Each of the HIV-1 pol-encoded enzymes, protease (PR), reverse transcriptase (RT) and integrase (IN) are released during virion maturation and are active only as dimers. Of the three, only RT comprises subunits of different mass. RT in mature infectious virions is a heterodimer of 66 kDa and 51 kDa subunits, even though its gene encodes a 66 kDa protein. The RT p51 subunit is formed by HIV-1 PR-catalyzed cleavage of RT p66, resulting in the removal of a ribonuclease H (RNH) domain. Given the existence of completely active recombinant p66/66 RT homodimers and alternative RT oligomers in other retroviruses, the apparent need for p66/51 RT heterodimers in the HIV-1 virion is unclear. To determine why the generation of active viral RT requires three processing events, we introduced mutations in the p51-RNH and RT-IN protease cleavage sites of an infectious HIV-1 molecular clone. Mutation of the RT-IN cleavage site had no effect on the activity or proteolytic stability of the p98/51 RT product, although infectivity was severely attenuated. This result was similar to findings previously reported for the PR-RT cleavage site. Surprisingly, mutation of the internal p51-RNH cleavage site did not increase RT p66 content, but resulted in attenuated virus containing greatly decreased levels of RT that was primarily RT p51. We further identified a compensatory second-site mutation T477A, found to restore RT activity and processing to p66/51 RT when introduced in the background of p51-RNH cleavage site mutations. These studies demonstrate that cleavage of the internal p51-RNH junction, not the flanking N-terminal or C-terminal junctions is essential for proteolytic stability of functional RT during virion maturation. These findings further emphasize the importance of the RNH domain in regulating proteolytic generation of p66/51 RT. The overall need for the RT heterodimer is attributable to the generation of its subunits. Formation of the 51 kDa subunit or cleavage of the p51-RNH junction is essential for RT stability in the virion, whereas formation of the 66 kDa subunit is important for efficient viral replication.

The herpes simplex virus type 1 immediate-early protein, ICP4, is the major transcriptional activator of viral early and late genes. ICP4 has been shown to bind DNA and interact with components of the general transcription machinery to activate or repress viral transcription, depending on promoter context. Microarray and northern blot analysis indicated that the abundance of the basal Pol II transcription machinery was significantly altered at late times of infection. Because viral early and late genes have very different promoter architectures, the cellular requirements for ICP4-mediated activation of early and late genes may differ. In testing this hypothesis using tk and gC as representative early and late promoters respectively, the general transcription factor TFIIA was found not to be required for ICP4 activation of the late gC promoter but was essential for activation of the early tk promoter. An intact INR element was required for TFIIA to be dispensable for ICP4 activation of the gC promoter. In the presence of TFIIA, ICP4 overcame the requirement for an intact INR on the gC promoter enhancing ICP4 activation of an INR-mutated gC promoter. When examining the binding properties of TFIIA, ICP4 and TBP/TFIID on early, late, and INR-mutated late promoters, ICP4 could only effectively substitute for TFIIA in stabilizing the binding of TFIID to the TATA box of the late promoter containing a functional INR. ICP4 required the additional activities of TFIIA to stabilize the binding of TFIID to the TATA box of an INR-mutated late promoter. Additionally, microarray and northern blot analysis indicated that TFIIA expression was reduced at late times of infection. The decrease in TFIIA expression during infection, the ability of ICP4 to substitute for TFIIA in stabilizing TFIID binding to an INR-containing late promoter, and its dispensability for activation of late but not early genes suggests a mechanism for the shutoff of early gene expression while allowing the continued expression of late genes.

A vaccine expressing virus-like particles is an attractive candidate for the development of an effective vaccine for human immunodeficiency virus type 1 (HIV-1). A single vaccine plasmid was constructed to express HIV-1 Gag, Pol, Env, Tat, Rev and Vpu. Safety mutations and deletions were introduced into the VLP DNA to generate a vaccine insert that was non-infectious. The 5¡¦and 3¡¦ long terminal repeats, integrase, vif, vpr and nef were removed to further enhance the safety of the vaccine insert. Moreover, mutations were introduced into nucleocapsid and reverse transcriptase to severely restrict viral RNA packaging and to abolish RT and RNase H activity. Virus-like particles were efficiently released from primate cells, but particles were not produced in rodent cells. Therefore, purified particles were used as the inoculum to test the immunogenicity of the VLP vaccines in a rodent system. Systemic and mucosal immune responses to HIV-1 were enhanced by intranasal immunization of purified VLPs expressed from the virally-regulated multi-gene DNA vaccine. VLPs were co-immunized with cytosine-phosphate-guanosine oligodeoxynucleotides (CpG ODNs) to enhance the immune response to HIV-1 gene products. VLPs elicited specific immunity to HIV-1 antigens in both the systemic and mucosal immune compartments. Anti-Env antibodies were detected in the sera, as well as in the washes from harvested lungs, intestines and vagina from immunized mice. In addition, Env- and Gag-specific IFN-×-secreting splenocytes were elicited in the mice vaccinated with VLPs. Co-inoculation of CpG ODNs with VLPs significantly enhanced both arms of the immune response. In addition, these particulate immunogens were compared to soluble proteins (Gag and Env). Mice immunized with soluble protein alone or co-vaccinated with CpG ODNs elicited much lower immune responses compared to VLP-vaccinated mice. Specifically, CTLs were induced by vaccination with VLPs, whereas the immune response elicited by soluble proteins (+/- CpG ODNs) was almost exclusively antibody-mediated. Overall, a vaccine expressing virus-like particles is one of the most promising alternatives to replication-competent virus in eliciting high levels of cross-reactive neutralizing antibodies in combination with a robust cell-mediated response against multiple viral antigens to protect an HIV-infected host from life-long infection or disease.

Cationic antimicrobial peptides (CAPs) are a very diverse group of amphipathic agents that demonstrate broad activity against Gram-positive and -negative bacteria. To overcome the obstacle of drug resistance among bacterial pathogens, CAPs have been extensively investigated as a potential source of new antimicrobials with novel mechanisms of action that may complement current antibiotic regimens. However, the suppression of antimicrobial activity in biological conditions (e.g., physiological salt concentrations, serum) constitutes a major challenge to the successful development of CAPs for clinical applications. We hypothesized that CAPs with optimized amphipathic structures can be designed de novo to enhance antibacterial activity and selective toxicity in environments that are generally challenging to host-derived peptides. Furthermore, the antibacterial efficacy will positively correlate with length, charge, Trp content, and helicity. Three specific aims were developed to address this hypothesis. The first aim addressed the design of amphipathic peptides to evaluate the influence of helicity, length, and Trp content on activity. Using a base unit peptide approach, we synthesized a multimeric series of 12-residue lytic base unit (LBU) composed of Arg and Val residues, positioned to form idealized amphipathic Ñ-helices. Another series of LBU derivatives (WLBU) was engineered by substituting Trp residues on the hydrophobic face. The correlation between length and helicity was established by circular dichroism analysis. Bacterial killing assays revealed no appreciable increase in activity for peptides longer than 24 residues. In addition, the inclusion of Trp residues in the hydrophobic face increased potency and selectivity in a novel co-culture system utilizing bacteria and primary cell lines, which led to the selection of WLBU2 as the shortest peptide (24 residues in length) with the highest potency. The second objective was to examine the selective toxicity of WLBU2 in biological or biologically-derived media. In contrast to the human CAP LL37, the peptide WLBU2 displayed potent activity against Pseudomonas aeruginosa in the presence of human serum and human blood ex-vivo, with no detectable red blood cell lysis or toxicity to human monocytes at all test concentrations. Finally, we demonstrated potent activity of WLBU2 in intraperitoneal and intravenous mouse models of Pseudomonas aeruginosa infection. WLBU2 not only protected mice prophylactically but also eradicated P. aeruginosa from the blood and other tissues at 3 to 4mg/kg. Furthermore, WLBU2 displayed only a minor stimulatory effect on inflammatory cytokines, notably IL1-£] and TNF-Ñ. Consistent with our in vitro studies, the in vivo data provide strong evidence for the potential application of WLBU2 in the treatment of systemic infection due to P. aeruginosa.

The interplay between host and virus that controls disease progression in HIV- and SIV-infected individuals is undoubtedly complex. Host factor(s) play a major role in this process, however, because in the SIV-infected macaque, differences in disease progression and survival persist even though identical virus stocks, doses, and routes of inoculation are used for infection. We have developed a simplified in vitro assay that significantly correlated with disease progression after infection in vivo. The amount of virus produced from primary CD4+ T cells obtained from uninfected monkeys and infected in vitro correlated significantly with the rate of disease progression and survival after inoculation of the animal. The goal of this dissertation was to conduct a detailed molecular and immunological analysis of why differential in vivo virus production occurs and how it relates to disease progression and survival. Analyses of the events occurring during virus infection in vivo revealed two findings: 1) low virus production was caused by a decreased efficiency of the early steps of reverse transcription, which affected virus dissemination in the culture, and 2) low virus production was associated with a decreased susceptibility to mitogen-induced apoptosis both the presence and absence of virus infection. Together, these data suggest that multiple, cooperative factors mediated by the host cell have a profound impact on virus production and the rate of virus spread in vivo. Subsequently, the impact of these events on infection and disease progression in vivo in these same animals was examined in detail. After infection, the four high producer animals exhibited characteristics of rapid progressors (higher viral loads during the set-point, faster decline in memory CD4+ T cells, weaker virus-specific immune responses), while the low producers, in contrast, appeared to be slow progressors. Interestingly, the number of infected cells in the peripheral blood and the timing of their emergence, rather than the peak viral load, was an accurate indicator of the eventual set point. We believe the results from acute infection in vivo are a direct result of slower viral replication during the very first few rounds of infection, as suggested by our in vitro data. The finding that incredibly early events during in vivo infection appeared to play a dominant role in the disease outcome underscores the studies of others who have shown that a better clinical prognosis can be achieved the earlier that antiretroviral treatment is initiated after infection. The studies presented here will hopefully enable not only the identification of novel host factors that influence disease outcome, but will also provide a framework for the design of new therapeutic agents that target specific host factors exerting their influence very early after infection.

Tuberculosis is a major cause of mortality worldwide. Elucidation of the host resistance against Mycobacterium tuberculosis infection and the pathogenesis of tuberculosis is a priority. The host response to M. tuberculosis in the lungs includes the formation of granulomas, focal accumulations of mononuclear cells coming together to fight infection. The signals required for the migration of cells into the lungs and those required for the formation of granulomas have not been well defined. In this thesis, we will describe the advances we have made in understanding chemokine expression and cell migration to the lungs in response to M. tuberculosis. We have determined that Tumor Necrosis Factor (TNF) partially controls cell migration by modulating chemokine expression by macrophages in vitro and CD11b+ cells in vivo. In addition, we have taken the subset of chemokines that are affected by TNF and addressed their functions in vivo through knockout and neutralization models. Specifically, we have used an aerosol model of tuberculosis in the mouse to address the roles of CXCL9 (MIG), CXCL10 (IP-10), chemokine receptor5 (CCR5) and chemokine receptor2 (CCR2). In our studies, there was no apparent, strong phenotype in mice without functional CXCL9 or CXCL10. CCR5-/- mice had altered pathology and increased inflammation in response to M. tuberculosis infection, but they were able to control the infection. The CCR5 -/- mice may be hyper-responsive due to higher antigen load in the lymph nodes in conjunction with increased dendritic cell migration and more primed lymphocytes. CCR2-/- mice had a clear defect in macrophage migration and a delay in T cell migration, and the course of the disease was governed by the initial dose. Specifically, when a low dose aerosol infection route was used, CCR2-/- mice were able to control infection with the reduced number of cells migrating into the lungs; however with a higher dose, the mice succumbed to infection. Our findings are relevant to understanding the immune response and granuloma formation during aerosol M. tuberculosis infection, and will contribute to an appreciation of the potential effects of anti-inflammatory or anti-chemokine therapies on infections.

Mycobacterium tuberculosis is the leading cause of death due to a single infectious agent worldwide. Immune response elicited against the bacterium is sufficient to control, but not to eliminate the pathogen leading to the establishment of asymptomatic state called latency. Currently very little is known about conditions that are required to prime robust, protective immune response in the initial stages of infection. We hypothesized that CD40 ligation is essential for the priming of strong Th1 response and for the subsequent control of M. tuberculosis infection. Furthermore, it is unknown how persistent exposure to mycobacterial antigens affects the functionality of T cells. Our initial thought was that the reason why immune response cannot eliminate M. tuberculosis and why reactivation occurs is due to T cell exhaustion as a result of continuous antigenic stimulation. Since the goal of every vaccine strategy is to elicit long-lasting and protective immunity, it is necessary to define the factors required for the development and maintenance of effector and memory T cell responses. We hypothesized that CD4 help and IL-15 are required for the development of functional CD8+ memory responses. Our results indicate that in murine model of tuberculosis CD40 ligation on antigen presenting cells either by host- or mycobacterium-derived ligands, is essential for the induction of robust IFN-× T cell response resulting in protection against disease and death. Once a strong immune response is elicited, long-term control of M. tuberculosis infection is mediated by dynamic changes in the frequency and types of T cell effector functions. We identified that IL-15 was not essential for the proliferation of CD4+ and CD8+ T cells or for the maintenance of their effector functions after primary and secondary M. tuberculosis infection. Our findings indicate that CD4 help was required during priming of effector CD8+ T cells and during secondary infection for the functional and durable memory CD8+ T cell responses. Collectively, the findings presented in this thesis broadened our understanding of what factors are essential for the generation and maintenance of functional effector and recall responses following M. tuberculosis infection.

The final degradation product of the third (C3) component of complement, C3d, is a natural adjuvant. The adjuvant properties of this protein have been studied using several antigens fused to C3d. To determine the ability of this protein to enhance the humoral immune response in mice with different genetic backgrounds, inbred and outbred mice were immunized with C3d conjugated antigens. C3d induced enhanced immune responses that were comparable in all mouse strains. Analysis of the isotype class switch suggested that C3d favored the development of humoral immune responses. The classic mechanism of enhancement of the immune response involves interaction of C3d with its natural ligand, complement receptor 2 (CR2). This molecule is expressed mainly by follicular dendritic cells and B-cells. Simultaneous interaction of the antigen and C3d with the surface immunoglobulin M (sIgM) and CR2 on the surface of B-cells triggers two signaling pathways that cross-talk and synergize in cell activation. Most previous studies stressed the importance of the C3d-CR2 interaction for the adjuvant effect. However, the data supporting this hypothesis was derived from in vitro studies. To determine the importance of the C3d-CR2 interaction for enhancement of the immune response in vivo, mice deficient in CR2 (CR2-/-) were immunized with antigens fused to C3d. Contrary to the predicted, CR2-/- mice immunized with antigens fused to C3d, developed almost similar humoral immune responses than wild-type mice. These results suggested that C3d enhances the immune responses by CR2-dependent and CR2-independent mechanisms. Finally, the ability of C3d to emulate the antigenic redundancy of T-cell independent antigens and thus induce class switch in the absence of CD4+ T-cells was explored in a MHC type II knock-out mouse model (MHC II-/-). A soluble form of hemagglutinin (sHA) fused to C3d inefficiently induced IgG class switch in MHC II-/- mice. This demonstrated that C3d requires CD4+ T-cells to optimally enhance the immune responses. However, despite of the mild secondary immune responses, MHC II-/- sHA-C3d3-vaccinated mice had reduced morbidity and enhanced survival following a lethal virus challenge, suggesting that besides B-cells, C3d activates other immune cells and that the final enhancement effect is the accumulation of various mechanisms.

Crucial to the function of polarized epithelial cells is the generation and maintenance of distinct protein populations at the apical and basolateral domains. Cellular mechanisms sustain this polarity by properly sorting apical and basolateral proteins in both the biosynthetic and postendocytic pathways via sorting signals inherent within the protein. N-glycosylation has been specifically implicated as one of several sorting signals responsible for delivery of proteins to the apical surface along the biosynthetic pathway. Whether this sorting signal is also used along the postendocytic pathway is currently not known. To examine N-glycan-dependent trafficking along the biosynthetic pathway and explore a possible role for N-glycans in sorting along the postendocytic pathway, this study focuses on the sialomucin endolyn. Characterization of the biosynthetic sorting signal revealed that two glycans at amino acid positions 68 and 74 were necessary and sufficient for apical delivery of endolyn. Terminal processing of N-glycans is also necessary for efficient apical delivery as incubation with kifunensine, a drug that inhibits terminal processing of N-glycans, resulted in the nonpolarized delivery of endolyn. A similar N-glycan-dependent sorting signal was also found to be important for apical recycling of endolyn along the postendocytic pathway. The use of the same sorting signal suggested that endolyn may be sorted in a compartment that is common to both the biosynthetic and postendocytic pathway. However, further studies revealed that endolyn is sorted in distinct compartments along these two pathways. These data enhance our understanding of how and where N-glycans mediate apical sorting to maintain the integrity of polarized sorting in epithelial cells.

Clostridium perfringens is a Gram-positive, anaerobic, spore-forming, rod-shaped bacterium that causes histotoxic infections and enterotoxaemias in humans and domestic animals. This bacterium owes its pathogenicity to the production of a large arsenal of toxins, including the C. perfringens enterotoxin (CPE) and beta2 toxin (CPB2). Type A, cpe-chromosomal isolates cause human food-poisoning (FP), whereas type A, cpe-plasmid isolates cause human antibiotic associated diarrhea (AAD). Symptoms of AAD are more severe and longer in duration than symptoms of FP. We hypothesized that AAD isolates may produce an accessory toxin, which would contribute to pathogenesis, as an explanation for these symptomatic differences. Consequently, the goal of this dissertation was to determine; 1) whether the recently discovered, plasmid-encoded, CPB2 toxin was produced by AAD isolates, 2) how this toxin functions in vitro, and 3) its role in vivo (in the context of human and animal disease). PCR analysis, pulsed-field gel electrophoresis/Southern blotting, and DNA sequencing determined that the cpb2 gene is preferentially associated with AAD isolates (but not FP isolates), in part, due to its presence on some cpe-encoding plasmids. Sequencing of cpe and cpe/cpb2 plasmids found a large region of plasmid DNA, encoding a conjugal-gene cluster, which appears conserved amongst most C. perfringens virulence plasmids. Subsequent sequencing of a number of cpb2 genes identified two CPB2 variants, both of which are produced (detected using Western blot). Neutral red cytotoxicity assays using purified CPB2 demonstrated that CPB2h1 is more cytotoxic for Caco-2 cells than CPB2h2 and this activity is heat labile. Further in vitro work (using 86Rubidium release assays and osmotic stabilizers) suggested that CPB2 acts by disrupting membrane permeability. In vivo experiments confirmed that our CPB2 preparations also were lethal using a mouse intravenous injection model. However, efforts to develop a more realistic model of pathogenesis (which initiates in the intestines) should be continued to further analyze the role of CPB2 in pathogenesis. Ultimately, the production of CPB2 by most AAD isolates, as opposed to FP isolates, could provide an explanation for the more severe symptoms associated with AAD cases and suggests that CPB2 may be involved in human gastrointestinal disease.