QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 12 September 2007
doi: 10.1242/dev.009308

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kim, B.-M. Articles by Shivdasani, R. A. Search for Related Content PubMed PubMed Citation Articles by Kim, B.-M. Articles by Shivdasani, R. A.

Independent functions and mechanisms for homeobox gene Barx1 in patterning mouse stomach and spleen

¹ Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA 02115, USA.
² Department of Medicine, Harvard Medical School, Boston, MA 02115, USA.
³ Department of Craniofacial Development, Dental Institute, Kings College, London SE1 9RT, UK.
⁴ Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA 02138, USA.
⁵ Department of Medicine, Brigham and Women's Hospital, Boston, MA 02115, USA.

View larger version (108K): [in this window] [in a new window]   Fig. 1. Gastrointestinal homeosis in the absence of Barx1. (A) Comparison of the differences in size and morphology of normal (left) and Barx1-/- (right) neonatal stomach, as observed in dozens of embryos and newborn mice. Es, esophagus; St, stomach; Du, duodenum. (B,D) Mucosal differences between normal (D) and Barx1-/- neonatal (B) stomach, revealing a crowded villiform epithelium in the latter. (C) Low-magnification micrograph of PAS staining of Barx1-/- neonatal stomach, shown to reveal the boundary within the stomach (dotted line) of strongly PAS+ gastric mucosa proximally and largely PAS- intestinal epithelium distally. (E-G) Normal epithelial histology (H&E stain) from the neonatal mouse forestomach (E), corpus (F) and antrum (G). (H,I) Gross architectural disorganization typifies the abnormal epithelium (H&E stain) of Barx1-/- neonatal mid-stomach. (J-O) Histochemical (PAS and Alcian Blue) and molecular [Pdx1, Cdx2, gastric intrinsic factor (If) and trefoil factor 3 (Tff3)] stains highlight the sharp boundary (dotted line) between epithelia of the gastric (top in J-M, left in N,O) and intestinal (bottom in J-M, right in N,O) types in Barx1-/- stomach. J-M represent consecutive tissue sections, and N-O are consecutive to each other but not to the others.

View larger version (72K): [in this window] [in a new window]   Fig. 2. Severe mucosal abnormalities in the proximal foregut of Barx1-/- mice. (A,B) Abnormally proximal expression of Pdx1 (near the squamous lining) and mixing of H+/K+-ATPase+ and Pdx1+ cells in the glandular mucosa of Barx1-/- neonatal stomach. The two panels show staining on adjacent sections. (C) Radial asymmetry of the most-proximal mucosa, with one surface lined by a flat squamoid epithelium (seen at the top of the image) and the other (bottom) by a cuboidal epithelium containing many ciliated cells (arrows). (D,E) PAS (D) and Muc5ac staining (E) reveal the presence in both surfaces of cuboidal cells with features of glandular pit cells. (F) Ultastructural confirmation of numerous apical cilia in cells lining neonatal Barx1-/- proximal foregut. (G,H) Immunohistochemical evidence for expression of Sox2, a squamous foregut marker, shown at low (G) and high (H) magnification. (I,J) Smooth muscle actin staining indicates smooth muscle differentiation in Barx1-/- stomach (J), although both signal strength and tissue continuity are reduced in comparison to littermate controls (I). (K) Graphic representation of gastrointestinal homeotic and regional anomalies that occur in the absence of Barx1.

View larger version (121K): [in this window] [in a new window]   Fig. 3. In vivo confirmation that Barx1 enables attenuation of Wnt signaling in prospective stomach epithelium. (A,B) lacZ staining in Barx1-/-;TOPGALTg (B) and control Barx1+/-;TOPGALTg (A) mouse fetal stomach at E16.5, when endogenous Wnt signaling is virtually abolished and there is no esophageal activity (dashed lines) in control embryos but strong signals remain in Barx1 mutants. (C,D) Histochemical confirmation of persistent ß-gal activity in rostral Barx1-/-;TOPGALTg stomach endoderm (D, arrowheads) in relation to negligible signal in corresponding Barx1+/-;TOPGALTg tissue (C). (E,F) Further confirmation by lacZ in situ hybridization of mRNA persistence in rostral E16.5 Barx1-/-;TOPGAL stomach endoderm (F), which contrasts with the absence of expression in control E16.5 Barx1+/+;TOPGAL tissue (E). (G-I) Persistent Wnt signaling is further revealed by nuclear localization of ß-catenin in the emerging squamous epithelium of Barx1-/- (G, boxed area shown at higher magnification in H) but not control (I) foregut. (J,K) Similarly, ß-gal activity persists in E18.5 Barx1-/-;Axin2lacZ fetal stomach (K) compared with Barx1+/-;Axin2lacZ controls (J). Dashed and solid lines in A and J demarcate normal esophagus (Es) and duodenum (Du), respectively. St, stomach; fore, forestomach; hind, hind-stomach (the blurred boundaries between fore- and hindstomach are represented by dotted lines). Scale bars: 75 µm in C; 60 µm in D; 75 µm in E,F.

View larger version (121K): [in this window] [in a new window]   Fig. 4. Independent confirmation of the adverse effects of unregulated Wnt signaling in developing stomach endoderm. Findings in E18.5 ShhCre/+;Catnb+/lox(ex3) mouse embryos. (A,B) Constitutive ß-catenin activation is confirmed by its nuclear staining (A), compared with membrane staining in areas that escaped Cre-mediated recombination (B). (C-E) The stomach is reduced in size and lined by a crowded, villiform epithelium (C,E; H&E stain) that shares features seen in Barx1 mutant stomach, including thick folds and mucosal invasion of the mesenchymal layer (E), which contrasts with relatively normal squamous differentiation (D) in many areas. (F) Radial asymmetry of E18.5 ShhCre/+;Catnb+/lox(ex3) esophagus, with squamous (blue arrowheads) and cuboidal (green arrowheads) epithelia on opposite surfaces. (G) Magnification of the area boxed in F. PAS staining reveals many cells abnormally producing mucin in the cuboidal epithelium of the esophagus. (H-J) Although intestinal villi are absent, there is considerable and ectopic expression of the intestinal marker Cdx2 (H,I), which is normally excluded from the squamous (J) and glandular (data not shown) stomach. H is a magnification of the area boxed in C.

View larger version (71K): [in this window] [in a new window]   Fig. 5. Abnormal spleen development in Barx1-/- mice. (A) Relationship of the spleen (Sp) in normal neonatal mice to adjacent stomach (St), duodenum (Du) and a fused pancreas (Pa). (B) Invariant size and location of the Barx1-/- spleen as a small tissue (marked with a dotted line) attached to the dorsal pancreas, which is separate from the ventral pancreatic bud. (C,D) Histologic elucidation of intimate association between splenic and pancreatic parenchyma by H&E stain (C) and insulin immunostaining of prospective pancreatic islets (D, arrowheads) on consecutive tissue sections. (E) Flow cytometric (left) and immunohistochemical (right) analysis of control (+/?) and Barx1-/- spleen (Sp), indicating normal hematopoiesis and lymphoid colonization (+/? refers to +/+ or +/-). Ter119, B220 and Gr1 are specific markers of red cells, B lymphocytes and granulocytes, respectively. Control and Barx1-/- spleen also showed identical flow cytometry profiles for T-cell and monocyte surface markers (data not shown).

View larger version (79K): [in this window] [in a new window]   Fig. 6. Correlates of spleen development and Barx1 expression. (A,B) The spleen (*) appears to be specified correctly in E9.5 Barx1-/- mouse embryos (B), as in control littermates (A). St, stomach; Pa, pancreas. (C,D) Differences in size and organ relationships between control (C) and Barx1-/- (D) embryos are evident by E12.5 and increase thereafter. The position of the mutant spleen is again highlighted with a dotted line in D. (E-G) Radioactive mRNA in situ hybridization at E10.5, E11.5 and E13.5, showing high Barx1 expression in stomach mesenchyme and mesothelium with exclusion from spleen mesenchyme throughout. Bright-field images are shown for the younger embryos and a dark-field image at E13.5. (H,I) Colorimetric mRNA in situ hybridization at E12.5, showing absence of Barx1 expression in intestinal (Int; arrowheads in H) mesentery and presence in the mesothelial lining of the caudal liver (Lv; arrowheads in I), adjacent to the stomach (St). (J,K) Immunohistochemical confirmation with Barx1 (J) and preimmune (K) antiserum of high expression in stomach mesenchyme (top) and mesothelium (arrowheads), with exclusion from the spleen (Sp) mesenchyme in E12.5 embryos.

View larger version (53K): [in this window] [in a new window]   Fig. 7. Barx1 effects on the spleen are not mediated through either Wnt signaling or a group of homeodomain transcription factors implicated in control of spleen development. (A-C) Lack of ß-gal activity in the pancreas or spleen of Barx1-/-;TOPGAL E18.5 mouse embryos, shown in situ (A) to contrast with residual activity in the stomach (St) and native signal in duodenum (Du). The dotted line in A marks the fused spleen-pancreas (Pa), where absence of ß-gal activity is further revealed in B and confirmed by microscopic analysis in C. (D) In contrast to Barx1, homeobox genes previously implicated in spleen development are mostly expressed in prospective spleen mesenchyme (Sp) and excluded from the mesothelium. The left column shows low-magnification images from each in situ hybridization, and the boxed area of each image is shown at higher magnification in the middle column, where the splenic capsule is demarcated by dotted lines. Images in the right-hand column reveal that expression of each of these homeobox genes is maintained in Barx1-/- spleen, which is recognized in part by juxtaposition to Pdx1+ pancreatic tissue (red box).

View larger version (60K): [in this window] [in a new window]   Fig. 8. Reduced expression of the multifunctional zinc-finger protein Wt1 in spleen mesothelium of Barx1-/- mice. (A) Among the genes previously implicated in spleen development, only expression of Wt1 is readily recognized in the mesothelial lining (arrowheads) of wild-type E13.5 stomach (St) and spleen (Sp). (B) Higher magnification images of Wt1 mRNA expression indicate enrichment in wild-type spleen capsule (the boxed area in the left panel is shown at even higher magnification in the middle panel, where arrowheads point to the outer surface) and substantially reduced levels in Barx1-/- spleen (right-hand panel). (C) Independent confirmation of reduced Wt1 transcript levels, assessed by RT-PCR on RNA extracted from isolated E13.5 spleen and surrounding mesothelium. Products of conventional RT-PCR are shown in the gel on the left, and quantitation by real-time PCR using different primers in the bar chart on the right. For quantitative RT-PCR, values were first normalized for Gapdh expression and then to a value of 1.0 for control samples (+/? refers to +/+ or +/-). (D) Confirmation of reduced Wt1 expression by immunohistochemistry. Control (+/+) E13.5 stomach and spleen are shown in the left-hand panels and mutant tissues in the right-hand panels; the boxed areas in the upper panels are shown at higher magnification in the lower panels. Arrowheads point to the mesothelium, where Wt1 levels are reduced in Barx1-/- embryos. St, stomach; Sp, spleen; Pa, pancreas; Kd, kidney; Lv, liver. (E) Wt1 expression in Barx1-/- E13.5 kidney is intact.