S-Space College of Natural Sciences (자연과학대학) Program in Genetic Engineering (협동과정-유전공학전공) Theses (Ph.D. / Sc.D._협동과정-유전공학전공)
Study on the function of Twist2 in the regulation of E protein target gene expression during thymocyte selection
흉선 세포 선별과정에서 E 단백질의 타겟 유전자 발현을 억제하는 Twist2 의 기능에 관한 연구
- 자연과학대학 협동과정 유전공학전공
- Issue Date
- 서울대학교 대학원
- 학위논문 (박사)-- 서울대학교 대학원 : 유전공학전공, 2016. 8. 성노현.
- Cellular differentiation is the process where a cell changes its state from one to another. The development of a multicellular organism involves a series of differentiation processes which result in a complex system of tissues and cell types. Differentiation is guided by the de novo binding of transcription factors to their cognate genomic sites. With the advent of technologies for investigating the genome-wide occupancy of transcription factors, a wealth of insights has emerged. Currently, ChIP-seq offers the most powerful means to locate genomic regions where critical regulatory interactions take place in a high resolution. Twist2, a bHLH transcription factor, is highly induced during thymocyte selection in response to positively selecting TCR signals. Despite the importance of the positive selection-specific regulation of Twist2 expression for the proper differentiation of T lymphocytes, our understanding has been limited. In an effort to uncover the role of Twist2 during thymocyte selection, I intensively investigated the identities of the genes regulated by Twist2, and here I report the first genome-wide information about Twist2 occupancy in mouse thymocytes.
By performing ChIP-seq with thymocytes from the Twist2-transgenic mice, a total of 3,822 highly confident Twist2-binding sites were identified. The gene-proximal distribution of Twist2-binding sites was revealed. In addition, strong evolutionary conservation over Twist2-binding sites was exhibited, suggesting that these regions could be under selective pressure and therefore might be functionally important. By using de novo motif-discovery algorithm, the canonical E-box sequence (5-CABMTGB-3) and two additional consensus sequences for the binding of ETS (5-MGGAAR-3) family and Runx (5-TGTGGTT-3) family transcription factors were discovered. E-box is the consensus DNA sequence to which bHLH transcription factors bind. In general, bHLH proteins dimerize preferentially with other members of the bHLH family to form hetero dimers. Before thymocyte selection, E proteins (HEB and E2A) are highly expressed in DP thymocytes and regulate the expression of critical genes by binding to E-box sequences located in the transcriptional regulatory regions.
I hypothesized that Twist2 may bind to the E-box sites pre-occupied by E proteins after being induced in post-DP thymocytes. To validate this tentative hypothesis, a recent study performed by Vanhille et al., which identified HEB-binding sites in DP thymocytes, was exploited and binding profiles of HEB and Twist2 were compared to count the number of regions shared by HEB and Twist2. Strikingly, 51% (842) of HEB-binding sites were overlapped with Twist2-binding sites. These overlapping genomic sites correspond to 22% (847/3,822) of Twist2-binding sites.
To identify putative genes directly regulated by Twist2 and/or E proteins in DP thymocytes, the occupancy peaks were linked to their adjacent genes. As results, 1,051 and 430 genes were associated with Twist2 and HEB, respectively. Among them, 239 genes were associated with the composite sites, binding of both Twist2 and HEB. To evaluate how the Twist2 occupancy correlates with the differential gene expression in DP thymocytes before (low level expression of Twist2) and immediately after (high level expression of Twist2) thymocyte selection, I analyzed the global gene expression profiles of pre-DP and post-DP thymocytes using microarray. The functional consequences of Twist2 binding to the composite sites during thymocyte selection were then estimated by Gene Set Enrichment Analysis (GSEA). As expected, genes associated with the composite sites were highly enriched among the group of genes down regulated upon thymocyte selection. Furthermore, genes associated with the composite sites were also highly enriched among the group of genes down regulated upon E protein deletion. This indicated that E-protein activity is indeed crucial for the proper expression of genes associated with the composite sites in pre-DP thymocytes. These analyses strongly supported that Twist2 might negatively regulate the E protein target genes during thymocyte selection by binding to the genomic regions preoccupied by E proteins.
By using DO11.10 TCR-transgenic mouse system, effects of Twist2 knock out on the expression of 23 genes were analyzed. In the absence of Twist2, genes that are known to be down regulated in response to thymocyte selection were up regulated. For example, Rorc and Cldn4, whose expression dropped dramatically upon receipt of positive selection signal, were up regulated in Twist2f/fCD4Cre+ DO11.10 cells. In contrast, Poll and Trim26, whose expressions were highly induced in positively selected DP cells, were down regulated in Twist2f/fCD4Cre+ DO11.10 cells.
Here, I suggest a novel mechanism of the transcriptional repression of E protein-mediated pre-selection programs in DP thymocytes. After being highly induced in response to positively selecting TCR signals, Twist2, forms heterodimer with E proteins and thereby negatively regulates the expression of E protein target genes, resulting transition to post-selection stage.