Neogenin, a regulator of adult hippocampal neurogenesis, prevents depressive-like behavior

Abstract

Adult neurogenesis in hippocampal dentate gyrus (DG) is a complex, but precisely controlled process. Dysregulation of this event contributes to multiple neurological disorders, including major depression. Thus, it is of considerable interest to investigate how adult hippocampal neurogenesis is regulated. Here, we present evidence for neogenin, a multifunctional transmembrane receptor, to regulate adult mouse hippocampal neurogenesis. Loss of neogenin in adult neural stem cells (NSCs) or neural progenitor cells (NPCs) impaired NSCs/NPCs proliferation and neurogenesis, whereas increased their astrocytic differentiation. Mechanistic studies revealed a role for neogenin to positively regulate Gli1, a crucial downstream transcriptional factor of sonic hedgehog, and expression of Gli1 into neogenin depleted NSCs/NPCs restores their proliferation. Further morphological and functional studies showed additional abnormities, including reduced dendritic branches and spines, and impaired glutamatergic neuro-transmission, in neogenin-depleted new-born DG neurons; and mice with depletion of neogenin in NSCs/NPCs exhibited depressive-like behavior. These results thus demonstrate unrecognized functions of neogenin in adult hippocampal NSCs/NPCs-promoting NSCs/NPCs proliferation and neurogenesis and preventing astrogliogenesis and depressive-like behavior, and suggest neogenin regulation of Gli1 signaling as a possible underlying mechanism.

Fig. 1. Neogenin expression in NSCs/NPCs and immature neurons in adult mouse hippocampal dentate gyrus

a X-gal staining (blue) analysis for lacZ gene expression (neogenin) in 2-month-old Neo^+/− brain sections. a’ a high magnification of the select region in a. Scale bars: 100μm in a and 50μm in a’. b Immunostaining of β-gal (green) in 2-month-old mouse hippocampal DG. The select region was amplified and shown in b’. Scale bars: 100 μm in b and b’. c Schematic illustration of various types of cells in adult mouse hippocampal SGZ. d–h Co-immunostaining analyses of β-gal (green) with different cell markers: Nestin or GFAP for radial NSCs; Mcm2 for non-radial NSCs; DCX for immature neurons, and NeuN for mature neurons. Scale bars: 10 μm in d–h

Fig. 2. Reduced adult hippocampal neurogenesis in Neo^Gli1CreER-CKO mice

a Schematic diagram of tamoxifen treatment and tissue (hippocampus) collection at indicated times after tamoxifen treatment. b Immunostaining analysis of tdTomato (red) at indicated times in the hippocampus of control (Neo^+/+;Gli1-CreER^T2;Ai9) and Neo^Gli1CreER-CKO mice. Scale bar = 100 μm. c Quantitative analysis of the numbers of tdTomato⁺ cells per DG. Data presented are mean ± SEM. (n = 5 per genotype). *P < 0.05; **P < 0.01. Student’s t-test. d Experimental scheme of tamoxifen injections and immunostaining analysis of DCX (white) in hippocampus of control and Neo^Gli1CreER-CKO mice. Scale bars = 100 μm. e Quantitative analysis of relative DCX⁺ cell density in d. f Quantitative analysis of relative process length of DCX⁺ neurons in d. g-h Western blot analysis of DCX protein levels in control and Neo^Gli1CreER-CKO hippocampus. Representative blots were shown in g, and quantification of the relative DCX protein levels was presented in h. Data in e, f, and h are presented as the mean ± SEM (n = 3 per genotype). *P < 0.05; **P < 0.01. Student’s t-test

Fig. 3. Decreased proliferation, whereas increased astrocytic differentiation, and no effect of neuronal differentiation from neogenin-deficient hippocampal NSCs/NPCs

a Model of the proliferation and differentiation of hippocampal NSCs/NPCs. b Double immunostaining of tdTomato (red) and DCX (green) in hippocampus of control and Neo^Gli1CreER-CKO mice. Scale bars = 50 μm. c Quantification of the ratio of tdTomato⁺; DCX⁺ over total tdTomato⁺ cells. d Double immunostaining of tdTomato (red) and Mcm2 (green). Scale bars = 50 μm. e Quantification of the ratio of tdTomato⁺ Mcm2⁺ over total tdTomato⁺ cells. f Double immunostaining analysis of tdTomato (red) and GFAP (green). f1–f4 showed a higher magnification of the selected regions. Radial glia-like NSCs (f1 and f3): Bipolar shape with a long radial process, located in SGZ of DG. Astrocytes (f4): stellate shape with star-like short processes, located in SGZ, granule cell layer and hilus. Scale bars = 50 μm. g, h Quantification of the radial glia-like NSCs g or astrocytes h ratio of tdTomato⁺; GFAP⁺ over total tdTomato⁺ cells. Data are presented as the mean ± SEM. (n ≥ 500 cells). *P < 0.05; **P < 0.01. Student’s t-test

Fig. 4. Reduced proliferation, and promoted their cell cycle exit, from neogenin deficient hippocampal NSCs/NPCs

a Schematic diagram of multiple tamoxifen and Brdu injections for indicated time. b Double immunostaining of Brdu (green) and Ki67 (red) in hippocampus of control and Neo^Gli1CreER-CKO mice. Scale bars = 100μm. c, d Quantification of Brdu⁺ cell density in c and Ki67⁺ cell density in d. Data are mean ± SEM. (n = 5 per genotype). *P < 0.05; *P < 0.01. Student’s t-test. e Quantification of the ratio of Brdu⁺; Ki67^- over total Brdu⁺ cells. Data are mean ± SEM. (n ≥ 500 cells). *P < 0.05. Student’s t-test. f Model of neogenin promotion of NSC/NPC proliferation, but suppression of their astrocytic differentiation

Fig. 5. Loss of neogenin reduced Gli1-mediated proliferation, but promoted astrocytic differentiation, in cultured NSCs

a Schematic drawing shows the isolation of DG-NSCs. b Brdu analysis for NSCs proliferation in vitro. Nestin (red), Brdu (white) and neogenin (green). (n ≥ 1000 cells from 5 slides). Scale bars = 50μm. c Quantification of Brdu incorporated ratio of Brdu⁺;Nestin⁺ over total Nestin⁺ cells. d, e Immunostaining of GFAP d or DCX e in differentiated cells in vitro. (n ≥ 800 cells from 5 slides). Scale bars = 20μm. f, g Quantification of the ratio of GFAP⁺f or DCX⁺g cells over total cells. h Real-time PCR (RT-PCR) analysis of relative gene levels of Shh signaling pathway (Normalized to NSCs of Neo^f/f mice). i Western blot analysis of Gli1 protein levels in Neo^f/f and Neo^Nestin-CKO NSCs. j Quantification of the relative Gli1 protein levels in I. (n = 3 per genotpye) k Brdu analysis for NSCs proliferation in three groups (Neo^f/f, Neo^Nestin-CKO and Neo^Nestin-CKO + Gli1). Scale bars = 10μm. l Quantification of the Brdu incorporated ratio, Brdu⁺;Nestin⁺ over total Nestin⁺ cells. Data are presented as the mean ± SEM. (n ≥ 1000 cells from 5 slides). *P < 0.05; **P < 0.01; *** P < 0.001. Student’s t-test or one-way ANOVA plus post-hoc analysis

Fig. 6. Impaired dendritic morphogenesis in neogenin deficient new-born neurons

a Retroviruses expressing d-cre or cre were injected into DG of Neo^f/f mice for 21days. Scale bars = 20 μm. b Tracing of representative new-born neurons in the DG from both groups. c Sholl analysis of the dendritic complexity in b by ImageJ software. (n ≥ 50 cells). d Quantification of the dendritic total length. (n ≥ 50 cells). e Quantification of the branch number. (n ≥ 50 cells). f Representative spine images of neurons infected with retrovirus expressing d-cre or cre for 1-month. Scale bars = 5 μm. g Quantitative analysis of the number of spines in d-cre or cre expressing new-born neurons. h Quantitative analysis of different types of spines in d-cre or cre expressing new-born neurons. Data are mean ± SEM. (n ≥ 30 cells). *P < 0.05; **P < 0.01. Student’s t-test

Fig. 7. Impairment of excitatory synaptic transmission and depressive-like behavior in mice that selectively knocked out neogenin in NSCs/NPCs

a Schematic drawing showed the strategy for electrophysiological recording. b Recording of tdTomato⁺ DG new-born neurons (red) in both Neo^+/+;Ai9 (control) and Neo^f/f;Ai9 (knockout neogenin) mice, Scale bars = 10 μm. c Representative traces of mEPSC of new-born neurons from Neo^+/+;Ai9 and Neo^f/f;Ai9 mice. d, e Quantitative analysis of the frequency d and amplitude e of mEPSC. (n ≥ 10 neurons). f Representative traces of mIPSC of new-born neurons from Neo^+/+;Ai9 and Neo^f/f;Ai9 mice. g, h Quantitative analysis of the frequency g and amplitude h of mIPSC. (n ≥ 10 neurons). i Schematic diagram of tamoxifen treatment and behavior test. j, k Control and Neo^Gli1CreER-CKO mice were subjected to the TST j and FST k, and quantification of the immobility time in both groups. l Control and Neo^Gli1CreER-CKO mice were subjected to sucrose preference test. The ratio of sucrose verse water consumed was quantified. Data are presented as the mean ± SEM. (n ≥ 9 per genotype). *P < 0.05; **P < 0.01. Student’s t-test. (m) Model of neogenin’s functions in adult hippocampal neurogenesis