Six-week-old C57BL/6 mice were housed in standard cages. Room conditions were kept at a 12:12-h light/dark cycle and an ambient temperature of 22±2°C. Animals were allowed free access to food and water. All animal procedures were performed in accordance with the guidelines for the Care and Use of Laboratory Animals of Tokushima University Graduate School of Dentistry.

Mice were fed a standard laboratory diet, MF (Oriental Yeast, Tokyo, Japan). To create a hard diet, MF chow was autoclaved. The hardness of the MF chow was significantly increased by 1.5-fold by autoclave treatment (Fig. 1). Therefore, autoclaved normal diet chow was used as the hard diet to enforce mastication in this study.

The time schedule of this experiment is shown in Fig. 2. At 6 post-natal weeks, the animals were placed in standard cages (4 or 5 animals/cage). Control mice were allowed free access to the standard pellet (Fig. 2A). Mice in the experimental group were given free access to a hard diet (autoclaved normal diet) (Fig. 2B). The animals were maintained in each experimental condition for 8 weeks and then received bromodeoxyuridine (BrdU; Sigma-Aldrich, St. Louis, MO, USA) injections (50 mg/g body weight) intraperitoneally once a day for 12 consecutive days. One day after the last BrdU injection, half the mice in each group were given an overdose of anesthetic and perfused transcardially with cold 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS) (11). The remaining animals in each group were kept in their respective experimental conditions for 5 additional weeks and were perfused in a similar manner.

Throughout the experiment, the nutritional status of the mice was monitored by weighing the mice regularly once a week until the day of sacrifice.

Mice were sacrificed and perfused transcardially with cold 4% paraformaldehyde in 0.1 M PBS as previously described (11). The brains were removed from the skulls and immersed in the same fixative for 5 h at 4°C with constant rotation followed by washing with PBS overnight at 4°C. Coronal 50-μm hippocampal sections were made using a vibratome and stored at −10°C in cryoprotectant reagent (30% sucrose, 30% ethylene glycol and 0.25 mM polyvinylpyrrolidone in PBS) (11). Vibratome sections were rinsed in 1X Tris-buffered saline (TBS), blocked with 10% normal goat serum (NGS), 1X Triton X-100 in TBS for 1 h, followed by incubation with primary antibodies overnight. The following primary antibodies were used: mouse monoclonal antibodies against neuronal nuclei (NeuN; 1:1,000; Millipore) and neuronal nitric oxide synthase (nNOS; 1:1,000; Santa Cruz Biotechnology, Inc.), rat-monoclonal antibodies against BrdU (1:20; Funakoshi), rabbit-polyclonal antibodies for c-Fos (1:10,000; Santa Cruz Biotechnology, Inc.). For evaluation of the changes in neural proliferation and survival in the DG, the BrdU in vivo labeling protocol was used. BrdU is incorporated only in the cells that are dividing and is integrated in the synthesized DNA enabling them to be detected immunohistochemically. The free-floating brain slices were initially treated with denatured solution (12 N HCl + 1X TBS) and incubated for 30 min at 37°C. After washing in TBS, the sections were incubated for 1 h at room temperature with the appropriate secondary antibodies: anti-mouse Alexa 594 (1:1,000) and anti-rabbit Alexa 488 (1:1,000). Finally, nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Slices were mounted, coverslipped in ProLong^® antifade reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) and examined under an epifluorescence microscope.

Brain slices stained with anti-NeuN antibody were examined and images of the 12 sections were acquired using an epifluorescence microscope. To calculate the DG volume, the area of the granular cell layer was measured using ImageJ software and multiplied by the thickness of the sections (50 μm) according to Cavalieri’s principle (13).

The DG was examined under a conventional epifluorescence microscope after the slices were immunohistochemically treated with specific antibodies. A total of 60 slices of 50-μm coronal sections were originally stored in 5 separated tubes containing 12 slices in each tube. Sections from every 200 μm across the whole hippocampus were selected (12 sections from anterior to posterior). BrdU, c-Fos and nNOS labeled cells were manually counted in each section by a blinded observer. A ×20 objective lens was used for all counting (11). The labeled cells within the DG were examined for both the group fed the normal diet and the group fed the hard diet.

The Morris water maze task was preformed to test spatial learning. The task used a circular pool filled with water containing white ink. A platform was placed in the pool so that it was invisible at water level. The experiment was carried out after the 5 weeks following BrdU injection. Tasks were conducted between 09:00 and 15:00 during the light cycle. After training, mice underwent 3 trials/day for 10 consecutive days. During testing, mice were required to locate the platform, and escape latency time was recorded. Once mice located the platform, they were permitted to remain on it for 10 sec. If mice did not locate the platform within 90 sec, they were placed on the platform for 10 sec and then removed from the pool, as previously described (14,15).

Mice were trained in the water maze as described above and then were subjected to a probe test. Each mouse was subjected to a probe test immediately after the final training trial on the 10th day. Before beginning the probe test, the platform was removed from the pool. Mice were released from the quadrant opposite to the previous platform quadrant, and the swimming was tracked for 90 sec, as previously described (16,17).

Total RNA was isolated from 21-week-old mice. Mice were sacrificed by an overdose of anesthetic and the hippocampus was dissected in RNAlater (Sigma). Total RNA of hippocampal cells was isolated using the TRIzol reagent (Invitrogen Life Technologies) according to the manufacturer’s instructions. RNA (1 μg) was reverse- transcribed to first-strand cDNA using a PrimeScript™ RT reagent kit (Takara Bio, Inc., Shiga, Japan) according to the manufacturer’s instructions. A Thermal Cycler Dice real-time system (Takara Bio, Inc.) was used for real-time PCR (18–20). The cDNAs were amplified with SYBR^® Premix Ex Taq and specific primer pairs for fibroblast growth factor (FGF)-1 (21), FGF-2 (22), epidermal growth factor (EGF) (23), vascular endothelial growth factor (VEGF) (24,25), insulin-like growth factor-1 (IGF-1) (26), nerve growth factor (NGF), neurotrophin-3 (NT-3), brain-derived neurotrophic factor (BDNF), glial cell line-derived neurotrophic factor (GDNF), neuropeptide Y, β-tubulin III, 2′,3′-cyclic-nucleotide phosphodiesterase (CNPase), microtubule-associated protein 2 (Map2), glial fibrillary acidic protein (GFAP), nestin, glutamate receptor 1 (GluR1) (27), glutamate receptor 2 (GluR2) (27), glutamate receptor 3 (GluR3) (27), glutamate receptor 4 (GluR4) (27), N-methyl D-aspartate receptor subtype 2B (NR2B) (28) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Primers for NGF, NT-3, GDNF, β-tubulin III, CNPase, Map2, GFAP, nestin and GAPDH were designed with Perfect Real-Time Primer Design software (Takara Bio, Inc.). The primers used in this study are listed in Table I. The PCR conditions were as follows: 10 sec at 95°C, followed by 40 cycles of 95°C for 5 sec and 60°C for 30 sec, and finally, 15 sec at 95°C and 30 sec at 60°C.

The results are presented as the means ± SEM. For statistical analysis of the results, the Student’s t-test and ANOVA (Scheffe’s post hoc test) were used to compare the group fed the normal diet with the group fed the hard diet.

Mice fed a normal diet between the ages of 6 and 20 weeks showed a significant difference in body weight compared to the mice fed a hard diet at 11 weeks (Fig. 3A) (P<0.01). However, at all other time-points, the difference in body weight between the groups was not statistically significant. Similarly, the amount of food intake between the group fed the normal diet and the group fed the hard diet showed no significant differences (Fig. 3B).

Representative images of BrdU-positive cells in the hippocampus 1 day after the BrdU injection are shown in Fig. 4 for the group fed the normal diet (Fig. 4A) and group fed the hard diet (Fig. 4B). The number of BrdU-positive cells in the DG 1 day after the BrdU injection was not significantly different between the groups (Fig. 4C).

Representative images of BrdU-positive cells in the hippocampus 5 weeks after BrdU injections are shown for the group fed the normal diet (Fig. 5A) and the group fed the hard diet (Fig. 5B). There was a significantly larger number of BrdU-positive cells in the group fed the hard diet compared to the group fed the normal diet 5 weeks after BrdU injection (P<0.001) (Fig. 5C).

As shown in Fig. 6A, the hippocampal volume of the mice fed the hard diet was significantly larger than the hippocampal volume of the mice fed the normal diet when measured 5 weeks after BrdU injection (P<0.05).

The number of c-Fos-positive cells in the group fed the hard diet was greater than that in the group fed the normal diet 5 weeks after BrdU injection (Fig. 6B) (P<0.05). Similarly, the number of nNOS-positive cells in the DG was significantly increased in the group fed the hard diet compared to the group fed the normal diet 5 weeks after BrdU injection (Fig. 6C) (P<0.01).

The results of the Morris water maze experiment are shown in Fig. 7. The escape latency in each group decreased as the number of trials progressed. No significant difference in the escape latency was found between the group fed the hard diet and the group fed the normal diet from day 1 to 9 (Fig. 7A). Ten days after the beginning of the experiment, we observed a significant difference between the groups (Fig. 7A) (P<0.005). The results of the water maze probe test are shown in Fig. 7B. After 10 days of water maze experiments, the probe test was performed. When mice fed the hard diet were allowed to search for the platform, the period during which the mice focally searched around the former platform location was significantly longer than searches in other locations (P<0.01). By contrast, when mice fed the normal diet were subjected to the probe test, the only difference observed in their searches was between the time spent around the former platform location and the left side of the arena (P<0.05).

As is shown in Fig. 8, we analyzed the mRNA expression of growth factors, neurotrophic factors, NSC markers and receptors in the hippocampus between the group fed the hard diet and the group fed the normal diet at the end of the experiment. Real-time PCR analysis revealed that the mRNA expression of the growth factors, FGF-1, FGF-2, EGF, VEGF and IGF-1, was not affected by food texture (Fig. 8A). In addition, the expression of neurotrophic factors (NGF, NT-3, BDNF, GDNF and neuropeptide Y) (Fig. 8B) and NSC markers (β-tubulin III, CNPase, Map2, GFAP and nestin) (Fig. 8C) did not differ between the group fed the hard diet and the group fed the normal diet. While the expression of the receptors, GluR2, GluR3, GluR4 and NR2B (Fig. 8D), was not affected by food texture, GluR1 mRNA expression was significantly increased in the group fed the hard diet (P<0.01).