XB-ART-37502Dev Biol 2008 May 01;3171:213-24. doi: 10.1016/j.ydbio.2008.02.011.
Show Gene links Show Anatomy links
Sox9 is required for invagination of the otic placode in mice.
The HMG-domain-containing transcription factor Sox9 is an important regulator of chondrogenesis, testis formation and development of several other organs. Sox9 is expressed in the otic placodes, the primordia of the inner ear, and studies in Xenopus have provided evidence that Sox9 is required for otic specification. Here we report novel and different functions of Sox9 during mouse inner ear development. We show that in mice with a Foxg1(Cre)-mediated conditional inactivation of Sox9 in the otic ectoderm, otic placodes form and express markers of otic specification. However, mutant placodes do not attach to the neural tube, fail to invaginate, and subsequently degenerate by apoptosis, resulting in a complete loss of otic structures. Transmission-electron microscopic analysis suggests that cell-cell contacts in the Sox9 mutant placodes are abnormal, although E-cadherin, N-cadherin, and beta-catenin protein expression are unchanged. In contrast, expression of Epha4 was downregulated in mutant placodes. In embryos with a Keratin-19(Cre)-mediated mosaic inactivation of Sox9, Sox9-negative and Sox9-positive cells in the otic ectoderm sort out from one another. In these embryos only Sox9-positive cells invaginate and form one or several micro-vesicles, whereas Sox9-negative cells stay behind and die. Our findings demonstrate that, in contrast to Xenopus, Sox9 is not required for the initial specification of the otic placode in the mouse, but instead controls adhesive properties and invagination of placodal cells in a cell-autonomous manner.
PubMed ID: 18377888
Article link: Dev Biol
Species referenced: Xenopus
Genes referenced: cdh1 cdh2 col2a1 dlx5 epha4 foxg1 pax2 sox10 sox9 tbx2
Article Images: [+] show captions
|Fig. 2. Otic specification occurs in the absence of Sox9. (a, b) Hematoxylin/eosin-stained transverse sections through the otic region of control and Sox9 mutant (mut) embryos at the 10-somite stage reveal a thickening of the ectoderm in both embryos (arrowheads). Panel a′ shows a hematoxylin/eosin-stained section through the same embryo as in panel a but at a level posterior to the otic region to show the thickness of non-otic head ectoderm (arrows) for comparison. (c–j) Whole mount in situ hybridizations with the indicated probes and at the indicated stages show similar expression of markers of otic specification in mutant and control embryos. (k, l) Expression of Tbx2 in the otic ectoderm is reduced in the absence of Sox9. In panels c–l, anterior is to the left, arrowheads label the otic placode. The scale bar in panel b corresponds to 50 μm in panels a, a′, b; the scale bar in panel l corresponds to 400 μm in panels c–l.|
|Fig. 3. Foxg1-Cre-mediated recombination of Sox9 blocks otic vesicle formation despite continuous expression of markers of otic specification, and results in loss of otic structures. (a, b) Whole mount view of the otic primordium of control and Sox9 mutant embryos at E9.5. Note that an otic vesicle has formed in the control embryo, whereas otic development is arrested at the placode stage in the Sox9 mutant. (c, d) Hematoxylin/eosin-stained sections through the otic region of control and Sox9 mutant embryos at E11.5. Note the complete absence of otic structures in the mutant. (e–l) Hematoxylin/eosin-stained sections through the otic region of control and Sox9 mutant (mut) embryos at the 13- to 22-somite stage (E8.5 to E9.0) reveal a failure of Sox9 mutant placodes to invaginate. Arrowheads in panels e and g indicate association between the otic placode and the hindbrain, asterisks in panels f and h the absence of this association in mutants. (m–t) Whole mount in situ hybridizations with the indicated probes and at the indicated stages. Except for Tbx2, patterns in mutant embryos resemble the patterns in control embryos prior to otic placode invagination. (f) facial (geniculate) ganglion; (hb) hindbrain; (op) otic placode; (ov) otic vesicle; (som) somites; (vc) vestibular cochlear ganglion. The scale bars in panels a–l correspond to 100 μm. the scale bar in panel t corresponds to 200 μm in panels m–t.|
|Fig. 5. Epha4 and Col2a1 expression is downregulated in Sox9 mutant placodes. (a–d) Whole mount in situ hybridization with an Epha4 probe reveals a slightly weaker and less defined Epha4 expression in mutant placodes at the 14 somite stage (a, b), and a strong reduction at the 16 somite stage (c, d). (e, f) Anti-Epha4 immunofluorescence on sections through the otic placode reveals robust membrane associated staining in control placodes and only weak staining in mutant placodes. The mutant placode in panel f is outlined with a dashed line. (g, h) Whole mount in situ hybridization with a Col2a1 probe. Col2a1 expression can not be detected in mutant placodes at the 15-somite stage. Arrowheads label the position of the otic placode. (nt) labels the neural tube, r3 and r5 label rhombomers-3 and -5, respectively. The scale bar in (h) corresponds to 250 μm in panels a–d, and 200 μm in panels g, h; the scale bar in panel f to 50 μm in panels e, f.|
|Fig. 7. Sox10 is not expressed in Sox9 mutant otic placodes. (a–d) Whole mount in situ hybridization of control (a, c) and Sox9 mutant (b, d) embryos with a Sox10 probe at the 13- (a, b) and 15-somite (c, d) stage. Arrowheads label the otic placode. (nc) labels Sox10 expression in neural crest cells. The scale bar in panel d corresponds to 200 μm in panels a–d.|
|Fig. 8. Phenotype of Sox9 mosaic embryos. (A) Sox9 mosaic embryos form one or several small Sox9-positive otic vesicles. (Aa, b) Whole mount view of the otic region of a control and Sox9 mosaic (mos) embryo at E9.5. Arrowheads label the otic vesicles. (Ac–f) Sections through the otic region of Sox9 mosaic embryos at E9.5 and E10.5, stained with a Sox9-specific antibody. Small arrows in panel c label individual Sox9-positive cells among Sox9-negative non-invaginating placode cells. (B) Sox9 immunofluorescence on sections through the otic region of E8.5 to E9.25 control (Ba, c, e) and Sox9 mosaic (Bb, d, f) embryos. Arrowheads in panel f label invaginating clusters of Sox9-positive cells. (C) Sox9-positive and -negative cells in E9.0 Sox9 mosaic placodes express Dlx5 and Pax2, but only Sox9-positive cells invaginate. (Ca) Sox9 immunohistochemistry, (Cb) Dlx5 in situ hybridization on a neighboring section to the section in panel Ca. Arrowheads label invaginating Sox9-positive cell clusters. (Cc–c′) Double-immunofluorescence against Pax2 (c, red) and Sox9 (c′, green) and overlay (c′). (D) Double-immunofluorescence against Epha4 (green) and Sox9 (red). (a, b) Overlays, (a′, b′) detection of Epha4 only. Asterisks label Sox9 negative regions of the placodes. (nt), neural tube. (E) Non-invaginating, Sox9-negative cells in the placode die by apoptosis. TUNEL assay (Ea) and anti-activated Caspase3 immunofluorescence (Eb) on sections through the otic region of Sox9 mosaic embryos. Note that no cell death is detected in the formed otic vesicle (Ea, arrowhead) or invaginating part of the placode (Eb, arrowhead). The sections shown in panels Ad, B, Eb) were counterstained with DAPI (blue) to label the nuclei. The scale bars in panels A, B, Ca, b, E) correspond to 100 μm, and to 20 μm in panels Cc′–c′, D).|