Type 3 Dearing (T3D) reovirus is an oncolytic virus able to lyse a majority of tumor types. The major drawback is that ∼20% of tumor types are resistant to lysis by T3D and therefore there is a need to improve the oncolytic ability of reovirus to include resistant tumors. When designing an oncolytic virus, an ideal virus would be sensitive to double stranded protein kinase (PKR) pathway involved in the anti-viral response of normal cells. Previous studies in our lab determined which gene segments were involved in generating high viral yields in the absence of PKR. The M1 gene segment from T3D controls PKR sensitivity and the Type 1 Lange (T1L) L3 gene segment controlled the increased viral yield in the absence of PKR. The hypothesis of this study is that improved oncolysis can be achieved by combining the T1L L3 gene segment with the T3D MI gene segment through the creation of T1L x T3D reassortants. The reassortant viruses with these gene segments and other unidentified gene segment combinations have the potential to have superior oncolytic ability compared to the parent serotypes. The Balb/c murine colon carcinoma tumor cell line, CT26, is resistant to lysis by serotype 1 Lang (T1L) and T3D following infection in vitro and is thus a demanding model in which to screen T1L x T3D reassortants for improved oncolytic ability. To identify genes involved in viral growth and protein synthesis a panel of T1L x T3D reassortants was monitored for replication and protein synthesis in CT26 cells as well as mouse embryo fibroblasts from wt and the PKR knockout (PKR-/-) Balb-c mice. In epithelial cells the primary gene segment involved in viral yield and protein production in CT26 cells was the T3D S1 gene segment, with secondary contribution from T3D L2, M2 and M3 gene segments for viral yield, and T3D L1 and M2 gene segments for protein production. However, in fibroblasts the T1L L3 gene segment was important for protein production in the presence of PKR and the T1L L1 and T3D M2 genes segments were involved in protein production in the absence of PKR. Different reovirus genes are involved in the productivity and permissiveness of infection in different cell types. In contrast to in vitro infection, 4 T1L x T3D reassortants, E1388, E1386, E1397, and EB123, possess enhanced ability to prolong survival of Balb/c mice bearing CT26 lung metastases. This finding demonstrated that ability to lyse tumors does not correlate with ability to lyse cells in vitro and that specific reovirus reassortants possess enhanced oncolytic properties. Virus infection of CT26 tumors in vivo with EB88, EB86, EB97, and EB123 was measured by immunofluorescence with anti-reovirus serum as well as apoptosis over 7 days following infection to show that all reassortants with enhanced oncolytic properties had infections of longer duration and generally more extensive TUNEL positive cells. I also observed that PKR-/- mice had a unique pattern of infection relative to T1L and T3D; bronchiolar epithelium was infected with T3D but not with T1L, whereas bronchiolar epithelium of wt Balb-c mice was resistant to infection with both viruses. The PKR dependent bronchiolar epithelial tropism mapped to the S1 genome segment. This finding, which contrasts with previous mapping of PKR dependent tropism for fibroblasts to the M1 genome segments, has identified a tissue specific aspect to the PKR dependent control of infection with reoviruses. As >95% of tumors are epithelial in origin, the difference in PKR dependent infection between fibroblast and epithelium is of relevance to oncolytic viruses which are thought to exploit defects in interferon responsiveness in tumor cells, which are largely epithelial in origin. On assessing the genotypes of reassortants that possess enhanced oncolytic abilities the T3D-L2 and T3D-M1 genome segments plus T1L-L3 and T1L-M2 genome segments appeared to be functioning to enhance tumor destruction through modulation of cell growth and protein production (L2 and L3-enhanced growth), and interferon/PKR sensitivity (M1-enhanced sensitivity). Although oncolytic properties appear multigenically controlled it is possible to obtain reovirus reassortants with enhanced combinations of proprieties that favor improved abilities to destroy tumors in intact host organisms. This study is thus foundational and has provided important data for further analysis of the mechanisms of reovirus tumor attack and its application to therapy.

In this thesis, chemical biology approaches were used to study Hepatitis C Virus (HCV) in non-infectious, subgenomic HCV replicon model systems. Specifically, small molecules were used as probes for studying host-virus interactions. Compared to genetic techniques, small molecules are interesting potential therapeutics for HCV because they can diffuse relatively quickly to their targets and also act on proteins after they have been posttranslationally modified. Small molecules are also much easier to deliver to cells. In our first study we looked on the effects of using a small molecule agonists and antagonist of the host receptor, peroxisome proliferator activated receptor alpha (PPARalpha) on the HCV viral life cycle. We found that impaired function of PPARalpha perturbs lipid homeostasis within the cells, causing inhibition of HCV replication. We confirmed this effect by silencing the PPARalpha gene using small interfering (si)RNA molecules. Our next project involved screening a library of abscisic acid (ABA) analogs for anti-HCV activity. ABA is a plant hormone, structurally similar to its counterpart in humans, retinoic acid. We found three ABA analogs that have anti-HCV activity comparable to known HCV inhibitors like lovastatin and 25-hydroxycholesterol. We identified the protein targets for one of the active ABA analogs by performing in-vivo experiments using an inhibitor-based profiling approach with a synthesized rhodamine fluorescent probe. Target identification allowed us to postulate a mechanism by which the ABA analog realizes its anti-HCV activity by interfering with the host protein responsible for both host and HCV protein folding. Finally, we demonstrated anti-HCV activity of the antitumor drug bleomycin (BLM). BLM causes HCV inhibition by directly targeting and degrading the HCV RNA. Although toxic to the host cell due to DNA degradation, we showed that BLM acts much faster on viral RNA located in the cytoplasm than on host DNA located in the nucleus.

Human parainfluenza virus type 3 (HPIV3), a member of the family Paramyxoviridae, is an enveloped, negative-sense single stranded RNA virus, and an important cause of upper and lower respiratory tract disease in children. Infection by HPIV3 requires fusion of the viral envelope with the plasma membrane of the host cell. Fusion occurs via the cooperative actions of the viral hemagglutinin-neuraminidase (HN) and fusion (F) glycoproteins. HN, the receptor-binding protein, binds sialic acid and undergoes a series of conformational changes that are believed to trigger characteristic conformational changes in the F protein that are necessary for membrane fusion. The specific objectives of this dissertation are to elucidate, in greater detail, features of HN-F interaction and F refolding during the entry process. Mutations were introduced at position I474 of the F protein, in heptad repeat 2 (HR2), a carboxy-proximal sequence that forms a highly stable structure with the amino-proximal heptad repeat 1 (HR1), known as the 6-helix bundle (6-HB). Formation of the 6-HB is believed to release energy required for fusion of virion and host cell membranes. Mutation of I474 to small, polar residues produced F proteins that are uncleaved and non-fusogenic; 6-HB formed by peptides based these mutants were thermodynamically unstable. Mutation of I474 to larger polar or nonpolar side chains, or to smaller polar side chains, resulted in F proteins that were cleaved and fusogenic to wild-type levels, although 6-HB stability was variable. The ability of HR2 peptides to inhibit virus infection and glycoprotein-driven fusion mirrored 6-HB stability. Tagging of HN with FLAG and hexahistidine tags permitted Ni-affinity purification of HN. Cotransfection of HN and F, including F I474D, S, and E mutants, showed that stable HN-F complexes could be isolated when cell membranes were solubilized with octyl glucoside. A distinct downregulation of HN expression was observed when HN and F were coexpressed in cells, likely as a co-translational or very early post-translational step. A series of three-part chimeric F sequences consisting of fragments of HPIV3 and bovine PIV3 F coding regions were constructed. When expressed in the presence of the HPIV3 HN protein, the chimeric F proteins showed no differences in level of expression or fusogenicity. Also, there were no differences in the stability of homologous and heterologous HR1/HR2 complexes, and inhibition of HPIV3 infection by HR2 peptides of the HPIV3 and BPIV3 F proteins was similar. HPIV3 resistant to inhibition by HR2 was produced by sequential passage of virus in increasing concentrations of a GST-HR2 fusion protein. Statistically significant resistance of populations as well as plaque-purified clones was demonstrated, although sequencing of resistant clones did not reveal mutations that could be clearly associated with resistance. A model for HN and F in the fusion process is presented, based on data produced in this dissertation and on previously published evidence.

Molecular basis of Influenza A virulence is not fully understood. This study focuses on the Hemagglutinin (HA) protein since it is known to be a critical determinant of virulence. The experimental approach used was mouse adaptation. The prototype clinical isolate A/HK/1/68 (H3N2) was subjected to serial mouse to mouse passages. Following longitudinal and parallel studies 11 mouse adapted populations were generated. Sequence analysis of all 11 populations identified a total of 24 mutations within the HA gene. These mutations clustered in two areas within the 3 dimensional structure. One adaptive region resides within the HA1 while the other is located in the HA2 domain. Four of the mouse adapted HA mutations exhibited evidence of convergent evolution. Three of these mutations (P162L/S, Q210R and G218W/E) reside in HA1 while one mutation (N154S/K) is located in HA2. Recombinant viruses possessing convergent HA mutations exhibited altered receptor binding and pH of fusion. These mutations increased infection and replication within the mouse lung (in vivo) and/or mTECs (in vitro). However, different infection patterns were observed indicating that distinct alpha2,3 SA receptors might be present in the tracheal, bronchial and alveolar cells. In addition, adaptive mutations in HA1 as well as HA2 were associated with enhanced virulence. Certain mouse adapted mutations parallel changes observed in other virulent variants. This identifies them as putative virulence determinants. Hence, these mutations can serve as predictors of virulence.

No description available.

Nonenveloped viruses must navigate a number of host cell membrane barriers during entry and after replication for release. Simian Virus 40 (SV40) pirates the endocytic pathway of the host cell for transport from the cell surface until it reaches the endoplasmic reticulum (ER). Within the ER lumen the capsid is thought to be disassembled releasing the viral genome which then must be transported across the ER and nuclear membrane barriers to initiate viral replication. Following replication the SV40 viral progeny must cross the nuclear plasma membrane without becoming membrane encapsulated. To overcome these obstacles during release, viruses such as SV40 utilize a lytic infectious cycle that results in the permeabilization of the host cell membranes through an unknown mechanism. The research presented here utilized numerous techniques to investigate the roles of the minor structural proteins in the SV40 life cycle. VP2 and VP3 were both shown to perform essential functions with VP2 contributing to cell binding while VP3 functioned downstream of this event. Both VP3 and VP2 showed the ability to post-translationally integrate into the ER membrane where VP3 attained a multi-membrane spanning topology. These findings combined with the propensity of VP2 and VP3 to oligomerize and permeabilize bacterial membranes provide the premise for the following hypothesis. We propose that upon disassembly in the ER lumen these minor structural oligomerize and integrate into the ER membrane where they act as a viroporin to aid in the transfer of the viral genome across the ER membrane. The SV40 mutant lacking both the VP2 and VP3 initiation codons revealed that in addition to VP3, another viral protein was synthesized from the VP2 reading frame. This protein we termed VP4 is expressed at the onset of lysis, it is not sequestered by VP1 and its removal inhibited the lytic property of VP3 towards E. coli and caused a significant 2 day delay in cell lysis during infection. In summary, my work provides new insight into how nonenveloped viruses have evolved to traverse the host cell membrane barriers during entry, when they must remain intact, and during release when they become dispensible.

Since it emerged as an infectious agent in 1981, the human immunodeficiency virus type 1 (HIV-1) is continually disseminated and remain fatal to the majority of those infected. Strategies including highly active retroviral therapies (HAART) with nucleoside analogues and protease inhibitors have shown limited success in therapy due to the virus' ability to evolve rapidly at every replication cycle as a consequence of it's highly error prone reverse transcriptase, generating resistant retroviral strains and in addition to latent HIV-1 reservoirs. Thirty years of research efforts to find a cure or to generate a vaccine has been met with failure. It is, therefore, of necessity to broaden our paradigm of therapy for the treatment and eventual cure of HIV-1 infection. In this study, I look beyond the current anti-retroviral strategies and instead rely on the mammalian host immune system to inhibit HIV-1 replication through molecular genetic manipulation. Here, we approach the inhibition of HIV-1 replication by up-regulation of the innate antiviral pathway that is natural to mammalian cells. HIV-1 derived self-inactivating lentiviral (SIN) vectors were designed and constructed to deliver the antiviral payloads of two antiviral enzymes, p68 kinase (PKR) and 2'-5' oligoadenlyate synthetase (2-5OAS), to target cell, SupT1 lymphoblastoid cells and CD4+ T lymphocytes under the control of a constitutive cytomegalovirus (CMV) promoter. These data here demonstrates a significant inhibition of HIV-1 replication in cells transduced with the anti HIV-1 transgenes PKR and 2-5OAS as determined by HIV-1 induced syncytia formation and HIV-1 p24 antigen capture assay. Furthermore, here demonstrated is an increase up-regulation of PKR and 2-5OAS 96 hr post cell transduction in all the clones when compared to pHIV empty vector control. These results demonstrate that the over-expression of PKR and 2-5OAS can inhibit HIV-1 replication and also confirm the involvement of PKR and 2-5OAS in the IFN-associated antiviral pathway against HIV-1 infection. / Microbiology and Immunology

Non-human primates are the best animal model to study the pathogenesis of HIV. The two common models for SIV pathogenesis are pathogenic and non-pathogenic SIV infections. The most widely used model of pathogenic infections is Rhesus macaques (Rh) infected with SIV strains derived from sooty mangabeys. Disease progression to AIDS occurs within a period of months to years, depending upon the SIV strain used. The best known example of non-pathogenic infections is African NHPs naturally infected with SIV. These animals rarely progress to AIDS despite maintaining levels of viremia that are at the same levels or higher than SIV viral loads in pathogenic infections. Little is known about the components of the immune system that are important for controlling SIV replication and disease progression in NHPs and HIV-infected humans. By understanding which immune correlates are important for controlling virus replication and progression to AIDS, it will be possible to design new HIV treatments. We intend to modulate SIV pathogenesis by separately depleting CD20, CD8, and regulatory-T cells in pathogenic and natural SIV infections of Rh and AGMs. Depletion of these cell types revealed that CD8 cells have a greater impact on SIV replication than depletion of CD20 cells. Moreover, we found that the experimental induction of immune activation, resulted in increases in viral replication, indicating that both immune activation and CD8+ cells are important controllers of disease progression to AIDS. These findings will be useful in the design of new therapies and treatments for HIV patients / acase@tulane.edu

CCR5 is the major co-receptor for HIV-1 entry and an important target for drug development. The recent finding that hematopoietic stem cell [1] transplantation from a CCR5-negative donor to an HIV-infected patient produces long-term virus control in the absence of antiretroviral drugs suggests the potential of stem cell and gene therapies targeting CCR5. To this end, we are developing a protocol to knock-out CCR5 in a patient's own HSC using CCR5-targeted zinc finger nucleases (ZFNs) ZFNs are sequence specific proteins that generate a double-stranded break in DNA, which is converted into a gene-disrupting lesion by host repair processes. We have optimized the delivery and function of ZFNs targeting CCR5 in human CD34-positive HSC, achieving up to 27% disruption of CCR5 alleles. We confirmed that ZFN-treated HSC remain fully functional by transplanting a mouse model of human hematopoiesis, the NOD/SCID/IL2gammacnull (NSG) mouse, where the modified HSC retained the ability to differentiate into multiple hematopoietic lineages Humanized NSG mice are additionally capable of supporting HIV-1 infection. Following challenge with an R5-tropic virus, control animals demonstrated altered CD4:CD8 ratios, profound loss of human cells in the thymus and GALT, and high viral loads in multiple tissues sampled. In contrast, ZFN-treated animals had significantly lower acute viral loads and very low levels of virus in tissues by 10-12 weeks post-infection. At this stage the numbers of human cells in tissues that are targets for HIV-1 infection had normalized, including the GALT and thymus. FACS and PCR analysis revealed a rapid and dramatic selection for CCR5-negative cells in these populations. These findings demonstrate that ZFN-treated HSC can generate HIV-resistant cells in vivo that rapidly replace cells depleted by HIV-1 infection, and importantly, preserve GALT populations. Transient ZFN treatment resulting in permanent disruption of CCR5 in autologous HSC could therefore represent a viable clinical approach for HIV-infected patients / acase@tulane.edu

Lassa hemorrhagic fever is a disease endemic to Western and Central Africa. It is also the most commonly imported hemorrhagic fever disease with numerous cases reported in travelers to the region throughout the forty one years since its first characterization in Nigeria. Lassa fever afflicts an estimated 300,000 or more people in Western Africa each year, with approximately 5,000 fatalities. These statistics may underreport the actual number of infections and overall fatality rate from the disease; in that modern, highly sensitive, specific, and rapid recombinant protein-based diagnostics recently implemented in Sierra Leone have demonstrated a higher prevalence amongst native populations. Despite significant advances in the diagnosis of Lassa fever control of the disease in resource poor and settings with vastly different social practices from those of western societies is both impractical and ecologically undesirable. The development of a robust vaccine platform that would confer long term protective immunity against Lassa fever and related Old world arenaviral agents is a highly desirable, more readily implementable and manageable means to control the disease. Such a vaccine would have applications both in the global health and biodefense arenas. To this end, critical aspects of Lassa virus gene expression were studied with the end goal of designing a non-infectious, non-replicative virus-like particle (VLP platform expressing all relevant immunogenic and protective arenaviral components. LASV VLP were immunogenic in mice, but ongoing and future studies will be aimed at determining their protective efficacy in susceptible animal models. A human vaccine comprised of LASV VLP is the ultimate endpoint / acase@tulane.edu