Specific expression of the Neurod1 gene in early committed neurogenic cells in adult hippocampus. (a) Immunohistochemical analysis of neurogenic dentate gyrus area in adult hippocampus of the transgenic mouse that has a Sox2 promoter–driven EGFP reporter. Top, Sox2 is shown in green, and we stained for NeuroD1 (red) and S100β (blue). Bottom left, a higher-magnification image is shown with DAPI staining (blue) Bottom right, section stained for NeuroD1 (red), S100β (green) and DAPI (blue). (b) Distinct population of neuroblast cells expressing the Neurod1 gene and mature neurons in adult mouse hippocampus. Top, Sox2 is shown in green in adult hippocampus of a Sox2 promoter–driven EGFP reporter transgenic mouse; we also stained for NeuroD1 (red) and NeuN (blue). Bottom, NeuroD1-positive cells (red) and NeuN-positive cells (green) were exclusive to the inner layer of the dentate gyrus. (c) Immunocytochemical analysis of NeuroD1-positive cells in adult rat hippocampus. NeuroD1-positive cells (red) colocalized with Ki67 (cyan). White arrows indicate colocalizing cells; the region in the white square is magnified in a separate window. Bottom left, NeuroD1-positive cells (red) colocalized calretinin (green) and DCX (green). Bottom right, some NeuroD1-positive cells (green) colocalized with nestin (red). White arrows indicate populations of colocalized cells for both markers. (d) Proliferative status of NeuroD1-positive cells in adult hippocampus. BrdU (100 mg per kg of body weight) was injected for 1 week into Fisher 344 rats (7–8 weeks old). Cells that were double positive for NeuroD1 (red) and BrdU (green) are indicated by white arrows. The colocalizing cell (NeuroD1 and BrdU positive) in the white square is magnified on the right.

Sox2/LEF DNA regulatory elements on the Neurod promoters. (a) Schematic representation of the binding sites of TCF/LEF and Sox transcription factors on the 3-kb promoters of the Neurod1 and Neurod2 genes. The sequences of the DNA regulatory elements recognized by Sox2 (black box) and TCF/LEF (white box) and the overlapping DNA regulatory consensus sequence (Sox/LEF; gray box) recognized by both Sox2 and TCF/LEF are shown (bottom). (b) Schematic representation of the Sox2/LEF-binding sites (gray boxes) in human, rat, and mouse Neurod1 promoters. (c) Comparison of expression levels between NSCs and differentiating neurons by quantitative real-time PCR (qRT-PCR) of genes related to Wnt signaling. The expression level of Gfap as a typical glial gene was assessed. Expression levels of β-tubulin III (Tubb3) and synapsin I (Syn1) were also analyzed as neuronal genes. Each mRNA value was normalized to that of Gapdh and then plotted as the fold increase of the sample of Neurod1 mRNA in NSCs (asterisk). (d) Simple reporter assay with the regulatory elements of Sox2, TCF/LEF and Sox/LEF transcription factors. The luciferase value was snormalized to a sample with a Sox-TATA reporter construct, with an FGF2 ligand (asterisk). (e) Activity of the Neurod1 promoter. A 1.5-kb Neurod1 promoter region (including a Sox/LEF site and a TCF/LEF site) was linked to the luciferase gene. The reporter construct with mutation at the Sox/LEF site on the Neurod1 promoter (NeuroD1 mutant luciferase) was also introduced to the adult NSCs. Luciferase value was normalized to sample with DMSO as a control ligand (asterisk).

Wnt signaling increases Neurod1 promoter activity during early neurogenesis. (a) Time course of Neurod1 mRNA expression during the early stages of neurogenesis in cultured adult NSCs. RT-PCR detection of Neurod1 and Gapdh is shown. Western blots of Sox2 and β-catenin during neurogenesis in cultured adult NSCs are shown in the lower panels. (b) Induction of a phosphorylated inactive form of GSK3β to stabilize β-catenin during Wnt signaling in early committed neurogenic cells. Western blots of both GSK3β and phosphorylated GSK3β were conducted using the same neuronal induction treatment. (c) The effect of Wnt signaling on the expression of Neurod1 mRNA. RT-PCR analysis using total RNA extracted from adult NSCs treated with Dkk1, DnWnt, Wnt3, TDZD8 or β-catenin shRNA. (d) The effect of Wnt signaling on the promoter activity of the Neurod1 gene. The luciferase value was normalized to that of cultured NSC sample with control vector and control ligand (DMSO, asterisk, white bar). (e) The effect of siRNAs targeting Neurog1, Neurog2, Sox2 and β-catenin on Wnt3a ligand–mediated induction on Neurod1 mRNA. qRT-PCR analysis for Neurod1 mRNA was plotted. The amount of mRNA present for each sample was normalized to that of Gapdh and then plotted as the fold increase over the control (control siRNA with DMSO). (f) ChIP analysis at the Neurod1 promoter in adult neurogenesis. PCR primers were designed to surround the Sox/LEF sequence on the rat Neurod1 promoter.

Effect of Wnt signaling on the expression of NeuroD1 and LINE-1 during adult neurogenesis. (a) DNA regulatory elements recognized by Sox2 and TCF/LEF transcription in LINE-1. A schematic representation of the Sox/LEF DNA regulatory elements (gray boxes) in the human, rat, and mouse retrotransposon LINE-1 is shown. (b) Promoter activity of the 5′ UTR and ORF fragment of LINE-1 during adult neurogenesis. Luciferase constructs with 5′ UTR and LINE-1 ORF2 sequences linked to the luciferase gene, in both forward (white) and reverse (black) orientations, were introduced into adult NSCs by lentivirus infection (left and middle panels). Partial fragments of the LINE-1 ORF2, the first 2.5 kb (black columns) and the last 1.5 kb (white columns), were also linked to the luciferase gene in the reporter assay (right). (c) ChIP analysis of rat LINE-1. PCR primers were designed to surround the Sox/LEF DNA regulatory elements. (d) Wnt3a-mediated induced production of LINE-1 ORF2 mRNA. The induction level was measured by qRT-PCR with several synthesized siRNAs. The mRNA level was normalized to that of Gapdh and then plotted as the fold increase over the control (control siRNA with DMSO). (e) The effect of Wnt signaling on the chromatin remodeling of LINE-1. Activation of the chromatin state (acetylation of histone H3) in the LINE-1 sequences was assessed using several Wnt-related constructs. (f) The activation and repression of the LINE-1 promoter. The effect of Wnt on the activity of LINE-1–based promoters was examined using the LINE-1 luciferase construct. After 1 d in the culture, luciferase assays were performed on adult NSCs treated with each construct.

Adult NSC cannot transition to immature and mature granule neurons in β-catenin cKO mice. Immunohistochemical analysis of the Sox2^CREGFP cells. (a) Sox2^CREGFP retrovirus was injected into the dentate gyrus of control mice (left) or β-catenin cKO mice (right). Immunohistochemical analysis of NeuroD1 (red), GFP (green) and DAPI (blue) in both groups is shown and GFP-positive cells colocalized with NeuroD1-positive cells in control mice (indicated by white arrows, left panels). In control mice, GFP-positive cells were present among the more differentiated neurons deeper in the granule cell layer (left). In β-catenin cKO mice, GFP-positive cells were observed more often near or in the hilus region, rather than among the more differentiated neurons deeper in the granule cell layer. (b) Numbers of GFP-positive cells in the dentate gyrus of control mice (white bars) and β-catenin cKO mice (black bars). (c) Numbers of marker and GFP double-positive cells in the dentage gyrus of control mice (white bars) and β-catenin cKO mice (black bars). (d) Percentages of marker-positive cells in the dentage gyrus of control mice (white bars) and β-catenin cKO mice (black bars).

Lineage tracing Sox2-positive NSCs in β-catenin cKO mice. Immunohistochemical analysis of the Sox2^CREGFP cells. (a,b) Sox2-positive cells were able to give rise to GFAP-positive (red, a) and to Sox2-positive NSCs (red, b). Magnified images show triple immunohistochemistry of GFAP (a) or Sox2 (b), GFP (green) and BrdU (blue, lower panels). (c) Immunohistochemical analysis of apoptotic cells labeled by AC3 in β-catenin cKO mice. Representative image of AC3 (red), GFP (green) and DAPI (blue) in control mice (left) and β-catenin cKO mice (right). The GFP-positive cells colocalizing with the AC3-positive cells are indicated by white arrows. The GFP and AC3 double-positive cells in the white dotted square are magnified in the bottom panels. (d) Quantification of GFP-positive and AC3-positive cells in dentate gyrus of control and β-catenin cKO mice. The numbers of the AC3-positive and AC3 and GFP double-positive cells in the dentate gyrus of control mice (white bars) and β-catenin cKO mice (black bars) are plotted.

Infection of lentivirus expressing β-catenin shRNA in vivo. To evaluate the effect of Wnt signaling in vivo, lentivirus expressing β-catenin shRNA or control lentivirus expressing only EGFP (lentivirus-GFP) was stereotactically injected into adult rat hippocampus. (a) Lentivirus-GFP in dentate gyrus. White arrows indicate the population of cells that were double positive for NeuroD1 (red) and EGFP (green). (b) Immunohistochemical analysis of the cells infected by the lentivirus encoding β-catenin shRNA and GFP. Cells expressing β-catenin shRNA and GFP (green) and NeuroD1-positive cells (red) are shown.

Activity of LINE-1 as a promoter in adult rat hippocampus. (a–d) We examined the activity and specificity of the LINE-1–based promoter in adult rat hippocampus. EGFP-expressing lentivirus, under the control of the LINE-1–based promoter, was stereotactically microinjected into the dentate gyrus of adult rats and the population of GFP-positive cells (green) was analyzed by immunohistochemistry using an antibody to NeuroD1 (red, a). Comparisons of the LINE-GFP population (green) to TUJ1 staining (red, b), NeuN staining (red, c) and Sox2 staining (magenta, d) are also shown.