Identification of a conserved alternative mRNA splicing program that supports hepatic growth and maturation during development and regeneration

Alternative splicing;Ribonucleic acid (RNA);Epithelial Splicing Regulatory Protein 2 (ESRP2);Liver development;Liver maturation

Bhate, Amruta

关键词: Alternative splicing;    Ribonucleic acid (RNA);    Epithelial Splicing Regulatory Protein 2 (ESRP2);    Liver development;    Liver maturation;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/99467/BHATE-DISSERTATION-2017.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Analysis of metazoan genomes has led to the extraordinary observation that mammals contain only ~25000 protein-coding genes. Therefore, post-transcriptional gene regulatory mechanisms (PTGRMs) are thought to be crucial for generating the diverse proteome required to support their high cellular complexity. Alternative splicing (AS), the most common PTGRM, is a process by which the exons of a pre-mRNA are spliced into different arrangements to produce structurally and functionally distinct mature mRNAs thereby contributing to the transcriptome complexity. Regulated AS events are known to play a determinative role in the brain, heart and skeletal muscle development; however, regulation of splicing or its function in hepatic growth and maturation is not well understood.The mammalian liver is a major metabolic organ responsible for a variety of functions and has an exceptional capacity for regeneration. Liver undergoes dramatic transitions with regards to structure and function during postnatal development and during regeneration implying that intricate regulatory mechanisms must control these processes.Signaling and transcriptional networks that regulate postnatal liver development are extensively studied, but the role of post-transcriptional mechanisms is poorly explored. My goal thus was to identify conserved AS networks in postnatal liver development and establish relationships between these splicing changes and their putative regulators, RNA Binding proteins, in normal liver maturation and regeneration.In the second chapter of the thesis, I studied the changes occurring in the liver transcriptome during postnatal periods of development.To characterize the conserved AS program during mammalian liver maturation, high throughput RNA-seq of mouse livers at E18 (embryonic day 18), P14 (postnatal day 14), P28 and P90 was performed. We observed ~5000 genes change in mRNA abundance, ~530 genes change in AS and ~200 genes change in alternative polyadenylation with minimal overlap among these categories. This indicates that postnatal liver maturation is accomplished by three separate modes of regulatory mechanisms. Analysis of AS at intervening timepoints (E16, E18, P0, P2, P7, P14, P28 and P90) showed that postnatal shift in AS is temporally coordinated and subsets of AS events follow distinct patterns of splicing change grouped as early (E16-P2), late (P14-P90) or biphasic (E18-P7 and P7-P90). As the liver is actively going through the process of maturation, we wanted to understand whether the changes occurring in the transcriptome are a result of the maturing parenchyma or due to a cell population change occurring in the liver. Comparison of AS in purified hepatocytes (Hep), major cell type of the liver, and the other cell types, non-parenchymal cells (NPCs) showed that a majority of AS are cell-type specific with most being Hep-specific. These AS transitions are evolutionarily conserved in mice and humans.RNA binding proteins (RBPs) play pivotal roles in regulating splicing transitions by binding near the variably used splice sites and modulating their accessibility to the spliceosome. Thus, using our RNA-sequencing data, we analyzed the expression of RBPs and found that the majority are down regulated at both mRNA and protein levels during postnatal liver maturation. The expression of these RBPs also follows distinct temporal patterns overlapping with subsets of AS transitions. The remarkable exception is ESRP2 (Epithelial Splicing Regulatory Protein 2), which is strongly upregulated postnatally. Mammalian ESRP2 belongs to a family of tissue-specific AS regulators, which are essential regulators that drive mesenchymal to epithelial transition by coordinating splicing of genes involved in cell-cell adhesion, cytoskeleton rearrangement, and intracellular signaling.The third chapter describes the studies done to establish that ESRP2 is a key regulator of postnatal splicing in maturing hepatocytes. ESRP2 was the only factor that exhibited significant up regulation in mRNA and protein levels during the first four weeks after birth. Upon assaying its relative mRNA and protein levels in purified Hep and NPC fractions from P0 and adult mouse livers, I found that ESRP2 levels are specifically higher in adult Hep. To ascertain whether ESRP2 plays a crucial role in fetal-to-adult shift of AS in liver, I obtained liver tissue from ESRP2 knock-out mice (ERSP2 KO) generated using full gene deletion and performed splicing analysis of 141 developmentally regulated AS events. I found that 31 AS events were strongly misspliced and showed a failure of fetal-to-adult splicing switch indicating that ESRP2 regulates these events during normal development. To investigate whether ESRP2 expression is sufficient for the fetal-to-adult switch of the AS transitions, I over expressed ESRP2 protein in liver cell lines HepG2 and AML12, using an adenoviral vector. These cell lines have minimal expression of ESRP2 and exhibit a neonatal splicing pattern. Forced expression of ESRP2 causes a fetal-to-adult splicing switch in the majority of these 31 targets, reinforcing the role of ESRP2. When evaluated for cell-type specificity, I found that 87% of ESRP2 targets were Hep-specific events. ESRP2 KO livers also showed a loss of mature hepatocyte markers like Cyp2b10, Alb and Chd1 and a persistent expression of fetal genes like Meg3 and Chd2. Histological analysis showed that ESRP2 KOs have a normal hepato-somatic index, but have smaller cells with increased number of mono- and bi-nucleated hepatocytes. Based on these results, I conclude that ESRP2 is a key regulator of post-natal splicing program in maturing hepatocytes.In chapter four, I describe the details of the project I initiated to characterize the AS program during liver regeneration. I used DDC (3, 5-diethoxycarbonyl-1, 4-dihydrocollidine), a hepatotoxin, to induce liver regeneration in mice. This established model of liver regeneration closely mimics the normal route through which liver regenerates in response to toxins. To characterize the AS program during liver regeneration, we performed RNA-seq of mouse livers which were treated with DDC for four weeks and mice which were on normal chow diet. The RNA-Seq data showed that ~3000 genes change in mRNA abundance and ~230 events change in AS upon toxin injury with minimal overlap. As the regenerating liver, has actively proliferating hepatocytes, it is plausible to have similar characteristics of an immature liver. Therefore, upon comparing the data set of developmentally regulated AS events with the list of genes that are differentially spliced in regeneration we find that there is a large subset of AS events common to both. These common AS transitions show splicing in reciprocal direction for development and regeneration, indicating that during regeneration these AS events exhibit a neonatal splicing pattern. To gain mechanistic information of the regulatory networks governing liver regeneration, it is necessary to look at the roles of RBPs. Our analysis has shown that many RBPs that are strongly down regulated during normal liver development are upregulated in regenerating livers. These RBPs exhibit similar expression in Hep isolated from livers treated with DDC indicating that these factors may play a role in governing the AS landscape during regeneration. Interestingly, I observed that ESRP2 mRNA and protein levels were downregulated in livers and hepatocytes treated with DDC for four weeks. Importantly, the expression levels of ESRP2 are restored in livers of animals that have been allowed to recover from the DDC treatment. Further studies should delineate the biological role of ESRP2 in controlling AS networks during liver regeneration.In chapter five, I lay out the future directions to take this research forward. One observation we had while performing RNA-Seq for four week DDC treated livers, was that these livers had massive inflammation and bile deposits due to the toxin-mediated injury. Therefore, it is probable that this RNA-Seq did not capture purely regeneration-based transcriptome changes, but also due to cell population change. One way to rectify this is to perform RNA-Seq on Hep isolated from DDC treated livers to avoid the background of the inflammatory cells. Another important area of research to develop is carefully characterizing the role of ESRP2 during mammalian liver regeneration. This is being undertaken by challenging ESRP2 KO and over expression mouse models with DDC to determine how gain and loss of ESRP2 expression affects the process of liver regeneration. In conclusion, I have identified a conserved, cell-type specific and temporally coordinated AS program operative postnatal liver development and demonstrated a direct role for ESRP2 in supporting a splicing regulatory network to support postnatal hepatocyte maturation.