Dulbecco’s modified Eagle medium/Ham’s nutrient mixture F-12 (DMEM/F-12 1:1 mixture), penicillin/streptomycin solution, NB (neurobasal medium), human insulin, and B27 were purchased from Gibco (Grand Island, NY, USA), and bFGF was from R&D Systems (Minneapolis, MN, USA). S3I-201 was obtained from Merck Calbiochem/EMD Biosciences (Darmstadt, Germany). Antibodies used were as follows: NT-3, NT-4/5, BDNF, Neurogenin 1, β-actin, and GAPDH were from Santa Cruz Biotechnology (Santa Cruz, CA, USA); p-STAT3 (Ser 727) and STAT3 were from Cell Signaling Technology (Beverly, MA, USA); Hpca and NeuroD were from Abcam (Cambridge, UK); Myc (Bethyl Laboratories, Montgomery, TX, USA); GFAP (Dako, Glostrup, Denmark); Alexa Fluor^® 488 secondary antibody goat anti-mouse IgG, Alexa Fluor^® 594 secondary antibody goat anti-rabbit IgG, and GFP monoclonal antibody were from Life Technologies (Paisley, UK). All other chemicals were of analytical grade.

Pregnant Sprague–Dawley (SD) rats were obtained from Orient Bio Inc. (Seoul, Korea). Briefly, embryonic hippocampi were mechanically dissected from E16.5 rat embryos in chilled Hank’s balanced salt solution (Gibco), seeded at 2 × 10⁵ cells on 10-cm culture dishes (Nunc A/S, Roskilde, Denmark) precoated with 15 μg/ml poly-L-ornithine (Sigma-Aldrich, St Louis, Mo, USA) and 1 μg/ml fibronectin (Invitrogen, Carlsbad, CA, USA), then cultured for 4 days in serum-free N2 medium supplemented with 20 ng/ml bFGF (R&D Systems). The medium was changed every other day, while the bFGF was supplemented every day in order to expand the population of proliferative precursors. For high density cultures, the cells were plated at 4 × 10⁴ cells per well on coated 24-well plate, 6 × 10⁵ cells per well on coated 6-well or 1 × 0⁶ cells on coated 6-cm culture dishes. The plated cells were induced to additional proliferation in N2+bFGF up to 70–80% cell confluence, before induction of differentiation.

Cells were fixed with 4% (w/v) paraformaldehyde in phosphate-buffered saline (PBS) for 20 min followed by washing with 0.1% BSA in PBS three times at room temperature. They blocked with 10% normal goat serum in 0.1% BSA in PBS containing 0.3% Triton X-100 for 1 h at room temperature. Next, the cells were immunostained with following primary antibodies; 1:400 rabbit polyclonal anti-GFAP (Dako), and 1:400 mouse monoclonal anti-GFP (Life Technologies) at 4°C. Cells were then labeled with 1:2000 Alexa Fluor^® 488 secondary antibody goat anti-mouse IgG (Life Technologies), and 1:2000 Alexa Fluor^® 594 secondary antibody goat anti-rabbit IgG (Life Technologies) for 1 h, before mounting in Vectashield (Vector Laboratories, Burlingame, CA, USA) with 4, 6-diamidino-2-phenylindole (DAPI) mounting medium. Immunoreactive cells were analyzed under an epifluorescence microscope (Nikon Instruments, Melville, NY, USA) at magnifications varying from ×200 to ×400.

Transfection of HNPCs was carried with the Amaxa Nucleofector^® Kit V (Amaxa Biosystems, Köln, Germany) according to the manufacturer’s optimized protocol. In brief, HNPCs were dissociated using 0.05% trypsin for 5 min followed by the addition of 10% FBS in N2 medium. The resuspended cells were counted using a hemocytometer and 5 × 10⁶ cells aliquots were placed in separate tubes (1 tube per nucleofection). The cells were centrifuged at 1000 rpm for 5 min and resuspended in 100 μl of pre-warmed Amaxa Nucleofector Solution. Each 100 μl of cell suspension was mixed with either 10 μg each of pMSCV-IRES-EGFP or pMSCV-Hpca-myc-IRES-EGFP. These DNA-cell suspensions were immediately transferred to Amaxa certified cuvettes and electroporated using the Nucleofector program G-13. The samples were examined by RT-PCR and western blot [see below].

For Hpca or Stat3 silencing experiments, Hpca siRNA (ON-TARGET plus SMART pool), and negative control siRNA (ON-TARGET plus non-targeting pool) were purchased from Dharmacon (Lafayette, CO, USA). Stat3 siRNA (5′-CUGUCUUUAGGCUGAUCAU-3′) was purchased from Bioneer (Daejeon, Korea). Transient siRNA transfections were performed in 6-well plates; cells were transfected with Lipofectamine RNAiMAX (Invitrogen) and 100 nM siRNA according to the manufacturer’s protocol (Invitrogen); and cells were harvested 72 h later.

Hippocalcin cDNA was cloned into a retrovirus vector containing IRES-EGFP (obtained from Dr. Hee Yong Chung, Department of Biomedical Science, Graduate School of Biomedical Science and Engineering, Hanyang University, Seoul, Republic of Korea). Virus particles were produced by transfecting the retrovirus packaging cell line 293gpg with the vector using Lipofectamine (Invitrogen), and supernatants containing viral particles were harvested after incubation for 48 h. For virus transduction, prepared HNPCs were incubated with a viral suspension (4 × 10⁶ particles/ml) containing polybrene (1 μg/ml; Sigma-Aldrich) for 4 h, followed by transfer to bFGF-supplemented N2 medium.

cDNA was prepared from total mRNA extracted from cells with TRIzol^® reagent (Thermo Fisher Scientific Inc., Waltham, MA, USA); 1 μg samples of RNA was reverse-transcribed using Superscript II and random primers (Invitrogen). The resulting cDNA was amplified by PCR using the primers listed in Table 1. PCR products were analyzed on 1.5% agarose gels and stained with loading star (Dyne Bio, Seoul, Korea). For real-time PCR, 5 μl of RT reaction product was amplified in triplicate in a final volume of 20 μl iQ^TM SYBR^® Green supermix (Bio-Rad, CA, USA). The primer sequences for real-time PCR were the same as for RT-PCR.

Cells were lysed in 20 mM Tris–HCl (pH 7.5) containing 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 2.5 mM sodium pyrophosphate, 1% Triton X-100, 1 mM PMSF, and 1 mM Na3VO4. Protein samples (20–30 μg) were loaded onto SDS–polyacrylamide gels and transferred to nitrocellulose membranes (Amersham Pharmacia Biotech, Amersham, UK) after electrophoresis. After blocking with 5% non-fat dried milk for 1 h, membranes were incubated with primary antibodies against Hpca anti-rabbit (1:500; Abcam), Myc anti-rabbit (1:2000; Bethyl Laboratories), NT-3 anti-rabbit (1:200; Santa Cruz Biotechnology), NT-4/5 anti-rabbit (1:200; Santa Cruz Biotechnology), BDNF anti-rabbit (1:250; Santa Cruz Biotechnology), NeuroD anti-rabbit (1:500; Abcam), Neurogenin 1 anti-goat (1:200; Santa Cruz Biotechnology), phospho-STAT3 anti-rabbit (1:1000; Cell Signaling Technology), STAT3 anti-mouse (1:1000; Cell Signaling Technology), β-actin anti-mouse (1:1000; Santa Cruz Biotechnology), GAPDH anti-rabbit (1:1000; Santa Cruz Biotechnology), GFAP anti-mouse (1:1000; Cell Signaling Technology) followed by HRP-conjugated secondary antibody (1:2000; New England Biolabs, Beverly, MA, USA), and specific bands were detected by ECL (Thermo Scientific, Rockford, IL, USA). Bands were quantified using ImageJ (v1.46r, National Institutes of Health, Bethesda, MA, USA).

Cell counting was performed in microscopic field, using an eyepiece grid at a final magnification of 200× or 400×. Immunostained and DAPI-stained cells were counted in 5–10 random areas of each culture coverslip. All values were confirmed in at least five independent cultures. Photos of the cells were taken with an epifluorescence microscope (Nikon Instruments), and morphological characteristics were quantitated using Image J software (NIH¹). The length of branches was defined as the distance from the cell body (proximal) to the tip of the longest branch (distal). For each graph, data on branch length were generated from at least five independent sets of astrocytes (in each condition) and greater than 100 cells were counted for each condition in each experiment.

Quantitative data are expressed as means ± SD of three to five independent experiments. The statistical significance of differences between groups was analyzed using Student’s t-test and was considered significant at p < 0.05.

Dissected and mechanically dissociated cells from hippocampi of E16.5 rat embryos were used to isolate NPCs. For growth of these cells, fresh bFGF (20 ng/ml) was present to prevent differentiation and promote proliferation. To investigate the role of Hpca in HNPCs differentiation, bFGF was removed for 48 h. We initially performed real-time PCR analyses and western blotting to determine the mRNA expression and protein level of Hpca. As shown in Figure 1A, Hpca and Gfap expression increased in differentiating cultures. When the cells were immunostained with GFAP, we also observed significantly increased GFAP positive cells during differentiation (Figure 1B). Increased levels of nerve-growth factors such as NT-3, NT-4/5, and BDNF, together with the proneural bHLH transcription factors neuroD and Ngn1 have been detected in the developing and adult hippocampus and they have major roles in development and maintenance of the hippocampal formation (Ip et al., 1993). Since these factors are closely associated with neural precursor cell differentiation, they can be used as markers of this process (Markus et al., 2002). To examine the effect of Hpca on differentiation, either pMSCV-IRES-EGFP or pMSCV-Hpca-myc-IRES-EGFP was transfected into the HNPCs for 2 days, and then cells were induced differentiation for 1 day. We showed that mRNA levels (Figures 1C,D) and protein expression (Figure 1E) of NT-3, NT-4/5, BDNF, NeuroD, and Ngn1 were significantly increased by Hpca overexpression in the presence of bFGF compared with vector control in the presence of bFGF. HNPCs have been considered as the primary progenitor cells for the neuronal and glial cell lineages during development (Rietze et al., 2001). Therefore, we examined the role of Hpca in expression of neuronal and glial markers during differentiation. HNPCs were transduced with either pMSCV-IRES-EGFP or pMSCV-Hpca-myc-IRES-EGFP for 2 days and allowed to differentiate for 3 days. We measured the percentage of GFAP-positive cells and Tuj1-positive cells under a fluorescence microscope. As shown in Figures 1F,G, Hpca overexpression resulted in markedly enhanced GFAP-positive astrocytes compared with vector control in the presence or absence of bFGF in HNPCs; at the same time Tuj1-positve neurons were not altered (data not shown). Moreover, Hpca overexpression significantly increased the length of branches compared with vector control under the either with bFGF or withdrawal bFGF (Figures 1F,H). We also found that overexpression of Hpca increased protein levels of GFAP compared with vector control in the presence or absence of bFGF in HNPCs (Supplementary Figure 1). These findings collectively imply that Hpca, which regulates the expression of neurotrophins and proneural bHLH transcription factors, promotes astrocytic differentiation of HNPCs.

To confirm the role of Hpca in the expression of GFAP, we examined the effects of Hpca knockdown. 3 days after transfection of HNPCs with Hpca siRNA, the cells were allowed to differentiate for 1 day, and Hpca expression was examined by western blotting. As shown in Figure 2A, Hpca siRNA markedly decreased expression of GFAP. We also found that knockdown of Hpca decreased expression of NT-3, NT-4/5, BDNF, NeuroD, and Ngn1 in the absence of bFGF in HNPCs (Figures 2B,C). To confirm the effect of Hpca on astrocytic differentiation and branch outgrowth of HNPCs, either control siRNA or Hpca siRNA was transfected into the HNPCs for 3 days. Similarly, as shown in Figures 2D,E, it reduced the number of GFAP-positive cells as well as branch outgrowth (Figures 2D,F).

Taken together, these results indicate that Hpca is crucial for the expression of GFAP and for branch formation.

We next asked which signaling pathway was responsible for Hpca-mediated astrocytic differentiation of HNPCs. Because the STAT3 activation is associated with GFAP expression in neural precursor cells (Sun et al., 2001), we investigated whether STAT3 was phosphorylated by Hpca overexpression. As shown in Figures 3A,B, phosphorylation of STAT3 at Ser727 was significantly increased by overexpression of Hpca compared to vector control in HNPCs. On the other hand, knockdown of Hpca decreased the phosphorylation of STAT3 (Ser727) (Figures 3C,D). These results suggest that Hpca has an important role in the activation of STAT3 (Ser727) during astrocytic differentiation of HNPCs.

Next, we determined whether STAT3 (Ser727) activation is required for astrocytic differentiation of HNPCs. Cells were transiently transfected with pMSCV-IRES-EGFP or pMSCV-Hpca-myc-IRES-EGFP for 2 days and treated with the STAT3 inhibitor, S3I-201, for 1 day. Inhibition of STAT3 not only decreased phosphorylation of STAT3 (Ser727) (Figures 4A,B) but also reduced the level of GFAP (Figures 4A,C) and transcript levels of NT-3, NT-4/5, BDNF, NeuroD, and Ngn1 (Figures 4D,E). We also found that treatment with 10 μM S3I-201 significantly reduced the number of GFAP-positive cells (31.8 ± 1.8% vs. 16.3 ± 1.3%, p < 0.05; Figures 4F,G) compared to the overexpression of Hpca. To test if STAT3 can also promote the branch outgrowth, we calculated the length of branches. The extent of branch outgrowth observed upon overexpression of Hpca was also reduced when STAT3 was inhibited (Figures 4F,H). To further determine the involvement of STAT3 in Hpca-mediated astrocytic differentiation, we knocked down STAT3 using siRNA. Stat3 siRNA completely abolished expression of GFAP (Figures 5A,B) and transcripts of NT-3, NT-4/5, BDNF, NeuroD, and Ngn1 compared with overexpression of Hpca (Figures 5C,D). We also found that knockdown of STAT3 markedly decreased the number of GFAP-positive cells (45.0 ± 4.0% vs. 15.0 ± 1.0%, p < 0.05; Figures 5E,F) compared with overexpression of Hpca. Furthermore, knockdown of STAT3 decreased branch outgrowth (Figure 5G). Taken together, our results demonstrate that STAT3 is the signaling molecule that triggers branch outgrowth leading to astrocytic differentiation of HNPCs.