**3. Results**

### *3.1. Neuroprotective E*ff*ect of 3,3*- *-Diindolylmethane on Glutamate-Treated HT-22 Cells*

DIM possesses some beneficial properties, such as antioxidant action [4] and neuroprotective action [7,8]. Considering this, we investigated the effect of DIM on glutamate-induced cytotoxicity in HT-22 cells. When HT-22 cells were incubated with DIM prior to glutamate exposure, DIM at the concentrations used (10–80 μM) enhanced cell viability up to a concentration of 80 μM (Figure 1A). In addition, DIM dose-dependently reduced the ROS level and also restored the GSH level in the cells

(Figure 1B,C). A DIM concentration of 40 μM was sufficient to restore the level of ROS or GSH to that of the control group. These results suggest that DIM exerts neuroprotective activity by suppressing oxidative stress in hippocampal neuronal cells.

**Figure 1.** Effect of 3,3- -diindolylmethane (DIM) on glutamate-induced cytotoxicity, and reduction of ROS and glutathione levels in HT-22 cells. HT-22 cells, seeded on a 96 well-plates and incubated for 24 h, were incubated with or without DIM (0–80 μM) for 30 min before glutamate challenge (5 mM). After 12 h, cell viability, ROS level, and GSH level were measured as described in Materials and Methods. (**A**) Cell viability, (**B**) ROS level, and (**C**) GSH level. Data are the mean ± SD values of triple determinations. \*\* *p* < 0.01 versus glutamate-treated group. – is absence, + is presence.

### *3.2. Inhibitory E*ff*ect of 3,3*- *-Diindolylmethane on Oxidative Stress-Induced Apoptosis*

After we found that DIM protects hippocampal neuronal cells against oxidative stress, we investigated the effect of DIM on the expression of apoptosis-related proteins in oxidative stress-exposed neuronal cells. In the previous report, we found that the elevation of ROS formation activated apoptosis signaling pathway in neuronal cells [12]. DIM downregulated the expression of pro-apoptotic factors such as Bax, cytochrome c, cleaved caspase-3, and AIF, whereas it upregulated that of Bcl-2, an anti-apoptotic factor (Figure 2). These findings suggest that DIM protects hippocampal neuronal cells against oxidative stress-induced apoptosis by regulating the expression of apoptosis-related proteins.

**Figure 2.** Suppressive effect of 3,3- -diindolylmethane on glutamate-induced apoptosis in HT-22 cells. HT-22 cells were seeded on a 60 mm dish, and then incubated for 24 h. The cells were challenged with glutamate after preincubation with or without DIM (0–40 μM) for 30 min. After 12 h, the expression of Bcl-2, Bax, cytochrome c, cleaved caspase-3, AIF, or β-actin was examined as described in Materials and Methods. The data were contained from three independent experiments. \*\* *p* < 0.01 versus glutamate-treated group. – is absence, + is presence.

### *3.3. Activatory E*ff*ect of 3,3*- *-Diindolylmethane on Both TrkB*/*CREB*/*BDNF Pathway and Akt*/*Nrf2*/*ARE Pathway*

It was reported that activation of the TrkB/CREB/BDNF pathway and the Akt/Nrf2/antioxidant response element (ARE) pathway contributed to the neuroprotective action of *N*-acetyl serotonin [12]; therefore, we supposed that both pathways could be activated by DIM, which was found to protect hippocampal neuronal cells against oxidative stress-induced apoptosis. DIM dose-dependently improved the expression of p-TrkB, p-CREB, BDNF, or p-Akt and antioxidant enzymes such as HO-1, GCLC, and NQO-1 in glutamate-treated HT-22 cells (Figure 3). In addition, DIM also promoted the nuclear translocation of Nrf2 in glutamate-treated HT-22 cells. Especially, DIM at 40 μM almost fully restored the expression of all the proteins. Based on these results, we suppose that DIM is able to promote the activation of both the TrkB/CREB/BDNF pathway and the Akt/Nrf2/ARE pathway in oxidative stress-exposed HT-22 cells. Therefore, the neuroprotective action of DIM may be closely associated with the activation of both signaling pathways.

**Figure 3.** Effect of 3,3- -diindolylmethane on phosphorylation of TrkB, Akt, or CREB, and expression of BDNF, Nrf2, or antioxidant enzymes. The experiments were performed as described in the Figure 2 legend. The data were obtained from three independent experiments. (**A**) p-TrkB, p-Akt, p-CREB, and BDNF and (**B**) nuclear and cytosolic Nrf2, HO-1, GCLC, and NQO-1. \*\* *p* < 0.01 versus glutamate-treated group. − is absence, + is presence.

### *3.4. Suppressive E*ff*ects of K252a and MK-2206 on Neuroprotective Action of 3,3*- *-Diindolylmethane*

Subsequently, we attempted to clarify the mechanism by which DIM promoted the activation of the TrkB/CREB/BDNF pathway and the Akt/Nrf2/ARE pathway in oxidative stress-exposed hippocampal neuronal cells. For this, we investigated the suppressive effects of K252a, a TrkB inhibitor, and MK-2206, a selective Akt inhibitor, on the neuroprotective action of DIM in oxidative stress-exposed hippocampal neuronal cells. When DIM was preincubated with HT-22 cells in combination with K252a or MK-2206 prior to glutamate treatment, the inclusion of either K252a or MK-2206 significantly attenuated the preventive action of DIM against oxidative stress-induced cell death (Figure 4A,B). In addition, the inhibitors reversed the inhibitory effect of DIM on the generation of ROS in glutamate-treated HT-22 cells (Figure 4C,D). Overall, these results support the notion that DIM protects hippocampal neuronal cells against oxidative stress by promoting the expression of both BDNF and antioxidant enzymes via activation of both the TrkB/CREB/BDNF pathway and the Akt/Nrf2/ARE pathway.

**Figure 4.** Inhibitory effect of K252a or MK-2206 on neuroprotective action of DIM. HT-22 cells were preincubated with or without DIM in combination with K252a or MK-2206 for 30 min before glutamate challenge. After 12 h, cell viability and ROS level were measured as described in Materials and Methods. (**A**), (**B**) Cell viability and (**C**), (**D**) ROS level. Data are the mean ± SD values of quintuple determinations. \*\* *p* < 0.01 versus glutamate-treated group; # *p* < 0.05 and ## *p* < 0.01 versus DIM with glutamate-treated group.

### *3.5. Improving E*ff*ect of 3,3*- *-Diindolylmethane on Scopolamine-Induced Memory Impairment in Mice*

Finally, to confirm the neuroprotective action of DIM in an in vivo system, we further investigated the neuroprotective action of DIM in scopolamine-treated mice. When mice were orally administrated with DIM (10 or 20 mg/kg) before scopolamine challenge, DIM at 20 mg/kg decreased the time of escape latency and increased the number of platform area crossings and latency time in scopolamine-treated mice (Figure 5). However, DIM at 10 mg/kg did not improve the latency time in the mice, which suggests that the administration of DIM at 10 mg/kg failed to reach the effective concentration in brain because of the rapid metabolism of DIM [25]. Consistent with its antioxidant action in an in vitro system, DIM reduced lipid peroxidation and also improved the level of GSH and the activities of GR and GPx in brain tissue (Figure 6A–D). Additionally, DIM reduced AChE activity and enhanced ChAT activity compared with the scopolamine group (Figure 6E,F), further supporting the antioxidant action of DIM. Also, DIM prevented scopolamine-induced oxidative damage of neuronal cells in the hippocampal CA1 and CA3 regions of mice brains (Figure 7). These findings suggest that DIM improves scopolamine-induced memory impairment by protecting neuronal cells against oxidative damage in the hippocampus of mice. Based on this, we suggest that the neuroprotective action of DIM may be largely ascribed to the activation of both the TrkB/CREB/BDNF pathway and the Akt/Nrf2/ARE pathway.

**Figure 5.** Inhibitory effect of 3,3- -diindolylmethane on memory impairment induced by scopolamine in mice. Mice were orally administrated with DIM (10 or 20 mg/kg) before scopolamine treatment (2 mg/kg, i.p.). After 1 h, the mice were tested for Morris water maze or passive avoidance (Acquisition trial). (**A**) Escape latency time, (**B**) number of platform area crossings, and (**C**) latency time. Data are the mean ± SEM values of sextuple determinations. \* *p* < 0.05 and \*\* *p* < 0.01 versus scopolamine-treated group.

**Figure 6.** Effect of 3,3- -diindolylmethane on antioxidant biomarkers and cholinergic enzymes in brain tissue of mice treated with scopolamine. Experiments were performed as described in the Figure 5 legend. Lipid peroxidation, GSH, and the activities of GPx, GR, AChE, or ChAT in brain tissues were determined as described in Materials and Methods. (**A**) Lipid peroxidation, (**B**) GSH, (**C**) GPx activity, (**D**) GR activity, (**E**) AChE activity, and (**F**) ChAT activity. Data are the mean ± SEM values of sextuple determinations. \* *p* < 0.05 and \*\* *p* < 0.01 versus scopolamine-treated group.

**Figure 7.** Neuroprotective effect of 3,3- -diindolylmethane on hippocampal CA1 and CA3 regions of mice treated with scopolamine. Experiments were performed as described in the Figure 5 legend. Neuronal cell staining of brain tissues was carried out as described in Materials and Methods. The results were obtained from three independent experiments. (**A**) CA1 region and (**B**) CA3 region. \*\* *p* < 0.01 versus scopolamine-treated group.
