The differentiation of naive T helper (Th) cells into effector/memory subsets and their interaction with invading pathogens ultimately determines the outcome of infectious human diseases. The research detailed herein is aimed to delineate the transcriptional regulation of human Th cell differentiation and to further understand how these cells interact with microbial pathogens. Utilizing a lentiviral transduction system, we genetically modify primary human Th cells to further elucidate the roles of T-bet and GATA-3 during human Th cell differentiation. Furthermore, we describe how two important human pathogens, human immunodeficiency virus (HIV)-1 and Helicobacter pylori (H. pylori), usurp or perturb Th cell function, possibly contributing to their evasion of human immune responses. Our results lend novel insight into the regulation of human Th cell differentiation, function and interaction with pathogens. Moreover, this work has implications in the development of novel therapeutic strategies to manipulate Th cell function during infectious diseases or immunopathologies.

Mammalian reoviruses are nonenveloped, double-stranded RNA viruses that serve as important models for studies of viral neuropathogenesis. Reovirus disease is initiated by binding of the virus to receptors on the surface of target cells. We identified junctional adhesion molecule-A (JAM-A), a tight junction protein, as a receptor for reovirus. Using a structure-function analysis we identified sequences in virus and receptor required for productive infection. Reovirus engagement of JAM-A in polarized epithelial cells during the entry phase induces a redistribution of JAM-A-associated proteins and increases paracellular permeability. The discovery that reovirus-JAM-A interactions alter tight junction integrity expands our knowledge of the physiologic effects of virus-receptor interactions and highlights a potential role for the tight junction in regulating the cellular response to viral infection.

NON-REDUNDANT FUNCTIONS OF THE MOUSE SSRP1 AND PFDN1 GENES REVEALED BY GENE TARGETING AND GENE ENTRAPMENT SHANG CAO Dissertation under the direction of Professor Henry Earl Ruley In the post-genome era, gene targeting and gene entrapment provide efficient strategies to analyze gene functions in mice. Conventional gene targeting ablates individual genes by homologous recombination while gene entrapment generates large numbers of disrupted genes from which specific mutations are selected for further studies. Mouse Ssrp1 and Pfdn1 gene were disrupted in murine embryonic stem cells, by conventional gene targeting and gene entrapment, respectively, and were subsequently introduced into the germ line. The Ssrp1 gene encodes a high-mobility-group (HMG) protein SSRP1, a member of a conserved chromatin-remodeling complex (FACT/DUF/CP) implicated in DNA replication, transcription and repair. Ssrp1 homozygous mutant embryos die soon after implantation, and pre-implantation blastocysts are defective for cell outgrowth and/or survival in vitro. Further loss of p53 does not influence growth and/or survival defects caused by loss of Ssrp1 function. Thus, Ssrp1 appears to encode non-redundant and p53-independent functions that are essential for cell viability. The Pfdn1 gene, coding for the first subunit of molecular chaperone prefoldin, was ablated by insertion of a gene trap retrovirus (GTR1.3). Null mutant mice are born at nearly Mendelian ratios, but most die before 5 weeks of age, after a period of stunted growth, physical wasting and neuromuscular abnormalities. Postnatal phenotypes of homozygous mutant mice include: (1) defective commissures in the central nervous system (CNS);(2) hematopoietic defects consisting of micronucleated erythrocytes and neutropils with abnormal nuclear segmentation; and (3) lymphopoietic defects include loss of B cell precursors and altered ratios of T cell subpopulations. Fibroblasts from mutant embryos (MEFs) are viable but display abnormal cytoskeletal organization, suggesting that Pfdn1 encodes functions required for cytoskeletal assembly. Thus, Pfdn1appears to encode non-redundant functions required for mouse development and survival, presumably by affecting cytoskeleton formation.

As inflammation is a major mechanism of disease, we investigated the signal transduction processes induced by the key inflammatory cytokine Interleukin (IL) 1 beta as well as the genome responses to pathogen-derived proinflammatory agonists. Myeloid differentiation primary response gene (MyD88) is the essential adaptor protein that transduces intracellular signals generated by the IL-1â receptor and multiple Toll-like receptors (TLRs) that recognize diverse pathogen surfaces. The IL-1â receptor complex interacts with MyD88 via the Toll/IL-1 receptor (TIR) domain. Here we identified the MyD88 TIR domain binding sites involved in IL-1â-induced protein-protein interactions. The MyD88 TIR domain required Box3 to act as dominant negative inhibitor of IL-1â signaling. Accordingly, mutations of residues 285-286 reversed the dominant negative effect of the MyD88 TIR domain on NFêB-dependent reporter gene activity and IL-6 production. Moreover, mutations of residues 171 in helix áA, 195-197 in Box2, and 275 in the âE strand had similar functional effects. Strikingly, mutations of residues 195-197 eliminated the TIR-TIR interaction of MyD88 and IL-1 receptor accessory protein (IL1RAcP). Mutations in Box2 and 3 prevented homotypic MyD88 oligomerization via the TIR domain. Overall, structure-function analysis produced a 3-dimensional docking model of TIR-TIR interaction between MyD88 and IL1RAcP. Animal models of systemic inflammation induced by staphylococcal enterotoxin B (SEB) and lipopolysaccharide (LPS) were used to study genome-wide transcriptional response. In vivo treatment with SEB induced 134 and 209 genes in spleen cells and T lymphocytes, respectively. Upregulation of these genes was inhibited by blocking NFêB signaling with a cell-penetrating nuclear import inhibitor cSN50 peptide or the IêBáÄN transgene. In vivo treatment of LPS induced upregulation of 1296 genes and downregulation of 1551 genes in the liver and correspondingly, 1109 and 402 genes in the spleen. The genome-wide response to LPS was ablated in TLR-4-deficent C3H/HeJ mice. The cSN50 peptide blocked 547 LPS-inducible and 669 LPS-downregulated genes in the liver, and 105 LPS-inducible and 230 LPS-suppressed genes in the spleen. Thus, nuclear import of NFêB and other stress-responsive transcription factors plays an important role in genome-wide response to microbial inducers of inflammation.

Autoimmune and atopic diseases are immune-mediated multigenic diseases. The work of this thesis tests the hypothesis of performing gene expression profiling using peripheral blood mononuclear cells as a common source to characterize distinct human diseases. Our studies demonstrate that distinct gene expression profiles exist in both human autoimmune and atopic diseases. The unique gene expression profile in autoimmune disease is composed of two parts; one is genetic, second results from disease onset. A unique gene expression profile is also present in atopic disease and a portion of this profile may be used to monitor responses to therapy. In addition, we demonstrate it is possible to use gene expression profiling as a prognostic factor. In parallel, a specific gene expression profile exists in T cells in a human type I diabetes murine model-NOD. However, the NOD profile is not shared with the common autoimmune disease profile, but predicts a common liability, lymphopenia, which may stimulate autoimmunity. In summary, the studies of this thesis demonstrate that the gene expression profiling method is a powerful tool to depict the molecular portraits of immune-mediated diseases.

Antigen receptor gene assembly is governed by transcriptional promoters and enhancers that communicate over large distances and modulate chromatin accessibility to V(D)J recombinase. The precise role of these cis-acting elements in opening chromatin at recombinase targets and the mechanisms underlying their collaboration remain unclear. I show that the TCR beta enhancer (Ebeta) directs long-range chromatin opening over both DbetaJbeta clusters. Strikingly, chromatin associated with the Dbeta1 gene segment is refractory to Ebeta-mediated opening. Accessibility at Dbeta1 is accompanied by the formation of a stable holocomplex between a Dbeta-proximal promoter, PDbeta1, and Ebeta. These findings indicate a stepwise process for DbetaJbeta recombination that relies on distinct aspects of Ebeta activity: an intrinsic function that directs general chromatin opening and a cooperative function with PDbeta1 that facilitates the unmasking of the Dbeta1 gene segment, triggering TCR beta gene assembly.

The immature B cell compartment in the spleen comprises of two subsets termed Transitional Type 1 (T1) and Transitional Type 2 (T2). Our laboratory has shown that T1 and T2 B cell subsets respond differently to crosslinking of the BCR; T1 cells die upon BCR stimulation whereas T2 cells proliferate and differentiate into a mature B cell phenotype. The signaling mechanisms that control these disparate biological outcomes in T1 versus T2 cells remain elusive. I hypothesize that the composition of the BCR signalosome is altered during T1 to T2 B cell development. This in turn activates distinct signaling pathways in response to BCR crosslinking. I performed cDNA microarrays and proteomic approaches to define the global genetic and biochemical signaling programs that control the differential BCR responses of T1, T2 and mature B cells. Through proteomics I identified the serine/threonine phosphatase PP2A, CD45 associated protein (CD45-AP) and a never before identified protein in B cells, Src Kinase Associated Phosphoprotein (SKAP55) specifically in T1 but not T2 or mature B cells. Through microarray analysis we found that pro-apoptotic genes did not show any significant difference among T1, T2 and mature B cells, but T1 cells did have lower levels of two anti-apoptotic genes, A1 and Bcl-2. This suggests that the effect in B cell survival of T1 cells may be due to lower expression of anti-apoptotic genes. In addition I found that T1 B cells had lower levels of NF-kappaB1 and NF-kappaB2. Thus, global genomics and proteomics are feasible approaches to help define the unique characteristics of the transitional B cell subsets that underlie their differentiation.

MICROBIOLOGY AND IMMUNOLOGY CHARACTERIZATION OF AN AUTOIMMUNE GENE EXPRESSION SIGNATURE KEVIN MAAS Dissertation under the direction of Professor Thomas Aune Autoimmune diseases affect approximately 5% of the human population and are believed to arise from immune mediated attack against self-antigen. Microarray studies demonstrate highly reproducible gene expression signatures in peripheral blood mononuclear cells (PBMC) of patients with a range of autoimmune disorders. Neither in vivo recapitulation of an immune response to foreign antigen or disease duration accounted for the autoimmune gene expression signature. However, unaffected first-degree relatives also contained this gene expression signature, suggesting that it arose from a heritable trait(s). The autoimmune signature includes, at least in part, under-expressed genes that encode proteins that inhibit cell cycle progression and stimulate apoptosis. We found that reduced TP53 gene expression and p53 protein levels are associated with selective loss of lymphocyte radiation-induced apoptosis in patients with rheumatoid arthritis. We hypothesize that these liabilities may contribute to autoimmunity. Approved______________________________________ Date_________

Self-reactive B lymphocytes are frequently produced as a consequence of B cell antigen receptor rearrangement. Autoreactive B cells that are not eliminated or inactivated by tolerance mechanisms survive and mature in the periphery. In the spleen, the marginal zone serves as a reservoir for autoreactive B lymphocytes. Marginal zone B cells are known for their rapid and robust responses to T-independent stimuli and serve functions in both the innate and adaptive arms of the immune system. Anti-insulin transgenic B cells are preferentially selected into the marginal zone and are functionally anergic. These cells provide an opportunity to study how autoreactive B cells mature into the marginal zone subset. Using the anti-insulin transgenic model, we find that multiple factors influence marginal zone B cell maturation. These elements include B cell receptor specificity, lineage regulators such as Notch2, and a differentially expressed transcriptional profile. Understanding the processes that regulate marginal zone B cell maturation and how anergy is maintained in this population will impact our ability to manage these cells in host defense and autoimmune disease.

Protein synthesis is a complex and poorly understood process that requires the concerted activity of ribosomes and a plethora of trans-acting protein factors. This dissertation used proteomic, biochemical, and functional analysis to conclude that the Saccharomyces cerevisiae ribosome contains the protein Asc1p. One of Asc1ps functions is to represses the synthesis of a subset of proteins in yeast. At elevated temperatures Asc1p is required for ribosome biogenesis. This dissertation also identified trans-acting proteins that activate the cap-independent translation of the ornithine decarboxylase mRNA. One of these proteins, ZNF9, is mutated in type-2 myotonic dystrophy. Future studies aimed at discovering the regulatory targets of ZNF9 may lead to an improved understanding of this diseases pathogenesis.