Fig. 3. Interference with CS-GAGs impairs neurogenesis and simultaneously favors gliogenesis in vitro. The distribution of ßIII-tubulin-positive neurons and GFAP-positive astrocytes was assessed in mouse Nsph cryosections by double immunohistochemistry. (A,B) Photomicrographs illustrating cortical (A) or striatal Nsphs (B) grown in the absence (w/o, upper panels) or presence (lower panels) of ChABC. The digestion of CS-GAGs by ChABC reduced ßIII-tubulin immunoreactivity and enhanced GFAP immunoreactivity. (C,D) These observations were quantified using acutely dissociated suspensions of control (dark gray) and ChABC-treated (light gray) cortical (C) and striatal (D) Nsphs and plotted as percentage fractions. An apparent 4-fold reduction in the frequency of ßIII-tubulin-positive neurons was recorded in both the cortical (Cor) and striatal (GE) ChABC-treated Nsphs that was accompanied by a 2-fold increase in GFAP-positive cells. (E,F) An analogous differentiation assay was performed using the 473HD-expressing subpopulation of radial glia immunopanned from acutely dissociated primary Nsphs. When the 473HD-immunoselected cells were cultivated for 2 days in vitro (div) in the presence of ChABC (light gray), the number of ßIII-tubulin-positive neurons diminished 4-fold (E) and GFAP-expressing astrocytes doubled in number both for cortical and striatal Nsphs (F), as compared with control conditions (dark gray). (G,H) The importance of CS-GAGs for fate decision was confirmed in Nsph differentiation assays in which ChABC treatment for 6 div caused a reduction in the number of ßIII-tubulin-positive neurons (G) and an elevation in the number of GFAP-positive astrocytes (H). (A,B,G,H) Cell nuclei were counterstained with bisbenzimide and are shown in blue (DAPI). ^**, P<0.01; ^***, P<0.001. Scale bars: 50 µm in A,B; 100 µm in G,H.