Cannabinoids promote embryonic and adult hippocampus neurogenesis and produce anxiolytic- and antidepressant-like effects

Increased proliferation of embryonic NS/PCs by HU210 and AEA.

To examine the effects of HU210 on NS/PC proliferation, cultured embryonic NS/PCs were incubated with different concentrations of HU210. With the WST-8 assay, changes in NS/PC proliferation between HU210- and vehicle-treated culture were significant at some concentrations of HU210, as evidenced by significant group effects with 1-way ANOVA (F_5,18 = 513.129, P < 0.01). Specifically, when 10 nM to 1 μM of HU210 were added to the culture medium containing the mitogenic growth factors bFGF and EGF, the WST-8 assay showed a significant increase in NS/PC proliferation (Tukey post-hoc tests, P < 0.05); 1 nM of HU210 exerted no significant effects (P = 0.072); 10 μM produced profound toxic effects on cultured NS/PCs (Figure ​(Figure2A).2A). Because HU210 can activate both CB1 and CB2 receptors, we next used the selective CB1 receptor antagonist AM281 to identify the possible involvement of CB1 in the action of HU210 on NS/PC proliferation. Although 1 nM to 1 μM of AM281 alone produced no significant effects on NS/PC proliferation, 10 nM to 1 μM of AM281 blocked the promoting effects of 10 nM to 1 μM of HU210 on NS/PC proliferation (1-way ANOVA for repeated measures, F_2,25.713 = 16.792, P < 0.01; pairwise comparisons, HU210-treated cells with or without AM281: P < 0.01) (Figure ​(Figure2A),2A), suggesting that HU210 specifically acts on CB1 receptors to promote NS/PC proliferation. While 10 μM of AM281 alone significantly inhibited NS/PC proliferation (P < 0.01), this concentration of AM281 did not exert significant effects in preventing the lethal effects of 10 μM of HU210 on NS/PCs (Figure ​(Figure2A),2A), indicating that the lethal effects of 10 μM of HU210 on NS/PC cells were caused nonspecifically or by another receptor.

To confirm the effects of 10 nM to 1 μM of HU210 on promoting NS/PC proliferation as previously assessed by the WST-8 assay, the BrdU incorporation assay was used. It measures cell proliferation by detecting dividing cells. Similar to the results of the WST-8 assay, 1-way ANOVA showed significant group effects (F_5,18 = 176.004; P < 0.01); Tukey post-hoc tests revealed that 10 nM to 1 μM of HU210 significantly increased NS/PC proliferation (P < 0.05), which was blocked by 10 nM to 1 μM of the selective CB1 receptor antagonist AM281 (1-way ANOVA for repeated measures, F_2,36 = 19.081, P < 0.01; pairwise comparisons, HU210-treated cells with or without AM281: P < 0.01) (Figure ​(Figure22B).

To determine the effects of the endogenous cannabinoid AEA on NS/PC proliferation, cultured NS/PCs were incubated with different concentrations of AEA. The WST-8 assay showed significant group effects with 1-way ANOVA (F_5,18 = 61.585, P < 0.01). Tukey post-hoc tests further showed that 1 μM to 10 μM of AEA significantly increased NS/PC proliferation (P < 0.05) in the presence of bFGF and EGF; 100 μM produced toxic effects (Figure ​(Figure22C).

To explore the possibility of whether HU210 itself is able to produce mitogenic effects, we further examined NS/PC proliferation by adding different concentrations of HU210 to the culture medium with or without the mitogenic growth factors bFGF and EGF. When bFGF and EGF were absent from the culture medium, a significant overall change in NS/PC proliferation was observed following HU210 application (F_5,30 = 219.076, P < 0.01) (Figure ​(Figure2D).2D). Specifically, 10 nM to 1 μM of HU210 without growth factors produced significant mitogenic effects on NS/PCs (Tukey post-hoc tests, P < 0.05), whereas 10 μM of HU210 killed the cells. Similar results were observed in the control culture when different concentrations of HU210 were added to the culture medium containing the mitogenic growth factors (F_5,30 = 194.429, P < 0.01; Tukey post-hoc tests, P < 0.05) (Figure ​(Figure2D).2D). Nevertheless, the basal proliferation levels with bFGF and EGF were significantly higher than those without bFGF and EGF (1-way ANOVA for repeated measures, F_1,30 = 214.703, P < 0.01; pairwise comparisons: P < 0.01) (Figure ​(Figure22D).

Intracellular signaling involved in HU210-induced NS/PC proliferation.

To investigate the mechanisms underlying the action of HU210 on NS/PC proliferation, we examined the intracellular signaling pathways. CB1 receptor stimulation activates G_i/o or G_s proteins (27, 28). To examine whether G_i/o protein mediates the effects of HU210, we added pertussis toxin, a selective blocker for G_i/o protein activation, to the culture medium 4 hours prior to HU210 treatment. Again, 10 nM to 1 μM of HU210 significantly increased NS/PC proliferation (1-way ANOVA, F_5,18 = 880.629, P < 0.01; post-hoc tests, P < 0.01 between control and each of the 3 concentrations of HU210), which was completely blocked by 100 ng/ml of pertussis (1-way ANOVA for repeated measures, F_1,18 = 41.64, P < 0.01; pairwise comparisons, HU210-treated cells with or without pertussis: P < 0.01) (Figure ​(Figure2E).2E). It has been shown that HU210 activates G_s proteins when G_i/o proteins are inhibited by pertussis toxin (27). Therefore, to determine whether the blockade effects of HU210-induced NS/PC proliferation following pertussis treatment is achieved by activation of G_s proteins, we examined the effects of cholera toxin, a G_s protein activator, on NS/PC proliferation. Incubation of NS/PCs with 1 mg/ml of cholera toxin stimulated about 14-, 80-, 90-, and 13-fold increase in cAMP accumulation in NS/PCs 0.5, 1, 2, and 24 hours after the addition of cholera toxin; cAMP production returned to the basal levels 48 hours after cholera toxin (1-way ANOVA, F_5,18 = 93.341, P < 0.01) (Figure ​(Figure2F).2F). These results indicate the effective activation of G_s proteins in NS/PCs by cholera toxin. However, there was no significant change in NS/PC proliferation 0.5, 1, 2, 24, and 48 hours after the addition of cholera toxin (1-way ANOVA, F_5,18 = 76.562, P = 0.86) (Figure ​(Figure2G).2G). These results together suggest the involvement of G_i/o proteins, but not G_s proteins, in HU210-induced NS/PC proliferation.

Since G_i/o protein activates PI3K/Akt and ERK signaling (29), which are well known to play an important role in cell growth and cell death, we studied whether HU210 could activate Akt and ERK1/2. There was no significant change in phosphorylation of phospho-Akt during the first 1 hour after HU210 application (F_4,10 = 1.693, P = 0.228) (Figure ​(Figure3A),3A), indicating that the PI3K/Akt signaling pathway is not involved in the action of HU210 on NS/PC proliferation. In contrast, changes in phosphorylation of phospho-ERK1/2 (pERK1/2) during the first 1 hour after HU210 application were dramatic at specific time points, as shown by 1-way ANOVA (with growth factors, F_4,15 = 33.698, P < 0.01; without growth factors, F_4,15 = 23.513, P < 0.01). As early as 5 minutes after addition of HU210 to culture medium with (Figure ​(Figure3B)3B) or without bFGF and EGF (Figure ​(Figure3C),3C), a 2.5-fold increase in phosphorylation of pERK1/2 was observed (P < 0.05). At 15 minutes after HU210 application, phosphorylation of pERK1/2 reached the peak level, which was about a 4-fold (with growth factors) or 7-fold increase (without growth factors) relative to control (P < 0.01). By 60 minutes after addition of HU210, phosphorylation of pERK1/2 either significantly decreased (P < 0.05) (Figure ​(Figure3B)3B) or returned to the pretreatment level (Figure ​(Figure3C).3C). We did not observe any significant changes in the total ERK1/2 during the first 1 hour after HU210 application. Thus, the significant increase in pERK1/2 in this period suggests an important involvement of ERK signaling pathway in the action of HU210 in promoting NS/PC proliferation. This hypothesis was supported by further experiments in which U0126, a specific inhibitor of the ERK pathway, was employed. Figure ​Figure3D3D shows an overall significant difference in pERK1/2 phosphorylation after application of vehicle or 100 nM of HU210 with or without 10 μM of U0126 (F_3,8 = 60.769, P < 0.01). Specifically, HU210 profoundly increased phosphorylation of pERK1/2 (P < 0.01), which was almost completely blocked by U0126 (P < 0.01). A parallel experiment demonstrated that U0126 blocked the promoting effects of 100 nM of HU210 on NS/PC proliferation (1-way ANOVA for repeated measures, F_1,17 = 6.356, P < 0.05; pairwise comparisons, HU210-treated cells with or without U0126: P < 0.05) (Figure ​(Figure33E).

HU210 and AEA do not affect neuronal differentiation of cultured NS/PCs.

To examine the effects of HU210 on neuronal differentiation of cultured NS/PCs, neurospheres were dissociated, plated, and cultured in the medium containing bFGF and EGF for 1 day and then in another medium containing different concentrations of HU210 without bFGF or EGF for 8 days. After fixation, immunofluorescence staining was performed using antibodies against the neuronal marker β-tubulin III (TuJ1), followed by Hoechst staining that detects all the cultured cells. Cell counting revealed no significant difference among the ratios of TuJ1-labeled neurons and Hoechst-labeled total cells following treatment with vehicle or 10 nM, 100 nM, or 1 μM of HU210 (1-way ANOVA, F_4,20 = 3.307, P = 0.324) (Figure ​(Figure4),4), suggesting that HU210 exerts no significant effects on neuronal differentiation of cultured NS/PCs. Similarly to HU210, AEA (1 and 5 μM) did not produce significant effects on neuronal differentiation of cultured NS/PCs (1-way ANOVA, F_2,9 = 0.177, P = 0.840) (Figure ​(Figure44B).

Increased hippocampal cell proliferation following HU210 treatment in adult rats.

BrdU labeling of dividing cells was used to test the acute effects of HU210 treatment on cell proliferation in adult hippocampus. Adult rats received a single dose of vehicle, AM281 (3 mg/kg, i.p.), or HU210 (25 or 100 μg/kg, i.p.), followed 2 hours later by BrdU administration and then perfusion 1 day later. BrdU-labeled cells showed fusiform or irregular shape and were clustered or aggregated in the SGZ (Figure ​(Figure5A)5A) throughout the whole hippocampus in all rats examined. Cell counting revealed no significant change in the number of BrdU-positive cells in the SGZ among rats treated with vehicle, AM281, or HU210 (1-way ANOVA, F_3,16 = 52.784, P = 0.58; n = 5) (Figure ​(Figure5B).5B). We then examined the effects of chronic HU210 injection on cell proliferation in adult hippocampus. Two hours after receiving the last dose of twice-daily injections of vehicle, AM281 (3 mg/kg, i.p.), or HU210 (25 or 100 μg/kg, i.p.) for 10 days, adult Long-Evans rats received BrdU administration and then were perfused 1 day later. Immunohistochemical staining showed an apparent increase in the density of BrdU-labeled cells in the SGZ following chronic administration of 100 μg/kg of HU210 (Figure ​(Figure5C).5C). One-way ANOVA revealed a significant overall difference in the mean ± SEM number of BrdU-positive cells in the SGZ (F_3,16 = 11.504, P < 0.001; n = 5) (Figure ​(Figure5D).5D). Tukey post-hoc test showed a significant increase (about 40%) in the number of BrdU-labeled cells following 100 μg/kg of HU210 (P < 0.05) but not 25 μg/kg of HU210 (P = 0.979), relative to vehicle (Figure ​(Figure5D).5D). AM281 injection seemingly decreased the number of BrdU-positive cells in the SGZ, but there was no significant difference relative to control (P = 0.099).

Increased newborn hippocampal neurons following chronic HU210 treatment in adult rats.

A recent study has demonstrated that newborn neurons in the dentate granule cell layer that had survived 4 weeks were stably integrated into the granule cell layer (30). To examine the survival, migration, and differentiation of HU210-induced newborn cells in the SGZ, we injected rats twice daily with HU210 (100 μg/kg, i.p.), AM281 (3 mg/kg), or vehicle for 10 days, followed 12 hours later by 4 BrdU injections at 12 hours intervals. One month after the last HU210, AM281, or vehicle injection, the majority of BrdU-labeled cells migrated and dispersed into the granule cell layer and showed size and morphology indistinguishable from both their neighboring granule neurons and from different treatment (Figure ​(Figure6A).6A). The number of BrdU-labeled dentate cells in HU210-treated rats was significantly higher than that in vehicle-treated rats (Student’s t test, P < 0.01; n = 5) (Figure ​(Figure6B),6B), indicating that most of chronic HU210–induced newborn cells survived. Immunofluorescence staining revealed that HU210- and vehicle-treated rats exhibited a similar proportion of BrdU/neuronal nuclear antigen (BrdU/NeuN) double-labeling cells to the total BrdU-labeled cells (Student’s t test, P = 0.977) (Figure ​(Figure6C),6C), suggesting that chronic HU210-induced newborn cells in the SGZ have neuronal differentiation ratio similar to that of vehicle-induced newborn cells in the SGZ. Nevertheless, because chronic HU210 treatment significantly increased the number of BrdU-labeled newborn cells in the dentate gyrus (Figure ​(Figure6B),6B), the total number of newborn neurons doubly labeled with BrdU/NeuN in the dentate gyrus also significantly increased following chronic HU210.

No hippocampal neuronal death following chronic HU210 treatment in adult rats.

Ample evidence has illustrated the increased hippocampal neurogenesis following ischemia, epileptic status, enriched environment, or exercise (15). It is therefore possible that increased hippocampal neurogenesis following chronic HU210 treatment in adult rats may result from the toxic effects of chronic HU210 treatment on hippocampal neurons. To explore this possibility, we examine the total number of the dentate granule and CA3 pyramidal neurons following twice-daily injections of HU210 (100 μg/kg) for 10 days. As depicted in Figure ​Figure7A,7A, HU210-treated rats did not show detectable loss of NeuN-immunopositive neurons in the hippocampus, relative to naive control rats. Stereological cell counting confirmed that no significant difference in the total number of the dentate granule cells (F_1,4 = 1.443, P = 0.782) and CA3 pyramidal neurons (F_1,4 = 5.099, P = 0.553) between naive and HU210-treated rats (Figure ​(Figure7B).7B). These results, however, do not exclude the possibility that some of NeuN-stain neurons following chronic HU210 treatment shown in Figure ​Figure7A7A are dying. Accordingly, we used TUNEL stain and Fluoro-Jade B stain to examine the degenerating hippocampal neurons (31) in rats receiving chronic HU210 treatment, with the naive rats as negative control and kainic acid–treated rats as positive control (31). We failed to detect any TUNEL- or Fluoro-Jade B–stained degenerating cells throughout the whole hippocampus in both naive rats and HU210-treated rats, whereas kainic acid–injected rats showing epileptic status exhibited numerous dying cells in the CA3 pyramidal cell layer and even dentate granule cell layer (Figure ​(Figure7,7, C and D).

Anxiolytic and antidepressant effects of chronic HU210.

Two recent studies employing novelty-suppressed feeding (NSF) tests and forced swimming test (FSTs) as measures of anxiety and depression have shown that chronic treatment with the antidepressant fluoxetine produced anxiolytic and antidepressant effects (18, 19), and the anxiolytic effects are likely achieved by promoting hippocampal neurogenesis (18). Therefore, we employed the same behavioral tests to examine the effects of chronic HU210 treatment on measures of anxiety and depression. Rats received twice-daily injections of vehicle, AM281 (3 mg/kg), or HU210 (100 μg/kg) for 10 days, followed 12 hours later by 4 BrdU injections at 12-hour intervals. Rats were subjected to behavioral testing 1 month later, based on the recent finding that hippocampal newborn neurons need 4 weeks to become functional (32). In the NSF test, 1-way ANOVA showed an overall significant difference in the latency to eat in the novel environment among the 3 groups of rats deprived of food for 48 hours (F_2,20 = 8.187, P < 0.01). As shown in Figure ​Figure8A,8A, relative to vehicle treatment, chronic HU210 (but not AM281) treatment significantly reduced the latency to eat food in the novel environment (P < 0.01). However, when returned to their home cages immediately after the test, rats receiving vehicle, chronic AM281, and chronic HU210 showed no significant difference in the latency to eat food (F_2,20 = 0.276, P = 0.762) (Figure ​(Figure8A)8A) or the amount of food consumed (F_2,20 = 0.839, P = 0.447). In the FST, there was an overall significant difference in the duration of immobility among vehicle-, AM281-, and HU210-treated rats (F_2,19 = 4.441, P < 0.05). Post-hoc test revealed that HU210 (but not AM281) significantly decreased immobility (P < 0.05) (Figure ​(Figure8B),8B), whereas neither AM281 nor HU210 produced significant effects on the number of rats climbing in the first 5 minutes in the pretest sessions of the FST (F_2,19 = 7.552, P = 0.887) (Figure ​(Figure8C).8C). Rats were killed for immunohistochemical staining after behavioral tests. The majority of BrdU-positive cells in vehicle-, AM281-, or HU210-treated rats were located in the granule cell layer, suggesting that they became granule neurons. Cell counting revealed an overall significant difference in the number of BrdU-stained cells in the dentate gyrus (F_2,19 = 3.896, P < 0.05). Post-hoc test showed results similar to those in Figure ​Figure5D:5D: namely, relative to vehicle-treated rats, HU210-treated rats displayed a significant increase (P < 0.05) in the number of BrdU-positive cells in the dentate gyrus, whereas AM281-treated rats exhibited no significant difference (P = 0.165). Thus, these data together suggest that chronic HU210 treatment promoted hippocampal neurogenesis and exerted anxiolytic- and antidepressant-like effects.

Primary NS/PC culture.

NS/PCs were isolated and propagated using a neurosphere method developed by Reynolds and Weiss (54) and modified by Gritti et al. (55). Hippocampi were dissected under a stereomicroscope from E17 Long-Evans rat embryos into HBSS without calcium or magnesium (Invitrogen Corp.). The tissues were then cut into small pieces, digested with 0.05% trypsin/0.53 mM EDTA (Sigma-Aldrich) for 10 minutes at 37°C and triturated with a fire-polished pipette into individual cells. The cells were collected by centrifugation, resuspended in DMEM/F12 medium (1:1 mixture) (Invitrogen Corp.), and gently forced through a 41-μm nylon net filter (Millipore) to form a suspension of disaggregated cells. The disaggregated cells were seeded into uncoated T25 culture flasks (TPP) at a density of 1 × 10⁵ viable cells/ml in DMEM/F12 medium supplemented with B27 supplement (Invitrogen Corp.), 20 ng/ml FGF-2 (Chemicon International), and 20 ng/ml EGF (Invitrogen Corp.). The cells were maintained in a humidified incubator at 37°C with 95% atmospheric air/5% CO₂, grown as free-floating aggregates (neurospheres), and passaged every 5–6 days. All experiments were carried out using the second passage of NS/PCs.

NS/PC proliferation assay.

Single-cell suspensions were prepared from neurospheres by enzymatic dissociation with trypsin-EDTA. The cells were then allowed to pass through a 40-μm nylon net filter and plated in 96-well microplates at a density of 1 × 10⁵ cells/ml, followed 12 hours later by the addition of reagents including 0, 1 nM, 10 nM, 100 nM, 1 μM, or 10 μM of HU210 [(6aR)-trans-3-(1,1-dimethylheptyl)-6a,7,10,10a-tetrahydro-1-hydroxy-6,6-dimethyl-6H-dibenzo[b,d]pyran-9-methanol] (dissolved in 0.1% [vol/vol] of DMSO; Tocris Bioscience) or AM281 [1-(2,4-dichlorophenyl)-5-(4-iodophenyl)-4-methyl-N-4-morpholinyl-1H-pyrazole-3-carboxamide] (dissolved in 0.1% [vol/vol] of DMSO; Tocris Bioscience), or 0, 10 nM, 100 nM, 1 μM, 10 μM, or 100 μM of AEA [N-(2-hydroxyethyl)-5Z,8Z,11Z,14Z-eicosatetraenamide] (dissolved in 0.1% [vol/vol] of ethanol; Sigma-Aldrich). The concentration range of HU210, AM281, and AEA was determined according to our pilot experiments. In separate experiments, 100 ng/ml of pertussis (Sigma-Aldrich) or 10 μM of U0126 [1,4-diamino-2,3-dicyano-1,4-bis(2-aminophenylthio) butadiene] (Cell Signaling Technology) were added 4 hours and 1 hour, respectively, prior to the addition of different concentrations of HU210. The concentrations of pertussis and U0126 were chosen based on the previous findings that these concentrations effectively inhibited G_i/o protein and ERK signaling in cultured NS/PCs (56, 57). After incubating the microplate at 37°C in 5% CO₂/95% air for 48 hours, cell proliferation was measured using the WST-8 assay with Cell Counting Kit-8 (Dojindo Molecular Technologies Inc.) or BrdU incorporation assay with Cell Proliferation ELISA BrdU kit (Roche Diagnostics Corp.). In additional experiments, 1 mg/ml of cholera toxin (Sigma-Aldrich) was added, and 0, 0.5, 1, 2, 24, or 48 hours later, the culture medium containing cholera toxin was replaced with the regular culture medium. Cell proliferation was measured 48 hours after addition of cholera toxin using the WST-8 assay. The concentration of cholera toxin was determined after a pilot experiment.

WST-8 assay.

The number of viable cells was estimated using the WST-8 assay, which provides effective and reproducible determination of the proliferative activity of NS/PCs (58). WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt] is reduced by the mitochondrial enzyme NAD-dependent succinate dehydrogenase to form a colored formazan product, which is soluble in the culture medium. The amount of formazan dye generated by the activity of the dehydrogenases in cells is known to be directly proportional to the number of living cells. To measure the proliferative activity of NS/PCs in 96-well microplates, 10 μl of the Cell Counting Kit solution were added into each well, followed by incubation of the microplates at 37°C in 5% CO₂/95% air for 5 hours. Absorbance was measured at 450 nm using a microplate reader (Molecular Devices) with a reference wavelength of 650 nm.

BrdU incorporation assay.

The cells were labeled with 10 μM of BrdU and incubated at 37°C in 5% CO₂/95% air for 4 hours. After centrifugation at 300 g for 10 minutes, the labeling medium was removed by aspiration and the cells were dried. Then, cellular DNA was denatured by adding 200 μl of FixDenat solution into each well, and the microplates were incubated at room temperature for 30 minutes. After removal of the FixDenat solution, 100 μl of anti-BrdU antibody conjugated with peroxidase solution was added to each well, followed by incubation of microplates at room temperature for 90 minutes. After 3 washes with washing solution, 100 μl of substrate solution was added to each well, and the microplates were incubated at room temperature for 40 minutes. The reaction was then stopped by adding 25 μl of 1 M H₂SO₄ solution into each well. Absorbance was measured at 450 nm using a microplate reader (Molecular Devices) with a reference wavelength of 650 nm.

Measurement of cAMP levels in cultured NS/PCs.

NS/PCs were incubated with 1 mg/ml of cholera toxin for 0, 0.5, 1, 2, 24, or 48 hours, followed by preparation of samples for the measurement of cAMP production using the Direct cAMP Enzyme Immunoassay Kit (Sigma-Aldrich) according to manufacturer’s manual.

Immunocytochemistry for cultured NS/PCs.

To detect CB1 receptor in cultured hippocampal NS/PCs, neurospheres were plated on 12-mm glass coverslips coated with 15 μg/ml of poly-l-ornithine and were cultured in DMEM/F12 medium supplemented with B27 supplement, 20 ng/ml FGF-2, and 20 ng/ml EGF for 24 hours. Cells on coverslips were fixed with 4% paraformaldehyde, rinsed in PBS, and blocked with 5% normal goat serum for 60 minutes. Then, cells were incubated in a cocktail solution containing goat anti-CB1 antibody (1:500; Santa Cruz Biotechnology Inc.) and mouse anti-nestin antibody (1:2,000; Chemicon International) at 4°C overnight. The cells were washed in PBS 3 times and incubated in another cocktail solution containing Alexa Fluor 488–conjugated rabbit anti-mouse IgG (1:400) and Alexa Fluor 568-–conjugated rabbit anti-goat IgG (1:400) secondary antibodies (Invitrogen Corp.) at room temperature for 3 hours. Cell nuclei were stained in PBS containing 5 μg/ml of Hoechst 33258 (Sigma-Aldrich) at room temperature for 20 minutes. The coverslips were washed and mounted with antifade medium (DakoCytomation). Immunohistochemical controls for CB1 antibody were performed using antiserum preabsorbed with the immunogen (1.0 mg/ml).

For differentiation experiments, dissociated cells were plated on 12-mm glass coverslips coated with 15 μg/ml of poly-l-ornithine within 24-well culture plate at a density of 3 × 10⁴ cells per well and were cultured in DMEM/F12 medium supplemented with B27 supplement, 20 ng/ml FGF-2, and 20 ng/ml EGF for 12 hours. Then, the culture medium was changed to FGF-2– and EGF-free DMEM/F12 medium containing B27 supplement and HU210 (10 or 100 nM or 1 μM) or AEA (1 or 5 μM). Cells were cultured further for 8 days, with complete change of culture medium and drugs once every 3 days. After fixation with 4% paraformaldehyde, cells were rinsed in PBS and incubated in 5% normal goat serum for 60 minutes before incubation in mouse anti-TuJ1 antibody (1:500; Sigma-Aldrich) at 4°C overnight. The primary antibody was visualized by incubation of cells in Alexa Fluor 568–conjugated goat anti-mouse secondary antibody (1:400; Invitrogen Corp.) at room temperature for 3 hours. Cell nuclei were stained with Hoechst 33258 solution for 20 minutes. The number of TuJ1-positive cells and Hoechst-stained total cells was counted using a 40× objective lens on an Olympus BX-51 light microscope. Cell counting was conducted in 10 randomly chosen fields in each coverslip by an observer who had no knowledge of the treatment conditions.

Western blotting.

NS/PCs were plated on 10-cm culture dishes coated with 15 μg/ml of poly-l-ornithine at 1 × 10⁶ cells/dish for 24 hours. Then, 100 nM of HU210 or vehicle was added to the culture medium, and the cells were collected 0, 5, 15, 30, and 60 minutes later. Total cellular proteins were extracted with lysis buffer (1% Triton X-100, 10% glycerol, 20 mM Tris-HCl, pH 7.4, 1 mM EDTA, 1 mM sodium orthovanadate, 1 mM sodium fluoride, and protease inhibitor mix). Equivalent amounts of extracted proteins (20 μg) were resolved on 10% SDS-PAGE and electroblotted at 4°C for 50 minutes to nitrocellulose membrane (Amersham Biosciences). After blocking the background staining with 5% (wt/vol) skim milk in PBS, the membranes were incubated in goat anti-CB1 antibody (1:500; Santa Cruz Biotechnology Inc.), rabbit anti-pAkt (1:1,000; Alpha Diagnostic International), mouse anti-pERK1/2 (1:1,000; Santa Cruz Biotechnology Inc.), and rabbit anti–total ERK1/2 (1:1,000; Cell Signaling Technology). Antibody against β-actin (1:1,000; Sigma-Aldrich) was used as an internal control for the concentration of protein loaded. Immunoreactive proteins were detected using HRP-conjugated secondary antibodies and an ECL kit (Amersham Biosciences) according to the manufacturer’s instructions. Quantification of the immunoblots was performed by densitometric analysis of chemiluminescence-exposed films, using Image-Pro Plus software (version 4.1; Carsen Group). Immunohistochemical controls were performed by using antiserum preabsorbed with the immunogen (1.0 mg/ml).

PCR.

NS/PCs were collected, and rat CB1 full encoding region was amplified in NS/PCs by PCR with primers (sense: 5′-TGGATCCATGAAGTCGATCCTAGATGGCC-3′; antisense: 5′-CGAATTCTCACAGAGCCTCGGCGGACGTG-3′). The amplification reaction was carried out in a PerkinElmer GeneAmp (PCR System 9600) for 30 cycles. Each cycle consisted of denaturation for 40 seconds at 94°C, annealing for 40 seconds at 55°C, and an extension for 120 seconds at 72°C. A final extension step at 72°C for 5 minutes terminated the amplification. PCR product was analyzed by electrophoresis using a 1% agarose gel.

HU210, AM281, and kainic acid treatment in adult rats.

Both HU210 and AM281 were dissolved in pure DMSO, and aliquots were stored at –80°C with the final working concentration of 1 ml/kg body weight. Two doses of HU210 were used, 25 or 100 μg/kg, which produce either nondetectable or severe behavioral changes, respectively, including ataxic walking and deep sleep. Only 1 dose of AM281, 3 mg/kg, was employed, because this dose effectively antagonizes the behavioral effects produced by HU210 (8). Two injection protocols were utilized. To detect the acute effects of cannabinoid on cell proliferation in the hippocampus, rats received a single injection of vehicle, 25 or 100 μg/kg of HU210, or 3 mg/kg of AM281. To examine the chronic effects, rats received twice daily injections of vehicle, 25 or 100 μg/kg of HU210, or 3 mg/kg of AM281 for 10 days. Three rats received kainic acid injection (10 mg/kg, i.p.), and 1 hour after epileptic status, rats received sodium pentobarbital to stop behavioral seizures (31). One day after kainic acid injection, rats were killed for TUNEL and Fluoro-Jade B staining.

BrdU labeling in adult rats.

Male Long-Evans, Wistar, and Fischer 344 rats weighing 220–250 g received i.p. injections of 50 mg/kg of BrdU (Sigma-Aldrich). Three BrdU injection paradigms were used. To identify the possible expression of CB1 receptors in dividing cells in the hippocampal dentate gyrus, rats received a single dose of BrdU 2 hours after HU210 and were perfused 2 hours later. To examine cell proliferation in the hippocampus, rats received 2 doses of BrdU (2 hours after the last HU210) with a 2-hour interval and were killed 24 hours after the last BrdU injection. To investigate differentiation of newborn cells in the hippocampus, rats received 4 doses of BrdU (12 hours after the last HU210) with 12-hour intervals and were allowed to survive for 1 month after the last BrdU injection.

Immunohistochemistry for adult rats.

Under anesthesia with sodium pentobarbital (50 mg/kg, i.p.), rats were perfused transcardially with 0.9% saline, followed by 4% paraformaldehyde in 0.1 M PBS (pH 7.4). The brains were removed, post-fixed for 2 hours in the same fixative, and placed in 20% sucrose until they sank. Coronal sections (40-μm thickness) through the entire hippocampus were cut on a sliding microtome and stored in PBS. Immunohistochemical detection of single BrdU labeling was conducted with previously described protocols (16). Briefly, free-floating sections were pretreated in 50% formamide/2× SSC buffer (0.3 M NaCl, 0.03 M sodium citrate) at 65°C for 2 hours and were incubated in 2 M HCl at 37°C for 30 minutes. After a 10-minute wash in 0.1 M borate buffer (pH 8.5) to neutralize the HCl, sections were incubated in mouse anti-BrdU antibody (1:1,000; Sigma-Aldrich) at 4°C for 36 hours. Sections were washed again and incubated in biotinylated goat anti-mouse secondary antibody (1:250; Vector Laboratories) for 2 hours at room temperature, followed by wash in PBS and an incubation in avidin-biotin-peroxidase complex for 2 hours. After a final wash, sections were reacted for peroxidase enzyme activity by using 3,3′-diaminobenzidine. Immunohistochemical detection of single NeuN labeling using mouse anti-NeuN antibody (1:2,000; Chemicon International) was conducted with protocols similar to those described above, except pretreatment with formamide and HCl was omitted. The specificity of immunolabeling was verified by controls in which the primary antibody was omitted.

For double immunofluorescence staining, DNA in the sections was first denatured to expose the antigen, as described above. Sections were then incubated at 4°C for 36 hours in a cocktail solution containing sheep anti-BrdU antibody (1:500; Cedarlane Laboratories Ltd.) and mouse anti-NeuN antibody (1:500; Chemicon International). After multiple washes in PBS, sections were incubated for 4 hours in Alexa Fluor 488–conjugated donkey anti-sheep secondary antibody preabsorbed with goat IgG (1:400; Invitrogen Corp.) and Alexa Fluor 568–conjugated goat anti-mouse secondary antibody (1:400; Invitrogen Corp.). Sections were mounted on gelatinized slides and coverslipped.

TUNEL stain and Fluoro-Jade B stain.

Three groups of 3 rats each receiving no treatment (control), chronic HU210 injection, and kainic acid injection were perfused with 4% paraformaldehyde as described above. Several series of coronal sections (40-μm) through the hippocampus were cut on a sliding microtome. One series of sections was stained with Fluoro-Jade B stain according to the manufacturer’s instruction (Histo-Chem Inc.) and our previous study (31). Adjacent sections were stained with a TUNEL staining kit according to the manufacturer’s instruction (Trevigen Inc.).

Quantification of BrdU and NeuN labeling.

Every sixth section through the dorsal hippocampus was processed for BrdU immunohistochemistry, resulting in each section being 240 μm apart, thus ensuring the same cell would not be counted twice. All BrdU-positive cells in the entire granule cell layer (superior and inferior blades), SGZ (defined as a 2-cell-body-wide zone along the border of the granule cell layer), and the hilus were counted in each section by an experimenter who was not informed of the group assignment. The total number of BrdU-labeled cells per section was then averaged. This cell profile counting method is theoretically comparable to the true stereology cell counting, because we conducted random sampling (i.e., every sixth section was sampled), avoided double counting of the same cells (each section was 240 μm apart), and counted all the cells labeled with BrdU regardless of their shape and size or of the volume of the dentate gyrus. Two recent studies have used similar cell counting methods (59, 60), which showed results indistinguishable from those provided by stereology cell counting (61).

For quantification of BrdU/NeuN double-labeling cells, immunofluorescence images were obtained under a confocal laser-scanning microscope (LSM 510 META; Zeiss). Fifty BrdU-positive cells in each rat were randomly selected and then analyzed by orthogonal reconstructions from z-series (z-step, 1 μm) to obtain the proportion of BrdU/NeuN double-labeling cells to BrdU-stained cells.

Stereological cell counting was performed for quantification of the total number of NeuN-stained neurons in the hippocampus of 2 groups of 3 rats each receiving no treatment (control) or chronic HU210 injection. We used an Olympus BX-51 microscope interfaced with a color digital camera (MicroFire; Optronics) and analyzed using the image analysis computer software Stereo Investigator (version 4.10; MicroBrightField Inc.). The analysis of the sections started from a random position at the origin of the hippocampus. Every sixth section was included such that the rostral 7 sections through the dorsal hippocampus in each animal were analyzed, and these 7 sections cover over 90% of the dorsal hippocampus. The contour delineations (left and right) of the dentate granule cell layer and CA3 pyramidal cell layer medial to an artificial line connecting the tips of the 2 granule cell blades were drawn using a 2.5× lens. Cells were counted using the optical fractionator method (62). Each counting frame was placed at an intersection of the lines forming a virtual grid that was randomly generated and randomly placed by the software within the granule or pyramidal cell layer. The software also controlled the motorized stage of the microscope that allowed scanning of the entire brain region by successively meandering from one counting frame to the other. Cells were counted using a 100× oil lens (NA 1.40) and were included in the measurements only when they came into focus within the dissector (dissector height, 11 μm; average thickness of mounted sections was 13 μm). The software calculated the estimated total number of cells in each brain structure. The precision of the individual estimations is expressed by the coefficient of error. This value expresses the intra-individual variation due to the stereological estimating procedure (i.e., sampling of sections and counting locations), thus providing the information necessary for determining whether more or less sampling should be performed. The inter-animal variation in total cell number is given by the coefficient of variation. Counting was performed blind to the treatment, and the sections were decoded only after completion of the analysis.

Behavioral testing.

Rats received twice-daily injections of vehicle, AM281 (3 mg/kg), or HU210 (100 μg/kg) for 10 days, followed 12 hours later by 4 BrdU injections at 12 hour intervals. Rats were subjected to behavioral testing 1 month later. NSF testing was conducted according to a previously described protocol (63). Food was removed from the home cage 48 hours before testing, but water was provided. Prior to testing, a single pellet of food weighing 3.5–4.5 g was placed on a platform in the center of the testing plastic box (52 × 48 × 32 cm) with the floor covered with a thin layer of wooden bedding. Rats were placed individually in a corner of the box and a stopwatch immediately started. The measure of interest was the latency to begin eating, defined as chewing of the food, not simply sniffing or playing with it. If animals had not eaten within 600 seconds, the test was terminated and the animals were assigned a latency score of 600 seconds. Immediately after the test, rats were transferred to their home cages, and the latency to feed and the amount of food consumed in 15 minutes were measured, according to our pilot experiments. The whole process was videotaped and the latency to begin eating in the novel environment was measured by 2 experimenters blind to the treatment conditions.

For the FST, rats were placed individually in the testing plastic cylinder (54 cm high × 35 cm in diameter) containing a 38-cm water column (22 ± 1°C), according to previous studies (41, 42). Water was replaced between every trial. Two swimming sessions were conducted: an initial 15-minute pretest, followed by a 5-minute test 24 hours later. Following swim sessions, rats were removed from the cylinder, dried with a towel, and placed underneath a heating lamp for approximately 30 minutes before being returned to their home cages. Both pretest and test sessions were videotaped. Two observers blind to the treatment conditions scored the number of rats climbing in the first 5 minutes in the pretest session and for time spent immobile in the test session. Immobility was defined as floating passively in the water and only making slight movements to keep the head above the water.

Irradiation.

Three groups of 5 rats each received twice-daily injections of vehicle (group 1), HU210 (group 2; 100 μg/kg), or HU210 (group 3; 100 μg/kg) for 10 days, followed 12 hours later by 4 BrdU injections at 12-hour intervals. All the rats were subjected to behavioral testing 1 month later. Group 3 rats also received irradiation using a Philips RT250 orthovoltage unit operated at 250 kV and 15 mA. Pre-irradiation dosimetry was performed using a PTW Unidos Electrometer type 10002 with Capintec Inc. Ionization Chamber model PR-06G. According to the method of Santarelli et al. (18), anesthetized rats were protected with lead shields that covered the entire body except the hippocampus. Dose rate measurements were performed with the ionization chamber (including its buildup cap) placed on the Perspex sheet. Readings were corrected for air temperature and pressure and indicated a dose rate of 67.2 cGy/min at a source-to-surface distance of 45 cm. The total dose was 5 Gy. Two 5 Gy doses were delivered 7 days apart, and the first dose was given 1 day after the first HU210 injection.

This work was supported by grants from the Canadian Institutes of Health Research (CIHR) and the Heart and Stroke Foundation of Saskatchewan to X. Zhang, who is the recipient of the CIHR New Investigator Award. W. Jiang and S.-P. Ji were supported by Postdoctoral Fellowship Award from the Saskatchewan Health Research Foundation. We thank Y. Li and G. Kort for technical assistance.