Portal de Periódicos da CAPES

Stomach Development Is Dependent on Fibroblast Growth Factor 10/Fibroblast Growth Factor Receptor 2b–Mediated Signaling

2006; Elsevier BV; Volume: 130; Issue: 4 Linguagem: Inglês

10.1053/j.gastro.2006.02.018

1528-0012

Bradley Spencer‐Dene, Frédéric Sala, Savério Bellusci, Stephen Gschmeissner, Gordon Stamp, Clive Dickson,

Congenital Diaphragmatic Hernia Studies

Background & Aims: Fibroblast growth factors (Fgfs) and their receptors (Fgfrs) are important intercellular signaling molecules that are essential to mammalian embryonic development. The signaling pathways between endoderm-derived gastric epithelium and the surrounding mesenchyme are largely unknown; however, the developmental expression profile of the IIIb isoform of Fgfr2 (Fgfr2b) and its main ligand, Fgf10, suggest that they may be strong candidates. Mice lacking either component (Fgfr2b−/− or Fgf10−/−) were examined to determine the role of Fgfr2b-mediated signaling during gastric organogenesis. Methods: Stomachs from embryonic day 13.5–18.5 Fgfr2b−/−, Fgf10−/−, and wild-type littermates were collected and analyzed by conventional histology, immunohistochemistry, in situ hybridization, and electron microscopy. Results: Fgfr2b−/− and Fgf10−/− fetuses had stomachs smaller than wild-type, consisting of relatively proportionate forestomach but disproportionately reduced glandular stomach, the mucosa of which has low cytoarchitectural complexity with a spiral arrangement of large mucosal folds. During mid to late fetal stages (embryonic day 15.5–18.5), epithelial differentiation to mucous and chief cell lineages was rudimentary, with no expression of several early cytodifferentiation markers including GATA4, GATA6, and H+/K+-adenosine triphosphatase and abnormal expression of members of the hedgehog family of signaling molecules. Conclusions: Fgfr2b and Fgf10 are part of a signaling network with Sonic hedgehog and Indian hedgehog that are essential to anterior-posterior and radial patterning in gastric development. Background & Aims: Fibroblast growth factors (Fgfs) and their receptors (Fgfrs) are important intercellular signaling molecules that are essential to mammalian embryonic development. The signaling pathways between endoderm-derived gastric epithelium and the surrounding mesenchyme are largely unknown; however, the developmental expression profile of the IIIb isoform of Fgfr2 (Fgfr2b) and its main ligand, Fgf10, suggest that they may be strong candidates. Mice lacking either component (Fgfr2b−/− or Fgf10−/−) were examined to determine the role of Fgfr2b-mediated signaling during gastric organogenesis. Methods: Stomachs from embryonic day 13.5–18.5 Fgfr2b−/−, Fgf10−/−, and wild-type littermates were collected and analyzed by conventional histology, immunohistochemistry, in situ hybridization, and electron microscopy. Results: Fgfr2b−/− and Fgf10−/− fetuses had stomachs smaller than wild-type, consisting of relatively proportionate forestomach but disproportionately reduced glandular stomach, the mucosa of which has low cytoarchitectural complexity with a spiral arrangement of large mucosal folds. During mid to late fetal stages (embryonic day 15.5–18.5), epithelial differentiation to mucous and chief cell lineages was rudimentary, with no expression of several early cytodifferentiation markers including GATA4, GATA6, and H+/K+-adenosine triphosphatase and abnormal expression of members of the hedgehog family of signaling molecules. Conclusions: Fgfr2b and Fgf10 are part of a signaling network with Sonic hedgehog and Indian hedgehog that are essential to anterior-posterior and radial patterning in gastric development. Patterning of the mammalian gastrointestinal tract during development starts with an invagination of the prospective gut endoderm to form a primitive gut tube. This is followed by rostrocaudal and subsequent radial differentiation into several distinct cell lineages. This process is orchestrated by a wide variety of inductive signaling molecules operating primarily through epithelial-mesenchymal interactions. The adult mouse stomach epithelium is patterned rostrocaudally, beginning with the keratinized stratified squamous forestomach epithelium, which starts differentiating from embryonic day (E) 16.5. This region is devoid of glandular structures and histologically similar to esophagus. The gastric body region is distal and is subdivided into zymogenic and mucous glandular zones, both commencing as columnar epithelia that begins to stratify and form intraepithelial vacuoles from E13.5 and then invaginating into the lamina propria, forming gastric units.1Fukamachi H. Mizuno T. Takayama S. Epithelial-mesenchymal interactions in differentiation of stomach epithelium in fetal mice.Anat Embryol. 1979; 157: 151-160Crossref PubMed Scopus (50) Google Scholar, 2Karam S.M. Li Q. Gordon J.I. Gastric epithelial morphogenesis in normal and transgenic mice.Am J Physiol. 1997; 272: G1209-G1220PubMed Google Scholar Zymogenic glands, whose presence defines the zymogenic region, contain pepsinogen-secreting chief cells, acid-secreting parietal cells, and mucous-secreting pit cells as well as enteroendocrine cells and presumptive gastric stem cells. Mucus glands, characteristic of the mucous region, lack chief cells, and only a proportion of them contain parietal cells. In the posterior region of the stomach is the pyloric antrum, lined by mucoparietal and simple mucous glands.3Lee E.R. Trasler J. Dwivedi S. Leblond C.P. Division of the mouse gastric mucosa into zymogenic and mucous regions on the basis of gland features.Am J Anat. 1982; 164: 187-207Crossref PubMed Scopus (84) Google Scholar, 4Lee E.R. Dynamic histology of the antral epithelium in the mouse stomach: I. Architecture of antral units.Am J Anat. 1985; 172: 187-204Crossref PubMed Scopus (34) Google Scholar Several candidate signaling molecules and transcription factors controlling stomach regionalization during vertebrate development have been identified, including 2 members of the hedgehog family, Sonic (Shh) and Indian (Ihh), and homeobox genes.5Bitgood M.J. McMahon A.P. Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo.Dev Biol. 1995; 172: 126-138Crossref PubMed Scopus (1198) Google Scholar, 6Kim S.K. Hebrok M. Li E. Oh S.P. Schrewe H. Harmon E.B. Lee J.S. Melton D.A. Activin receptor patterning of foregut organogenesis.Genes Dev. 2000; 14: 1866-1871PubMed Google Scholar, 7Ramalho-Santos M. Melton D.A. McMahon A.P. Hedgehog signals regulate multiple aspects of gastrointestinal development.Development. 2000; 127: 2763-2772Crossref PubMed Google Scholar, 8Van den Brink G.R. Hardwick J.C. Tytgat G.N. Brink M.A. Ten Kate F.J. Van Deventer S.J. Peppelenbosch M.P. Sonic hedgehog regulates gastric gland morphogenesis in man and mouse.Gastroenterology. 2001; 121: 317-328Abstract Full Text Full Text PDF PubMed Scopus (220) Google Scholar, 9Aubin J. Dery U. Lemieux M. Chailler P. Jeannotte L. Stomach regional specification requires Hoxa5-driven mesenchymal-epithelial signaling.Development. 2002; 129: 4075-4087Crossref PubMed Google Scholar, 10Kim B.M. Buchner G. Miletich I. Sharpe P.T. Shivdasani R.A. The stomach mesenchymal transcription factor Barx1 specifies gastric epithelial identity through inhibition of transient Wnt signalling.Dev Cell. 2005; 8: 611-622Abstract Full Text Full Text PDF PubMed Scopus (166) Google Scholar The fibroblast growth factor (Fgf) family of intercellular signaling molecules comprises 22 members, many of them essential to vertebrate development as regulators of processes such as cell proliferation, migration, and differentiation. They specifically bind to and activate a family of 4 tyrosine kinase transmembrane receptors (fibroblast growth factor receptors [Fgfrs]) in the presence of heparan sulfate proteoglycan.11McKeehan W.L. Wang F. Kan M. The heparan sulfate-fibroblast growth factor family diversity of structure and function.Prog Nucleic Acid Res Mol Biol. 1998; 59: 135-176Crossref PubMed Scopus (364) Google Scholar, 12Ornitz D.M. FGFs, heparan sulfate and FGFRs complex interactions essential for development.Bioessays. 2000; 22: 108-112Crossref PubMed Scopus (637) Google Scholar, 13Ornitz D.M. Itoh N. Fibroblast growth factors.Genome Biol. 2001; 2 (reviews3005.1–reviews3005.12)Crossref PubMed Google Scholar In this report, we are concerned with effects dependent on the b isoform of one member of this family, Fgfr2. Through alternative splicing, the Fgfr2 gene generates 2 tissue-specific receptor isoforms: (1) Fgfr2b, which is expressed on most epithelia, and (2) Fgfr2c, which is usually expressed in mesenchyme. Fgfr2b is activated by the ligands Fgf7 and Fgf10, usually expressed in adjacent mesenchyme.14Ornitz D.M. Xu J. Colvin J.S. McEwen D.G. MacArthur C.A. Coulier F. Gao G. Goldfarb M. Receptor specificity of the fibroblast growth factor family.J Biol Chem. 1996; 271: 15292-15297Crossref PubMed Scopus (1455) Google Scholar Within the developing gastrointestinal tract, Fgfr2b is expressed in the epithelial lining of the esophagus, stomach, and intestine.15Orr-Urtreger A. Bedford M.T. Burakova T. Arman E. Zimmer Y. Yayon A. Givol D. Lonai P. Developmental localization of the splicing alternatives of fibroblast growth factor receptor-2 (FGFR2).Dev Biol. 1993; 158: 475-486Crossref PubMed Scopus (490) Google Scholar, 16Peters K.G. Werner S. Chen G. Williams L.T. Two FGF receptor genes are differentially expressed in epithelial and mesenchymal tissues during limb formation and organogenesis in the mouse.Development. 1992; 114: 233-243Crossref PubMed Google Scholar, 17Matsubara Y. Ichinose M. Tatematsu M. Ichinose M. Oka M. Yahagi N. Kurokawa K. Kageyama T. Miki K. Fukamachi H. Stage-specific elevated expression of the genes for hepatocyte growth factor, keratinocyte growth factor, and their receptors during the morphogenesis and differentiation of rat stomach mucosa.Biochem Biophys Res Commun. 1996; 222: 669-677Crossref PubMed Scopus (23) Google Scholar In contrast, Fgf10 is expressed in the mesenchyme but in a more restricted region corresponding to the developing glandular stomach, cecum, and rectum.9Aubin J. Dery U. Lemieux M. Chailler P. Jeannotte L. Stomach regional specification requires Hoxa5-driven mesenchymal-epithelial signaling.Development. 2002; 129: 4075-4087Crossref PubMed Google Scholar, 18Burns R.C. Fairbanks T.J. Sala F. De Langhe S. Mailleux A. Thiery J.P. Dickson C. Itoh N. Warburton D. Anderson K.D. Bellusci S. Requirement for fibroblast growth factor receptor 2-IIIb signaling for cecal development in mouse.Dev Biol. 2004; 265: 61-74Crossref PubMed Scopus (64) Google Scholar Fgfr2b−/− mice exhibit multiple visceral abnormalities, including lung, pancreas, and cecum defects,18Burns R.C. Fairbanks T.J. Sala F. De Langhe S. Mailleux A. Thiery J.P. Dickson C. Itoh N. Warburton D. Anderson K.D. Bellusci S. Requirement for fibroblast growth factor receptor 2-IIIb signaling for cecal development in mouse.Dev Biol. 2004; 265: 61-74Crossref PubMed Scopus (64) Google Scholar, 19De Moerlooze L. Spencer-Dene B. Revest J. Hajihosseini M. Rosewell I. Dickson C. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signaling during mouse organogenesis.Development. 2000; 127: 483-492Crossref PubMed Google Scholar, 20Revest J.M. Spencer-Dene B. Kerr K. De Moerlooze L. Rosewell I. Dickson C. Fibroblast growth factor receptor 2-IIIb acts upstream of Shh and Fgf4 and is required for limb bud maintenance but not for the induction of Fgf8, Fgf10, Msx1, or Bmp4.Dev Biol. 2001; 231: 47-62Crossref PubMed Scopus (246) Google Scholar, 21Pulkkinen M.A. Spencer-Dene B. Dickson C. Otonkoski T. The IIIb isoform of fibroblast growth factor receptor 2 is required for proper growth and branching of pancreatic ductal epithelium but not for differentiation of exocrine or endocrine cells.Mech Dev. 2003; 120: 167-175Crossref PubMed Scopus (54) Google Scholar but there has been no thorough characterization of the gastric phenotype. Fgf10−/− mice exhibit many but not all of the phenotypes seen in Fgfr2b−/− mice.22Mailleux A.A. Spencer-Dene B. Dillon C. Ndiaye D. Savona-Baron C. Itoh N. Kato S. Dickson C. Thiery J.P. Bellusci S. Role of FGF10/FGFR2b signaling during mammary gland development in the mouse embryo.Development. 2002; 129: 53-60PubMed Google Scholar, 23Min H. Danilenko D.M. Scully S.A. Bolon B. Ring B.D. Tarpley J.E. DeRose M. Simonet W.S. Fgf-10 is required for both limb and lung development and exhibits striking functional similarity to Drosophila branchless.Genes Dev. 1998; 12: 3156-3161Crossref PubMed Scopus (746) Google Scholar, 24Sekine K. Ohuchi H. Fujiwara M. Yamasaki M. Yoshizawa T. Sato T. Yagishita N. Matsui D. Koga Y. Itoh N. Kato S. Fgf10 is essential for limb and lung formation.Nat Genet. 1999; 21: 138-141Crossref PubMed Scopus (1011) Google Scholar, 25Bhushan A. Itoh N. Kato S. Thiery J.P. Czernichow P. Bellusci S. Scharfmann R. Fgf10 is essential for maintaining the proliferative capacity of epithelial progenitor cells during early pancreatic organogenesis.Development. 2001; 128: 5109-5117Crossref PubMed Google Scholar Gastric abnormalities have been reported for Fgf10−/− mice, namely a reduction in stomach size, absence of gastric smooth muscle layers, and abnormal chief cell localization within the gastric pit at birth.26Ohuchi H. Hori Y. Yamasaki M. Harada H. Sekine K. Kato S. Itoh N. FGF10 acts as a major ligand for FGF receptor 2 IIIb in mouse multi-organ development.Biochem Biophys Res Commun. 2000; 277: 643-649Crossref PubMed Scopus (528) Google Scholar We have analyzed in some detail the role of Fgf10/Fgfr2b signaling in fetal stomach development and found that while smooth muscle development is normal, loss of either the ligand or the receptor leads to a radical alteration of stomach growth and patterning. All animal experiments were performed according to UK Home Office regulations. The targeted disruption of Fgfr2b19De Moerlooze L. Spencer-Dene B. Revest J. Hajihosseini M. Rosewell I. Dickson C. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signaling during mouse organogenesis.Development. 2000; 127: 483-492Crossref PubMed Google Scholar and Fgf1023Min H. Danilenko D.M. Scully S.A. Bolon B. Ring B.D. Tarpley J.E. DeRose M. Simonet W.S. Fgf-10 is required for both limb and lung development and exhibits striking functional similarity to Drosophila branchless.Genes Dev. 1998; 12: 3156-3161Crossref PubMed Scopus (746) Google Scholar has been described previously. Both Fgfr2b and Fgf10 heterozygotes were intercrossed to generate homozygous null embryos at the required embryonic stage, and yolk sacs were used for genotyping by polymerase chain reaction.19De Moerlooze L. Spencer-Dene B. Revest J. Hajihosseini M. Rosewell I. Dickson C. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signaling during mouse organogenesis.Development. 2000; 127: 483-492Crossref PubMed Google Scholar The morning of vaginal plug appearance was considered E0.5. Control mice consisted of wild-type (wt) and heterozygous littermates, the latter being phenotypically indistinguishable from wt mice. E13.5–15.5 wt and mutant embryos and isolated stomachs from E16.5–E18.5 mice were fixed in 10% neutral buffered formalin, processed into paraffin, and sectioned for light microscopy. Sections were either stained with H&E or immunostained by the avidin-biotin-peroxidase technique. Parietal cells were identified using a mouse monoclonal antibody to the α subunit of H+/K+–adenosine triphosphatase (ATPase) 4A6 (a gift from Prof Sven Mardh, Linköping University, Linköping, Sweden) and zymogenic cells using a sheep polyclonal anti-pepsinogen (The Binding Site, Birmingham, United Kingdom). Gastric muscle was stained with a peroxidase-conjugated mouse monoclonal antibody to α–smooth muscle actin (U7033; Dako, Cambridgeshire, United Kingdom). Mucus-producing foveolar pit cells were identified using periodic acid–Schiff/alcian blue staining. Endocrine cells were identified using rabbit anti-chromogranin A+B (AB8205; Abcam, Cambridge, United Kingdom). Forestomach epithelium was identified using rabbit anti-mouse keratin 14 (PRB-155P; Covance Research Products, Princeton, NJ). Proliferating cells were detected using rabbit anti-phosphorylated histone H3 (06570; Upstate Biotechnology, Hampshire, United Kingdom)27Siegel P.M. Shu W. Cardiff R.D. Muller W.J. Massague J. Transforming growth factor β signalling impairs Neu-induced mammary tumorigenesis while promoting pulmonary metastasis.Proc Natl Acad Sci U S A. 2003; 100: 8430-8435Crossref PubMed Scopus (380) Google Scholar and rat anti-mouse Ki67 (M7269; Dako). Apoptotic cells were detected using the ApopTagPlus In Situ Apoptosis Detection Kit (Oncor, Gaithersburg, MD) and rabbit anti-mouse active caspase 3 (AF835; R&D Systems, Oxfordshire, United Kingdom). Where necessary for antigen retrieval, slides were microwaved for 10 minutes in sodium citrate buffer (pH 6). Images were obtained using a Nikon Eclipse E1000 microscope and a Nikon DXM1200F digital camera (Nikon, Surrey, United Kingdom). For scanning electron microscopy, E17.5 and E18.5 Fgfr2b−/−, Fgf10−/−, and wt stomachs were dissected in cold phosphate-buffered saline and fixed in 2.5% glutaraldehyde/2% paraformaldehyde in Sorenson's phosphate buffer. Samples were dehydrated, critical point dried, coated with platinum, and viewed in a JEOL 6700F microscope (JEOL, Hertfordshire, United Kingdom). In situ hybridization was performed on frontal and sagittal serial sections through mutant and wt littermate embryos at E15.5 essentially as described previously.28Kettunen P. Karavanova I. Thesleff I. Responsiveness of developing dental tissues to fibroblast growth factors expression of splicing alternatives of FGFR1, -2, -3, and of FGFR4; and stimulation of cell proliferation by FGF-2, -4, -8, and -9.Dev Genet. 1998; 22: 374-385Crossref PubMed Scopus (181) Google Scholar The sections were hybridized with 106 cpm of [35S]-labeled antisense riboprobes to mouse GATA4, mouse GATA6, mouse H+/K+-ATPase β subunit (from Dr D. Wilson, Washington University, St Louis, MO), mouse SHH (from Dr C. Tickle, University of Dundee, Dundee, Scotland), mouse IHH, mouse Ptc (from Dr S. Bellusci, University of Southern California, Los Angeles, CA), mouse Fgf10 (from Dr S. Werner, ETH, Zurich, Switzerland), and mouse Fgfr2b.19De Moerlooze L. Spencer-Dene B. Revest J. Hajihosseini M. Rosewell I. Dickson C. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signaling during mouse organogenesis.Development. 2000; 127: 483-492Crossref PubMed Google Scholar Nonradioactive in situ hybridization was also performed using a digoxigenin-labeled antisense riboprobe to the messenger RNA sequence corresponding to the tyrosine kinase domain of Fgfr2.19De Moerlooze L. Spencer-Dene B. Revest J. Hajihosseini M. Rosewell I. Dickson C. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signaling during mouse organogenesis.Development. 2000; 127: 483-492Crossref PubMed Google Scholar Briefly, rehydrated sections were mildly digested with proteinase K, fixed in 4% paraformaldehyde, and hybridized overnight at 60°C for 10 minutes. Posthybridization washes with 2× standard saline citrate and 0.2× standard saline citrate at 60°C were followed by incubation with an alkaline phosphatase–conjugated sheep anti-digoxigenin antibody (Roche, Hertfordshire, United Kingdom). Signal was developed for several hours with NBT/BCIP (Roche). A sense probe was used as a control. Whole-mount in situ hybridization was performed on isolated E16.5–18.5 stomachs as previously described.20Revest J.M. Spencer-Dene B. Kerr K. De Moerlooze L. Rosewell I. Dickson C. Fibroblast growth factor receptor 2-IIIb acts upstream of Shh and Fgf4 and is required for limb bud maintenance but not for the induction of Fgf8, Fgf10, Msx1, or Bmp4.Dev Biol. 2001; 231: 47-62Crossref PubMed Scopus (246) Google Scholar A digoxigenin-labeled antisense riboprobe was synthesized from mouse Nkx2.5 complementary DNA (from Dr M Schinke, Harvard University, Cambridge, MA). Stomachs were dissected from wt C57BL6 embryos at E12.5 and cultured for 48 hours in vitro on filters (Nucleopore Track-Etch Membrane, 13 mm, 0.8 μm; Whatman, Clifton, NJ) laid on the surface of 1 mL of Dulbecco's modified Eagle medium/F12 medium (Invitrogen, Paisley, Scotland) containing 50 U/mL penicillin/streptomycin (Invitrogen). Beads were washed in phosphate-buffered saline, incubated for 1 hour at 37°C in 10 μL of 100 ng/mL human recombinant Fgf10 or bovine serum albumin (BSA), and then washed again for 1 hour in culture medium. Controls (BSA bead) were placed in (n = 3) or next to (n = 3) the stomachs. Experimentals (Fgf10 bead) were placed inside (n = 4) or next to (n = 5) the stomachs. After 48-hour culture, stomachs were fixed in 4% paraformaldehyde for 30 minutes and stored in 70% ethanol before being processed into paraffin for histology and immunohistochemistry. Stomachs were dissected from Fgf10−/− (n = 2) and wt (n = 5) E14.5 embryos and placed briefly in cold Dulbecco's modified Eagle medium/F12 culture medium. Stomachs were then grafted under the flank skin of nude mice (NU/NU, CRL:NU-FOXX1 ). After 3 weeks, recipient mice were culled and the grafts were dissected out and fixed in 4% paraformaldehyde for 24 hours and stored in 70% ethanol before processing for histology and immunohistochemistry. Stomachs were dissected from wt, Fgfr2b−/−, and Fgf10−/− mice between E16.5 and E18.5. Mutant stomachs were small, although roughly in proportion to the reduced size of the mouse as a whole (Figure 1 A). In both cases, the forestomach appeared to show a slight loss of definition of the longitudinal folds but no other gross changes. The muscular wall of the stomach appeared normal. The glandular epithelial lining of the corpus region of both Fgfr2b−/− and Fgf10−/− mice appeared devoid of normal complex branched glands, being replaced by a broad region of columnar epithelium similar in appearance to the narrow band of transitional epithelium separating the forestomach and corpus in wt mice. The normal complex glandular structure of the antral region of the wt stomach was replaced in both Fgfr2b−/− and Fgf10−/− mice with simple large mucosal protrusions (Figure 1 B–D). Histologic analyses revealed differences between the stomachs of wt and mutant mice. At E16.5, wt mice had closely packed developing body glands with differentiating chief cells that occupied the lamina propria. In the antral region, branching tubular glands were lined by pale mucinous cells. By E18.5, there was clearly established differentiation between body and antrum with chief and parietal cells identifiable in the pits (Figure 2 A and 2D). In contrast, sectioning of stomachs at E17.5 (n = 4 of each mutant genotype and wt) and E18.5 (n = 10 of each mutant genotype and 4 wt) revealed in the mutant embryos a greatly simplified glandular epithelium (Figure 2 B, C, E, and F). The stomach surface was lined by mucosa containing separated simple gastric epithelial infolds forming simple angulated pits. In the antrum of mutant embryos, some of the pits showed some tubular bifurcation but no complex glands were observed (Figure 2 E and F). These observations indicate that the epithelium lining the gastric body in both mutant lines is histologically abnormal. A more detailed analysis using scanning electron microscopy and serial sectioning of both Fgfr2b−/− and Fgf10−/− stomachs showed that the transition zone of the forestomach-body junction, characterized by cuboidal epithelium with occasional ciliated cells, comprised approximately 40% of the surface glandular epithelia. The proximal half of the foreshortened glandular stomach showed mucus-producing cells identified by periodic acid–Schiff staining and eosinophilic cells in the pits/crypts comprising chief and endocrine cells (Figure 3 and data not shown). Immunohistochemistry confirmed there were chief cells (pepsinogen positive) and endocrine cells, but no ATPase-positive parietal cells were detected (Figure 3). The absence of parietal cells and the reduced abundance of chief cells suggest that differentiation of the gastric mucosa in these mutants is severely compromised. In mice, the gastric epithelium starts to stratify and differentiate from E12.5 onward, and we investigated whether proliferation or apoptosis rates were abnormal in mutant mice at this key stage.10Kim B.M. Buchner G. Miletich I. Sharpe P.T. Shivdasani R.A. The stomach mesenchymal transcription factor Barx1 specifies gastric epithelial identity through inhibition of transient Wnt signalling.Dev Cell. 2005; 8: 611-622Abstract Full Text Full Text PDF PubMed Scopus (166) Google Scholar We found that both mutant lines exhibited a clear decrease in the number of proliferating cells compared with wt littermates (Figure 4 A), which may be responsible for the overall reduction in distal stomach size seen more obviously in older embryos. At the same stage, there was no increase in the number of gastric epithelial cells undergoing apoptosis (Figure 4 B). In the normal mouse stomach, Shh is strongly expressed in the forestomach epithelium and weakly in the glandular stomach epithelium lining the corpus and antrum, whereas Ihh shows a reciprocal distribution with expression in corpus and antral glandular epithelium5Bitgood M.J. McMahon A.P. Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo.Dev Biol. 1995; 172: 126-138Crossref PubMed Scopus (1198) Google Scholar, 7Ramalho-Santos M. Melton D.A. McMahon A.P. Hedgehog signals regulate multiple aspects of gastrointestinal development.Development. 2000; 127: 2763-2772Crossref PubMed Google Scholar (Figure 5). In both Fgfr2b−/− and Fgf10−/− mice at E15.5, Shh is aberrantly expressed throughout the gastric epithelium, whereas Ihh is not detected. Ptc, the receptor for Shh and Ihh, is expressed normally in the developing gastric mesenchyme29Marigo V. Scott M.P. Johnson R.L. Goodrich L.V. Tabin C.J. Conservation in hedgehog signaling induction of a chicken patched homolog by Sonic hedgehog in the developing limb.Development. 1996; 122: 1225-1233Crossref PubMed Google Scholar in both mutants and wt mice (Figure 5). At E15.5, Fgfr2 expression normally matches that of Shh. Interestingly, in Fgf10−/− embryos at E15.5, there is an aberrant Fgfr2 expression pattern with transcripts distributed throughout the entire stomach epithelium matching the aberrant ectopic Shh expression (Figure 6). This indicates that Fgf10 and Fgfr2b regulate hedgehog signaling pathways previously shown to be important for radial and anterior-posterior patterning.Figure 6(A) Nonisotopic in situ hybridization showing that the expression of Fgfr2 is normally confined to the developing forestomach epithelium (f) in wt mice. (B) However, in Fgf10−/− mice, the receptor is expressed ectopically in the developing glandular epithelium of the corpus region (c) (bar = 250 μm).View Large Image Figure ViewerDownload Hi-res image Download (PPT) The transcription factors GATA4 and GATA6 are essential for early endoderm formation but are also expressed at later stages in the developing stomach.30Jacobsen C.M. Narita N. Bielinska M. Syder A.J. Gordon J.I. Wilson D.B. Genetic mosaic analysis reveals that GATA-4 is required for proper differentiation of mouse gastric epithelium.Dev Biol. 2002; 241: 34-46Crossref PubMed Scopus (75) Google Scholar Wt mice at E15.5 express GATA4 and GATA6 in the nascent epithelium of the corpus and antrum but not the forestomach (Figure 7). However, in Fgfr2b−/− and Fgf10−/− embryos, neither GATA4 nor GATA6 is expressed in the stomach epithelium at this stage (Figure 7). Similarly, H+/K+-ATPase normally present in parietal cells was expressed in the wt corpus but not that of either mutant (Figure 7). This supports histology and immunohistochemistry data that normal differentiation of the glandular stomach fails in both Fgfr2b−/− and Fgf10−/− mice. The muscular wall of the stomach appeared proportionate and all layers were present, as confirmed by α–smooth muscle actin immunohistochemistry (Figure 3 D and H). Also, the nerve plexuses were present and no abnormality was observed by conventional microscopy (data not shown). The homeobox gene Nkx-2.5 is expressed in the mesenchyme of the pyloric sphincter domain in vertebrate embryos9Aubin J. Dery U. Lemieux M. Chailler P. Jeannotte L. Stomach regional specification requires Hoxa5-driven mesenchymal-epithelial signaling.Development. 2002; 129: 4075-4087Crossref PubMed Google Scholar, 31Lints T.J. Parsons L.M. Hartley L. Lyons I. Harvey R.P. Nkx-2.5a novel murine homeobox gene expressed in early heart progenitor cells and their myogenic descendants.Development. 1993; 119: 419-431Crossref PubMed Google Scholar and was expressed normally in both mutants at E17.5–E18.5 (Figure 8), indicating that development of the pyloric sphincter is normal in both mutants. Because both mutant mice die at birth, any phenotypic changes normally apparent in adult mice could not be readily determined. However, using an established isografting technique,32Rubin D.C. Swietlicki E. Gordon J.I. Use of isografts to study proliferation and differentiation programs of mouse stomach epithelia.Am J Physiol. 1994; 267: 27-39PubMed Google Scholar we were able to determine whether the abnormal phenotype seen in the glandular stomachs of Fgf10−/− mutants was not due to developmental delay. Intact stomachs were dissected from mutant and wt E14.5 embryos and grown as isografts, implanted subcutaneously in the flanks of nude mice, and harvested after 3 weeks. Isografts were then fixed and processed into paraffin for histologic and immunohistochemical staining for the presence of parietal, endocrine, and foveolar mucus-producing cells. From surviving grafts, the Fgf10−/− stomachs remained abnormal in appearance, especially the glandular epithelium, and ATPase-positive parietal cells were still not present, although chromogranin-positive endocrine cells were evident (Figure 9). In the wt grafts, both parietal and endocrine cells were present. In the wt grafts, neutral mucin-positive foveolar cells were clearly seen; however, in the Fgf10−/− grafts, these were seen only in patches with large areas of unstained epithelial cells, suggesting that formation of this lineage is only partially successful (data not shown). To corroborate these findings in vitro, wt embryonic stomachs were cultured with beads soaked in recombinant Fgf10 or BSA embedded within the mesenchyme close to the forestomach epithelium or positioned outside the forestomach for 48 hours and subsequently examined histologically. The forestomach epithelium grew around the internal beads expressing rFgf10. Murine keratin 14 immunostaining was used as a marker of mature stratified squamous epithelia including forestomach epithelium, which is not normally expressed until late fetal stages of development. The stomach cultures containing an internal Fgf10 bead expressed keratin 14 throughout the enveloping epithelium and in the adjacent epithelia, indicative of an acceleration of the normal differentiative program (Figure 10). This was not seen in control BSA beads or when the bead was placed outside the stomach. Proliferation was also assessed by immunohistochemically staining sections with anti-Ki67. The Fgf10 bead–containing cultures appeared to have proportionally fewer proliferating cells than the controls (Figure 10). We report the gastric developmental defects exhibited by mice null for Fgf10 or its receptor, Fgfr2b, during fetal development. Mutant stomachs from both mutants were reduced in size when compared with those from normal littermates. This finding is consistent with an overall proportionate reduction in size of mutants versus controls and may also be due to a decrease in proliferation observed here at E13.5.19De Moerlooze L. Spencer-Dene B. Revest J. Hajihosseini M. Rosewell I. Dickson C. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signaling during mouse organogenesis.Development. 2000; 127: 483-492Crossref PubMed Google Scholar, 23Min H. Danilenko D.M. Scully S.A. Bolon B. Ring B.D. Tarpley J.E. DeRose M. Simonet W.S. Fgf-10 is required for both limb and lung development and exhibits striking functional similarity to Drosophila branchless.Genes Dev. 1998; 12: 3156-3161Crossref PubMed Scopus (746) Google Scholar, 24Sekine K. Ohuchi H. Fujiwara M. Yamasaki M. Yoshizawa T. Sato T. Yagishita N. Matsui D. Koga Y. Itoh N. Kato S. Fgf10 is essential for limb and lung formation.Nat Genet. 1999; 21: 138-141Crossref PubMed Scopus (1011) Google Scholar Histologic analysis showed that the glandular epithelial architecture was severely malformed in the corpus and antral regions. Both radial and anterior-posterior patterning appear to be affected, leading to loss of the normal mucosal architecture. Alterations in the normal expression domains of several signaling and transcription factors previously shown to be essential for normal gastric development, including members of the hedgehog and GATA families, were consistently altered in both Fgfr2b−/− and Fgf10−/− mice. The normal glandular complexity and branching seen in wt mice was replaced in both mutants by simple unbranched pits instead of distinct tubular invaginations. At E13.5, a clear reduction in proliferating gastric epithelial cells was observed in both mutants. This proliferative deficiency occurs at a key stage in gastric organogenesis when the glandular epithelium is just starting to develop.10Kim B.M. Buchner G. Miletich I. Sharpe P.T. Shivdasani R.A. The stomach mesenchymal transcription factor Barx1 specifies gastric epithelial identity through inhibition of transient Wnt signalling.Dev Cell. 2005; 8: 611-622Abstract Full Text Full Text PDF PubMed Scopus (166) Google Scholar One marker of parietal cell differentiation is H+/K+-ATPase, hence the absence of this protein or the messenger RNA encoding it, in the stomachs of Fgfr2b−/− and Fgf10−/− embryos at E15.5 and E18.5, which suggests that there is not simply a delay in parietal cell differentiation. This was further confirmed by transplanting E14.5 Fgf10−/− stomachs subcutaneously into nude mice as isografts and harvesting 3 weeks later. The Fgf10−/− isografts did not go on to develop a normally patterned glandular epithelium containing parietal cells, in contrast to wt isografts that did. Parietal cells have been proposed to control the fate of precursor cells, including the presumptive gastric stem cell. The preparietal and parietal cell lineages have previously been selectively ablated in transgenic mice by expressing cytotoxins driven by the H+/K+-ATPase promoter.33Canfield V. West A.B. Goldenring J.R. Levenson R. Genetic ablation of parietal cells in transgenic mice a new model for analyzing cell lineage relationships in the gastric mucosa.Proc Natl Acad Sci U S A. 1996; 93: 2431-2435Crossref PubMed Scopus (102) Google Scholar, 34Li Q. Karam S.M. Gordon J.I. Diphtheria toxin-mediated ablation of parietal cells in the stomach of transgenic mice.J Biol Chem. 1996; 271: 3671-3676Crossref PubMed Scopus (188) Google Scholar Ablation of parietal cells led to the disintegration of the gastric glands and the loss of chief and mucus-producing cells.33Canfield V. West A.B. Goldenring J.R. Levenson R. Genetic ablation of parietal cells in transgenic mice a new model for analyzing cell lineage relationships in the gastric mucosa.Proc Natl Acad Sci U S A. 1996; 93: 2431-2435Crossref PubMed Scopus (102) Google Scholar However, we observe these other lineages in the absence of the parietal cell lineage in the Fgfr2b−/− and Fgf10−/− mutants, despite abnormalities in architectural patterning. These observations suggest that these other lineages are not entirely dependent on parietal cell differentiation. A recent study using GATA4 chimeras has shed light on the roles it plays during gastric differentiation.30Jacobsen C.M. Narita N. Bielinska M. Syder A.J. Gordon J.I. Wilson D.B. Genetic mosaic analysis reveals that GATA-4 is required for proper differentiation of mouse gastric epithelium.Dev Biol. 2002; 241: 34-46Crossref PubMed Scopus (75) Google Scholar Glandular stomach epithelium failed to form, being replaced instead with squamous epithelium. Several differentiative markers were not detectable, including GATA6, pepsinogen, and H+/K+-ATPase, whereas the Shh expression domain was extended. This observation would be entirely consistent with the present analysis of the Fgfr2b−/− and Fgf10−/− mutants, placing Fgf10/Fgfr2b signaling upstream of GATA4 and GATA6 (Figure 11). Mice deficient for isoforms of activin receptor II have been shown to have defective foregut patterning, including an apparent anteriorization of the posterior stomach. Mutant stomach epithelium was entirely comprised of squamous epithelium, whereas Shh expression appeared to extend into the posterior epithelium.6Kim S.K. Hebrok M. Li E. Oh S.P. Schrewe H. Harmon E.B. Lee J.S. Melton D.A. Activin receptor patterning of foregut organogenesis.Genes Dev. 2000; 14: 1866-1871PubMed Google Scholar, 35Li Q. Karam S.M. Coerver K.A. Matzuk M.M. Gordon J.I. Stimulation of activin receptor II signaling pathways inhibits differentiation of multiple gastric epithelial lineages.Mol Endocrinol. 1998; 12: 181-192Crossref PubMed Google Scholar These results also suggested that down-regulation of Shh signaling may be a key requirement for normal gut patterning. Shh−/− mice exhibited early defects in foregut development36Litingtung Y. Lei L. Westphal H. Chiang C. Sonic hedgehog is essential to foregut development.Nat Genet. 1998; 20: 58-61Crossref PubMed Scopus (588) Google Scholar and at E18.5 showed overgrowth of the gastric epithelium, reduced gastric smooth muscle, intestinal metaplasia, and other intestinal defects.7Ramalho-Santos M. Melton D.A. McMahon A.P. Hedgehog signals regulate multiple aspects of gastrointestinal development.Development. 2000; 127: 2763-2772Crossref PubMed Google Scholar Moreover, mice null for Ihh also had severe gastrointestinal abnormalities, including gut malrotation and aganglionic dilated colon.7Ramalho-Santos M. Melton D.A. McMahon A.P. Hedgehog signals regulate multiple aspects of gastrointestinal development.Development. 2000; 127: 2763-2772Crossref PubMed Google Scholar These studies indicate that Shh and Ihh act early to pattern the gut endoderm and may have multiple functions both upstream and downstream of Fgf signaling. As the forestomach begins to form, the simple anterior epithelium starts to thicken, ultimately forming stratified squamous epithelium that expresses differentiation markers including keratin 14.37Vassar R. Rosenberg M. Ross S. Tyner A. Fuchs E. Tissue-specific and differentiation-specific expression of a human K14 keratin gene in transgenic mice.Proc Natl Acad Sci U S A. 1989; 86: 1563-1567Crossref PubMed Scopus (322) Google Scholar To further investigate the roles played by Fgf10 in stomach development, embryonic stomachs were cultured containing beads soaked in Fgf10 or BSA within the forestomach mesenchyme adjacent to the forestomach epithelium and cultured for 48 hours. After staining for keratin 14 and the proliferation marker Ki67, the ectopic Fgf10 is able to locally accelerate the normal differentiative pathways taken by nascent forestomach. However, because the forestomach forms normally in both mutant mice, the Fgf10/Fgfr2b signaling pathway is not required for forestomach development. The muscle layers surrounding the stomachs of both mutants stained positively for α–smooth muscle actin and were histologically normal. These findings are contrary to those reported previously using a different line of Fgf10−/− mice, on a C57BL/6 × CBA strain background, in which the gastric smooth muscle layers were reportedly absent at birth.24Sekine K. Ohuchi H. Fujiwara M. Yamasaki M. Yoshizawa T. Sato T. Yagishita N. Matsui D. Koga Y. Itoh N. Kato S. Fgf10 is essential for limb and lung formation.Nat Genet. 1999; 21: 138-141Crossref PubMed Scopus (1011) Google Scholar, 26Ohuchi H. Hori Y. Yamasaki M. Harada H. Sekine K. Kato S. Itoh N. FGF10 acts as a major ligand for FGF receptor 2 IIIb in mouse multi-organ development.Biochem Biophys Res Commun. 2000; 277: 643-649Crossref PubMed Scopus (528) Google Scholar The changes in phenotype may be due to strain differences because the Fgf10−/− line in this study23Min H. Danilenko D.M. Scully S.A. Bolon B. Ring B.D. Tarpley J.E. DeRose M. Simonet W.S. Fgf-10 is required for both limb and lung development and exhibits striking functional similarity to Drosophila branchless.Genes Dev. 1998; 12: 3156-3161Crossref PubMed Scopus (746) Google Scholar has been maintained as C57BL/6 × C57BL/6. In addition, the pyloric sphincter is essentially normal in both Fgfr2b−/− and Fgf10−/− embryos. In this study, we have investigated the roles played by Fgfr2b and Fgf10 in the development of the stomach. Using Fgfr2b−/− and Fgf10−/− embryos, we showed that the absence of either the receptor or the ligand component of this signaling pathway is associated with impaired development of foregut endoderm into glandular but not squamous epithelium. The inability to correctly differentiate into gastric glandular epithelial lineages correlates with the ectopic expression of Shh and the loss of Ihh, GATA4, and GATA6, suggesting that Fgf10/Fgfr2b signaling is involved in the regulation of these genes. We suggest that in the glandular stomach, mesenchymal Fgf10 negatively regulates the expression of both its receptor Fgfr2b and Shh (Figure 11) while positively regulating Ihh, GATA4, and GATA6 in the epithelium. The authors thank the following individuals and departments at Cancer Research UK: the Experimental Pathology Laboratory, G. Elia and the Histopathology Laboratory, Biological Resources, and Prof Julian Lewis for critically reviewing this paper.