Genes are typically expressed in equal amounts from both parentally inherited chromosomes. However, recent studies have demonstrated that genes can be preferentially transcribed from a locus. Non-random preferential expression of alleles can occur in a parent-of-origin pattern, known as imprinting, where epigenetic factors regulate their transcription. Alternatively, it can occur in a haplotype-specific pattern, where cis-acting polymorphisms in regulatory regions are thought to underlie the phenomenon. Both forms of unequal allelic expression have been associated with human disease. Consequently, it is important to identify genes subject to unequal allelic expression and characterize mechanisms that regulate differential transcription. This thesis presents the results of a screen for unequal allelic expression where approximately 50 murine transcripts homologous to genes on human chromosome 7 were analyzed. Human chromosome 7 was selected due to its association with several human disorders that show parent-of-origin effects. The screen identified non-imprinted preferential allelic expression in numerous transcripts and demonstrated that such patterns can occur in tissue specific patterns. Paraoxonase-1 (Pon1), a gene implicated in arthrosclerosis, was identified as having a dynamic pattern of allelic expression which varies throughout embryonic development. This finding represents the first report of a developmentally regulated pattern of allelic variance. Carboxypeptidase-A4 (Cpa4) was identified as having a tissue-specific imprinted pattern of expression, where the maternal allele was preferentially expressed in all embryonic tissues, with the exception of the brain. The Krüppel-like factor 14 gene (Klf14), a novel imprinted transcript, was found to have ubiquitous maternal expression in all human and murine tissues analyzed. A differentially methylated region, generally associated with imprinted transcripts, was not found in the gene’s CpG island, nor was a differential pattern of histone modifications identified. However, it was determined that maternal methylation regulates the transcript. The data in this thesis contribute to our understanding of the numerous patterns of allelic expression that exist in nature and the diverse mechanisms that regulate them. Ultimately, quantitative analyses of allelic expression patterns and the identification of their underlying genomic DNA sequences will become standard protocol in all biomedical studies.

Allelic expression imbalance, or AEI, is the term given to differences in the expression levels of the two alleles of a gene. AEI has been previously identified in a number of species using various techniques. Here, the genome-wide extent of allelic expression imbalance in the pathogenic yeast species, Candida albicans, was examined through use of RNA sequencing in combination with a novel computational pipeline based around the diploid reference genome. Techniques for validating these results were investigated, and the difficulties surrounding specificity and quantification are discussed. As C. albicans is a highly heterozygous species, it was hypothesised that polymorphisms within alleles lead to differences in allele expression, which are further linked to differences in allele function. The functional consequences of AEI were therefore interrogated through investigation of Gene Ontology, identification of condition specific responses in AEI, and targeted construction and phenotypic screening of heterozygous knockout strains. Together, these results strongly suggest that divergence in allele expression is not linked to differences in allele function. Investigations of the possible control mechanisms behind the differences in allele expression were considered, with a focus upon structural factors such as chromosomal location, GC content, allele length and codon usage. However, issues with establishing causality are present, and difficulties lie in distinguishing between functional differences and consequences of bias in sequencing technologies. This piece of research has advanced the understanding of gene expression mechanisms within a medically important pathogen, paving the way for further investigations into the functional consequences of allelic expression imbalance in Candida albicans.

Genes are typically expressed in equal amounts from both parentally inherited chromosomes. However, recent studies have demonstrated that genes can be preferentially transcribed from a locus. Non-random preferential expression of alleles can occur in a parent-of-origin pattern, known as imprinting, where epigenetic factors regulate their transcription. Alternatively, it can occur in a haplotype-specific pattern, where cis-acting polymorphisms in regulatory regions are thought to underlie the phenomenon. Both forms of unequal allelic expression have been associated with human disease. Consequently, it is important to identify genes subject to unequal allelic expression and characterize mechanisms that regulate differential transcription. This thesis presents the results of a screen for unequal allelic expression where approximately 50 murine transcripts homologous to genes on human chromosome 7 were analyzed. Human chromosome 7 was selected due to its association with several human disorders that show parent-of-origin effects. The screen identified non-imprinted preferential allelic expression in numerous transcripts and demonstrated that such patterns can occur in tissue specific patterns. Paraoxonase-1 (Pon1), a gene implicated in arthrosclerosis, was identified as having a dynamic pattern of allelic expression which varies throughout embryonic development. This finding represents the first report of a developmentally regulated pattern of allelic variance. Carboxypeptidase-A4 (Cpa4) was identified as having a tissue-specific imprinted pattern of expression, where the maternal allele was preferentially expressed in all embryonic tissues, with the exception of the brain. The Krüppel-like factor 14 gene (Klf14), a novel imprinted transcript, was found to have ubiquitous maternal expression in all human and murine tissues analyzed. A differentially methylated region, generally associated with imprinted transcripts, was not found in the gene’s CpG island, nor was a differential pattern of histone modifications identified. However, it was determined that maternal methylation regulates the transcript. The data in this thesis contribute to our understanding of the numerous patterns of allelic expression that exist in nature and the diverse mechanisms that regulate them. Ultimately, quantitative analyses of allelic expression patterns and the identification of their underlying genomic DNA sequences will become standard protocol in all biomedical studies.

Polymorphisms in the gene encoding SORL1, involved in cellular trafficking of APP, have been implicated in late-onset Alzheimer's disease, by a mechanism thought to affect mRNA expression. To search for regulatory polymorphisms, we have measured allele-specific mRNA expression of SORL1 in human autopsy tissues from the prefrontal cortex of 26 Alzheimer's patients, and 51 controls, using two synonymous marker SNPs (rs3824968 in exon 34 (11 heterozygous AD subjects and 16 controls), and rs12364988 in exon 6 (8 heterozygous AD subjects)). Significant allelic expression imbalance (AEI), indicative of the presence of cis-acting regulatory factors, was detected in a single control subject, while allelic ratios were near unity for all other subjects. We genotyped 7 SNPs in two haplotype blocks that had previously been implicated in Alzheimer's disease. Since each of these SNPs was heterozygous in several subjects lacking AEI, this study fails to support a regulatory role for SORL1 polymorphisms in mRNA expression.

No description available.

We aimed to assess whether the breast-cancer intermediate-risk genes CHEK2, ATM ant TP53 were subject to differential allelic expression (DAE) in lymphoblastoid cell lines (LCLs) from high-risk breast cancer patients for whom no mutation in BRCA1 or BRCA2 had been identified.We implemented an assay based on high-resolution melting curve analysis (HRM) of single labeled fluorescent probes to detect allelic expression imbalance. The method relies on the distinction of the two alleles of an exonic marker SNP in heterozygous individuals with a fluorescent signal correlated to the relative abundance of each transcript. We developed an analysis tool for HRM data processing, specifically dedicated to DAE assessment. In our series, we found evidence for DAE for CHEK2, in carriers of the truncating mutation 1100delC. When combining mutation-screening data and assessment of DAE, we did not identify functional regulatory variant located in cis of the studied genes that would lead to DAE, in the transcriptional regulatory milieu of freely proliferating LCLs. Our results support that HRM is a method with high sensitivity and accuracy that can be used for DAE assessment. This approach can be applied to study breast and blood tissue samples. The latter would be of great interest for high-throughput mutation screening projects aiming to identify dysfunctional regulatory variants in candidate genes.

Le domaine Dlk1-Dio3 est l’un des rares domaines imprimés contrôlés par une région de contrôle d'impression méthylée sur le chromosome paternel, nommée IG-DMR. Dans l’embryon, au niveau du domaine Dlk1, Rtl1 et Dio3 les gènes codant pour des protéines sont exprimés à partir du chromosome paternel, tandis que les ARNs non-codants dont Meg3, les snoRNAs à boite C/D et les micro-ARNs sont exprimés à partir du chromosome maternel.Il a été montré que la copie maternelle de l'IG-DMR est nécessaire pour l'expression des gènes imprimés de ce domaine et que les ARNs de types enhancer (de la même région) activent la transcription des ARNs non-codants. Cependant, les mécanismes qui régulent l'expression imprimée de gènes codant pour des protéines restent indéterminés. Dans ce projet, nous avons cherché à élucider les mécanismes qui contrôlent l'expression spécifiquement paternelle des gènes codant pour des protéines ainsi que le rôle possible des ARNs non-codants dans ce processus.Pour nos études alléliques, nous avons utilisé des cellules ES hybrides qui ont été obtenues en croisant des lignées de M. musculus domesticus et M. musculus molossinus. Ces cellules ont été différenciées in vitro dans des lignées neurales. Dans les cellules ES, l'expression Dlk1 est détectée à partir des deux chromosomes parentaux à des niveaux très bas. Lors de la différenciation, l'allèle paternel de Dlk1 devient actif tandis que le niveau d'expression de l'allèle maternel reste faible. Nos études de la chromatine ont montré que cette surexpression est due à l’activation de la chromatine sur l'allèle paternel de Dlk1.L'un de nos objectifs était d'explorer le rôle de Meg3 (un long ARN non-codant) dans la régulation de l’empreinte de Dlk1. A cet effet, nous avons généré des cellules souches embryonnaires déficientes en Meg3. Dans toutes les lignées déficientes, de suppressions maternelles ou bi-alléliques, nous avons constaté une perte d’expression de tous les ANRs non-codants. De plus, l’expression de Dlk1 devient bi-allélique dans ces cellules. Pour élucider le mécanisme de l'empreinte de ce gène, nous avons décidé d'étudier les caractéristiques de la chromatine au niveau du promoteur Dlk1 dans les cellules déficientes en Meg3. Nous avons examiné les modifications activatrices et répressives des histones ainsi que l'occupation de l'ARN Pol II. Nous avons observé l'acquisition des marques d’une chromatine active sur les deux chromosomes ainsi que le recrutement bi-allélique de l'ARN Pol II.Bien que nous n’ayons pas pu détecter une perte de la marque répressive H3K27me3 suite à la surexpression de Dlk1, nous avons observé un gain d'acétylation sur ce résidu lysine. Afin de comprendre davantage le rôle de la marque H3K27me3 sur l’empreinte de Dlk1, nous avons généré des cellules ES dépourvues de EZH2, la méthyltransférase de H3K27. L’expression de Dlk1 dans les cellules différenciées dépourvues de H3K27me3 est bi-allélique.Enfin, ces données suggèrent que l'expression des ARNs non-codant empêche l'activation de Dlk1 sur le chromosome maternel via l’activité de EZH2 au cours du développement. / The Dlk1-Dio3 imprinted domain is one of the few imprinted domains that are controlled by a paternally methylated imprinting control region, IG-DMR. Protein-coding genes of the domain, Dlk1, Rtl1 and Dio3 are expressed from the paternal chromosome, and non-coding RNAs (ncRNAs) including Meg3, C/D box snoRNAs and microRNAs are expressed from the maternal chromosome exclusively in the embryo. Maternal copy of the IG-DMR is required for the imprinted gene expression at this domain. Enhancer RNAs transcribed from this region are involved in activation of ncRNA expression on the maternal chromosome. However, the regulation of imprinted expression of protein-coding genes remains unknown. In this project, we aimed to elucidate the mechanisms controlling the paternal specific expression of protein-coding genes and a possible role of ncRNAs in this process.For our allelic studies, we made use of hybrid ES cells that were obtained by crossing M. musculus domesticus and M. musculus molossinus strains. These cells were differentiated in vitro into neural lineages. In ES cells, Dlk1 expression is detected from both parental chromosomes at very low levels. Upon differentiation, paternal allele of Dlk1 gets activated while low level of expression is detected from maternal allele. Our chromatin studies showed that this upregulation is through the acquisition of active chromatin on the paternal allele of Dlk1.One of our aims was to explore the role of Meg3 long non-coding RNA (lncRNA) in the regulation of Dlk1 imprinting. For this purpose, we generated ES cells deficient in Meg3. In all maternal or biallelic deletion lines, we observed complete loss of all ncRNA expression. Interestingly, in these cells Dlk1 expression becomes biallelic. To elucidate the mechanism of imprinting of this gene, we set out to study the chromatin features at the Dlk1 promoter in Meg3 deficient cells. We looked into active and repressive histone modifications and RNA Pol II occupancy. We observed acquisition of active chromatin marks on both chromosomes as well as biallelic recruitment of RNA Pol II.Although we could not detect a loss of repressive mark H3K27me3 upon Dlk1 upregulation on the paternal allele, we observed gain of acetylation on this lysine residue. To further investigate the role of H3K27me3 mark on Dlk1 imprinting, we generated ES cells that lack functional EZH2, the H3K27 methyltransferase. Dlk1 is biallelically expressed in the differentiated cells that are devoid of H3K27me3.Combined, these data suggest a model in which non-coding RNA expression prevents the developmental activation of Dlk1 on the maternal chromosome by a process that also requires the activity of EZH2.

No description available.

No description available.

Cell phenotypic transitions, or cell fate decision making processes, are regulated by complex regulatory networks composed of genes, RNAs, proteins and metabolites. The regulation can take place at the epigenetic, transcriptional, translational, and post-translational levels to name a few. Epigenetic histone modification plays an important role in cell phenotype maintenance and transitions. However, the underlying mechanism relating dynamical histone modifications to stable epigenetic cell memory remains elusive. Incorporating key pieces of molecular level experimental information, we built a statistical mechanics model for the inheritance of epigenetic histone modifications. The model reveals that enzyme selectivity of different histone substrates and cooperativity between neighboring nucleosomes are essential to generate bistability of the epigenetic memory. We then applied the epigenetic modeling framework to the differentiation process of olfactory sensory neurons (OSNs), where the observed 'one-neuron-one-allele' phenomenon has remained as a long-standing puzzle. Our model successfully explains this singular behavior in terms of epigenetic competition and enhancer cooperativity during the differentiation process. Epigenetic level events and transcriptional level events cooperate synergistically in the OSN differentiation process. The model also makes a list of testable experimental predictions. In general, the epigenetic modeling framework can be used to study phenotypic transitions when histone modification is a major regulatory element in the system. Post-transcriptional level regulation plays important roles in cell phenotype maintenance. Our integrated experimental and computational studies revealed such a motif regulating the differentiation of definitive endoderm. We identified two RNA binding proteins, hnRNPA1 and KSRP, which repress each other through microRNAs miR-375 and miR-135a. The motif can generate switch behavior and serve as a noise filter in the stem cell differentiation process. Manipulating the motif could enhance the differentiation efficiency toward a specific lineage one desires. Last we performed mathematical modeling on an epithelial-to-mesenchymal transition (EMT) process, which could be used by tumor cells for their migration. Our model predicts that the IL-6 induced EMT is a stepwise process with multiple intermediate states. In summary, our theoretical and computational analyses about cell phenotypic transitions provide novel insights on the underlying mechanism of cell fate decision. The modeling studies revealed general physical principles underlying complex regulatory networks. / Ph. D.