Role of Notch signaling in epidermis and appendages

Résumé Dans la peau, il a été montré que Notch1 induit l'arrêt de la prolifération et la différentiation des keratinocytes. L'inactivation de Notch1 cause une hyperplasie de l'épiderme et la formation de carcinomes basaux cellulaires. Notre groupe a principalement identifié deux voies de signalisations, la voie Shh et la voie Wnt, qui sont dérégulées en conséquence de l'inactivation de Notch1 dans la peau. Nous avons démontré l'habilité de Notch1 à réprimer la voie Wnt induite par ß-catenin dans les keratinocytes primaires ainsi que dans d'autres types de cellules épithéliales humaines. De plus, nous avons pu déterminer que Notch1 régule cette voie, probablement en favorisant la phosphorylation de ß-catenin par le complexe axin/APC/GSK-3ß. La protéine faisant partie de la voie Wnt, ou la protéine affectant la voie Wnt, qui est régulée par Notch1 est sujette à de plus amples investigations. Un autre but de cette étude a été l'identification de potentiels gènes cibles de Notch1 autres que ceux faisant partie des voies de signalisation Shh et Wnt précédemment évoquées. Ce projet fut abordé par l'analyse de puces à ADN (ISREC et Affymetrix) qui ont été utilisées pour des expériences de gain et de perte de fonction de Notch1 dans des keratinocytes prúmaires. En plus de l'hyperplasie épidermale, les souris Notch1 déficiente ont une perte importante de poils. Nous avons montré que Notch1 est nécessaire pour le développement et l'homéostasie des follicules pileux. En effet, l'inactivation du gène Notch1 mediée par l'activation des kératines 5 ou 14 dans l'épiderme, cause des défauts du cycle ainsi que de la structure des poils. De plus, d'autres appendices de la peau, comme les glandes sudoripares et de Meibomius, ont une structure anormale et sont non fonctionnelles dans les souris Notch1 déficiente. Finalement, nous avons observé que la déficience de Notch1 dans l'épithélium cornéen mène à la formation d'une plaque épidermale opaque sur la cornée. Basé sur l'hypothèse que le défaut des glandes de Meibomius des souris Notch1 déficientes cause des lésions de la surface oculaire, nous avons montré que Notch1 est essentiel pour la cicatrisation de la cornée. Lorsque Notch1 est absent, les cellules souches de l'épithélium cornéen ne sont plus capables de se différentier en cellules cornéennes, mais réparent la blessure en se différentiant en épiderme. Ce résultat indique que Notch1 est essentiel pour la différentiation de cellules souches de la cornée qui sont spécifiquement impliquées dans la réparation de la cornée. De plus, nous avons montré que l'expression de CRBP1 dans l'épithélium cornéen est diminuée en l'absence de Notch1, ceci étant possiblement à l'origine de la formation de la plaque épidermale. Abstract: In the skin, Notch1 has been shown to trigger cell growth arrest and differentiation of keratinocytes. Notch1 inactivation results in epidermal hyperplasia and subsequent formation of basal cell carcinoma-like (BCC-like) tumors. So far our group has identified two main pathways, the Shh and the Wnt pathway, that are deregulated as a consequence of Notch1 inactivation in the skin. We showed the ability of Notch1 to represses ß-catenin-mediated Wnt signaling in primary keratinocytes as well as in other types of human epithelial cells. In addition we were able to determine that Notch1 regulates this pathway possibly by enhancing ß-catenin phosphorylation by the axin/APC/GSK-3ß complex. The exact target protein of the Wnt pathway or target protein that affects the Wnt pathway, and that is regulated by Notch1, is subject of current investigation. Another aim of this study was the identification of possible Notch1 target genes in addition to those of the Shh and Wnt signaling pathways. This was addressed by gene chip analysis using ISREC as well as Affymetrix microarrays for gain and loss of function of Notch1 in mouse primary keratinocytes. In addition to epidermal hyperplasia, Notch1 deficient mice show an important hair loss. We showed that Notch1 is required for postnatal development and homeostasis of hair follicles. Indeed, keratin5 or keratinl4-driven Cre recombinase-mediated inactivation of the Notch1 gene in the epidermis causes perturbations of the hair cycle and structural defects of the hair follicle. Moreover, other skin appendages, like the sweat and Meibomian glands show abnormal morphology and are not functional in the Notch 1 deficient mice. Finally, we observed that Notch1 deficiency in the corneal epithelium leads to the formation of an epidermal corneal plaque. Based on the hypothesis that the Meiboinian gland defect in the Notch1 deficient mice results in lesions of the eye surface, we showed that Notch1 is essential for wound-healing of the cornea. In absence of Notch1 the stem cells of the corneal epithelium are no longer able to differentiate in the corneal fate but instead repair the wound by differentiating into skin-like epidermis. This result indicated that Notch1 is essential for the differentiation of corneal stem cells specifically implicated in corneal wound-healing. Moreover, we showed that CRBP1 expression in the corneal epithelium was lost in the absence of Notch1, possibly being at the origin of plaque formation.

The tumor suppressive role of WIF1 in glioblastoma

Expression based prediction of gene alterations identified WNT inhibitory factor I (WIF1) as a new candidate tumor suppressor gene involved in glioblastoma. WIF1 encodes a secreted WNT antagonist and it is strongly down-regulated in most glioblastoma as compared to normal brain both by genomic deletion and WIF1 promoter hypermethylation. WIF1 expression in glioblastoma cell lines inhibited cell proliferation in vitro and in vivo and strongly reduced migration capability. Interestingly, WIF1 expression induced a senescence-like phenotype characterized by the appearance of enlarged, flattened and multinucleated cells positive for the presence of senescence associated ß-galactosidase, a late marker of senescence. It is of note that WIF1 induced senescence, in glioma cell lines, is independent of either p53 or pRB, two pathways that have been widely associated with this process. The analysis of the signaling pathways downstream of WIF1 brought some interesting results. WIF1 expression inhibited the canonical pathway but alteration of this pathway alone couldn't explain all the WIFl-induced effects. Some WIF1-related changes were attributed to inhibition of the non-canonical pathway, as we could prove by downregulation of WNT5a, the main ligand of the non-canonical WNT pathway. For example, a drastic reduction of phosphorylation of both ERK and p38 was detected when either overexpressing WIF1 or downregulating WNT5a. Due to the complexity of the non-canonical pathway is difficult to define the precise mechanism of signal transduction. We have excluded the involvement of the WNT5a-JNK-APl pathway and preliminary results suggest the implication of the WNT-calcium signaling, but further evidence is needed. Moreover, from the analysis of the gene expression profile of WIF1 expressing cells we could select a very interesting candidate: MALATI, a non-coding RNA widely associated with migratory capability in many different types of tumors. We found MALATI to be overexpressed in glioblastoma specimens compared to normal brain and to be associated with total tumor volume. The downregulation of MALATI by RNAi (RNA interference] drastically impairs migration, thus it is a very interesting potential target in the context of invasive tumors such as glioblastoma. Résumé WIFl a été sélectionné en tant que putatif suppresseur de tumeurs dans le cadre des glioblastomes par une analyse qui a était conduit à partir des données d'expression de gènes provenant d'environ 80 glioblastomes. WIF1 code pour une protéine destinée à la sécrétion qui antagonise la voie de WNT et son expression est fortement sous-exprimé dans la plupart des glioblastome par rapport à tissu cérébral normal. Cette sous-expression est due à deux mécanismes différents: à la délétion de la partie génomique codant pour WIF1 et à l'hyper méthylation de son promoteur. La surexpression de WIF1 réduit la capacité de prolifération des cellules de glioblastome in vitro ainsi que in vivo et elle réduit aussi leur capacité migratoire. Il est intéressant de remarquer que l'espression de WIF1 induit un phénotype sénescent caractérisé par l'apparition de cellules aplaties, multi nucléées et positives pour l'activité de l'enzyme ß-galactosidase associée à la sénescence, un marqueur tardif de la sénescence. Il est à noter que le phénotype sénescent qui est induit par WIF1 est indépendant de p53 et pRB, deux voies qui ont été largement associées à ce processus. L'analyse des les voies de signalisation en aval de WIFl a apporté des résultats intéressants. L'expression de WIF1 inhibe la voie canonique de WNT, mais l'altération de cette voie seule ne pouvait pas expliquer tous les effets induits par WIF1. Nous avons pu prouver que certains changements sont liés à l'inhibition de la voie non-canonique qui est activée par WNT5cc. Par exemple, une réduction drastique de la phosphorylation de ERK et p38 à la fois a été détectée lorsque WIFl a été surexprimé ou WNT5a sous- exprimé. En raison de la complexité de la voie non-canonique, il est difficile de définir le mécanisme précis de la transduction du signal. Nous avons exclu l'implication de la voie JNK-WNT5a-APl et les résultats préliminaires suggèrent l'implication de la voie de signalisation appelée WNT-calcium. En plus, l'analyse du profil d'expression génique de cellules sur-exprimant WIF1 nous a permis d'identifier un candidat très intéressant: MALATI, un ARN non- codants largement associés à la capacité migratoire dans nombreux types de tumeurs. Nous avons trouvé que MALATI est surexprimé dans les échantillons de glioblastome par rapport à tissu cérébral normal et il est associé au volume total de la tumeur. La sous-expression de MALATI altère considérablement la migration des cellules tumorales. Donc, MALATI, est une cible potentielle très intéressante dans le cadre d'une tumeur invasive telle que le glioblastome.

The role of WNT and mTOR in human memory T cell formation

Cancer is one of the world's leading causes of death with a rising trend in incidence. These epidemiologic observations underline the need for novel treatment strategies. In this regard, a promising approach takes advantage of the adaptive effector mechanisms of the immune system, using T lymphocytes to specifically target and destroy tumour cells. However, whereas current approaches mainly depend on short-lived, terminally differentiated effector T cells, increasing evidence suggests that long lasting and maximum efficient immune responses are mediated by low differentiated memory T cells. These memory T cells should display characteristics of stem cells, such as longevity, self-renewal capacity and the ability to continuously give rise to further differentiated effectors. These stem celllike memory T (TSCM) cells are thought to be of key therapeutic value as they might not only attack differentiated tumour cells, but also eradicate the root cause of cancer, the cancer stem cells themselves. Thus, efforts are made to characterize TSCM cells and to identify the signalling pathways which mediate their induction. Recently, a human TSCM cell subset was described and the activation of the Wnt-ß-catenin signalling pathway by the drug TWS119 during naive CD8+ T (TN) cell priming was suggested to mediate their induction. However, a precise deciphering of the signalling pathways leading to TSCM cell induction and an in-depth characterization of in vitro induced and in vivo occurring TSCM cells remain to be performed. Here, evidence is presented that the induction of human and mouse CD8+ and CD4+ TSCM cells may be triggered by inhibition of mechanistic/mammalian target of rapamycin (mTOR) complex 1 with simultaneously active mTOR complex 2. This molecular mechanism arrests a fraction of activated TN cells in a stem cell-like differentiation state independently of the Wnt-ß-catenin signalling pathway. Of note, TWS119 was found to also inhibit mTORCl, thereby mediating the induction of TSCM cells. Suggesting an immunostimulatory effect, the acquired data broaden the therapeutic range of mTORCl inhibitors like rapamycin, which are, at present, exclusively used due to their immunosuppressive function. Furthermore, by performing broad metabolic analyses, a well-orchestrated interplay between intracellular signalling pathways and the T cells' metabolic programmes could be identified as important regulator of the T cells' differentiation fate. Moreover, in vitro induced CD4+ TSCM cells possess superior functional capacities and share fate-determining key factors with their naturally occurring counterparts, assessed by a first-time full transcriptome analysis of in vivo occurring CD4+ TN cell, TSCM cells and central memory (TCM) cells and in vitro induced CD4+ TSCM cells. Of interest, a group of 56 genes, with a unique expression profile in TSCM cells could be identified. Thus, a pharmacological mechanism allowing to confer sternness to activated TN cells has been found which might be highly relevant for the design of novel T cell-based cancer immunotherapies.

Increased Wnt signaling triggers oncogenic conversion of human breast epithelial cells by a Notch-dependent mechanism.

Wnt and Notch signaling have long been established as strongly oncogenic in the mouse mammary gland. Aberrant expression of several Wnts and other components of this pathway in human breast carcinomas has been reported, but evidence for a causative role in the human disease has been missing. Here we report that increased Wnt signaling, as achieved by ectopic expression of Wnt-1, triggers the DNA damage response (DDR) and an ensuing cascade of events resulting in tumorigenic conversion of primary human mammary epithelial cells. Wnt-1-transformed cells have high telomerase activity and compromised p53 and Rb function, grow as spheres in suspension, and in mice form tumors that closely resemble medullary carcinomas of the breast. Notch signaling is up-regulated through a mechanism involving increased expression of the Notch ligands Dll1, Dll3, and Dll4 and is required for expression of the tumorigenic phenotype. Increased Notch signaling in primary human mammary epithelial cells is sufficient to reproduce some aspects of Wnt-induced transformation. The relevance of these findings for human breast cancer is supported by the fact that expression of Wnt-1 and Wnt-4 and of established Wnt target genes, such as Axin-2 and Lef-1, as well as the Notch ligands, such as Dll3 and Dll4, is up-regulated in human breast carcinomas.

Role of β-catenin and γ-catenin in haematopoiesis : Generation of floxed Delta like 4 gene targeted mice

Résumé La voie de signalisation de Wnt est extrêmement conservée au cours de l'évolution. Les protéines Wnt sont des molécules sécrétées qui se lient à la famille de récepteurs Frizzled. Cette interaction mène à la stabilisation de la protéine β-caténine, qui va s'accumuler dans le cytoplasme puis migrer dans le noyau où elle peut s'hétérodimériser avec les facteurs de transcription de la famille TCF/LEF. Il a été démontré que cette voie de signalisation joue un rôle important durant la lymphopoïèse et de récents résultats suggèrent un rôle clé de cette voie dans le renouvellement des Cellules Souches Hématopoïétique (CSH). Des études se basant sur un système de surexpression de protéines montrent clairement que la voie Wnt peut influencer l'hématopoïèse. Cependant, le rôle de la protéine β-caténine dans le système hématopoïétique n'a jamais été testé directement. Ce projet de thèse se propose d'étudier la fonction de la protéine β-caténine par sa délétion inductible via le système Cre-loxP. De façon surprenante, nous avons pu démontrer que les progéniteurs de la moelle osseuse, déficients en β-caténine, ne montrent aucune altération dans leur capacité à s'auto-renouveler et/ou à reconstituer toutes les lignées hématopoïétiques (myéloïde, érythroïde et lymphoïde) dans les souris-chimères. De plus, le développement, la survie des thymocytes ainsi que la prolifération des cellules T périphériques induite par un antigène, sont indépendants de β-caténine. Ces résultats suggèrent soit que la protéine β-caténine ne joue pas un rôle primordial dans le système hématopoiétique, soit que son absence pourrait être compensée par une autre protéine. Un candidat privilégié susceptible de se substituer à β-caténine, serait plakoglobine, aussi connu sous le nom de γ-caténine. En effet, ces deux protéines partagent de multiples caractéristiques structurelles. Afin de démontrer que la protéine γ-caténine peut compenser l'absence de β-caténine, nous avons généré des souris dans lesquelles, le système hématopoïétique est déficient pour ces deux protéines. Cette déficience combinée de β- caténine et γ-caténine ne perturbe pas la capacité des Cellules Souche Hématopoïétique-Long Terme (CSH-LT) de se renouveler, par contre elle agit sur un progéniteur précoce déjà différencié de la moelle osseuse. Ces résultats mettent en évidence que la protéine γ-caténine est capable de compenser l'absence de protéine β-caténine dans le système hématopoïétique. Par conséquent, ce travail contribue à une meilleure connaissance de la cascade Wnt dans l'hématopoïèse. Summary The canonical Wnt signal transduction pathway is a developmentally highly conserved. Wnts are secreted molecules which bind to the family of Frizzled receptors in a complex with the low density lipoprotein receptor related protein (LRP-5/6). This initial activation step leads to the stabilization and accumulation of β-catenin, first in the cytoplasm and subsequently in the nucleus where it forms heterodimers with TCF/LEF transcription factor family members. Wnt signalling has been shown to be important during early lymphopoiesis and has more recently, been suggested to be a key player in self-renewal of haematopoietic stem cells (HSCs). Although mostly gain of function studies indicate that components of the Wnt signalling pathway can influence the haematopoietic system, the role of β-catenin has never been directly investigated. The aim of this thesis project is to investigate the putatively critical role of β-catenin in vivo using the Cre-loxP mediated conditional loss of function approach. Surprisingly, β-catenin deficient bone marrow (BM) progenitors arc not impaired in their ability to self-renew and/or to reconstitute all haematopoietic lineages (myeloid, erythroid and lymphoid) in both mixed and straight bone marrow chimeras. In addition, both thymocyte development and survival, and antigen-induced proliferation of peripheral T cells are β- catenin independent. Our results do not necessarily exclude the possibility of an important function for β-catenin mediated Wnt signalling in the haematopoietic system, it rather raises the question that β-catenin is compensated for by another protein. A prime candidate that may take over the function of β-catenin in its absence, is the close relative plakoglobin, also know as γ-catenin. This protein shares multiple structural features with β-catenin. In order to investigate whether γ-catenin can compensate for the loss of β-catenin we have generated mice in which the haematopoietic compartment is deficient for both proteins. Combined deficiency of β-catenin and γ-catenin does not perturb Long Term-Haematopoietic Stem Cells (LT-HSC) self renewal, but affects an already lineage committed progenitor population within the BM. Our results demonstrate that y-catenin can indeed compensate for the loss of β-catenin within the haematopoietie system.

Yellow submarine of the Wnt/Frizzled signaling: submerging from the G protein harbor to the targets.

The Wnt/Frizzled signaling pathway plays multiple functions in animal development and, when deregulated, in human disease. The G-protein coupled receptor (GPCR) Frizzled and its cognate heterotrimeric Gi/o proteins initiate the intracellular signaling cascades resulting in cell fate determination and polarization. In this review, we summarize the knowledge on the ligand recognition, biochemistry, modifications and interacting partners of the Frizzled proteins viewed as GPCRs. We also discuss the effectors of the heterotrimeric Go protein in Frizzled signaling. One group of these effectors is represented by small GTPases of the Rab family, which amplify the initial Wnt/Frizzled signal. Another effector is the negative regulator of Wnt signaling Axin, which becomes deactivated in response to Go action. The discovery of the GPCR properties of Frizzled receptors not only provides mechanistic understanding to their signaling pathways, but also paves new avenues for the drug discovery efforts.

BCL9 is an essential component of canonical Wnt signaling that mediates the differentiation of myogenic progenitors during muscle regeneration.

Muscle stem cells and their progeny play a fundamental role in the regeneration of adult skeletal muscle. We have previously shown that activation of the canonical Wnt/beta-catenin signaling pathway in adult myogenic progenitors is required for their transition from rapidly dividing transient amplifying cells to more differentiated progenitors. Whereas Wnt signaling in Drosophila is dependent on the presence of the co-regulator Legless, previous studies of the mammalian ortholog of Legless, BCL9 (and its homolog, BCL9-2), have not revealed an essential role of these proteins in Wnt signaling in specific tissues during development. Using Cre-lox technology to delete BCL9 and BCL9-2 in the myogenic lineage in vivo and RNAi technology to knockdown the protein levels in vitro, we show that BCL9 is required for activation of the Wnt/beta-catenin cascade in adult mammalian myogenic progenitors. We observed that the nuclear localization of beta-catenin and downstream TCF/LEF-mediated transcription, which are normally observed in myogenic progenitors upon addition of exogenous Wnt and during muscle regeneration, were abrogated when BCL9/9-2 levels were reduced. Furthermore, reductions of BCL9/9-2 inhibited the promotion of myogenic differentiation by Wnt and the normal regenerative response of skeletal muscle. These results suggest a critical role of BCL9/9-2 in the Wnt-mediated regulation of adult, as opposed to embryonic, myogenic progenitors.

Platforms for high-throughput screening of Wnt/Frizzled antagonists.

Signaling cascades initiated by Wnt lipoglycoproteins and their receptors of the Frizzled family regulate many aspects of animal development and physiology. Improper activation of this signaling promotes carcinogenic transformation and metastasis. Development of agents blocking the Wnt-Frizzled signaling is of prime interest for drug discovery. Despite certain progress no such agents are as yet brought to the market or even to clinical trials. One reason for these delays might be the use of suboptimal readout assays. In this article we overview existing and developing assay platforms to screen for Wnt-Frizzled antagonists. Among those, G protein-activating assays built on the emerging GPCR properties of Frizzleds are highlighted.

PROX1 Promotes Metabolic Adaptation and Fuels Outgrowth of Wnt(high) Metastatic Colon Cancer Cells.

The Wnt pathway is abnormally activated in the majority of colorectal cancers, and significant knowledge has been gained in understanding its role in tumor initiation. However, the mechanisms of metastatic outgrowth in colorectal cancer remain a major challenge. We report that autophagy-dependent metabolic adaptation and survival of metastatic colorectal cancer cells is regulated by the target of oncogenic Wnt signaling, homeobox transcription factor PROX1, expressed by a subpopulation of colon cancer progenitor/stem cells. We identify direct PROX1 target genes and show that repression of a pro-apoptotic member of the BCL2 family, BCL2L15, is important for survival of PROX1(+) cells under metabolic stress. PROX1 inactivation after the establishment of metastases prevented further growth of lesions. Furthermore, autophagy inhibition efficiently targeted metastatic PROX1(+) cells, suggesting a potential therapeutic approach. These data identify PROX1 as a key regulator of the transcriptional network contributing to metastases outgrowth in colorectal cancer.

Cancer du sein: radiothérapie et signalisation oestrogénique [Breast cancer: radiotherapy and estrogen signaling].

Hormone receptors are expressed in more than 75% of breast cancer. Therefore, two prescription modalities of endocrine therapy could be proposed: either sequential or concomitant to breast cancer irradiation. If combined to radiotherapy, is endocrine therapy a radiosensitizer? Does endocrine therapy enhance the risk factor of radio-induced toxicity? Here, we will distinguish the interaction of ionizing radiation combined with therapies targeting oestrogen receptor (REα) from the interaction of ionizing radiation with oestrogen. This review aims at making clear all these items.

The Wnt receptor FZD1 mediates chemoresistance in neuroblastoma through activation of the Wnt/beta-catenin pathway.

The development of chemoresistance represents a major obstacle in the successful treatment of cancers such as neuroblastoma (NB), a particularly aggressive childhood solid tumour. The mechanisms underlying the chemoresistant phenotype in NB were addressed by gene expression profiling of two doxorubicin (DoxR)-resistant vs sensitive parental cell lines. Not surprisingly, the MDR1 gene was included in the identified upregulated genes, although the highest overexpressed transcript in both cell lines was the frizzled-1 Wnt receptor (FZD1) gene, an essential component of the Wnt/beta-catenin pathway. FZD1 upregulation in resistant variants was shown to mediate sustained activation of the Wnt/beta-catenin pathway as revealed by nuclear beta-catenin translocation and target genes transactivation. Interestingly, specific micro-adapted short hairpin RNA (shRNAmir)-mediated FZD1 silencing induced parallel strong decrease in the expression of MDR1, another beta-catenin target gene, revealing a complex, Wnt/beta-catenin-mediated implication of FZD1 in chemoresistance. The significant restoration of drug sensitivity in FZD1-silenced cells confirmed the FZD1-associated chemoresistance. RNA samples from 21 patient tumours (diagnosis and postchemotherapy), showed a highly significant FZD1 and/or MDR1 overexpression after treatment, underlining a role for FZD1-mediated Wnt/beta-catenin pathway in clinical chemoresistance. Our data represent the first implication of the Wnt/beta-catenin pathway in NB chemoresistance and identify potential new targets to treat aggressive and resistant NB.

Epigenetic alteration of the Wnt inhibitory factor-1 promoter occurs early in the carcinogenesis of Barrett's esophagus.

The role of Wnt antagonists in the carcinogenesis of esophageal adenocarcinoma (EAC) remains unclear. We hypothesized that downregulation of the Wnt inhibitory factor-1 (WIF-1) might be involved in the neoplastic progression of Barrett's esophagus (BE). We analyzed the DNA methylation status of the WIF-1 promoter in normal, preneoplastic, and neoplastic samples from BE patients and in EAC cell lines. We investigated the role of WIF-1 on EAC cell growth and the chemosensitization of the cells to cisplatin. We found that silencing of WIF-1 correlated with promoter hypermethylation. EAC tissue samples showed higher levels of WIF-1 methylation compared to the matched normal epithelium. In addition, we found that WIF-1 hypermethylation was more frequent in BE samples from patients with EAC than in BE samples from patients who had not progressed to EAC. Restoration of WIF-1 in cell lines where WIF-1 was methylation-silenced resulted in growth suppression. Restoration of WIF-1 could sensitize the EAC cells to the chemotherapy drug cisplatin. Our results suggest that silencing of WIF-1 through promoter hypermethylation is an early and common event in the carcinogenesis of BE. Restoring functional WIF-1 might be used as a new targeted therapy for the treatment of this malignancy.

Epigenetic alterations and constitutive Wnt signaling are early events in the progression from Barrett's esophagus to esophageal adenocarcinoma

SUMMARY Barrett's esophagus (BE) is an acquired condition in which the normal squamous epithelium in the distal esophagus is replaced by a metaplastic columnar epithelium, as a complication of chronic gastroesophageal reflux. The clinical significance of this disease is its associated predisposition to esophageal adenocarcinoma (EAC). EAC is a highly lethal disease. Better understanding of the pathogenesis of columnar metaplasia and its progression to cancer might allow the identification of biomarkers that can be used for early diagnosis, which will improve the patient survival. In this study, an improved protocol for methylation-sensitive single-strand conformation analysis, which is used to analyze promoter methylation, is proposed and a methylation-sensitive dot blot assay is described, which allows a rapid, easy, and sensitive detection of promoter methylation. Both methods were applied to study the methylation pattern of the APC promoter in histologically normal appearing gastric mucosa. The APC promoter showed monoallelic methylation, and because the methylated allele differed between the different gastric cell types, this corresponded to allelic exclusion. The APC methylation pattern was frequently altered in noimal gastric mucosa associated with neoplastic lesions, indicating that changes in the pattern of promoter methylation might precede the development of neoplasia, without accompanying histological manifestations. An epigenetic profile of 10 genes important in EAC was obtained in this study; 5 promoter genes (APC, TIMP3, TERT, CDKN2A and SFRP1) were found to be hypermethylated in the tumors. Furthermore, the promoter of APC, TIMP3 and TERT was frequently methylated in BE samples from EAC patients, but rarely in BE samples that did not progress to EAC. These three biomarkers might therefore be considered as potential predictive markers for increased EAC risk. Analysis of Wnt pathway alterations indicated that WNT2 ligand is overexpressed as early as the low-grade dysplastic stage and downregulation by promoter methylation of the SFRP1 gene occurrs already in the metaplastic lesions. Moreover, loss of APC expression is not the only factor involved in the activation of the Wnt pathway. These results indicate that a variety of biologic, mostly epigenetic events occurs very early in the carcinogenesis of BE. This new information might lead to improved early diagnosis of EAC and thus open the way to a possible application of these biomarkers in the prediction of increased EAC risk progression. RESUME L'oesophage de Barrett est une lésion métaplasique définie par le remplacement de la muqueuse malpighienne du bas oesophage par une muqueuse cylindrique glandulaire, suite à une agression chronique par du reflux gastro-esophagien. La plus importante signification clinique de cette maladie est sa prédisposition au développement d'un adénocarcinome. Le pronostic de l'adénocarcinome sur oesophage de Barrett est sombre. Seule une meilleure compréhension de la pathogenèse de l'épithélium métaplasique et de sa progression néoplasique permettrait l'identification de biomarqueurs pouvant être utilisés pour un diagnostic précoce ; la survie du patient serait ainsi augmentée. Dans cette étude, un protocole amélioré pour l'analyse de la méthylation par conformation simple brin est proposé. De plus, une technique d'analyse par dot blot permettant une détection rapide, facile et sensible de la méthylation d'un promoteur est décrite. Les deux méthodes ont été appliquées à l'étude de la méthylation du promoteur du gène APC dans des muqueuses gastriques histologiquement normales. Le promoteur APC a montré une méthylation monoallélique et, parce que les allèles méthylés différaient entre les différents types de cellules gastriques, celle-ci correspondait à une méthylation allélique exclusive. La méthylation d'APC a été trouvée fréquemment altérée dans la muqueuse gastrique normale associée à des lésions néoplasiques. Ceci indique que des changements dans la méthylation d'un promoteur peuvent précéder le développement d'une tumeur, et cela sans modification histologique. Un profil épigénétique des adénocarcinomes sur oesophage de Barrett a été obtenu dans cette étude. Cinq promoteurs (APC, TIMP3, TERT, CDKN2A et SFRP1) ont été trouvés hyperméthylés dans les tumeurs. Les promoteurs d'APC, TIMP3 et TERT étaient fréquemment méthylés dans l'épithélium métaplasique proche d'un adénocarcinome et rarement dans l'épithélium sans évolution néoplasique. Ces trois biomarkers pourraient par conséquent être considérés comme marqueur prédicatif d'un risque accru de développer une tumeur. L'analyse des altérations de la voie Wnt a montré que WNT2 est surexprimé déjà dans des dysplasies de bas-grade et que la dérégulation de SFRP1 par méthylation de son promoteur intervenait dans les lésions métaplasiques. Une perte d'expression d'APC n'est pas le seul facteur impliqué dans l'activation de cette voie. Ces résultats montrent qu'une grande diversité d'événements biologiques, principalement épigénétiques, surviennent très tôt lors de la carcinogenèse de l'oesophage de Barrett. Ces nouveaux éléments pourraient améliorer le diagnostic précoce et rendre possible l'application de ces biomarqueurs dans la prédiction d'un risque accru de développer un adénocarcinome sur un oesophage de Barrett.

Replicating adenoviruses targeting tumours with constitutive activation of the Wnt signalling pathway

Abstract Activation of the Wnt pathway through mutation of the adenomatous polyposis coli and 13-catenin genes is a hallmark of colon cancer. These mutations lead to constitutive activation of transcription from promoters containing binding sites for Tcf/LEF transcription factors. Tumour-selective replicating oncolytic viruses are promising agents for cancer therapy. They can in principle spread throughout a tumour mass until all the cancerous cells are killed, and clinical trials have shown that they are safe except at very high doses. Adenoviruses are interesting candidates for virotherapy because their biology is well understood and their small genome can be rapidly mutated. Adenoviruses with Tcf binding sites in the E2 early promoter replicate selectively in cells with an active Wnt pathway. Although these viruses can replicate in a broad panel of colon cancer cell lines, some colorectal cancer cells are only semi-permissive for Tcf-virus replication. The aim of my thesis was to increase the safety and the efficacy of Tcf-viruses for colon cancer virotherapy. I replaced the endogenous ElA viral promoter by four Tcf binding sites and showed that transcription from the mutant promoter was specifically activated by the Wnt pathway. A virus with Tcf binding sites in the ElA and E4 promoters was more selective for the Wnt pathway than the former Tcf-E2 viruses. Moreover, insertion of Tcf binding sites into all early promoters further increased viral selectivity, but reduced viral activity. I showed that Tcf-dependent transcription was inhibited by the interaction between ElA and p300, but deletion of the p300-binding site of ElA generally led to viral attenuation. In the semi-permissive cell lines, replication of Tcf-viruses remained lower than that of the wild-type virus. The E2 promoter was the most sensitive to the cell type, but I was unable to improve its activity by targeted mutagenesis. To increase the toxicity of the Tcf-E1A/E4 virus, I decided to express a suicide gene, yeast cytosine deaminase (yCD), late during infection. This enzyme converts the prodrug 5-FC to the cytotoxic agent 5-FU. yCD was expressed in a DNA replication-dependent manner and increased viral toxicity in presence of 5-FC. Tcf-ElA and yCD adenoviruses are potentially useful vectors for the treatment of liver metastases from colorectal tumours. Résumé Dans la quasi-totalité des cancers du côlon, la voie Wnt est activée par des mutations dans les gènes codant pour APC ou pour la (3-caténine. Ces mutations activent de façon constitutive la transcription de promoteurs contenant des sites de liaison pour les facteurs de transcription Tcf/LEF. Les virus réplicatifs spécifiques aux tumeurs sont des agents prometteurs pour la thérapie cancéreuse. En principe, ces vecteurs peuvent se propager dans une masse tumorale jusqu'à destruction de toutes les cellules cancéreuses, et des études cliniques ont démontré que de tels vecteurs n'étaient pas toxiques, sauf à de très hautes doses. Les adénovirus sont des candidats intéressants pour la thérapie virale car leur biologie est bien définie et leur petit génome peut être rapidement modifié. Des adénovirus comportant des sites de liaison à Tcf dans leur promoteur précoce E2 se répliquent sélectivement dans les cellules qui possèdent une voie Wnt active. Ces virus sont capables de se répliquer dans un grand nombre de cellules cancéreuses du côlon, bien que certaines de ces cellules ne soient que semi-permissives pour la réplication des virus Tcf. Le but de ma thèse était d'augmenter la sécurité et l'efficacité des virus Tcf. Le promoteur viral endogène ElA a été remplacé par quatre sites de liaison à Tcf, ce qui a rendu son activation dépendante de la voie Wnt. Un virus comportant des sites de liaison pour Tcf dans les promoteurs ElA et E4 était plus sélectif pour la voie Wnt que les précédents virus Tcf-E2, et un virus comportant des sites Tcf dans tous les promoteurs précoces était encore plus sélectif, mais moins actif. J'ai montré que l'interaction entre ElA et p300 inhibait la transcription dépendante de Tcf, mais la délétion du domaine concerné dans ElA a eu pour effet d'atténuer les virus. Dans les cellules semi-permissives, la réplication des virus Tcf était toujours plus basse que celle du virus sauvage. J'ai identifié le promoteur E2 comme étant le plus sensible au type cellulaire, mais n'ai pas pu augmenter son activité par mutagenèse. Pour augmenter la toxicité du virus Tcf-E1A/E4, j'ai décidé d'exprimer un gène suicide, la cytosine déaminase (yCD), pendant la phase tardive de l'infection. Cette enzyme transforme la procirogue 5-FC en l'agent cytotoxique 5-FU. yCD était exprimée après réplication de l'ADN viral et augmentait la toxicité virale en présence de 5-FC. Les virus Tcf-ElA et yCD sont des vecteurs potentiellement utiles pour le traitement des métastases hépatiques de cancers colorectaux.