Search

Human cytomegalovirus (HCMV) is a ubiquitous herpesvirus residing latently in a majority of the population. While generally asymptomatic in healthy individuals, primary infection or reactivation of latent infection in the immunocompromised, immunosuppressed, and immunonaïve can lead to end organ failure if left untreated. HCMV encodes four G protein-coupled receptors (GPCRs), which play roles in both lytic and latent infection by manipulating the host cell environment, in some cases through altering cell signaling. The objective of these studies was to gain further insight into the role the HCMV-encoded GPCRs, US28 and UL33, play in the molecular mechanisms of lytic and latent infection, and in turn viral dissemination and transmission.In the first study, we demonstrate a new mechanism by which US28 controls HCMV latency. Our lab previously showed US28 is necessary for the establishment and maintenance of HCMV latency. To gain a more complete understanding of the cellular factors US28 manipulates during latency, we evaluated cellular transcriptomic changes in response to US28 expression and found this viral GPCR upregulates the cellular gene, myeloid nuclear differentiation antigen (MNDA). Previous findings revealed MNDA increases DNA-binding of the cellular transcription factor, Yin Yang 1 (YY1), which is a repressor of the major immediate early (MIE) enhancer/promoter, a region whose regulation of immediate early gene expression, in part, controls the balance between HCMV latency and reactivation. We show US28 upregulates MNDA, leading to a US28-dependent interaction between MNDA and YY1. This complex binds the MIE enhancer/promoter, resulting in transcriptional silencing of this locus, thereby aiding in the maintenance of latent infection. These studies reveal a new mechanism by which US28 regulates the MIE enhancer, through the control of cellular proteins MNDA and YY1.Secondly, we evaluated the HCMV-encoded GPCR UL33 and its role in viral replication. Previous studies showed UL33 is dispensable for lytic infection in fibroblasts, however the role of UL33 in cell types required for viral dissemination and transmission, such as epithelial cells, remains unknown. Rodent homologs of HCMV UL33, murine CMV M33 and rat CMV R33, play an important role in viral pathogenesis in their respective hosts, as deletion of these homologs leads to a replication defect in salivary gland epithelial cells in their respective species. To determine if UL33 plays a similar role in HCMV infection of epithelial cells, we evaluated the impact of UL33 on lytic replication in epithelial cells. In our studies, we found UL33 is required for efficient viral growth, as well as viral mRNA and viral protein expression in epithelial cells. Further, we determined UL33 is required for efficient binding of the HCMV virion to epithelial cells, which in turn leads to decreased entry and virion delivery. These studies suggest the HCMV-encoded GPCR, UL33, is important for efficient lytic replication in epithelial cells.

Coronaviruses (CoVs) are a diverse family of enveloped viruses responsible for causing respiratory and/or enteric diseases across a wide range of species. The expansive animal host range of CoVs can be attributed to their ability to adapt to diverse cellular environments (i.e. pH, ions, proteases, receptors), and exploit different entry pathways to mediate viral-host cell membrane fusion and infect cells. Ions have arisen as an important factor in the viral membrane fusion mechanism. Within the CoV family, our team has found that calcium availability leads to increased virus infection of cells. Previously, we observed that calcium availability also promotes insertion of SARS FP into the host cell bilayer and subsequent membrane lipid ordering, pointing to the possible role of calcium in interacting directly with the FP during CoV membrane fusion activity. Due to the highly conserved nature of the FP within the CoV family, we chose to investigate the specific binding pockets of Ca2+ within SARS- and MERS-CoV. We used site-directed mutagenesis and infectivity assays to pinpoint specific residues that lead to changes in infectivity when calcium is present or not. We identified potential calcium-binding residues by substituting the charged residues (i.e. aspartic acid, glutamic acid) in the FP with non-charged amino residue, alanine. We compared the infectivity of mutant and wild-typed CoV pseudoparticles under calcium-rich or poor environments using available FDA-approved calcium blocking drugs. Our data suggests very similar residue coordination between SARS-CoV and SARS-CoV-2 as both FPs binds to two Ca2+. Overall, these results demonstrate that Ca2+ have specific interactions with the CoV FP and opens the possibility of utilizing FDA-approved calcium blocking drugs as a potential treatment against COVID-19.

Despite nearly three decades of research, not much is known regarding the early stages of development for KSHV lymphoproliferative disorders, hindering our ability to develop prophylactic measures or effective treatments. This dissertation will focus on the host and viral factors influencing the magnitude and dynamics of KSHV infection in the human tonsil to pave the way for future interventions directed at limiting person-to-person transmission of KSHV. To understand the contribution of host factors to KSHV susceptibility in B lymphocytes, we generated a library of 40 tonsillar specimens. Our results indicate that the immunological composition of tonsillar lymphocytes varies across our donor samples, and KSHV possess a diverse B lymphocyte tropism. Furthermore, the highly specific targeting of plasma cells is not due to CD138 (Syndecan-1) being used as an attachment factor, and heparan sulfates, in general, do not play an important role in KSHV infection of B cells. Finally, the donor-dependent immunological factors and immune status of individual samples influences the overall susceptibility, as well as specific targeting of B cell lineages. To understand the significance of viral gH/gL glycoprotein interaction with some of the key EphA family of receptors in the process of KSHV entry into B lymphocytes, we demonstrate that the expression of EphA2, EphA4, and EphA7 are donor and subset-specific and gH/gL-EphA interactions are important for KSHV-WT infection of primary tonsil lymphocytes. These interactions are critical for establishing KSHV-WT infection of plasma cells and germinal center cells, and KSHV exploits alternative mechanism of entry in absence of gH. Finally, we develop a model for cell-to-cell transmission of KSHV within human tonsils by examining KSHV spread within and between primary cell types derived from tonsil specimens, located at the proximity of the crypt lumen, which are in direct contact with external pathogens present in saliva. We show that a variety of primary tonsillar cells are susceptible to KSHV infection and can differentially transmit the infection into B cells. We demonstrate that KSHV spread between and within tonsil cell types is directed, suggesting, this tropism may be influenced by differential virion composition arising from each cell type.

The Paramyxoviridae family includes many RNA viruses that collectively have a broad host range, including birds, fish, and mammals. Paramyxoviruses can productively infect a wide range of animal hosts and may be transmitted through the respiratory tract. The mammalian paramyxoviruses, Nipah virus (NiV) and Feline Morbillivirus (FeMV) belong to the Henipavirus and Morbillivirus genera, respectively. In addition to the highly virulent NiV, over twenty henipa-like viruses have recently been discovered globally. There are three additional species in the Henipavirus genus, with less than a decade since their discovery, including Cedar virus (CedV), Ghana virus (GhV), and Mojiang virus (MojV). Henipaviral host cell entry is coordinated by two viral surface glycoproteins: the tetrameric receptor binding protein (G) and the trimeric fusion glycoprotein (F). G consists of four domains: the cytoplasmic tail, transmembrane, extracellular stalk, and globular head. For henipavirus entry to occur, G interacts with the host cell receptor and undergoes receptor-induced conformational changes during which the stalk of G is exposed to F. The interaction between the head of F and G stalk domains leads to a rearrangement of the central region of the G stalk. This triggers F to initiate the fusion cascade, resulting in virus-cell membrane fusion, which is essential for viral entry or cell-cell membrane fusion. The G stalk is involved in maintaining an oligomeric structure, mediated by a distinct cluster of three cysteine residues. The roles of the stalk domain in other henipaviral glycoproteins, including MojV, remain unknown. Our main objective is to understand the role of the stalk domain within members of the Paramyxoviridae family during fusion activation and host cell entry. Overall, the results obtained from this research inform about the novel role G has in addition to receptor binding and fusion triggering in the fusion pore expansion step for NiV. Additionally, we discovered the MojV stalk cysteine residues are involved in oligomerization and modulate fusogenicity. By deepening our understanding of the mechanisms for paramyxovirus glycoprotein-driven membrane fusion, this research supports new methods to stop the infection and syncytia-associated pathogenesis and provide new targets for antivirals.

The innate immune system is critical in blocking the first steps of infection, thus preventing robust pathogen growth. Innate immune responses are primarily mediated by pathogen recognition receptors (PRRs), which sense molecular patterns that warn of microbial infection or nearby tissue damage. One indicator of tissue damage is the release and oxidation of the common membrane lipid 1-palmitoyl-2-arachidonoyl-sn-phosphatidylcholine (PAPC) to form oxPAPC, a heterogenous mix of oxidized lipids. While oxPAPC is well known to induce hyperactive states in primed professional immune cells, very little is known about how it interacts with non-immune cells and with the adaptive immune system as a whole. To explore the impact of oxidized phospholipids on these compartments, we assessed alterations in cytokine production and antibody expression in in vitro and in vivo settings induced by oxidized phospholipids, along with viral replication in cells that were pretreated with these oxidized phospholipids. Our studies indicated that oxidized phospholipids can, in vitro, lead to increased production of adaptive immune stimulating cytokines such as interleukin 1. Further, when included in an experimental admixture in OVA sensitization experiments in mice, oxidized phospholipids increase antibody titers. This augmentation of the antibody response was observed in the blood, in the gut mucosa, and in the pulmonary mucosa. When epithelial cells from oxygenated tissues such as the lung epithelium and the mouth were treated with oxidized phospholipids, they became refractory to infection by the RNA viruses Influenza and Vesicular Stomatitis Virus. This viral restriction occurred in a manner independent of interferon signaling. These data revealed a limitation of viral entry, restricting viral infection by preventing viruses from infecting the cells in the first place rather than inhibiting viral processes within the cell. This resulted in a smaller initial pool of infected cells and resulted in a kinetic defect in viral replication. Together, these data suggest a multifaced role for oxidized phospholipids in modulating both the innate and adaptive immune responses.

Cells are complex biomechanical systems that constantly adapt to and modify their physical environment. The cell membrane is the boundary that separates organelles and other molecular machineries in a cell from the extracellular environment and plays a major role in maintaining the biochemical and biomechanical homeostasis of cells. The constant deformation and remodeling of cell membranes is key in biological processes like cell division, polarization and migration. It also enables the trafficking of a diverse ensemble of cargos in and out of cells by endocytosis and exocytosis, respectively. This remodeling process is regulated by mechanical factors like membrane tension, rigidity, substrate stiffness, morphology of the cargo,…etc. Clathrin-mediated endocytosis (CME) is an endocytic pathway by which cells internalize cargos from the extracellular environment by extensive remodeling of the plasma membrane from a flat to a bud morphology supported by clathrin-coat assembly. Epsin is an adaptor protein in CME that aids in membrane bending and is recruited prior to clathrin to clathrin-coated pit (CCP) nucleation sites. Epsin has a structured N-terminal ENTH (epsin N-terminal homology) region and an unstructured C-terminal IDP (intrinsically disordered protein) region that are involved in membrane bending via amphipathic helix insertion and steric repulsion, respectively. Membrane tension plays an inhibitory role in CME by destabilizing CCP assembly. My investigation shows that overexpressing epsin rescues the stability of CCPs under high membrane tension. The recruitment of epsin to CCPs increases with elevated membrane tension. Masking the N-terminus of ENTH with EGFP or deletion of ENTH in epsin delays recruitment of epsin to CCP nucleation sites. Interfering with the activity of H0 helix abrogates the elevated recruitment of epsin at high tension. The ENTH domain, while necessary for tension sensitivity of epsin, is not necessary for epsin puncta formation. Deletion of adaptor protein 2 binding sites and clathrin binding sites in the IDP domain of epsin renders it cytosolic. Further, the IDP domain itself is sufficient to impart stability to CCPs as CCPs recruiting epsin with only IDP domain shows reduction in the fraction of abortive pits. My findings conclude that H0 alpha helix in the ENTH domain of epsin acts as a tension sensor and the IDP domain acts as a tether stabilizing CCPs. Our work reveals the complimentary action of structured ENTH and unstructured IDP domains of epsin supporst CME at high tension. CME is also utilized by pathogens like Influenza A virus (IAV) to gain access to host cells. Epsin is a known cargo-specific adaptor for IAVs. Epsin interacts with ubiquitinated surface receptors bound to IAVs via its ubiquitin-interacting motifs. My investigation shows that the ENTH domain in epsin biomechanically regulates curvature generation around spherical IAVs thereby initiating CME of the virion. Mutations in H0 helix negatively regulate the recruitment of epsin to CCPs containing spherical IAVs and their subsequent internalization. However, internalization of filament-forming IAVs were not affected by the inhibition of H0 helix formation in ENTH domain of epsin. In summary, the findings of this dissertation define the biomechanical role of epsin in CME under high membrane tension conditions and during viral entry. By improving our understanding of biomechanics of endocytosis and viral entry, I hope that this research will lead to development of better therapies targeting abnormal endocytosis pathways and viral diseases, as well as advances in viral vector and nanoparticle-based drug delivery systems.

Henipaviral entry and infection occurs by direct virus-host cell membrane fusion by a similar mechanism, the formation of pathological syncytia occurs by cell-cell membrane fusion. Both viral-cell and cell-cell membrane fusion are coordinated by the receptor-binding attachment (G) and fusion (F) glycoproteins. Typically, the attachment protein binds the cognate host-cell receptor (ephrinB2 or ephrinB3) and undergoes a series of conformational changes that ultimately leads to exposure of the stalk domain that in turn triggers the fusion protein. In due course, the fusion protein proceeds to undergo a series of conformational changes as it progresses through the fusion cascade. Using henipaviruses as a model for the broader Paramyxovirus family, we identified novel domains between the henipaviral G and F proteins that affect fusion. Further, we identified conserved henipaviral regions within the fusion glycoprotein that affect distinct steps of the fusion cascade. First, we examine the fusion capacities between homologous and heterologous pairs of Nipah (NiV) and Hendra (HeV) G and F proteins and found that the heterologous pair of NiV-G and HeV-F is hyperfusogenic. We further determine the overall domains in G and F that contribute to affects in fusion through construction of NiV and HeV G and F chimeras. Secondly, we examine the roles of several F regions: the HR3, N1 and N4 in membrane fusion. These regions are initially in close proximity to the fusion peptide and/or undergo secondary structural changes between the pre-fusion and post-fusion F conformations. Here, by performing alanine scan mutagenesis and various functional analyses, we report that specific residues within these domains modulate the fusogenic capacity of NiV-F. Importantly, we defined the steps in the fusion cascade that these mutants affect. The roles of these domains and regions in the membrane fusion cascade have never been described before and may be common amongst other paramyxoviruses, thus providing promising targets for drug design. Moreover, our studies provide a new important understanding of how certain domains between G and F and certain regions within F modulate fusion.

Enveloped viruses rely heavily on their ability to merge the viral membrane to the host cell membrane during viral entry. Therefore, improving the understanding of membrane components utilized in viral fusion is crucial. Additionally, the development of broad-spectrum antivirals capable of inhibiting the entry of zoonotic viruses that cause a tremendous burden to human health is critically important. Nipah virus is a deadly enveloped virus with a high mortality rate in humans and agriculturally important animals. Despite the high case-fatality rates of the Nipah virus and many other enveloped viruses (i.e., Ebola, SARS, MERS), there is a lack of approved antiviral therapeutics. Various chemical agents capable of altering the viral membrane composition have revealed a valuable target in antiviral drug design. Therefore, studying the role of viral membrane components is vital in understanding fusion mechanics and for the development of potential antivirals. Herein, I present a series of experiments used to study virus-induced membrane fusion. First, to understand the role of membrane cholesterol in Nipah virus fusion and egress, I performed various biochemical assays that revealed that membrane cholesterol influences the levels of virus-cell and cell-cell membrane fusion. Second, I found that modulating the concentration of membrane cholesterol alters both the incorporation of Nipah virus proteins into lipid rafts and the budding efficiency of Nipah virus-like particles. Lastly, we characterized a novel class of broad-spectrum antivirals, the XM series, aimed at inhibiting viral membrane fusion. We found that XM-01 increases order in the viral membrane, which significantly reduced or completely inhibited viral entry. These inhibitors can cause various enveloped viruses to be fusogenically inert while leaving immunogenic sites intact. This effect allowed us to test the ability of these compounds to produce inactivated-virus vaccines, tested for the Influenza virus. Altogether, these findings progress the understanding of membrane components in viral entry and release, and the development of effective antivirals.

Middle East Respiratory Syndrome coronavirus (MERS-CoV) infections are associated with a significant mortality rate, and existing drugs show poor efficacy. Identifying novel targets/pathways required for MERS-CoV infectivity is therefore important for developing novel therapeutics. As an enveloped virus, translocation through the endolysosomal system provides one pathway for cellular entry of MERS-CoV. In this context, Ca2+-permeable channels within the endolysosomal system regulate both the luminal environment and trafficking events, meriting investigation of their role in regulating processing and trafficking of MERS-CoV. Knockdown of endogenous two-pore channels (TPCs), targets for the Ca2+ mobilizing second messenger NAADP, impaired infectivity in a MERS-CoV spike pseudovirus translocation assay. This effect was selective as knockdown of the lysosomal cation channel mucolipin-1 (TRPML1) was without effect. Pharmacological inhibition of NAADP-evoked Ca2+ release using several bisbenzylisoquinoline alkaloids also blocked MERS pseudovirus translocation. Knockdown of TPC1 (biased endosomally) or TPC2 (biased lysosomally) decreased the activity of furin, a protease which facilitates MERS-CoV fusion with cellular membranes. Pharmacological or genetic inhibition of TPC1 activity also inhibited endosomal motility impairing pseudovirus progression through the endolysosomal system. Following this initial validation of NAADP-gated TPCs supporting pseudotyped MERS-CoV infectivity, a high throughput drug screen of 1534 compounds was conducted in a sea urchin egg homogenate system, yielding eighteen‘hits’ exhibiting >80% inhibition of NAADP-evoked Ca2+ release. A validation pipeline for these candidates yielded seven drugs that inhibited NAADP-evoked Ca2+ release without depleting acidic Ca2+ stores in a human cell line. These candidates displayed a similar penetrance of inhibition in both the sea urchin system and the human cell line, and the extent of inhibition of NAADP-evoked Ca2+ signals strongly correlated with observed inhibition of infectivity of a MERS-CoV pseudovirus. This work highlights the utility of targeting regulators of intracellular trafficking as a novel therapeutic strategy, and provide strong support for the execution of a higher throughput screening campaign using the sea urchin egg homogenate system to discover new ligands for manipulation of NAADP-gated TPC signaling.

Abstract: Most antivirals target viral proteins and are specific for only one virus, or even one viral genotype. Whereas viral proteins are encoded in the plastic viral genome, virion lipids are not and their rearrangements during fusion of the virion envelope to cellular membranes are conserved among otherwise unrelated enveloped viruses. Antivirals that inhibit these lipid rearrangements could thus pose a high barrier to resistance and have broad-spectrum activities. Fusion occurs through a hemifusion stalk in which only the outer leaflets are fused, bent with their polar heads forming a smaller radius than their hydrophobic tails (negative curvature). Outer leaflets enriched in phospholipids with head groups of larger cross sections than their lipid tails (“inverted cone”) disfavor this negative curvature, inhibiting fusion. The rigid amphipathic fusion inhibitors (RAFIs) are synthetic compounds of inverted cone molecular geometry. They inhibit the infectivity of otherwise unrelated enveloped viruses. The leading RAFI aUY11 has an ethynyl-perylene hydrophobic and an uracil-arabinose polar moiety. aUY11 intercalates in viral envelopes to directly inhibit virion-to-cell fusion. Previous studies showed that amphipathicity, rigidity, and inverted cone molecular geometry are required. My hypothesis is that the inverted cone molecular geometry of the RAFIs increases the energy barrier for the hemifusion stalk, inhibiting fusion. Then, chemically distinct compounds with similar amphipathicity, rigidity, and inverted cone shape would have similar antiviral potencies, regardless of specific chemical groups. Alternatively, the perylene group exposed to visible light may induce viral lipid peroxidation. Then, the perylene group and absorbance at visible spectrum would be required. I evaluated the activities of twenty-five chemically distinct RAFIs with similar amphipathicity, rigidity, and inverted cone shape. The perylene moiety and absorption at visible spectrum were not required, but a minimum length of the hydrophobic moiety of 10.3 Å was. The arabino moiety could be modified or replaced by other groups. Cytidine replacement for uracil was not tolerated, indicating the O8 carbonyl group is required. Bilayer intercalation was required but not sufficient. The vast majority of RAFIs had no overt cytotoxicity (CC50 > 20 μM; TI > 250−1200). Carbonyl or butylamide substitutions for arabino or cytidine replacement for uracil, increased cytotoxicity. Cytotoxicity was mainly determined by the polar moiety and there was no correlation between antiviral and cytostatic activities. The definition of the effects of shape and chemical groups on the activity of the RAFIs opens the possibility for rational design of lipid-acting antivirals active against a broad spectrum of enveloped viruses.

Human cytomegalovirus (HCMV) causes morbidity and mortality in congenitally infected newborns, transplant recipients, and AIDS patients. Currently, there is no approved CMV vaccine to address these issues. In an effort to develop an alternative treatment to CMV we tested our hypothesis that heparan sulfate binding D-peptides would be effective against multiple HCMV strains in vitro and it would be effective in vivo against murine CMV (MCMV) (Chapter 1). We show that the D-peptide is able to reduce CMV infection in vitro and in vivo. Another approach to combating CMV infections is to neutralize pathogenic factors that contribute to CMV spread and/or pathogenesis. HCMV contains genes with homology to human immune modulators. These genes have been implicated in establishing lifelong HCMV latency. One of these factors is an HCMV expressed viral chemokine (vCXCL-1). CXCL-1 binds to chemokine receptors expressed on host immune cells. In order to understand how vCXCL-1 is involved in CMV pathogenesis we used recombinant MCMV overexpressing chimpanzee CMV vCXCL-1 (vCXCL-1CCMV) and murine CXCL-1 (KC). We hypothesized that CMV encodes vCXCL-1 to aid in viral dissemination (Chapter 2). In contrast we found that over expression of CXC chemokines recruits innate immune cells to the site of infection and primary dissemination organs leading to a dissemination blockade. vCXCL-1 is encoded by the HCMV UL146 gene, which has shown variation among clinical isolates. This variation may attribute contribute to varying clinical outcomes of HCMV infections. We hypothesize that variation in vCXCL-1 leads to differential activation of neutrophils (Chapter 3). Throughout this thesis we have explored how the virus spreads, causes disease and a potential peptide that could be used to prevent the initial CMV infection.

Abstract: Viral entry is an attractive antiviral target. Entry inhibitors prevent infection of healthy cells and inhibit viral replication before viruses establish persistent reservoirs. Two entry steps, the primary attachment to cellular glycans and the lipid rearrangements during fusion, are conserved among many unrelated viruses. Therefore, inhibitors of these entry steps, acting through appropriate mechanisms, are likely to have broad-spectrum antiviral activity. Using small molecules as probes, I identified and characterized three mechanisms by which it is possible to inhibit the entry of unrelated viruses. The majority of human viruses, including enveloped and nonenveloped viruses, initially bind to cellular glycans. Epigallocatechin gallate (EGCG), a green tea polyphenol, competes for virion binding to heparan sulfate or sialic acid moieties in cellular glycans to inhibit the infectivity of most human viruses. All enveloped viruses rely on lipid rearrangements during entry steps. Rigid amphipathic fusion inhibitors (RAFIs) act through biophysical mechanisms to inhibit the formation of the negative membrane curvature required for the fusion of enveloped viruses. Enveloped viruses constitute a large group of human viruses, including most clinically important pathogens. Curcumin and 25-hydroxycholesterol (25HC) modulate the fluidity and composition of lipid membranes to interfere with the replication of hepatitis C virus (HCV) and other enveloped viruses. None of these small molecules are ideal candidates for antiviral drugs. More importantly, the identification of these mechanisms opens the possibility for the rational design of small molecule entry inhibitors with broad-spectrum antiviral activities and appropriate pharmacological properties. Furthermore, the small molecules described in this thesis are useful as probes to characterize viral entry steps.

The entry of HIV-1 into target cells and viral spread among cells of the immune system are two critical events in the viral life cycle. This study aimed to further characterise HIV-1 pathogenesis three-fold. Firstly, development of fluorescent markers for HIV-1 content and lipid resulted in a novel system to generate fluorescent HIV-1 particles with near wild-type infectivity for use in single particle imaging. Through stringent analysis of the viral progeny, two key caveats of HIV-1 particle tracking were identified. Firstly, the viral content was compartmentalised both within and outside of the HIV-1 capsid core. Also, although improvements were made to the lipid label, significant numbers of fluorescent microvesicles that co-associate with HIV-1 were recognised. These key findings influence the interpretation of these particles when used for single particle tracking. Secondly, this fluorescent virus was used to characterise one of the most potent forms of viral spread, transfer across the dendritic cell and CD4+ T cell synapse. In this context, a new pathogenic structure was identified, which we now term HIV-1 filopodia. These finger-like structures protrude from the membrane and are used by the newly formed virions as a means of cell-cell transfer. The spread of HIV-1 via viral filopodia was found to be reliant on the viral protein Nef, specifically in the domain that interacts with the cellular protein Dynamin-II. Finally, using a panel of novel small molecule inhibitors, along with well characterised virological methods, it was investigated whether endocytosis was a pre-requisite for entry into primary cell targets of HIV-1. Significant post-entry blocks of viral infection were detected when using Dynamin-II inhibitors. In studies of primary cells, limited inhibition of HIV-1 entry was observed in macrophages and resting CD4+ T cells, while viral entry was blocked in HeLa-based and activated CD4+ T cells. This thesis provides new, stringent techniques for visualising HIV-1 entry and spread at the cellular level. Using this foundation, new viral pathogenic structures we now call HIV-1 filopodia were identified. In addition, this foundation has further demonstrated the limits to using single particle tracking to dissect viral entry. Given these limitations, an alternative approach was developed to study HIV-1 entry and it was ultimately observed that endocytosis of HIV-1 is not a pre-requisite for HIV-1 fusion and subsequent infection.

Respiratory syncytial virus (RSV), a paramyxovirus, is the most significant respiratory pathogen in infants and causes 90,000 emergency hospitalizations in the United States and 160,000 deaths worldwide every year. It is also a leading respiratory pathogen in the elderly. RSV infection provides weak immunity, and the virus infects individuals repeatedly throughout life. The only effective antiviral compound commercially available is a monoclonal antibody, palivizumab (Synagis®), which is given prophylactically to at-risk infants. Despite intense efforts, no other therapeutic and no vaccines have been approved for use, although several are in development. Most drugs developed against RSV target the fusion (F) glycoprotein. The F protein is responsible for fusing the host cell and viral membranes together to initiate infection. The RSV F protein is unique among other paramyxoviruses in that it does not require its partner attachment (G) glycoprotein to function as do most other paramyxovirus F proteins. It is the subject of extensive research in hopes of determining its mechanism of action and developing anti-RSV compounds. The mature, cleaved RSV F protein contains N-glycans at asparagine residues 27, 70, and 500. It has previously been reported that none of these N-glycans are required for protein processing or cell surface expression, but that the N500 glycan is required for cell-to-cell fusion in an assay serving as a proxy for virion-cell fusion. In this study, we replicated these results in the full-length protein. We also built N27Q, N70Q, N500Q, N27/70Q, and N/27/70/500 mutations into a soluble version of the RSV F protein to determine the role, if any, that these N-glycans have in triggering. Following purification, each of the N-glycan mutants migrated near the top of a sucrose velocity sedimentation gradient, similar to the wild type (WT) sF protein, indicating that these proteins were produced and secreted from the cell in the prefusion form. We determined that none of these glycans were responsible for maintaining the prefusion form, and that premature triggering is not the reason for the fusion deficiency of the N500Q mutant. We have previously determined that exposure to low molarity triggers the RSV F protein, resulting in the exposure of its hydrophobic fusion peptide and resultant aggregation. Following dialysis of the WT and mutant sF proteins into a 50 mM Hepes buffer, each migrated to the bottom of a sucrose velocity sedimentation gradient, indicating that the proteins had triggered and aggregated by their hydrophobic fusion peptides. All of the N-glycan sF proteins were triggerable, indicating that the inability of the N500Q mutant to fuse is not the result of an inability to trigger and is likely due instead to an inability to refold properly following triggering. The N500Q glycan is attached to the HRB domain of the RSV F protein and may guide the HRB as it completes formation of the 6-helix bundle, a characteristic form of the postfusion F protein that brings the virion and cell membranes into close proximity and results in the fusion of these two membranes.

Influenza is an important human pathogen that causes fatal disease in 250,000-500,000 people worldwide each year. Because of high levels of variation between influenza strains, vaccines are not always effective and must be administered annually. Influenza virus, which replicates primarily in the lung epithelium, encodes only 10 proteins and relies heavily on host products to replicate. Determining which cellular factors are important for influenza replication represents an important area of virology and cellular biology research, and could elucidate proteins or pathways to target for antiviral therapies. We developed a high throughput screening method in primary human bronchial epithelium (HBECs) to identify novel regulators of influenza replication. We first used this method to functionally examine 1745 genes that were identified as potential influenza regulators due to transcriptional regulation by virus or viral products, direct interaction with viral proteins via yeast two-hybrid, or through computational analysis. This screen confirmed some known regulators of influenza replication while identifying novel viral interactors as influenza regulators (e.g. USHBP1, ZMAT4). We also found that the WNT, p53, and ER stress pathways, among others, affect viral replication and interferon production. The life cycle of influenza involves extensive intracellular trafficking of viral components. We again used RNAi to systematically examine the roles of vesicle, RNA, and protein trafficking genes in the production of infectious influenza A virus in primary lung cells. Among the factors that significantly impact viral infection, we identify a set of five genes with strong antiviral effects that are mutated in patients with Hermansky-Pudlak syndrome (HPS). Depletion of HPS genes leads to elevated viral RNA at an early stage of influenza infection prior to transcription. In contrast, depletion of these genes does not alter the innate immune response to virus or interferon. Using an HPS-1 patient cell line, we find an increase in viral fusion to endosomal compartments but no change in viral binding to the cell surface or entry into the early endosome. Our studies uncover a potential role for many trafficking factors in the influenza life cycle, and point to an HPS1-dependent process that inhibits viral entry prior to viral membrane fusion.

To enter and infect cells, viruses must overcome the barrier presented by the cell membrane. Enveloped viruses, which possess their own lipid bilayer, fuse their viral membrane with the cell membrane. Non-enveloped viruses, whose outer surface is composed of proteins, penetrate through the hydrophobic interior of the cell membrane. Viruses accomplish the processes by coupling conformational changes in viral "entry proteins" to membrane disruption. This dissertation investigates the membrane disruption mechanisms of rotavirus, a non-enveloped virus, and vesicular stomatitis virus (VSV), an enveloped virus. Rotavirus uses proteins of its outer capsid to penetrate the membrane and deliver a transcriptionally-active core particle into the cell cytoplasm. \(VP5^*\), an outer capsid protein, undergoes a foldback rearrangement that translocates three clustered hydrophobic loops by \(\sim 180^{\circ}\). This rearrangement resembles the foldback rearrangements of enveloped virus fusion proteins. In the first half of my dissertation, I show that the hydrophobicity of the \(VP5^*\) apex is required for membrane disruption during rotavirus cell entry by mutating hydrophobic residues within the loop to hydrophilic residues. One particular mutation diminishes liposome interaction by the protein, blocks membrane penetration by virus particles in cells, and reduces particle infectivity by 10,000-fold. VSV uses its fusion protein, G, to fuse at low pH. Unlike other viral fusion proteins, pH-induced conformational changes in G are reversible. In the second half of my dissertation, I measure the fusion kinetics of individual VSV particles using a single-particle fusion assay previously developed for influenza virus. I find that hemifusion by VSV consists of at least two steps, an initial step that is pH-dependent and reversible, and a second step that is pH-independent. At low pHs, the second step becomes the sole rate-limiting step. I also show that at pH 6.6, the VSV particle enters a stable intermediate state that binds tightly to membranes but does not precede to fusion. This dissertation uses a variety of experimental approaches to arrive at a more detailed understanding of how viruses use their entry proteins to either penetrate or fuse with the cell membrane.

Efficient entry is a critical determinant of reovirus oncolysis. While a model for the entry of reovirus virions has been described, the method by which partially uncoated intermediate subviral particles (ISVPs) enter host cells remains unknown. Biochemical studies have suggested that ISVPs can directly penetrate the plasma membrane since these particles can cause membrane damage and hemolysis at high multiplicities of infection. We used a combination of techniques to show that these particles can use endocytosis to enter and productively infect host cells. We found that uptake of reovirus virions and ISVPs was inhibited in cells treated with a small molecule inhibitor of dynasore. Analysis of reovirus replication in cells treated with genistein and in cells expressing dominantnegative caveolin-1 revealed that ISVPs use dynamin-dependent caveolar endocytosis to enter host cells. These studies also showed that reovirus virions were able to take advantage of caveolar endocytosis to infect cells. Because many viruses have recently been found to enter cells through dynamindependent and –independent endocytic pathways, we also assessed the role of cholesterol in reovirus infection. The growth of reovirus virions, but not ISVPs, was inhibited in cells treated with methyl-β-cyclodextrin, which extracts membrane cholesterol. We also wanted to learn whether reovirus entry had cell type-specific differences. Analysis of reovirus infection in a rat embryo fibroblast (CREF) and Ras-transformed CREF cell line suggested that reovirus particles lose the ability to use clathrin-mediated endocytosis in CREF cells following Ras transformation. Reovirus can also infect cells that lack caveolae. To assess whether reovirus used different endocytic pathways to infect cells devoid of caveolae, we analyzed reovirus entry and infection in polarized Caco-2 cells. While reovirus infection of these cells was dynamin-independent, particle internalization occurred through both dynamin-dependent and –independent pathways. Visualization of particle uptake showed that particles internalized through dynamin-dependent endocytosis were transcytosed to the basolateral surface. These findings have demonstrated that reovirus can use a diverse set of endocytic pathways to enter and infect cells. The data presented in this thesis show that reovirus entry is a complex process that can vary in different cell types. In addition, it has revealed that particles internalized into various endocytic pathways are trafficked to different destinations in polarized epithelial cells.

Abstract: The chimeric Alb-uPA SCID mouse that has been transplanted with human hepatocytes is a model to facilitate in vivo study of HCV. We explored further development of the model by using repopulation with immortalized human hepatocytes (IHH) in place of primary human hepatocyte (PHH) transplantation to support HCV infection. In vitro HCV studies typically utilize a human hepatoma cell line (Huh7) and rely on transfection with transcribed genomic RNA derived from a unique HCV strain (JFH1). Unfortunately, this system has not been successful in support of infection with serum-derived HCV (HCVser). IHH may offer an alternative since their differentiation status remains close to that of PHH. IHH transfected with HCV RNA (H77 or JFH1) or infected with HCVser showed stable intracellular and supernatant HCV RNA by real-time RT-PCR. IHH showed intracellular HCV NS3 proteins. HCV transfected or infected IHH secrete infectious HCVcc for in vivo and vitro.

Vesicular stomatitis virus (VSV) is an enveloped, nonsegmented, negative-sense, RNA virus belonging to the Rhabdoviridae family. VSV is considered the prototypic Rhabdovirus due to its simple genetic organization, broad host cell tropism, and ability to be easily grown in cell culture. Therefore, VSV has been used as the prototype to study viral entry, transcription, replication, and assembly. Viral entry, a critical step in the lifecycle of the virus, is mediated by the outer surface protein, G and will be the focus of this dissertation. We hypothesize that the highly conserved residues in the membrane-proximal region of VSV G protein are critical to membrane fusion through participation in low pH-induced conformational changes and stability of theses structures. This hypothesis was tested initially through the use of site directed mutagenesis and then by studying rapidly acquired second site mutations. Residues that are conserved completely or conserved for their biochemical properties were either deleted or mutated to alanines. The mutated G proteins were examined through the use of transient transfection assays for their surface expression levels, fusion capacity, and ability to undergo pH dependent conformational changes. Mutation of a conserved HPH motif (H423/P424/H425) resulted in a dramatic decrease in surface expression. ΔH423/P424/H425 was completely fusion defective as assessed by syncytium formation assays that measured cell-cell fusion. Likewise, the cell-cell fusion activity for ΔH425 was affected in that the pH threshold required to trigger the fusion event was decreased. These findings support previous reports suggesting a requirement for histidine protonation in order for the pH dependent conformational changes needed for membrane fusion to occur. Recombinant viruses encoding the mutated G proteins were recovered and all of the viruses expressing G proteins with reduced cell-cell fusion capacity, as compared to wild type, grew to lower titers, with the exception of one mutant: D435A/D436A/E437A. Mutation of the DDE motif resulted in a G protein with a reduced capacity to fuse in transiently transfected cells. However, recombinant virus encoding D435A/D436A/E437A was recovered and unexpectedly grew to wild type titers, suggesting that the virus is overcoming the fusion defect caused by the mutation. Wild type VSV virions have been shown to express a truncated G protein on their surface, called G-stem (GS). GS was detected on the virion of this mutant at levels higher than that of wild type suggesting the truncated G protein was contributing to enhancing the infectivity of the virus. Mutation of conserved phenylalanines (F440A/F441A) also resulted in a similar phenotype as the ΔH425 mutant. Reduced surface expression was observed when this mutated G protein was transiently expressed in cells and a decrease in the threshold pH required for fusion was also observed. Recombinant VSVs encoding the mutated G proteins were recovered to analyze the impact of the mutations on the ability of the viruses to grow and spread. Interestingly, viruses encoding the ΔH425 and the F440A/F441A mutant G proteins rapidly acquired additional mutations that made the viruses better able to grow and spread as indicated by increased titers and plaque sizes. Plaque isolates were obtained and subjected to sequence analysis revealing several variants: ΔH425/S438I, ΔHI426S/S438I, F440V/F441A, F440T/F441A, F440A/F441A/S438I, and F440A/F441A/CT9. To identify the advantage the additional mutations were providing the recombinant viruses (if any), the mutations were cloned into expression vectors and the mutant G proteins were examined for their surface expression levels and fusion capacity. The mutations appeared to enhance cell surface expression levels, with the exception of the ΔH425/S438I and F440A/F441A/CT9 mutants. The F440A/F441A mutants still were reduced in cell-cell fusion activity at pH 6.0 as compared to wild type G. However, the ΔH425 mutants’ fusion activity was partially restored at pH 6.0. Results from these experiments suggest that when both the ΔH425 and F440A/F441A mutant G acquire a S438I mutation the fusion capacity at pH 6.0 is restored. The findings in these studies indicate residues in the membrane-proximal region of VSV G are critical to the fusion process, suggesting a likely contribution to the stabilization of the pre- and postfusion structures.