The observations that substance P (SP) is localized in the anterior pituitary gland (AP) and is regulated by the hormonal status of the animal, as well as the demonstration of SP binding sites in the AP, have led to the idea that SP may participate in the regulation of AP function. Numerous and sometimes contradictory reports of SP effects on AP hormone secretion, particularly on luteinizing hormone (LH), left the question of whether SP acts directly at the level of the AP to regulate LH secretion still unanswered. To investigate a possible physiological function of SP in the AP, the effects of exogenous SP on LH secretion from AP cells from adult and prepubertal male and female rats in short term culture were studied. It was found that SP (100nM-1μM) significantly stimulates LH release in cultured AP cells and that this effect varies as a function of age and sex. SP has no significant effect on LH release from AP cells of male and female prepubertal rats. After day 30 a sharp increase in the response to SP occurs in both sexes. This level of responsiveness continues through adulthood in AP cells from the female rat. In contrast, AP cells from male rats failed to respond during adulthood (over 50 days of age) but were highly responsive during the peripubertal period (30-35 days). The possibility that the responsiveness to SP is influenced by the endocrine status of the animal was investigated by exposing AP cells from responding animals to androgens in vivo and in vitro. It was found that AP cells from female rats treated with androgen were less responsive to 100nM SP but did respond at higher doses of SP. SP effects on AP function were further analyzed in experiments using radioligand binding assays to assess possible changes in SP receptor number or affinity as related to age and sex. In AP membranes from female rats, maximum binding is 8-fold higher (Bmax=4.2 pmo1/mg membrane protein) than in AP membranes from male rats (Bmax=560fmo1/ mg membrane protein). These studies suggest a role for SP as a secondary regulator of LH secretion with possible physiological significance for reproductive function.

T lymphocytes are essential components of the immune system and as such are continually regulated by a variety of factors. Every step of their development, survival and function is tightly monitored to ensure their ability to recognize most foreign agents and mount adaptive immune responses during pathogenic infections, while remaining tolerant to self-antigens. Among the many factors that participate in the regulation of T cell development and function are the cytokines. Cytokines that signal through the common gamma (γс) chain and the Janus kinase 3 (Jak3) include IL-2, -4, -7, -9, -15, and -21 and have been implicated in the regulation of every stage in the life of a T cell. Therefore, it is not surprising that mutations in the γс chain or Jak3 lead to a SCID condition in humans and mice. Specifically, Jak3-deficient mice are characterized by a reduction in thymic cellularity and dysregulated T cell homeostasis. They have an expansion of memory-like CD4+ mature T cells and an almost complete absence of mature CD8+ T cells. By investigating the TCR repertoire of CD4+ T cells in the thymus and spleen of Jak3-/- mice, I deduced that the CD4+ T cell activation and expansion is TCR-specific and takes place in the periphery of the mice. After crossing Jak3-deficient mice to Bcl-2 transgenic mice I showed that the developmental block observed in Jak3-/- mice could not be rescued by the anti-apoptotic factor, despite the fact that its expression did increase, slightly, the total numbers of developing thymocytes. The enforced expression of Bcl-2 was also not sufficient to revert the dysregulation of T cell homeostasis in Jak3-/- mice. Finally, in order to further understand the role played by γс cytokines during T cell function, I investigated the ability of mature Jak3-/- CD8+ T cells to become activated and differentiate into effector cells in response to a viral infection. My results indicate that CD8+ T cells are activated and proliferate in response to a viral infection, but their survival, as well as their ability to proliferate and differentiate into effector cells are greatly impaired, resulting in the inability of Jak3-deficient mice to mount a protective response.

This thesis details my examination of several mechanisms for modulation of N-type calcium channels in neonatal rat superior cervical ganglion (SCG) neurons. The first part of this work characterizes cross-talk between two distinct mechanisms of modulation: readily-reversible inhibition induced by activation of heterotrimeric G-proteins (termed G-protein-mediated inhibition), and phosphorylation of the channel by protein kinase C (PKC). Data previously presented by other groups suggested that one effect of activating PKC is to prevent G-protein-mediated inhibition. The goal of this project was to confirm this hypothesis by testing functional competition between these two pathways. My findings show that G-protein-mediated inhibition blocks the effects of activating PKC, and that phosphorylation by PKC blocks G-protein-mediated inhibition, confirming that these two mechanisms are mutually exclusive. In addition, I investigated the effect of activating PKC on whole-cell barium currents in the absence of G-protein-mediated inhibition. When endogenous G-proteins were inactivated by dialyzing the cell with GDP-β-S, a guanine nucleotide that prevents activation of the G-protein's α subunit, activation of PKC with phorbol esters was without obvious effect on whole-cell current amplitude, fast and holding potential-dependent inactivation, and voltage-dependent activation, suggesting that PKC's principal role in modulating these currents is to prevent G-protein-mediated inhibition. From these results, I advanced Bean's 1989 model of reluctant and willing gating (induced by G-protein-mediated inhibition and relief of that inhibition, respectively). In this expanded model, reluctant channels, inhibited by G-proteins, are resistant to phosphylation by PKC (reluctant/P-resistant). Unmodulated channels are called willing/available, as they exhibit willing gating, and are available for either binding to a G-protein or phosphorylation by PKC. Finally, phosphorylation of a willing/available channel by PKC drives the channel into the willing/G-resistant state, in which the channel gates willingly, and is resistant to G-protein-mediated inhibition. These results are published in the Journal of General Physiology(2000; 115:277-286), and are presented in this thesis as Chapter II. In addition to membrane-delimited inhibition, N-type calcium channels are also subject to inhibition via a diffusible second-messenger pathway. In SCG neurons, this inhibition can be observed following stimulation of M1 muscarinic receptors by the agonist oxotremorine-M. Our lab previously hypothesized that the diffusible messenger involved might be the polyunsaturated fatty acid arachidonic acid (AA). To test this hypothesis, our lab examined the effect of bath-applied AA on whole-cell SCG calcium currents, and demonstrated that AA induces inhibition with similar properties as M1 muscarinic inhibition. An analysis of AA's effects on unitary N-type calcium currents, published by Liu and Rittenhouse in Journal of Physiology(2000; 525:391-404), revealed that this inhibition is mediated, at least in part, by both a significant increase in the occurrence of null-activity sweeps and a significant decrease in mean closed dwell time. Based on these results, our lab conducted an examination of AA's effects on whole-cell currents in SCG neurons, and found that AA-induced inhibition is mediated by an increase in holding potential-dependent inactivation and appears independent of AA metabolism. When I examined AA's effects in greater detail, I discovered that, in addition to inhibition, AA also appeared to cause significant enhancement of whole-cell currents. The results characterizing AA's general effects on whole-cell calcium currents in SCG neurons have been published in American Journal of Physiology - Cell Physiology(2001; 280:C1293-C1305). Because my finding that AA enhances whole-cell neuronal calcium currents revealed a novel pathway through which this current can be modulated, I proceeded to characterize this effect. My results showed that enhancement develops significantly faster than inhibition, suggesting different mechanisms or pathways. In addition, dialyzing the cell with BSA, a protein that binds fatty acids, blocked the majority of AA-induced inhibition, but did not reduce enhancement, suggesting that enhancement is independent of inhibition and might be mediated at an extracellular site. Using fatty acid analogs that cannot cross the cell membrane, I confirmed that enhancement occurs extracellularly. My data also indicate that AA-induced enhancement of whole-cell currents does not require metabolism of AA, consistent with enhancement being mediated directly by AA. I also examined the biophysical characteristics of enhancement, and found that both an increase in the voltage sensitivity of activation and an increase in activation kinetics underlie this effect. Finally, using both pharmacological agents and a recombinant cell line, I presented the first demonstration that AA enhances N-type calcium current. These findings are described in detail in a paper recently published in American Journal of Physiology - Cell Physiology(2001; 280:C1306-C1318), and are presented in this thesis as Chapter III. In our investigation of AA's effects on whole-cell calcium currents, we utilized a voltage protocol, in conjunction with pharmacology, to enhance the level of L-type current in these cells. Since whole-cell calcium currents in SCG neurons are comprised of mostly (80-85%) N-type current, with the remaining current comprised of mostly L-type current, this approach allowed us to examine both N- and L-type currents. When currents are recorded in the presence of 1 μM FPL 64174 (FPL), a benzoyl pyrrole L-type calcium channel agonist first described in 1989, stepping the membrane potential to -40 mV following a test pulse to +10 mV generates a slowly-deactivating ("tail") current. This tail current is made up entirely of L-type current, and allows us to readily investigate the effect of various modulatory mechanisms on this current type. Although FPL has been used for almost a decade to study L-type calcium currents, activity of FPL on N-type calcium currents has not been investigated. Because our lab routinely uses micromolar concentrations of FPL to measure whole-cell and unitary calcium currents in neuronal cells, I tested whether FPL has any effects on N-type calcium current. Therefore, I examined the effect of FPL on whole-cell calcium currents in an HEK 293 cell line that expresses recombinant N-type calcium channels. Application of 1 and 10 μM FPL caused significant, voltage-independent inhibition of currents, demonstrating that FPL inhibits N-type calcium current. Thus, at micromolar concentrations, FPL is not selective for L-type calcium current, and any examination of its effects on whole-cell calcium currents should take this into account. The results describing FPL's effects on L- and N-type calcium currents are included in a manuscript currently in preparation, and are presented as Chapter IV.

This thesis details my examination of several mechanisms for modulation of neuropeptide release via voltage-dependent and voltage-independent intraterminal signaling in isolated neurohypophysial terminals. The first part of this work characterizes depolarization-induced neuropeptide release in the absence of extracellular calcium. The goal of this project was to examine the relationship between depolarization-induced release of intracellular calcium stores and depolarization-secretion coupling of neuropeptides. We demonstrate that depolarization in the absence of extracellular calcium induced by either High K+ or electrical stimulation induces a rise in [Ca2+]i and subsequent neuropeptide release from Hypothalamic Neurohypophysial System (HNS) terminals. A portion of extracellular calcium-independent neuropeptide release is due to intraterminal calcium, but the remaining depolarization-induced release may be due to calcium-independent voltage-dependent (CIVD) release (Zhang and Zhou, 2002; Zhang et al., 2004; Yang et al., 2005). Nevertheless, our results clearly show that extracellular calcium is notnecessary for depolarization-induced neuropeptide secretion from these CNS terminals. In addition, I investigated the role of internal calcium stores in mediating μ-opioid inhibition of voltage-gated calcium channels (VGCCs). Inhibition of VGCCs via μ-opioid agonists has been shown to reduce neuropeptide release in response to High K+ stimulation of isolated terminals (Bicknell et al., 1985b; Russell et al., 1993; van Wimersma Greidanus and van de Heijning, 1993; Munro et al., 1994; Ortiz-Miranda et al., 2003; Russell et al., 2003; Ortiz-Miranda et al., 2005). My findings show μ-opioid inhibition, of VGCC and High K+-mediated rise in [Ca2+]i, are via a voltage-independent diffusible second-messenger targeting release of calcium from ryanodine-sensitive stores, possibly mediated via the cyclic ADP ribose signaling pathway. Furthermore, I detail a different intracellular messenger pathway mediating the κ-opioid inhibition of VGCC and High K+-mediated rise in [Ca2+]ii. In contrast to the μ-opioid inhibition, κ-receptor activation is coupled to a voltage-dependent membrane-delimited pathway. Inhibition of neuropeptide release via both endogenous and exogenous κ-opioid agonists has been extensively studied (Bicknell et al., 1985a; Nordmann et al., 1986a; Wammack and Racke, 1988; Munro et al., 1994; Ingram et al., 1996; Rusin et al., 1997a). My investigation shows that the κ-inhibition of VGCC is voltage-dependent and is furthermore, relieved within the context of a physiological burst of action potentials (APs). This physiologically-evoked, activity-dependent modulation of VGCC and subsequent release, represents an important mechanism for short-term synaptic plasticity at the level of the terminals. Given the ubiquitous nature of voltage-dependent G-protein signaling in the CNS, our results may prove important in understanding modulatory effects of specific bursting patterns throughout the CNS. In the last 30 years the neurohypophysial system has proven to be an excellent system to study the complexities of depolarization-secretion coupling (DSC). There have been many advances in our understanding of the underlying mechanisms involved and their physiological implications. The current work focuses on two important features of DSC; voltage and calcium. Although in many ways these two are intrinsically linked through VGCC activation, we have found that in isolated HNS terminals that is not always the case. We have further found that when voltage and calcium influx are linked during DSC, modulation by opioids may or may not be linked to activity-dependent relief depending on the opioid receptor activated. This finding has important implications in neuropeptide release during patterned stimulation in vivo. As I will discuss further, many factors play into the complexities of the regulatory mechanisms involving release. As investigations into this remarkable field continue, I hope to have contributed a valuable piece to the puzzle.

C1q TNF Related Protein 3 (CTRP3), is a cytokine that is primarily secreted from adipose tissue, which classifies it as an adipokine. Our previous research has shown that CTRP3 prevents alcoholic fatty liver disease (ALD) in male mice. However, even when accounting for confounding factors such as absolute and relative alcohol intake, females are more sensitive to the effects of consumption compared to male mice. Therefore, the goal of this project was to determine whether CTRP3 prevented ALD in female mice. Methods: Female wild type (WT) and female CTRP3 transgenic over expressing (Tg) mice were fed an ethanol containing liquid diet (5% v/v) for 6 weeks. Daily weight and food intake measurements were taken and external heat-pads were placed under a portion of the cage to facilitate thermoregulation. Hepatic steatosis was determined by total triglyceride quantification and lipid droplet quantitation in liver sections. Data were analyzed by repeated measures ANOVA, t-test, or Log-rank (Mantel-Cox) test as appropriate. Results: There was no difference between WT and Tg mice in food intake or body weight. There was no difference in survival between WT and Tg mice, however, Tg mice trended towards a reduced rate of survival compared with WT mice (78% in WT versus 44% in Tg, p=0.13). Stereological analysis indicated no difference in the percent of lipid liver volume between the two groups (WT 7.2±3.6 vs Tg 5.1±4.1%). This finding was consistent with no difference in total hepatic triglyceride accumulation observed between WT and Tg mice (12.7±4.4 vs. 13.1±6.8 mg triglycerides/gram liver protein). Conclusion: Combined these data indicate that unlike previous studies with male mice, CTRP3 is not protective against alcohol-induced hepatic steatosis in female mice. Combined, these data indicate that the adipokines such as CTRP3 contribute to physiology in a sex-specific manner.

Since the first cases were reported over thirty years ago, great strides have been made to control disease progression in people living with HIV/AIDS. However, current estimates report that there are about 34 million individuals infected with HIV worldwide. Critical in the ongoing fight against this pandemic is the continuing development of highly active anti-retroviral therapies, ideally those with novel mechanisms of action. Currently, there are no medications approved for use that exploit the HIV-1 MA protein, despite its central role in multiple stages of the virus life cycle. This thesis sought to examine whether a highly conserved glutamate residue at position 99 in the understudied C-terminal helix of MA is required for HIV-1 replication. I characterized a panel of mutant viruses that contain different amino acid substitutions at this position using viral infectivity studies, virus-cell fusion assays, and immunoblotting. In doing so, I found that substitution of this glutamate with either a valine (E99V) or lysine (E99K) residue disrupted Env incorporation into nascent HIV particles, and abrogated their ability to fuse with target-cell membranes. In determining that the strain of HIV could affect the magnitude of E99V-associated defects, I identified a compensatory substitution at MA residue 84 that rescued both E99V- and E99K-associated impairments. I further characterized the MA E99V and E99K mutations by truncating HIV Env and pseudotyping with heterologous envelope proteins in an attempt to overcome the Env incorporation defect. Unexpectedly, I found that facilitating fusion at the plasma membrane was not sufficient to reverse the severe impairments in virus infectivity. Using quantitative PCR, I determined that an early post-entry step is disrupted in these particles that contain the E99V or E99K MA substitutions. However, allowing entry of mutant virus particles into cells through an endosomal route conferred a partial rescue in infectivity. As the characterization of this post-entry defect was limited by established virological methods, I designed a novel technique to analyze post-fusion events in retroviral infection. Thus, I present preliminary data regarding the development of a novel PCR-based assay that monitors trafficking of the viral reverse transcription complex (RTC) in an infected cell. The data presented in this thesis indicate that a single residue in MA, E99, has a previously unsuspected and key role in multiple facets of HIV-1 MA function. The pleiotropic defects that arise from specific substitutions of this amino acid implicate a hydrophobic pocket in MA in Env incorporation and an early post-entry function of the protein. These findings suggest that this understudied region of MA could be an important target in the development of a novel antiretroviral therapy.

Membrane excitability, gene expression, and neurotransmitter release are all controlled by voltage-gated L-type Ca2+ (L- )channels. In turn, Ca2+ channels are highly regulated by signal transduction cascades initiated by G protein-coupled receptor (GPCR) activation. In medium spiny neurons of the striatum, both the muscarinic M1 receptors (M1R) and dopaminergic D2 receptors (D2R) specifically inhibit the Cav1.3 L-channel. In Chapters III and IV, the pathways downstream of M1Rs and D2Rs are examined to determine whether an overlap or intersection in inhibition of Cav1.3 occurs by these two receptors. Transient transfection of Cav1.3 channels in HEK 293 cells, stably transfected with the M1R, and in ST14A cells were used as model systems. While a further characterization of ST14A cells determined that they exhibit a striatal profile, D2Rs or M1Rs did not inhibit Cav1.3. Lack of current inhibition may be due to the finding of no detectable expression of phospholipase Cβ-1 protein in ST14A cells. Ca2+ channels are multiprotein complexes comprised of α1, β, and α2δ subunits. While the actions of arachidonic acid (AA) have been shown to mimic M1R inhibition of L-current in superior cervical ganglion neurons, the precise identity of the L-channel in these neurons -either Cav1.2 or Cav1.3 or both- is not known. The transfected model systems allowed for the analysis of whole-cells currents with different β subunit combinations as well as the study of only Cav1.3 channels. In Chapter III, I show that activation of M1Rs with the agonist Oxo-M inhibited Cav1.3 channels coexpressed with either β1b, β2a, β3, or β4 subunits. Surprisingly, the magnitude of Cav1.3, β2a currents was inhibited less than Cav1.3 currents with other β subunits. In Chapter V, AA is shown to mimic the profile of M1R stimulation on Cav1.3 currents. The magnitude of Cav1.3, β2a currents was inhibited less than Cav1.3 currents with other β subunits by AA. This discovery points to a novel role for accessory β subunits in altering the magnitude of AA inhibition and kinetic changes of Cav1.3. Arachidonic acid (AA) inhibits Ca2+ channels by an unknown mechanism at an unknown site. In Chapter V, I found that Cavl.3 inhibition by AA was state-dependent and most likely stabilizes a closed channel conformation. The finding that the Ca2+ channel accessory β subunit alters the magnitude of AA inhibition and kinetic changes of Cav1.3 suggests that AA could alter processes which rely on L-channels such as Ca2+-dependent gene expression, secretion and membrane excitability.

One of the hallmarks of an infection with pathogenic HIV-1 is the elevated level of immune activation that leads to rapid progression to AIDS. Surprisingly, nonhuman primates naturally infected with SIV do not exhibit an augmented activation phenotype nor severe immunodeficiency. One of the viral components implicated in determining the state of immune activation is the accessory protein Nef which has been demonstrated to affect T cell signaling pathways from within the intracellular compartment and for Nef from SIV, to downregulate TCR surface expression. Recently, Nef from HIV-1 and SIV have been demonstrated to bind the ζ chain of the TCR which functions as the primary signaling subunit of the receptor. However, the molecular details of the Nef-TCRζ interaction as well as the role of complex formation in modulation of immune activation remain largely unknown. This thesis describes work directed at elucidating the biochemical and structural features of the Nef-TCRζ interaction and the functional consequences of complex formation relevant to T cell activation. Chapter I provides a brief introduction on HIV/SIV classification and pathogenesis with an emphasis on Nef and its pleiotropic function in T cells. Chapter II describes the biochemical characterization of the interaction of the conserved core domain of Nef proteins from HIV-1, HIV-2 and SIV with the cytoplasmic domain of TCRζ. The core domains of HIV-2 ST and SIVmac239 are demonstrated to bind the cytoplasmic domain of TCRζ at two distinct regions and with different affinities. In contrast, the core domain of HIV-1 isolate ELI Nef only binds to one region and with the weakest calculated affinity among the HIV-1, HIV-2 and SIV Nef proteins studied. In addition, both the N-terminal domain and the strong TCRζ-binding core domain of SIVmac239 Nef each are demonstrated to be necessary but not sufficient for downregulation of TCR surface expression. Chapter III describes the crystallization and structure determination methods used to solve the crystal structures of the core domain of SIVmac239 Nef in complex with two overlapping TCRζ polypeptides. Crystals of Nef in complex with the longer TCRζDP1 (L51-D93) polypeptide grew in a tetragonal space group but only diffracted to low resolution. In contrast, crystals of the Nefcore-TCRζA63-R80 complex grew in an orthorhombic space group and diffracted to high resolution but were nearly perfectly pseudo-merohedrally twinned thus complicating structure determination. Following identification of the twin law relating the twin domains, the structure of the Nefcore-TCRζA63-R80 complex was determined using refinement procedures that accounted for crystal twinning to 2.05 Å. The structure of the Nefcore-TCRζDP1 complex was solved to 3.7 Å from a single non-twinned crystal. The altered crystal packing induced by the shorter TCRζA63-R80polypeptide is postulated to have led to a reduction in crystal symmetry and increase in proneness to crystal twinning. Chapter IV provides a detailed analysis of the structure of the Nefcore-TCRζA63-R80 complex and demonstrates its effect on modulation of Src family kinase activity. The TCRζ polypeptide adopts an alpha helical conformation and occupies a hydrophobic crevice on Nef not shared by any of Nef’s reported interaction partners. The interaction of Nefcore with TCRζ is mediated primarily by the burial of hydrophobic residues on TCRζ (L75, L77) in a hydrophobic pocket on Nef and a salt bridge between a glutamic acid (E74) on TCRζ and a basic patch on Nef consisting of two conserved arginines (R105, R106). The TCRζ polypeptide additionally orders the N-terminus of Nefcore into a polyproline type II helix that has been described to bind the SH3 domain of Src family kinases. We demonstrate that in vitro phosphorylation of TCRζcyt by Fyn and Src is specifically augmented by HIV-1 and SIV Nefcoreand suggest that Nef-TCRζ complex formation cooperatively enhances kinase activity. Chapter V contains overall conclusions, future directions and a model illustrating the proposed role of the Nef-TCRζ interaction in immune activation modulation. The Appendices contain sequences of the proteins, gene constructs and primers used in this work.

The instructive signals received by T cells during the programming stages of activation will determine the fate of effector and memory populations generated during an immune response. Members of the tumor necrosis factor (TNF) superfamily play an essential role in influencing numerous aspects of T cell adaptive immune responses including cell activation, differentiation, proliferation, survival, and apoptosis. My thesis dissertation describes the involvement of two such members of the TNF superfamily, TNF and FasL, and their influence on the fate of T cells early during responses to viral infections and to the induction of transplantation tolerance. TNF is a pleiotropic pro-inflammatory cytokine that has an immunoregulatory role in limiting the magnitude of T cell responses during a viral infection. Our laboratory discovered that one hallmark of naïve T cells in secondary lymphoid organs is their unique ability to rapidly produce TNF after activation and prior to acquiring other effector functions. I hypothesized that T cell-derived TNF will limit the magnitude of T cell responses. The co-adoptive transfer of wild type (WT) P14 and TNF-deficient P14 TCR transgenic CD8+ T cells, that recognize the GP33 peptide of lymphocytic choriomeningitis virus (LCMV), into either WT or TNF-deficient hosts demonstrated that the donor TNF-deficient P14 TCR transgenic CD8+ T cells accumulate to higher frequencies after LCMV infection. Moreover, these co-adoptive transfer experiments suggested that the effect of T cell-derived TNF is localized in the microenvironment, since the TNF produced by WT P14 TCR transgenic CD8+ T cells did not prevent the accumulation of TNF-deficient P14 TCR transgenic CD8+ T cells. To determine if T cell-produced TNF is acting on professional APC to suppress the generation of virus-specific T cell responses, I performed co-adoptive transfer experiments with WT P14 TCR transgenic CD8+ and TNF-deficient P14 TCR transgenic CD8+ T cells into TNFR1/2 (1 and 2) deficient mice. These experiments demonstrated that the absence of TNFR1/2 signaling pathway in the host cells resulted in a greater accumulation of WT P14 TCR transgenic CD8+ T cells, thereby considerably diminishing the differences between donor WT P14 TCR transgenic CD8+ and donor TNF-deficient P14 TCR transgenic CD8+ T cells. The increased frequency and absolute numbers of WT P14 TCR transgenic CD8+ T cells in TNFR1/R2 deficient recipients suggests that one mechanism for the suppressive effect of T cell-derived TNF on antigen-specific T cells occurs as a result of TNFR signaling in the host cells. However, the donor TNF-deficient P14 TCR transgenic CD8+T cells still accumulated to higher frequency and numbers compared to their donor WT transgenic counterparts. Together, these findings indicate that T cell-produced TNF can function both in an autocrine and a paracrine fashion to limit the magnitude of anti-viral T cell responses. Given the immunoregulatory role of TNF and the ability of peripheral naïve T cells to produce this cytokine, I questioned at what stage of development do T cells become licensed to produce this cytokine. The peripheral naïve T cell pool is comprised of a heterogeneous population of cells at various stages of development, a process that begins in the thymus and is completed after a post-thymic maturation phase in the periphery. I hypothesized that naïve T cells emigrating from the thymus will be competent to produce TNF only after undergoing a maturation process in the periphery. To test this hypothesis, I compared cytokine profiles of CD4+ and CD8+single positive (SP) thymocytes, recent thymic emigrants (RTEs) and mature-naïve (MN) T cells during TCR activation. SP thymocytes exhibited a poor ability to produce TNF when compared to splenic T cells despite expressing similar TCR levels and possessing comparable activation kinetics with respect to the upregulation of CD25 and CD69 following stimulation. The reduced ability of SP thymocytes to produce TNF correlated with a decreased level of detectable TNF message following stimulation when compared to splenic counterparts. Stimulation of SP thymocytes in the context of a splenic environment did not fully enable TNF production, suggesting an intrinsic defect in their ability to produce TNF as opposed to a defect in antigen presentation. Using a thymocyte adoptive transfer model, I demonstrate that the ability of T cells to produce TNF increases progressively with time in the periphery as a function of their maturation state. RTEs identified by the expression of green fluorescent protein (GFP) (NG-BAC transgenic mice), showed a significantly enhanced ability to express TNF relative to SP thymocytes, but not to the extent of MN T cells. Together, these findings suggest that TNF expression by naïve T cells is regulated via a gradual licensing process that requires functional maturation in peripheral lymphoid organs. This highlights the functional heterogeneity of the naïve T cell pool (with respect to varying degrees of TNF production) during early T cell activation that can contribute to the many subsequent events that shape the course of an immune response. The productive activation of naïve T cells requires at least initial two signals; the first being through the TCR and the second is the engagement of co-stimulatory molecules on the surface of the T cells. T cells activated in the absence of co-stimulation become anergic or undergo cell death. Agents that block co-stimulation of antigen-specific T cells are emerging as an alternative to immunosuppressive drugs to prolong allograft survival in transplant recipients. Targeted blockade of CD154-CD40 interactions using a αCD154 monoclonal antibody (MR1) with a simultaneous transfusion of allogeneic splenocytes (donor specific transfusion or DST) efficiently induces tolerance to allografts. This co-stimulation blockade-induced tolerance is characterized by the deletion of host alloreactive T cells within 24 hours of treatment. Toll-like receptor (TLR) agonists abrogate tolerance induced by co-stimulation blockade by impairing the deletion of host alloreactive T cells and resulting in allograft rejection. The goal of my study was to determine the underlying molecular mechanisms that protect host alloreactive T cells from early deletion after exposure to TLR agonists. I hypothesized that TLR ligands administered during co-stimulation blockade regimen differentially regulate the expression of pro- and anti-apoptotic molecules in alloreactive T cells, during the initial stages of activation thereby preventing deletion. To test this hypothesis, I used syngeneic bone marrow chimeric mice containing a trace population of alloreactive KB5 TCR transgenic CD8+ T cells (KB5 Tg CD8+ T cells) that recognize H-2Kb as an alloantigen. I show here that KB5-CD8+ T cells downregulate CD127 (IL-7R!) and become apoptotic as early as 12 hrs after co-stimulation blockade. In contrast, KB5 Tg CD8+ T cells from mice treated with bacterial lipopolysaccaride (LPS) during co-stimulation blockade failed to become apoptotic, although CD127 was downregulated. Examination of the mRNA expression profiles of several apoptotic genes in purified KB5 CD8+ T cells from mice treated with DST+anti-CD154 for 12 hrs revealed a significant upregulation of FasL mRNA expression compared to the untreated counterparts. However, in vitro FasL blockade or in vivo cytotoxicity experiments with mice deficient in Fas or FasL indicated that the Fas-FasL pathway might not be crucial for tolerance induction. Another pro-apoptotic molecule BIM was upregulated in alloreactive T cells during co-stimulation blockade. This suggests that both the Fas pathway and BIM may be playing complementary roles in inducing deletional tolerance. Although FasL expression was diminished in alloreactive T cells in the presence of LPS, BIM expression was not diminished, suggesting that alloreactive T cells may still be vulnerable to undergo apoptosis. Concomitantly, I also found that LPS treatment during co-stimulation blockade resulted in non-specific upregulation of Fas expression in alloreactive T cells and non-transgenic T cells (CD4+ and CD8+). I demonstrate here that treatment with Fas agonistic antibody in vitrofor 4 hours can selectively induce apoptosis of alloreactive T cells that were believed to be refractory to apoptosis during LPS treatment. I speculate that under these conditions, deletion may be occurring due to the involvement of both Fas and BIM. Further, the mRNA expression profile revealed interleukin-10 (IL-10) as a molecule induced in alloreactive T cells during LPS treatment. Analysis of serum confirmed the systemic expression of IL-10 protein in mice treated with LPS during co-stimulation blockade. I hypothesized that LPS-induced IL-10 can have an anti-apoptotic role in preventing the deletion of alloreactive T cells and mediating allograft rejection. Contrary to my hypothesis, I found that IL-10 KO mice rejected allogeneic target cells similar to their WT counterparts, suggesting that IL-10 may not be required for LPS-mediated abrogation of tolerance induction. In addition to the systemic induction of IL-10, LPS also induced cytokines such as interleukin-6 (IL-6), TNF and interferon-γ (IFN-γ). These findings suggest that both Fas-FasL and BIM mediated apoptotic pathways may play complementary roles in inducing the early deletion of activated alloreactive T cells during tolerance induction. On the other hand, the mechanism of LPS mediated abrogation of tolerance induction can not be attributed to IL-10 alone as it may be playing a synergistic role along with other proinflammatory cytokines that may in turn result in the prevention of alloreactive T cell death during this process. Most importantly, these findings indicate that despite emerging from a pro-inflammatory cytokine milieu, alloreactive T cells are still susceptible to undergo Fas-mediated apoptosis during the first 24 hours after co-stimulation blockade and LPS treatment. Therefore, targeting the Fas-FasL pathway to induce deletion of alloreactive T cells during the peri-transplant period may still be a potential strategy to improve the efficacy of co-stimulation blockade induced transplantation tolerance during an environmental perturbation such as inflammation or infection.

Intracellular pathogens are recognized by a specialized subset of lymphocytes known as CD8+ T cells. Pathogen recognition by CD8+ T cells occurs through binding of T cell receptors (TCR) to processed antigens in complex with major histocompatibility complex (MHC) class I proteins. TCR engagement of antigens in complex with MHC class I typically lead to cytotoxic CD8+ T cell responses, which result in pathogen clearance. Due to the large number of foreign antigens that might be encountered by any given host a diverse repertoire of TCRs must be available for immune recognition. The main source of TCR diversity is generated by somatic recombination of the TCR genes. However, it has been suggested that selection eliminates so many recombined TCR sequences, that a high degree of TCR cross-reactivity must occur for the immune system to be able to recognize a large set of foreign pathogens. The work presented in this thesis was directed towards the understanding of the molecular mechanisms of CD8+ T cell recognition and cross-reactivity. Chapter I of this thesis gives an overview of the immune system, with a focus on CD8+ T cells. Chapter II of this thesis describes the development of novel bi-specific MHC heterodimers that are specific towards cross-reactive CD8+ T cells. Classically, MHC tetramers have been used for phenotypic characterization of antigen-specific T cells. However, identification of cross-reactive T cells requires the simultaneous use of two MHC tetramers, which was found to result in MHC tetramer cross-competition. For this reason, we generated bi-specific MHC heterodimers, which would not be affected by the affinity between the component peptide-MHC complexes for TCR. We generated T cell lines, which cross-react with antigens from lymphocytic choriomeningitis virus (LCMV) and vaccinia virus (VV), to test our bi-specific MHC heterodimers. We show that the heterobifunctional cross-linking utilized to generate bi-specific MHC heterodimers does not affect specific binding onto cross-reactive CD8+ T cells. Chapter III describes a mechanism for a cross-reactive CD8+ T cell response between the disparate antigens, lymphocytic choriomeningitis virus (LCMV)-GP34 (AVYNFATM) and vaccinia virus (VV)-A11R (AIVNYANL), which share the three underlined residues. The recognition determinants for LCMV-GP34 and VV-A11R were compared by an alanine/lysine scanning approach for both epitopes. Functional analysis of the mutated peptides clearly indicates that the shared P4N residue between LCMV-GP34 and VV-A11R is an important TCR contact for the recognition of both epitopes. In addition, we determined the crystal structures of both Kb-VV-A11R and Kb-LCMV-GP34. Structural analysis revealed that the two complexes are nearly identical structural mimics, which was unexpected due to the primary sequence disparity. Together with the functional studies, our results highlight that structural similarities between different peptide-MHC complexes can mediate cross-reactive T cell responses. Chapter IV of this thesis includes additional discussion, overall conclusions and future directions. Chapter V includes the protocols and the gene constructs that were used in this work. Also included in Chapter V are results from two unrelated incomplete projects which have yielded significant findings.