JNK signalling modulates intestinal homeostasis and tumourigenesis in mice
2009; Springer Nature; Volume: 28; Issue: 13 Linguagem: Inglês
10.1038/emboj.2009.153
ISSN1460-2075
AutoresRocı́o Sancho, Abdolrahman S. Nateri, Amaya García de Vinuesa, Cristina Aguilera, Emma Nye, Bradley Spencer‐Dene, Axel Behrens,
Tópico(s)Genetic factors in colorectal cancer
ResumoArticle11 June 2009free access JNK signalling modulates intestinal homeostasis and tumourigenesis in mice Rocio Sancho Rocio Sancho Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Search for more papers by this author Abdolrahman S Nateri Abdolrahman S Nateri Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Cancer Genetics Lab, Wolfson Digestive Diseases Centre, School of Medical and Surgical Sciences, Nottingham, UK Search for more papers by this author Amaya Garcia de Vinuesa Amaya Garcia de Vinuesa Departamento de Biologia Celular, Fisiologia e Inmunologia, Universidad de Cordoba, Cordoba, Spain Search for more papers by this author Cristina Aguilera Cristina Aguilera Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Search for more papers by this author Emma Nye Emma Nye Experimental Pathology Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Search for more papers by this author Bradley Spencer-Dene Bradley Spencer-Dene Experimental Pathology Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Department of Histopathology, Imperial College, London, UK Search for more papers by this author Axel Behrens Corresponding Author Axel Behrens Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Search for more papers by this author Rocio Sancho Rocio Sancho Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Search for more papers by this author Abdolrahman S Nateri Abdolrahman S Nateri Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Cancer Genetics Lab, Wolfson Digestive Diseases Centre, School of Medical and Surgical Sciences, Nottingham, UK Search for more papers by this author Amaya Garcia de Vinuesa Amaya Garcia de Vinuesa Departamento de Biologia Celular, Fisiologia e Inmunologia, Universidad de Cordoba, Cordoba, Spain Search for more papers by this author Cristina Aguilera Cristina Aguilera Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Search for more papers by this author Emma Nye Emma Nye Experimental Pathology Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Search for more papers by this author Bradley Spencer-Dene Bradley Spencer-Dene Experimental Pathology Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Department of Histopathology, Imperial College, London, UK Search for more papers by this author Axel Behrens Corresponding Author Axel Behrens Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK Search for more papers by this author Author Information Rocio Sancho1,‡, Abdolrahman S Nateri1,2,‡, Amaya Garcia de Vinuesa3, Cristina Aguilera1, Emma Nye4, Bradley Spencer-Dene4,5 and Axel Behrens 1 1Mammalian Genetics Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK 2Cancer Genetics Lab, Wolfson Digestive Diseases Centre, School of Medical and Surgical Sciences, Nottingham, UK 3Departamento de Biologia Celular, Fisiologia e Inmunologia, Universidad de Cordoba, Cordoba, Spain 4Experimental Pathology Laboratory, CRUK London Research Institute, Lincoln's Inn Fields Laboratories, London, UK 5Department of Histopathology, Imperial College, London, UK ‡These authors contributed equally to this work *Corresponding author. Mammalian Genetics Laboratory, Cancer Reasearch UK, London Research Institute, Lincoln's Inn Fields Laboratories, 44, Lincoln's Inn Fields, London WC2A 3PX, UK. Tel.: 44 207 269 3361; Fax: 44 207 269 3581; E-mail: [email protected] The EMBO Journal (2009)28:1843-1854https://doi.org/10.1038/emboj.2009.153 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Wnt signalling is a crucial signalling pathway controlling intestinal homeostasis and cancer. We show here that the JNK MAP kinase pathway and one of its most important substrates, the AP-1 transcription factor c-Jun, modulates Wnt signalling strength in the intestine. Transgenic gut-specific augmentation of JNK signalling stimulated progenitor cell proliferation and migration, resulting in increased villus length. In the crypt, c-Jun protein was highly expressed in progenitor cells and the absence of c-Jun resulted in decreased proliferation and villus length. In addition to several known c-Jun/AP-1 target genes, expression of Wnt target genes Axin2 and Lgr5 were stimulated by JNK activation, suggesting a cross talk of JNK to Wnt signalling. Expression of the Wnt pathway component TCF4 was controlled by JNK activity, and chromatin immunoprecipitation and reporter assays identified tcf4 as a direct c-Jun target gene. Consequently, increased JNK activity accelerated tumourigenesis in a model of colorectal carcinogenesis. As c-jun is a direct target of the TCF4/β-catenin complex, the control of tcf4 expression by JNK/c-Jun leads to a positive feedback loop that connects JNK and Wnt signalling. This mechanism regulates the physiological function of progenitor cells and oncogenic transformation. Introduction The intestinal tissue consists of adjoining villi and crypts. The villus protrudes into the intestinal lumen and contains terminally differentiated cells. Villi consist of three main epithelial cell types that fulfil the main functions of the gut. Enteroendocrine cells release gastrointestinal hormones, enterocytes absorb nutrients and goblet cells secrete a protective mucus barrier. The crypt is formed by an invagination of the epithelial sheet into the underlying connective tissue. Most of the cells present in the crypts are immature, with the notable exception of differentiated Paneth cells, which are located at the crypt base and secrete antibacterial peptides into the crypt lumen (Humphries and Wright, 2008). Although it has been clear that intestinal crypts harbour the stem cells of this tissue, the exact location of these cells has remained controversial (Barker et al, 2008). Using long-term label retention, a classical assay for stem cells, the cell located at position +4 relative to the crypt bottom, above the terminally differentiated Paneth cells, was suggested to be the intestinal stem cell (Potten et al, 1974). The stem cell properties of +4 cells have recently been substantiated. Bmi-1 was identified as a marker for +4 stem cells and lineage tracing studies using a knock-in allele driving an inducible creER construct from the Bmi-1 promoter showed that +4 stem cells are capable of giving rise to all intestinal cell lineages (Sangiorgi and Capecchi, 2008). In addition, using clonal marking techniques also the crypt base columnar (CBC) cells, which are located at the crypt bottom in between paneth cells, were suggested to have stem cell like properties (Cheng and Leblond, 1974a, 1974b). CBC cells are easily identified due to their elongated shape and wedge shaped nuclei, in contrast to the +4 cells, which are difficult to distinguish morphologically from their neighbouring cells. An important finding that lends molecular support for the stem cell nature of CBC cells was the identification of Lgr5, also known as Gpr49, as a CBC cell marker gene. Moreover, elegant lineage-tracing experiments using a knock-in of an inducible creER allele showed that CBC cells could differentiate into all the intestinal cell lineages (Barker et al, 2007). Thus, there seem to co-exist two spatially separated stem cell compartments in the intestine. Whether the +4 and the CBC stem cells fulfill similar or different functions, or whether the molecular pathways regulating these two cell population are similar or distinct, is not clear. The Wnt signalling pathway controls the proliferation and mediates developmental signals between cells. In the absence of Wnt signal, in unstimulated cells, a degradation complex consisting of the adenomatous polyposis coli (APC) tumour-suppressor protein, axin, and the glycogen synthase kinase, phosphorylates β-catenin marking it for subsequent ubiquitination and degradation. On Wnt ligand binding to its Frizzled receptor, a signalling cascade (termed canonical Wnt signalling) is triggered that destabilizes the degradation complex, allowing unphosphorylated β-catenin levels to accumulate and translocate to the nucleus where β-catenin functions as a co-factor for transcription factors of the T-cell factor/lymphoid-enhancing factor (TCF/LEF) family (Giles et al, 2003; Radtke and Clevers, 2005). Wnt signalling can also activate β-catenin-independent pathways. An important pathway for non-canonical Wnt signal transduction is the activation of the c-Jun N-terminal kinases (JNK) (Veeman et al, 2003). JNKs are serine/threonine kinases that belong to the group of MAP kinases, which are activated by a plethora of extracellular signals and are essential mediators of signal transduction (Davis, 2000). JNK was originally identified, as the name suggests, as an activity that phosphorylated the c-Jun N-terminus (Derijard et al, 1994). The proto-oncoprotein c-Jun belongs to the AP-1 group of transcription factors, which is a crucial regulator of cellular proliferation, apoptosis and tumourigenesis (Shaulian and Karin, 2001; Eferl and Wagner, 2003). Using a leucine zipper interaction interface, c-Jun heterodimerizes and forms functional transcription factors with a number of interacting partners, including all members of the Fos and ATF families of proteins (Mechta-Grigoriou et al, 2001). AP-1 activity is strongly induced in response to numerous signals, including growth factors, cytokines and extracellular stresses (Davis, 2000). AP-1 stimulation is mediated, in part, through the phosphorylation of c-Jun by the JNKs (Davis, 2000). c-Jun N-terminal phosphorylation at the serine residues 63 and 73 and threonine residues 91 and 93 within its transactivation domain is thought to increase the transcription of target genes, one of which is the c-jun gene itself (Angel et al, 1988). The Wnt signalling pathway is believed to be the major signalling pathway controlling intestinal homeostasis and cancer. c-jun is a well-characterized Wnt target gene (Mann et al, 1999; Staal et al, 2004) and absence of c-jun significantly delays tumourigenesis in mice heterozygous for a nonsense mutation at codon 850 of the Apc gene (ApcMin/+), which develop multiple intestinal neoplasias due to excessive canonical Wnt signalling (Moser et al, 1993; Nateri et al, 2005). Although Wnt signalling seems to be essential for intestinal stem cell function and maintenance, little is known about the contribution and function of other signalling pathways. In this study, we show that the JNK MAP kinase signalling pathway and its main substrate, the c-Jun transcription factor, have an important role in intestinal progenitor cells and oncogenic transformation. Results Transgenic JNK pathway activation in the gut Loss-of-function studies in the mouse have yielded important clues about JNK function (Yang et al, 1997; Kuan et al, 1999; Sabapathy et al, 1999; Chang et al, 2003), but the interpretation of the mutant phenotypes has been complicated by genetic redundancy as the JNK protein family is encoded by three genes jnk1, jnk2 and jnk3 (Gupta et al, 1996; Weston and Davis, 2002). It is conceivable that essential roles of MAP kinase signalling have been masked by redundancy and have remained undetected. Therefore, we devised a strategy that allows to reproducibly activate JNK signalling in mice. To investigate the significance of JNK activation in the intestine, we generated transgenic mice allowing the overexpression of a constitutively active JNKK2-JNK1 fusion protein (Zheng et al, 1999). A cassette encoding a fusion protein of β-galactosidase and NeoR (β-geo) flanked by loxP sites prevents the expression of constitutively active JNK1 (JNKK2-JNK1) before Cre-mediated recombination (β-geo–JNKK2-JNK1) (Figure 1A). Floxed single transgenic mice were crossed with Villin-cre transgenic mice previously shown to provide efficient gut-specific Cre activity (el Marjou et al, 2004). Figure 1.JNK signalling increases progenitor cell proliferation and villus length. (A) Scheme of the β-geo–JNKK2-JNK1 construct before and after Cre recombination. When these single transgenic mice are crossed to Cre transgenic mice, the β-geo cassette is excised and the JNKK2-JNK1 cDNA is expressed. (B) Protein lysates from unrecombined β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestines were analysed for JNK1, c-Jun, ser 73 phosphorylated c-Jun (p-c-Junser73), total JNK and β-actin (loading control) expression. (C) Haematoxylin and eosin staining of duodenum epithelium from β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG mice. Green shading denotes proliferative zone (crypt, C) and yellow shading marks the zone of differentiation (villus, V). All animals were killed between 8–12 weeks. Scale bar represents 50 μM. (D) Quantification of the villus length from the base of the villus to the villus apex of β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestines. Histogram represents the villus length as mean ± s.e.m. in different regions of the gut (D=duodenum, J=jejunum and I=Ileum) (*P⩽0.05; student's t test). (E) Immunohistochemistry for BrdU on representative crypts of β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestines 2 h BrdU post-injection. Black arrowheads represent BrdU+ proliferative progenitors whereas red arrow heads indicate BrdU+ columnar base cells (CBC). Scale bar represents 30 μM. (F) Quantification of BrdU+ and Ki67+ cells in β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG crypts (*P⩽0.05; student's t test). (G) BrdU+ CBC quantification in β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG crypts (*P⩽0.05; student's t test). (H) Immunohistochemistry for BrdU on representative sections of β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestines 24 h BrdU post-injection. (I) Quantification of BrdU+ cells in the villi of β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestines 24 h BrdU post-injection (*P⩽0.05; student's t test). (n and P values are detailed in Supplementary Table S1). Download figure Download PowerPoint In the absence of Cre recombinase, JNKK2-JNK1 fusion protein was not detectable in the intestine of β-geo–JNKK2-JNK1 single transgenic mice (Figure 1B), but JNKK2-JNK1 expression was induced in β-geo–JNKK2-JNK1;Villin-cre+ double transgenic mice (designated JNKK2-JNK1ΔG mice). c-Jun N-terminal phosphorylation was stimulated in JNKK2-JNK1ΔG mice, and presumably because of the autoregulation of the c-jun promoter by c-Jun, led to increased c-Jun protein levels (Figure 1B). JNKK2-JNK1ΔG mice developed normally and were indistinguishable from their control littermates. Histological analysis showed that the gross morphology of the intestine was normal in JNKK2-JNK1ΔG mice, and that all four differentiated cell types of the intestine (paneth cells, enterocytes, goblet cells and enteroendocrine cells) were present (Figure 1C and Supplementary Figure 1). JNK signalling increases intestinal cell proliferation and villus length Although intestinal tissue architecture and cell differentiation was unaffected by increased JNK signalling, we noted that the average villus length seemed to be increased in JNKK2-JNK1ΔG mice, which was corroborated by morphometric quantification (Figure 1C, D; and Supplementary Figure 2). Average villus length was slightly but consistently increased throughout the intestine in JNKK2-JNK1ΔG mice compared with control mice, albeit not to the same extent. This suggested that JNK signalling increased the total number of cells present per villus unit. JNK/c-Jun signalling controls proliferation in a number of cell types (Johnson et al, 1993; Hess et al, 2004), therefore we determined whether the increase in villus length might be caused by a pro-proliferative role of JNK signalling. At the crypt–villus junction, rapidly proliferating transit-amplifying (TA) cells are present that are capable of differentiating towards the intestinal cell lineages. Both bromodeoxyuridine (BrdU) labelling (Figure 1E) and immunohistochemistry (IHC) for the proliferation marker Ki67 (data not shown) indicated that the number of cycling transit-amplifying cells per crypt was increased by augmented JNK signalling in the small intestine (Figure 1E black arrowheads and Figure 1F; Supplementary Figure 2) and also in the colon (Supplementary Figure 3). In contrast to several other stem cell populations, including the +4 cells, CBC cells are frequently cycling and incorporate BrdU (Figure 1E red arrowheads). Activation of the JNK pathway in JNKK2-JNK1ΔG mice significantly increased the percentage of CBC cells incorporating BrdU by 50% (Figure 1G). JNK signalling also has well-established functions in the regulation of cell death and cell migration (Shaulian and Karin, 2002; Xia and Karin, 2004). However, survival of intestinal cells was not altered in JNKK2-JNK1ΔG mice (Supplementary Figure 4). In contrast, long-term BrdU labelling showed an increased rate of migration across the crypt–villus axis (Figure 1H, I). Thus, it seems that stimulation of JNK signalling leads to increased cell number and villus length by regulating proliferation and migration of progenitor cells. JNK signalling controls both AP-1 and Wnt target genes To investigate in more detail the function of JNK signalling in intestinal homeostasis, we carried out IHC for a crucial substrate of the JNK family of kinases, the c-Jun transcription factor. Staining for c-Jun protein was detected at the apex of the villi (data not shown), but in the crypt c-Jun protein was highly expressed in CBC cells (Figure 2A red arrowheads). Figure 2.Wnt target gene activation in JNKK2-JNK1ΔG mice. (A, B) Immunohistochemistry for c-Jun (A) or p-c-Junser73 (B) on representative crypts of β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestines. Red arrowheads represent c-Jun positively labelled columnar base cells (CBC). Scale bar represents 30 μM. (C) qRT–PCR analysis of c-jun, ccdn1, cd44, axin2, lgr5, tcf4 and gapdh transcripts in JNKK2-JNK1ΔG intestines compared with β-geo–JNKK2-JNK1. The data are normalized to β-actin and represented as fold induction over β-geo–JNKK2-JNK1 mice. (D) Immunohistochemistry for CD44 on representative crypts of β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestines. Red arrowheads indicate CD44 staining inCBC whereas black arrowheads indicate CD44+ proliferative progenitors. Scale bar represents 10 μM. (E) Lgr5 in situ hybridization in comparable regions from β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestines. (F) Quantification of CBCs in β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG crypts. (G) Western blot analysis of protein lysates from unrecombined β-geo–JNKK2-JNK1 and JNKK2-JNK1ΔG intestine and liver for cyclinD1, TCF4 and β-catenin expression. Download figure Download PowerPoint Moreover, phosphorylated c-Jun at serine 73 (p-c-Junser73) as a readout for JNK activity was detected in both CBC and TA progenitor cells and was elevated in JNKK2-JNK1ΔG mice compared with controls (Figure 2B). Importantly, staining for phosphorylated c-Jun was absent in junAA homozygous mutant mice, in which the JNK phosphoacceptor serines 63 and 73 are mutated to alanines and which served as a negative control for the IHC (Behrens et al, 1999) (Supplementary Figure 5). The expression of c-Jun and p-c-Junser73 indicates that JNK activity is present in progenitor cells and increased in JNKK2-JNK1ΔG mice. To understand the molecular mechanism of how JNK signalling controls intestinal stem cell function, we investigated the expression of known target genes of the JNK/c-Jun signalling pathway. c-Jun autoregulates the transcription of the c-jun gene and accordingly c-jun mRNA levels were increased in JNKK2-JNK1ΔG intestine (Figure 2C)(Angel et al, 1988). The expression of cyclinD1 (ccnd1) is regulated by both JNK and Wnt signalling (Tetsu and McCormick, 1999; Wisdom et al, 1999; Wulf et al, 2001), and cyclinD1 expression was also increased in JNKK2-JNK1ΔG mice. CD44 is expressed in the crypt and induced in intestinal tumours (Wielenga et al, 1999; Sansom et al, 2004). cd44 is a c-Jun target gene both in the intestine and in neurons (Raivich et al, 2004; Nateri et al, 2005), and augmentation of JNK signalling significantly induced cd44 mRNA expression (Figure 2C). IHC showed that CD44-protein expression was greatly augmented in CBC cells and transient amplifying cells in the crypts of JNKK2-JNK1ΔG mice (Figure 2D arrowheads). As Wnt signalling is a key pathway in intestinal stem cells, we also investigated the regulation of bona fide Wnt target genes. Unexpectedly, JNK signalling also substantially increased the expression of the classical Wnt target genes axin2 and lgr5 (also called gpr49) (Jho et al, 2002; Van der Flier et al, 2007). Lgr5 is highly expressed in CBC cells and has been used as a marker gene to define the CBC cells (Barker et al, 2007). It was conceivable that the increased lgr5 mRNA abundance was due to an increased number of CBC cells, or alternatively, activation of JNK signalling might change the expression pattern of lgr5 and stimulate lgr5 expression inappropriately in differentiated cell types. However, in-situ hybridization for lgr5 mRNA showed that the cell type specificity of lgr5 expression was not altered in JNKK2-JNK1ΔG mice, which as in controls was largely confined to the bottom of the crypt. Instead, lgr5 levels in CBC cells were noticeably increased in JNKK2-JNK1ΔG mice (Figure 2E arrowheads). Moreover, quantification showed normal numbers of CBC cells in JNKK2-JNK1ΔG mice (Figure 2F). Therefore, activation of JNK signalling stimulates both AP-1 and Wnt target gene expression in progenitor cells. Wnt target gene induction in the gut is mainly controlled by the TCF4 transcription factor. tcf4 (encoded by the tcf7l2 gene) mRNA levels were augmented in JNKK2-JNK1ΔG mice (Figure 2C), suggesting a potential mechanism of JNK-mediated Wnt target gene induction. Western blot analysis showed higher levels of cyclinD1 and TCF4 protein in the gut of JNKK2-JNK1ΔG mice (Figure 2F). The expression of cyclinD1 and TCF4 in the liver, where Villin-cre is not active and the JNKK2-JNK1 fusion protein is not expressed (data not shown), was comparable (Figure 2G). Decreased progenitor cell proliferation and villus length in the absence of c-Jun The expression of c-Jun in progenitor cells indicated a potential role of c-Jun in this cell population. To directly test the importance of c-Jun in intestinal homeostasis, we used mice with a tissue-specific inactivation of c-jun in the gut (c-junf/f; villin-cre+ mice or c-junΔG mice)(Behrens et al, 2002; el Marjou et al, 2004; Nateri et al, 2005). IHC staining for c-Jun was not detected in the intestine of c-junΔG mice, confirming the deletion of the floxed c-jun gene in progenitor cells and the specificity of the IHC staining (Figure 3A arrowheads). As in JNKK2-JNK1ΔG mice, the gross intestinal architecture and histological appearance of c-junΔG intestine was normal. However, the average villus length in c-junΔG mice was reduced (Figure 3B). Morphometric quantification showed that there was subtle, but reproducible and statistically significant reduction of villus length in the absence of c-jun (Figure 3C). Not all regions of the gut were affected to the same extent by increased JNK signalling, possibly because of the different histological structure and biology of distinct regions of the intestine. Inactivation of c-jun also led to an appreciable reduction in the percentage of cells incorporating BrdU (Figure 3D, E) whereas c-jun deletion had no significant effect on CBC cell number (Figure 3F, G). Moreover, quantitative PCR and western blot analysis showed diminished levels of cyclinD1, TCF4 and Lgr5 mRNA, and TCF4 and cyclinD1 protein in c-junΔG gut (Figure 3H, I). Thus, inactivation of c-jun leads to the opposite phenotype than augmentation of JNK signalling. Figure 3.Absence of c-Jun decreases crypt cell proliferation and villus length. (A) Immunohistochemistry for c-Jun on representative crypts from c-junf/f and c-junΔG intestines. Red arrow heads represent c-Jun positively labelled columnar base cells (CBC), black arrow heads represent c-Jun negative CBCs. Scale bar represents 30 μM. (B) Haematoxylin and eosin staining of jejunum epithelium from c-junf/f and c-junΔG mice. Green shading denotes proliferative zone (crypt, C) and yellow shading marks the zone of differentiation (villus, V). All animals were killed between 8–12 weeks. Scale bar represents 50 μM. (C) Quantification of the villus length from the base of the villus to the villus apex of c-junf/f and c-junΔG intestines. Histogram represents the villus length as mean ± s.e.m. in different regions of the gut (D=duodenum, J=jejunum and I=Ileum) (*P⩽0.05; student's t test). (D) Immunohistochemistry for BrdU on representative crypts from c-junf/f and c-junΔG intestines. Black arrowheads represent BrdU+ proliferative progenitors. Scale bar represents 25 μM. (E) Quantification of BrdU+ cells is represented in the histogram and expressed as % of positive cells per crypt, considering c-junf/f as 100% (*P⩽0.05; student's t test). (F) Lgr5 in situ hybridization in comparable crypts from c-junf/f and c-junΔG intestines. (G) Quantification of CBCs in c-junf/f and c-junΔG crypts. (H) qRT–PCR analysis of c-jun, ccdn1 and tcf4 transcripts in c-junΔG intestines compared with c-junf/f. The data are normalized to β-actin and represented as fold induction over c-junf/f mice. (I) Western analysis of protein lysates from c-junf/f and c-junΔG intestines for c-Jun, TCF4, cyclinD1 and β-actin (loading control). * indicates non-specific band. (n and P values are detailed in Supplementary Table S1). Download figure Download PowerPoint tcf4 is a direct c-Jun target gene The deregulation of tcf4 expression in JNKK2-JNK1ΔG mice suggested a potential mechanism of cross talk between JNK and Wnt signalling. Therefore we examined the regulation of TCF4 by JNK signalling in more detail. Treatment of HCT116 colon cancer cells with the JNK activator anisomycin (Ans) stimulated c-Jun and TCF4 expression (Figure 4A), whereas a pharmacological JNK inhibitor (JNKi) reduced c-jun and tcf4 mRNA and protein levels (Figure 4B, C). Thus, tcf4 expression like c-jun expression, is controlled by JNK activity. c-Jun overexpression resulted in increased TCF4 protein levels (Figure 4D). Moreover, to investigate whether endogenous c-Jun is required for tcf4 expression, we generated a stable polyclonal HCT116 cell line expressing a lentivirus-mediated c-jun knockdown shRNA hairpin construct and a matched control cell line expressing an irrelevant shRNA hairpin. c-jun depletion significantly reduced tcf4 mRNA and TCF4 protein (Figure 4E, F). Moreover, also cyclinD1 and, importantly, axin2 mRNA levels were noticeably reduced (Figure 4E). Consequently the JNK/c-Jun pathway was required for efficient proliferation of HCT116 tumour cells as both c-jun knock down and pharmacological JNK inhibition reduced growth of these cells (Supplementary Figure 6). Figure 4.tcf4 is a direct c-Jun target gene. (A) Western blot analysis of c-Jun and TCF4 protein levels in HCT116 cells treated with anisomycin (Ans) for the indicated time periods. (B) qRT–PCR analysis of c-jun and tcf4 transcripts in HCT116 cells treated with SP600125 (JNKi) for the indicated time periods. (C) Western analysis of protein lysates from HCT116 cells treated with SP600125 (JNKi) for c-Jun, TCF4 and β-actin (loading control). (D) Western blot analysis of c-Jun and TCF4 protein levels in HCT116 transfected with an expression plasmid for c-Jun (c-Jun-Ires–gfp) or empty vector (Ires–gfp). (E) qRT–PCR analysis of c-jun, tcf4, axin2 and ccdn1 transcripts in HCT116-si-c-jun (stable cell line that express an shRNA against c-jun) or HCT116-si-control (expressing an irrelevant shRNA) (F) Western blot analysis of c-Jun, TCF4, cyclinD1 and β-actin (loading control) in HCT116-si-c-jun and HCT116-si-control cell lines. (G) Schematic representation of TCF4 promoter. Black rectangles denote AP1 consensus binding sites. Arrows indicate approximate positions of qPCR primers used in ChIPs. (H) ChIP was carried out using HCT116 cells treated for 2 h with JNK inhibitor (JNKi) or Ans and c-Jun binding to the TCF4-1 region was determined by qPCR. Data is represented as IP/INPUT (%). Rabbit IgG antibody was used as an isotype control. (I) Schematic representation of pGL3–TCF4-1. (J) HCT116 cells were transfected with pGL3–TCF4-1, pGL3–uPA or pGL3–TATA together with c-Jun, c-Jun∼ATF2 and c-Jun∼Fos overexpression vector (or empty vector as control). Data represent luciferase activity relative to pGL3–TATA+empty-vector-transfected cells. (K–M) Double immunofluorescence for c-Jun (red; K) and TCF4 (green; L) and the merge image (M) on paraffin section from ApcMin/+ tumour. Download figure Download PowerPoint To investigate the regulation of tcf4 expression, we screened the tcf4 promoter for the presence of AP-1/c-Jun binding sites. Computational transcription factor binding site predictions identified two clusters of AP-1/c-Jun binding sites. At position −996 to −833 bp relative to the TATA box, three AP-1 sites were predicted closely adjacent to each other, and a second cluster of binding sites within the first exon of the tcf4 gene (Figure 4G). To investigate the relevance of these
Referência(s)