Fig. 4. Overexpression of c-Gcm1 promotes neuronal but not glial differentiation in spinal cord. (A,B) E1.5 neural tube electroporated with control or c-Gcm1 expression vectors (green) and analysis of O4 expression (red) three days later (E5/E5.5). (A) Electroporation of control vector does not modify the pattern of O4 decorating oligodendrocyte progenitors in the ventral neuroepithelium and no ectopic expression of O4 is observed. (B) Note that the endogenous O4-positive domain is strongly reduced (arrow) and no ectopic expression of O4 is detected when the c-Gcm1-expressing vector is used. (C-I) Electroporation with control (F,H) or c-Gcm1 (D,E,G,I) expression vector was performed in E4.5/E5 embryonic spinal cord that was further dissected, opened dorsally and plated in culture with neuroepithelial precursors up, as depicted in C. Glial and neuronal differentiation was assessed three days later by immunolabeling using Glast (D, red), O4 (E,F,G, red) and Lim1/2 (H,I, red). (D,E) Transverse sections of open-book spinal cords showing that c-Gcm1 overexpression does not induce ectopic Glast-positive (D) or O4-positive (E) cells. Note in E that the O4-positive domain is strongly reduced in the ventral spinal cord. (F,G) High magnifications show that some cells electroporated with control vector have adopted an O4-positive fate (F, arrows), whereas no c-Gcm1-overexpressing cells adopt such a fate (G). (H,I) High magnifications of explants showing that most cells electroporated with control vector are located in the neuroepithelium, and only a few of them reach the mantle layer and express Lim1/2 (H). By contrast, most c-Gcm1-overexpressing cells have left the neuroepithelium and all of them express Lim1/2 (I). ne, neuroepithelium; ML, mantle layer; FP, floor plate. Scale bars: 120µm in A-E; 40µm in F-I.