Fetal Hypothalamic Neuroprogenitor Cell Culture: Preferential Differentiation Paths Induced by Leptin and Insulin

Animals

Studies were approved by the Animal Research Committee of the Los Angeles Biomedical Research Institute at Harbor-University of California Los Angeles (LABioMed, Torrance, CA) and were in accordance with the American Association for Accreditation of Laboratory Care and National Institutes of Health guidelines. Four pregnant Sprague Dawley rats were used. Fetal brains were collected at e20 to integrate developmental neurogenesis and effective leptin/insulin signaling (26). Six fetal brains per litter were pooled.

Hypothalamic dissection

The brain was placed ventral side up. A coronal slice (∼4-mm thickness, which contains whole hypothalamus) was cut. The caudal margin of the slice was mamillary body, and rostral edge was optic chiasm. Then, semispheres adjunct to two sides of hypothalamus were cut, and finally, the dorsal part of the slice was cut off. The ventral part (∼2 mm), which is the hypothalamus, was used for NPC isolation.

Culture of hypothalamic NPC: NS

After Hanks' solution wash, the brains were transferred to DMEM/F12 medium, and hypothalami were minced and digested by 0.1% trypsin-EDTA (Invitrogen, Carlsbad, CA) plus 0.01% deoxyribonuclease (Sigma, St. Louis, MO) for 30 min at 37 C in CO₂ incubator. Trypsin was inactivated with 10% fetal bovine serum and the digest filtered and centrifuged at 1000 rpm for 5 min. Cells were resuspended (∼5 × 10⁴ cells/ml) in complete medium (CM) [neurobasal medium containing 1% anti-anti (Invitrogen), 2% B27 (catalog no. 17504-044; Invitrogen), 20 ng/ml fibroblast growth factor (FGF)2 (Sigma), 20 ng/ml epidermal growth factor (EGF) (Sigma), 1 μg/ml Heparin (Lilly, Indianapolis, IN), and 2.5 μg/ml l-glutamine (Invitrogen)], and 4 μg/ml insulin (because total absence of insulin has been shown to cause NPC cell death (27) with no leptin or IGF-I and were cultured for 8 d in CO₂ incubator. FGF2 (10 ng/ml) was added to the culture every 3 d, and the NS at this stage represented passage (P)0. P0 NS were collected and disassociated into single cells by 0.1% trypsin-EDTA digestion at 37 C for 10 min. The disassociated cells were centrifuged at 1000 rpm for 5 min, reseeded in CM at same cell density as above, and represented P1 cells. P1 cells were used for treatment (see below) or after 8 d underwent further trypsinization and reseeding, representing P2 cells.

Induction of NPC differentiation

The following experiments on NPC proliferation and differentiation used only P1 and P2 cells. For the induction of differentiation, the disassociated cells were resuspended in differentiating medium (DM) (in absence of FGF2, EGF, and Heparin) and seeded in culture dishes precoated with 0.01% poly-l-lysine (Sigma).

Leptin and insulin treatments

For studying leptin and insulin effects, the NS or NPC were cultured in both CM and DM and treated with leptin (0, 20, and 40 ng/ml; Sigma) or insulin (0, 10, 20, and 40 μg/ml) every 48 h for 8 d. Subsequently, the cells were harvested for cell proliferation or protein expression analysis. Leptin and insulin concentrations used in the present study have been previously shown to be effective in inducing leptin-mediated effects in primary hypothalamic neuronal culture (26, 28) and insulin-mediated effects on brain slices (29).

Chemical inhibitors of Notch, extracellular signal-regulated kinase (ERK), and signal transducer and activator of transcription (STAT)3

NPC cultured in both CM and DM were treated every 48 h for 8 d with Notch inhibitor (γ-secretase inhibitor X, 2 μm; Sigma), ERK inhibitor (PD98059, 20 μg/ml; Chemicom, Billerica, MA) or STAT3 inhibitor (Tyrphostin AG490, 5 μm; Sigma) in presence or absence of leptin (40 ng/ml) or insulin (40 μg/ml). Cell proliferation or protein expression was determined.

NPC proliferation [3-(4,5-dimethylthiazol-2-yl)-diphenyl tetrazolium bromide (MTT) assay]

NPC proliferation rate was determined by the MTT (Sigma) colorimetric assay (30, 31). Disassociated NS cells in CM were seeded in 24-well-culture plate (1 ml/well). At d 8 of culture, 50 μl MTT solution (5 mg/ml in BPS) was added to each cell culture well and incubated for 1 h at 37 C in CO₂ incubator. The cultured cells were harvested, and MTT reaction product formazan was extracted with 300 μl acidic isopropanol (isopropanol in 0.04 n HCl). The optical density of the formazan solution was measured on an ELISA plate reader (VICTOR 1420 Multilabel Counter, Perkin Elmer Wallac Inc., Gaithersburg, MD) at 570 nm. Cell proliferating index was expressed as value of OD 570 nm.

Western blot analysis

The disassociated NS cells in CM or DM were seeded in six-well plates. At d 8 of culture, the cells were harvested and dissolved in RIPA solution (Cell Signaling, Danvers, MA) with protease inhibitor cocktail (Thermo, Rockford, IL). After a brief sonication, the cell lysates were precleared by centrifugation (10,000 rpm) for 10 min at 4 C. Protein content of the cell lysates was determined by bicinchoninic acid Protein Assay kit (Thermo). Then the cell lysates were denatured in XT sample buffer (Bio-Rad, Hercules, CA) with XT Reducing Agent (Bio-Rad) at 100 C for 3 min, and 20 μg of proteins for each sample were used to run Critrion XT Precast Gel. After electro-transferring the fragmented proteins to Nitrocellulose Membrane (Bio-Rad), the blots were blocked for 1 h in Tris-buffered saline containing 0.05% Tween 20 (TBST) and 5% fat-free milk powder. Primary antibodies were diluted in blocking solution and incubated with blots overnight at 4 C. The blots were then washed in TBST for 10 min (three cycle washes) and incubated for 1 h in solution of secondary antibodies conjugated horseradish peroxidase (HRP) at room temperature. The secondary antibody included antimouse IgG-HRP (1:2000; Cell Signaling), antirabbit IgG-HRP (1:2000; Cell Signaling), and antigoat IgG-HRP (1:1000; Santa Cruz Biotechnology, Inc., Santa Cruz, CA). After three cycle washes by TBST, blots were applied with SuperSignal West Pico Chemiluminescence Substrate (Pierce, Rockford, IL) to produce chemiluminescence, which was visualized by exposing blots to x-ray film (HyBlot CL Autoradiography Film; Denville Scientific, Inc., Metuchen, NJ). Densities of _target protein bands were determined by densitometer (Alpha Digidoc 1000; Alpha Innotech Corp., Santa Clara, CA) and normalized against glyceraldehyde-3-phosphate dehydrogenase loading controls. The primary antibodies used included: rabbit antinestin (1:5000; Sigma), rabbit anticlass IIIβ-tubulin (Tuj1) (1:5000; Sigma), rabbit antiglial fibrillary acidic protein (GFAP) (1:10,000; Dako North America Inc., Carpinteria, CA), rabbit anticleaved Notch1 (1:1000; Cell Signaling), rabbit antihairy and enhancer of split 1 (Hes1) (1:1000; Santa Cruz Biotechnology, Inc.), rabbit antiphosphorylated ERK1/2 (1:2000; Cell Signaling), mouse antiphosphorylated STAT3 (1:1000; Upstate, Waltham, MA), and mouse antiglyceraldehyde-3-phosphate dehydrogenase (1:10,000; Chemicon). The antibody concentrations were diluted according to manufacturer's recommendation. For Tuj1, GFAP, and nestin, range of dilutions were further undertaken to determine optimal dilution to reduce background and increase band intensity.

Immunostaining

For staining cultured cells, four separate NPC cultures were used. P0 disassociated NS cells in CM or DM were seeded in 12-well plates, in which round cover glasses (catalog no. 12-545-84 18CIR-1D; Fisher Scientific, Pittsburgh, PA) were preplaced. For each culture, P1 NPC were cultured in DM and were either untreated or treated with leptin (20 ng/ml) or insulin (20 μg/ml). At d 8, the cells were fixed in 4% paraformaldehyde in PBS for 30 min and were immunostained with rabbit antinestin (Sigma), rabbit anti-Tuj1 (1:500; Sigma), rabbit anti-GFAP (1:500; Dako), or 4′,6-diamidino-2-phenylindole (DAPI). Secondary antibodies were donkey antirabbit IgG-Alexa Fluor 488 (1:250) or donkey antimouse IgG-Alexa Fluor 568 (1:250). Images shown were taken at ×40. Approximately 3000 total cells (DAPI-stained nucleus) from four separate cultures were counted using ImageJ software. Stained images of Tuj1 or GFAP were superimposed on related DAPI image were. Cells with nuclei surrounded by Tuj1 or GFAP signals were counted as neurons or astrocytes, respectively. Thus, Tuj1 + DAPI and GFAP + DAPI positive cells were counted and are expressed as percentage of total DAPI cells. For negative control for NS immunostaining, the primary antibody against nestin, Tuj1, or GFAP was replaced by normal rabbit IgG. The rest of the procedure was identical to nestin/Tuj1/GFAP staining.

Data analysis

Differences between treatments were analyzed by ANOVA with Dunnett's post hoc test. Values (mean ± se) are presented as percentage of control. For clarity purposes, one representative blot is shown for protein data.

NS growth and differentiation

Fetal hypothalamic NS were consistently cultured in complete media and successfully produced P1 and P2 NS under controlled conditions. As demonstrated in Fig. 1, both leptin and insulin induced significant NS proliferation, in a dose-dependent response. For confirmation of NS proliferation, both control and leptin/insulin-stimulated NS were stained for DAPI (nucleus stained) and nestin, a neuroprogenitor cell marker and IgG (negative control) (Fig. 2, A and B). When measured by Western blot analysis, both leptin and insulin demonstrated increased nestin expression (Fig. 2C). Although the leptin-induced proliferation index was greater than that of insulin, insulin demonstrated a markedly increased level of nestin stimulation compared with leptin.

As evident by immunostaining, both leptin (20 ng/ml) and insulin (20 μg/ml) stimulated hypothalamic NS differentiation into neurons (Tuj1) and astrocytes (GFAP), when cultured in differentiation media, respectively (Figs. 3A and 4A). Due to potential limitation of precise quantitative assessment by immunostaining, we further confirmed the changes by determining protein expression using varying doses of leptin and insulin. The neuronal differentiation marker (Tuj1) demonstrated significant increases in response to increasing doses of leptin, although modest increases in response to insulin (Fig. 3, B and C). In contrast to neuronal differentiation, there was no significant increase in astrocyte GFAP expression in response to leptin, although insulin induced a significant increase in a dose-dependent fashion (Fig. 4, B and C).

Determination of leptin and insulin signaling pathways

Both leptin and insulin induced Notch1 signaling, as evidenced by increased activation of the Notch1 intracellular domain (ICD) and Hes1 expression when cultured in complete (Fig. 5A) media. There was a relative absence of dose-response effects in complete media, with the lowest doses of both insulin and leptin demonstrating peak effects. There was no increase in Notch1 or Hes1 expression in differentiation media (data not shown). In contrast to the Notch1 pathway, leptin induced activation of the ERK and Janus Kinase/STAT pathway in both complete and differentiation media, as evidenced by increased pERK1/2 and pSTAT3 (Fig. 5B). Insulin induced pERK1/2 phosphorylation in complete media and differentiation media but had no significant effect on pSTAT3 (Fig. 5B).

To investigate the role of these signaling pathways on NPC proliferation, differentiation, and signaling responses, we used selective Notch, ERK, and STAT3 inhibitors. As demonstrated in Table 1, leptin and insulin-mediated NS proliferation was inhibited by Notch, ERK, and STAT3 inhibitors. Notably, the Notch and ERK inhibitors suppressed control NS proliferation and reduced leptin and insulin-stimulated proliferation to below control levels. Although the STAT3 inhibitor had no effect on basal proliferation, it suppressed leptin and insulin-stimulated proliferation to control levels. Similar to the measure of proliferation index, the ERK inhibitor reduced both control and leptin/insulin-mediated hypothalamic nestin expression for NS cultured in complete media. The STAT inhibitor did not affect control nestin expression but significantly suppressed leptin/insulin-mediated nestin expression (Table 2). To further explore any cross talk between the signaling pathways, we measured the effects of PD98059 and AG490 on Notch activation. Neither the ERK nor STAT3 inhibitor affected leptin or insulin-induced Notch1 ICD or Hes1 (data not shown).

When examined in differentiation media, the ERK inhibitor had no effect on control neuronal differentiation or on leptin or insulin-induced neuronal marker expression. However, it completely blocked insulin-induced astrocyte differentiation (GFAP) (Table 3). In contrast, the STAT inhibitor significantly suppressed both control and leptin/insulin-induced Tuj1 neuronal marker expression, although there was no effect of the STAT inhibitor on control or insulin-induced astrocyte differentiation (Table 3). The effect of the Notch inhibitor on NS differentiation was not studied due to the lack of leptin or insulin-induced ICD/Hes1 activation in differentiation media.