Ethanol changes Nestin-promoter induced neural stem cells to disturb newborn dendritic spine remodeling in the hippocampus of mice

Abstract

Adolescent binge drinking leads to long-lasting disorders of the adult central nervous system, particularly aberrant hippocampal neurogenesis. In this study, we applied in vivo fluorescent tracing using NestinCreER^T2::Rosa26-tdTomato mice and analyzed the endogenous neurogenesis lineage progression of neural stem cells (NSCs) and dendritic spine formation of newborn neurons in the subgranular zone of the dentate gyrus. We found abnormal orientation of tamoxifen-induced tdTomato⁺ (tdTom⁺) NSCs in adult mice 2 months after treatment with EtOH (5.0 g/kg, i.p.) for 7 consecutive days. EtOH markedly inhibited tdTom⁺ NSCs activation and hippocampal neurogenesis in mouse dentate gyrus from adolescence to adulthood. EtOH (100 mM) also significantly inhibited the proliferation to 39.2% and differentiation of primary NSCs in vitro. Adult mice exposed to EtOH also exhibited marked inhibitions in dendritic spine growth and newborn neuron maturation in the dentate gyrus, which was partially reversed by voluntary running or inhibition of the mammalian target of rapamycin-enhancer of zeste homolog 2 pathway. In vivo tracing revealed that EtOH induced abnormal orientation of tdTom⁺ NSCs and spatial misposition defects of newborn neurons, thus causing the disturbance of hippocampal neurogenesis and dendritic spine remodeling in mice.

Introduction

Adolescent binge drinking can cause persistence of adolescent-like synaptic physiology, abnormal neurogenesis and behavior, and sensitivity to alcohol into adulthood (Alaux-Cantin et al., 2013; Mathews et al., 2016; Ji et al., 2018; Spear, 2018; Kwon et al., 2022). It also reduces the size of the adult basal forebrain and olfactory bulb (Coleman et al., 2011). Acute ethanol (EtOH) induces different effects on neural and behavioral function during adolescence and adulthood, with adolescent rats showing greater memory impairment than adults in the Morris water maze and on an odor discrimination task (Land and Spear, 2004). During adolescence, hippocampal neurogenesis has more of a developmental function, whereas in adulthood, it has more of a homeostatic function (Sahay and Hen, 2007; Rolando and Taylor, 2014). Some studies have shown that EtOH exposure inhibits the proliferation of neural stem cells (NSCs) and neurogenesis in the hippocampus of adolescent mice, as judged by bromodeoxyuridine (BrdU) and proliferating-cell nuclear antigen staining (Anderson et al., 2012; Campbell et al., 2014; Geil et al., 2014; Squeglia et al., 2014). Although the hippocampus is susceptible to alcohol neurotoxicity, it also recovers with abstinence (Bartels et al., 2007; Wilson et al., 2017), and similar neurotoxicity and recovery patterns can be seen in the effects of alcohol on adult hippocampal neurogenesis (Nixon and Crews, 2002; Taffe et al., 2010; Golub et al., 2015). Neuroimmune responses, neuroinflammation, and/or epigenetics could be related to these effects (Vetreno and Crews, 2015; Vetreno et al., 2018; Dobs and Ali, 2019). Nevertheless, how binge drinking affects dentate lineage progression and newborn dendritic spine remodeling remains unclear.

Voluntary running can increase hippocampal neurogenesis via activity-dependent increases in differentiation and survival of NSCs (Vetreno et al., 2016, 2018, 2020). This activity thus improves recovery of adult hippocampal plasticity and is accompanied by abrogation of memory decline in adult mice (Wong-Goodrich et al., 2010). Voluntary exercise has been reported to promote an increase of newborn neurons in the dentate gyrus (DG) and increase the hippocampal expression of brain-derived vascular endothelial growth factor (VEGF) and insulin-like growth factor 1 (IGF-1) (Peng et al., 2023). It also induces adult hippocampal neurogenesis in a rodent model of fetal alcohol spectrum disorders (Boehme et al., 2011). Furthermore, EtOH increased the translocation of the mammalian target of rapamycin (mTOR) into lysosomes, resulting in mTOR overactivation (Mazan-Mamczarz et al., 2015; Chao et al., 2018). Hyperactivity of mTOR has been shown to elevate phosphorylation of ribosomal protein S6 (pS6) (Feliciano et al., 2012), while the mTOR inhibitor rapamycin rescued defective neurogenesis and spine pruning in adolescent mice (Montesinos et al., 2018; Sun et al., 2023). However, whether voluntary running or inhibition of mTOR can rescue the aberrant hippocampal neurogenesis induced by EtOH remains unknown.

The current study used in vivo fluorescent tracing in NestinCreER^T2::Rosa26-tdTomato mice to identify the effects of EtOH on the endogenous neurogenic lineage progression of NSCs and on dendritic spine remodeling of newborn neurons in the DG. Additionally, we aimed to determine whether voluntary running or inhibition of the mTOR-enhancer of zeste homolog 2 (EZH2) pathway ameliorates these impairments in the mouse hippocampus.

Methods

Animals

The NestinCreER^T2::Rosa26-tdTomato mice were generated by crossing NestinCreER^T2 mice (Shanghai Jiao Tong University, China) with homozygous/heterozygous CAG tdTomato mice (B6.Cg-Gt(ROSA)26Sor^{tm14(CAG-tdTomato)Hze}/J, The Jackson Laboratory, Bar Harbor, ME, USA, RRID: IMSR_JAX:007914). Genotyping of the transgenic mice at 4 weeks showed that approximately 50% were NestinCreER^T2 positive and 75% were tdTomato positive. Primers used for mouse genotyping are listed in Table 1. The transgenic mice were then backcrossed with C57BL/6J once per 6 months. C57BL/6J mice at the ages of 3–4 and 7 weeks (obtained from Beijing HFK Bioscience Co., Beijing, China, license No. SCXY (Jing) 2020-0004) were accommodated for 1 week before enrolment in the experiments. The age of adolescent and adult mice used in this study were 4 and 8 weeks old, respectively. Mice were housed in a specified pathogen-free room at a maximum of five mice per cage, with a 12-hour light/dark cycle and food and water ad libitum. All animal procedures applied in this study were performed in accordance with protocols approved by the Experimental Animal Research Committee of Shenyang Pharmaceutical University (approval No. SYPU-IACUC-S2020-11.16-104, and No. SYPU-IACUC-S2020-12.16-105 on November 12, 2020 and December 12, 2020) and conformed to the National Research Council Guide for the Care and Use of Laboratory Animals (National Research Council, 2011). The animals used in the experiment were randomly divided into different groups.

Table 1.

Primers for genotyping

Oligonucleotides	Primer sequences
NestinCreERT2 mouse genotyping	Forward: 5′-GCC TGC ATT ACC GGT CGA TGC-3′
	Reverse: 5′-CAG GGT GTT ATA AGC AAT CCC-3′
tdTomato mouse genotyping	oIMR 9020: 5′-AAG GGA GCT GCA GTG GAG TA-3′
	oIMR 9021: 5′-CCG AAA ATC TGT GGG AAG TC-3′
	oIMR 9103: 5′-GGC ATT AAA GCA GCG TAT CC-3′
	oIMR 9105: 5′-CTG TTC CTG TAC GGC ATG G-3′
Actin	Forward: 5′-CAC CGG AGA ATG GGA AGC CGA-3′
	Reverse: 5′-TCC ACA CAG ATG GAG CGT CCA-3′

All the primers were from the Jackson Laboratory (Bar Harbor, ME, USA).

Experimental schedule

Fate mapping was performed in NestinCreER^T2::Rosa26-tdTomato mice, in which Tamoxifen (TAM)-induced NSC-specific expression of tdTomato facilitated in vivo dentate NSC tracing in the DG. EtOH (5.0 g/kg, intraperitoneally (i.p.)) was given to 8-week-old mice for 7 consecutive days and TAM (100 mg/kg, i.p.) was injected to induce tdTomato expression. Tissue was collected and analyzed 1 and 2 months after the first TAM injection (Figure 1A and B).

Figure 1.

EtOH induces abnormal activation of tdTom⁺ NSCs in the DG.

(A, B) Schematic diagrams of the experimental design. EtOH (5.0 g/kg, i.p.) was given to mice and TAM (100 mg/kg, i.p.) was injected to induce tdTomato expression. (C) Representative confocal images of DAPI⁺tdTom⁺ NCs (DAPI, blue; tdTom, red). Scale bars: 200 μm. (D) Quantification of DAPI⁺tdTom⁺ NCs. (E, F) Representative confocal images of tdTom⁺MCM2⁺ PCs, GFAP⁺MCM2⁺ aNSCs, and GFAP⁺tdTom⁺ NSCs (GFAP, green, Alexa Fluor 488; MCM2, white, Alexa Fluor 647). The boxed regions are shown at a higher magnification in the right panels. Scale bars: 100 μm. (G, H, J) Quantifications of GFAP⁺tdTom⁺ NSCs (G), GFAP⁺tdTom⁺MCM2⁺ aNSCs (H), and tdTom⁺MCM2⁺ PCs (J). (I, K, L) The proportions of GFAP⁺tdTom⁺ NSCs (I), tdTom⁺MCM2⁺ PCs (K), and GFAP⁺tdTom⁺MCM2⁺ aNSCs (L) in total tdTom⁺ NCs. Data are presented as mean ± SEM (n = 5 mice/group). *P < 0.05, **P < 0.01, ***P < 0.001, vs. 1 month Ctrl group; ###P < 0.001 (two-way analysis of variance with Fisher’s least significant difference test). 1M and 2M represent the analysis, which was done 1 or 2 months after the last EtOH administration. aNSCs: Activated NSCs; Ctrl: control; DAPI: 4′,6-diamidino-2-phenylindole; DG: dentate gyrus; EtOH: ethanol; GFAP: glial fibrillary acidic protein; MCM2: minichromosome maintenance protein 2; NCs: neural cells; NSCs: neural stem cells; PCs: proliferative cells; tdTom⁺: tdTomato⁺.

EtOH treatment

For the in vivo experiment, EtOH (Tianjin Yongda Chemical Reagents, Tianjin, China, Cat# 64-17-5) was dissolved in normal saline at a stock concentration of 20% (v/v) and was administered (5.0 g/kg, i.p.) once per day for 7 consecutive days. For the in vitro experiment, EtOH (100 mM) was prepared in 1‰ dimethyl sulfoxide. NSCs were treated with EtOH 6 hours after plating the cells. Cells were harvested after 3 days for analysis.

Tamoxifen administration

To induce tdTomato fluorescence in the CreER/LoxP system of NestinCreER^T2::Rosa26-tdTomato mice, TAM (Merck, Darmstadt, Germany, T5648) was prepared in a 9:1 corn oil (Merck, C8267) to EtOH mixture at a stock concentration of 10 mg/mL (Feng et al., 2013). TAM (100 mg/kg, i.p.) was administered to mice twice per day for 5 consecutive days, and mice were analyzed 1 or 2 months after the first TAM injection (Zhou et al., 2018).

Rapamycin administration

To inhibit the overactivation of mTOR, rapamycin (APExBIO, Houston, TX, USA, A8167) was first dissolved in EtOH at a stock concentration of 25 mg/mL and was further diluted to a final concentration of 1 mg/mL in 4% EtOH, 5% Tween 80, and 5% polyethylene glycol 400 (Zhou et al., 2018). Rapamycin (20 mg/kg, i.p.) was delivered to mice once every 2 days for a total of five doses.

5’-Bromo-2’-deoxyuridine administration

For analysis of cell maintenance in the hippocampal DG (Sun et al., 2015), BrdU (Merck, B5002) was dissolved in normal saline at a stock concentration of 10 mg/mL and was injected (50 mg/kg, i.p.) into mice once per day for 7 consecutive days. Mice were analyzed 14 days after the first 5′-bromo-2′-deoxyuridine (BrdU) injection (Zhou et al., 2018). To investigate EtOH-induced changes in cell proliferation, pulse BrdU (300 mg/kg, i.p.) was injected only once 2 hours before mice were sacrificed (Nixon et al., 2008). Quantification of BrdU⁺ cells within the subgranular zone (SGZ) and granule cell layer (GCL) was carried out as previously described (Zhou et al., 2018).

Voluntary running

Mice were divided randomly and put into cages with running wheel (running cages) or not (sedentary cages) at the same time. In the running cages, mice were free to access to the running wheel (Shandong Xinhua, Zibo, Shandong, China) for 1 month (van Praag et al., 1999; Feng et al., 2013) after EtOH or saline injection. During the first 5 days in running cages, the animals were injected with TAM twice per day. The animals were sacrificed 1 month after the first TAM injection (Naylor et al., 2008).

Primary cell isolation and culture

NSCs were isolated from the DG of 3–4-week-old C57BL/6J mice based on the published method (n = 6/experiment, at least three independent experiments, half male, half female, specified pathogen-free level) (Lupatov et al., 2017). Mice were anesthetized with 2.0–2.5% (v/v) isoflurane (RWD Life Science, Shenzhen, China, Cat# R510-22-10) and euthanized by inhalation of 70–100% excess carbon dioxide. Cells from the second passage were resuspended in NSC growth medium: Dulbecco’s modified Eagle medium/Nutrient Mixture F-12 (DMEM/F12, Gibco, Waltham, MA, USA, 11330-032) supplemented with 2% B27 (Gibco, 12587-010), 1.3% glucose (Merck, G8769), 0.5% 4-(2-hydroxyethyl) piperazine-1-ethanesulfonic acid (HEPES, Gibco, 15630-106), 0.1% progesterone (Merck, 5341-25GM), 1% putrescine (Merck, P5780), and 0.1% ITS liquid media supplement (ITSS, Merck, I3146). The following growth factors were freshly added before the experiment: 20 ng/mL epidermal growth factor (EGF, Merck, E4127), 10 ng/mL fibroblast growth factor-2 (FGF-2, R&D, Minneapolis, MN, USA, 233-FB-025/CF) and 1.83 mg/μL Heparin (Merck, H3149). For passaging, the neurospheres were treated with Accutase (Merck, A6964), and after 3 days they were counted and harvested. For the monolayer culture, the NSCs were seeded on poly-L-lysine (Merck, P2636) and laminin (Roche, Basel, Switzerland, 11243217001)-coated coverslips. For NSC differentiation, the NSCs were first cultured for 3 days in NSC medium with 2% fetal bovine serum (Gibco, 10099141), 5 ng/mL FGF-2, and no EGF. Then, the cells were further cultured in NSC medium without either EGF or FGF-2 for 11 days. The medium was renewed every 3 days throughout the procedure.

Sample processing and immunofluorescence

After 1 or 2 months of EtOH treatment, mice were anesthetized with 2.0–2.5% (v/v) isoflurane for 3–5 minutes by an animal anesthesia ventilator system (RWD Life Science). Mice were then perfused with 4% paraformaldehyde, and whole brains were postfixed overnight at 4°C. Coronal sections (40 μm) were prepared with a vibratome (Leica, Wetzlar, Germany, 1200). Sections were blocked and permeabilized in 5% normal swine serum (Solarbio, Beijing, China, S9060) in PBST (phosphate buffered saline [PBS] + 0.2% Triton X-100) for 1 hour before being incubated overnight at 4°C with primary antibodies. The slices were then washed three times (10 minutes each) with PBST and incubated with corresponding secondary antibodies for 1 hour at room temperature (20 ± 5°C) in darkness. A series of six sections from the DG (anterior, middle, and posterior) were analyzed from each mouse.

For the immunofluorescence analysis, cells were fixed with 3% paraformaldehyde and 4% sucrose for 15 minutes at room temperature and then permeabilized with 0.5% NP-40 (Diamond, Shanghai, China, A100109-0100) in PBS for 5 minutes at room temperature. The samples were then incubated with primary antibodies at 4°C overnight. Next, the cells were washed twice (10 minutes each) with PBS and incubated with corresponding secondary antibodies for 1 hour at room temperature in darkness.

We used the following primary antibodies: Nestin (mouse, 1:500, Merck, Cat# MAB353, RRID: AB_94911), glial fibrillary acidic protein (GFAP; chicken, 1:1500, Abcam, Cambridge, UK, Cat# Ab4674, RRID: AB_304558), anti-beta III Tubulin (TuJ1; mouse, 1:1500, Abcam, Cat# Ab78078, RRID: AB_2256751), Ki67 (mouse, 1:1000, Abcam, Cat# Ab8191, RRID: AB_306346), minichromosome maintenance protein 2 (MCM2; mouse, 1:1000, BD, Franklin Lakes, NJ, USA, Cat# 610701, RRID: AB_398024), doublecortin (DCX; rabbit, 1:1000, Abcam, Cat# Ab207175, RRID: AB_2894710), BrdU (rat, 1:500, Abcam, Cat# Ab6326, RRID: AB_305426), neuronal nuclei (NeuN; rabbit, 1:500, Merck, Cat# MAB377, RRID: AB_2298772), pS6 (rabbit, 1:400, CST, Danvers, MA, USA, Cat# 2211, RRID: AB_331679), trimethylated Lys 27 of histone 3 (H3K27me3; rabbit, 1:1000, CST, Cat# 9733, RRID: AB_2616029), trimethylated Lys 9 of histone 3 (H3K9me3; rabbit, 1:1000, CST, Cat# 13969, RRID: AB_2798355), trimethylated Lys 36 of histone 3 (H3K36me3; rabbit, 1:1000, CST, Cat# 4909, RRID: AB_1950412), trimethylated Lys 4 of histone 3 (H3K4me3; rabbit, 1:1000, CST, Cat# 9751, RRID: AB_2616028), and H3 (rabbit, 1:1000, CST, Cat# 9715, RRID:AB_331563). Secondary antibodies were as follows: donkey anti-mouse 647 (1:500, Abcam, Cat# Ab150107, RRID: AB_2890037), goat anti-chicken 488 (1:500, Abcam, Cat# Ab150169, RRID: AB_2636803), donkey anti-rabbit 647 (1:500, Abcam, Cat# Ab150075, RRID: AB_2752244), donkey anti-rabbit 488 (1:500, Abcam, Cat# Ab150073, RRID: AB_2636877), donkey anti-mouse 488 (1:500, Abcam, Cat# Ab150105, RRID: AB_2732856), and donkey anti-mouse 555 (1:500, Abcam, Cat# Ab150106, RRID: AB_2857373).

Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI; Beyotime, Shanghai, China, C1006). Fluorescent images were taken by confocal laser-scanning microscope (Nikon A1R Plus, Nikon, Tokyo, Japan). ImageJ software (Version 1.52e; National Institutes of Health, Bethesda, MD, USA) (Schneider et al., 2012) with the Nikon ND2 reader plugin was used for analyzing the positive cells in the defined area of the hippocampus.

Western blotting

Proteins from cultured NSCs were separated by 10–12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and electro-transferred to polyvinylidene fluoride membranes. The membranes were incubated in the primary antibodies (H3K27me3, H3K9me3, H3K36me3, H3K4me3, or H3: rabbit, 1:1000, CST) overnight at 4°C and incubated in the secondary antibody (Goat anti-Rabbit IgG (H+L), 1:5000, Thermo Scientific, Waltham, MA, USA, Cat# 31460, RRID: AB_228341) at room temperature for 1 hour. Immunoreactive bands were detected by incubation with chemiluminescent horseradish peroxidase substrates and exposed in a FluorChem E imager (PerkinElmer. Inc, Wellesley, MA, USA, NEL105001EA). The grey values of the protein bands were determined using ImageJ software and normalized to their respective controls.

Quantitative polymerase chain reaction

Total RNA from cultured NSCs was isolated using a Trizol reagent (Thermo Scientific, K0731), and cDNA was synthesized using a Reverse Transcription Kit (Takara, Kyoto, Japan, RR047A) according to the manufacturer’s instructions. cDNA was amplified using SYBR Green PCR mix (Thermo Scientific, 25743) and quantitative gene expression analysis was run in a real-time thermal cycler (CFX Opus 96, Bio-Rad, Hercules, CA, USA). The RNA expression levels were normalized to Gapdh. The relative levels of mRNA were calculated using the comparative 2^–ΔΔCt method (Jozefczuk and Adjaye, 2011). Primers are listed in Table 2.

Table 2.

Primers for quantitative polymerase chain reaction analyses

Gene	Forward primer (5’–3’)	Reverse primer (5’–3’)
Nes	TTG CAG ACA CCT GGA AGA AG	CCT CTG GTA TCC CAA GGA AA
Pax6	ATG AGA GTA GCG ACT CCA GAA GTT G	ATG CGA CTT CAG CTG AAG CGG
Rbfox3	GGC AAA TGT TCG GGC AAT TCG	TCA ATT TTC CGT CCC TCT ACG AT
Dcx	TTT GGA CAT TTT GAC GAA CGA GA	GTG GGC ACT ATG AGT GGG AC
Sox4	GAC AGC GAC AAG ATT CCG TTC	GTT GCC CGA CTT CAC CTT C
Sox9	CGG AAC AGA CTC ACA TCT CTC C	GCT TGC ACG TCG GTT TTG G
Ezh1	TCA AAC GGC TCC AGG CAA ATA	GGA CAC GAA GTT TCT TCC ACT C
Ezh2	CCC TGA CCT CTG TCT TAC TTG TGG A	ACG TCA GAT GGT GCC AGC AAT A
Suz12	TAT GGT GAA GTC CTA CTC GTT GC	GCT GTA AAC GCC TGT TTT TGT C
Bmi1	AAA TCC CCA CTT AAT GTG TGT CC	CTT GCT GGT CTC CAA GTA ACG
Gapdh	CCC ATC ACC ATC TTC CAG GAG	CTT CTC CAT GGT GGT GAA GAC A

Bmi1: B lymphoma Mo-MLV insertion region 1 homolog; Dcx: doublecortin; Ezh1: enhancer of zeste homolog 1; Ezh2: enhancer of zeste homolog 2; Gapdh: glyceraldehyde-3-phosphate dehydrogenase; Nes: nestin; Pax6: paired box gene 6; Rbfox3: RNA binding fox-1 homolog 3; Sox4: SRY-related HMG-box 4; Sox9: SRY-related HMG-box 9; Suz12: suppressor of zeste-12 protein.

Imaging and analysis

Fluorescence images were captured using a confocal laser-scanning microscope (Nikon A1R Plus) with 20× (0.31 pixel) and 60× (0.10 pixel) objective lenses. tdTom⁺ newborn granule cells were imaged with Z-stack scanning at 0.5-μm intervals by confocal microscopy. To analyze the dendritic structure of tdTom⁺ neurons, reconstructions of the dendritic processes were made from the Z stacks. The projection images were semiautomatically traced by ImageJ software with the Neuron J plugin. Images of dendritic spines were deconvoluted with Nikon Software (NIS-Elements AR, Nikon). The total dendritic length and branch number of each tdTom⁺ neuron were subsequently used for analysis. The Sholl analysis for dendritic complexity was carried out by counting the number of traced dendrites that crossed a series of concentric circles which were located at 10-μm intervals from the cell soma (Feng et al., 2013). tdTom⁺ dendritic spines were classified into four groups: thin, stubby, mushroom, and branched. These groups were established by the following criteria. Thin: 1 μm < length ≤ 5 μm and neck width/head width < 1.5 μm; stubby: length ≤ 1 μm and neck width/head width < 1.5 μm; mushroom: neck width/head width ≥ 1.5 and length ≤ 5 μm; branched: 2 or more heads (Swanger et al., 2011; Bian et al., 2015). A series of six sections from the bilateral hippocampus (anterior, middle, and posterior) of each mouse were analyzed. The number of positive cells was counted by ImageJ. The Mander’s coefficients of the integrated optical densities were analyzed by Image-Pro Plus 6.0 software (Media Cybernetics, Silver Spring, MD, USA).

Assessment of tdTom⁺ cell orientation

A compass mapping method was performed to determine the orientation of tdTom⁺ cells in the SGZ (Naylor et al., 2008). The normal leading process of a tdTom⁺ cell projects vertically through the GCL into the molecular layer. The angle of a perfect process leaving the cell body perpendicular to the SGZ was defined as 0°. We measured the leading process of at least 80 tdTom⁺ cells in the bilateral hippocampus from six brain slices (anterior, middle, and posterior) from each mouse and calculated the average orientation of all the tdTom⁺ cells (growing upwards perpendicular to SGZ in neat rows).

Statistical analysis

All experimental results are reported as mean ± standard error of mean (SEM). All the graphs were made using GraphPad Prism 8.4.0 (GraphPad Software, San Diego, CA, USA, www.graphpad.com). Data from control (Ctrl) + running and EtOH + running groups were analyzed by Student’s t-test. Data from vehicle or EtOH treatment for 1 or 2 months were analyzed by two-way analysis of variance (ANOVA) followed by Fisher’s least significant difference test for multiple comparisons. The total dendritic length and the total branch number of newborn neurons were mapped by cumulative distribution and the Kolmogorov-Smirnov test. Outliers were not tested for, and no data were excluded from the analysis. The level of statistical significance was set at P < 0.05. The optimal sample size was not calculated in this study. Instead, the numbers of animals for the experiments were chosen based on the numbers used in similar published experiments, including our own experiments (Li et al., 2015; Vetreno et al., 2017; Zhou et al., 2018; Liu et al., 2021).

Results

EtOH induces abnormal activation of tdTom⁺ NSCs in the DG

For these tests, we compared control mice with those that received EtOH treatment for 7 days. Genotyping showed that the NestinCreER^T2::Rosa26-tdTomato mice were successfully constructed (Additional Figure 1 ^{(412.8KB, tif)} ). A decrease (28.5 ± 2.8%) in the density of tdTom⁺ neural cells (NCs) was observed 1 month after EtOH treatment. The density had decreased further (37.3 ± 1.6% overall decrease) by 2 months after EtOH treatment (Figure 1C and D, two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 7.514, P = 0.0254). We examined whether EtOH disturbed the lineage progression of NSCs in adult DG. Fate mapping revealed a significant EtOH-induced decrease in the generation of GFAP⁺tdTom⁺ NSCs, and GFAP⁺tdTom⁺MCM2⁺ activated NSCs (aNSCs) at 1 month after EtOH treatment. At 2 months after EtOH treatment, the reduction was more obvious for GFAP⁺tdTom⁺ NSCs (decreased by 52.0 ± 1.4%) (two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 64.23, P < 0.0001) than for GFAP⁺tdTom⁺MCM2⁺ aNSCs (decreased by 43.8 ± 1.2%) (two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 18.33, P = 0.0027) (Figure 1E–H). The percentage of GFAP⁺tdTom⁺ NSCs among tdTom+ NCs did not differ between EtOH-treated mice and controls at 1 month after treatment, but was significantly lower after 2 months (controls: 26.8 ± 1.3%; EtOH: 20.5 ± 1.7%) (Figure 1I, two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 35.83, P = 0.0003). The density of MCM2⁺tdTom⁺ proliferative cells (PCs) was lower 2 months after EtOH injections than it was in the control group (Figure 1J, two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 0.01340, P = 0.9107). The percentage of MCM2⁺tdTom⁺ PCs among tdTom+ NCs did not differ between controls and EtOH-treated mice at 2 months after treatment (Figure 1K, two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 16.21, P = 0.0038). Meanwhile, in these mice, there was no decrease in the proportion of aNSCs in tdTom+ NCs (Figure 1L, two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 0.06852, P = 0.8001).

NSCs have the capacity for self-renewal and activation (Lu et al., 2017). In primary culture of NSCs, noticeable reductions in Nestin⁺ NSCs (controls: 82.4 ± 1.7%, EtOH: 63.1 ± 1.8%) and Ki67⁺ PCs (controls: 76.2 ± 2.2%, EtOH: 39.8 ± 1.4%) were induced by EtOH (100 mM). Quantification of Nestin⁺Ki67⁺ aNSCs among Nestin⁺ NSCs showed a marked decrease of aNSCs in EtOH-treated mice (Additional Figure 2A ^{(419.9KB, tif)} –D ^{(419.9KB, tif)} ). Moreover, a significant decrease was observed in the mRNA levels of the NSC marker Nes and the progenitor marker paired box gene 6 (Pax6) after EtOH treatment (Additional Figure 2E ^{(419.9KB, tif)} ). SRY-related HMG-box 4 (Sox4) helps determine neuronal properties (Bergsland et al., 2006; Mu et al., 2012). Sox9 is reported to be crucial for the maintenance of multipotent NSCs (Scott et al., 2010). We discovered that the mRNA levels of both Sox4 and Sox9 were markedly decreased in primary NSCs (Additional Figure 2F ^{(419.9KB, tif)} and G ^{(419.9KB, tif)} ). These data therefore suggest that EtOH inhibits the self-renewal and activation of primary NSCs, thus leading to defects in NSC differentiation into neurons. In addition, we also evaluated EtOH-induced proliferative changes with BrdU labeling. The results showed that while EtOH significantly reduced BrdU⁺ PCs in the DG of both adolescent and adult mice, it was more obvious in adolescents (24.4 ± 4.7% in adulthood, 49.5 ± 10.4% in adolescence, Additional Figure 2H ^{(419.9KB, tif)} –P ^{(419.9KB, tif)} ).

EtOH induces the blockade of dentate newborn neurons maturation

Developing neuroblasts (NBs) are particularly vulnerable to EtOH-induced cell death shortly after neural tube closure (Dan et al., 2014). To determine whether EtOH negatively affects hippocampal neurogenesis, we examined the production of NBs in the DGs of control and EtOH-treated mice. Although the production of DCX⁺tdTom⁺ NBs in the control group increased sharply from 4780 ± 676 cells/mm³ at 1 month to 12,919 ± 688 cells/mm³ at 2 months, it was significantly lower after both 1 and 2 months of EtOH treatment (two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 7.445, P = 0.0259; Figure 2A–C). In contrast, the proportion of DCX⁺tdTom⁺ NBs among all tdTom⁺ NCs was significantly greater 1 month after EtOH treatment, and this increase was maintained for 2 months (two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 0.01327, P = 0.9111; Figure 2D). Moreover, compared with the control group, the density of DCX⁺tdTom⁺MCM2⁺ proliferative NBs declined considerably in the EtOH group, especially at 2 months after EtOH exposure (two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 9.170, P = 0.0164; Figure 2E). However, the proportion of DCX⁺tdTom⁺MCM2⁺ proliferative NBs among all tdTom⁺ NCs remained unchanged (two-way ANOVA, interaction of treatment × times, F_{(1, 8)} = 6.193, P = 0.0376; Figure 2F). The specific patterns of EtOH-induced NSC lineage progression in the DG were then summarized. The cell types of tdTom⁺ cells in DG were analyzed, including GFAP⁺tdTom⁺ NSCs, DCX⁺tdTom⁺ immature neurons (IMNs), tdTom⁺ mature neurons (MNs) with neuronal morphology and located in GCL and remaining tdTom⁺ neural progenitor cells. At both 1 and 2 months after EtOH exposure, there were substantial increases in the proportions of neural progenitor cells and IMNs, which contrasted to an obvious decline in the proportion of MNs among tdTom⁺ new NCs (two-way ANOVA, interaction of groups × types, F_{(9, 48)} = 55.32, P < 0.0001; Figure 2G and H). Thus, EtOH inactivates NSCs, which results in the accumulation of neural progenitor cells, which in turn prevents IMNs from maturing into neurons in the DG. This blockade can be persistent and long-lasting, continuing up to 2 months after EtOH exposure.

Figure 2.

EtOH induces the blockade of dentate newborn neuron maturation.

(A, B) Representative confocal images of DCX⁺tdTom⁺MCM2⁺ tdTom⁺ PNBs, DCX⁺MCM2⁺ PNBs, tdTom⁺MCM2⁺ PCs, and tdTom⁺DCX⁺ NBs (DCX, green, Alexa Fluor 488; tdTom, red; MCM2, white, Alexa Fluor 647). The four differently shaped arrowheads point to tdTom⁺ PNBs, PNBs, PCs, and NBs, as shown in the right panels. Scale bars: 50 μm. (C, E) Quantifications of DCX⁺tdTom⁺ NBs (C) and DCX⁺tdTom⁺MCM2⁺ tdTom⁺ PNBs (E). (D, F) The proportions of DCX⁺tdTom⁺ NBs (D) and DCX⁺tdTom⁺MCM2⁺ tdTom⁺ PNBs (F) in total tdTom⁺ NCs. (G) Representative confocal images of tdTom⁺ NCs. Boxed areas are magnified on the right. Scale bars: 40 μm. (H) Lineage progression analysis of NSCs, as indicated by quantification of tdTom⁺ NCs in the DG. Data are presented as mean ± SEM (number of animals, n = 5/group). **P < 0.01, ***P < 0.001, vs. 1 month Ctrl group; ##P < 0.01, ###P < 0.001 (two-way analysis of variance with Fisher’s least significant difference test). 1M and 2M represent the analysis, which was done 1 or 2 months after the last EtOH administration. Ctrl: Control; DCX: doublecortin; DG: dentate gyrus; EtOH: ethanol; MCM2: minichromosome maintenance protein 2; NBs: neuroblasts; NCs: neural cells; NSCs: neural stem cells; PCs: proliferative cells; PNBs: proliferative NBs; tdTom⁺: tdTomato⁺.

Next, we analyzed the numbers of neurons and astrocytes that differentiated from NSCs in vitro after EtOH treatment. The percentage of TuJ1⁺ neurons decreased (controls: 17.8 ± 0.8%, EtOH: 12.6 ± 1.0%), while GFAP⁺ astrocytes did not change significantly (Additional Figure 3A ^{(162.4KB, tif)} –C ^{(162.4KB, tif)} ). Quantitative PCR analysis indicated a significant decrease in expression of the NB marker Dcx and the neuron marker RNA binding fox-1 homolog 3 (Rbfox3) after EtOH treatment (Additional Figure 3D ^{(162.4KB, tif)} ). The mRNA levels of Sox4 and Sox9 were also significantly reduced in neurons (Additional Figure 3E ^{(162.4KB, tif)} and F ^{(162.4KB, tif)} ). Therefore, these data suggest that EtOH prevents the differentiation and maturation of NSCs into neurons within the DG.

EtOH disturbs the development of dentate BrdU⁺ neurons from adolescence to adulthood

To further explore how EtOH affects the development of BrdU⁺ neurons, adult (8-week-old) mice were exposed to EtOH for 7 days and then injected with BrdU (50 mg/kg, i.p.) once daily for 7 days, and analyzed 2 weeks after the first BrdU injection (Figure 3A). We found that EtOH induced a significant increase in DCX⁺NeuN^– NBs, accompanied by marked reductions of more developed DCX⁺NeuN⁺ IMNs and DCX^–NeuN⁺ MNs (Figure 3B–F). The pie charts in Figure 3G show a similar pattern. Compared with controls, a dramatic increase in the percentage of DCX⁺NeuN^– NBs was observed (controls: 23.2%, EtOH: 39.4%), with concurrent decreases in the percentages of more developed DCX⁺NeuN⁺ IMNs (controls: 55.2%; EtOH: 44.2%) and DCX^–NeuN⁺ MNs (controls: 14.3%; EtOH: 10.0%) among all BrdU⁺ neuronal cells after EtOH treatment (Figure 3G). Thus, EtOH disrupts the maturation of BrdU⁺ neurons in the DG of adult mice.

Figure 3.

EtOH disturbs the development of dentate BrdU+ neurons from adolescence to adulthood.

(A) Schematic diagram of the experimental procedure in adult mice. (B) Representative confocal images of EtOH or vehicle treatment. The four differently shaped arrowheads point to BrdU⁺DCX^–NeuN^– non-neurons (NNs), BrdU⁺DCX⁺NeuN^– NBs, BrdU⁺DCX⁺NeuN⁺ IMNs, and BrdU⁺DCX^–NeuN⁺ MNs, as shown in the right panels (DCX, green, Alexa Fluor 488; BrdU, red, Alexa Fluor 555; NeuN, blue, Alexa Fluor 647). Scale bars: 20 μm. (C–F) Quantification of total BrdU⁺ cells (C), NBs (D), IMNs (E), and MNs (F) in the DG of EtOH-treated and control (Ctrl) adult mice. (G) Pie charts showing the proportions of the different cell types in the total population of BrdU⁺ cells. (H) Schematic diagram of the experimental procedure in adolescent mice. (I) Representative confocal images of EtOH and vehicle treatment. The four different arrowheads point to the different cell types as in B. Scale bars: 20 μm. (J–M) Quantifications of total BrdU⁺ cells (J), NBs (K), IMNs (L), and MNs (M) in the DG of EtOH-treated and Ctrl adult mice. (N) Pie charts showing the proportions of different cell types in the total BrdU⁺ cell population. Data are presented as mean ± SEM (number of animals, n = 6/group). *P < 0.05, **P < 0.01, ***P < 0.001, vs. Ctrl group (Student’s t-test). BrdU: Bromodeoxyuridine; Ctrl: control; DCX: doublecortin; DG: dentate gyrus; EtOH: ethanol; IMNs: immature neurons; MNs: mature neurons; NBs: neuroblasts; NeuN: neuronal nuclei; NNs: non-neurons.

Adolescent (4-week-old) mice were also injected with BrdU after EtOH exposure and analyzed as adults 2 weeks after the first BrdU injection, as described above (Figure 3H). Analysis showed that the percentage of DCX⁺NeuN^– NBs among BrdU⁺ cells in adolescent mice was also dramatically increased by EtOH treatment (Figure 3I–N). This reveals that exposure to EtOH in adolescence progressively disturbs the development of BrdU⁺ neurons, which persists into adulthood.

To explore whether EtOH further impairs the development of BrdU⁺ neurons, adult mice were also injected with BrdU as described above and analyzed 1 month after the first BrdU injection (Additional Figure 4A ^{(306.1KB, tif)} ). In accordance with the above findings, the EtOH-treated mice had a marked increase of DCX⁺NeuN^– NBs and a marked decrease of DCX^–NeuN⁺ MNs among BrdU⁺ neuronal cells (Additional Figure 4B ^{(306.1KB, tif)} –G ^{(306.1KB, tif)} ). Together, these results of long-term tracing revealed that EtOH further disrupts the maturation of BrdU⁺ neurons in the DG.

EtOH sustains to suppress temporal and spatial dendritic development and spine formation in dentate newborn neurons

To further characterize the development of newborn neurons in the DG, we examined the dendritic morphology of tdTom⁺ newborn neurons in vivo. We used a compass system (Naylor et al., 2008) to determine the angle of the leading process in relation to each cell’s position. The tdTom⁺ NSCs with dendritic clusters, which were located in the SGZ, were observed to be more randomized and disordered in EtOH-treated mice, and these defects were more severe 2 months after EtOH injection (Figures 1C, 4A and 4B, two-way ANOVA, interaction of treatment × times, F_{(1, 158)} = 0.2998, P = 0.5848). Furthermore, tdTom⁺ newborn neurons in control mice were mostly located at the proximal border of the GCL 1 month after TAM injection, and more soma of tdTom⁺ newborn neurons migrated into the distal border of the GCL at 2 months after TAM treatment. In EtOH-treated mice, no such migration was observed, and the orientation tuning of newborn neurons was markedly disrupted. These defects were more severe at 2 months after EtOH exposure (Figures 1C, 4A, and 4C). The results indicate that EtOH disturbs the spatial arborization of NSCs and newborn neurons in the DG. Sholl analysis revealed that EtOH significantly reduced both the total dendritic length and the number of dendritic branches in tdTom+ newborn neurons. Moreover, the effects were more severe at 2 months than at 1 month after EtOH exposure (Figure 4D–G). In addition, the data showed a significant decline in tdTom⁺ dendritic spine density in EtOH-treated mice (Figure 4H and I, two-way ANOVA, interaction of treatment × times, F_{(1, 158)} = 18.22, P < 0.0001). Furthermore, the proportions of mushroom and branched spines, which represent more mature spines (Risher et al., 2014), were sharply reduced at both 1 and 2 months after EtOH injection compared with controls (Figure 4J, two-way ANOVA, interaction of groups × types, F_{(9, 48)} = 39.25, P < 0.0001). Thus, EtOH leads to temporal and spatial dysregulation of multiple processes of neuronal maturation during adult hippocampal neurogenesis, ranging from dendritic outgrowth to dendritic spine formation in dentate newborn neurons of the DG.

Figure 4.

EtOH sustains to suppress temporal and spatial dendritic development and spine formation in dentate newborn neurons.

(A–C) Representative confocal image (A) and statistical analysis of tdTom⁺ NSCs in the SGZ (B, two-way analysis of variance with Fisher’s least significant difference test) and tdTom⁺ newborn neurons in the GCL (C, Student’s t-test). A mapping scheme was used to measure the orientation tuning and migration distance of tdTom⁺ NCs in the DG. (D) Representative confocal images of reconstructions of newborn neurons from tdTom⁺ cells in the GCL at 1 or 2 months after EtOH treatment. Scale bars: 20 μm. (E, F) Cumulative distribution plots of total dendrite length (E) and total numbers of branches (F) in the GCL (Kolmogorov-Smirnov test). (G) Sholl analysis of the dendritic complexity of tdTom⁺ newborn neurons in the GCL (the same groups of cells were analyzed as in E and F. (H) Representative confocal images of spines in tdTom⁺ newborn neurons in the DG. Scale bars: 10 μm. (I, J) Quantification of the numbers of spines per μm (I, two-way analysis of variance with Fisher’s least significant difference test) and the percentage of different types of spines in tdTom⁺ newborn neurons (J, two-way analysis of variance with Fisher’s least significant difference test). Data are presented as mean ± SEM (B, C, E, F: number of traced neurons: 1 month Ctrl group: n = 28, 1-month EtOH group: n = 27, 2 months Ctrl group: n = 32, 2 months EtOH group: n = 30; G, I, J: n = 5). *P < 0.05, **P < 0.01, ***P < 0.001, vs. 1 month Ctrl group; #P < 0.05, ##P < 0.01, ###P < 0.001. Ctrl: Control; DG: dentate gyrus; EtOH: ethanol; GCL: granule cell layer; NCs: neural cells; NSCs: neural stem cells; SGZ: subgranular zone; TAM: tamoxifen; tdTom⁺: tdTomato⁺.

Voluntary running ameliorates the negative effect of EtOH on the lineage progression of dentate NSCs into MNs

To explore whether voluntary running reverses the EtOH-induced dysregulation described above, we performed the same basic experiment as described above. However, in this case, mice in both EtOH and control groups were then put into running cages or control cages for 1 month (Figure 5A). After voluntary running, the density of tdTom⁺ NCs in control mice increased significantly from 20,848 ± 4568 cells/mm³ to 41,825 ± 3380 cells/mm³. These results are consistent with the published literature (Fuss et al., 2010; Ma et al., 2017). Importantly, the density of tdTom⁺ NCs in the running EtOH mice was considerably higher than that in the non-running EtOH mice, though it was still much lower than that in the running control mice (Figures 1D, 5B and 5C). Voluntary running also significantly increased the density of GFAP⁺tdTom⁺ NSCs in the control mice. The density of NSCs remained lower in the running EtOH mice than in the running control mice (Figures 1G, and 5D). Similar results were found for the density changes of MCM2⁺tdTom⁺ PCs and GFAP⁺MCM2⁺tdTom⁺ aNSCs in the control and EtOH mice that were allowed to run (Figure 5E and F). In parallel, the proportions of NSCs, PCs, and aNSCs among all NCs in the running mice were similar in the control and EtOH groups (Figure 5G–I). These findings indicate that voluntary running facilitates the activation of NSCs in the DG and partially rescues the defects induced by EtOH.

Figure 5.

Voluntary running ameliorates the negative effect of EtOH on the lineage progression of dentate NSCs into MNs.

(A) Schematic diagram of the design of the voluntary running experiment in mice. EtOH (5.0 g/kg, i.p.) and TAM (100 mg/kg, i.p.) were injected to induce tdTomato expression and voluntary running for 1 month. (B) Representative confocal images tdTom⁺MCM2⁺ PCs, GFAP⁺MCM2⁺ aNSCs, and GFAP+tdTom+ NSCs (GFAP, green, Alexa Fluor 488; tdTom, red; MCM2, white, Alexa Fluor 647). The boxed regions are shown at a higher magnification in the right panels. Scale bars: 100 μm. (C) Quantification of DAPI⁺tdTom⁺ NCs (Student’s t-test). (D–F) Quantification of NSCs (D), PCs (E), and aNSCs (F) (Student’s t-test). (G–I) Percentages of NSCs (G), PCs (H), and aNSCs (I) in total tdTom⁺ NCs (Student’s t-test). (J) Representative confocal images of DCX⁺tdTom⁺ NBs, tdTom⁺ NCs, and DCX⁺ NBs (DCX, green, Alexa Fluor 488; tdTom, red) for two groups (Ctrl⁺Running and EtOH⁺Running). The three differently shaped arrows point to tdTom⁺ NBs, NCs, and NBs, as indicated in the right panels. Scale bars: 50 μm. (K) Quantification of DCX⁺tdTom⁺ NBs (Student’s t-test). (L) Percentage of DCX⁺tdTom⁺ NBs in total tdTom⁺ NCs (Student’s t-test). (M) Representative confocal images of tdTom⁺ NCs. Scale bars: 55 μm. (N) Quantification of lineage progression of tdTom+ NCs in the DG (two-way ANOVA with Fisher’s least significant difference test). Data are presented as mean ± SEM (number of animals, n = 6). *P < 0.05, **P < 0.01, ***P < 0.001, vs. Ctrl + running group. aNSCs: Activated NSCs; Ctrl: control; DAPI: 4′,6-diamidino-2-phenylindole; DCX: doublecortin; DG: dentate gyrus; EtOH: ethanol; GFAP: glial fibrillary acidic protein; MCM2: minichromosome maintenance protein 2; MNs: mature neurons; NBs: neuroblasts; NCs: neural cells; NSCs: neural stem cells; PCs: proliferative cells; TAM: tamoxifen; tdTom⁺: tdTomato⁺.

We also examined the density of NBs after voluntary running. Quantification of DCX⁺tdTom⁺ NBs revealed an increase from 4780 ± 676 cells/mm³ in the non-running control mice to 14,058 ± 1401 cells/mm³ in the running control mice. Meanwhile, the density of NBs was also higher in the running EtOH group than in the non-running EtOH group (non-running EtOH group: 3316 ± 408 cells/mm³, running EtOH group: 10,205 ± 771 cells/mm³; Figures 2C and 5K). Nevertheless, even when exposed to running, NBs density remained lower in the EtOH mice than in the running control mice (Figure 5J and K). We also noticed that the proportion of NBs among all tdTom+ NCs did not change with running (Figure 5L). Interestingly, the distribution pattern of NSC lineage cell types was similar in the control and EtOH mice after running (Figure 5M and N, two-way ANOVA, interaction of groups × types, F_{(3, 30)} = 2.090, P = 0.1225). These results suggest that voluntary running can partially reverse the EtOH-induced deficiency in the production of newborn neurons in the DG.

To further explore whether voluntary running affects the EtOH-induced deficiency in the development of newborn neurons, the mice were given BrdU for long-term neuronal tracing. BrdU was injected once daily for 7 days and mice were analyzed 1 month after the first BrdU injection (Additional Figure 5A ^{(341.8KB, tif)} ). We discovered that when exposed to running, EtOH mice had significantly fewer BrdU⁺ cells and DCX^–NeuN⁺ MNs (Additional Figure 5B ^{(341.8KB, tif)} –F ^{(341.8KB, tif)} ) compared with the Ctrl mice. Interestingly, the proportion of different cell types was similar in both groups, indicating a similar process of dentate newborn neuron development (Additional Figure 5G ^{(341.8KB, tif)} ).

Thus, these findings reveal that voluntary running rescues the EtOH-induced defects in dentate NSC lineage distribution, but only partially reverses the decreased activation of NSCs and the production of newborn neurons.

Voluntary running suppresses the dendritic development and spine formation induced by EtOH in dentate newborn neurons

By Sholl analysis, we observed significant increases both in the total dendritic length and the branch number of tdTom⁺ newborn neurons after voluntary running in Ctrl mice. The complexity of dendrites of EtOH mice was still less than in Ctrl mice after running (Figures 4D–G and 6A–D). Control mice also exhibited a significant increase in the density of tdTom⁺ dendritic spines and the maturity of the spines (mushroom and branched spines) after voluntary running (Figures 4J and 6G, two-way ANOVA, interaction of groups × types, F_{(3, 30)} = 74.79, P < 0.0001). However, the running EtOH mice showed reduced tdTom+ dendritic spine density and maturation compared to running Ctrl mice (Figure 6E–G). These results suggest that voluntary running can promote the maturation of dendritic spines of DG newborn neurons, but can only partially reverse the inhibitory effects of EtOH.

Figure 6.

Voluntary running suppresses the dendritic development and spine formation induced by EtOH in dentate newborn neurons.

(A) Representative reconstructions, based on confocal images, of tdTom⁺ newborn neurons in the GCL of mice exposed to voluntary running. Scale bars: 20 μm. (B, C) Cumulative distribution plots of total dendrite length (B) and total number of branches (C) in the GCL (number of traced neurons: Ctrl + running: n = 32, EtOH + running: n = 34; Kolmogorov-Smirnov test). (D) Sholl analysis of the dendritic complexity of tdTom⁺ newborn neurons in the GCL (the same groups of cells were analyzed as in B and C). (E) Representative confocal images of spines in tdTom⁺ newborn neurons in the DG. Scale bars: 10 μm. (F, G) Quantification of the number of spines per μm (F, Student’s t-test) and percentage of different types of spines in tdTom⁺ newborn neurons (G, two-way analysis of variance with Fisher’s least significant difference test). Data are presented as mean ± SEM (number of animals, n = 6). *P < 0.05, **P < 0.01, ***P < 0.001, vs. Ctrl + running group. Ctrl: Control; DG: dentate gyrus; EtOH: ethanol; GCL: granule cell layer; tdTom⁺: tdTomato⁺.

Rapamycin alleviates the EtOH-induced retardation of dendritic spine maturation through inhibition of the mTOR-EZH2 pathway

To investigate whether EZH2 is involved in EtOH-induced inhibition of hippocampal neurogenesis, we first examined the levels of H3K27me3, a catalytic substrate for EZH2. In this study, the levels of H3K27me3 and H3K9me3 were decreased (Figure 7A–C), while the levels of H3K36me3 and H3K4me3 remained unchanged after EtOH (100 mM) induction in primary NSCs (Figure 7A, D and E). Similar results were found with respect to changes in H3K27me3 density in Nestin⁺ NSCs in vitro (Figure 7F and G). Meanwhile, the expression levels of enhancer of zeste homolog 1 (Ezh1), Ezh2, suppressor of zeste-12 protein (Suz12), and B lymphoma Mo-MLV insertion region 1 homolog (Bmi1) were down-regulated in NSCs and up-regulated in neurons of the EtOH group (Additional Figures 6A ^{(413.1KB, tif)} –H ^{(413.1KB, tif)} ). Immunostaining of H3K9me3, H3K36me3, and H3K4me3 in NSCs also confirmed the above findings in vitro (Additional Figure 6I ^{(413.1KB, tif)} –N ^{(413.1KB, tif)} ). These data suggest that EtOH inhibits the activation and differentiation of NSCs by regulating EZH2.

Figure 7.

Rapamycin alleviates the EtOH-induced retardation of dendritic spine maturation through inhibition of the mTOR-EZH2 pathway.

(A–E) Sample western blot images (A) and statistical analysis of primary NSC cultures exposed to EtOH (100 mM) or vehicle, including H3K27me3 (B), H3K9me3 (C), H3K36me3 (D) and H3K4me4 (E) (number of independent experiments, n = 4, *P < 0.05, vs. Ctrl group; Student’s t-test). (F) Sample confocal images of Nestin⁺ (green, Alexa Fluor 488), H3K27me3⁺ (red, Alexa Fluor 555), and DAPI (blue) cells. Scale bars: 45 μm. (G) Mander’s coefficients of colocation of H3K27me3 and DAPI (number of independent experiments, n = 6, about 1000 cells/group, ***P < 0.001, vs. Ctrl group; Student’s t-test). (H) Schematic diagram of the experiment to test the effect of the mTOR inhibitor rapamycin (Rap) treatment. (I, J) Sample confocal images after rapamycin treatment and EtOH injection along with quantification of H3K27me3⁺ (red, Alexa Fluor 488) and DAPI (blue) cells in the SGZ (number of animals, n = 5, **P < 0.01, ***P < 0.001, vs. Ctrl + Veh group; #P < 0.05; two-way analysis of variance with least significant difference test). Scale bars: 50 μm. (K, L) Representative confocal images after rapamycin treatment and EtOH injection along with quantification of pS6⁺ (red, Alexa Fluor 488) and DAPI (blue) cells in mouse DG (number of animals, n = 5, **P < 0.01, ***P < 0.001, vs. Ctrl + Veh group; ###P < 0.001; two-way analysis of variance with least significant difference test). Scale bars: 50 μm. (M) Representative confocal images of spines in tdTom⁺ newborn neurons in the DG for two groups (Ctrl+Rap and EtOH+Rap). Scale bars: 10 μm. (N, O) Quantification of the number of spines per μm (number of animals, n = 6, ***P < 0.001, vs. Ctrl + Rap group; Student’s t-test) and the percentage of different types of spines in tdTom⁺ newborn neurons (O, number of animals, n = 6, **P < 0.01, vs. Ctrl + Rap group; two-way analysis of variance with Fisher’s least significant difference test). Data are presented as mean ± SEM. Ctrl: Control; DAPI: 4′,6-diamidino-2-phenylindole; DG: dentate gyrus; EtOH: ethanol; EZH2: enhancer of zeste homolog 2; H3K27me3: trimethylated Lys 27 of histone 3; H3K36me3: trimethylated Lys 36 of histone 3; H3K4me3: trimethylated Lys 4 of histone 3; H3K9me3: trimethylated Lys 9 of histone 3; mTOR: the mammalian target of rapamycin; NSCs: neural stem cells; pS6: protein S6; Rap: rapamycin; SGZ: subgranular zone; TAM: tamoxifen; Veh: vehicle.

To explore whether inhibition of mTOR reverses EtOH-induced dysregulation of neurogenesis, we treated 8-week-old mice with the mTOR inhibitor rapamycin (20 mg/kg, i.p.) five times with a 2-day interval, overlapping with EtOH injection for 7 consecutive days, followed by TAM injection. After 1 month the tissue was collected (Figure 7H). Consistent with the above findings, mice exposed to EtOH (but not rapamycin) exhibited a significant decrease in H3K27me3 levels in the SGZ, while rapamycin treatment led to significantly greater H3K27me3 levels (Figure 7I and J, two-way ANOVA, interaction of treatment × times, F_{(1, 10)} = 450.1, P < 0.0001). Mice treated with EtOH (but not rapamycin) exhibited a significant increase in pS6⁺ cells in the DG, and rapamycin exposure significantly reduced this increase (Figure 7K and L, two-way ANOVA, interaction of treatment × times, F_{(1, 10)} = 521.8, P < 0.0001). Furthermore, tdTom⁺ dendritic spine density was considerably lower in the rapamycin-EtOH mice than in the rapamycin-control mice. At the same time, significant increases in the numbers of mushroom and branched spines were observed in the rapamycin-EtOH mice (two-way ANOVA, interaction of groups × types, F_{(3, 30)} = 61.92, P < 0.0001); nevertheless, the spines in the rapamycin-EtOH mice were still not as mature as those in the rapamycin-control mice (Figure 7M–O). These results suggest that EtOH inhibits the polycomb repressive complex 2 (PRC2), while rapamycin blocks EtOH-induced overactivation of mTOR, ultimately promoting the maturation of newborn dendritic spines and partially reversing the inhibition caused by EtOH.

Discussion

In the current study, long-term in vivo fluorescence tracing revealed that EtOH inhibited the neurogenesis lineage progression of endogenous Nestin promoter-induced NSCs and dendritic spine formation of tdTom⁺ newborn neurons in the DG of adult mice. Our work is the first to show that EtOH induces significant abnormal orientation of tdTom⁺ NSCs and retards hippocampal neurogenesis from adolescence to adulthood. In addition, EtOH markedly prevented dendritic growth and dendritic spine maturation in tdTom+ newborn neurons. It is worth noting that this study mainly focuses on Nestin-promoter tdTom⁺ NSCs and their lineage progression, showing that binge intake of EtOH ultimately leads to abnormal differentiation and maturation of tdTom⁺ neurons. Studies have shown that Nestin promoter-induced NSCs and their progeny cells account for less than 1% of tdTom⁺ NCs in the GCL of the adult DG (Lagace et al., 2007). After 28 days, these NSCs differentiate into newborn neurons, and the stemness of the NSCs remained unchanged in the Nestin-promoter transgenic mice (Park et al., 2007; Ma et al., 2009; Ferreira et al., 2018). The heterogeneity of radial glial cells includes Gli-promoter (Sun et al., 2001), Nestin-promoter (Sun et al., 2014; Chow et al., 2015), and Glast-promoter (DeCarolis et al., 2013). Pax6-promoter IPCs promote polarity and migration of newborn neurons during brain development (Gan et al., 2014; Thakurela et al., 2016). DCX-promoter NBs are also involved in the neuronal differentiation of mouse embryonic stem cells (Piens et al., 2010). These lineage progressions may also be involved in EtOH-induced changes in hippocampal neurogenesis, and need to be further elucidated.

The effect of EtOH on hippocampal neurogenesis remains controversial (Vetreno and Crews, 2015; Crews et al., 2021). This controversy may be due to the age of the animals, the dosage of EtOH, and the use of ex vivo staining which cannot continuously and specifically label a given cell population (Kee et al., 2002; Nixon and Crews, 2002; Balthazart and Ball, 2014; Susick et al., 2014). The development of fluorescence tracing of specific cells in transgenic mice allows us to study sustained endogenous hippocampal neurogenesis (Zhou et al., 2018). NestinCreER^T2::Rosa26-tdTomato transgenic mice were used in this study to specifically trace the lineage progression of hippocampal tdTom⁺ NSCs in vivo. Our data suggest that compared with adult mice, adolescent mice are more sensitive to EtOH, which significantly retards the proliferation of NSCs and neuronal maturation. This results in more BrdU⁺ NCs stagnating in the IMN stage. The brains of patients with Alzheimer’s disease are reported to contain abnormally oriented hippocampal NSCs and excess neurons at an early stage of development, which could be related to the memory loss observed in these patients (Barnett et al., 2022; Yang et al., 2022). These above reports indicates that the growth and orientation of hippocampal NSCs are closely related to brain cognitive function. Our study is the first to show that EtOH induces sustained and severe abnormal orientation of hippocampal tdTom⁺ NSCs. This further impairs the lineage progression of dentate NSCs, leading to more NSCs stagnating as progenitor cells or IMNs in adult mice, even 2 months after EtOH exposure. Research on dendritic spines showed that they are important for synapse formation (Hlushchenko et al., 2016). We found that EtOH exposure induces abnormally oriented NSCs, which would contribute to long-term abnormal orientation and mispositioning of dendritic newborn neurons and the impaired formation of dendritic spines. Therefore, this study provides important experimental evidence for the mechanisms underlying EtOH-induced deficits in new synapse formation and related memory impairment.

Studies have shown that alcohol exposure during adolescence can cause long-term structural and functional abnormalities in the adult brain (Helfer et al., 2009; Vetreno and Crews, 2015; Ji et al., 2018), particularly hippocampal-related brain development. The abnormalities include reduced neurogenesis, reduced synaptic plasticity, lower complexity of newborn neurons, and ultimately susceptibility to psychiatric disorders in adulthood (Vetreno and Crews, 2015). Consistent with the above results, our findings showed that EtOH exposure for 7 consecutive days in adolescent mice significantly blocks differentiation of adult-born neurons and led to more BrdU⁺ neurons stagnating as IMNs compared with adult mice. EtOH exposure in adolescent mice also significantly increased the percentage of NBs and IMNs, while significantly decreasing the density and proportion of MNs in the hippocampus of adult mice. In addition, EtOH-induced NSCs proliferation was more severely inhibited in adolescence than in adulthood. Thus, EtOH severely impaired the development of hippocampal newborn neurons in adolescent mice after they progresses into adulthood, further confirming that adolescent alcohol abuse impairs adult hippocampal neurogenesis (Coleman et al., 2021) and possibly impairs hippocampal-related learning and memory.

Notably, the EtOH-induced defects described above were partially reversed by voluntary running or inhibition of mTOR (Vetreno et al., 2016, 2018, 2020). Indeed, exercise is reported to promote hippocampal neurogenesis, which helps overcome and rescue aberrant hippocampal neurogenesis and memory loss (Zhou et al., 2018). Reports show that the mice run spontaneously on a total distance of 4–20 km/day and a total activity time of 3–7 hours/day (Manzanares et al., 2018). There are many mechanisms through which voluntary running promotes neurogenesis in the DG, such as overactivation of the Akt-mTOR pathway (Zhang et al., 2014, 2021), inhibition of new cell survival (Morris et al., 2010), epigenetic gene expression, including those triggered by EZH2 (Zhang et al., 2014; Stragier et al., 2015; Macht et al., 2020), and neuroimmune (Vetreno and Crews, 2015) and ERK pathway activity (Zhang et al., 2016). mTOR regulates hippocampal neurogenesis in adult mice by maintaining the pool of hippocampal NSCs (Raman et al., 2013), while inhibition of mTOR overactivation improves aberrant hippocampal neurogenesis (McDaniel and Wong, 2011). Our results demonstrate that inhibition of mTOR partially reverses the EtOH-impaired dendritic spine maturation of newborn neurons in adult mice. Furthermore, both voluntary running and inhibition of mTOR partially reverse the EtOH-induced damage to dendritic spines of hippocampal newborn neurons. Therefore, voluntary running or inhibition of mTOR may provide an important strategy for the treatment of EtOH-induced central nervous system diseases.

The mechanisms underlying the effect of EtOH-induced aberrant hippocampal neurogenesis have been widely reported, including neuroinflammation (Vetreno et al., 2016; Zhang et al., 2021). EtOH was found to significantly increase the translocation of mTOR into lysosomes, resulting in mTOR overactivation in hepatocytes in an adult mouse model of alcoholic binge-drinking (Mazan-Mamczarz et al., 2015; Chao et al., 2018). Conversely, other reports have shown that EtOH induced a significant decrease in mTOR activation in muscle cells or cancer cells in vitro (Hong-Brown et al., 2010; Mazan-Mamczarz et al., 2015). In this study, we explored the effects of EtOH on mTOR and found that pS6 was significantly increased after EtOH exposure, indicating the overactivation of mTOR by binge EtOH exposure. Furthermore, the abnormal activation of mTOR was significantly ameliorated by rapamycin, an mTOR inhibitor, which suggests that mTOR-mediated neuroinflammation may be involved in EtOH-induced aberrant hippocampal neurogenesis in adult mice. The mechanism by which EtOH causes the overactivation of mTOR, and thus impairs dendritic spine formation of newborn neurons, remains unclear. It could be related to neuroinflammation or have an epigenetic origin (Zhang et al., 2014; Macht et al., 2020), but future studies remain necessary.

In this study, we mainly focused on the mTOR-EZH2 pathway. Further studies of the phosphatase and tensin homolog (PTEN)-AKT pathway or other epigenetic regulators involved in EtOH-modulated hippocampal neurogenesis will be necessary in the future. Here, we explored the effects of EtOH on the endogenous neurogenic lineage progression of NSCs and dendritic spine remodeling of newborn neurons with in vivo fluorescent tracing in NestinCreER^T2::Rosa26-tdTomato mice. Other NSC lineage-specific transgenic mice, such as Pax6-promotered and Dcx-promotered, will be the next targets for our investigation. Our in vitro and in vivo results both support the claim that EtOH disturbs newborn dendritic spine remodeling in the hippocampus. Due to the specific populations of NSCs and their lineage cells, our methods were restricted to endogenous fluorescent imaging, immunofluorescent staining, and quantitative polymerase chain reaction. However, other approaches could be applied in the future, such as fluorescence activated cell sorting and in situ hybridization.

In conclusion, this study identified for the first time the abnormal orientation of tdTom⁺ NSCs and aberrant hippocampal neurogenesis both 1 and 2 months after EtOH exposure in mice. This aberrant orientation of NSCs further retarded dentate NSC lineage progression and blocked the growth of tdTom⁺ newborn neurons, ultimately disturbing newborn dendritic spine remodeling. Notably, voluntary running or inhibition of mTOR partially reverses the negative effects of EtOH. This study provides important experimental evidence for the effects of EtOH on hippocampal neurogenesis, and also provides an important theoretical basis for the treatment of central nervous system diseases caused by EtOH.

Additional files:

Additional Figure 1 ^{(412.8KB, tif)} : The establishment of the NestinCreER^T2::Rosa26-tdTomato mice after TAM induction for 1 month.

Additional Figure 1

The establishment of the NestinCreER^T2::Rosa26-tdTomato mice after TAM induction for 1 month.

(A) Representative agarose gel showing the results of genomic DNA genotyping assays in NestinCreERT2::Rosa26-tdTomato mice and wild-type mice. (B, C) Representative confocal images of tdTom+ cells (red) and DAPI (blue) (B), and quantification of total tdTom⁺ cells (C) in DG of NestinCreERT2::Rosa26-tdTomato mice. Scale bars: 200 μm. Data are presented as mean ± SEM (number of animals, n = 14/group; Student’s t-test). DAPI: 4’,6-Diamidino-2-phenylindole; DG: dentate gyrus; AM: tamoxifen; tdTom+: tdTomato⁺.

Additional Figure 2 ^{(419.9KB, tif)} : EtOH inhibits activation and proliferation of NSCs.

Additional Figure 2

EtOH inhibits activation and proliferation of NSCs.

(A) Representative confocal images of Nestin+ cells (green, Alexa Fluor 488), Ki67+ cells (red, Alexa Fluor 555) and DAPI (blue). Scale bars: 75 μm. (B, C) The percentages of Nestin+ cells (B) and Ki67+ cells (C). (D) The percentage of Nestin+Ki67+ aNSCs among Nestin+ NSCs. (E) qPCR analysis of the mRNA levels of Pax6 and Nes in NSCs treated with EtOH (100 mM). The data were normalized by the Ctrl group. (F, G) qPCR analysis of the mRNA levels of Sox4 (F) and Sox9 (G) in NSCs treated with EtOH (100 mM) or 0.1% dimethyl sulfoxide. The data were normalized by the Ctrl group. (H) Schematic diagrams of the experimental design of BrdU pulse labeling. The mice were injected with BrdU (300 mg/kg, i.p.) after EtOH or normal saline treatments and sacrificed 2 hours later. (I-K) Representative confocal images (I) and statistical analysis of DAPI+MCM2+ PCs in adult (J) and adolescent (K) mice after EtOH administration (DAPI, blue; MCM2, red, Alexa Fluor 488). The arrows point to DAPI+MCM2+ PCs. Scale bars: 55 μm. (L) The proportions of MCM2+ PCs in adult and adolescent mice before and after EtOH treatment. (M-O) Representative confocal images (M) and statistical analysis of BrdU+ PCs in adult (N) and adolescent (O) mice after EtOH administration (BrdU, green, Alexa Fluor 488). The arrows point to BrdU+ PCs and the dotted lines refer to the GCL of DG. Scale bars: 55 μm. (P) The proportions of BrdU+ PCs in adult and adolescent mice before and after EtOH treatment. Data are expressed as mean ± SEM (number of independent experiments, n = 6). *P < 0.05, **P < 0.01, ***P < 0.001, vs. Ctrl group (Student’s t-test). aNSCs: Activated NSCs; BrdU: bromodeoxyuridine; Ctrl: control; DAPI: 4’:6-diamidino-2-phenylindole; DG: dentate gyrus; EtOH: ethanol; GCL: granule cell layer; MCM2: minichromosome maintenance protein 2; Nes: nestin; NSCs: neural stem cells; Pax6: paired box gene 6; PCs: proliferative cells; qPCR: quantitative polymerase chain reaction; Sox4: SRY-related HMG-box 4; Sox9: SRY-related HMG-box 9.

Additional Figure 3 ^{(162.4KB, tif)} : EtOH significantly inhibits the differentiation of primary cultured NSCs.

Additional Figure 3

EtOH significantly inhibits the differentiation of primary cultured NSCs.

(A) Representative confocal images of TuJ1+ neurons (green, Alexa Fluor 488) and GFAP+ astrocytes (red, Alexa Fluor 555) and DAPI (blue). Scale bars: 55 μm. (B, C) Percentage of TuJ1+ neurons (B), and GFAP+ astrocytes (C). (D) qPCR analysis of mRNA levels of Dcx and Rbfox3 in neurons treated with EtOH (100 mM). (E, F) qPCR analysis of the mRNA levels of Sox4 (E) and Sox9 (F) in neurons treated with EtOH (100 mM) or 0.1% dimethyl sulfoxide. The data were normalized by the Ctrl group. Data are presented as mean ± SEM (number of independent experiments, n = 6/group). *P < 0.05, **P < 0.01, ***P < 0.001, vs. Ctrl group (Student’s t-test). Ctrl: Control; DAPI: 4’,6-diamidino-2-phenylindole; Dcx: doublecortin; EtOH: ethanol; GFAP: glial fibrillary acidic protein; NSCs: neural stem cells; qPCR: quantitative polymerase chain reaction; Rdfox3: RNA binding fox-1 homolog 3; Sox4: SRY-related HMG-box 4; Sox9: SRY-related HMG-box 9; TuJ1: anti-beta III tubulin.

Additional Figure 4 ^{(306.1KB, tif)} : EtOH inhibits the maturation of newborn neurons in the DG of adult mice.

Additional Figure 4

EtOH inhibits the maturation of newborn neurons in the DG of adult mice.

(A) Schematic diagram of the experimental design of BrdU maintenance labeling in adult mice. (B) Representative confocal images. The four different-shaped arrowheads point to BrdU+DCX-NeuN- NNs, BrdU+DCX+NeuN- NBs, BrdU+DCX+NeuN+ IMNs and BrdU+DCX-NeuN+ MNs (DCX, green, Alexa Fluor 488; BrdU, red, Alexa Fluor 555; NeuN, blue, Alexa Fluor 647), as shown in the right-hand panels. Scale bars: 20 μm. (C-F) Quantifications of total BrdU+ newborn neural cells (C), BrdU+DCX+NeuN- NBs (D), BrdU+DCX+NeuN+ IMNs (E) and BrdU+DCX-NeuN+ MNs (F) in adult DG following EtOH treatment. (G) Pie charts showing the proportions of different cell types in total BrdU+ cells. Data are presented as mean ± SEM (number of animals, n = 6/group). **P < 0.01, ***P < 0.001, vs. Ctrl group (Student’s t-test). BrdU: Bromodeoxyuridine; Ctrl: control; DCX: doublecortin; DG: dentate gyrus; IMNs: immature neurons; MNs: mature neurons; NBs: neuroblasts; NeuN: neuronal nuclei; NNs: non-neurons.

Additional Figure 5 ^{(341.8KB, tif)} : Voluntary running promotes the maturation of newborn neurons in the DG of adult mice.

Additional Figure 5

Voluntary running promotes the maturation of newborn neurons in the DG of adult mice.

(A) Schematic diagram of the design of the voluntary running experiment in adult mice. (B) Representative confocal images. The four different-shaped arrowheads point to different cell types as in B (DCX, green, Alexa Fluor 488; BrdU, red, Alexa Fluor 555; NeuN, blue, Alexa Fluor 647). Scale bars: 20 μm. (C-F) Quantifications of total BrdU+ newborn neural cells (C), BrdU+DCX+NeuN- NBs (D), BrdU+DCX+NeuN+ IMNs (E) and BrdU+DCX-NeuN+ MNs (F) in adult DG after EtOH treatment. (G) Pie charts showing the proportions of different cell types in total BrdU+ cells. Data are presented as mean ± SEM (number of animals, n = 6/group). ***P < 0.001, vs. Ctrl group (Student’s t-test). BrdU: Bromodeoxyuridine; Ctrl: control; DCX: doublecortin; DG: dentate gyrus; IMNs: immature neurons; MNs: mature neurons; NBs: neuroblasts; NeuN: neuronal nuclei; NNs: non-neurons.

Additional Figure 6 ^{(413.1KB, tif)} : EtOH inhibits the activation and differentiation of NSCs by regulating EZH2 pathway.

Additional Figure 6

EtOH inhibits the activation and differentiation of NSCs by regulating EZH2 pathway.

(A-D) qPCR analysis of the mRNA levels of Ezh1 (A), Ezh2 (B), Suz12 (C) and Bmi1 (D) in NSCs treated with EtOH (100 mM) or 0.1% dimethyl sulfoxide. (E-H) qPCR analysis of the mRNA levels of Ezh1 (E), Ezh2 (F), Suz12 (G) and Bmi1 (H) in neurons treated with EtOH (100 mM) or 1‰ dimethyl sulfoxide. The data were normalized by the Ctrl group. (I-K) Confocal images of H3K9me3+ (white, Alexa Fluor 555) (I) H3K36me3+ (white, Alexa Fluor 555) (J) H3K4me3+ (white, Alexa Fluor 555) (K) and DAPI (blue). Scale bars: 50 μm. (L-N) Mander’s coefficients of colocation of H3K9me3 (L), H3K36me3 (M), H3K4me3 (N) and DAPI. Values represent mean ± SEM (number of independent experiments, n = 6, about 1000 cells/group). *P < 0.05, **P < 0.01, ***P < 0.001, vs. Ctrl group (Student’s t-test). Ctrl: Control; DAPI: 4’,6-diamidino-2-phenylindole; EtOH: ethanol; Ezh1: enhancer of zeste homolog 1; Ezh2: enhancer of zeste homolog 2; H3K36me3: trimethylated Lys 36 of histone 3; H3K4me3: trimethylated Lys 4 of histone 3; H3K9me3: trimethylated Lys 9 of histone 3; NSCs: neural stem cells; qPCR: quantitative polymerase chain reaction; Suz12: suppressor of zeste-12 protein.

Additional file 1: Open peer review reports 1 ^{(133.3KB, pdf)} and 2 ^{(97.3KB, pdf)} .