Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Sep 15;144(18):3336-3348.
doi: 10.1242/dev.152843. Epub 2017 Aug 14.

Transgelin-expressing myofibroblasts orchestrate ventral midline closure through TGFβ signalling

Affiliations

Transgelin-expressing myofibroblasts orchestrate ventral midline closure through TGFβ signalling

Bashar Aldeiri et al. Development. .

Abstract

Ventral body wall (VBW) defects are among the most common congenital malformations, yet their embryonic origin and underlying molecular mechanisms remain poorly characterised. Transforming growth factor beta (TGFβ) signalling is essential for VBW closure, but the responding cells are not known. Here, we identify in mouse a population of migratory myofibroblasts at the leading edge of the closing VBW that express the actin-binding protein transgelin (TAGLN) and TGFβ receptor (TGFβR). These cells respond to a temporally regulated TGFβ2 gradient originating from the epithelium of the primary body wall. Targeted elimination of TGFβR2 in TAGLN+ cells impairs midline closure and prevents the correct subsequent patterning of the musculature and skeletal components. Remarkably, deletion of Tgfbr2 in myogenic or chondrogenic progenitor cells does not manifest in midline defects. Our results indicate a pivotal significance of VBW myofibroblasts in orchestrating ventral midline closure by mediating the response to the TGFβ gradient. Altogether, our data enable us to distinguish highly regulated epithelial-mesenchymal signalling and successive cellular migration events in VBW closure that explain early morphological changes underlying the development of congenital VBW defects.

Keywords: Exomphalos; Midline defect; Mouse; Myofibroblast; TGFβ; Transgelin.

PubMed Disclaimer

Conflict of interest statement

Competing interestsThe authors declare no competing or financial interests.

Figures

Fig. 1.
Fig. 1.
Tagln-Cre expression in the ventral midline and mitotic activity of TAGLN+ cells. (A) Transverse section at a thoracic level in an E12.5 wild-type (WT) mouse embryo stained for TAGLN, showing expression in the primary VBW (area between arrows). (B) Transverse section at an abdominal level in an E13.5 WT embryo stained for TAGLN showing expression in the primary abdominal wall (area between arrows) that is encircling the umbilical hernia. (C) Whole-mount β-galactosidase staining in Tagln-Cre:Rosa26-NGZ at three embryonic stages. The expression of TAGLN is evident in the somite at E11.5 and localises to the midline area when VBW closure is complete. Dotted lines delineate forelimb and hindlimb. (D) (Left) Numbered lines indicate the level of transverse sections shown in (1) A,E,H, (2) B,F and (3) G. (Right) Schematic of midline (red) and para-midline (grey) areas presented in the KI67 analysis in H,I. (E,F) Expression of Tagln-Cre:Rosa26-tdTom in the thoracic (E) and abdominal (F) ventral midline over a 4 day time window during the closure process and at postnatal day (P) 20. TAGLN expression becomes restricted to the midline area with advanced gestation and this expression is maintained postnatally. Inset in E15.5 shows high magnification of the midline. Arrowheads indicate internal mammary/superior epigastric vessels and asterisk indicates the xiphisternum. (G) TUNEL assay for apoptosis in the ventral midline at E15.5. There is no obvious pattern of apoptosis in TAGLN+-derived cells in the midline. Boxes show examples of individual TUNEL+ cells in the midline and para-midline areas. (H) KI67 staining of the ventral midline at E14.5. Primary body wall remnant at this stage shows limited mitotic activity, which is evident in the KI67 channel. (I) Comparison of KI67 expression between midline (ML) primary VBW cells (tdTom+) and para-midline (PML) secondary body wall cells (tdTom) in the thoracic and abdominal regions. Comparison was made on 200 cells from three different sections at each level; data presented as mean±s.e.m. **P<0.001, two-tailed t-test. H, heart; L, lungs; LV, liver; UH, umbilical hernia. Scale bars: 500 µm in A,B,F; 1000 µm in C and P20 in E,F; 200 µm in E,G,H.
Fig. 2.
Fig. 2.
Directional migration of TAGLN+ cells towards the ventral midline. Still images from 9 h time-lapse (time shown bottom right) of ex vivo body wall explant culture. The VBW is located at the righthand side of each panel and dorsally located tdTom+ cells (white arrows) are in the left top corner. (A) Time zero, showing location of tdTom+ cells. (B) Time zero, with added tracks and migration paths. Each tdTom+ cell centre is labelled with a grey square and the path and time course of the journey are marked with a colour-coded line. (C) At 3 h VBW cells show directional migration towards the ventral midline, whereas dorsal cells show little change in position. (D,E) At 6 and 9 h, respectively, midline directional migration continues in VBW cells. (F) Trajectories and journey length in the analysed cells. Grey arrows indicate the direction and length of each migration path. VBW cells show consistent directional migration towards the midline, whereas dorsal cells show little change in position.
Fig. 3.
Fig. 3.
TAGLN protein expression in the ventral midline during the closure process. Transverse sections are shown of thoracic (A-C) and abdominal (D-G) VBW from Tagln-Cre:Rosa26-tdTom embryos stained with TAGLN antibody. (A) E12.5, showing the left primary body wall. Complete overlap between Tagln-Cre (tdTom) and TAGLN signals is seen. (B) At E14.5 there is still near complete overlap between the tdTom and TAGLN signals. A magnified view of the closing midline (boxed area in B) is shown to the right. (C) At E16.5 the thoracic midline has completely closed. The tdTom signal is still seen as a narrow line in the midline, but TAGLN signal cannot be identified. The magnified view of the midline (boxed area in C) shows the fine line of tdTom+ cells that have now become negative for TAGLN. (D) The abdominal ventral midline at E12.5, Tagln-Cre and TAGLN signals show complete overlap. (E) At E13.5 the Tagln-Cre-derived cells (tdTom+) of the primary body wall still express TAGLN. (F) At E14.5 the TAGLN signal area in the primary ventral midline is restricted compared with the tdTom signal area of the Tagln-Cre cells. (G) The ventral midline, labelled by tdTom, at E15.5 (same level as in F) has largely downregulated TAGLN. The magnified view of the midline (boxed area in G) shows tdTom+ cells of the midline that have now become negative for TAGLN. H, heart; LV, liver; IN, intestine. Scale bars: 200 µm, except 25 µm in higher magnification images in C,G.
Fig. 4.
Fig. 4.
Characterisation of ventral midline cells in Tagln-Cre:Rosa26-tdTom during VBW closure. Expression of smooth muscle contractile proteins (A-D,H) in the primary wall is more evident at early stages of midline closure. (A) αSMA and vimentin are expressed in the thoracic primary body wall at E12.5 and correlate with tdTom signal. Insets are magnified views (at cellular level) of the boxed areas. (B) At E14.5 primary body wall cells labelled by tdTom are still strongly positive for vimentin and express the smooth muscle intermediate filament protein desmin. (C) E15.5 midline cells are immunopositive for the fibroblast marker ER-TR7. Inset shows the ventral midline area (boxed) at higher magnification. (D) When the thoracic midline is fully closed at E16.5 the residual primary midline cells still labelled by tdTom have now downregulated αSMA. As shown in the higher magnification inset, only a small number of cells (arrow) of the midline show expression of αSMA. (E) Numbered lines indicate the level of transverse sections shown in (1) A-D,F,G and (2) H-J. (F,G) Tendon markers are absent in the primary body wall. (F) Tendon marker tenascin-C is expressed at E13.5 around the rib primordium and just lateral to primary elements (bottom box), and sporadic low-level expression is seen in the primary body wall (top box). (G) At E14.5 no tenascin-C expression is seen in the primary body wall in the midline. Sternal primordium cells express tenascin-C and are seen encircling the primary body wall cells. (H-J) Abdominal primary body wall is made of myofibroblasts. (H) In the abdominal midline at E14.5, primary body wall cells express vimentin and desmin. (I) At E15.5 the cells of the abdominal midline are immunopositive for the fibroblast marker ER-TR7. (J) At E16.5 the ventral midline has fully closed and resident tdTom+ cells are seen in the midline. Tenascin-C expression can be detected in the edges of the falciform ligament, but not at the midline. Scale bars: 100 µm.
Fig. 5.
Fig. 5.
TGFβ2 and TGFβR2 in the VBW. (A) Transverse section in the abdominal VBW at E14.5 showing expression of TGFβR2 focused in the primary body wall area (labelled by tdTom) in the ventral midline. (A′) Confocal image of the boxed area in A, showing high-level TGFβR2 expression in tdTom+ cells beneath the epithelium. (B) Transverse section in the mid thoracic area at E12.5 Tagln-Cre:Rosa26-tdTom mouse embryo stained for TGFβ2 and E-cadherin to label epithelium. TGFβ2 protein is abundant in the midline area of the primary body wall (tdTom channel is removed to expose the TGFβ2 signal). (B′) Confocal image of the primary body wall area (box P) showing strong TGFβ2 expression in the epithelium (arrows) and weaker signalling in the subdermal layer (arrowheads). (B″) Confocal image of the secondary body wall area (box S) showing weak TGFβ2 signal in the subdermal layer (arrows). (C) Midline (ML) and para-midline (PML) ventral wall dissection in an E12.5 WT mouse embryo. (Ca) The embryo was decapitated and the tail excised. (Cb) The dorsal body wall was opened para-sagittal and the thoracic and abdominal organs were exposed. (Cc) The embryo was eviscerated, taking care to preserve the thin primary body wall. (Cd) The thin primary (midline) body wall was carefully dissected from the secondary (para-midline body) wall and sufficient margins were removed from both segments to avoid transitional areas. (D) RT-qPCR comparing Tgfb2 expression in the midline and para-midline of WT mouse embryos between E11.5 and E15.5. There is an anatomical and temporal Tgfb2 gradient in the midline during the closure period. Error bars are s.e.m.; each time point presented is from at least three biological replicates each containing tissue from at least five embryos. (E) Schematic of E14.5 embryo. The VBW delineated by the dashed line was dissected from Tagln-Cre:Rosa26-tdTom embryos and FACS sorted for tdTom signal. (E′) The FACS-sorted cohort. tdTom+ cells only accounted for an average of 15% of the total cell population of the VBW (as shown in E). (F) RT-qPCR on the FACS-sorted cells showed higher expression of Tgfbr2 in tdTom+ ventral midline cells. Error bars indicate s.e.m.; data presented are from three biological replicates each containing cells from tissue derived from at least seven embryos. **P<0.001; *P<0.05; NS, non-significant; two tailed t-test. H, heart; LV, liver; P, primary body wall; S, secondary body wall. Each experiment shown in D, E and F was repeated at least three times. Scale bars: 200 µm, except 10 µm in A′,B′,B″.
Fig. 6.
Fig. 6.
Tagln-Cre:Tgfbr2flx/flx embryos develop VBW closure defects. (A,B) Morphological comparison between Tagln-Cre:Tgfbr2flx/flx and Tagln-Cre:Tgfbr2flx/wt mouse embryos. (A) E13.5 Tagln-Cre:Tgfbr2flx/flx embryos show a translucent ventral midline, a more lateral limit to the secondary body wall (arrow) and absence of midline raphe (arrowhead) when compared with Tagln-Cre:Tgfbr2flx/wt. (B) The ventral midline closure defect in Tagln-Cre:Tgfbr2flx/flx. A thin membrane covers the VBW cavities, as compared with the nearly closed thoracic midline in the WT (arrow) and the embryos show a large exomphalos compared with the physiological umbilical hernia in the WT (arrowhead). (C) Transverse section in mid-thorax at E14.5 in WT (left) and Tagln-Cre:Tgfbr2flx/flx (right), with Alcian Blue staining to delineate ribs and counterstaining with Nuclear Fast Red. The VBW is composed of a thin sac in the mutant, whereas the two lateral sternebrae are nearly meeting in the midline in the WT. (D) Transverse section at level of the umbilical hernia at E14.5 in WT (left) and Tagln-Cre:Tgfbr2flx/flx (right), with Alcian Blue staining to delineate ribs and counterstaining with Nuclear Fast Red. In the WT only a small physiological umbilical hernia is present and the small intestine is returning to the abdominal cavity, whereas the mutant shows a large exomphalos defect and very few bowel loops are present in the abdominal cavity. (E-H) Characterisation of cell type in Tagln-Cre:Tgfbr2flx/flx thoracic (right) and abdominal (left) body wall by immunohistochemistry. (E) E13.5 mutant embryos show normal lateral body wall muscles (MF20+) and ribs (SOX9+), whereas the ventral midline is made of a thin sac. Condensations of SOX9+ and MF20+ cells (arrow) are seen just lateral to the VBW in the thoracic and abdominal areas, respectively. (F) E14.5 mutant embryo shows very little progression in secondary element migration, and the condensation of chondrocyte and myocyte (arrow) is still seen lateral to the VBW in the thoracic and abdominal compartments, respectively. (G) The VBW of Tagln-Cre:Tgfbr2flx/flx still expresses TAGLN. (H) The skin covering the premature VBW in Tagln-Cre:Tgfbr2flx/flx is made of a single layer of squamous epithelial cells (insets P), while in the secondary elements multilayered cuboid epithelium covers the lateral body wall (insets S). Bottom row of insets shows E-cadherin channel. H, heart; L, lungs; LV, liver; IN, intestine; P, primary body wall; S, secondary body wall; TA, transverses abdominis; IO, internal oblique; EO, external oblique; PC, panniculus carnosus; IC, intercostal muscles; R, rib. Scale bars: 1000 µm in A,B; 500 µm in C-G; 200 µm in H, 50 µm in insets.
Fig. 7.
Fig. 7.
Tgfbr2 knockout in myogenic and chondrogenic cells does not affect midline closure. (A,B) Transverse sections in the thoracic and abdominal region of an E14.5 MyoD-Cre:Tgfbr2flx/flx embryo. (A) Alcian Blue (AB) and Nuclear Fast Red (NFR) staining show normal developmental milestones, comparable to the WT (see Fig. 6C,D). (B) Normal muscle (MF20+) and chondrocyte [NG2 (CSPG4)+] development in the midline area of the mutant mouse. (C) Whole-mount MF20 staining of a 2-day-old pup, showing normal muscle development in the midline postnatally in the mutant. The umbilicus site is marked with a dotted circle. (D,E) Transverse sections in the thoracic and abdominal region of an E15.5 NG2-CreER™:Tgfbr2flx/flx embryo. (D) Alcian Blue and Nuclear Fast Red staining show normal developmental milestones, comparable to WT. (E) Normal muscle (MF20+) and chondrocyte (NG2+) development in the midline area of the mutant mouse. (F) Whole-mount Alizarin Red and Alcian Blue showing normal rib cage development and fused sternum in the midline at the fetal stage in the mutant. Scale bars: 500 µm in A,B,D,E; 1000 µm in C,F.

References

    1. Adam P. J., Regan C. P., Hautmann M. B. and Owens G. K. (2000). Positive- and negative-acting Krüppel-like transcription factors bind a transforming growth factor beta control element required for expression of the smooth muscle cell differentiation marker SM22alpha in vivo. J. Biol. Chem. 275, 37798-37806. 10.1074/jbc.M006323200 - DOI - PubMed
    1. Assinder S. J., Stanton J.-A. L. and Prasad P. D. (2009). Transgelin: an actin-binding protein and tumour suppressor. Int. J. Biochem. Cell Biol. 41, 482-486. 10.1016/j.biocel.2008.02.011 - DOI - PubMed
    1. Belenkaya T. Y., Han C., Yan D., Opoka R. J., Khodoun M., Liu H. and Lin X. (2004). Drosophila Dpp morphogen movement is independent of dynamin-mediated endocytosis but regulated by the glypican members of heparan sulfate proteoglycans. Cell 119, 231-244. 10.1016/j.cell.2004.09.031 - DOI - PubMed
    1. Boland S., Boisvieux-Ulrich E. and Houcine O. (1996). TGF beta 1 promotes actin cytoskeleton reorganization and migratory phenotype in epithelial tracheal cells in primary culture. J. Cell Sci. 109, 2207-2219. - PubMed
    1. Brewer S. and Williams T. (2004). Loss of AP-2α impacts multiple aspects of ventral body wall development and closure. Dev. Biol. 267, 399-417. 10.1016/j.ydbio.2003.11.021 - DOI - PubMed

Publication types