Five weeks-old male Wistar rats weighing 160~180 g were used. Immobilization was used to induce chronic stress according to the previously described method [8,13,14]. Fluoxetine·HCl (10 mg/kg/day, p.o.) or DHED (10 mg/kg/day, p.o., Jeil Pharm., Korea) dissolved in 0.3% CMC (carboxymethylcellulose, Sigma, USA) solution was administered via oral route to animals according to the treatment schedule in Fig. 1A (n=4~5 per group). All behavior tests were scored by an investigator blinded to treatment conditions and all manipulations were performed in accordance with the Guidelines for Animal Experiments of Ethics Committee of Seoul National University.

Locomotion was measured after stress. Rats were freely placed in the center of the 40×50×60 cm open filed divided into 30 equal squares by black lines. The numbers of crossings of the black lines were used as measures of locomotion for 5 min [10]. Y-maze test was used for testing spatial working memory during the 8-min testing session according to the previously described method [10]. The apparatus was an opaque plastic maze and each arm was 80 cm long, 35 cm high, and 15 cm wide, and converged at an equilateral angle. Food and water were restricted. Three continuous entries into three different arms were counted as an alternation. Forced swimming test has been widely used to evaluate potential effectiveness of antidepressant drugs according to the previously described method [15]. The rats were gently placed individually in plexiglass cylinders (40 cm height; 18 cm in diameter, 30 cm water depth), and contained water was at a temperature of 25℃, and then the total duration of swimming mobility (vigorous attempts to escape) time was measured for a 10 min test session.

After perfusion, rat brains were lysed in RIPA buffer with cocktail of protease inhibitors (Roche). The proteins were separated by SDS-PAGE and transferred to a PVDF membrane as described [14]. The blot was probed with NCAM (Millipore, USA, AB5032) and GAPDH (Santa Cruz Biotechnology, USA, SC-257788) antibodies. After being incubated with horseradish peroxidase-conjugated secondary antibody, the membrane was visualized by ECL plus solution (Amersham Pharmacia).

Primary neurons were derived from the embryos (E17 days) cortex of 6- to 7-week-old pregnant SD rat (Koatech, South Korea). Briefly, an anesthetized pregnant rat was exposed to an overdose inhalation of 70~100% CO₂ gas for euthanasia, and then the uterus was extracted into cold phosphate buffered saline (PBS). We collected embryos in cold PBS and then moved them into a filtered flask to administer euthanasia using an overdose inhalation of 70~100 % CO₂ gas. Next, we placed them in a fresh dish with HBSS. The cerebral cortex was dissected from embryo rat brains and dissociated by gently triturate. Using curved, sharp forceps, the brains were collected from the embryos under a microscope. All animal experimental procedures were performed in accordance with 'the Guidelines of the Ethics Committee at Seoul National University'.

Cells were cultured in a specified medium for neurons (neurobasal medium supplemented with B27 [Gibco BRL, NY] and penicillin-streptomycin (100 U/ml/100 mg/ml) [Life Technology] at 37℃ and 5% CO₂ and plated at a density of 5×10⁴ cells/well on Deckgläser cover glasses (18 mm Φ, marienfeld) coated with PLL (1 mg/ml) in the 12 well plates.

Experiments were performed on 10-day cultures. After DHED (0.5 and 1 µM) and fluoxetine (0.5 µM) treatment for 4 hr, staurosporin (1 µM) were added for 1 hr.

After three washes with PBS, cells were fixed with 4% p-formaldehyde for 30 min. After three washes with PBS, cells were blocked and permeabilized in PBS added with 2% FBS, 2% BSA, 4% normal goat serum and 0.2% Triton X-100. Cells were incubated with primary antibodies for overnight at 4℃. After washing with PBS, proteins were visualized using secondary species-specific IgG conjugated with fluorescence. Cells were observed by confocal laser-scanning microscopy using an LSM-510 imaging system (Carl Zeiss Inc., Germany).

These results were analyzed by ANOVA followed by the LSD test. Statistical significance was set at ^*p<0.05, ^**p<0.01. Data are expressed as means±SEM. All data were analyzed using one-way analysis of variance (ANOVA, IBM SPSS statistics version 21).

We examined the effects of DHED and fluoxetine on locomotor activity in open field test. There were no noticeable differences in the numbers of crossings among five groups, indicating that locomotor activity is not affected by DHED or fluoxetine (Fig. 1B). Then, we investigated whether stress-induced spatial memory deficits could be improved by the treatment with DHED or fluoxetine in Y-maze test. As shown in Fig. 1C, the percentage of alternation behavior of the stressed rats (47.37±5.85) was more decreased than that of the vehicle-treated control (86.67±2.80) or the DHED-treated (82.03±1.88) control. But, the stress-induced decline in spontaneous alternations was significantly elevated to the vehicle control level by DHED (76.33±3.04) or fluoxetine treatment (71.08±6.18). Forced swimming test was done [15] to determine whether stress-induced reduction in the mobility time could be changed by the treatment with DHED or fluoxetine. As shown in Fig. 2, the climbing mobility time of the vehicle-treated stressed rats (136.0±13.71) was decreased in comparison to that of the vehicle-treated (200.5±21.20) or DHED-treated control (236.0±26.88). But, the stress-induced reduction in the mobility time was significantly increased by DHED (249.75±17.56) or fluoxetine treatment (262.8±31.39).

To assess DHED's therapeutic effects, we examined the levels of NCAM (180, 140, or 120 kDa isoform) that involved with synaptic plasticity and development of neuron [12] in the brain total homogenates. NCAM was detected by western blot with a specific antibody against NCAM and quantified (Fig. 3A). The levels of total NCAM in the vehicle-treated stressed rats were decreased in comparison to those of the vehicle- (1.588±0.202) or DHED-treated (1.995±0.228) control rats (Fig. 3B). However, the reduction of total NCAM levels (0.714±0.096) induced by stress was prevented by DHED (1.879±0.274) or fluoxetine (1.036±0.191) treatment.

To further confirm the preventive effects of DHED or fluoxetine in the level of NCAM reduced by stress, we checked the NCAM levels in primary cortical neurons treated with staurosporine which is a prototypical ATP-competitive kinase inhibitor in that it binds to many kinases with high affinity.

NCAM levels were also decreased by the treatment with staurosporine (1 µM) in primary cortical neurons (Supplementary Fig 1). The pretreatment with DHED (0.5 and 1 µM) or fluoxetine (0.5 µM) restored the level of NCAM reduced by sturosporine.