In the mid-gestation embryo, the first definitive transplantable hematopoietic stem cells (dHSCs) emerge by embryonic day E10.5-E11 in the aorta-gonadomesonephros (AGM) region, as a result of a step-wise maturation of precursors called pre-HSCs. The analysis of several Notch mutants suggests that Notch signalling is essential for the execution of the definitive hematopoietic programme in the AGM. Mouse embryos deficient for Notch1, RBP-Jk or Jagged1 cannot efficiently generate intra-embryonic hematopoeitic progenitors. It has also been reported that knockdown of Notch target genes (Hes1, Hes5) results in hematopoietic impairment. However a clear picture of the role of Notch pathway in HSC development is still missing. In this work we characterised precise stages and cell types during HSC development in which Notch signalling is involved. First we used a Hes1-dEGFP reporter mouse line that allowed us to monitor Notch pathway activity in a narrow window of time. The results suggest that the level of Notch activity fluctuates in HSC lineage in the AGM region and is down-regulated in dHSCs in the foetal liver (where dHSCs migrate after generation in the AGM region). By using transplantation assay, we further showed that fluctuations of Notch activity are essential for HSC development, and that this pattern in the HSC lineage might work as a switch between maturation and proliferation of PreHSC1, PreHSC2 and dHSC, in which temporary decrease might be required to mature from one type to another, both in vitro and in vivo. These findings might need to be taken into consideration for in vitro generation of haematopoietic stem cells, where a fine tuning of Notch signalling activity could greatly improve their emergence.

We recently reported that intestinal-specific overexpression of the POZ-ZF transcription factor Kaiso produced two prominent phenotypes in 1-year old mice: Kaiso transgenic (KaisoTg) mice presented with chronic intestinal inflammation, and an increase in secretory cell types – a trait typical of Notch signalling inhibition. Despite these findings however, the factor(s) responsible for Kaiso-mediated inflammation and secretory cell increases had not been elucidated. The primary goal of this thesis was to begin filling in this knowledge gap, by shedding mechanistic insight on Kaiso’s role in governing these two prominent phenotypes. First, we elucidated Kaiso’s role in the Notch signalling pathway and found that Kaiso inhibited the expression of the Notch1 receptor, and its ligand Dll-1, but promoted the expression of the Jagged-1 ligand. We postulated that the Kaiso-mediated reduction in Dll-1 might be responsible for the increase in secretory cell types, whereas Kaiso-mediated regulation of Jagged-1, which is dispensable for cell fate decisions, may be implicated in colon cancer progression. Importantly, we also found that Kaiso’s effects on Notch pathway inhibition occurred prior to the onset of chronic intestinal inflammation. Our analyses of the chronic inflammatory phenotype in KaisoTg mice demonstrated that Kaiso overexpression drives pathogenic neutrophil-specific recruitment (as evidenced by increases in neutrophil-specific enzymatic activity, the formation of crypt abscesses, and augmented expression levels of the neutrophils-specific chemokine, MIP2); an increase in the pore-forming Claudin-2; reduction of the cell adhesion protein E-cadherin; and abnormal intestinal epithelial repair mechanisms. Together, these findings imply that the pathogenesis of Kaiso-mediated intestinal inflammation is a multi-factorial process. A secondary goal of this thesis was to initiate studies to elucidate how the Kaiso binding partner, Znf131, might play a role Kaiso-mediated transcriptional regulation. We found that Znf131 indirectly associated with several Kaiso target genes, including Cyclin D1 (CCND1). Importantly, Znf131 activated a minimal CCND1 promoter previously shown to be inhibited by Kaiso. Moreover, Kaiso overexpression attenuated Znf131-mediated transcriptional activation and Znf131 expression in intestinal cells. Together, these findings hint that Znf131 and Kaiso may exert opposing biological functions, which may have implications in Kaiso-mediated intestinal homeostasis and disease. / Thesis / Doctor of Philosophy (PhD)

Intrahepatic cholangiocarcinoma (ICC) is an aggressive malignancy with a dismal prognosis. Few patients present with disease amenable to resection and chemotherapy is not curative. The incidence of ICC is rising worldwide and new therapeutic approaches are urgently required. Notch signalling is critical for the embryological development and regeneration of the biliary tree in the mammalian liver. Dysregulation of Notch is known to drive tumorigenesis in a range of solid and haematological malignancies and the aim of this work was to define its contribution to the pathogenesis of ICC. Transgenic overexpression of Notch1 has been described to result in the formation of biliary lineage tumours in the liver. I have used resected human tissue, a chemically-induced model of ICC in rat and a novel transgenic murine model in which the tumour suppressor p53 is conditionally deleted from biliary epithelia, to demonstrate that endogenous Notch signalling is acting via the Notch3 receptor to drive tumorigenesis. I use multiple independent methods of Notch3 blockade to establish that Notch3 promotes epithelial cell survival and self-renewal in ICC and demonstrate that Notch3 inhibition significantly attenuates tumour growth in vivo. My data suggest that Notch3 promotes activity through the PI3K/AKT cell survival cascade via a mechanism independent of the effector of canonical Notch, RBPJκ. Given the significant toxicity associated with gamma-secretase inhibitors these findings offer a novel and specific target for further investigation and future therapeutic development in ICC.

<p> The Notch signalling pathway is a lateral inhibition pathway that serves to limit the number of cells in a proneural cluster (a group of equipotent cells) that will adopt a neural cell fate during neurogenesis in Drosophila. The proper segregation of neural and epidermal progenitor cells during neurogenesis requires the expression of both the proneural genes and the neurogenic genes. Expression of proneural genes, such as achaete, gives cells the potential to commit to a neural cell fate. The neurogenic genes encode proteins that act in the Notch signalling cascade and are required for cell fate determination during Drosophila neurogenests. Notch and Delta are neurogenic genes that encode large transmembrane proteins. Interaction between the extracellular domains of Notch and Delta is thought to transmit a signal to the nucleus by way of the DNAbinding Suppressor of Hairless protein. In response to Notch activation Suppressor of Hairless is translocated to the nucleus where it activates the transcription ofthe neurogenic genes ofthe Enhancer of split complex (E(spl)-C). The products of the E(spl)-C are bHLH transcription factors. They possess a Cterminal tryptophan-arginine-proline-tryptophan (WRPW) motif that interacts with the product of another neurogenic gene, groucho. The groucho gene product encodes a protein containing a WD40 repeat element. When bound to Groucho, E(spl) bHLH proteins are able to repress transcription of proneural genes, such as achaete, thereby directing the cell to adopt a non-neural cell fate.</p> <p> A number of murine groucho homologues have been identified and named Grg's (Groucho related genes). Three full length Grg proteins have been identified which contain all five domains found in the Drosophila Groucho protein. Two short Grg proteins have also been identified which only contain one of the domains found in the full-length Grg proteins. A number of murine homologues of the Drosophila E(spl)-C have also been identified and named Hes (Hairy/Enhancer of split) proteins. Like the gene products of the Drosophila E(spl)-C, the Hes proteins are bHLH proteins containing a C-terminal WRPW motif. One of the Hes proteins, Hes3, is lacking a basic domain and therefore lacks the DNA-binding activity possessed by the other Hes proteins. </p> <p> Attempts were made to detect interactions between Grg and Hes proteins using co-immunoprecipitation techniques. The anti-WD40 antibody, which recognizes the long WD40-containing Grg proteins, was able to specifically immunoprecipitate 35S-labelled Grgl . This antibody was also able to recognize WD40-containing Grg proteins present in Pl9 cell extracts. However, attempts to co-immunoprecipitate radiolabelled Hesl and AMLlb proteins with Grg proteins present in P19 cell extract were unsuccessful due to the low affinity of the antiWD40 antibody and the background caused by the binding of the test proteins to Sepharose. A second method of co-immunoprecipitation was attempted using an HA-tagged Grgl fusion protein and a commercially available anti-HA antibody. The attempt to co-immunoprecipitate 35S-labelled Hesl with radiolabelled HAtagged Grg 1 was unsuccessful due to a high degree of background caused by Hesl binding to protein G Agarose. Using the Yeast Two-Hybrid interaction assay, the WD40-containing Grg proteins, Grgl and Grg4, were found to interact with Hesl. However, using the same assay WD40-containing Grg proteins were found not to interact with Hes3, which lacks DNA-binding activity. A Western blot was performed to determine if the Hes3 fusion proteins were being expressed in transformed yeast but none were detected. This may have been due to the poor affinity of the anti-GAL4 activation domain antibody. A similar Western blot demonstrated that the Grg proteins, fused to the GAL4 DNA binding domain, were being expressed in transformed yeast extract. The WD40-containing Grg proteins, Grgl and Grg4, were also found not to interact with AMLlb, a protein which contains a C-terminal VWRPY domain which is reminiscent of the Cterminal WRPW interaction domain found in Hes proteins and Drosophila E(spl) proteins. However, WD40-containing Grg proteins were able to interact with an AML 1 b mutant in which the VWRPY motif was mutated to VWRPW in the Yeast Two Hybrid assay. </p> / Thesis / Master of Science (MSc)

Thymic epithelial cells (TECs) are indispensable for the development of T cells in the thymus. Two subtypes of TECs exist in the thymus, medullary mTECs and cortical cTECs. Both mTECs and cTECs originate from endodermal thymic epithelial progenitor cells (TEPCs) in the embryo, but how the differentiation of TEPCs is regulated is not well understood. The aims of this thesis were to establish the role of Notch signalling in TEPC differentiation, and how it interacts with known regulators such as FOXN1 and the NFκB pathway. Gene expression data showed that Notch is active in TEPCs and exhibits a correlation with the mTEC lineage. Loss of Notch function led to a significant reduction in the number of mTECs in the thymus, and this can be attributed to aberrant mTEC specification. Furthermore, the duration of Notch activity in determining mTEC number appears limited to the early phase of organogenesis, and precedes RANK/NFκB mediated mTEC proliferation. Gain of Notch function resulted in a considerable shift to a primitive, TEPC-like phenotype, and subsequently a latent increase in mTEC frequency. Finally, transcriptomic and functional analyses pointed to a cross-repressive mechanism between Notch and FOXN1 in TEPCs. Taken together, these results identified Notch as a novel regulator of mTEC specification, likely through maintaining the potency of fetal TEPCs, a prerequisite for mTEC lineage commitment.

Le cœur des vertébrés se développe à partir du tube cardiaque et de la participation des cellules progénitrices mésodermiques du second champ cardiaque (SHF). Une perturbation de l’addition des cellules du SHF conduit à des malformations cardiaques congénitales (MCC). Chez l’embryon, l’outflow tract (OFT) dérivé du seul SHF est formé par deux domaines complémentaires qui formeront le myocarde sous-aortique et sous-pulmonaire. Ce travail analyse les cellules progénitrices du SHF qui contribuent aux deux domaines de l’OFT pour former la base de l’aorte et du tronc pulmonaire, l’identité transcriptionnelle des domaines et leur régulation. Nous avons mis en évidence une sous-population de cellules progénitrices Notch-dépendantes, situées en région antérieure du mésoderme pharyngé, qui contribue au myocarde sous-aortique. Nous avons démontré que des cascades de régulation croisées impliquant Notch/Hes1 et Tbx1/Pparg sont importantes pour former les deux domaines fonctionnels régionalisés de l’OFT. Des expériences de culture d’explants et d’embryons ont démontré que Pparg est nécessaire au déploiement des cellules du SHF et pour la régulation transcriptionnelle du futur myocarde sous-pulmonaire. Dans le domaine complémentaire, futur myocarde sous-aortique, nous avons observé l’expression de Dlk1, un régulateur négatif de Pparg. Dlk1 est en amont de la voie de régulation Notch et participe probablement à l’identité régionale de l’OFT. Dans son ensemble, ce travail identifie de nouvelles voies de signalisation et gènes qui régulent l'identité régionale du mésoderme cardio-pharyngé et de nouvelles cibles pour l’étude clinique des MCC. / The vertebrate heart develops from the heart tube and the contribution of mesodermal progenitors termed second heart field (SHF). Perturbation in SHF addition leads to congenital heart defects (CHD). The outflow tract (OFT) myocardium is entirely derived from the SHF. Distinct regions of the embryonic OFT have been shown to give rise to subaortic and subpulmonary myocardium of the heart. The work described here focuses on SHF progenitor subpopulations in mouse giving rise to distinct OFT domains and characterizes the regional transcriptional identity and regulation of future subaortic and subpulmonary myocardium. We identified Notch-dependent subaortic myocardial SHF progenitors in anterior pharyngeal mesoderm. We demonstrated that Notch/Hes1 and Tbx1/Pparg cross regulatory cascades are important to establish functionally important OFT regional domains. Explant and embryo culture experiments revealed that Pparg is required for both the deployment of SHF cells and transcriptional regulation of the future subpulmonary myocardial domain. We also found that Dlk1, a negative regulator of Pparg, is expressed in the complementary subaortic domain upstream of Notch receptor activation and potentially participates in the establishment of OFT regional identity. We also report an overlapping transcriptional profile between future subaortic myocardium and subpopulation of epicardial cells at fetal stages. Finally, we provide evidence for the existence of conserved bipotential myogenic progenitors in cardiopharyngeal mesoderm coexpressing Nkx2-5 and Tbx1. Overall this work identifies novel pathways and genes in cardiopharyngeal mesoderm that may contribute to clinically relevant CHD.

T lymphocytes are considered an essential and advanced component of the immune system, since these cells are able to discriminate self from non-self, start up an immune reaction and further develop into memory cells. However, therapies based on the use of patient derived newly generated T cells reinoculated into humans do not exist. This is due to difficulties in replicating the peculiar conditions required for T cell development in vitro. The systems developed so far are based on the use of animal or unrelated human thymic tissue and therefore they would not be adequate to be used in any clinical application. Having conjectured that human skin cells, rearranged in a threedimensional fashion, would be able to support the development of human T lymphocytes from hematopoietic stem cells, we developed a model consisting of human skin keratinocytes and fibroblasts arrayed on a synthetic matrix so to create a prototype suitable to be translated into the clinic. In this way we were able to induce few hundred cord blood CD34⁺ haematopoietic stem cells to entirely develop into mature CD4⁺ or CD8⁺ T lymphocytes in vitro. However, circulating adult peripheral CD34⁺ precursors failed to survive in the same conditions. Finally we were able to explain our success as consequence of strong induction of the Notch delta ligand Dll-4 by the keratinocytes cultured in the construct. In synthesis, we report here for the first time that skin keratinocytes, in the presence of fibroblasts and reconfigured in a three-dimensional arrangement, are able to induce the differentiation of a minimal amount of cord but not adult blood stem cells into fully differentiated T cells by acting through the Dll-4 Notch signaling pathway in vitro.

<p>The Notch signalling pathway is an evolutionary conserved pathway, named after the Notch receptors, Notch1-4 in mammals, which upon cell-cell contact and ligand binding releases the intracellular domain (NICD). NICD translocates into the nucleus where it binds the transcriptional repressor RBP-Jk, which together with co-activators belonging to the Mastermind-like family of proteins form a transcriptional activation complex. This complex activates genes controlling cell fate decision, embryonic development, proliferation, differentiation, adult homeostasis and stem cell maintenance. On the other hand, disrupted Notch signalling may result in pathological conditions like cancer, although the mechanisms behind the disruption are often complex and in many cases largely unknown.</p><p>Notch1 drives the lymphocyte differentiation towards a T-cell fate and activating mutations in the gene have been suggested to be involved in T-cell lymphoma. In <em>paper I, </em>genetic alterations in <em>Notch1 </em>and the Notch1 regulating gene <em>CDC4 </em>were investigated in tumours from murine T-cell lymphoma induced with phenolphthalein, 1,3-butadiene or 2’,3’-dideoxycytidine. We identified activating <em>Notch1</em> mutations in 39% of the lymphomas, suggesting that <em>Notch1 </em>is<em> </em>an important target gene for mutations in chemically induced lymphomas.<em></em></p><p>While it is known that constitutively activated Notch signalling has a clear oncogenic function in several solid malignancies as well, the molecular mechanisms are less known in this context. Unpublished data of our lab, together with other recent studies, suggest that mutations of Notch and Notch-related genes <em>per se</em> are uncommon in solid malignancies including colorectal cancer, while a growing body of evidence indicates that aberrant Wnt/b-catenin signalling may result in pro-tumoural Notch activation in these contexts. In <em>paper II</em>, we therefore investigated potential transcriptional interactions between the Notch and Wnt signalling pathways in colorectal cancer cell lines. The proximal Notch and Wnt pathway gene promoters were bioinformatically identified and screened for putative TCF/LEF1 and RBP-Jk sites. In canonical Wnt signalling, Apc negatively regulates b-catenin leading to repression of TCF/LEF1 target genes. Upon repression of the Wnt pathway we observed that several genes in the Notch pathway, including <em>Notch2</em>, were transcriptionally downregulated. We also confirmed binding of Lef1 to <em>Notch2</em> as well as other Notch pathway gene promoters and luciferase assays showed an increased activity for at least one LEF1/TCF-site in the <em>Notch2</em> promoter upon co-transfection of HT29 or HCT116 cells with mutated b-catenin. HT29 cell lines were also treated with the g-secretase inhibitor DAPT, leading to inactivation of the Notch pathway by preventing release of NICD. However, results showed no effects on Apc, b-catenin or their target <em>cyclin D1</em>. Taken together, these results indicate that the Wnt pathway may function as a regulator of the Notch pathway through the TCF/LEF1 target gene program in colon cancer cell lines.</p><p>In summary, Notch pathway deregulation is of importance in both murine T-cell lymphoma and human colorectal cancer, although the mechanisms differ. The current results give new insights in Notch pathway alterations as well as the signalling networks in which the Notch pathway interacts, and thus increase the understanding of Notch’s involvement in malignant diseases.</p> / Studies on molecular genetic alterations in colorectal cancer

The aim of this thesis was to examine role of Sirt1 in the Notch signalling pathway, using Drosophila as a model organism. Based on in vivo and in vitro studies, we conclude that Sirt1 plays a positive role in Notch signalling. In embryonic S2N cells, Sirt1 is responsible for the protection from metabolic stress-induced down-regulation of subset of E(Spl) genes. During development, Sirt1 is responsible for proper Notch-dependent specification of SOPs and wing development. Sirt1 can regulate the Notch signalling on multiple levels via deacetylation of various substrates involved in the Notch signalling revealed by our proteomic survey.

The Notch signalling pathway is an evolutionary conserved pathway, named after the Notch receptors, Notch1-4 in mammals, which upon cell-cell contact and ligand binding releases the intracellular domain (NICD). NICD translocates into the nucleus where it binds the transcriptional repressor RBP-Jk, which together with co-activators belonging to the Mastermind-like family of proteins form a transcriptional activation complex. This complex activates genes controlling cell fate decision, embryonic development, proliferation, differentiation, adult homeostasis and stem cell maintenance. On the other hand, disrupted Notch signalling may result in pathological conditions like cancer, although the mechanisms behind the disruption are often complex and in many cases largely unknown. Notch1 drives the lymphocyte differentiation towards a T-cell fate and activating mutations in the gene have been suggested to be involved in T-cell lymphoma. In paper I, genetic alterations in Notch1 and the Notch1 regulating gene CDC4 were investigated in tumours from murine T-cell lymphoma induced with phenolphthalein, 1,3-butadiene or 2’,3’-dideoxycytidine. We identified activating Notch1 mutations in 39% of the lymphomas, suggesting that Notch1 is an important target gene for mutations in chemically induced lymphomas. While it is known that constitutively activated Notch signalling has a clear oncogenic function in several solid malignancies as well, the molecular mechanisms are less known in this context. Unpublished data of our lab, together with other recent studies, suggest that mutations of Notch and Notch-related genes per se are uncommon in solid malignancies including colorectal cancer, while a growing body of evidence indicates that aberrant Wnt/b-catenin signalling may result in pro-tumoural Notch activation in these contexts. In paper II, we therefore investigated potential transcriptional interactions between the Notch and Wnt signalling pathways in colorectal cancer cell lines. The proximal Notch and Wnt pathway gene promoters were bioinformatically identified and screened for putative TCF/LEF1 and RBP-Jk sites. In canonical Wnt signalling, Apc negatively regulates b-catenin leading to repression of TCF/LEF1 target genes. Upon repression of the Wnt pathway we observed that several genes in the Notch pathway, including Notch2, were transcriptionally downregulated. We also confirmed binding of Lef1 to Notch2 as well as other Notch pathway gene promoters and luciferase assays showed an increased activity for at least one LEF1/TCF-site in the Notch2 promoter upon co-transfection of HT29 or HCT116 cells with mutated b-catenin. HT29 cell lines were also treated with the g-secretase inhibitor DAPT, leading to inactivation of the Notch pathway by preventing release of NICD. However, results showed no effects on Apc, b-catenin or their target cyclin D1. Taken together, these results indicate that the Wnt pathway may function as a regulator of the Notch pathway through the TCF/LEF1 target gene program in colon cancer cell lines. In summary, Notch pathway deregulation is of importance in both murine T-cell lymphoma and human colorectal cancer, although the mechanisms differ. The current results give new insights in Notch pathway alterations as well as the signalling networks in which the Notch pathway interacts, and thus increase the understanding of Notch’s involvement in malignant diseases. / Studies on molecular genetic alterations in colorectal cancer