Transkripsiyon faktörü aktivitesi ve gen ifadesinde RNA ve protein etkenleri üzerine bir çalışma

Transcription factors can activate or repress expression of their target genes. Given this critical role, their activity is decisive in health and disease. Transcription factor function is affected by numerous factors that include co-activators, enhancer RNA, and protein oligomerization. Combining functional genomic and proteomic techniques with computational methods, the research presented here investigated how these factors influence transcription factors and modify gene regulation, and the role of a transcriptional repressor in cancer. Using the androgen receptor (AR) as a model transcription factor, we characterized how the enhancer long non-coding RNA KLK3e affects expression of AR regulated genes. Combining in vitro RNA pulldown and mass spectrometry, we identified proteins that interact with KLK3e. These included known AR regulators such as DDX5, DDX17 and HSPA8 that were known to affect prostate cancer pathogenesis. Our results and computational modeling of the RNA-protein complex suggest that KLK3e RNA may influence AR mediated transcription through acting as a scaffold to gather AR co-activators near the transcription site. Such RNA scaffolding effects may be paralleled in other nuclear receptors and transcription factors, as DDX5 has been implicated to interact with other steroid receptors such as the estrogen and vitamin D receptors. Next we explored how co-activator proteins influence gene transcription. We used four novel small molecule inhibitors that target different sites on the AR and disrupt co-activator binding. To identify the disrupted interactions, we conducted rapid immunoprecipitation mass spectrometry of endogenous proteins (RIME) and incorporated RNA sequencing to understand the effects on gene expression. Inhibition of either a co-activator binding site, dimerization site, DNA binding site, or ligand binding site appeared to have a similar negative impact on the number of protein interactors. Also, all of the inhibitors that target different sites affected the gene expression profile similarly and reduced transcription by AR. To understand how transcription factor oligomerization occurs, we conducted in silico modeling with the glucocorticoid receptor (GR). Experimental observations suggested that GR interacts with the mineralocorticoid receptor (MR) and can also form homo-tetramers upon DNA binding. Our computational models explained the variation in experimental observations and demonstrated that GR and MR could interact through several alternative interfaces shared by close relatives of the nuclear receptor family. Predictions demonstrated that ligand or antagonist binding did not prevent interaction but rather changed the interface preference. This can leave different sites available for binding of co-activators, explaining why different ligands can produce different cellular outcomes. We also proposed a mechanism of action for GR tetramerization through the ligand binding domain, and the chromosomal looping model of interaction through the DNA binding domains. Finally, circadian clock regulation also relies on transcriptional factors. Cryptochrome is a circadian transcriptional repressor, and its deletion was observed to delay cancer and extend the lifespan of p53 mutant mice. With RNA sequencing analyses we investigated the transcriptomic changes following UV damage response upon cryptochrome deletion in a p53 mutant background. Gene Set Enrichment Analysis of differentially expressed genes demonstrated enrichment in IFN-γ immune surveillance and TNFα signaling via NF-κB. Protein network analysis pinpointed p21, Sirt1 and Jun as key players. Differentially expressed genes also contained a high ratio of non-coding RNAs. In short, we show that the KLK3e enhancer RNA interacts with AR co-activators, and separately we observed that inhibition of co-activator binding can have the same effect on AR transcription factor activity as inhibition of ligand activation, DNA binding, or dimerization. Our structural models showed how GR might oligomerize with MR through alternative interfaces, and that ligand or antagonist binding can affect interface preference. We also demonstrated how cryptochrome deletion in p53 mutants enhance apoptotic and anti-tumorigenic responses to UV damage at the transcriptome level.