There is an increasing need to understand how inflammation is initiated by endogenous factors in the absence of infection. Various diseases such as atherosclerosis and arthritis are shaped by endogenous mediators. In situations such as transplant or trauma where there is extensive amount of tissue damage, endogenous factors can be released to influence inflammation. The modes of activation in which immune cells liberate endogenous factors for incurring immune responses remain elusive. Adenosine triphosphate (ATP) activation of the puringeric receptor P2X7 has been implicated in several immune responses. P2X7 promotes the shedding of microvesicles (MV) and the secretion of inflammatory mediators. I hypothesized that P2X7-induced MV containing some of these inflammatory mediators would promote the activation of innate immune cells such as macrophages. Using murine bone marrow derived macrophages as a model for macrophage function, I describe that harvested P2X7-induced MV from myeloid cells promote macrophage activation including pro-inflammatory cytokine secretion and co-stimulatory ligand upregulation. Phospholipids from P2X7-induced MV are partially responsible for the observed macrophage activation. Isolated phospholipids from P2X7-induced MV activate TLR4. Secondly, I describe mature cathepsin D release into P2X7-induced MV from myeloid cells. P2X7-induced MV from myeloid cells contain both intermediate and mature forms of cathepsin D. Furthermore, P2X7 stimulation of myeloid cells promotes the peripheral displacement of cathepsin D and dynamin. Dynasore, a selective and potent dynamin inhibitor, significantly reduced the secretion of mature but not intermediate cathepsin D. Lastly, I describe a novel morphological alteration following P2X7 activation of myeloid cells. ATP stimulates de novo filopodia production. These filopodia are the result of actin polymerization, Rho kinases, and phospholipases. Furthermore, P2X7 promotes the re-localization of lipids and actin-based machinery to the periphery of ATP treated cells. Collectively, these results demonstrate that P2X7-induced MV possess stimulatory cargo including phospholipids that can activate macrophages and cathepsins that are potentially capable of degrading extracellular matrix components. This data would suggest a provocative role for P2X7-induced MV and actin-based processes in promoting sterile disease.

The T cell Immunoglobulin Domain and Mucin Domain 3 (TIM-3) is a type I glycoprotein expressed primarily on the surface of activated T cells and myeloid cells. The extracellular domain of TIM-3 consists of an IgV domain and a mucin domain with several sites for N- and O-linked glycosylation. The IgV domain is important for binding of TIM-3 to two of its known ligands, a β-galactoside binding lectin known as galectin-9 (Gal-9) and phosphatidylserine, a marker of early apoptosis. The cytoplasmic tail of TIM-3 has six conserved tyrosines, although their role in modulating downstream signaling pathways has yet to be determined. TIM-3 is widely regarded as a negative regulator of effector T cell function and viability. TIM-3 is also upregulated on exhausted T cells and is postulated to have a role in the development and/or maintenance of T cell exhaustion. However, the exact regulation of T cells by TIM-3 has not been fully established for several reasons. TIM-3 and at least one of its ligand is expressed on both T cells and antigen presenting cells (APC). Therefore, it is not clear whether TIM-3 antibodies or Tim-3 Ig fusion proteins block the ligation of TIM-3 on T cells or on APCs to enhance effector T cell function. Additionally, gal-9 can also induce apoptosis in cells lacking the expression of TIM-3 and has been shown to positively regulate other cell types such as dendritic cells and mast cells. As TIM-3 is becoming an increasingly attractive therapeutic target because of its ability to reverse exhaustion in T cells, it is important to determine the regulatory nature of TIM-3 on T cells. To do this, we expressed Tim-3 ectopically in Tim-3- Jurkat T cells and observed that Tim-3 enhances instead of inhibits signaling downstream of the T-cell receptor and co-stimulator, CD28. Then, using a series of truncation and point mutants of Tim-3, we determined that Y256 and Y263 are the most crucial of the six conserved tyrosines in mediating Tim-3 signaling. Another unexpected finding was that in addition to apoptosis, gal-9 also induces the secretion of pro-inflammatory cytokines from T helper subsets independently of Tim-3.

The microenvironment of aerodigestive cancers contains tumor promoting inflammatory signals often involved in innate immunity. SCCHN is an epithelial malignancy characterized by the secretion of inflammatory mediators that can promote tumorigenesis and lymph node metastasis. The chemokine receptor CCR7 is a key molecule whose aberrant expression in SCCHN has been linked to pro-survival, invasive and metastatic pathways. Indeed, the selective upregulation of CCR7 in metastatic SCCHN tumors has been previously described. However, the mechanisms of CCR7 expression have not yet been elucidated. Inflammatory cytokines are known to upregulate CCR7 in immune cells through downstream NF-κB dependent mechanisms. In addition, antimicrobial peptides such as human β-defensin 3 (HBD3) are capable of promoting an inflammatory microenvironment and may possess tumor-promoting properties. Given the frequent overexpression NF-κB in SCCHN and its association with a more aggressive SCCHN phenotype, I hypothesized that NF-κB may be a key mediator of invasive and metastatic disease by promoting CCR7 expression in SCCHN tumors. Indeed, I identified and studied four potential NF-κB binding sites in the promoter region upstream of the CCR7 gene and report on their relative contribution to CCR7 expression in metastatic SCCHN. Furthermore, I demonstrate that HBD3 induces CCR7 expression in dendritic cells as well as primary SCCHN tumors in an NF-κB-dependent fashion. Interestingly, HBD3 stimulation provides anti-apoptotic signals to SCCHN cells, as evidenced by tumor resistance to cisplatin-induced cell death. As presented in this dissertation, these findings suggest that HBD3 represents a novel, NF-κB-regulated mediator of CCR7 expression and anti-apoptotic pathways, which may be exploited by developing SCCHN tumors to enhance their growth, survival and evolution into a metastatic phenotype. NF-κB appears to be a key regulator of basal and inducible CCR7 expression. The observed NF-κB induction of CCR7 and its subsequent downstream pathways provide clinically important therapeutic targets to control the progression and metastasis of SCCHN tumors.

Aspergillus fumigatus (A. fumigatus) is commonly associated with allergic bronchopulmonary aspergillosis (ABPA), which represents one of the most extreme manifestations of fungal allergy. Eosinophils are considered to be effector cells mediating lung dysfunction, but there is increasing appreciation that pulmonary neutrophilia also contributes to ABPA pathology. In our efforts to recapitulate this fungal allergic condition, we found that persistent exposure to A. fumigatus resulted in neutrophil and eosinophil-biased responses in BALB/c and C57BL/6 mice, respectively, and that the former mimicked the inflammation pattern observed in ABPA. By performing a comparative study, we found substantially higher lung TNF-á levels in neutrophil-rich BABL/c mice compared to C57BL/6 mice. TNF-á blockade or deficiency in BALB/c mice switched the response from neutrophilia to eosinophilia, implicating TNF-á as the key mediator. We identified CD11c+CD11b+Ly6C+ inflammatory dendritic cells (DCs) and macrophages as the primary sources of TNF-á, and that depletion of these CD11c+ cells in CD11c-DTR BALB/c mice dramatically reduced lung TNF-á levels. Importantly, BALB/c DCs demonstrated a stronger TNF-á-producing capacity than C57BL/6 DCs, strongly suggesting that this was the basis for the strain-associated TNF-á difference. As compared to TNF-áhigh BALB/c DCs, TNF-á low C57BL/6 DCs contained more repressive NF-êB p50 homodimers at the TNF-á promoter at early time points following A. fumigatus infection, and expressed notably less pattern recognition receptors, in particular TLR2 following persistent fungal exposure. These differences explained the strain-specific differential TNF-á production by DCs. In addition, TNF-á deficiency itself blunted the accumulation of Ly6c+CD11b+ DCs, implicating a positive feedback loop to amplify the cellular sources of TNF-á. Functionally, higher amounts of IL-5 in C57BL/6 and TNF-á-/- mice were associated with higher eosinophil counts, while collaboration between TNF-á and IL-17A triggered significantly higher levels of the neutrophil chemoattractants KC and MIP-2 in the BALB/c mice. In summary, our study identifies that TNF-á, acting as a molecular switch, orchestrates a sequence of events in DCs and CD4+ T cells and promotes pulmonary neutrophilia.

Oropharyngeal candidiasis (OPC; thrush) is an opportunistic oral infection caused by the commensal fungus Candida albicans that afflicts immunosuppressed and immunocompromised individuals. Although patients with HIV/AIDS are the prototypical population associated with development of thrush, there are a host of conditions that lead to susceptibility to OPC, including Hyper-IgE Syndrome (HIES). The autosomal dominant form of HIES (AD-HIES) is characterized by inherited dominant negative mutations in the STAT3 transcription factor, which is crucial for physiologic homeostasis and development in many tissues, including immune cells. Due to the ubiquitous nature of STAT3, mutations in this gene lead to a highly varied set of clinical sequelae, including oral thrush. Numerous cytokines utilize STAT3 to mediate downstream signaling, including IL-6 and IL-23, which are crucial for the differentiation and expansion of T helper 17 (Th17) cells, respectively. IL-23 is necessary for 1) in vivo expansion of Th17 cells, and 2) the secretion of IL-17 and other proinflammatory cytokines. Recently, the Th17 lineage has been shown to play a major role in host defense against OPC. Our lab has previously shown that mice deficient in the IL-17 receptor (IL-17RAKO) and IL-23p19 (IL-23KO) are susceptible to OPC. Furthermore, saliva from both AD-HIES patients and IL-23KO mice exhibits reduced ability to kill C. albicans ex vivo. However, despite the requirement for both IL-17 and IL-23 in protecting against OPC, we found that CD4-specific STAT3 knockout (CD4(stat3KO)) mice are not susceptible to our model of OPC. These data suggested that STAT3 mediates the initial response to oral C. albicans challenge by inducing an IL-17-producing subset other than Th17. Experiments carried out in this study revealed that the initial source of IL-17 is lymphocytic, and is likely multifactorial, involving both γδ and αβ T cells. Furthermore, in this study, natural killer and natural killer T cells were shown to have no apparent role in IL-17 secretion in response to C. albicans challenge in the oral cavity. In addition, while a memory or rechallenge response may involve CD4+ Th17 cells, this lineage appears to be unessential for the initial response to C. albicans in the oral cavity. Although we also attempted to examine the immune response to OPC using a mouse model of HIES harboring a transgenic STAT3 mutation, these mice proved resistant to C. albicans challenge at the buccal mucosa. However, despite the OPC resistance, they did exhibit a skin phenotype consistent with STAT3 deficiency, including eczematous lesions and delayed wound healing. These findings suggest that this mouse model may be used to study the role of STAT3 in both eczema as well as the immune response to cutaneous Staphylococcus aureus and C. albicans skin infections in the context of STAT3.

Natural killer cells (NK) belong to a heterogenous group of cells primarily thought to function against tumors. However, additional studies have indicated a broader role for these cells in immune regulation, including the regulation of B cells. Peripherial blood lymphocytes from rheumatoid arthritis (RA) patients and normal individuals were purified into three groups: T, B, and NK cells. Purity of these populations was assessed by Facs analysis (94%, 90% and 86% respectively.) These cells were used in a proliferative assay to determine the affects of NK cells on polyclonal proliferation of B cells. The normal and rheumatoid arthritis patients were previously assessed in terms of a proliferative response to recall antigen and designated as either anergic (non-responsive) or non-anergic (responsive). Inhibition of B cell proliferative response was seen in the normal (-35% ± 4.5) and non-anergic subgroup (-25.6 ± 3.8), but not in the anergic subgroup (+11.2 ± 76.8). While removal of the CD57⁺ NK cells did not reverse the inhibition seen in the normal or non-anergic individuals, elimination of the CD16⁺ cells did (-1 ± 0.816 and 6.4 ± 6.5 respectively), with no significant effect on the anergic subpopulation. To investigate the dependence of this NK regulation on T cells, CD8⁺ T cells were removed. While there was no significant difference seen in the inhibition levels in the normal and non-anergic groups, inhibition in the anergic population significantly increased. This work suggests that NK cells are involved in limiting the proliferative response of activated B cells and this is dependent on the CD16⁺ subset of NK cells. While non-anergic rheumatoid arthritis individuals show no difference from the normal group in this function, anergic patients do, with the CD8⁺ T cells (suppressor/cytotoxic) subset apparently involved. Functional differences seen in the RA group were reflected by abnormal expression of cell surface antigens, as assessed in peripherial blood lymphocytes by Facs analysis. An increase in the expression of the CD8 marker in both the anergic and non-anergic individuals supports the hypothesis of an underlying T suppressor cell defect. Compensation by the two RA subgroups appears to be different. The anergic group had higher numbers of CD16⁺ cells, while the non-anergic group had NK cells that functioned at a higher level, as determined by the decrease expression of the CD56 marker in the surface of these cells. These data serve to reinforce the idea that individuals with RA belong to a heterogenous group.

Peripheral blood lymphocytes cultured in interleukin 2 (IL-2) acquire the ability to lyse tumor cell targets in a non-major histocompatability complex (MHC) restricted manner. These lymphokine activated killer (LAK) cells release tumor necrosis factor-alpha (TNF) and interferon-gamma (IFN-γ) in culture and when stimulated by interaction with tumor cells in vitro. The capacity of several suppressive factors to affect the release of TNF and IFN-γ from LAK cells which have been stimulated with K562 erythroleukemia cells was investigated. A number of agents known to inhibit mononuclear phagocyte secretion of TNF were tested, including prostaglandin E₂ (PGE₂), alpha-globulins and the synthetic protease inhibitor tosyl-arginine methyl ester (TAME). The alpha globulins and TAME are thought to suppress TNF release from monocytes by inhibiting proteolytic cleavage of the cytokine from the cell surface. The addition of PGE₂, whole plasma alpha globulins, purified alpha-1-acid glycoprotein, and TAME inhibited TNF and IFN-γ release in a concentration dependent manner. These inhibitory factors appear to act directly on the lymphocytes to suppress cytokine secretion, as the presence of monocytes or metabolically active tumor cells was not required. The effects of alpha-1-acid glycoprotein on LAK cells were also investigated. The addition of this alpha globulin fraction inhibits the incorporation of ³[H] thymidine by LAK cells stimulated with tumor cells, and results in a detectable decrease in TNF mRNA. The capacity of alpha-1-acid glycoprotein to suppress the release of TNF production also resulted in an inhibition of LAK cell generation which was reversed by the addition of exogenous TNF. In contrast to their suppressive effect on peripheral blood monocytes, the protease inhibitors alpha-1-antitrypsin and alpha-2-macroglobulin enhanced TNF secretion from LAK cells exposed to leukemia cells. Analysis of cell surface TNF expression by fluorescence activated cell sorting (FACS) suggest that the observed differences in regulation reflect the capacity of particular phenotypes to express TNF as a transmembrane molecule. The data presented here indicate that the regulation of TNF release by LAK cells stimulated with tumor cells differs significantly from that previously observed in monocytes and suggest a regulatory role for alpha-1-acid glycoprotein in TNF secretion by LAK cells.

One third of the world’s population is infected with Mycobacterium tuberculosis, causing two million deaths annually. The bacteria avoid immune clearance by persisting within macrophages by subverting normal phagosome maturation and acidification. In order to spread, the bacteria induce necrotic death of its host macrophage, broadcasting the infection into neighboring cells. However, it has long been appreciated that the apoptotic, rather than necrotic death of an infected macrophage results in bacterial growth suppression, improved adaptive immune response and survival. The mechanism for apoptosis-mediated bacterial suppression has hitherto remained unknown. In this dissertation we report that apoptosis itself is not intrinsically bactericidal. We find that following apoptosis, the M. tuberculosis-infected macrophage is engulfed by bystander macrophages through the process of efferocytosis. Efferocytosis, or apoptotic cell clearance, is a critical function of macrophages; however, little is known regarding efferocytosis of infected apoptotic cells. We find that M. tuberculosis-infected macrophages die by apoptosis more commonly than found previously. By confocal microscopy we observed that apoptotic macrophages are rapidly engulfed by uninfected macrophages. Efferocytosis of M. tuberculosisinfected macrophages occurs in vitro with all macrophage types tested and in vivo- specifically in the lung, indicating that efferocytosis could play an important role during infection. We developed an uninfected macrophage co-culture system in which we observe efferocytosis and define conditions in which it occurs. Using this co-culture system we observe a suppression of bacterial growth. By blocking efferocytosis, we have found that the engulfment of infected cells is required for M. tuberculosis control in the macrophage co-culture system, demonstrating that efferocytosis is a novel antibacterial mechanism. We then demonstrated using transmission electron microscopy that the M. tuberculosis-containing efferocytic phagosome is structurally distinct from the traditional M. tuberculosis phagosome. Bacteria from within the efferocytic phagosome are unable to halt its maturation, and as such are delivered to lysosomes. Furthermore, we find that following efferocytosis, M. tuberculosis are killed. While efferocytosis is recognized as a constitutive housekeeping function of macrophages, our work indicates that is should also be viewed as an antimicrobial effector mechanism.

The ability for CD4 T cells to efficiently search for and subsequently respond to microbial pathogens is essential for protective immunity, but mechanisms controlling these responses are not completely understood. In this thesis I study the regulation of CD4 T cell responses at two different stages during an immune response. First, I analyze one of the most basic mechanisms by which T cells search for and become activated by an antigenic stimulus during the initial events in an adaptive immune response. Using human memory CD4 T cells in vitro I have identified a novel role for actin-rich invadapodia-like protrusions (ILPs) in overcoming the energy barrier required for the T cell receptor (TCR) to send signals into T cells when interacting with peptide-loaded MHC II. My studies show that ILPs, which are used during migration, are also essential for surveying the surface of other cells during cellular communication. Secondly, I explore the costimulatory requirements and function of T follicular regulatory \((T_{FR})\) cells, a newly identified subset of regulatory T \((T_{REG})\) cells. Using mouse models, I have discovered that the costimulatory receptor PD-1 inhibits the differentiation and function of \(T_{FR}\) cells in vivo. My work also has revealed that \(T_{FR}\) cells can circulate within the blood and that blood TFR cells can potently inhibit B cell mediated antibody production in vivo. Taken together, the studies presented here not only provide insights into the very initial events leading to adaptive immunity, but also demonstrate how adaptive immunity is controlled during the effector phase of an immune reaction.

The ATM protein senses DNA double-stranded breaks (DSBs) and facilitates proper repair. B and T lymphocytes of ATM-deficient patients have increased antigen receptor locus translocations associated with aberrant V(D)J recombination. Correspondingly, ATM-deficient humans are predisposed to both T and B cell malignancies. However, ATM-deficiency in mice only leads to T cell lymphomas, all of which harbor T cell receptor locus translocations resulting from aberrant V(D)J recombination. The first goal of this study was to assess whether ATM-deficient B cell lymphomas occur in mice in which V(D)J recombination is targeted to the c-myc oncogene or in which B cell survival is increased via enforced expression of the anti-apoptotic Bcl2 protein. We found that, in the absence of ATM, either inserting the V(D)J substrate into c-myc or enhancing B cell survival led to the development of mature B cell lymphomas in a subset of mice. Moreover, combining both genetic alterations led to complete penetrance of mature B cell lymphomas in the ATM-deficient background.