Bone Morphogenetic Protein Signaling Is Essential for Terminal Differentiation of the Intestinal Secretory Cell Lineage
2007; Elsevier BV; Volume: 133; Issue: 3 Linguagem: Inglês
10.1053/j.gastro.2007.06.066
ISSN1528-0012
AutoresBenoît A. Auclair, Yannick D. Benoit, Nathalie Rivard, Yuji Mishina, Nathalie Perreault,
Tópico(s)Cancer-related gene regulation
ResumoBackground & Aims: Bone morphogenetic proteins (Bmps) are morphogens known to play key roles in gastrointestinal development and pathology. Most Bmps are produced primarily by the mesenchymal compartment and activate their signaling pathways following a paracrine or autocrine route. The aim of this study was to investigate the role of epithelial Bmp signaling in intestinal morphogenesis and maintenance of adult epithelial cell functions. Methods: With the use of tissue-specific gene ablation, we generated mice lacking the Bmp receptor type IA (Bmpr1a) exclusively in the intestinal epithelium. Bmpr1a mutant and control mice were sacrificed for histology, immunofluorescence, Western blot analysis, electron microscopy, and quantitative polymerase chain reaction. Results: As well as showing increased proliferation and altered intestinal epithelial morphology, Bmpr1a mutant mice revealed that epithelial Bmp signaling is associated with impaired terminal differentiation of cells from the secretory lineage but not with the determination of cell fate. Loss of Bmp signaling exclusively in the epithelial compartment is not sufficient for the initiation of the de novo crypt phenomenon associated with juvenile polyposis syndrome. Conclusions: Epithelial Bmp signaling plays an important role in the terminal differentiation of the intestinal secretory cell lineage but not in de novo crypt formation. These findings emphasize the importance of delineating the contribution of the stroma vs the epithelium in gastrointestinal physiology and pathology. Background & Aims: Bone morphogenetic proteins (Bmps) are morphogens known to play key roles in gastrointestinal development and pathology. Most Bmps are produced primarily by the mesenchymal compartment and activate their signaling pathways following a paracrine or autocrine route. The aim of this study was to investigate the role of epithelial Bmp signaling in intestinal morphogenesis and maintenance of adult epithelial cell functions. Methods: With the use of tissue-specific gene ablation, we generated mice lacking the Bmp receptor type IA (Bmpr1a) exclusively in the intestinal epithelium. Bmpr1a mutant and control mice were sacrificed for histology, immunofluorescence, Western blot analysis, electron microscopy, and quantitative polymerase chain reaction. Results: As well as showing increased proliferation and altered intestinal epithelial morphology, Bmpr1a mutant mice revealed that epithelial Bmp signaling is associated with impaired terminal differentiation of cells from the secretory lineage but not with the determination of cell fate. Loss of Bmp signaling exclusively in the epithelial compartment is not sufficient for the initiation of the de novo crypt phenomenon associated with juvenile polyposis syndrome. Conclusions: Epithelial Bmp signaling plays an important role in the terminal differentiation of the intestinal secretory cell lineage but not in de novo crypt formation. These findings emphasize the importance of delineating the contribution of the stroma vs the epithelium in gastrointestinal physiology and pathology. See editorial on page 1035. See editorial on page 1035. The adult intestinal mucosa is composed of undifferentiated pluripotent stem cells as well as undifferentiated daughter cells located in the lower portion of the intestinal crypt and differentiated functional epithelial cells found along the villus. The commitment of the daughter cell to differentiate is acquired in the upper third of the crypt where the cell loses its ability to divide.1Ménard D. Beaulieu J.-F. Boudreau F. et al.Cell signaling and growth factors in development: from molecules to organogenesis.in: Unsicker K. Krieglstein K. Gastrointestinal tract. Volume 2. Wiley-Vch, Weinheim2006: 755-790Google Scholar, 2Bjerknes M. Cheng H. Gastrointestinal stem cells II. Intestinal stem cells.Am J Physiol. 2005; 289: G381-G387Crossref PubMed Scopus (113) Google Scholar There are 4 types of terminally differentiated intestinal cells derived from stem cells that can be divided into 2 categories. First are the absorptive cells, which play a role in the absorption of nutrients, and, second, cells of the secretory lineage including the mucin-secreting goblet cells, the hormone-secreting enteroendocrine cells, and the antimicrobial peptide–secreting Paneth cells.1Ménard D. Beaulieu J.-F. Boudreau F. et al.Cell signaling and growth factors in development: from molecules to organogenesis.in: Unsicker K. Krieglstein K. Gastrointestinal tract. Volume 2. Wiley-Vch, Weinheim2006: 755-790Google Scholar, 2Bjerknes M. Cheng H. Gastrointestinal stem cells II. Intestinal stem cells.Am J Physiol. 2005; 289: G381-G387Crossref PubMed Scopus (113) Google Scholar Such cellular dynamics and specification of cell fate requires a complex and precise regulation of gene expression along the anterior/posterior and crypt–villus axes.1Ménard D. Beaulieu J.-F. Boudreau F. et al.Cell signaling and growth factors in development: from molecules to organogenesis.in: Unsicker K. Krieglstein K. Gastrointestinal tract. Volume 2. Wiley-Vch, Weinheim2006: 755-790Google Scholar, 2Bjerknes M. Cheng H. Gastrointestinal stem cells II. Intestinal stem cells.Am J Physiol. 2005; 289: G381-G387Crossref PubMed Scopus (113) Google Scholar The defined expression of certain genes at a specific time or at a precise location requires a specialized microenvironment that depends on a variety of growth factors, cell–cell interactions, and epithelial–mesenchymal interactions.3Perreault N. Sackett S.D. Katz J.P. et al.Foxl1 is a mesenchymal modifier of min in carcinogenesis of stomach and colon.Genes Dev. 2005; 19: 311-315Crossref PubMed Scopus (65) Google Scholar, 4Madison B.B. Braunstein K. Kuizon E. et al.Epithelial hedgehog signals pattern the intestinal crypt-villus axis.Development. 2005; 132: 279-289Crossref PubMed Scopus (307) Google Scholar, 5Roberts D.J. Molecular mechanisms of development of the gastrointestinal tract.Dev Dyn. 2000; 219: 109-120Crossref PubMed Scopus (181) Google Scholar Recent studies have identified the Hedgehog, platelet-derived growth factor, and bone morphogenetic protein (Bmp) signaling pathways as key mediators in crypt and villus morphogenesis as well as maintenance of intestinal stem cells.4Madison B.B. Braunstein K. Kuizon E. et al.Epithelial hedgehog signals pattern the intestinal crypt-villus axis.Development. 2005; 132: 279-289Crossref PubMed Scopus (307) Google Scholar, 6Karlsson L. Lindahl P. Heath J.K. et al.Abnormal gastrointestinal development in PDGF-A and PDGFR-(alpha) deficient mice implicates a novel mesenchymal structure with putative instructive properties in villus morphogenesis.Development. 2000; 127: 3457-3466PubMed Google Scholar, 7Haramis A.P. Begthel H. van den Born M. et al.De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine.Science. 2004; 303: 1684-1686Crossref PubMed Scopus (636) Google Scholar, 8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar The Bmps are morphogens belonging to the transforming growth factor β superfamily. Bmps play active roles in many cellular functions in adult animals,9Mishina Y. Function of bone morphogenetic protein signaling during mouse development.Front Biosci. 2003; 8: d855-d869Crossref PubMed Google Scholar signaling through the serine/threonine kinase receptor subtypes I and II. The main Bmps found in the small intestine, Bmp2 and Bmp4,5Roberts D.J. Molecular mechanisms of development of the gastrointestinal tract.Dev Dyn. 2000; 219: 109-120Crossref PubMed Scopus (181) Google Scholar perform their signaling through the type IA receptor (Bmpr1a).9Mishina Y. Function of bone morphogenetic protein signaling during mouse development.Front Biosci. 2003; 8: d855-d869Crossref PubMed Google Scholar Recently, 3 research groups generated mice with impaired Bmp signaling affecting the epithelial and stromal compartments of the gut simultaneously.4Madison B.B. Braunstein K. Kuizon E. et al.Epithelial hedgehog signals pattern the intestinal crypt-villus axis.Development. 2005; 132: 279-289Crossref PubMed Scopus (307) Google Scholar, 7Haramis A.P. Begthel H. van den Born M. et al.De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine.Science. 2004; 303: 1684-1686Crossref PubMed Scopus (636) Google Scholar, 8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar All showed the phenomenon of de novo crypt in the intestinal mucosa because the intestinal epithelium of these mice resembles the mucosa of patients with juvenile polyposis syndrome. This shows a significant contribution of both epithelial and mesenchymal Bmp signaling in intestinal pathology. Interestingly, a recent study by Kim et al10Kim B.G. Li C. Qiao W. et al.Smad4 signalling in T cells is required for suppression of gastrointestinal cancer.Nature. 2006; 441: 1015-1019Crossref PubMed Scopus (270) Google Scholar revealed that the selective loss of Smad4 in the T cell, a prominent stromal component, led to epithelial tumors throughout the gastrointestinal tract, whereas its epithelial-specific deletion did not. Deletion of Smad4 in T cells suggests that loss of Smad4-dependent signaling in cells within the epithelial microenvironment has an important role in the evolution of juvenile polyposis syndrome. However, a specific role and the molecular relevance of excluding the Bmp pathway from one particular cell compartment over the other in gastrointestinal maintenance of adult epithelial cell functions remain unknown. This study shows that the loss of Bmp signaling exclusively in the intestinal epithelium has a dramatic consequence on epithelial proliferation and terminal differentiation and maturation of cells from the secretory lineage, but not on the initiation of, de novo crypts. SVEV-Bmpr1aloxP/loxP mice were provided by Dr. Mishina and the C57BL/6 12.4KbVilCre transgenic line was provided by Dr. Gumucio. Both mutations were genotyped following protocols already published.11Madison B.B. Dunbar L. Qiao X.T. et al.Cis elements of the villin gene control expression in restricted domains of the vertical (crypt) and horizontal (duodenum, cecum) axes of the intestine.J Biol Chem. 2002; 277: 33275-33283Crossref PubMed Scopus (601) Google Scholar, 12Mishina Y. Hanks M.C. Miura S. et al.Generation of Bmpr/Alk3 conditional knockout mice.Genesis. 2002; 32: 69-72Crossref PubMed Scopus (229) Google Scholar Total RNA was isolated and processed using the Totally RNA extraction kit (Ambion, Austin, TX). Reverse-transcription polymerase chain reaction analysis was performed using AMV-RT (Roche Diagnostic, Laval, Quebec, Canada) following the manufacturer's protocol. Quantitative real-time polymerase chain reaction was performed using the Quantitech SYBR Green Kit from Qiagen (Mississauga, Ontario, Canada) with the Roche Diagnostic Light Cycler. Primer sequences are available on request. Jejunum were fixed in 4% paraformaldehyde overnight at 4°C, then dehydrated and embedded in paraffin. Alcian blue and H&E stainings were performed as previously described.13Lee C.S. Perreault N. Brestelli J.E. et al.Neurogenin 3 is essential for the proper specification of gastric enteroendocrine cells and the maintenance of gastric epithelial cell identity.Genes Dev. 2002; 16: 1488-1497Crossref PubMed Scopus (249) Google Scholar Mice were injected with 10 μL of bromodeoxyuridine (BrdU) (Invitrogen, Burlington, Ontario, Canada) per gram of body weight 90 minutes before death. Immunofluorescence staining was performed as previously described.13Lee C.S. Perreault N. Brestelli J.E. et al.Neurogenin 3 is essential for the proper specification of gastric enteroendocrine cells and the maintenance of gastric epithelial cell identity.Genes Dev. 2002; 16: 1488-1497Crossref PubMed Scopus (249) Google Scholar The following antibodies were used: anti-BrdU (1:50; Roche Diagnostic), anti-lysozyme (1:2; DakoCytomation, Mississauga, Ontario, Canada), anti–phosphorylated-Smad 1-5-8 (1:200; Cell Signaling, Boston, MA), anti–chromogranin A (1:2; Zymed), anti–intestinal fatty acid binding protein (1:5000; kindly provided by Dr. Gordon), anti-Bmpr1a (1:50; Abgent, San Diego, CA), anti–activated-caspase3 (1:1500; Cell Signaling), and fluorescein isothiocyanate–conjugated anti-rabbit or mouse immunoglobulin G (1:200; Vector, Burlington, Ontario, Canada). In situ hybridization was performed according to the Wilkinson14Wilkinson D.G. In situ hybridization: a practical approach. Oxford University Press, IRL Press, NY1992Google Scholar procedure, modified by Lantz et al,15Lantz K.A. Vatamaniuk M.Z. Brestelli J.E. et al.Foxa2 regulates multiple pathways of insulin secretion.J Clin Invest. 2004; 114: 512-520Crossref PubMed Scopus (145) Google Scholar and sucrase-isomaltase riboprobes were synthesized as previously described.16Traber P.G. Regulation of sucrase-isomaltase gene expression along the crypt-villus axis of rat small intestine.Biochem Biophys Res Commun. 1990; 173: 765-773Crossref PubMed Scopus (84) Google Scholar Total protein was isolated from the jejunum as described previously.3Perreault N. Sackett S.D. Katz J.P. et al.Foxl1 is a mesenchymal modifier of min in carcinogenesis of stomach and colon.Genes Dev. 2005; 19: 311-315Crossref PubMed Scopus (65) Google Scholar Nuclear proteins were isolated using the ProteoExtract Subcellular Proteome Extraction kit (EMD Biosciences, San Diego, CA) following the manufacturer's protocol. Fifty (total) or 15 (nuclear) micrograms of protein were separated as described previously.3Perreault N. Sackett S.D. Katz J.P. et al.Foxl1 is a mesenchymal modifier of min in carcinogenesis of stomach and colon.Genes Dev. 2005; 19: 311-315Crossref PubMed Scopus (65) Google Scholar The following antibodies were used and incubated overnight at 4°C: anti–c-myc (1:200), anti–cyclin D1/D2 (1:1000), anti-actin (1:5000), anti–histone H1 (1:1000), anti-rabbit or goat–horseradish peroxidase (1:3000) (all from Santa Cruz Biotech, Santa Cruz, CA), and anti–β-catenin (1:2500) (Cell Signaling). Images of alcian blue–labeled goblet cells and chromogranin A–stained enteroendocrine cells were acquired and positive cells were counted per crypt/villus axis. Nuclei were counted using 4′,6-diamidino-2-phenylindole. Variation of crypt number and crypt fission was assessed by counting the total number of crypts, villi, and crypt fissions in a blind manner. Paneth cells were quantified with lysozyme immunofluorescence. Proliferation was measured by counting BrdU-labeled cells per crypt. Statistics were calculated using the Mann–Whitney or the Student 2-tailed t test. All statistical analyses were performed using Systat10 (Systat software Inc., San Jose, CA). To determine the specific contribution of epithelial Bmp signaling on intestinal physiology and cell function, we used the tissue-specific gene deletion of Bmpr1a using the Cre-loxP system. To delete the Bmpr1a gene exclusively in the epithelium, we crossed Bmpr1aloxP/loxP mice with a Villin-Cre transgenic line, which directs Cre expression in the epithelium of the small intestine and colon, but not in the mesenchymal compartment.11Madison B.B. Dunbar L. Qiao X.T. et al.Cis elements of the villin gene control expression in restricted domains of the vertical (crypt) and horizontal (duodenum, cecum) axes of the intestine.J Biol Chem. 2002; 277: 33275-33283Crossref PubMed Scopus (601) Google Scholar, 12Mishina Y. Hanks M.C. Miura S. et al.Generation of Bmpr/Alk3 conditional knockout mice.Genesis. 2002; 32: 69-72Crossref PubMed Scopus (229) Google ScholarVillin-Cre;Bmpr1aloxP/loxP mice survived to at least 1 year of age and grew normally. Bmpr1a immunofluorescence of control mice (Figure 1A) revealed a strong expression in the upper portion of the crypt unit and villus, with weaker staining in cells at the bottom of the crypt (Figure 1E). Bmpr1a expression was lost exclusively in the intestinal epithelium of the crypt-villus axis in Villin-Cre;Bmpr1aloxP/loxP mice but not in the mesenchymal cells of the lamina propria (Figure 1B and F). Loss of Bmp signaling exclusively in the intestinal epithelium was assessed by immunofluorescence of phosphorylated-Smad 1-5-8. Phosphorylation and nuclear translocation of these Bmp effectors are features of an active Bmp signaling pathway.9Mishina Y. Function of bone morphogenetic protein signaling during mouse development.Front Biosci. 2003; 8: d855-d869Crossref PubMed Google Scholar Phosphorylated-Smad 1-5-8 immunostaining in control mice is observed mainly in villus epithelial cells, as well as at the crypt–villus junction, and to a lesser extent at the bottom of the crypt near the stem and Paneth cell region (Figure 1C and G). Immunoreactivity to phosphorylated-Smad 1-5-8 is lost in Villin-Cre;Bmpr1aloxP/loxP mice (Figure 1D and H). The expression pattern of phosphorylated-Smad 1-5-8 is predominantly in regions of the crypt–villus axis composed of differentiating and terminally differentiated cells, suggesting a possible role for Bmp in intestinal epithelial cytodifferentiation. Histologic analysis showed that the intestine of Villin-Cre;Bmpr1aloxP/loxP mice displayed abnormal epithelial morphology with elongated villi and multiplication of crypt units (Figure 1J). Statistical analysis of the number of crypt units over the number of villi shows a significant 1.37-fold increase of crypt units (Figure 1K) in Villin-Cre;Bmpr1aloxP/loxP mice, suggesting a role for Bmp signaling in crypt fission. A statistical analysis of the percentage of crypt fission found in mutant compared with control animals revealed a 4.62-fold increase in crypt fission in Villin-Cre;Bmpr1aloxP/loxP mice (Figure 1L). We noticed the absence of de novo crypts along the villus as well as polyp growth in Villin-Cre;Bmpr1aloxP/loxP mice, even in 1-year-old animals. The re-initiating of crypt morphogenesis in the villi associated with juvenile polyposis syndrome has been reported in other mouse models with impaired Bmp signaling.4Madison B.B. Braunstein K. Kuizon E. et al.Epithelial hedgehog signals pattern the intestinal crypt-villus axis.Development. 2005; 132: 279-289Crossref PubMed Scopus (307) Google Scholar, 7Haramis A.P. Begthel H. van den Born M. et al.De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine.Science. 2004; 303: 1684-1686Crossref PubMed Scopus (636) Google Scholar, 8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar These results suggest different roles for mesenchymal as opposed to epithelial Bmp signaling in crypt morphogenesis and epithelial cell function. We next analyzed whether proliferation and/or programmed cell death were altered in these mice. Proliferation assays in Villin-Cre;Bmpr1aloxP/loxP mice showed a significant increase in the number of proliferating cells, which was not confined to the bottom but was scattered throughout the crypt (Figure 2B). Statistical analysis revealed a significant increase in proliferative cells of 1.76-fold in Bmpr1a mutant mice (Figure 2E). No difference in activated–caspase-3 immunostaining was observed between control and mutant mice (data not shown). These results confirmed that the thickening of the intestinal mucosa seen in the Villin-Cre;Bmpr1aloxP/loxP mice was a direct outcome of deregulation of the proliferation program of the epithelial cells. To verify the involvement of the Wnt/β-catenin signaling pathway in the deregulation of epithelial cell proliferation in Villin-Cre;Bmpr1aloxP/loxP mice, we performed Western blot analysis with nuclear epithelial intestinal extracts. As shown in Figure 2F, loss of epithelial Bmp signaling does not affect β-catenin nuclear expression, suggesting no increase in Wnt/β-catenin signaling activity. We then analyzed c-myc and cyclin D1 (Figure 2G) and showed that c-myc is not modulated, whereas cyclin D1 is up-regulated in Villin-Cre;Bmpr1aloxP/loxP mice. Cyclin D1 expression is up-regulated not only by the Wnt/β-catenin pathway, but also by many other signaling pathways such as the Ras/Erk/mitogene activated protein kinase pathway.17Lavoie J.N. L'Allemain G. Brunet A. et al.Cyclin D1 expression is regulated positively by the p42/p44MAPK and negatively by the p38/HOGMAPK pathway.J Biol Chem. 1996; 271: 20608-20616Crossref PubMed Scopus (1101) Google Scholar To characterize changes in the differentiation and maturation of the 4 intestinal cell types in Villin-Cre;Bmpr1aloxP/loxP mice, we stained the mucosa with specific markers for each cell type. Goblet cells were stained with alcian blue (Figure 3A and B) and counted. No significant change was found between Bmpr1a mutant mice and control littermates (Figure 3K), however, the goblet cells were consistently smaller in the mutant mice, suggesting a lack of maturity of these cells. Indeed, smaller mucigen granules and reduction in size were observed in the mutant animals (Figure 3H). No significant change was observed in the expression of the goblet cell–specific marker Muc2 between the mutant and control animals (Table 1). Comparative analysis by quantitative polymerase chain reaction of both Klf4 and Tff3, factors that regulate goblet cell terminal differentiation,18Katz J.P. Perreault N. Goldstein B.G. et al.The zinc-finger transcription factor Klf4 is required for terminal differentiation of goblet cells in the colon.Development. 2002; 129: 2619-2628Crossref PubMed Google Scholar, 19Podolsky D.K. Lynch-Devaney K. Stow J.L. et al.Identification of human intestinal trefoil factor Goblet cell-specific expression of a peptide targeted for apical secretion.J Biol Chem. 1993; 268: 6694-6702Abstract Full Text PDF PubMed Google Scholar revealed a decrease in expression of 3.47-fold and 2.06-fold, respectively, in the mutant animals (Table 1). Hence, these data show that mice with impaired Bmp signaling show alterations in goblet cell terminal differentiation and maturity.Table 1Gene Expression Changes in the Jejunum of Villin-Cre;Bmpr1aloxP/loxP MiceGene descriptionGene symbolFoldPCell determination genes Math 1Math 1−1.95.05 Hairy and enhancer of split 1Hes1.32NSEnterocyte markers E74-like factor 3Elf31.14NS Sucrase-isomaltaseSI−1.75NSEnteroendocrine cell markers Neurogenin 3Ngn3−2.25.025 Neurogenic differentiation 1Neurod1−3.33.025 Chromogranin AChga−2.04.025Paneth cell markers CryptdinDefcr−5.24.025 LysozymeLyz−4.80.003 Matrix metallopeptidase 7MMP7−2.25.025Goblet cell markers Mucin 2Muc2−1.19NS Kruppel-like factor 4Klf4−3.47.024 Trefoil factor 3Tff3−2.06.05Hedgehog signaling pathway Indian HedgehogIHH1.49NS GLI-Kruppel family member 1Gli1−2.82NS Patched 1Ptc−1.89NSNOTE. Fold changes represent the ratio of mean expression values (control/mutant). Negative values indicate reduction in Villin-Cre;Bmpr1aloxP/loxP intestines.NS, nonsignificant fold change (Mann–Whitney test). Open table in a new tab NOTE. Fold changes represent the ratio of mean expression values (control/mutant). Negative values indicate reduction in Villin-Cre;Bmpr1aloxP/loxP intestines. NS, nonsignificant fold change (Mann–Whitney test). Bmp signaling activity at the bottom of the crypt near the stem and Paneth cell regions (Figure 1G) has been associated exclusively with stem cell renewal.8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar Herein, we observed a more compact staining of lysozyme, a Paneth cell marker, in Bmpr1a mutant mice (Figure 3D) as compared with the control (Figure 3C), however, the location of these cells was not affected by the loss of epithelial Bmp signaling. Statistical analysis revealed no significant decrease in the number of Paneth cells in the mutant animals (Figure 3K), suggesting that the determination of Paneth precursor cells is not affected by the loss of epithelial Bmp signaling. Thus, we hypothesized that Paneth cells were not differentiated fully in the mutant mice. Indeed, a reduction in the amount of Paneth cell secretory granule content was noted in mutant mice (Figure 3J). Finally, a decrease of 2.28-fold in matrilysin, 4.8-fold in lysozyme, and 5.24-fold in cryptdin messenger RNA (mRNA) expressions, all important markers for Paneth cell functionality and maturation (Table 1),20Weeks C.S. Tanabe H. Cummings J.E. et al.Matrix metalloproteinase-7 activation of mouse paneth cell pro-alpha-defensins: SER43 down arrow ILE44 proteolysis enables membrane-disruptive activity.J Biol Chem. 2006; 281: 28932-28942Crossref PubMed Scopus (38) Google Scholar, 21van Es J.H. Jay P. Gregorieff A. et al.Wnt signalling induces maturation of Paneth cells in intestinal crypts.Nat Cell Biol. 2005; 7: 381-386Crossref PubMed Scopus (528) Google Scholar was observed in the mutant mice. Expression of Indian Hedgehog, which negatively regulates the differentiation of Paneth precursor cells into fully mature cells,22Varnat F. Heggeler B.B. Grisel P. et al.PPARbeta/delta regulates paneth cell differentiation via controlling the hedgehog signaling pathway.Gastroenterology. 2006; 131: 538-553Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar as well as its downstream targets Gli-1 and Patched-1, were not affected by the loss of epithelial Bmp signaling (Table 1). These results support that Paneth cell terminal differentiation as well as maintenance can be linked to the instructive interaction of epithelial Bmp signaling. We next analyzed whether enteroendocrine cells were affected by the loss of epithelial Bmp signaling. Immunostaining with chromogranin A revealed fewer positive cells in Villin-Cre;Bmpr1aloxP/loxP mice (Figure 3F). We found a significant decrease of 75% of enteroendocrine cells per crypt–villus axis in the mutant mice (Figure 3K), strongly suggesting that the enteroendocrine precursor is affected by the loss of epithelial Bmp signaling. Accordingly, we noted a 2.04-fold decrease in chromogranin A mRNA levels in the mutant animals (Table 1). Comparative analysis by quantitative polymerase chain reaction of proenteroendocrine factors, such as Neurogenin 3 (Ngn3) and BETA2/NeuroD1,13Lee C.S. Perreault N. Brestelli J.E. et al.Neurogenin 3 is essential for the proper specification of gastric enteroendocrine cells and the maintenance of gastric epithelial cell identity.Genes Dev. 2002; 16: 1488-1497Crossref PubMed Scopus (249) Google Scholar, 23Mutoh H. Fung B.P. Naya F.J. et al.The basic helix-loop-helix transcription factor BETA2/NeuroD is expressed in mammalian enteroendocrine cells and activates secretin gene expression.Proc Natl Acad Sci U S A. 1997; 94: 3560-3564Crossref PubMed Scopus (155) Google Scholar revealed a significant decrease of 2.25-fold and 3.33-fold, respectively, in the mutant animals (Table 1). Thus, loss of epithelial Bmp signaling caused a reduction in Ngn3 and BETA2/NeuroD expression, leading to improper specification and differentiation of enteroendocrine cells. We then investigated whether absorptive cell differentiation also was affected in Bmpr1a mutant mice. Differentiated cells of the absorptive lineage express specific markers such as intestinal fatty acid binding protein and sucrase-isomaltase.1Ménard D. Beaulieu J.-F. Boudreau F. et al.Cell signaling and growth factors in development: from molecules to organogenesis.in: Unsicker K. Krieglstein K. Gastrointestinal tract. Volume 2. Wiley-Vch, Weinheim2006: 755-790Google Scholar In situ hybridization for sucrase-isomaltase (Figure 4A and B) and immunostaining for intestinal fatty acid binding protein (Figure 4C and D) revealed no modulation in these markers. No ultrastructural modification was noted between absorptive cells from mutant or control mice (Figure 4E and F) and no significant modulation of sucrase-isomaltase mRNA levels was observed (Table 1). Thus, lack of terminal differentiation in mice with intestinal epithelial impaired Bmp signaling is restricted to cells of the secretory lineage. Given the central role of hairy enhancer of split-1 and Math1 in intestinal epithelial specification,24Yang Q. Bermingham N.A. Finegold M.J. et al.Requirement of Math1 for secretory cell lineage commitment in the mouse intestine.Science. 2001; 294: 2155-2158Crossref PubMed Scopus (768) Google Scholar and that Bmpr1a mutant mice have impaired terminal differentiation of the secretory cell lineage, we analyzed their levels of expression. A slight 1.95-fold decrease of Math1 but no significant modulation of hairy enhancer of split mRNA levels were noted in the mutant animals (Table 1). Hence, loss of epithelial Bmp signaling is not sufficient to directly affect the secretory precursor cells or to change the fate of cells from the secretory to the absorptive cell lineage. Intercellular communications are decisive for intestinal morphogenesis as well as for the establishment and functional maintenance of the crypt–villus axis.3Perreault N. Sackett S.D. Katz J.P. et al.Foxl1 is a mesenchymal modifier of min in carcinogenesis of stomach and colon.Genes Dev. 2005; 19: 311-315Crossref PubMed Scopus (65) Google Scholar, 4Madison B.B. Braunstein K. Kuizon E. et al.Epithelial hedgehog signals pattern the intestinal crypt-villus axis.Development. 2005; 132: 279-289Crossref PubMed Scopus (307) Google Scholar, 5Roberts D.J. Molecular mechanisms of development of the gastrointestinal tract.Dev Dyn. 2000; 219: 109-120Crossref PubMed Scopus (181) Google Scholar Studies have identified the Bmp signaling pathway as a possible key mediator in crypt and villus morphogenesis as well as maintenance of intestinal stem cells.4Madison B.B. Braunstein K. Kuizon E. et al.Epithelial hedgehog signals pattern the intestinal crypt-villus axis.Development. 2005; 132: 279-289Crossref PubMed Scopus (307) Google Scholar, 7Haramis A.P. Begthel H. van den Born M. et al.De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine.Science. 2004; 303: 1684-1686Crossref PubMed Scopus (636) Google Scholar, 8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar However, the specific roles and contributions of the Bmp signaling pathway on one particular cell compartment compared with another in the maintenance of adult epithelial intestinal cell function has not yet been delineated. The current study generated a murine model with impaired Bmp signaling exclusively in the epithelial compartment of the intestine. In this study, we provide evidence that epithelial intestinal Bmp signaling plays a critical role in the maintenance of epithelial architecture within the crypt–villus axis. We showed that loss of epithelial Bmp signaling leads to elongated villi and multiplication of the crypt units. The latter observation suggests a possible role for Bmp signaling in crypt fission. The role of Bmp signaling in stem cell renewal and consequently on crypt fission already has been addressed in previous work.8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar We have found the same occurrence in our Villin-Cre;Bmpr1aloxP/loxP mice, suggesting that the regulation of the crypt fission found at the bottom of the crypt unit is associated with epithelial Bmp signaling. Interestingly, and in major contrast with previous studies,4Madison B.B. Braunstein K. Kuizon E. et al.Epithelial hedgehog signals pattern the intestinal crypt-villus axis.Development. 2005; 132: 279-289Crossref PubMed Scopus (307) Google Scholar, 7Haramis A.P. Begthel H. van den Born M. et al.De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine.Science. 2004; 303: 1684-1686Crossref PubMed Scopus (636) Google Scholar, 8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar there is an absence of the de novo crypt phenomenon as well as polyp growth in Villin-Cre;Bmpr1aloxP/loxP mice even in 1-year-old animals. The key difference between the present study and those referred to previously is that herein, the Bmp signaling pathway is affected exclusively in the intestinal epithelium, whereas in previous studies both the epithelial and mesenchymal compartments were affected. These results suggest that the Bmp signaling pathway in the mesenchyme plays a significant role in crypt morphogenesis and juvenile polyposis syndrome whereas epithelial signalization plays a role in cell proliferation, stem cell renewal,8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar and maintenance of epithelial cell differentiation. We have investigated how epithelial Bmp signaling can suppress or restrain epithelial cell proliferation. Unexpectedly, we showed that the loss of epithelial Bmp signaling does not lead to an increase in β-catenin nuclear expression, suggesting no increase in Wnt/β-catenin signaling activity. These results are contradictory to those recently reported by He et al,8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar who reported an increase in nuclear β-catenin levels associated with an activation of target genes involved in cell-cycle progression. This increase in nuclear β-catenin appears to be the consequence of the inhibition of phosphatase and tensin homologue deleted on chromosome 10 (PTEN) activity in the intestinal crypt epithelium of Mx1-Cre;Bmpr1aloxP/loxP mice, resulting in the activation of the phosphatidylinositol 3-kinase (PI3K)/Akt pathway and the inhibition of Gsk3β activity. He et al suggested that Bmp signaling inhibits intestinal stem cell self-renewal through activation of PTEN, leading to suppression of Wnt/β-catenin signaling. However, their conclusions have raised some controversy in this field of research. Bjerknes and Cheng25Bjerknes M. Cheng H. Re-examination of P-PTEN staining patterns in the intestinal crypt.Nat Genet. 2005; 37: 1016-1018Crossref PubMed Scopus (14) Google Scholar have shown that the phosphorylated form of PTEN (p-PTEN)-positive cells identified by He et al8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar as intestinal stem cells are more likely to be the postmitotic secreting cells frequently found at the crypt base. After our result, we analyzed 2 known target genes of the Wnt/β-catenin pathway in our Villin-Cre;Bmpr1aloxP/loxP mice, namely c-myc and cyclin D1. Interestingly, we showed that c-myc is not modulated, whereas cyclin D1 is up-regulated in Villin-Cre;Bmpr1aloxP/loxP mice. Of note, cyclin D1 expression is enhanced not only by the Wnt/β-catenin pathway but also by many other signaling pathways such as the Ras/Erk/mitogene activated protein kinase pathway,17Lavoie J.N. L'Allemain G. Brunet A. et al.Cyclin D1 expression is regulated positively by the p42/p44MAPK and negatively by the p38/HOGMAPK pathway.J Biol Chem. 1996; 271: 20608-20616Crossref PubMed Scopus (1101) Google Scholar which also controls intestinal epithelial crypt cell proliferation.26Rivard N. Boucher M.J. Asselin C. et al.MAP kinase cascade is required for p27 downregulation and S phase entry in fibroblasts and epithelial cells.Am J Physiol. 1999; 277: C652-C664PubMed Google Scholar, 27Aliaga J.C. Deschenes C. Beaulieu J.F. et al.Requirement of the MAP kinase cascade for cell cycle progression and differentiation of human intestinal cells.Am J Physiol. 1999; 277: G631-G641PubMed Google Scholar Hence, one could speculate that epithelial Bmp signaling may inhibit proliferation of intestinal epithelial cells by interfering with the activation of this pathway instead of the Wnt/β-catenin pathway. In addition, epithelial Bmp signaling may directly control intestinal cell-cycle progression by modulating gene expression of cell inhibitors such as p21Cip/Waf, p27Kip, and p57 as reported previously in colon cancer cells28Beck S.E. Jung B.H. Del Rosario E. et al.BMP-induced growth suppression in colon cancer cells is mediated by p21(WAF1) stabilization and modulated by RAS/ERK.Cell Signal. 2007; 19: 1465-1472Crossref PubMed Scopus (51) Google Scholar and in human keratinocytes.29Gosselet F.P. Magnaldo T. Culerrier R.M. et al.BMP2 and BMP6 control p57(Kip2) expression and cell growth arrest/terminal differentiation in normal primary human epidermal keratinocytes.Cell Signal. 2007; 19: 731-739Crossref PubMed Scopus (50) Google Scholar Because the Bmpr1a gene ablation in the study with the Mx1-Cre;Bmpr1aloxP/loxP mice was in both the epithelial and stromal compartments of the gut, we think it is possible that the activation of the Wnt/β-catenin pathway is associated with the loss of mesenchymal Bmp signaling. Such action of the mesenchyme on the epithelial Wnt/β-catenin pathway has been shown previously and contributes to gastrointestinal epithelial tumor initiation.3Perreault N. Sackett S.D. Katz J.P. et al.Foxl1 is a mesenchymal modifier of min in carcinogenesis of stomach and colon.Genes Dev. 2005; 19: 311-315Crossref PubMed Scopus (65) Google Scholar Our study revealed an important and novel function for epithelial intestinal Bmp signaling in the proper terminal differentiation of cells from the intestinal secretory lineage. Our data suggest that key regulators of goblet cell terminal differentiation such as Tff3 and Klf4 are down-regulated in the mutant mice. Interestingly, Klf4 has been shown to be a target of Bmp signaling and to play a role in the maintenance of the vascular smooth muscle cell phenotype.30King K.E. Iyemere V.P. Weissberg P.L. et al.Kruppel-like factor 4 (KLF4/GKLF) is a target of bone morphogenetic proteins and transforming growth factor beta 1 in the regulation of vascular smooth muscle cell phenotype.J Biol Chem. 2003; 278: 11661-11669Crossref PubMed Scopus (110) Google Scholar Here, we show that epithelial loss of Bmp signaling results in a significant decrease of Klf4 mRNA and protein levels (data not shown), which leads to an important reduction in terminal differentiation of intestinal goblet cells. Our analysis exposed a dramatic decrease of 75% of enteroendocrine cells per crypt–villus axis in the jejunum of Bmpr1a mutant mice as well as a significant decrease in proenteroendocrine factors such as Ngn3 and BETA2/NeuroD. These results strongly suggest that the enteroendocrine precursor cell, which is derived from the secretory precursor, is affected by the loss of epithelial Bmp signaling. This could be explained partially by the reduction in Ngn3 because this proenteroendocrine factor is known to play a role in the proliferation of the endocrine precursor.13Lee C.S. Perreault N. Brestelli J.E. et al.Neurogenin 3 is essential for the proper specification of gastric enteroendocrine cells and the maintenance of gastric epithelial cell identity.Genes Dev. 2002; 16: 1488-1497Crossref PubMed Scopus (249) Google Scholar Our data show a previously unknown role for epithelial Bmp signaling in enteroendocrine cell determination, as well as final specification through its involvement with the proenteroendocrine factors Ngn3 and BETA2/NeuroD. The Bmp signaling activity found at the bottom of the crypts thus far has been associated with stem cell renewal.8He X.C. Zhang J. Tong W.G. et al.BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling.Nat Genet. 2004; 36: 1117-1121Crossref PubMed Scopus (887) Google Scholar The differentiated Paneth cells found at the bottom of the crypts do not arise from direct differentiation of nearby stem or progenitor cells, but arrive there by the downward migration of already committed cells of the Paneth phenotype.2Bjerknes M. Cheng H. Gastrointestinal stem cells II. Intestinal stem cells.Am J Physiol. 2005; 289: G381-G387Crossref PubMed Scopus (113) Google Scholar Our study shows that Paneth cell terminal differentiation can be linked to the instructive interaction of Bmp signaling. Because we observed no modulation in the number of Paneth cells in the mutant animals, this strongly suggests that the determination of Paneth precursor cells is not affected by the loss of epithelial Bmp signaling. Interestingly, Indian Hedgehog, the main representative of the hedgehog pathway in the gut, negatively regulates differentiation of Paneth precursor cells into fully mature cells.22Varnat F. Heggeler B.B. Grisel P. et al.PPARbeta/delta regulates paneth cell differentiation via controlling the hedgehog signaling pathway.Gastroenterology. 2006; 131: 538-553Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar The investigation of Indian Hedgehog expression revealed no significant modulation for this gene in mice with impaired epithelial Bmp signaling. Hence, Bmp signaling affects terminal differentiation and maturation of Paneth cells by a different mechanism than either the hedgehog or Wnt pathways. However, the decrease in matrilysin expression, which is also an important regulator of Paneth cell terminal differentiation,20Weeks C.S. Tanabe H. Cummings J.E. et al.Matrix metalloproteinase-7 activation of mouse paneth cell pro-alpha-defensins: SER43 down arrow ILE44 proteolysis enables membrane-disruptive activity.J Biol Chem. 2006; 281: 28932-28942Crossref PubMed Scopus (38) Google Scholar could provide some explanation for a possible mechanism for the impaired Paneth cell terminal differentiation. Thus, our results reveal a distinct role for Bmp signaling at the bottom of the crypt in the terminal differentiation of Paneth cells. The phenotypes observed in the Bmpr1a mutant may suggest a greater role for Bmp signaling in the specification of intestinal precursor cells. Still, the change seen in Math1 expression in animals with impaired epithelial Bmp signaling might not be sufficient to directly affect precursor cells or to change the fate of cells from the secretory to the absorptive cell lineage. These observations clearly show that the loss of epithelial Bmp signaling affects the differentiation potential of the precursor cells, preventing them from fully acquiring their mature and functional phenotype. In summary, we show that epithelial Bmp signaling actively plays a role in terminal differentiation and maturation of cells from the secretory lineage. Most importantly, we have shown that the loss of Bmp signaling exclusively in the intestinal epithelial compartment is not responsible for the de novo crypt phenomenon, suggesting a greater role for mesenchymal Bmp signaling in juvenile polyposis syndrome. The authors thank Pierre Chailler for his contribution to Figure 5, Elizabeth Herring for critical reading of the manuscript, Dr. Gumucio for providing the 12.4KbVilCre transgenic line, Dr. Gordon for providing the intestinal fatty acid binding protein antibody, and Denis Martel for assistance with electron microscopy. BMP Signaling in the Intestine: Cross-Talk Is KeyGastroenterologyVol. 133Issue 3PreviewThe transforming growth factor-β (TGF-β) superfamily of cytokines—including members of the TGF-β, bone morphogenetic protein (BMP), and activin proteins—plays important roles in development and tissue homeostasis. TGF-β and BMP ligands bind their respective receptors to activate SMAD-mediated transcription. These signaling pathways regulate growth, apoptosis, and differentiation of intestinal epithelial cells,1 and therefore are often mutated in intestinal diseases such as familial juvenile polyposis (FJP). Full-Text PDF
Referência(s)