The thymus in mice and humans originates from the third pharyngeal pouch endoderm. This process is divided into early Foxn1-independent stages and later Foxn1-dependent stages. Foxn1 is indispensible for the differentiation of thymic epithelial progenitor cells (TEPCs) as the development of thymus in Foxn1 mutant mice is arrested around E12.5. The transcriptional changes associated with the developmental of the thymus are poorly understood. In particular, the transcriptional regulation of Foxn1 in the developing thymic rudiment has not been definitively identified. Recently, Pax1, Pax9, Tbx1, and E2Fs have been implicated in transcriptional regulation of Foxn1. However, with the exception of E2Fs, evidence regarding their direct involvement in regulating Foxn1 expression is missing. Therefore, the aims of this thesis were to study the transcriptional regulation of Foxn1 through identification of its regulatory regions and studying the transcriptional changes associated with the developing thymus. These aims were addressed through the use of chromatin-immunoprecipitation technique combined with next-generation sequencing and gene expression analyses of the developing TEPCs. The data presented in this thesis identified H3K4me3 and H3K27ac marked Foxn1 promoter and five H3K4me1 and H3K27ac marked putative enhancer regions. The combination of gene expression analyses and transcription factor binding sites within the above regions suggested Ets1, Isl1, Foxc1, Nfia, Nfib, Srf, Foxo1, Nfatc2, Ing4, Foxa2, Hes1, E2Fs, and p53 as candidate transcriptional regulators of Foxn1. Nfatc2 appears also to be a target of Foxn1 that could play an important role in thymus development by regulating a large set of genes. Comparison of wild type and Foxn1 null thymus showed that Foxn1 could act as positive regulator of Pax1 and negative regulator of Gata3 and Eya1, genes important for third pharyngeal pouch development. The comparison of transcriptome of E10.5 and E11.5 third pharyngeal pouch cells and E12.5 TEPCs showed that genes involved in tissue development are downregulated while those involved in antigen presentation, a process important for thymus function, are upregulated during development. These results also demonstrated a decrease in the activity of transcription factor network involving Hox genes and an increase in the activity of a network involving Nfkb, Rela, and Irf genes. Analysis of signalling pathways suggested that the NFκB signalling pathway could be important for thymus development after E12.5 while TGFβ signalling pathway appeared to be detrimental to Foxn1 expression in thymic epithelial cells. Together, I identified several transcription factors that could be involved in transcriptional regulation of Foxn1 in TEPCs, several genes that could be a target of FOXN1, changes in transcription factor network and signalling pathways associated with the developing thymic rudiment.

The overall goal of this research is to understand roles of gut microbiota metabolites and adipocyte transcription factor (TF) network in health and disease by developing systematic analysis methods. As microbiota can perform diverse biotransformation reactions, the spectrum of metabolites present in the gastrointestinal (GI) tract is extremely complex but only a handful of bioactive microbiota metabolites have been identified. We developed a metabolomics workflow that integrates in silico discovery with targeted mass spectrometry. A computational pathway analysis where microbiota metabolisms are modeled as a single metabolic network is utilized to predict a focused set of targets for multiple reaction monitoring (MRM) analysis. We validated our methodology by predicting, quantifying in murine cecum and feces and characterizing tryptophan (TRP)-derived metabolites as ligands for the aryl hydrocarbon receptor. The adipocyte process of lipid droplet accumulation and differentiation is regulated by multiple TFs that function together in a network. Although individual TF activation is previously reported, construction of an integrated network has been limited due to different measurement conditions. We developed an integrated network model of key TFs - PPAR, C/EBP, CREB, NFAT, FoxO1, and SREBP-1c - underlying adipocyte differentiation. A hypothetic model was determined based on literature, and stochastic simulation algorithm (SSA) was applied to simulate TF dynamics. TF activation profiles at different stages of differentiation were measured using 3T3-L1 reporter cell lines where binding of a TF to its DNA binding element drives expression of the Gaussia luciferase gene. Reaction trajectories calculated by SSA showed good agreement with experimental measurement. The TF model was further validated by perturbing dynamics of CREB using forskolin, and comparing the predicted response with experimental data. We studied the molecular recognition mechanism underlying anti-inflammatory function of a bacterial metabolite, indole in DC2.4 cells. The indole treatment attenuated the fraction of cells that were producing the pro-inflammatory cytokine, TNFα and knockdown of nuclear receptor related 1 (Nurr1; NR4A2) resulted in less indole-derived suppression of TNFα production. The first discovery of NR4A2 as a molecular mediator of the endogenous metabolite, indole is expected to provide a new strategy for treatment of inflammatory disorders.

<p>Sensory neuron diversity is a common theme in the animal kingdom. It provides the cellular infrastructure that supports the accurate perception of the external world. Among all sensory systems, the olfactory system demonstrates an extreme in the extraordinarily diversified neuronal classes it holds. The system-wide cellular diversity is in sharp contrast with the individual specialization of olfactory receptor neurons (ORNs) per se. How the nervous system, particularly the olfactory system, uses limited genetic information to generate a huge variety of neurons with distinct properties remains elusive. </p><p>The adult Drosophila olfactory system is an excellent model to address this question due to its conserved organizational principles and reduced complexity. The fly olfactory appendages contain 50 ORN classes, each of which expresses a single receptor gene from a family of ~80 genes. Stereotyped clusters of 1-4 ORN classes define about 20 sensilla subtypes, belonging to 3 major morphological types. All cellular components within a sensillum are born by a single sensory organ precursor (SOP) via asymmetric divisions. The molecular mechanisms that determine SOP differentiation potentials to develop into distinct sensilla subtypes and the associated ORN classes are unknown.</p><p>From a genetic screen, we identified two mutant alleles in the rotund (rn) gene locus, which has a critical function in diversifying ORN classes. Rn is required in a subset of SOPs to confer novel sensilla subtype differentiation potentials from otherwise default ones within each sensilla type lineage. In rn mutants, ORNs in rn-positive sensilla subtypes are converted to lineage-specific default rn-negative fates, resulting in only half of the normal ORN diversity. This work is described in Chapter 2.</p><p>Based on an unbiased time-course transcriptome analysis, we discovered two critical downstream targets of Rn, Bric-à-brac (Bab) and Bar. In light of the knowledge about leg development, we found these genes, along with Apterous (Ap) and Dachshund (Dac), are part of the conserved proximal-distal (PD) gene network that play a crucial role in patterning the antennal precursor field prior to proneural gene-mediated SOP selection. Interactions between these PD genes under the influence of morphogen gradients separate the developing antennal disc into 7 concentric domains. Each ring is represented by a unique combination of the aforementioned transcription factors, coding the differentiation potentials for a limited number of sensilla subtypes. Genetic perturbations of the network lead to predictable changes in the ratios of different sensilla subtypes and corresponding ORN classes. In addition, using CRISPR/Cas9 technology, we were able to add tags to specific rn isoforms in the endogenous locus, and show positive regulation of Bab and negative regulation of Bar by the direct binding of Rn to the promoters in vivo. This work is presented in Chapter 3.</p><p>We proposed a three-step mechanism to explain ORN diversification, starting from pre-patterning of the precursor field by PD genes, followed by SOP selection by proneural genes, and ended with Notch-mediated neurogenesis. The final outcomes are greatly determined by the pre-patterning phase, which may be modified during evolution to compensate special olfactory needs by individual species. In our model, each step serves a single purpose, which displays context-dependent functions. By changing contexts, reassembly of the same logical steps may guide neuronal diversification in parallel systems with completely different identities. This step-wise mechanism seems to be a common strategy that is used by many other systems to generate neuronal diversity.</p> / Dissertation