Francisella tularensis, the causative agent of tularemia, is a pathogen capable of survival and growth in a vast array of environments ranging from arthropod vectors to over one hundred different mammalian hosts, including humans. An understanding of the mechanisms that this bacterium uses to adapt to these varied environments is vital to fully understanding its pathogenesis. The environmental signals that Francisella responds to include low iron concentrations, oxidative stress, and temperature. The work described in this dissertation constitutes a significant step forward in our understanding of Francisella adaptation specifically to the host intracellular environment. We have shown that bacterial growth conditions have a great impact on bacterial phenotypes, particularly on the ability of Francisella to induce or inhibit macrophage cytokine production. We have identified a specific eukaryotic molecule, spermine, which is abundant in the intracellular environment and leads to significant changes in bacterial phenotypes and gene expression. We also present the first evidence of a role for the abundant Francisella IS elements in the regulation of transcription by functioning as spermine-responsive promoters. The ability of the bacterium to adapt to this signal is vital to its survival and a mutant lacking this sensing mechanism is highly attenuated. A thorough analysis of different culture conditions has lead to the identification of both known and putative virulence factors that may be important for altering host cell responses to F. tularensis. Our results demonstrate a novel mechanism of host-pathogen interaction and could have significant implications for other intracellular pathogens.

Select members of the group Borrelia burgdorferi sensu lato are the causative agents of Lyme disease (LD), a multisystem, potentially chronic disorder with debilitating clinical manifestations including Lyme arthritis, carditis, and neuroborreliosis. Current knowledge regarding the expression of virulence factors encoded by B. burgdorferi and the breadth of their distribution amongst Borrelia species within or beyond the sensu lato group is limited. Some genes historically categorized into paralogous gene family (pgf) 54 have been suggested to be important during transmission to and/or infection of mammalian hosts. By studying the factors affecting the expression of this gene family and its encoded proteins, their distribution, and the disease profile of a bba66 deletion isolate, we aimed to determine the importance of pgf 54 genes in Lyme disease and their conservation amongst diverse Borrelia species. The culmination of the studies discussed in this thesis describe the association of select genes historically categorized into pgf 54 with infectious phenotypes, with the borrelial sigma factor cascade, and the localization of their encoded proteins to the outer surface of the bacterial cell. Together, the expression profiles and localization of these genes/proteins demonstrates that they are regulated by the ãN-ãS cascade, similarly to the known virulence factor, OspC, and that they are found on the outer surface of the cell where they would have the potential to interact with or sense host factors. Moreover, putative orthologs of these genes were detected by Southern blotting and PCR in diverse Borrelia species associated with both Lyme disease and relapsing fever, some of which expressed pH-responsive proteins that were cross-reactive with antibodies specific for orthologs expressed by B. burgdorferi isolate B31. Finally, an insertion-deletion of one of these genes, bba66, was examined in vivo and was found to be infectious in C3H/HeJ mice. Though bba66 was not found to be absolutely required for murine infection in the study presented here, we and other groups hypothesize that bba66 may instead be important during dissemination or adherence to murine cardiac tissue. Thus, future studies are aimed to determine the function and putative importance of BBA66 beyond the establishment of murine infection.

Herpes simplex virus type 1 (HSV-1) infected cell polypeptide 4 (ICP4) is a critical regulator of viral gene expression that is required for productive infection. ICP4 has been shown to act, depending on the promoter structure, as both an activator or repressor of viral genes. ICP4 has two broad transcriptional regulatory domains, the N-terminal domain, which exhibits both repression and transactivation, and the C-terminal domain, which is involved exclusively in transactivation. ICP4 regulates transcription through interactions between ICP4s transcriptional regulatory domains and cellular general transcription factors including components of TFIID. Although it has been shown that a region in the amino-terminus of ICP4 corresponding to aa 30-210 is necessary for appropriate transactivation and repression of viral genes, specific domains responsible for these activities remain uncharacterized. Using deletion mutants spanning this region, we show that the entire region is necessary for ICP4 function, but that loss of aa 30-142 has a greater detrimental impact on the ability of ICP4 to transactivate E and L genes. Neither deletion had a significant impact on ICP4 repression, however. The protein interactions made by ICP4 in vivo during infection were also studied. We show ICP4 forms complexes with TFIID, thus verifying previous in vitro data. Novel interactions between TFII-I and components of the Mediator complex were also identified. In addition, Mediator was found to colocalize with ICP4 starting at early and continuing into late times of infection. Mediator was also recruited to viral promoters in an ICP4-dependent manner, showing a direct role for Mediator in ICP4-mediated transcription. Together, the data show that ICP4 regulates transcription through interactions between its complex regulatory domains and a diverse set of cellular protein complexes, including components of the basal transcription machinery and coactivator complexes.

CD8+ T cells have a little understood noncytolytic activity that suppresses human immunodeficiency type 1 (HIV-1) replication in an antigen-independent and MHC-unrestricted manner. This activity specifically inhibits transcription of the HIV-1 proviral genome. Little is understood about the molecular nature of the factor(s) mediating this potent antiviral activity of CD8+ T cells. It is known that a factor secreted by CD8+ T cells can suppress the transcription of HIV-1. However, the antiviral mechanism appears most potent with cell-to-cell contact. Previous investigations by several groups into the nature of this secreted factor have been largely based on a presumption that noncytolytic suppression of HIV-1 by CD8+ T cells is exclusively mediated by a soluble protein. Based on several lines of evidence suggesting the specific involvement of cell-contact determinants in eliciting the noncytolytic CD8+ T cell effector function against HIV-1, a novel approach to the problem was utilized based on the hypothesis that a membrane-bound factor is the prime mediator suppressing HIV-1 transcription. In the ensuing investigation, evidence was uncovered demonstrating the existence of a membrane-localized HIV-1 suppressing factor that was secreted as 30-100nm spherical vesicles termed exosomes. Exosomes from a CD8+ T cell line inhibited the replication of R5 and X4 HIV-1 isolates and were shown to specifically suppress of HIV-1 transcription in acute and chronic models of infection. A much greater degree of complexity to the CD8+ T cell secreted antiviral activity was found than a soluble protein alone could account for. The evidence presented in this study suggests that CD8+ T cell suppression of HIV-1 is predominantly mediated by a membrane-bound protein factor that can be cleaved into a soluble isoform with the secreted CD8+ cell antiviral activity being largely exosome-driven. The results presented in this study provide a much more concrete understanding of the mechanisms underlying CD8+ T cells suppression of HIV-1 transcription and outline new approaches to conclusively identifying the molecular factor mediating potent inhibition of the HIV-1 transcritional promoter.

The vast diversity of HIV-1 infections has greatly impeded the development of a successful HIV-1/AIDS vaccine. Previous vaccine work has demonstrated limited levels of protection against SHIV/SIV infection, but protection was only observed when the challenge virus was directly matched to the vaccine strain. As it is likely impossible to directly match the vaccine strain to all infecting strains in nature, it is necessary to develop an HIV-1 vaccine that can protect against a heterologous viral challenge. Envelope-based vaccines were developed containing gp120 monomers, gp140 trimers and gp160 on the surface of a virus-like particle. Vaccination studies indicated that similar systemic levels of Env-specific antibodies were generated by each vaccine. However the VLP-based vaccine led to increased recognition of Env epitopes as well as a significant increase in mucosal Env antibody responses. Also the VLP-based vaccine led to the development of strong cellular responses to both Env and Gag. Therefore the VLP platform was moved forward to develop broadly reactive vaccines. Consensus and Polyvalent clade B and clade C vaccines were developed and investigated for cellular responses in mice. The results indicated that both Polyvalent and Consensus VLP vaccines led to an increase in the number of cellular Env epitopes recognized as compared with primary Envelope-based vaccines. These findings were true for both clade B and clade C vaccines. To determine the level of protection generated by VLP based vaccines, consensus clade B as well as a polyvalent clade B vaccine were investigated in non-human primates. The polyvalent clade B vaccine led to the protection of 3 out of 4 challenged animals with decreased viral burden observed in the infected individual. Overall these studies indicate that a virus-like particle vaccine encoding multiple primary envelopes is a promising HIV-1/AIDS vaccine strategy for protecting against heterologous HIV-1 viruses.

Although more doses have been given of the tuberculosis vaccine (BCG) than any other vaccine in history, tuberculosis remains the second leading cause of death due to an infectious agent worldwide. Our understanding of the immune response to the bacterium is incomplete, which has partly been due to a lack of known bacterial T cell epitopes. In recent years, several epitopes have been identified, which has enabled the use of new reagents and assays to study T cell responses. Furthermore, development of recombinant bacteria expressing model antigens has facilitated the use of many more reagents and techniques, as well as allowing us to compare M. tuberculosis-specific T cell responses to those induced by other infectious agents. The aim of this thesis was to better understand the CD8 T cell response to M. tuberculosis and identify factors that regulate the response. We showed that, unexpectedly, the primary CD8 T cell response was more effective than the memory CD8 T cell response to M. tuberculosis, in that primary cells secreted higher levels of IFN-g and were more cytotoxic. We also showed that cytotoxicity and secretion of IFN-g were carried out by distinct M. tuberculosis-specific CD8 T cells, which indicates that there is a functional defect in these cells that is not seen in other infections. Contrary to chronic viral infections, however, M. tuberculosis-specific CD8 T cells did not appear to become exhausted, despite the chronicity of the infection. We further showed that the function of CD8 T cells was influenced by various factors, such as interactions with pre-existing memory cells, the quality of CD4 T cell help and the environment that the cells were primed in. Manipulation of these factors can increase the quality of the immune response, giving multifunctional CD4 and CD8 T cells, as well as CD8 T cells capable of simultaneously being cytotoxic and secreting IFN-g, without altering the bacterium. This indicates that modifications in delivery of tuberculosis vaccines may result in improved efficacy. The work presented may contribute to advances in the development of new tuberculosis vaccines, as well as improvement in efficacy of the existing vaccine.

Multiple B cell abnormalities have been described in humans infected with HIV. These abnormalities include hypergammaglobulinaemia, diminished B cell response to mitogenic stimuli, lymphoma and a depletion of the memory B cell population. There is also evidence to suggest that B cells in HIV infected patients are functionally impaired. The initial antibody response to HIV infection is slow to appear and antibody responses to B cell mediated vaccines in HIV infected individuals are less robust and less durable than in uninfected individuals. Although B cell abnormalities have been characterized in humans, efforts to link these abnormalities to a specific defect within the B cell compartment have not been entirely successful. The SIV/macaque model of HIV infection of humans provides a means for addressing questions about the naïve and memory B cell populations, whose activity may be differentially compromised by HIV infection, but lacking is the ability to resolve these functionally relevant B cell populations in the rhesus macaque. In this study, we established CD27 as a definitive memory B cell marker in the rhesus macaque. Further, we demonstrated that the naïve and memory B cell populations are depleted from the periphery within 14 days of SIV infection and that the memory B cell population recovered more quickly. We also showed that chronic SIV infection resulted in a loss of CD40 mediated naïve B cell survival, indicating a potential mechanism through which SIV infection may lead to the production of non-reactive or self reactive antibody producing cells. Together, these findings demonstrated that B cell dysfunctions associated with SIV infection are not limited to the memory B cell population as previously thought, but rather that naïve B cell deficits may be more severe than what has been observed in the memory compartment. Increased focus on abrogating alterations that occur within the naïve compartment have the potential to improve viral control in infected individuals. This study of phenotypic and functional B cell changes over the course of infection will aid in the development of strategies that have the potential to improve prophylactic and therapeutic B cell mediated vaccine efficacy.

The fungal opportunistic pathogen, Pneumocystis jirovecii (formerly Pneumocystis carinii f. sp. hominis) (Pc) is the causative agent of Pneumocystis Pneumonia (PcP) in immunocompromised persons. Despite improvements in anti-retroviral treatments and Pc prophylaxis, Pc remains an important pathogen in immunocompromised populations. Pc colonization, the presence of Pc in subjects without clinical signs or symptoms of PcP, is common in HIV+ subjects; however, the clinical consequences of colonization are undefined. The non-human primate model of Pc infection in simian immunodeficiency virus (SIV)- or chimeric simian-human immunodeficiency virus (SHIV)-infected macaques has been developed to study Pc colonization pathogenesis in the context of AIDS immunosuppression. Using this model, immunologic parameters associated with natural Pc colonization of macaques were evaluated to gain understanding of protective immune responses to Pc. Humoral immunity to the recombinant Pc-antigen, kexin (KEX1), correlated with protection from subsequent Pc colonization, despite declining CD4+ T cells. Furthermore, macaques that remained Pc-negative were protected against lung injury observed in macaques that became Pc-colonized, supporting a role for Pc in pulmonary obstruction development. These experiments suggest KEX1-specific antibodies may provide protection of immunocompromised individuals from developing obstructive pulmonary disease. Because B cell deficits and dysfunctions are reported in HIV+ subjects, we examined peripheral blood B cell populations in SHIV-infected macaques. We report declines in total (CD20+), memory (CD20+CD27+) and IgM+ memory B cell numbers, increased percentages of activated (CD95+) B cells, and hypergammaglobulinemia in SHIV-infected macaques, similar to what has been reported for HIV+ patients, suggesting the relevance of this model for studying HIV-related B cell dysfunctions. Pc colonization status did not correlate with deficits in total B cell populations. Rather, protection from Pc-colonization appears associated with a KEX1-specific memory B cell pool, despite early loss of total CD27+ B cells. These results suggest exposure to Pc prior to immunosuppression, resulting in high levels of circulating antibodies/plasma cells, contributes to maintenance of a Pc-specific memory B cell pool following immunosuppression. These results demonstrate importance of a Pc-specific humoral response in protection from Pc colonization and pulmonary damage, thereby providing a rationale for Pc-KEX1 vaccine development to protect at-risk populations against this opportunistic pathogen.

Infection with high-risk human papillomaviruses (HPVs) is the main cause of cervical cancer, the second most common cause of cancer-related mortality in women worldwide. High-risk HPV types, such as HPV-16, express two oncoproteins, E6 and E7, which function to subvert critical host cell cycle control mechanisms in order to promote viral genome amplification. Disruption of the pRB signaling axis and the p53-mediated stress response by the HPV E7 and E6 oncoproteins, respectively, results not only in aberrant proliferation but also in host cellular changes that can promote genomic instability. The high-risk HPV-16 E7 oncoprotein was found to induce centrosome abnormalities thereby disrupting mitotic fidelity and increasing the risk for chromosome missegregation and aneuploidy. Aneuploidy is frequently found in pre-malignant high-risk HPV-associated lesions and is a critical factor for malignant progression. This thesis was designed to determine the molecular mechanisms behind the ability of HPV-16 E7 to rapidly induce centriole overduplication. This rapid induction was found to be possible through the simultaneous formation of more than one daughter centriole at single maternal centrioles (centriole multiplication). It was previously discovered that the centriole multiplication pathway relied on cyclin E, CDK2 and PLK4. However, it was not known before how these molecular players cooperate in the centriole multiplication pathway or how HPV-16 E7 expression promotes the activation of this pathway. Here, we report that cyclin E/CDK2 mediates the aberrant recruitment of PLK4 to maternal centrioles. This initial recruitment step was not sufficient to induce centriole multiplication unless PLK4 protein levels were increased. We found that PLK4 protein levels were controlled by proteolysis, specifically by CUL1-based E3 ubiquitin ligase complexes localized at maternal centrioles. SCF activity was found to control not only baseline PLK4 protein stability but its activity-dependent degradation following cyclin E/CDK2 overexpression. High-risk HPV-16 E7 is known to deregulate cyclin E/CDK2 complexes and we found that ectopic expression of HPV-16 E7 promoted the aberrant recruitment of PLK4 to maternal centrioles. Since our previous experiments have shown that aberrant recruitment of PLK4 is not sufficient to drive centriole overduplication, we determined whether HPV-16 E7 may also disrupt PLK4 expression. We found that HPV-16 E7, but not low-risk HPV proteins or mutants of HPV-16 E7 that lack the ability to induce centriole overduplication, causes a moderate but significant upregulation of PLK4 mRNA. Besides centriole duplication control, we discovered that proteolysis also regulates other aspects of centriole synthesis such as regulation of daughter centriole length. Defining the precise molecular circuitry of centriole biogenesis will aid not only in deepening the current understanding of centriole biogenesis but also aid in identification of novel targets, such as CDK2 or PLK4, for small molecules to prevent centriole abnormalities, mitotic infidelity and malignant progression in pre-invasive high-risk HPV-associated lesions.

Th1-polarizing cytokine IL-12 exhibits potent anti-tumor activity in multiple cancer models; however, therapeutic use of this cytokine is limited due to severe IFN-ã-mediated toxicity. To reduce the amount of IL-12 needed to elicit a therapeutic response, and thereby decrease associated toxicity, it is necessary to characterize novel cytokines to use in conjunction with IL-12. Newly described IL-12 family member IL-23 shares the IL-12 p40 subunit and promotes Th1 immunity by inducing IFN-ã expression and specifically stimulating proliferation of memory CD4+ T-cells. I have demonstrated that injection of an adenovirus expressing IL-23 (Ad.IL-23) into the tumor microenvironment results in significantly enhanced survival and tumor rejection in 40 percent of animals, with concomitant induction of protective anti-tumor immunity. Furthermore, the anti-tumor activity of IL-23 is dependent on IL-12, IFN-ã and CD4+ and CD8+ T-cells, indicating generation of a Th1 response. Delivery of adenovirus expressing IL-12 (Ad.IL-12) into the tumor microenvironment also results in enhanced survival and tumor rejection in up to 90 percent of animals. In contrast to Ad.IL-23, Ad.IL-12 anti-tumor activity requires only IFN-ã and induces protective immunity only 50 percent of the time, suggesting activation primarily of innate immunity. Due to the similar, yet divergent, effector mechanisms used in Ad.IL-12 and Ad.IL-23 anti-tumor effects, I hypothesized that use of these two viruses together would result in synergistic enhancement of tumor eradication. Surprisingly, Ad.IL-12 and Ad.IL-23 do not synergistically enhance anti-tumor effects over use of either cytokine alone, possibly due to IL-23 p19 sequestration of p40. To circumvent this possibility, adenovirus expressing single chain IL-23 (Ad.scIL-23) was constructed and characterized. Ad.scIL-23 expresses greater levels of cytokine than Ad.IL-23 and treatment of tumor bearing mice results in tumor rejection in 90 percent of animals. However, Ad.scIL-23 does not synergize with Ad.IL-12. Overall, I have shown that IL-23 does possess therapeutic anti-tumor effects, but does not synergize with IL-12 to enhance tumor eradication. Future studies could include characterizing CD4+ T-cell infiltrate of tumors treated with both Ad.IL-12 and Ad.IL-23 to elucidate the mechanism behind the lack of synergy between these two cytokines.