Genomic Analysis of microRNAs in Mouse Skin: Quantification, Biogenesis, Target Recognition and Regulatory Functions

MiRNAs are a class of small RNAs, approximately 20~22 nucleotides in length. These broadly conserved molecules represent a novel layer of gene regulation that fine-tunes the output of the transcriptome and contributes to the robustness of biological pathways. However, the quantitative picture of the miRNA pathway is unclear. Fundamental questions concerning the biogenesis pathway, target recognition and regulatory functions of individual miRNAs within mammalian tissues remain poorly understood. In this thesis, I employ mouse skin as a model system to examine the miRNA pathway with genomic approaches. I profiled the expression of miRNAs in mouse skin with a small RNA cloning protocol that has been widely used in the field. I observed a significant disagreement between the expression levels of individual miRNAs and the frequencies of corresponding clones in cDNA libraries. I established a system to benchmark the biases, optimized each step and achieved a highly efficient cDNA library construction strategy. Collectively, I have established a quantitative pipeline for accurately profiling miRNA expression levels with deep sequencing. I next investigated two important aspects of miRNA biogenesis, sorting of miRNAs to individual Argonaute proteins and the diversity of mature miRNA sequences. I demonstrated that mammalian miRNAs are randomly sorted to one of four Argonaute proteins and their distribution among Argonaute proteins is correlated with the abundance of each Argonaute protein. I also characterized many miRNA isoforms that potentially have important implications for the function of miRNAs. I then focused on the target recognition and regulatory function of miR-203, one of the most highly expressed miRNAs in the skin. With an inducible mouse model and microarray analysis, I identified many novel targets of miR-203. My results revealed that miR-203 recognizes a large number of targets with its seed sequences located in the 5’ of this miRNA. Furthermore, I showed that these miR-203 targets are highly enriched in the regulation of cell cycle and cell division as well as in the response to DNA damage. These results establish the widespread role for miR-203 in suppressing self-renewal of stem cells during epidermal differentiation. Finally, I studied the phenotypes of animals with genetic deletion of the miR-143/145 cluster. However, no discernible phenotypes were observed during normal skin development. This observation is consistent with many studies where individual miRNAs are shown to be not required for animal development