It has recently become apparent that cells encode a large number of novel non-protein-coding genes called long non-coding RNAs (lncRNAs). Whilst the biological function of many lncRNAs remains unknown, recent evidence has suggested that lncRNAs may be important regulators of cellular growth, differentiation and may play a significant role in cancer. Epidermal growth factor (EGF) – an activator of the ERK1/2 signalling cascade – is an important spatio-temporal regulator of transcription and, ultimately, of cellular growth and movement. EGF stimulation triggers a wave-like expression of immediate-early genes (IE genes), followed by delayed-early genes (DE genes) and secondary-response genes (SR genes). Over the years, considerable effort has been made to unravel the regulatory loops downstream of EGF signalling. This study investigated whether lncRNAs are sensitive to EGF signalling and whether they play a role in the transcriptional programme associated with EGF signalling. In order to identify lncRNAs regulated by EGF signalling, I sequenced nuclear RNA in the presence or absence of EGF stimulation. RNA-seq data showed that 173 lncRNAs are upregulated by EGF, of which 89 were intergenic lncRNAs (lincRNAs). The time-dependent expression profile of EGF-upregulated lincRNAs followed the well-established expression pattern of IE genes. Finally, investigation of the expression of lincRNAs in primary breast and lung cancer cells showed that EGF-upregulated lincRNAs were differentially expressed in cancer. The EGF-dependent induction profile and cancer enrichment were particularly strong for one of the transcripts – EGF-induced lncRNA 1 (EIN1) – and I selected it for further studies. Firstly, using bioinformatics and biochemical approaches, I confirmed the non-coding status of the EIN1 transcript. Secondly, I confirmed that EIN1 transcription is ERK1/2-dependent and is independent of protein synthesis. Investigation of EIN1 expression in normal tissues showed its high enrichment in the human cardiovascular system. At the cellular level, the EIN1 transcript was predominantly found in the nucleus. Functionally, the depletion of endogenous EIN1 transcripts (using the newly developed CRISPRi approach) led to changes in the EGF-dependent transcription programme. EIN1 downregulation resulted in the addition of normally EGF-independent genes into the EGF-dependent expression programme. Collectively, these results show that EGF (via the ERK1/2 pathway) can regulate transcription of lincRNAs. The EIN1 example suggests that lincRNAs may play a crucial role in the modulation of the EGF-dependent expression programme by limiting of the scope of the programme.

Myoclonus-Dystonia (MD) is an inherited, rare, autosomal dominant movement disorder characterized by quick, involuntary muscle jerking or twitching (myoclonus) and involuntary muscle contractions that cause twisting and pulling movements, resulting in abnormal postures (dystonia). The first MD locus was mapped to 7q21-q31 and called DYT11; this locus corresponds to the SGCE gene. Our group previously identified a second MD locus (DYT15) which maps to a 3.18 Mb region on 18p11. Two patients were chosen to undergo next-generation sequencing, which identified 2,292 shared novel variants within the critical region. Analysis of these variants revealed a 3 bp duplication in a transcript referred to as CD108131, which is believed to be a long non-coding RNA. Characterization of this transcript determined that it is 863 bp in size, it is ubiquitously expressed, with high expression in the cerebellum, and it accounts for ~3% of MD cases.

Le dogme central de la biologie repose sur la production de protéines à partir de notre ADN. L’ADN est d’abord transcrit en ARN et celui-ci est ensuite traduit en protéine. C’est donc en cette dernière qu’est localisé le “pouvoir exécutif” de la cellule, ce qui explique le fait que les protéines soient devenues le centre d’attention de la recherche. L’ARN, quant à lui, est donc depuis longtemps considéré comme une molécule intermédiaire, dont l’unique raison d’être est le transfert d’information entre l’ADN et les protéines. Pourtant, ces dernières années, les avancées technologiques ont révélé qu’une majeure partie de notre génome, notre ADN, est transcrit en ARNs dits « noncodants » ne donnant pas lieu à une protéine. Ceux-ci sont impliqués dans de nombreux processus cellulaires et de ce fait participent aux pathologies. D’autre part, de nouvelles technologies ont aussi mené à l’observation que le métabolisme des ARNs, codants ou non, est la cible de nouveaux mécanismes de régulation: les modifications chimiques des ribonucléosides. Analysées de manière conjointe, ces découvertes poussent à la révision du rôle des ARNs au sein des processus cellulaires. Dès lors, dans le cadre de cette thèse nous avons voulu mieux comprendre la fonction et la régulation des molécules d’ARN afin d’en révéler le rôle plus central qu’ils jouent dans les processus cellulaire et en particulier, la cancérogenèse. Pour ce faire cette thèse comporte deux parties, la première décrit comment certains ARNs, dit “longs ARNs non-codants” participent au développement et à l’hétérogénéité du cancer colorectal. En effet ces ARNs exercent des fonctions “exécutives” sans être la source d’une protéine. Nous avons identifié 282 long ARNs non-codants dont les profils d’expression reflètent les différentes caractéristiques rencontrées au travers des différents sous-types de tumeurs colorectales. De plus, nos analyses informatiques ont indiqué que ces ARNs font partie intégrante des réseaux de signalisations les plus importants et les plus souvent dérégulés dans les différents sous-types que présente ce cancer. Enfin, et ce via des expériences in vitro nous soutenons la validité de nos analyses informatiques en confirmant le rôle de lncBLID-5, un long ARN non-codant, dans la régulation du cycle cellulaire et de la transition épithéliale vers mésenchymale un processus cellulaire très important dans les cancers colorectaux. Dans la deuxième partie nous avons étudié la méthylation des cytosines de l’ARN, une modification très récemment identifiée. Nous avons découvert que la protéine SRSF2, un facteur général de l’épissage des ARNs, est capable de se lier aux cytosines méthylées et ce plus fortement qu’aux cytosines non-méthylées. Enfin, nous montrons que la mutation P95H de SRSF2, très fréquente chez les patients atteints de leucémie, empêche SRSF2 de favoriser sa liaison aux cytosines méthylées laissant entrevoir de nouvelles explications à l’épissage défectueux conduisant à ce type de cancer. En conclusion nos travaux apportent de nouvelles informations quant à l’implication et la régulation des ARNs codants et non-codants dans le cadre du cancer. Ces résultats devraient nous mener à revoir le rôle qu’occupe l’ARN au sein des processus cellulaires sains ainsi que pathologiques, ouvrant la porte sur une nouvelle dimension de cibles diagnostiques et thérapeutiques. / Doctorat en Sciences biomédicales et pharmaceutiques (Médecine) / info:eu-repo/semantics/nonPublished

京都大学 / 0048 / 新制・課程博士 / 博士(エネルギー科学) / 甲第22797号 / エネ博第411号 / 新制||エネ||79(附属図書館) / 京都大学大学院エネルギー科学研究科エネルギー基礎科学専攻 / (主査)教授 片平 正人, 准教授 小瀧 努, 教授 森井 孝 / 学位規則第4条第1項該当 / Doctor of Energy Science / Kyoto University / DGAM

Les cancers bronchopulmonaires non à petites cellules (CBNPC) sont la première cause de décès par cancer, les adénocarcinomes étant la forme la plus fréquente. Malgré une prise en charge précoce, ils constituent, par leur taux de récidive important et leur mauvais pronostic, un véritable problème de santé publique. Nous nous intéressons à l'hypoxie, un facteur agressivité des ADC, et à la famille des longs ARNs non codants (lncARNs), dérégulés dans de nombreux cancers et en réponse à l'hypoxie, mais peu caractérisés sur le plan structural et fonctionnel. Ces transcrits représentent un espoir pour le développement de nouvelles thérapies. Au cours de ma thèse, j'ai identifié une dizaine de lncARNs régulés par l'hypoxie in vitro et in vivo dans des ADC de stades précoces. j'ai caractérisé NLUCAT1, un transcrit nucléaire induit par l'hypoxie. L'invalidation de ce transcrit par le système CRISPR/Cas9 a révélé une diminution de la prolifération et de l'invasion cellulaires, et une augmentation du stress oxydatif et de la sensibilité au cisplatine, traduisant un potentiel rôle pro-oncogénique de ce transcrit dans les ADC. L'analyse du transcriptome a révélé une répression des réseaux de gènes contrôlés par NRF2, HIF et NFkβ dans les cellules déficientes pour NLUCAT1. Nous avons notamment identifié des gènes de la réponse anti-oxydante régulés par NRF2 dont l'ARN interférence mime en partie les conséquences de l'inactivation de NLUCAT1 sur l'apoptose. Nos résultats démontrent que NLUCAT1 exerce des activités pro-tumorales dans les ADC et suggère qu'il pourrait représenter une cible thérapeutique potentielle dans ce type de cancer. / Non Small Cell Lung Cancer (NSCLC) is the leading cause of cancer death worldwide, with poor prognosis and a high rate of recurrence despite early surgical removal. It is therefore essential to identify new prognostic markers and new therapeutic targets. We are interested in gene regulation related to hypoxia, a factor associated with relapse of lung adenocarcinomas (LUAD). The roles of long non coding RNAs (incRNAs) in cancer development and hypoxic response are largely unexplored. A transcriptome profiling of early-stage LUAD samples indicated that a set of incRNAs was correlated to a metagene hypoxic signature. Some of these transcripts were also sensitive to hypoxia in LUAD cell lines. We focused on a new "hypoxaLinc", named NLUCAT1 that is strongly up-regulated by hypoxia in vitro and correlated to hypoxic markers and bad prognosis in LUAD samples. Full molecular charactherization of NLUCAT1 showed that LUCAT1 is mainly regulated by NF-kβ and NRF2 transcription factors. Targered deletion of NLUCAT using CRISPR/CAS9 in A549 LUAD cell line, revelated a decrase in proliferative and invasive properties, an increase in oxidative stress and a higher sensisivity to displatin-induced apoptosis. We identified genes of the NRF2-regulated and anti-oxidant response whose RNA interference partially mimicked the consequences of NLUCAT1 inactivation on ROS-dependent caspase activation. Overall, our data strongly demonstrate that NLUCAT1 exerts pro-tumoral activities in early stages hypoxic LUADs ans suggest it could represent a new potential therapeutic target in lung cancer.

Wound healing is a complex physiological process. Effective wound healing enables the skin barrier function to be restored once the skin is injured. However, due to the complex nature of wounds, the mechanisms underlying tissue repair are still poorly understood. This has hindered the development of treatment for chronic wound, which is posing threat to both human health system and economy. Long non-coding RNAs (lncRNAs) have been identified as important gene expression regulators and to play functional roles in many biological processes.  The aim of this study was to unravel the gene regulatory network in human skin wound healing, in particular, to identify lncRNAs that may play a functional role in skin repair. Here we performed RNA sequencing to profile gene expression in fibroblasts and keratinocytes isolated from matched skin and day-7 acute wounds of five healthy donors. We predicted a total of 1974 and 3444 mRNA–lncRNA correlated pairs in wound fibroblasts and wound keratinocytes, respectively. By integrating the results from gene ontology enrichment and weighted co-expression network analysis, we shortlisted lncRNAs that may play a functional role in human skin wound healing.

Indiana University-Purdue University Indianapolis (IUPUI) / Non-coding RNAs across the genome have been associated with various biological processes, ranging from regulation of splicing to remodeling of chromatin. Amongst the repertoire of non-coding sequences lies a critical species of RNAs called long non-coding RNAs (lncRNAs). LncRNAs significantly contribute to a large spectrum of human phenotypes, including cancers, Heart failure, Diabetes, and Alzheimer’s disease. This dissertation emphasizes the need to characterize the functional role of lncRNAs to improve our understanding of human diseases. This work consolidates a resource from multiple computational genomics and natural language processing-based approaches to advance our ability to functionally annotate hundreds of lncRNAs and their interactions, providing a one-stop lncRNA functional annotation and dynamic interaction network and multi-facet omics data visualization platform. RNA interactions are vital in various cellular processes, from transcription to RNA processing. These interactions dictate the functional scope of the RNA. However, the multifaceted functional nature of RNA stems from its ability to form secondary structures. Therefore, this work establishes a computational method to characterize RNA secondary structure by integrating SHAPE-seq and long-read sequencing to enhance further our understanding of RNA structure in modulating the post-transcriptional regulatory processes and deciphering the influence at several layers of biological features, ranging from structure composition to consequent protein occupancy. This study will potentially impact the research community by providing methods, web interfaces, and computational pipelines, improving our functional understanding of long non-coding RNAs. This work also provides novel integration methods of technologies like Oxford Nanopore-based long-read sequencing, RNA structure-probing methods, and machine learning. The approaches developed in this dissertation are scalable and adaptable to investigate further the functional and regulatory role of RNA and its structure. Overall, this study accelerates the development of RNA-based diagnostics and the identification of therapeutic targets in human disease.

Receptor-interacting protein kinase 1 (RIPK1) is a major upstream mediator of inflammation and cell death. These processes are key to common inflammatory diseases such as atherosclerosis, where macrophages play an important role in their progression. Closely linked to the expression of downstream genes, long non-coding RNAs (lncRNAs) are critical to controlling cellular processes in health and disease. As post-transcriptional regulatory mechanisms for RIPK1 are largely unknown, this project seeks to study the stability of Ripk1 mRNA and RIPK1 protein, along with Ripk1 mRNA interactions with relevant lncRNAs under various conditions. Using transcription and translation inhibitors, we determined that both Ripk1 mRNA and RIPK1 protein are relatively unstable with half-lives of approximately 3 h. Their turnover in macrophages is further influenced by the timing and duration of inflammation. We also implemented a novel RNA pull-down procedure to capture Ripk1 mRNA and attached lncRNAs for next-generation sequencing. Through differential expression analysis, we discovered significant upregulation of known lncRNA AC125611 and novel lncRNA MSTRG.5894.1 in Ripk1-targeted samples subject to inflammation. MSTRG.7477.1 was upregulated during necroptosis, while MSTRG.5684.5 was upregulated during both inflammation and necroptosis. GapmeR-mediated knockdowns of AC125611 and MSTRG.5684.5 under inflammatory conditions resulted in decreased Ripk1 mRNA expression and RIPK1 protein expression, respectively. Meanwhile, MSTRG.7477.1 knockdowns were connected to decreased RIPK1 at both the mRNA and protein levels. Our research ultimately advances the current understanding of RIPK1 regulation by focusing on Ripk1 mRNA-lncRNA associations and turnover of its mRNA and protein in macrophages, paving the way for future investigations into their capacity to act as therapeutic targets.

No description available.

Epigenetics was a term first coined to understand how cells with the same genetic make up can differentiate into various cell types. Elegant research over the past 30 years has shown that these mechanisms include heritable marks such as DNA methylation and histone modifications along with stable expression of non- coding RNAs. Within the realm of epigenetics is a phenomenon known as genomic imprinting. Imprints are marks that distinguish the maternal from the paternal chromosomes in the form of methylation. Methylation marks can influence transcript expression, resulting in only one allele being expressed. One imprinted domain is the Kcnq1 domain located on chromosome 11p15.5 in humans and chromosome 7 in the mouse. This domain is thought to be under the control of a paternally expressed long noncoding RNA (ncRNA) Kcnq1ot1. The Kcnq1ot1 ncRNA is expressed on the paternal chromosome due to a differentially methylation region located within its promoter. The promoter is methylated on the maternal allele thus inhibiting ncRNA expression, whereas the promoter is unmethylated on the paternal allele. In the placenta, a most of the genes located within a one mega-basepair region are exclusively expressed from the maternal chromosome, whereas the transcripts on the paternal chromosome are silenced by the ncRNA. The placenta seems to follow the classic idea of an imprinted domain. However, in the embryo and more specifically, in the embryonic heart, this is not the case. In the embryonic heart, only a 400kb region is restricted to maternal expression. In addition, one the genes, Kcnq1, starts out expressed exclusively from the maternal allele in early development but switches to biallelic expression during mid-gestation. The purpose of my research is to determine the underlying complexities that are involved in the regulation of transcripts within the Kcnq1 domain. This involves the Kcnq1 gene itself, which has been shown to transition from mono- to biallelic expression during mid-gestation and the Kcnq1ot1 ncRNA per se. I hypothesize that regulation by the Kcnq1ot1 ncRNA is not occurring in a uniform manner in the embryo; rather, the amount of regulation by the ncRNA is dependent on the developmental stage and specific tissue. In addition, this regulation involves complex interactions between enhancers, insulators and other regulatory elements to control the amount of silencing by the Kcnq1ot1 ncRNA. First, through a series of experiments looking at the Kcnq1 promoter, the mechanism of Kcnq1 paternal expression was determined. It was confirmed that Kcnq1 becomes biallelic during mid-gestation in the heart. Bisulfite mutagenesis and methylation sensitive chromatin immunoprecipitation were used to test the hypothesis that the Kcnq1 promoter was methylated in early development and then lost its methylation mark. However, a lack of methylation disproved this mechanism of paternal Kcnq1 activation. Rather, chromosome conformation capture (3C) determined that the Kcnq1 promoter interacts in a tissue-specific manner with regions within the domain that have enhancer activity. The role of the ncRNA within our system was also investigated. Interestingly, when Kcnq1ot1 allelic expression was profiled throughout development in heart, it transitioned to biallelic expression during heart development but remained monoallelic in the liver and brain. Several possibilities could account for this phenomenon, including loss of promoter methylation and/or an alternative transcript start site. Both of these options were explored using bisulfite mutagenesis and 5' RACE. However, the Kcnq1ot1 promoter region retained its methylation mark even after the maternal transcript was turned on, disproving this idea. Rather, a maternal specific transcript was found in the heart to start downstream of the CpG islands. Lastly, to gain a better understand of the Kcnq1ot1 ncRNA, experiments were carried out on a mutant mouse in which a truncated form of the ncRNA was transmitted paternally; this is dubbed the "Kterm" mouse. Unexpectedly, Kcnq1 still followed the same mono- to biallelic transition as seen in the wild-type, whereas the head and body counterparts from the same stage embryos were biallelic for Kcnq1. Also, the immediate upstream genes, Cdk1nc and Slc22a18, lost their mono-allelic expression in neonatal heart, liver and brain when the Kterm mutation was transmitted. This suggested that Kcnq1ot1 did not function as a silencer for Kcnq1 paternal expression in the heart, but rather had an alternative and previously unknown function. From qRT-PCR, 3C and ChIP assays, it was determined that the Kcnq1ot1 ncRNA plays a role in regulating Kcnq1 gene expression in the heart by limiting its interaction to specific cis-acting enhancers. When the ncRNA was absent, the Kcnq1 promoter interacted with non-native sites along the domain, possibly causing the increase in transcript expression. This phenomenon was specific to the heart and was not seen in other tissues. These findings showed that Kcnq1 paternal expression is the result of strong developmental and tissue specific enhancers. Chromatin interactions in cis put a strong enhancer in contact with the Kcnq1 promoter to increase its expression in later development. In addition, a truncation mutation model identified a key role for the Kcnq1ot1 ncRNA in regulating Kcnq1 expression. Instead of regulating the imprinting status of Kcnq1, the ncRNA regulates the amount of Kcnq1 transcript being produced in the heart by regulating chromatin interactions. Finally, these studies identified a maternally expressed Kcnq1ot1 transcript whose role in heart development is still not fully understood. Taken together, these findings support a model where an inhibitory factor(s) silence the paternal Kcnq1 transcript and maternal Kcnq1ot1 transcript and in later development, this factor is released allowing for expression and chromatin interactions to occur. / Molecular Biology and Genetics