Male C57BL/6J mice (n = 68) were purchased from the Jackson Laboratory (USA). The animals were used at 8 weeks of age, and housed at 22℃ with 60% humidity and a 12-h light/dark cycle. All mice had free access to food and tap water. Animal handling and care conformed to guidelines that comply with current international laws and policies (National Institutes of Health [NIH] Guide for the Care and Use of Laboratory Animals, NIH Publication No. 85-23, 1985, revised 1996), and the experimental protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of Seoul National University (Approval No. SNU-110412-2). All experiments were conducted to minimize both the number of mice used and suffering due to the procedures performed in the present study.

Six-week-old mice were individually caged and allowed to adapt to a chow diet for 1 week. After this time, the mice were fed a commercial LFD (D12450Bi used as a control diet for D12492, n = 34; Research Diets, USA) or HFD (D12492i, n = 34; Research Diets) for 8 weeks. Four weeks after initiating the LFD and HFD feeding, vehicle (0.1% methyl cellulose) or 2 mg/kg RSG (Avandia; GlaxoSmithKline, USA) was orally administered to mice in both groups using a feeding needle (Kent Scientific, USA) once a day for 4 weeks. The RSG dose was chosen because it is equivalent to the doses clinically used in humans. With this concentration, we previously observed a reduction of cell proliferation and neuroblast differentiation in normal mice [26]. The experimental schedule was adopted because a previous study detected an HFD-induced reduction of cell proliferation and neuroblast differentiation at 4 weeks after the start of HFD consumption [12]. In addition, DCX is exclusively expressed in immature neurons in cells 1 to 28 days old [2].

Eight weeks after starting LFD or HFD feeding (4 weeks after vehicle or RSG treatment), the mice (n = 7 in each group) were anesthetized with 30 mg/kg Zoletil 50 (Virbac, France) and perfused transcardially with 0.1 M phosphate-buffered saline (PBS, pH 7.4) followed by 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4). The brains were removed and postfixed by 4% paraformaldehyde in 0.1 M PB for 12 h. The brain tissues were cryoprotected by overnight infusion with 30% sucrose, and 30-µm-thick coronal sections were serially cut using a cryostat (Leica, Germany). The sections were transferred to six-well plates containing PBS for further processing.

Immunohistochemistry was performed under the same conditions for tissues from each group to obtain consistent results. We selected brain sections collected between 1.6 mm and 1.8 mm posterior to the bregma as determined by a mouse brain atlas [8]. The sections were incubated with goat anti-DCX antibodies (1 : 50 dilution; Santa Cruz Biotechnology, USA) or rabbit anti-Ki67 (1 : 1,000; Abcam, UK) overnight at room temperature, and subsequently exposed to biotinylated donkey anti-goat or goat anti-rabbit IgG (1 : 200; Vector Laboratories, USA) for 2 h and streptavidin peroxidase complex (1 : 200, Vector Laboratories) for 30 min at room temperature. Antibody binding was detected using 0.05% 3,3'-diaminobenzidine tetrachloride (Sigma, USA) in 0.1 M Tris-HCl buffer (pH 7.2), and the sections were mounted on gelatin-coated slides.

The numbers of Ki67- and DCX-positive cells in samples from all groups were determined using an image analysis system equipped with a computer-based charge-coupled device (CCD) camera (Optimas 6.5; Media Cybernetics, USA). In addition, images of all DCX-immunoreactive structures in the dentate gyrus were obtained with a BX51 light microscope (Olympus, Japan) equipped with a digital camera (DP71; Olympus) connected to a PC monitor. Ki67- and DCX-positive cells in the dentate gyrus of each section were counted using Optimas 6.5 software (Media Cybernetics). Cell counts for all the sections from every mouse were averaged and are presented as a percentage.

To confirm the effects of RSG on neuroblast differentiation, five mice from each group were sacrificed and used for Western blot analysis [39]. After the brains were removed, the dentate gyrus was removed with a surgical blade. The dentate gyrus was homogenized in 50 mM PBS (pH 7.4) containing 0.1 mM ethylene glycol bis-(2-aminoethyl ether)-N,N,N',N' tetraacetic acid (EGTA, pH 8.0; Sigma), 0.2% Nonidet P-40 (Sigma), 10 mM ethylendiamine-tetraacetic acid (EDTA, pH 8.0; Sigma), 15 mM sodium pyrophosphate (Sigma), 100 mM β-glycerophosphate (Sigma), 50 mM NaF (Sigma), 150 mM NaCl (Sigma), 2 mM sodium orthovanadate (Sigma), 1 mM phenylmethylsulfonyl fluoride (PMSF, Sigma), and 1 mM dithiothreitol (DTT; Sigma). After centrifugation, protein concentration of the supernatants was determined using a Micro BCA protein assay kit (Pierce Chemical, USA) with bovine serum albumin as the standard. Aliquots containing 50 µg of total protein were boiled in a loading buffer that contained 150 mM Tris (pH 6.8), 3 mM DTT, 6% sodium dodecylsulfate (SDS), 0.3% bromophenol blue (Sigma), and 30% glycerol. The aliquots were then loaded onto a 8% polyacrylamide gel (Sigma). After electrophoresis, the proteins were transferred to nitrocellulose membranes (Pall Corporation, USA). To reduce background signals, the membranes were blocked with 5% non-fat dry milk (Sigma) in PBS containing 0.1% Tween 20 for 45 min. Next, the blots were incubated with with goat anti-DCX (1 : 100) and then peroxidase-conjugated anti-goat IgG (Vector Laboratories). Antibody binding was detected with an enhanced luminol-based chemiluminescent (ECL) kit (Pierce Chemical). The bands was densitometrically scanned to quantify the relative optical density (ROD) using Scion Image software (Scion Corporation, USA). The obtained data were normalized against that for β-actin.

In order to confirm changes in BDNF levels in the dentate gyrus [38], five mice from each group were anesthetized with 100 mg/kg Zoletil 50 (Virbac) and decapitated. The hippocampus was removed from the brain and stored in liquid nitrogen. BDNF levels in the hippocampus were measured using a BDNF Emax immunoassay kit (Promega, USA). The tissue samples were weighed and 300 µL of lysis buffer was added to each sample. The samples were then sonicated for 30 sec and centrifuged at 4℃ for 20 min. The supernatant was stored at -20℃ until it was analyzed. All samples were assayed in duplicate and the absorbance was read with an enzyme-linked immunosorbent assay (ELISA) plate reader (Bio Tek, USA). The concentration for each sample was calculated by plotting the absorbance values on a standard curve with known concentrations generated by the assay.

Data are presented as the mean for each experiment. Differences between the mean values were analyzed with a one-way analysis of variance followed by Tukey's multiple range test. Statistical significance was considered at p < 0.05.

In all groups, Ki67-positive nuclei were detected in the subgranular zone of the dentate gyrus. Among all mice, the LFD-fed vehicle-treated group had the highest number of Ki67-positive cells (Figs. 1A and E). The number of Ki67-positive cells was decreased in the LFD-fed RSG-treated group compared to the LFD-fed vehicle-treated animals (Figs. 1B and E). Ki67-positive nuclei were rarely detected in the dentate gyrus of the HFD-fed vehicle-treated group, and the number of Ki67-positive cells was significantly decreased in these mice compared to the LFD-fed vehicle-treated animals (Figs. 1C and E). The number of Ki67-positive cells was markedly increased in the HFD-fed RSG-treated mice compared to the HFD-fed vehicle-treated group (Figs. 1D and E).

In all groups, DCX-immunoreactive neuroblasts were detected in the subgranular zone of the dentate gyrus. The dendrites extended into the molecular layer of the dentate gyrus. Compared to the other three groups, the LFD-fed vehicle-treated mice had a greater number of DCX-positive neuroblasts in the dentate gyrus (Figs. 2A and B). The number of DCX-immunoreactive neuroblasts and their dendrites were decreased in the dentate gyrus of the LFD-fed RSG-treated group compared to the LFD-fed vehicle-treated animals (Figs. 2C, D, and I). The number of DCX-positive neuroblasts was also markedly decreased in the HFD-fed vehicle-treated group compared to the LFD-fed vehicle-treated group. In addition, the abundance of DCX-immunoreactive dendrites was prominently decreased in the HFD-fed vehicle-treated group (Figs. 2E and F). However, the number of DCX-immunoreactive neuroblasts was increased in the dentate gyrus in the HFD-fed RSG-treated group compared to the HFD-fed vehicle-treated mice (Figs. 2G, H, and I).

Western blot analysis showed that DCX protein expression in the dentate gyrus was significantly lower in the LFD-fed RSG-treated mice than in the LFD-fed vehicle-treated group. Compared to the LFD-fed vehicle-treated group, the HFD-fed vehicle-treated group had significantly decreased levels of DCX protein. However, the expression of DCX protein in the HFD-fed RSG-treated group was markedly increased compared to that in the HFD-fed vehicle-treated group (Fig. 2J).

In the LFD-fed vehicle-treated group, the average level of BDNF in the hippocampal homogenates was 96.1 pg/mg of protein. BDNF protein levels were lower in the LFD-fed RSG-treated group than in the LFD-fed vehicle-treated mice. In contrast, BDNF levels in the HFD-fed RSG-treated animals were significantly increased compared to those in the HFD-fed vehicle-treated group (Fig. 3).