Regulation of CDX2 and intestinal differentiation in homeostasis and carcinogenesis: emphasis on the role of MEX3A

Faculty of Medicine of the University of Porto Doctoral Program in Biomedicine Regulation of CDX2 and intestinal differentiation in homeostasis and carcinogenesis: emphasis on the role of MEX3A Regulação do CDX2 e diferenciação intestinal em homeostasia e carcinogénese: relevância funcional do MEX3A Bruno Miguel Correia Pereira IPATIMUP – Differentiation and Cancer group Porto, Portugal 2013 Orientação Professora Doutora Raquel Maria da Silva Graça Almeida, Investigadora e líder do grupo Diferenciação e Cancro do IPATIMUP Professora Afiliada da Faculdade de Medicina da Universidade do Porto Co-orientação Professora Doutora Maria Leonor Martins Soares David, Investigadora do grupo Diferenciação e Cancro do IPATIMUP Professora Catedrática da Faculdade de Medicina da Universidade do Porto DISSERTAÇÃO DE CANDIDATURA AO GRAU DE DOUTOR APRESENTADA À FACULDADE DE MEDICINA DA UNIVERSIDADE DO PORTO iii Artigo 48º, § 3º – A Faculdade não responde pelas doutrinas expendidas na Dissertação (Regulamento da Faculdade de Medicina da Universidade do Porto - Decreto-Lei nº 19337, de 29 de Janeiro de 1931). v JÚRI DAS PROVAS DE DOUTORAMENTO DISSERTATION DEFENSE COMMITTEE Presidente: Doutor José Agostinho Marques Lopes, Director da Faculdade de Medicina da Universidade do Porto (por delegação reitoral) Vogais: Doutor Marc Billaud, Investigador do Institut Albert Bonniot, CRI INSERM/UJF U823, Grenoble, França Doutor Peter Jordan, Investigador do Instituto Nacional de Saúde Dr. Ricardo Jorge de Lisboa Doutor Manuel Alberto Coimbra Sobrinho Simões, Professor Catedrático da Faculdade de Medicina da Universidade do Porto Doutora Carla Isabel Gonçalves de Oliveira, Professora Afiliada da Faculdade de Medicina da Universidade do Porto Doutor João António Pinto de Sousa, Professor Associado da Faculdade de Medicina da Universidade do Porto Doutora Raquel Maria da Silva Graça Almeida, Professora Afiliada da Faculdade de Medicina da Universidade do Porto, e orientadora da tese vii LISTA DE PUBLICAÇÕES LIST OF PUBLICATIONS Ao abrigo do disposto no nº 2 do Art. 8º do Decreto-Lei nº 388/70, constituem parte integrante desta Dissertação os seguintes trabalhos já publicados: Pereira B, Oliveira C, David L, Almeida R. (2009) CDX2 promoter methylation is not associated with mRNA expression. International Journal of Cancer, 125(7):1739-1742 Pereira B, Sousa S, Barros R, Carreto L, Oliveira P, Oliveira C, Chartier NT, Plateroti M, Rouault JP, Freund JN, Billaud M, Almeida R. (2013) CDX2 regulation by the RNA-binding protein MEX3A: impact on intestinal differentiation and stemness. Nucleic Acids Research, 41(7):3986-3999 Pereira B, Le Borgne M, Chartier NT, Billaud M, Almeida R. (2013) MEX-3 proteins: recent insights on novel post-transcriptional regulators. Trends in Biochemical Sciences, 38(10):477-479 Em cumprimento com o disposto no Decreto-Lei supramencionado, o autor declara que participou activamente na concepção e execução do trabalho experimental, na interpretação e discussão dos resultados, e na redacção dos respectivos artigos científicos. ix TABLE OF CONTENTS LIST OF PUBLICATIONS FUNDING ACKNOWLEDGEMENTS TABLE OF CONTENTS ABSTRACT RESUMO ABBREVIATIONS CHAPTER 1. GENERAL INTRODUCTION AND AIMS 1.1 Gastrointestinal carcinogenesis 1.1.1 An overview on gastric cancer 1.1.2 Intestinal metaplasia 1.1.3 Colorectal carcinogenesis 1.1.4 Intestinal homeostasis 1.2 The intestinal transcription factor CDX2 1.2.1 CDX2 homeobox gene and its targets 1.2.2 Role in homeostasis 1.2.3 Involvement in carcinogenesis 1.2.4 CDX2 regulatory network 1.3 Post-transcriptional regulation of gene expression 1.3.1 From transcripts to proteins 1.3.2 RNA-binding proteins and their biological functions 1.4 The MEX-3 family of RNA-binding proteins 1.4.1 Caenorhabditis elegans MEX-3 1.4.2 The mammalian MEX3 proteins 2 4 7 8 11 13 15 17 24 25 27 29 xxv ix x xi xvii xxiii xxi 1 xvii 1.5 Long-term goal and specific aims CHAPTER 2. CDX2 PROMOTER METHYLATION Paper I – CDX2 promoter methylation is not associated with mRNA expression. International Journal of Cancer, 2009 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A Paper II – CDX2 regulation by the RNA-binding protein MEX3A: impact on intestinal differentiation and stemness. Nucleic Acids Research, 2013 Paper II – Supplementary Data CHAPTER 4. THE MAMMALIAN MEX3 PROTEIN FAMILY Paper III – MEX-3 proteins: recent insights on novel posttranscriptional regulators, Trends in Biochemical Sciences, 2013 CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS 5.1 CDX2 regulation by promoter methylation 5.2 3D AGS cell model and loss of CDX2 expression 5.3 Mechanistic aspects of MEX3A function 5.4 MEX3A and carcinogenesis 5.5 A conserved function for MEX3A in mammalian embryogenesis? 5.6 Concluding remarks 37 45 59 67 75 77 78 82 84 87 35 43 65 73 30 REFERENCES APPENDIX Appendix A. List of antibodies and primers used in this study Appendiz B. List of differentially expressed genes between 3D and 2D AGS cell cultures 119 91 xix ABSTRACT The homeodomain transcription factor CDX2 is a key player in intestinal differentiation. Therefore, it is not surprising to find its expression significantly altered in carcinogenic processes of the gastrointestinal tract. In the stomach, de novo CDX2 expression drives a preneoplastic lesion known as intestinal metaplasia that increases the risk of gastric cancer development. Quite the opposite, in the context of colorectal cancer, CDX2 has been classically described as a tumour-suppressor, although this view has been gradually challenged in the last few years. Notwithstanding, it is known that tumours retaining CDX2 expression are more differentiated and show better outcome in terms of survival. Thus, a comprehensive understanding of CDX2 regulation stands out as a research priority, for in the longrun, the acquired knowledge could be integrated into alternative therapeutic strategies that might benefit the patient. Taking into consideration that the extracellular microenvironment is also important to the cellular malignant behavior as genetic constrains, we decided to tackle its influence over CDX2 expression in two fronts:, analyze the possible role of DNA epigenetic changes, hypothesizing that infection by Helicobacter pylori and the generation of chronic inflammation could in some way affect the CDX2 gene methylation pattern and induce its transcription; and in a more exploratory approach, establish 3D culture models that better mimic the in vivo milieu, hypothesizing that they would provide new insights regarding signaling pathways affecting CDX2 levels. We show here that there is a lack of correlation between CDX2 methylation status and its expression in intestinal metaplasia and gastric cancer cell lines, excluding this mechanism as a CDX2 regulatory feature during gastric malignancy. On the other hand, through genomewide screening of the 3D culture system, comprising a gastric cancer cell line with CDX2 expression and the basement matrix matrigel, we were able to identify the RNA-binding protein MEX3A as putatively involved in CDX2 regulation. Further supported by an evolutionary functional link to its Caenorhabditis elegans orthologue MEX-3, we demonstrate that MEX3A maintains a conserved repressive function over CDX2 in gastric and intestinal cellular models. This is dependent on the interaction with a specific binding determinant present in CDX2 mRNA 3`untranslated region. Moreover, MEX3A overexpression carries phenotypic consequences, impairing intestinal differentiation and cellular polarization, while promoting gain of stemness properties. As a result, we describe for the first time a xxi mechanism of CDX2 post-transcriptional control, likely contributing to intestinal homeostasis and carcinogenesis, and add a new layer to the already intricate CDX2 regulatory network. RESUMO O factor de transcrição homeobox CDX2 é um elemento crítico na diferenciação intestinal. Assim, não é inesperado o facto de apresentar níveis de expressão significativamente alterados em processos carcinogénicos do tracto gastrointestinal. No estômago, a expressão de novo de CDX2 induz uma lesão pré-neoplásica denominada metaplasia intestinal que confere um risco acrescido de desenvolvimento de carcinoma gástrico. Pelo contrário, no contexto colorectal, este gene tem sido geralmente descrito como um supressor tumoral, embora esta classificação tenha vindo a ser posta em causa nos últimos anos. Mesmo assim, reconhece-se que tumores com expressão de CDX2 são mais diferenciados e detêm melhor prognóstico em termos de sobrevida. Torna-se, então, fundamental obter uma compreensão aprofundada dos processos de regulação de CDX2, pois a longo prazo o conhecimento adquirido pode ser integrado em estratégias terapêuticas alternativas que poderão beneficiar o paciente. Tendo em consideração que o microambiente extracelular é, tal como as alterações genéticas, importante para o comportamento celular maligno, decidimos confrontar a sua influência sobre a expressão de CDX2 em duas partes: analisar o possível papel de alterações epigenéticas ao nível do promotor do gene, colocando a hipótese de que o processo de infecção por Helicobacter pylori e a subsequente resposta inflamatória poderão criar condições que afectam o perfil de metilação de CDX2 induzindo a sua transcrição; e numa abordagem de carácter mais exploratório, estabelecer modelos de cultura 3D que mimetizem o ambiente in vivo, colocando a hipótese de que estes permitiriam obter novos dados no que respeita às vias de sinalização que afectam os níveis de CDX2. Demonstrámos que não existe uma correlação entre o estado de metilação do promotor de CDX2 e a sua expressão na metaplasia e em linhas celulares de cancro gástrico, excluindo este mecanismo de regulação no processo de carcinogénese gástrica. Por outro lado, através de um varrimento transcricional do sistema de cultura 3D, constituído por uma linha celular gástrica e a matriz de membrana basal matrigel, identificámos uma proteína de ligação ao ARN designada MEX3A como estando potencialmente envolvida na regulação de CDX2. Apoiados numa base funcional evolutiva em relação ao papel do ortólogo MEX-3 em Caenorhabditis elegans, provámos que o MEX3A mantém uma função conservada enquanto repressor da expressão de CDX2 em modelos celulares gástricos e intestinais. Este efeito é dependente da xxiii interacção com um motivo de ligação específico presente na região 3` nãotraduzida do ARN mensageiro de CDX2. Adicionalmente, a sobrexpresão de MEX3A acarreta consequências fenotípicas, afectando a diferenciação intestinal e polaridade celular, promovendo simultaneamente o ganho de propriedades estaminais. Em conclusão, descrevemos pela primeira vez um mecanismo póstranscricional de controlo de CDX2, com relevância para a homeostasia intestinal e provavelmente carcinogénese, adicionando um novo elemento à complexa rede de regulação do gene CDX2. ABBREVIATIONS The following list describes relevant abbreviations and acronyms used. Bcd Bicoid BMP Bone morphogenetic protein Cad Caudal CDX Caudal type homeobox CIN Chromosomal instability CpG Cytosine guanine dinucleotide CRC Colorectal cancer eIF Eukaryotic initiation factor GI Gastrointestinal GLD-1 Defective in germline development hnRNP K Heterogeneous nuclear ribonucleoprotein K HOX Homeobox IM Intestinal metaplasia KH K-homology LGR5 Leucine rich repeat-containing G-protein coupled receptor 5 LRC Label-retaining cell mex Muscle excess MEX3 Mex-3 homologue miR microRNA MRE MEX-3 recognition element mRNP Messenger ribonucleoprotein MSI Microsatellite instability MSI1 Musashi-1 OLFM4 Olfactomedin-4 P body Processing body QKI Quaking RBP RNA-binding protein RING Really interesting new gene RKHD RING finger and KH domain-containing TE Trophectoderm UTR Untranslated region xxv propose that bone marrow-derived cells might be the initiator cells in the gastric carcinogenic process, including IM and dysplasia, by homing in and engrafting to the injury sites (Houghton et al. 2004; Varon et al. 2012). Although enticing as a theory, there has been limited translation of this model into the human setting. Whichever the case, IM is most probably the outcome of an adaptive response that goes amiss, in which the tissue tries to cope with several aggressive inputs, but ends up with the creation of susceptible ground for neoplastic transformation (Mesquita et al. 2006). In terms of classification, two major gastric IM types can be recognized, taking into account not only morphological, but also molecular alterations, like loss and gain of mucinous differentiation markers and mucin-associated carbohydrate antigens (Reis et al. 1999; Silva et al. 2002). The complete IM or intestinal type (previously denominated type I) reflects a complete switch in the differentiation program, with loss of gastric mucins MUC1, MUC5AC, and MUC6, and de novo expression of the intestinal mucin MUC2. It is characterized by the presence of absorptive cells, Paneth cells and goblet cells secreting sialomucins, similar to the small intestinal phenotype. The incomplete IM or GI-mixed type reflects a mixture of gastric and intestinal components, both at the glandular and cellular level (Niwa et al. 2005). It is characterized by the presence of columnar and goblet cells secreting sialomucins and/or sulphomucins (previously denominated types II and III, respectively), similar to the colonic phenotype. Remarkably, the ectopic intestinal glands still preserve a normal cell migration pattern, which is achieved by a redeployment of the proliferative niche from the middle to the base of the gland (Inada et al. 2001; Sakamoto et al. 2011). Reciprocal interactions between the epithelium and the mesenchyme might be behind this architectural remodelling (Mutoh et al. 2005a), as the induction of a mesenchymal intestinal phenotype during the early stages of IM might lead to the establishment of a positive regulatory loop involving paracrine signals (Sakagami et al. 1984). It is not yet clear if the two IM types can be considered sequential steps in a shared process of gradual intestinalization or if they arise independently. The possibility exists that the incomplete type is biologically more unstable as it reflects an aberrant differentiation program without phenotypic parallel in the adult organism. In agreement, epidemiological data, though scarce, shows that it confers increased risk of gastric cancer development compared to the complete type (Filipe and Jass 1986; Rokkas CHAPTER 1. GENERAL INTRODUCTION AND AIMS et al. 1991; Filipe et al. 1994), highlighting the importance of stratification concerning prognostic significance (González et al. 2013). Curiously, bacterial colonization is typically absent in foci of complete IM (Bravo and Correa 1999), favouring a presumably protective function. 1.1.3 Colorectal carcinogenesis Colorectal cancer (CRC) is currently the third most common cancer worldwide, as 10% of the newly diagnosed cases are malignancies of the colon or rectum. Almost 60% of the cases occur in developed regions and the areas with highest incidence are North America, Australia/New Zealand and Europe (Ferlay et al. 2010). More than 600.000 people die each year, making it the fourth leading cause for cancer-related death. In Portugal, considering the total number of cases, it places first in terms of incidence and mortality (IARC 2010). As for the small intestine, it has a remarkably low incidence of primary carcinomas, especially considering its length and surface area, and those that do occur are often related to hereditary syndromes. About 90% of sporadic tumours occur in individuals over the age of fifty. Other risk factors include family history, a diet low in fibers and high in red meat, alcohol, smoking, and sedentary occupation (Johnson et al. 2013). A present estimate is that 15-30% of CRCs have a familial component. Less than 5% of these happen in a recognizable setting of highly penetrant cancer syndromes due to germline mutations, being the most common the hereditary nonpolyposis CRC and familial adenomatous polyposis (Fearon 2011). Despite advances in surgical techniques and adjuvant therapy, there has been only a modest improvement in survival for patients with advanced neoplasms (Edwards et al. 2012). Hence, effective primary and secondary preventive approaches must be developed to reduce morbidity and mortality. Since the description of the classic adenoma-carcinoma pathway (Fearon and Vogelstein 1990), defining CRC as the result of a gradual accumulation of changes that transform normal glandular epithelial cells into adenoma, followed by invasive carcinoma and eventually metastatic cancer, our understanding of its molecular pathogenesis has advanced and led to numerous revisions of this linear tumour progression model. It is now recognized that loss of genomic stability is a hallmark CHAPTER 1. GENERAL INTRODUCTION AND AIMS 7 feature of colorectal carcinogenesis (Grady and Carethers 2008), and this property led to the establishment of a classification system: (i) the chromosomal instability (CIN) phenotype, found in as many as 85% of tumours and defined by the presence of aneuploidy or structural aberrations; (ii) the MSI phenotype, defined by the presence of unstable loci due to inactivation of genes in the DNA mismatch repair family; and (iii) the CpG island methylator phenotype, exhibiting both global DNA hypomethylation and hypermethylation of gene promoters that contain CpG islands (Pritchard and Grady 2011). It is the accumulation of mutations combined with multiple cycles of clonal selection that results in the deregulation of signalling pathways controlling cell proliferation, differentiation, apoptosis, angiogenesis and invasion, ultimately culminating in cancer development. 1.1.4 Intestinal homeostasis The intestinal tract is anatomically divided into two well-defined segments, the small intestine and large intestine (or colon), lined by a specialized single layer of cells organized into two morphologically and functionally distinct compartments: flask-shaped submucosal invaginations known as crypts of Lieberkühn, and fingershaped luminal protrusions termed villi, which dramatically increase the absorptive surface area of the small intestine. The crypt constitutes the proliferative compartment of the intestinal epithelium; it is monoclonal and maintained by four to six multipotent stem cells located in the lower third (Bjerknes and Cheng 1999) that give rise to intermediate descendants referred to as transit-amplifying cells. The villus represents the differentiated compartment and is polyclonal as its cells derive from several crypts (Potten and Loeffler 1990). Absorptive enterocytes, mucous-producing goblet cells, and hormone-secreting enteroendocrine cells migrate upwards along the basement membrane to the epithelium apex, where they undergo apoptosis, being subsequently exfoliated into the intestinal lumen (Hall et al. 1994). Paneth cells are unusual in that they settle at the crypt base and are the only cell type migrating downwards (Bry et al. 1994). The modular organization of the small intestine and colon is globally comparable. Histologically, there are, however, two important differences between them. The colon carries no villi; instead, it has a flat surface epithelium and larger colonic crypts extending deep into the submucosa. The relative abundance of each of the main cell types also varies markedly within the CHAPTER 1. GENERAL INTRODUCTION AND AIMS different intestinal segments. Enterocytes are highly polarized cells responsible for absorbing and transporting nutrients across the epithelium, secreting a cocktail of hydrolytic enzymes into the gut. They make up more than 80% of all intestinal epithelial cells. Goblet cells secrete protective mucins and trefoil proteins required for the movement and effective removal of luminal contents, while providing protection against shear stress and chemical damage. Accordingly, their numbers increase from the proximal (4%) to distal (16%) intestine (Karam 1999). Enteroendocrine cells coordinate gut physiology through specific hormone production (Höcker and Wiedenmann 1998). They are scattered as individual cells throughout the mucosa, representing a small proportion (<1%) of the cells lining the epithelium (Schonhoff et al. 2004). Finally, Paneth cells have a function in innate immunity, synthesizing bactericidal agents such as defensins and lysozyme (Porter et al. 2002). They are absent from the colon and have a life expectancy of six to eight weeks (van der Flier and Clevers 2009), much longer than that of their terminally differentiated villus counterparts, with a turnover rate of roughly three to five days (Wright and Irwin 1982). Currently, some controversy exists as to the presence of distinct types of intestinal stem cells. The “+4 position” model assumes the crypt base is exclusively populated by terminally differentiated Paneth cells and that stem cells are located just above them, on average at the +4 position (Potten et al. 1974). These cells were shown to divide once every day and to be unusually sensitive to radiation, possibly preventing the accumulation of deleterious genomic changes (Potten 1977). They were described to retain DNA labelling, and are also called labelretaining cells (LRCs), a feature suggested as being the result of asymmetric strand segregation (Potten et al. 2002). The “stem cell zone” model states that small, undifferentiated, cycling cells called crypt base columnar (CBC) cells, residing in a stem cell-permissive environment and wedged between the Paneth cells at the base of the crypts, are likely to be the true stem cells (Cheng and Leblond 1974; Bjerknes and Cheng 1999). Yet, definitive proof of stemness requires putative stem cells to be experimentally linked to their progeny, and this has proven elusive due to lack of specific markers (Barker et al. 2012). The musashi-1 (MSI1) gene encodes a RNAbinding protein initially described as a regulator of asymmetric division in neural stem cells (Glazer et al. 2012). Later studies showed that it denotes multipotent stem cells in other settings, being also highly expressed at the crypt base, with expression CHAPTER 1. GENERAL INTRODUCTION AND AIMS 9 evident on the CBC cells, as well as in some LRCs (Potten et al. 2003). However, its broad expression domain suggests that it is a marker of early committed progenitors alike. Lineage tracing techniques in inducible mouse models led to the identification of the leucine rich repeat-containing G-protein coupled receptor 5 (LGR5), the first specific CBC cell marker (Barker et al. 2007). The LGR5 protein acts as the receptor for a small family of Wnt agonists called R-spondins. Lgr5+ cells are highly uniform in morphology, invariably touch Paneth cells, and divide each day. Another robust marker of Lgr5+ stem cells called olfactomedin-4 (OLFM4) has recently emerged. It is a member of the olfactomedin domain-containing family and encodes a secreted glycoprotein that appears to have anti-apoptotic and cell cycle regulatory characteristics (Grover et al. 2010). Molecular in situ hybridization revealed that OLFM4 is highly expressed in normal CBC cells in human small intestine and colon (van der Flier et al. 2009). The most reliable candidate as an LRC marker to date is BMI1, which encodes a component of the polycomb repressing complex 1 that acts as a chromatin modifier, being implicated in the stable maintenance of gene repression (Valk-Lingbeek et al. 2004). It has attracted attention due to its role in regulating self-renewal of neural and hematopoietic progenitors. Bmi1 was found to mark rare cells at the +4 cell position uniquely in about 10% of the proximal small intestine (Sangiorgi and Capecchi 2008). Modelling of the intestinal epithelium is based on a delicate balance between self-renewal and differentiation, which must be maintained throughout life. Noteworthy, in all species studied, the crypt-villus axis junction represents the physical threshold from which intestinal cells acquire their final functional characteristics, arguing for conserved molecular mechanisms involved in this process. In fact, intestinal homeostasis is dependent on autocrine and paracrine interactions between the mucosa and the underlying mesenchyme, and many of the intervening signalling pathways, such as Wnt, Bone Morphogenetic Protein (BMP), and Notch, have been identified (Crosnier et al. 2006). These molecular signals provide some of the basic principles through which intestinal architecture is organized, but it is still not clear at what point stem cell progeny loses its potency and becomes irreversibly committed to differentiation. Irrespectively of how this occurs, cell fate decisions need to be closely timed in relation to the pattern of cell divisions. While the mechanisms controlling intestinal cell transitions are far from completely CHAPTER 1. GENERAL INTRODUCTION AND AIMS understood, it is obvious that they involve transcription factors conferring compartment-specific gene expression. When considering the molecular basis of intestinal differentiation, a common point of convergence exists, independently of the biological setting – the caudal (cad) type homeobox 2 gene (CDX2). This transcription factor is exclusively present in the intestine and de novo expressed in every foci of ectopic intestinal differentiation in the body, associated with carcinogenic processes. Its embryonic requirement and transcriptional activity over intestine-specific genes incorporate the properties of a “master regulator”. 1.2.1 CDX2 homeobox gene and its targets One of the earliest isolated homeobox genes in Drosophila showed for the first time maternal as well as zygotic expression, accumulating in a concentration gradient spanning the antero-posterior axis of the embryo. During later embryogenesis, it was expressed in more posterior structures, therefore it was named cad (Mlodzik et al. 1985). Soon after, three murine orthologues designated Cdx1 (Duprey et al. 1988), Cdx2 (James and Kazenwadel 1991) and Cdx4 (Gamer and Wright 1993) were characterized. With the exception of Cdx4, they are confined to the posterior gut endoderm during later development and the mature intestine after birth. The human CDX2 gene was independently cloned from an adult jejunal cDNA library (Drummond et al. 1997) and by differential screening of mRNA from CRC (Mallo et al. 1997). It maps to the ParaHox gene cluster in chromosome 13q12.3, has three exons and encodes a 313 amino acid protein containing a nuclear translocation and activation domain in the amino terminus (Trinh et al. 1999), and a highly conserved helix-turn-helix DNA-binding motif called the homeodomain towards the carboxyl terminus. The CDX2 homeodomain shares 100% homology with CHAPTER 1. GENERAL INTRODUCTION AND AIMS 11 1.2 THE INTESTINAL TRANSCRIPTION FACTOR CDX2 the hamster CDX3 homeodomain, 96% with the mouse CDX2, and even 88% with the homeodomain from Drosophila Cad, implying recognition of similar DNA targets. CDX proteins have been demonstrated to bind as a monomer or dimer to one or more CDX-responsive elements, typically consisting of the consensus sequence (A/C)TTTAT(A/G), in direct or reverse orientation both in promoter and gene enhancer regions (Margalit et al. 1993; Taylor et al. 1997; Verzi et al. 2010). These motifs are frequently juxtaposing (Troelsen et al. 1997) or even intersecting (Lambert et al. 1996) a TATA-box sequence, usually part of the RNA polymerase II binding site. Although some reports demonstrate that CDX2 can act as an indirect repressor, counteracting CDX1 or other transcription factor-mediated activation, by competing for the same binding sites (Gautier-Stein et al. 2003; Furumiya et al. 2013) or through interaction with basal components of the transcriptional machinery (Chun et al. 2007; Mutoh et al. 2010), most studies provide strong evidence of a role in activating gene transcription (Verzi et al. 2011). Indeed, CDX2 is determinant for the expression of numerous intestinal differentiation markers, and it has been shown to regulate absorptive lineage-specific factors such as sucrase-isomaltase (Suh et al. 1994; Boudreau et al. 2002), lactase phlorizin hydrolase (Troelsen et al. 1997; Fang et al. 2000), carbonic anhydrase 1 (Drummond et al. 1996), guanylyl cyclase C (Park et al. 2000), calbindin-D9K (Lambert et al. 1996; Wang et al. 2004a), liver-intestine cadherin (Hinoi et al. 2002), and villin (Yamamichi et al. 2009); goblet cell-specific factors such as MUC2 (Mesquita et al. 2003; Yamamoto et al. 2003) and trefoil factor 3 (Shimada et al. 2007); and enteroendocrine cell-specific proglucagon (Jin and Drucker 1996). On the contrary, CDX2 levels consistently seem lowest in Paneth cells, a lineage recently shown to be suppressed by CDX2 overexpression-mediated loss of nuclear β-catenin (Crissey et al. 2011). CDX2 is also involved in controlling the expression of additional molecules that contribute to cellular dynamics, including processes of proliferation (Uesaka et al. 2002), growth arrest (Bai et al. 2003; Aoki et al. 2011), migration (Coskun et al. 2010), adhesion (Lorentz et al. 1997; Sakaguchi et al. 2002; Hinkel et al. 2012), metabolism and transport (Modica et al. 2009; Kakizaki et al. 2010), inflammation (Wang et al. 2005), glycoproteome modulation (Isshiki et al. 2003), and apoptosis (Mallo et al. 1998). In any case, regulatory outcome ultimately results from the cooperative balance with other important transcription factors, being mostly relevant the hepatocyte nuclear factor (HNF1 and HNF4) and GATA-binding factor (GATA4/5/6) families (Boudreau et al. 2002; Verzi et al. 2010; CHAPTER 1. GENERAL INTRODUCTION AND AIMS Verzi et al. 2013). On the other hand, these and other transcription factors, such as KLF4, can be directly regulated by CDX2 (Mahatan et al. 1999; Boyd et al. 2010). Hence, the complex hierarchies of control that govern gene expression during intestinal differentiation and development are determined by the stoichiometry of different transcription factors and cofactors within intestinal cells at any given time, as well as by the type, number, and arrangement of cis-acting elements in the regulatory regions of intestinal genes. 1.2.2 Role in homeostasis During embryogenesis, CDX homologues participate in patterning of the vertebral column (Subramanian et al. 1995; Chawengsaksophak et al. 1997; van Nes et al. 2006), as well as in haematopoiesis (Wang et al. 2008), via HOX gene regulation and with a certain degree of functional redundancy (van den Akker et al. 2002; Savory et al. 2009). Both Cdx1 and Cdx2 are expressed during endoderm development and in adult intestine; however, the specific role of each member and the extent of their functional equivalence is still not completely understood. In the later embryo, CDX1 and CDX2 levels vary quantitatively along the rostrocaudal axis, with highest expression of CDX1 in the distal portion of the colon and highest CDX2 expression in the proximal colon, diminishing in either direction (James et al. 1994; Silberg et al. 2000). A fairly complementary gradient of expression has also been described along the vertical crypt-villus axis, with the former primarily localized to the crypt, and the latter primarily along the villus, although with less staining towards the tip (Silberg et al. 1997; Rings et al. 2001; Kim et al. 2002; Silberg et al. 2002). These patterns endure throughout the lifespan of the animal and are also observed in human tissue (Walters et al. 1997; Boulanger et al. 2005). They are thought to reflect intrinsically different functionalities, with CDX1 associated to a more proliferative phenotype and CDX2 to a more differentiated one, conceding a partially interchangeable activity in certain contexts (Verzi et al. 2011). Cdx1−/− mutants are viable and fertile, and like transgenics overexpressing Cdx1, display no overt intestinal phenotype (Subramanian et al. 1995; Bonhomme et al. 2008; Crissey et al. 2008). Cdx2 precedes and is needed for Cdx1 onset during intestinal development (Eda et al. 2002; Silberg et al. 2002; Mutoh et al. 2009), and Cdx1 requirement is only unmasked upon Cdx2 loss, as their combined absence in the adult stage CHAPTER 1. GENERAL INTRODUCTION AND AIMS 13 significantly enhances lethality effects of isolated Cdx2 deficiency (Verzi et al. 2010; Verzi et al. 2011; Stringer et al. 2012). It therefore follows that CDX2 function in the adult intestine is more vital than that of CDX1. The role of CDX2 started to be established in vitro, when it was observed that forced expression in the undifferentiated intestinal cell line IEC-6 arrests proliferation and initializes an epithelial polarity program (Suh and Traber 1996). Cdx2-/- mice embryos display pre-implantation lethality and die in utero between embryonic day 3.5 and 5.5, whereas the Cdx2+/- mutants are viable, but exhibit skeletal anomalies or stunted growth (Chawengsaksophak et al. 1997; Tamai et al. 1999). Crucial evidence that CDX2 functions in specifying intestinal cell fate was initially obtained by observing that 90% of the heterozygous mice develop multiple polyps within the first three months of life, particularly in the proximal colon. These polyps present the remaining Cdx2 allele inactivated and reveal a homeotic reversion towards anterior differentiation (Beck et al. 1999). This process of intercalary regeneration means that local sporadic Cdx2 haploinsufficiency conveys a signal for a pathway of rostral phenotype; hence, a gradient of positional information is observed, with areas of stratified squamous epithelium similar to that seen in the oesophagus, areas resembling the gastric mucosa, and even areas reminiscent of the small intestine. Wild-type male hosts with Cdx2–/– cells from mutant female donors also develop chimaeric intestinal patches of organotypically normal stomach epithelium (Beck et al. 2003). Of interest, the underlying host stroma concurrently assumes a gastric phenotype, proving that endodermal expression of CDX2 initiates endodermal/mesodermal cross-talk and is the primary signal for gut differentiation, subsequently involving appropriate feedback loops (Stringer et al. 2008). A complementary approach, relying in transgenic expression of Cdx2 in the stomach, demonstrated that CDX2 is sufficient to induce intestinal enterocytes with enzymatic and absorptive functions in vivo (Mutoh et al. 2005b). This is reinforced by these mice being able to survive over one month after extensive small bowel resection, when compared to the short seven day lifespan of surgery-controls. Other animal models have been independently reported, in which embryonic lethality of Cdx2 ablation was circumvented by conditionally targeting its knockout at posterior developmental stages. Pending on temporal restrictions, this inactivation differentially impacts global intestinal morphology. Early loss results in anterior transformation of the small intestine to an oesophageal phenotype (Gao et al. 2009), CHAPTER 1. GENERAL INTRODUCTION AND AIMS CHAPTER 1. GENERAL INTRODUCTION AND AIMS whereas in latter induction the intestine shows a more gastric-like identity, with impaired endo-lysosomal trafficking and inefficient segregation of apico-basal membrane domains (Gao and Kaestner 2010; Grainger et al. 2010). Conditional inactivation in the already adult mice compromises enterocyte function, causing severe malnutrition and death in about three weeks (Verzi et al. 2010; Verzi et al. 2011). A limited ablation strategy compatible with long-term survival has been applied in this setting revealing the presence of partial gastric-nature metaplasias associated with loss of all differentiated intestinal cell types (Hryniuk et al. 2012; Stringer et al. 2012). Hence, CDX2 stands out as being the most critical element in the development, differentiation, and maintenance of the intestinal phenotype. 1.2.3 Involvement in carcinogenesis Beyond its homeotic function, CDX2 is also involved in processes of leukemogenesis (Lengerke and Daley 2012) and GI carcinogenesis. Conflicting data regarding the nature of CDX2 function in tumour initiation and development exist, and most probably CDX2 assumes different roles in dissimilar contexts. IM and gastric cancer An association between IM and intestinal-type gastric carcinoma has been previously established (Correa 1992). It is now widely recognized that the main molecular driver of this preneoplastic condition is de novo expression of CDX2. Several studies have described the presence of both CDX1 and CDX2 in nearly all gastric IM foci (Bai et al. 2002; Almeida et al. 2003; Kim et al. 2006; Barros et al. 2008). But proof of concept supporting CDX2 involvement in this process was provided by two transgenic mouse models achieving inappropriate Cdx2 expression in the gastric epithelium. These mice develop IM, characterized by the presence of absorptive, goblet, and enteroendocrine cell-types (Mutoh et al. 2002; Silberg et al. 2002). One model even progressed to gastric cancer after long-term induction (Mutoh et al. 2004), reinforcing the implication of IM in the genesis of gastric carcinoma. Besides being ectopically expressed in IM, it has been repeatedly demonstrated that CDX2 expression is downregulated in the progression from IM to gastric cancer (about 50% are positive), and that CDX2 positivity significantly correlates with a more differentiated (intestinal) histology and better prognosis (Kaimaktchiev et al. 2004; 15 CHAPTER 1. GENERAL INTRODUCTION AND AIMS As for the influence of stromal elements, it was observed early on that epithelialmesenchymal cellular interactions can differentially affect the expression of CDX1 and CDX2 homeobox genes (Duluc et al. 1997). For example, different components of the basement membrane matrix, which lie at the interface of and are synthesized by epithelial and mesenchymal cells, such as Laminin-α1, can positively modulate CDX2 expression (Lorentz et al. 1997). A stimulation of CDX2 expression by subepithelial colonic myofibroblasts was also reported, mediated by the noncanonical Wnt family member WNT5A and the epithelial receptor ROR2, contributing to inhibition of the canonical Wnt signalling and intestinal differentiation (Pacheco and Macleod 2008). On the other hand, collagen type I induces phenotypic changes in CRC cells through the β1-integrin/FAK signalling pathway that involve reduced Cdx2 promoter activity and mRNA expression (Brabletz et al. 2004). In agreement with this, a model of orthotopic and heterotopic xenografs in nude mice demonstrated that CDX2 expression is adaptable and strongly dependent on the microenvironment (Benahmed et al. 2007). Post-transcriptional regulation by microRNAs Due to the absence of CDX2 expression in about half of the gastric cancer cases and heterogeneous loss in CRCs, it was hypothesized that microRNAs (miRs) could also be associated with CDX2 silencing in these contexts. Computational prediction of miR-binding sites, using independent databases, led to the selection of miR-9 and miR-204 as putative candidates for CDX2 regulation (Rotkrua et al. 2011). A comparative analysis of CDX2 and miR-9 expression in a panel of gastric cancer tissues revealed an overall inverse correlation between both, and in vitro experimental data, confirmed that miR-9 interacts directly and specifically with the CDX2 3`untranslated region (UTR), leading to downregulation of CDX2 target genes and promotion of cell growth. More recently, it was demonstrated that exogenous CDX1-induced expression of miR-9, miR-16 and miR-22 suppresses CDX2 mRNA by targeting its 3`UTR in a CRC cell line (Tagawa et al. 2012). Post-translational regulation Phosphorylated forms of p38 MAPKs were shown to be mostly retained in the nuclei of villus cells, with differential p38α kinase activity constituting an early and necessary event for the initiation of the intestinal differentiation program (Houde et CHAPTER 1. GENERAL INTRODUCTION AND AIMS al. 2001). This stimulatory effect is accomplished by CDX2 phosphorylation, enhancing its transcriptional ability without involvement in the loss of proliferative potential or cell survival. On the other hand, phosphorylation by ERK1/2 MAPKs of serine 60 within the amino-terminal activation domain of CDX2, reduces its transcriptional capability and impairs certain intestinal differentiation properties (Rings et al. 2001; Lemieux et al. 2011). This modification was mainly found in the proliferating compartment of the intestine, while the nonphosphorylated and more active CDX2 was predominantly found in the differentiated region, in agreement with the localization of pERK1/2 (Aliaga et al. 1999). In addition, activation of the ERK1/2 MAPK cascade was shown to promote ubiquitin-proteasome-dependent turnover of the CDX2 protein (Krueger et al. 2009). Two independent studies have elucidated how CDX2 regulation can be coordinated with the cell cycle machinery. It was observed that CDX2 undergoes CRM1-dependent nuclear export and subsequent proteolytic degradation by interaction with CDK2 in proliferative intestinal cells (Boulanger et al. 2005). The CDK2-mediated phosphorylation of CDX2 was found to occur downstream of the homeodomain at serine 283, identified as being part of a conserved motif of four evenly spaced serines called the 4S motif, similar to the one controlling β-catenin degradation by the proteasome (Gross et al. 2005). Preventing phosphorylation through this site blocked polyubiquitination and stabilized CDX2, with impact on overall cell behaviour. In conclusion, CDX2 contains multiple phosphorylation sites that either positively or negatively balance its activity and/or stability in response to different signalling pathways, being conceivable that their combined action is required for the onset of the complete differentiation process. Presently, it is well established that gene expression is regulated at multiple levels, as previously demonstrated for CDX2, and that the diverse processes involved are integrated within each other. Transcriptional control is one of the most important steps within the gene regulation cascade. Nevertheless, the significance of post23 1.3 POST-TRANSCRIPTIONAL REGULATION OF GENE EXPRESSION CHAPTER 1. GENERAL INTRODUCTION AND AIMS transcriptional control is evolving, as the ability to reprogram protein synthesis is a common theme in embryonic development, wound healing, inflammation, metabolic stress and aging. An overview of the main concepts of post-transcriptional regulation is presented, focusing on RNA-binding proteins (RBPs) and their roles in development and disease. 1.3.1 From transcripts to proteins An increasing number of publications show that a poor correlation between steady-state transcript abundances and corresponding protein pools generally exists in almost every organism (de Sousa Abreu et al. 2009; Maier et al. 2009; Vogel et al. 2010). In accordance, a recent study comprehensively analysed mRNA and protein levels, half-lives, transcription and translation rate constants for thousands of genes in NIH3T3 mouse fibroblasts, and found that mRNA levels only explain around 40% of the variability in protein levels (Schwanhäusser et al. 2011). Whether this exact percentage is valid for other cell types is unknown. Thus, it is clear that a significant fraction of Eukaryotic gene regulation is post-transcriptional in nature. Post-transcriptional regulation encompasses RNA processing, localization, translation and decay, as well as RNA stability throughout. These interconnected mechanisms provide complementary quality-control layers that collectively define the fate of every transcript (Moore 2005). Adding to the complexity, RNAs do not dwell alone in the cell, as they are ever accompanied by trans-acting factors, namely RBPs and non-coding RNAs, such as miRs, that bind cis-elements usually present in the 3´UTR (Kuersten and Goodwin 2003; Huntzinger and Izaurralde 2011). These are capable of adjusting the amount of gene product more rapidly, precisely and with a tactical reversibility option that transcriptional regulation alone cannot offer. It is this unique escort, their relative positions, and interactions that create a highly dynamic web of messenger ribonucleoprotein (mRNP) complexes, ruling RNA life. Indeed, Eukaryotic mRNPs have been functionally considered “posttranscriptional operons” that markedly expand the regulatory elasticity of our surprisingly small genomes (Keene and Tenenbaum 2002; Keene 2007). Evidence in favour of this enticing RNA operon model on a genome-wide scale is mounting, as studies show that discrete subsets of mRNAs with shared sequence elements and CHAPTER 1. GENERAL INTRODUCTION AND AIMS encoding proteins with common functions and locations are coordinately regulated by specific mRNPs (Gerber et al. 2004; Hogan et al. 2008). RNA processing starts in the nucleus, where precursor mRNAs acquire a 7methylguanosine cap structure at the 5' terminus (Lewis and Izaurralde 1997), a specialized poly(A) tail at the 3` end (Mangus et al. 2003), and introns are removed by splicing (Matlin et al. 2005). A major remodelling in mRNP composition occurs as mature mRNAs pass through the nuclear-pore complex (Köhler and Hurt 2007), with some factors remaining stably associated, whereas others are dynamically replaced by cytoplasmic counterparts. Upon export, several mRNAs are integrated in the translationally active pool, a process that remodels the protein coat and assembles polysomes. The canonical mechanism of translation is intricate and involves three coordinated events: initiation, elongation and termination (Jackson et al. 2010). On the other hand, some mRNAs are programmed for delayed translation, which allows transcripts to be transported to a specific subcellular localization or even stored until developmental or environmental cues call for their protein synthesis. These mRNAs are packaged into cytoplasmic mRNP granules that lack a limiting membrane and are visible under light microscopy, called processing bodies (P bodies) and stress granules (Anderson and Kedersha 2006). 1.3.2 RNA-binding proteins and their biological functions An extensive computational analysis established that RBPs comprise 8 to 15% of the protein coding repertoire in eukaryotic genomes, highlighting both their ancient origin and the importance of RNA regulation in cell function (Anantharaman et al. 2002). One possible explanation is that as highly specific processes to fine-tune gene expression evolved, a concomitant expansion of the number of RBPs needed has occurred. For example, at least 74% of human genes express multiple isoforms by using different exonic combinations through alternative splicing (Johnson et al. 2003). Its emergence during evolution drove the need for a corresponding increase in the number of RBPs, and it is also, in itself, a mechanism by which cells can expand their RBP repertoire. A subset of these proteins recognizes common features to almost every message, such as the 5`cap or the 3`poly (A) tail. However, the majority have a requirement for a primary sequence or type of secondary structure in which the 25 CHAPTER 1. GENERAL INTRODUCTION AND AIMS former is embedded, that are present in some mRNAs but not others. Their specificity is mediated by unique structural arrangements of individual RNA-binding domains whose properties are further moderated by auxiliary domains. This property, coined as “cooperative modularity”, accommodates different functionalities and allows for enormous combinatorial potential, further increasing affinity towards RNA (Lunde et al. 2007). To date, more than 40 different binding domains have been proposed, being the most frequent the RNA recognition motif. Other common classes include the K-homology (KH) domain, zinc-fingers of the CCCH and CCHC type, and the double-stranded RNA-binding domain. RBPs capacity to virtually regulate every aspect of RNA biogenesis and function is remarkable. In fact, they play pivotal roles during embryonic development (Kuersten and Goodwin 2003), and in a breath of several homeostatic processes such as synaptic plasticity (Luo et al. 2010), immune responses (Anderson 2008), epithelial cell proliferation (Yang et al. 2011), differentiation (Yang et al. 2010) and polarity (Nagaoka et al. 2012). With increasing knowledge on their importance, the more apparent it becomes that tampering with RBPs expression or function underlies the onset of several pathological conditions, including cancer (Lukong et al. 2008). Members of the signal transduction and activation of RNA (STAR) family of RBPs play vital roles in cell proliferation and differentiation. SAM68, for instance, is overexpressed in breast (Lukong et al. 2005) and prostate cancer cells (Busà et al. 2007). On the contrary, Quaking (QKI) seems to function as a tumour suppressor in CRCs (Yang et al. 2010) and glioblastomas (Chen et al. 2012). Members of an evolutionarily-conserved family of RBPs, the MEX-3 family, are emerging as important post-transcriptional regulators of several cellular processes in diverse physiological settings. Their functions and underlying molecular mechanisms, particularly well-known for the ancestral mex-3 in the model organism Caenorhabditis elegans, are now described. 1.4 THE MEX-3 FAMILY OF RNA-BINDING PROTEINS CHAPTER 1. GENERAL INTRODUCTION AND AIMS 1.4.1 Caenorhabditis elegans MEX-3 In the worm embryo, normal development requires precise spatial and temporal expression patterns of maternal mRNAs, such as the anteriorly localized membrane receptor glp-1 and the posteriorly localized transcription factor pal-1, the Drosophila notch and cad orthologues, respectively (Evans and Hunter 2005). The correct distribution of these messages that act to direct lineage-specific commitment patterns from individual blastomeres, is determined by partitioningdefective (par) genes, which are required to establish polarity and whose disruption cause the earliest and most extensive embryonic abnormalities (Kemphues et al. 1988; Goldstein and Macara 2007). Molecular evidence places C. elegans MEX-3 amid both classes, an intermediate regulator of cell fate determinants under pardependent control, thus acting as a critical component in the link between cell polarity and asymmetric gene expression (Huang et al. 2002). mex-3 gene encodes a protein with two seventy-amino acid regions that are 40% identical to each other. The repeated sequences correspond to putative RNAbinding motifs, initially identified in the pre-mRNA-binding heterogeneous nuclear ribonucleoprotein (hnRNP) K and named KH domains (Siomi et al. 1993). MEX-3 interacts with a minimal MEX-3 recognition element (MRE), defined as the degenerate consensus sequence (A/G/U)(G/U)AGN(0-8)U(U/A/C)UA (Pagano et al. 2009). Mutations disrupting the mex-3 locus are fully penetrant, recessive, strict maternal-effect and embryonically lethal, resulting in embryos that abnormally generate body-wall muscle from the anterior blastomere (AB), hence the name mex for “muscle excess” (Draper et al. 1996). This homeotic transformation is due, in part, to the function MEX-3 exerts over the transcription factor pal-1, the orthologue of CDX genes in mammals, and the somatic determinant of posterior identity (Hunter and Kenyon 1996). mex-3 mRNA and protein accumulate cytoplasmically and are unevenly distributed in early embryos (Draper et al. 1996). After fertilization and between the two and four-cell stage, MEX-3 is preferentially localized to the AB lineage, disappearing afterwards in a pattern similar to that described for other maternal mRNAs (Seydoux and Fire 1994). On the other hand, while pal-1 mRNA is present throughout the early embryo, PAL-1 protein is asymmetrically localized and only detected from the four-cell stage onwards in descendants of the posterior (P1) blastomere lineage, thus correlating with low MEX-3 levels (Figure 3A). In accordance, PAL-1 is ectopically expressed in all cells of mex-3 mutant embryos and 27 CHAPTER 1. GENERAL INTRODUCTION AND AIMS MEX-3 has been shown to repress translation of a lacZ RNA reporter construct containing the pal-1 3`UTR in anterior blastomeres (Hunter and Kenyon 1996). Therefore, MEX-3 plays a crucial role as a translational repressor of pal-1, specifying proper blastomere identity during early embryogenesis. Post-transcriptional regulation of maternal RNAs is also a primary mechanism to control gene expression in C. elegans germline, and MEX-3 contributes to the maintenance of germ cell totipotency in the adult worm (Ciosk et al. 2006). The hermaphrodite germline is a highly dynamic organ presenting a stereotypical organization of gametes commitment: the most distal end of the gonad is a proliferative compartment containing mitotic cells that progressively enter meiosis in a central syncytial region. Spontaneous germ cell death by apoptosis or differentiation into sperm cells (late L4 larval stage) and oocytes (adulthood) occurs in the proximal gonad. Switches between these different stages involve spatially nonoverlapping translational regulation of target mRNAs by the RBPs defective in germline development (GLD-1) and MEX-3 (Lee and Schedl 2001; Mootz et al. 2004) (Figure 3B). In maturing oocytes, MEX-3 also mediates pal-1 inhibition; combined with GLD-1 action, this ensures that embryos do not inherit maternal PAL-1 protein and develop properly (Draper et al. 1996; Mootz et al. 2004). In accordance, double mutant strains for mex-3 and gld-1 develop a germline tumour containing numerous cells of muscular, neuronal and intestinal nature, reminiscent of human CHAPTER 1. GENERAL INTRODUCTION AND AIMS teratomas (Ciosk et al. 2006). This germ cell transdifferentiation seems to be the end result of an abnormal expression of somatic determinants such as PAL-1. 1.4.2 The mammalian MEX3 proteins C. elegans mex-3 was identified and characterized in humans as a family of four homologous genes called MEX3A to MEX3D (Buchet-Poyau et al. 2007). These are located at chromosomic positions 1q22, 15q25.2, 18q21.2 and 19p13.3, respectively, and are composed of two exons and one intron. The closely related encoded proteins contain two tandemly repeated KH domains that putatively provide RNA-binding properties. Additionally, they possess a carboxy-terminal really interesting new gene (RING) finger module, not present in the nematode MEX-3 and believed to mediate protein-protein interactions. Four mouse orthologues were also identified displaying strong similarity at the amino acid level to their human counterparts. For these characteristics they are also known as RING finger and KH domain-containing (RKHD) proteins. Expression analysis of the MEX3 transcripts in different human tissues showed that MEX3D is ubiquitous, while the others have a varied expression pattern, with the highest level found in fetal brain and testis (Buchet-Poyau et al. 2007). Overexpression experiments demonstrated that MEX3 are phosphoproteins that shuttle between the nucleus and the cytoplasm by the CRM1-dependent export pathway and via an N-terminally located export signal. Confocal microscopy analysis revealed co-localization of MEX3A and MEX3B with DCP1A and AGO family members, being the latter catalytic components of the RNA-induced silencing complex (RISC), the key effector of miR and RNA interference (RNAi) pathways. This association was restricted to P bodies, which are known sites of mRNA turnover (Eulalio et al. 2007). These results together with the ability of MEX3 proteins to interact in vitro with poly(A) ribonucleotide homopolymers and certain cellular mRNAs, indicate that MEX3 proteins constitute novel human RBPs potentially involved in posttranscriptional regulatory networks (Donnini et al. 2004; Buchet-Poyau et al. 2007; Courchet et al. 2008). Nevertheless, the specific biological role of each member was basically unknown during the course of this work, and only recently started to be explored in more detail. 29 CHAPTER 1. GENERAL INTRODUCTION AND AIMS The homeobox transcription factor CDX2 is determinant for proper embryonic development, as well as for endodermal induction and adult maintenance of the intestinal epithelium. Besides this homeostatic function, in vitro experiments, animal models and human lesions show that de novo expression of CDX2 is associated with ectopic foci of intestinal differentiation, which in turn conveys an increased risk of cancer development. A paradigmatic example of this link is provided by gastric IM, a lesion that encompasses a tissue adaptive response to H. pylori infection and subsequent chronic inflammatory reaction. The role of bacterial eradication as a common preventive strategy against gastric cancer development remains debatable and largely dependent on the extent of preneoplastic changes at the time of treatment, given the apparent stability of the CDX2-dependent IM phenotype. Moreover, the nature of CDX2 involvement in the establishment and/or progression of CRC is far from being consensual. Early data suggested that loss of CDX2 was a common event; however, ensuing studies have shown that its expression is retained and in some cases even increased. Therefore, defining a thorough portrayal of the CDX2 regulatory network stands out as a leading investigation priority, not only to advance its specific contribution to these carcinogenic processes, but most importantly to apply the acquired knowledge in the design of alternative and perhaps more target-oriented therapeutic strategies. The present work, framed within the abovementioned conceptions, intends to uncover new molecular mechanisms of CDX2 regulation focusing on the role of yet poorly-described microenvironmental influences as a research guiding line. In this regard, the following specific objectives are proposed: 1. Evaluate if methylation at the promoter level is a mechanism involved in the regulation of CDX2 gene expression DNA methylation is an epigenetic mechanism of transcriptional regulation and a direct measure of the environmental effect over the genome, with an involvement in cancer attributed to the inappropriate silencing of tumour suppressor genes (hypermethylation), or loss of oncogene repression (hypomethylation). Previous studies reported that the methylation status of CDX2 might have a relevant regulatory role in different contexts. But they present discrepancies, both concerning the region of query and the methods employed to analyse it. The lack of 1.5 LONG-TERM GOAL AND SPECIFIC AIMS CHAPTER 1. GENERAL INTRODUCTION AND AIMS a comprehensive analysis led us to evaluate if methylation at the promoter level could be a mechanism involved in the regulation of CDX2 expression, and the obtained results are presented in Chapter 2. We hypothesized that in intestine, where CDX2 is expressed, the gene would be unmethylated, while in the normal gastric mucosa, where CDX2 is not detected, the gene would be methylated. The process of H. pylori infection and accompanying inflammatory response would in some way be responsible for promoter demethylation, constituting the triggering event leading to the development of IM. The alteration in CDX2 methylation profile along the gastric carcinogenic process would constitute a proof of concept, undoubtedly establishing the relevance of this mechanism and contributing to settle a contentious question in the field. 2. Assess the role of cell-matrix interactions in the regulation of CDX2 expression For many years now, traditional methods of cell culture have produced important conceptual advances in cancer research. Nevertheless, cells grown on flat substrates can differ considerably in their morphology, differentiation, cell-cell and cell-matrix interactions from those growing in more physiological conditions. At the other end of the spectrum we have whole-animal models, which frequently corroborate the importance of particular processes. But these are complex, expensive and not easily manipulated. By mimicking features of the in vivo microenvironment and taking advantage of the same tools used to study cells in standard culture, 3D cell/tissue models have been bridging the gap between both, providing unique perspectives on the behaviour of stem cells, epithelial tissues and tumours. Earlier publications described the role of epithelial-mesenchymal interactions as determinant for the expression of CDX2, not only for proper spatio-temporal induction of intestinal differentiation, but also in modulating its levels during carcinogenesis. However, the direct molecular mediators responsible for this adaptable response remained for the most part elusive. In Chapter 3, we present results regarding the establishment of in vitro 3D cell culture models comprising gastric cancer cell lines and an extracellular matrix. We hypothesized that this approach, exploratory in nature, would allow pinpointing regulatory factors and signalling pathways involved in the control of CDX2 expression. 31 CHAPTER 2. CDX2 PROMOTER METHYLATION 1740 PEREIRA ET AL. F IGURE 1 – Methylation status of CDX2 in a panel of human gastric cancer cell lines and functional relation with CDX2 mRNA expression. (a) Schematic representation of the 5`proximal flanking region of CDX2 gene. Thin vertical lines mark the location of CpG dinucleotides. A box indicates the first exon, including non-coding (white) and coding (black) sequences. Two putative CpG islands were identified with the online bioinformatic tool CpGPlot (www.ebi.ac.uk/tools/emboss/cpgplot), one in the promoter region (–1603bp to –1197bp) and another in the first exon, including the coding region (–22bp to +660bp). Arrow-heads indicate the position of bisulfite specific primers. (b) RT-PCR analysis of CDX2 mRNA expression in gastric cancer cell lines. GAPDH mRNA expression was used as an internal loading control and NTC indicates no template control. (c) Methylation profile of CDX2 in gastric cancer cell lines obtained by direct sequencing of bisulfite-treated genomic DNA. M, methylated; U, unmethylated. (d) Real-time PCR for CDX2 expression in human gastric cancer cell lines after treatment with the demethylating agent 5-aza-2`-deoxycytidine (5azaDc, Sigma) during different time-points. Each experiment was carried out in triplicate at least twice; the results are expressed as mean ± SD of biological replicas. The values obtained with vehicle-treated cells were referred to as 1. CDX2 mRNA levels were normalized with the corresponding 18S rRNA levels. CHAPTER 2. CDX2 PROMOTER METHYLATION CDX2 PROMOTER METHYLATION AND mRNA EXPRESSION 1741 F IGURE 2 – Bisulfite DNA sequencing of the first and second fragments of CDX2 CpG island 2 in two specimens of the gastric mucosa and adjacent IM foci (S1 and S2) and in two specimens of normal colonic mucosa (S3 and S4). Only clearly identifiable normal gastric glands and metaplastic glands were selected using a PALM Microbeam Microscope (Zeiss). With this methodology we were able to minimize stromal cell contamination in our tissue samples. PCR amplicons from the tissues were cloned into pCR4-TOPO vector using the TOPO TA Cloning kit (Invitrogen), and at least five individual clones from each sample were sequenced using the ABI Prism BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems) with a reverse M13 primer. Each horizontal row of circles represents analysis in a single clone of bisulfite-treated DNA of the 42 CpG sites contained in the mentioned region. Solid and open circles represent methylated and unmethylated CpG sites, respectively. 39 CHAPTER 2. CDX2 PROMOTER METHYLATION 1742 PEREIRA ET AL. References 1. Yuasa Y, Nagasaki H, Akiyama Y, Hashimoto Y, Takizawa T, Kojima K, Kawano T, Sugihara K, Imai K, Nakachi K. DNA methylation status is inversely correlated with green tea intake and physical activity in gastric cancer patients. Int J Cancer 2009;124:2677–82. 2. Almeida R, Silva E, Santos-Silva F, Silberg DG, Wang J, De Bol'os C, David L. Expression of intestine-specific transcription factors. CDX1 and CDX2, in intestinal metaplasia and gastric carcinomas. J Pathol 2003;199:36–40. 3. Mesquita P, Raquel A, Nuno L, Reis CA, Silva LF, Serpa J, van Seuningen I, Barros H, David L. Metaplasia—a transdifferentiation process that facilitates cancer development: the model of gastric intestinal metaplasia. Crit Rev Oncog 2006;12:3–26. 4. Guo M, House MG, Suzuki H, Ye Y, Brock MV, Lu F, Liu Z, Rustgi AK, Herman JG. Epigenetic silencing of CDX2 is a feature of squamous esophageal cancer. Int J Cancer 2007;121:1219–26. 5. Yuasa Y, Nagasaki H, Akiyama Y, Sakai H, Nakajima T, Ohkura Y, Takizawa T, Koike M, Tani M, Iwai T, Sugihara K, Imai K, et al. Relationship between CDX2 gene methylation and dietary factors in gastric cancer patients. Carcinogenesis 2005;26:193–200. 6. Kawai H, Tomii K, Toyooka S, Yano M, Murakami M, Tsukuda K, Shimizu N. Promoter methylation downregulates CDX2 expression in colorectal carcinomas. Oncol Rep 2005;13:547–51. 7. Liu T, Zhang X, So CK, Wang S, Wang P, Yan L, Myers R, Chen Z, Patterson AP, Yang CS, Chen X. Regulation of Cdx2 expression by promoter methylation, and effects of Cdx2 transfection on morphology and gene expression of human esophageal epithelial cells. Carcinogenesis 2007;28:488–96. 8. Fujii S, Ochiai A. Enhancer of zeste homolog 2 downregulates E-cadherin by mediating histone H3 methylation in gastric cancer cells. Cancer Sci 2008;99:738–46. 41 43 CDX2 post-transcriptional regulation by the RNA-binding protein MEX3A Chapter 3 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A The results concerning this chapter are published in: Pereira B, Sousa S, Barros R, Carreto L, Oliveira P, Oliveira C, Chartier NT, Plateroti M, Rouault JP, Freund JN, Billaud M, Almeida R. (2013) CDX2 regulation by the RNAbinding protein MEX3A: impact on intestinal differentiation and stemness. Nucleic Acids Research, 41(7):3986-3999 doi: 10.1093/nar/gkt087 This work was supported by Fundação para a Ciência e a Tecnologia (FCT) – Programa Operacional Ciência e Inovação 2010 do Quadro Comunitário de Apoio III and FEDER [PTDC/SAU-OBD/64490/2006]. __________________________________________________________________________________ The author declares that he performed and/or was involved in all the experimental work, participated in the study design and wrote the manuscript. Microarray experiments were conducted at NFDM (National Facility for DNA Microarrays, Universidade de Aveiro, Portugal) and 3D cell cyst assays at Institut Albert Bonniot (INSERM-UJF U823, Grenoble, France). CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A 3986–3999 Nucleic Acids Research, 2013, Vol. 41, No. 7 Published online 13 February 2013 doi:10.1093/nar/gkt087 CDX2 regulation by the RNA-binding protein MEX3A: impact on intestinal differentiation and stemness Bruno Pereira 1 , Sofia Sousa 1 , Rita Barros 1 , Laura Carreto 2 , Patrícia Oliveira 1 , Carla Oliveira 1,3 , Nicolas T. Chartier 4 , Michelina Plateroti 5 , Jean-Pierre Rouault 6 , Jean-Noël Freund 7 , Marc Billaud 4 and Raquel Almeida 1,3, * 1IPATIMUP – Institute of Molecular Pathology and Immunology of the University of Porto, 4200-465 Porto, Portugal, 2Department of Biology and CESAM, RNA Biology Laboratory, University of Aveiro, 3810-193 Aveiro, Portugal, 3FMUP – Faculty of Medicine, University of Porto, 4200-319 Porto, Portugal, 4INSERM-UJF U823, Institut Albert Bonniot, BP 170, 38042 Grenoble Cedex 9, France, 5Centre de Génétique et de Physiologie Moléculaire et Cellulaire, Université Claude Bernard Lyon 1, UMR5534, 69622 Villeurbanne, France, 6Institut de Génomique Fonctionnelle de Lyon, UMR5242 CNRS/INRA/UCBL/ENS Ecole Normale Supérieure de Lyon, 69364 Lyon Cedex 07, France and 7INSERM UMR_S1113, Université de Strasbourg, Fédération de Médecine Translationnelle, 67200 Strasbourg, France Received November 5, 2012; Revised January 22, 2013; Accepted January 23, 2013 ABSTRACT The homeobox transcription factor CDX2 plays a crucial role in intestinal cell fate specification, both during normal development and in tumorigenic processes involving intestinal reprogramming. The CDX2 regulatory network is intricate, but it has not yet been fully uncovered. Through genome-wide screening of a 3D culture system, the RNA-binding protein MEX3A was identified as putatively involved in CDX2 regulation; therefore, its biological relevance was addressed by setting up cell-based assays together with expression studies in murine intestine. We demonstrate here that MEX3A has a repressive function by controlling CDX2 levels in gastric and colorectal cellular models. This is dependent on the interaction with a specific binding determinant present in CDX2 mRNA 3`untranslated region. We have further determined that MEX3A impairs intestinal differentiation and cellular polarization, affects cell cycle progression and promotes increased expression of intestinal stem cell markers, namely LGR5, BMI1 and MSI1. Finally, we show that MEX3A is expressed in mouse intestine, supporting an in vivo context for interaction with CDX2 and modulation of stem cell properties. Therefore, we describe a novel CDX2 post-transcriptional regulatory mechanism, through the RNA-binding protein MEX3A, with a major impact in intestinal differentiation, polarity and stemness, likely contributing to intestinal homeostasis and carcinogenesis. INTRODUCTION The homeodomain transcription factor CDX2 is a critical determinant of intestinal homeostasis, both during development and throughout adult life. CDX2 is involved in the antero-posterior patterning of the mammalian embryo and is the key molecular mediator of intestinal differentiation (1–4). Furthermore, multiple evidences substantiate CDX2 crucial role in carcinogenesis of the digestive tract. It was shown to inhibit cell growth and migration in vitro as well as dissemination of colon tumour cells in vivo (5). CDX2 heterogeneous loss has also been observed in colorectal carcinomas (CRCs), particularly in invasive cells at the tumour edge (6). Moreover, CDX2 reduction increases the progression of chemically induced CRCs (7). Conversely, under certain pathological conditions, CDX2 becomes abnormally expressed in other organs of the digestive tract besides intestine, namely the esophagus *To whom correspondence should be addressed. Tel: +351 2255 70700; Fax: +351 2255 70799; Email: [email protected] Present address: Sofia Sousa, School of Pharmacy, Faculty of Health Sciences, University of Eastern Finland, Yliopistonranta 1 C, FIN-70211 Kuopio, Finland. © The Author(s) 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/3.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. 45 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A Nucleic Acids Research, 2013, Vol. 41, No. 7 3987 (8) and stomach (9,10), driving a precancerous lesion known as intestinal metaplasia, a process confirmed in transgenic mouse models (11,12). Owing to the essential function in intestinal development, differentiation and carcinogenesis, CDX2 regulation has been extensively studied. We have previously identified different mechanisms involved in the transcriptional regulation of this gene such as the Bone morphogenetic protein (BMP) pathway (13), SOX2 (14) and a CDX2 autoregulatory loop (15). Several transcription factors including HNF4a, GATA6, TCF4 and β-catenin were shown to interact with Cdx2 promoter fragments (16). However, mutations at the CDX2 locus are a rare event in CRC (17), and its expression does not depend on methylation of the proximal promoter (18). On the other hand, CDX2 protein phosphorylation has also been shown to modify its activity in intestinal cells (19,20). These and other studies support the notion that CDX2 regulation is intricate and strictly controlled. During the past two decades, post-transcriptional regulation emerged as a fundamental mechanism guiding gene expression in higher eukaryotic cells, being at the core of normal cellular processes but also cancer initiation and development. It is now increasingly clear that RNA maturation, localization, translation and stability provide multiple layers of spatio-temporal control determining a transcript’s fate (21,22). These coupled events are generally dependent on the cooperation between cis-regulatory elements and trans-acting factors, such as non-coding RNAs and RNA-binding proteins (RBPs). RBPs have been implicated in virtually every aspect of RNA metabolism (22,23), particularly, their repressive role is critical to establish precise translational patterns that define developmental and differentiation switches in many organisms. In Caenorhabditis elegans, MEX-3 is a translational repressor that regulates blastomere identity during early embryogenesis (24) and germline totipotency in the adult worm (25). MEX-3 has two K homology domains, which are conserved single-stranded RNA-binding motifs (26). Mutations disrupting mex-3 locus are lethal, resulting in embryos that inappropriately generate body-wall muscle from the anterior blastomere; hence, the name mex for ‘muscle excess’. This is specified, in part, by the repressive function that MEX-3 exerts over the transcription factor pal-1, the C. elegans orthologue of caudal in Drosophila and CDX in mammals (24,27). In humans, mex-3 was identified and characterized as four homologous genes, MEX3A–D (28,29), whose biological relevance is starting to be explored. Recently, the functional role of MEX3C as a RNA-binding ubiquitin E3 ligase was established, mediating the post-transcriptional decay of HLA-A allotypes (30). It was also shown to be necessary for normal postnatal growth in mutant mice by enhancing the local expression of insulin-like growth factor 1 in bone (31) and appears to be involved in metabolic regulation of energy balance (32), through yet unknown effectors. A variant form of MEX3D called TINO was shown to negatively regulate the antiapoptotic protein BCL-2 in HeLa cells (33). Finally, knockdown of MEX3A by siRNAs was shown to suppress cell proliferation and migration in human gastric cancer cells, but the molecular mechanisms behind these findings were not addressed (34). Pursuing the aim of uncovering new CDX2 regulatory mechanisms, we explored a putative translational repression by MEX3A, which was inversely correlated with CDX2 in a 3D experimental model. By studying MEX3A expression in vivo and using a cell line-based approach to modulate its levels in vitro, our study describes a novel post-transcriptional process by which CDX2 expression is impaired in the gastrointestinal setting and intestinal-like homeostasis compromised, through alterations in differentiation, polarity and stemness features. Another layer of control is thus added to the complex CDX2 regulatory network, involving MEX3A as a key regulator of intestinal homeostasis, which might have significant implications to gastrointestinal carcinogenesis. MATERIALS AND METHODS Cell culture and treatments Human gastric carcinoma cell line AGS (ATCC, American Type Culture Collection) and CRC cell line Caco-2 (ATCC) were cultured under standard conditions in RPMI-1640 medium and Dulbecco’s modified Eagle’s medium, respectively, containing 10% fetal bovine serum, 100 U/ml of penicillin and 100 µg/ml of streptomycin (Life Technologies). For the AGS 3D culture, flasks were coated with 50 µl/cm2 of matrigel basement membrane (BD Biosciences) at a 1.5:1 proportion to serum-free medium. A 2 x 104 cells/cm2 suspension was seeded on top and maintained for 14 days with medium change every 2 days. For the Caco-2 3D culture, coverslips were coated with 60 µl/cm2 of matrigel. A 6 x 103 cells/cm2 suspension plus 2% matrigel was seeded on top and maintained for 8 days with medium change every 2 days. To quantify lumen formation, >100 cysts were microscopically examined. For proteasome inhibition, cells were treated with 25 µM MG132 (Calbiochem) or vehicle treated with DMSO. To inhibit transcription, cells were exposed to 10 µg/ml of Actinomycin D (Sigma). Constructs and site-directed mutagenesis The previously published pCMV-MEX3A expression vector (28) was used together with a pCMV-Tag3B empty vector (Agilent Technologies) in transfections. A pRLControl construct containing a humanized Renilla luciferase (Rluc) coding sequence (35) was used as a backbone to create the pRLCDX2 vector, encoding a luciferase fusion transcript to the parental CDX2 3`untranslated region (UTR). The QuickChange sitedirected mutagenesis kit (Stratagene) was then used to introduce specific mutations in the previous plasmid, to generate the pRL∆CDX2 construct, with a mutated MEX-3 recognition element (MRE). Oligonucleotides containing the desired mutations were designed according to the manufacturer’s instructions (Supplementary Table S1). Taking into account the degenerate consensus sequence described for MEX-3 binding in C. elegans, a mutational background of Cytidine was used because presumably this base is not tolerated in the MRE (36). Nucleic Acids Research, 2013, Vol. 41, No. 7 3987 3988 Nucleic Acids Research, 2013, Vol. 41, No. 7 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A Transfections, RNAi and luciferase assays Transient transfections were done using Lipofectamine 2000 reagent according to the manufacturer’s guidelines (Life Technologies). A DNA (µg) to Lipofectamine 2000 reagent (µl) ratio of 1:1.5 in OPTI-MEM medium was used for routine transfection experiments of expression vectors. For stable transfections, selection was initiated 48h post-transfection in medium supplemented with 0.6 mg/ml of G418 (Sigma). Neomycin-resistant positive clones obtained through limiting dilution were routinely maintained with 0.2 mg/ml of G418. A commercial set of three Stealth Select siRNA duplexes (HSS150674, HSS150675 and HSS150676) directed against human MEX3A (Life Technologies) and a custom set of three siRNA duplexes directed against CDX2 were used with scrambled controls. An siRNA duplexes (pmol) to Lipofectamine 2000 reagent (µl) ratio of 20:1 in OPTIMEM medium was used for inhibition experiments. Luciferase reporter assays were performed with the Rluc Assay System (Promega), and β-galactosidase activity was used for normalization of experimental variations. RNA isolation and quantitative real-time PCR Total RNA was extracted using TRI Reagent (Sigma) and reverse transcribed using the Superscript II Reverse Transcriptase kit (Life Technologies). Analysis of BMI1, CDX2, GAPDH, LGR5, MEX3A and Rluc mRNA expression was performed in an ABI Prism 7500 system using SYBRgreen reagent (Life Technologies) and specific primer pairs (Supplementary Table S1). Each sample was amplified in triplicate and specificity confirmed by dissociation analysis. The 18S rRNA expression was measured for normalization of target gene abundance. Microarrays and data processing Experiments were performed at the National Facility for DNA Microarrays (Aveiro, Portugal). Three independent 2D and 3D AGS cell cultures were selected for total RNA extraction using TRI Reagent. RNA quantity and quality were assessed using the Nanodrop ND-1000 (Thermo Scientific) and 2100 Bioanalyzer (Agilent Technologies) systems, and only samples with a RNA integrity number above nine were considered for further study. Preparation and labelling of the RNA was performed using the One-Color Microarray-Based Gene Expression Analysis Quick Amp Labelling kit (Agilent Technologies). Briefly, 600 ng of total RNA from 2D and 3D biological replicas was used as input together with spike-in controls to generate Cyanine 3-labelled cRNA. The amplified cRNA samples were purified with RNeasy Mini kit (Qiagen) and hybridized to Human Gene expression 4 x 44K v2 Microarray slides (Agilent Technologies) at 65 o C for 17h. After washing procedures, images of the hybridizations were acquired using a G2505B Microarray Scanner (Agilent Technologies). The Feature Extraction software was used for spot identification, background subtraction and quantification of the fluorescent signal. Raw expression data were processed using BRB-ArrayTools v3.8.1 software (37). After base 2 log transformation and average of probe replicates, normalization was applied using the median intensity over the entire array to minimize systematic variance. Differentially expressed transcripts between the two culture conditions were filtered using the Class Comparison tool, performing an unpaired sample t-test with a P < 0.01 and considering a minimal 1.5-fold change. Hierarchical clustering analysis of significantly altered genes was conducted in TIGR MultiExperiment Viewer v4.8.1 software (38) using Euclidean correlation and average linkage clustering, and expression values indicated colorimetrically. Functional annotation of differentially expressed genes, identified by extending the unpaired t-test parameter to P < 0.05 and maintaining the minimal fold change was performed with DAVID program (39). Microarray data have been submitted to the ArrayExpress database (http://www.ebi. ac.uk/arrayexpress/) and assigned the identifier E-MTAB1234. The complete list of differentially expressed genes is detailed in Supplementary Table S2. Protein extraction and western blot analysis Cells were lysed for 30 min on ice in a lysis buffer containing 20 mM Tris–HCl (pH 7.5), 150 mM NaCl, 2 mM ethylenediaminetetraacetic acid and 1% IGEPAL (Sigma), supplemented with Complete protease inhibitor cocktail (Roche Applied Science), 1 mM PMSF and 1 mM Na 3 VO 4 . Lysates were centrifuged at 12 000 rpm for 20 min at 4 o C and the supernatant recovered. Protein concentration was determined using the BCA Protein Assay Reagent (Thermo Scientific). Protein extracts (30–50 µg) were run on 10% sodium dodecyl sulphate–polyacrylamide gel electrophoresis, transferred to a nitrocellulose membrane, blotted overnight with appropriate antibodies (Supplementary Table S1), and signals revealed with ECL detection kit (GE Healthcare Life Sciences). Actin levels were used to normalize protein expression, and quantification of western blots was performed using Fiji software (40). Flow cytometry Caco-2 cells were harvested 48 h after transfection at confluence and fixed with 1% paraformaldehyde for 20 min at room temperature followed by permeabilization with 0.1% Triton X-100 (Sigma) for 5 min on ice. Staining was performed with anti-MEX3A antibody and visualized with goat anti-rabbit FITC-conjugated secondary antibody. For DNA content assessment, cells were incubated with a propidium iodide (50 µg/ml) and RNAse A (200 µg/ml) solution (Sigma) for 30 min at room temperature. Samples were read in a FACSCanto II (BD Biosciences) flow cytometer, and analysis was performed using the FlowJo software. RNA-immunoprecipitation assay Cells were lysed 48 h after transfection as indicated for protein extraction but with the addition of 20 U/ml of RNAseOUT Ribonuclease inhibitor (Life Technologies). Before the lysis procedure, cells were washed with phosphate buffered saline (PBS) and subjected to ultraviolet crosslink (254-nm wavelength) at an energy level 47 Nucleic Acids Research, 2013, Vol. 41, No. 7 3995 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A Figure 7. Polarity alterations induced by MEX3A in Caco-2 cells. (A) Immunofluorescence showing ZO-1 expression in Caco-2 mock and MEX3A stably transfected cell lines at day -2 of culture (original magnification, x400). (B) Morphology of Caco-2 mock and MEX3A cysts in bright field microscopy during 3D culture (original magnification, x100) and CDX2/MEX3A expression (original magnification, x630). (C) Quantification of cysts with lumen or no lumen at culture day 8. (D) Expression of E-cadherin and Phalloidin staining in Caco-2 cysts (original magnification, x630; all scale bars 20 µm). 3D cell culture systems have been shown to enable physiological and functional differentiation of several epithelial cell types (47,48), constituting a promising alternative to overcome standard cell culture limitations. Accordingly, our transcriptomic analysis performed on a 3D model of AGS cells, which have a significant level of endogenous CDX2 responsive to different molecular stimuli (13–15), allowed us to disclose MEX3A as a molecular player involved in the regulation of CDX2 translation. The array presented an increased MEX3A expression in 3D culture and no alteration in the levels of the other MEX3 family members. We further demonstrated that MEX3A overexpression leads to marked CDX2 protein decrease in two cell lines. We tried to overcome the limitation of using overexpression systems by using two different cell lines that were both transiently and stably transfected with a MEX3A plasmid, which gave concordant results. In addition, CDX2 negative regulation by MEX3A was confirmed in a more physiological context using a siRNA approach towards endogenous MEX3A in Caco-2 cells. We proved that MEX3A is able to interact with CDX2 mRNA through a canonical MRE present CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A 3996 Nucleic Acids Research, 2013, Vol. 41, No. 7 Figure 8. Expression patterns of MEX3A and CDX2 in mouse normal intestine. Representative immunofluorescence data for MEX3A and CDX2 in small intestine and colon are shown (original magnification, x200; scale bars 50 µm; insert original magnification, x400). in the 3´UTR. This sequence, which we now show to be functionally relevant in humans, seems to be the single determinant of MEX3A binding, independently of the upstream 5`coding region. In fact, bioinformatics analysis shows that the degenerate MRE motif has been evolutionarily conserved in different CDX2 homologues (Supplementary Figure S3), namely in chimpanzee, mouse, rat, zebrafish, fruit fly and frog, suggesting that MEX3A is critical for CDX2 regulation. Endogenously, MEX3A was mainly localized in the nuclear compartment of intestinal epithelial cells, both in vitro and in vivo. This result is crucial to show that a biological background for the regulation of CDX2 by MEX3A exists. In this regard, a variant form of MEX3D 55 Nucleic Acids Research, 2013, Vol. 41, No. 7 3997 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A called TINO, which has been shown to negatively regulate BCL-2 expression by transcript destabilization, is also predominantly localized in the nuclei of HeLa cells (33). On the other hand, MEX3A transfectants showed predominant cytoplasmic staining. A potential problem could be antibody specificity; however, this was successfully evaluated by transfection of siRNA duplexes directed against MEX3A (Supplementary Figure S4). Therefore, the differential localization in distinct expression backgrounds might be because a certain threshold in expression levels has to be achieved for cytoplasmic translocation to occur, given that MEX3 proteins are capable of performing nucleocytoplasmic shuttling (28). On the other hand, specific subcellular distribution might be related with protein phosphorylation, in accordance with published data describing MEX3 members as phosphoproteins (28,29). Whichever the case, this does not contradict post-transcriptional regulation, as it can be elicited at multiple points of the transcript lifespan, including pre-mRNA processing in the nucleus, export from the nucleus to the cytoplasm and subsequent coordinated trafficking of the mature mRNA to the translation machinery. It also remains to be fully clarified the consequence of partial accumulation in P bodies, structures involved in processes of mRNA degradation, nonsensemediated mRNA decay, translational repression and RNA-mediated gene silencing (49), although there are several RBPs with established roles in translational regulation known to colocalize with P bodies (50,51). We used the Caco-2 cell line model to modulate MEX3A expression and assess its phenotype. Surprisingly, MEX3A inhibition produced distinct effects over CDX2 levels depending on cellular confluence. MEX3A might selectively regulate unique subsets of targets in different culture conditions. In agreement, it is known that transcriptomic and proteomic changes occur during the progression from the proliferative state to spontaneous differentiation of Caco-2 cells. Conversely, MEX3A might have a dual role, acting both as a repressor or enhancer contingent on the cellular microenvironment, as observed for other RBPs, like HuR (52). It is possible that intricate associations of MEX3A with other molecular effectors, namely other proteins or microRNAs, determine divergent regulatory endings. MEX3A overexpression in Caco-2 cells resulted in pronounced phenotypic alterations. CDX2 and Villin downregulation were indicative of a loss in intestinal differentiation, which was further confirmed by the flow cytometric profile, showing reduced G0/G1 population, usually associated with less-differentiated cells. Another hallmark feature of MEX3A overexpression was the altered cellular polarization in standard culture, as well as the impaired ability to form polarized structures in the presence of matrigel. These effects may be mediated by CDX2, as this transcription factor was previously shown to regulate intestinal Villin through recruitment of the Brm-type SWI/SNF complex to its promoter (53), and MEX3A-expressing cysts closely resemble the ones obtained with Caco-2 cells in which CDX2 suppression was achieved by lentiviral short-hairpin RNA particles (4). It is, therefore, important to ascertain the biological setting where this regulation might be determinant, which is suggested by the predominant expression of MEX3A in the stem, transit-amplifying and migrating post-mitotic cells of the intestine. Although CDX2 protein can be detected in most of these cells, its level is lower in the crypts compared with the uppermost differentiated cells of the villi (2,19,20). A similar increasing bottom-up gradient has been described along the colonic gland axis in the distal colon epithelium. By contrast, in situ hybridization revealed that CDX2 mRNA was homogeneously distributed along the entire crypt-villus axis (20). Given this CDX2 protein expression gradient, that we also show, and lack of correlation with mRNA, it is likely that MEX3A fine-tunes CDX2 levels in vivo as well, in a transcription-independent manner, providing swift availability of the protein to meet the physiological requirements of the continuously renewed gut epithelium. Moreover, it has been shown that intestinal stem cells cannot differentiate into any of the intestinal lineages in a background of Cdx2 ablation (54), revealing the need for tight CDX2 regulation as determinant to proper phenotype switching. Most interestingly, we observed higher expression of different intestinal stem cell markers when we overexpressed MEX3A in Caco-2, suggesting that this protein is associated with stem cell features. Our AGS 3D model also showed a significant upregulation of OLFM4 (Supplementary Table S2), previously identified as a marker for LGR5+ stem cells in human intestine (46), although this gene was not directly upregulated in Caco-2 cells, which might be due to intrinsic properties of each cell line. Strengthening the hypothesis for a role of MEX3A in stem cell potential, a recent publication showed that MEX3A is part of the molecular signature of the LGR5+ intestinal stem cells, presenting a 1.64-fold increase in relation to daughter cells, along with MSI1, LGR5 and OLFM4 (55). Furthermore, our study reinforces the increasing knowledge that other regulatory mechanisms in addition to transcriptional ones have important functions in stem cells. Although we do not know yet how MEX3A relates with the intestinal stem cell phenotype, we hypothesize that MEX3A overexpression in itself, together with the induction of a lessened differentiated phenotype and polarity defects mediated by CDX2 downregulation, might be critical to allow a permissive environment for the appearance of stemness features. Furthermore, the MEX3A–CDX2 axis might be important during early embryogenesis where CDX2 is required for correct trophectoderm differentiation, while absent from the inner cell mass, in a process that requires tight regulation (56). Interestingly, the pattern of expression of the C. elegans orthologues of both proteins was found to be mutually exclusive in early stages of embryogenesis (24,27). Still, it remains to be assessed the relevance of MEX3A in multiple pathological contexts of the gastrointestinal tract where differentiation abnormalities directed by CDX2 are key events. In conclusion, we have identified a novel role for MEX3A protein in the regulation of intestinal differentiation, polarity and stemness features, partially mediated by the repression of CDX2. This is the first description of a CDX2 regulatory mechanism based on its mRNA control by an RBP, having a significant impact in CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A 3998 Nucleic Acids Research, 2013, Vol. 41, No. 7 intestinal homeostasis and likely in gastrointestinal carcinogenesis. ACCESSION NUMBERS The microarray data from this publication have been submitted to the ArrayExpress database (http://www.ebi. ac.uk/arrayexpress/) and assigned the identifier E-MTAB-1234. SUPPLEMENTARY DATA Supplementary Data are available at NAR Online: Supplementary Tables 1 and 2 and Supplementary Figures 1–4. ACKNOWLEDGEMENTS The authors are grateful to Professor Leonor David for fruitful discussion, valuable experimental advice throughout the project and critical reading of the manuscript. They thank Dr Catarina Leitão for technical assistance in flow cytometry and Dr Patrícia Castro for technical assistance in confocal microscopy. The overall study was coordinated by R.A. with contributions from M.B.; B.P., S.S., R.B. and L.C. performed the experiments; M.P., JP.R. and JN.F. provided essential reagents; B.P. and R.A. conceived and designed the experiments; B.P. and R.A. wrote the manuscript; all authors contributed to different aspects of data analysis and critical revision of the manuscript. FUNDING Fundação para a Ciência e a Tecnologia (FCT)— Programa Operacional Ciência e Inovação 2010 do Quadro Comunitário de Apoio III and FEDER [PTDC/ SAU-OBD/64490/2006]. IPATIMUP is an Associate Laboratory of the Portuguese Ministry of Science, Technology and Higher Education and is partially supported by FCT. B.P. and R.B. acknowledge FCT for financial support [SFRH/BD/43168/2008 and SFRH/BPD/ 68276/2010, respectively]. M.B. was supported by a grant from the Fondation ARC, and N.T.C. was the recipient of a fellowship from the Fondation pour la Recherche Médicale. Funding for open access charge: FCT [SFRH/ BD/43168/2008]. Conflict of interest statement. None declared. REFERENCES 1. Beck,F., Chawengsaksophak,K., Waring,P., Playford,R.J. and Furness,J.B. (1999) Reprogramming of intestinal differentiation and intercalary regeneration in Cdx2 mutant mice. Proc. Natl Acad. Sci. USA, 96 , 7318–7323. 2. Silberg,D.G., Swain,G.P., Suh,E.R. and Traber,P.G. (2000) Cdx1 and cdx2 expression during intestinal development. Gastroenterology, 119 , 961–971. 3. Gao,N., White,P. and Kaestner,K.H. (2009) Establishment of intestinal identity and epithelial-mesenchymal signaling by Cdx2. Dev. Cell, 16, 588–599. 4. Gao,N. and Kaestner,K.H. (2010) Cdx2 regulates endo-lysosomal function and epithelial cell polarity. Genes Dev., 24 , 1295–1305. 5. Gross,I., Duluc,I., Benameur,T., Calon,A., Martin,E., Brabletz,T., Kedinger,M., Domon-Dell,C. and Freund,J.N. (2008) The intestine-specific homeobox gene Cdx2 decreases mobility and antagonizes dissemination of colon cancer cells. Oncogene, 27 , 107–115. 6. Brabletz,T., Spaderna,S., Kolb,J., Hlubek,F., Faller,G., Bruns,C.J., Jung,A., Nentwich,J., Duluc,I., Domon-Dell,C. et al. (2004) Down-regulation of the homeodomain factor Cdx2 in colorectal cancer by collagen type I: an active role for the tumor environment in malignant tumor progression. Cancer Res., 64 , 6973–6977. 7. Bonhomme,C., Duluc,I., Martin,E., Chawengsaksophak,K., Chenard,M.P., Kedinger,M., Beck,F., Freund,J.N. and DomonDell,C. (2003) The Cdx2 homeobox gene has a tumour suppressor function in the distal colon in addition to a homeotic role during gut development. Gut, 52 , 1465–1471. 8. Eda,A., Osawa,H., Satoh,K., Yanaka,I., Kihira,K., Ishino,Y., Mutoh,H. and Sugano,K. (2003) Aberrant expression of CDX2 in Barrett’s epithelium and inflammatory esophageal mucosa. J. Gastroenterol., 38, 14–22. 9. Almeida,R., Silva,E., Santos-Silva,F., Silberg,D.G., Wang,J., De Bolós,C. and David,L. (2003) Expression of intestine-specific transcription factors, CDX1 and CDX2, in intestinal metaplasia and gastric carcinomas. J. Pathol., 199 , 36–40. 10. Barros,R., Camilo,V., Pereira,B., Freund,J.N., David,L. and Almeida,R. (2010) Pathophysiology of intestinal metaplasia of the stomach: emphasis on CDX2 regulation. Biochem. Soc. Trans., 38, 358–363. 11. Silberg,D.G., Sullivan,J., Kang,E., Swain,G.P., Moffett,J., Sund,N.J., Sackett,S.D. and Kaestner,K.H. (2002) Cdx2 ectopic expression induces gastric intestinal metaplasia in transgenic mice. Gastroenterology, 122 , 689–696. 12. Mutoh,H., Hakamata,Y., Sato,K., Eda,A., Yanaka,I., Honda,S., Osawa,H., Kaneko,Y. and Sugano,K. (2002) Conversion of gastric mucosa to intestinal metaplasia in Cdx2-expressing transgenic mice. Biochem. Biophys. Res. Commun., 294, 470–479. 13. Barros,R., Pereira,B., Duluc,I., Azevedo,M., Mendes,N., Camilo,V., Jacobs,R.J., Paulo,P., Santos-Silva,F., van Seuningen,I. et al. (2008) Key elements of the BMP/SMAD pathway co-localize with CDX2 in intestinal metaplasia and regulate CDX2 expression in human gastric cell lines. J. Pathol., 215, 411–420. 14. Camilo,V., Barros,R., Sousa,S., Magalhães,A.M., Lopes,T., Santos,A.M., Pereira,T., Figueiredo,C., David,L. and Almeida,R. (2012) Helicobacter pylori and the BMP pathway regulate CDX2 and SOX2 expression in gastric cells. Carcinogenesis, 10 , 1985– 1992. 15. Barros,R., da Costa,L.T., Pinto-de-Sousa,J., Duluc,I., Freund,J.N., David,L. and Almeida,R. (2011) CDX2 autoregulation in human intestinal metaplasia of the stomach: impact on the stability of the phenotype. Gut., 60 , 290–298. 16. Benahmed,F., Gross,I., Gaunt,S.J., Beck,F., Jehan,F., Domon-Dell,C., Martin,E., Kedinger,M., Freund,J.N. and Duluc,I. (2008) Multiple regulatory regions control the complex expression pattern of the mouse Cdx2 homeobox gene. Gastroenterology, 135 , 1238–1247. 17. Woodford-Richens,K.L., Halford,S., Rowan,A., Bevan,S., Aaltonen,L.A., Wasan,H., Bicknell,D., Bodmer,W.F., Houlston,R.S. and Tomlinson,I.P. (2001) CDX2 mutations do not account for juvenile polyposis or Peutz-Jeghers syndrome and occur infrequently in sporadic colorectal cancers. Br. J. Cancer, 84, 1314–1316. 18. Pereira,B., Oliveira,C., David,L. and Almeida,R. (2009) CDX2 promoter methylation is not associated with mRNA expression. Int. J. Cancer, 125 , 1739–1742. 19. Rings,E.H., Boudreau,F., Taylor,J.K., Moffett,J., Suh,E.R. and Traber,P.G. (2001) Phosphorylation of the serine 60 residue within the Cdx2 activation domain mediates its transactivation capacity. Gastroenterology, 121 , 1437–1450. 20. Boulanger,J., Vézina,A., Mongrain,S., Boudreau,F., Perreault,N., Auclair,B.A., Lainé ,J., Asselin,C. and Rivard,N. (2005) Cdk2-dependent phosphorylation of homeobox transcription 57 Nucleic Acids Research, 2013, Vol. 41, No. 7 3999 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A factor CDX2 regulates its nuclear translocation and proteasome-mediated degradation in human intestinal epithelial cells. J. Biol. Chem., 280 , 18095–18107. 21. Moore,M.J. (2005) From birth to death: the complex lives of eukaryotic mRNAs. Science, 309 , 1514–1518. 22. Besse,F. and Ephrussi,A. (2008) Translational control of localized mRNAs: restricting protein synthesis in space and time. Nat. Rev. Mol. Cell. Biol., 9 , 971–980. 23. Dreyfuss,G., Kim,V.N. and Kataoka,N. (2002) Messenger-RNAbinding proteins and the messages they carry. Nat. Rev. Mol. Cell. Biol., 3 , 195–205. 24. Draper,B.W., Mello,C.C., Bowerman,B., Hardin,J. and Priess,J.R.(1996) MEX-3 is a KH domain protein that regulates blastomere identity in early C. elegans embryos. Cell, 87 , 205– 216. 25. Ciosk,R., DePalma,M. and Priess,J.R. (2006) Translational regulators maintain totipotency in the Caenorhabditis elegans germline. Science, 311 , 851–853. 26. Siomi,H., Matunis,M.J., Michael,W.M. and Dreyfuss,G. (1993) The pre-mRNA binding K protein contains a novel evolutionary conserved motif. Nucleic Acids Res., 21, 1193–1198. 27. Hunter,C.P. and Kenyon,C. (1996) Spatial and temporal controls target pal-1 blastomere-specification activity to a single blastomere lineage in C. elegans embryos. Cell, 87 , 217–226. 28. Buchet-Poyau,K., Courchet,J., Le Hir,H., Séraphin,B., Scoazec,J.Y., Duret,L., Domon-Dell,C., Freund,J.N. and Billaud,M. (2007) Identification and characterization of human Mex-3 proteins, a novel family of evolutionarily conserved RNA-binding proteins differentially localized to processing bodies. Nucleic Acids Res., 35, 1289–1300. 29. Courchet,J., Buchet-Poyau,K., Potemski,A., Brès,A., Jariel-Encontre,I. and Billaud,M. (2008) Interaction with 14-3-3 adaptors regulates the sorting of hMex-3B RNA-binding protein to distinct classes of RNA granules. J. Biol. Chem., 283 , 32131–32142. 30. Cano,F., Bye,H., Duncan,L.M., Buchet-Poyau,K., Billaud,M., Wills,M.R. and Lehner,P.J. (2012) The RNA-binding E3 ubiquitin ligase MEX-3C links ubiquitination with MHC-I mRNA degradation. EMBO J., 31 , 3596–3606. 31. Jiao,Y., Bishop,C.E. and Lu,B. (2012) Mex3c regulates insulin-like growth factor 1 (IGF1) expression and promotes postnatal growth. Mol. Biol. Cell, 23 , 1404–1413. 32. Jiao,Y., George,S.K., Zhao,Q., Hulver,M.W., Hutson,S.M., Bishop,C.E. and Lu,B. (2012) Mex3c mutation reduces adiposity and increases energy expenditure. Mol. Cell. Biol., 32 , 4350–4362. 33. Donnini,M., Lapucci,A., Papucci,L., Witort,E., Jacquier,A., Brewer,G., Nicolin,A., Capaccioli,S. and Schiavone,N. (2004) Identification of TINO: a new evolutionary conserved BCL-2 AU-rich element RNA-binding protein. J. Biol. Chem., 279 , 20154–20166. 34. Jiang,H., Zhang,X., Luo,J., Dong,C., Xue,J., Wei,W., Chen,J., Zhou,J., Gao,Y. and Yang,C. (2012) Knockdown of hMex-3A by small RNA interference suppresses cell proliferation and migration in human gastric cancer cells. Mol. Med. Report., 6 , 575–580. 35. Pillai,R.S., Bhattacharyya,S.N., Artus,C.G., Zoller,T., Cougot,N.,Basyuk,E., Bertrand,E. and Filipowicz,W. (2005) Inhibition of translational initiation by Let-7 MicroRNA in human cells. Science, 309 , 1573–1576. 36. Pagano,J.M., Farley,B.M., Essien,K.I. and Ryder,S.P. (2009) RNA recognition by the embryonic cell fate determinant and germline totipotency factor MEX-3. Proc. Natl Acad. Sci. USA, 106, 20252–20257. 37. Simon,R., Lam,A., Li,M.C., Ngan,M., Menenzes,S. and Zhao,Y. (2007) Analysis of gene expression data using BRB-array tools. Cancer Inform., 3 , 11–17. 38. Saeed,A.I., Sharov,V., White,J., Li,J., Liang,W., Bhagabati,N., Braisted,J., Klapa,M., Currier,T., Thiagarajan,M. et al. (2003) TM4: a free, open-source system for microarray data management and analysis. Biotechniques, 34 , 374–378. 39. Huang,D.W., Sherman,B.T. and Lempicki,R.A. (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc., 4 , 44–57. 40. Schindelin,J., Arganda-Carreras,I., Frise,E., Kaynig,V., Longair,M., Pietzsch,T., Preibisch,S., Rueden,C., Saalfeld,S., Schmid,B. et al. (2012) Fiji: an open-source platform for biological-image analysis. Nat. Methods, 9 , 676–682. 41. Sheth,U. and Parker,R. (2003) Decapping and decay of messenger RNA occur in cytoplasmic processing bodies. Science, 300 , 805–808. 42. Pinto,M., Robine-Lé on,S., Appay,M.D., Kedinger,M., Triadou,N., Dussaulx,E., Lacroix,B., Simon-Assmann,P., Haffen,K., Fogh,J. et al. (1983) Enterocyte-like differentiation and polarization of the human colon carcinoma cell line Caco-2 in culture. Biol. Cell, 47, 323–330. 43. Potten,C.S., Booth,C., Tudor,G.L., Booth,D., Brady,G., Hurley,P., Ashton,G., Clarke,R., Sakakibara,S. and Okano,H. (2003) Identification of a putative intestinal stem cell and early lineage marker; musashi-1. Differentiation, 71 , 28–41. 44. Barker,N., van Es,J.H., Kuipers,J., Kujala,P., van den Born,M., Cozijnsen,M., Haegebarth,A., Korving,J., Begthel,H., Peters,P.J. et al. (2007) Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature, 449 , 1003–1007. 45. Sangiorgi,E. and Capecchi,M.R. (2008) Bmi1 is expressed in vivo in intestinal stem cells. Nat. Genet., 40 , 915–920. 46. van der Flier,L.G., Haegebarth,A., Stange,D.E., van de Wetering,M. and Clevers,H. (2009) OLFM4 is a robust marker for stem cells in human intestine and marks a subset of colorectal cancer cells. Gastroenterology, 137 , 15–17. 47. Bissell,M.J., Radisky,D.C., Rizki,A., Weaver,V.M. and Petersen,O.W. (2002) The organizing principle: microenvironmental influences in the normal and malignant breast. Differentiation, 70 , 537–546. 48. Ootani,A., Toda,S., Fujimoto,K. and Sugihara,H. (2003) Foveolar differentiation of mouse gastric mucosa in vitro. Am. J. Pathol., 162, 1905–1912. 49. Eulalio,A., Behm-Ansmant,I. and Izaurralde,E. (2007) P bodies: at the cross-roads of post-transcriptional pathways. Nat. Rev. Mol. Cell. Biol., 8 , 9–22. 50. Chu,C.Y. and Rana,T.M. (2006) Translation repression in human cells by microRNA-induced gene silencing requires RCK/p54. PLoS Biol., 4, e210. 51. Yang,W.H., Yu,J.H., Gulick,T., Bloch,K.D. and Bloch,D.B. (2006) RNA-associated protein 55 (RAP55) localizes to mRNA processing bodies and stress granules. RNA, 12 , 547–554. 52. Kim,H.H., Kuwano,Y., Srikantan,S., Lee,E.K., Martindale,J.L. and Gorospe,M. (2009) HuR recruits let-7/RISC to repress c-Myc expression. Genes Dev., 23, 1743–1748. 53. Yamamichi,N., Inada,K., Furukawa,C., Sakurai,K., Tando,T., Ishizaka,A., Haraguchi,T., Mizutani,T., Fujishiro,M., Shimomura,R. et al. (2009) Cdx2 and the Brm-type SWI/SNF complex cooperatively regulate villin expression in gastrointestinal cells. Exp. Cell Res., 315, 1779–1789. 54. Stringer,E.J., Duluc,I., Saandi,T., Davidson,I., Bialecka,M., Sato,T., Barker,N., Clevers,H., Pritchard,C.A., Winton,D.J. et al. (2012) Cdx2 determines the fate of postnatal intestinal endoderm. Development, 139, 465–474. 55. Muñoz,J., Stange,D.E., Schepers,A.G., van de Wetering,M., Koo,B.K., Itzkovitz,S., Volckmann,R., Kung,K.S., Koster,J., Radulescu,S. et al. (2012) The Lgr5 intestinal stem cell signature: robust expression of proposed quiescent ‘+4’ cell markers. EMBO J., 31 , 3079–3091. 56. Strumpf,D., Mao,C.A., Yamanaka,Y., Ralston,A., Chawengsaksophak,K., Beck,F. and Rossant,J. (2005) Cdx2 is required for correct cell fate specification and differentiation of trophectoderm in the mouse blastocyst. Development, 132 , 2093–2102. CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A SUPPLEMENTARY DATA Supplementary Figure S1. Subcellular localization of MEX3A in P bodies (A) Immunofluorescence of MEX3A stably transfected AGS cells for myc-tag and for endogenous DCP1A and EDC4 proteins. (B) Immunofluorescence of MEX3A stably transfected Caco-2 cells for myc-tag and endogenous DCP1A and EDC4 proteins. White arrow-heads point to different sites of co-localization (original magnification, x630; scale bars 20 µm). 59 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A Supplementary Figure S2. Study of the expression profile of endogenous MEX3A in Caco-2 cells. (A) Western blot of MEX3A and CDX2 expression during Caco-2 differentiation. The time-point in which confluence is achieved is referred to as day 0 (culture day 6). (B) qPCR of MEX3A and CDX2 mRNA expression during the same time-points. Values for CDX2 and MEX3A at day -2 were referred to as 1. CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A Pan troglodytes CDX2 - One predicted MRE in the 3`UTR bp Sequence 1991 AGAGTTTTTA Mus musculus Cdx2 - Two predicted MREs in the CDS bp Sequence 873 GGAGTTTCACTTTA 1171 GGAGGGGTTTTA Rattus norvegicus Cdx2 - Three predicted MREs, one in the CDS and two in the 3`UTR bp Sequence 1010 GGAGGGGTTTTA 1228 TTAGATTTTTTTTTTA 2027 GGAGCTTTA Danio rerio cdx1a - One predicted MRE in the 3`UTR bp Sequence 1176 TGAGTTTA Xenopus laevis cdx2 - Seven predicted MREs in the 3`UTR bp Sequence 1269 ATAGACTTTTA 1640 TTAGATCCCTTCTA 1860 GTAGCATTTTA 2251 GTAGGCAGCCTTTA 2355 GGAGACGTTTA 2488 TGAGTGTTTTATA 2546 GGAGGATATATTTA Drosophila melanogaster cad - Four predicted MREs in the 3´UTR bp Sequence 2235 ATAGCCGCATATATA 2251 AGAGTTTTAACGTTTA 2298 GTAGTTAATATA 2469 ATAGCTATTTA Supplementary Figure S3. Bioinformatics search for MREs in CDX2 homologues of several species. Different MRE sequences located in the coding sequence (CDS) or 3`UTR of the transcripts are shown. 61 CHAPTER 3. CDX2 POST-TRANSCRIPTIONAL REGULATION BY THE RNA-BINDING PROTEIN MEX3A Supplementary Figure S4. Inhibition of endogenous MEX3A in Caco-2 cells. (A) Immunofluorescence for endogenous MEX3A at different time-points, upon inhibition with specific siRNAs for 24h (original magnification, x400; scale bar 20 µm). (B) Corresponding western blot analysis of MEX3A downregulation. 63 71 73 Discussion and concluding remarks Chapter 5 CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS GI malignancies remain a leading cause of morbidity worldwide, affecting almost four million individuals per year. This health burden has been intensively tackled during the last decades in the search for new approaches of prevention and management, and though slow, some progress has been achieved. A major advancement in understanding cancer aetiology is the recognition that although intrinsically genetic in origin, its heterogeneous nature stems from complex interactions between environmental and host factors. At the cellular level, this implies that cancer cells are not isolated entities solely defined by the linear accumulation of irreversible genetic alterations. Carcinogenesis is a rather dynamic process characterized by ongoing adaptations to the surrounding microenvironment. The homeobox transcription factor CDX2 is a master regulator of intestinal differentiation. Not surprisingly, CDX2 deregulation is associated with GI carcinogenic processes, including IM onset during gastric cancer development and CRC, though its role in the latter is not consensual. Since structural alterations in the CDX2 locus are rare, it has become evident that regulation must account for the major alterations reported in CDX2 levels in different pathological conditions and consequently, for maintaining its normal homeostatic balance. Henceforth, efforts have been concentrated in defining the regulatory mechanisms underlying CDX2 expression. Indeed, several transcriptional, post-transcriptional and post-translational mechanisms have been described to control CDX2, making this a highly complex and tightly organized regulatory system. With the objective of unveiling new molecular mechanisms of CDX2 regulation, we focused on the role of microenvironmental influences as a guiding line. In this regard, we have studied the effect of methylation at the CDX2 promoter level and of MEX3A protein over CDX2 expression, an RNA-binding factor uncovered in a cellmatrix interaction model. The different parts of this work are now debated in distinct sections, being pointed out within each one future research avenues of interest. At the end, a general conclusion is provided, integrating all data from a molecular and evolutionary standpoint. CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS Our analysis focused on two classically defined CpG islands, identified in the 5`proximal flanking region, and thus, more likely involved in CDX2 regulation given the transcription start site proximity. We initially chose a panel of gastric cancer cell lines displaying differential expression of CDX2 mRNA to validate a bisulfite-modified DNA sequencing protocol and establish possible correlations with the CDX2 gene methyl-pattern. The choice of strategy was based on the fact that methylationspecific PCR uses different pairs of primers to specifically amplify methylated or unmethylated sequences, providing a semi-quantitative output that might not be representative of the whole CpG island, while sequencing of individual PCR amplicons with an unbiased primer pair generates extensive methylation maps at single-nucleotide resolution. Herewith, we were able to determine that promoter methylation status does not correlate with CDX2 expression levels, which has been confirmed in following studies (Varon et al. 2012; Zhang et al. 2013). This is in accordance with the CDX2 methylation frequency not showing any relationship to different clinicopathological characteristics, like tumour histological type or invasion (Yuasa et al. 2005), a fact at odds with the knowledge that gastric cancers retaining CDX2 expression are less aggressive and more differentiated (Liu et al. 2007a). Contradictory results have also been published in oesophageal tissues, with the presence of methylated and unmethylated alleles in cancer, and unmethylated in matched normal epithelia, despite no detectable mRNA or protein in the latter (Vaninetti et al. 2009). We were interested in assessing methylation in IM pre-malignant condition compared to the normal gastric mucosa. We noticed the upper CpG island was heavily methylated in all cell lines tested, consistent with a previous description (Kawai et al. 2005), and hence could not be discriminative. As a result, we limited the tissue analysis to the lower CpG island only. Analysis of normal colonic mucosa, IM foci and matched adjacent normal epithelium showed a methylation pattern deemed inconsistent with the CDX2 tissue-type expression. Hence, CDX2 expression in IM is not attributable to demethylation, but to transcriptional activation by other mechanisms. This lack of correlation has also been observed in stomachs from normal mice and IM cases from Cdx2-transgenic mice (Mutoh et al. 2009). Concerning the intestinal background, a recent report stated that CDX2 promoter hypermethylation is rare in CRC cases and cell lines (Salari et al. 2012). Therefore, it 75 5.1 CDX2 REGULATION BY PROMOTER METHYLATION CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS seems CDX2 promoter methylation is not causative for its own levels in intestine either, and cannot account for the subset of CRCs presenting heterogeneous CDX2 expression. Most studies, including our own, assumed that functionally relevant DNA methylation occurs in proximal promoter areas, but it is possible that other sites might have been neglected. In fact, bioinformatic analysis reveals that at least three additional CpG islands exist in more distal 5` intergenic locations, approximately 6kb, 8kb and 11kb from the CDX2 transcriptional start site, the last two overlapping the neighbouring gene PRHOXNB. CpG shore methylation is an emerging concept that might also be worthwhile exploring in this context. These are regions of lower CpG density that lie in close proximity, but often not within, CpG islands. Notably, and at least for colon, most cancer-associated alterations in methylation were predominantly observed in these areas, rather than restricted to CpG islands (Irizarry et al. 2009). In principle, these putative methyl-sites should be responsive to inhibitors of DNA methyltransferase activity as well. Nevertheless, the broad range nature of this type of treatments is expected to generate indiscriminate effects on multiple targets, whose net balance might not reflect increased CDX2 transcription. Thus, methylation profile analysis of these alternative regions stands as a future research goal. Another key point to be considered is the role of chromatin state. Cell typespecific chromatin organization enables differential access to and activity of regulatory elements and the manifestation of unique cellular phenotypes. Alterations in these signatures are a common feature of cancer initiation and development (Suvà et al. 2013). Active promoter regions include heightened nuclease sensitivity implying nucleosome depletion, and histone modifications associated with transcriptional activation, such as methylation of histone H3 at lysine 4 (H3K4me1 or H3K4me3) and histone H3 acetylated at lysine 27 (H3K27Ac). Quite the opposite, nucleosome compaction, methylation of histone H3 at lysine 9 (H3K9me2 or H3K9me3) and of histone H3 at lysine 27 (H3K27me3) are essentially repressive traits (Zhou et al. 2011). In this regard, chromatin changes associated with CDX1 and CDX2 expression were only described in one publication and in colon cancer cell lines (Lu et al. 2008). The in vivo pattern of these epigenetic marks, particularly in IM development, remains to be examined. CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS Given that previous reports described the role of epithelial-mesenchymal crosstalk as relevant for CDX2 expression, we established in vitro 3D culture models in order to study interactions between gastric epithelial cells and stromal elements that might regulate CDX2. These systems have now been applied with success to generate and maintain gastric and intestinal architecture in the form of organoids from single Lgr5+ stem cells (Sato et al. 2009; Barker et al. 2010). Our transcriptomic analysis comparing 3D with standard AGS cell culture revealed a remodeling of the gene expression program. Such adaptation was particularly noticeable by alterations in actin reorganization mediated by increased expression of GTPases of the Rho family, like ARHGEF9, which acts as a guanine nucleotide exchange factor for the polarity-related protein CDC42. An increase in COL4A5 and LAMB1 transcripts, for example, was also observed, which encode for type IV collagen and laminin, respectively, components of the basal lamina that has been shown to be primarily a product of epithelial cells, pointing towards an inductive feedback loop between the matrix and epithelial cells. Interestingly, we noticed a significant increase in cell cycle related elements, for instance in SGOL1, which encodes a protein that shields the cohesin complex from cleavage and is crucial for faithful chromosome segregation during mitosis and meiosis (Kahyo et al. 2011); ERCC6L, a DNA helicase that is essential for maintenance of genome integrity (Baumann et al. 2007); and the polymerase POLD1, which functions in DNA replication/repair and whose mutations affecting the proofreading domains have recently been shown to increase the risk for colorectal carcinoma (Palles et al. 2013). Transcripts encoding inflammatory response factors, like chemokine ligands, interferons and interleukins, were all downregulated. Furthermore, an increase in Notch1, DLL1, and FZD1 was detected. Combinatorial control between these Notch and Wnt pathway members seems necessary to maintain intestinal stem cells (Fre et al. 2009). In line, the array showed a 60-fold increase in the expression of OLFM4, a specific marker of Lgr5+ intestinal stem cells (van der Flier et al. 2009). The previous results, together with the decreased CDX2 expression suggest that our AGS 3D cells resemble undifferentiated intestinal cells. Since AGS was first established from an intestinal-type gastric adenocarcinoma, we speculate that a partial reversion of the malignant towards an intestinal phenotype might occur under a physiologically more relevant microenvironment, reminiscent of what has been 77 5.2 3D AGS CELL MODEL AND LOSS OF CDX2 EXPRESSION CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS observed in a mammary model (Weaver et al. 1997). The full potential of this model thus remains to be explored, as it might provide additional insights on factors involved in the regulation of intestinal differentiation. Through microarray analysis to dissect the transcriptional regulatory circuitry of AGS cells in culture with an extracellular matrix, and further supported by an evolutionary link, we identified and validated MEX3A as a CDX2 repressor. Considering the distinct experimental settings established, our interpretation is that MEX3A actively controls CDX2 expression during proliferative states, when fine-tuning of CDX2 at the transcript level seems more relevant to couple cell cycle arrest with the emergence of a differentiation program (Bai et al. 2003). In subsequent differentiated states, CDX2 expression has already been demonstrated to rely mostly in its protein stability (Boulanger et al. 2005). Hence, MEX3A might exhibit target specificity, and this hypothesis is favoured by the fact that it is still detected in the differentiated section of the crypt-villus unit, albeit weaker than in the lower portions, arguing for some activity in this compartment that might not be CDX2-oriented. Regarding the spatial compartmentalization in P bodies, we did not address if these are functionally relevant or if MEX3A effect on mRNA takes place as efficiently in the cytoplasm. Nevertheless, evolutionary comparative data supports this specific localization. C. elegans MEX-3 is involved in the correct segregation of large foci known as P granules to the final germline precursor (Draper et al. 1996; Schisa et al. 2001; Jud et al. 2007). P granules are membrane-free ribonucleoprotein cytoplasmic structures that primarily contain maternally expressed mRNAs alongside components shared with both P bodies and stress granules (Updike and Strome 2010). These three discrete microdomains are dynamically and reversibly induced in response to environmental signals. Thus, their nature seems to be in accordance with a function in maintaining or modifying certain RNA properties, delaying or even preventing the translational process. Inhibition of specific P body components through an RNAi strategy will help to shed light over MEX3A functional requirements. We showed that MEX3A interacts with CDX2 mRNA through a canonical MRE present in the 3`UTR. This 5.3 MECHANISTIC ASPECTS OF MEX3A FUNCTION CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS does not exclude the possibility of MEX3A recognizing either different variations of the degenerate sequence described for MEX-3 binding in C. elegans (Pagano et al. 2009) or even other unknown sequences; however, it provides an anchor point to search for new targets. We have not assessed if the interaction is direct or mediated by another RBP that complexes with MEX3A, either through its RING domain or even through the KH domains, as these have also been shown to mediate protein oligomerization in some instances (Chen et al. 1997). RNA electrophoretic mobility gel shift assays with a probe containing the CDX2 3`UTR MRE sequence together with a MEX3A recombinant protein will provide a definitive answer concerning this issue. Another aspect that was not advanced is MEX3A own sequence and/or structural interaction determinants. MEX-3 proteins possess type I KH domains, which consist of three-stranded antiparallel β-sheets, oriented against three α-helices. The stable β-α-α-β-β-α conformation exposes a flexible loop between the first two helices, occupied by semi-conserved positively charged residues in a Gly-X-X-Gly motif (Valverde et al. 2008). Even though the total number of solved structures with 79 CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS eight-cell stage, which precedes establishment of distinct cell populations (Ralston and Rossant 2008; Jedrusik et al. 2010). Since these cells will give rise to both the inner cell mass and TE, the corollary of this delay is that Cdx2 must be repressed during initial apolar divisions, and upon induction it must be silenced and maintained as cells are allocated to inside or outside positions, respectively. Transcriptional control in response to lineage segregation is an already established means of regulating Cdx2 (Nishioka et al. 2009), but it is not well understood how this is imposed in the first place. Asymmetric cell divisions produce daughter cells with distinct fates, and rely on the distinct segregation of key determinants, including localized mRNAs. Accordingly, it was reported that Cdx2 transcripts become preferentially localized apically at the late eight-cell stage and asymmetrically inherited during mouse development at the eightto sixteen-cell stage transition (Jedrusik et al. 2008). This was not seen for transcripts encoding other cell commitment transcription factors, like Nanog. A compelling hypothesis is that MEX3A might contribute to the initial Cdx2 mRNA repression and possibly to its apical enrichment, recapitulating C. elegans MEX-3 effect over pal-1. This possibility is strengthened by a report showing that in mouse Cdx2 asymmetric localization depends on a minimal cis-element comprising the last 97 nucleotides of the transcript open reading frame (Skamagki et al. 2013), where we had already identified one MRE and another one close upstream. Assuming this scenario, MEX3A might be involved in the regulation of the Cdx2 transcript translation in an embryonic setting or even in its transport, as it was demonstrated that intact microtubules and actin cytoskeleton, together with the activity of motor proteins of the kinesin superfamily are a requirement for this asymmetric localization (Skamagki et al. 2013). Thus, MEX3 proteins might act as conserved components of the cell polarization pathway contributing to CDX2 regulation during mammalian embryogenesis, thereby linking gene regulation and positional information. CHAPTER 5. DISCUSSION AND CONCLUDING REMARKS With this work, we have advanced the knowledge on the molecular mechanisms of CDX2 regulation. On one hand, our results support the lack of evidence in favor of a role of methylation on de novo CDX2 expression in the gastritis-metaplasia-carcinoma sequence, contributing to solve an ambiguous topic in the field. On the other hand, we provide evidence of a new mechanism of CDX2 post-transcriptional regulation mediated by the RNA-binding protein MEX3A. The relevance of MEX3A just started to be unraveled, but our data points towards its involvement in intestinal homeostasis and possibly gastrointestinal carcinogenesis, with effects in cellular polarity and stemness. In this regard, we will now ascertain the role of MEX3A by studying its expression in preneoplastic and neoplastic lesions, as well as its correlation with CDX2 and clinicopathological parameters. In parallel, we will establish a transgenic mouse model with conditional and inducible MEX3A expression in the villi compartment of the intestine, in order to assess the effects of ectopic MEX3A over the intestinal phenotype. This will be complemented by a transcriptome-wide analysis of MEX3A regulatory interactions. Overall, we expect to significantly contribute to the understanding of the impact of MEX3A posttranscriptional control on gene expression, both in homeostasis and disease. Even considering MEX-3 well-described role in C. elegans embryonic development and maintenance of cell totipotency, it is remarkable to find such a marked degree of functional parallelism in higher organisms, a feature that one can predict to reflect the importance of this protein family. Consequently, it seems we have disclosed a case of evolutionary conservation that might have major implications for basic and clinical research. 87 5.6 CONCLUDING REMARKS 89 References 91 REFERENCES Aliaga JC, Deschênes C, Beaulieu JF, Calvo EL, Rivard N. 1999. Requirement of the MAP kinase cascade for cell cycle progression and differentiation of human intestinal cells. Am J Physiol 277: G631-641. Almeida R, Silva E, Santos-Silva F, Silberg DG, Wang J, De Bolós C, David L. 2003. Expression of intestine-specific transcription factors, CDX1 and CDX2, in intestinal metaplasia and gastric carcinomas. J Pathol 199: 36-40. Anantharaman V, Koonin E, Aravind L. 2002. Comparative genomics and evolution of proteins involved in RNA metabolism. Nucleic Acids Res 30: 1427-1464. Anderson P. 2008. Post-transcriptional control of cytokine production. Nat Immunol 9: 353-359. Anderson P, Kedersha N. 2006. RNA granules. J Cell Biol 172: 803-808. Aoki K, Kakizaki F, Sakashita H, Manabe T, Aoki M, Taketo MM. 2011. Suppression of colonic polyposis by homeoprotein CDX2 through its nontranscriptional function that stabilizes p27Kip1. Cancer Res 71: 593-602. Aoki K, Tamai Y, Horiike S, Oshima M, Taketo MM. 2003. Colonic polyposis caused by mTOR-mediated chromosomal instability in Apc+/Delta716 Cdx2+/- compound mutant mice. Nat Genet 35: 323-330. Ariz M, Mainpal R, Subramaniam K. 2009. C. elegans RNA-binding proteins PUF-8 and MEX-3 function redundantly to promote germline stem cell mitosis. Dev Biol 326: 295-304. Asonuma S, Imatani A, Asano N, Oikawa T, Konishi H, Iijima K, Koike T, Ohara S, Shimosegawa T. 2009. Helicobacter pylori induces gastric mucosal intestinal metaplasia through the inhibition of interleukin-4-mediated HMG box protein Sox2 expression. Am J Physiol Gastrointest Liver Physiol 297: G312-322. Bai YQ, Miyake S, Iwai T, Yuasa Y. 2003. CDX2, a homeobox transcription factor, upregulates transcription of the p21/WAF1/CIP1 gene. Oncogene 22: 79427949. Bai YQ, Yamamoto H, Akiyama Y, Tanaka H, Takizawa T, Koike M, Kenji Yagi O, Saitoh K, Takeshita K, Iwai T et al. 2002. Ectopic expression of homeodomain protein CDX2 in intestinal metaplasia and carcinomas of the stomach. Cancer Lett 176: 47-55. Barker N, Huch M, Kujala P, van de Wetering M, Snippert HJ, van Es JH, Sato T, Stange DE, Begthel H, van den Born M et al. 2010. Lgr5(+ve) stem cells drive selfrenewal in the stomach and build long-lived gastric units in vitro. Cell Stem Cell 6: 25-36. Barker N, Ridgway RA, van Es JH, van de Wetering M, Begthel H, van den Born M, Danenberg E, Clarke AR, Sansom OJ, Clevers H. 2009. Crypt stem cells as the cells-of-origin of intestinal cancer. Nature 457: 608-611. REFERENCES Barker N, van Es JH, Kuipers J, Kujala P, van den Born M, Cozijnsen M, Haegebarth A, Korving J, Begthel H, Peters PJ et al. 2007. Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449: 1003-1007. Barker N, van Oudenaarden A, Clevers H. 2012. Identifying the stem cell of the intestinal crypt: strategies and pitfalls. Cell Stem Cell 11: 452-460. Barros R, da Costa LT, Pinto-de-Sousa J, Duluc I, Freund JN, David L, Almeida R. 2011. CDX2 autoregulation in human intestinal metaplasia of the stomach: impact on the stability of the phenotype. Gut 60: 290-298. Barros R, Freund JN, David L, Almeida R. 2012. Gastric intestinal metaplasia revisited: function and regulation of CDX2. Trends Mol Med 18: 555-563. Barros R, Marcos N, Reis CA, De Luca A, David L, Almeida R. 2009a. CDX2 expression is induced by Helicobacter pylori in AGS cells. Scand J Gastroenterol 44: 124125. Barros R, Mendes N, Howe JR, Reis CA, de Bolos C, Carneiro F, David L, Almeida R. 2009b. Juvenile polyps have gastric differentiation with MUC5AC expression and downregulation of CDX2 and SMAD4. Histochem Cell Biol 131: 765-772. Barros R, Peleteiro B, Almeida R, Figueiredo C, Barros H, David L, Lunet N. 2010. Relevance of high virulence Helicobacter pylori strains and futility of CDX2 expression for predicting intestinal metaplasia after eradication of infection. Scand J Gastroenterol 45: 828-834. Barros R, Pereira B, Duluc I, Azevedo M, Mendes N, Camilo V, Jacobs RJ, Paulo P, Santos-Silva F, van Seuningen I et al. 2008. Key elements of the BMP/SMAD pathway co-localize with CDX2 in intestinal metaplasia and regulate CDX2 expression in human gastric cell lines. J Pathol 215: 411-420. Baumann C, Körner R, Hofmann K, Nigg EA. 2007. PICH, a centromere-associated SNF2 family ATPase, is regulated by Plk1 and required for the spindle checkpoint. Cell 128: 101-114. Beck F, Chawengsaksophak K, Luckett J, Giblett S, Tucci J, Brown J, Poulsom R, Jeffery R, Wright NA. 2003. A study of regional gut endoderm potency by analysis of Cdx2 null mutant chimaeric mice. Dev Biol 255: 399-406. Beck F, Chawengsaksophak K, Waring P, Playford RJ, Furness JB. 1999. Reprogramming of intestinal differentiation and intercalary regeneration in Cdx2 mutant mice. Proc Natl Acad Sci U S A 96: 7318-7323. Beck F, Erler T, Russell A, James R. 1995. Expression of Cdx-2 in the mouse embryo and placenta: possible role in patterning of the extra-embryonic membranes. Dev Dyn 204: 219-227. Benahmed F, Gross I, Gaunt SJ, Beck F, Jehan F, Domon-Dell C, Martin E, Kedinger M, Freund JN, Duluc I. 2008. Multiple regulatory regions control the complex 93 REFERENCES expression pattern of the mouse Cdx2 homeobox gene. Gastroenterology 135: 1238. Benahmed F, Gross I, Guenot D, Jehan F, Martin E, Domon-Dell C, Brabletz T, Kedinger M, Freund JN, Duluc I. 2007. The microenvironment controls CDX2 homeobox gene expression in colorectal cancer cells. Am J Pathol 170: 733744. Bjerknes M, Cheng H. 1999. Clonal analysis of mouse intestinal epithelial progenitors. Gastroenterology 116: 7-14. Blache P, van de Wetering M, Duluc I, Domon C, Berta P, Freund JN, Clevers H, Jay P. 2004. SOX9 is an intestine crypt transcription factor, is regulated by the Wnt pathway, and represses the CDX2 and MUC2 genes. J Cell Biol 166: 37-47. Bleuming SA, Kodach LL, Garcia Leon MJ, Richel DJ, Peppelenbosch MP, Reitsma PH, Hardwick JC, van den Brink GR. 2006. Altered bone morphogenetic protein signalling in the Helicobacter pylori-infected stomach. J Pathol 209: 190-197. Bonhomme C, Calon A, Martin E, Robine S, Neuville A, Kedinger M, Domon-Dell C, Duluc I, Freund JN. 2008. Cdx1, a dispensable homeobox gene for gut development with limited effect in intestinal cancer. Oncogene 27: 4497-4502. Bonhomme C, Duluc I, Martin E, Chawengsaksophak K, Chenard MP, Kedinger M, Beck F, Freund JN, Domon-Dell C. 2003. The Cdx2 homeobox gene has a tumour suppressor function in the distal colon in addition to a homeotic role during gut development. Gut 52: 1465-1471. Bornschein J, Wex T, Peitz U, Kuester D, Roessner A, Malfertheiner P. 2009. The combined presence of H pylori infection and gastro-oesophageal reflux disease leads to an up-regulation of CDX2 gene expression in antrum and cardia. J Clin Pathol 62: 254-259. Bosman FT, Carneiro F, Hruban RH, Theise ND. 2010. WHO classification of tumours of the digestive system, 4th edition. IARC Press, Lyon. Boudreau F, Rings EH, van Wering HM, Kim RK, Swain GP, Krasinski SD, Moffett J, Grand RJ, Suh ER, Traber PG. 2002. Hepatocyte nuclear factor-1 alpha, GATA-4, and caudal related homeodomain protein Cdx2 interact functionally to modulate intestinal gene transcription. Implication for the developmental regulation of the sucrase-isomaltase gene. J Biol Chem 277: 31909-31917. Boulanger J, Vézina A, Mongrain S, Boudreau F, Perreault N, Auclair BA, Lainé J, Asselin C, Rivard N. 2005. Cdk2-dependent phosphorylation of homeobox transcription factor CDX2 regulates its nuclear translocation and proteasomemediated degradation in human intestinal epithelial cells. J Biol Chem 280: 18095-18107. Boyd M, Hansen M, Jensen TG, Perearnau A, Olsen AK, Bram LL, Bak M, Tommerup N, Olsen J, Troelsen JT. 2010. Genome-wide analysis of CDX2 binding in intestinal epithelial cells (Caco-2). J Biol Chem 285: 25115-25125. REFERENCES Brabletz T, Spaderna S, Kolb J, Hlubek F, Faller G, Bruns CJ, Jung A, Nentwich J, Duluc I, Domon-Dell C et al. 2004. Down-regulation of the homeodomain factor Cdx2 in colorectal cancer by collagen type I: an active role for the tumor environment in malignant tumor progression. Cancer Res 64: 6973-6977. Bravo JC, Correa P. 1999. Sulphomucins favour adhesion of Helicobacter pylori to metaplastic gastric mucosa. J Clin Pathol 52: 137-140. Brown S, Fellers J, Shippy T, Denell R, Stauber M, Schmidt-Ott U. 2001. A strategy for mapping bicoid on the phylogenetic tree. Curr Biol 11: R43-44. Bry L, Falk P, Huttner K, Ouellette A, Midtvedt T, Gordon JI. 1994. Paneth cell differentiation in the developing intestine of normal and transgenic mice. Proc Natl Acad Sci U S A 91: 10335-10339. Buchet-Poyau K, Courchet J, Le Hir H, Séraphin B, Scoazec JY, Duret L, Domon-Dell C, Freund JN, Billaud M. 2007. Identification and characterization of human Mex-3 proteins, a novel family of evolutionarily conserved RNA-binding proteins differentially localized to processing bodies. Nucleic Acids Res 35: 1289-1300. Busà R, Paronetto MP, Farini D, Pierantozzi E, Botti F, Angelini DF, Attisani F, Vespasiani G, Sette C. 2007. The RNA-binding protein Sam68 contributes to proliferation and survival of human prostate cancer cells. Oncogene 26: 4372-4382. Camilo V, Barros R, Sousa S, Magalhães AM, Lopes T, Mário Santos A, Pereira T, Figueiredo C, David L, Almeida R. 2012. Helicobacter pylori and the BMP pathway regulate CDX2 and SOX2 expression in gastric cells. Carcinogenesis 33: 1985-1992. Chawengsaksophak K, James R, Hammond VE, Köntgen F, Beck F. 1997. Homeosis and intestinal tumours in Cdx2 mutant mice. Nature 386: 84-87. Chen AJ, Paik JH, Zhang H, Shukla SA, Mortensen R, Hu J, Ying H, Hu B, Hurt J, Farny N et al. 2012. STAR RNA-binding protein Quaking suppresses cancer via stabilization of specific miRNA. Genes Dev 26: 1459-1472. Chen T, Damaj BB, Herrera C, Lasko P, Richard S. 1997. Self-association of the singleKH-domain family members Sam68, GRP33, GLD-1, and Qk1: role of the KH domain. Mol Cell Biol 17: 5707-5718. Cheng H, Leblond CP. 1974. Origin, differentiation and renewal of the four main epithelial cell types in the mouse small intestine. V. Unitarian Theory of the origin of the four epithelial cell types. Am J Anat 141: 537-561. Cho PF, Poulin F, Cho-Park YA, Cho-Park IB, Chicoine JD, Lasko P, Sonenberg N. 2005. A new paradigm for translational control: inhibition via 5'-3' mRNA tethering by Bicoid and the eIF4E cognate 4EHP. Cell 121: 411-423. Chun SY, Chen F, Washburn JG, MacDonald JW, Innes KL, Zhao R, Cruz-Correa MR, Dang LH, Dang DT. 2007. CDX2 promotes anchorage-independent growth by transcriptional repression of IGFBP-3. Oncogene 26: 4725-4729. 95 REFERENCES Jackson RJ, Hellen CU, Pestova TV. 2010. The mechanism of eukaryotic translation initiation and principles of its regulation. Nat Rev Mol Cell Biol 11: 113-127. James R, Erler T, Kazenwadel J. 1994. Structure of the murine homeobox gene cdx-2. Expression in embryonic and adult intestinal epithelium. J Biol Chem 269: 1522915237. James R, Kazenwadel J. 1991. Homeobox gene expression in the intestinal epithelium of adult mice. J Biol Chem 266: 3246-3251. Jedrusik A, Bruce AW, Tan MH, Leong DE, Skamagki M, Yao M, Zernicka-Goetz M. 2010. Maternally and zygotically provided Cdx2 have novel and critical roles for early development of the mouse embryo. Dev Biol 344: 66-78. Jedrusik A, Parfitt DE, Guo G, Skamagki M, Grabarek JB, Johnson MH, Robson P, Zernicka-Goetz M. 2008. Role of Cdx2 and cell polarity in cell allocation and specification of trophectoderm and inner cell mass in the mouse embryo. Genes Dev 22: 2692-2706. Jiang H, Zhang X, Luo J, Dong C, Xue J, Wei W, Chen J, Zhou J, Gao Y, Yang C. 2012. Knockdown of hMex-3A by small RNA interference suppresses cell proliferation and migration in human gastric cancer cells. Mol Med Rep 6: 575-580. Jin T, Drucker DJ. 1996. Activation of proglucagon gene transcription through a novel promoter element by the caudal-related homeodomain protein cdx-2/3. Mol Cell Biol 16: 19-28. Johnson CM, Wei C, Ensor JE, Smolenski DJ, Amos CI, Levin B, Berry DA. 2013. Metaanalyses of colorectal cancer risk factors. Cancer Causes Control 24: 1207-1222. Johnson JM, Castle J, Garrett-Engele P, Kan Z, Loerch PM, Armour CD, Santos R, Schadt EE, Stoughton R, Shoemaker DD. 2003. Genome-wide survey of human alternative pre-mRNA splicing with exon junction microarrays. Science 302: 2141-2144. Jud M, Razelun J, Bickel J, Czerwinski M, Schisa JA. 2007. Conservation of large foci formation in arrested oocytes of Caenorhabditis nematodes. Dev Genes Evol 217: 221-226. Kahyo T, Iwaizumi M, Shinmura K, Matsuura S, Nakamura T, Watanabe Y, Yamada H, Sugimura H. 2011. A novel tumor-derived SGOL1 variant causes abnormal mitosis and unstable chromatid cohesion. Oncogene 30: 4453-4463. Kaimaktchiev V, Terracciano L, Tornillo L, Spichtin H, Stoios D, Bundi M, Korcheva V, Mirlacher M, Loda M, Sauter G et al. 2004. The homeobox intestinal differentiation factor CDX2 is selectively expressed in gastrointestinal adenocarcinomas. Mod Pathol 17: 1392-1399. Kakizaki F, Aoki K, Miyoshi H, Carrasco N, Aoki M, Taketo MM. 2010. CDX transcription factors positively regulate expression of solute carrier family 5, member 8 in the colonic epithelium. Gastroenterology 138: 627-635. REFERENCES Karam SM. 1999. Lineage commitment and maturation of epithelial cells in the gut. Front Biosci 4: D286-298. Kawai H, Tomii K, Toyooka S, Yano M, Murakami M, Tsukuda K, Shimizu N. 2005. Promoter methylation downregulates CDX2 expression in colorectal carcinomas. Oncol Rep 13: 547-551. Kazumori H, Ishihara S, Rumi M, Kadowaki Y, Kinoshita Y. 2006. Bile acids directly augment caudal related homeobox gene Cdx2 expression in oesophageal keratinocytes in Barrett's epithelium. Gut 55: 16-25. Keene JD. 2007. RNA regulons: coordination of post-transcriptional events. Nat Rev Genet 8: 533-543. Keene JD, Tenenbaum SA. 2002. Eukaryotic mRNPs may represent posttranscriptional operons. Mol Cell 9: 1161-1167. Kemphues KJ, Priess JR, Morton DG, Cheng NS. 1988. Identification of genes required for cytoplasmic localization in early C. elegans embryos. Cell 52: 311-320. Kim BM, Buchner G, Miletich I, Sharpe PT, Shivdasani RA. 2005. The stomach mesenchymal transcription factor Barx1 specifies gastric epithelial identity through inhibition of transient Wnt signaling. Dev Cell 8: 611-622. Kim HS, Lee JS, Freund JN, Min KW, Lee JS, Kim W, Juhng SW, Park CS. 2006. CDX-2 homeobox gene expression in human gastric carcinoma and precursor lesions. J Gastroenterol Hepatol 21: 438-442. Kim S, Domon-Dell C, Wang Q, Chung DH, Di Cristofano A, Pandolfi PP, Freund JN, Evers BM. 2002. PTEN and TNF-alpha regulation of the intestinal-specific Cdx-2 homeobox gene through a PI3K, PKB/Akt, and NF-kappaB-dependent pathway. Gastroenterology 123: 1163-1178. Kirchner T, Müller S, Hattori T, Mukaisyo K, Papadopoulos T, Brabletz T, Jung A. 2001. Metaplasia, intraepithelial neoplasia and early cancer of the stomach are related to dedifferentiated epithelial cells defined by cytokeratin-7 expression in gastritis. Virchows Arch 439: 512-522. Köhler A, Hurt E. 2007. Exporting RNA from the nucleus to the cytoplasm. Nat Rev Mol Cell Biol 8: 761-773. Krueger F, Madeja Z, Hemberger M, McMahon M, Cook SJ, Gaunt SJ. 2009. Downregulation of Cdx2 in colorectal carcinoma cells by the Raf-MEK-ERK 1/2 pathway. Cell Signal 21: 1846-1856. Kuersten S, Goodwin EB. 2003. The power of the 3′ UTR: translational control and development. Nat Rev Genet 4: 626-637. Lambert M, Colnot S, Suh E, L'Horset F, Blin C, Calliot ME, Raymondjean M, Thomasset M, Traber PG, Perret C. 1996. cis-Acting elements and transcription factors involved in the intestinal specific expression of the rat calbindin-D9K gene: 103 REFERENCES binding of the intestine-specific transcription factor Cdx-2 to the TATA box. Eur J Biochem 236: 778-788. Lauren P. 1965. The two histological main types of gastric carcinoma: diffuse and socalled intestinal-type carcinoma. An attempt at a histo-clinical classification. Acta Pathol Microbiol Scand 64: 31-49. Lee MH, Schedl T. 2001. Identification of in vivo mRNA targets of GLD-1, a maxi-KH motif containing protein required for C. elegans germ cell development. Genes Dev 15: 2408-2420. Lemieux E, Boucher MJ, Mongrain S, Boudreau F, Asselin C, Rivard N. 2011. Constitutive activation of the MEK/ERK pathway inhibits intestinal epithelial cell differentiation. Am J Physiol Gastrointest Liver Physiol 301: G719-730. Lengerke C, Daley GQ. 2012. Caudal genes in blood development and leukemia. Ann N Y Acad Sci 1266: 47-54. Lewis JD, Izaurralde E. 1997. The role of the cap structure in RNA processing and nuclear export. Eur J Biochem 247: 461-469. Liu Q, Teh M, Ito K, Shah N, Ito Y, Yeoh KG. 2007a. CDX2 expression is progressively decreased in human gastric intestinal metaplasia, dysplasia and cancer. Mod Pathol 20: 1286-1297. Liu T, Zhang X, So CK, Wang S, Wang P, Yan L, Myers R, Chen Z, Patterson AP, Yang CS et al. 2007b. Regulation of Cdx2 expression by promoter methylation, and effects of Cdx2 transfection on morphology and gene expression of human esophageal epithelial cells. Carcinogenesis 28: 488-496. Lorentz O, Cadoret A, Duluc I, Capeau J, Gespach C, Cherqui G, Freund JN. 1999. Downregulation of the colon tumour-suppressor homeobox gene Cdx-2 by oncogenic ras. Oncogene 18: 87-92. Lorentz O, Duluc I, Arcangelis AD, Simon-Assmann P, Kedinger M, Freund JN. 1997. Key role of the Cdx2 homeobox gene in extracellular matrix-mediated intestinal cell differentiation. J Cell Biol 139: 1553-1565. Lu X, Freund JN, Muller M, Ravey J, Nicolas JP, Gueant JL, Namour F. 2008. Differential regulation of CDX1 and CDX2 gene expression by deficiency in methyl group donors. Biochimie 90: 697-704. Lukong KE, Chang KW, Khandjian EW, Richard S. 2008. RNA-binding proteins in human genetic disease. Trends Genet 24: 416-425. Lukong KE, Larocque D, Tyner A, Richard S. 2005. Tyrosine phosphorylation of sam68 by breast tumor kinase regulates intranuclear localization and cell cycle progression. J Biol Chem 280: 38639-38647. Lunde BM, Moore C, Varani G. 2007. RNA-binding proteins: modular design for efficient function. Nat Rev Mol Cell Biol 8: 479-490. REFERENCES Luo Y, Shan G, Guo W, Smrt RD, Johnson EB, Li X, Pfeiffer RL, Szulwach KE, Duan R, Barkho BZ et al. 2010. Fragile x mental retardation protein regulates proliferation and differentiation of adult neural stem/progenitor cells. PLoS Genet 6: e1000898. Mahatan CS, Kaestner KH, Geiman DE, Yang VW. 1999. Characterization of the structure and regulation of the murine gene encoding gut-enriched Krüppel-like factor (Krüppel-like factor 4). Nucleic Acids Res 27: 4562-4569. Maier T, Güell M, Serrano L. 2009. Correlation of mRNA and protein in complex biological samples. FEBS Lett 583: 3966-3973. Mallo GV, Rechreche H, Frigerio JM, Rocha D, Zweibaum A, Lacasa M, Jordan BR, Dusetti NJ, Dagorn JC, Iovanna JL. 1997. Molecular cloning, sequencing and expression of the mRNA encoding human Cdx1 and Cdx2 homeobox. Downregulation of Cdx1 and Cdx2 mRNA expression during colorectal carcinogenesis. Int J Cancer 74: 35-44. Mallo GV, Soubeyran P, Lissitzky JC, André F, Farnarier C, Marvaldi J, Dagorn JC, Iovanna JL. 1998. Expression of the Cdx1 and Cdx2 homeotic genes leads to reduced malignancy in colon cancer-derived cells. J Biol Chem 273: 1403014036. Mangus DA, Evans MC, Jacobson A. 2003. Poly(A)-binding proteins: multifunctional scaffolds for the post-transcriptional control of gene expression. Genome Biol 4: 223. Marchetti M, Caliot E, Pringault E. 2003. Chronic acid exposure leads to activation of the cdx2 intestinal homeobox gene in a long-term culture of mouse esophageal keratinocytes. J Cell Sci 116: 1429-1436. Margalit Y, Yarus S, Shapira E, Gruenbaum Y, Fainsod A. 1993. Isolation and characterization of target sequences of the chicken CdxA homeobox gene. Nucleic Acids Res 21: 4915-4922. Matlin AJ, Clark F, Smith CW. 2005. Understanding alternative splicing: towards a cellular code. Nat Rev Mol Cell Biol 6: 386-398. Matsuda K, Yamauchi K, Matsumoto T, Sano K, Yamaoka Y, Ota H. 2008. Quantitative analysis of the effect of Helicobacter pylori on the expressions of SOX2, CDX2, MUC2, MUC5AC, MUC6, TFF1, TFF2, and TFF3 mRNAs in human gastric carcinoma cells. Scand J Gastroenterol 43: 25-33. Merlos-Suárez A, Barriga FM, Jung P, Iglesias M, Céspedes MV, Rossell D, Sevillano M, Hernando-Momblona X, da Silva-Diz V, Muñoz P et al. 2011. The intestinal stem cell signature identifies colorectal cancer stem cells and predicts disease relapse. Cell Stem Cell 8: 511-524. Mesquita P, Jonckheere N, Almeida R, Ducourouble MP, Serpa J, Silva E, Pigny P, Silva FS, Reis C, Silberg D et al. 2003. Human MUC2 mucin gene is 105 REFERENCES transcriptionally regulated by Cdx homeodomain proteins in gastrointestinal carcinoma cell lines. J Biol Chem 278: 51549-51556. Mesquita P, Raquel A, Nuno L, Reis CA, Silva LF, Serpa J, Van Seuningen I, Barros H, David L. 2006. Metaplasia--a transdifferentiation process that facilitates cancer development: the model of gastric intestinal metaplasia. Crit Rev Oncog 12: 326. Milano F, van Baal JW, Buttar NS, Rygiel AM, de Kort F, DeMars CJ, Rosmolen WD, Bergman JJ, Van Marle J, Wang KK et al. 2007. Bone morphogenetic protein 4 expressed in esophagitis induces a columnar phenotype in esophageal squamous cells. Gastroenterology 132: 2412-2421. Mizoshita T, Inada K-i, Tsukamoto T, Nozaki K, Joh T, Itoh M, Yamamura Y, Ushijima T, Nakamura S, Tatematsu M. 2004. Expression of the intestine-specific transcription factors, Cdx1 and Cdx2, correlates shift to an intestinal phenotype in gastric cancer cells. J Cancer Res Clin Oncol 130: 29-36. Mlodzik M, Fjose A, Gehring WJ. 1985. Isolation of caudal, a Drosophila homeo boxcontaining gene with maternal expression, whose transcripts form a concentration gradient at the pre-blastoderm stage. EMBO J 4: 2961-2969. Modica S, Morgano A, Salvatore L, Petruzzelli M, Vanier MT, Valanzano R, Esposito DL, Palasciano G, Duluc I, Freund JN et al. 2009. Expression and localisation of insulin receptor substrate 2 in normal intestine and colorectal tumours. Regulation by intestine-specific transcription factor CDX2. Gut 58: 1250-1259. Moore MJ. 2005. From Birth to Death: The Complex Lives of Eukaryotic mRNAs. Science 309: 1514-1518. Mootz D, Ho DM, Hunter CP. 2004. The STAR/Maxi-KH domain protein GLD-1 mediates a developmental switch in the translational control of C. elegans PAL-1. Development 131: 3263-3272. Moskaluk CA, Zhang H, Powell SM, Cerilli LA, Hampton GM, Frierson HFJr. 2003. Cdx2 protein expression in normal and malignant human tissues: an immunohistochemical survey using tissue microarrays. Mod Pathol 16: 913-919. Mutoh H, Hakamata Y, Sato K, Eda A, Yanaka I, Honda S, Osawa H, Kaneko Y, Sugano K. 2002. Conversion of gastric mucosa to intestinal metaplasia in Cdx2expressing transgenic mice. Biochem Biophys Res Commun 294: 470-479. Mutoh H, Hayakawa H, Sakamoto H, Sashikawa M, Sugano K. 2009. Transgenic Cdx2 induces endogenous Cdx1 in intestinal metaplasia of Cdx2-transgenic mouse stomach. FEBS J 276: 5821-5831. Mutoh H, Hayakawa H, Sashikawa M, Sakamoto H, Sugano K. 2010. Direct repression of Sonic Hedgehog expression in the stomach by Cdx2 leads to intestinal transformation. Biochem J 427: 423-434. REFERENCES Mutoh H, Sakurai S, Satoh K, Osawa H, Tomiyama T, Kita H, Yoshida T, Tamada K, Yamamoto H, Isoda N et al. 2005a. Pericryptal fibroblast sheath in intestinal metaplasia and gastric carcinoma. Gut 54: 33-39. Mutoh H, Sakurai S, Satoh K, Tamada K, Kita H, Osawa H, Tomiyama T, Sato Y, Yamamoto H, Isoda N et al. 2004. Development of gastric carcinoma from intestinal metaplasia in Cdx2-transgenic mice. Cancer Res 64: 7740-7747. Mutoh H, Satoh K, Kita H, Sakamoto H, Hayakawa H, Yamamoto H, Isoda N, Tamada K, Ido K, Sugano K. 2005b. Cdx2 specifies the differentiation of morphological as well as functional absorptive enterocytes of the small intestine. Int J Dev Biol 49: 867-871. Nagaoka K, Udagawa T, Richter JD. 2012. CPEB-mediated ZO-1 mRNA localization is required for epithelial tight-junction assembly and cell polarity. Nat Commun 3: 675. Niessing D, Blanke S, Jäckle H. 2002. Bicoid associates with the 5'-cap-bound complex of caudal mRNA and represses translation. Genes Dev 16: 2576-2582. Nishioka N, Inoue K-i, Adachi K, Kiyonari H, Ota M, Ralston A, Yabuta N, Hirahara S, Stephenson RO, Ogonuki N et al. 2009. The Hippo signaling pathway components Lats and Yap pattern Tead4 activity to distinguish mouse trophectoderm from inner cell mass. Dev Cell 16: 398-410. Niwa T, Ikehara Y, Nakanishi H, Tanaka H, Inada K-i, Tsukamoto T, Ichinose M, Tatematsu M. 2005. Mixed gastricand intestinal-type metaplasia is formed by cells with dual intestinal and gastric differentiation. J Histochem Cytochem 53: 75-85. Olsen AK, Coskun M, Bzorek M, Kristensen MH, Danielsen ET, Jørgensen S, Olsen J, Engel U, Holck S, Troelsen JT. 2013. Regulation of APC and AXIN2 expression by intestinal tumor suppressor CDX2 in colon cancer cells. Carcinogenesis 34: 13611369. Ostareck DH, Ostareck-Lederer A, Shatsky IN, Hentze MW. 2001. Lipoxygenase mRNA silencing in erythroid differentiation: The 3'UTR regulatory complex controls 60S ribosomal subunit joining. Cell 104: 281-290. Pacheco II, Macleod RJ. 2008. CaSR stimulates secretion of Wnt5a from colonic myofibroblasts to stimulate CDX2 and sucrase-isomaltase using Ror2 on intestinal epithelia. Am J Physiol Gastrointest Liver Physiol 295: G748-759. Pagano JM, Farley BM, Essien KI, Ryder SP. 2009. RNA recognition by the embryonic cell fate determinant and germline totipotency factor MEX-3. Proc Natl Acad Sci U S A 106: 20252-20257. Pallafacchina G, François S, Regnault B, Czarny B, Dive V, Cumano A, Montarras D, Buckingham M. 2010. An adult tissue-specific stem cell in its niche: a gene profiling analysis of in vivo quiescent and activated muscle satellite cells. Stem Cell Res 4: 77-91. 107 REFERENCES Palles C, Cazier JB, Howarth KM, Domingo E, Jones AM, Broderick P, Kemp Z, Spain SL, Guarino E, Guarino Almeida E et al. 2013. Germline mutations affecting the proofreading domains of POLE and POLD1 predispose to colorectal adenomas and carcinomas. Nat Genet 45: 136-144. Park J, Schulz S, Waldman SA. 2000. Intestine-specific activity of the human guanylyl cyclase C promoter is regulated by Cdx2. Gastroenterology 119: 89-96. Porter EM, Bevins CL, Ghosh D, Ganz T. 2002. The multifaceted Paneth cell. Cell Mol Life Sci 59: 156-170. Potten CS. 1977. Extreme sensitivity of some intestinal crypt cells to X and gamma irradiation. Nature 269: 518-521. Potten CS, Booth C, Tudor GL, Booth D, Brady G, Hurley P, Ashton G, Clarke R, Sakakibara S-i, Okano H. 2003. Identification of a putative intestinal stem cell and early lineage marker; musashi-1. Differentiation 71: 28-41. Potten CS, Kovacs L, Hamilton E. 1974. Continuous labelling studies on mouse skin and intestine. Cell Tissue Kinet 7: 271-283. Potten CS, Loeffler M. 1990. Stem cells: attributes, cycles, spirals, pitfalls and uncertainties. Lessons for and from the crypt. Development 110: 1001-1020. Potten CS, Owen G, Booth D. 2002. Intestinal stem cells protect their genome by selective segregation of template DNA strands. J Cell Sci 115: 2381-2388. Pritchard CC, Grady WM. 2011. Colorectal cancer molecular biology moves into clinical practice. Gut 60: 116-129. Ralston A, Rossant J. 2008. Cdx2 acts downstream of cell polarization to cellautonomously promote trophectoderm fate in the early mouse embryo. Dev Biol 313: 614-629. Reis CA, David L, Correa P, Carneiro F, de Bolós C, Garcia E, Mandel U, Clausen H, Sobrinho-Simões M. 1999. Intestinal metaplasia of human stomach displays distinct patterns of mucin (MUC1, MUC2, MUC5AC, and MUC6) expression. Cancer Res 59: 1003-1007. Rhoads RE. 2009. eIF4E: new family members, new binding partners, new roles. J Biol Chem 284: 16711-16715. Rings EH, Boudreau F, Taylor JK, Moffett J, Suh ER, Traber PG. 2001. Phosphorylation of the serine 60 residue within the Cdx2 activation domain mediates its transactivation capacity. Gastroenterology 121: 1437-1450. Rokkas T, Filipe MI, Sladen GE. 1991. Detection of an increased incidence of early gastric cancer in patients with intestinal metaplasia type III who are closely followed up. Gut 32: 1110-1113. Rotkrua P, Akiyama Y, Hashimoto Y, Otsubo T, Yuasa Y. 2011. MiR-9 downregulates CDX2 expression in gastric cancer cells. Int J Cancer 129: 2611-2620. REFERENCES Sakagami Y, Inaguma Y, Sakakura T, Nishizuka Y. 1984. Intestine-like remodeling of adult mouse glandular stomach by implanting of fetal intestinal mesenchyme. Cancer Res 44: 5845-5849. Sakaguchi T, Gu X, Golden HM, Suh E, Rhoads DB, Reinecker HC. 2002. Cloning of the human claudin-2 5'-flanking region revealed a TATA-less promoter with conserved binding sites in mouse and human for caudal-related homeodomain proteins and hepatocyte nuclear factor-1alpha. J Biol Chem 277: 21361-21370. Sakamoto H, Mutoh H, Hayakawa H, Sashikawa M, Sugano K. 2011. Cell lineage dynamics in the process leading to intestinal metaplasia. J Gastroenterol 46: 620-628. Sakamoto H, Mutoh H, Ido K, Satoh K, Hayakawa H, Sugano K. 2007. A close relationship between intestinal metaplasia and Cdx2 expression in human gallbladders with cholelithiasis. Hum Pathol 38: 66-71. Salari K, Spulak ME, Cuff J, Forster AD, Giacomini CP, Huang S, Ko ME, Lin AY, van de Rijn M, Pollack JR. 2012. CDX2 is an amplified lineage-survival oncogene in colorectal cancer. Proc Natl Acad Sci U S A 109: E3196-3205. Sampath P, Pritchard DK, Pabon L, Reinecke H, Schwartz SM, Morris DR, Murry CE. 2008. A hierarchical network controls protein translation during murine embryonic stem cell self-renewal and differentiation. Cell Stem Cell 2: 448-460. Sangiorgi E, Capecchi MR. 2008. Bmi1 is expressed in vivo in intestinal stem cells. Nat Genet 40: 915-920. Sato T, Vries RG, Snippert HJ, van de Wetering M, Barker N, Stange DE, van Es JH, Abo A, Kujala P, Peters PJ et al. 2009. Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459: 262-265. Satoh K, Mutoh H, Eda A, Yanaka I, Osawa H, Honda S, Kawata H, Kihira K, Sugano K. 2002. Aberrant expression of CDX2 in the gastric mucosa with and without intestinal metaplasia: effect of eradication of Helicobacter pylori. Helicobacter 7: 192-198. Satou Y. 1999. posterior end mark 3 (pem-3), an ascidian maternally expressed gene with localized mRNA encodes a protein with Caenorhabditis elegans MEX-3-like KH domains. Dev Biol 212: 337-350. Savory JG, Pilon N, Grainger S, Sylvestre JR, Béland M, Houle M, Oh K, Lohnes D. 2009. Cdx1 and Cdx2 are functionally equivalent in vertebral patterning. Dev Biol 330: 114-122. Schisa JA, Pitt JN, Priess JR. 2001. Analysis of RNA associated with P granules in germ cells of C. elegans adults. Development 128: 1287-1298. Schonhoff SE, Giel-Moloney M, Leiter AB. 2004. Minireview: Development and differentiation of gut endocrine cells. Endocrinology 145: 2639-2644. 109 REFERENCES Schoppmeier M, Fischer S, Schmitt-Engel C, Löhr U, Klingler M. 2009. An ancient anterior patterning system promotes caudal repression and head formation in ecdysozoa. Curr Biol 19: 1811-1815. Schwanhäusser B, Busse D, Li N, Dittmar G, Schuchhardt J, Wolf J, Chen W, Selbach M. 2011. Global quantification of mammalian gene expression control. Nature 473: 337-342. Selgrad M, Bornschein J, Rokkas T, Malfertheiner P. 2010. Clinical aspects of gastric cancer and Helicobacter pylori--screening, prevention, and treatment. Helicobacter 15: 40-45. Semba S, Satake S, Matsushita M, Yokozaki H. 2009. Phosphatase activity of nuclear PTEN is required for CDX2-mediated intestinal differentiation of gastric carcinoma. Cancer Lett 274: 143-150. Seydoux G, Fire A. 1994. Soma-germline asymmetry in the distributions of embryonic RNAs in Caenorhabditis elegans. Development 120: 2823-2834. Shimada T, Koike T, Yamagata M, Yoneda M, Hiraishi H. 2007. Regulation of TFF3 expression by homeodomain protein CDX2. Regul Pept 140: 81-87. Silberg DG, Furth EE, Taylor JK, Schuck T, Chiou T, Traber PG. 1997. CDX1 protein expression in normal, metaplastic, and neoplastic human alimentary tract epithelium. Gastroenterology 113: 478-486. Silberg DG, Sullivan J, Kang E, Swain GP, Moffett J, Sund NJ, Sackett SD, Kaestner KH. 2002. Cdx2 ectopic expression induces gastric intestinal metaplasia in transgenic mice. Gastroenterology 122: 689-696. Silberg DG, Swain GP, Suh ER, Traber PG. 2000. Cdx1 and cdx2 expression during intestinal development. Gastroenterology 119: 961-971. Silva E, Teixeira A, David L, Carneiro F, Reis CA, Sobrinho-Simões J, Serpa J, Veerman E, Bolscher J, Sobrinho-Simões M. 2002. Mucins as key molecules for the classification of intestinal metaplasia of the stomach. Virchows Arch 440: 311317. Siomi H, Matunis MJ, Michael WM, Dreyfuss G. 1993. The pre-mRNA binding K protein contains a novel evolutionarily conserved motif. Nucleic Acids Res 21: 11931198. Skamagki M, Wicher KB, Jedrusik A, Ganguly S, Zernicka-Goetz M. 2013. Asymmetric localization of Cdx2 mRNA during the first cell-fate decision in early mouse development. Cell Rep 3: 442-457. Slack JM. 2007. Metaplasia and transdifferentiation: from pure biology to the clinic. Nat Rev Mol Cell Biol 8: 369-378. Stebbins-Boaz B, Cao Q, de Moor CH, Mendez R, Richter JD. 1999. Maskin is a CPEBassociated factor that transiently interacts with elF-4E. Mol Cell 4: 1017-1027. REFERENCES Stringer EJ, Duluc I, Saandi T, Davidson I, Bialecka M, Sato T, Barker N, Clevers H, Pritchard CA, Winton DJ et al. 2012. Cdx2 determines the fate of postnatal intestinal endoderm. Development 139: 465-474. Stringer EJ, Pritchard CA, Beck F. 2008. Cdx2 initiates histodifferentiation of the midgut endoderm. FEBS Lett 582: 2555-2560. Strumpf D, Mao CA, Yamanaka Y, Ralston A, Chawengsaksophak K, Beck F, Rossant J. 2005. Cdx2 is required for correct cell fate specification and differentiation of trophectoderm in the mouse blastocyst. Development 132: 2093-2102. Subramanian V, Meyer BI, Gruss P. 1995. Disruption of the murine homeobox gene Cdx1 affects axial skeletal identities by altering the mesodermal expression domains of Hox genes. Cell 83: 641-653. Subtil C, Guérin E, Schneider A, Chenard MP, Martin E, Domon-Dell C, Duluc I, Brabletz T, Kedinger M, Duclos B et al. 2007. Frequent rearrangements and amplification of the CDX2 homeobox gene in human sporadic colorectal cancers with chromosomal instability. Cancer Lett 247: 197-203. Suh E, Chen L, Taylor J, Traber PG. 1994. A homeodomain protein related to caudal regulates intestine-specific gene transcription. Mol Cell Biol 14: 7340-7351. Suh E, Traber PG. 1996. An intestine-specific homeobox gene regulates proliferation and differentiation. Mol Cell Biol 16: 619-625. Suvà ML, Riggi N, Bernstein BE. 2013. Epigenetic reprogramming in cancer. Science 339: 1567-1570. Tagawa T, Haraguchi T, Hiramatsu H, Kobayashi K, Sakurai K, Inada K-I, Iba H. 2012. Multiple microRNAs induced by Cdx1 suppress Cdx2 in human colorectal tumour cells. Biochem J 447: 449-455. Tamai Y, Nakajima R, Ishikawa T, Takaku K, Seldin MF, Taketo MM. 1999. Colonic hamartoma development by anomalous duplication in Cdx2 knockout mice. Cancer Res 59: 2965-2970. Taupin D, Podolsky DK. 1999. Mitogen-activated protein kinase activation regulates intestinal epithelial differentiation. Gastroenterology 116: 1072-1080. Taylor JK, Levy T, Suh ER, Traber PG. 1997. Activation of enhancer elements by the homeobox gene Cdx2 is cell line specific. Nucleic Acids Res 25: 2293-2300. Tosh D, Slack JM. 2002. How cells change their phenotype. Nat Rev Mol Cell Biol 3: 187-194. Trinh KY, Jin T, Drucker DJ. 1999. Identification of domains mediating transcriptional activation and cytoplasmic export in the caudal homeobox protein Cdx-3. J Biol Chem 274: 6011-6019. 111 119 Appendix ANTIBODIES Antigen Application Description Dilution Source Commercial reference Actin WB Rabbit polyclonal 1:8000 Santa Cruz Biotechnology sc-1616-R CDX2 IF and WB Mouse monoclonal IgG1 (Clone CDX2-88) 1:50 and 1:1000 Biogenex MU392A-UC Cyclin D1 WB Rabbit monoclonal (Clone SP4) 1:500 Thermo Scientific RM-9104 DCP1A IF Rabbit polyclonal 1:500 Sigma Aldrich D5444 E-cadherin IF Mouse monoclonal IgG1 (Clone HECD-1) 1:300 Takara Biochemicals M106 EDC4 IF Rabbit polyclonal 1:400 Cell Signaling Technology 2548 F-actin IF Tetramethylrhodamine B isothiocyanate (TRITC)-conjugate of phalloidin 1:10000 Sigma Aldrich 77418 MEX3A IF and WB Rabbit polyclonal 1:600 and 1:1000 Sigma Aldrich PRS4869 MSI1 WB Rabbit polyclonal 1:1000 Millipore AB5977 c-Myc IF and WB Mouse monoclonal IgG1 (Clone 9E10) 1:600 and 1:1000 Clontech 631206 Villin WB Mouse monoclonal 1:3000 Kindly provided by Dr. Sylvie Robine, Institute Curie, Paris, France ― ZO-1 IF Rabbit polyclonal 1:50 Invitrogen 61-7300 Secondary Goat anti-Mouse IgG IF Alexa Fluor 488 or Alexa Fluor 594 conjugated 1:100 Invitrogen A-11001 or A-11005 Goat anti-Mouse IgG WB Horseradish peroxidase (HRP) conjugated 1:2000 Santa Cruz Biotechnology sc-2005 Goat anti-Rabbit IgG IF Alexa Fluor 488 or Alexa Fluor 594 conjugated 1:100 Invitrogen A-11008 or A-11012 Goat anti-Rabbit IgG WB Horseradish peroxidase (HRP) conjugated 1:2000 Santa Cruz Biotechnology sc-2004 PRIMER SEQUENCES (5`- 3`) Bisulfite sequencing CpG1 F: TTTTAGAAATGATAGGATGAAGG R: ACCAAAAAACCTAAAACTAAAAA CpG2.1 F: GTAGGTTAGAGGGAGGGAT R: TCCTTATCCAAAAAATAACTCAC CpG2.2 F: GAGTTATTTTTTGGATAAGGA R: TAACCATTCCAATCCTCCC CpG2.3 F: GGGAGGATTGGAATGGTTA qPCR R: TTTACAACAACCCAAAAAC 18S F: CGCCGCTAGAGGTGAAATTC R: CATTCTTGGCAAATGCTTTCG BMI1 F: GTCTACATTCCTTCTGTAAAACG R: CTTGGAGAGTTTTATCTGACC CDX2 F: TTCACTACAGTCGCTACATCACCAT R: TTGTTGATTTTCCTCTCCTTTGCT GAPDH F: TCAAGGCTGAGAACGGGAAG R: AGAGGGGGCAGAGATGATGA LGR5 F: CAGCGTCTTCACCTCCTACCTA R: CCTTGGGAATGTATGTCAGAGC MEX3A F: CAAGCTCTGCGCTCTCTACAAA R: GGCCTTAATCTTGCAGCCTTG Rluc F: TGCAAGCAAATGAACGTGCTG R: TCTAGCCACGGGCTCGATGT Site-directed mutagenesis pRL∆CDX2 vector*S: GTAACATCCAAGCCAGCCTTTTCCAAGCCTTCTGGATCC AS:GGATCCAGAAGGCTTGGAAAAGGCTGGCTTGGATGTTAC * Mutated nucleotides are in bold and underlined Appendix A. List of antibodies and primers used in this study.