NPCs were isolated from the rat embryonic day 16 (E16) embryonic brain as described previously [11,12]. Time-pregnant Sprague Dawley rats were purchased from Orient Co., Ltd., a branch of Charles River Laboratories (Gyeonggi-do, Korea). NPCs expressing green fluorescent protein (GFP) were isolated from the transgenic mice expressing GFP under the control of paired box protein 6 (pax6) promoter (pax6-GFP mice), which were kindly offered by Drs. Jiha Kim and James D. Lauderdale, Dept. of Cellular Biology, University of Georgia, Athens, GA 30602, USA [13,14]. Briefly, hippocampal eminence was dissected from the rat E16 embryonic forebrain aseptically using fine forceps under a dissecting microscope. NPCs expressing GFP were isolated from the dorsal forebrain of the E14 embryo of pax6-GFP transgenic mice. The tissues were collected and dissociated mechanically in Ca²⁺/Mg²⁺-free Hank's Balanced Salt Solution (HBSS, Cat. No. 88284, Invitrogen) and the cells were plated at 30,000 cells/cm² on a 60 mm culture dish pre-coated with 15 µg/ml poly-L-ornithine (Cat. No. P4957, Sigma-Aldrich, St. Louis, MO, USA) and 1 µg/ml fibronectin (Cat. No. F0635, Sigma-Aldrich, St. Louis, MO, USA), and allowed to proliferate in serum-free N2 media with basic fibroblast growth factor (bFGF; 10 ng/ml; Cat. No. P15655, R&D system, USA) for 3~4 days up to 70~80% confluences. bFGF is known to stimulate cell division and suppress the differentiation of NPCs [15,16]. To induce differentiation into neurons, NPCs were grown in N2 media with bFGF for an additional 1 day after plating on new dishes and changed with N2 media without bFGF and subsequently incubated for appropriate times (differentiation media). In order to examine the effect of aucubin on neuronal differentiation, NPCs were incubated in differentiation media including various concentrations of aucubin (0, 0.01, 0.1, 1, and 10 µM, Cat. No. 55561, Sigma-Aldrich, St. Louis, MO, USA).

For immunocytochemical assays, NPCs were grown on glass coverslips (Cat. No. 1916-91012, Bellco, USA) for 7 more days for differentiation including aucubin to detect specific neuron markers. NPCs were passaged using 0.05% trypsin–EDTA (Cat. No. 25300054, Gibco by Thermo Fisher Scientific Inc.) on 12 mm glass coverslips (Cat. No. 1916-91012, Bellco, Vineland, NJ, USA). Rat hippocampal NPCs (3×10⁴ cells/cm²) or mouse NPCs from dorsal forebrain (1.5×10⁴ cells/cm²) were grown in N2 media with bFGF for an additional 1 day. Rat hippocampal NPCs were grown for 6 more days and mouse NPCs from dorsal forebrain for 2 more days without fibroblast growth factors (FGF) to detect specific neuronal cell markers (glutamic acid dehydrogenase 65/67 (GAD65/67), vesicular glutamate transporter 1 (vGluT1), and GFP). For immunoblot assays, NPCs (3×10⁴ cells/cm²) were grown on a 60 mm dish for 3 more days in differentiation media with or without aucubin.

NPCs cultured on coverslips were fixed with 4% paraformaldehyde (Sigma) for 30 min and washed with phosphate-buffered saline (PBS). For GABA antibody reactivity, the cells were fixed with 4% paraformaldehyde, 0~0.5% glutaraldehyde, and 0.5% potassium dichromate in 0.1 M phosphate buffer at pH 6.5. After fixing, the cells were incubated with primary antibodies against GFP (rabbit anti-GFP antibody, 1:500, Cat. No. AB3080, Merck KGaA, Darmstadt, Germany), gamma-aminobutyric acid (mouse anti-GABA antibody, 1:1000, Cat. No. ab86186, Abcam), glutamic acid dehydrogenase 65/67 (rabbit anti-GAD65/67 antibody, 1:500, Cat. No. Ab1511, Chemicon International Inc.), and vesicular glutamate transporter 1 (mouse anti-vGluT1 antibody, 1:500, Cat. No. MAB5502, Chemicon International Inc.). Cells were subsequently incubated with FITC- or Alexa fluor-conjugated secondary antibodies (1:200 for each, Molecular Probes). Slides were counterstained with 4,6-di-amidine-2-phenylindole dihydrochloride (DAPI, 1 µg/ml, Cat. No. D9542, Sigma-Aldrich). At least five fields of view (objective magnification: ×200) from an average of five slides per experimental group were selected randomly and neuronal differentiation marker-positive cells were then counted. Stained slides were examined under a confocal laser scanning biological microscope (LSM 510, Carl Zeiss).

NPCs in 60 mm dishes incubated for 3 days in differentiation media with or without aucubin were lysed in 80 µl of ice-cold radioimmunoprecipitation assay (RIPA) buffer (cat # R0278, Sigma-Aldrich) containing a protease inhibitor cocktail and cleared via 10 min of centrifugation at 13,000×g. Following determination of protein concentration using Bradford protein assay kits (Cat. No. B6916, Bio-Rad, Hercules, California, USA), 10~20 µg of proteins were separated on 10 % sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, transferred to a nitrocellulose membrane, and probed for various neuronal markers and β-actin. The primary antibodies used were mouse anti-vGluT1 antibody (1:400, Chemicon), rabbit anti-GAD65/67 antibody (1:500, Cat. No. PA5-36080, Chemicon), rabbit anti-NeuN antibody (1:5000, Cat. No. ab104225, Abcam) and mouse anti-β-actin antibody (1:1000, Cat. No. sc-47778, Santa Cruz, Delaware, California, USA). Signals were visualized using enhanced chemiluminescence (LumiGLO, KPL Europe, Guildford, UK) as recommended by the manufacturer. The intensity of each band was quantified with a densitometric scanner using an image analysis program.

All quantitative data are expressed as mean±standard deviation. A one-way analysis of variance (ANOVA) using the Bonferroni test was performed to determine significant differences. The assumptions of ANOVA were found to satisfy Levene's test for homogeneity of variance and to pass tests for normality. A value of p<0.05 was considered statistically significant.

To search for the effect of aucubin on neuronal differentiation, we isolated NPCs primarily from the rat E16 hippocampus when precursors of pyramidal cells initiate proliferation [17,18]. Neuronal differentiation of NPCs was induced by incubating in chemically defined N2 medium without bFGF. In the differentiation condition, NPCs were treated with various concentrations, such as 0.01, 0.1, 1 and 10 µM of aucubin. We investigated whether aucubin facilitates neuronal differentiation into specific types of neuronal cells by examining GAD65/67- and vGluT1-expressing cells using an immunocytochemical assay. GAD65/67, a GABA synthesizing enzyme and vGluT1 are expressed specifically in GABAergic and glutamatergic neurons, respectively, which are the representative inhibitory and excitatory neurons in the mammalian brain. After 7 days of differentiation, GAD immunoreactive cells elongated a number of long neurites with the addition of aucubin, while vGluT1 immunoreactive cells contained mostly one short neurite. The length of the neurites of differentiated cells immunostained with anti-GAD65/67 antibodies seemed to be increased when the aucubin concentration was increased (Fig. 1A and 1C). While the numbers of both GAD65/67 and vGluT1 immunoreactive cells increased statistically significantly in aucubin-treated NPCs compared to NPCs with vehicle-treated control, the increase was much more prominent in GAD65/67 immunoreactive cells (Fig. 1A and 1D). Aucubin treatment at a low concentration of 0.1 µM increased GAD65/67 immunoreactive cells about 1.8 times compared to control, and a bigger increase was observed at 10 µM concentration (Fig. 1A). In contrast, vGluT1 immunoreactive cells increased at most by 1.1-fold with 0.1 µM of aucubin (Fig. 1C and 1D), without a bigger increase at higher concentrations of aucubin. These data indicate that aucubin promotes the neuronal differentiation of NPCs more into GAD65/67-expressing cells than to vGluT1-expressing cells.

To determine whether aucubin promotes neuronal differentiation specific to GABAergic neurons, we examined the protein expression of GAD65/67, vGluT1, and a mature neuron marker, NeuN, in NPCs differentiated with or without aucubin by immunoblotting analysis. Fig. 2 shows that the treatment of aucubin in differentiation conditions upregulated the protein expression of NeuN, GAD65/67, and vGluT1. When the intensity of the expressed proteins was accessed in relation to β-actin, a substantial increase in GAD65/67 expression was detected in 0.01 µM aucubin-treated NPCs and a dose-dependent increase of up to 3.8-fold was observed in the 1 µM concentration of aucubin (Fig. 2A and 2C). However, the increase in vGluT1 expression in NPCs treated with a concentration of aucubin between 0.01 and 10 µM was much less than that of GAD65/67 (Fig. 2C and 2D). NeuN, a mature neuron marker, was also expressed more in differentiated NPCs treated with 1 and 10 µM of aucubin, similar to vGluT1 (Fig. 2B). Taken together, these results suggest that aucubin has a differential effect to promote neuronal differentiation into GABAergic neurons rather than glutamatergic neurons.

Glutamate decarboxylase (GAD65 and GAD67) is responsible for synthesizing GABA, an inhibitory neuronal transmitter. To confirm the effects of aucubin on the cell type differentiation of cortical NPCs, we next examined if aucubin enhances GABAergic differentiation and produces GABA. We isolated cortical NPCs from the dorsal forebrain of pax6-GFP transgenic mice, which express GFP under the control of pax6 promoter. In the differentiation condition, cortical NPCs were treated with the same concentrations of aucubin (0.01, 0.1, 1, and 10 µM) as hippocampal NPCs and the proportion of cells producing GABA was determined by immunocytochemical analysis. As shown in Fig. 3A, treatment with aucubin at various concentrations for 3 days increased the number of pax6-expressing NPCs, determined as GFP-positive cells. Aucubin at a very low concentration of 0.01 µM showed the effect of increasing the differentiation of NPCs, and treatment with 1 µM of aucubin increased the number of GFP-positive NPCs more than two-fold compared to NPCs with vehicle treatment (Fig. 3A and 3B). When the proportion of GABA-producing neurons among DAPI-positive cells was determined by immunostaining using anti-GABA antibodies, increasing values were observed with aucubin treatment during neuronal differentiation (Fig. 3B). In particular, in the differentiation condition including 0.1 µM aucubin, it exhibited a more than 1.5-fold increase over controls in the proportion of GABA-producing neurons. This indicates that aucubin facilitates differentiation into neurons producing GABA, an inhibitory neurotransmitter, during the neuronal differentiation of NPCs.