Fig. 4: Dentate gyrus neuronal FGFR1 activation identifies a novel FGFR1–Notch–BDNF axis.

a A schematic representation of viral transduction and activation in hippocampal neuronal culture. b A graphical summary of mRNA expression level upon FGFR1 activation. c–l qRT–PCR results for Notch1 (c), Rest (d), Hes1 (e), Hes5 (f), Hey1 (g), Hey2 (h), Runx1 (i), Sdc4 (j), FGF2 (k) and BDNF (l) in optoFGFR1-transduced neurons exposed to blue light illumination, GFP-transduced neurons were treated with soluble basic FGF. The data are presented as means ± s.e.m. The sample sizes were follow: for c n = 7 for 0 h light stimulation, n = 4 for 10 min light stimulation, n = 3 for 30 min light stimulation, n = 4 for 2 h light stimulation, n = 4 for 4 h light stimulation, n = 3 for 6 h light stimulation, n = 4 for 12 h light stimulation, n = 3 for 16 h light stimulation, n = 3 for 24 h light stimulation, n = 4 for 30 min light stimulation with EGFP expression and n = 4 for 30 min ligand treatment with EGFP expression. A one-way ANOVA revealed a significant effect of optoFGFR1 activation over time (F(10, 32) = 9.286, P < 0.0001). Post hoc comparisons using Tukey’s multiple comparisons test showed significant differences between control and optoFGFR1 activation (0 h versus 30 min, P = 0.003; 0 h versus bFGF, P = 0.0002; 30 min versus bFGF, P = 0.0012). For d n = 7 for 0 h light stimulation, n = 3 for 10 min light stimulation, n = 3 for 30 min light stimulation, n = 5 for 2 h light stimulation, n = 3 for 4 h light stimulation, n = 3 for 6 h light stimulation, n = 3 for 12 h light stimulation, n = 3 for 16 h light stimulation, n = 4 for 24 h light stimulation, n = 3 for 2 h light stimulation with EGFP expression and n = 3 for 2 h ligand treatment with EGFP expression. A one-way ANOVA revealed a significant effect of optoFGFR1 activation over time (F(10, 29) = 6.01, P < 0.0001). Post hoc comparisons using Tukey’s multiple comparisons test showed significant differences between control and optoFGFR1 activation (0 h versus 2 h, P < 0.0001; 0 h versus bFGF, P = 0.0011; 2 h versus bFGF, P = 0.0173). For e n = 11 for 0 h light stimulation, n = 7 for 10 min light stimulation, n = 7 for 30 min light stimulation, n = 7 for 2 h light stimulation, n = 9 for 4 h light stimulation, n = 9 for 6 h light stimulation, n = 8 for 12 h light stimulation, n = 6 for 16 h light stimulation, n = 3 for 24 h light stimulation, n = 3 for 2 h light stimulation with EGFP expression and n = 5 for 2 h ligand treatment with EGFP expression. A one-way ANOVA revealed a significant effect of optoFGFR1 activation over time (F(10, 64) = 7.018, P < 0.0001). Post hoc comparisons using Tukey’s multiple comparisons test showed significant differences between control and optoFGFR1 activation (0 h versus 2 h, P < 0.0001; 0 h versus bFGF, P = 0.001; 2 h versus bFGF, P = 0.0097). For f n = 7 for 0 h light stimulation, n = 3 for 10 min light stimulation, n = 3 for 30 min light stimulation, n = 8 for 2 h light stimulation, n = 3 for 4 h light stimulation, n = 3 for 6 h light stimulation, n = 3 for 12 h light stimulation, n = 3 for 16 h light stimulation, n = 5 for 24 h light stimulation, n = 4 for 2 h light stimulation with EGFP expression and n = 4 for 2 h ligand treatment with EGFP expression. A one-way ANOVA revealed a significant effect of optoFGFR1 activation over time (F(10, 35) = 8.159, P < 0.0001). Post hoc comparisons using Tukey’s multiple comparisons test showed significant differences between control and optoFGFR1 activation (0 h versus 2 h, P = 0.0002; 0 h versus bFGF, P = 0.0021; 2 h versus bFGF, P = 0.009). For g n = 4 for 0 h light stimulation, n = 3 for 10 min light stimulation, n = 3 for 30 min light stimulation, n = 5 for 2 h light stimulation, n = 3 for 4 h light stimulation, n = 3 for 6 h light stimulation, n = 3 for 12 h light stimulation, n = 3 for 16 h light stimulation, n = 4 for 24 h light stimulation, n = 3 for 2 h light stimulation with EGFP expression and n = 3 for 2 h ligand treatment with EGFP expression. A one-way ANOVA revealed no significant effect of optoFGFR1 activation over time (F(10, 26) = 1.016, P = 0.4571). Post hoc comparisons using Tukey’s multiple comparisons test showed no significant differences between control and optoFGFR1 activation pairs. For h n = 8 for 0 h light stimulation, n = 3 for 10 min light stimulation, n = 3 for 30 min light stimulation, n = 4 for 2 h light stimulation, n = 3 for 4 h light stimulation, n = 3 for 6 h light stimulation, n = 3 for 12 h light stimulation, n = 4 for 16 h light stimulation, n = 4 for 24 h light stimulation, n = 3 for 2 h light stimulation with EGFP expression and n = 3 for 2 h ligand treatment with EGFP expression. A one-way ANOVA revealed a significant effect of optoFGFR1 activation over time (F(10, 30) = 10.22, P < 0.0001). Post hoc comparisons using Tukey’s multiple comparisons test showed significant differences between control and optoFGFR1 activation (0 h versus 2 h, P < 0.0001; 0 h versus bFGF, P < 0.0001; 2 h versus bFGF, P = 0.0006). For i n = 4 for 0 h light stimulation, n = 3 for 10 min light stimulation, n = 3 for 30 min light stimulation, n = 5 for 2 h light stimulation, n = 3 for 4 h light stimulation, n = 3 for 6 h light stimulation, n = 3 for 12 h light stimulation, n = 3 for 16 h light stimulation, n = 4 for 24 h light stimulation, n = 3 for 2 h light stimulation with EGFP expression and n = 3 for 2 h ligand treatment with EGFP expression. A one-way ANOVA revealed no significant effect of optoFGFR1 activation over time (F(10, 26) = 0.7205, P = 0.6983). Post hoc comparisons using Tukey’s multiple comparisons test showed no significant differences between control and optoFGFR1 activation pairs. For j n = 8 for 0 h light stimulation, n = 5 for 10 min light stimulation, n = 5 for 30 min light stimulation, n = 5 for 2 h light stimulation, n = 5 for 4 h light stimulation, n = 5 for 6 h light stimulation, n = 5 for 12 h light stimulation, n = 5 for 16 h light stimulation, n = 3 for 24 h light stimulation, n = 3 for 4 h light stimulation with EGFP expression and n = 3 for 4 h ligand treatment with EGFP expression. A one-way ANOVA revealed a significant effect of optoFGFR1 activation over time (F(10, 41) = 3.781, P = 0.0012). Post hoc comparisons using Tukey’s multiple comparisons test showed significant differences between control and optoFGFR1 activation (0 h versus 4 h, P = 0.0212; 0 h versus bFGF, P = 0.0064; 4 h versus bFGF, P = 0.0139). For k n = 11 for 0 h light stimulation, n = 8 for 10 min light stimulation, n = 8 for 30 min light stimulation, n = 8 for 2 h light stimulation, n = 8 for 4 h light stimulation, n = 8 for 6 h light stimulation, n = 9 for 12 h light stimulation, n = 8 for 16 h light stimulation, n = 3 for 24 h light stimulation, n = 3 for 12 h light stimulation with EGFP expression and n = 4 for 12 h ligand treatment with EGFP expression. A one-way ANOVA revealed a significant effect of optoFGFR1 activation over time (F(10, 67) = 6.426, P < 0.0001). Post hoc comparisons using Tukey’s multiple comparisons test showed significant differences between control and optoFGFR1 activation (0 h versus 12 h, P = 0.002; 0 h versus bFGF, P < 0.0001; 12 h versus bFGF, P = 0.016). For l n = 11 for 0 h light stimulation, n = 7 for 10 min light stimulation, n = 7 for 30 min light stimulation, n = 7 for 2 h light stimulation, n = 9 for 4 h light stimulation, n = 7 for 6 h light stimulation, n = 11 for 12 h light stimulation, n = 4 for 16 h light stimulation, n = 3 for 24 h light stimulation, n = 3 for 12 h light stimulation with EGFP expression and n = 3 for 12 h ligand treatment with EGFP expression. A one-way ANOVA revealed a significant effect of optoFGFR1 activation over time (F(10, 61) = 12.08, P < 0.0001). Post hoc comparisons using Tukey’s multiple comparisons test showed significant differences between control and optoFGFR1 activation (0 h versus 12 h, P < 0.0001; 0 h versus bFGF, P = 0.001; 12 h versus bFGF, P = 0.0062). ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. m qRT–PCR results for Notch1, Notch-related transcription factors and growth factors treated with shRNA-Scramble (gray) or shNotch1 (blue). Each component was optogenetically activated at its maximal expression time identified in b. The data are presented as means ± s.e.m.; n = 3 for Notch1 gene shScramble, n = 6 for Notch1 gene shRNA-Notch1, n = 3 for Hes1 gene shScramble, n = 6 for Hes1 gene shRNA-Notch1, n = 3 for Hey2 gene shScramble, n = 6 for Hey2 gene shRNA-Notch1, n = 4 for FGF2 gene shScramble, n = 4 for FGF2 gene shRNA-Notch1, n = 4 for BDNF gene shScramble and n = 4 for BDNF gene shRNA-Notch1. A two-way ANOVA revealed significant effect of genes (F(4, 33) = 6.035, P = 0.0009), shNotch1 (F(1, 33) = 130.8, P < 0.0001) and a significant interaction of the genes and shNotch1 (F(4, 33) = 6.035, P = 0.0009). Post hoc comparisons using Šídák’s multiple comparisons test showed that shRNA‐mediated Notch1 knockdown significantly reduced mRNA levels of Notch1 (P < 0.0001), Hes1 (P = 0.0137), Hey2 (P = 0.0322), FGF2 (P < 0.0001) and BDNF (P = 0.0015) compared with scramble control. *P < 0.05; **P < 0.01; ****P < 0.0001. n A schematic representation of optoFGFR1 viral injection and activation. o Western blots for Notch1, Hes1, FGF2 and BDNF using hippocampal dentate gyrus tissue lysate. Gray, no light activation; blue, light activation at each component’s maximal expression time identified in b. The data are presented as means ± s.e.m.; n = 6 for Notch1 protein dark, n = 7 for Notch1 protein light, n = 5 for Hes1 protein dark, n = 4 for Hes1 protein light, n = 6 for FGF2 protein dark, n = 6 for FGF2 protein light, n = 5 for BDNF protein dark and n = 6 for BDNF protein light. A two‐way ANOVA with target genes and light activation revealed significant main effects of light versus dark illumination (F(1, 37) = 153.1, P < 0.0001), target genes (F(3, 37) = 2.964, P = 0.0445) and a target genes and light activation interaction (F(3, 37) = 2.964, P = 0.0445). Post hoc comparisons using Šídák’s multiple comparisons test showed that light significantly increased levels of Notch1 (P < 0.0001), Hes1 (P < 0.0001), FGF2 (P = 0.0002) and BDNF (P < 0.0001) compared with dark controls. ***P < 0.001; ****P < 0.0001.