Position weight matrices (PWMs) have become a tool of choice for the identification of transcription factor binding sites in DNA sequences. PWMs are compiled from experimentally verified and aligned binding sequences. PWMs are then used to computationally discover novel putative binding sites for a given protein. DNA-binding proteins often show degeneracy in their binding requirement, the overall binding specificity of many proteins is unknown and remains an active area of research. Although PWMs are more reliable predictors than consensus string matching, they generally result in a high number of false positive hits. A previous study introduced a novel method to PWM training based on the known motifs to sample additional putative binding sites from a proximal promoter area. The core idea was further developed, implemented and tested in this thesis with a large scale application. Improved mono- and dinucleotide PWMs were computed for Drosophila melanogaster. The Matthews correlation coefficient was used as an optimization criterion in the PWM refinement algorithm. New PWMs keep an account of non-uniform background nucleotide distributions on the promoters and consider a larger number of new binding sites during the refinement steps. The optimization included the PWM motif length, the position on the promoter, the threshold value and the binding site location. The obtained predictions were compared for mono- and dinucleotide PWM versions with initial matrices and with conventional tools. The optimized PWMs predicted new binding sites with better accuracy than conventional PWMs.

Embryonic stem cells (ESCs) are characterized by their unlimited capacity for self-renewal and the ability to contribute to every lineage of the developing embryo. The promoters of developmentally regulated loci within these cells are marked by coincident epigenetic modifications of gene activation and repression, termed bivalent domains. Trithorax group (TrxG) and Polycomb Group (PcG) proteins respectively place these epigenetic marks on chromatin and extensively colocalize with Oct4 in ESCs. Although it appears that these cells are poised and ready for differentiation, the switch that permits this transition is critically held in check. The derepression of bivalent domains upon knockdown of Oct4 or PcG underscores their respective roles in maintaining the pluripotent state through epigenetic regulation of chromatin structure. The mechanisms that facilitate the recruitment and retention of Oct4, TrxG, and PcG proteins at developmentally regulated loci to maintain the pluripotent state, however, remain unknown. Oct4 may function as either a transcriptional activator or repressor. Prevailing thought holds that both of these activities are required to maintain the pluripotent state through activation of genes implicated in pluripotency and cell-cycle control with concomitant repression of genes required for differentiation and lineage-specific differentiation. More recent evidence however, suggests that the activator function of Oct4 may play a more critical role in maintaining the pluripotent state (Hammachi et al., 2012). The purpose of the studies described in this dissertation was to clarify the underlying mechanisms by which Oct4 functions in transcriptional activation and repression. By so doing, we wished to contextualize its role in pluripotent cells, and to provide insight into how changes in Oct4 function might account for its ability to facilitate cell fate transitions. As a result of our studies we find that Oct4 function is dependent upon post-translational modifications (PTMs). We find through a combination of experimental approaches, including genome-wide microarray analysis, bioinformatics, chromatin immunoprecipitation, functional molecular, and biochemical analyses, that in the pluripotent state Oct4, Akt, and Hmgb2 participate in a regulatory feedback loop. Akt-mediated phosphorylation of Oct4 facilitates interaction with PcG recruiter Hmgb2. Consequently, Hmgb2 functions as a context dependent modulator of Akt and Oct4 function, promoting transcriptional poise at Oct4 bound loci. Sumoylation of Oct4 is then required to maintain Hmgb2 enrichment at repressed loci and to transmit the H3K27me3 mark in daughter progeny. The expression of Oct4 phosphorylation mutants however, leads to Akt inactivation and initiates the DNA Damage Checkpoint response. Our results suggest that this may subsequently facilitate chromatin reorganization and cell fate transitions. In summary, our results suggest that controlled modulation of Oct4, Akt, and Hmgb2 function is required to maintain pluripotency and for the faithful induction of transcriptional programs required for lineage specific differentiation.

Little is known about higher order chromatin structures in the human brain and their function in transcription regulation. We employed chromosome conformation capture (3C) to analyze chromatin architecture within 700 Kb surrounding the transcription start site (TSS) of the NMDA receptor and schizophrenia susceptibility gene, GRIN2B, in human and mouse cerebral cortex. Remarkably, both species showed a higher interaction between the TSS and an intronic sequence, enriched for (KRAB) Krueppel associated Box domain binding sites and selectively targeted by the (H3K9) histone 3 lysine 9 specific methyltransferase ESET/SETDB1. Transgenic mice brain cortical nuclei over-expressing Setdb1 showed increased heterochromatin-protein 1 signal at the interacting regions coupled with decreased Grin2b expression. 3C further revealed three long distant chromatin loop interactions enriched with functional enhancer specific (H3K27Ac) histone 3 lysine 27 acetylation signal in GRIN2B expressing tissue (human cortical nuclei and Human Embryonic Kidney - HEK cells). Doxycycline-induced SETDB1 over-expression decreased 2 out of 3 loop interaction frequencies suggesting a possible SETDB1-mediated transcription repression. We also report a specific looping interaction between a region 50Kb upstream of the (GAD1) Glutamic Acid Decarboxylase – 1 gene TSS and the GAD1 TSS in human brain nuclei. GAD1 catalyzes the rate limiting step in (GABA) gamma amino-butyric acid synthesis and is quintessential for inhibitory signaling in the human brain. Clinical studies in schizophrenia brain samples reveal a decreased looping interaction frequency in correspondence with a decrease in gene expression. Our findings provide evidence for the existence of transcription relevant higher order chromatin structures in human brain.