XB-ART-48755J Cell Sci 2014 Jun 01;127Pt 11:2542-53. doi: 10.1242/jcs.146811.
Show Gene links Show Anatomy links
GEF-H1 functions in apical constriction and cell intercalations and is essential for vertebrate neural tube closure.
Rho family GTPases regulate many morphogenetic processes during vertebrate development including neural tube closure. Here we report a function for GEF-H1/Lfc/ArhGEF2, a RhoA-specific guanine nucleotide exchange factor that functions in neurulation in Xenopus embryos. Morpholino-mediated depletion of GEF-H1 resulted in severe neural tube defects, which were rescued by GEF-H1 RNA. Lineage tracing of GEF-H1 morphants at different developmental stages revealed abnormal cell intercalation and apical constriction, suggesting that GEF-H1 regulates these cell behaviors. Molecular marker analysis documented defects in myosin II light chain (MLC) phosphorylation, Rab11 and F-actin accumulation in GEF-H1-depleted cells. In gain-of-function studies, overexpressed GEF-H1 induced Rho-associated kinase-dependent ectopic apical constriction - marked by apical accumulation of phosphorylated MLC, γ-tubulin and F-actin in superficial ectoderm - and stimulated apical protrusive activity of deep ectoderm cells. Taken together, our observations newly identify functions of GEF-H1 in morphogenetic movements that lead to neural tube closure.
PubMed ID: 24681784
PMC ID: PMC4038946
Article link: J Cell Sci
Species referenced: Xenopus
Genes referenced: actl6a arhgef2 h1-3 mlc1 myc rab11a rho rho.2 rhoa shroom3 sox2 srf
Morpholinos: arhgef2 MO2 arhgef2 MO3 arhgef2 MO4
Article Images: [+] show captions
|Fig. 1. Neural tube defects in embryos depleted of GEF-H1. (A) Uninjected embryo at stage 16. (B–D) Four-cell embryos were unilaterally injected with control (CoMO) (B), or GEF-H1a (GEF-H1a MO) (C,D) morpholino oligonucleotides (20 ng each) and LacZ RNA (40 pg). (D) Human GEF-H1 RNA (60 pg) rescued the neural fold defect that was caused by GEF-H1a MO. The dorsal view is shown, anterior is to the top. Arrows point to the neural folds. (E,F) Neural fold morphology in stage-17 (st 17) embryos that had been injected with CoMO (40 ng) (E) or GEF-H1 MO (F) (40 ng) was visualized by using β-catenin staining (green). Membrane-associated RFP RNA (160 pg) (F) marks the injected side of the embryo. N, notochord; S, somite. Tissue boundaries are demarcated by broken lines. Arrowhead points to constricting cells, * lack of apical constriction. (G,H) Constricted cell morphology in the cross-sections of neurula embryos (stage 15) that had been unilaterally injected with GEF-H1 MO (40 ng). (G) control (co) embryo. (H) GFP (green) is a lineage tracer, (H′) red channel only of H. The broken lines in G and H′ show the cell boundaries. M, midline position, dorsal is at the top. Arrows point to constricting cells, * lack of apical constriction. (I) Quantification of neural fold defects caused by the injection of GEF-H1a MO and the rescue with human or Xenopus GEF-H1 RNA. Morpholino oligonucleotides and RNAs were injected at the indicated doses. Neural tube defects were scored at stages 16–18 as described in supplementary material Fig. S1. The number of embryos that were scored is shown along the top. (J) Quantification of cell shape in superficial neuroectoderm from GEF-H1-MO-injected or control uninjected sides. Results are expressed as the ratios of apical-basal length over the apical width (ABL/AW) in four to six superficial cells that were located adjacent to the midline. Means±s.d. are shown. Cells were scored in 8–11 sections from three embryos, with the total cell number shown on top of each group. Statistical significance was assessed by using Student's t-test, P<0.0001. Scale bars: 100 µm (E); 20 µm (G).|
|Fig. 2. Changes in the Sox2-positive domain in GEF-H1 morphants. (A–E) Embryos were injected laterally into two sites at the four-cell stage with LacZ RNA as lineage tracer (40 pg) and morpholino oligonucleotides or RNAs as indicated. Doses were as follows: (A–C) GEF-H1a MO and (D) CoMO (control) (20 ng), (E) GEF-H1 MO (40 ng) and human GEF-H1 RNA (GEF-H1; wt) (60 pg) or human GEF-H1-Y393A RNA (Y393A) (800 pg). Whole-mount in situ hybridization with a Sox2 probe at stages 14–15 is shown, β-galactosidase activity (red) marks the injected side. A dorsal view is shown, the anterior is top. (B) Wild-type human GEF-H1 but not the (C) inactive GEF-H1-Y393A RNA reverses the morphant phenotype (A). (A–E) White bars indicate the width of the Sox2 expression domain at the level of the mid-hindbrain boundary. (F) Quantification of embryo phenotypes. The Sox2 domain was measured at the mid-hindbrain boundary as indicated in A–D. The ratios of the Sox2 domain width on the injected side over that on the uninjected side are shown. A ratio of 1.0 indicates equal width of the Sox2 domain on both sides. Statistical significance was assessed by using Student's t-test. ***P<0.0001. The n values are shown across the top of the graph and indicate the number of embryos that were measured.|
|Fig. 3. Depletion of GEF-H1 interferes with apical constriction in the neural plate. Four-cell embryos were unilaterally co-injected with 40 ng of morpholino oligonucleotide against GEF-H1 (GEF-H1 MO) and 200 pg of GFP RNA (as a lineage tracer; green). Immunostaining of transverse neural fold sections (in stage-16 embryos) for apical constriction markers. The uninjected side served as a control (co st 16). Arrows point to the uninjected area, * cells that had been injected with GEF-H1 MO. The midline (M) is indicated by the broken line. (A,B) phosphorylated myosin II light chain (pMLC; red), (C,D) F-actin, (E,F) Rab11. Control-MO-injected embryos (supplementary material Fig. S1D–F) were indistinguishable from the uninjected embryos. B′, D′, F′ show the images in the corresponding panel in the red channel only; B′′ and D′′ show the images in the corresponding panel in the green channel only. Scale bar: 20 µm.|
|Fig. 5. GEF-H1 induces ectopic apical constriction in ectodermal cells. (A–C) Ectopic apical constriction induced by GEF-H1 and Shroom in ectoderm. Four-cell embryos were injected with RNAs and cultured until late-blastula or early-gastrula stages (9–10+). (A) Uninjected control (Co) embryo. (B) Embryo injected with Xenopus Myc–GEF-H1 RNA (50 pg). (C) Embryo injected with Shroom RNA (200 pg). Arrows point to the areas undergoing apical constriction. Top view is shown. (D–G) Cross-sections of embryos that had been injected with human GEF-H1-Y393A (D) or human GEF-H1-C53R (E) or Shroom (F) RNA. Ectoderm cells accumulate pigment at the apical surface and reduce their apical surfaces (arrows). Human GEF-H1-Y393A-expressing cells are indistinguishable from control uninjected ectoderm. (G) Scheme of the experiments showing the tissue sectioned in (D–J), apical is up. (H) Quantification of cell shape in GEF-H1-expressing and control (uninj) cells. Results are expressed as ratios of apical-basal cell length over apical width (ABL/AW) in the superficial layer cells (as shown in D and E). Means±s.d. are shown. The numbers above the bars indicate the number of cells counted per group (n). Five GEF-H1-expressing and two control embryos were examined. Statistical significance was assessed by Student's t-test, P<0.0001. (I) Morphology of cells that underwent apical constriction (arrows) from embryos that had been injected with Xenopus Myc–GEF-H1 (50 pg) or Shroom (200 pg) RNAs as indicated. The cell membrane is marked by co-expression of membrane-targeted GFP (mGFP), which was detected by using an antibody against GFP (green). Co, control ectoderm. (J) En face views of ectoderm from the experiment in I. Arrows point to cells that express Xenopus Myc-GEF-H1 or Shroom with membrane-associated GFP (mGFP). Live embryos are shown. * uninjected cells. Scale bars: 20 µm.|
|Fig. 6. GEF-H1 activates actomyosin contractility in superficial ectoderm. (A,B) Cross-sections of superficial ectoderm that expressed Xenopus Myc-GEF-H1 RNA (25–50 pg). RNA injections were as described in the Fig. 5 legend. (A) Brightfield image of uninjected ectoderm (Co). (B) Brightfield view of GEF-H1-expressing cells that are elongated and highly pigmented. (B′) Myc-staining shows apical enrichment of Xenopus Myc-GEF-H1 in constricting cells, which are elongated and heavily pigmented (arrow). (C,D) Accumulation of apical F-actin (red) revealed by Phalloidin staining (arrow) in cross-sections of Xenopus Myc-GEF-H1-expressing ectoderm (D), but not in uninjected control tissue (C). (E) GEF-H1 and ROCK activate MLC phosphorylation (pMLC). Immunoblot analysis of lysates of stage-11 embryos, which had been injected with GEF-H1 and ROCK RNA, using antibodies against phosphorylated MLC. No change in total GFP–MLC levels is detected (antibody against GFP). (F,G) Accumulation of pMLC (green) in GEF-H1-expressing cells. (F) Control uninjected embryonic ectoderm. (G) Accumulation of pMLC at constricted apical surfaces (arrows) of Xenopus Myc-GEF-H1-expressing cells. G′, Brightfield reveals elongated ‘bottle cell’ morphology (arrows) with reduced apical surfaces and increased pigmentation. (H,I) GEF-H1 induces γ-tubulin redistribution (red). (H) Control ectoderm at stage 10.5 showing centrosomal localization (arrowheads). (I,I′) Myc-GEF-H1 RNA-injected ectoderm with constricting cells (arrow). I′ shows the red channel only. Scale bars: 20 µm.|
|Fig. 7. ROCK is required for GEF-H1-induced ectopic apical constriction. (A,B) Four-cell embryos were injected with GEF-H1-C53R (60 pg) and/or dominant-negative ROCK (DN-ROCK, 400 pg) RNAs into two animal-ventral blastomeres. The animal pole view of stage-8.5 embryos is shown. (A) Intense pigmentation (arrows) accompanies apical constriction in response to expression of GEF-H1-C53R. (B) Dominant-negative ROCK completely blocks the effect of active GEF-H1. (C) Quantification of phenotypic changes for embryos that had been injected with the indicated RNAs. Scoring: ++, hyper-pigmentation; +, weak hyperpigmentation; no change, normal morphology; −, de-pigmentation. Pigment accumulation was associated with apical surface reduction and constriction. The y-axis shows the percentage of embryos with the indicated phenotype. The numbers across the top indicate the total number of embryos examined. (D) Protein levels for GFP-GEF-H1-C53R and Myc-tagged dominant-negative ROCK were assessed by using antibodies against GFP and Myc in embryo lysates taken from experiments performed in A and B. (E) Dominant-negative ROCK suppresses GEF-H1-dependent MLC phosphorylation (pMLC). Four-cell embryos were injected into four animal blastomeres with RNAs encoding GFP–MLC (250 pg), GEF-H1-C53R (C53R, 60 pg), GEF-H1-Y393A (Y393A, 500 pg), dominant-negative ROCK (400 pg, H, high dose; 200 pg, L, low dose), or wild-type ROCK (ROCK, 250 pg) as indicated. Injected embryos were lysed at stage 8.5 for western blotting analysis with antibodies against pMLC. Antibody staining against Myc and GFP shows the levels of dominant-negative ROCK and GFP–MLC, respectively. α-tubulin is loading control. (F) ROCK is required for GEF-H1-mediated SRF-Luc reporter activation. 20 pg of SRF reporter (3DA-SRF-Luc or mutated 2MA-SRF-Luc) DNAs were co-injected with RNAs encoding GEF-H1-C53R (60 pg) and dominant-negative ROCK (250 pg) into two dorsal-animal blastomeres of four-cell embryos, as indicated. The relative luciferase activity was measured in lysates of embryos harvested at stage 10.5. Means±s.d. are shown. Every group included four independent sets, each containing four embryos.|
|Fig. 8. GEF-H1 regulates protrusive behavior and the polarized distribution of phosphorylated myosin light chain in inner ectoderm cells. (A–D) Immunostaining of transverse ectoderm sections of stage-11 (st 11) embryos that had been injected with the indicated morpholino oligonucleotides (40 ng) or Myc–GEF-H1 mRNA (5–10 pg). GFP is a lineage tracer. (A) In control (Co) ectoderm, phosphorylated MLC (pMLC) is enriched at the apico-lateral boundaries of inner cells (arrowhead). The apical–basal axis is indicated. (B) Overexpression of Myc–GEF-H1 leads to the formation of apical GEF-H1-positive protrusions in inner ectoderm cells (arrow, n>25). pMLC staining is shown (red). B′ shows the staining of Myc (green) in the same field as that shown in B. B′ shows the merged image of B and B′. (C,D) The GEF-H1 morpholino oligonucleotide (GEF-H1 MO, D) but not the control morpholino oligonucleotide (CoMO, C) interferes with pMLC (red) polarization and causes disorganized inner cell morphology in the ectoderm of injected embryos (n>25). C′ and D′ show merged images of the pMLC staining from the corresponding panel with localization of GFP. * lack of pMLC staining in GEF-H1-depleted cells as compared with that of uninjected cells (arrow). Scale bar: 20 µm.|
References [+] :
Aijaz, Binding of GEF-H1 to the tight junction-associated adaptor cingulin results in inhibition of Rho signaling and G1/S phase transition. 2005, Pubmed