by several million years of evolution (François et al., 2010) poses certain limitations to the use of MHV68 as an animal model to study natural gammaherpesvirus infections.

In order to identify unknown rodent-associated herpesviruses, Ehlers and colleagues captured Mus musculus and 14 other rodent species from several locations in Germany, the United Kingdom, and Thailand and used degenerate PCR to identify herpesviruses that naturally infect them. During the study, 38 novel rodent herpesviruses were detected. 31% of all tested samples were positive for a novel murine rhadinovirus which was tentatively named Mus musculus

1. Introduction

rhadinovirus 1 (MmusRHV1). The data confirmed MmusRHV1 to be the first gammaherpesvirus that naturally infects Mus musculus, as wood mice, bank vole and field voles (Microtus agrestis) were tested negative for this virus (Ehlers et al., 2007).

MmusRHV1 displays 45% sequence identity on the nucleotide level and 55% sequence identity on the amino acid level with MHV68 (Figure 3). To date, 8 kbp of MmusRHV1 have been sequenced. This sequence comprises ORF6 which is the major DNA-binding protein (MDBP), ORF7, glycoprotein B (gB) encoded by ORF8, and the DNA polymerase (DPOL) encoded by ORF9. Of the 8 kbp, 3432 bp of the DPOL and gB gene sequence have been published (Ehlers et al., 2007).

Figure 3: Phylogenetic analysis of MmusRHV1. A phylogenetic tree was constructed using the nucleic acid sequences of gB-DPOL segments of MmusRHV1 (depicted in red) and MHV68 (depicted in blue) and those of other known gammaherpesviruses. The results of bootstrap analysis (100-fold) are presented to the right of the second vertical divider. The original study and construction of the phylogenetic tree was performed by Ehlers and colleagues (Ehlers et al., 2007).This figure only serves as an illustrative example of the published phylogenetic tree. EBV: Epstein Barr virus; CalHV-3: Calitrichine herpesvirus 3; KSHV: Kaposi’s sarcoma-associated herpesvirus; RRV: Rhesus monkey rhadinovirus; MHV68: murine gammaherpesvirus 68; BoHV-4; bovine herpesvirus 4; HVS: herpesvirus saimiri; HVA: herpesvirus ateles; EHV-2: equine herpesvirus 2; MmusRHV1: Mus musculus rhadinovirus 1; AlHV-1: alcelaphine herpesvirus 1; PLHV-1: porcine lymphotropic herpesvirus 1.

1. Introduction

1.2 Host defense

The immune system consists of two major arms: the in-born innate immune system that confers protection within several hours upon infection, and the adaptive or acquired immune system that provides protection several days after infection. The adaptive immune response consists of humoral immunity, mediated by antibodies produced by B lymphocytes, and cell-mediated immunity, mediated by T lymphocytes. The adaptive immune system also creates a memory, making future responses against specific antigens highly efficient. The main components of the innate immune system are phagocytic leukocytes, dendritic cells, natural killer cells and the complement system (Janeway Immunology, 8th edition). Activation of the innate immune system leads to the induction of the adaptive immune system by the secretion of cytokines and upregulation of the expression of costimulatory molecules on antigen presenting cells (APC) (Schenten and Medzhitov, 2011).

Dendritic cells (DC) are APC that connect innate and adaptive immunity. They do so by initiating the adaptive immune response to invading pathogens (Banchereau and Steinman, 1998). DC arise from both myeloid and lymphoid progenitors within the bone marrow, and migrate via the blood to different tissues and lymphoid organs (Kupiec-Weglinski et al., 1988). DC are found in two distinct functional states. Immature DC are located in skin mucosa and other lymphoid organs, and phagocytose antigens. On the contrary, mature dendritic cells are poor phagocytes but are highly efficient stimulators of naïve T cell responses (Banchereau et al., 1998).

In the bone marrow the common myeloid precursors (CMP) develop into monocyte/dendritic cell precursors (MDP) and give rise to common monocyte precursors (cMOP) or common DC progenitors (CDP) (Figure 4). CMP are heterogeneous in cell surface expression of the FMS- like tyrosine kinase 3 (Flt3) receptor. Flt3 ligand (Flt3L) binds to the Flt3 receptor expressed on myeloid and lymphoid progenitors in the bone marrow, and induces their differentiation (Lyman et al., 1998). A majority of the DC subsets originate from CDP and develop into plasmacytoid DC (pDC) and pre-DC. Pre-DC circulate in the blood and give rise to conventional DC (cDC) (Figure 4).

1. Introduction

Figure 4: Murine dendritic cell development from bone marrow hematopoietic stem cells and lineage committed progenitors. cDC and pDC can be generated from early myeloid precursors (CMP) expressing the cell surface localized Flt3 receptor. CMP, common myeloid progenitors; MDP, monocyte/dendritic cells precursors; CDP, common DC progenitors; cMOP, common monocyte precursors; pDC, plasmacytoid dendritic cells; cDC, conventional dendritic cells. The figure is adapted from Brinkmann et al., 2015.

In vitro, CMP are heterogeneous in Flt3 expression, and are able to produce both cDC and pDC subsets upon stimulation with FLt3L (Wu and Liu et al., 2007; O’Keeffe et al., 2002). Granulocyte-macrophage colony-stimulating factor (GM-CSF) on the other hand favorably expands the myeloid DC subset in vivo (Pulendran et al., 1999). In the presence of GM-CSF, bone marrow myeloid cell precursors produce macrophages, granulocytes and cDC (Inaba et al., 1993). Though both GM-CSF and Flt3L play an important role in the production of cDC, GM- CSF alone can suppress FLt3L-driven pDC development, hence rendering Flt3L critical for the development of pDC (Wu and Liu., 2007).

Plasmacytoid dendritic cells (pDC)

Plasmacytoid dendritic cells were originally described as interferon-producing cells (IPC). Upon activation, pDC produce type I interferon (IFN) (reviewed by Liu et al., 2005), and are additionally endowed with a T-cell stimulatory function (Grouard et al., 1997). pDC are capable of secreting IFNα even in the absence of a positive feedback loop (Barchet et al., 2002). This is

1. Introduction

due to the constitutive expression of endogenous interferon regulatory factor (IRF) 7 which facilitates rapid production of type I interferon (Barchet et al., 2002). In addition to interferons, pDC secrete other cytokines like GM-CSF, tumor necrosis factor (TNF), IL-6, IL-8 and IL-12 (Bauer et al., 2001; Cella et al., 2000). Through the secretion of IL-6, pDC promote the differentiation of B cells into plasma cells, which leads to the production of pathogen specific antibodies (Jego et al., 2003; Poeck et al., 2004). Through the secretion of IL-12, pDC induce IFNγ secretion by NK and CD8+ T cells (Krug et al., 2004; Cella et al., 2000). pDC also produce chemokines enabling them to recruit leukocytes to sites of inflammation (Penna et al., 2001). To modulate and maintain normal production of IFN, pDC express an array of surface receptors as for example sialic-acid-binding immunoglobulin-like lectin H (Siglec-H), a DaP12-associated receptor that reduces IFN production both in vitro and in vivo (Blasius et al., 2006). pDC also express cell surface receptors such as B220, CD45RA and intermediate levels of CD11c (O’Keeffe et al., 2002).

Conventional dendritic cells (cDC) cDC can be divided into three subsets according to their expression of CD8α homodimer and the CD4 molecule: CD4+CD8α- cDC, CD4-CD8α+cDC and CD4-CD8α- cDC (Vremec et al., 1992; Vremec et al., 2000). CD8α+ cDC produce cytokines like TNF, IL-6 (Sathe et al., 2011) and IL- 12. Upon activation, CD8α+ cDC secrete large amounts of IL-12p70 (Hochrein et al., 2001) + which leads to TH1 priming of CD4+ T cells (Maldonado-Lopez et al., 1999). CD8α cDC also produce type I IFN (Hochrein et al., 2001). CD4+ cDC produce high levels of chemokines like macrophage inflammatory protein (MIP) 3α, MIP3β and chemokine (C-C motif) ligand 5 (CCL5) + + (Proietto et al., 2004). Contrary to the TH1 response induced by CD8α cells, CD4 cells induce - - a TH2 response (Maldonado-Lopez et al., 1999). CD4 CD8α cDC or double negative (DN) cDC cross-present exogenous antigens poorly, but are as efficient as CD4+ cells in direct major histocompatibility complex (MHC) class II presentation (Schnorrer et al., 2006).

Many studies have shown that differential expression of pattern recognition receptors by DC subsets provides them with diverse mechanisms of distinguishing and responding to pathogens (Fuchsberger et al., 2005). The differential expression of Toll-like receptors (TLR) on DC subsets isolated from mouse spleen is summarized in Table 1. The data are either mRNA levels determined by semi-quantitative PCR (Edwards et al., 2003, Krug et al., 2001; Okada et al., 2003) or proteomics (Luber et al., 2010) and, therefore are subject to a high degree of variations.

1. Introduction

Table 1: TLR expression pattern among various murine DC subsets Murine TLR cDC pDC Reference CD4+ CD8+ CD4- CD8- TLR1 ++ ++ ++ ++ Edwards et al., 2003; Krug et al., 2001; Edwards et al., 2003; Krug et al., 2001; TLR2 ++ ++ ++ ++ Okada et al., 2003; Luber et al., 2010 Edwards et al., 2003; Krug et al., 2001; TLR3 + +++ + - Okada et al., 2003; Luber et al., 2010 Edwards et al., 2003; Krug et al., 2001; TLR4 + + + +/- Okada et al., 2003 Edwards et al., 2003; Krug et al., 2001; TLR5 +++ + ++ + Okada et al., 2003 TLR6 +++ +++ ++ ++ Edwards et al., 2003; Krug et al., 2001; Edwards et al., 2003; Krug et al., 2001; TLR7 ++ + ++ +++ Okada et al., 2003; Luber et al., 2010 TLR8 ++ ++ ++ ++ Edwards et al., 2003; Krug et al., 2001 Edwards et al., 2003; Krug et al., 2001; TLR9 ++ ++ ++ +++ Okada et al., 2003; Luber et al., 2010 TLR11 - + - - Luber et al., 2010 TLR12 - + + + Luber et al., 2010 TLR13 - + - - Luber et al., 2010

1.2.1 Pattern recognition receptors

Danger or pathogen associated molecular patterns (DAMPs or PAMPs) can induce an immune response after their detection by germline-encoded receptors called pattern recognition receptors (PRR). Toll-like receptors (TLR) are one of the classes of PRR (Lemaitre et al., 1996; Medzhitov et al., 1997) and are located on the cell surface and endosomal compartments. They recognize nucleic acids and several components unique to bacteria, fungi and parasites. C-type lectin receptors (CLR) are transmembrane proteins, which function as PRR. They recognize carbohydrates and play an important role in the detection of fungal pathogens (Hardison et al., 2012). Retinoic acid-inducible gene-I (RIG-I)-like receptors (RLR) and cytosolic DNA sensors are localized mainly in the cytosol and are responsible for the recognition of nucleic acids from both microbial and host origin in the cytoplasm. RLR sense cytoplasmic RNA, while cytosolic DNA sensors recognize cytoplasmic DNA (Takeuchi et al., 2000). Upon ligand binding, PRR

1. Introduction

activate signaling cascades leading to production of type I IFN and proinflammatory cytokines (Takeuchi et al., 2010). Another group of PRR in the cytoplasm is a group of nucleotide-binding oligomerization domain (NOD) - like receptors (NLR), which sense PAMP and DAMP in the cytoplasm leading to the formation of inflammasomes. Caspase-1 is activated upon the formation of multiprotein complexes called inflammasomes and facilitates the cleavage of pro- IL-1β and pro-IL-18 into the mature forms of these cytokines that get secreted (Schroder et al., 2010).

1.2.1.1 Toll-like receptors Toll receptors are evolutionarily conserved between insects and humans (Anderson KV, 2000). Toll is a type I transmembrane receptor essential for the dorsal-ventral development in the Drosophila embryo and was first discovered by Christiane Nüsslein-Volhard. For this work Nüsslein-Volhard was awarded the Nobel Prize in 1995. Lemaitre and colleagues constructed a recessive mutation in Drosophila for immune deficiency (imd) and monitored the expression of antifungal peptide drosomycin. Interestingly, drosomycin remained fully inducible in imd mutants indicating the existence of a different pathways leading to the expression of antifungal peptide genes (Lemaitre et al., 1995). Lemaitre and colleagues generated Drosophila lines homozygous for both imd and Toll-signaling mutations. Expression of antifungal and antibacterial genes was severely affected in these double mutants. Thus, their study contributed to the idea that Toll- signaling is involved in the immune response (Lemaitre et al., 1996).

The first homologue of Toll that was discovered in mammalian cells was TLR4 (Medzhitov et al., 1997). The ligand of TLR4 was identified by genetic studies. A mutant strain of mice (C3H/HeJ) with a recessive autosomal mutation in the lps locus was found to be hyporesponsive to lipopolysaccharide (LPS), a cell wall component of Gram-negative bacteria (Poltorak et al., 1998). The corresponding chromosomal location to the lps locus in the human genome (chromosome 9q32–33) was found to be the same region to which human TLR4 has been mapped (Rock et al., 1998). Mice deficient of TLR4 behaved similar to C3H/HeJ mice and failed to respond to stimulation with LPS. Upon further examination, a single point mutation at amino acid residue 712 in the intracellular domain of the TLR4 gene of C3H/HeJ mice was identified as the cause of the hyporesponsiveness to LPS (Hoshino et al., 1999), indicating that binding of LPS to TLR4 is essential for its recognition.

TLR are type I transmembrane receptors, with an N-terminal domain which mediates recognition of PAMP, and a C-terminal cytoplasmic domain that mediates downstream signaling (Bell et al., 2003). Due to its shared homology with interleukin-1 receptor (IL-1R) the

1. Introduction

cytoplasmic domain of TLR is referred to as Toll/IL-1R homology (TIR) domain. TLR undergo dimerization upon binding of their agonists (Jin 2007; Kim et al 2007; Liu et al 2008). Ligand induced dimerization of the cytoplasmic TIR domains leads to the recruitment of the adaptor proteins myeloid differentiation primary response gene 88 (MyD88), Toll-interleukin 1 receptor [TIR] domain-containing adapter protein (TIRAP), TIR-domain-containing adapter-inducing interferon-b (TRIF) and TRIF related adaptor molecule (TRAM) (O’Neill et al., 2007). The MyD88 and TRIF signaling adapters activate signaling cascades leading to activation of the transcription factors NFκB and IFN regulatory factors (IRF) leading to the production of proinflammatory cytokines and type I IFN, respectively (reviewed by Kawai and Akira, 2007 (Table 2, Figure 5).

To date, there are 10 and 12 TLR identified in humans and mice, respectively (Takeda et al., 2003; Yamamoto and Takeda, 2010). TLR1-9 are conserved between humans and mice. Though initial studies suggested that TLR8 is not biologically active in mice (Heil et al., 2004) (Wang et al., 2007), more recent reports suggest that TLR8 is dynamically expressed, and functional in mice (Ma et al., 2006; Demaria et al., 2010). TLR10 is not expressed in mice due to retroviral insertion in the gene (Hasan et al., 2005) while TLR11, 12 and 13 are exclusively present in mice (Yamamoto and Takeda, 2010).

Bacterial components are recognized by cell surface TLR1, TLR2, TLR4, TLR5 and TLR6. TLR2 recognizes the most diverse set of PAMP such as peptidoglycans, lipoteichoic acid, lipoarabinomanan, lipoproteins from Gram-negative bacteria, yeast, spirochetes and fungi. Additionally, TLR2 recognizes hemagglutinin (HA) proteins from Bordetella pertussis (Asgarian- Omran et al., 2015), and wild-type measles virus (Bieback et al., 2002). The recognition of such diverse structures has mainly been attributed to its ability to heterodimerize with TLR1 and TLR6 (Ozinsky et al., 2000). TLR2 homodimerization has been proposed to occur in the absence of TLR1 and TLR6, but has not been observed with current techniques (Jin et al., 2007). TLR1 and TLR6 are present on the cell surface, and are required to be dimerized with TLR2 in order to be functional (Takeuchi et al., 2002). The TLR1/2 dimer recognizes outer surface protein A (OspA), the outer-surface lipoprotein of the pathogen Borrelia burgdorferi and synthetic lipoprotein, N-palmitoyl-S-dipalmitoylglyceryl (Pam3) Cys-Ser-(Lys)4 (CSK4)

(Pam3CSK4) (Takeuchi et al., 2001; Alexopoulou et al., 2002). The TLR2/6 dimer recognizes macrophage-activating lipoprotein 2 kD (MALP-2) which is derived from mycoplasma (Takeuchi et al., 2001). Human TLR10 is an orphan member of the TLR family and studies have indicated that TLR10 can not only homodimerize, but also heterodimerize with TLR1 and TLR2 (Hasan et

1. Introduction

al., 2005; Guan et al., 2010). Although a ligand for TLR10 remains to be identified, Regan et al identified TLR10 as a key mediator of the inflammatory response to Listeria monocytogenes in macrophages and intestinal epithelial cells (Regan et al., 2013).

TLR4 is a cell surface TLR, and recognizes bacterial LPS as well as envelope proteins from respiratory syncytial virus (RSV) and mouse mammary tumor virus (MMTV) (reviewed by Kawai and Akira, 2007) and synthetic monophosphoryl lipid A (MPLA) (Evans et al., 2003). TLR5 is essential for the inflammatory response to flagellin from Gram-negative bacteria (Hayashi et al., 2001) (Table 2).

Murine TLR11 and TLR12 are endosomal TLR capable of forming a heterodimeric complex. TLR11 and TLR12 are involved in the recognition of profilin protein of the protozoan parasite Toxoplasma gondii and induce the expression of IL-12 in DC (Raetz et al., 2013). In addition to the recognition of profilin, TLR11 recognizes flagellin of uropathogenic bacteria (Zhang et al., 2004) (Table 2).

TLR3, 7, 8, 9 and 13 recognize nucleic acids and are localized in the endoplasmic reticulum and endosomes (Broz and Monack, 2013). TLR3 recognizes double-stranded ribonucleic acid (dsRNA) that is produced during viral replication (Alexopoulou et al., 2001) or derived from RNA viruses (Akira et al., 2006). TLR3 also recognizes the dsRNA analog polyinosonic-polycytidylic acid (Poly I:C). TLR7 and TLR8 (human) recognize single-stranded RNA from viruses (Diebold et al., 2004). Synthetic ligands for TLR7 are polyuridylic acid (poly (U)), imidazoquinoline and resiquimod (Miller et al., 1999). TLR9 recognizes dsDNA and oligodeoxynucleotides that contain unmethylated CpG sequences (Kumar et al., 2009). TLR13 recognizes bacterial RNA (Hidmark et al., 2012), specifically a conserved 23s ribosomal RNA (rRNA) sequence that serves as the binding site of macrolide, lincosamide, and streptogramin group (which includes erythromycin) antibiotics in bacteria (Oldenburg et al., 2012). Hence, bacteria resistant to erythromycin escape recognition via TLR13 (Oldenburg et al., 2012). TLR13 is also involved in the recognition of vesicular stomatitis virus (VSV) (Shi et al., 2011) (Table 2).

1. Introduction

Table 2: Localization, ligands and adapter proteins of TLR Cell surface TLR Endosomal TLR TLR1/2 TLR2/6 TLR4 TLR5 TLR3 TLR7/8 TLR9 TLR11 and TLR12 TLR13 Un- Triacyl- Diacyl- 23s ribosomal Bacteria LPS, LTA Flagellin methylated Flagellin lipopeptide Lipopeptide,LTA RNA CpG DNA Envelope Viruses dsRNA ssRNA dsDNA VSV Proteins Parasites Zymosan Hemozoin Profilin Fungus Poly(U) Synthetic Malp2 LPS Imi- Pam CSK Flagellin Poly(I:C) CpG ORN Sa19 compounds 3 4 FSL-1 quimod R848

Pam3CSK4, N-palmitoyl-S-dipalmitoylglyceryl (Pam3) Cys-Ser-(Lys)4 (CSK4); Malp2, macrophage-activating lipopeptide 2;LPS, lipopolysaccharide; LTA, lipoteichoic acid; dsRNA, double-stranded RNA; Poly(I:C), polyinosinic-polycytidylic acid; ssRNA, single-stranded RNA; Poly(U), polyuridylic acid; dsDNA, double-stranded DNA; VSV, vesicular stomatitis virus.

1. Introduction

With the exception of TLR3, all TLR recruit the signaling adapter protein MyD88 (Medzhitov et al., 1998). TLR4 uses all four adapters, the signaling adapters MyD88 and TRIF as well as the sorting adapters TIRAP and TRAM (Horng et al., 2002; Hoebe et al., 2003; Yamamoto et al., 2003). TLR3 solely uses TRIF (O’Neill et al., 2007).

In MyD88-deficient mice NFkB activation and inflammatory cytokine induction upon stimulation of TLR5, TLR7 and TLR9 signaling is defective (Hemmi et al., 2002; Hoshino et al., 2002). However, the response is not disrupted in mice deficient of TIRAP, TRIF or TRAM, indicating that MyD88 is the only adaptor used by these TLR (Yamamoto et al., 2002; Horng et al., 2002) (Yamamoto et al., 2003; Hoebe et al., 2003; Yamamoto et al., 2003). TLR2, which also utilizes MyD88 as an adapter, requires TIRAP to bridge between MyD88 and TLR2 (Yamamoto et al., 2002; Horng et al., 2002). Based on the adapters utilized by the TLR, TLR signaling is divided into two pathways: MyD88-dependent and TRIF-dependent.

MyD88-dependent signaling of TLR

Via its death domain, MyD88 interacts with the death domains of the IL-1R-associated kinase (IRAK) family members IRAK1, IRAK2 and IRAK4 (Martin and Wesche, 2002). This association between IRAK and MyD88 triggers hyperphosphorylation of IRAK1 and other kinases, which in turn leads to MyD88 interacting with the downstream adaptor tumour necrosis factor (TNF) receptor-associated factor 6 (TRAF6) (Burns et al., 2000). TRAF6 recruits a protein kinase complex involving transforming growth factor-β-activated kinase-1 (TAK1) and TAK1 binding proteins (TABs) which leads to the activation of a distinct pathway involving the IκB kinase (IKK) complex. The IKK complex phosphorylates IκB which enables the translocation of NFκB to the nucleus and the activation of proinflammatory cytokine expression (Kanayama et al., 2004), (Medzhitov et al., 1997; reviewed by Kawai and Akira, 2007) (Figure 5). MyD88 is also capable of inducing the production of IFN via the formation of a multiprotein signaling complex including IRAK1, IRAK2, IRAK4, TRAF3 and TRAF6 (Shinohara et al., 2006). The formation of this complex leads to phosphorylation and activation of IRF7 (Shinohara et al., 2006) (Figure 5).

TRIF-dependent signaling of TLR

TRIF recruits the multiprotein signaling complex composed of TRAF6, the TNF receptor 1- associated protein DEATH domain protein (TRADD) and the E3 ubiquitin-protein ligase pellino homolog 1 (Pellino1). TRAF6 recruits a protein kinase complex involving TAK1 and TABs which leads to the activation of a distinct pathway involving the IKK complex. The IKK complex

1. Introduction phosphorylates IκB which enables the translocation of NFκB to the nucleus and the activation of proinflammatory cytokine expression to activate NFκB (Sato et al., 2003; Kawai and Akira, 2007) (Figure 5). TRIF is also capable of inducing the production of IFN via activation of TBK1 and the kinases IKKi and IKKγ (also called NF-kappa-B essential modulator (NEMO)) via TRAF3 (Figure 5). The formation of this complex leads to phosphorylation and activation of IRF3. Phosphorylated IRF3 translocates to the nucleus to induce the expression of type I IFN (Tseng et al., 2010).

1. Introduction

Figure 5: Schematic overview of TLR signaling. The adaptor protein MyD88 is used by all TLRs, except TLR3. TLR3 and TLR4 use the adaptor molecule TRIF which initiates the production of proinflammatory cytokines and activates a type I IFN response via IRF3. MyD88-dependent TLR initiate signaling cascades by activating IRAK and TRAF6 which lead to the production of proinflammatory cytokines via NFκB. Recruitment of MyD88 by endosomal TLR also activates the production of type I IFN via IRF7.

1. Introduction

1.2.2 UNC93B and TLR trafficking

By screening macrophages from C57BL/6 mice that were treated with the mutagen N-ethyl-N- nitrosourea for the defects in TLR function, Tabeta and colleagues identified mice that failed to produce normal amounts of TNF in response to the TLR3 stimulus Poly(I:C), the TLR7 stimulus resiquimod, and the TLR9 stimulus unmethylated DNA oligonucleotides bearing CpG motifs. Notably, the TNF response to the TLR4 agonist LPS was not reduced in these mice. The production of a homozygous mutant stock revealed a recessive mutant phenotype. A single point mutation at position 412 from histidine to arginine (H412R) located within the transmembrane domain 9 of UNC93B was noted to cause this particular phenotype in mice (Tabeta et al., 2006).

These mice were designated ‘triple D’ (3d) mice due to their defect in nucleic acid sensing by three endosomally located TLR (Tabeta et al., 2006). 3d mice were shown to be highly susceptible to infection with MCMV, Listeria monocytogenes, and Staphylococcus aureus (Tabeta et al., 2006). A defect in the exogenous processing of antigen was observed in 3d mice and cross-presentation of antigen for MHC class I activation was strongly inhibited by the 3d mutation, but the mechanism by which this occurs was not addressed by Tabeta and colleagues.

Tabeta and colleagues observed that UNC93B showed a localization pattern reminiscent of the endoplasmic reticulum (ER) (Tabeta et al., 2006). Subsequent studies revealed murine UNC93B to be a glycosylated ER-resident protein predicted to have 12 transmembrane domains with both N- and C-terminus of the protein present in the cytoplasm (Brinkmann et al., 2007) (Figure 6).

1. Introduction

Figure 6: Model of the UNC93B protein. The UNC93B protein is composed of 598 amino acids, with both the N- and C-terminus predicted to localize in the cytoplasm. UNC93B is predicted to have 12 transmembrane domains. Based on this prediction, the single point mutation at position 412 from histidine to arginine (H412R) is located within the transmembrane domain 9 (●) and causes the 3d phenotype in mice. There are two predicted N-linked glycosylation (*) sites (N251HT and N272KT). Figure adapted from (Brinkmann et al., 2007).

Preparative immunoprecipitation and mass spectrometry analysis identified interaction of TLR3, 7, 9 and 13 with WT, but not mutant UNC93B. Additional biochemical analyses confirmed the interaction of WT UNC93B and TLR3, 7 and 9, and the use of chimeric TLR revealed that the transmembrane domains of TLR3 and 9 are responsible for binding to UNC93B (Brinkmann et al., 2007). In WT cells, TLR7 and TLR9 traffic from the ER to the endolysosome when cells are stimulated with synthetic ligands. However, in DC of 3d mice, TLR7 and TLR9 do not exit the ER and fail to produce TNF when stimulated with imiquimod or CpG. This phenotype could be corrected upon ectopic expression of WT UNC93B. This suggested that UNC93B is involved in trafficking of nucleic acid sensing TLR from the ER to the endolysomal compartment (Kim et al., 2008). Further studies showed that TLR7 and TLR9 traffic through the general secretory pathway before they reach the endosome (Ewald et al., 2008; Ewald et al., 2011). In the endolysosome, for TLR9 signaling to occur, the ectodomain of TLR9 must be cleaved by cathepsins (Edwald et al., 2008; Park et al., 2008). Though proteolytic cleavage is not necessary for binding of TLR9 to its ligand, only processed TLR9 recruits the adapter protein MyD88. Similar processing mechanisms have been shown for TLR3 and TLR7 (Garcia- Cattaneo et al., 2012; Edwald et al., 2008; Qi et al., 2012).

UNC93B exhibits a disparity in the trafficking of TLR7 and TLR9. In cells expressing UNC93B carrying the point mutation D34A in the N terminus of UNC93B, the TLR7 response is increased due to a stronger interaction between TLR7 and UNC93B. TLR9 responses are reduced in presence of UNC93B D34A. Mice homozygous for the D34A mutation develop spontaneous

1. Introduction lethal autoimmunity due to enhanced TLR7 responses and diminished TLR9 responses (Fukui et al., 2009; Fukui et al., 2011).

There is further evidence for differential regulation of TLR trafficking. Trafficking of TLR7 is independent of the adaptor protein-2 (AP-2), with TLR7 recruiting AP-4 directly for transport from the ER to the endosomes (Lee et al., 2013) (Figure 7A). AP-3 is reported to target TLR7 and TLR9 to lysosome related organelles specialized for type I IFN induction, and UN93B and TLR9 trafficking to these compartments is impaired in cells deficient of AP-3 (Sasai et al., 2010) (Figure 7). UNC93B-TLR9 complexes can be internalized from the cell surface to the endolysosome with the help of UNC93B-dependent recruitment of AP-2 at the cell surface (Lee et al., 2013). Mutating Y539A in UNC93B disrupts the interaction between UNC93B and AP-2 and in turn leads to the accumulation of TLR9 at the plasma membrane (Figure 7B).

In addition to TLR3, 7 and 9, confocal imaging analysis demonstrated that UNC93B colocalizes with TLR8 in HeLa cells (Itoh et al., 2011). In HEK93T cells TLR11 and 12 were shown to colocalize with UNC93B in the ER (Andrade et al., 2013). Lee et al demonstrated based on the acquisition of Endo-H resistant glycans that CD4-TLR chimeric proteins for TLR3, 7, 9, 11 and 13 require UNC93B to exit the ER in HEK293T cells (Lee et al., 2013). Though signaling of cell surface TLR4 is not dependent on UNC93B (Tabeta et al., 2006), a recent study revealed that trafficking of cell surface localized TLR5 is dependent on UNC93B. TLR5 physically interacts with UNC93B, and cells from 3d mice or UNC93B-/- mice lack TLR5 at the cell surface and fail to secrete cytokines upon stimulation with the TLR5 agonist flagellin (Huh et al., 2014).

1. Introduction

Figure 7: UNC93B differentially regulates trafficking of TLR7 and TLR9 to endosomal compartments. (A) The TLR7-UNC93B complex directly traffics from the ER to the endolysosome. This process is dependent on AP-4 and facilitates binding of TLR7 to ssRNA in the endolysosome. TLR9 uses the secretory pathway and traffics together with UNC93B via the cell surface to reach the endolysosomal compartment. (B) UNC93B recruits AP-2 to mediate endocytosis and endosomal delivery of the UNC93B- TLR9 complex. In the endolysosome, TLR9 binds to dsDNA and induces a proinflammatory cytokine response via NFκB. AP-3-mediated delivery of TLR9 from the endolysosomal to the lysosomal compartment leads to the induction of type I IFN production via activation of IRF7. Introduction of a single point mutation H412R (3d) in UNC93B disrupts the interaction between UNC93B and endosomal TLR leading to retainment of both proteins in the ER.

1. Introduction

UNC93B and infection Autosomal recessive deficiency of UNC93B in human patients has been linked to development of herpes simplex virus-1 (HSV-1) encephalitis (HSE). Peripheral blood mononuclear cells (PBMC) from patients with autosomal recessive deficiency in UNC93B show impaired cellular IFNα/β and -λ responses but produce normal levels of IFNγ, TNF, IL-1β and IL-6 upon stimulation with HSV-1. As in 3d mice, cells from UNC93B deficient patients are defective in signaling via TLR3, 7, 8 and 9 (Casrouge et al., 2006). Deficiency of TLR3 in particular has been associated with HSE recurrence with 66% of TLR3-deficient patients having at least 1 relapse of HSE (Lim et al., 2014).

Upon infection with MCMV, 3d mice show a reduced production of IL-12, TNF, IFN-α and IFN-γ, transient hepatitis, and impairment in viral clearance (Crane et al., 2012). Using N-ethyl-N- nitrosourea mutagenesis, Lafferty et al created a novel mutation in Unc93b1 that results in an in-frame deletion of exon 4 following transcript splicing. Mice with a loss of function mutation are designated as loss of endosomal TLR response (Letr) mice. Unc93b1Letr/Letr mice intranasally infected with 400 PFU of influenza A/PR/8/34 (H1N1) show delayed higher viral load in the lung (Lafferty et al., 2014). Inflammation and mortality during ssRNA coxsackievirus strain B serotype 3 (CVB3) infection is also regulated by UNC93B dependent endosomal TLR signaling (Lafferty et al., 2015). Unc93b1Letr/Letr mice displayed higher viral load in organs, increased inflammation, necrosis and fibrosis in the cardiac tissue during acute CVB3 infection (Lafferty et al., 2015).

Besides increased susceptibility to viral infections, 3d mice or mice deficient of UNC93B show enhanced susceptibility to parasitic infection. Deficiency of functional UNC93B abolishes

TLR11- dependent secretion of IL-12 and attenuates TH1 responses against Toxoplasma gondii (Pifer et al., 2011). 3d mice are susceptible to infection with Trypanosoma cruzi, and absence of TLR7, which recognizes parasite RNA, leads to reduced IL-12 and IFNγ production (Caetano et al., 2011).

1.2.3 TLR-mediated detection and control of MHV68

In vitro detection of MHV68 by TLR

Viral proteins, DNA and RNA are the three different classes of herpesvirus PAMPs that are recognized by TLR (Paludan et al., 2013). Upon recognition of stimuli by TLR, signaling via various adapters and activation factors lead to the production of proinflammatory cytokine and IFN. Primary bone marrow-derived macrophages (BMDM) produce TNFα and IL-6 in response to TLR agonists such as CpG B and MCMV, but fail to produce detectable amounts of either of

1. Introduction the cytokines in response to MHV68 infection at MOI 1 and MOI 2 (Bussey et al., 2014). Mouse embryonic fibroblasts (MEF) express functional TLR, and secrete a range of cytokines in response to TLR stimulation with specific ligands (Kurt Jones et al., 2004). Upon MHV68 infection with an MOI of 1, WT MEF produce the inflammatory cytokine IL-6, and additionally exhibit a weak IFNα response (Michaud et al., 2010). In contrast, MyD88-/- MEF show complete abolishment of IL-6 and IFNα production upon infection with MHV68 (Michaud et al., 2010). TLR2 signals via the MyD88-dependent pathway, and TLR2-/- MEF show a reduction in IL-6 and IFNα secretion when compared to WT MEF, suggesting a possible role for TLR2 in the detection of MHV68 in MEF (Michaud et al., 2010). TRIF-/- MEF on the contrary still produce IFNα in response to MHV68 infection, indicating that the MHV68 induced production of IFNα in MEF occurs via the MyD88-dependent, but not the TRIF dependent TLR pathway (Michaud et al., 2010).

Macrophages and MEF are not major IFN producing cells, whereas DC as described in section 1.2 are the major IFN producing cells. In addition to IFN, FLDC consisting of both pDC and cDC from WT mice produce IL-12 and IL-6 in response to MHV68 infection (Guggemoos et al., 2007). FLDC from WT and TLR9 mice were infected with MHV68 at an MOI of 0.1 and 18 hours after infection, supernatants were analyzed for the presence of IL-12 p40/p70, IL-6 and IFNα. MHV68 infected TLR9-/- FLDC failed to produce IL-12 and IL-6, and showed a partial abolishment of IFNα levels. These results strongly advocate the role of TLR9 in the detection of MHV68 in DC (Guggemoos et al., 2007).

Detection of lytic replication of MHV68 by TLR in vivo

Intranasal (i.n.) infection with 103 PFU of MHV68 does not affect acute replication of the virus in the lungs of TLR3-/- mice (Gargano et al., 2008). Five days post i.n. infection with 105 PFU of MHV68, higher titers are observed in the lungs of TLR2-/- mice, accompanied by slightly lower IL-6 levels in the lung homogenates of TLR2-/- mice when compared to controls, which might indicate a role of TLR2 in the early defense against MHV68 after i.n. infection (Michaud et al., 2010). TLR9-/- mice infected intraperitoneally with 5 x 105 PFU of MHV8 display higher splenic titers when compared to WT mice 6 days post infection (Guggemoos et al., 2007). This increase in lytic viral load was observed in the spleen after intraperitoneal (i.p.) infection, but not in the lungs of TLR9-/- mice after i.n. infection, suggesting that TLR9 may be important in the detection of MHV68 in the spleen after i.p. infection (Guggemoos et al., 2007).

Two studies have analyzed MHV68 acute infection in mice lacking the signaling adapter MyD88, which is crucial for signaling of all TLR except TLR3. Upon i.n. infection with 103 PFU of MHV68,

1. Introduction replication of the virus in the lungs of MyD88-/- is not affected (Gargano et al., 2008). However, upon i.n. infection with 105 PFU of MHV68, MyD88-/- mice show an increase in titer in the lung on days 3 and 5 post infection, and reduced IL-6 levels in lung homogenates of MyD88-/- mice when compared to the controls (Michaud et al., 2010). This difference in the two studies might be attributed to the difference in viral dose or the use of different mouse strains.

MyD88 is essential for the induction of NFκB, and signaling via MyD88 leading to NFκB activation is defective in MyD88-/- mice (Hemmi et al., 2002; Hoshino et al., 2002). Mice infected intraperitoneally with 103 PFU of the recombinant MHV68-IκBαM.1, which expresses a constitutively active form of the NFκB inhibitor IκBα, display significantly lower viral titers in the spleens 9 days post infection when compared to spleens from mice infected with WT MHV68 (Krug et al., 2007). However, 7 days post i.n. infection with 103 PFU of MHV68-IκBαM.1, viral titers in the lungs and spleens of mice are comparable to titers from mice infected with the revertant virus (Krug et al., 2007; Gargano et al., 2008). These findings indicate that NFκB activation plays a role in the acute replication of MHV68 in the spleen post i.p. infection (Krug et al., 2007), but acute viral replication in the spleen and lungs post i.n. infection with 103 PFU are not affected by the absence of MyD88 or NFκB (Krug et al., 2007; Gargano et al., 2008).

Detection of MHV68 by TLR during latency in vivo

Signaling via TLR is an important factor in the control of the balance between latency and reactivation of MHV68 (Gargano et al., 2009). An activated TLR7 pathway plays an important role in the establishment of MHV68 latency (Haas et al., 2014). Haas and colleagues infected mice intraperitoneally with 105 PFU MHV68, and stimulated them with the TLR7 ligand R848 starting on day 1 post infection and continued treatment with R848 for 19 days post infection. On day 20, spleens were collected and splenocytes were analyzed for the presence of mRNA from MHV68 ORF73, which encodes latency associated nuclear antigen (LANA). Splenocytes from mice treated with R848 showed a 7-fold increase in MHV68 ORF73 expression when compared to untreated mice. This finding suggests that triggering TLR7 during the primary infection of MHV68 promotes the establishment of latency by increasing the reservoir size of latent MHV68 (Haas et al., 2014). Though the study showed the involvement of TLR7 in the establishment of MHV68, they did not observe an increased reactivation of MHV68 from TLR7-/- B cells after i.p. infection with 1 x 105 PFU of MHV68 (Haas et., 2014). A separate study has analyzed the role of TLR9 in the establishment of MHV68 latency, and MHV68 reactivation from latency. Upon i.p. infection with 5 x 104 PFU of MHV68, higher rates of reactivation were observed in splenocytes from TLR9-/- mice when compared to those from WT mice 17 days post

1. Introduction infection. In accordance with the observation, a higher viral load was observed in spleens from TLR9-/- mice when compared to WT mice (Guggemoos et al., 2007). This finding suggests that TLR9 plays an important role in the reactivation of MHV68.

Apart from its role during acute replication of MHV68, MyD88 also plays an important role in B- cell activation, germinal-center formation, and class switching in MHV68 infected splenocytes and contributes to the control of the establishment of MHV68 latency in B cells (Gargano et al., 2008). Upon i.n. infection with 103 PFU of MHV68, spleens from MyD88-/- mice show a decrease in the frequency of MHV68 viral genome positive B cells (Gargano et al., 2008). Gargano et al intranasally infected mice with 1000 PFU of MHV68, and 42 days post infection administered either PBS, 15 µg of LPS or 20 µg of CpG DNA intraperitoneally. Splenocytes from infected mice were analyzed on days 1, 3, 7 and 14 after treatment with TLR agonists. Treatment with LPS and CpG, which stimulate MyD88-dependent TLR 4 and 9, increased the frequency of activated splenic B cells 1 day after administration when compared to mice treated with PBS, and lead to significantly higher reactivation of MHV68 from latency 14 days post stimulation (Gargano et al., 2009).

NFκB which is downstream of TLR signaling has also been shown to be crucial for the establishment and maintenance of latent MHV68 infection (Haas et al., 2014). In vivo, MHV68- IκBαM.1 is impaired in its ability to establish latent infection (Krug et al., 2007). Additionally, the frequency of splenocytes harboring MHV68 genome 16 days post i.n. infection with 103 PFU of MHV68 is approximately 50 fold lower in mice infected with MHV68-IκBαM.1 than the viral load in splenocytes of mice infected with WT MHV68 (Krug et al., 2007). However, following i.p. infection with 103 PFU of MHV68-IκBαM.1, the frequency of peritoneal exudate cells (PEC) which include macrophages harboring MHV68-IκBαM.1 is nearly identical to the frequency of PEC harboring WT MHV68. This suggests that NFκB activation is dispensable for the establishment of MHV68 latency in macrophages in peritoneal exudates after i.p. infection, but is necessary for the establishment of latency and reactivation of the virus from splenocytes after i.n. infection (Krug et al., 2007).

Though the role of MyD88, TLR2, TLR7 and TLR9 have been addressed with respect to MHV68 infection, there have been no studies analyzing the role of endosomal TLR exclusively, or the synergistic role of both TLR7 and 9 in the detection and control of MHV68.

1. Introduction

1.3 Aims of the study

Herpesviruses are double-stranded DNA viruses that establish latent infections in their respective hosts. Murine gammaherpesvirus 68 (MHV68), originally isolated from bank voles, is currently the only gammaherpesvirus used to study gammaherpesvirus infections in mice. Though the natural host of MHV68 is the wood mouse, the virus is currently studied in the house mouse, Mus musculus. The fact that the house mouse and the wood mouse are separated by several million years of evolution, and that their respective herpesviruses coevolved with their hosts, poses certain limitations to infection studies with MHV68 in house mice. The possible differences of MHV68 with regards to infection kinetics in house mice compared to other gammaherpesviruses in their natural hosts might be attributed to the fact that MHV68 does not naturally infect the house mouse. Therefore, we postulated that a gammaherpesvirus that naturally infects house mice would be a valuable tool to study aspects of a natural infection in the host and serve as a novel tool for the research community. Two only recently discovered herpesviruses, namely the betaherpesvirus murine cytomegalovirus 2 (MCMV2) and the gammaherpesvirus Mus musculus rhadinovirus 1 MmusRHV1, that both naturally infect Mus musculus, were detected in mouse colonies originally captured in the wild. Isolation and sequencing of the complete genome of these two viruses has not yet been reported. One major aim of this study was the isolation and whole genome sequence analysis of MmusRHV1. As a side aspect of the project, we aimed to isolate and sequence the complete genome of MCMV2.

The second major aim was to study the role of Toll-like receptors (TLR) for the innate immune response to MHV68 infection in vitro and in vivo. Through this study, we wanted to analyze the individual roles of cell surface TLR2 and endosomal TLR7, 9 and 13 in the detection of MHV68. In addition, we addressed the synergistic role of endosomal TLR by using UNC93B-/- mice which lack functional TLR3, 7, 9, 11, 12 and 13, and mice lacking both TLR7 and TLR9, in the detection of MHV68. For the in vitro studies, type I interferon levels were analyzed upon MHV68 infection of different dendritic cell subsets. For the in vivo studies, the role of TLR during three different stages of MHV68 infection was assessed: (i) lytic replication, (ii) establishment of latency and (iii) reactivation from latency.

2. Materials and Methods

2 Materials and Methods 2.1 Buffers and solutions

FACS buffer, HBS, LB medium, lysis buffer, MACS buffer and PBS were prepared under sterile conditions with sterile filtered or autoclaved materials. All solutions were prepared using Milli-Q water. Table 3: Buffers and solutions Buffers and solutions Components Agarose gel 1% agarose in TAE buffer FACS buffer 2% Fetal calf serum (FCS) in PBS 20 mM HEPES HBS 150 mM NaCl

in H2O, pH 7.4 IFNα wash buffer 0.05% Tween 20 in PBS 10 g/L Tryptone LB medium 5 g/L Yeast extract 7 g/L NaCl

ddH2O, pH 7.5 10 mM Tris-HCl, pH 8.5

1.5 mM MgCl2 Lysis buffer (PCR) 1% NP-40

1% Tween 20 in H2O, pH 7.4 0.5% BSA MACS buffer 2 mM EDTA in PBS 137 mM NaCl 2.7 mM KCl

PBS 10 mM Na2HPO4 x 2 H2O 1.8 mM KH2PO4 in H2O, pH 7.4 40 mM Tris base 20 mM sodium acetate TAE buffer 1 mM EDTA

in ddH20 100 mM Tris-HCl, pH 8.5 200 mM NaCl Tail lysis buffer 5 mM EDTA 0.2% SDS TNFα wash buffer 0.01% Triton X-100 in PBS

2. Materials and Methods

2.2 Commercially available kits

Table 4: List of commercially available kits Kit Source Catalogue number DNeasy Blood & Tissue kit (50) Qiagen 69504 NucleoSpin Gel and PCR Clean-up Macherey-Nagel 740609.50 Pan B Cell Isolation Kit II, mouse Miltenyi Biotec GmbH 130-104-443 PCR Master Mix 2x Thermo Fisher Scientific K0172 Phusion Blood Direct PCR Kit Thermo Fisher Scientific F-547S Plasmacytoid Dendritic Cell Isolation Kit II Miltenyi Biotec GmbH 130-092-786 Premix Ex Taq Hot Start Version Takara Clontech RR030A QIAGEN Plasmid Purification Maxi Kit Qiagen 12963

2.3 Cloning of plasmid controls

In order to create positive controls for polymerase chain reactions (PCR), specific regions of murine cytomegalovirus 1 (MCMV1), Mus Musculus rhadinovirus 1 (MmusRHV1), murine cytomegalovirus 2 (MCMV2) and Mus musculus beta-2-microglobulin (B2M) from previously published sequences were amplified using primers listed in Table 5. MmusRHV1 and MCMV2 positive organs kindly provided by Bernhard Ehlers (Robert Koch Institute, Berlin) were used as starting material to clone MmusRHV1 and MCMV2. pGEM-T vector (Promega) was used for cloning of inserts. The PCR was performed with Thermus aquaticus (Taq) polymerase, which due to its non-template dependent terminal transferase activity adds a single 3’ deoxyadenonsine overhang to each of the PCR products. The 3’ A overhang facilitated high efficiency cloning of the PCR products into pGEM-T vector, which is a linearized vector with single 3’-deoxythymidine at both ends. The PCR products were cloned into pGEM-T vector using T4 DNA ligase (Promega). Plasmid maps for plasmids described in Table 5 were created using SnapGene software (GSL Biotech; available at snapgene.com) and are listed in the appendix. For the detection of MHV68 ORF50 in latently infected splenocytes, MHV68 ORF50 plasmid was used as a positive control. Helene Todt subcloned intronless full length MHV68 RTA (ORF50) in pCMV Flag. Intron-containing full length MHV68 RTA in pCMV Flag (VIMM code L237) and the intronless C-terminal RTA trunctation mutant pCMV Flag Rd1 (VIMM code L239) (Wu et al. 2001) were kindly provided by Ren Sun (University of California, Los Angeles). L237 and L239 were digested with BglII and XbaI. The 1637 bp fragment from L237, containing the C-terminal portion of RTA, was ligated using T4 DNA ligase to the 4786 nt fragment of L239, which contained an intronless N-terminal portion of RTA and the vector backbone. All constructs generated during this study are listed in Table 6. All constructs were transformed into chemically competent DH10B strain of E.coli. Resulting constructs in pGEM-T were verified by restriction digests with NcoI and NotI, and intronless RTA in pCMV Flag was verified by colony screen PCR using plasmid-specific primers (CMV for and hGH-PA rev) (Table 7). Correct plasmids were then prepared using the Plasmid Maxi Kit (Qiagen) according to the manufacturer’s instructions. All inserts were fully sequenced in both directions for verification; the oligonucleotides used for sequencing are listed in Table 7.

2. Materials and Methods

2.3.1 Bacteria

A chemically competent DH10B strain of Escherichia coli (genetoype: F- mcrA Δ (mrr-hsdRMS- mcrBC) Φ80dlacZΔM15 ΔlacX74 endA1 recA1 deoR Δ (ara,leu)7697 araD139 galU GalK nupG rpsL λ-) was used for cloning. Transformed bacteria containing plasmid insert where cultivated in LB medium, or grown on LB agar supplemented with 100 µg/ml ampicillin (Sigma Aldrich) overnight at 37°C.

2. Materials and Methods

Table 5: List of oligonucleotides used for cloning inserts into pGEM-T vector Oligo Accession numbers number Designation Sequence 5’  3’ Tm (°C) Region of published (VIMM) sequence (NCBI) 370 MmusRHV1-gB For TCGGGAGTATAACTATTACAC 54 464 bp – 975 bp of published AY854167 367 MmsRHV1- gB Rev GCGATTACGCAACCCTTCC 58.8 MmusRHV1 genome 403 MCMV1-gB For AGCGTCCATGATGTTTCGGA 57.3 716 bp – 1611 bp of published NC_004065.1 404 MCMV1-gB Rev GCGATACGTGACCACCAAGA 59.4 MCMV1 glycoprotein B (gB) 443 MCMV2-gB For GGTATAGGAGCGGTGTTGGG 61.4 604 bp – 1249 bp of published GU017485.1 444 MCMV2-gB Rev TTAACCCCGTTGCACCATCA 57.3 MCMV2 genome 449 B2M For TCAAGTATACTCACGCCACCC 59.5 3199 bp – 3905 bp of the NC_000068.7 450 B2M Rev AGACATCCAGGGGTTGTAAGC 59.7 published B2M gene

Table 6: List of constructs used as positive controls for polymerase chain reactions Code Antibiotic Construct Vector Description Insert length (VIMM) resistance Full length MHV68 ORF50 pCMV Flag MHV68 ORF50 L274 pCMV Flag Ampicillin 1752 bp (RTA), without intron Glycoprotein B region of pGEM-T-MCMV1 gB L141 pGEM-T Ampicillin 895 bp MCMV1 Glycoprotein B region of pGEM-T-MCMV2 gB L142 pGEM-T Ampicillin 645 bp MCMV2 pGEM-T-Mmus Beta -2 L144 pGEM-T Ampicillin beta-2-microglobulin 706 bp Microglobulin Glycoprotein B region of pGEM-T-MmusRHV1gB L140 pGEM-T Ampicillin 511 bp MmusRHV1

2. Materials and Methods

Table 7 : Oligonucleotides used for colony PCR and sequencing Oligo No. (VIMM) Designation Sequence 5’  3’ 384 SP6_pGEMT ATTTAGGTGACACTATAG 385 T7_Promoter_ pGEMT TAATACGACTCACTATAGG 376 CMV For CGCAATGGGCGGTAGGCGTG 377 hGH-PA rev CCAGCTTGGTTCCCAATAGA 987 MHV68 ORF50 1109 Rev AACTGGAACTCTTCTGTGGC

2.3.2 Determination of copy number to assess PCR sensitivity

In order to better assess the sensitivity of our PCR reactions, DNA from plasmid controls was serially diluted in order to have an exact number of viral copies per reaction. Copy number of plasmid DNA was calculated using the following formula:

푋 푛푔 ∗ 6.0221 푥 1023푚표푙푒푐푢푙푒푠/푚표푙푒 푁푢푚푏푒푟 표푓 푐표푝푖푒푠 (푚표푙푒푐푢푙푒푠) = 660푔 1 푥 109푛푔 (푁 ∗ ) ∗ 푚표푙푒 푔

Where X = total amount of plasmid in ng, N = length of dsDNA, 660 g/mole is the average molar mass of dsDNA (eu.idtdna.com). Serial 10-fold dilutions ranging from 108 viral copies/µl to 1 viral copy/µl were prepared using nuclease-free water.

2.4 Extraction of DNA from tissue samples

DNA from mouse organ samples was prepared using the DNeasy Blood & Tissue kit (Qiagen) according to the manufacturer’s instructions. Briefly, tissue samples were cut into small pieces and placed into 1.5 ml microcentrifuge tubes. 180 µl of tissue lysis buffer (Qiagen) and 20 µl of proteinase K (600 mAU/ml) (Qiagen) was added to the sample. The sample was mixed thoroughly by vortexing before being centrifuged. Samples were then incubated in a thermomixer at 56°C overnight. For samples from co-culture experiments were organ snippets, or homogenized organs from mice were co-cultured with permissible cell lines like M2-10B4 and NIH3T3, DNA was extracted from co-cultured cells by centrifuging up to 5 x 106 cells at room temperature (RT) for 5 min at 300 g, and resuspending cells in 200 µl PBS, before the addition of 20 µl proteinase K. Cultured cells and organ samples were vortexed for 15 s before the addition of 200 µl of lysis buffer (Qiagen) to the sample. Next, 200 µl of ethanol (96-100%) was added and samples were mixed thoroughly by vortexing. The 600 µl mixture was then pipetted into the DNeasy mini spin column placed in 2 ml collection tubes. Samples were then centrifuged at 6000 g for 1 min at RT, and the collection tube along with the flow-through was discarded. The DNeasy columns were then placed into new collection tubes before the addition of 500 µl of wash buffer 1 (Qiagen), and centrifuged at RT for 1 min at 6000 g. After discarding flow through, the column was placed into a new 2 ml collection tube, 500 µl of wash buffer 2 (Qiagen) was added and the columns in collection tubes were centrifuged for 3 min at 20,000 g to dry DNeasy membrane.

2. Materials and Methods

After centrifugation, columns were placed into 1.5 ml microcentrifuge tubes, and 100 µl of DNAse free water was added directly onto the DNeasy membrane. The samples were incubated at RT for 1 min and eluted by centrifugation at 6000 g for 1 min. DNA concentration was measured using a NanoDrop spectrophotometer (Thermo Scientific) at 260 nm. Purity of

DNA was determined by the ratio of the absorbance at 260 nm to that at 280 nm (A260/A280).

2.5 PCR analysis for the detection of MmusRHV1, MCMV1, MCMV2 and MHV68

PCR reactions were performed using 2x PCR Master Mix (Thermo scientific, #K0172) or Phusion Blood Direct PCR kit (Thermo scientific, F-547S) and T3 or Tgradient thermocyclers (Biometra). Amplified PCR products were visualized on 1% agarose gels.

2.5.1 Nested PCR to detect the presence of MmusRHV1, MCMV1 and MCMV2

MmusRHV1, MCMV1 and MCMV2 from organs or tissue culture samples were detected with nested PCR using primers as listed in Table 8. DNA was prepared as described in 2.4. Purified DNA was diluted to a final concentration of 100 ng/µl and used as a template for the first round of PCR amplification. The amplicon from the first reaction was used as the template for the second amplification round. The reactions were prepared as follows:

Round I Round II 2X PCR Master Mix (Thermo scientific) 13 µl 10 µl Forward primer (10 µM) 1 µl 1 µl Reverse primer (10 µM) 1 µl 1 µl Template (100 ng/µl) 1 µl 1 µl of 1st round amplicon

Nuclease free H2O 14 µl 7 µl

The nested PCR was carried out similarly for all targets using the following thermal cycling conditions, where X indicates the annealing temperature as listed in Table 8. Denaturation 2 min 95°C Amplification 30 s 95°C Annealing 1 min x°C 40 cycles Extension 1 min 72°C Final extension 15 min 72°C Cooling ∞ 4°C

2.5.2 Non-invasive blood PCR to screen for the presence of MmusRHV1

Up to 30 µl of blood was collected in EDTA coated microcentrifuge tubes from anesthetized mice through eye bleeds, and diluted in sterile PBS to reduce the total concentration of EDTA. Alternatively, up to 800 µl of blood was obtained by cardiac puncture and collected in heparin coated tubes (Sarstedt, 41.1504.005). For the detection of MmusRHV1, PCR no 1 and 2 were performed with primers as listed in Table 8.

2. Materials and Methods

The recommended starting amount of template (5% of total reaction) was added directly into the reaction. After the first round of PCR amplification, 1 µl of the amplicon was used as the template for the second round of PCR amplification. The PCR was prepared as follows:

Round I Round II 2X Phusion Blood PCR Buffer (Thermo scientific) 10 µl 10 µl Forward primer (10 µM) 1 µl 1 µl Reverse primer (10 µM) 1 µl 1 µl Phusion Blood II DNA polymerase 0.4 µl 0.4 µl Template 1 µl blood 1 µl (1st round amplicon)

Nuclease free H2O 6.6 µl 6.6 µl

The PCR was carried out using the following thermal cycling conditions:

Lysis of cells 5 min 98°C Denaturation 1 min 98°C Annealing 5 s x°C 40 cycles Extension 30 s 72°C Final extension 1 min 72°C Cooling ∞ 4°C

Where x indicates the appropriate annealing temperature as listed in Table 8.

2. Materials and Methods

Table 8: List of oligonucleotides used for polymerase chain reaction

Oligo PCR T Annealing Product size Construct No. Designation Sequence 5’  3’ m Nested PCR No. (°C) Temp ( X°C) (bp) (VIMM) MmusRHV1 370 MmusRHV1-gB For TCGGGAGTATAACTATTACAC 54.0 1 49 Round I 514 367 MmsRHV1- gB Rev GCGATTACGCAACCCTTCC 58.8 371 MmusRHV1-gB For CGCGACCTCGACAACAAC 58.2 2 52/56* Round II 314 366 MmsRHV1- gB Rev ACCTCCCGAGACTTACTC 56.0 783 MmusRHV1-Dpol For CGACGTCCCTCTCCGTTTT 59.5 3 55 Round I 477 784 MmusRHV1-Dpol Rev CAGTCAAACTCCAGGGCCG 61.4

785 MmusRHV1-Dpol For CGTCAACGTCTTCCGCCA 58.4 4 55 Round II 246 786 MmusRHV1-Dpol Rev CTGAAGCCCGAGTGTTCCA 59.5 766 MmusRHV1-gB For AACGGCACCACGGTCTTTAT 59.9 5 55 Round I 960 767 MmsRHV1- gB Rev CGGACGCCCGGGATAAATAG 60.3 775 MmusRHV1-gB For TCGATGGAGGTCTGTGAGGA 59.6 6 55 Round II 410 776 MmsRHV1- gB Rev TATCTAGGTTTCCGCCCCCA 60.0 MCMV1 403 MCMV1-gB For AGCGTCCATGATGTTTCGGA 57.3 7 52 Round I 895 404 MCMV1-gB Rev GCGATACGTGACCACCAAGA 59.4

362 MCMV1-gB For AACAGAAGCAGCAGTGGTAC 57.3 8 52 Round II 389 363 MCMV1-gB Rev CGGTTTCAAGGGCGATACG 58.8 MCMV2 386 MCMV2-gB For GTCGGAAAATATTTGTTGCTGG 56.5 9 50 Round I 1652 402 MCMV2-gB Rev GCGAGAACGTCTCCTACTTC 58.0

443 MCMV2-gb For GGTATAGGAGCGGTGTTGGG 61.4 10 50 Round II 645 444 MCMV2-gb Rev TTAACCCCGTTGCACCATCA 57.3

*Annealing temperature used for blood PCR.

2. Materials and Methods

2.5.3 Limiting dilution nested PCR for the detection of MHV68 ORF50

The frequency of cells harboring MHV68 ORF50 was determined by limiting dilution nested PCR (single-copy sensitivity). Mice infected intravenously with MHV68 were sacrificed using a

CO2 chamber and spleens from infected animals were processed as described in section 2.17. Splenocytes were processed and serially diluted to obtain 8 dilutions of splenocytes ranging from 150,000 cells/100 µl to 69 cells/100 µl (Figure 10B). 12 duplicates of each of the dilutions were plated in a background of 104 NIH3T3 cells in 96 well plates. Cells were lysed overnight at

56°C in a lysis buffer (10 mM Tris-HCl, pH 8.5, 1.5 M MgCl2, 1% NP-40, 1% Tween 20, 0.2 mg/ml proteinase K) and proteinase K was inactivated at 95°C for 15 min as described by Weck et al., 1999. Next, 10 µl of this lysed cell suspension was used as a template for the first round of PCR amplification which was performed in 96 well PCR plates (Greiner Bio-One, 652250) as follows:

Cell lysate 10 µl PCR master mix 8 µl Forward primer (10 µM) 1 µl Reverse primer (10 µM) 1 µl

Plates were then sealed with aluminum sealing foil (Greiner Bio-One, 676090) and centrifuged at RT for 2 min at 720 g. The first round of PCR amplification was carried out in a Tgradient thermocycler (Biometra) with the following thermal cycling conditions:

Denaturation 2 min 95°C Amplification 30 s 95°C Annealing 30 s 55°C 44 cycles Extension 30 s 72°C Final extension 5 min 72°C Cooling ∞ 4°C

For the second round of PCR amplification, 1µl of the first round of amplification was added to the following mixture.

1st round amplicon 1 µl PCR master mix 4 µl Forward primer (10 µM) 0.5 µl Reverse primer (10 µM) 0.5 µl

Nuclease free H2O 4 µl

Every PCR plate contained control reaction mixtures: uninfected splenocytes, and 1 copy of MHV68 ORF50 calculated as described in section 2.3.2. in a background of 104 NIH3T3 cells. All assays performed demonstrated single-copy sensitivity, with no false-positive results. Primers used this assay are listed in Table 9.

2. Materials and Methods

Table 9: List of oligonucleotides used for limiting dilution nested PCR for the detection of MHV68 ORF50 Oligo Nested Tm Annealing Product no. Designation Sequence 5’  3’ PCR (°C) temp (°C) size (bp) (VIMM) MHV68 987 ORF50 1109 AACTGGAACTCTTCTGTGGC 58.4 Rev Round I 55 586 MHV68 988 ORF50 523 GGCCGCAGACA TTTAATGAC 58.4 For MHV68 989 ORF50 1056 CCCCAATGGTTCATAAGTGG 58.4 Rev Round II 55 382 MHV68 990 ORF50 674 ATCAGCACGCCATCAACATC 58.4 For

2.6 Cell culture

Table 10: List of cell lines used in this study Cell line Description Source Medium M2-10B4 Murine bone marrow stroma- ATCC (CRL-1972) DMEM high glucose derived cell line 10% FCS 2 mM Glutamine 1% Penicillin/Streptomycin (10,000 U/mL) (pen/strep) NIH3T3 Immortalized NIH/- Swiss ATCC (CRL-1658) DMEM high glucose mouse embryo fibroblasts 10% FCS 2 mM Glutamine 1% pen/strep 1 mM sodium pyruvate Primary cell Description Source Medium Dendritic Bone marrow derived dendritic This study RPMI 1640 cells cells 10% FCS 2 mM glutamine 1% pen/strep 50 µM β-mercaptoethanol

M2-10B4 and NIH3T3 cells were cultured at 37°C in a humidified 7.5% CO2 incubator, and the cells were maintained at a subcultivation ratio of 1:6 or 1:10 every two to three days.

2.6.1 Primary cells – preparation of dendritic cells with Flt3 ligand and GM-CSF

For the preparation of pDC and cDC, femur and tibia of mice were harvested and placed in PBS before being extruded using a 0.5 x 16 mm, 25G x 5/8’’ needle attached to a 10 ml syringe containing 10 ml of DC medium (RPMI 1640, 10% FCS, 2 mM glutamine, 1% pen/strep, 50 µM β-mercaptoethanol). The bone marrow cells were resuspended in DC medium and collected in a 50 ml conical tube. Cells were centrifuged at 300 g for 6 min at RT and the supernatant was

2. Materials and Methods

removed. The cell pellet was resuspended in 500 µl of red blood cell (RBC) lysis buffer (Sigma Aldrich, R7757) per mouse and incubated for 90 s at RT before the addition of 10 ml of medium/mouse. The cell suspension was centrifuged at RT for 6 min at 300 g. The cell pellet was then resuspended in 1 ml of DC medium and brought up to a final volume of 10 ml/mouse. To count cells, 20 µl of cell suspension was combined with 180 µl medium and 20 µl of this diluted cell suspension was combined with 20 µl of Tuerk’s solution for a final dilution of 1:20. For preparation of conventional dendritic cells (cDC), cells were diluted to a concentration of 1x106 cells/ml in DC medium, and 1x107 cells in 10 ml were plated into 10 cm dishes before the addition of granulocyte macrophage colony stimulating factor (GM-CSF) (PeproTech, 315-03) diluted to a final concentration of 20 ng/ml. After 4, 6 and 7 days, the medium was replenished. For this, two thirds of the cell suspension was centrifuged and the pellet was resuspended in 7 ml of fresh DC medium with 20 ng/ml GM-CSF and the cell suspension was returned to the 10 cm dish. On day 8, plates were swirled once, and nonadherent cells were collected, centrifuged at 260 g for 6 min at RT and resuspended in 5 ml DC medium and then counted. 100,000 cells/well or 300,000 cells/well in 100 µl were then plated in a U-bottom 96 well tissue culture- treated plate.

For preparation of FMS-like tyrosine kinase 3 ligand (Flt3L) induced dendritic cells (FLDC), cells were counted and adjusted to a concentration of 1.5 x 106 cells/ml. 2.5% of Flt3L was added to DC medium, and cells in medium containing Flt3L were pipetted into tissue culture flasks of different sizes (Table 11), placed into an incubator (37°C, 7.5% CO2) and left undisturbed for 8 days. On day 8, the cells were pipetted out of the flasks after carefully washing the growth surface of the flasks and collecting all cells in suspension. Cells were then centrifuged at 300 g for 6 min at RT and the pellet was resuspended in 10 ml of medium/mouse. Cells were diluted 1:10 in 180 µl of DC medium and then 1:2 in trypan blue, for a final counting dilution of 1:20. Total FLDC were used for antibody staining, MACS or FACS purification, or directly plated (300,000 cells/well) for stimulation.

Table 11: Working volumes for cultivation of FLDC in tissue culture flasks Flask Maximum capacity (ml) Working volume (ml) T-175 550 50-55 T-75 250 18-24 T-25 50 7-10

2.7 Viruses

MCMV-GFP was amplified and purified as described by Scheibe et al., 2013, and the titers were determined by standard plaque assay on M2-10B4 cells. The MCMV-GFP virus stock preparation and plaque assay was performed by Baca Chan, Vladimir Magalhães and Georg Wolf. MHV68 WUMS (MHV68) strain was purchased from ATCC (VR-1465) and purified from infected M210B4 cells as described by Bussey et al., 2014. Titers of purified viruses were determined by 50% Tissue Culture Infective Dose (TCID50) as described by Reed and Muench (Reed and Muench, 1938). MHV68 virus stock preparation and determination of titers was performed by Kendra Bussey. NDV was obtained from the laboratory of Andrea Kroeger, HZI.

2. Materials and Methods

2.8 Cell sorting

2.8.1 Fluorescence-activated cell sorting (FACS)

To determine the contribution of pDC and cDC to detection of MHV68, FLDC were sorted by FACS based on specific markers with antibodies against CD11b and B220. 5 x 107 FLDC were centrifuged at RT for 6 min at 300 g in 15 ml falcon tubes and supernatants were decanted and cells were washed with 15 ml of PBS. After centrifuging, the cells were resuspended in 4 ml of PBS containing premixed antibodies in specific dilutions (Table 12) and incubated for 15 min at 4°C. Ten ml of FACS buffer was then added and cells were centrifuged at RT for 6 min at 300 g, then again washed with 15 ml of PBS. Supernatants were poured off and the cell pellet was resuspended in 1 ml of DC medium. CD11bhighB220low cDC and CD11blowB220high pDC were sorted on Aria-II SORP by Lothar Groebe of the Flow Cytometry and Cell Sorting unit at HZI.

Table 12: List of antibodies used for FACS and flow cytometry Antibody Source Order number Dilution

CD45R-B220 PerCp-Cy5.5 eBioscience 45-0452-82 1:150

Anti-mouse CD11b PE eBioscience 12-0112 1:4000

Anti-mouse Siglec-H FITC eBioscience 11-0333 1:1000

Anti-mouse CD11c APC eBioscience 17-0114 1:400

2.8.2 Magnetic-activated cell sorting (MACS)

2.8.2.1 B cell isolation

For isolation of B cells from spleens, mice were sacrificed by the administration of CO2 and spleens were excised under sterile conditions. Spleens were then cut into small pieces with surgical scissors and dissociated using the rubber stopper of a sterile 2 ml and a 70 mm cell strainer. Occasional washing with cell culture medium (DMEM supplemented with 10% FCS) eluted single splenocytes. Cells were then pelleted by centrifuging the suspension at 300 g for 10 min at RT. Supernatants were completely aspirated from cell pellets. Next, RBC were lysed for 90 seconds in 1 ml of RBC lysis buffer (Sigma Aldrich, R7757), before diluting the lysis buffer by the addition of 35 ml of medium. Cells were centrifuged at RT for 10 min at 300 g, resuspended in fresh medium, and counted. Then, up to 107 cells were centrifuged at 300 g for 10 minutes at RT before the supernatant was completely aspirated. All splenic B cells were isolated using pan B cell isolation kit II, mouse (Miltenyi Biotec) according to the manufacturer’s instructions. Cell pellets were resuspended in 40 µl of MACS buffer (0.5% BSA, 2 mM EDTA in PBS), and 10 µl of pan B cell biotin-antibody cocktail (Miltenyi Biotec) was added to the cell suspension and thoroughly mixed. The cell suspension was incubated at 5-8°C for 5 min before the addition of 30 µl of cold MACS buffer, and 20 µl of anti-biotin microBeads. The cell suspension was mixed well, and then incubated for an additional 10 min at 5-8°C. After the

2. Materials and Methods

incubation, the total volume was adjusted to 500 µl by the addition of 400 µl of cold MACS buffer. Next, LD columns were placed in the magnetic field of a MACS separator and primed by rinsing with 2 ml of MACS buffer. After priming, new collection tubes were placed under the columns, and the cell suspension was applied to the columns. Unlabeled cells that passed through the column were collected into the tubes, and then the column was washed twice with 1 ml wash buffer. The total effluent was considered the unlabeled Pan B cell fraction. The collected cells were centrifuged at 300 g for 10 min at RT before counting, and 200,000 B cells/well were plated onto feeder layers of NIH3T3 cells (10,000 cells/well) in 96 well plates, which had been plated 24 hrs previously.

2.8.2.2 Isolation of plasmacytoid dendritic cells (pDC) FLDC were prepared as described in section 2.6.1. On day 8, cells were centrifuged at 300 g for 6 min at RT, resuspended in 5 ml of DC medium and counted. Up to 108 cells were pelleted by centrifuging the cell suspension at 300 g for 10 min at RT. Supernatant was aspirated completely and the cell pellet was re-suspended in 350 µl of MACS buffer. Using the plasmacytoid dendritic cell isolation kit II (Miltenyi Biotec), 50 µl of FcR blocking reagent was added per cell pellet and mixed thoroughly before the addition of 100 µl of plasmacytoid dendritic cell biotin cocktail to cells. The cell suspension was mixed well before incubating at 5- 8°C for 10 min. After incubation, cells were washed by adding 10 ml of MACS buffer and centrifuged at RT for 10 min at 300 g. The supernatant was aspirated completely before resuspending the pellet with 800 µl of MACS buffer and 200 µl of anti-biotin microbeads. After thorough mixing, the cell suspension was incubated for 15 min at 5-8°C. Cells were then washed with 10 ml of MACS buffer and centrifuged at 300 g for 10 min at RT. Supernatants were completely aspirated, and cell pellets were resuspended in 500 µl of buffer. LS columns (Miltenyi Biotec) were placed into a QuadroMACS separator (Miltenyi Biotec) and primed by the addition of 3 ml of MACS buffer. After priming, new collections tubes were placed under the columns to collect effluent, and the 500 µl of the cell suspension was passed through the column. The columns were then rinsed with 3 ml MACS buffer. The rinsing was repeated twice once the column reservoir was empty. All of the approximately 9.5 ml effluent was collected and marked as the enriched plasmacytoid dendritic cell fraction. The cells retained in the column were eluted by removing the column from the magnetic field and using a plunger and additional buffer to elute the labelled non-pDC. pDC were counted and plated for ELISA (300,000 cells/well), or antibody labelled for analysis by flow cytometry. Analysis of DC populations by flow cytometry In order to verify the DC subtypes obtained by Flt3L culture and MACS purification, bone marrow derived FLDC and sorted pDC mice were analyzed by flow cytometry. 5 x 105 of total FLDC and pDC after MACS were centrifuged in 5 ml polystyrene capped tubes at 260 g for 6 min at RT and then resuspended in PBS containing diluted antibodies. FITC-conjugated Siglec- H was diluted 1:1000 in PBS, while APC conjugated CD11c was diluted 1:400 (Table 12). To differentiate between CD11chiSiglec-Hlow cDC and CD11clowSiglec-Hhigh pDC in total FLDC populations, cells were dual stained with a mixture of both antibodies. Additional samples were stained with single antibodies to set laser powers and compensation. All stained samples alongside an unstained control were incubated for 15 min at 5-8°C after which 1 ml of PBS was

2. Materials and Methods

added and the cell suspension was centrifuged for 300 g for 6 min at RT. The supernatant was poured off and the washing step was repeated before resuspending the cells in 500 µl of PBS. Cells were stored at 5-8°C until they were analyzed. Data were acquired with a LSR II flow cytometer and FACSDiva software (BD Biosciences). FCS files were exported and data were analyzed with FlowJo software (Version 10, Tree Star, Inc.). 2.9 Stimulation of MmusRHV1 reactivation

In order to promote reactivation of MmusRHV1, B cells were stimulated with the chemical agents 12-O-tetradecanoylphorbol-13-acetate (TPA), sodium (Na) butyrate and the TLR4 agonist LPS. Additionally, cells were also stimulated with Schistosoma mansoni soluble egg antigen (SEA) which was kindly provided by Dania Richter (Institute for Pathology/Charité, Berlin, Germany) (Table 13).

Table 13: List of stimulants used for MmusRHV1 reactivation Reagent Stock concentration Final concentration Company Product (per well) number 12-O- tetradecanoylphorbol- 50 mg/ml 50 ng/ml Sigma P1585 13-acetate (TPA) sodium (Na) butyrate 0.5 M 0.1 mM Sigma B5887 LPS 10 mg/ml 100 ng/ml Sigma L2630 SEA 40 mg/ml 50 µg/ml - -

2.10 Stimulation of TLR with viruses and TLR ligands for TNFα and IFNα measurement by enzyme-linked immunosorbent assay (ELISA)

To analyze the response to TLR stimulation, 300,000 FLDC or pDC or 100,000 cDC were seeded in 100 µl per well in 96-well U-bottom plates. Cells were infected with viruses for 18, 24, 48 or 72 hpi as indicated in figure legends, or stimulated for 18 h with agonists for cell surface and intracellular TLR. The amount of virus needed to infect DC at an MOI of 0.5 or 2 was calculated using the following formula: 푀푂퐼 × 푁푢푚푏푒푟 표푓 푐푒푙푙푠 푉표푙푢푚푒 푠푡표푐푘 (푚푙) = 푃퐹푈/푚푙

For NDV infections, the NDV stock was diluted 1:250 or 1:500 in Opti-MEM as per previous qualitative assessment, and 100 µl of diluted virus was added to 100 µl of the cell suspension.

For stimulation of cells with the TLR1/TLR2 agonist Pam3CSK4 and the TLR9 agonist CpG 2336 (Table 14) stimuli were diluted in medium creating a 2 x working dilution, and 100 µl of the 2x dilution was added directly to the 100 µl of cell suspension in 96 well plates to give the final concentration (conc.) as indicated in Table 15. For transfection with DOTAP (Roche), per sample 1.2 µl DOTAP was diluted in 2.8 µl of HEPES buffered saline (HBS), and the mixture was further diluted by the addition of 2 µl of HBS. After incubation at RT for 15 min, 94 µl DC medium was added to the mixture, and this 100 µl was added per well. Similarly, for stimulation with PolyU and ORN Sa19, 0.2 µl of TLR agonists were diluted in 1.8 µl of HBS, before

2. Materials and Methods

combining with 1.2 µl DOTAP diluted in 2.8 µl HBS. The transfection mixes were incubated at RT for 15 min before the addition of 94 µl medium and 100 µl of this mixture were added per well. After stimulation for 18 hours, supernatants were collected and stored at -20°C for analysis of tumor necrosis factor α (TNFα) and interferon α (IFNα) by ELISA as described in section 2.11.

Table 14: List of synthetic TLR ligands Product Stock Final TLR Ligand Company Sequence (5’- 3’)/description number conc. conc. ALX- TLR1/2 Pam CSK Alexis Synthetic triacylated lipoprotein 1 mg/ml 1 µg/ml 3 4 165-066 TLR7 PolyU InvivoGen tlrl-sspu Synthetic ssRNA 1 mg/ml 1 µg/ml CpG ODN TLR9 MWG - GGGGACGACGTCGTGGGGGGG 1 mM 0.5 µM 2336 tlrl- TLR13 ORN Sa19 InvivoGen GGACGGAAAGACCCCGUGG 1 mg/ml 1 µg/ml orn19

Table 15: List of stimuli and dilution factor of the supernatant used for ELISA Cell type TLR ligand/Virus Dilution of supernatant ELISA FACS purified DC MHV68 1:25 IFNα (FLDC, pDC, cDC) GM-CSF derived cDC MCMV 1:2 TNFα MHV68 1:2 TNFα NDV 1:2 TNFα pDC (MACS) CpGa 1:25 IFNα MCMV 1:25 IFNα MHV68 1:2 IFNα NDV 1:2 IFNα PolyU 1:25 IFNα pDC (MACS) MCMV 1:2 TNFα ORN Sa19 1:2 TNFα Pam3CSK4 1:2 TNFα

2.11 ELISA

TNFα and IFNα ELISA were performed to ascertain the amounts of TNFα and IFNα in cell culture supernatants from FLDC, pDC and cDC collected 18, 24, 48 and/or 72 hours post stimulation/infection. Supernatants were diluted as listed in Table 15. The antibodies used for TNFα and IFNα ELISA are listed in Table 16.

2. Materials and Methods

Table 16: List of antibodies used for ELISA Stock Final Product Antibody Company concentration concentration number

TNFα ELISA

Purified hamster anti- 0.5 mg/ml 2 µg/ml 557516 BD Biosciences mouse/rat TNF antibody

Biotin rabbit anti-mouse/rat 0.5 mg/ml 1 µg/ml 557432 BD Biosciences TNF antibody

HRP-conjugated streptavidin 1.24 mg/ml 0.31 µg/ml N100 Thermo Scientific antibody

Stock Final Product Antibody Company concentration concentration number

IFNα ELISA

Rat anti-mouse Interferon PBL Assay 250 µg 2.5 µg/ml 22100-1 alpha antibody Science

Rabbit anti-mouse interferon PBL Assay 0.31 mg/ml 16 ng/ml 32100-1 alpha antibody Science

Peroxidase-conjugated Jackson AffiniPure F(ab’) fragment 0.8 mg/ml 80 ng/ml 711-036-152 2 Immunoresearch donkey anti-rabbit IgG (H+L)

2.11.1 TNFα ELISA

TNFα ELISA was performed in 96 well plates (NUNC, #442404, flat bottom, Maxisorp). Purified Hamster Anti Mouse/Rat TNFα antibody was used as the capture antibody and diluted in PBS to a concentration of 2 µg/ml. 50 µl of diluted antibody was plated per well and incubated overnight at 4°C. The antibody was washed off by rinsing twice with 200 µl/well TNFα wash buffer (PBS/0.01% Triton X-100), and plates were blocked with 200 µl/well of PBS/2% BSA and incubated for an hour at 37°C. Plates were then washed twice with wash buffer. In order to determine the amount of TNFα produced, serial 1:2 dilutions of recombinant mouse TNF (BD Bioscience, product no. 554589), ranging from 2000 pg/ml to 31.25 pg/ml were plated in duplicates. Supernatants from experimental samples were diluted 1:2 in PBS/2% BSA and were plated in duplicates alongside standards. Plates were incubated for an hour at 37°C, and washed 4 times with wash buffer. Next, the biotin rabbit anti-mouse/rat TNF antibody used for detection was diluted in PBS/2% BSA, and 50 µl were added to each well. ELISA plates were incubated for 45 min at RT. Following four washes with wash buffer, 50 µl of HRP-conjugated streptavidin antibody diluted 1:4000 in PBS/2% BSA was added to every well and incubated for 30 min at RT. Plates were then washed 4 times with wash buffer. 50 µl of 3,3’,5,5’5-

2. Materials and Methods

tetramethylbenzidine (TMB) substrate (Sigma-Aldrich, #T0440) was added per well and incubated in the dark for 10-30 min to allow color development, after which the reaction was stopped by the addition of 50 µl/well of stop solution (1 N H2SO4). The absorbance was measured at 450 nm on an ELISA plate reader (Tecan Infinite M200).

2.11.2 IFNα ELISA

For IFNα ELISA, rat monoclonal antibody against mouse interferon alpha was diluted in PBS to a final concentration of 2.5 µg/ml, 50 µl were added per well in 96 well plates (NUNC, #442404, flat bottom, Maxisorp) and plates were incubated overnight at 4°C. Plates were washed three times with IFNα wash buffer (0.05% Tween 20 in PBS) and blocked for 2 hours at RT with 200 µl/well of blocking buffer (PBS/1% BSA). Plates were washed 3 times before the addition of standards and supernatants from experimental samples. Mouse interferon alpha A (PBL Assay Science, product no. 12100-1) was diluted in PBS/0.1% BSA to obtain a final concentration of 5 ng/ml and serial two-fold dilutions were prepared in PBS/1% BSA with concentrations ranging from 5000 pg/ml to 78 pg/ml. Supernatants were diluted in PBS/1% BSA as indicated in Table 15 and 50 µl were added per well. Standards and samples were incubated overnight at 4°C and then wells were washed thrice with wash buffer. For detection of IFNα, rabbit polyclonal antibody against mouse interferon was diluted 1:5000 in PBS/1% BSA, 50 µl of diluted detection antibody was added per well, and plates were then incubated for 4 hr at RT. After the incubation, plates were washed thrice and incubated with 50 µl per well of peroxidase- conjugated donkey anti-rabbit antibody, which had been diluted 1:10,000 in PBS/1% BSA. Plates were incubated at RT for 90 min and washed 5 times after the incubation period. 50 µl of TMB substrate was added per well and incubated in the dark for 20-30 min to allow color development. To stop the reaction, 50 µl of 1 N H2SO4 was added per well. The absorbance was measured at 450 nm on an ELISA plate reader (Tecan Infinite M200).

2.12 Mice

Mouse colonies Ennigerloh (Enni), Hamm, Horst and Monheim were originally captured in the wild (Table 17) and bred in captivity. Enni, Horst and Monheim strains are Mus musculus domesticus, and the Hamm strain is genetically Mus musculus domesticus with naturally acquired genomic material from Mus spretus via adaptive introgressive hybridization. Before their arrival at HZI, mice from each colony were maintained in interconnected cages with up to 40 mice in one cage system. Samples and mice from all four mouse colonies were kindly provided by Dania Richter and Franz-Rainer Matuschka (Institute for Pathology/Charité, Berlin, Germany). A few animals from the Horst mouse colony were transferred to the quarantine facility at HZI, all mice were ear marked and housed individually, as breeding pairs or in groups of up to 6 mice in individually ventilated cages (IVC). Mice were treated with Ivermectin (#A112I3220, ssniff Spezialdiäten GmbH, Soest, Germany), an anti-parasitic drug administered via mouse feed, for 12 weeks. Health monitoring was performed according to FELASA recommendations by mfg Diagnostics GmbH, Wendelsheim, Germany.

2. Materials and Methods

Table 17: Origin of wild Mus musculus colonies Origin Genetic Collection Mouse colony background date Ennigerloh (Enni) Ennigerloh M.m.domesticus - M.m.domesticus + Hamm Hamm 2007 M.spretus Horst Leverkusen M.m.domesticus 1996 Monheim Monheim M.m.domesticus 2001

Wild type (WT) control, 3d, UNC93B-/-, TLR and IFNAR deficient mice used in the study were bred under specific pathogen-free (SPF) conditions in the HZI central animal facility. Health monitoring was performed according to FELASA recommendations. C57BL/6N and C57BL/6J were purchased from Charles River Laboratory, Inc., or bred at the HZI central animal facility.

All mice housed at the central animal facility were monitored on a quarterly basis and tested negative for the following pathogens:

Mouse hepatitis virus, mouse rotavirus, mouse norovirus, minute virus of mice, mouse parvovirus, Theiler virus, lymphocytic choriomeningitis virus, mouse Viruses adenovirus type 1, mouse adenovirus type 2, ectromelia virus, pneumonia virus of mice, reovirus type 3, Sendai virus, hantavirus and murine cytomegalovirus Helicobacter spp.,Pasteurella pneumotropic, Streptococci β-haemolytic, Streptococcus pneumoniae, Citrobacter rodentium, Clostridium piliforme, Corynebacterium kutscheri, Mycoplasma pulmonis, Salmonella spp., Bacteria Streptobacillus moniliformi, CAR bacillus, Pseudomonas aeruginosa, Campylobacter, Bordetella bronchiseptica, Bordetella hinzii, Staphylococcus aureus, Corynebacterium bovis, β-haemolytic Streptococcus gr. C,G, Proteus mirabilis, Klebsiella pneumoniae, Klebsiella oxytoca Myobia, Aspiculuris, Syphacia, Giardia, Spironucleus muris, Cryptosporidium, Parasites Entamoeba, Pneumocystis murina

The UNC93B-/- mouse strain was created as part of the NCRR-NIH supported Knockout Mouse Project (KOMP) repository at UC Davis (https://www.komp.org) from the VGB6 ES cell clone 10049A-G9, which was originally generated by Regeneron Pharaceuticals, Inc. A chromosome 19 deletion of 13,703 bp, from positions 3,935,240 to 3,948,942 on the + strand, resulted in a complete deletion of the UNC93B coding sequence. Methods used to create the Velocigene targeted alleles have been published (Valenzuela et al., 2003). High-throughput engineering of the mouse genome coupled with high-resolution expression analysis. For further details about the UNC93B-/- mouse strain generated by KOMP see https://www.komp.org/geneinfo.php?geneid=85731.

To obtain a breeding colony of UNC93B knockout mice at the HZI, frozen sperm from UNC93B-/- mice Unc93b1tm1(KOMP)Vlcg background C57BL/6NTac were purchased from the KOMP repository. In vitro fertilization (IVF) was performed by the HZI animal facility. Briefly, oocytes of superovulated C57BL/6N females obtained from Charles River Laboratory were in vitro fertilized

2. Materials and Methods

with sperm from UNC93B-/- mice overnight. The resulting 2-cell embryos were transferred into pseudopregnant C57BL/6N mice purchased from Charles River Laboratory. Heterozygous offspring were interbred and maintained as a homozygous colony at the HZI.

TLR13-deficient mice (B6N Tlr13 (tm1(KOMP)Vlcg) were created from the VGB6 ES cell clone TLR13_BC7, which contains a targeted deletion of exons ENSMUSE00000463707 and ENSMUSE00000281209. The gene targeting methods have been described (Valenzuela et al., 2003; Skarnes et al., 2011). The ES cells were purchased from Velocigene by the Schughart group and injected into B6N embryos at the HZI. Resulting germline-positive mice were backcrossed for 2 generations onto a C57BL/6NTac background, then interbred and maintained as a homozygous colony at the HZI.

TLR7/9-/- mice were generated during this study by crossing TLR7-/- and TLR9-/- mice. Heterozygous siblings were bred to obtain 1st generation homozygous mice lacking both TLR7 and 9. Homozygous TLR7/9-/- mice were then bred for colony maintenance and use in experiments.

Table 18: List of all SPF mouse strains used MHV68 studies

2. Materials and Methods

Number of Laboratory Stock ID at Strain Reference Supplier Genotype Nomenclature backcrossed name HZI background generations

Charles River C57BL/6J Laboratory, C57BL/6J C57BL/6J C57BL/6J C57BL/6JHZI wild type HZI Charles River C57BL/6N Laboratory C57BL/6NCrl C57BL/6NCrl C57BL/6N C57BL/6NCrl wild type (Strain code 027), HZI Generated -/- B6.Unc93b1(tm1(KOM UNC93B HZI MBR-13 C57BL/6N unc93b1 : ko/ko during this P) Vlcg) - 10049A-G9 study TLR13 : Bastian 2 x onto -/- Hatesuer, ko/Y (♂) B6N TLR13 KLS-31 C57BL/6N C57BL/6N Klaus Tlr13(tm1(KOMP)Vlcg)

Schughart, HZI ko/ko (♀)

B6.129SV/EV- IFNAR-/- HZI SZL-22 C57BL/6J IFNAR : ko/ko IfnarI(tm1 Aguet) - 129SV/EV Tabeta et al., 3d HZI MBR-6 C57BL/6J 3d : tg/tg B6J-Unc93b1(3d) 2006 Loof et al., -/- 9 x onto 2008; Wooten TLR2 HZI SZL-16 C57BL/6J TLR2 : ko/ko B6.129-Tlr2tm1Kir/J x C57BL/6J et al., 2002

TLR7 : -/- Dunja Bruder, 10 x onto Hemmi et al., TLR7 MGK-11 C57BL/6J B6.129P2-Tlr7(tm1Aki) HZI ko/Y (♂)ko/ko (♀) C57BL/6J 2002

-/- B6.129P2-Tlr9(tm1Aki) Hemmi et al., TLR9 HZI MBR-8 C57BL/6 TLR9 : ko/ko 12 x onto BL/6 - E14.1 2000 Generated -/- TLR7: ko/Y, ko/ko B6.129P2-Tlr9(tm1Aki) TLR7/9 This study, HZI MBR-28 C57BL/6J during this TLR9 : ko/ko Tlr7(tm1Aki) study

2. Materials and Methods

2.13 Genotyping of mouse strains used in the study of MHV68 infection both in vitro and in vivo

In order to confirm the genotype of mice listed in Table 18, tail snippets were obtained from WT and TLR-deficient mice and mouse genomic DNA was isolated following overnight digestion at 56°C in tail lysis buffer. PCR was performed with a mixture of either 2 or 3 oligos (primer A, B and C) as listed in Table 19 in order to differentiate among WT, heterozygous and homozygous mice. PCR was performed with primers A, B and C for TLR2, TLR9, TLR13 and UNC93B, while two different sets of primers were used in separate reactions to distinguish between homozygous TLR7 deficient and homozygous 3d mice from their respective WT mice (Table19). 2x PCR Master Mix (Thermo scientific, #K0172) or Ex Taq Hot start (Takara, #RR030) and T3 thermocyclers (Biometra) were used for the PCR. Amplified PCR products were visualized on 1% agarose gels. The PCR was prepared as follows:

TLR2 TLR7 (and WT) TLR9 TLR13 UNC93B 3d (and WT) PCR 5,5 µl 5,5 µl 5,5 µl 5,5 µl 6 µl 5,5 µl Master mix (Thermo Fisher) (Thermo Fisher) (Thermo Fisher) (Thermo Fisher) (Ex Taq Hot start) (Thermo Fisher) Primer A (10µM) 1 µl 0.5 µl 0.5 µl 0.5 µl 0.5 µl 0.5 µl Primer B (10µM) 1 µl 0.5 µl 0.5 µl 1 µl 0.5 µl 0.5 µl Primer C (10µM) 1 µl - 0.5 µl 0.5 µl 0.5 µl - DNA 1 µl 1 µl 1 µl 1 µl 1 µl 1 µl

Nuclease free H2O 1,5 µl 3,5 µl 3 µl 2,5 µl 3 µl 3,5 µl

The PCR was carried out using the following thermal cycling conditions:

TLR2 TLR7 TLR9 TLR13 UNC93B 3d Denaturation 3 min 94°C 3 min 94°C 2 min 94°C 2 min 94°C 5 min 94°C 3 min 94°C Amplification 30 sec 94°C 30 sec 94°C 1 min 94°C 30 sec 94°C 15 sec 94°C 15 sec 94°C Annealing 45 sec 63°C 45 sec 65°C 2 min 60°C 45 sec 58°C 30 sec 65/55°C 30 sec 64°C Extension 45 sec 72°C 2 min 72°C 3 min 72°C 1 min 72°C 1 min 72°C 1 min 72°C Final extension 2 min 72°C 15 min 72°C 10 min 72°C 4 min 72°C 5 min 72°C 10 min 72°C Cooling ∞ 4°C ∞ 4°C ∞ 4°C ∞ 4°C ∞ 4°C ∞ 4°C

2. Materials and Methods

Table 19: List of oligonucleotides used for genotyping Oligo No. T Genotype Primer Designation Sequence 5’  3’ m Cycles Product size (VIMM) (°C) A 839 TLR2com CTTCCTGAATTTGTCCAGTACA 58.4 WT – 499 bp TLR2 B 840 TLR2mut GGGCCAGCTCATTCCTCCCAC 67.3 35 Het – 499/334 bp Hom – 334 bp C 841 TLR2WT ACGAGCAAGATCAACAGGAGA 59.5 A 12 TLR7XT CCAGATACATCGCCTACCTACTAGACC 66.5 40 WT – 1350 bp B 13 TLR7WT ACGTGATTGTGGCGGTCAGAGGATAAC 66.5 TLR7 A 12 TLR7XT CCAGATACATCGCCTACCTACTAGACC 66.5 40 Hom – 1350 bp B 18 Neo Primer ATCGCCTTCTATCGCCTTCTTGACGAG 66.5 A 15 WT2 GTGGAGGGACTTTTGGCCACATTCTATA 65.1 WT – 1600 bp TLR9 B 14 Extra GCAATGGAAAGGACTGTCCACTTTGTG 65 40 Hom – 1250 bp C 18 Neo Primer ATCGCCTTCTATCGCCTTCTTGACGAG 66.5 A 842 TLR13 SU GTTGAGAGCCAAGGGAGAAG 60.5 WT – 800 bp TLR13 B 843 TLR13 Rev TTCCTTTCACTGAAGCACTAACTC 62 35 Het – 800/500 bp C 844 Ifit LacZrev TCGTAACCGTGCATCTGCCAG 63.2 Hom – 500 bp A 176 Reg-10049-F1 GCCTTGTACTCAGAGACTGCCTTGG 66.3 10 (65°C) WT – 939 bp UNC93B B 174 Reg-10049-R1 CATATTCTCCACGGCGTAACTGAACC 64.8 30 (55°C) Hom – 652 bp C 175 3’Universal Neo GCAGCCTCTGTTCCACATACACTTCA 64.8 A 3 WTgeno412for2 GGCTGGGGCAGGACTGCA 62.8 35 WT – 631 bp B 2 3dgenorev2 GCTGAGGCTACATAGTGAGAC 59.8 3d A 1 3dgeno412for2 GGCTGGGGCAGGACTGCG 65.1 35 Hom – 631 bp B 2 3dgenorev2 GCTGAGGCTACATAGTGAGAC 59.8

2. Materials and Methods

2.14 In vivo infection

Age-matched C57BL/6 mice were purchased from Charles River Laboratories or bred in the HZI animal facility. All mice were housed in the infection unit for MHV68 infection experiments.

For intranasal infections, 8-10 week old, age matched mice of mixed gender (unless stated otherwise) of WT and TLR deficient mice were anesthetized intraperitoneally using ketamine/xylazine (0.1 ml/10 g weight). Mice were infected intranasally with 2 x 106 PFU of MHV68 in 30 µl of sterile PBS. Cages containing anesthetized animals were placed on pre- warmed heat pads (SnuggleSafe), and mice were monitored until they recovered to their original state of activity. Starting on the day of infection, body weight of the mice was recorded daily until the experiment was terminated. Mice were sacrificed 3, 6, 10, and 17 post infection. Salivary gland, lung, and spleen were collected and processed as described in 2.16.

Alternatively, mice were infected intraperitoneally with 2 x 106 PFU of MHV68 in 200 µl of sterile PBS, and animals were sacrificed 3 and 6 dpi to collect salivary gland, lung, liver, and spleen. Mice were infected intravenously with 2 x 106 PFU of MHV68 in 100 µl of sterile PBS. Body weights of mice were monitored daily, and salivary gland, lung, liver, and spleen were collected 3 dpi, or spleens were collected 22 dpi and processed as described in 2.17.

2.15 Perfusion

In order to ensure minimal contamination of tissue with blood, intravenously infected mice were perfused before obtaining the brain. Mice were anesthetized using CO2, and under sterile conditions the diaphragm was cut along both sides to expose the heart. A 0.5 x 16 mm, 25G x 5/8’’ needle attached to a 10 ml syringe was used to introduce heparin (10 U/ml) (Sigma- Aldrich) diluted 1:1000 in saline (0.9% NaCl in H2O) into the left ventricle of intact hearts. A small incision was made in the right atrium to facilitate the flow of blood out of the heart. Each mouse was perfused to remove blood with a total of 30 ml heparin containing saline. Successful perfusion was confirmed by yellowing of the liver. The brain was excised from the skull, and the olfactory bulb, cerebrum, cerebellum, and brain stem (Figure 8) were dissected using sterile surgical instruments. Each region was further sliced into smaller pieces, placed into homogenizing tubes (MP Biomedicals) containing 0.8 ml (cerebrum and cerebellum) or 0.2 ml (olfactory bulb and brain stem) of cell culture medium and processed as described in 2.16.

Figure 8: Regions of a mouse brain

2. Materials and Methods

2.16 Determination of viral titers in organs by tissue culture infective dose (TCID50)

Salivary gland, lung, liver, spleen, and different parts of the brain were weighed, homogenized using a FastPrep-24 instrument (MP Biomedicals) at 6.5 m/s for 60 seconds, and then frozen at

-70°C. One day prior to TCID50 sample preparation, 6,250 M2-10B4 cells were plated in 50 µl per well in 96-well plates. On the day of the assay, organ samples frozen at -70°C were thawed in a thermomixer at 25°C, then kept on ice at all times. Organ samples were centrifuged at 5000 g at 4°C for 5 min, and clarified supernatant was then transferred into fresh 1.5 ml microfuge tubes.

Figure 9: Schematic representation of virus titration in organs using TCID50

To aid in the preparation of serial 10-fold dilutions (10-1 to 10-8), 180 µl medium was added per well of a 96 well plate. Clarified homogenates from organs were gently mixed, then 20 µl were added to 180 µl of medium to obtain a 10-1 dilution. Homogenates (12 per plate) at a dilution of 10-1 were carefully mixed with a multichannel pipette while gently stirring to avoid introducing bubbles. 20 µl of the 10-1 sample was then added into the next row containing 180 µl of medium to make the 10-2 dilution, and samples were mixed. The process was repeated with fresh pipette tips until a final dilution of 10-8 was reached. The 8 dilutions were then transferred onto M2-10B4 cells, with each dilution plated as 6 replicates of 25 µl each. The same procedure was repeated for all other organ homogenates. M2-10B4 cells were incubated with the inoculum at 37°C for 90 min. 150 µl of medium was then added drop wise into wells, before returning the plates to the incubator (Figure 9).

Plates were rocked intermittently from days 1-10. On day 10, wells were analyzed for cytopathic effect (CPE) under a light microscope, and compared to plates with uninfected organs to examine for toxicity. Positive wells were counted for each dilution and viral titers in TCID50/ml were calculated by the method of Reed and Muench (Reed and Muench 1938). The viral titer per gram of tissue (TCID50/g) was determined by multiplying TCID50/ml value by the volume of homogenate (0.8 ml for salivary gland, lung, liver, spleen, cerebellum and cerebrum, or 0.2 ml for brain stem and olfactory bulb) and dividing this value by the respective organ weight in grams.

2. Materials and Methods

2.17 Ex vivo reactivation assay

One day prior to reactivation assay sample preparation, 10,000 NIH3T3 cells were plated in 100 µl per well in 96-well plates, with 3 plates per sample. Uninfected and intravenously infected mice were sacrificed with a CO2 and spleens were removed using ethanol-sterilized forceps and scissors rinsed in PBS.

Spleens were then placed into pre-weighed tubes containing M2-10B4 medium and kept on ice. Spleen weights were determined, and then spleens were placed into 70 µM cell strainers on 50 ml conical tubes. Two spleens separated by gender were pooled per experiment. The spleens were disrupted using the rubber end of the piston of a 2 ml syringe, while occasionally washing with medium to elute cells in a total volume of 16 ml. The cells were then centrifuged at 260 g for 10 min at RT, the supernatant was gently decanted, and the cell pellet was resuspended in 2ml of RBC lysis buffer (Sigma-Aldrich, R7757).

The cells were gently mixed for 60 seconds in the lysis buffer before the addition of approximately 30 ml of medium. Cells were then centrifuged at RT for 10 min at 260 g, the supernatant was decanted, cells were resuspended in 20 ml medium, and aggregated cells were removed via a second cell strainer. The single cell suspension was placed on ice, and cells were diluted using serial two-fold dilutions in medium followed by a final dilution in Trypan blue (1:32 final dilution). Cells were counted and diluted in medium to obtain 15 ml of 1.5 x 106 cells/ml. Starting with 1.5 x 106 cells/ml, serial 3-fold dilutions were prepared, with a total of 8 dilutions ranging from 150,000 cells/100 µl to 69 cells/100 µl medium (Figure 10A). Uninfected splenocytes served as a control and were plated alongside splenocytes from infected mice.

Starting with the most dilute cells, splenocytes were transferred into 96 well plates containing NIH3T3 cells. 24 replicates of each of the 8 dilutions (100 µl of splenocytes/well) were transferred using multichannel pipettes (Figure 10B) for the reactivation assay, and 12 replicates of each of the 8 dilutions were plated for limiting dilution nested PCR assay and processed as described in section 2.5.3. Next, approximately 5 ml of each dilution was subjected to two freeze-thaw cycles (-70°C and RT) and plated on NIH3T3 cells analogous to intact splenocytes, to determine the prevalence of preformed virus particles. Splenocyte-NIH3T3 co-cultures were incubated at 37°C with 7.5% CO2 for 14 days with intermittent rocking. After 14 days, samples were analyzed with light microscopy for the presence of cytopathic effects on the feeder layer. The percentage positivity was calculated by dividing the number of positive wells per dilution (x) by the number of total wells plated (24) per dilution and multiplying by 100. 푥 푃푒푟푐푒푛푡푎푔푒 푝표푠푖푡푖푣푖푡푦 = ( ) × 100 24

2. Materials and Methods

Figure 10: Schematic representation of preparation and plating of splenocyte dilutions for ex vivo MHV68 reactivation assay

2.18 Statistical analyses

All data were analyzed with Prism software version 6.00 for Windows (GraphPad, San Diego California, USA). ELISA data were analyzed using the two-tailed unpaired t test. Organ titer data were analyzed with the two-tailed Mann-Whitney nonparametric test. To determine frequencies of reactivation and genome-positive cells, the variable slope sigmoidal dose-response equation was utilized for a nonlinear regression curve fit of the data, and based on the Poisson distribution, the regression line intersect was set at 63.2%.

The variable slope sigmoidal dose-response equation is as follows:

(푇표푝 − 퐵표푡푡표푚) 푦 = 퐵표푡푡표푚 + 1 + 10(퐿표푔퐸퐶50−푥)(퐻푖푙푙푠푙표푝푒)

To determine the frequency of cells harboring the virus (x), the sigmoidal dose-response equation was solved for x:

1 푇표푝 − 푦 푥 = 퐿표푔퐸퐶50 − ( ) 푙표푔 ( ) 퐻푖푙푙푠푙표푝푒 푦 − 퐵표푡푡표푚

Where Bottom is the y value at the bottom plateau, Top is the y value at the top plateau, EC50 is the x value where 50% of wells were positive for the virus, Hillslope describes the steepness of the curve, and the y regression line intersect is set to 63.2% (GraphPad.com).

3. Results

3 Results 3.1 Attempt to establish a novel autologous mouse model for herpesvirus infection

Murine gammaherpesvirus 68 (MHV68) is currently used in animal models to study gammaherpesvirus infections in vivo. MHV68 was originally isolated from bank voles (Blaskovic et al., 1980) and its natural host of MHV68 was found to be the wood mouse (Blasdell et al., 2003). However, the virus is currently studied in M.musculus. As these different hosts of MHV68 are separated by several million years (François et al., 2010) it poses certain limitations to the use of MHV68 as an animal model.

Ehlers and coworkers have provided evidence for a rhadinovirus which naturally infects M.musculus and the novel virus was tentatively named Mus musculus rhadinovirus 1 (MmusRHV1) (Ehlers et al., 2007). An additional study identified the presence of a novel beta herpesvirus tentatively named Mus musculus cytomegalovirus 2 (MmusCMV-2) (designated as MCMV2 in this study) in the same mouse colonies that were positive for MmusRHV1 (Teterina et al., 2009).

This led us to two important aims:

 As most of the commercially available mice and reagents currently used by the scientific community are derived from M.musculus, we wanted to establish an autologous infection model with MmusRHV1 in its natural host M.musculus. Therefore, we aimed to isolate MmusRHV1 and additionally characterize the natural infection and transmission of MmusRHV1.  As a side aspect of the project, we aimed to isolate and sequence the complete genome of the novel MCMV2.

For our study, we first received organ samples from mice of wild colonies named Ennigerloh (Enni), Hamm, Horst and Monheim. These mice were used for the identification of MmusRHV1 and MCMV2 (Ehlers et al., 2007; Teterina et al., 2009). Since the discovery of two novel viruses, these wild mice colonies originally captured in North Rhine-Westphalia, Germany, are maintained in a laboratory setting by Dania Richter and Franz-Rainer Matuschka from the Institute for Pathology at the Charité in Berlin. We wanted to verify that these mice were still positive for MmusRHV1 and MCMV2 and could serve as a source for the isolation of these novel viruses, hence we received lung and spleen samples from mice belonging to all four colonies. In addition to the organ samples, we received mice from the Horst colony from D. Richter and F. Matuschka to set up our own breeding colony at the HZI.

3. Results

3.1.1 Colony screen verified presence of MmusRHV1 and MCMV2 in wild mouse colonies

In previous studies, MmusRHV1 was detected in 75 samples (31% of all tested samples) from 33 different mice of different M.musculus colonies (Ehlers et al., 2007). MCMV2 on the contrary was only found in lung samples from one mouse (Teterina et al., 2009). MCMV (designated as MCMV1 for the purpose of this study) was found to be highly prevalent in spleen, lungs, inguinal lymph nodes and salivary glands (Teterina et al., 2009).

To determine the prevalence of MmusRHV1, MCMV1 and MCMV2 within the four colonies, we designed specific nested polymerase chain reactions (PCR) for each of the viruses (Table 8). Appropriate positive plasmid controls for each PCR were included that contain regions of glycoprotein B (gB) of MmusRHV1, MCMV2 or MCMV1. Nuclease free water was used a negative control for every PCR experiment to identify potential cross-contamination of samples by plasmid DNA.

DNA was extracted from lungs and spleens of 34 mice in total (Enni, n = 3; Horst, n = 11; Monheim, n = 3; Hamm, n = 17) and used for analysis by nested PCR. Nested PCR (PCR no. 1 and 2, Table 8) was used to analyze the presence of MmusRHV1 within the colonies. Colony Horst had the highest prevalence for MmusRHV1 with 86.36% of the animals testing positive for the virus when compared to Enni where 0% of the animals were positive. Monheim and Hamm colonies showed the presence of MmusRHV1 with 16.6% and 29.4% of the animals being positive, respectively (Figure 11A). Presence of MCMV1 was analyzed by PCR no. 7 and 8 (Table 8) and the virus was found in all four colonies, albeit with different rates of prevalence. In the Enni and Monheim colonies 100% of the mice were positive for MCMV1, whereas in the Horst colony 72.7% and in the Hamm colony 82.3% were positive (Figure 11B). Presence of MCMV2 was analyzed using PCR 9 and 10 (Table 8) and in contrast to MCMV1, MCMV2 was found in only in the Horst colony of the four colonies tested. Only one out of four lung samples from the Horst colony was positive for MCMV2 (Figure 11C).

3. Results

Figure 11: Presence of both RHV1 and MCMV2 was verified in the Horst colony. Lung and spleen samples from animals from the four wild colonies Enni (n = 3), Horst (n = 11), Monheim (n = 3) and Hamm (n = 17) were analyzed by nested PCR for the presence of MmusRHV1 (A), MCMV1 (B), and MCMV2 (C). Figures depict the relative fractions of mice that showed the presence of virus in either the lung or spleen, or both organs.

In summary, MmusRHV1 was present in three of the four colonies analyzed, and colony Horst showed the highest rates of positivity for MmusRHV1. As expected, MCMV1 was highly prevalent in all four colonies, with colony Horst showing the lowest prevalence. On the contrary, MCMV2 was present in only one single animal belonging to the Horst colony. As colony Horst showed the highest prevalence of MmusRHV1, lowest prevalence of MCMV1, and evidence for the presence of MCMV2, it was selected for further studies.

3.1.2 Efforts to isolate MmusRHV1 in an in vitro setting

Cytomegaloviruses can infect a wide range of cells within their host, and predominantly target epithelial and endothelial cells, fibroblasts and smooth muscle cells (Sinzger et al., 2008). On the contrary, the viral genome of MHV68 is present in bone marrow pre-B cells and immature B cells during early latency and in immature B cells during long-term latency (Coleman et al., 2010). In concurrence with data from Teterina and colleagues our colony screen for MCMV1, MCMV2 and MmusRHV1 showed the presence of co-infection with the betaherpesviruses MCMV1 and MCMV2, alongside MmusRHV1 in Horst mice (Figure 11). In order to overcome the presence of betaherpesviruses, which should not be present in B cells, and to increase the probability of isolating MmusRHV1, which possibly resides in B cells, we isolated B cells from spleens of the HZI Horst colony.

3. Results

Briefly, Pan B cells were isolated from total B cells by negative selection using magnetic- activated cell sorting (MACS) and DNA was extracted from purified B cells and analyzed for the presence of MmusRHV1. Nested PCR of B cells from Horst mice revealed that the virus was below the detection limit of the PCR which is 103 copies of MmusRHV1/µl. We hypothesized that a latent infection of MmusRHV1 which results in low copy numbers of the virus might still be prevalent. Therefore, we tried to facilitate spontaneous reactivation from isolated B cells. For this, B cells were explanted on to NIH3T3 fibroblasts. Sorted B cells from uninfected C57BL/6J (B6J) mice were plated alongside samples from Horst mice as control. Plates were regularly analyzed by light microscopy for the presence of possible cytopathic effects (CPE) caused by MmusRHV1. In agreement with the PCR data, no cytopathic effect was observed in co-cultured samples. In conclusion, we did not observe any evidence for spontaneous reactivation in B cells isolated from mice of the Horst colony.

In order to promote reactivation of MmusRHV1, we stimulated B cells with the chemical agents 12-O-tetradecanoylphorbol-13-acetate (TPA), sodium (Na) butyrate and the TLR4 agonist LPS. TPA is a well-established tumor promoting agent that efficiently induces reactivation of EBV and KSHV (Hausen et al., 1978; Davis et al., 2001). Na-butyrate induces reactivation of KSHV in vitro (Gregory et al., 2009). LPS has been shown to result in a significant increase in the frequency of reactivating cells (Moser et al., 2004). In addition to chemical stimuli, we used Schistosoma mansoni soluble egg antigen to stimulate B cells as SEA has been shown to induce herpesvirus reactivation (Everts et al., 2009; Reese et al., 2014).

However, upon stimulation of co-cultures, there were no signs of lytic viral replication or reactivation characterized by CPE of NIH-3T3 cells. This was in concurrence with the inability to detect MmusRHV1 in B cells by nested PCR (data not shown).

Next, we utilized salivary glands and lungs from Horst mice as these organs had tested positive for the presence of MmusRHV1 in preliminary experiments (data not shown). Salivary glands and lungs were homogenized, and clarified supernatants were added to M2-10B4 cells. As infection can be enhanced by centrifugation, M2-10B4 cells were centrifuged for 5 min at 260 g at room temperature before incubation at 37°C. Plates were monitored daily for CPE. Wells showing presence of rounded cells or plaques were marked and monitored on a daily basis and compared to M2-10B4 cells that were inoculated with supernatants prepared from SG and spleens of uninfected B6J mice (Figure 12A). Supernatants from wells showing a CPE were transferred onto fresh M2-10B4 cells in bigger well formats for expansion. DNA was extracted from cells in wells that showed CPE and subjected to nested PCR to test for the presence of

3. Results

MCMV1 and MmusRHV1 (PCR 7 and 4, respectively: Table 8). In addition to these samples, DNA extracted from lung and salivary gland of donor Horst mice was compared alongside.

As depicted in Figure 11B, the lungs from the Horst mouse analyzed was negative for MCMV1 and therefore there was no evidence of the virus in cell culture samples. However, the DNA extracts from the salivary gland and cell-free supernatant after expansion showed evidence for the presence of MCMV1 (Figure 12B). PCR amplified products from the salivary gland were extracted and sequenced. All samples analyzed showed sequence similarity to MCMV1. MmusRVH1 was absent from both lung and salivary gland of Horst mice, and no evidence of MmusRHV1 replication was found in cell culture samples (Figure 12B).

Figure 12: Inoculation of M2-10B4 cells with cell free homogenate of spleens or salivary gland from Horst mice resulted in the development of viral plaques that tested positive for MCMV1. (A) Culture of cell free homogenate from salivary gland samples from B6J mice and Horst mice with M2-10B4 cells resulted in the appearance of viral plaques in samples from Horst mice. (B) Nested PCR to test for the presence of MCMV1 in lung and salivary gland revealed the presence of MCMV1 (895 bp) in the salivary gland but not in the lung of Horst mice. DNA was extracted from M2-10B4 cells showing cytopathic effect after incubation with cell free homogenate from lung or salivary gland of Horst mice, and MCMV1 was present in M2-10B4 cells cultured with homogenate from salivary gland.

In summary, our co-culture experiments of sorted B cells with NIH3T3 cells did not result in spontaneous or induced reactivation of MmusRHV1 or MCMV2. However, culture of cell-free homogenates of lung and salivary gland from Horst animals on M2-10B4 cells revealed the presence of MCMV1.

3.1.3 Maintenance of wild mouse colony as a resource of MmusRHV1 at the HZI

Health monitoring revealed the presence of endo- and ectoparasites

In order to study the transmission of MmusRHV1 in animals naturally harboring MmusRHV1, a colony of Horst mice was set up in the quarantine area at the central animal facility of the Helmholtz Centre for Infection Research (HZI). According to standard HZI protocol, two sentinel Horst mice underwent a hygiene status test according to FELASA recommendations to test for a

3. Results

panel of mouse pathogens. The examination was performed by mfg diagnostics GmbH (Wendelsheim, Germany).

Based on ELISA and PCR results obtained from the diagnostic center, both mice from the Horst colony tested positive for a variety of pathogens including bacteria, fungi, viruses, ecto- and endo-parasites (Table 19). A large panel of DNA and RNA viruses, including Hanta virus, murine cytomegalovirus (MCMV), mouse hepatitis virus (MHV), murine norovirus and reovirus, among others, tested positive by ELISA using serum samples collected from the mice. In accordance to our finding of MCMV1 and MCMV2 (Figure 11B and C), one of the two mice analyzed was positive for MCMV by ELISA (Table 20). The mice were also analyzed for the presence of various other microorganisms including bacteria, mycoplasma, and fungi by either cultures, ELISA or PCR. PCR tests revealed the presence of Helicobacter rodentium in one of the two mice tested, while both mice tested positive for the presence of Helicobacter spp., Helicobacter typhlonius and Mannheimia haemolytica. Mice were also positive for Mannheimia haemolytica and Staphylococcus spp. by culture. In addition to the serological evidence of active or previous viral infections and the presence of bacterial infections, ecto- and endoparasites were detected by light microscopy. The ectoparasite Acarina was found during coat examination of both mice (Figure 13A). One of the two mice was positive for endoparasitic Syphacia spp. (Figure 13B) and both mice were positive for Trichomonas spp. (Figure 13C).

Figure 13: Health monitoring revealed the presence of endo- and ectoparasites in Horst mice. Two Horst mice were tested for ecto and endoparasites and light microscopy revealed the presence of the ectoparasite Acarina (A) in the coat of both Horst mice. The endoparasites Syphacia spp (B) and Trichomonas spp (C) were detected in one or both Horst mice tested, respectively. Microscopy was performed by mfg diagnostics, GmbH, Wendelsheim, Germany.

3. Results

Table 20: List of pathogens tested in the two mice of the Horst colony. The tests were performed by mfg diagnostics GmbH (Wendelsheim, Germany). Viruses Number positive Test method Ectromelia virus 0/2 ELISA Epizootic diarrhea of infant mice virus 0/2 ELISA Theiler’s encephalomyelitis virus strain 0/2 ELISA GDVII Hanta virus 0/2 ELISA Lymphocytic choriomeningitis virus 0/2 ELISA Mouse adenovirus 1 and 2 0/2 ELISA Murine cytomegalovirus 1/2 ELISA Mouse hepatitis virus 2/2 ELISA Murine norovirus 0/2 ELISA Mouse parvovirus 0/2 ELISA Protoparvovirus 0/2 ELISA Pneumonia virus of mice 0/2 ELISA Reovirus 0/2 ELISA Sendai virus 0/2 ELISA Bacteria/Mycoplasma/Fungus Number positive Test method Bordetella bronchiseptica 0/2 Culture Citrobacter rodentium 0/2 Culture Clostridium piliforme 0/2 ELISA Corynebacterium kutscheri 0/2 Culture Helicobacter bilis 0/2 PCR Helicobacter hepaticus 0/2 PCR Helicobacter muridarum 0/2 PCR Helicobacter rodentium 1/2 PCR Helicobacter spp. 2/2 PCR Helicobacter typhlonius 2/2 PCR Mannheimia haemolytica 2/2 Culture Mycoplasma pulmonis 0/2 ELISA Pasteurella spp. 0/2 Culture Proteus spp. 0/2 Culture Salmonella spp. 0/2 Culture

3. Results

Bacteria/Mycoplasma/Fungus Number positive Test method Staphylococcus aureus 0/2 Culture Staphylococcus spp. 2/2 Culture Streptobacillus moniliformis 0/2 Culture Streptococci 0/2 Culture Streptococcus pneumoniae 0/2 Culture Parasites Number positive Test method Acarina (mites) 2/2 Microscopy Syphacia spp. 1/2 Microscopy Trichomonas spp. 2/2 Microscopy

Due to these test results, the Horst mouse colony permanently remained in the quarantine facility of the central animal facility of the HZI. To treat the colony for the ecto- and endoparasites, the colony was fed a special diet for 12 weeks containing Ivermectin (13 ppm), an anti-parasitic drug.

In addition to a change in diet, the housing system of the mice was reformed upon arrival at the HZI. Before their arrival at the central animal facility, mice from each colony were housed in interconnected cages with up to 40 mice in one cage system. Mice were free to move from one cage to the other via a connecting tube. This housing system showed evidence of bites among mice indicating the presence of fights within the colony simulating a wild mouse colony. Contrary to their previous housing system, upon being transferred to the central animal facility, all mice were ear marked and placed into individually ventilated cages (IVC). The ear mark code of the animals was entered into an internal database for the purpose of keeping track of breeding, birth and death of mice within the colony. Horst mice were either placed in cages with other mice of the same gender with a maximum of 6 mice per cage, or into breeding pairs with two adult mice per cage. Adult mice in a breeding pair with a new litter were housed with their young until the pups were weaned on day 21. After weaning, male pups were either co-housed with their biological father, or by themselves to avoid territorial conflicts. Female pups were co- housed with either their biological mother or other female mice. Seven different breeding pairs were set up with the available adult mice, and each breeding pair was categorized based on the male in the breeding pair.

3. Results

3.1.4 Establishment of a non-invasive blood PCR screen for MmusRHV1

In order to screen and verify the presence of MmusRHV1 in the new colony setting, a non- invasive blood PCR screen specific to the gB region was set up to test blood samples from living mice (Section 2.5.2). Blood from all adult mice of the colony was collected either through retro-orbital bleeds from living animals or cardiac puncture from animals that were sacrificed and used as a template for MmusRHV1 nested PCR. A total of 7 breeding pairs and their respective offspring constituting 7 different “families” comprised a total of 101 animals, and were analyzed systematically. The screen was conducted cage wise and in a semi-blinded fashion without prior knowledge of the breeding pairs, or the viral status of the family that the mice came from. MmusRHV1 plasmid control DNA was diluted in whole blood obtained from an uninfected mouse from a herpesvirus-free, specific pathogen free (SPF) breeding unit to obtain 103 copies/µl of MmusRHV1 which served as the positive control. Whole blood from herpesvirus- free SPF mice served as ‘biological’ negative control in addition to nuclease free water controls (PCR no.1 and 2, Table 8). Ten mice were analyzed from family 1 (father ID 413433), and 20% of the mice were found to be positive for MmusRHV1. In family 2 (father ID 413503) 25 mice were analyzed, with 52% of them testing positive. Family 3 (father ID 413507) consisted of 27 mice, of which 7 mice were positive, showing 25.9 % positivity, while family 4 (father ID 413509) with 7 mice showed a 57% positivity. Family 5 (father ID 413510) included 12 animals and none was tested positive by blood PCR for MmusRHV1. On the contrary, family 6 (father ID 413512), consisting of 12 mice, showed a percentage positivity of 75% with 9 positive animals (Figure 14). Family 7 (father ID 434436) showed 75% positivity with 6 of 8 animals testing positive by nested PCR. The complete results from the non-invasive blood screen for MmusRHV1 are summarized in Table 21. Figure 14 is exemplary of the screen for family 6.

3. Results

Figure 14: Systematic non-invasive screen to determine the presence of MmusRHV1 in Horst mice bred at the HZI. Blood samples were collected by retro-orbital bleeds from the offspring of family 6 (father: 413512, mother: 413535) right after weaning when the offspring was 21 days of age. Nested PCR for MmusRHV1 gB (314 bp) was performed using heparin blood as template. Whole blood spiked with plasmid containing nucleotides 464-975 of gB from MmusRHV1 (103 copies/µl) was used as positive control. Whole blood from herpesvirus-free SPF mice and nuclease free water served as negative controls. The father tested negative for MmusRHV1, whereas the mother was positive. The offspring of this family included 6 male and 4 female mice. Of these, 5 males and 3 females were tested positive for the presence of MmusRHV1. The agarose gel is exemplary and shows the result of the MmusRHV1 PCR for gB from three male offspring. Base pair ladder: lane 1 and 8, three male offspring: lanes 2-4, PCR controls: lane 5-7.

Table 21: Summary of the results of the non-invasive blood screen of 7 Horst families Total Father Viral status Mother Viral status number of Number of (internal by blood (internal by blood animals animals Percentage Family ID) PCR ID) PCR analyzed positive positive (%) 1 413433 Negative 413452 Positive 10 2 20

2 413503 Positive 413524 Positive 25 13 52

3 413507 Negative 413533 Positive 27 7 25.9

4 413509 Positive 413532 Negative 7 4 57

5 413510 Negative 413536 Negative 12 0 0

6 413512 Negative 413535 Positive 12 9 75

7 434436 Positive 414268 Positive 8 6 75

Total 101 41 40.5

3. Results

In summary, the non-invasive colony screen for the presence of MmusRHV1 revealed the presence of the virus in the colony maintained in IVC cages. This finding suggests that MmusRHV1 is transmitted from the parents to their offspring in the new caging system and group set up. Moreover, the presence of MmusRHV1 in young adults after weaning indicates that MmusRHV1 is possibily transmitted from parent to offspring before weaning at 21 days after birth.

3.1.5 Co-housing strategy and design to study natural transmission of MmusRHV1

Analysis of the time frame for vertical transmission of MmusRHV1 within the colony

Vertical transmission of gammaherpesviruses has been addressed by multiple studies which show evidence of anti-KSHV antibodies in infants of seropositive mothers (Wilkinson et al., 1999; Bourboulia et al., 1998). Adding to the studies of KSHV, vertical transmission of MHV68 via the placenta from pregnant mice to fetuses and consequently to newborn mice has previously been shown (Štiglincová et al., 2011). Our non-invasive blood screen revealed presence of MmusRHV1 in young mice right after weaning at 21 days of age, indicating the possibility of vertical transmission of the virus and hence is in concurrence with the published data for KSHV and MHV68.

In order to better understand this finding with respect to a natural and autologous gammaherpesvirus infection, we wanted to further investigate the time frame of vertical transmission of MmusRHV1 from positive parent to offspring. For this purpose, we set up 8 different breeding pairs with two Horst animals that tested positive by gB PCR in our blood screen, 1 breeding pair with a Horst female and a B6J male and 1 breeding pair with a Horst male and B6J female. Litters from 2 of the Horst/Horst breeding pairs were collected on day 1 and 19 after birth, and litters from 2 Horst/B6J breeding pairs were collected on day 10 and 19 after birth (Table 22) DNA was extracted from spleens of a total of eight 1 day old animals, eight 10 day old animals and twenty seven 19 day old animals. A nested PCR for MmusRHV1 gB (PCR 5 and 6, Table 8) was performed to analyze the viral status of the animals. The results are summarized in Table 22.

Contrary to our prior evidence by blood PCR for the possible transfer of the virus from positive parents to their offspring before the age of 21 days (Figure 14 and Table 21), all offspring (n = 43) from previously positive parent/parents tested negative for MmusRHV1 by nested PCR (Table 22). Presence of DNA in the extracted samples was confirmed by PCR for the housekeeping gene beta-2-microglobulin (B2M).

3. Results

To verify the viral status of previously positive parents, all breeding pairs were terminated and spleens were excised from the animals for DNA extraction. Contrary to our previous findings by PCR on blood samples, MmusRHV1 was not detected by nested PCR. These findings suggest that MmusRHV1 was present either below the detection limit, which is 103 copies/µl, absent from the spleen or the infection was cleared in the course of the 4 months between the analyses.

Table 22: Summary of the results of the non-invasive blood screen to analyze vertical transmission of MmusRHV1

Number of Internal ID Status of Internal ID Status of Age of Animals offspring of father the father of mother the mother offspring positive analyzed

8 1 day 0 464696 434449 Positive Positive (Horst) (Horst) 5 19 days 0

425387 434442 Positive Positive 7 19 days 0 (Horst) (Horst) 9 19 days 0 434439 514560 Positive Negative (Horst) (B6J) 8 10 days 0

514536 420060 Negative Positive 6 19 days 0 (B6J) (Horst)

In summary, the results from the experiment to determine the time frame of vertical transmission of MmusRHV1 were inconclusive as parents that had previously tested positive by blood PCR tested negative by nested PCR of spleen samples four months later. As these experiments were performed 4 months after the initial blood screen, and coincided with the termination of treatment with Ivermectin, we hypothesized that the absence of parasites after the treatment with Ivermectin might contribute to low MmusRHV1 copies in samples from Horst mice. This hypothesis is further discussed in section 4.2.

Analysis of horizontal transmission of MmusRHV1 within the colony

A recent study revealed the sexual transmission of MHV68 from experimentally infected female BALB/c mice to co-housed naïve male mice (François et al., 2013). In this study, six week-old female BALB/c mice were infected intranasally with 104 PFU of luciferase+ MHV68 and 21 days

3. Results

post infection MHV68 could be detected in the genital tract by bioluminescence imaging. The co-housed male mice showed the presence of luciferase+ MHV68 in the genital region around 4 days after contact with the female mice. The signal peaked around 10 days and was maintained for 3 weeks. In addition to sexual transmission from infected females to naïve males, François and colleagues observed presence of MHV68 in a naïve female after breeding with an experimentally infected male (François et al., 2013). Multiple studies also addressed the possibility of horizontal gammaherpesvirus transfer through saliva (Hricova and Mistrikova, 2007; Vieira et al., 1997; Pauk et al., 2000; Martro et al., 2007).

To study the possible sexual transmission of MmusRHV1, 3 breeding pairs containing females that were tested positive for MmusRHV1 by blood PCR and naïve males were set up (Figure 15A). An additional breeding pair with a MmusRHV1 positive male by blood PCR and a naïve B6J female was set up to study possible transmission from male mice to female mice (Figure 15B). To study possible non-sexual horizontal transmission of MmusRHV1, a positive Horst female was co-housed with 5 naïve B6J females (Figure 15C). Two control breeding pairs with B6J mice were maintained in the same area as the other breeding pairs to test for possible transmission of the virus via handling by animal caretakers or aerosols (Figure 15D).

Figure 15: Co-housing cage systems to study horizontal transmission of MmusRHV1 in naturally infected animals.

Similar to our findings from the experiment to determine the time frame of possible vertical transmission of the virus from parents to offspring (Table 22), MmusRHV1 was not detectable by nested PCR in DNA extracted from spleen of previously positive Horst mice. Not surprisingly, there were no detectable amounts of MmusRHV1 in samples from co-housed naïve B6J mice. Organs from B6J adults and offspring from control breeding pairs were also negative for MmusRHV1 by nested PCR.

3. Results

In summary, experimental results from nested PCR to determine the possibility of horizontal transmission of MmusRHV1 were inconclusive as MmusRHV1 was not detected in organ samples from animals previously positive for the virus by blood PCR. This finding too might be attributed to the possible contribution of parasites to the change in viral status of the mice, and is further discussed in section 4.2.

3. Results

3.2 The role of Toll-like receptors in the detection of MHV68

3.2.1 Identification of cell types responsible for the production of type I interferons upon MHV68 infection in vitro

Murine gammaherpesvirus 68 (MHV68) is used to study the interactions between the gammaherpesviridae and their hosts in an amenable laboratory model. Innate immune responses of the host play an important role in the control of MHV68. Guggemoos and colleagues identified a role for endosomal Toll-like receptor (TLR) 9 in the detection of MHV68 infection in dendritic cells (DC) that were cultured in the presence of FMS-like tyrosine kinase 3 ligand (Flt3L), designated Flt3L-DC (FLDC). The study showed that the production of IFNα in TLR9-/- FLDC was severely reduced compared to WT FLDC, but was not completely abolished. This suggested that additional TLR may be involved in the detection of MHV68 in vitro in FLDC (Guggemoos et al., 2007). Other than the study on TLR9, there has been no extensive research on the contribution of other cell surface or endosomal TLR or UNC93B, an escort protein for endosomal TLR, to the in vitro detection of MHV68. Through this study, we wanted to better understand the individual roles of cell surface TLR2 and endosomal TLR7, 9 and 13, and the synergistic role of endosomal TLR by using UNC93B-/- FLDC which lack functional TLR3, 7, 9, 11, 12 and 13, in the detection of MHV68 in different subsets of dendritic cells (DC). Since TLR7 and 9 are the major type I IFN inducing pattern recognition receptors in plasmacytoid dendritic cells, and we wanted to ascertain the combined role of these two TLR, we intercrossed TLR7-/- and TLR9-/- mice to obtain TLR7/9-/- mice lacking both endosomal TLR7 and 9.

FLDC are composed of plasmacytoid DC (pDC) and conventional DC (cDC) with differential expression of TLR. Therefore it becomes important to study the role of these specific DC subsets for MHV68 detection after their purification from FLDC cultures. A study showed that in bone marrow (BM) cultures supplemented with Flt3L alone, pDC started to emerge at day 3–5 post culture, whereas CD8α+ cDC appeared 7 days post culture (Zhan et al., 2012). The authors further suggested that the optimal yield of both cell types stabilized around day8-9, but that cell death increased after day 9 (Zhan et al., 2012). Thus, we selected day 8 for preparation and stimulation of cells from FLDC cultures. To understand which of the DC subset was responsible for the detection of MHV68 in FLDC cultures, bone marrow cells from WT mice were cultivated for 8 days in the presence of Flt3L before separating bulk FLDC populations into pDC and cDC subsets. In addition to FLDC, we wanted to analyze the type I IFN response in differently differentiated conventional DC, namely granulocyte macrophage colony stimulating factor (GM- CSF)-derived cDC.

3. Results

ELISA was used to measure IFNα levels upon MHV68 infection of (i) FLDC, (ii) pDC and cDC that were FACS-purified from FLDC cultures, (iii) GM-CSF-derived DC and (iv) pDC that were MACS purified from FLDC cultures.

In vitro detection of MHV68 by FLDC and FACS-purified pDC and cDC

Upon recognition of its ligand, Flt3 induces a major extracellular signaling pathway regulating steady-state pDC and cDC generation from BM progenitors in vivo (McKenna et al., 2000). By the addition of Flt3L to bone marrow cells, comprehensive subsets of splenic DC equivalents develop, including type I IFN producing CD11clow CD11b+CD45RA+B220highGr1low pDC (Asselin- Paturel et al., 2003) and antigen presenting CD11chigh CD45RA− cDC, which can be further divided into CD4+ cDC, CD8α+ cDC and CD4−CD8α- cDC subsets (Shortman and Naik, 2007). Due to the presence of these and other specific surface markers, these subpopulations can be separated using fluorescence-activated cell sorting (FACS).

Figure 16: FACS-purified pDC sorted from Flt3L-differentiated bone marrow cells produce IFNα upon MHV68 infection. FLDC were sorted by FACS into pDC and cDC subsets. FLDC, pDC and cDC were then plated at equal cell numbers and infected with MHV68 at an MOI of 2 or left uninfected (-). ELISA was performed on supernatants collected 18 hours post infection (hpi) to determine IFNα levels (ng/ml). The experiment was performed twice, and results are shown as mean +S.D. of two replicates from one representative experiment. FLDC: FMS-like tyrosine kinase 3 ligand (Flt3L) induced dendritic cells, pDC: plasmacytoid dendritic cells, cDC: conventional dendritic cells. The experiment was performed together with Kendra Bussey.

After 8 days of culture in the presence of Flt3L, FLDC were collected and labelled with APC-A conjugated CD11c antibody and PerCP-Cy5-5-A conjugated B220 antibody. The antibody labelled cells were then sorted into CD11chi cDC and B220+ pDC. After sorting, CD11chiB220low cDC cells constituted 26.4% of the total live cells sorted, while 8.2% of the total live cells were CD11clowB220high pDC. Unsorted FLDC and sorted cDC and pDC were plated at a cell density of 300,000 cells/well and were infected with MHV68 at an MOI of 2 or left uninfected. 18 hours later, supernatants were analyzed by ELISA for production of IFNα. A relatively high background

3. Results

level of IFNα, 0.9 ng/ml was detected from uninfected FLDC, pDC and cDC for this specific ELISA as the supernatants were diluted to a factor of 1:25 (Figure 16). MHV68 infected FLDC produced 2.3 ng/ml of IFNα (Figure 16). Sorted pDC produced 38.9 ng/ml of IFNα, while only background levels of IFNα, 0.72 ng/ml, were detected upon infection of cDC with MHV68 (Figure 16). This indicates that the fraction of pDC present in the total FLDC population is the major producer of IFNα by MHV68-infected FLDC, and cDC do not contribute to the production of IFNα after MHV68 infection. In summary, pDC were identified as the major producers of IFNα in FLDC upon MHV68 infection.

In vitro detection of MHV68 by GM-CSF-derived cDC

As we did not see a contribution of Flt3L-derived cDC to the IFNα response to MHV68, we next analyzed the type I IFN response in GM-CSF-derived cDC. For this purpose we stimulated bone marrow cells from WT, UNC93B-/-, TLR2-/-, TLR7-/-, TLR9-/- and TLR13-/- mice with GM-CSF which produces a homogenous cDC subset (Zhan et al., 2012) alongside monocyte-derived macrophages (Helft et al., 2015) and granulocytes (Lutz et al., 1999).

After stimulation with GM-CSF for 8 days, cells were plated and infected with Newcastle disease virus (NDV), MCMV or MHV68. IFNα levels in the supernatants were then determined by ELISA 18 hours after infection. As with FACS purified cDC, we detected background levels of IFNα upon infection of GM-CSF-derived cDC from WT and TLR deficient mice with MHV68 (Figure 17A). In contrast, NDV infection led to the production of 8.6 ng/ml of IFNα from WT GM-CSF- derived cDC, and cDC from UNC93B-/-, TLR2-/-, TLR7-/- and TLR9-/- mice produced similar amounts of IFNα. TLR13-/- cDC produced 7.1 ng/ml of IFNα upon NDV infection which was slightly lower when compared to WT GM-CSF-derived cDC (Figure 17B).

Upon infection with MCMV, WT cDC produced 2.1 ng/ml of IFNα. There was no difference between WT and UNC93B-/- cDC with respect to their response to MCMV. cDC from TLR2-/- mice produced 3 ng/ml IFNα, while cDC from TLR7-/-, TLR9-/- and TLR13-/- produced 2.5, 1.8 and 1.8 ng/ml of IFNα, respectively (Figure 17C). These results show that UNC93B-dependent TLR as well as TLR2 do not play a role in IFNα response to NDV and MCMV in GM-CSF- derived cDC.

3. Results

Figure 17: GM-CSF differentiated cDC do not produce IFNα after MHV68 infection. cDC from wild type (WT), UNC93B-/-, TLR2-/-, TLR7-/-, TLR9-/- and TLR13-/- mice were infected with MHV68 at an MOI of 2 (A), NDV (B) or MCMV at an MOI of 0.5 (C) or left uninfected (-). Levels of IFNα in supernatants were determined by ELISA 18 hpi. The experiment was performed twice, and results are shown as mean +S.D. of two replicates from one representative experiment. The experiments were performed together with Kendra Bussey.

In summary, GM-CSF-derived cDC produce IFNα in response to the single-stranded RNA virus NDV and the double-stranded DNA virus MCMV, but do not produce a measureable IFNα response when infected with MHV68.

3.2.2 Characterization of the role of TLR expressed by pDC for sensing MHV68 in vitro

Our results showed that FACS sorted pDC are the major producers of IFNα upon MHV68 infection (Figure 16). Therefore, we wanted to study the role of TLR in the detection of MHV68 in pDC. Previous studies showed that stress imposed by prolonged sorting time reduced cell viability and altered their gene expression profile (Gomes et al., 2001; Li et al., 2013). In addition, FACS sorted cells are labelled with antibodies which may affect their biological activity. We therefore followed an alternative approach to purify pDC from FLDC. We chose to purify pDC from a total FLDC population by magnetic labelling using negative selection using MACS. Specific antibodies were used to magnetically label non-pDC which were then retained on a MACS column placed in a magnetic field. Non-antibody labelled pDC passed through the column and were collected for further analysis.

In order to analyze the purity of the magnetically sorted pDC by flow cytometry, total FLDC and sorted pDC were labelled with the pDC specific antibodies Siglec-H and CD11c. pDC, which are defined as CD11cloSiglec-H+, were gated within the living cell population. Before MACS purification, 38.3% of the total live cell populations were CD11cloSiglec-H+ (Figure 18A), while after MACS purification 70.9% of the live cell population were identified as pDC (Figure 18B).

3. Results

Figure 18: Enrichment of pDC population by magnetic cell sorting. pDC from FLDC culture of WT mice were magnetically sorted using negative selection. Magnetically sorted cells were labelled with CD11c and SiglecH antibodies and analyzed by flow cytometry. The pDC population (CD11cloSiglec-H+) in FLDC before magnetic cell sorting (A) were compared to the pDC population in FLDC after magnetic cell sorting of FLDC (B). Cell debris and dead cells were excluded from the analysis based on scatter signals. The experiment was performed twice, and results are shown from one representative experiment. The experiments were performed together with Kendra Bussey. pDC express a limited repertoire of TLR: cell surface localized TLR2 and endosomally located TLR 7, 9 and 12 (Luber et al., 2010). In order to define the roles of UNC93B, TLR7 and 9 in the in vitro recognition of MHV68, total FLDC were generated from TLR7, TLR9, TLR7/9 and UNC93B deficient mice and the pDC population was enriched using the MACS technology. As stimulation of pDC with the TLR2/6 agonist FSL-1 and the TLR12 agonist profilin did not induce an IFNα or TNFα response in pDC in our experiments (data not shown), pDC were transfected with single stranded Poly (U) RNA to specifically activate TLR7, stimulated with the TLR9 agonist CpGa 2336, or infected with NDV or MCMV, and production of IFNα was determined by ELISA 18 hours later. Upon stimulation with Poly (U), 40 ng/ml of IFNα was secreted by pDC from WT and 51 ng/ml from TLR9-/- mice. As expected, pDC from UNC93B-/-, TLR7-/- and TLR7/9-/- mice did not to respond to Poly (U) (Figure 19A).

Upon stimulation with the TLR9-specific agonist CpGa, 18 ng/ml of IFNα were secreted by pDC from WT and 27 ng/ml from TLR7-/- mice. As expected, pDC from UNC93B-/-, TLR9-/- and TLR7/9-/- mice failed to respond to CpGa stimulation (Figure 19B).

3. Results

Figure 19: pDC deficient of TLR7 and 9 fail to produce IFNα in response to their respective synthetic ligands. pDC from wild type (WT), UNC93B-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice were transfected with Poly(U) (A), stimulated with CpG 2336 (B), or infected with NDV (C) or MCMV-GFP MOI 0.5 (D) for 18 h. Levels of IFNα in supernatants were determined by ELISA. The experiment was performed thrice, and results are shown as mean +S.D. of two replicates from one representative experiment. The experiments were performed together with Kendra Bussey.

As shown in figure 19C, NDV infected WT pDC produced 12.5 ng/ml of IFNα which was comparable to 12.2 ng/ml of IFNα produced by infected TLR9-/- pDC. UNC93B-/- and TLR7/9-/- pDC showed a significant reduction in IFNα production when compared to WT pDC by producing 5.5 and 6.3 ng/ml of IFNα, respectively, while TLR7-/- pDC showed a further decrease by producing 2.5 ng/ml of IFNα (Figure 19C). These results were surprising as though the absence of TLR7 alone lead to a reduced IFNα response, and pDC from TLR9-/- did not show an impaired IFNα response, pDC from UNC93B-/- and TLR7/9-/- mice lacking functional TLR7 did not produced higher amount of IFNα than pDC from TLR7-/- mice.

Supernatants from pDC infected with MCMV were diluted 25-fold, and therefore a background level of 7.9 ng/ml IFNα was detected from uninfected pDC for the ELISA (Figure 19D). WT pDC produced 47 ng/ml of IFNα. TLR7-/- pDC produced 61.5 ng/ml of IFNα which was significantly higher when compared to WT pDC. TLR9-/- pDC on the other hand produced 14.5 ng/ml of IFNα which was significantly lesser than that from WT pDC. Interestingly, we only detected background levels of IFNα, 7.9 ng/ml, from pDC from UNC93B-/- and TLR7/9-/- mice lacking functional TLR7 and TLR9. These results were surprising as although the absence of TLR7 alone did not lead to a reduced IFNα response, and pDC from TLR9-/- showed an impaired IFNα

3. Results

response, pDC from mice lacking both TLR7 and 9 were completely unresponsive to MCMV infection. These results indicate a synergistic role of TLR7 and TLR9 in the detection of MCMV in pDC prepared from FLDC cultures, in agreement with a publication showing a combined role for TLR7 and TLR9 in the IFN response to MCMV in vivo (Zucchini et al., 2008).

In summary, pDC responded to stimulation with synthetic TLR7 and 9 agonists as well as infection with NDV and MCMV, showing that the cells were viable and functional after magnetic cell sorting. The TLR7 and TLR9 genotypes could be verified by stimulation with synthetic agonists. As expected, TLR7 is involved in the type I IFN response to NDV. In addition, we could show that TLR7 and 9 play a synergistic role in MCMV detection in pDC.

As pDC did not respond to stimulation with TLR2 and TLR12 agonists, we could not induce IFNα production from pDC of all genotypes. We therefore stimulated total FLDC containing both pDC and cDC from WT, UNC93B-/-, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and TLR13-/- mice with the TLR1/2 ligand Pam3CSK4, the TLR13 ligand ORN Sa19 or infected the cells with MCMV and analyzed TNFα production by ELISA.

Figure 20: FLDC produce TNFα upon stimulation with Pam3CSK4, ORN Sa19 and infection with MCMV. FLDC from wild type (WT), UNC93B-/-, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and TLR13-/- mice were stimulated with 1 µg/ml of the TLR1/2 ligand Pam3CSK4 (A), transfected with 1 µg/ml of the TLR13 ligand ORN Sa19 (B) or infected with MCMV at an MOI of 0.5 (C). As negative controls cells were left unstimulated (A) or uninfected (C) or transfected with the transfection reagent DOTAP only (B). Levels of TNFα in supernatants were determined by ELISA 18 hours post stimulation/infection. Results are shown as mean +S.D. of 3 (A and C) or 2 (B) independent experiments. Statistical significance compared to wild type controls was determined by two-tailed, unpaired t test. The experiments were performed together with Kendra Bussey.

Upon stimulation with the TLR1/TLR2 agonist Pam3CSK4, TNFα levels secreted by WT FLDC, 1.1 ng/ml, were similar to those secreted by TLR7-/-, TLR9-/-, TLR7/9-/- and TLR13-/- FLDC (Figure 20A). As expected, TLR2-/- FLDC failed to produce TNFα, and the difference was statistically significant when compared to WT (p < 0.0001) (Figure 20A). Uncharacteristically, FLDC from UNC93B-/- mice with functional TLR2 produced 0.6 ng/ml of TNFα, which was slightly lower than WT FLDC and this difference was statistically significant (p = 0.0401) (Figure 20A).

3. Results

Upon DOTAP transfection of the TLR13 agonist ORN Sa19, FLDC from WT, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- secreted comparable amounts of TNFα (Figure 20B). As expected, UNC93B-/- and TLR13-/- mice that both lack TLR13 signaling failed to respond to stimulation with ORN Sa19 (Figure 20B). Upon infection with MCMV, WT, TLR2-/-, TLR7-/- and TLR13-/- FLDC produced similar amounts of TNFα (Figure 20C). In contrast, TNFα production was abolished in -/- -/- -/- FLDC obtained from UNC93B , TLR9 and TLR7/9 mice (Figure 20C).

In summary, FLDC produce TNFα in response to the TLR2 agonist Pam3CSK4, TLR13 agonist ORN Sa19 and MCMV. As expected, TLR2-/- FLDC did not produce TNFα in response to -/- -/- stimulation with Pam3CSK4 and UNC93B and TLR13 FLDC failed to respond to stimulation with ORN Sa19. In addition, we have shown that TLR9 contributes to the TNFα response to MCMV in FLDC.

Absence of UNC93B and in particular TLR7 and 9 impairs the production of IFNα in response to MHV68 infection

In order to ascertain which of the TLR expressed by pDC play a role in the detection of MHV68 and subsequent production of type I interferon, WT, UNC93B-/-, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- MACS purified pDC were infected with MHV68 WUMS at an MOI of 0.5 and 2 for 24, 48 and 72 hours, and the amount of IFNα produced was determined by ELISA.

IFNα was detected at all three time points upon infection with both MOI, albeit at different levels (Figure 21). pDC from WT mice secreted 1 ng/ml of IFNα when infected with MHV68 at an MOI of 0.5 and 1.4 ng/ml of IFNα at an MOI of 2 24 hpi (Figure 21A). At 48 hpi, WT pDC secreted 2.6 and 2.3 ng/ml of IFNα when infected with MHV68 at an MOI of 0.5 and 2, respectively (Figure 24B). At 72 hpi, 2.3 and 2 ng/ml of IFNα were produced upon infection with MHV68 at an MOI of 0.5 and 2, respectively (Figure 21C). Notably there was no dose dependent effect observed between the two different MOI at all time points analyzed (Figure 21). pDC from WT, TLR2-/- and TLR7-/- mice produced comparable amounts of IFNα upon infection with MHV68 at both MOI at all three time points analyzed (Figure 21). In contrast, the production of IFNα by pDC from UNC93B-/- and TLR7/9-/- was completely abolished at both MOI and all time points analyzed, and the reduction was statistically significant compared to the IFNα response from WT pDC (Figure 21 and Table 23). On the contrary, the response of TLR9 pDC to infection with MHV68 differed between the different time points post infection. At 24 hpi, pDC from TLR9-/- mice exhibited only background levels of 0.4 ng/ml of IFNα upon infection with MHV68 at both MOI of 0.5 and 2, which was comparable to 0.36 ng/ml of IFNα detected from

3. Results

uninfected pDC and significantly reduced compared to the response from WT pDC (Figure 21A and Table 23). However, at 48 hpi, TLR9-/- pDC produced 0.8 and 1 ng/ml of IFNα at an MOI of 0.5 and 2, respectively and the response was significantly reduced in comparison to IFNα produced by WT pDC (Figure 21B and Table 23). TLR9-/- pDC produced higher levels of INFα when compared to UNC93B-/- and TLR7/9-/- pDC (Figure 21B). Similarly, at 72 hpi, TLR9-/- pDC produced 0.8 and 0.7 ng/ml of IFNα at an MOI of 0.5 and 2, respectively, and the response was reduced in comparison to IFNα produced by WT pDC (Figure 21C and Table 23). TLR9 deficient pDC produced higher levels of INFα when compared to pDC from UNC93B-/- and TLR7/9-/- mice (Figure 21C).

Figure 21: Type I IFN responses of pDC upon MHV68 infection are dependent on UNC93B and endosomal TLR7 and 9. pDC from wild type (WT), UNC93B-/-, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice were infected with MOI 0.5 and 2 of MHV68 for 24 h (A), 48 h (B) and 72 h (C). Levels of IFNα in supernatants were determined by ELISA. Results are shown as mean + S.D. of two independent experiments. IFNα levels were statistically different among TLR deficient pDC when compared to WT pDC upon MHV68 infection and p values at different times after infection with MHV68 are shown in table 23. The experiments were performed together with Kendra Bussey.

In summary, detection of MHV68 by pDC is completely dependent on the endosomal transport protein UNC93B and the combination of TLR7 and TLR9. TLR9 deficiency leads to complete abolishment of IFNα response at 24 hpi, suggesting TLR9 is the major contributor to detection of MHV68 in pDC early after infection. However, at 48 and 72 hpi, the IFNα response from TRL9-/- pDC is reduced compared to WT pDC, but not completely abolished, indicating the contribution of another TLR to the response. While pDC deficient in TLR7 alone did not exhibit a defect in the IFNα response to MHV68 compared to WT pDC, our data with UNC93B-/- and TLR7/9-/- pDC clearly indicates that TLR7 and 9 acts synergistically in the recognition of MHV68 in vitro.

3. Results

Table 23: Statistical significance of IFNα levels produced by TLR-deficient pDC when compared to WT pDC upon MHV68 infection 24 hpi 48 hpi 72 hpi MOI 0.5 MOI 2 MOI 0.5 MOI 2 MOI 0.5 MOI 2 WT vs UNC93B-/- 0.0130 0.0042 0.0001 0.0001 0.0001 0.0002 WT vs TLR2-/- 0.0481 ns ns ns ns ns WT vs TLR7-/- Ns ns ns ns ns ns WT vs TLR9-/- 0.0173 0.0052 0.0001 0.0001 0.003 0.0026 WT vs TLR7/9-/- 0.0135 0.0044 0.0001 0.0001 0.0001 0.0002

3.3 Characterization of the role of TLR in sensing MHV68 infection in vivo

As illustrated in the scheme shown in Figure 22, we analyzed the impact of TLR during three different stages of MHV68 infection: (i) acute infection/lytic replication (section 3.3), (ii) establishment of latency (section 3.4.1) and (iii) reactivation from latency (section 3.4.2).

Toll-like receptors (TLR) are involved in the detection of MHV68 infection in vivo. TLR2 may contribute to the production of proinflammatory cytokines and type I IFN and help control viral burden in the lung after intranasal (i.n.) infection with MHV68 (Michaud et al., 2010). Among the endosomal TLR, TLR9 has been shown to be important in the detection of MHV68 after intraperitoneal (i.p.) infection with higher viral loads in the spleen observed both during lytic and latent infection (Guggemoos et al., 2007). Other than studies on TLR2 and 9, there has been no extensive research on the contribution of other TLR or UNC93B, an escort protein for endosomal TLR, to the detection of MHV68. Based on our in vitro studies we hypothesized that endosomal TLR may act in concert in these different stages of MHV68 infection. In order to better understand the role of UNC93B in the detection of MHV68 in vivo, we used UNC93B-/- mice lacking functional endosomal TLR, and triple defect (3d) mice which are unresponsive to stimulation of endosomal TLR due to a single point mutation in the UNC93B gene (Tabeta et al., 2006). Additionally, we used TLR2, TLR7, TLR9, TLR13 and TLR7/9 deficient mice for our in vivo studies to carefully dissect which individual TLR are responsible for detection and initiation of an antiviral immune response upon MHV68 infection. Mice were infected intranasally, intraperitoneally or intravenously, and organs from these animals were examined for the presence of acute infection/lytic replication at various time points. Viral titers from salivary gland, lung, liver and spleen obtained early after infection were determined using median tissue culture infective dose (TCID50) (Figure 22).

3. Results

MHV68 infects the spleen via the mediastinal lymph nodes (MLN), where latency is predominately established within splenic B cells (Flano et al., 2000). The period of latency is characterized by the maintenance of a nonintegrated viral episome and the limited transcription of viral gene expression in the absence of infectious virion production (Krug et al., 2010). In 2008, Gargano and colleagues have shown that MHV68 latency is preferentially established in activated germinal-center B cells in the spleen. They further showed that TLR signaling via MyD88, an adapter protein for all TLR with the exception of TLR3, plays an important function for the establishment of MHV68 latency (Gargano et al., 2008). As previous studies on the role of individual TLR are limited, and studies have not addressed the synergistic roles of multiple endosomal TLR in the establishment of MHV68 latency, we performed a nested PCR assay against the MHV68 ORF50 gene on splenocytes from UNC93B-/-, TLR2-/-, TLR7-/-, TLR9-/-, TLR13-/-, or TLR7/9-/- mice to determine the number of latently infected splenocytes (Figure 22).

In addition to affecting the establishment of latency, TLR have been associated with reactivation of MHV68 from latency. Previous studies have demonstrated that stimulation of TLR with the TLR4 agonist LPS or the TLR9 agonist CpG DNA can drive MHV68 reactivation from latency (Gargano et al., 2009). Guggemoos and colleagues reported that there is a significantly higher number of reactivating splenocytes present from TLR9 deficient mice infected intraperitoneally with 5 x 105 PFU of MHV68 (Guggemoos et al., 2007). As the previous studies did not address the role of other TLR, or the synergistic roles of multiple endosomal TLR, we wanted to additionally analyze if there is enhancement or impairment of MHV68 reactivation from latently infected splenocytes lacking multiple endosomal TLR, and did so by limiting dilution reactivation assay. The assay was performed as described previously (Weck et al., 1999). To distinguish between infectious virus in the samples and latently infected cells, serial 3-fold dilutions of the splenocytes were plated before and after disruption of viable cells by two freeze-thaw cycles (Guggemoos et al., 2007) (Figure 22).

As studies comparing C57BL/6J (B6J) and C57BL/6N (B6N) mice demonstrated a range of phenotypic differences between the two strains (Simon et al., 2013), the mice used in this study were divided into two groups based on their genetic background. UNC93B-/- and TLR13-/- mice are on the B6N background and were compared to WT B6N mice, while TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice are on the B6J background and were compared to WT B6J mice.

3. Results

Figure 22: Overview of MHV68 in vivo experiments conducted in this study. 8-12 week old mice of the indicated genotypes were infected intranasally (i.n.), intraperitoneally (i.p.) or intravenously (i.v.) with 2 x 106 PFU of MHV68. Viral titers in salivary gland, lung, liver and spleen at different time points post infection were analyzed by TCID50. The latent viral load in spleens was analyzed by limiting dilution nested PCR assay on MHV68 ORF50 22 days post i.v. infection. The rate of reactivation of MHV68 from latently infected splenocytes after i.v. infection was determined using reactivation assay.

WT (B6J and B6N), UNC93B-/-, TLR2-/-, TLR7-/-, TLR9-/-, TLR13-/-, TLR7/9-/- and 3d mice were infected intranasally, intraperitoneally or intravenously with 2 x 106 PFU of MHV68 and mice were weighed during the course of the experiment. Salivary gland, lung, spleen and liver from these animals were examined for the presence of lytic replication at various time points.

3.3.1 Role of TLR in the detection of MHV68 after intranasal infection

Typically, infection of mice with MHV68 via the intranasal route involves productive virus replication in organs including the lung and spleen (Sunil-Chandra et al., 1992). Weight loss has been reported in WT BALB/c mice with after intranasal infection (Sunil-Chandra et al., 1992). Additionally, weight loss has been reported for immunocompromised mice after intranasal infection. For example, 8-12 week old C57BL/6 IFNγR−/− mice of mixed gender when infected intranasally with 105 PFU of MHV68 have been shown to lose up to 20% of their total body mass (Krug et al., 2010).

3. Results

Mice infected intranasally with MHV68 recover from weight loss

To analyze whether the absence of TLR might have an effect on weight loss in mice after i.n. infection with MHV68, 8-12 week old mice lacking various TLR and their respective background control were infected intranasally with 2 x 106 PFU of MHV68, and their weight was monitored starting on the day of the infection, day 0, until day 17, when the experiment was terminated. UNC93B-/- and TLR13-/- mice as well as B6N mice lost weight starting on day 6, began recovering by day 10 and had recovered by day 12 (Figure 23A). UNC93B-/- mice did not show an increase in weight loss when compared to B6N mice. While TLR13-/- mice lost more weight on day 8 compared to B6N mice, the difference was not statistically significant (Figure 23A). 3d, TLR2-/-, TLR7-/- and TLR9-/- mice were compared to their corresponding B6J control mice for weight loss. Though all mice began to lose weight on day 6, they returned to their initial weight by day 12. There was no significant difference in weight loss among the different TLR-deficient mice studied (Figure 23B).

Figure 23: Mice infected intranasally with MHV68 recover from weight loss. 8-12 week old B6N, UN93B-/-, TLR13-/-, B6J, 3d, TLR2-/-, TLR7-/- and TLR9-/- mice were infected intranasally with 2 x 106 PFU of MHV68. Body weight (g) was noted at indicated days post infection. Difference in body weight (% initial) was calculated based on initial body weight of the mouse on day 0. Closed circles: mice of B6N genetic background (A), closed squares: mice of B6J genetic background (B). A minimum of 9 mice were used for the analysis. Results are shown as mean ± S.D. and are combined from 5 independent experiments. The experiments were performed together with Kendra Bussey.

During the course of the experiment, other aspects of the animal’s condition like appearance of fur, response to external stimuli, and behavior in the cage were also noted. There were no noteworthy signs of visible clinical illness other than weight loss. In summary, there was no significant difference in weight loss of TLR-deficient mice or WT controls after intranasal infection with MHV68.

3. Results

Lytic replication in the lung is not TLR dependent

Upon intranasal infection MHV68 replicates in the lung. Within 2 weeks, lytic viral replication is undetectable in the lungs though latent virus persists in splenic B cells, macrophages, dendritic cells and lung epithelial cells life-long (Krug et al., 2010). Michaud and colleagues showed that MyD88-/- and TLR2-/- mice displayed higher viral titers in the lung 24 hours post i.n. infection with 1 x 105 PFU of MHV68 when compared to WT mice (Michaud et al., 2010). The study did not analyze the effect of other MyD88 dependent TLR or the effect of TLR2 in the recognition of MHV68 at other time points post infection. Hence, we infected WT B6N, UNC93B-/-, TLR13-/-, WT B6N, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice intranasally with 2 x 106 PFU of MHV68. On days 3, 6, 10 or 17 post infection lungs were obtained from mice and assayed using

TCID50. Uninfected organs were used as controls to identify homogenate toxicity to the monolayer.

On day 3 post infection, WT B6N, UNC93B-/-, TLR13-/-, WT B6J, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice showed similar lung viral titers with varied distribution within groups (Figure 24A). Two outliers were present, one B6J mouse and one TLR9-/- mouse. Lungs from TLR2-/- mice appeared to have slightly lower viral titers than lungs from B6J mice, but the difference was not statistically significant. Mice lacking TLR7 showed a slight increase in viral titers when compared to B6J, while lung titers from TLR9-/- and TLR7/9-/- mice were comparable to those from B6J mice. Lungs from 3d mice showed slightly elevated titers in comparison to lungs from B6J mice, but the increase was not statistically significant (Figure 24A).

In general, titers of MHV68 in the lungs increased from day 3 to day 6 post infection (Figure 24A and B). On day 6 after infection, all mice showed similar viral titers in the lung (Figure 24B). Lungs obtained from UNC93B-/- mice on average appeared to have higher titers in comparison to B6N lungs due to increased viral titers from two mice out of 20 mice, but as the majority of mice showed similar titers as WT B6N mice, this observation was not statistically significant (Figure 24B). MHV68 titers from lungs of mice lacking TLR2 showed a divide, with one group showing slightly elevated titers compared to WT B6J, while the other set showed similar titers (Figure 24B). The clusters formed in the TLR2-/- groups were not based on gender. Overall, the average viral titer from lungs of TLR2-/- mice was similar to that from B6J mice. Similarly, titers from lungs of TLR7-/- and TLR9-/- mice were comparable to titers from B6J mice, and though the titers from TLR7/9-/- and 3d mice had a wide distribution with respect to titers, the difference was not statistically different from B6J mice (Figure 24B).

3. Results

On day 10, the viral loads in the lungs were below the detection limit, with the exception of two outliers: one B6N mouse and one TLR13-/- mouse (Figure 24C). By day 17, there were no detectable viral titers in lung homogenates from all genotypes analyzed (Figure 24D).

Figure 24: Replication of MHV68 in the lung does not show TLR dependency on days 3 and 6 post infection. 8-12 week old B6N, UNC93B-/-, TLR13-/-, B6J, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice were infected intranasally with 2 x 106 PFU of MHV68, and lungs were excised on day 3 (A), 6 (B), 10 (C) and 17 (D) post infection. Closed circles: mice of B6N genetic background, closed diamonds: mice of B6J genetic background, open triangles: 3d mice (B6J genetic background). Each symbol corresponds to an individual mouse, black horizontal lines indicate means, and the dotted red line represents the detection limit. Results are combined from 3, 4, 2 or 2 independent experiments for day 3, 6, 10 and 17, respectively. At least 7 mice were used per data set shown in A, B and C. A minimum of 9 mice were used for day 17 post infection (D). The experiments were performed together with Kendra Bussey.

In summary, after i.n. infection, acute replication of MHV68 in the lung on day 3 and 6 post infection was not dependent on cell surface TLR2, UNC93B, or endosomal TLR7, 9 or 13. By day 10, we observed the onset of viral clearance in the lung and on day 17 lytic MHV68 was cleared from the lungs of all mice examined.

Lytic replication in the salivary gland is dependent on TLR9 10 days post intranasal infection

Saliva is one of the main routes of transmission of KSHV (Minhas and Wood, 2014). Additionally, bioluminescence imaging studies with recombinant MHV68 expressing firefly

3. Results

luciferase revealed infection of the salivary gland area starting at day 10 post infection (Hwang et al., 2008). Previous studies have not analyzed MHV68 titers in the salivary gland, the role of TLR for the spread of MHV68 to the salivary glands or in the control of virus replication in this organ, or the clearance of lytic virus from the salivary gland after i.n. infection. We thus determined viral titers in the salivary gland from 8-12 week old mice intranasally infected with 2 x 106 PFU of MHV68 on days 3, 6, 10 and 17 post infection.

On day 3 post infection, viral titers from salivary gland were mostly under the detection limit with the exception of a few positive salivary glands from WT (B6N and B6J), UNC93B-/-, TLR13-/- and TLR2-/- mice (Figure 25A). On day 6 post infection, viral titers were above the detection limit (Figure 25B). At day 6 slightly higher titers were observed in the salivary glands of UNC93B-/- and TLR13-/- mice in comparison to salivary glands from B6N mice, but the difference was not statistically significant (Figure 25B). Salivary glands from TLR2-/- mice did not show a difference in titers when compared to salivary glands from B6J mice. Salivary glands from TLR7-/- mice on the other hand showed lower titers when compared to B6J mice and this difference was statistically significant with a p value of 0.0253. Titers from salivary glands of TLR9-/- and TLR7/9-/- mice were slightly higher than in salivary glands of B6J mice, but not significantly. Salivary gland from 3d mice showed a slight decrease in titers in comparison to those from WT mice, but the difference was not statistically significant (Figure 25B).

3. Results

Figure 25: Replication of MHV68 in the salivary gland is TLR9 dependent 10 days post infection. 8- 12 week old B6N, UNC93B-/-, TLR13-/-, B6J, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice were infected intranasally with 2 x 106 PFU of MHV68, and their salivary glands were excised on day 3 (A), 6 (B), 10 (C) and 17 (D) post infection. Closed circles: mice of B6N genetic background, closed diamonds: mice of B6J genetic background, open triangles: 3d mice (B6J genetic background). Each symbol corresponds to the value from one individual mouse, black horizontal lines indicate means, and dotted red line represents the detection limit. Results are combined from 3, 4, 2 or 1 independent experiment(s) for day 3, 6, 10 and 17, respectively. At least 7 mice were used per data set shown in A, B and C. A minimum of 3 mice were used for viral titer analysis on day 17 post infection (D). The experiments were performed together with Kendra Bussey.

In contrast to the lung, where viral titers were below the detection limit by day 10 post intranasal infection, MHV68 was detectable in the salivary gland 10 days post infection (Figure 25C). There was no statistically significant difference among viral titers from the salivary gland of B6N, UNC93B-/- and TLR13-/- mice. As on day 6, salivary gland from TLR7-/- mice 10 days post infection showed slightly decreased viral titers in comparison to WT mice, but the difference was not statistically significant. In contrast, viral titers of salivary gland from TLR9-/- (p = 0.0012) and TLR7/9-/- (p = 0.0431) mice were significantly elevated when compared to those from B6J mice (Figure 25C). By day 17, salivary gland from all genotypes were negative indicating viral clearance with the exception of a single TLR9-/- mouse which showed the presence of MHV68, albeit at very low titers (Figure 25D).

In summary, MHV68 replicates in the salivary gland at day 6 and 10 post i.n. infection and is cleared by day 17 post i.n. infection. On day 6, salivary glands from TLR7-/- and 3d mice displayed lower titers when compared to B6J mice. On the contrary, on day 10 post i.n. infection we observed an increase in MHV68 titers in salivary glands from TLR9-/- and TLR7/9-/- mice. This suggests that the detection of MHV68 in the salivary gland on day 10 is dependent on TLR9. However, TLR9 signaling is also absent in UNC93B deficient mice where we do not see elevated viral loads on day 10 post infection.

Lytic replication in the spleen after intranasal infection is not TLR dependent

B lymphocytes within the spleen encompass the main reservoir of latent MHV68 after intranasal infection (Flano et al., 2003). In order to elucidate on the role of TLR in controlling splenic viral titers at early time points after i.n. infection, 8-12 week old mice were infected with 2 x 106 PFU of MHV68 and sacrificed on 3, 6 and 10 days post infection. MHV68 titers were determined in spleen homogenates by TCID50 assay and are depicted in Figure 26.

3. Results

Figure 26: MHV68 titers in the spleen on day 10 post intranasal infection are not dependent on endosomal TLR. 8-12 week old B6N, UNC93B-/-, TLR13-/-, B6J, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice were infected intranasally with 2 x 106 PFU of MHV68 and their spleens were obtained on day 3 (A), 6 (B) and 10 (C) post infection. Closed circles: mice of B6N genetic background, closed diamonds: mice of B6J genetic background, open triangles: 3d mice (B6J genetic background). Each symbol corresponds to the value from one individual mouse, black horizontal lines indicate means, and the dotted red line represents the detection limit. At least 7 mice per genotype are included. Results are combined from 3, 4 or 2 independent experiments for day 3, 6 and 10, respectively. The experiments were performed together with Kendra Bussey.

Three days post intranasal infection very low amounts of virus were present in the spleen, with the exception of several spleens from TLR7/9-/- and 3d mice, which showed a slight increase in splenic viral titers compared to WT B6J mice (Figure 26A). On day 6, a number of spleens with viral titers above the detection limit was higher than those on day 3 (Figure 26B). Spleens from TLR13-/- mice showed a slight but not significant increase in titers when compared to WT B6N mice. Splenic titers from TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice were comparable to those from B6J mice (Figure 26B). By day 10, splenic titers were well above the detection limit for all genotypes (Figure 26C). There was no difference in splenic viral titers between WT B6N mice and UNC93B-/- mice. Interestingly, the slight increase in splenic titers from TLR13-/- mice that was detected on day 6 was still observed on day 10, but the increase remained statistically insignificant. Splenic titers from TLR2-/- and TLR7-/- mice were comparable to titers from WT B6J mice. Spleens from TLR9-/- mice showed slightly elevated titers when compared to spleens from WT B6J mice but the difference was not statistically significant. However, since the titers in

3. Results

spleens from TLR7/9-/- and UNC93B-/- mice which both lack TLR9 signaling were comparable to titers from WT mice, the role of TLR9 for MHV68 replication in the spleen after intranasal infection is difficult to interpret (Figure 26C).

In summary, we did not detect MHV68 in the spleen on day 3 post i.n. infection. On day 6 and 10 post i.n. infection, when viral titers could be determined, MHV68 replication was not dependent on TLR2 or any of the UNC93B dependent endosomal TLR.

Absence of UNC93B post intranasal infection decreases splenomegaly

As with the human gammaherpesvirus Epstein Barr virus (EBV), splenic B cells are a principal reservoir of latent MHV68 infection (Flano et al., 2000). The establishment of latency in mice is associated with an increase in the number of splenic leukocytes and thus a large increase in spleen size (splenomegaly), with a peak of splenomegaly around day 14 post intraperitoneal infection of BALB/c mice (Usherwood et al., 1996). In order to determine the extent of splenomegaly after intranasal infection with 2 x 106 PFU of MHV68 in WT B6N, UNC93B-/-, TLR13-/-, WT B6J, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice, mice were sacrificed 17 days post infection and spleens from infected animals were weighed. Spleens from uninfected B6J mice were used as controls.

Spleens from MHV68 infected WT B6N mice exhibited splenomegaly with a threefold increase in spleen weight when compared to average spleen weight from uninfected B6J mice (Figure 27). UNC93B-/- spleens from MHV68 infected mice were significantly smaller (p < 0.0001) than spleens from infected B6N mice (Figure 27). Spleens from MHV68 infected TLR13-/- mice on the other hand did not show a difference in weight when compared with spleens of B6N mice. Similar to WT B6N mice, WT B6J mice exhibited splenomegaly with a threefold increase in spleen weight when compared to average spleen weight from uninfected B6J mice (Figure27). Spleens from TLR7-/- and TLR7/9-/- mice were significantly smaller in comparison to infected spleens from B6J mice (p =0.0176 and 0.0002, respectively) (Figure 27). Notably, though spleens from mice lacking TLR9 did not show a difference in spleen weight when compared to WT B6J mice, spleens from TLR7/9-/- mice deficient of both TLR7 and 9 showed lower spleen weights (p = 0.0002) when compared to weights from spleens of mice lacking only TLR7 (p = 0.0176).

3. Results

Figure 27: Mice lacking UNC93B and specifically endosomal TLR7 show a decrease in splenomegaly 17 days after i.n. infection with MHV68. 8-12 week old B6N, UNC93B-/-, TLR13-/-, B6J, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice were infected intranasally with 2 x 106 PFU of MHV68 and intact spleens were removed and weighed 17 dpi. Grey hexagons: uninfected mice (B6J genetic background), closed circles: mice of B6N genetic background, closed diamonds: mice of B6J genetic background. Each symbol corresponds to the value from one individual mouse, and black horizontal lines indicate means. Results are combined from three independent experiments and include at least 9 mice per genotype. Statistical significance was determined using Mann-Whitney test. The experiments were performed together with Kendra Bussey.

In summary, mice lacking UNC93B, and in particular TLR7, showed a reduced extent of splenomegaly 17 days post intranasal infection when compared to their respective WT controls. This indicates the importance of endosomal TLR7 in the progression of splenomegaly. TLR7/9-/- mice showed lower spleens weights when compared to spleens from TLR7-/- mice, indicating a possible synergistic role of TLR7 and 9 for MHV68-induced splenomegaly. This is also supported by our observation that UNC93B-/- mice, lacking functional among others TLR7 and TLR9, show a similar level of splenomegaly as TLR7/9-/- mice.

3.3.2 Role of TLR in the detection of MHV68 after intraperitoneal infection

Vidy and colleagues showed that the route of infection affects MHV68 dissemination pathways (Vidy et al., 2013). We therefore, in addition to examining the role of TLR in the detection of MHV68 after i.n. infection, analyzed the role of TLR in detection of MHV68 after intraperitoneal (i.p.) infection. After i.p. infection, TLR9 is involved in the detection of MHV68 in the spleen (Guggemoos et al., 2007). However, this study did not examine the role of TLR9 in other organs or the role of other TLR in the detection and control of MHV68 after i.p. infection. Thus, we verified if loss of TLR9 or of TLR2, 7 or 13 would affect MHV68 titers in the salivary gland, lung, spleen and liver on days 3 and 6 post infection. We also included UNC93B-/-, 3d and TLR7/9-/- in

3. Results

our analysis to understand the synergistic role of endosomal TLR in the detection of MHV68 after i.p. infection.

Mice show no signs of weight loss upon intraperitoneal infection with MHV68

Similar to our i.n. infection studies, we first analyzed the change in weight after i.p. infection with 2 x 106 PFU of MHV68. 8-12 week old WT (B6J), TLR2-/- and 3d mice were infected intraperitoneally and monitored for weight loss alongside other clinical signs of illness. Mice were weighed starting on the day of infection (day 0) and regularly monitored at the indicated time points post infection as shown in Figure 28. Loss of endosomal TLR function in the 3d mice, or the absence of cell surface TLR2 did not lead to weight loss in the mice analyzed. On the contrary, mice steadily gained weight over the duration of this experiment up to day 18 post infection (Figure 28). None of the mice analyzed showed any other noteworthy signs of illness.

Figure 28: Mice infected intraperitoneally with MHV68 do not lose weight. 8-12 week old B6J, 3d and TLR2-/- mice were infected intraperitoneally with 2 x 106 PFU of MHV68. Body weight (g) was noted at indicated days post infection. Difference in body weight (% initial) was calculated based on initial body weight of the mouse on day 0. A minimum of 3 mice were used per genotype for the analysis. Results are shown as mean ± S.D. The data is from one experiment, and was performed together with Kendra Bussey.

In summary, contrary to the observed weight loss after i.n. infection with 2 x 106 PFU of MHV68 (Figure 23), mice infected intraperitoneally with 2 x 106 PFU of MHV68 did not lose weight and showed no signs of illness.

3. Results

Lytic replication is dependent on UNC93B in the lung, spleen and liver on day 3 post intraperitoneal infection

Next we analyzed lytic viral titers in i.p. infected mice 3 days post infection and the results are depicted in Figure 29. Viral titers were well above the detection limit in lung, spleen and liver (Figure 29B, C and D). However, only a few salivary glands showed signs of acute replication of MHV68 and the majority of the animals were negative (Figure 29A). Therefore, interpretation of the results with the salivary glands was not possible.

MHV68 titers from lungs of UNC93B-/- mice were significantly higher (p = 0.0048) than those from B6N mice (Figure 29B). However, there was no difference in viral titers from the lung among the various single TLR knockout mice, 3d mice and their respective WT controls.

Figure 29: Early replication of MHV68 in the lung, spleen and liver is UNC93B dependent 3 days post intraperitoneal infection. 8-12 week old B6N, UNC93B-/-, TLR13-/-, B6J, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice were infected intraperitoneally with 2 x 106 PFU of MHV68, and their salivary gland (A), lung (B), spleen (C) and liver (D) were obtained on day 3 post infection. Closed circles: mice of B6N genetic background, closed diamonds: mice of B6J genetic background, open triangles: 3d mice (B6J genetic background). Each symbol corresponds to the value from an individual mouse, and black horizontal lines indicate means. Results are combined from two independent experiments and include at least 4 mice per genotype. Statistical significance was determined using Mann-Whitney test. The experiments were performed together with Kendra Bussey.

3. Results

Similar to the lung, there was a statistically significant increase (p = 0.0093) in viral titers from spleens of UNC93B-/- mice when compared to B6N mice (Figure 29C). Yet, titers of spleen from single TLR knockout mice did not significantly differ when compared to control mice (Figure 29C). Spleens from TLR7/9-/- and 3d mice however appeared to have slightly higher titers than spleens from B6J mice, but the differences were not significant (Figure 29C).

In the liver we observed a similar trend to titers from the spleen, with UNC93B-/- mice having significantly higher titers than B6N mice, and the increase was statistically significant (p value = 0.0205) (Figure 29D). On the contrary, TLR7-/- mice showed lower titers when compared to B6J mice (p = 0.0104) (Figure 29D). There was no statistically significant difference between the TLR2-/-, TLR9-/-, TLR7/9-/- and 3d mice and B6J WT control mice.

In summary, 3 days post i.p. infection, MHV68 could be detected in lung, liver and spleen, but not in the salivary gland. The control of MHV68 replication in lung, spleen and liver was UNC93B dependent. However, we did not observe higher titers in TLR7/9-/- or 3d mice, which both also lack multiple functional TLR as UNC93B-/- mice. Notably, only seven TLR7/9-/- and four 3d mice had been included in the experiments, and the inclusion of more mice may result in significant differences.

In order to better understand how the infection progresses after the mice have been infected intraperitoneally with MHV68, we analyzed the salivary gland, lung, spleen and liver 6 days post infection. The viral titers from salivary glands on day 6 were negative from all genotypes analyzed (Figure 30A). We observed a slight increase in viral titers in the lungs of UNC93B-/- mice, but this was not statistically significant. Lungs from TLR13-/- mice showed similar viral titers as WT B6N controls (Figure 30B). The viral titers in the lungs from TLR2-/- mice showed a wide variation within the group, with only one animal showing higher titers than the average of B6J mice (Figure 30B). Lungs from TLR7-/- mice showed viral titers comparable to WT B6J mice, with the exception of one animal that showed an elevated titer (Figure 30B). Viral titers in the lungs of TLR9-/-, TLR7/9-/- and 3d mice were comparable to titers of lungs from WT B6J mice (Figure 30B).

The statistically significant increase in viral titers that was evident in UNC93B-/- spleens 3 days post infection (Figure 29C) was absent on day 6 (Figure 30C). Similar to spleens from UNC93B-/- mice, spleens from TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice did not show an increase in viral titer when compared to WT mice (Figure 30C). On the contrary, viral titers in the liver on day 6 showed differences among the genotypes (Figure 30D). Though viral titers from livers of UNC93B-/- mice showed no statistically significant difference to those of B6N mice,

3. Results

TLR13-/- mice showed a statistically higher viral titer when compared to WT control (p = 0.0290). Viral titers from livers of two TLR2-/- mice and one TLR7-/- mouse were higher when compared to titers from livers of WT B6J mice, but the increase in viral titers on average was not statistically significant. Viral titers from liver samples of TLR9-/- and TLR7/9-/- mice were higher when compared to viral titers from livers of WT B6J mice and the change in titer was statistically significant (p = 0.0142 and 0.0079, respectively) (Figure 30D). Livers from 3d mice also showed an increase in viral titers, but the difference was not significant, possibly due to the lack of sufficient mice.

In summary, 6 days post i.p. infection, viral replication in the salivary gland was below the detection limit. TLR13-/-, TLR9-/- and TLR7/9-/- mice showed higher titers in the liver after i.p. infection indicating a role for these endosomal TLR in the detection of MHV68 in the liver.

Figure 30: Lytic replication of MHV68 in the liver is TLR9 dependent 6 days post intraperitoneal infection. 8-12 week old B6N, UNC93B-/-, TLR13-/-, B6J, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice were infected intraperitoneally with 2 x 106 PFU of MHV68, and their salivary gland (A), lung (B), spleen (C) and liver (D) were obtained on day 6 post infection. Closed circles: mice of B6N genetic background, closed diamonds: mice of B6J genetic background, open triangles: 3d mice (B6J genetic background). Each symbol corresponds to the value from one individual mouse, and horizontal lines indicate means. The dotted red line represents the detection limit. Results are combined from three independent experiments, or are indicative of a single experiment and include at least 3 mice per genotype. Statistical significance was determined using Mann-Whitney test. The experiments were performed together with Kendra Bussey.

3. Results

3.3.3 Role of TLR in the detection of MHV68 after intravenous infection

Systemic infections after organ transplantation and blood transfusion have been described for KSHV (Hladik et al., 2006). Yet, acute replication and manifestation of MHV68 and the role of TLR in control of MHV68 replication after i.v. infection has not been studied. In order to understand the role of TLR in the detection of MHV68 after i.v. infection, we infected mice with 2 x 106 PFU of MHV68 and monitored their weight loss, analyzed lytic viral titers in salivary gland, lung, spleen and liver 3 days post infection and determined the role of TLR in the establishment of latency and reactivation of MHV68.

B6J and B6N mice show differences in recovery from weight loss upon intravenous infection with MHV68

A study comparing WT B6J and WT B6N mice demonstrated a range of phenotypic differences between the two mouse strains including differences in components of the immune system (Simon et al., 2013). We first compared weight loss of WT B6N and WT B6J mice after i.v. infection with MHV68. As with intranasal and intraperitoneal infection, mice were weighed on a daily basis and analyzed for other signs of clinical illness alongside weight loss starting on day 0 for consecutive 10 days (Figure 31).

Figure 31: B6J mice show increased weight loss when compared to B6N mice upon intravenous MHV68 infection. 8-10 week old B6N and B6J mice were infected intravenously with 2 x 106 PFU of MHV68. Body weight (g) was noted at indicated days post infection. Difference in body weight (% initial) was calculated based on initial body weight of the mouse on day 0. Results are shown as mean ± S.D. and are combined from three (B6N) or two (B6J) independent experiments and include at least 8 mice per data point. Statistical significance was determined using Mann-Whitney test. The experiments were performed together with Kendra Bussey.

3. Results

Mice on the B6N and B6J background began losing weight between 5 and 6 days post infection. B6N mice started to recover between day 6 and 7, and reached their initial weight by day 9. B6J mice on the other hand showed increased weight loss up to day 7 and started to recover between day 7 and 8 and took longer than B6N mice to reach their initial weight. There was statistically no difference in weight loss between B6N and B6J mice between day 0 and day 6. On day 7 B6J mice showed a decrease in weight which was statistically significant (p = 0.0031) when compared to B6N mice. The weight loss of B6J mice when compared to B6N mice remained significantly different on days 8 (p = 0.0007), 9 (p = 0.0005) and day 10 (p = 0.0004) (Figure 31). In summary, WT mice with B6J background lost more weight after i.v. infection with MHV68 when compared to B6N mice, and additionally took longer to gain weight post infection.

Mice lacking endosomal TLR display a significant decrease in body weight when infected with MHV68 intravenously

We next compared weight loss of WT B6N mice with UNC93B-/- and TLR13-/- mice infected with i.v. with 2 x 106 PFU of MHV68. For comparison, weight loss of WT B6N shown in Figure 31 was graphed together with UNC93B and TLR13 deficient mice shown in Figure 32.

Figure 32: UNC93B and TLR13 deficient mice show increased weight loss than WT B6N mice upon MHV68 infection. 8-10 week old B6N, UNC93B-/- and TLR13-/- mice were infected intravenously with 2 x 106 PFU of MHV68. Body weight (g) was noted on indicated days post infection. Difference in body weight (% initial) was calculated based on initial body weight of the mice on day 0. Results are shown as mean ± S.D. and are combined from three (UNC93B-/-) or two independent (WT B6N and TLR13-/-) experiments and include at least 8 mice per genotype. Statistical significance was determined using Mann-Whitney test. The experiments were performed together with Kendra Bussey.

We did not observe statistically significant differences in weight loss between WT B6N and UNC93B-/- mice between day 0 and day 6 (Figure 32A). On day 7 however, UNC93B-/- mice showed a decrease in weight which was statistically significant (p < 0.0001) when compared to WT B6N mice (Figure 32A). The weight loss of UNC93B-/- mice when compared to WT B6N

3. Results

mice remained significantly different on days 8 (p = 0.0006), 9 (p = 0.0003) and day 10 (p = 0.0005) (Figure 32A). We did not observe statistically significant differences in weight loss between WT B6N and TLR13-/- mice between day 0 and day 6 (Figure 32B). On day 7 however, TLR13-/- mice showed a decrease in weight which was statistically significant (p < 0.00041) when compared to WT B6N mice (Figure 32B). The weight loss of TLR13-/- mice when compared to WT B6N mice remained significantly different on days, 8 (p = 0.0011), 9 (p = 0.0021) and day 10 (p = 0.0062) (Figure 32B).

We next compared weight loss of WT B6J mice with TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice infected with i.v. with 2 x 106 PFU of MHV68. For comparison, weight loss of WT B6J shown in Figure 31 was graphed together with TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice shown in Figure 33.

3. Results

Figure 33: TLR2-/- and TLR9-/- show less weight loss than B6J mice. 8-12 week old B6J, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice were infected intravenously with 2 x 106 PFU of MHV68. Body weight (g) was noted at indicated time points. Change in body weight was calculated based on initial body weight of the mice (on day 0) (A). Weight loss in B6J (shown in Figure 31) mice was compared to TLR2-/- mice (B), TLR7-/- and TLR7/9-/- mice (C). Additionally, weight loss of TLR7-/- and TLR9-/- mice (D) and TLR9-/- and TLR7/9-/- mice (E) was compared to WT B6J. Weight loss of B6J mice depicted in grey (D and E) serves as a guideline and was not taken into consideration for statistical analysis. Results are shown as mean ± S.D. and are combined from two independent experiments and include at least 8 mice per genotype. Statistical significance was determined using Mann-Whitney test. The experiments were performed together with Kendra Bussey.

As shown earlier in Figure 31, WT B6J mice began to lose weight between days 5 and 6, and this weight loss continued up to day 7 (Figure 33A, D and E). WT B6J mice lost more weight and recovered more slowly than TLR2-/- mice (Figure 33B). TLR2-/- mice showed significantly higher weights than WT B6J mice on days 7, 8 and 9 post infection with p values of 0.0207, 0.0047 and 0.0379, respectively (Figure 33B). TLR7-/- and WT B6J mice showed a very similar

3. Results

weight loss pattern with no statistically significant difference (Figure 33C). TLR7/9-/- mice showed a similar weight loss pattern as TLR7-/- mice with no difference to WT B6J mice, with the exception of day 3 post infection when TLR7/9-/- showed significantly lower weights than B6J mice (p = 0.0499) (Figure 33C). Conversely, TLR9 deficiency did not result in weight loss compared to WT B6J mice (Figure 33A, D and E). In contrast, the TLR9-/- mice showed less weight loss in comparison to B6J mice on days 7, 8 and 9 (Figure 33A, D and E) but this was not statistically significant (data not shown). Though the difference between WT B6J and TLR9-/- was not significant, there was a statistically significant difference between TLR7 and TLR9 deficient mice on days 7 and 8 (p = 0.0003 and 0.0006). TLR7 mice lost significantly more weight than TLR9 mice (Figure 33D). TLR7/9-/- mice lost more weight when compared to mice lacking only TLR9-/- (Figure 33E). This difference was significant on days 7, 8 and 9 with p values of 0.0002, 0.0019 and 0.0171, respectively (Figure 33E).

In summary, UNC93B-/- and TLR13-/- mice showed a significant loss of weight when compared to WT B6N mice. With respect to mice on a B6J background, TLR2-/- mice showed less weight loss when compared to WT B6J mice, while TLR7-/- and TLR7/9-/- mice did not show a statistically significant weight loss when compared to B6J mice. Similar to TLR2-/- mice, mice lacking TLR9 did not lose as much weight as their WT counterparts. Additionally, there was a statistically significant difference in weight loss between WT B6N and WT B6J mice indicating that the genetic background of the mice plays an important role in the control of MHV68 post i.v. infection.

Lytic replication of MHV68 in the lung, spleen and liver is dependent on UNC93B post intravenous infection

We next determined viral titers in salivary gland, lung, spleen and liver of mice that were intravenously infected with MHV68. Three days post infection, all mice showed presence of MHV68 in the salivary gland (Figure 34A), lung (Figure 34B), spleen (Figure 34C) and liver (Figure 34D). There were no statistically significant differences in MHV68 titers between salivary gland from B6N mice and those from UNC93B-/- mice, and while salivary gland from TLR13-/- mice showed slightly lower titers in comparison to B6N, the difference was not statistically significant (Figure 34A). Salivary glands from TLR2-/- and TLR7-/- mice did not show a difference in MHV68 titers when compared to B6J mice. Mice deficient of TLR9 showed a slight increase in titers when compared to B6J mice, but the difference was not statistically significant. Salivary gland from TLR7/9-/- mice showed a large distribution of viral titers, and a number of mice showed an increase in viral titers when compared to the controls, but the increase was not

3. Results

statistically significant when compared to titers from salivary gland of B6J mice. Titers of salivary gland from 3d mice, on the other hand, clustered together and were statistically higher than those from B6J mice (p = 0.0470) (Figure 34A). This indicates that functional endosomal TLR contribute to the control of MHV68 in salivary gland 3 days post i.v. infection, at least in mice on a B6J background.

Figure 34: Replication of MHV68 in the lung, spleen and liver is UNC93B dependent 3 days post intravenous infection. 8-10 week old B6N, UN93B-/-, TLR13-/-, B6J, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and 3d mice were infected intravenously with 2 x 106 PFU of MHV68, and their salivary gland (A), lung (B), spleen (C) and liver (D) were obtained 3 days post intravenous infection. Titers were determined by TCID50 assay. Closed circles: mice of B6N genetic background, closed diamonds: mice of B6J genetic background, open triangles: 3d mice (B6J genetic background). Each symbol corresponds to the value from one individual mouse, and horizontal lines indicate means. Results are combined from three experiments (WT B6N and UNC93B-/-) or two independent experiments and include at least 7 mice per genotype. Statistical significance was determined using Mann-Whitney test. The experiments were performed together with Kendra Bussey.

We next determined the presence of MHV68 in the lung 3 days post infection and the results are shown in Figure 34B. Similar to intranasal infection, the highest viral titers after i.v. infection were found in the lung. Similar to titers post i.p. infection, lungs from UNC93B-/- mice showed higher viral titers than those from B6N mice (p = 0.0089). Lungs from TLR13-/- and TLR2-/- mice did not show any difference in titers when compared to titers from lungs of the respective WT mice. Though lungs from TLR7-/- mice displayed slightly elevated viral titers when compared to

3. Results

titers from lungs of the B6J controls, the increase was not statistically significant. Lungs from TLR9-/- mice on the other hand did not show any increase in titers when compared to lungs from B6J controls. Lungs from TLR7/9-/- mice showed slightly higher titers when compared to mice lacking only TLR9, and similar titers as TLR7-/- mice. Similar to lungs from UNC93B-/- mice, lungs from 3d mice showed significantly higher titers when compared to B6J mice (p = 0.0201). This suggests that multiple endosomal TLR are involved in the control of MHV68 infection in the lung 3 days post i.v. infection.

Next, we analyzed splenic titers days post i.v. infection (Figure 34C). Similar to the lung, UNC93B-/- mice showed significantly higher splenic viral titers than WT B6N mice (p = 0.0006). There were no differences in viral titers between spleens from TLR13-/- mice and WT B6N mice. While splenic titers from TLR2-/- were slightly but not significantly lower when compared to B6J mice, spleens from TLR7-/- mice had similar titers as B6J mice. MHV68 titers from spleens obtained from TLR9-/- mice were significantly higher than WT mice (p = 0.0014). Similarly, mice lacking both TLR7 and TLR9 showed higher splenic viral titers when compared to B6J (p = 0.0302). As the spleens from UNC93B-/- mice, those from 3d mice showed a statistically significant increase in titers when compared to B6J mice (p = 0.0099). This suggests that UNC93B and in particular TLR9 are essential for the detection of MHV68 in the spleen 3 days post i.v. infection.

Additionally, we examined the liver for the presence of MHV68 and the results are depicted in Figure 34D. Similar to titers from the other organs, livers of UNC93B-/- mice showed significantly higher viral load (p = 0.0033), while titers from TLR13-/- livers were comparable to those from B6N mice. TLR2-/- and TLR7-/- mice showed slightly lower viral titers in the liver when compared to B6J mice, but the difference was not statistically significant. Livers from TLR9-/- mice showed similar titers to those from B6J mice. On the contrary, livers from TLR7/9-/- and 3d mice showed comparatively higher viral titers than those from B6J mice, though the difference was not statistically significant. These results suggest that multiple endosomal TLR might play a synergistic role in the control of MHV68 in the liver 3 days after intravenous infection.

In summary, multiple endosomal TLR play a role in the control of MHV68 infection via the intravenous route in salivary gland, lung, and liver, while TLR9 is involved in the control of MHV68 in the spleen. In the lung and liver, the lack of individual TLR2, 7, 9 or 13 did not result in elevated viral titers, while organs from UNC93B deficient mice showed an increase in viral titers. TLR7/9-/- and 3d mice displayed a close to significant increase in viral titers in the liver when compared to WT B6J mice. Overall our findings suggest that multiple endosomal TLR are

3. Results

involved in the detection and control of MHV68. Additionally, in the salivary gland we observed an apparent difference in viral titers between UNC93B-/- (B6N) and 3d (B6J) mice, both lacking functional endosomal TLR, which suggests differences in spread of MHV68 to the salivary gland between mice on a B6N and B6J background.

Lytic replication of MHV68 in the brain 3 days post intravenous infection is not dependent on UNC93B

UNC93B and type I IFNs are known to be important in the detection and control of herpes simplex virus type 1 (HSV-1) infection in the brain (Wang et al., 2012). Kang and colleagues monitored MHV68 infection in the brains of 9-10 week old intracerebroventrically infected BALB/c mice. By using a recombinant MHV68 virus expressing firefly luciferase they observed persistence of MHV68 both inside and outside the CNS (Kang et al., 2012). Outside the CNS MHV68 was present particularly in the salivary gland, lung, stomach, liver, spleen and pancreas (Kang et al., 2012). Since the role of UNC93B and individual TLR has not been determined in the context of MHV68 infection, we intravenously infected 8-10 week old WT B6N and UNC93B-/- mice with 2 x 106 PFU of MHV68. As a previous study showed presence of MHV68 in brains from i.v. infected IFNAR-/- mice (Julia Spanier, unpublished data), we included IFNAR-/- mice in our study. Mice were perfused before the brains were excised to remove blood which may contain MHV68. Brains were harvested and viral titers were determined by TCID50 from olfactory bulb, cerebrum, cerebellum and the brain stem. Toxicity of brain homogenate was excluded by using uninfected brain sections as negative controls.

100

3. Results

Figure 35: Replication of MHV68 in the olfactory bulb, cerebrum, cerebellum and brain stem is not UNC93B dependent. 8-10 week old B6N, UNC93B-/- and IFNAR-/- mice were infected intravenously with 2 x 106 PFU of MHV68. Olfactory bulb (A), cerebrum (B), cerebellum (C) and brain stem (D) were obtained from perfused mice on day 3 post infection. Each symbol corresponds to the values from one individual mouse. B6N (n=9), UNC93B-/- (n=8), IFNAR-/- (n=3). The horizontal lines across the data sets indicate means. The experiments were performed together with Kendra Bussey.

As shown in Figure 35, viral titers were above the detection limit in all brain sections analyzed. The olfactory bulbs showed the highest viral titers in comparison to the cerebrum, cerebellum and olfactory bulb. There was no significant difference in viral titers between olfactory bulbs of UNC93B-/- and B6N mice (Figure 35A). As expected, olfactory bulbs from IFNAR-/- mice had a much higher titer when compared to olfactory bulbs from B6N mice (Figure 35A). The cerebrum and the cerebellum (Figure 35B and C) showed a comparable trend to the olfactory bulb, with similarities in viral titers between B6N and UNC93B-/- mice. The viral titers from brain stems of UNC93B-/- mice were slightly elevated when compared to brain stems of B6N mice, but the difference was not statistically significant (Figure 35D). Cerebrum, cerebellum and brain stem from IFNAR-/- mice showed an increase in viral titers when compared to those from B6N mice (Figure35B, C and D). In summary, UNC93B does not seem to contribute to the control of MHV68 replication in the brain 3 days post intravenous infection in adult mice.

101

3. Results

Absence of UNC93B after intravenous infection decreases splenomegaly

As we observed splenomegaly after intranasal infection (Figure 27), we analyzed the contribution of TLR to the induction of splenomegaly after i.v. infection with MHV68. Intact spleens from adult WT B6N, UNC93B-/-, TLR13-/-, TLR2-/-, TLR7-/-, TLR9-/-, and TLR7/9-/- mice were excised and weighed 22 days after infection with MHV68. Spleens from uninfected B6J mice were used as controls and the results are summarized in Figure 36.

Figure 36: Mice lacking UNC93B show less splenomegaly 22 days after i.v. infection with MHV68. 11-12 week old B6N, UN93B-/-, TLR13-/-, B6J, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice were infected intravenously with 2 x 106 PFU of MHV68, and intact spleens from mice were weighed 22 dpi. Grey hexagons: uninfected mice (B6J genetic background), closed circles: mice of B6N genetic background, closed diamonds: mice of B6J genetic background. Each symbol corresponds to the value from one individual mouse, and horizontal lines indicate means. Results are combined from three independent experiments and include at least 8 mice per genotype. Statistical significance was determined using Mann-Whitney test. The experiments were performed together with Kendra Bussey.

Spleens of WT B6N mice showed signs of splenomegaly with a 2.5 times increase in spleen size when compared to uninfected B6J mice. On the contrary, splenomegaly was reduced in UNC93B-/- mice when compared to spleens from MHV68 infected WT B6N mice (p = 0.0008), with spleens from UNC93B-/- mice (n = 14) weighing on average 0.17791 g while those from WT B6N mice (n = 10) weighed on average 0.23593 g. Spleens from MHV68 infected TLR13-/- mice on average did not show a difference in size when compared with spleens of WT B6N mice. Spleens from infected WT B6J mice weighed 0.17921 g on average and showed significantly lower (p = 0.0012) spleen weights when compared to MHV68 infected WT B6N spleens. Spleens from TLR2-/- and TLR7-/- mice infected with MHV68 were of similar sizes as spleens from WT B6J mice. TLR9-/- mice showed slightly enhanced splenomegaly when compared to WT B6J upon MHV68 infection, though the difference was not statistically significant. Spleens

102

3. Results

from infected TLR7/9-/- showed no difference when compared to spleen sizes of WT B6J mice (Figure 36).

In summary, spleens from UNC93B-/- mice lacking the function of all endosomal TLR showed a statistically significant decrease in splenomegaly when compared to spleens from WT B6N mice upon MHV68 infection. In addition, WT B6N mice latently infected with MHV68 had larger spleens than WT B6J mice.

3.4 Role of TLR in the establishment of latency and reactivation of MHV68 after intravenous infection

After lytic infection, MHV68 establishes latency in the splenocytes, particularly in B cells. Haas and colleagues intraperitoneally infected C57BL/6, TLR7-/- and TLR9-/- mice with 1 x 105 PFU of MHV68 to assess the frequency of MHV68 genome-positive cells and the rate of reactivation from splenic B cells 34 days post infection. Lack of TLR7 or TLR9 did not have an impact on the establishment of latency in B cells, but TLR9-/- B cells showed higher rates of reactivation ex vivo (Haas et al., 2014). Similarly, TLR9 deficient splenocytes from mice infected intraperitoneally with 5 x 105 PFU of MHV68 showed an increase in reactivation when compared to reactivation from splenocytes of WT mice (Guggemoos et al., 2007).

In vitro stimulation of S11 latently MHV68-infected murine B cells with the TLR7 ligand R848 and the TLR9 ligand CpG has been shown to activate NFκB and lead to the suppression of MHV68 reactivation (Haas et al., 2014). Additionally, when C57BL/6 mice were infected intraperitoneally with 105 PFU MHV68, and then treated with the TLR7 ligand R848 from day 1 post infection to day 19 post infection, ex vivo splenocytes showed a 7-fold increase in MHV68 ORF73 expression when compared to splenocytes from untreated mice. This finding suggests that triggering TLR7 during the primary infection of MHV68 promotes the establishment of latency by increasing the reservoir size of latent MHV68 (Haas et al., 2014).

A study utilizing i.n. infection of MyD88 deficient mice showed a decrease in the frequency of B cells harboring MHV68 and hence the establishment of splenic latency when compared to those from WT mice (Gargano et al., 2008). MyD88 deficient splenocytes from mice infected intranasally with 103 PFU of MHV68 further showed a decrease in reactivation of MHV68 when compared to those from WT splenocytes (Gargano et al., 2008). In summary, TLR7 and 9, as well as other Myd88-dependent TLR, may be important for the establishment of latency, as well as reactivation from latency (Gargano et al., 2008; Guggemoos et al., 2007; Haas et al., 2014).

103

3. Results

In order to expand on these studies, clarify the role of individual TLR, and analyze the synergistic role of endosomal TLR in the establishment of latency and reactivation of MHV68, WT B6N and B6J, TLR2-/-, TLR7-/-, TLR9-/-, TLR7/9-/- and TLR13-/- mice were infected intravenously with 2 x 106 PFU of MHV68, and spleens from infected animals were harvested 22 dpi. Up to 3 spleens were combined based on gender and genotype of the infected mice and intact splenocytes were prepared. A series of eight, three-fold serial dilutions starting from 1.5 x 105 splenocytes/well were prepared and replicates were plated in a background of uninfected NIH3T3 cells in 96 well plates (Figure 37). To assess the presence of MHV68 genome copies in latently infected splenocytes, we utilized a nested PCR assay for the open reading frame (ORF) 50 sequence of MHV68 (Weck et al., 1999). Briefly, splenocytes and background NIH3T3 cells were lysed and 10 replicates of each of the lowest 5 dilutions, containing 5556 splenocytes/ well to 69 splenocytes/well, were analyzed by ORF50 nested PCR (Figure 37). The frequency of viral genomes was calculated as described (Section 2.5.3).

Figure 37: Preparation of splenocytes for limiting dilution nested PCR and ex vivo reactivation assay

To determine the frequency of splenocytes containing virus that are capable of reactivating from latency, we used an ex vivo reactivation assay. Twenty four replicates of all 8 splenocyte dilutions were plated on NIH3T3 cells (Figure 37). Fourteen days after plating, wells were inspected using light microscopy for cytopathic effect caused by reactivating. The frequency of

104

3. Results

reactivation was calculated as described in section 2.17. To detect the presence of preformed infectious virus, all dilutions of infected splenocytes were subjected to two freeze-thaw cycles, plated on NIH3T3 cells, and cytopathic effect was examined. Results show for reactivation assays exclude preformed virus.

3.4.1 The role of TLR in the establishment of latency

To quantify the frequency of latently infected splenocytes, we used a limiting-dilution PCR for ORF50 as described (Figure 37). At 22 dpi, we detected more MHV68 viral genome copies in female B6N mice compared to female UNC93B-/- mice in only 1 dilution of splenocytes (Figure 38A). However, the sigmoidal dose-response curves differed (Figure 38A), resulting in a calculated MHV68 genome load of 1 viral genome for every 373 splenocytes from female B6N mice, compared to 1 in 867 for female UNC93B-/- mice (Table 24). The number of viral genome copies differed between male B6N mice and male UNC93B-/- mice at multiple dilutions of splenocytes (Figure 38A). Based on the sigmoidal dose-response curve, an MHV68 genome load of 1 viral genome for every 580 splenocytes from male B6N mice, and 1 in 618 for male UNC93B-/- mice was observed (Table 24). We detected more MHV68 viral genome copies in female B6J mice compared to female TLR7-/- mice in only 1 dilution of splenocytes (Figure 38B). However, the sigmoidal dose-response curves differed (Figure 38B), resulting in a calculated MHV68 genome load of 1 viral genome for every 279 splenocytes from female B6J mice, compared to 1 in 458 for female TLR7-/- mice (Table 24). We detected similar MHV68 viral genome copies from splenocytes from male TLR7-/- mice as we did for splenocytes from male B6J mice, with frequencies of 1 in 173 and 1 in 237, respectively (Figure 38B and Table 24).

The viral genome load in splenocytes from female TLR9-/- mice, 1 in 294, was similar to that from splenocytes from female B6J mice, which was 1 in 279 (Figure 38C and Table 24). Contrary to the observation from female TLR9-/- mice, TLR 9-deficient male mice showed a reduction in frequency of splenocytes harboring the virus when compared to cells from male WT mice (Figure 38C). Based on the sigmoidal dose response curve, the viral genome load in splenocytes from male TLR9-/- mice, 1 in 589, was 2.5 times lower than that from splenocytes from male B6J mice, which was 1 in 237 (Table 24).

The viral genome load in splenocytes from female TLR7/9-/- mice, 1 in 399, was lower than that from splenocytes from female B6J mice, which was 1 in 279 (Figure 38D and Table 24). Male TLR7/9-/- mice showed a reduction in frequency of splenocytes harboring MHV68 when compared to splenocytes from male WT mice (Figure 38D). Based on the sigmoidal dose

105

3. Results

response curve, the viral genome load in splenocytes from male TLR7/9-/- mice, 1 in 610, was 2.5 times lower to that from splenocytes from male B6J mice, which was 1 in 237 (Table 24). To ensure that the frequency of MHV68 DNA copies detected by PCR represent latent infection, we controlled for the presence of preformed virus by plating disrupted splenocytes onto NIH3T3 cells and scoring for CPE after 14 days. The amount of CPE observed was negligible, indicating that preformed virus was not present and the viral genome amplified by PCR stemmed from latent infection.

106

3. Results

107

3. Results

Figure 38: Absence of UNC93B, TLR7, TLR9 and TLR7/9 promote establishment of latent MHV68 infection in vivo. 8-12 week old WT (B6N and B6J) UNC93B-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice were infected intravenously with 2 x 106 PFU of MHV68 and on day 22 post infection spleens were harvested. The number of splenocytes harboring viral genomes was determined by limiting dilution PCR targeting MHV68 ORF50. A minimum of 2 mice were used per experiment and two experiments were combined per data set with the exception of UNC93B males (A) where only one experiment is represented. All data are represented as mean percentages of positive reactions ± S.D. A sigmoidal dose-response curve was fitted by nonlinear regression analysis using Prism 6.0. The experiments were performed together with Kendra Bussey.

In summary, we observed a difference between male and female mice with respect to the establishment of MHV68 latency after intravenous infection. Female UN93B-/- and TLR7-/- mice showed a decrease in latently infected splenocytes when compared to female B6N and B6J mice, respectively. This decrease in latently infected cells was not observed in male UN93B-/- and TLR7-/- mice. The establishment of MHV68 latency in splenocytes was dependent on endosomal TLR9. Gender played an important role in the establishment of latency, as male mice lacking TLR9 had fewer splenocytes positive for MHV68 genomic DNA, which was not observed in splenocytes from female mice. A similar gender dimorphism was observed in TLR7/9-/- mice, with male mice showing a decrease in latently infected cells when compared to WT mice.

3.4.2 The role of TLR in reactivation of MHV68

We next analyzed the role of TLR in reactivation of MHV68. First we examined ex vivo reactivation of MHV68 from splenocytes of female and male B6N, UNC93B-/- and TLR13-/- mice. In female UNC93B-/- mice, MHV68 reactivation was observed from 1 in 30,086 splenocytes resulting in a 1.7 times increase in the frequency of MHV68 reactivation when compared to female B6N mice where reactivation was observed from 1 in every 52,896 splenocytes (Figure 39A and Table 24). UNC93B-deficient male mice showed a 2.9 times increase in frequency of MHV68 reactivation (Figure 39A). One in every 34,430 splenocytes from male UNC93B-/- mice reactivated, compared to those from male B6N mice which showed reactivation from 1 in 98,437 splenocytes (Table 24). Contrary to the enhanced reactivation observed in UNC93B-deficient splenocytes (Figure 39A), TLR13-/- splenocytes from female mice did not display a vast difference in reactivation when compared to splenocytes from female B6N mice, with reactivation frequencies of 1 in 74,474 and 1 in 52,896, respectively (Figure 39B and Table 24). TLR13-/- splenocytes from male mice reactivated similarly to splenocytes from male B6N mice, with reactivation frequencies of 1 in 77,118 and 1 in 98,437, respectively (Figure 39B and Table 24).

108

3. Results

Figure 39: Reactivation of MHV68 from splenocytes differs among WT B6N mice, UNC93B-/- and TLR13-/- mice. 8-12 week old WT (B6N) UNC93B-/- and TLR13-/- mice were infected intravenously with 2 x 106 PFU of MHV68 and on day 22 post infection spleens were harvested. MHV68 reactivation from splenocytes was determined by ex vivo reactivation assay. A minimum of 2 mice were used per experiment. Three experiments were combined for UNC93B-/- mice and controls and two experiments were combined for WT B6N and TLR13-/- mice and controls. All data are represented as mean percentages of positive reactions ± S.D. A sigmoidal dose-response curve was fitted by nonlinear regression analysis using Prism. The experiments were performed together with Kendra Bussey.

In summary, the increase in reactivation from UNC93B-/- splenocytes from female mice (Figure 39A), complimented by the decrease in the frequency of viral genome-positive UNC93B-/- splenocytes from female mice (Figure 38A) indicates that UNC93B plays an essential role in the control of MHV68 of latency in female mice. An increase in reactivation from UNC93B deficient splenocytes from male mice indicates that following intravenous infection, the absence of UNC93B in both male and female mice contributes to MHV68 reactivation, albeit to a higher extent in females (Figure 39). The absence of TLR13, on the other hand, does not contribute to reactivation of MHV68 in intravenously infected male and female mice (Figure 39B).

109

3. Results

Next, we examined ex vivo reactivation of MHV68 from splenocytes of female and male B6J, TLR2-/- , TLR7-/- , TLR9-/- and TLR7/9-/- mice. The absence of TLR2 in splenocytes from female mice only very slightly affected the frequency of reactivating splenocytes (Figure 40A). One in 77,342 splenocytes from female TLR2-/- mice showed reactivation of MHV68 when compared to 1 in 87,550 splenocytes of female WT B6J mice. Splenocytes from male TLR2-/- mice on the contrary showed a 2 times decrease in frequency of MHV68 reactivation with 1 in 102,430 splenocytes reactivating, compared to 1 in 48,861 splenocytes from male B6J mice (Table 24).

The absence of TLR7 in splenocytes from female mice led to enhancement of reactivation. TLR7 deficient splenocytes reactivated 2.3 times more efficiently than those from female B6J mice with reactivation frequencies of 1 in 37,276 and 1 in 87,550, respectively. On the contrary, splenocytes from male TLR7-/- mice showed a similar frequency of MHV68 reactivation to those from B6J mice, with reactivation frequencies of 1 in 55,898 and 1 in 48,861, respectively (Figure 40B).

Similar to the enhanced reactivation observed in TLR7-deficient splenocytes from female mice, TLR9-deficient splenocytes from female mice reactivated 3.6 times more efficiently than B6J mice, with reactivation frequencies of 1 in 24,283 and 1 in 87,550, respectively (Figure 40C) (Table 24). Similarly, splenocytes from male TLR9-/- mice reactivated 1.5 times more efficiently than B6J mice, with reactivation frequencies of 1 in 31,128 and 1 in 48,861, respectively (Figure 40C and Table 24). As observed for splenocytes from female UNC93B-/-, TLR7-/- and TLR9-/- mice, TLR7/9-/- splenocytes from female mice reactivated 5.7 times more efficiently than B6J mice (Figure 40D), with reactivation frequencies of 1 in 15,377 and 1 in 87,550, respectively (Table 24). On the contrary, the absence of TLR7 and 9 did not greatly alter the frequency of reactivating splenocytes in from male mice, when compared to male those from WT B6J mice, with reactivation frequencies of 1 in 39,121 and 1 in 48,861, respectively (Figure 40D and Table 24).

110

3. Results

111

3. Results

Figure 40: Frequency of MHV68 reactivation from splenocytes differs among WT B6J, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice. 8-12 week old WT (B6J) TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice were infected intravenously with 2 x 106 PFU of MHV68 and on day 22 post infection spleens were harvested. MHV68 reactivation from splenocytes was determined by ex vivo reactivation assay. A minimum of 2 mice were used per experiment and two experiments were combined for all genotypes. All data are represented as mean percentages of positive reactions ± S.D. A sigmoidal dose-response curve was fitted by nonlinear regression analysis using Prism. The experiments were performed together with Kendra Bussey.

In summary, the absence of TLR2 did not contribute to reactivation of MHV68 in female mice, but in splenocytes from male mice, slightly less reactivation was observed in the absence of TLR2. Endosomal TLR7 and 9 were found to play an essential role in the reactivation of MHV68. The increase in reactivation of MHV68 from splenocytes from female TLR7-/- mice (Figure 40B), intensified by the decrease in the frequency of viral genome-positive TLR7-/- splenocytes from female mice (Figure 38B), suggests that following intravenous infection, the absence of TLR7 in female mice contributes to MHV68 reactivation. This finding is not reflected in splenocytes from male TLR7-/- mice, which display neither a vast difference in number of latently infected cells (Figure 38C) nor an increase in reactivation (Figure 40C). Therefore, this result is indicative that gender may play an important role in the establishment of latency and reactivation of MHV68 in TLR7-/- mice.

The increase in reactivation of MHV68 from splenocytes from female and male TLR9-/- splenocytes (Figure 38C) and the decrease in latently infected splenocytes in the absence of TLR9 in male mice (Figure 39C) suggest that TLR9 plays a crucial role in the establishment of latency and virus reactivation. The increase in reactivation from splenocytes lacking TLR7 or TLR9 is well reflected in data from mice deficient of both endosomal TLR. Splenocytes from female TLR7/9-/- mice show an increased rate of reactivation when compared to splenocytes from B6J mice. The frequency of reactivation from male TLR7/9-/- mice is comparable to that from B6J mice, but as the cells show a lower frequency of latently infected cells when compared to B6J mice, there appears to be a contribution of TLR7 and 9 to virus reactivation in male mice infected intravenously with MHV68.

112

3. Results

Table 24: Comparison of frequency of latently infected cells and frequency of MHV68 reactivation Splenocytes Frequency of latently infected cells Reactivating cell frequency B6N 1 in 373 1 in 52,896 UNC93B-/- 1 in 867 1 in 30,086

TLR13-/- n.d 1 in 74,474

B6J 1 in 279 1 in 87,550 TLR2-/- n.d 1 in 77,342

Females TLR7-/- 1 in 458 1 in 37,276 TLR9-/- 1 in 294 1 in 24,283 TLR7/9-/- 1 in 399 1 in 15,377 B6N 1 in 580 1 in 98,437 UNC93B-/- 1 in 618 1 in 34,430 TLR13-/- n.d 1 in 77,118 B6J 1 in 237 1 in 48,861

TLR2-/- n.d 1 in 102,430 Males TLR7-/- 1 in 173 1 in 55,898 TLR9-/- 1 in 589 1 in 31,128 TLR7/9-/- 1 in 610 1 in 39,121 n.d: not determined.

113

4. Discussion

4 Discussion

4.1 Establishment of an autologous model to study gammaherpesvirus infections

We aimed to establish an autologous infection model with MmusRHV1, a novel murine gammaherpesvirus, in its natural host Mus musculus, and to characterize the natural infection and transmission of MmusRHV1. Furthermore, we wanted to isolate and characterize the murine betaherpesvirus MCMV2. For this purpose we utilized wild mouse colonies bred in captivity, and evaluated the presence of MmusRHV1 and MCMV2 in these colonies. A non- invasive blood screen was established to analyze the presence of MmusRHV1 in mice and to study natural transmission of the virus within a colony. Our non-invasive screen bore evidence of vertical transmission of MmusRHV1 from positive parents to offspring, though the exact time frame of the transfer was inconclusive. As the study involved mice originally captured in the wild harboring their natural infections, it served as a model to decipher the role of co-infections. The results revealed that the presence of parasitic co-infections might influence the gammaherpesvirus life cycle and consequently its detection by PCR. These data suggest that studying gammaherpesviruses in a naturally infected, autologous mouse model might reveal novel insights about natural transmission of the virus and the effects of co-infections on the status of herpesvirus infections.

4.1.1 Isolation of MmusRHV1

In order to characterize the genome and study the biology of the two novel viruses MmusRHV1 and MCMV2, we aimed to isolate the virus and propagate it in cell culture. For this purpose, we co-cultured B cells with the permissive murine cell lines M2-10B4 and NIH3T3 or transferred organ homogenates from lung and salivary glands of Horst mice onto the cells. We did not observe the presence of MmusRHV1 or MCMV2 in either of the experiments (Figure 12). However, presence of MCMV1 was observed in co-culture supernatants from salivary glands (Figure 12A and B) and confirmed by sequencing. This finding suggests that MCMV1 is likely to be more dominant in a cell culture system when compared to MmusRHV1 and MCMV2.

As 72.72% of the Horst colony analyzed tested positive for MCMV1 (Figure 11), it is highly unlikely that an animal positive for MmusRHV1 or MCMV2 but negative for MCMV1 is chosen. Therefore, it might be efficient to block MCMV1 in cell culture specifically, by using external guide sequence (EGS), a technique that consists of targeting and cleaving specific mRNA by using derived tRNA and RNase P (Forster A.C and Altman S, 1990) Jiang and colleagues constructed an efficient EGS to target the mRNA encoding the protease of murine

114

4. Discussion

cytomegalovirus (MCMV) (mPR), which is essential for viral DNA encapsidation and replication (Jiang et al., 2012). This novel technology could be utilized to specifically inhibit MCMV1 in cell culture while allowing MmusRHV1 to replicate. Whether this would serve as an efficient strategy to isolate MCMV2 is difficult to speculate without further sequence information of the novel virus which might share sequence homology to MCMV1.

In order to chemically induce reactivation of MmusRHV1, we stimulated the isolated B cells in a co-culture system with multiple stimulants like 12-O-tetradecanoylphorbol-13-acetate (TPA) (Hausen et al., 1978; Davis et al., 2001; Adler et al., 2002), sodium butyrate (Gregory et al., 2009) and lipopolysaccharide (LPS) (Cook et al., 2006) which have all been shown to reactivate herpesviruses from latently infected cells. In addition to commercially available stimulants, co- cultured cells were stimulated with soluble egg antigens (SEA) which has been shown to induce herpesvirus reactivation (Everts et al., 2009; Reese et al., 2014). Intriguingly, inducing reactivation of co-cultures did not lead to the detection of MmusRHV1 (data not shown).

First, as we could not detect the presence of MmusRHV1 by PCR (data not shown) it is difficult to speculate if the animal was negative for the virus, or if the viral load was below the detection limit of our nested PCR. If the animal was indeed positive of the virus, one of the probable reasons for the absence of MmusRHV1 in the co-culture system after stimulation might be attributed to the activation of NFκB. Though stimulation of latently infected splenocytes with LPS, which leads to the activation of NFκB via TLR4 has been shown to result in a significant increase in the frequency of MHV68 reactivation (Moser et al., 2004), it is worth noting that induction of NFκB has been previously shown to inhibit reactivation of gammaherpesviruses MHV68, EBV and HHV-8 (Brown et al., 2003; Haas et al., 2014). It is possible that similar to stimulation with LPS, stimulation of latently infected cells with TPA which has been previously shown to activate NFκB signaling (Kundu et al., 2006; Osborn et al., 1989) might favor inhibition of MmusRHV1 over its reactivation. However, it is noteworthy that MCMV has been shown to reactivate via NFκB signaling (Cook et al., 2006). Studies suggest that a mediator or signal that is capable of activating NFκB might be involved in reactivating CMV from latency, as both human and murine major immediate early (MIE) enhancer/promoter regions of CMV contain numerous NFκB consensus sequences (Dorsch-Häsler et al., 1985; Sambucetti et al., 1989). Therefore, stimulating NFκB might lead to the reactivation of MCMV from latency. As a result of working with samples that were probably co-infected with both MCMV1 and MmusRHV1, these variations in pathways of the different subfamilies of herpesviruses might have led to the

115

4. Discussion

presence of MCMV1 in cell culture. Absence of MCMV2 in co-cultured samples might be attributed to low prevalence (9%) of the virus within the Horst colony (Figure 11C).

Contrary to LPS and TPA, sodium butyrate inhibits NFκB expression. This is accomplished by upregulating T helper 2 (TH2) cytokines, and preventing the translocation of NFκB to the nucleus and hence inhibiting the transcription of TNFα (Meijer et al., 2010). Similar to sodium butyrate,

SEA has also been shown to upregulate TH2 cytokines (Reese et al., 2014). Absence of MmusRHV1 growth in the co-culture stimulated with sodium butyrate and SEA might be attributed to the lacking permissibility of M2-10B4 and NIH3T3 cells for the reactivation of MmusRHV1. As the growth of this novel virus has not been studied in a cell culture system previously, it is difficult to speculate whether these cell lines support the growth of the virus or otherwise.

Taking these factors into consideration, it is difficult to conclude from our attempts to isolate MmusRHV1 that the virus fails to reactivate and replicate in a cell culture system. To determine whether reactivation of MmusRHV1 in a cell culture setting is inhibited by the presence of NFκB, further experiments cultivating B cells from mice naturally harboring the virus with cells deficient in NFκB signaling could be performed.

4.1.2 Characterization of the natural transmission of MmusRHV1 in Mus musculus

As transmission of gammaherpesviruses has never been studied in an autologous model, we designed experiments to study the natural transmission of MmusRHV1.

Vertical transmission of MmusRHV1

While the colony was still maintained with up to 40 animals in a single cage system, our initial screen showed a high prevalence of MmusRHV1 within the Horst colony (86.36%) (Figure 11A). Multiple studies have previously shown evidence of vertical transmission of gammaherpesviruses from positive mothers to their offspring (Bourboulia et al., 1998; Wilkinson et al., 1999; Hricova and Mistrikova, 2008). Therefore, we hypothesized that the most probable route of transmission was likely to be vertical transmission from positive parents to their offspring. We analyzed “families” of Horst mice (established in November 2013) for the presence of virus in the members by using a non-invasive blood screen (March 2014, Figure 41). The screen revealed the presence of MmusRHV1 in offspring that were weaned from their mothers at 3 weeks of age (Figure 14 and Table 21). This finding indicates that MmusRHV1 can be transmitted from parent to child before weaning and the data are consistent with data from Štiglincová and colleagues who chronically infected six week-old female BALB/c mice

116

4. Discussion

intranasally with 2 x 104 PFU of MHV68 and found that newborns from infected mothers were positive for the virus. This represents proof of vertical transmission via the placenta of MHV68 from pregnant females to fetuses and consequently to newborn mice (Štiglincová et al., 2011).

Determination of the time frame of vertical transmission of MmusRHV1

In order to better understand the time frame of this vertical transmission of MmusRHV1 from positive parent to offspring, we determined the presence of MmusRHV1 in litters from seropositive parents on day 1, 10 and 19 after birth (July 2014) (Figure 41). Contrary to our findings from the non-invasive blood screen (Figure 14), presence of MmusRHV1 in offspring was not detected (Table 22). Upon further investigation, parents who were previously seropositive did not show presence of MmusRHV1 in blood or organs. The absence of MmusRHV1 in the offspring might hence be attributed to the failure in detection of MmusRHV1 from previously positive parents. Telfer and colleagues tested blood samples from wood mice for antibodies against MHV68. Mice which were previously positive, tested positive in subsequent captures, interestingly, up to 65% of the positive animals tested negative on at least one occasion after the positive test result (Telfer et al., 2007). Though our data indicating change in viral status in the blood over time are in agreement with data from Telfer and colleagues, PCR analysis of spleen samples from both parents and offspring were negative. This led us to consider that change in viral status in the blood over time could not be the only cause for the lack of transmission of MmusRHV1 from parents to offspring. We speculated that treatment with Ivermectin, an antiparasitic drug might play a vital role in the change of viral status of the animals.

Breeding pairs with positive parents were designed based on the initial non-invasive blood screen conducted in March 2014, which was performed immediately after completion of the 12- week long treatment with Ivermectin, which is indicated in red in figure 41. In July 2014, nested PCR for MmusRHV1 failed to detect presence of the virus in the blood and organs of previously seropositive mice and not surprisingly, their new born litter. Possible repercussions due to treatment with Ivermectin are further discussed in section 4.2.

117

4. Discussion

Horizontal transmission of MmusRHV1

We hypothesized horizontal transmission of MmusRHV1 as another possible route, and to test our hypothesis, we set up two types of breeding pairs (May 2014): one type of breeding pair comprising a positive (by blood PCR) Horst female and a naïve C57BL/6J male, and another type of breeding pair with a positive (by blood PCR) Horst male and a naïve C57BL/6J female (Figure 15A and B). We examined the blood and organ samples (in July 2014) from these animals for the presence of MmusRHV1 and did not observe seroconversion of naïve partners in either direction. This finding was in contrast to a recent study where MHV68 was revealed to be sexually transmitted between experimentally infected laboratory mice and their naïve sexual partners (François et al., 2013). Six week-old female BALB/c mice infected intranasally with 104 PFU of luciferase+ MHV68 showed the presence of infectious virus in the genital tract, with evidence of sexual transmission of MHV68 to naïve males, where males revealed light emission in the male genital region around 4 days after contact with female, and the signal peaked around 10 days and remained for 3 weeks (François et al., 2013).

A couple of factors might contribute to the dissimilarities between the two studies. First, MmusRHV1 and MHV68 are different viruses and have been shown to branch distantly from each other on phylogenetic trees (Ehlers et al., 2007). The content of nonconserved genes which represents the individual makeup and the basis for their unique pathogenic properties may vary substantially between MmusRHV1 and MHV68. Second, it is to be expected that mice infected with MHV68 intranasally with 104 PFU exhibit a different transmission pattern compared to the transmission of naturally prevalent MmusRHV1 in our Horst mice. Third, while François and colleagues used a laboratory model of 6 week old BALB/c mice infected with MHV68, we examined the transmission of a naturally infected virus in C57BL/6 mice, which have been shown to be more resistant to MHV68 infection than BALB/c mice due to their differences in levels of viral gene expression between the two mouse strains (Weinberg et al., 2004). Additionally, the age of the mice used might also contribute to the observed differences. All adult mice used in our horizontal transmission study were definitively older than 6 weeks (age was not defined for adults that were born before arrival at the quarantine unit).

As with the experiment to determine the time frame of vertical transmission of MmusRHV1, it is important to note that the final analysis for evidence of horizontal transmission was conducted in September 2014, and that the treatment with Ivermectin might have contributed to our observations. The possible mechanism by which treatment with an anti-parasitic drug might lead to change in viral status is discussed in the following section.

118

4. Discussion

119

4. Discussion

Figure 41: Timeline depicting course of experiments. Region in red represents 12 weeks of treatment with Ivermectin, region in blue represents time period when animals tested seropositive by nested PCR and region in grey represents time period when animals tested seronegative by nested PCR.

4.2 Characterization of co-infections and its effect on herpesvirus latency in naturally infected mouse models

Although standardized mouse studies are generally conducted in specific pathogen-free animals of different hygienic status, our data suggest that there is additional complexity due to natural co-infections of the host, particularly in situations where ecto- and endoparasites are present and may have the ability to interfere or regulate chronic viral infections.

A hygiene status test by mfg diagnostics, Wendelsheim, Germany revealed a variety of pathogens in the Horst mice (Table 20). Alongside co-infections with MCMV and mouse hepatitis virus (MHV), the mice were positive for multiple species of bacteria like Helicobacter spp, and Staphylococcus spp. Ectoparasite Acarina was found during coat examination of both mice, and endoparasites Syphacia spp and Trichomonas spp were detected by light microscopy (Figure 13). Due to the presence of both ecto- and endoparasites, mice were treated with 13 ppm of Ivermectin for 12 weeks (time period is depicted in red) via their diet (Figure 41). Animals which tested positive for MmusRHV1 in the non-invasive blood screen in March 2014, tested negative by August 2014. Eradication of ecto- and endoparasites might be causal in the failure to detect MmusRHV1 in samples from previously positive animals and multiple hypotheses need to be taken into consideration in order to thoroughly evaluate the role of parasitic co-infections and its effect on the reactivation of gammaherpesviruses.

Promoter arthropod hypothesis

The ‘promoter arthropod hypothesis’ suggests that insect blood feeding increases KSHV transmission through viral reactivation via inflammatory mechanisms associated with the bite (Ascoli et al., 2009; Coluzzi et al., 2002). Similarly, hypersensitivity to arthropod bites has been reported to induce viral reactivation of EBV (Asada, 2007). The saliva of these arthropods assists their hematophagy and contains a range of cytokine modulators, that trigger itching and the development of species-specific local and systemic inflammatory responses in the host

(Titus et al., 2006), suggesting that the bite polarizes adaptive immune response to a TH2 profile

(Ferreira and Silva, 1999). TH2 cells cause an allergic reaction with the secretion of IL-4, IL-5,

IL-9 and IL-13. T helper 1 and 17 (TH1 and TH17) on the contrary, drive autoimmunity with the aid of macrophages and DC which produce an abundance of pro-inflammatory cytokines like IFNγ, TNFα, IL-6, IL-12 and IL-17 (reviewed by Khan and Fallon, 2013). It is likely that before

120

4. Discussion

the treatment with Ivermectin, the presence of mites (Figure 13) in the colony helped to induce a

TH2 response in the mice, which in turn facilitated the lytic state of MmusRHV1, resulting in the detection of lytic virus by nested PCR of blood samples (Table 21) (Summarized in Figure 42).

Other studies indicate that insect bites can potentiate herpesvirus transmission via saliva of the host (when insect bite lesions are licked by another host whose saliva carries the virus) (Coluzzi et al., 2002). This ancestral behavior has been hypothesized to play a role in the transmission of HHV-8 transmission (Coluzzi et al., 2002; Coluzzi et al., 2003). It is also noteworthy that a sharp decrease in HHV-8 seroprevalence was observed between the 1945–1950 and 1951–1955 birth cohorts among Sassari blood donors and coincided with vector control measures carried out at that time (Coluzzi et al., 2003). Although this ecological level data is interesting, the link between arthropods and gammaherpesviruses may be confounded by other environmental or behavioral factors, and additional detailed studies are needed to test this hypothesis. However, it is possible that a similar phenomenon occurs among mice, and grooming, biting and fighting among animals in large cage settings before their transfer to the quarantine facility might have led to the continuous spread of the virus among animals co-housed in large groups. After transfer into new IVC systems, with only up to 6 mice per cage, mice continued to exhibit signs of extensive grooming and bites, but did not indicate spread of virus among co-housed animals. At present, we can only speculate about the contribution of the different housing regiments to the loss of detectable MmusRHV1 as the change in housing regiment coincided with the treatment with Ivermectin (Figure 41). Specific controlled studies would be necessary to be able to experimentally distinguish the potential contributing factors.

Presence of intestinal helminths promotes "two signal" mechanism and induces reactivation of gammaherpesviruses

Another hypothesis involves co-infection of gammaherpesviruses with intestinal helminths (endoparasites). A study showed that seroprevalence of KSHV was higher in women with malaria parasitaemia, hookworm and Mansonella perstans co-infections and the seroprevalence correlated with increasing intensity of hookworm infection, likely due to the impact of these parasites on the immune function of the host (Wakeham et al., 2011).

Reese and colleagues examined the effects of acute infection with nematode Heligmosomoides polygyrus and trematode Schistosomiasis mansoni in C57BL6/J mice latently infected intraperitoneally with MHV68-M3-FL (a recombinant MHV68 expressing a firefly luciferase gene driven by the viral M3 promoter (M3FL)) and imaged between days 47 and 53 following infection. Both acute H.polygyrus infection and S. mansoni egg challenge led to better

121

4. Discussion

reactivation of MHV68 when compared to untreated mice (Reese et al., 2014). Intestinal helminths generate a strong TH2-driven cytokine response, which counters biological effects of IFNγ and IL-12 and drives the activation of macrophages with an M2 (immunoregulatory) instead of an M1 (proinflammatory) phenotype (McSorley et al., 2013). It is possible that presence of the intestinal parasites Syphacia spp and Trichomonas spp (Figure 13) led to natural reactivation of MmusRHV1 in Horst mice via helminth-induced TH2 polarization.

Interlukin 4 (IL-4) is another cytokine that induces TH2 differentiation and inhibits TH1 differentiation (reviewed by Paludan.S.R, 1998). Bone marrow derived macrophages (BMDM) treated with IL-4 after MHV68 infection show increased viral replication. This suggests that IL-4 acts on replication of MHV68, rather than by increasing the number of infected cells. The observed enhancement in replication is dependent on the TH2-associated transcription factor

STAT6, and also occurs upon stimulation with IL-13, another TH2-associated cytokine (Reese et al., 2014). IL-4 and IL-13 were shown to transactivate N4/N5 promoter of MHV68 gene 50, a homolog of KSHV RTA/ORF50, and antagonize IFNγ-mediated suppression of gene 50. Upon being induced by IL-4 and IL-13, STAT6 promotes viral replication by binding to the N4/N5 promoter of MHV68 gene 50 (Reese et al., 2014).

Similar to the observation by Reese and colleagues about the "two signal" mechanism by which co-infection can induce reactivation inhibiting IFNγ and inducing IL-4 (summarized in Figure 42), it is possible that the presence of Syphacia spp and Trichomonas spp (Figure 13) in the Horst mice before treatment with Ivermectin might have contributed to the reactivation of MmusRHV1, and hence resulted in its detection by nested PCR and continuous spread within the colony (Table 21).

However, not all parasitic co-infections lead to reactivation of gammaherpesviruses (summarized in Figure 42). Intriguingly, induction of EBV latent membrane protein 1 (LMP-1) a viral mimic of the CD40–CD40 ligand interaction is known to suppress virus reactivation due to IL-4 and IL-13 mediated by STAT6 (Adler et al., 2002). Additionally, in murine malarial infections, Plasmodium spp provide constant stimulation of endosomal TLR (Pichyangkul et al., 2004), via MyD88, an essential adaptor molecule which induces IL-12 leading to the activation of EBV (Adachi et al., 2001; Coban C et al., 2005; Zauner et al., 2010). As we speculate that the failure to detect MmusRHV1 infection in Horst mice is due to the absence of ecto- and endoparasites as seen with KSHV, it is notable that the available amino acid sequences of MmusRHV1 clusters closer to the rhadinoviruses consisting of KSHV (Figure 3), than to

122

4. Discussion

lymphocryptoviruses consisting of EBV on phylogenetic trees (Ehlers et al., 2007) and hence might contribute to MmusRHV1 behaving similar to HHV-8 in vivo.

Figure 42: Influence of parasitic co-infections on latent herpesvirus infections

As we did not re-evaluate the health status of Horst mice in July 2014, it is important to note that our speculations about the correlation between parasitic co-infections and viral status in the animals are not conclusive and would need more thorough and systematic studies to address the hypotheses. Another confounding factor is the change in housing from large groups of mice with up to 40 animals in open conventional cages to smaller groups of mice in individually ventilated cages. It would be beneficial to obtain new mice from the original (captive) wild colony, determine the status of MmusRHV1 by the blood screen, the presence of parasites, and then monitor change in viral status of co-infected mice periodically upon treatment with Ivermectin. Co-infected mice that are left untreated should serve as a control to test if presence of parasites promotes viral replication. If a change in viral status would be observed in treated animals, a second health monitoring to confirm the absence of parasitic co-infections would help to ascertain if presence of parasites is detrimental to reactivation status of the virus. The experimental setup to determine the role of parasitic co-infections will certainly be less

123

4. Discussion

complicated once MmusRHV1 is isolated and can be used for experimental infection of mice. Studies similar to those conducted by Reese and colleagues could be performed in mice experimentally infected with MmusRHV1 to confirm our speculation.

As a closing remark, successful isolation and characterization of MmusRHV1 will serve as an important tool to understand autologous gammaherpesvirus infections in mouse models, and help comprehend the co-factors influencing viral infection in an autologous mouse model. More research needs to be done to investigate whether our initial findings can be verified or not.

4.3 Role of UNC93B and specifically TLR7 and 9 in the detection of MHV68 in vitro

In this study we utilized MHV68, a murine gammaherpesvirus, to investigate the role of TLR in the detection and control of MHV68 in vitro.

Our initial study of FACS sorted DC revealed that pDC are the major producers of IFNα upon MHV68 infection (Figure 16). MHV68 infected pDC produced 17 times more IFNα when compared to MHV68 infected FLDC. These results are in line with cell sorting data that indicated that only 38.3% of the total FLDC population stained as CD11cloSiglec-H+ pDC (Figure 18). Taking these data together in addition to the fact that FLDC and pDC were plated at similar cell numbers of 300,000 cells/well, it is evident that IFNα produced by total FLDC arises from the pDC present within the population. On the contrary, 300,000 cells/well of cDC purified from FLDC cultures did not lead to the production of IFNα upon infection with MHV68 (Figure 16). This finding indicates that MHV68 might be capable of inhibiting PRR responses and/or is not recognized efficiently by PRR in cDC. Another possible explanation might be the differential retainment of ligands in the endosomal compartments of pDC and cDC. While pDC are known for their unique character of retaining TLR-bound CpG in the endosomal compartments together with the MyD88–IRF7 complex for extended periods, in cDC, CpG is rapidly transferred to lysosomal vesicles to be degraded (Honda et al., 2005). This retention of CpG in pDC allows for a robust downstream signaling response from the endosome before its fusion with the lysosome and subsequent degradation (Honda et al., 2005). Therefore, it is possible that MHV68 is retained in the endosomal compartments of pDC longer than it is in cDC and hence induces a better response in pDC.

As observed with FACS sorted cDC, infection of GM-CSF derived cDC from WT, UNC93B-/-, TLR2-/-, TLR7-/- TLR9-/- and TLR13-/- mice with MHV68 at an MOI of 2, did not lead to IFNα production (Figure 17A). Our findings are congruent to those from a study by Sun and colleagues who failed to detect IFN bioactivity in the supernatant from GM-CSF DC infected

124

4. Discussion

with MHV68 (Sun et al., 2015). The failure of MHV68 to elicit an IFNα response in cDC might be due to its capability to evade detection by pattern recognition receptors. Notably, at high doses of MHV68 infection ranging from an MOI of 10 to 1000, virus infection stimulates the expression of TNFα, IL-6 and IL-1β in GM-CSF derived cDC (Sun et al., 2015). As we only monitored the production of IFNα (Figure 17A) and TNFα (data not shown) after infection with MHV68 at an MOI of 2, additional experiments using virus infection at higher MOI could be performed to analyze differences in IFNα and cytokine levels from various TLR-deficient GM-CSF derived cDC.

Contrary to GM-CSF derived cDC, pDC sorted from FLDC cultures from WT mice produced significant amounts of IFNα when infected with MHV68 at an MOI of 0.5 and 2 (Figure 21). Nevertheless, our observations indicate MHV68 is a poor inducer of TLR dependent IFNα when compared to the IFNα response from pDC upon infection with NDV and MCMV (Figure 19C and D). This might be attributed to the deamination of the MHV68 genome which specifically removes TLR9 stimulatory CpG motifs (Pezda et al., 2011).This property has been shown to be unique to only gammaherpesviruses but not alphaherpesviruses or betaherpesviruses which do not show CpG suppression (Karlin et al., 1994). However, the cause for the difference in IFNα levels among UNC93B-/-, TLR7-/- and TLR7/9-/- pDC upon infection with NDV (Figure 19C) is uncertain and additional experiments have to be performed to elucidate the possible role of other endosomal TLR in the detection of NDV in pDC.

Upon infection of TLR-deficient pDC with MHV68, we did not observe a decrease in IFNα production by TLR2-/- pDC when compared to WT pDC. This finding suggests that TLR2 does not play an important role in the detection of MHV68 in pDC at 24, 48 and 72 hpi (Figure 21). Similarly, we did not observe a reduction in IFNα levels from TLR7-/- pDC at 24, 48 and 72 hpi with MHV68 (Figure 21). Guggemoos and colleagues demonstrated that at 24 hours post infection with MHV68, IFNα levels was partially abolished from TLR9-/- FLDC (Guggemoos et al., 2007). In MHV68 infected pDC, we observed a complete dependency on TLR9 signaling for IFNα production at 24 hours post infection, and a partial, although strong, dependency on TLR9 signaling for IFNα was observed at 48 and 72 hours post infection. UNC93B-/- mice showed complete abolishment of IFNα at 24, 48 and 72 hours post infection. As pDC from TLR7/9-/- mice behaved as UNC93B-/- pDC, this finding suggests that besides TLR9, TLR7 is also involved in the detection of MHV68 in pDC.

The involvement of TLR7 raises an important question regarding the mechanism by which MHV68 is detected by ssRNA sensing TLR7 present in the endosomal compartment of pDC.

125

4. Discussion

Multiple hypotheses need to be carefully examined to determine the exact mechanism by which such recognition occurs. One possible scenario is that pDC engulf apoptotic debris from MHV68 infected cells that contains cellular mRNA which is delivered to the endosome leading to TLR7 triggering. However, it has been reported that pDC do not support lytic infection of MHV68 (Pezda et al., 2011). If that be the case, it is unlikely that the presence of viral nucleic acids in pDC is sensed within the endosomal compartments without infection-induced cell death. However, to confirm or deny this hypothesis, one could perform MHV68 growth curves in pDC and additionally monitor cell viability of pDC. A major limitation for such an approach will be the sparse cell numbers that are obtained after sorting for pDC from a total FLDC population.

It has recently been reported that TLR7 is synergistically activated by guanosine (G), 2′- deoxyguanosine (dG), 7-MetG, 8-hydroxyguanosine (8-OHG) and 8-hydroxy-2′-deoxyguanosine (8-OHdG) in the presence of ssRNA (Shibata et al., 2015). The study showed that the interaction between TLR7 and guanosine was enhanced by the presence of ssRNA, and TLR7 was activated in the presence of guanosine and poly U without the use of the transfection reagent DOTAP. This finding indicates that TLR7 is able to detect extracellular guanosine (Shibata et al., 2015). Therefore, it is probable that TLR7 recognizes G/dG rich regions of MHV68 in the presence of oligoribonucleotides (ORN). The exact mechanism of how such recognition might occur upon infection of pDC with MHV68 needs to be further elucidated.

A third possibility involves the presence of MHV68 encoded RNA that arrives with the virus, and hence can be detected in the endosome upon virus engulfment. Multiple studies report the detection of DNA viruses by RNA sensing TLR like TLR3 and TLR7. Upon infection with MCMV we observe partial abolishment of IFNα from TLR9-/- pDC, but complete abolishment of IFN response from UNC93B-/- and TLR7/9-/- pDC (Figure 19D). Our data is in concurrence with Zucchini and colleagues who report that TLR7 is involved in the detection of MCMV (Zucchini et al., 2008). Another study showed that KSHV is detected by TLR3, an endosomal TLR which detects dsRNA (West and Damania, 2008). Bechtel and colleagues have shown that KSHV encapsidates several viral transcripts in its virion (Bechtel et al., 2005). These transcripts could activate TLR3 upon virus entry and uncoating in the endosome. Other studies address the possibility of the presence of KSHV microRNA (miRNA) (Cai et al., 2005; Samols et al., 2005). These miRNAs, which are contained in the virion, might activate RNA sensing by TLR3 or TLR7 in the endosomal compartment. Recently, MHV68 encoded miRNA has also been reported (Feldman et al., 2014). MHV68 miRNAs demonstrate differential expression which is time- dependent. Feldman and colleagues monitored 17 mature miRNAs, and noted that 7 them

126

4. Discussion

miRNA were induced at high levels within 48 hours. Five of the miRNAs including mghv-miR- M1-2-5p, mghv-miR-M1-3-3p, mghv-miR-M1-8-5p, mghv-miR-M1-13-5p, and mghv-miR-M1-15- 5p, peaked between 16 or 24 hours post infection, while mghv-miR-M1-5-5p and mghv-miR-M1- 15-3 peaked at 48 hours post infection (Feldman et al., 2014). The detection of differential expression of MHV68 miRNAs by TLR7 might be the most probable possibility.

As we observe complete abolishment of IFNα from TLR9-/- pDC at 24 hpi, it is likely that TLR9 plays a crucial role in the initial detection of MHV68 in pDC. At 48 and 72 hpi however, IFNα production from TLR9-/- pDC is not completely abolished. pDC from TLR7/9 double deficient mice show complete abolishment at 24, 48 and 72 hpi. If MHV68 miRNA is expressed in pDC, this might explain the possible role of TLR7 in the detection of MHV68. Detection of miRNA by TLR7 expressed in TLR9-/- pDC might lead to the production of IFNα albeit at lower levels than the WT pDC which have functional TLR9 and TLR7. Feldmann and colleagues generated a miRNA mutant virus MHV68.Zt6, with significantly reduced expression of all 14 pre-miRNAs by two-step lambda Red-mediated recombination (Feldman et al., 2014). To verify our hypothesis, it would be ideal to perform IFNα ELISA on WT, UNC93B-/-, TLR7-/-, TLR9-/- and TLR7/9-/- pDC infected with WT MHV68 and MHV68.Zt6. Another approach could be to compare IFNα levels from Mcoln1-/- pDC which show impaired TLR7 responses to ssRNA (Li et al., 2015) and those from TLR7-/- pDC after infection with MHV68.

4.4 Role of TLR during MHV68 infection in vivo upon different routes of infection

4.4.1 Intranasal infection

Clinical signs of illness

After intranasal infection with 2 x 106 PFU of MHV68 there was no significant difference in weight loss of TLR-deficient mice or WT controls (Figure 23). Mice lost weight beginning on day 6 and recovered by day 12. Similar to our observations, Sunil-Chandra and colleagues reported that BALB/c mice infected intranasally with 4 x 105 PFU of MHV68 developed clinical signs of infection but recovered from their symptoms by days 10-12 (Sunil-Chandra et al., 1992). The failure to induce an inflammatory response is often linked with the absence of weight loss in mice (Moyron-Quiroz et al., 2004). As we do not see a significant decrease in viral titers from TLR-deficient mice after intranasal infection, it is possible that inflammatory cytokines are produced in a TLR independent manner and therefore fail to cause a difference in weight loss between TLR-deficient and WT mice.

127

4. Discussion

Organ titers

We monitored viral titers in the lungs, salivary gland and spleen on days 3, 6, 10 and 17 after infection. High viral titers were present in the lungs on day 3 and 6 post i.n. infection, and MHV68 was not detected either on day 10 or day 17 after infection (Figure 24). The spleen (Figure 26) did not show presence of MHV68 3 and 17 days post i.n. infection, but detectable viral titers were present by day 6 and still present on day 10. These results are similar to observations by Hwang and colleagues who infected BALB/c mice with 5 x 106 PFU of MHV68 expressing firefly luciferase by a viral M3 promoter (M3FL) intranasally, and monitored luciferase activity 5 days post infection (Hwang et al., 2008). Viral replication in the lung was increased days 1-7 and cleared by day 12 after infection. In the spleen, presence of virus was observed starting on day 10, and showed a peak 14 days post infection (Hwang et al., 2008). The salivary gland did not show presence of MHV68 3 and 17 days post i.n. infection, but detectable viral titers were present by day 6 and still present on day 10 (Figure 25). Intriguingly, these results disagree with observations by Hwang and colleagues who reported the presence of M3FL in BALB/c mice between days 10 and 18 post infection (Hwang et al., 2008). Further experiments must be performed to validate our observation of viral dissemination.

To analyze TLR involved in the detection of MHV68 after acute infection, we compared viral titers in the lung, salivary gland and spleen among WT and various TLR deficient mice. Though we did not observe any statistically significant differences in titers among the various mice in the lung, lungs from TLR7-/- and 3d mice showed slightly elevated titers on day 3 p.i. when compared to B6J mice. To confirm the role of TLR7 in the lungs after intranasal infection with MHV68, further experiments including more mice should be performed. Lungs from TLR2-/- mice showed slightly lower titers when compared to B6J mice (Figure 24A), leading our observation to contradict those by Michaud et al. Michaud and colleagues observed increased viral titers in lung homogenates of 4-6 week old TLR2-/- mice on C57BL/6 background 5 days post intranasal infection with 105 PFU of MHV68 (Michaud et al., 2010). The increase in viral titers from TLR2-/- mice was accompanied by a decrease in IL-6 levels from lung homogenates. The difference in data between the two studies might first be attributed to the age of the mice. While Michaud and colleagues utilized 4-6 week old mice, we used 8-10 week old mice for our experiments. Second, the representative experiment selected by Michaud et al was from a set of 3 experiments which included 5 mice each where they observed only a 3 fold increase in titers from lungs of TLR2-/- mice.

128

4. Discussion

On day 6 post intranasal infection, lungs from UNC93B-/- , TLR9-/- and 3d mice showed slightly elevated titers when compared to their respective WT control mice, however, the difference was not statistically significant (Figure 24B). We did not observe a statistically significant difference in titers among mice deficient in TLR2, 7, 7/9 or 13 mice when compared to their respective WT controls (Figure 24B). Our data is in concurrence with Guggemoos et al., who observed no difference in viral titers in the lungs of TLR9-/- mice 6 days after intranasal infection with 5 x 104 PFU of MHV68 (Guggemoos et al., 2007). This observation might be attributed to lower levels of TLR9 being expressed on DC in lungs when compared to the spleen for example (Chen et al., 2006). Furthermore, the absence of TLR dependence in the detection of MHV68 after intranasal infection might be attributed to the fact that alveolar macrophages are among the initial innate immune cells to encounter inhaled viruses. Upon infection with respiratory syncytial virus, macrophages are known to become highly phagocytic and produce large amounts of cytokines to induce the adaptive immune system (Becker et al., 1991). Sun and colleagues observed elevated levels of MHV68 DNA in the lungs of STINGgt/gt mice 3 days post intranasal infection with 5 x 104 PFU of MHV68 (Sun et al., 2015). There was no increase in the expression of IFN-β mRNA in the lungs of infected mice, but a STING dependent elevation in CXCL10 mRNA levels. The increase in CXCL10 levels correlated with recruitment of macrophages to the lungs through a STING dependent process (Sun et al., 2015). If aveolar macrophages are involved in the intranasal infection with MHV68, it is possible that these macrophages use cGAS/STING as opposed to TLR as the primary detection pathway for MHV68.

Viral titers in the spleens were not detectable on day 3 (Figure 26A). Virus was detected in the spleen 6 and 10 days after infection (Figure 26). TLR deficient mice did not show significant differences in titers when compared to their respective WT control mice (Figure 26B and C) with the exception of TLR13-/- mice which showed slightly elevated titers in the spleen on days 6 and 10 post infection when compared to titers from B6N mice (Figure 26C). A study involving quantitative real-time RT-PCR revealed that TLR13 is expressed strongly only in the spleen (Shi et al., 2011). This finding might suggest a possible role for TLR13 in the detection of MHV68 in the spleen 6-10 days post intranasal infection, however further experiments including more mice are needed to elucidate the role of TLR13 in the spleen on specific days post i.n. infection. Similar to titers from spleens of TLR13-/- mice, spleens from TLR9-/- mice also showed slightly elevated titers when compared to spleens from B6J mice on 10 days after i.n. infection. Surprisingly, the same was not observed in mice that were deficient of both TLR7 and TLR9 which showed comparable titers to WT B6J and TLR7-/- mice (Figure 26C). One probable reason for this difference might be attributed to the number of mice used for each genotype.

129

4. Discussion

Additional experiments must be performed to postulate a possible role of other endosomal TLR in the spleen.

To our knowledge, this is the first study to examine the role of TLR in the detection of MHV68 in the salivary gland post intranasal infection with MHV68. Though there was no detectable viral titer in the salivary gland on days 3 and 17 after intranasal infection, there was detectable virus 6 and 10 days post infection (Figure 25) On day 6, TLR7-/- showed significantly lower viral titers when compared to B6J, and lower titers were also observed in salivary glands from 3d mice, though the difference was not statistically significant (Figure 25B). Though there was high variation among the mice, on an average the titers from TLR9-/- and TLR7/9-/- were comparable to B6J mice. However, by day 10, TLR9-/- and TLR7/9-/- mice showed a statistically significant increase in viral titers when compared to WT B6J mice (Figure 25C).This suggests a possible role for TLR9 in the detection of MHV68 in the salivary gland 10 days post intranasal infection with MHV68.

MHV68 induced splenomegaly after intranasal infection

On day 17 after intranasal infection with 2 x 106 PFU of MHV68, we observed a decrease in splenomegaly in UNC93B-/-, TLR7-/- and TLR7/9-/- mice (Figure 27). Previous studies have shown that splenomegaly induced by MHV68 involves the activation of the T cell compartment which leads to the proliferation of lymphocytes (Ehtisham et al., 1993; Usherwood et al., 1996). After intranasal infection of 3-4 week old female BALB/c with 4 x 105 PFU of MHV68, Ehtisham and colleagues observed reduced splenomegaly in mice depleted of the CD4 T cell population when compared to non-depleted mice (Ehtisham et al., 1993). A similar finding was reported by Usherwood and colleagues who intraperitoneally infected BALB/c with 5 x 106 PFU of MHV68, and treated them with anti-CD4 depleting antibody, observed a reduction in infective centers in the spleen 10 days after infection (Usherwood et al., 1996). Though depletion of CD8 T cells did not affect splenomegaly, an increase in splenic B cells and CD8 T cells to over 450% above normal spleen levels was associated with the increase in spleen size (Usherwood et al., 1996). Furthermore, the presence of IFNα facilitates the maturation of DC to generate effective antigen presenting cells (APC) which have an increased capacity to stimulate CD4 and CD8 T cells (Palucka et al., 2002; Dalod et al 2003; Blanco et al., 2001). Reduced splenomegaly in UNC93B-/- and TLR7/9-/- mice is in agreement with absence of IFNα production by UNC93B-/- and TLR7/9-/- pDC upon infection with MHV68 (Figure 21). This observation does not hold true for TLR7, as TLR7-/- pDC produce IFNα upon MHV68 infection (Figure 21), and does not explain why there is no reduction in spleen weights of TLR9-/- mice (Figure 27). Therefore, the possible

130

4. Discussion

role of TLR7 and other endosomal TLR in the stimulation of CD4 and CD8 T cells after infection with MHV68 and the contribution of IFNα to this process needs to be examined directly in future studies. Cell type specific histopathological studies of the spleen or flow cytometry analysis of splenocytes from MHV68 infected spleens of WT and TLR deficient mice could be performed at multiple time points post infection to analyze change in cell populations, and the contribution of TLR to this process.

B6.Sle1yaa are C57BL/6J-congenic animals with systemic lupus erythmatosus susceptibility (Sle) 1 locus and Yaa-containing Y chromosome (Yaa). It is noteworthy that up-regulation of TLR7 in these B6.Sle1yaa male mice causes splenomegaly (Fairhurst et al., 2008). This phenomenon is reversed in B6.Sle1YaaTLR7- mice, where TLR7 is absent (Fairhurst et al., 2008). In an initial experiment where we compared spleen weights of uninfected mice, we did not observe a difference in spleen weights between uninfected WT B6J and uninfected TLR7-/- mice. However, as the result is only from an initial further experiments must be performed to confirm that there is no difference between uninfected WT and TLR7-/- mice. As there is a higher reduction in spleen weights of UNC93B-/- and TLR7/9-/- mice when compared to spleen weights from TLR7-/- mice, it is evident that absence of TLR7 is not the only cause for lower spleen weights.

4.4.2 Intraperitoneal infection

Clinical signs of illness

WT B6J, TLR2-/- and 3d mice infected intraperitoneally with 2 x 106 PFU of MHV68 did not lose weight and showed no signs of illness (Figure 28). On the contrary, all mice gained weight during the course of the infection. This observation might correlate with the route of infection, and the cell types involved in the initial control and recognition of the virus.

Organ titers

In order to better understand the role of TLR in the detection of MHV68 after intraperitoneal infection, we infected WT and TLR deficient mice with 2 x 106 PFU of MHV68, and examined viral titers in the salivary gland, spleen, lung and liver 3 and 6 days post infection. On days 3 and 6 post i.p. infection viral titers in the salivary gland were mostly below the detection limit (Figure 29A and 30A). The absence of MHV68 in the salivary gland early after intraperitoneal infection is in concurrence with observations by Hwang and colleagues who infected BALB/c intraperitoneally with 500 PFU of MHV68, and did not note the presence of MHV68 until day 8 post infection (Hwang et al., 2008). Therefore, it might be beneficial to assess viral titers at time

131

4. Discussion

points later than 6 days after i.p. infection. On day 3 post infection MHV68 titers in the lung, spleen and liver were increased only in UNC93B-/- mice (Figure 29B, C and D). This finding suggests that UNC93B dependent TLR act synergistically in the detection and control of MHV68 in the lung, spleen and liver after i.p. infection. The synergistic role of endosomal TLR after i.p. infection is further exemplified in viral titers in the spleen, where spleens from TLR7/9-/- and 3d mice show slightly elevated, but not statistically significant titers in comparison to those from TLR7-/- and TLR9-/- mice. These results should be further elucidated with additional experiments that include more mice. Interestingly, we observed a significant reduction of MHV68 titers in the livers of TLR7-/- mice (Figure 29D), similar to reduced titers in the salivary gland on day 6 after i.n. infection (Figure 25B). Further experiments are needed to address the organ specific role of TLR7 after intraperitoneal infection with MHV68.

Intriguingly, by day 6, MHV68 replication in lung, spleen and liver was not UNC93B dependent (Figure 30B, C and D). A potential reason for the lack of TLR dependency in our study might be attributed to the involvement of other DNA sensors such as cGAS. A recent study demonstrated a modest increase in MHV68 viral load in the lungs and spleen of cGAS deficient 8-9 week old B6J mice 7 days after i.p. infection with 106 PFU of MHV68 (Schoggins et al., 2014). Contrary to our observation of the lack of TLR9 dependency in the spleen after i.p. infection, Guggemoos and colleagues observed a significant increase in viral titers in the spleens of TLR9-/- mice after i.p. infection with 5 x 105 PFU of MHV68 (Guggemoos et al., 2007). It is important to note that Guggemoos and colleagues observe a large variation in titers from spleens (n = 9) from three independent experiments. However, it is possible that the 8-10 week old mice used in our study were older than mice used by Guggemoos and colleagues who do not specify the age or gender of the mice used in their experiments.

In the liver we observed higher viral titers from TLR13-/-, TLR9-/- and TLR7/9-/- mice (Figure 30D). 3d mice (n = 3) lacking functional TLR7, TLR9 and TLR13 signaling also displayed higher titers in the liver, but the difference was not statistically significant. However, as livers from TLR7/9-/- and 3d mice displayed similar titers which were not higher than titers observed in livers from TLR9-/- mice, the difference might be attributed solely to the absence of TLR9.

4.4.3 Intravenous infection

Clinical signs of illness

Contrary to prior observations of no clinical illness post intravenous infection with 4 x 105 PFU of MHV68 (Sunil-Chandra et al., 1992), we observed weight loss in mice infected with 2 x 106

132

4. Discussion

PFU of MHV68 between days 5 - 9 after infection (Figure 31). This difference in our observation might be attributed to the use of a higher viral inoculum. Interestingly, we observed a difference in recovery after weight loss between WT B6N and WT B6J mice (Figure 31). The difference between these mice might be attributed to the difference in IFNγ producing NK cells. Simon and colleagues established that a larger fraction of NK cells were activated by IL-12 in B6J mice when compared to B6N mice and B6J mice displayed a higher percentage of IFNγ positive cells (Simon et al., 2013). Endogenous IFNγ has been shown to be instrumental in increased body weight loss which is accompanied by hypoglycemia (Matthys et al., 1995). Additionally, NK cell depletion has been shown to alleviate severe weight loss caused by immunopathological T cell reponse to a virus infection (Waggoner et al., 2014). Therefore, it is likely that infection of WT B6J mice with MHV68 leads to an increase in IFNγ positive cells and lead to increased weight loss when compared to WT B6N mice.

We observed a statistically significant difference in weight loss between UNC93B-/- and TLR13-/- mice compared to WT B6N mice (Figure 32). This finding might indicate the importance of TLR13 in the detection of MHV68 infection, but as we did not observe an increase in organ titers, or differences regarding the establishment of latency or reactivation, further experiments will have to be performed to confirm the possible contribution to TLR13.

Contrary to the increased weight loss observed with UNC93B-/- and TLR13-/- mice, TLR2-/- and TLR9-/- mice exhibit lesser weight loss when compared to WT B6J mice (Figure 33B and D). Interestingly, we did not observe a significant difference in weight loss between WT B6J mice when compared to TLR7-/- and TLR7/9-/- mice (Figure 33C). This finding might be attributed to IFNγ production by NK cells. Martinez et al showed that murine NK cells express TLR2, 4, 8 and 9, but do not express TLR3 and TLR7 (Martinez et al., 2012). Therefore, the absence of TLR2 or TLR9 on NK cells might lead to reduced IFNγ production, in turn leading to reduced weight loss when compared to WT B6J mice. Additionally, the absence of type I IFN signaling after MHV68 infection has been shown to lead to an increase in short-lived CD8 T cells with a reduced TNFα, IFNγ and IL-2 production (Jennings et al., 2014). As we observe reduced IFNα levels from TLR9-/- pDC (Figure 21), it might explain the reduction in weight loss. This however does not explain the decreased weight loss observed with TLR2-/- mice and the slightly increased weight loss observed in UNC93B-/- and TLR7/9-/- mice.

In summary, as we observe a significant difference among mice based on the genetic background, we cannot hypothesize about the contribution of individual endosomal TLR 7 and 9 for the weight loss observed in UNC93B-/- mice. Additional experiments with 3d mice which are

133

4. Discussion

on the B6J background must be performed to elucidate which of the endosomal TLR apart from TLR13 contribute to the observed weight loss. Intriguingly, on day 3 we observed significantly difference in body weights of B6J mice when compared to TLR2-/- and TLR7/9-/- mice (Figure 33B and C). This variation might be attributed to difference in inflammatory cytokine levels. As our in vitro studies were limited to the analysis of IFNα production by pDC after MHV68 infection, it would be useful to obtain a broader overview of IFN and cytokine levels in vivo during acute infection. Additional experiments measuring IFNα, TNFα and IL-6 levels in plasma and supernatant from organ homogenates during acute replication of MHV68 after i.n., i.p. and i.v. infection could be performed to understand how differences in IFN and cytokine production correlates to observed weight loss.

Organ titers

On day 3 after intravenous infection with 2 x 106 PFU of MHV68, we observed viral titers in salivary gland, lung, spleen and liver (Figure 34). A similar observation was made by Sunil- Chandra and colleagues upon intravenous infection of BALB/c mice with 4 x 105 PFU of MHV68, where they observed the presence of MHV68 in the lung, spleen, thymus, heart and liver (Sunil-Chandra et al., 1992).

Though there was no increased titer observed in salivary glands from UNC93B-/- mice, salivary glands from TLR7/9-/- and 3d mice displayed elevated titers (Figure 34A). The difference in titers might be attributed to differences between B6N and B6J mice. However, this finding did not hold true for the lungs, where the increase in viral titers from UNC93B-/- and 3d mice was consistent irrespective of the genetic background of the mice (Figure 34B). The spleen displayed a clear TLR9 dependency for the control of MHV68 (Figure 34C). It is possible that the difference between TLR9 dependency in the lung and spleen is partly due to the physiological absence of TLR9 in the lungs. Chen and colleagues reported that cDC and pDC in the spleen express TLR9, but that there is no detectable expression of TLR9 on either of the subsets in the lung (Chen et al., 2006). From a functional hypothesis’ stand point, the only drawback is that during our preliminary experiment we did not observe a significant difference in splenic viral titers between WT and UNC93B-/- mice (n = 9) after perfusion (data not shown). As no experimental parameters apart from perfusion were altered between the two i.v. infection experiments, it is probable that the observed dependency on endosomal TLR after intravenous infection of MHV68 is not particularly organ specific and rather dependent on the presence of viremia. This possibility however needs to be examined directly by investigating differences in organ titers of perfused control and TLR deficient mice 3 days post intravenous infection.

134

4. Discussion

Livers from UNC93B-/- mice displayed significantly increased titers when compared to WT B6J controls, while TLR7/9-/- and 3d mice displayed slightly higher viral titers, but the difference was not significant (Figure 34D). These findings suggest a possible synergistic role of endosomal TLR in the detection of MHV68 in the liver. Replication of MHV68 in the liver early after i.v. infection might reflect viral replication within the vasculature without being specific to the organ. The same can also be true for virus production in the lungs after i.v. infection as it is vastly different in its timing from lung infection after i.n. infection with 2 x 106 PFU of MHV68.

Role of TLR in the detection of MHV68 in the brain 3 days after infection

Multiple studies since the initial isolation of MHV68 have addressed possible neuro- invasiveness of the virus and analyzed its interactions with the central nervous system (CNS) (Chan et al., 1999; Chan et al., 2000; Weinberg et al., 2005; Karatas et al., 2008). Furthermore, UNC93B and type I IFN have been shown to be important in the detection and control of herpes simplex virus type 1 (HSV-1) in the brain (Wang et al., 2012). Therefore, we looked at the role of TLR in the detection of MHV68 in the brain.

Three days after intravenous infection with 2 x 106 PFU of MHV68, we did not observe a difference in viral titers in the brains of 8-10 week old UNC93B-/- mice when compared to titers from WT B6N mice after perfusion (Figure 35). The absence of viral titers in the brain 3 days after i.v. infection might be attributed to multiple factors.

First, age of the mice and time point of analysis might influence MHV68 titers in the brain. BALB/c mice infected intranasally with 2 x 104 PFU of MHV68 at 1, 2, 3 and 4 weeks of age showed a direct correlation between their age and the onset of clinical sings of infection (Terry et al., 2000). In concurrence with data by Terry and colleagues, intranasal inoculation of newborn (24 hrs old) BALB/c mice with 2 x 105 PFU of MHV68 was seen to cause inflammation and cerebral infection in 50% of the animals (Haüsler et al., 2005). The patterns of inflammation including hydrocephaly, meningitis, cerebellitis and encephalitis were similar to those known from infection with EBV. Ten days post infection, cerebral viral loads of up to Log4 was reached and lowered to Log2 by day 40 post infection (Haüsler et al., 2005). Furthermore, a study of 4-5 week old 126, CBA and BALB/c mice intranasally infected with 2 x 104 PFU of MHV68 revealed presence of viral nucleic acids in histology samples of perfused brains 3 weeks post infection (Terry et al., 2000). Therefore, it might be necessary to analyze brain samples at later time points after infection. Second, the route of infection might contribute to the control of MHV68 infection. Upon intracerebral infection with 2 x 104 PFU of MHV68, 1, 2, 3 and 4 week old BALB/c succumb to viral infection and became moribund and do not show an age dependency

135

4. Discussion

(Terry et al., 2000). Additionally, upon intracerebroventricular infection of 8-9 week old BALB/c mice, a survival rate of approximately 80% is observed (Kang et al., 2011; Cho et al., 2009).

Taken together, this finding indicates that the neuroinvasiness of MHV68 is dependent on the route of infection and the age of the mice at inoculation. Further experiments including histopathology studies and viral titers will be required to elucidate the role of UNC93B and endosomal TLR in the detection of MHV68 in the brain.

MHV68 induced splenomegaly after intravenous infection

Since it was evident that UNC93B−/− mice had higher viral titers, and showed higher weight loss than WT B6N mice, we anticipated that these mice deficient of endosomal TLR would have higher virus-induced pathology. Surprisingly, similar to reduced splenomegaly after intranasal infection, we found that UNC93B−/− mice actually had decreased splenomegaly after intravenous infection with 2 x 106 PFU of MHV68. However, the same was not observed for TLR7-/- or TLR7/9-/- mice on a B6J background (Figure 36). Future experiments including 3d mice which are on a B6J genetic background will be helpful to further elucidate the synergistic role of endosomal TLR apart from TLR7 and 9 in mice on a B6J genetic background.

Therefore, it is evident that the development of a viral infection in mice and its manifestation as clinical symptoms depends on the route of inoculation, and the spread of the virus from the initial site of replication. The differences in MHV68 replication based on route of infection are summarized in Table 25. Additional experiments must be performed to confirm some of our findings, to further elucidate mechanisms and address other unanswered questions.

Table 25: Summary of route dependent replication of MHV68

136

4. Discussion

Days after infection 0 3 4-5 6 7-9 10 17-22 Intranasal Slightly elevated No significant WT and TLR WT and TLR WT and TLR Spleens from 2 x 106 PFU of titers from TLR7-/- difference in deficient mice deficient mice deficient mice UNC93B-/-, MHV68 WUMS and 3d lungs. No weight loss of start losing show a recover from TLR7-/- and (30µl) detectable titers in TLR deficient weight reduction in weight loss TLR7/9-/- mice the salivary gland mice or WT Salivary glands weight MHV68 replication display reduced and spleen control mice from TLR7-/- in the salivary splenomegaly mice show gland is dependent 17 dpi reduced viral on TLR9 titers Intraperitoneal Lung, spleen and Mice steadily Livers from Mice steadily Mice steadily gain Mice steadily 2 x 106 PFU of liver from UNC93B- gain weight over TLR13-/-, TLR9-/- gain weight over weight over the gain weight until MHV68 WUMS /- mice show the duration of and TLR7/9-/- the duration of duration of the termination of (200µl) enhanced viral the experiment show elevated the experiment experiment experiment on titers, while liver titers day 18 from TLR7-/- mice show a reduction in titer Intravenous B6J mice show Mice on B6N WT B6N mice, WT B6N mice Mice on B6J Spleens from 2 x 106 PFU of increase in weight and B6J TLR2-/- and recover from background do not UNC93B-/- mice MHV68 WUMS when compared to background lose TLR9-/- mice weight loss and completely recover display reduced (200µl) TLR deficient mice. weight starting start to recover return to initial from weight loss splenomegaly Salivary gland and on day 5 from weight weight by day 9. 22 dpi lung from 3d mice loss. UNC93B-/-, Latency and show elevated viral TLR13-/- and WT reactivation of titers, lungs and B6J, TLR7-/- and MHV68 is liver from UNC93B- TLR7/9-/- mice summarized in /- mice show show increased Table 24 enhanced viral weight loss on replication. Acute day 7 and begin replication of to recover on MHV68 in the day 8 spleen is TLR9 dependent

137

4. Discussion

4.5 Role of TLR and gender for establishment of MHV68 latency

The immune system exhibits gender dimorphism which results from the direct effects of sex hormones such as estrogen and androgens (reviewed by Ngo et al., 2014). Multiple studies describe gender dimorphism in viral infection and are summarized in Table 26.

Table 26: Gender induced differences in mortality following viral infections Greater Virus (strain) Route Inoculum dose Mouse strain Reference mortality HSV-1 (strain 17) i.p. 3.16 x 107 PFU B6 Females Geurs et al., 2012

HSV-1 (strain F) c.s. 105-107 PFU 129 sv/EV Males Han et al., 2001

HSV-1 (strain 17) c.s. 10 x LD50 129 x (B6 x 129)F1 Males Lundberg et al., 2003

MCMV i.p. 105-107 PFU B6 Females Geurs et al., 2012

VV i.p. 106-108 PFU B6 Females Geurs et al., 2012

2 Influenza A i.n. 10 TCID50 B6 Females Robinson et al., 2011

Coxsackievirus i.p. 100 PFU B6 Males Robinson et al., 2011 i.p. – intrapertoneal infection; i.n. – intranasal infection; c.s. – corneal scarification; HSV – Herpes simplex virus; MCMV – Murine cytomegalovirus; VV – Vaccinia virus.

One common factor between the different viruses listed in Table 26 is the involvement of UNC93B and endosomal TLR in the detection and control of infection. Therefore, it becomes necessary to clarify the possible gender specific role of UNC93B and individual TLR in the establishment of MHV68 latency.

We observed a reduction in the establishment of MHV68 latency in splenocytes deficient of endosomal TLR. Female UNC93B-/- and TLR7-/- mice displayed a lower frequency of latently infected cells, and while no difference in frequency of latently infected cells from TLR9-/- female mice was observed, female TLR7/9-/- mice showed a slightly reduced frequency when compared to cells from WT controls (Table 24). On the contrary, male UNC93B-/- and TLR7-/- mice displayed similar frequencies of latently infected splenocytes when compared to their respective WT controls, and cells from male TLR9-/- and TLR7/9-/- mice showed lower frequencies of MHV68 ORF50 positive cells when compared to WT controls (Figure 38 and Table 24).

As female TLR9-/- mice did not show a reduction in the frequency of splenocytes harboring MHV68 (Table 24), absence of functional TLR7 signaling in UNC93B-/- female mice might contribute majorly to the observed effect. The reduction in the frequency of latently infected

138

4. Discussion

splenocytes in the absence of TLR7 in female mice is in agreement with a previous study which showed that in vivo stimulation of TLR7 with R848 after intraperitoneal infection with 1 x 105 PFU of MHV68 increased the frequency of infected splenocytes by 7 fold when compared to mock-treated samples (Haas et al., 2014). The difference in frequencies between female and male mice deficient of TLR7 might be attributed to a gender dimorphism as TLR7 stimulation has been shown to induce higher IFNα production in females (Berghofer et al., 2006), and more specifically, pDC from women induce higher IFNα production upon stimulation with HIV- encoded TLR7 ligands when compared to pDC from men (Meier et al., 2009). It is probable that in the absence of TLR7, pDC from female mice produce lesser amounts of IFNα. As IFNα is involved in positive feedback signaling with the estrogen receptor and UNC93B (Panchanathan et al., 2013), the absence of IFNα might result in standard levels of estrogen and UNC93B in cells as opposed to an upregulation in the presence of excess IFNα. Low levels of UNC93B also lead to reduced activation of MyD88 and NFκB dependent pathways which might further facilitate the reduction in the frequency of latently infected splenocytes.

As absence of TLR7 in splenocytes from male mice does not lead to a reduction in the frequency of latently infected cells (Table 24), the effect seen in TLR7/9-/- male mice must be due to the absence of TLR9. S11 cells which are latently infected with MHV68 show basal levels of NFκB, and downregulation of TLR9 but not TLR7 by shRNA shows a reduction in NFkB p65 levels (Haas et al., 2014). This finding might imply that the activation of TLR9 and not TLR7 leads to the translocation of NFκB p65 to the nucleus and hence contributes to basal levels of NFκB in latently infected cells. Thus the reduced basal levels of NFκB in TLR9-/- splenocytes might contribute to their ability to harbor a latent infection. The dissimilarity in frequencies between female and male mice deficient of TLR9 might be due to a gender disparity. For instance, following stimulation with ODN 2006, peripheral blood mononuclear cells (PBMC) from males induce higher levels of IL-10 in comparison to PBMC from females of reproductive age and post-menopausal females. This difference in IL-10 levels is not observed between PBMC from male and females after stimulation with the TLR7 ligand Imiquimod (Torcia 2012). Therefore, absence of TLR9 in splenocytes from male mice might result in reduced levels of IL- 10 production, leading to an increased activation of MHV68 specific cytotoxic T cells, and consequently a reduced frequency of latently infected cells.

Our findings of the involvement of endosomal TLR7 and 9 in the establishment of MHV68 latency is consistent with a study where i.n. infection of MyD88 deficient mice with 103 PFU of MHV68 resulted in a decrease in the frequency of B cells harboring MHV68 (Gargano et al.,

139

4. Discussion

2008). However, our data is not in line with observations by Haas and colleagues who reported that 34 days post intraperitoneal infection with 1 x 105 PFU of MHV68, the absence of TLR7 or TLR9 did not have an impact on the establishment of latency in B cells (Haas et al., 2014). Apart from differences in experimental parameters including the route of infection, the dose of the virus, and days after infection, it is important to note that we did not observe a consistent decrease in latently infected cells between the females and male mice. Therefore, it is possible that pooling spleens from mixed gender, or using only one particular gender of mice for experiments might confound the analysis.

4.6 Influence of gender and TLR on reactivation of MHV68

As we wanted to clarify the possible gender specific role of individual TLR in MHV68 reactivation, mice were separated based on gender for this assay. Absence of TLR13 in female and male mice did not lead to an increase in reactivation of MHV68 from splenocytes (Figure 39B). Surprisingly, male TLR2-/- mice showed a slightly reduced rate of reactivation when compared to WT controls (Figure 40A). This was not true in the case of female TLR2-/- mice (Figure 40A). Additional reactivations assays supplemented with ORF50 limiting dilution PCR must be performed to confirm the reduction of MHV68 reactivation from male TLR2-/- mice.

UNC93B-/- mice displayed a higher frequency of reactivating splenocytes with male UNC93B-/- mice showing slightly elevated frequencies when compared to female UNC93B-/- mice (Figure 39A). Similar to the enhanced reactivation observed in UNC93B deficient splenocytes, TLR9-/- splenocytes from female and male mice also showed an enhancement in reactivation (Figure 40C). This data is in agreement with observations by Guggemoos and colleagues who showed an increase in reactivation from splenocytes from mice infected intraperitoneally with 5 x 105 PFU of MHV68 (Guggemoos et al., 2007).

Gargano and colleagues reported that in mice intranasally infected with 1000 PFU of MHV68 the decrease in frequency of viral genome-positive cells is reflected by a decrease in reactivation of MHV68 from MyD88-/- splenocytes (Gargano et al., 2008). Contrary to those observations, we observed an enhanced reactivation from splenocytes deficient of endosomal TLR7 (only females) and 9 (both males and females) though they displayed a lower frequency of latently infected cells. This finding exemplifies the importance of UNC93B and TLR7 and 9 in reactivation of MHV68.

TLR7-/- and TLR7/9-/- splenocytes from female mice and not male mice showed an increase in reactivation (Figure 40B and D). This difference between the two genders might be attributed to

140

4. Discussion

the presence of estrogen. Clinical research suggests a probable role of estrogen as a risk factor for the reactivation of herpesviruses like HSV from latency (Miguel et al., 2010). Additionally, reactivation of MHV68 has been shown to be controlled by a ‘two-signal’ mechanism where induction of IL-4 and inhibition of TH1 responses by blocking IFNγ induces reactivation of

MHV68 (Reese et al., 2014). The presence of estrogen can change the balance between TH1 and TH2 cells by reducing TH1 cells and increasing the presence of TH2 cells. On the contrary, androgens decrease levels of both TH1 and TH2 cells, and hence do not alter the balance between the two (reviewed by Ngo et al., 2014). Therefore, it is possible that the presence of estrogen in female mice reduces TH1 cells, and increases the rate of reactivation.

In summary, during the course of this study we could unequivocally demonstrate that in vitro, UNC93B and more specifically endosomal TLR7 and 9 are involved in the detection of MHV68 in pDC. Moreover, using age and gender matched WT B6 and TLR deficient mice we demonstrated that TLR dependency during acute replication of MHV68 is dependent on the route of infection and is organ specific. We utilized intravenous infection to elucidate the role of TLR in the establishment of latency and reactivation of MHV68, and could demonstrate a gender dimorphism in the immune response to MHV68 in TLR deficient mice. In the absence of UNC93B and UNC93B dependent TLR7 and 9, there was a decrease in the frequency of latently infected splenocytes and an increase in MHV68 reactivation.

141

Abstract

Abstract Herpesviruses are double-stranded DNA viruses that display strict host specificity. A hallmark of herpesviruses is their ability to establish chronic, lifelong infections. After acute infection, herpesviruses enter a dormant state known as latency, during which very few viral genes are transcribed. Latency can be intermitted by reactivation of the virus, resulting in the transcription of most viral genes and production of novel viral progeny. While primary infections with the human gammaherpesviruses Kaposi’s sarcoma-associated herpesvirus (KSHV) and Epstein– Barr virus (EBV) are usually asymptomatic, tumors and lymphomas may manifest in individuals. As herpesviruses are species specific, the gammaherpesvirus murine herpesvirus 68 (MHV68) is currently used and widely accepted for in vivo studies in mice. However, Mus musculus, which is widely used in the scientific community, is not the natural host of MHV68. Instead, the natural hosts of MHV68 are the wood mouse (Apodemus sylvaticus) or the bank vole (Myodes glareolus), which have separated millions of years ago from Mus musculus.

We aimed to study gammaherpesvirus infection in the natural host Mus musculus. For these studies, a Mus musculus colony captured in the wild and harboring the novel murine gammaherpesvirus Mus musculus rhadinovirus 1 (MmusRHV1), the novel betaherpesvirus Mus musculus cytomegalovirus 2 (MCMV2) and the well characterized MCMV1, was made available to us. The presence of these three herpesviruses was confirmed by polymerase chain reaction. Next, a non-invasive blood PCR screen was established to analyze the presence of MmusRHV1 and thereafter its transmission from naturally infected animals to co-housed naïve animals. The experiments revealed evidence of transmission of MmusRHV1 from positive parents to offspring before the age of day 21. In vitro, it proved difficult to isolate MmusRHV1 and MCMV2 due to the more robust growth of MCMV1 in cell culture. Furthermore, while MmusRHV1 was detected early in our study, it was absent from later samples. We observed that the treatment of pre- existing parasitic co-infections of the mice housed at the HZI animal facility with ivermectin coincided with the subsequent failure to detect MmusRHV1. We speculate that parasitic infections may contribute to lytic MmusRHV1 replication and viral shedding.

Toll-like receptors (TLR) are an integral part of the innate immune system. We used TLR2-/-, TLR7-/-, TLR9-/- and TLR13-/- mice to analyze the role of individual TLR in the context of MHV68 infection. In addition, we addressed the synergistic role of endosomal TLR by using UNC93B-/- and 3d mice, which both lack functional TLR3, 7, 9, 11, 12 and 13, as well as mice lacking both TLR7 and TLR9. First, we analyzed the type I interferon (IFN) response of dendritic cell (DC) subsets upon MHV68 infection in vitro. Next, we assessed the control of lytic MHV68 infection

142

Abstract

after different routes of infection. In addition, we analyzed the establishment of latency and reactivation after intravenous infection.

Our in vitro results clearly show that plasmacytoid DC (pDC), but not conventional DC, are the primary type I IFN producers upon MHV68 infection in vitro. The type I IFN response upon MHV68 infection was significantly reduced in pDC from TLR9 deficient mice, and completely abolished in mice lacking both TLR7 and TLR9. As TLR7 has not been implicated in MHV68 detection yet, this study reveals unequivocally that in addition to TLR9, TLR7 is involved in the detection of MHV68 in pDC.

In vivo, the route of MHV68 infection strongly influenced weight loss and viral titers in various organs of mice. Lytic viral titers after intranasal infection with MHV68 were similar across all genotypes studied, and no genotype-dependent effects were observed. In contrast, upon intraperitoneal infection, we observed enhanced MHV68 replication in the lung, spleen and livers of UNC93B deficient mice. Infection via the intravenous route also resulted in enhanced viral titers in UNC93B deficient mice. Acute replication of MHV68 in the spleen upon intravenous infection was dependent on endosomal TLR9. Twenty-two days post intravenous infection the gender of the mice strongly influenced the establishment of latency and reactivation. Furthermore, MHV68 reactivation from spleens of UNC93B deficient animals and more specifically from those deficient in endosomal TLR7 and 9 was enhanced. Our studies highlight the importance of multiple factors such as endosomal TLR, the route of infection and gender of the mice in control of MHV68 in vivo.

143

Abstract

Herpesviren sind strikt wirtsspezifische Viren mit einem Genom aus doppelsträngiger DNA. Ein Merkmal der Herpesviren ist ihre Eigenschaft, chronische und lebenslange Infektionen zu verursachen. Nach der akuten Infektion etablieren Herpesviren den sogenannten Zustand der Latenz. Während dieser Phase werden nur sehr wenige virale Transkripte exprimiert. Die Latenz kann durch Phasen der Reaktivierung, in der die meisten viralen Gene transkribiert und neue Virusnachkommen gebildet werden, unterbrochen werden. Während die Primärinfektion mit den humanen Gammaherpesviren Kaposi-Sarkom-assoziiertes Herpesvirus (KSHV) und Epstein–Barr-Virus (EBV) in den meisten Fällen asymptomatisch verläuft, können diese Viren langzeitig Tumore verursachen. Da Herpesviren strikt wirtsspezifisch sind wird das murine Herpesvirus 68 (MHV68), das große Homologien zu EBV und KSHV aufweist, als gängiges Modell für die Studien von Gammaherpesviren in Mausmodellen verwendet. Jedoch ist der natürliche Wirt von MHV68 nicht die weitläufig für Forschungen verwendete Hausmaus, Mus musculus, sondern entfernte Verwandte wie die Waldmaus (Apodemus sylvaticus) oder die Rötelmaus (Myodes glareolus). Diese sind durch mehrere Millionen Jahre der Evolution von Mus musculus getrennt. Ein Ziel dieser Arbeit war, gammaherpesvirale Infektionen im natürlichen Wirt Mus musculus zu untersuchen. Dazu wurde uns eine Mus musculus Kolonie zur Verfügung gestellt, die in der Wildnis eingefangen und positiv für das neue Gammaherpesvirus Mus musculus Rhadinovirus (MmusRHV1) getestet wurde. Die verwendete Mus musculus Kolonie wurde ebenfalls positiv für das neue Betaherpesvirus Murines Cytomegalievirus 2 (MCMV2) und das bereits bekannte MCMV1 getestet. Das Vorhandensein dieser drei Viren in den Mäusen wurde in dieser Arbeit erfolgreich mit molekularbiologischen Methoden nachgewiesen. Im Rahmen dieser Arbeit wurde ein nicht invasiver auf der Polymerasekettenreaktion basierender Bluttest entwickelt, um die Transmission von MmusRHV1 auf nicht infizierte Mäuse zu untersuchen. Diese Experimente lieferten Anhaltspunkte dafür, dass MmusRHV1 von positiv getesteten Eltern im Laufe von 21 Tagen nach der Geburt auf ihre Nachkommen übertragen wurde. Die Isolation des MmusRHV1 und MCMV2 in der Zellkultur erwies sich als schwierig, und es gelangen lediglich der Nachweis und die Isolation des bereits bekannten, schnell replizierenden MCMV1. Hinzu kam, dass zu Beginn der Arbeiten MmusRHV1 in den Mäusen sehr gut nachgewiesen werden konnte, später jedoch nicht mehr nachweisbar war. Diese Beobachtung korrelierte zeitlich mit der notwendigen Parasiten Behandlung der Mäuse mit Ivermectin, die am HZI im Laufe unserer Studie durchgeführt wurde. Daher vermuten wir, dass Infektionen mit Parasiten eine wichtige

144

Abstract

Voraussetzung für die Replikation, Ausscheidung und Transmission von MmusRHV1 darstellen könnten. Toll-ähnliche Rezeptoren (TLR) sind eine essentielle Komponente des angeborenen Immunsystems. Um die Rolle der Toll-ähnlichen Rezeptoren 2, 7, 9 und 13 im Kontext der MHV68 Infektion zu untersuchen, wurden in vitro und in vivo Studien durchgeführt. Um eine mögliche synergistische Rolle der Toll-ähnlichen Rezeptoren im Rahmen der MHV68 Infektion aufzuzeigen, wurden UNC93B-defiziente Mäuse sowie Mäuse mit der H412R Punktmutation im UNC93B Protein, sogenannte 3d Mäuse, in die Studie eingeschlossen. In den UNC93B- defizienten sowie in den 3d Mäusen sind die endosomalen Toll-ähnlichen Rezeptoren 3, 7, 9, 11, 12 and 13 nicht funktionell. Zusätzlich wurden während dieser Arbeit Mäuse generiert, denen TLR7 und TLR9 (TLR7/9) fehlen. Zunächst wurde die durch MHV68 induzierte Typ I Interferon (IFN) Antwort von verschiedenen dendritischen Zellen in vitro analysiert. Danach wurde die Kontrolle der lytischen MHV68 Infektion nach intranasaler, intraperitonealer und intravenöser Infektion in vivo charakterisiert. Ferner wurden die Latenzetablierung und Reaktivierung nach intravenöser Infektion analysiert. Die durgeführten in vitro Studien zeigten, dass MHV68 in plasmazytoiden, nicht jedoch in konventionellen, dendritischen Zellen eine Typ I IFN Antwort induzierte. Die Typ I IFN Antwort war stark reduziert in plasmazytoiden dendritischen Zellen von TLR9 defizienten Mäusen, jedoch nur komplett abwesend in Zellen von TLR7/9 defizienten Mäusen. Da für TLR7 bislang keine Beteiligung bei der MHV68 Infektion gezeigt wurde, ist dies die erste Studie die gezeigt hat, dass TLR7 gemeinsam mit TLR9 eine wichtige Rolle in der Detektion von MHV68 in plasmazytoiden dendritischen Zellen hat. In vivo konnte gezeigt werden, dass die MHV68 Infektion abhängig von der Infektionsroute einen sehr unterschiedlichen Verlauf aufweist. Lytische virale MHV68 Titer nach intranasaler Infektion waren in allen untersuchten TLR- defizienten und Wildtyp Mäusen gleich. Nach intraperitonealer Infektion hingegen zeigten UNC93B-defiziente Mäuse erhöhte MHV68 Titer in Lunge, Milz und Leber. Nach intravenöser Infektion zeigten UNC93B-defiziente Mäuse ebenfalls erhöhte MHV68 Titer, wohingegen die akute lytische Replikation in der Milz TLR9-abhängig war. Zweiundzwanzig Tage nach intravenöser Infektion wurden geschlechtsspezifische Unterschiede hinsichtlich der Etablierung der Latenz und Reaktivierung beobachtet. Die Reaktivierung von MHV68 aus der Milz von UNC93B-defizienten und TLR7/9-defizienten Mäusen war erhöht. Zusammenfassend zeigt die MHV68 Studie, dass mehrere Faktoren wie Infektionsroute, Geschlecht und endosomale Toll-ähnliche Rezeptoren einen Einfluss auf die Kontrolle der MHV68 Infektion haben.

145

List of abbreviations

List of abbreviations µl milli liter µM micro molar 3d triple defect 8-OHdG 8-hydroxy-2-deoxyguanosine 8-OHG 8-hydroxyguanosine AIDS acquired immune deficiency syndrome AlHV-1 Alcelaphine herpesvirus 1 AP-2 adapter protein-2 APC antigen presenting cells APC allophycocyanin ATCC American type culture collection B2M beta-2-microglobulin B6J C57BL/6J B6N C57BL/6N BMDM bone marrow-derived macrophages BoHV-4 bovine herpesvirus 4 BSA bovine serum albumin C centigrade CalHV-3 Calitrichine herpesvirus 3 CD cluster of differentiation cDC conventional dendritic cells CDP common DC progenitors cGAS cyclic GMP-AMP synthase CLR c-type lectin receptors cMOP common monocyte precurors CMP common myeloid precursors CMV cytomegalovirus CNS central nervous system CPE cytopathic effect CpG 5'-c-phosphate-g-3' CTL cytotoxic t-lymphocyte CVB3 Coxsackievirus strain B serotype 3 DAMP danger associated molecular pattern DC dendritic cells dG deoxyguanosine DMEM dulbecco’s modified eagle medium DN double negative DNA deoxyribonucleic acid N-[1-(2,3-Dioleoyloxy)propyl]-N,N,N-trimethylammonium DOTAP methyl-sulfate DPOL DNA polymerase dsDNA double-stranded DNA dsRNA double-stranded RNA EBV Epstein-Barr virus EDTA ethylenediaminetetraacetic acid EGS external guide sequence EHV-2 Equine herpesvirus 2 ELISA enzyme-linked immunosorbent assay FACS fluroscence-activated cell sorting

146

List of abbreviations

FCS fetal calf serum FITC fluorescein isothiocyanate FLDC FMS-like tyrosine kinase 3 ligand (Flt3L) induced dendritic cells Flt3 FMS-like tyrosine kinase 3 Flt3L FMS-like tyrosine kinase 3 ligand G guanosine gB glycoprotein b GM-CSF granulocyte-macrophage colony-stimulating factor H2O dihydrogen monoxide HA hemagglutinin HBS HEPES buffered saline HCMV human cytomegalovirus HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HHV human herpesvirus HIV human immunodeficiency virus hpi hours post infection HRP horse radish peroxidase HSE herpes simplex virus-1 (hsv-1) encephalitis HSV-1 herpes simplex virus type 1 HSV-2 herpes simplex virus type 2 HVA herpesvirus ateles HVS herpesvirus saimiri HZI Helmholtz center for infection research i.n. intranasal i.p. intraperitoneal i.v. intravenous IFN type i interferon IFNAR interferon-α/β receptor IFNα interferon alpha IFNγ interferon gamma IFNγR interferon gamma receptor IKK IκB kinase IL interlukin IL-1R interleukin-1 receptor imd Immune deficiency IPC interferon-producing cells IRAK IL-1R-associated kinase IRF interferon regulatory factor IVC individually ventilated cages kb kilo basepair Kbp kilo base pair KCl potassium chloride KS Kaposi’s sarcoma KSHV Kaposi’s sarcoma-associated herpesvirus LD lethal dose Letr loss of endosomal tlr response LMP-1 latent membrane protein 1 LPS lipopolysaccharide LTA lipoteichoic acid MACS magnetic-activated cell sorting MALP-2 macrophage-activating lipoprotein 2

147

List of abbreviations

MCD Multicentric Castleman’s disease MCMV murine cytomegalovirus (MCMV1) MDBP major dna-binding protein MDP monocyte/dendritic cell precursors MEF mouse embryonic fibroblasts MHC major histocompatibility complex MHV mouse hepatitis virus MHV68 murid herpesvirus 4 type 68 MIE major immediate early min minute Mip3 macrophage inflammatory protein 3 miRNA micro RNA ml micro liter MLN mediastinal lymph nodes MMTV mouse mammary tumor virus MmusCMV- Mus musculus cytomegalovirus 2 2 (MCMV2) MmusRHV1 Mus musculus rhadinovirus 1 MOI multiplicity of infection MPLA monophosphoryl lipid a mPR protease of mcmv MuHV-4 murid herpesvirus 4 MyD88 myeloid differentiation primary response gene 88 NaCl sodium chloride NCBI national center for biotechnology information NDV Newcastle disease virus NEMO NF-kappa-b essential modulator NFkB nuclear factor kappa b ng nano gram NK natural killer NLR nod like receptors nm nano meter NOD nucleotide-binding oligomerization domain ORF open reading frame ORN oligoribonucleotides OsPA outer surface protein a Pam3CSK4 n-palmitoyl-s-dipalmitoylglyceryl cys-ser-(lys)4 PAMP pathogen associated molecular pattern PBMC peripheral blood mononuclear cells PBS phosphate buggered saline PCR polymerase chain reaction pDC plasmacytoid dendritic cells PE phycoerythrin PEL primary effusion lymphoma PFU plaque forming unit PLHV-1 porcine lymphotropic herpesvirus 1 Poly (I:C) polyinosonic-polycytidylic acid Poly (U) polyuridylic acid PRR pattern recognition receptors RCMV-E rat cytomegalovirus english isolate

148

List of abbreviations

RCMV-E rat cytomegalovirus English isolate RCMV-M rat cytomegalovirus Maastricht isolate RIG-I retinoic acid-inducible gene-i RNA ribonucleic acid RPMI Roswell park memorial institute medium rRNA ribosomal rna RRV rhesus rhadinovirus RSV respiratory syncytial virus RT room temperature RTA replication and transcription activator s second S.D standard deviation SDS sodium dodecyl sulfate SEA Schistosoma mansoni soluble egg antigen shRNA short hairpin rna Siglec-H sialic-acid-binding immunoglobulin-like lectin h SIV simian immunodeficiency virus Sle systemic lupus erythmatosus susceptibility SPF specific pathogen-free STAT signal transducer and activator of transcription TAB tak1 binding proteins TAE tris base, acetic acid and edta TAK transforming growth factor-β-activated kinase TCID50 tissue culture infective dose TH1 t helper cells type i TH2 t helper cells type 2 TIR toll-interleukin 1 receptor TIRAP tir domain-containing adapter protein TLR toll-like receptors Tm melting temperature TNF tumor necrosis factor TPA 12-o-tetradecanoylphorbol-13-acetate TRADD tnf receptor 1-associated protein death domain protein TRAF tnf receptor-associated factor TRAM trif related adaptor molecule TRIF tir-domain-containing adapter-inducing interferon-b USA United States Of America VIMM viral immune modulation VSV vesicular stomatitis virus VV vaccinia virus VZV varicella-zoster virus WT wild type

149

List of tables

List of tables Table 1: TLR expression pattern among various murine DC subsets ...... 12 Table 2: Localization, ligands and adapter proteins of TLR ...... 16 Table 3: Buffers and solutions ...... 29 Table 4: List of commercially available kits ...... 30 Table 5: List of oligonucleotides used for cloning inserts into pGEM-T vector ...... 32 Table 6: List of constructs used as positive controls for polymerase chain reactions ...... 32 Table 7: Oligonucleotides used for colony PCR and sequencing ...... 33 Table 8: List of oligonucleotides used for polymerase chain reaction ...... 36 Table 9: List of oligonucleotides used for limiting dilution nested PCR for the detection of MHV68 ORF50 ...... 38 Table 10: List of cell lines used in this study ...... 38 Table 11: Working volumes for cultivation of FLDC in tissue culture flasks ...... 39 Table 12: List of antibodies used for FACS and flow cytometry ...... 40 Table 13: List of stimulants used for MmusRHV1 reactivation ...... 42 Table 14: List of synthetic TLR ligands ...... 43 Table 15: List of stimuli and dilution factor of the supernatant used for ELISA ...... 43 Table 16: List of antibodies used for ELISA ...... 44 Table 17: Origin of wild Mus musculus colonies ...... 46 Table 18: List of all SPF mouse strains used MHV68 studies ...... 47 Table 19: List of oligonucleotides used for genotyping ...... 50 Table 20: List of pathogens tested in the two mice of the Horst colony...... 61 Table 21: Summary of the results of the non-invasive blood screen of 7 Horst families ...... 64 Table 22: Summary of the results of the non-invasive blood screen to analyze vertical transmission of MmusRHV1 ...... 66 Table 23: Statistical significance of IFNα levels produced by TLR-deficient pDC when compared to WT pDC upon MHV68 infection ...... 78 Table 24: Comparison of frequency of latently infected cells and frequency of MHV68 reactivation...... 113 Table 25: Summary of route dependent replication of MHV68 ...... 136 Table 26: Gender induced differences in mortality following viral infections ...... 138

150

List of figures

List of figures Figure 1: Herpesviridae virion structure...... 1 Figure 2: Phylogenetic analysis of MCMV2...... 4 Figure 3: Phylogenetic analysis of MmusRHV1...... 8 Figure 4: Murine dendritic cell development from bone marrow hematopoietic stem cells and lineage committed progenitors...... 10 Figure 5: Schematic overview of TLR signaling...... 19 Figure 6: Model of the UNC93B protein...... 21 Figure 7: UNC93B differentially regulates trafficking of TLR7 and TLR9 to endosomal compartments...... 23 Figure 8: Regions of a mouse brain ...... 51

Figure 9: Schematic representation of virus titration in organs using TCID50 ...... 52 Figure 10: Schematic representation of preparation and plating of splenocyte dilutions for ex vivo MHV68 reactivation assay ...... 54 Figure 11: Presence of both RHV1 and MCMV2 was verified in the Horst colony...... 57 Figure 12: Inoculation of M2-10B4 cells with cell free homogenate of spleens or salivary gland from Horst mice resulted in the development of viral plaques that were tested positive for MCMV1...... 59 Figure 13: Health monitoring revealed the presence of endo- and ectoparasites in Horst mice...... 60 Figure 14: Systematic non-invasive screen to determine the presence of MmusRHV1 in Horst mice bred at the HZI...... 64 Figure 15: Co-housing cage systems to study horizontal transmission of MmusRHV1 in naturally infected animals...... 67 Figure 16: FACS-purified pDC sorted from Flt3L-differentiated bone marrow cells produce IFNα upon MHV68 infection...... 70 Figure 17: GM-CSF differentiated cDC do not produce IFNα after MHV68 infection...... 72 Figure 18: Enrichment of pDC population by magnetic cell sorting...... 73 Figure 19: pDC deficient of TLR7 and 9 fail to produce IFNα in response to their respective synthetic ligands...... 74

Figure 20: FLDC produce TNFα upon stimulation with Pam3CSK4, ORN Sa19 and infection with MCMV...... 75 Figure 21: Type I IFN responses of pDC upon MHV68 infection are dependent on UNC93B and endosomal TLR7 and 9...... 77

151

List of figures

Figure 22: Overview of MHV68 in vivo experiments conducted in this study...... 80 Figure 23: Mice infected intranasally with MHV68 recover from weight loss...... 81 Figure 24: Replication of MHV68 in the lung does not show TLR dependency on days 3 and 6 post infection...... 83 Figure 25: Replication of MHV68 in the salivary gland is TLR9 dependent 10 days post infection...... 85 Figure 26: MHV68 titers in the spleen on day 10 post intranasal infection are not dependent on endosomal TLR...... 86 Figure 27: Mice lacking UNC93B and specifically endosomal TLR7 show a decrease in splenomegaly 17 days after i.n. infection with MHV68...... 88 Figure 28: Mice infected intraperitoneally with MHV68 do not lose weight...... 89 Figure 29: Early replication of MHV68 in the lung, spleen and liver is UNC93B dependent 3 days post intraperitoneal infection...... 90 Figure 30: Lytic replication of MHV68 in the liver is TLR9 dependent 6 days post intraperitoneal infection...... 92 Figure 31: B6J mice show increased weight loss when compared to B6N mice upon intravenous MHV68 infection...... 93 Figure 32: UNC93B and TLR13 deficient mice show increased weight loss than WT B6N mice upon MHV68 infection...... 94 Figure 33: TLR2-/- and TLR9-/- show less weight loss than B6J mice...... 96 Figure 34: Replication of MHV68 in the lung, spleen and liver is UNC93B dependent 3 days post intravenous infection...... 98 Figure 35: Replication of MHV68 in the olfactory bulb, cerebrum, cerebellum and brain stem is not UNC93B dependent...... 101 Figure 36: Mice lacking UNC93B show less splenomegaly 22 days after i.v. infection with MHV68...... 102 Figure 37: Preparation of splenocytes for limiting dilution nested PCR and ex vivo reactivation assay ...... 104 Figure 38: Absence of UNC93B, TLR7, TLR9 and TLR7/9 promote establishment of latent MHV68 infection in vivo...... 108 Figure 39: Reactivation of MHV68 from splenocytes differs among WT B6N mice, UNC93B-/- and TLR13-/- mice...... 109 Figure 40: Frequency of MHV68 reactivation from splenocytes differs among WT B6J, TLR2-/-, TLR7-/-, TLR9-/- and TLR7/9-/- mice...... 112

152

List of figures

Figure 41: Timeline depicting course of experiments...... 120 Figure 42: Influence of parasitic co-infections on latent herpesvirus infections ...... 123 Figure 43: Plasmid map for pGEM-T-Mmus Beta-2-microglobulin ...... 174 Figure 44: Plasmid map for pGEM-T-MCMV1-gB ...... 175 Figure 45: Plasmid map for pGEM-T-MCMV2-gB ...... 176 Figure 46: Plasmid map for for pGEM-T-MmusRHV1-gB ...... 177

153

List of references

List of references Adachi, K, Tsutsui, H, Kashiwamura, S, Seki, E, Nakano, H, Takeuchi, O, Takeda, K, Okumura, K, Van Kaer, L, Okamura, H, Akira, S, and Nakanishi, K (2001): Plasmodium berghei infection in mice induces liver injury by an IL-12- and toll-like receptor/myeloid differentiation factor 88-dependent mechanism. J Immunol 167, 5928-34.

Adler, B, Schaadt, E, Kempkes, B, Zimber-Strobl, U, Baier, B, and Bornkamm, GW (2002): Control of Epstein-Barr virus reactivation by activated CD40 and viral latent membrane protein 1. Proc Natl Acad Sci U S A 99, 437-42.

Ahmed, EH, and Baiocchi, RA (2016): Murine Models of Epstein-Barr Virus-Associated Lymphomagenesis. Ilar j 57, 55-62.

Akira, S, Uematsu, S, and Takeuchi, O (2006): Pathogen recognition and innate immunity. Cell 124, 783- 801.

Alexopoulou, L, Holt, AC, Medzhitov, R, and Flavell, RA (2001): Recognition of double-stranded RNA and activation of NF-[kappa]B by Toll-like receptor 3. Nature 413, 732-738.

Anderson, KV, Jürgens, G, and Nüsslein-Volhard, C Establishment of dorsal-ventral polarity in the Drosophila embryo: Genetic studies on the role of the Toll gene product. Cell 42, 779-789.

Andrade, WA, Souza Mdo, C, Ramos-Martinez, E, Nagpal, K, Dutra, MS, Melo, MB, Bartholomeu, DC, Ghosh, S, Golenbock, DT, and Gazzinelli, RT (2013): Combined action of nucleic acid-sensing Toll-like receptors and TLR11/TLR12 heterodimers imparts resistance to Toxoplasma gondii in mice. Cell Host Microbe 13, 42-53.

Asada, H (2007): Hypersensitivity to mosquito bites: a unique pathogenic mechanism linking Epstein-Barr virus infection, allergy and oncogenesis. J Dermatol Sci 45, 153-60.

Ascoli, V, Senis, G, Zucchetto, A, Valerio, L, Facchinelli, L, Budroni, M, Dal Maso, L, and Coluzzi, M (2009): Distribution of 'promoter' sandflies associated with incidence of classic Kaposi's sarcoma. Med Vet Entomol 23.

Asgarian-Omran, H, Amirzargar, AA, Zeerleder, S, Mahdavi, M, van Mierlo, G, Solati, S, Jeddi-Tehrani, M, Rabbani, H, Aarden, L, and Shokri, F (2015): Interaction of Bordetella pertussis filamentous hemagglutinin with human TLR2: identification of the TLR2-binding domain. Apmis 123, 156-62.

Asselin-Paturel, C, Brizard, G, Pin, JJ, Briere, F, and Trinchieri, G (2003): Mouse Strain Differences in Plasmacytoid Dendritic Cell Frequency and Function Revealed by a Novel Monoclonal Antibody. The Journal of Immunology 171, 6466-6477.

Banchereau, J, and Steinman, RM (1998): Dendritic cells and the control of immunity. Nature Nature 392, 245-252.

Barchet, W, Blasius, A, Cella, M, and Colonna, M Plasmacytoid dendritic cells. Immunologic Research 32, 75-83.

Barton, E, Mandal, P, and Speck, SH (2011): Pathogenesis and host control of gammaherpesviruses: lessons from the mouse. Annu Rev Immunol 29, 351-97.

154

List of references

Bauer, S, Kirschning, CJ, Hacker, H, Redecke, V, Hausmann, S, Akira, S, Wagner, H, and Lipford, GB (2001): Human TLR9 confers responsiveness to bacterial DNA via species-specific CpG motif recognition. Proc Natl Acad Sci U S A 98, 9237-42.

Bechtel, J, Grundhoff, A, and Ganem, D (2005): RNAs in the Virion of Kaposi's Sarcoma-Associated Herpesvirus. J Virol 79, 10138-46.

Becker, S, Quay, J, and Soukup, J (1991): Cytokine (tumor necrosis factor, IL-6, and IL-8) production by respiratory syncytial virus-infected human alveolar macrophages. The Journal of Immunology 147, 4307- 12.

Bell, JK, Mullen, GED, Leifer, CA, Mazzoni, A, Davies, DR, and Segal, DM (2003): Leucine-rich repeats and pathogen recognition in Toll-like receptors. Trends in Immunology 24, 528-533.

Berghofer, B, Frommer, T, Haley, G, Fink, L, Bein, G, and Hackstein, H (2006): TLR7 Ligands Induce Higher IFN- Production in Females. The Journal of Immunology 177, 2088-2096.

Bieback, K, Lien, E, Klagge, IM, Avota, E, Schneider-Schaulies, J, Duprex, WP, Wagner, H, Kirschning, CJ, ter Meulen, V, and Schneider-Schaulies, S (2002): Hemagglutinin Protein of Wild-Type Measles Virus Activates Toll-Like Receptor 2 Signaling. J Virol 76, 8729-36.

Blanco, P, Palucka, AK, Gill, M, Pascual, V, and Banchereau, J (2001): Induction of dendritic cell differentiation by IFN-alpha in systemic lupus erythematosus. Science 294, 1540-3.

Blasdell, K, McCracken, C, Morris, A, Nash, AA, Begon, M, Bennett, M, and Stewart, JP (2003): The wood mouse is a natural host for Murid herpesvirus 4. Journal of General Virology 84, 111-113.

Blasius, AL, Cella, M, Maldonado, J, Takai, T, and Colonna, M (2006): Siglec-H is an IPC-specific receptor that modulates type I IFN secretion through DAP12. Blood 107, 2474-6.

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 Virol 24, 468.

Bourboulia, D, Whitby, D, Boshoff, C, Newton, R, Beral, V, Carrara, H, Lane, A, and Sitas, F (1998): Serologic evidence for mother-to-child transmission of Kaposi sarcoma-associated herpesvirus infection. Jama 280, 31-2.

Bowden, RJ, Simas, JP, Davis, AJ, and Efstathiou, S (1997): Murine gammaherpesvirus 68 encodes tRNA-like sequences which are expressed during latency. Journal of General Virology 78, 1675-1687.

Brinkmann, MM, Dag, F, Hengel, H, Messerle, M, Kalinke, U, and Cicin-Sain, L (2015): Cytomegalovirus immune evasion of myeloid lineage cells. Med Microbiol Immunol 204, 367-82.

Brinkmann, MM, Spooner, E, Hoebe, K, Beutler, B, Ploegh, HL, and Kim, YM (2007): The interaction between the ER membrane protein UNC93B and TLR3, 7, and 9 is crucial for TLR signaling. J Cell Biol 177, 265-75.

Brown, HJ, Song, MJ, Deng, H, Wu, TT, Cheng, G, and Sun, R (2003): NF- B Inhibits Gammaherpesvirus Lytic Replication. J Virol 77, 8532-8540.

155

List of references

Brown, MG, Dokun, AO, Heusel, JW, Smith, HR, Beckman, DL, Blattenberger, EA, Dubbelde, CE, Stone, LR, Scalzo, AA, and Yokoyama, WM (2001): Vital involvement of a natural killer cell activation receptor in resistance to viral infection. Science 292, 934-7.

Broz, P, and Monack, DM (2013): Newly described pattern recognition receptors team up against intracellular pathogens. Nat Rev Immunol 13, 551-65.

Burns, K, Clatworthy, J, Martin, L, Martinon, F, Plumpton, C, Maschera, B, Lewis, A, Ray, K, Tschopp, J, and Volpe, F (2000): Tollip, a new component of the IL-1RI pathway, links IRAK to the IL-1 receptor. Nat Cell Biol 2, 346-51.

Bussey, KA, Reimer, E, Todt, H, Denker, B, Gallo, A, Konrad, A, Ottinger, M, Adler, H, Sturzl, M, Brune, W, and Brinkmann, MM (2014): The gammaherpesviruses Kaposi's sarcoma-associated herpesvirus and murine gammaherpesvirus 68 modulate the Toll-like receptor-induced proinflammatory cytokine response. J Virol 88, 9245-59.

Caetano, BC, Carmo, BB, Melo, MB, Cerny, A, dos Santos, SL, Bartholomeu, DC, Golenbock, DT, and Gazzinelli, RT (2011): Requirement of UNC93B1 reveals a critical role for TLR7 in host resistance to primary infection with Trypanosoma cruzi. J Immunol 187, 1903-11.

Cai, X, Lu, S, Zhang, Z, Gonzalez, CM, Damania, B, and Cullen, BR (2005): Kaposi's sarcoma- associated herpesvirus expresses an array of viral microRNAs in latently infected cells. Proc Natl Acad Sci U S A 102, 5570-5.

Cannon, MJ, Hyde, TB, and Schmid, DS (2011): Review of cytomegalovirus shedding in bodily fluids and relevance to congenital cytomegalovirus infection. Rev Med Virol 21, 240-55.

Casrouge, A, Zhang, SY, Eidenschenk, C, Jouanguy, E, Puel, A, Yang, K, Alcais, A, Picard, C, Mahfoufi, N, Nicolas, N, Lorenzo, L, Plancoulaine, S, Senechal, B, Geissmann, F, Tabeta, K, Hoebe, K, Du, X, Miller, RL, Heron, B, Mignot, C, de Villemeur, TB, Lebon, P, Dulac, O, Rozenberg, F, Beutler, B, Tardieu, M, Abel, L, and Casanova, JL (2006): Herpes simplex virus encephalitis in human UNC-93B deficiency. Science 314, 308-12.

Cella, M, Facchetti, F, Lanzavecchia, A, and Colonna, M (2000): Plasmacytoid dendritic cells activated by influenza virus and CD40L drive a potent TH1 polarization. Nat Immunol 1, 305-10.

Cesarman, E, Chang, Y, Moore, PS, Said, JW, and Knowles, DM (1995): Kaposi's sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas. N Engl J Med 332, 1186-91.

Chan, PK, Ng, HK, Hui, M, Ip, M, Cheung, JL, and Cheng, AF (1999): Presence of human herpesviruses 6, 7, and 8 DNA sequences in normal brain tissue. J Med Virol 59, 491-5.

Chan, PKS, Ng, H-K, Cheung, JLK, and Cheng, AF (2000): Survey for the Presence and Distribution of Human Herpesvirus 8 in Healthy Brain. Journal of Clinical Microbiology 38, 2772-2773.

Chen, L, Arora, M, Yarlagadda, M, Oriss, TB, Krishnamoorthy, N, Ray, A, and Ray, P (2006): Distinct Responses of Lung and Spleen Dendritic Cells to the TLR9 Agonist CpG Oligodeoxynucleotide. The Journal of Immunology 177, 2373-2383.

156

List of references

Cho, H-J, Kim, S, Kwak, S-E, Kang, T-C, Kim, H-S, Kwon, H-J, Kim, Y-W, Kim, Y-S, Choi, E-K, and Song, MJ (2009): Age-dependent pathogenesis of murine gammaherpesvirus 68 infection of the central nervous system. Molecules and Cells 27, 105-111.

Coban, C, Ishii, KJ, Kawai, T, Hemmi, H, Sato, S, Uematsu, S, Yamamoto, M, Takeuchi, O, Itagaki, S, Kumar, N, Horii, T, and Akira, S (2005): Toll-like receptor 9 mediates innate immune activation by the malaria pigment hemozoin. J Exp Med 201, 19-25.

Coluzzi, M, Calabrò, ML, Manno, D, Chieco-Bianchi, L, Schulz, TF, and Ascoli, V (2003): Reduced seroprevalence of Kaposi's sarcoma-associated herpesvirus (KSHV), human herpesvirus 8 (HHV8), related to suppression of Anopheles density in Italy. Medical and Veterinary Entomology 17, 461-464.

Coluzzi, M, Manno, D, Guzzinati, S, Tognazzo, S, Zambon, P, Arca, B, Costantini, C, and Ascoli, V (2002): The bloodsucking arthropod bite as possible cofactor in the transmission of human herpesvirus-8 infection and in the expression of Kaposi's sarcoma disease. Parassitologia 44.

Cook, CH, Trgovcich, J, Zimmerman, PD, Zhang, Y, and Sedmak, DD (2006): Lipopolysaccharide, tumor necrosis factor alpha, or interleukin-1beta triggers reactivation of latent cytomegalovirus in immunocompetent mice. J Virol 80, 9151-8.

Crane, MJ, Gaddi, PJ, and Salazar-Mather, TP (2012): UNC93B1 Mediates Innate Inflammation and Antiviral Defense in the Liver during Acute Murine Cytomegalovirus Infection. PLoS One 7, e39161.

Dalod, M, Hamilton, T, Salomon, R, Salazar-Mather, TP, Henry, SC, Hamilton, JD, and Biron, CA (2003): Dendritic cell responses to early murine cytomegalovirus infection: subset functional specialization and differential regulation by interferon alpha/beta. J Exp Med 197, 885-98.

Davis, DA, Rinderknecht, AS, Zoeteweij, JP, Aoki, Y, Read-Connole, EL, Tosato, G, Blauvelt, A, and Yarchoan, R (2001): Hypoxia induces lytic replication of Kaposi sarcoma–associated herpesvirus. Blood 97, 3244-3250.

Demaria, O, Pagni, PP, Traub, S, de Gassart, A, Branzk, N, Murphy, AJ, Valenzuela, DM, Yancopoulos, GD, Flavell, RA, and Alexopoulou, L TLR8 deficiency leads to autoimmunity in mice. J Clin Invest 120, 3651-3662.

Diebold, SS, Kaisho, T, Hemmi, H, Akira, S, and Reis e Sousa, C (2004): Innate Antiviral Responses by Means of TLR7-Mediated Recognition of Single-Stranded RNA. Science 303, 1529-1531.

Dorsch-Häsler, K, Keil, GM, Weber, F, Jasin, M, Schaffner, W, and Koszinowski, UH (1985): A long and complex enhancer activates transcription of the gene coding for the highly abundant immediate early mRNA in murine cytomegalovirus. Proc Natl Acad Sci U S A 82, 8325-9.

Edwards, AD, Diebold, SS, Slack, EMC, Tomizawa, H, Hemmi, H, Kaisho, T, Akira, S, and Sousa, CRe (2003): Toll-like receptor expression in murine DC subsets: lack of TLR7 expression by CD8α+ DC correlates with unresponsiveness to imidazoquinolines. Eur J Immunol 33, 827-833.

Ehlers, B, Dural, G, Yasmum, N, Lembo, T, de Thoisy, B, Ryser-Degiorgis, M-P, Ulrich, RG, and McGeoch, DJ (2008): Novel Mammalian Herpesviruses and Lineages within the Gammaherpesvirinae: Cospeciation and Interspecies Transfer. J Virol 82, 3509-3516.

Ehlers, B, Kuchler, J, Yasmum, N, Dural, G, Voigt, S, Schmidt-Chanasit, J, Jakel, T, Matuschka, FR, Richter, D, Essbauer, S, Hughes, DJ, Summers, C, Bennett, M, Stewart, JP, and Ulrich, RG (2007):

157

List of references

Identification of novel rodent herpesviruses, including the first gammaherpesvirus of Mus musculus. J Virol 81, 8091-100.

Ehtisham, S, Sunil-Chandra, NP, and Nash, AA (1993): Pathogenesis of murine gammaherpesvirus infection in mice deficient in CD4 and CD8 T cells. J Virol 67, 5247-5252.

Epstein, MA, Achong, BG, and Barr, YM (1964): Virus particles in cultured lymphoblasts from burkitt's lymphoma. Lancet 1, 702-3.

Evans, JT, Cluff, CW, Johnson, DA, Lacy, MJ, Persing, DH, and Baldridge, JR (2003): Enhancement of antigen-specific immunity via the TLR4 ligands MPL adjuvant and Ribi.529. Expert Rev Vaccines 2, 219- 29.

Everts, B, Perona-Wright, G, Smits, HH, Hokke, CH, van der Ham, AJ, Fitzsimmons, CM, Doenhoff, MJ, van der Bosch, J, Mohrs, K, Haas, H, Mohrs, M, Yazdanbakhsh, M, and Schramm, G (2009): Omega-1, a glycoprotein secreted by Schistosoma mansoni eggs, drives Th2 responses. J Exp Med 206, 1673-80.

Ewald, SE, Engel, A, Lee, J, Wang, M, Bogyo, M, and Barton, GM (2011): Nucleic acid recognition by Toll-like receptors is coupled to stepwise processing by cathepsins and asparagine endopeptidase. J Exp Med 208, 643-51.

Ewald, SE, Lee, BL, Lau, L, Wickliffe, KE, Shi, G-P, Chapman, HA, and Barton, GM (2008): The ectodomain of Toll-like receptor 9 is cleaved to generate a functional receptor. Nature 456, 658-662.

Fairhurst, AM, Hwang, SH, Wang, A, Tian, XH, Boudreaux, C, Zhou, XJ, Casco, J, Li, QZ, Connolly, JE, and Wakeland, EK (2008): Yaa autoimmune phenotypes are conferred by overexpression of TLR7. Eur J Immunol 38, 1971-8.

Feldman, ER, Kara, M, Coleman, CB, Grau, KR, Oko, LM, Krueger, BJ, Renne, R, van Dyk, LF, and Tibbetts, SA (2014): Virus-encoded microRNAs facilitate gammaherpesvirus latency and pathogenesis in vivo. MBio 5, e00981-14.

Ferreira, BR, and Silva, JS (1999): Successive tick infestations selectively promote a T-helper 2 cytokine profile in mice. Immunology 96, 434-9.

Fields, B. N., Knipe, D. M., Howley, P. M., & Griffin, D. E. (2001). Fields virology. Philadelphia: Lippincott Williams & Wilkins

Flano, E, Husain, SM, Sample, JT, Woodland, DL, and Blackman, MA (2000): Latent murine gammaherpesvirus infection is established in activated B cells, dendritic cells, and macrophages. J Immunol 165, 1074-81.

Flano, E, Jia, Q, Moore, J, Woodland, DL, Sun, R, and Blackman, MA (2005): Early establishment of gamma-herpesvirus latency: implications for immune control. J Immunol 174, 4972-8.

Flano, E, Kim, IJ, Moore, J, Woodland, DL, and Blackman, MA (2003): Differential gamma-herpesvirus distribution in distinct anatomical locations and cell subsets during persistent infection in mice. J Immunol 170, 3828-34.

Forster, AC, and Altman, S (1990): External guide sequences for an RNA enzyme. Science 249, 783-6.

158

List of references

Francois, S, Vidick, S, Sarlet, M, Desmecht, D, Drion, P, Stevenson, PG, Vanderplasschen, A, and Gillet, L (2013): Illumination of murine gammaherpesvirus-68 cycle reveals a sexual transmission route from females to males in laboratory mice. PLoS Pathog 9, e1003292.

Francois, S, Vidick, S, Sarlet, M, Michaux, J, Koteja, P, Desmecht, D, Stevenson, PG, Vanderplasschen, A, and Gillet, L (2010): Comparative study of murid gammaherpesvirus 4 infection in mice and in a natural host, bank voles. J Gen Virol 91, 2553-63.

Frazer, KA, Eskin, E, Kang, HM, Bogue, MA, Hinds, DA, Beilharz, EJ, Gupta, RV, Montgomery, J, Morenzoni, MM, Nilsen, GB, Pethiyagoda, CL, Stuve, LL, Johnson, FM, Daly, MJ, Wade, CM, and Cox, DR (2007): A sequence-based variation map of 8.27 million SNPs in inbred mouse strains. Nature 448, 1050-3.

Fuchsberger, M, Hochrein, H, and O'Keeffe, M (2005): Activation of plasmacytoid dendritic cells. Immunol Cell Biol 83, 571-7.

Fukui, R, Saitoh, S, Kanno, A, Onji, M, Shibata, T, Ito, A, Onji, M, Matsumoto, M, Akira, S, Yoshida, N, and Miyake, K (2011): Unc93B1 restricts systemic lethal inflammation by orchestrating Toll-like receptor 7 and 9 trafficking. Immunity 35, 69-81.

Fukui, R, Saitoh, S, Matsumoto, F, Kozuka-Hata, H, Oyama, M, Tabeta, K, Beutler, B, and Miyake, K (2009): Unc93B1 biases Toll-like receptor responses to nucleic acid in dendritic cells toward DNA- but against RNA-sensing. J Exp Med 206, 1339-50.

Gallo, RC (1998): The enigmas of Kaposi's sarcoma. Science 282, 1837-9.

Garcia-Cattaneo, A, Gobert Fç, X, Müller, M, Toscano, F, Flores, M, Lescure, A, Del Nery, E, and Benaroch, P (2012): Cleavage of Toll-like receptor 3 by cathepsins B and H is essential for signaling. Proc Natl Acad Sci U S A 109, 9053-8.

Gargano, LM, Forrest, JC, and Speck, SH (2009): Signaling through Toll-like receptors induces murine gammaherpesvirus 68 reactivation in vivo. J Virol 83, 1474-82.

Gargano, LM, Moser, JM, and Speck, SH (2008): Role for MyD88 signaling in murine gammaherpesvirus 68 latency. J Virol 82, 3853-63.

Geurs, TL, Hill, EB, Lippold, DM, and French, AR (2012): Sex differences in murine susceptibility to systemic viral infections. J Autoimmun 38, J245-53.

Gomes, I, Sharma, TT, Mahmud, N, Kapp, JD, Edassery, S, Fulton, N, Liang, J, Hoffman, R, and Westbrook, CA (2001): Highly abundant genes in the transcriptosome of human and baboon CD34 antigen-positive bone marrow cells. Blood 98, 93-99.

Gregory, SM, West, JA, Dillon, PJ, Hilscher, C, Dittmer, DP, and Damania, B (2009): Toll-like receptor signaling controls reactivation of KSHV from latency. Proc Natl Acad Sci U S A 106, 11725-30.

Grouard, G, Rissoan, MC, Filgueira, L, Durand, I, Banchereau, J, and Liu, YJ (1997): The enigmatic plasmacytoid T cells develop into dendritic cells with interleukin (IL)-3 and CD40-ligand. J Exp Med 185, 1101-11.

159

List of references

Grunewald, K, Desai, P, Winkler, DC, Heymann, JB, Belnap, DM, Baumeister, W, and Steven, AC (2003): Three-dimensional structure of herpes simplex virus from cryo-electron tomography. Science 302, 1396- 8.

Guan, Y, Ranoa, DR, Jiang, S, Mutha, SK, Li, X, Baudry, J, and Tapping, RI (2010): Human TLRs 10 and 1 share common mechanisms of innate immune sensing but not signaling. J Immunol 184, 5094-103.

Guggemoos, S, Hangel, D, Hamm, S, Heit, A, Bauer, S, and Adler, H (2007): TLR9 Contributes to Antiviral Immunity during Gammaherpesvirus Infection. The Journal of Immunology 180, 438-443.

Guo, H, Shen, S, Wang, L, and Deng, H (2010): Role of tegument proteins in herpesvirus assembly and egress. Protein Cell 1, 987-98.

H Speck, S, and W Virgin Iv, H (1999): Host and viral genetics of chronic infection: a mouse model of gamma-herpesvirus pathogenesis. Current Opinion in Microbiology 2, 403-409.

Haas, F, Yamauchi, K, Murat, M, Bernasconi, M, Yamanaka, N, Speck, RF, and Nadal, D (2014): Activation of NF-kappaB via endosomal Toll-like receptor 7 (TLR7) or TLR9 suppresses murine herpesvirus 68 reactivation. J Virol 88, 10002-12.

Han, X, Lundberg, P, Tanamachi, B, Openshaw, H, Longmate, J, and Cantin, E (2001): Gender influences herpes simplex virus type 1 infection in normal and gamma interferon-mutant mice. J Virol 75, 3048-52.

Hardison, SE, and Brown, GD (2012): C-type Lectin Receptors Orchestrate Anti-Fungal Immunity. Nat Immunol 13, 817-22.

Hasan, U, Chaffois, C, Gaillard, C, Saulnier, V, Merck, E, Tancredi, S, Guiet, C, Briere, F, Vlach, J, Lebecque, S, Trinchieri, G, and Bates, EE (2005): Human TLR10 is a functional receptor, expressed by B cells and plasmacytoid dendritic cells, which activates gene transcription through MyD88. J Immunol 174, 2942-50.

Hausen, HZ, O'Neill, FJ, Freese, UK, and Hecker, E (1978): Persisting oncogenic herpesvirus induced by the tumour promoter TPA. Nature 272, 373-375.

Hausler, M, Sellhaus, B, Scheithauer, S, Engler, M, Alberg, E, Teubner, A, Ritter, K, and Kleines, M (2005): Murine gammaherpesvirus-68 infection of mice: A new model for human cerebral Epstein-Barr virus infection. Ann Neurol 57, 600-3.

Hayashi, F, Smith, KD, Ozinsky, A, Hawn, TR, Yi, EC, Goodlett, DR, Eng, JK, Akira, S, Underhill, DM, and Aderem, A (2001): The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 410, 1099-1103.

He, J, Bhat, G, Kankasa, C, Chintu, C, Mitchell, C, Duan, W, and Wood, C (1998): Seroprevalence of human herpesvirus 8 among Zambian women of childbearing age without Kaposi's sarcoma (KS) and mother-child pairs with KS. J Infect Dis 178, 1787-90.

Heil, F, Hemmi, H, Hochrein, H, Ampenberger, F, Kirschning, C, Akira, S, Lipford, G, Wagner, H, and Bauer, S (2004): Species-specific recognition of single-stranded RNA via toll-like receptor 7 and 8. Science 303, 1526-9.

160

List of references

Helft, J, Bottcher, J, Chakravarty, P, Zelenay, S, Huotari, J, Schraml, BU, Goubau, D, and Reis e Sousa, C (2015): GM-CSF Mouse Bone Marrow Cultures Comprise a Heterogeneous Population of CD11c(+)MHCII(+) Macrophages and Dendritic Cells. Immunity 42, 1197-211.

Hemmi, H, Kaisho, T, Takeuchi, O, Sato, S, Sanjo, H, Hoshino, K, Horiuchi, T, Tomizawa, H, Takeda, K, and Akira, S (2002): Small anti-viral compounds activate immune cells via the TLR7 MyD88-dependent signaling pathway. Nat Immunol 3, 196-200.

Hemmi, H, Takeuchi, O, Kawai, T, Kaisho, T, Sato, S, Sanjo, H, Matsumoto, M, Hoshino, K, Wagner, H, Takeda, K, and Akira, S (2000): A Toll-like receptor recognizes bacterial DNA. Nature 408, 740-5.

Hidmark, A, von Saint Paul, A, and Dalpke, AH (2012): Cutting edge: TLR13 is a receptor for bacterial RNA. J Immunol 189, 2717-21.

Hjalgrim H, Friborg J, Melbye M. The epidemiology of EBV and its association with malignant disease. In: Arvin A, Campadelli-Fiume G, Mocarski E, et al., editors. Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge: Cambridge University Press; 2007. Chapter 53.

Hladik, W, Dollard, SC, Mermin, J, Fowlkes, AL, Downing, R, Amin, MM, Banage, F, Nzaro, E, Kataaha, P, Dondero, TJ, Pellett, PE, and Lackritz, EM (2006): Transmission of human herpesvirus 8 by blood transfusion. N Engl J Med 355, 1331-8.

Hochrein, H, Shortman, K, Vremec, D, Scott, B, Hertzog, P, and O'Keeffe, M (2001): Differential Production of IL-12, IFN- , and IFN- by Mouse Dendritic Cell Subsets. The Journal of Immunology 166, 5448-5455.

Hoebe, K, Janssen, EM, Kim, SO, Alexopoulou, L, Flavell, RA, Han, J, and Beutler, B (2003): Upregulation of costimulatory molecules induced by lipopolysaccharide and double-stranded RNA occurs by Trif-dependent and Trif-independent pathways. Nat Immunol 4, 1223-9.

Honda, K, Ohba, Y, Yanai, H, Negishi, H, Mizutani, T, Takaoka, A, Taya, C, and Taniguchi, T (2005): Spatiotemporal regulation of MyD88-IRF-7 signalling for robust type-I interferon induction. Nature 434, 1035-40.

Horng, T, Barton, GM, Flavell, RA, and Medzhitov, R (2002): The adaptor molecule TIRAP provides signalling specificity for Toll-like receptors. Nature 420, 329-33.

Hoshino, K, Kaisho, T, Iwabe, T, Takeuchi, O, and Akira, S (2002): Differential involvement of IFN-beta in Toll-like receptor-stimulated dendritic cell activation. Int Immunol 14, 1225-31.

Hoshino, K, Takeuchi, O, Kawai, T, Sanjo, H, Ogawa, T, Takeda, Y, Takeda, K, and Akira, S (1999): Cutting Edge: Toll-Like Receptor 4 (TLR4)-Deficient Mice Are Hyporesponsive to Lipopolysaccharide: Evidence for TLR4 as the Lps Gene Product. The Journal of Immunology 162, 3749-3752.

Hricova, M, and Mistrikova, J (2007): Murine gammaherpesvirus 68 serum antibodies in general human population. Acta Virol 51, 283-7.

Hudson, JB, Chantler, JK, Loh, L, Misra, V, and Muller, MT (1979): Model systems for analysis of latent cytomegalovirus infections. Can J Microbiol 25, 245-53.

Hudson, JB, Walker, DG, and Altamirano, M Analysis in vitro of two biologically distinct strains of murine cytomegalovirus. Arch Virol 102, 289-295.

161

List of references

Hughes, DJ, Kipar, A, Sample, JT, and Stewart, JP (2010): Pathogenesis of a model gammaherpesvirus in a natural host. J Virol 84, 3949-61.

Huh, JW, Shibata, T, Hwang, M, Kwon, EH, Jang, MS, Fukui, R, Kanno, A, Jung, DJ, Jang, MH, Miyake, K, and Kim, YM (2014): UNC93B1 is essential for the plasma membrane localization and signaling of Toll- like receptor 5. Proc Natl Acad Sci U S A 111, 7072-7.

Hutter, JA, Scott, J, Wreghitt, T, Higenbottam, T, and Wallwork, J (1989): The Importance of Cytomegalovirus in Heart-Lung Transplant Recipients. Chest 95, 627-631.

Hwang, S, Wu, TT, Tong, LM, Kim, KS, Martinez-Guzman, D, Colantonio, AD, Uittenbogaart, CH, and Sun, R (2008): Persistent gammaherpesvirus replication and dynamic interaction with the host in vivo. J Virol 82, 12498-509.

Itoh, H, Tatematsu, M, Watanabe, A, Iwano, K, Funami, K, Seya, T, and Matsumoto, M (2011): UNC93B1 physically associates with human TLR8 and regulates TLR8-mediated signaling. PLoS One 6, e28500.

Jego, G, Palucka, AK, Blanck, JP, Chalouni, C, Pascual, V, and Banchereau, J (2003): Plasmacytoid dendritic cells induce plasma cell differentiation through type I interferon and interleukin 6. Immunity 19, 225-34.

Jennings, RN, Grayson, JM, and Barton, ES (2014): Type I interferon signaling enhances CD8+ T cell effector function and differentiation during murine gammaherpesvirus 68 infection. J Virol 88, 14040-9.

Jiang, X, Gong, H, Chen, Y-C, Vu, G-P, Trang, P, Zhang, C-Y, Lu, S, and Liu, F (2012): Effective inhibition of cytomegalovirus infection by external guide sequences in mice. Proceedings of the National Academy of Sciences 109, 13070-13075.

Jin, MS, Kim, SE, Heo, JY, Lee, ME, Kim, HM, Paik, SG, Lee, H, and Lee, JO (2007): Crystal structure of the TLR1-TLR2 heterodimer induced by binding of a tri-acylated lipopeptide. Cell 130, 1071-82.

Jin, MS, and Lee, JO (2008): Structures of the toll-like receptor family and its ligand complexes. Immunity 29, 182-91.

Kanayama, A, Seth, RB, Sun, L, Ea, C-K, Hong, M, Shaito, A, Chiu, Y-H, Deng, L, and Chen, ZJ TAB2 and TAB3 Activate the NF-κB Pathway through Binding to Polyubiquitin Chains. Molecular Cell 15, 535-548.

Kang, HR, Cho, HJ, Kim, S, Song, IH, Lee, TS, Hwang, S, Sun, R, and Song, MJ (2012): Persistent infection of a gammaherpesvirus in the central nervous system. Virology 423, 23-9.

Karatas, H, Gurer, G, Pinar, A, Soylemezoglu, F, Tezel, GG, Hascelik, G, Akalan, N, Tuncer, S, Ciger, A, and Saygi, S (2008): Investigation of HSV-1, HSV-2, CMV, HHV-6 and HHV-8 DNA by real-time PCR in surgical resection materials of epilepsy patients with mesial temporal lobe sclerosis. J Neurol Sci 264, 151-6.

Karlin, S, Doerfler, W, and Cardon, LR (1994): Why is CpG suppressed in the genomes of virtually all small eukaryotic viruses but not in those of large eukaryotic viruses? J Virol 68, 2889-97.

Kawai, T, and Akira, S (2007): Signaling to NF-kappaB by Toll-like receptors. Trends Mol Med 13, 460-9.

162

List of references

Kenneson, A, and Cannon, MJ (2007): Review and meta-analysis of the epidemiology of congenital cytomegalovirus (CMV) infection. Rev Med Virol 17, 253-76.

Khan, AR, and Fallon, PG (2013): Helminth therapies: translating the unknown unknowns to known knowns. Int J Parasitol 43, 293-9.

Kim, HM, Park, BS, Kim, JI, Kim, SE, Lee, J, Oh, SC, Enkhbayar, P, Matsushima, N, Lee, H, Yoo, OJ, and Lee, JO (2007): Crystal structure of the TLR4-MD-2 complex with bound endotoxin antagonist Eritoran. Cell 130, 906-17.

Kim, KS, and Carp, RI (1971): Growth of murine cytomegalovirus in various cell lines. J Virol 7, 720-5.

Kim, YM, Brinkmann, MM, Paquet, ME, and Ploegh, HL (2008): UNC93B1 delivers nucleotide-sensing toll-like receptors to endolysosomes. Nature 452, 234-8.

Krug, A, Luker, GD, Barchet, W, Leib, DA, Akira, S, and Colonna, M (2004): Herpes simplex virus type 1 activates murine natural interferon-producing cells through toll-like receptor 9. Blood 103, 1433-7.

Krug, A, Towarowski, A, Britsch, S, Rothenfusser, S, Hornung, V, Bals, R, Giese, T, Engelmann, H, Endres, S, Krieg, AM, and Hartmann, G (2001): Toll-like receptor expression reveals CpG DNA as a unique microbial stimulus for plasmacytoid dendritic cells which synergizes with CD40 ligand to induce high amounts of IL-12. Eur J Immunol 31, 3026-37.

Krug, LT, Moser, JM, Dickerson, SM, and Speck, SH (2007): Inhibition of NF-?B Activation In Vivo Impairs Establishment of Gammaherpesvirus Latency. PLoS Pathog 3, e11.

Krug, LT, Torres-González, E, Qin, Q, Sorescu, D, Rojas, M, Stecenko, A, Speck, SH, and Mora, AL (2010): Inhibition of NF-κB Signaling Reduces Virus Load and Gammaherpesvirus-Induced Pulmonary Fibrosis. Am J Pathol 177, 608-21.

Kumar, H, Kawai, T, and Akira, S (2009): Pathogen recognition in the innate immune response. Biochem J 420, 1-16.

Kundu, JK, Shin, YK, Kim, SH, and Surh, YJ (2006): Resveratrol inhibits phorbol ester-induced expression of COX-2 and activation of NF-kappaB in mouse skin by blocking IkappaB kinase activity. Carcinogenesis 27, 1465-74.

Kupiec-Weglinski, JW, Austyn, JM, and Morris, PJ (1988): Migration patterns of dendritic cells in the mouse. Traffic from the blood, and T cell-dependent and -independent entry to lymphoid tissues. J Exp Med 167, 632-45.

Kurt-Jones, EA, Chan, M, Zhou, S, Wang, J, Reed, G, Bronson, R, Arnold, MM, Knipe, DM, and Finberg, RW (2004): Herpes simplex virus 1 interaction with Toll-like receptor 2 contributes to lethal encephalitis. Proc Natl Acad Sci U S A 101, 1315-20.

Lacoste, V, Lavergne, A, de Thoisy, B, Pouliquen, JF, and Gessain, A (2010): Genetic diversity and molecular evolution of human and non-human primate Gammaherpesvirinae. Infect Genet Evol 10, 1-13.

Lafferty, EI, Flaczyk, A, Angers, I, Homer, R, d'Hennezel, E, Malo, D, Piccirillo, CA, Vidal, SM, and Qureshi, ST (2014): An ENU-induced splicing mutation reveals a role for Unc93b1 in early immune cell activation following influenza A H1N1 infection. Genes Immun 15, 320-332.

163

List of references

Lafferty, EI, Wiltshire, SA, Angers, I, Vidal, SM, and Qureshi, ST (2015): Unc93b1 -Dependent Endosomal Toll-Like Receptor Signaling Regulates Inflammation and Mortality during Coxsackievirus B3 Infection. J Innate Immun 7, 315-30.

Lee, BL, Moon, JE, Shu, JH, Yuan, L, Newman, ZR, Schekman, R, and Barton, GM (2013): UNC93B1 mediates differential trafficking of endosomal TLRs. Elife 2.

Lemaitre, B, Kromer-Metzger, E, Michaut, L, Nicolas, E, Meister, M, Georgel, P, Reichhart, JM, and Hoffmann, JA (1995): A recessive mutation, immune deficiency (imd), defines two distinct control pathways in the Drosophila host defense. Proc Natl Acad Sci U S A 92, 9465-9.

Lemaitre, B, Nicolas, E, Michaut, L, Reichhart, JM, and Hoffmann, JA (1996): The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 86, 973-83.

Li, Q, Zhang, X, Peng, Y, Chai, H, Xu, Y, Wei, J, Ren, X, Wang, X, Liu, W, Chen, M, and Huang, D (2013): Comparison of the sorting efficiency and influence on cell function between the sterile flow cytometry and immunomagnetic bead purification methods. Prep Biochem Biotechnol 43, 197-206.

Li, X, Saitoh, S, Shibata, T, Tanimura, N, Fukui, R, and Miyake, K (2015): Mucolipin 1 positively regulates TLR7 responses in dendritic cells by facilitating RNA transportation to lysosomes. Int Immunol 27, 83-94.

Lim, HK, Seppänen, M, Hautala, T, Ciancanelli, MJ, Itan, Y, Lafaille, FG, Dell, W, Lorenzo, L, Byun, M, Pauwels, E, Rönnelid, Y, Cai, X, Boucherit, S, Jouanguy, E, Paetau, A, Lebon, P, Rozenberg, F, Tardieu, M, Abel, L, Yildiran, A, Vergison, A, Roivainen, R, Etzioni, A, Tienari, PJ, Casanova, JL, and Zhang, SY (2014): TLR3 deficiency in herpes simplex encephalitis: High allelic heterogeneity and recurrence risk. Neurology 83, 1888-97.

Liu, L, Botos, I, Wang, Y, Leonard, JN, Shiloach, J, Segal, DM, and Davies, DR (2008): Structural basis of toll-like receptor 3 signaling with double-stranded RNA. Science 320, 379-81.

Liu, YJ (2005): IPC: professional type 1 interferon-producing cells and plasmacytoid dendritic cell precursors. Annu Rev Immunol 23, 275-306.

Loof, TG, Goldmann, O, and Medina, E (2008): Immune recognition of Streptococcus pyogenes by dendritic cells. Infect Immun 76, 2785-92.

Luber, CA, Cox, J, Lauterbach, H, Fancke, B, Selbach, M, Tschopp, J, Akira, S, Wiegand, M, Hochrein, H, O'Keeffe, M, and Mann, M (2010): Quantitative proteomics reveals subset-specific viral recognition in dendritic cells. Immunity 32, 279-89.

Lundberg, P, Welander, P, Han, X, and Cantin, E (2003): Herpes Simplex Virus Type 1 DNA Is Immunostimulatory In Vitro and In Vivo. J Virol 77, 11158-69.

Lutz, MB, Kukutsch, N, Ogilvie, AL, Rossner, S, Koch, F, Romani, N, and Schuler, G (1999): An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods 223, 77-92.

Ma, Y, Li, J, Chiu, I, Wang, Y, Sloane, JA, Lu, J, Kosaras, B, Sidman, RL, Volpe, JJ, and Vartanian, T (2006): Toll-like receptor 8 functions as a negative regulator of neurite outgrowth and inducer of neuronal apoptosis. J Cell Biol 175, 209-15.

164

List of references

Maldonado-López, R, De Smedt, T, Michel, P, Godfroid, J, Pajak, B, Heirman, C, Thielemans, K, Leo, O, Urbain, J, and Moser, M (1999): CD8α+ and CD8α− Subclasses of Dendritic Cells Direct the Development of Distinct T Helper Cells In Vivo. J Exp Med 189, 587-592.

Manicklal, S, Emery, VC, Lazzarotto, T, Boppana, SB, and Gupta, RK (2013): The “Silent” Global Burden of Congenital Cytomegalovirus. Clin Microbiol Rev 26, 86-102.

Martinez, J, Huang, X, and Yang, Y (2010): Direct TLR2 signaling is critical for NK cell activation and function in response to vaccinia viral infection. PLoS Pathog 6, e1000811.

Martro, E, Esteve, A, Schulz, TF, Sheldon, J, Gambus, G, Munoz, R, Whitby, D, and Casabona, J (2007): Risk factors for human Herpesvirus 8 infection and AIDS-associated Kaposi's sarcoma among men who have sex with men in a European multicentre study. Int J Cancer 120, 1129-35.

Matthys, P, Mitera, T, Heremans, H, Van Damme, J, and Billiau, A (1995): Anti-gamma interferon and anti-interleukin-6 antibodies affect staphylococcal enterotoxin B-induced weight loss, hypoglycemia, and cytokine release in D-galactosamine-sensitized and unsensitized mice. Infect Immun 63, 1158-64.

Mayama, S, Cuevas, LE, Sheldon, J, Omar, OH, Smith, DH, Okong, P, Silvel, B, Hart, CA, and Schulz, TF (1998): Prevalence and transmission of Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) in Ugandan children and adolescents. Int J Cancer 77, 817-20.

Medzhitov, R, Preston-Hurlburt, P, and Janeway, CA (1997a): A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 388, 394-397.

Medzhitov, R, Preston-Hurlburt, P, and Janeway, CA, Jr. (1997b): A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 388, 394-7.

Medzhitov, R, Preston-Hurlburt, P, Kopp, E, Stadlen, A, Chen, C, Ghosh, S, and Janeway, CA, Jr. (1998): MyD88 is an adaptor protein in the hToll/IL-1 receptor family signaling pathways. Mol Cell 2, 253-8.

Meier, A, Chang, JJ, Chan, ES, Pollard, RB, Sidhu, HK, Kulkarni, S, Wen, TF, Lindsay, RJ, Orellana, L, Mildvan, D, Bazner, S, Streeck, H, Alter, G, Lifson, JD, Carrington, M, Bosch, RJ, Robbins, GK, and Altfeld, M (2009): Sex differences in the Toll-like receptor-mediated response of plasmacytoid dendritic cells to HIV-1. Nat Med 15, 955-9.

Meijer, K, de Vos, P, and Priebe, MG (2010): Butyrate and other short-chain fatty acids as modulators of immunity: what relevance for health? Curr Opin Clin Nutr Metab Care 13, 715-21.

Mesri, EA, Cesarman, E, and Boshoff, C (2010): Kaposi's sarcoma and its associated herpesvirus. Nat Rev Cancer 10, 707-19.

Michaud, F, Coulombe, F, Gaudreault, E, Kriz, J, and Gosselin, J (2010): Involvement of TLR2 in recognition of acute gammaherpesvirus-68 infection. PLoS One 5, e13742.

Miller, RL, Imbertson, LM, Reiter, MJ, and Gerster, JF (1999): Treatment of primary herpes simplex virus infection in guinea pigs by imiquimod. Antiviral Res 44, 31-42.

Minhas, V, and Wood, C (2014): Epidemiology and transmission of Kaposi's sarcoma-associated herpesvirus. Viruses 6, 4178-94.

165

List of references

Moore, PS, and Chang, Y (2010): Why do viruses cause cancer? Highlights of the first century of human tumour virology. Nat Rev Cancer 10, 878-89.

Moser, JM, Upton, JW, Gray, KS, and Speck, SH (2005): Ex vivo stimulation of B cells latently infected with gammaherpesvirus 68 triggers reactivation from latency. J Virol 79, 5227-31.

Moyron-Quiroz, JE, Rangel-Moreno, J, Kusser, K, Hartson, L, Sprague, F, Goodrich, S, Woodland, DL, Lund, FE, and Randall, TD (2004): Role of inducible bronchus associated lymphoid tissue (iBALT) in respiratory immunity. Nat Med 10, 927-34.

Murphy, K., Travers, P., Walport, M., & Janeway, C. (2008). Janeway's immunobiology. New York: Garland Science.

Nash, AA, Dutia, BM, Stewart, JP, and Davison, AJ (2001): Natural history of murine gamma-herpesvirus infection. Philos Trans R Soc Lond B Biol Sci 356, 569-79.

Nash, AA, and Sunil-Chandra, NP (1994): Interactions of the murine gammaherpesvirus with the immune system. Curr Opin Immunol 6, 560-3.

Ngo, ST, Steyn, FJ, and McCombe, PA (2014): Gender differences in autoimmune disease. Frontiers in Neuroendocrinology 35, 347-369.

Nicola, AV, McEvoy, AM, and Straus, SE (2003): Roles for endocytosis and low pH in herpes simplex virus entry into HeLa and Chinese hamster ovary cells. J Virol 77, 5324-32.

O'Keeffe, M, Hochrein, H, Vremec, D, Pooley, J, Evans, R, Woulfe, S, and Shortman, K (2002): Effects of administration of progenipoietin 1, Flt-3 ligand, granulocyte colony-stimulating factor, and pegylated granulocyte-macrophage colony-stimulating factor on dendritic cell subsets in mice. Blood 99, 2122 - 2130

O'Neill, LA, and Bowie, AG (2007): The family of five: TIR-domain-containing adaptors in Toll-like receptor signalling. Nat Rev Immunol 7, 353-64.

Okada, T, Lian, Z-X, Naiki, M, Ansari, AA, Ikehara, S, and Gershwin, ME (2003): Murine thymic plasmacytoid dendritic cells. Eur J Immunol 33, 1012-1019.

Oldenburg, M, Krüger, A, Ferstl, R, Kaufmann, A, Nees, G, Sigmund, A, Bathke, B, Lauterbach, H, Suter, M, Dreher, S, Koedel, U, Akira, S, Kawai, T, Buer, J, Wagner, H, Bauer, S, Hochrein, H, and Kirschning, CJ (2012): TLR13 Recognizes Bacterial 23S rRNA Devoid of Erythromycin Resistance–Forming Modification. Science 337, 1111-1115.

Orzechowska, BU, Powers, MF, Sprague, J, Li, H, Yen, B, Searles, RP, Axthelm, MK, and Wong, SW (2008): Rhesus macaque rhadinovirus-associated non-Hodgkin lymphoma: animal model for KSHV- associated malignancies. Blood 112, 4227-34.

Osborn, L, Kunkel, S, and Nabel, GJ (1989): Tumor necrosis factor alpha and interleukin 1 stimulate the human immunodeficiency virus enhancer by activation of the nuclear factor kappa B. Proceedings of the National Academy of Sciences 86, 2336-2340.

Ozinsky, A, Underhill, DM, Fontenot, JD, Hajjar, AM, Smith, KD, Wilson, CB, Schroeder, L, and Aderem, A (2000): The repertoire for pattern recognition of pathogens by the innate. Proc Natl Acad Sci U S A 97, 13766-71.

166

List of references

Palucka, AK, Banchereau, J, Blanco, P, and Pascual, V (2002): The interplay of dendritic cell subsets in systemic lupus erythematosus. Immunol Cell Biol 80, 484-8.

Paludan (1998): Interleukin-4 and Interferon-γ: The Quintessence of a Mutual Antagonistic Relationship. Scandinavian Journal of Immunology 48, 459-468.

Paludan, SR, and Bowie, AG (2013): Immune sensing of DNA. Immunity 38, 870-80.

Panchanathan, R, Liu, H, and Choubey, D (2013): Expression of murine Unc93b1 is up-regulated by interferon and estrogen signaling: implications for sex bias in the development of autoimmunity. Int Immunol 25, 521-9.

Park, B, Brinkmann, MM, Spooner, E, Lee, CC, Kim, YM, and Ploegh, HL (2008): Proteolytic cleavage in an endolysosomal compartment is required for activation of Toll-like receptor 9. Nat Immunol 9, 1407-14.

Pauk, J, Huang, ML, Brodie, SJ, Wald, A, Koelle, DM, Schacker, T, Celum, C, Selke, S, and Corey, L (2000): Mucosal shedding of human herpesvirus 8 in men. N Engl J Med 343, 1369-77.

Pellet, P. and Roizman, B. (2007) Herpesviridae: A Brief Introduction. In: Howley, P., Ed., Fields Virology, 5th Edition, Lippincott, Philadelphia, 2480-2499.

Pembrey, L, Raynor, P, Griffiths, P, Chaytor, S, Wright, J, and Hall, AJ (2013): Seroprevalence of Cytomegalovirus, Epstein Barr Virus and Varicella Zoster Virus among Pregnant Women in Bradford: A Cohort Study. PLoS One 8.

Penna, G, Sozzani, S, and Adorini, L (2001): Cutting edge: selective usage of chemokine receptors by plasmacytoid dendritic cells. J Immunol 167, 1862-6.

Pezda, AC, Penn, A, Barton, GM, and Coscoy, L (2011): Suppression of TLR9 immunostimulatory motifs in the genome of a gammaherpesvirus. J Immunol 187, 887-96.

Pichyangkul, S, Yongvanitchit, K, Kum-arb, U, Hemmi, H, Akira, S, Krieg, AM, Heppner, DG, Stewart, VA, Hasegawa, H, Looareesuwan, S, Shanks, GD, and Miller, RS (2004): Malaria Blood Stage Parasites Activate Human Plasmacytoid Dendritic Cells and Murine Dendritic Cells through a Toll-Like Receptor 9- Dependent Pathway. The Journal of Immunology 172, 4926-4933.

Pifer, R, and Yarovinsky, F (2011): Innate responses to Toxoplasma gondii in mice and humans. Trends Parasitol 27, 388-93.

Poeck, H, Wagner, M, Battiany, J, Rothenfusser, S, Wellisch, D, Hornung, V, Jahrsdorfer, B, Giese, T, Endres, S, and Hartmann, G (2004): Plasmacytoid dendritic cells, antigen, and CpG-C license human B cells for plasma cell differentiation and immunoglobulin production in the absence of T-cell help. Blood 103, 3058-3064.

Poltorak, A, He, X, Smirnova, I, Liu, M-Y, Huffel, CV, Du, X, Birdwell, D, Alejos, E, Silva, M, Galanos, C, Freudenberg, M, Ricciardi-Castagnoli, P, Layton, B, and Beutler, B (1998): Defective LPS Signaling in C3H/HeJ and C57BL/10ScCr Mice: Mutations in Tlr4 Gene. Science 282, 2085-2088.

Raetz, M, Hwang, SH, Wilhelm, CL, Kirkland, D, Benson, A, Sturge, CR, Mirpuri, J, Vaishnava, S, Hou, B, Defranco, AL, Gilpin, CJ, Hooper, LV, and Yarovinsky, F (2013): Parasite-induced TH1 cells and intestinal dysbiosis cooperate in IFN-gamma-dependent elimination of Paneth cells. Nat Immunol 14, 136-42.

167

List of references

Ramanan, P, and Razonable, RR (2013): Cytomegalovirus Infections in Solid Organ Transplantation: A Review. Infect Chemother 45, 260-71.

Reddehase, MJ (2002): Antigens and immunoevasins: opponents in cytomegalovirus immune surveillance. Nat Rev Immunol 2, 831-844.

Reed, JL, and Muench, H (1938): A simple method of estimating fifty per cent endpoints. The American Journal of Hygiene 27.

Reese, TA, Wakeman, BS, Choi, HS, Hufford, MM, Huang, SC, Zhang, X, Buck, MD, Jezewski, A, Kambal, A, Liu, CY, Goel, G, Murray, PJ, Xavier, RJ, Kaplan, MH, Renne, R, Speck, SH, Artyomov, MN, Pearce, EJ, and Virgin, HW (2014): Helminth infection reactivates latent gamma-herpesvirus via cytokine competition at a viral promoter. Science 345, 573-7.

Regan, T, Nally, K, Carmody, R, Houston, A, Shanahan, F, Macsharry, J, and Brint, E (2013): Identification of TLR10 as a key mediator of the inflammatory response to Listeria monocytogenes in intestinal epithelial cells and macrophages. J Immunol 191, 6084-92.

Robinson, DP, Huber, SA, Moussawi, M, Roberts, B, Teuscher, C, Watkins, R, Arnold, AP, and Klein, SL (2011a): Sex chromosome complement contributes to sex differences in coxsackievirus B3 but not influenza A virus pathogenesis. Biology of Sex Differences 2, 1-11.

Robinson, DP, Lorenzo, ME, Jian, W, and Klein, SL (2011b): Elevated 17?-Estradiol Protects Females from Influenza A Virus Pathogenesis by Suppressing Inflammatory Responses. PLoS Pathog 7, e1002149.

Robinson, DP, Lorenzo, ME, Jian, W, and Klein, SL (2011c): Elevated 17β-estradiol protects females from influenza A virus pathogenesis by suppressing inflammatory responses. PLoS Pathog 7.

Rock, FL, Hardiman, G, Timans, JC, Kastelein, RA, and Bazan, JF (1998): A family of human receptors structurally related to Drosophila Toll. Proceedings of the National Academy of Sciences 95, 588-593.

Sambucetti, LC, Cherrington, JM, Wilkinson, GW, and Mocarski, ES (1989): NF-kappa B activation of the cytomegalovirus enhancer is mediated by a viral transactivator and by T cell stimulation. Embo j 8, 4251- 8.

Samols, MA, Hu, J, Skalsky, RL, and Renne, R (2005): Cloning and identification of a microRNA cluster within the latency-associated region of Kaposi's sarcoma-associated herpesvirus. J Virol 79, 9301-5.

Sasai, M, Linehan, MM, and Iwasaki, A (2010): Bifurcation of Toll-Like Receptor 9 Signaling by Adaptor Protein 3. Science 329, 1530-4.

Sathe, P, and Wu, L (2011): The network of cytokines, receptors and transcription factors governing the development of dendritic cell subsets. Protein Cell 2, 620-630.

Scalzo, AA, Fitzgerald, NA, Simmons, A, La Vista, AB, and Shellam, GR (1990): Cmv-1, a genetic locus that controls murine cytomegalovirus replication in the spleen. J Exp Med 171, 1469-83.

Scheibe, E., Lienenklaus, S., May, T., Magalhaes, V. G, Weiss, S., Brinkmann M. M. Measurement of MCMV induced interferon-β with immortalized luciferase reporter cells. Methods Book „ Virus-Host- Interaction” in the series “Methods in Molecular Biology” – Springer Verlag, Human Press, 2013.

168

List of references

Schenten, D, and Medzhitov, R (2011): Chapter 3 - The Control of Adaptive Immune Responses by the Innate Immune System. In W. A. Frederick (Ed.): Advances in Immunology, Vol. Volume 109. Academic Press, pp. 87-124.

Schnorrer, P, Behrens, GMN, Wilson, NS, Pooley, JL, Smith, CM, El-Sukkari, D, Davey, G, Kupresanin, F, Li, M, Maraskovsky, E, Belz, GT, Carbone, FR, Shortman, K, Heath, WR, and Villadangos, JA (2006): The dominant role of CD8(+) dendritic cells in cross-presentation is not dictated by antigen capture. Proc Natl Acad Sci U S A 103, 10729-34.

Schoggins, JW, MacDuff, DA, Imanaka, N, Gainey, MD, Shrestha, B, Eitson, JL, Mar, KB, Richardson, RB, Ratushny, AV, Litvak, V, Dabelic, R, Manicassamy, B, Aitchison, JD, Aderem, A, Elliott, RM, Garcia- Sastre, A, Racaniello, V, Snijder, EJ, Yokoyama, WM, Diamond, MS, Virgin, HW, and Rice, CM (2014): Pan-viral specificity of IFN-induced genes reveals new roles for cGAS in innate immunity. Nature 505, 691-5.

Schroder, K, and Tschopp, J The Inflammasomes. Cell 140, 821-832.

Shi, Z, Cai, Z, Sanchez, A, Zhang, T, Wen, S, Wang, J, Yang, J, Fu, S, and Zhang, D (2011): A novel Toll-like receptor that recognizes vesicular stomatitis virus. J Biol Chem 286, 4517-24.

Shibata, T, Ohto, U, Nomura, S, Kibata, K, Motoi, Y, Zhang, Y, Murakami, Y, Fukui, R, Ishimoto, T, Sano, S, Ito, T, Shimizu, T, and Miyake, K (2015): Guanosine and its modified derivatives are endogenous ligands for TLR7. Int Immunol.

Shinohara, ML, Lu, L, Bu, J, Werneck, MBF, Kobayashi, KS, Glimcher, LH, and Cantor, H (2006): Osteopontin expression is essential for interferon-α production by plasmacytoid dendritic cells. Nat Immunol 7, 498-506.

Shortman, K, and Naik, SH (2007): Steady-state and inflammatory dendritic-cell development. Nat Rev Immunol 7, 19-30.

Simas, JP, Bowden, RJ, Paige, V, and Efstathiou, S (1998): Four tRNA-like sequences and a serpin homologue encoded by murine gammaherpesvirus 68 are dispensable for lytic replication in vitro and latency in vivo. Journal of General Virology 79, 149-153.

Simon, MM, Greenaway, S, White, JK, Fuchs, H, Gailus-Durner, V, Wells, S, Sorg, T, Wong, K, Bedu, E, Cartwright, EJ, Dacquin, R, Djebali, S, Estabel, J, Graw, J, Ingham, NJ, Jackson, IJ, Lengeling, A, Mandillo, S, Marvel, J, Meziane, H, Preitner, F, Puk, O, Roux, M, Adams, DJ, Atkins, S, Ayadi, A, Becker, L, Blake, A, Brooker, D, Cater, H, Champy, M-F, Combe, R, Danecek, P, di Fenza, A, Gates, H, Gerdin, A-K, Golini, E, Hancock, JM, Hans, W, Hölter, SM, Hough, T, Jurdic, P, Keane, TM, Morgan, H, Müller, W, Neff, F, Nicholson, G, Pasche, B, Roberson, L-A, Rozman, J, Sanderson, M, Santos, L, Selloum, M, Shannon, C, Southwell, A, Tocchini-Valentini, GP, Vancollie, VE, Westerberg, H, Wurst, W, Zi, M, Yalcin, B, Ramirez-Solis, R, Steel, KP, Mallon, A-M, de Angelis, M, Herault, Y, and Brown, SD (2013): A comparative phenotypic and genomic analysis of C57BL/6J and C57BL/6N mouse strains. Genome Biology 14, 1-22.

Sinzger, C, Digel, M, and Jahn, G (2008): Cytomegalovirus cell tropism. Curr Top Microbiol Immunol 325, 63-83.

Skarnes, WC, Rosen, B, West, AP, Koutsourakis, M, Bushell, W, Iyer, V, Mujica, AO, Thomas, M, Harrow, J, Cox, T, Jackson, D, Severin, J, Biggs, P, Fu, J, Nefedov, M, de Jong, PJ, Stewart, AF, and

169

List of references

Bradley, A (2011): A conditional knockout resource for the genome-wide study of mouse gene function. Nature 474.

Smith, HRC, Heusel, JW, Mehta, IK, Kim, S, Dorner, BG, Naidenko, OV, Iizuka, K, Furukawa, H, Beckman, DL, Pingel, JT, Scalzo, AA, Fremont, DH, and Yokoyama, WM (2002): Recognition of a virus- encoded ligand by a natural killer cell activation receptor. Proceedings of the National Academy of Sciences 99, 8826-8831.

Smith, LM, McWhorter, AR, Masters, LL, Shellam, GR, and Redwood, AJ (2008): Laboratory strains of murine cytomegalovirus are genetically similar to but phenotypically distinct from wild strains of virus. J Virol 82, 6689-96.

Smith, MG (1954): Propagation of Salivary Gland Virus of the Mouse in Tissue Cultures. Experimental Biology and Medicine 86, 435-440.

Soulier, J, Grollet, L, Oksenhendler, E, Cacoub, P, Cazals-Hatem, D, Babinet, P, d'Agay, MF, Clauvel, JP, Raphael, M, Degos, L, and et al. (1995): Kaposi's sarcoma-associated herpesvirus-like DNA sequences in multicentric Castleman's disease. Blood 86, 1276-80.

Štiglincova, V, Chalupkova, A, Hrabovska, Z, ČIpkova, J, Wagnerova, M, and Mistrikova, J (2011): Vertical transmission of murine gammaherpesvirus 68 in mice. Acta Virologica 55, 55-59.

Sun, C, Schattgen, SA, Pisitkun, P, Jorgensen, JP, Hilterbrand, AT, Wang, LJ, West, JA, Hansen, K, Horan, KA, Jakobsen, MR, O'Hare, P, Adler, H, Sun, R, Ploegh, HL, Damania, B, Upton, JW, Fitzgerald, KA, and Paludan, SR (2015): Evasion of innate cytosolic DNA sensing by a gammaherpesvirus facilitates establishment of latent infection. J Immunol 194, 1819-31.

Sunil-Chandra, NP, Arno, J, Fazakerley, J, and Nash, AA (1994): Lymphoproliferative disease in mice infected with murine gammaherpesvirus 68. Am J Pathol 145, 818-26.

Sunil-Chandra, NP, Efstathiou, S, Arno, J, and Nash, AA (1992a): Virological and pathological features of mice infected with murine gammaherpesvirus 68. Journal of General Virology 73, 2347-2356.

Sunil-Chandra, NP, Efstathiou, S, and Nash, AA (1992b): Murine gammaherpesvirus 68 establishes a latent infection in mouse B lymphocytes in vivo. Journal of General Virology 73, 3275-3279.

Swanson, EC, and Schleiss, MR (2013): Congenital cytomegalovirus infection: new prospects for prevention and therapy. Pediatr Clin North Am 60, 335-49.

Tabeta, K, Hoebe, K, Janssen, EM, Du, X, Georgel, P, Crozat, K, Mudd, S, Mann, N, Sovath, S, Goode, J, Shamel, L, Herskovits, AA, Portnoy, DA, Cooke, M, Tarantino, LM, Wiltshire, T, Steinberg, BE, Grinstein, S, and Beutler, B (2006): The Unc93b1 mutation 3d disrupts exogenous antigen presentation and signaling via Toll-like receptors 3, 7 and 9. Nat Immunol 7, 156-64.

Takeda, K, Kaisho, T, and Akira, S (2003): Toll-like receptors. Annu Rev Immunol 21, 335-76.

Takeuchi, O, and Akira, S (2010): Pattern Recognition Receptors and Inflammation. Cell 140, 805-820.

Takeuchi, O, Kaufmann, A, Grote, K, Kawai, T, Hoshino, K, Morr, M, Mühlradt, PF, and Akira, S (2000): Cutting Edge: Preferentially the R-Stereoisomer of the Mycoplasmal Lipopeptide Macrophage-Activating Lipopeptide-2 Activates Immune Cells Through a Toll-Like Receptor 2- and MyD88-Dependent Signaling Pathway. The Journal of Immunology 164, 554-557.

170

List of references

Takeuchi, O, Kawai, T, Mühlradt, PF, Morr, M, Radolf, JD, Zychlinsky, A, Takeda, K, and Akira, S (2001): Discrimination of bacterial lipoproteins by Toll-like receptor 6. Int Immunol 13, 933-940.

Takeuchi, O, Sato, S, Horiuchi, T, Hoshino, K, Takeda, K, Dong, Z, Modlin, RL, and Akira, S (2002): Cutting edge: role of Toll-like receptor 1 in mediating immune response to microbial lipoproteins. J Immunol 169, 10-4.

Telfer, S, Bennett, M, Carslake, D, Helyar, S, and Begon, M (2007): The dynamics of murid gammaherpesvirus 4 within wild, sympatric populations of bank voles and wood mice. J Wildl Dis 43, 32- 9.

Terry, LA, Stewart, JP, Nash, AA, and Fazakerley, JK (2000): Murine gammaherpesvirus-68 infection of and persistence in the central nervous system. Journal of General Virology 81, 2635-2643.

Teterina, A, Richter, D, Matuschka, FR, Ehlers, B, and Voigt, S (2009): Identification of a novel betaherpesvirus in Mus musculus. Virol J 6, 225.

Titus, RG, Bishop, JV, and Mejia, JS (2006): The immunomodulatory factors of arthropod saliva and the potential for these factors to serve as vaccine targets to prevent pathogen transmission. Parasite Immunol 28, 131-41.

Torcia, MG, Nencioni, L, Clemente, AM, Civitelli, L, Celestino, I, Limongi, D, Fadigati, G, Perissi, E, Cozzolino, F, Garaci, E, and Palamara, AT (2012): Sex differences in the response to viral infections: TLR8 and TLR9 ligand stimulation induce higher IL10 production in males. PLoS One 7, e39853.

Tseng, PH, Matsuzawa, A, Zhang, W, Mino, T, Vignali, DAA, and Karin, M (2010): Different modes of TRAF3 ubiquitination selectively activate type I interferon and proinflammatory cytokine expression. Nat Immunol 11, 70-5.

Tsutsui, Y, Kosugi, I, and Kawasaki, H (2005): Neuropathogenesis in cytomegalovirus infection: indication of the mechanisms using mouse models. Rev Med Virol 15, 327-45.

Usherwood, EJ, Stewart, JP, Robertson, K, Allen, DJ, and Nash, AA (1996): Absence of splenic latency in murine gammaherpesvirus 68-infected B cell-deficient mice. Journal of General Virology 77, 2819-2825.

Valenzuela, DM, Murphy, AJ, Frendewey, D, Gale, NW, Economides, AN, Auerbach, W, Poueymirou, WT, Adams, NC, Rojas, J, Yasenchak, J, Chernomorsky, R, Boucher, M, Elsasser, AL, Esau, L, Zheng, J, Griffiths, JA, Wang, X, Su, H, Xue, Y, Dominguez, MG, Noguera, I, Torres, R, Macdonald, LE, Stewart, AF, DeChiara, TM, and Yancopoulos, GD (2003): High-throughput engineering of the mouse genome coupled with high-resolution expression analysis. Nat Biotechnol 21, 652-9.

Vicetti Miguel, RD, Sheridan, BS, Harvey, SA, Schreiner, RS, Hendricks, RL, and Cherpes, TL (2010): 17- beta estradiol promotion of herpes simplex virus type 1 reactivation is estrogen receptor dependent. J Virol 84, 565-72.

Vidy, A, Sacher, T, Adler, H, Jordan, S, Koszinowski, UH, and Ruzsics, Z (2013): Systemic and Local Infection Routes Govern Different Cellular Dissemination Pathways during Gammaherpesvirus Infection In Vivo. J Virol 87, 4596-4608.

Vieira, J, Huang, ML, Koelle, DM, and Corey, L (1997): Transmissible Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) in saliva of men with a history of Kaposi's sarcoma. J Virol 71, 7083-7.

171

List of references

Vremec, D, Pooley, J, Hochrein, H, Wu, L, and Shortman, K (2000): CD4 and CD8 Expression by Dendritic Cell Subtypes in Mouse Thymus and Spleen. The Journal of Immunology 164, 2978-2986.

Vremec, D, Zorbas, M, Scollay, R, Saunders, DJ, Ardavin, CF, Wu, L, and Shortman, K (1992): The surface phenotype of dendritic cells purified from mouse thymus and spleen: investigation of the CD8 expression by a subpopulation of dendritic cells. J Exp Med 176, 47-58.

Waggoner, SN, Daniels, KA, and Welsh, RM (2014): Therapeutic depletion of natural killer cells controls persistent infection. J Virol 88, 1953-60.

Wakeham, K, Webb, EL, Sebina, I, Muhangi, L, Miley, W, Johnson, WT, Ndibazza, J, Elliott, AM, Whitby, D, and Newton, R (2011): Parasite infection is associated with Kaposi's sarcoma associated herpesvirus (KSHV) in Ugandan women. Infectious Agents and Cancer 6, 1-7.

Wang, G-H, Garvey, TL, and Cohen, JI (1999): The murine gammaherpesvirus-68 M11 protein inhibits Fas- and TNF- induced apoptosis. Journal of General Virology 80, 2737-2740.

Wang, HY, and Wang, R-F (2007): Regulatory T cells and cancer. Curr Opin Immunol 19, 217-223.

Wang, JP, Bowen, GN, Zhou, S, Cerny, A, Zacharia, A, Knipe, DM, Finberg, RW, and Kurt-Jones, EA (2012): Role of specific innate immune responses in herpes simplex virus infection of the central nervous system. J Virol 86, 2273-81.

Weck, KE, Kim, SS, Virgin, HW, and Speck, SH (1999): Macrophages Are the Major Reservoir of Latent Murine Gammaherpesvirus 68 in Peritoneal Cells. J Virol 73, 3273-3283.

Weinberg, A, Bloch, KC, Li, S, Tang, YW, Palmer, M, and Tyler, KL (2005): Dual infections of the central nervous system with Epstein-Barr virus. J Infect Dis 191, 234-7.

Weinberg, JB, Lutzke, ML, Alfinito, R, and Rochford, R (2004): Mouse strain differences in the chemokine response to acute lung infection with a murine gammaherpesvirus. Viral Immunol 17, 69-77.

West, J, and Damania, B (2008): Upregulation of the TLR3 pathway by Kaposi's sarcoma-associated herpesvirus during primary infection. J Virol 82, 5440-9.

Wilkinson, D, Sheldon, J, Gilks, CF, and Schulz, TF (1999): Prevalence of infection with human herpesvirus 8/Kaposi's sarcoma herpesvirus in rural South Africa. S Afr Med J 89, 554-7.

Wooten, RM, Ma, Y, Yoder, RA, Brown, JP, Weis, JH, Zachary, JF, Kirschning, CJ, and Weis, JJ (2002): Toll-like receptor 2 is required for innate, but not acquired, host defense to Borrelia burgdorferi. J Immunol 168, 348-55.

Yamamoto, M, Sato, S, Hemmi, H, Hoshino, K, Kaisho, T, Sanjo, H, Takeuchi, O, Sugiyama, M, Okabe, M, Takeda, K, and Akira, S (2003a): Role of Adaptor TRIF in the MyD88-Independent Toll-Like Receptor Signaling Pathway. Science 301, 640-643.

Yamamoto, M, Sato, S, Hemmi, H, Sanjo, H, Uematsu, S, Kaisho, T, Hoshino, K, Takeuchi, O, Kobayashi, M, Fujita, T, Takeda, K, and Akira, S (2002a): Essential role for TIRAP in activation of the signalling cascade shared by TLR2 and TLR4. Nature 420, 324-9.

172

List of references

Yamamoto, M, Sato, S, Hemmi, H, Uematsu, S, Hoshino, K, Kaisho, T, Takeuchi, O, Takeda, K, and Akira, S (2003b): TRAM is specifically involved in the Toll-like receptor 4-mediated MyD88-independent signaling pathway. Nat Immunol 4, 1144-50.

Yamamoto, M, Sato, S, Mori, K, Hoshino, K, Takeuchi, O, Takeda, K, and Akira, S (2002b): Cutting edge: a novel Toll/IL-1 receptor domain-containing adapter that preferentially activates the IFN-beta promoter in the Toll-like receptor signaling. J Immunol 169, 6668-72.

Yamamoto, M, and Takeda, K (2010): Current Views of Toll-Like Receptor Signaling Pathways. Gastroenterology Research and Practice 2010, 8.

Yang, H, Bell, TA, Churchill, GA, and Pardo-Manuel de Villena, F (2007): On the subspecific origin of the laboratory mouse. Nat Genet 39, 1100-7.

Zauner, L, Melroe, GT, Sigrist, JA, Rechsteiner, MP, Dorner, M, Arnold, M, Berger, C, Bernasconi, M, Schaefer, BW, Speck, RF, and Nadal, D (2010): TLR9 triggering in Burkitt's lymphoma cell lines suppresses the EBV BZLF1 transcription via histone modification. Oncogene 29, 4588-98.

Zhan, Y, Vega-Ramos, J, Carrington, EM, Villadangos, JA, Lew, AM, and Xu, Y (2012): The inflammatory cytokine, GM-CSF, alters the developmental outcome of murine dendritic cells. Eur J Immunol 42, 2889- 900.

Zhang, D, Zhang, G, Hayden, MS, Greenblatt, MB, Bussey, C, Flavell, RA, and Ghosh, S (2004): A toll- like receptor that prevents infection by uropathogenic bacteria. Science 303, 1522-6.

Zucchini, N, Bessou, G, Traub, S, Robbins, SH, Uematsu, S, Akira, S, Alexopoulou, L, and Dalod, M (2008): Cutting Edge: Overlapping Functions of TLR7 and TLR9 for Innate Defense against a Herpesvirus Infection. The Journal of Immunology 180, 5799-5803.

173

Appendix

Figure 43: Plasmid map for pGEMT-T-Mmus Beta-2-microglobulin

174

Appendix

Figure 44: Plasmid map for pGEM-T-MCMV1-gB

175

Appendix

Figure 45: Plasmid map for pGEM-T-MCMV2-gB

176

Appendix

Figure 46: Plasmid map for for pGEM-T-MmusRHV1-gB

177

Acknowledgements

Firstly, I would like to express my sincere gratitude to my advisor Prof. Melanie M. Brinkmann for her continuous support during the course of my PhD. Thank you, Melanie, for giving me the opportunity to not only complete an essential part of my scientific journey in your lab, but for providing a platform for immense learning opportunities that will serve as a stepping stone for the future.

I am extremely grateful to my examiners for finding time to review my thesis.

I cannot express enough thanks to my thesis committee members: Prof. Dr. Thomas Schulz, Prof. Dr. André Bleich, Dr. Bernhard Ehlers and Dr Matthias Ottinger for their advice during my research. Special thanks to the international graduate school for partially funding my PhD and providing constant support.

I would like to thank Dr. Matthias Ottinger and Dr. Kendra A. Bussey for supporting my experimental work during the course of my PhD. It was a pleasure to work with you during the course of these two very special projects. I will always be grateful for the excellent troubleshooting and guidance provided throughout my experimental work and the writing process. I would also like to thank Kirsten M. Kleemann for her help with the Horst mice.

Special thanks go out to my colleagues at VIMM. Christine, Georg and Leon, thank you for your immense technical help during the course of my project. Baca, thank you for your help with the i.v. experiments. Margit, Elisa, Ulrike, Markus and Brigitte, thank you for all the shared lunches and laughter. Vladimir and Helene, a small note in my acknowledgement section is not enough to express my gratitude. I really couldn’t have done it without your support. All in all, it was an enormous pleasure working with all of you.

My stay in Braunschweig was brightened by many wonderful people. Aniruddha, Arne, Björn, Inga, Keerthi, Marco, Melissa, Nadiia, Niharika, René, Sonia and Shiwani, thank you! (It’s alphabetical! Don’t worry.) Your encouragement when times got rough are truly appreciated and duly noted.

A special shout out to Martin at Vapiano and the people from Asia Imbiss. The countless times I have returned to have tasty meals during my hectic schedule here will not be forgotten.

Needless to say, I wouldn’t have gotten here without the immense support and encouragement from my family. Thanks to my parents, my sister and her family for being there by my side even while being separated by continents.

179

Erklärung

Hiermit erkläre ich, dass ich die Dissertation „Detection of murine herpesvirus 68 by the innate immune system and studies on Mus musculus rhadinovirus 1 in its natural host“ selbstständig verfasst habe. Die Experimente zu dem Thema „Detection of murine herpesvirus 68 by the innate immune system“ wurden gemeinsam mit Kendra Bussey, Postdoktorandin in der Arbeitsgruppe Brinkmann am Helmholtz Zentrum für Infektionsforschung in Braunschweig, durchgeführt.

Ich habe keine entgeltliche Hilfe von Vermittlungs- bzw. Beratungsdiensten in Anspruch genommen. Niemand hat von mir unmittelbar oder mittelbar entgeltliche Leistungen für Arbeiten erhalten, die im Zusammenhang mit dem Inhalt der vorgelegten Dissertation stehen. Ich habe die Dissertation am Helmholtz Zentrum für Infektionsforschung in Braunschweig angefertigt.

Die Dissertation wurde bisher nicht für eine Prüfung oder Promotion oder für einen ähnlichen Zweck zur Beurteilung eingereicht.

Ich versichere, dass ich die vorstehenden Angaben nach bestem Wissen vollständig und der Wahrheit entsprechend gemacht habe.

Berlin, den 27.10.2016

180