DISSERTATION ROLE OF ENDOGENOUS RETROVIRUS IN CONTROL OF FELINE LEUKEMIA VIRUS INFECTION AND IMPLICATIONS FOR CROSS SPECIES TRANSMISSION Submitted by Elliott S. Chiu Department of Microbiology, Immunology, and Pathology In partial fulfillment of the requirements For the Degree of Doctor of Philosophy Colorado State University Fort Collins, Colorado Summer 2019 Doctoral Committee: Advisor: Sue VandeWoude Edward A. Hoover Gregory D. Ebel W. Chris Funk Copyright by Elliott Sinming Chiu 2019 All Rights Reserved ii ABSTRACT ROLE OF ENDOGENOUS RETROVIRUS IN CONTROL OF FELINE LEUKEMIA VIRUS INFECTION AND IMPLICATIONS FOR CROSS SPECIES TRANSMISSION Endogenous retroviruses (ERV) are markers of ancient retroviral infections, though evolutionary forces have limited the capacity for ERV replication and virulence. While they are seldom considered infectious alone, they maintain the ability to interact with their exogenous retroviral (XRV) progenitors. In Chapter One, we review the interactions that exist between ERV and XRV dyads. One such couplet includes feline leukemia virus (FeLV), a common domestic cat pathogen. In Chapter Two, we review FeLV subgroup taxonomy and the methods used from which they were originally characterized. Though the domestic cat is regarded as the natural host for the virus, recent reports have documented FeLV infections in wild felids with pathogenic consequences. Chapter Three examines the root of a contemporary FeLV outbreak in Florida panthers (Puma concolor coryi), a species that lacks endogenous FeLV. Our phylogenetic analysis of the contemporary FeLV outbreak has further implicated domestic cats (Felis catus) as the origin of FeLV infections in wild felids. Furthermore, we detected a recombinant oncogenic variant in Florida panthers that is believed to be non-horizontally transmissible. These field studies have prompted us to examine the cellular basis of infection and intrinsic resistance to the virus. In Chapter Four, we interrogate the cellular basis of FeLV infections between puma (P. concolor) and domestic cat cells using in vitro approaches. We demonstrated that puma cells support greater infection and replication. Additionally, we documented enFeLV long terminal repeats (LTR) in domestic cats are negatively correlated to FeLV infection outcomes in vitro. Natural FeLV infections in both Florida panther and domestic cat tissues offered us the opportunity to examine end stage disease dynamics, which demonstrate that Florida panthers have the ability to produce more virus despite having lower proviral loads than domestic cats. The results of both in vivo and in vitro experiments prompted us to further iii investigate enFeLV-LTRs and their role in FeLV infection. Chapter Five took advantage of the publicly available data in the NCBI Sequence Read Archive (SRA) to evaluate enFeLV-LTR basal transcription levels. Data-mining the domestic cat transcriptome showed that lymphoid cells, which are relatively resistant to in vitro FeLV infection, transcribe more enFeLV elements than relatively susceptible cells (i.e., fibroblasts). We also identified microRNA transcripts are produced that have the potential ability to down-regulate FeLV RNA transcripts. In Chapter Six, we innovated a new methodology to characterize the enFeLV-LTR integration sites across the entire genome of 20 related and unrelated domestic cats in an attempt to uncover genes that may be influenced by LTR enhancement of gene expression. We found one LTR integration site in a limited number of cats that is within 1MB of APOBEC1, an antiviral gene, and that the most common gene found in close proximity to LTR integration sites are zinc fingers, a broad-acting class of regulatory proteins. Collectively, this groundwork provides future directions to uncover direct and indirect mechanisms of enFeLV-mediated restriction of FeLV infection. We conclude that because wild felids lack enFeLV, they may be more vulnerable to FeLV infection. As urbanization forces niche overlap and contact between wild and domestic felids, the risk of infection of these species is likely to increase, and thus it will be important to consider contacts between FeLV-infected domestic cats and wild felid populations during development of conservation action plans. iv ACKNOWLEDGEMENTS This work could not have been completed without the intellectual contributions and emotional support from an ever-growing cadre of collaborators. They have helped me immeasurably by providing samples, developing protocols in addition to reviewing and providing constructive criticism to my experimental approaches and writing. Below is a list of collaborators who contributed to my work separated by chapter. Chapter 1 – Roderick Gagne, Sue VandeWoude Chapter 2 – Edward Hoover, Sandra Quackenbush, Sue VandeWoude Chapter 3 –Simona Kraberger, Mark Cunningham, Lara Cusack, Dave Onorato, Melody Roelke, Sue VandeWoude Chapter 4 – Ryan Troyer, Ryan Mackie, Lisa Wolfe, Karen Fox, Mark Fisher, Ivy Levan, Jennifer Malmberg, Gary Mason, Alex Byas, Laura Hoon-Hanks, Lauren Harris, Devin von Stade, Esther Musselman, Mary Nehring, Craig Miller, Melody Roelke, Sue VandeWoude Chapter 5 – Matthew Moxcey, Mark Stenglein, Justin Lee, Sue VandeWoude Chapter 6 – Roderick Gagne, Matthew Moxcey, Mark Stenglein, Justin Lee, Susan Bailey, Miles McKenna, David Maranon, Sue VandeWoude Finally, I would like to thank my friends, family, lab-mates, and DVM/PhD colleagues whose encouragement I have leaned on every step of the way. I would not be at this point in my professional and personal development if it were not for them and am privileged to have them in my life. v TABLE OF CONTENTS ABSTRACT ………………………………………………………………………………………………. ii ACKNOWLEDGEMENTS ……………………………………………………………………………… iv CHAPTER ONE: Interactions Between Endogenous Retroviruses and Their Exogenous Counterparts … 1 Figures ……………………………………………………………………………………………….. 19 LITERATURE CITED…………………………………………………………………………………… 22 CHAPTER TWO: A Retrospective Examination of Feline Leukemia Subgroup Characterization: Viral Interference Assays to Deep Sequencing ………………………………………………………………... 31 Figures ………………………………………………………………………………………………. 42 LITERATURE CITED…………………………………………………………………………………… 45 CHAPTER THREE: Multiple Introductions of Domestic Cat (Felis catus) Feline Leukemia Virus in Endangered Florida Panthers (Puma concolor coryi)……………………………………………………. 52 Introduction ………………………………………………………………………………………….. 52 Methods ……………………………………………………………………………………………… 54 Results ……………………………………………………………………………………………….. 57 Discussion …………………………………………………………………………………………... 59 Figures ………………………………………………………………………………………………. 63 Tables ……………………………………………………………………………………………….. 70 LITERATURE CITED…………………………………………………………………………………… 73 CHAPTER FOUR: Comparisons of feline leukemia virus (FeLV) Susceptibility of puma (Puma concolor) and domestic cat (Felis catus) cells in relation to endogenous FeLV elements ……………… 76 Introduction ………………………………………………………………………………………….. 76 Methods ……………………………………………………………………………………………… 78 Results ……………………………………………………………………………………………….. 82 Discussion …………………………………………………………………………………………… 84 Figures ………………………………………………………………………………………………. 89 Tables ………………………………………………………………………………………………... 95 LITERATURE CITED…………………………………………………………………………………… 97 vi CHAPTER FIVE: Endogenous Feline Leukemia Virus (enFeLV) siRNA May Drive Direct Interference with FeLV Infection ……………………………………………………………………………………. 101 Introduction ………………………………………………………………………………………… 101 Methods ……………………………………………………………………………………………. 103 Results ……………………………………………………………………………………………… 104 Discussion …………………………………………………………………………………………. 106 Figures ……………………………………………………………………………………………… 111 Tables ………………………………………………………………………………………………. 117 LITERATURE CITED………………………………………………………………………………….. 119 CHAPTER SIX: Characterization of Endogenous Feline Leukemia Virus (enFeLV) Long Terminal Repeat (LTR) Integration Site Diversity ……………………………………………………………….. 121 Introduction ………………………………………………………………………………………… 121 Methods …………………………………………………………………………………………….. 122 Results ……………………………………………………………………………………………… 126 Discussion …………………………………………………………………………………………. 128 Figures ……………………………………………………………………………………………… 132 Tables ………………………………………………………………………………………………. 139 LITERATURE CITED………………………………………………………………………………….. 216 CONCLUSION ………………………………………………………………………………………… 218 1 CHAPTER ONE Interactions Between Endogenous Retroviruses and Their Exogenous Counterparts INTRODUCTION Animal genomes are riddled with evolutionary fossils left over from ancient viral infections; notable among these are remnants of retroviruses. Retroviruses require conversion of RNA genomes into DNA that stably integrate into host DNA (Coffin, 1992). Following integration, retroviruses hijack host cellular machinery for replication. In normal somatic cell infections, this process allows for the establishment of life long infections, as integrated viral genomes are not easily excised. In the special case of retroviral germline infections, integrated viral genomes may be passed along to progeny by vertical transmission (Figure 1). This results in permanent viral penetrance into the animal’s cellular genome, resulting in an endogenous retrovirus (ERV), in contrast to horizontally transmitted counterparts, referred to as exogenous retroviruses (XRV) (Coffin, 2004). Early on in infection, ERVs are identical to their XRV progenitors. At this stage, they may be transcribed and can be infectious. After generations, ERV mutations accumulate, mitigating deleterious impacts of fully functional viruses. Defective ERVs are selected through generations of inheritance and become fixed in host populations (Lober et al., 2018). This phenomenon is correlated with diminished virulence, marked by mutations and deletions of the exogenous virus. Through the process of retrotransposition, ERV genomes may also be translocated, or ‘retrotranspose’ to other loci, leaving behind solo-long terminal repeats (LTRs), which harbor enhancer and promoter functions (Britten, 1997) and other remnant ERVs (Boeke and Stoye, 1997). Once ERVs integrate and become fixed within a population, they can be regarded as markers of ancient retroviral infections (Coffin, 2004) and become integrated as a component of normal host genomic function. Analysis of retroviral-like sequences within host genomes has provided evidence that retroviruses were present during the Paleozoic Era (Aiewsakun and Katzourakis, 2017). Exogenous retroviruses that are responsible for seeding ERVs often become extinct, but in some cases, related 2 viruses still exist and interact with their endogenous counterparts in a variety of ways. Evaluation of exogenous-endogenous retroviral interactions provides a unique opportunity to witness host ‘evolution in action’, and document ways that viral infections can perpetuate through means other than horizontal transmission. In consideration of the interesting relationship between ERVs and their exogenous counterparts, this review will examine and catalogue reported interactions between endogenous and exogenous retroviruses in avian and mammalian hosts, predominantly domestic animals which account for most of the scientific literature in this area (Garcia-Etxebarria et al., 2014). This work will update a thorough review conducted in 1998 (Rasmussen, 1997) and build upon a recent review detailing the origin and evolutionary consequences of endogenous retroviruses (Johnson, 2019). Endogenous retroviruses in non- avian and non-mammalian hosts have only recently been documented (Xu et al., 2018), and there are likely many other species and systems where similar mechanisms have not yet been described. We will review the interactions between ERVs and their extant XRV counterparts. This includes ERV-XRV interactions in Jaagsiekte sheep retrovirus (JSRV), Rous sarcoma virus and avian sarcoma leukosis virus (RSV/ASLV), murine leukemia Virus (MMLV), mouse mammary tumor virus (MMTV), feline leukemia virus (FeLV), and koala retrovirus (KoRV) infections (Table 1). Understanding mechanisms by which ERV-XRV pairs interact provide a basis for study of ERVs that constitute approximately 8% of human genomes, relating potential interactions with past, present or future XRV epidemics. This is of particular importance as newly introduced retroviruses, such as HIV, have the significant potential of becoming endogenized. Beyond that, ERV-XRV interactions have and will continue to shape animal disease and biology. This review will not specifically discuss the role of endogenous retroviruses in placentation and mammalian evolution, normal biological processes, immune modulation, oncogenesis, and disease progression outside the scope of ERV interactions, which have all been thoroughly reviewed recently (Bannert et al., 2018; Denner, 2016; Frank and Feschotte, 2017; Grandi and Tramontano, 2018; Johnson, 2019; Morris et al., 2018). Readers are also referred to a recent review on the biology and early actions of endogenization (Greenwood et al., 2018). 3 NATURE OF ENDOGENOUS/EXOGENOUS RETROVIRAL INTERACTIONS Endogenous and exogenous retroviruses exist in a unique system in which a virally derived host genetic component has the potential to come in contact with a related transmissible virus. Due to their high nucleotide similarity, these two entities can often interact in predictable ways that transcend the normal paradigms of host immunity. As a primer to the specific interactions detailed below, ERV-XRV interactions can be classified in a number of different groups (Table 2). Briefly, recombination is commonly encountered as the retroviral life cycle establishes the conditions for two genetically related viral RNAs to be co-packaged in virions. ERVs have the capacity to modulate viral recognition by immune or host cells through receptor interference and immunological self-tolerance. Finally, ERVs have the capacity to restrict and/or promote viral restriction (occasionally simultaneously) using different cellular and molecular mechanisms. ROUS SARCOMA VIRUS/AVIAN SARCOMA LEUKOSIS VIRUS (RSV/ASLV; Alpharetrovirus) In 1910, Rous sarcoma virus became the first oncogenic virus that was described in any species after Dr. Peyton Rous demonstrated that cell-free filtrates from avian sarcomas were capable of transmitting sarcomas in chickens (Rous, 1911). RSV and ASLV nomenclature has evolved as different genetic elements and functions have been identified (Rubin, 2011). RSV is a replication-competent rapidly transforming virus closely related to the ASLV complex. It was not until 1953 that additional ASLVs were described by using focus formation assays. Initially these agents were called resistance- inducing factors (RIFs) due to their ability to restrict RSV infection (Rubin, 1955). This ultimately led to the identification of additional ASLV subgroups including subgroup E and J that comprise endogenous ASLV. Subgroups F through I are also endogenous ASLVs found in species other than the domesticated chicken (Gallus gallus) (Payne and Nair, 2012; Venugopal, 1999). The major difference between RSV and other members of the ASLV complex is that ASLVs lack the v-src gene that conveys the rapidly transforming properties of RSV (Swanstrom and Wills, 1997). LTR sequences from the endogenous ASLVs are also defective. While they contain motifs that resemble enhancer regions, these regions are 4 not complete and as a result, are insufficient for enhancer function (Habel et al., 1993). Barring these differences, RSV is indistinguishable from the ASLV complex viruses. Chickens have on average, 5 endogenous ASLV integration sites per individual with up to 2 dozen identified loci, though only one locus exists in the red jungle fowl (G. gallus wild ancestor) reference genome (Benkel and Rutherford, 2014). Certain chicken cell lines (i.e., line 0) important in the agricultural production industry were specifically developed to delete ERVs from the genome. As noted above, ASLV subgroup E loci harbor mutations that render the ERV incapable of generating infectious virus. One notable exception is loci ev-2, which is capable of producing Rous-associated virus-0 (RAV- 0), a non-oncogenic ASLV (Astrin et al., 1980). While not completely replication-defective, RAV-0 U3 LTR and surrounding segments vary from RSV, which is believed to account for a reduction in virulence (Hughes, 1982). Endogenous ASLV has been documented to both promote and protect against disease following exogenous virus RSV/ASLV infection. Interactions between RAV-0 and RSV were described following quail cell infections with non-producing foci RSV (RSV(O)). Japanese quail cells lack RAV-0, which limited replication of RSV(O), a phenomenon that was not observed in chicken cells that harbor RAV-0. Co-inoculation of RAV-0 with RSV(O) in Japanese quail cells rescued the ability for RSV replication, demonstrating its helper virus activity (Figure 2) (Vogt and Friis, 1971). Unsurprisingly, recombination occurs frequently between exogenous retroviruses and their related endogenous counterparts as nucleotide similarities predispose polymerases resuming stalled transcription on a different strand that resembles the original template (Luo and Taylor, 1990). These recombinants may increase virulence in many cases as mutations that have been accumulated in ERVs may be rescued by replacing the defunct sequence with functional protein-encoding sequences. ASLV subgroup E-RAV-60 recombinants were shown to induce lymphoid leukosis in response to exogenous virus-origin c regions in recombinant viruses (Crittenden et al., 1980). More recently, a rapidly evolving oncogenic subgroup (subgroup J) has been identified, and is believed to have emerged through recombination of a previously uncharacterized exogenous ASLV and an endogenous ERV distantly related to ASLVs (Sacco et al., 2004; Smith et al., 1999; Venugopal, 1999). 5 Interference between endogenous and exogenous ASLV/RSV is dependent ERV structure and age at time of infection. Chicken embryos inoculated with RAV-0 challenged with exogenous RAV-1 and RAV-2 (RSV pseudotype viruses) supported greater infection with RAV-1 or RAV-2, as measured length of viremia, or delayed or absent development of an adaptive immune response to the challenge virus. The mechanism proposed was that chicks were immunologically tolerant to envelope group-specific glycoproteins similar between both endogenous and exogenous viruses (Crittenden et al., 1987). Particular ev loci have since been identified to support this hypothesis. While some loci may exacerbate infection (e.g. ev12, ev21), others may work to help restrict immune tolerance (e.g. ev6, ev9); others still have no influence on ASLV infection (e.g., ev1, ev3) (Gavora et al., 1995; Kuhnlein et al., 1992; Smith et al., 1990). ev21 tolerance is most restrictive if the endogenous virus is genetically transmitted (versus experimentally horizontally transmitted) and if ASLV infection occurs prior to 4 weeks of age (Fadly and Smith, 1997). Specific endogenous integration sites have been characterized to protect against disease. Crittenden et al. demonstrated that variation in ev gene expression and the RAV strain inoculated influenced ALV shedding rates and horizontal transmission characteristics of the exogenous infection (Crittenden et al., 1984). They measured relationship between genomic variation at ev2 and ev3 loci versus infection with four strains of ALV (RAV-1, RAV-2, RPL-40, and RPL-42). Animals lacking both ev loci exhibited higher mortality from non-neoplastic syndromes following RAV-1 infection. Line-0 chickens lacking both ev loci (which were developed separately from ev2-/ev3- negative chickens described above) had lower viremia and higher antibody production following RAV-1 infection. Inexplicably, line-0 chickens did not develop non-neoplastic syndromes. The other three ALV strains tested were more variable, generating fewer non-neoplastic syndromes. ev-positive chickens developed less robust neutralizing antibody responses. In a separate experiment, Denesvre et al. indicated that Subgroup J ASLVs expression of endogenous ev/J 4.1Rb envelope protein completely interfered with exogenous infection (Denesvre et al., 2003). These studies provide strong evidence for immunological tolerance as a mechanism for ERV-XRV interference. 6 New developments in transcriptomic analysis has allowed for the investigation of endogenous ASLVs in light of disease progression and infection. Interestingly, ALV-E env expression is increased in embryonic fibroblasts infected with Mareck’s disease virus (herpesvirus), but is decreased in cells infected with ALV subgroup J viruses and reticuloendotheliosis virus (Hu et al., 2017; Hu et al., 2016). The significance of these results has yet to be determined, but indicate a possible role for ALV expression during development in generation of immunological tolerance to a variety of exogenous infections. JAAGSIEKTE SHEEP RETROVIRUS (JSRV; Betaretrovirus) Ovine pulmonary adenocarcinoma (OPA) was first recorded in domesticated sheep herds in the 1800’s; however, its etiological agent, JSRV, was not identified until 1995 (Palmarini et al., 1996; Sharp and De Martini, 2003). OPA, originally known as jaagsietke for the respiratory clinical symptoms in affected animals, is the cause of transmissible neoplastic disease that occurs in approximately 30% of JSRV-infected individuals. Prevalence of JSRV varies from 40-80% of sheep and is found rarely in goats. Much like other retroviruses that exist both exogenously and endogenously, early molecular detection was hampered by the presence of related endogenous entities due to their high similarity (Palmarini et al., 1996). Historical perspective of the natural history of JSRV in sheep and in depth examination of JSRV induced oncogenesis is reviewed in (Sharp and De Martini, 2003) and (Leroux et al., 2007). Fluorescent in situ hybridization and qPCR have identified approximately 30 copies of the endogenous virus in both sheep and goats (both domestic and wild breeds). Chromosomes 6 and 9 appear to harbor multiple copies of endogenous JSRV (enJSRV) (Carlson et al., 2003). Two chromosomal integration sites (1q45 and 2q41) are common between the two species (Carlson et al., 2003; Hecht et al., 1996; Palmarini et al., 2000). enJSRV appear to be necessary in placenta formation and are required for normal reproductive function in sheep (Dunlap et al., 2006), as reviewed in (Spencer and Palmarini, 2012a; Spencer and Palmarini, 2012b). While enJSRV and JSRV share 90-98% amino acid similarity (Palmarini et al., 2000) across the majority of its genome, there is extensive difference within U3 region that shares 56% sequence identity, 7 and is markedly longer in enJSRV (Bai et al., 1996). Aside from being useful as makers of differentiation by molecular methods, the differences in the LTR have been discovered to drive differential expression of JSRV and enJSRV and underlie cellular tropism. JSRV LTR is specifically active in lung epithelia through the recruitment of lung-specific transcription factors, including hepatocyte nuclear factor-3β and NF-κB. enJSRV LTR, on the other hand, contain regions that favor transcription in the reproductive tract relating to reproductive competence (McGee-Estrada and Fan, 2007; Palmarini et al., 2000). Unlike ASLV, interactions between enJSRV and JSRV described to date are predominately antagonistic, and recombination between the enJSRV and JSRV has not been documented. Expression of enJSRV during ontology of the embryo may induce immunologic tolerance to JSRV (DeMartini et al., 2003; Varela et al., 2009). Spencer et al., demonstrated transcription of enJSRV in developing fetal sheep thymuses, and showed that animals that developed OSA lacked virus-specific antibody responses (Spencer et al., 2003). There is evidence that enJSRV may mediate protection against the exogenous form of the virus both in early and late stages on infection. enJSRV restricts JSRV’s ability to enter the cell through receptor interference by saturating hyaluronidase-2, the entry receptor for JSRV, limiting the number of receptors that are expressed on the surface of the cell (Spencer et al., 2003). The action of one enJSRV ERV (enJS56A1) represents a unique restriction factor in the realm of ERV/XRV interactions. In the course of a normal JSRV (betaretrovirus) infection, JSRV Gag organizes around the microtubule organization center/pericentriolar region for viral assembly and budding. Whole provirus enJS56A1- expressed Gag co-assemble with JSRV Gag and these aggregates block normal intracellular trafficking of JSRV Gag, leading to reduced formation of mature virions (Arnaud et al., 2007b; Murcia et al., 2007). Aggregates that form in the cytoplasm are then degraded through proteasomal machinery(Arnaud et al., 2008). The amino acid residue responsible for centrosomal trafficking is residue 21, determined as mutated JSRV Gag at this site modifies this activity (Arnaud et al., 2007a). Wild-type (R21) and mutant (W21) variants of the Gag are found in normal individual sheep in a breed-specific manner (Viginier et al., 2012). 8 MOUSE MAMMARY TUMOR VIRUS (MMTV; Betaretrovirus) Mouse mammary tumor virus is an exogenous virus that is transmitted in the milk of lactating animals and as the name suggests, can result in mammary tissue tumorigenesis. The endogenous form of the virus (Mtv) is found at varying copy numbers between 2 and 8 copies in laboratory mouse strains. There are more than 30 different characterized Mtv. MMTV-Mtv interactions have been extensively reviewed in (Holt et al., 2013). To date, recombination between MMTV and Mtv has not been recorded. This section will briefly address what has been established as a comparison to other systems described in this review, and focus on updates since the most recent review (Holt et al., 2013). Indirect impacts of Mtv on MMTV infection relate to immunological alterations that impact MMTV infection parameters and disease progression non-specifically. With respect to direct MMTV-Mtv interactions, two main areas have been investigated in detail: tumorigenesis and Mtv superantigen (Sag) production as well as its impacts on subsequent infection. Immunological tolerance driven by immune response development in the presence of Mtv is perhaps the clearest evidence of this mechanism underlying ERV interference diminishing the clinical and virologic outcome of ERV infection. Both MMTV and Mtv encode for Sag, a type-2 transmembrane protein (Choi et al., 1991; Marrack et al., 1991). With respect to MMTV, Sag is a viral accessory protein encoded for by the 3’ LTR and is necessary for dissemination from infected gut-associated lymphoid tissue to mammary glands (Golovkina et al., 1992). Mouse lineages that are defective in processing MMTV Sag are thus resistant to MMTV infection and dissemination (Holt et al., 2013). Sag also mediates B-cell-T-cell interactions and immune cell activation (Held et al., 1993). Some endogenous Mtv delete specific T-cell Vβ through negative T-cell selection during thymus organogenesis, offering a protective advantage against MMTV infection (Holt et al., 2013). A library of transposable repetitive elements (TREome) of 56 individual laboratory mouse strains has been characterized, revealing high amounts of diversity among strain specific Sag coding sequences that may drive strain immune phenotypes (Lee et al., 2016). This has led to the characterization of strain-specific responses to stress and infection based on Mtv loci and genotypes (Hsu et al., 2017). In 2017, additional Mtv Sags have been implicated in the clonal deletion of specific T- 9 cell populations with variable regions Vβ5.1, Vβ6, Vβ8.1, Vβ8.2, Vβ9, and Vβ11 in NC/Nga mice, a model for atopic dermatitis (Ohkusu-Tsukada et al., 2017). While the direct restriction of MMTV through immune modulation following Mtv SAG production is possible since MMTV infects T-cells, it is possible that other systems undergo the same clonal deletion, but leads instead towards immune tolerance and increased virulence to the exogenous virus that are not lymphotropic such as bovine viral diarrhea virus (Bolin, 1995). Mtv has been implicated in tumorigenesis via Sag-dependent and Sag-independent mechanisms. For example, mammary tumor development has been shown to be dependent upon Mtv-1, 2, and 4 expression, where presence is associated with an increased incidence of mammary tumors, potentially through the up-regulation of host proto-oncogenes (Bruno et al., 2013; Imai et al., 1983; Matsuzawa et al., 1990; Van Nie and Vaerstraeten, 1975). Even though Sag-reactive T-cells may be deleted, in some cases Mtv-Sag engenders the conditions for tumorigenesis, by activating specific T-cell populations that support the development of certain neoplasms. Specifically, Mtv-29 Sag stimulates Vβ16 CD+ T cells preempting B cell lymphomagenesis (Sen et al., 2001). Mtv-knock out mouse strains infected with MMTV had decreased tumor development by 90% (Bhadra et al., 2006), and deletion of Mtv is associated with increased resistance to tumorigenesis by MMTV strains that encode ‘strong’ Sag. Thus, Mtv appears to increase pathogenicity of MMTV by increasing the likelihood of the development of tumors, however exact mechanisms are not yet known. MURINE LEUKEMIA VIRUS (MuLV; Gammaretrovirus) Much like MMTV, MuLVs has been thoroughly characterized in mouse strains used in biomedical research. The origin of endogenous MuLVs and the interactions that have been documented with exogenous MuLV have been extensively reviewed (Kozak, 2014). Ecotropic MuLVs (E-MuLVs) are found in both lab strains and wild mice, and are unable to infect other species. Polytropic MuLVs (P- MuLVs/MCF-MuLV) are found in lab strains and wild mice that also have the ability to infect a limited host range group in vitro, including human and mink cells. Xenotropic MuLVs (X-MuLVs) have not been 10 identified in lab strains, but are found in wild mice and have a very broad host range that include human, rabbits, cat, bat, and dog cells (Kozak, 2014). Endogenous MuLVs from all three subgroups have been documented in C57BL mouse genomes (>50 loci identified) and other mouse strains have diverse ERV MuLV genotypes at a wide variety of integration sites (Kozak, 2014). Endogenous MuLV is capable of producing infectious MuLV, which is a relatively unique property among ERVs. The vast majority of endogenous E-MuLVs (Emv) are produced from fully functional MuLVs with few mutations resulting in defective replication (Jenkins et al., 1982). Some laboratory mouse strains such as NZB and F/St also have the ability to produce infectious X-MuLVs from their endogenous X-MuLV (Xrv) (Kozak, 2014). As such, MuLV viral particles may exist simultaneously as endogenous and exogenous retroviruses. In contrast, endogenous P-MuLVs (Pmvs) are incapable of producing infectious virus, despite having coding regions with open reading frames (Jern et al., 2005; Kozak, 2014). The transcriptional activity of many endogenous MuLVs has resulted in high levels of recombination between viral strains, and this has become a very important part of their biology and relates to disease outcoms. While Pmvs do not generate viruses alone, they may still generate transmissible P-MuLV when opportunistically packaged along with E-MuLVs (Evans et al., 2009). Transmission of P-MuLV may induce leukemia through insertional mutagenesis following activation of Myc or deactivation of Trp53 (Kozak, 2014). Infectious P-MuLVs appear to be generated de novo in infected mice and do not appear to be horizontally transmitted. Mice that do not generate P-MuLVs do not develop viral-induced lymphomas or leukemia (Kozak, 2014). Other examples of MuLV recombination result in other neoplastic syndromes. A recombination event has been documented in xenografts of human tumors passaged in immunosuppressed mice; these recombinant viruses may result in neoplasms on their own (Kozak, 2014). Historically, recombination of P-MuLV was believed to occur primarily between Pmv and E- MuLV within env and LTR sequences. Recently, additional sequencing targeting the origins of pathogenic and nonpathogenic recombinant MuLVs has shown that recombination occurs across the entire E-MuLV genome, although recombination in gag is limited due to host antiviral restriction factors. 11 Pathogenicity was not linked with specific changes to LTR, yet, all pathogenic P-MuLV LTRs had undergone recombination, duplication, or mutation (Bamunusinghe et al., 2017). Beyond recombination, MuLV ERVs have been implicated in restriction against exogenous MuLV infection. This has been documented through a number of different mechanisms. Fv1, the first antiviral host restriction factor described, is related to the gag gene of a murine endogenous retrovirus (Young et al., 2018). While it only shares 43% amino acid identity to its closest ERV relative, it has been shown to target MuLV capsid proteins, inhibiting viral replication (Bénit et al., 1997). Specific alleles for Fv1 display differential abilities to restrict different MuLV subgroups. When exposed at physiologic expression levels in cell culture, Fv1 n and Fv1 b restrict B-tropic MuLV and N-tropic MuLV, respectively. When concentrations of these gene products are increased above physiologic expression levels, Fv1 b gains the ability to partially restrict B-tropic MuLV, whereas Fv1 n restriction profiles remain the same regardless of expression level (Li et al., 2016). This may not be biologically significant, but points at the need to consider individual loci and alleles of endogenous retroviral elements and the extent that these genes are transcribed with respect to their antiviral restriction capabilities. Specific MuLV ERV env genes have been co-opted by the host to function as restriction factors against retroviruses, including MuLV. A number of restriction genes (e.g., Fv-4, Rmcf, Rmcf2) have MuLV origins and restrict both E-MuLV and P-MuLV infection (Kozak, 2014). Other endogenous MuLV derived structures (i.e., X-MuLV solo LTR) modulate host gene expression. For example, X- MuLV LTR insertions in the mouse APOBEC3 (mA3) intron increase mA3 gene expression, leading to restricted viral replication (Sanville et al., 2010). APOBEC3 is a potent antiviral restriction factor that acts as a cytidine deaminase that interferes with viral replication and this interaction represents one example in which a specific LTR integration site indirectly restricts viral infection through host gene exaptation (Harris and Dudley, 2015). Recent investigations have characterized the diversity and interrelationship of endogenous MuLVs in laboratory and wild mice, (Bamunusinghe et al., 2016; Lee et al., 2017). Next generation sequencing techniques have revealed the penetrance of endogenous MuLVs in wild and domesticated 12 Mus musculus (Hartmann et al., 2015). These analyses have revealed a high degree of variation in the location and copy number of endogenous MuLVs across and even within individuals (Lee et al., 2017). Variation in endogenous MuLVs across tissues in single individuals indicated that some MuLV ERVs are transcriptionally active and maintain the ability to replicate infectious virions in a pattern that is strain- specific. Friend MuLV infection has been documented to mobilize endogenous MuLVs (Boi et al., 2016). Endogenous MuLV transcripts increase one-day post infection with Friend-MuLV, and transcription remains higher throughout the course of infection. It is hypothesized that this process may represent an increased risk of recombination which contributes to pathogenesis and oncogenesis (Boi et al., 2016). This discovery is similar to previous investigations of Mtv viral infections; following certain viral infections, Mtv expression is increased, while in other cases, Mtv expression is decreased. FELINE LEUKEMIA VIRUS (FeLV; Gammaretrovirus) Feline leukemia virus (FeLV) is a common pathogenic retrovirus that is capable of causing fulminant disease in domestic cats (Hartmann, 2012) originally discovered in a cluster of group housed cats that developed feline malignant lymphoma (Schneinder et al., 1967). Endogenous FeLV (enFeLV) is in genomes of domestic cats (F. catus) and related small felids (Felis spp.) at variable copy numbers (8– 12 copies per haploid genome and up to 19 per diploid genome), but does not exist in other Felidae (Chiu et al., 2018; Polani et al., 2010). While the genetics and mechanisms of interactions between enFeLV and its exogenous counterpart are not well understood, we perhaps know the most about the clinical outcomes of exogenous FeLV infection on its host. The availability of several effective vaccines, and the ability of cats to ‘self-cure’ from infection are unique aspects of FeLV infection that could be exploited to elucidate better understanding of exogenous-endogenous retroviral interactions and the role of host immunity in controlling retroviral disease. The group of viruses known as FeLV is actually made up of 6 recognized subgroups in addition to an endogenous form of the virus (enFeLV) (Chiu et al., 2018) . FeLV is found worldwide in feral and 13 outdoor cats at prevalence of 3-18% (Bandecchi et al., 2006; Gleich et al., 2009; Muirden, 2002; Yilmaz and Ilgaz, 2000). The most common exogenous subgroup of the virus (FeLV-A) is believed to be the only transmissible subgroup and is found in nearly 100% of FeLV clinical cases (Jarrett et al., 1978; Jarrett and Russel, 1978). enFeLV has long been thought to interact with its exogenous counterparts resulting in recombination, oncogenesis, or interference (Polani et al., 2010). Approximately 50% of cats that are infected with FeLV-A develop a recombinant form (FeLV-B) that arises from recombination between FeLV-A and enFeLV env and 3’-LTR regions (Powers et al., 2018). This recombination event leads to a change in cellular tropism as well as accelerated disease progression and formation of lymphoid and other tumors (Anderson et al., 2001; Bechtel et al., 1999). FeLV-B is known as an oncogenic variant, leading to lymphosarcoma in cats that are infected with the recombinant due to insertional mutagenesis (Fujino et al., 2008). FeLV-B is believed to primarily arise in individual hosts versus being spread by horizontal transmissible in domestic cats, though there have been two FeLV-B cases in domestic cats documented in the absence of FeLV-A infection (Stewart et al., 2013). FeLV-B has also been detected in the endangered Florida panthers that lack enFeLV sequences, the subject of Chapter 3 of my PhD studies, (Chiu et al., 2019), and was also potentially detected in a jaguar (Silva et al., 2016). While enFeLV is responsible for generating virulent FeLV-B, it also has been hypothesized to provide protection against horizontal transmission. One hypothesis for restricted FeLV-B horizontal transmission is the receptor interference. As enFeLV imparts a full or partial env gene to FeLV-B, the envelope proteins produced are very similar, and expression of defective enFeLV-env may neutralize FeLV-B receptors (McDougall et al., 1994), as has been described in JSRV. While recombination and receptor neutralization have been documented prior to the early 2000’s, mystery still surrounds how enFeLV may interact with FeLV-A infection. Given the fact that experimental FeLV infection can result in a wide range of disease outcomes, the possibility of immune self-tolerance has been hypothesized, but has not yet been proven (Charreyre and Pedersen, 1991). Thorough examination of infection outcomes has defined at least 4 types of infection – persistent, regressive, latent, and abortive (Torres et al., 2005). Experimental infection of SPF cats has previously 14 shown that higher enFeLV proviral copy numbers bolster FeLV viral replication (Tandon et al., 2008); however, a natural infection of a colony of domestic cat-Asian leopard cat (Prionailurus bengalensis) backcrossed hybrids demonstrated the opposite, where higher enFeLV proviral led to lower infection and viral replication (Powers et al., 2018). Do puma cells lacking enFeLV support greater FeLV infection when compared to domestic cat cells? This is the question posed in Chapter 4 of this dissertation. We discovered that FeLV infection displayed a negative correlation with increasing enFeLV copy number, suggesting a protective capacity of enFeLV in light of FeLV-A infection. When primary PBMCs (high levels of enFeLV expression) were infected with exogenous FeLV-A, they supported less infection that fibroblasts, which transcribe 10-fold less enFeLV. We confirmed that enFeLV transcription is tissue-specific and enFeLV transcription is highest in lymphoid tissues. We are in the process of testing two hypotheses in Chapters 5 and 6 to explain these observations: 1) enFeLV transcription interacts in a virus specific manner due to direct interference not yet described in other ERV-XRV systems, and/or, 2) solo-LTRs may be positioned near anti-viral host genes that restrict FeLV infection indirectly, as has been documented in MuLV. KOALA RETROVIRUS (KoRV; Gammaretrovirus) Koala retrovirus represents the most recent ERV described in the literature and is currently undergoing the process of endogenization, with earliest estimation of invasion set at 22,200-49,900 years (Ishida et al., 2015). As such, it has been studied as a model to understand how early retroviral endogenization impacts host biology (Greenwood et al., 2018; Lober et al., 2018). Koala retrovirus was first associated in a case of spontaneous leukemia in a captive koala (Canfield et al., 1988) and identified as a gammaretrovirus closely related to gibbon ape leukemia virus present in koala populations dating back to the 19 th century (Hanger et al., 2000). While the virus has endogenized in koala populations in Northern Australia, the virus has not entirely penetrated the koala populations in South Australia. Prevalence is reported at 14.8% in Kangaroo Island koalas and is apparently absent in Phillip Island koalas (Simmons et al., 2012). To date, 10 subgroups of KoRV (A-J) have been identified, although only 15 one exists as an ERV (KoRV-A) with 39-133 distinct loci described (Chappell et al., 2017; Hobbs et al., 2017; Ishida et al., 2015; Shojima et al., 2013; Xu et al., 2015; Xu et al., 2013). Much like MuLV, KoRV- A as a newly endogenizing retrovirus behaves like an exogenous virus in that it maintains the ability to generate infectious virions and can be transmitted horizontally; however, it has been confirmed to be endogenous as germ cells giving rise to inherited transmission, and all cells contain proviral DNA (Tarlinton et al., 2006). Despite this, there is high levels of variation displayed in copy number which may be due to the maintained ability for the ERV to replicate (Tarlinton et al., 2006). KoRV biology, evolution, and disease association is reviewed further in (Denner and Young, 2013) and (Xu and Eiden, 2015). As a newly endogenizing virus, not enough time has elapsed on an evolutionary scale for endogenous KoRV-A genomes to become fixed in all koala genomes. KoRV-A is not pathogenic, nor is it associated with an altered immune profile (Maher and Higgins, 2016). However, KoRV-B is associated with chylamydiosis (Waugh et al., 2017). Much like other ERVs, KoRV-A induces immune tolerance, deleting lymphocytes that recognize KoRV-A (Fiebig et al., 2015). This has implications for vaccination and for recombination, as KoRV-A replication is not restricted by host adaptive immune mechanisms. Recombination of endogenous KoRV-A and exogenous KoRV has not been detected, but detection is difficult due to the high number of endogenous proviruses and the degree of variation that exists amongst these endogenous KoRV-As. Recombination has, however, been described between KoRV-A and a unique ERV, Phascolarctos endogenous retroelements (PhER) (Hobbs et al., 2017; Lober et al., 2018). While there are few consistent established ERV-XRV interactions with respect to KoRV, this new virus remains an important model of endogenization, which may give us more insight as to how ERV-XRV interactions evolve during the early phase of viral endogenization. SIGNIFICANCE While we no longer regard non-coding DNA (ncDNA) as genomic junk, there is an enormous amount to learn about the variety of ways that ncDNA impacts an animal’s biology. Here, we have provided an 16 overview of documented interactions between endogenous and exogenous retroviruses as one step in considering ERV impacts on host biology. While the endogenous retroviral genes are believed to emerge and proliferate following similar mechanisms, the interactions that exist are dependent on the biology of specific viruses highlighted in Table 1. Following initial description in chickens, ERV-XRV interactions have been most thoroughly characterized in mouse models. The primary interactions between endogenous and exogenous viral pairs can be characterized as recombination, receptor interference, immunological self-tolerance, superantigen interference, and action as an antiviral restriction factor (Figure 3; Table 2). On the surface, endogenous retroviruses appear to have common ways in which they interact with their exogenous counterparts. Almost all display the ability to both restrict and promote exogenous infection, though the mechanisms by which this action occurs is virus-specific and act independently from one another. Most virus dyads do experience receptor interference, recombination, and immune self-tolerance. Mouse mammary tumor virus is the only retrovirus shown to interfere via superantigens, and murine leukemia virus is the only retrovirus reviewed where its corresponding ERV hijacks host gene expression to restrict broadly against XRV infection, including its exogenous counterpart. However, more in depth investigations of other retroviruses may reveal these as more general mechanisms. Despite the fact that approximately 8% of human genomes are made of ERV elements, humans ERVs with related extant exogenous retroviruses have not been defined (Seifarth et al., 2005; van der Kuyl, 2012). One possible exception is HTLV and HTLV type I-related endogenous sequences (HRES) identified in 1989, sharing low sequence homology (>30% in the gag region) encoding a 28-kDa expressed in H9 human T cells that is cross-reactive with HTLV-I gag antigens (Banki et al., 1992; Perl et al., 1989). While it remains entirely possible that exogenous retroviruses will be discovered that have human ERVs correlates, the fact that we have not identified such circumstances raises the tantalizing question of whether ERV establishment ultimately allows a host to overcome the exogenous viral infection that preempted its existence. This is incredibly important, as approximately 70% of infectious diseases are zoonotic in nature that may have the potential to interact with human ERVs (Blancou et al., 2005). Furthermore, unfixed human ERVs are continually undergoing transposition, which may lead to 17 new host interactions (Wildschutte et al., 2016). It is for this reason that some groups are looking to remove ERVs in domesticated pigs with sights of reducing risks of ERV interactions in xenografts and organ transplants (Niu et al., 2017). Ultimately, understanding how our genome elements relate directly to susceptibility or resistance to infection through ERV interactions may be key in preparation for defending against new retroviral threats, assessing individual risks to infections, and overcoming many infectious and non-infectious diseases 18 Figure 1.1: Retrovirus infection results in the integration of reverse transcribed viral DNA into the host genome that results in lifelong infection. In a small proportion of infections, infection can occur in the germline. When infected germ cells result in a developing embryo, all fetal cells are infected with the new endogenous retrovirus. Over many generations, the virus undergoes multiple mutations including single nucleotide polymorphisms, insertions, and deletions, often times leading to the inability for endogenous retrovirus to produce infections virus. Exogenous Retrovirus Primary infection Horizontal Transmission Nucleus Host cell Uncoating Reverse transcription and integration Infection Retroviral proviral DNA Germline Infection Vertical Transmission Viral RNA Proviral DNA Endogenous Retrovirus Multiple Generations 19 Figure 1.2: Rous Sarcoma Virus (RSV) Associated Virus (0) (RAV-0) is an endogenous retrovirus that does not produce infectious virus. In C/A strain chickens harboring RAV-0, infections with RSV leads to replication of the virus. Japanese quail are susceptible to RSV infection, but they have minimal RSV replication. When Japanese quail are co-inoculated with RAV-0 and RSV, RAV-0 acts as a helper virus that rescues Japanese quail cells to produce replicate. Rous Sarcoma Virus C/A Chicken Japanese Quail Host cell Budding Nucleus RSV provirus RAV-0 Replication Packaging Host cell Inefficient Replication Nucleus RSV provirus Replication Packaging Host cell Budding Nucleus RSV provirus RAV-0 Replication Packaging + RAV-0 Host cell Nucleus RAV-0 Cell culture No infection Outcome No viral production RSV production Minimal RSV production Rescued RSV production 20 Endogenous Retrovirus Receptor Interference Restrict Exogenous Retrovirus Infection Promote Exogenous Retrovirus Infection Recombination Immune self-tolerance Copackaging of exogenous and endogenous retrovirus RNA Integration of Recombinant virus X X Negative selection of self-reactive T-cells ERV-positive ERV-negative Downregulation or competition with ERV Env Host cell Budding Nucleus RSV provirus RAV-0 Replication Packaging Host cell Inefficient Replication Nucleus RSV provirus Replication Packaging ERVs act as a helper virus for XRV replication Template switching during reverse transcription NucleusJSRV enJSRV Gag Gag enJSRV Gag blocks normal trafficking of JSRV Gag 21 Figure 1.3: Endogenous retroviruses have many functions that impact normal host cell biology. A number of these functions specifically interact with exogenous retroviruses. Major classes of interaction include receptor interference, immune self-tolerance, recombination, as well as the simultaneous action of restriction and promotion of exogenous retrovirus infection through different mechanisms. Above, particular examples are shown, but are not exhaustive as particular interactions virus-specific. 22 Table 1.1: Six endogenous/exogenous retroviral dyads have been identified. ERV = endogenous retrovirus; XRV = exogenous retrovirus; RSV = Rous sarcoma virus; ASLV = avian sarcoma leucosis virus; JSRV = Jaagsiekte sheep retrovirus; MMTV = mouse mammary tumor virus; MuLV = murine leukemia virus; FeLV = feline leukemia virus; KoRV = koala retrovirus; ND = not determined. Virus Host ERV terminology Date of earliest ERV Integration Endogenous copy number Shared identity with XRV Common diseases associated with XRV RSV/ ASLV Galliform birds endogenous ASLV ND average 5 >97% (pol and env;nt); 59% (gag;nt) Sarcoma, leukosis JSRV Sheep, Goats enJSRV 0.9-1.8 MYA average 30 90-98% (AA) Ovine pulmonary adenocarcinoma MMTV Mice Mtv ND 2-8 ND Mammary adenoma MuLV Mice Emv, Xmv, Pmv ~1 MYA >50 100%* (nt) Leukemia FeLV Felid cats enFeLV 3-4 MYA 8-24 86% (nt) Leukemia, lymphoma, anemia, immune suppression KoRV Koalas KoRV-A 22,200-49,900 0-165 ND Associations with neoplasia and secondary, opportunistic infections, including Chlamydiosis 23 Table 1.2: Endogenous/exogenous retroviral dyads interact in five main categories. The specifics of interactions are specific to the respective couplets. ND = not determined Virus ERV terminology Immunological self-tolerance Recombination Receptor interference Restrict exogenous infection Promote exogenous infection RSV/ASLV endogenous ASLV Yes Yes Yes Yes Yes JSRV enJSRV Yes ND Yes Yes ND MMTV Mtv Yes ND ND Yes Yes MuLV Emv, Xmv, Pmv ND Yes ND Yes Yes FeLV enFeLV ND Yes Yes Yes Yes KoRV KoRV-A Yes Yes ND ND ND 24 LITERATURE CITED Aiewsakun, P., Katzourakis, A., 2017. Marine origin of retroviruses in the early Palaeozoic Era. Nat Commun 8, 13954. Anderson, M.M., Lauring, A.S., Robertson, S., Dirks, C., Overbaugh, J., 2001. Feline Pit2 functions as a receptor for subgroup B feline leukemia viruses. J Virol 75, 10563-10572. Arnaud, F., Caporale, M., Varela, M., Biek, R., Chessa, B., Alberti, A., Golder, M., Mura, M., Zhang, Y.P., Yu, L., Pereira, F., Demartini, J.C., Leymaster, K., Spencer, T.E., Palmarini, M., 2007a. A paradigm for virus-host coevolution: sequential counter-adaptations between endogenous and exogenous retroviruses. PLoS pathogens 3, e170. Arnaud, F., Murcia, P.R., Palmarini, M., 2007b. Mechanisms of late restriction induced by an endogenous retrovirus. J Virol 81, 11441-11451. Arnaud, F., Varela, M., Spencer, T.E., Palmarini, M., 2008. Coevolution of endogenous betaretroviruses of sheep and their host. Cell Mol Life Sci 65, 3422-3432. Astrin, S.M., Crittenden, L.B., Buss, E.G., 1980. ev 2, a genetic locus containing structural genes for endogenous virus, codes for Rous-associated virus type 0 produced by line 72 chickens. J Virol 33, 250-255. Bai, J., Zhu, R.-Y., Stedman, K., Cousens, C., Carlson, J., Sharp, J.M., DeMartini, J.C., 1996. Unique long terminal repeat U3 sequences distinguish exogenous Jaagsiekte sheep retroviruses associated with ovine pulmonary carcinoma from endogenous loci in the sheep genome. J Virol 70, 3159- 3168. Bamunusinghe, D., Liu, Q., Plishka, R., Dolan, M.A., Skorski, M., Oler, A.J., Yedavalli, V.R.K., Buckler-White, A., Hartley, J.W., Kozak, C.A., 2017. Recombinant origins of pathogenic and nonpathogenic mouse gammaretroviruses with polytropic host range. J Virol 91, 1-19. Bamunusinghe, D., Naghashfar, Z., Buckler-White, A., Plishka, R., Baliji, S., Liu, Q., Kassner, J., Oler, A.J., Hartley, J., Kozak, C.A., 2016. Sequence Diversity, Intersubgroup Relationships, and Origins of the Mouse Leukemia Gammaretroviruses of Laboratory and Wild Mice. J Virol 90, 4186-4198. Bandecchi, P., Dell’Omodarme, M., Magi, M., Palamidessi, A., Prati, M.C., 2006. Feline leukaemia virus (FeLV) and feline immunodeficiency virus infections in cats in the Pisa district of Tuscany, and attempts to control FeLV infection in a colony of domestic cats by vaccination. Vet Rec 158, 555- 557. Banki, K., Maceda, J., Hurley, E., Ablonczy, E., Mattson, D.H., Szegedy, L., Hung, C., Perl, A., 1992. Human T-cell lymphotropic virus (HTLV)-related endogenous sequence, HRES-1, encodes a 28- kDa protein: A possible autoantigen for HTLV-I gag-reactive autoantibodies. Proceedings of the National Academy of Science 89, 1939-1943. Bannert, N., Hofmann, H., Block, A., Hohn, O., 2018. HERVs New Role in Cancer: From Accused Perpetrators to Cheerful Protectors. Front Microbiol 9, 178. Bechtel, M.K., Hayes, K.A., Mathes, L.E., Pandey, R., Stromberg, P.C., Roy-Burman, P., 1999. Recombinant feline leukemia virus (FeLV) variants establish a limited infection with altered cell tropism in specific-pathogen-free cats in the absence of FeLV Subgroup A Helper Virus. Veterinary Pathology 36, 91-99. Bénit, L., de Pearseval, N., Casella, J.-F., Callebaut, I., Cordonnier, A., Heidmann, T., 1997. Cloning of a new murine endogenous retrovirus, MuERV-L, with strong similarity to the human HERV-L element and with a gag coding sequence closely related to the Fv1 restriction gene. J Virol 71, 5652-5657. Benkel, B., Rutherford, K., 2014. Endogenous avian leukosis viral loci in the Red Jungle Fowl genome assembly. Poult Sci 93, 2988-2990. Bhadra, S., Lozano, M.M., Payne, S.M., Dudley, J.P., 2006. Endogenous MMTV proviruses induce susceptibility to both viral and bacterial pathogens. PLoS pathogens 2, e128. 25 Blancou, J., Chomel, B.B., Belotto, A., Meslin, F.X., 2005. Emerging or re-emerging bacterial zoonoses: factors of emergence, surveillance and control. Vet Res 36, 507-522. Boeke, J.D., Stoye, J.P., 1997. Retrotransposons, endogenous retroviruses, and the evolution of retroelements, in: Coffin, J.M., Hughes, S.H., Varmus, H.E. (Eds.), Retroviruses. Cold Spring Harbor Laboratory Press, Cold Spring Harbor (NY), pp. 343-346. Boi, S., Rosenke, K., Hansen, E., Hendrick, D., Malik, F., Evans, L.H., 2016. Endogenous retroviruses mobilized during friend murine leukemia virus infection. Virology 499, 136-143. Bolin, S.R., 1995. The Pathogenesis of Mucosal Disease. Veterinary Clinics of North America: Food Animal Practice 11, 489-500. Britten, R.J., 1997. Mobile elements inserted in the distant past have taken on important functions. Gene 205, 177-182. Bruno, R.D., Rosenfield, S.M., Smith, G.H., 2013. Late developing mammary tumors and hyperplasia induced by a low-oncogenic variant of mouse mammary tumor virus (MMTV) express genes identical to those induced by canonical MMTV. Molecular Cancer 12, 1-6. Canfield, P.J., Sabine, J.M., Love, D.N., 1988. Virus particles associate with leukaemia in a koala. Australian Veterinary Journal 65, 327-328. Carlson, J., Lyon, M., Bishop, J., Vaiman, A., Cribiu, E., Mornex, J.F., Brown, S., Knudson, D., DeMartini, J., Leroux, C., 2003. Chromosomal Distribution of Endogenous Jaagsiekte Sheep Retrovirus Proviral Sequences in the Sheep Genome. J Virol 77, 9662-9668. Chappell, K.J., Brealey, J.C., Amarilla, A.A., Watterson, D., Hulse, L., Palmieri, C., Johnston, S.D., Holmes, E.C., Meers, J., Young, P.R., 2017. Phylogenetic Diversity of Koala Retrovirus within a Wild Koala Population. J Virol 91. Charreyre, C., Pedersen, N.C., 1991. Study of feline leukemia virus immunity. Journal of American Veterinary Medical Association 199, 1316-1324. Chiu, E.S., Hoover, E.A., VandeWoude, S., 2018. A retrospective examination of feline leukemia subgroup characterization: viral interference assays to deep sequencing. Viruses 10, doi:10.3390/v10010029. Chiu, E.S., Kraberger, S., Cunningham, M., Cusack, L., Roelke, M., VandeWoude, S., 2019. Multiple Introductions of Domestic Cat Feline Leukemia Virus in Endangered Florida Panthers. Emerg Infect Dis 25, 92-101. Choi, Y., Kappler, J.W., Marrack, P., 1991. A superantigen encoded in the open reading frame of the 3’ long terminal repeat of mouse mammary tumour virus. Nature 350, 203-207. Coffin, J.M., 1992. Genetic Diveristy and Evolution of Retroviruses. Springer, Berlin Heidelberg. Coffin, J.M., 2004. Evolution of retroviruses: Fossils in our DNA. Proceedings of the American Philosophical Society 148, 264-280. Crittenden, L.B., Hayward, W.S., Hanafusa, H., Fadly, A.M., 1980. Induction of neoplasms by subgroup E recombinants of exogenous and endogenous avian retroviruses (Rous-associated virus type 60). J Virol 33, 915-919. Crittenden, L.B., McMahon, S., Halpern, M.S., Fadly, A.M., 1987. Embryonic infection with the endogenous avian leukosis virus Rous-associated virus-0 alters responses to exogenous avian leukosis virus infection. J Virol 61, 722-725. Crittenden, L.B., Smith, E.J., Fadly, A.M., 1984. Influence of endogenous viral (ev) gene expression and strain of exogenous avial leukosis virus (ALV) on Mortality and ALV infection and shedding in chickens. Avian Diseases 28, 1037-1056. DeMartini, J.C., Carlson, J.O., Leroux, C., Spencer, T.E., Palmarini, M., 2003. Endogenous retrovirus related to jaagsiekte sheep retrovirus. Current Topics in Microbiology and Immunology 275, 117- 137. Denesvre, C., Soubieux, D., Pin, G., Hue, D., Dambrine, G., 2003. Interference between avian endogenous ev/J 4.1 and exogenous ALV-J retroviral envelopes. J Gen Virol 84, 3233-3238. Denner, J., 2016. Expression and function of endogenous retroviruses in the placenta. APMIS 124, 31-43. 26 Denner, J., Young, P.R., 2013. Koala retroviruses: characterization and impact on the life of koalas. Retrovirology 10, 1-7. Dunlap, K.A., Palmarini, M., Varela, M., Burghardt, R.C., Hayashi, K., Farmer, J.L., Spencer, T.E., 2006. Endogenous retroviruses regulate periimplantation placental growth and differentiation. PNAS 103, 14390-14395. Evans, L.H., Alamgir, A.S., Owens, N., Weber, N., Virtaneva, K., Barbian, K., Babar, A., Malik, F., Rosenke, K., 2009. Mobilization of endogenous retroviruses in mice after infection with an exogenous retrovirus. J Virol 83, 2429-2435. Fadly, A.M., Smith, E.J., 1997. Role of contact and genetic transmission of endogenous virus-21 in the susceptibility of chickens to avian leukosis virus infection and tumors. Poultry Science 76, 968- 973. Fiebig, U., Keller, M., Moller, A., Timms, P., Denner, J., 2015. Lack of antiviral antibody response in koalas infected with koala retroviruses (KoRV). Virus Res 198, 30-34. Frank, J.A., Feschotte, C., 2017. Co-option of endogenous viral sequences for host cell function. Curr Opin Virol 25, 81-89. Fujino, Y., Ohno, K., Tsujimoto, H., 2008. Molecular pathogenesis of feline leukemia virus-induced malignancies: insertional mutagenesis. Vet Immunol Immunopathol 123, 138-143. Garcia-Etxebarria, K., Sistiaga-Poveda, M., Begoña, M.J., 2014. Endogenous retroviruses in domestic animals. Current Genomics 15, 256-265. Gavora, J.S., Spencer, J.L., Benkel, B., Gagnon, C., Emsley, A., Kulenkamp, A., 1995. Endogenous viral genes influence infection with avian leukosis virus. Avian Pathol 24, 653-664. Gleich, S.E., Krieger, S., Hartmann, K., 2009. Prevalence of feline immunodeficiency virus and feline leukaemia virus among client-owned cats and risk factors for infection in Germany. J Feline Med Surg 11, 985-992. Golovkina, T.V., Chervonsky, A., Dudley, J.P., Ross, S.R., 1992. Transgenic mouse mammary tumor virus superantigen expression prevents viral infection Cell 69, 637-645. Grandi, N., Tramontano, E., 2018. Human Endogenous Retroviruses Are Ancient Acquired Elements Still Shaping Innate Immune Responses. Front Immunol 9, 2039. Greenwood, A.D., Ishida, Y., O’Brien, S.P., Roca, A.L., Eiden, M.V., 2018. Transmission, evolution, and endogenization: lessons learned from recent retroviral invasions. Microbiology and Molecular Biology Reviews 82, e00044. Habel, D.E., Dohrer, K.L., Conklin, K.F., 1993. Functional and defective components of avian endogenous virus long terminal repeat enhancer sequences. J Virol 67, 1545-1554. Hanger, J.J., Bromham, L.D., McKee, J.J., O’Brien, T.M., Robinson, W.F., 2000. The nucleotide sequence of koala (Phascolarctos cinereus) retrovirus: a novel type C endogenous virus related to gibbon ape leukemia virus. J Virol 74, 4264-4272. Harris, R.S., Dudley, J.P., 2015. APOBECs and virus restriction. Virology 479-480, 131-145. Hartmann, K., 2012. Clinical aspects of feline retroviruses: a review. Viruses 4, 2684-2710. Hartmann, S., Hasenkamp, N., Mayer, J., Michaux, J., Morand, S., Mazzoni, C.J., Roca, A.L., Greenwood, A.D., 2015. Endogenous murine leukemia retroviral variation across wild European and inbred strains of house mouse. BMC genomics 16, 613. Hecht, S.J., Stedman, K.E., Carlson, J.O., DeMartini, J.C., 1996. Distribution of endogenous type B and type D sheep retrovirus sequences in ungulates and other mammals. Proceedings of the National Academy of Science 93, 3297-3302. Held, W., Shakhov, A.N., Izui, S., Waanders, G.A., Scarpellino, L., MacDonald, H.R., Acha-Orbea, A., 1993. Superantigen-reactive CD4+ T cells are required to stimulate B cells after infection with mouse mammary tumor virus. Journal of Experimental Medicine 177, 359-366. Hobbs, M., King, A., Salinas, R., Chen, Z., Tsangaras, K., Greenwood, A.D., Johnson, R.N., Belov, K., Wilkins, M.R., Timms, P., 2017. Long-read genome sequence assembly provides insight into ongoing retroviral invasion of the koala germline. Sci Rep 7, 15838. 27 Holt, M.P., Shevach, E.M., Punkosdy, G.A., 2013. Endogenous mouse mammary tumor viruses (mtv): new roles for an old virus in cancer, infection, and immunity. Front Oncol 3, 287. Hsu, K., Lee, Y.K., Chew, A., Chiu, S., Lim, D., Greenhalgh, D.G., Cho, K., 2017. Inherently variable responses to glucocorticoid stress among endogenous retroviruses isolated from 23 mouse strains. Biochim Biophys Acta Mol Basis Dis 1863, 2594-2600. Hu, X., Zhu, W., Chen, S., Liu, Y., Sun, Z., Geng, T., Song, C., Gao, B., Wang, X., Qin, A., Cui, H., 2017. Expression patterns of endogenous avian retrovirus ALVE1 and its response to infection with exogenous avian tumour viruses. Arch Virol 162, 89-101. Hu, X., Zhu, W., Chen, S., Liu, Y., Sun, Z., Geng, T., Wang, X., Gao, B., Song, C., Qin, A., Cui, H., 2016. Expression of the env gene from the avian endogenous retrovirus ALVE and regulation by miR-155. Arch Virol 161, 1623-1632. Hughes, S.H., 1982. Sequence of the long terminal repeat and adjacent segments of the endogenous avian virus rous-associated virus 0. J Virol 43, 191-200. Imai, S., Tsubura, Y., Hilgers, J., Michalides, R., 1983. A new locus (Mtv-4) for endogenous mammary tumor virus expression and early mammary tumor development in the SHN mous strain. J. Natl. Cancer Inst. 71, 517-521. Ishida, Y., Zhao, K., Greenwood, A.D., Roca, A.L., 2015. Proliferation of Endogenous Retroviruses in the Early Stages of a Host Germ Line Invasion. Mol Biol Evol 32, 109-120. Jarrett, O., Hardy, W.D., Jr., Golder, M.C., Hay, D., 1978. The frequency of occurrence of feline leukaemia virus subgroups in cats. Int. J. Cancer 21, 334-337. Jarrett, O., Russel, P.H., 1978. Differential growth and transmission in cats of feline leukaemia viruses of subgroups A and B. Int. J. Cancer 1978, 466-472. Jenkins, N.A., Copeland, N.G., Taylor, B.A., Lee, B.K., 1982. Organization, distribution, and stability of endogenous ecotropic murine leukemia virus DNA sequences in chromosomes of Mus musculus. J Virol 43, 26-36. Jern, P., Stoye, J.P., Coffin, J., 2005. Role of APOBEC3 in Genetic Diversity among Endogenous Murine Leukemia Viruses. PLoS Genetics preprint. Johnson, W.E., 2019. Origins and evolutionary consequences of ancient endogenous retroviruses. Nat Rev Microbiol. Kozak, C.A., 2014. Origins of the endogenous and infectious laboratory mouse gammaretroviruses. Viruses 7, 1-26. Kuhnlein, U., Fairfull, R.W., Gowe, R., Kulenkamp, A., Mou, L., Zadworny, D., 1992. Synergism between the endogenous viral loci ev6 and ev9 in inducing immunological tolerance to avian leukosis virus. British Poultry Science 34, 93-104. Lee, K.-H., Lim, D., Greenhalgh, D., Cho, K., 2017. Highly Variable Genomic Landscape of Endogenous Retroviruses in the C57BL/6J Inbred Strain, Depending on Individual Mouse, Gender, Organ Type, and Organ Location. International Journal of Genomics 2017, 1-10. Lee, K.H., Lim, D., Chiu, S., Greenhalgh, D., Cho, K., 2016. Genomic landscapes of endogenous retroviruses unveil intricate genetics of conventional and genetically-engineered laboratory mouse strains. Exp Mol Pathol 100, 248-256. Leroux, C., Girard, N., Cottin, V., Greenland, T., Mornex, J.F., Archer, F., 2007. Jaagsiekte Sheep Retrovirus (JSRV): from virus to lung cancer in sheep. Vet Res 38, 211-228. Li, W., Yap, M.W., Voss, V., Stoye, J.P., 2016. Expression levels of Fv1: effects on retroviral restriction specificities. Retrovirology 13, 42. Lober, U., Hobbs, M., Dayaram, A., Tsangaras, K., Jones, K., Alquezar-Planas, D.E., Ishida, Y., Meers, J., Mayer, J., Quedenau, C., Chen, W., Johnson, R.N., Timms, P., Young, P.R., Roca, A.L., Greenwood, A.D., 2018. Degradation and remobilization of endogenous retroviruses by recombination during the earliest stages of a germ-line invasion. Proc Natl Acad Sci U S A. Luo, G.X., Taylor, J., 1990. Template switching by reverse transcriptase during DNA synthesis. J Virol 64, 4321-4328. 28 Maher, I.E., Higgins, D.P., 2016. Altered Immune Cytokine Expression Associated with KoRV B Infection and Season in Captive Koalas. PloS one 11, e0163780. Marrack, P., Kushnir, E., Kappler, J., 1991. A maternally inherited superantigen encoded by a mammary tumour virus. Nature 349, 524-526. Matsuzawa, A., Sayama, K., Tsubura, A., Murakami, A., 1990. A congenic line of the DDD mouse strain, DDD?1-Mtv-2/Mtv-2: establishment and mammary tumorigenesis. Japanese Journal of Cancer Research 81, 639-644. McDougall, A.S., Terry, A., Tzavaras, T., Cheney, C., Rojko, J., Neil, J.C., 1994. Defective endogenous proviruses are expressed in feline lymphoid cells: evidence for a role in natural resistance to subgroup B feline leukemia viruses. J Virol 68, 2151-2160. McGee-Estrada, K., Fan, H., 2007. Comparison of LTR enhancer elements in sheep beta retroviruses: insights into the basis for tissue-specific expression. Virus Genes 35, 303-312. Morris, G., Maes, M., Murdjeva, M., Puri, B.K., 2018. Do Human Endogenous Retroviruses Contribute to Multiple Sclerosis, and if So, How? Mol Neurobiol. Muirden, A., 2002. Prevalence of feline leukaemia virus and antibodies to feline immunodeficiency virus and feline coronavirus in stray cats sent to an RSPCA hospital. Vet Rec 150, 621-625. Murcia, P.R., Arnaud, F., Palmarini, M., 2007. The transdominant endogenous retrovirus enJS56A1 associates with and blocks intracellular trafficking of Jaagsiekte sheep retrovirus Gag. J Virol 81, 1762-1772. Niu, D., Wei, H.J., Lin, L., George, H., Wang, T., Lee, I.H., Zhao, H.Y., Wang, Y., Kan, Y., Shrock, E., Lesha, E., Wang, G., Luo, Y., Qing, Y., Jiao, D., Zhao, H., Zhou, X., Wang, S., Wei, H., Guell, M., Church, G.M., Yang, L., 2017. Inactivation of porcine endogenous retrovirus in pigs using CRISPR-Cas9. Science 357, 1303-1307. Ohkusu-Tsukada, K., Tsukada, T., Takahashi, K., 2017. Clonal deletion of T cell repertoires with specific T cell receptor Vbeta chains by two endogenous superantigens in NC/Nga mice. Biosci Biotechnol Biochem 81, 2160-2163. Palmarini, M., Cousens, C., Dalziel, R.G., Bai, J., Stedman, K., Demartini, J.C., Sharp, J.M., 1996. The exogenous form of Jaagsiekte retrovirus is specifically associated with a contagious lung cancer of sheep. J Virol 70, 1618-1623. Palmarini, M., Hallwirth, C., York, D., Murgia, C., de Oliveira, T., Spencer, T.E., Fan, H., 2000. Molecular cloning and functional analysis of three Type D endogenous retroviruses of sheep reveal a different cell tropism from that of the highly related exogenous Jaagsiekte sheep retrovirus. J Virol 74, 8065-8076. Payne, L.N., Nair, V., 2012. The long view: 40 years of avian leukosis research. Avian Pathol 41, 11-19. Perl, A., Rosenblatt, J.D., Chen, I.S.Y., DiVincenzo, J.P., Bever, R., J., P.B., NAbraham, G.N., 1989. Detection and cloning of new HTLV-related endogenous sequences in man. Nuclei Acids Research 17, 6841-6854. Polani, S., Roca, A.L., Rosensteel, B.B., Kolokotronis, S.O., Bar-Gal, G.K., 2010. Evolutionary dynamics of endogenous feline leukemia virus proliferation among species of the domestic cat lineage. Virology 405, 397-407. Powers, J.A., Chiu, E.S., Kraberger, S.J., Roelke-Parker, M., Lowery, I., Erbeck, K., Troyer, R., Carver, S., VandeWoude, S., 2018. Feline leukemia virus disease outcomes in a domestic cat breeding colony: Relationship to endogenous FeLV and other chronic viral infections. J Virol. Rasmussen, H.B., 1997. Interactions between exogenous and endogenous retroviruses. Journal of Biomedical Science 4, 1-8. Rous, P., 1911. A sarcoma of the fowl transmissible by an agent separable from the tumor cells. Journal of Experimental Medicine 13, 397-411. Rubin, H., 1955. Quantitative relations between causative virus and cell in the Rous No. 1 chicken sarcoma. Virology 1, 445-473. Rubin, H., 2011. The early history of tumor virology: Rous, RIF, and RAV. Proc Natl Acad Sci U S A 108, 14389-14396. 29 Sacco, M.A., Howes, K., Smith, L.P., Nair, V.K., 2004. Assessing the roles of endogenous retrovirus EAV-HP in avian leukosis virus subgroup J emergence and tolerance. J Virol 78, 10525-10535. Sanville, B., Dolan, M.A., Wollenberg, K., Yan, Y., Martin, C., Yeung, M.L., Strebel, K., Buckler-White, A., Kozak, C.A., 2010. Adaptive evolution of Mus Apobec3 includes retroviral insertion and positive selection at two clusters of residues flanking the substrate groove. PLoS pathogens 6, e1000974. Schneinder, R., Frye, F.L., Taylor, D.O.N., Dorn, C.R., 1967. A household cluster of feline malignant lymphoma. Cancer Research 27, 1316-1322. Seifarth, W., Frank, O., Zeilfelder, U., Spiess, B., Greenwood, A.D., Hehlmann, R., Leib-Mosch, C., 2005. Comprehensive analysis of human endogenous retrovirus transcriptional activity in human tissues with a retrovirus-specific microarray. J Virol 79, 341-352. Sen, N., Simmons, W.J., Thomas, R.M., Erianne, G., Zhang, D.J., Jaeggli, N.S., Huang, C., Xiong, X., Tsiagbe, V.K., Ponzio, N.M., Thorbecke, G.J., 2001. META-Controlled env-Initiated Transcripts Encoding Superantigens of Murine Mtv29 and Mtv7 and Their Possible Role in B Cell Lymphomagenesis. The Journal of Immunology 166, 5422-5429. Sharp, J.M., De Martini, J.C., 2003. Natural history of JSRV in sheep. Current Topics in Microbiology and Immunology 275, 55-79. Shojima, T., Yoshikawa, R., Hoshino, S., Shimode, S., Nakagawa, S., Ohata, T., Nakaoka, R., Miyazawa, T., 2013. Identification of a novel subgroup of Koala retrovirus from Koalas in Japanese zoos. J Virol 87, 9943-9948. Silva, C.P., Onuma, S.S., de Aguiar, D.M., Dutra, V., Nakazato, L., 2016. Molecular detection of Feline Leukemia Virus in free-ranging jaguars (Panthera onca) in the Pantanal region of Mato Grosso, Brazil. Braz J Infect Dis 20, 316-317. Simmons, G.S., Young, P.R., Hanger, J.J., Jones, K., Clarke, D., McKee, J.J., Meers, J., 2012. Prevalence of koala retrovirus in geographically diverse populations in Australia. Aust Vet J 90, 404-409. Smith, E.J., Fadly, A.M., Crittenden, L.B., 1990. Interactions between endogenous virus loci ev6 and ev21. 1. Immune response to exogenous avian leukosis virus infection. Poultry Science 69, 1244- 1250. Smith, L.M., Toye, A.A., Howes, K., Bumstead, N., Payne, L.N., Venugopal, K., 1999. Novel endogenous retroviral sequences in the chicken genome closely related to HPRS-103 (subgroup J) avian leukosis virus. J. Gen. Virol 80, 261-268. Spencer, T.E., Mura, M., Gray, C.A., Griebel, P.J., Palmarini, M., 2003. Receptor Usage and Fetal Expression of Ovine Endogenous Betaretroviruses: Implications for Coevolution of Endogenous and Exogenous Retroviruses. J Virol 77, 749-753. Spencer, T.E., Palmarini, M., 2012a. Application of next generation sequencing in mammalian embryogenomics: Lessons learned from endogenous betaretroviruses of sheep. Animal Reproduction Science 134, 95-103. Spencer, T.E., Palmarini, M., 2012b. Endogenous retroviruses of sheep: a model system for understanding physiological adaptation to an evolving ruminant genome. Journal of Reproduction and Development 58, 33-37. Stewart, H., Jarrett, O., Hosie, M.J., Willett, B.J., 2013. Complete genome sequences of two feline leukemia virus subgroup B isolates with novel recombination sites. Genome Announc 1, e00036- 00012. Swanstrom, R., Wills, J.W., 1997. Synthesis, assembly, and process of viral proteins, in: Coffin, J.M., Hughes, S.H., Varmus, H.E. (Eds.), Retroviruses. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, pp. 263-334. Tandon, R., Cattori, V., Pepin, A.C., Riond, B., Meli, M.L., McDonald, M., Doherr, M.G., Lutz, H., Hofmann-Lehmann, R., 2008. Association between endogenous feline leukemia virus loads and exogenous feline leukemia virus infection in domestic cats. Virus Res 135, 136-143. Tarlinton, R.E., Meers, J., Young, P.R., 2006. Retroviral invasion of the koala genome. Nature 442, 79- 81. 30 Torres, A.N., Mathiason, C.K., Hoover, E.A., 2005. Re-examination of feline leukemia virus: host relationships using real-time PCR. Virology 332, 272-283. van der Kuyl, A.C., 2012. HIV infection and HERV expression: a review. Retrovirology 9, 6. Van Nie, R., Vaerstraeten, A.A., 1975. Studies of genetic transmission of mammary tumour virus by C3Hf mice. Int. J. Cancer 16, 922-931. Varela, M., Spencer, T.E., Palmarini, M., Arnaud, F., 2009. Friendly viruses: the special relationship between endogenous retroviruses and their host. Ann N Y Acad Sci 1178, 157-172. Venugopal, K., 1999. Avian leukosis virus subgroup J- a rapidly evolving group of oncogenic retroviruses. Research in Veterinary Science 67, 113-119. Viginier, B., Dolmazon, C., Lantier, I., Lantier, F., Archer, F., Leroux, C., Terzian, C., 2012. Copy number variation and differential expression of a protective endogenous retrovirus in sheep. PloS one 7, e41965. Vogt, P.K., Friis, R.R., 1971. An avian leukosis virus related to RSV(O): properties and evidence for helper activity. Virology 43, 223-234. Waugh, C.A., Hanger, J., Loader, J., King, A., Hobbs, M., Johnson, R., Timms, P., 2017. Infection with koala retrovirus subgroup B (KoRV-B), but not KoRV-A, is associated with chlamydial disease in free-ranging koalas (Phascolarctos cinereus). Scientific Reports 7. Wildschutte, J.H., Williams, Z.H., Montesion, M., Subramanian, R.P., Kidd, J.M., Coffin, J.M., 2016. Discovery of unfixed endogenous retrovirus insertions in diverse human populations. Proc Natl Acad Sci U S A. Xu, W., Eiden, M.V., 2015. Koala Retroviruses: Evolution and Disease Dynamics. Annu Rev Virol 2, 119-134. Xu, W., Gorman, K., Santiago, J.C., Kluska, K., Eiden, M.V., 2015. Genetic diversity of koala retroviral envelopes. Viruses 7, 1258-1270. Xu, W., Stadler, C.K., Gorman, K., Jensen, N., Kim, D., Zheng, H., Tang, S., Switzer, W.M., Pye, G.W., Eiden, M.V., 2013. An exogenous retrovirus isolated from koalas with malignant neoplasias in a US zoo. Proc Natl Acad Sci U S A 110, 11547-11552. Xu, X., Zhao, H., Gong, Z., Han, G.Z., 2018. Endogenous retroviruses of non-avian/mammalian vertebrates illuminate diversity and deep history of retroviruses. PLoS pathogens 14, e1007072. Yilmaz, H., Ilgaz, A., 2000. Prevalence of FIV and FeLV infections in cats in Istanbul. J Feline Med Surg 2, 69-70. Young, G.R., Yap, M.W., Michaux, J.R., Steppan, S.J., Stoye, J.P., 2018. Evolutionary journey of the retroviral restriction gene Fv1. Proc Natl Acad Sci U S A 115, 10130-10135. 31 CHAPTER TWO A Retrospective Examination of Feline Leukemia Subgroup Characterization: Viral Interference Assays to Deep Sequencing INTRODUCTION In the early 1960’s, William Jarrett described feline leukemia virus (FeLV) as the infectious agent responsible for approximately half of observed cases of feline leukemia and lymphoma (Jarrett et al., 1964). The discovery of this pathogenic gammaretrovirus launched the field of feline retrovirology and discoveries relating to mechanisms of retroviral-induced cancers and oncogenes (Benveniste et al., 1975; Willett and Hosie, 2013). FeLV was historically a common domestic cat pathogen, and remains one of the few retroviral diseases for which there is an effective vaccine (Hoover et al., 1996; Pedersen et al., 1979; Willett and Hosie, 2013). As the incidence of FeLV decreased via effective quarantine and vaccination procedures, and with the discovery of feline and simian immunodeficiency viruses as alternate and more analogous models for HIV research, studies of FeLV biology and pathogenesis diminished. Therefore, most of the significant FeLV literature was generated before the development of ‘modern’ molecular techniques. In this retrospective, we review the traditional assays used to establish classical virus subgroups, examine how modern molecular techniques may be used to re-evaluate FeLV subgroup classification schemes, and provide new information to unravel interactions between exogenous and endogenous retroviruses. FeLV GENOME ORGANIZATION The genome structure of retroviruses includes three genes flanked by un-translated regulatory sequences known as long terminal repeats (LTR). Gag encodes group-specific capsid antigens, pol encodes protease, integrase, and reverse transcriptase enzymes, and env encodes the envelope proteins (Coffin et al., 1992). FeLV is approximately 8.4-kb in length and lacks accessory genes characteristic of complex feline retroviruses such as feline immunodeficiency virus (FIV) and feline foamy virus (FFV, 32 also referred to as feline spumavirus, FSV). FeLV contains two reading frames, one for gag and pol genes and a second that encodes the env transcript (Figure 1) (Willett and Hosie, 2013). ENDOGENOUS FeLV (enFeLV) As part of the retroviral infection cycle, viral RNA is reverse transcribed into DNA, which enters the nucleus and integrates within the host genome. This process leads to an integrated provirus in host cell DNA, a hallmark of retroviral infection that is a required component of the viral lifecycle. If integration occurs in a germ cell, the provirus can be transmitted vertically through simple Mendelian inheritance (Lavialle et al., 2013). As retrotransposable elements, endogenized retroviruses have duplicate flanking LTRs, and thus can be excised and relocate to other areas of the genome via recombination. Endogenized viruses may acquire mutations that impair productive viral replication, yet remain as endogenous genomic elements fixed in the host genome (Boeke and Stoye, 1997). enFeLV appears to have invaded the feline genome prior to the speciation of the Felis genus (Polani et al., 2010). While enFeLVs do not induce disease in the host, they are highly relevant to domestic cat FeLV biology. enFeLV is expressed in many tissue types and is associated with FeLV infection (Krunic et al., 2015; McDougall et al., 1994; Tandon et al., 2008a; Tandon et al., 2008b). enFeLV integration site and copy numbers vary among individual cats (8-12 copies per haploid genome; up to 19 per diploid genome) due to viral transposition events and multiple independent integrations (Boeke and Stoye, 1997; Koshy et al., 1980; Polani et al., 2010; Roca et al., 2005; Roca et al., 2004). Increased enFeLV proviral copies have been correlated with both increased (Tandon et al., 2008a; Tandon et al., 2008b) and decreased (Powers et al.) susceptibility to FeLV infection, but not with disease progression (Tandon et al., 2008a). Endogenous and exogenous FeLVs are approximately 86% similar at the nucleotide level. Differences between enFeLV and exFeLV occur in gag and env, and feature insertions and deletions (INDELs), frameshifts, nonsense mutations, and changes to the unique 3’ regions of the LTR (Figure 1) (Willett and Hosie, 2013). As noted below, enFeLV recombination with exFeLV results in novel FeLV subgroups (Soe et al., 1985), though the relationship between enFeLV and exFeLV infection has not been 33 extensively studied. Because most felid species do not harbor enFeLV, naturally occurring FeLV infections in non-domestic felids provide an opportunity to interrogate protection or promotion of exogenous FeLV by enFeLV in a biologically relevant system. EXOGENOUS FeLV (exFeLV) It is postulated that FeLV arose from a rodent-derived virus that evolved to infect cats as a consequence of predator/prey relationship between cats and mice (Benveniste et al., 1975). Exogenous (horizontally transmissible/infectious) FeLVs have been classified as subgroups based upon functional and genetic relatedness. The first three FeLV subgroups identified (FeLV-A, B, and C) were characterized using viral interference (VI) assays, and eventually were associated with subgroup-specific clinical phenotypes by O. Jarrett and colleagues (Jarrett, 1992; Sarma and Log, 1971, 1973). Definition of FeLV subgroups was an early area of intense FeLV study because of their relation to differences in disease progression and prognosis. FeLV-A is the most common horizontally transmitted subgroup (Jarrett et al., 1978; Jarrett and Russel, 1978). While FeLV-A has been reported to be less pathogenic than other FeLV subgroups, it has been associated with macrocytic anemia, immunosuppression, and lymphoma (Hartmann, 2012; Willett and Hosie, 2013). FeLV-B, a recombinant of FeLV-A with endogenous FeLV (enFeLV), has been reported to occur in approximately half of cats infected with FeLV-A. It arises by recombination between FeLV-A and enFeLV subsequent to co-packaging of expressed enFeLV and exFeLV transcripts into a single virion, followed by strand displacement during reverse transcription (Roy-Burman, 1996; Stewart et al., 2011; Stewart et al., 1986). FeLV-B is tumorigenic (Hartmann, 2012), and is considered to be incapable of horizontal transmission unless it is co-transmitted with FeLV-A (Sarma and Log, 1973), with rare exception (Silva et al., 2016; Stewart et al., 2013). FeLV-C is a less common subgroup that arises from de novo mutations in env of FeLV-A and has been associated with the development of aplastic anemia (Abkowitz et al., 1987; Hoover et al., 1974; Mackey et al., 1975; Onions et al., 1982; Overbaugh et al., 1988; Riedel et al., 1986; Willett and Hosie, 2013). 34 VIRAL INTERFERENCE ASSAYS (VI) Viral interference assays test the ability of one viral strain to limit infection with a second viral isolate. Viral interference occurs via both intrinsic and extrinsic mechanisms resulting from cellular pathways that are perturbed during viral infection. Extrinsic VI is caused by competitive blockage of virus receptor by proteins or other viruses that bind and occlude receptor-mediated entry for subsequent viruses. Intrinsic VI refers to multiple processes including intra-cellular receptor fatigue (Breiner et al., 2001; McDougall et al., 1994; Piguet et al., 1999; Rasmussen, 1997), interferon-mediated interference in response to viral genetic material (Haller et al., 2006), and superinfection exclusion (Folimonova, 2012). Viral interference assays were used to distinguish and initially define FeLV subgroups A, B, and C, presumably via intrinsic mechanisms. FeLV viruses that “interfere” with one another (i.e., virus A precludes superinfection with virus B) were tested by a classical method to identify viral groups of the same subgroup (which interfere) versus viruses of different subgroups (which do not interfere) (Marcus and Carver, 1967; Rott et al., 1972). In 1971, Sarma and Log used interference assays to establish the first three recognized FeLV subgroups: A, B, and C (Figure 2) (Sarma and Log, 1971). Focus-forming FeLV/murine sarcoma virus (MSV) pseudotypes (viral chimeric constructs in which MSV envelope proteins have been replaced by FeLV env) were produced by rescue of 9 natural tumorigenic FeLV isolates following co-culture on Harvey MSV-infected hamster tumor cells and feline embryonic fibroblasts. Subsequent in vitro infection of feline embryo fibroblasts with one subgroup resulted in the blockage of the corresponding pseudotype. Cell cultures were considered to demonstrate viral interference if a 2-log drop in focus forming titer was measured. For example, when feline embryo fibroblast cultures were infected with FeLV-A, they were still susceptible to FeLV-B and C pseudotypes (i.e., foci were present following secondary infection). Additionally, cells infected with FeLV-C were susceptible to FeLV-B pseudotype infection, and vice versa (Figure 2). These experiments led to the conclusion that FeLV-A, B and C were genetically different and capable of superinfection in cells. 35 Curiously, primary infection with FeLV-B or FeLV-C virus blocked subsequent infection of FeLV-A pseudotype. This unexpected display of viral interference between different strains subgroups provided evidence for co-infection between FeLV-A and other. This led to the hypothesis that FeLV-A is a necessary precursor for the development of more pathogenic FeLV subgroups and is an essential helper virus for other subgroups. Subgroups were further described by demonstrating that neutralizing antibodies raised in goats and cats inoculated with different strains demonstrated subgroup neutralizing specificity, further elucidating variation among subgroups (Russell and Jarrett, 1978). Using this criterion, FeLV-A was more monotypic compared to FeLV-B and C, which displayed more antigenic variation. On a functional level, VI among FeLV subgroups may be explained by variation in receptor use (extrinsic interference). FeLV-A uses thiamine transporter receptors (ThTR-1) (Mendoza et al., 2006) while FeLV-B uses a common retroviral entry receptor, the phosphate transporter receptors (PiT-1/2) (Anderson et al., 2001; Johann et al., 1992; O'Hara et al., 1990; Takeuchi et al., 1992). FeLV-A env would bind ThTR-1, which would not preclude binding to PiT-1/2, but cells infected with FeLV-B would not be permissive to an additional FeLV-A infection as FeLV-B infections almost always involve a FeLV-A co-infection. FeLV-C uses a heme exporter receptor (FLVCR-1/2) along with ThTR-1/2 (Quigley et al., 2000; Shalev et al., 2009; Tailor et al., 1999). SANGER SEQUENCING Now a fundamental technique in molecular biology, Sanger sequencing was developed in 1977, after FeLV was discovered and classified by VI assays (Sanger et al., 1977; Sarma and Log, 1971, 1973). Sanger sequencing introduced nucleotide analysis allowing researchers to understand and associate FeLV genetic sequences with functional proteins (Tailor and Kabat, 1997). Additionally, other FeLV subgroups marked by relatively minor genetic variations were identified , making subgroup identification more complicated. In 1980, Rosenberg et al. conducted a sequence-level comparative analysis of FeLV-A, B, and C. Homology indices based on 2D PAGE fingerprinting were low among all subgroups (37-66%) 36 (Rosenberg et al., 1980). Modern sequencing technologies have allowed full genome analyses of FeLV, and documented homology among all subgroups and enFeLV by pairwise comparison. Figure 3 illustrates strain similarities using SDTv1.2 nucleotide pairwise comparison tool following MAFFT multiple sequence alignment (Muhire et al., 2014). FeLV-A displays the strongest sequence conservation among distinct FeLV-A isolates, with some genes having 98% homology (Boomer et al., 1994; Donahue et al., 1988). Other subgroups are less well conserved. For instance, FeLV-B was first characterized as having up to ten variable regions with respect to FeLV-A (Nunberg et al., 1984; Riedel et al., 1986; Riedel et al., 1988; Roy-Burman, 1996) (J. I. M., unpublished results). The sequences of the variable region depend on the enFeLV source. enFeLVs have not been rigorously examined at the nucleotide level; as a result few FeLV-B sequences have been recorded to allow for detailed nucleotide comparisons (Boomer et al., 1997; Borjatsch et al., 1992; Miyake et al., 2016; Nunberg et al., 1984; Riedel et al., 1986; Riedel et al., 1988; Rigby et al., 1992; Rohn et al., 1998; Stewart et al., 1986; Sugai et al., 2001; Tailor and Kabat, 1997). Variable regions 1-5 (vr1-5) and potentially the C-terminus domain are believed to be responsible for altering cellular tropism due to changes in the receptor binding protein (gp70) based upon phylogenetic analysis (Faix et al., 2002; Riedel et al., 1988; Stewart et al., 1986). Few studies have been performed to document consequences of amino acid variation in other variable regions. Alignment and comparative analyses of enFeLV, FeLV-A and FeLV-B sequences identify a relatively conserved 5’ recombination site in the 5’ gp70 gene. A 3’ recombination site region is also evident, but is more variable (Boomer et al., 1994; Stewart et al., 2011; Watanabe et al., 2013). Variation in recombination sites between enFeLV and exFeLV results in nucleotide divergence among FeLV-B genotypes, particularly in the envelope gene. However, FeLV-B’s still share significant pairwise identity to the closely related FeLV-A’s. Work from the laboratories of Roy-Burman and Overbaugh examining exFeLV/enFeLV recombination during in vitro infections has revealed that replication efficiency and cellular tropism depends on the length and region of the enFeLV sequence incorporated into the FeLV-B recombinant (Pandey et al., 1991). Amino acid changes localized to two variable regions (VRA and VRB) mediate the ability of FeLV-B to bind to receptors Pit1 and/or Pit 2 (Boomer et al., 1997; Sugai et al., 2001). Aside from changes to the env gene, 37 FeLV-B recombinants have been described that incorporate enFeLV sequences in the LTR region and gag gene (Tzavaras et al., 1990). Curiously, while enFeLV is seen as a necessary progenitor for the generation of FeLV-B, it has also been posited that truncated enFeLV Env may act to interfere with FeLV-B infection (McDougall et al., 1994). Sequence analysis of FeLV-C linked genotypic determinants to disease phenotypes (Riedel et al., 1986). Changes in the FeLV-C 3’ pol and 5’ env gene are associated with aplastic anemia and expand the host range to other species in cell culture (Riedel et al., 1988). Naturally occurring FeLV-C isolates demonstrate that FeLV-C is the result of amino acid changes in the N-terminal portion of the surface protein. Further studies indicated that an 886-bp fragment from FeLV-C encompassing the 3’ end of pol (73 amino acids) and the 5’ end of env (241 amino acids) to a recipient FeLV-A were necessary to confer the fatal aplastic anemia phenotype (Riedel et al., 1988). Subsequent analysis indicated that a three-codon deletion within the first variable region of the vr1 of the 5’ env gene and nine adjacent substitutions may be sufficient to confer virulent phenotype (Borjatsch et al., 1992; Rigby et al., 1992). These findings suggest precise mutations at specific loci may dictate disease phenotypes typically ascribed to FeLV-C. FeLV-61C (aka FeLV-T), is a T-cytopathic FeLV subgroup capable of forming syncytia in 3201 cells, was first isolated in a natural thymic lymphoma (Mullins et al., 1989). FeLV-T induces a fatal immunosuppressive disorder described as FeLV-FAIDS (72). The subgroup was characterized following experimental infections of a domestic cats with a transmissible FeLV clone, 61E (35, 72, 73).. An infected cat subsequently developed thymic lymphoma, atypical of FeLV-A infection, and tissues were analyzed for mutations underlying this phenotype (35). Sequence analysis revealed a variant of primary FeLV-A env containing a 6-amino acid insertion and 6-amino acid deletion (71). Another FeLV variant with a 4-amino acid insertion, (81T), was shown to be sufficient to induce the FeLV-T phenotype (Rohn et al., 1998). This variant, like 61C, was replication-defective (Gwynn et al., 2000). Chimeras generated from 61E and 81T generated tissue culture-adapted isolates with compensatory mutations at positions 7 and 375, rescuing the Env processing ability. These changes both occur outside of the receptor-binding domain (Gwynn et al., 2000). Further research documented that FeLV-T is incapable of membrane fusion 38 to its receptor (Pit1) due to a histidine-aspartate substitution at the N-terminus (Sakaguchi et al., 2015). Infection is possible only in the presence of FeLIX, a truncated envelope protein constitutively produced by enFeLV, which shares greater than 90% identity to FeLV-B env (Anderson et al., 2000). Ultimately, the progressive FeLV-FAIDS disease progression and augmented cellular tropism led to classification the FeLV-T subgroup (Donahue et al., 1991; Rohn et al., 1998). In the late 1980’s, Levesque et al. examined naturally occurring FeLV from a group of cats experiencing lymphomas. One animal had developed a multicentric lymphoma that was non-B-cell non-T-cell in origin (Levesque et al., 1990). LTR recombinants of FeLV-945 and a closely related retrovirus, Moloney murine leukemia virus, were identified in tumor tissue (Starkey et al., 1998). Variant FeLV-945 was shown to have a specific 21-bp tandem triplication repeatedly identified in independent multicentric lymphomas, conferring a replicative advantage in feline cells (Chandhasin et al., 2004; Levy, 2008; Prabhu et al., 1999). The two most recent additions to the FeLV subgroup family include the less characterized FeLV subgroups D and TG35. FeLV-D was identified concurrent with the discovery of a novel domestic cat endogenous retrovirus (ERV-DC) that is divergent from enFeLV (Anai et al., 2012). Transduction of the ERV-DC env gene into FeLV produced FeLV-D that displayed novel receptor interference patterns (Ito et al., 2013). As has been hypothesized with FeLV-B, FeLV-D appears to be restricted by an ERV-DC envelope-like antiretroviral factor termed, Refrex-1 (Ito et al., 2015; Ito et al., 2013). FeLV-TG35 was identified in a 1-year-old castrated male cat. One of several env clones (TG35-2) harbored a 7- amino acid substitution and two amino acid insertions in the vr1. Although the sequence bore resemblance to FeLV-A, interference assays confirmed that TG35-2 Env targeted a different receptor, potentially constituting a new subgroup (Miyake et al., 2016). This review of novel FeLV variants and subgroups demonstrates a wide range of sequence heterogeneity. Some subgroups represent infrequent point mutations, while others represent recombination events resu