Nearly half of the human genome is composed of sequence encoding for elements known as retrotransposons. Although these non-coding RNA elements were initially considered to be “junk,” the last few decades have brought about a new appreciation for the importance of retrotransposons as dynamic players in a variety of host pathways. In Chapter I, I review the evolution of retrotransposons, and their architecture in mammalian genomes. I focus on the Short Interspersed Nuclear Elements (SINEs), a class of retrotransposon derived from host RNA polymerase III (RNAPIII)-transcribed genes. I describe their transcription by RNAPIII, as well as their functional relevance pre- and post-transcriptionally. I introduce B2 SINEs, a prominent murine SINE often used as a model to study SINE biology. These elements are highly induced during times of cellular stress, such as that caused by viral infection. Murine gammaherpesvirus 68 (MHV68) leads to a dramatic upregulation of B2 SINEs, which can influence viral gene transcription. In Chapter II, I present our investigation of the pathway to induction of B2 SINEs by MHV68 during infection, and our finding of the conserved herpesvirus kinase ORF36 as being able to induce these elements independently. In Chapter III, I discuss attempts to elucidate the functional relevance of B2 SINEs induced during infection. I describe an RNA-sequencing experiment done to examine changes in global transcription in the presence or absence of B2 SINEs during infection, as well as an experimental method to isolate B2 ncRNAs induced during infection, for potential use in downstream functional applications. In Chapter IV, I discuss perspectives on the presented data, and future directions for continued elucidation of the link between ORF36 and B2 SINE induction, as well as future directions for studying the functional relevance of induced elements. This study highlights the important role that retrotransposons, in particular SINE elements, play in host and viral pathways, and provides the first description of a herpesviral ORF that is able to induce these elements.

Lytic infection with gamma herpesviruses such as Epstein Barr Virus (EBV) and Kaposi's sarcoma-associated herpesvirus (KSHV) results in a global depletion of cellular mRNA, which manifests in a general suppression of host gene expression, termed host shutoff. The importance of targeting cellular gene expression at the posttranscriptional level is highlighted by the observation that many changes in important cellular gene programs result in drastic changes in RNA stability. Hence, virus targeting of host gene expression at this level likely represents an efficient way to change the cellular environment into one the supports maximal viral production. In KSHV, the host shutoff phenotype is caused by virally encoded multifunctional nuclease, SOX, that is conserved throughout all gamma-herpesviruses. Here I provide detailed mechanistic insight into how SOX targets host messages to enact host shutoff activity in cells. I show that with in the cytoplasm, SOX specifically targets conserved mRNA features, which normally promote their efficient expression. Furthermore, I demonstrate that this mechanism of mRNA targeting is conserved across various other viruses, likely demonstrating an evolutionarily convergent strategy. Characterizing how viruses co-op specific cellular factors involved in gene expression will yield a better understanding of the signals that initiate activation of RNA turnover pathways needed for viral replication success.

Gene expression is generally thought of as a linear pathway, beginning with transcription and processing of messenger RNA (mRNA), followed by export to the cytoplasm for translation and decay. Viruses have long been a key to discovering novel pathways and regulation in cells in their modulation of these pathways and gene expression is no different. Here we describe how infection with gammaherpesviruses is characterized by widespread mRNA degradation that leads to reduced RNA Polymerase II (Pol II) occupancy of host promoters by chromatin immuno- precipitation and sequencing (ChIP-seq) of infected cells. Pol II occupancy depends on RNA decay as infection with a decay-deficient virus does not broadly alter Pol II occupancy. In contrast to host promoters, viral promoters continue to recruit Pol II and transcribe their genes even at late stages in infection. This transcriptional escape of viral genes is all the more notable in light of a massive reduction in components of Pol II and general transcription factors (GTF) protein levels that we describe during infection. We further isolate the transcriptional capacity of viral genes to their presence on the replicating viral genome.

In Chapter 1 of this thesis we provide an overview of RNA decay, transcription and the literature that investigates mechanistic links between these two processes. This situates our work as distinct from the majority of literature that finds RNA decay and transcription are modulated to maintain constant RNA levels in a cell; instead, during pathogenic stress RNA decay feeds forward to further inhibit transcription. Chapter 2 characterizes the transcriptional repression phenotype during murine herpesvirus 68 infection and highlights how viral genes are able to robustly transcribe their genome irrespective of widespread decay. In Chapter 3 we further investigate the protein depletions of Pol II subunits and GTFs that exacerbate transcriptional repression of the host that we uncover in Chapter 2. Chapter 4 delves into the dual roles of SOX, the viral nuclease that instigates RNA decay, as both an RNase and a DNase.

In summary, this dissertation links disparate stages of the gene expression pathway and provides mechanistic details for how this occurs during MHV68 infection.

Alterations in global mRNA decay broadly impact upstream and downstream stages of gene expression, although signals that connect these processes are incompletely defined. Here, we used tandem mass tag labeling coupled with mass spectrometry to reveal that changing the mRNA decay landscape, as frequently occurs during viral infection, results in subcellular redistribution of RNA binding proteins. Accelerating Xrn1-dependent mRNA decay through expression of a gammaherpesviral endonuclease drove nuclear translocation of many RBPs, including poly(A) tail-associated proteins. Conversely, cells lacking Xrn1 exhibited changes in the localization or abundance of numerous factors linked to mRNA turnover. Using these data, we uncovered a new role for cytoplasmic poly(A) binding protein in repressing mammalian TATA-binding protein and RNA polymerase II transcription upon its mRNA decay-induced translocation to the nucleus.

We identified PABPC nuclear-specific protein-protein interactions, and found a number of interactions with proteins involved in the ubiquitin-proteasome system. Two of these protein partners were required to maintain the mRNA decay-transcription feedback pathway. Furthermore, mRNA decay initiated using the decapping enzyme, D10 from vaccinia virus, revealed that this system did not depend on the process by which the accelerated mRNA decay was initiated and instead is responsive to broad changes in mRNA abundance. In addition, we found that PABPC nuclear translocation was not affected by Dis3L2. Together with the finding that PABPC nuclear accumulation alone was not sufficient to repress nascent mRNA synthesis by RNAPII, these data provide evidence that another unknown factor is required, along with PABPC, to fulfill the feedback loop.

Collectively, our results show that changes in cytoplasmic mRNA decay can directly impact protein subcellular localization, providing a mechanism to connect seemingly distal stages of gene expression.

In the beta- and gammaherpesviruses, a specialized complex of viral transcriptional activators (vTAs) coordinate to direct expression of virus-encoded late genes, which are critical for the production of infectious virions. The vTAs in Kaposi’s sarcoma-associated herpesvirus (KSHV) are ORF18, ORF24, ORF30, ORF31, ORF34, and ORF66. While the general organization of the vTA complex has been mapped, the individual roles of these proteins, and how they coordinate to activate late gene promoters, remains largely unknown. Thus, we set out to determine the roles of ORF18, which is a highly interconnected vTA component and of ORF24, which is a putative structural and functional TATA-binding protein (TBP) mimic.

We first performed a comprehensive mutational analysis of the conserved residues in ORF18. The mutants were largely selective for disrupting the interaction with ORF30 but not the other three ORF18 binding partners: ORF31, ORF34, and ORF66. Furthermore, disrupting the ORF18-ORF30 interaction weakened the vTA complex as a whole, and an ORF18 point mutant that failed to bind ORF30 was unable to complement an ORF18 null virus. Thus, contacts between individual vTAs are critical, as even small disruptions in this complex result in profound defects in KSHV late gene expression. These findings underscore how individual interactions between the late gene transcription components are critical for both the stability and function of the complex.

Next, we examined the mechanism by which ORF24 binds RNA polymerase II (Pol II). Previous work in our lab found that ORF24 is a modular protein that binds to viral late gene promoters through its central TBP-like domain. Residues in the N-terminus of ORF24 bind and recruit human Pol II to activate the expression of late genes. We found that ORF24 interacts directly with the heptapeptide repeats of the carboxy terminal domain of Pol II, suggesting that ORF24 may be involved in bringing Pol II to sites of active viral late gene transcription. Collectively, our data highlight how KSHV (and likely other gamma and betaherpesviruses) direct robust transcription of late genes via a unique and streamlined mechanism of pre-initiation complex assembly.

More than 100 types of RNA modifications are known to exist, regulating key aspects of cellular biology and metabolism. First discovered to be present on viral RNAs in the 1970s, methylation at the N6 position of adenosine (m6A) is the most abundant internal modification within eukaryotic mRNAs, and is proposed to be dynamically regulated during times of cellular stress. However, the transcriptome-wide distribution and function of m6A in the lifecycle of a dsDNA virus had not been explored. In Chapter 1, I review how human herpesviruses manipulate host pathways in order to replicate, with particular emphasis on gammaherpesviruses. I also discuss the diverse roles played by RNA modifications in gene expression, focusing on enzymes that install m6A (writers), and proteins that bind m6A modified mRNA (readers). In Chapter 2, I present a detailed investigation of how the m6A pathway impacts the lifecycle of the oncogenic human virus Kaposi’s sarcoma-associated herpesvirus (KSHV). Mass spectrometry and transcriptome-wide m6A-sequencing revealed enrichment of m6A in viral transcripts upon lytic reactivation, including in the lytic transactivator ORF50. Depletion of the writer METTL3 and reader YTHDF2 in three different KSHV reactivation models had differential pro- and anti-viral impacts on viral gene expression depending on the cell-type analyzed. Finally, in Chapter 3, I put my findings in the broader context of the literature examining m6A modifications in viral lifecycles. I discuss technical challenges that have hindered research to-date, and potential reasons for the diverse pro and anti-viral effects ascribed to the m6A pathway, including a discussion of why the m6A pathway may have cell-type specific effects. I highlight the impact of m6A deposition on host transcripts, including those of the innate immune system. Overall, this study demonstrates the critical importance of m6A modifications in regulating KSHV lytic gene expression and sets the stage for future studies to elucidate more mechanistic details, including the potential for discrete impacts at different stages of the viral lifecycle.