Eight-week-old male C57BL/6 mice were obtained from Orient Bio. (Korea). The mice were fed a standard animal diet (Samyang Feed, Korea). All animal experiments followed a protocol approved by the Committee for Animal Experimentation at Chonnam National University, Korea.

Fast neutrons were generated by the Korea Cancer Center Hospital (Korea) cyclotron using a proton-beryllium reaction. The estimated forward neutron spectrum had a peak energy of 35 MeV. For fast neutrons, the mean dose rate was 94 mGy/min. The irradiation doses were measured with an IC-17 ionization chamber (Far West Technology, USA) and a RDM 2A electrometer (Therados, Sweden).

Behavioral dysfunction of the mice was evaluated after cranial irradiation with 0.8 Gy of fast neutrons by open field analysis (n = five mice/group), object recognition memory test (n = seven mice/group), and contextual fear conditioning (n = seven mice/group) on 1, 7, and 14 days post-irradiation (p.i.). Sham-irradiated control mice were also transported to the irradiation facility but were not irradiated (1 day p.i.).

To observe the detrimental effect of fast neutrons on neurogenesis in adult hippocampi, the mice were cranially irradiated with 0.8 Gy of fast neutrons. The mice were sacrificed 24 h later and hippocampi from the mice in each group were removed (n = three mice/group). Samples of the hippocampi were fixed in 4% paraformaldehyde (Sigma-Aldrich, USA) in phosphate buffered saline (PBS, pH 7.4) and embedded in paraffin for histological analysis.

Open field analysis was used to measure the activity of the mice in a novel environment. Total moving distance (cm), movement time (sec), number of movement episodes, and rest time (sec) were determined over 5 min using the TruScan Photo Beam Activity System (Coulbourn Instruments, USA).

Eight-week-old mice were acclimated to the training/testing acryl chamber (42 cm length, 28 cm width, and 20 cm height) for 24 h. The objects used for recognition testing were made from plastic and had three different shapes: cubes, pyramids, and cylinders. Each object was 3.5 cm high and could not be displaced by the mice. The chamber area and objects were cleaned with 75% ethanol between trials to prevent the buildup of olfactory cues. During training, two randomly selected objects with different shapes were presented to each mouse for 10 min. At 24 h after training, another set of objects (one previously presented object and one novel object) was presented to the trained mice. For example, if the cube- and pyramid-shaped objects were presented during training, the cylinder-shaped object was used as a novel object during testing. Interactions of the mouse with each object, including approaches and sniffing, were scored [7,10,16,17]. If the mouse retained the memory of a previously encountered object, it would show preference for the novel object during testing. The percentage preference was defined as the number of interactions with a specific object divided by the total number of interactions with both objects.

During training for fear conditioning, mice were placed in an operant chamber with a metal grid floor (TruScan chamber with photobeam sensors; Coulbourn Instruments, USA) for 30 sec after which a mild electric foot shock was delivered (0.3 mA for 1 sec). At 24 h after training, mice were tested and scored to determine the percentage of freezing behavior (defined as immobility except for breathing) every 2 sec for 2 min.

Coronal sections (5 µm-thick) were cut using a microtome (SM2010R; Leica, Germany) and deparaffinized using routine protocols before being incubated citrate buffer (0.01 M, pH 6.0) and heated in an autoclave for 10 min. All subsequent steps were performed at room temperature. The sections were treated with 0.3% hydrogen peroxide in methyl alcohol for 20 min to block endogenous peroxidase activity. After three washes with PBS, the sections were blocked with 10% normal goat serum (Vector Laboratories, USA) in PBS for 1 h and then incubated for 2 h with immunohistochemical markers for proliferating cells (monoclonal rabbit anti-Ki-67 antibody, 1 : 200 dilution; Acris Antibodies GmbH, Germany) and immature progenitor cells (polyclonal rabbit anti-DCX antibody, 1 : 400 dilution; Cell Signaling Technology, USA). After three washes in PBS, the sections were incubated with a 1 : 100 dilution of biotinylated goat anti-rabbit IgG (Vector Laboratories, USA) for 45 min. After three washes in PBS, the sections were incubated for 45 min with an avidin-biotin peroxidase complex (Vector Laboratories, USA) prepared according to the manufacturer's instructions. The sections were then washed three times in PBS, and antibody binding was detected by incubating for 3 min in a diaminobenzidine substrate (Vector Laboratories, USA) prepared according to the manufacturer's instructions. As a negative control, the primary antibodies were omitted when processing a few test sections from each experiment. After completion of color development, the sections were counterstained with Harris's hematoxylin (Sigma-Aldrich, USA) for 20 sec, washed in running tap water for 20 min, dehydrated by incubating in a graded series of ethanol solutions, washed with xylene, and mounted with Canadian balsam (Sigma-Aldrich, USA).

The numbers of cells showing the specific characteristics of proliferating (immunopositive for Ki-67) and immature progenitor (immunopositive for DCX) cells in the hippocampus were scored using a histomorphometric approach [6,7,10,16,17]. The observer was blind to the sample identities. Brain samples from each mouse were taken approximately 2.12 mm caudal to the bregma. For each mouse, a standardized counting area included the DG regions in three 5 µm-thick coronal sections chosen from a one-in-ten series of sections representing the rostral/mid-hippocampus. One coronal section was chosen from each of the three regions of the hippocampus (about 50 µm apart) such that the sections did not overlap. All positively immunolabeled cells within the subgranular zone of the supra- and infra-pyrimidal blades of the DG were counted. The number of immunopositive cells was averaged across the DG region in each of the three brain sections so that the mean number of immunopositive cells in the three sections of each mouse was n = 1.

Data are reported as the mean ± SE and were analyzed using a one-way analysis of variance (ANOVA) followed by Student-Newman-Keuls post hoc test for multiple comparisons. In all cases, a p-value < 0.05 was considered significant.

The basal locomotor activity of sham-irradiated controls and mice 1, 7, and 14 days p.i. was examined in a novel environment using open field analysis (n = five in each group) because different levels of anxiety can affect the motivation and performance in learning and memory tests. There were no significant differences in movement distance (p = 0.854 [1 day], p = 0.521 [7 days], and p = 0.922 [14 days] vs. the sham-irradiated controls), movement time (p = 0.741 [1 day], p = 1.000 [7 days], and p = 0.587 [14 days] vs. the sham-irradiated controls), movement episodes (p = 0.158 [1 day], p = 0.347 [7 days], and p = 0.129 [14 days] vs. the sham-irradiated controls), and resting time (p = 0.741 [1 day], p = 1.000 [7 days], and p = 0.587 [14 days] vs. the sham-irradiated controls) between the sham-irradiated controls and mice 1, 7 and 14 days p.i. All mice had comparable moving distances, movement times and episodes, and resting times (Table 1). This suggested that acute cranial irradiation with 0.8 Gy of fast neutrons did not alter basal locomotor activity that can contribute to potential differences in hippocampus-related learning and memory behavior test results.

Mice (n = seven in each group) were examined using the object recognition memory sensitive hippocampus-dependent paradigm [7,10,16,17]. The sham-irradiated controls and mice examined 1, 7, and 14 days after acute irradiation displayed an equal preference for a pair of two objects with different shapes during training (Fig. 1A). Object recognition memory deficits were evident during testing in mice trained 1 day (p < 0.01 vs. the sham-irradiated controls) and 7 days (p < 0.05 vs. the sham-irradiated controls) p.i. whereas the mice trained 14 days p.i. did not show any object recognition memory deficit (Fig. 1B). During the test, preferences (mean ± SE) for a novel object were 68.6 ± 1.9% for the sham-irradiated controls, 57.6 ± 2.2% for the mice trained 1 day p.i., 56.0 ± 4.9% for mice trained 7 days p.i., and 64.6 ± 2.7% for mice trained 14 days p.i. There was no significant difference in the total number of interactions during training between the sham-irradiated control and mice examined on 1 (p = 0.105), 7 (p = 0.568), and 14 (p = 0.718) days p.i. (Table 2). This suggests that all mice had comparable levels of attention, motivation, and visual perception.

Contextual fear conditioning, another hippocampus-related learning paradigm [10], was used to examine seven mice in each group. During training, both sham- and fast neutron-irradiated mice displayed minimal freezing (Fig. 2). Sensitivity to electric foot-shock was further tested and no significant differences were observed in the threshold current required to elicit stereotypic responses, including flinching, vocalization, and jumping/vocalization, between the sham- and fast neutron-irradiated mice (data not shown). This finding suggests that the mice had comparable levels of sensitivity to the electric foot-shock. The memory retention trials (testing) were carried out 24 h after the acquisition trial (training). Sham-irradiated mice displayed a significant increase in freezing (61.3 ± 6.2%; mean ± SE) when tested 24 h after training (1 day after sham irradiation with 0 Gy; Fig. 2). The mice trained 1 and 7 days p.i. displayed significantly lower levels of freezing during the test (1 day p.i., 35.2 ± 5.8%, p < 0.01 vs. the sham-irradiated control; 7 days p.i., 26.3 ± 2.8%, p < 0.001 vs. the sham-irradiated control; Fig. 2). However, the mice trained 14 days p.i. did not show a significant deficit in contextual fear conditioning (Fig. 2).

Ki-67- (Figs. 3A and B; 19.0 ± 1.5 cells/DG in three mice) and DCX-positive cells (Figs. 3C and D; 62.7 ± 6.7 cells/DG in three mice) were consistently observed in the hippocampal DG of the adult control mice. The number of Ki-67-positive cells in the DG markedly declined 24 h following irradiation with 0.8 Gy of fast neutrons (3.7 ± 0.7 cells/DG, p < 0.001 vs. the sham-irradiated control; Figs. 3A and B). In comparison, the number of DCX-positive cells in the DG declined 24 h p.i. (18.3 ± 1.5 cells/DG, p < 0.01 vs. the sham-irradiated control; Figs. 3C and D).