Le Glypican-3 (GPC3) est surexprimé dans 72% des carcinomes hépatocellulaire (CHC). C’est un co-récepteur membranaire du récepteur WNT, qui appartient à la famille des protéoglycanes à sulfates d'héparane. L'objectif général de ma thèse vise à étudier les mécanismes de régulation post-transcriptionnelle de l’expression du GPC3 dans le CHC. Pour cela, j’ai développé un test fonctionnel qui m’a permis de cribler une bibliothèque de 876 microARNs humains. Ceci a conduit à l’identification de 5 microARNs régulateurs de l’expression de l’ARNm codant pour le GPC3 via sa région 3’ non traduite (NT). Mon travail de thèse porte plus particulièrement sur le miR-1271 et le miR-1291 car ils sont dérégulés dans le CHC et sont respectivement inhibiteur et inducteur de l’expression du GPC3. Dans un premier projet, j’ai démontré que le miR-1271 cible directement la région 3’NT du GPC3 et diminue la stabilité de son ARNm. Ce microARN est sous-exprimé dans le CHC et son expression corrèle négativement avec celle de l'ARNm du GPC3 dans les CHC associés à une infection par le virus de l’hépatite B. Dans un deuxième projet, j’ai démontré que le miR-1291 régule positivement l’expression du GPC3 en inhibant un facteur intermédiaire. Une analyse in silico a permis d’identifier IRE1α comme candidat. IRE1α est une protéine transmembranaire du réticulum endoplasmique (RE) qui participe à « l’Unfolded Protein Response », une réponse adaptative activée lors de l’accumulation de protéines mal conformées dans le RE. J’ai démontré qu’IRE1α clive l’ARNm codant pour le GPC3 grâce à son activité endoribonucléase. D’autre part, le miR-1291 cible directement l’ARNm codant pour IRE1α dans sa région 5’NT ce qui inhibe son expression et induit une surexpression du GPC3. Le miR-1291 est surexprimé dans le CHC et son expression corrèle positivement avec celle de l’ARNm du GPC3. En conclusion, mon travail de thèse m’a permis de mettre en évidence et de caractériser deux nouveaux microARNs (miR-1271 et miR-1291) contrôlant l’expression du GPC3 par des mécanismes directs ou indirects. La pertinence physiopathologique de ces régulations dans le CHC est en accord avec les niveaux d’expression respectifs de ces microARNs, qui pourraient contribuer à la surexpression du GPC3 dans ces tumeurs. / Glypican-3 (GPC3) is overexpressed in 72% of hepatocellular carcinoma (HCC). It is a co-receptor for WNT receptor and belongs to the heparan sulfate proteoglycans family. The general objective of my PhD thesis was to study the mechanisms by which GPC3 is post-transcriptionnally regulated in HCC. To this end, I developed a functional test that allowed me to screen a library of 876 human microRNAs. This led me to identify 5 microRNAs that regulate the expression of GPC3 mRNA through its 3’Untranslated Region (UTR). The work presented in this thesis particulary focuses on miR-1271 and miR-1291 as both microRNAs present a deregulated expression in HCC and are respectively inhibitor and activator of GPC3 mRNA expression. In a first project, I demonstrated that miR-1271 directly binds to GPC3 mRNA 3’UTR and affects its stability. This microRNA is underexpressed in HCC and its expression negatively correlates with that of GPC3 mRNA in a subgroup of HCC corresponding to those associated with hepatitis B virus infection. In a second project, I demonstrated that miR-1291 postively regulates the expression of GPC3 mRNA by targeting an intermediate factor. An in silico analysis led to the identification of the Inositol Requiring Enzyme 1 alpha (IRE1α) as a potential candidate. IRE1α is an endoplasmic reticulum (ER) resident type I transmembrane protein and contributes to the signaling of the Unfolded Protein Response (UPR). The UPR is an adaptive response activated upon accumulation of improperly folded proteins in the ER. I showed that IRE1α cleaves GPC3 mRNA through its endoribonuclease activity. Moreover I demonstrated that miR-1291 directly targets IRE1α mRNA through its 5’UTR, thereby decreasing its expression and contributing to GPC3 mRNA overexpression. MiR-1291 is overexpressed in HCC and its expression positively correlates with that of GPC3 mRNA. In summary, the work carried out during my PhD allowed the identification and the characterization of two new microRNAs (miR-1271 and miR-1291) that control the expression of GPC3 mRNA through direct or indirect mechanisms. The pathophysiological relevance of these regulatory mechanisms is in agreement with the respective expression levels of these microRNAs in HCC, which could therefore contribute to the overexpression of GPC3 in those tumors.

Glypican-3 (GPC3), the cellular membrane proteoglycan, has been established as a tumor biomarker for early diagnosis of hepatocellular carcinoma (HCC). GPC3 is highly expressed in more than 70% HCC tissues detected by antibody-based histopathological systems. Recently, aptamers, a short single-strand DNA or RNA generated from systematic evolution of ligands by exponential enrichment (SELEX), were reported as potential alternatives in tumor-targeted imaging and diagnosis. In this study, a total of 19 GPC3-bound aptamers were successfully screened by capillary electrophoresis (CE)-SELEX technology. After truncated, AP613-1 was confirmed to specifically target GPC3 with a dissociation constant (KD) of 59.85 nM. When modified with a phosphorothioate linkage, APS613-1 targeted GPC3 with a KD of 15.48 nM and could be used as a specific probe in living Huh7 and PLC/PRF/5 imaging, GPC3-positive cell lines, but not in L02 or A549, two GPC3-negative cell lines. More importantly, Alexa Fluor 750-conjugated APS613-1 could be used as a fluorescent probe to subcutaneous HCC imaging in xenograft nude mice. Our results indicated that modified AP613-1, especially APS613-1, was a potential agent in GPC3-positive tumor imaging for HCC early diagnosis.

This body of work is concerned with the genetics of craniofacial morphology and specifically with that of the cranial sutures which form fibrous articulations between the calvarial bones. The premature fusion of these sutures, known as craniosynostosis, is a common developmental abnormality and has been extensively utilised here as a tool through which to study the genetics of suture morphogenesis and craniofacial diversity. Investigations began with a search for polymorphisms associated with normal variation in human craniofacial characteristics. Denaturing High-Performance Liquid chromatography was used to identify polymorphisms in two genes causative for craniosynostosis by analysing DNA from a large cohort of individuals from four ethnogeographic populations. A single nucleotide polymorphism in fibroblast growth factor receptor 1 was identified as being associated with variation in the cephalic index, a common measure of cranial shape. To further, and specifically, investigate the molecular processes of suture morphogenesis gene expression was compared between unfused and prematurely fusing/fused suture tissues isolated from patients with craniosynostosis. Two approaches, both utilising Affymetrix gene expression microarrays, were used to identify genes differentially expressed during premature suture fusion. The first was a novel method which utilised the observation that explant cells from both fused and unfused suture tissue, cultured in minimal medium, produce a gene expression profile characteristic of minimally differentiated osteoblastic cells. Consequently, gene expression was compared between prematurely fused suture tissues and their corresponding in vitro de-differentiated cells. In addition to those genes known to be involved in suture morphogenesis, a large number of novel genes were identified which were up-regulated in the differentiated in vivo state and are thus implicated in premature suture fusion and in vivo osteoblast differentiation. The second microarray study involved an extensive analysis of 16 suture tissues and compared gene expression between unfused (n=9) and fusing/fused sutures (n=7). Again, both known genes and a substantially large number of novel genes were identified as being differentially expressed. Some of these novel genes included retinol binding protein 4 (RBP4), glypican 3 (GPC3), C1q tumour necrosis factor 3 (C1QTNF3), and WNT inhibitory factor 1 (WIF1). The known functions of these genes are suggestive of potential roles in suture morphogenesis. Realtime quantitative RT PCR (QRT-PCR) was used to verify the differential expression patterns observed for 11 genes and Western blot analysis and confocal microscopy was used to investigate the protein expression for 3 genes of interest. RBP4 was found to be localised on the ectocranial surface of unfused sutures and in cells lining the osteogenic fronts while GPC3 was localised to suture mesenchyme of unfused sutures. A comparison between each unfused suture (coronal, sagittal, metopic, and lambdoid) demonstrated that gene expression profiles are suture-specific which, based on the identification of differentially expressed genes, suggests possible molecular bases for the differential timing of normal fusion and the response of each suture to different craniosynostosis mutations. One observation of particular interest was the presence of cartilage in unfused lambdoid sutures, suggesting a role for chondrogenesis in posterior skull sutures which have generally been thought to develop by intramembranous ossification without a cartilage precursor. Finally, the effects of common media supplements used in in vitro experiments to stimulate differentiation of calvarial suture-derived cells were investigated with respect to their ability to induce in vivo-like gene expression. The response to standard differentiation medium (ascorbic acid + β-glycerophosphate) with and without dexamethasone was measured by both mineralisation and matrix formation assays and QRT-PCR of genes identified in the above described microarray studies. Both media induced collagen matrix and bone nodule formation indicative of differentiating osteoblasts. However, the genes expression profiles induced by both media differed and neither recapitulated the levels and profiles of gene expression observed in vivo for cells isolated from both fused and unfused suture tissues. This study has implications for translating results from in vitro work to the in vivo situation. Significantly, the dedifferentiation microarray study identified differentially expressed genes whose products may be considered candidates as more appropriate osteogenic supplements that may be used during in vitro experiments to better induce in vivo-like osteoblast differentiation. This study has made a substantial contribution to the identification of novel genes and pathways involved in controlling human suture morphogenesis and craniofacial diversity. The results from this research will stimulate new areas of inquiry which will one day aid in the development of better diagnostics and therapeutics for craniosynostosis, and other craniofacial and more general skeletal abnormalities.