Colorectal cancer (CRC) is a major source of morbidity and mortality in the Western world. Approximately 15% of all CRCs develop via the mutator pathway, which results from a deficiency of mismatch repair (MMR) system and leads to genome-wide microsatellite instability (MSI). MLH1 promoter hypermethylation accounts for the majority of MSI CRCs. Numerous single nucleotide polymorphisms have been identified in MMR genes, however their functional roles in affecting MMR system, and therefore susceptibility to MSI CRCs, are unknown. This study uses a multidisciplinary approach combining molecular genetics, epigenetics, and epidemiology to examine the contribution of MMR gene polymorphisms in CRC. Among a panel of MMR SNPs examined, the MLH1 (-93G>A) promoter polymorphism (rs1800734) was shown to be associated with increased risk of MSI CRCs in two Canadian populations, Ontario and Newfoundland. Functional studies of the MLH1-93G>A polymorphism indicate that it has weak effects on the core promoter activity, although it dramatically reduces activity of the shorter promoter constructs in a panel of cell lines. Furthermore, MLH1 gene shares a bi-directional promoter with EMP2AIP1 gene, and the MLH1-93G>A polymorphism increases the activity of the reverse, EPM2AIP1 promoter. Examination of alternative role of the MLH1-93G>A polymorphism in MSI-H CRCs led to evaluation of a 500-kilobase pair chromosome 3 region around the MLH1 gene and identification of two additional SNPs, rs749072 and rs13098279, which are in strong linkage disequilibrium with rs1800734. All three SNPs showed strong associations with MLH1 promoter methylation, loss of MLH1 protein expression, and MSI-H CRCs in three populations, Ontario, Newfoundland, and Seattle. Such findings potentially implicate genetic susceptibility to DNA methylation. Logistic regression models for MSI-H versus non-MSI-H CRCs demonstrate that models including MLH1 IHC status and MLH1 promoter methylation status fit the data most parsimoniously in all three populations combined, however, when rs1800734/rs749072/rs13098279 was added to this model, polymorphisms no longer remained significant indicating that the observed associations of these polymorphisms with the MSI-H CRCs occur through their effect on DNA methylation. This study identified a novel mechanism in which common missense alterations may contribute to complex disease.

Domestication of animals has given rise to a great phenotypic divergence in selected animals and rapidly generated species of animals more accustomed to human contact and social interactions within the species. Previous studies in chickens (Gallus gallus) have managed to find behavioral and adaptive differences between Red Junglefowl (RJF) and White Leghorn (WL), differences inherent to the domestication process. These phenotypic changes could spawn from a variety of different genomic factors, including an epigenetic gene expression regulatory mechanism known as CpG methylation, a DNA modification of CpG dinucleotides that in turn affect nucleosome formation. In this study we investigated the methylation differences between RJF and WL. This by selecting genes that has previously been shown to be both differentially expressed (DE) and differentially methylated (DM) between RJF and WL, and had shown the same kind of differences in both parental animals and their offspring. By using methylation-sensitive high-resolution melting (MSHRM) we tried to confirm previous DM result, and four genes; FUCA1, RUFY3, PCDHAC1 and TXNDC16 were tested and verified to be DM between RJF and WL.

<p>Lichens are symbioses between a fungus and a photosynthesizing partner such as a green alga or a cyanobacterium. Unlike mycorrhizal or rhizobial symbioses, the lichen symbiosis is not well understood either morphologically or molecularly. The lichen symbiosis has been somewhat neglected for several reasons. Lichens grow very slowly in nature (less than 1 cm a year), it is difficult to grow the fungus and the alga separately and, moreover, it remains difficult to resynthesize the mature symbiosis in the laboratory. It is not yet possible to delete genes, nor has any transformation method been established to introduce genes into the genomes of either the fungus or the alga. However, the lack of genetic tools for these organisms has been partially compensated for by the sequencing of the genomes of the lichenizing fungus <italic>Cladonia grayi</italic> and its green algal partner <italic>Asterochloris</italic> sp. This work uses the model lichen system <italic>Cladonia grayi</italic> and the associated genomes to explore one evolutionary and one developmental question concerning the lichen symbiosis.</p><p>Chapter One uses data from the genomes to assess whether there was evidence of horizontal gene transfer between the lichen symbionts in the evolution of this very intimate association; that is, whether genes of algal origin could be found in the fungal genome or vise versa. An initial homology search of the two genomes demonstrated that the fungus had, in addition to ammonium transporter/ammonia permease genes that were clearly fungal in origin, ammonium transporter/ammonia permease genes which appeared to be of plant origin. Using cultures of various lichenizing fungi, plant-like ammonium transporter/ammonia permease genes were identified by degenerate PCR in ten additional species of lichen in three classes of lichenizing fungi including the Lecanoromycetes, the Eurotiomycetes, and the Dothidiomycetes. Using the sequences of these transporter genes as well as data from publically available genome sequences of diverse organisms, I constructed a phylogy of 513 ammonium transporter/ammonia permease sequences from 191 genomes representing all main lineages of life to infer the evolutionary history of this family of proteins. In this phylogeny I detected several horizontal gene transfer events, including the aforementioned one which was demonstrated to be not a transfer from plants to fungi or vise versa, but a gene gain from a group of phylognetically unrelated hyperthermophilic chemoautolithotrophic prokaryotes during the early evolution of land plants (Embryophyta), and an independent gain of this same gene in the filamentous ascomycetes (Pezizomycotina), which was subsequently lost in most lineages but retained in even distantly related lichenized fungi. Also demonstrated was the loss of the native fungal ammonium transporter and the subsequent replacement of this gene with a bacterial ammonium transporter during the early evolution of the fungi. Several additional recent horizontal gene transfers into lineages of eukaryotes were demonstrated as well. The phylogenetic analysis suggests that what has heretofore been conceived of as a protein family with two clades (AMT/MEP and Rh) is instead a protein family with three clades (AMT, MEP, and Rh). I show that the AMT/MEP/Rh family illustrates two contrasting modes of gene transmission: AMT family as defined here exhibits standard parent-to-offspring inheritance, whereas the MEP family as defined here is characterized by several ancient independent horizontal gene transfers (HGTs) into eukaryotes. The clades as depicted in this phylogenetic study appear to correspond to functionally different groups, with ammonium transporters and ammonia permeases forming two distinct and possibly monophyletic groups.</p><p>In Chapter Two I address a follow-up question: in key lichenizing lineages for which ammonium transporter/ammonia permease (AMTP) genes were not found in Chapter One, were the genes lost? The only definitive infomation which can demonstrate absence of a gene from a genome is a full genome sequence. To this end, the genomes of eight additional lichenizing fungi in the key clades including the Caliciales (sensu Gaya 2011), the Peltigerales, the Ostropomycetidae, the Acarosporomycetidae, the Verrucariales, the Arthoniomycetidae and the Lichinales were sequenced using the Ilumina HiSeq technology and assembled with the short reads assembly software Velvet. These genomes were searched for ammonium transporter/ammonia permease sequences as well as 20 test genes to assess the completeness of each assembly. The genes recovered were included in a refined phylogenetic analysis. The hypothesis that lichens symbiotic with a nitrogen-fixing cyanobacteria as a primary photobiont or living in high nitrogen environments lose the plant-like ammonium transporters was upheld, but did not account for additional losses of ammonium transporters/ammonia permeases in the Acarosporomyetidae and Arthoniomycetes. In addition, the four AMTP genes from <italic>Cladonia grayi</italic> were shown to be functional by expression of the lichen genes in a strain of <italic>Saccharomyces cerevisiae</italic> in which all three native ammonium transporters were deleted, and assaying for growth on limiting ammonia as a sole nitrogen source. </p><p>In Chapter Three I use genome data to address a developmental aspect of the lichen symbiosis. The finding that DNA in three genera of lichenizing fungi is methylated in symbiotic tissues and not methylated in aposymbiotic tissues or in the free-living fungus (Armaleo & Miao 1999a) suggested that epigenetic silencing may play a key role in the development of the symbiosis. Epigenetic silencing involves several steps that are conserved in many eukaryotes, including methylation of histone H3 at lysine 9 (H3K9) in nucleosomes within the silenced region, subsequent binding of heterochromatin-binding protein (HP1) over the region, and the recruitment of DNA methyltransferases to methylate the DNA, all of which causes the underlying chromatin to adopt a closed conformation, inhibiting the transcriptional machinery from binding. In this chapter I both identify the genes encoding the silencing machinery and determine the targets of the silencing machinery. I use degenerate PCR and genome sequencing to identify the genes encoding the H3K9 histone methyltransferase, the heterochromatin binding protein, and the DNA methyltransferases. I use whole genome bisulfite sequencing of DNA from the symbiotic structures of <italic>Cladonia grayi</italic> including podetia, squamules and soredia as well as DNA from cultures of the free-living fungus and free-living alga to determine which regions of the genome are methylated in the symbiotic and aposymbiotic states. In particular I examine regions of the genomes which appear to be differentially methylated in the symbiotic versus the aposymbiotic state. I show that DNA methylation is uncommon in the genome of the fungus in the symbiotic and aposymbiotic states, and that the genome of the alga is methylated in the symbiotic and aposymbiotic states.</p> / Dissertation

Diss. (sammanfattning) Stockholm : Karolinska institutet, 2009. / Härtill 6 uppsatser.

Thesis (Ph. D.)--University of Washington, 2000. / Vita. Includes bibliographical references (leaves 126-144).

Le cancer du col utérin est un des cancers les plus prévalents dans le monde et représente une cause majeure de mortalité par cancer chez la femme. Le dépistage de ce cancer est actuellement basé sur l’examen clinique, la colposcopie, la cytologie du col utérin et l’histopathologie dont la reproductibilité inter observateur est médiocre surtout en ce qui concerne les lésions de bas grade (CIN 1). Bien qu’amélioré par de nouveaux marqueurs (p16INK4A), la recherche de nouveaux marqueurs spécifiques des lésions du col utérin nécessite l’élucidation des mécanismes de la cancérogenèse parmi lesquels les altérations épigénétiques, but de notre recherche. Nous avons évalué au cours des différentes étapes de la cancérogenèse du col utérin la méthylation globale de l'ADN par immunohistochimie et HPLC. L’hyperméthylation des îlots de CpG de trois gènes candidats impliqués dans la signalisation cellulaire (CDH13, DAPK1, TWIST1) a été étudié par une technique de PCR – méthylation spécifique (MSP) après modification de l’ADN au bisulfite de sodium. L’étude immunohistochimique à l'aide d’anticorps dirigés contre la 5-MeCyd objective une hypométhylation globale de l’ADN tardive n’apparaissant que dans les cancers invasifs. Ces résultats ont été confirmés par HPLC. Par contre, l’hyperméthylation des îlots de CpG des gènes candidats étudiés est présente dés le stade lésions précancéreuses. Ces résultats suggèrent qu’au cours de la cancérogenèse du col utérin, l’hypométhylation globale et l’hyperméthylation des îlots de CpG sont deux mécanismes indépendants. L’étude du profil de méthylation des régions promotrices des gènes CDH13, DAPK1, TWIST1 pourrait constituer un marqueur potentiel des lésions précancéreuses et cancéreuses du col de l’utérus / Background: The uterine cervix cancer is one of the main cancer among women worldwide and represents an important cause of death in women especially in developing countries. The screening of this cancer is based on clinic tests, colposcopy, cytology and histopathology which have a low inter observer reproducibility, specially concerning CIN1 lesions. Although improved by new markers (p16INK4A), the discovery of new specific markers requires the elucidation of mechanisms of carcinogenesis such as epigenetic alterations, aim of our study. Material and methods: We evaluated during uterine cervix carcinogenesis global DNA methylation by immunohistochemistry and HPLC. CpGs islands hypermethylation of three genes implied in cellular signalizing (CDH13, DAPK1, TWIST1) were evaluated using specific methylation PCR (MSP) after bisulfite DNA modification. Results: Immunohistochemistry study using anti 5-MeCyd antibodies objectives a late global DNA hypomethylation appearing only in invasive uterine cervix carcinoma. These results were confirmed by HPLC. However, CpGs islands hypermethylation of the promoter of CDH13, DAPK1 and TWIST1 genes studied were detected in precancerous lesions. Conclusions: Our results suggest that during uterine cervix carcinogenesis, global DNA hypomethylation and gene hypermethylation were two independent mechanisms. The methylation profile of the promoter regions of CDH13, DAPK1 and TWIST1 genes would represent a potential marker for the diagnosis of the precancerous and cancerous uterine cervix lesions

published_or_final_version / Medical Sciences / Master / Master of Medical Sciences

published_or_final_version / Medicine / Doctoral / Doctor of Philosophy

Epigenetic dysfunction is a known contributor in carcinogenesis, and is emerging as a mechanism involved in toxicant-induced malignant transformation for environmental carcinogens such as arsenicals. In addition to aberrant DNA methylation of single genes, another manifestation of epigenetic dysfunction in cancer is agglomerative DNA methylation, which can participate in long-range epigenetic silencing that targets many neighboring genes and has been shown to occur in several types of clinical cancers. Using in vitro model systems of toxicant-induced malignant transformation, we found hundreds of aberrant DNA methylation events that emerge during malignant transformation, some of which occur in an agglomerative fashion. In an arsenite-transformed prostate epithelial cell line, the protocadherin (PCDH), HOXC and HOXD gene family clusters are targeted for agglomerative DNA methylation. Aberrant DNA methylation in general occurred more often within H3K27me3 stem cell domains. We found a striking association between enrichment of H3K9me3 stem cell domains and toxicant-induced agglomerative DNA methylation. Global gene expression profiling of the arsenite-transformed prostate epithelial cells showed that gene expression changes and DNA methylation changes were negatively correlated, but less than 10% of the hypermethylated genes were down-regulated. These studies confirm that a majority of the DNA hypermethylation events occur at transcriptionally repressed, H3K27me3 marked genes. In contrast to aberrant DNA methylation targeting H3K27me3 pre-marked silent genes, we found that actively expressed ZNF genes marked with H3K9me3 on their 3' ends, are preferred targets of DNA methylation linked gene silencing. H3K9me3 mediated gene silencing of ZNF genes was widespread, occurring at individual ZNF genes on multiple chromosomes and across ZNF gene family clusters. At ZNF gene promoters, H3K9me3 and DNA hypermethylation replaced H3K4me3, resulting in a widespread down-regulation of ZNF gene expression which accounted for 8% of all the down-regulated genes in the arsenical-transformed cells. In summary, these studies associate arsenical exposure with agglomerative DNA methylation of gene family clusters and widespread silencing of ZNF genes by DNA hypermethylation-linked H3K9me3 spreading, further implicating epigenetic dysfunction as a driver of arsenical-induced carcinogenesis.

The tumor microenvironment consists of multiple cells types, including endothelial cells that line the tumor vasculature. Tumor vasculature is often abnormal and results in development of tissue ischemia, another contributing factor to the tumor microenvironment. Previous studies have demonstrated that ischemia influences epigenetic programming, but the mechanisms remained unclear and required further investigation. First, we profiled DNA methyltransferase (DNMT) expression and activity in human colorectal cancer cells (HCT116) under hypoxia or hypoglycaemia (mimicking ischemia). We found that DNMT1 and DNMT3b were significantly downregulated by hypoxia and hypoglycaemia, and DNMT3a was downregulated by hypoglycaemia. However, DNMT1 downregulation was p53-dependent. To examine if the changes in DNMT expression and activity translated to changes in DNA methylation patterns, we used bisulfite sequencing and examined the promoter region of p16. Hypoglycaemia significantly demethylated this region in both p53 wild-type and p53-null cells. Next, we used a genome-wide approach to discover what additional genes are hypomethylated by ischemia. Methylated DNA was immunoprecipitated and analysed with an Affymetrix promoter array, in parallel with an expression array. Ingenuity pathway analysis software revealed that a significant proportion of genes which were hypomethylated and upregulated were involved in cellular movement, including PLAUR and CYR61. We believe that hypoxia and hypoglycaemia may be driving changes in DNA methylation through dysregulation of DNMTs, resulting in cells acquiring a more mobile phenotype in ischemic regions. DNMT and histone deacetylase inhibitors are commonly used in research and some cancer therapies. Modifying epigenetic patterning with these inhibitors has been widely studied in cancer cells, but only briefly explored in the tumor’s vascular endothelium. We profiled the effect of these inhibitors on endothelial cell (EC) behaviour, and tested if combining them with a targeted anti-angiogenic therapy would augment the inhibition of angiogenesis. When the DNMT inhibitor 5-aza-2’-deoxycytidine was combined with sunitinib, inhibition of EC proliferation was enhanced compared to treatment with sunitinib alone. EC migration was also inhibited by the combination of these two inhibitors, but not in an additive manner. These studies have improved our understanding of how altering epigenetic patterning with ischemia and therapeutic inhibitors can influence colorectal cancer and endothelial cell behaviour. / Canadian Cancer Society Research Institute. The Cancer Research Society.