Fig 1: Conditional GSK-3ß knock-out in NSCs induces a time-dependent AHN deficit. (A) GSK-3ß was conditionally knocked-down in NSCs by cross-breeding Nes-Cre/ERT2 with GSK-3ßflox/flox mice, and then knocked-out by cross-breeding GSK-3ß+/- mice. (B) Representative immunofluorescent images showing BrdU- and DCX- labeled AHN at 4- and 8-wpt. Scale bars were 50 µm. (C) Both GSK-3ß knock-down and knock-out increased the number of BrdU-labeled newborn cells at 4-wpt (left), while GSK-3ß knock-out significantly decreased the proportion of BrdU/DCX co-labeled cells (i.e. immature newborn neurons). One-way ANOVA followed by Tukey's multiple comparisons tests (Left, F (2,12) = 8.301, P = 0.0055. Right, F (2,6) = 3.750, P = 0.0878). * P < 0.05, ** P < 0.01. n= 3~5 mice/group. (D) GSK-3ß knock-out but not knock-down reduced DCX-positive immature neurons at 8-wpt. One-way ANOVA followed by Tukey's multiple comparisons tests (F (2, 9) = 4.917, P = 0.0361). * P < 0.05, n= 3~5 mice/group. (E) Experimental scheme illustrating the labeling of newborn neurons with ROV-EGFP. (F) Representative images showing the dendrite maturation of ROV-EGFP-labeled newborn neurons. (G) GSK-3ß-/- mice exhibited decreased dendrite complexity of newborn neurons both at 4- and 8-wpt but not at 3-wpt, compared with the age-matched GSK-3ßflox/flox or GSK-3ß+/- littermates. Repeated measures ANOVA followed by Tukey's multiple comparisons tests (3-wpt, genotype effect: F (5, 156) = 76.29, P < 0.0001; distance effect: F (2, 156) = 0.7106, P = 0.4929; interaction: F (10, 156) = 1.036, P = 0.4155. 4-wpt, genotype effect: F (5, 156) = 112.9, P < 0.0001; distance effect: F (2, 156) = 11.44, P < 0.0001; interaction: F (10, 156) = 2.609, P = 0.0059. 8-wpt, genotype effect: F (5, 132) = 54.16, P < 0.0001; distance effect: F (2, 132) = 6.314, P = 0.0024; interaction: F (10, 132) = 2.700, P = 0.0048). * P < 0.05. n = 3 mice/group. Data were represented as mean ± SEM.