Neuroprotective Compounds from Salix pseudo-lasiogyne Twigs and Their Anti-Amnesic Effects on Scopolamine-Induced Memory Deficit in Mice

Abstract

Bioassay-guided fractionation of an 80 % methanolic extract of Salix pseudo-lasiogyne twigs has resulted in the isolation of two new compounds (1–2) along with ten known ones (3–12). The new compounds were determined to be 3′-O-acetylsalicin (1) and 2′,6′-O-acetylsalicortin (2) by using spectroscopic analyses. Compounds (3–12) were identified as salicin (3), 2′-O-acetylsalicin (4), salicortin (5), 2′-O-acetylsalicortin (6), 3′-O-acetylsalicortin (7), 6′-O-acetylsalicortin (8), 2′-O-(E)-ρ-coumaroylsalicortin (9), grandidentatin (10), isograndidentatin (11), and saligenin (12). Among the isolated compounds, compounds 2, 5, 6, 7, and 8 bearing 1-hydroxy-6-oxo-2-cyclohexenecarboxylate moiety significantly inhibited lipopolysaccharide-induced nitric oxide production in BV2 microglial cells in vitro. Further, we studied anti-amnesic activities of the 80 % methanolic extract, the EtOAc fraction, and compound 6 from S. pseudo-lasiogyne. They exerted a significant cognitive-enhancing effect on scopolamine-induced memory deficit in mice. In addition, they also significantly increased the reduced activities of glutathione reductase and superoxide dismutase and the glutathione content in the hippocampus and cortex of scopolamine-induced amnesic mice.

Key words

Salix pseudo-lasiogyne - Salicaceae - salicortin - 1-hydroxy-6-oxo-2-cyclohexenecarboxylate - BV2 microglial cell - passive avoidance test

The genus Salix (Salicaceae) comprises approximately 400 species of deciduous trees distributed in cold and temperate regions of the Northern hemisphere. The willow tree is the most abundant among the Salix spp. and has been used to relieve pain and inflammation for thousands of years [1]. Salix pseudo-lasiogyne H. Lev., distributed over several Asian countries, has been used for the treatment of pain and fever in Korean traditional medicine [2]. While searching for anti-inflammatory natural products using BV2 microglial cells, which are widely employed in in vitro assay systems [3], [4], it was found that an 80 % methanolic extract of S. pseudo-lasiogyne significantly inhibited NO production induced by lipopolysaccharide (LPS) in BV2 microglial cells. Thus, we attempted to isolate active compounds from S. pseudo-lasiogyne and to evaluate those compoundsʼ anti-inflammatory activity in BV2 microglia. It has also been reported that excessive NO-induced inflammation can increase oxidative stress in the brain, which consequently can induce neurodegenerative disorder accompanied by memory deficit [5]. Thus, we also examined the memory-enhancing effects of an extract of S. pseudo-lasiogyne and of a major compound from this plant that showed the most potent anti-inflammatory activity in vitro on scopolamine–induced amnesic mice using the passive avoidance test.

Dried and pulverized S. pseudo-lasiogyne twigs were extracted with 80 % methanol by using an ultrasonic apparatus. The 80 % methanolic extract was suspended in distilled water and successively partitioned with n-hexane, EtOAc, and n-BuOH. The EtOAc fraction, which showed the most potent inhibitory activity on NO production of LPS-induced BV2 cells, was subjected to repeated column chromatography (CC) and high-performance liquid chromatography (HPLC), resulting in the isolation of two new compounds, 1–2, and ten known ones (3–12) ([Fig. 1]). Structures of these compounds were unequivocally determined by 1D, 2D NMR experiments, MS analyses, as well as by comparison with reference materials of known compounds.

Fig. 1 Structures of compounds 1–12 isolated from S. pseudo-lasiogyne twigs.

Compound 1 was isolated as yellowish needles, [α]_D ²⁵ − 60.1 (c 0.80, EtOH), and its molecular formula (C₁₅H₂₀O₈) was established by positive HRFABMS (m/z 329.1234 [M + H]⁺, calcd. for 329.1236). In the ¹H NMR spectrum, the chemical shifts and coupling constants of the characteristic signals at δ _H 7.76 (1H, d, J = 6.9 Hz, H-3), 7.54 (1H, d, J = 10.9 Hz, H-6), 7.22 (1H, td-like, J = 8.5, 1.5 Hz, H-5), and 7.09 (1H, t, J = 7.3 Hz, H-4) indicated the presence of a 1, 2-disubstituted benzene moiety. Also, the signals at δ _H 5.52 (1H, d, J = 7.9 Hz, H-1′) suggested that compound 1 had an anomeric proton. In the HMBC spectrum of 1, the anomeric proton at δ _H 5.52 and two methylene protons at δ _H 5.25 (1H, d, J = 13.8 Hz, H-7a) and 5.08 (1H, d, J = 13.8 Hz, H-7b) correlated with quaternary carbons at δ _C 156.3 (C-1) and at δ _C 133.0 (C-2), respectively ([Fig. 2]). Thus, a sugar residue and an oxygenated methylene moiety were present at C-1 and C-2. From the above information, compound 1 was deduced to be similar to a known compound, salicin (3), except for the signals at δ _H/δ _C 1.98 (3H, s, H-2′′)/21.1 (C-2′′) and at δ _C 170.7 (C-1′′) [6], [7]. HMBC spectrum correlation between the proton at δ _H 5.89 (1H, t, J = 9.5 Hz, H-3′) and a carbonyl carbon inacetyl group (δ _C 170.7) determined the position of the acetyl group at C-3′. It was previously reported that an acetyl moiety was located at C-2′ or C-6′ instead of at C-3′ in the structure of the same aglycone [8], [9]. On the basis of the above-described information, compound 1 was determined to be 3′-O-acetylsalicin.

Fig. 2 Key correlations of HMBC spectral data of compounds 1 and 2.

Compound 2 was isolated as a whitish, amorphous powder, [α]_D ²⁵ − 130.1 (c 1.12, EtOH). The positive HRFABMS of 2 exhibited m/z 531.1481 [M + Na]⁺ (calcd. for 531.1478) indicating C₂₄H₂₈O₁₂ as its molecular formula. The ¹H and ¹³C NMR spectra and the 2D NMR analysis showed features similar to those of compound 3 except for characteristic peaks that indicated the structure of 1-hydroxy-6-oxo-2-cyclohexenecarboxylate and two acetyl moieties. In the HMBC spectrum, the position of 1-hydroxy-6-oxo-2-cyclohexenecarboxylate was determined by the cross-peaks from δ _H 5.13 (1H, m, H-7a) and 5.12 (1H, m, H-7b) to δ _C 172.2 (C-8) ([Fig. 2]). Correlation peaks from δ _H 4.99 (1H, t, J = 9.6 Hz, H-2′) to δ _C 172.7 (C-1′′) and from δ _H 4.40 (1H, m, H-6′a) and 4.27 (1H, m, H-6′b) to δ _C 173.4 (C-1′′′) confirmed that the position of the two acetyl moieties were at OH-1′′ and OH-6′′, respectively. Thus, compound 2 was identified as 2′,6′-O-acetylsalicortin. Although it was previously reported in an organic synthetic study, this is the first report of its occurrence in nature [10].

The ten known compounds were identified as salicin (3) [6], [7], 2′-O-acetylsalicin (4) [8], salicortin (5) [7], [11], 2′-O-acetylsalicortin (6) [8], 3′-O-acetylsalicortin (7) [12], 6′-O-acetylsalicortin (8) [13], 2′-O-(E)-ρ-coumaroylsalicortin (9) [13], grandidentatin (10) [14], isograndidentatin (11) [14], and saligenin (12) [15] ([Fig. 1]).

We examined the inhibitory activity of an 80 % methanolic extract of S. pseudo-lasiogyne and its n-hexane, EtOAc, and n-BuOH fractions on LPS-stimulated NO production in BV2 microglia. The EtOAc fraction which showed the most potent inhibitory activity (IC₅₀ 17.6 mg/mL; data not shown) was subjected to repeated chromatographic techniques. The two new compounds 1–2 along with the ten known compounds 3–12 were isolated, and their inhibitory activity on LPS-stimulated NO production in BV2 cells was determined ([Table 1]). Compounds 2, 5, 6, 7, and 8 exhibited more potent inhibitory activities than the other compounds. Compound 3 and its derivatives (1 and 4) showed weak inhibitory activity, while saligenin (12), a metabolite of 3, showed no inhibitory activity. The key difference between the salicin- (1, 3, and 4) and salicortin-type (2, 5, 6, 7, and 8) compounds was the presence of a 1-hydroxy-6-oxo-2-cyclohexenecarboxylate moiety at OH-7. These data suggest that the moiety could be an important factor in elucidating the inhibitory activity of LPS-induced NO production in BV2 microglia.

Table 1 Inhibitory effects of compounds 1–12 from S. pseudo-lasiogyne twigs on LPS-induced NO production in BV2 microglial cells.

Compound	IC50 (µM)	Compound	IC50 (µM)
IC50 means the 50 % inhibitory concentration (µM) on LPS-induced NO production in BV2 cells. The nitrite concentration in vehicle- and LPS-treated cells was 6.3 ± 0.2 and 54.1 ± 0.2 μM, respectively. L-NIL was used as the positive control
1	> 100	8	11.4 ± 2.1
2	13.2 ± 1.3	9	> 100
3	> 100	10	> 100
4	> 100	11	> 100
5	15.3 ± 3.1	12	> 100
6	11.4 ± 2.2	L-NIL	65.1 ± 4.1
7	11.9 ± 0.5

Further, we aimed to determine whether the 80 % methanolic extract, the EtOAc fraction, and a compound from S. pseudo-lasiogyne, which all had anti-inflammatory effects in vitro, had cognitive–enhancing activity in mice with memory deficits induced by scopolamine. Moreover, we attempted to preliminarily examine the action mechanisms in vivo. Since compound 6 was the most abundant among the obtained compounds 1–12 and had 1-hydroxy-6-oxo-2-cyclohexenecarboxylate moiety, which significantly inhibited LPS-induced NO production in BV2 microglial cells, we investigated the effect of compound 6 on attenuated memory deficits induced by scopolamine in mice. The cognitive-enhancing effect of the 80 % methanolic extract, EtOAc fraction of S. pseudo-lasiogyne, and compound 6 was evaluated using the passive avoidance test ([Table 2]). The step-through latency of the scopolamine-treated mice (1 mg/kg body weight s. c.; 28.1 s) was significantly reduced compared to that of the 0.5 % carboxylmethyl cellulose-treated control mice (176.1 s). However, the short step-through latency induced by scopolamine was significantly reversed by treatment with the 80 % methanolic extract (100 mg/kg body weight p. o.), the EtOAc fraction (100 mg/kg body weight p. o.), and compound 6 (1 and 2 mg/kg body weight p. o.). Donepezil (2 mg/kg body weight p. o.), an acetylcholinesterase inhibitor and the most widely used treatment for Alzheimerʼs disease, was used as a positive control, and it restored step-through latency time by 119.7 s. In comparison with the cognitive-enhancing activity of donepezil, the 80 % methanolic extract (109.3 s at 100 mg/kg body weight p. o.) and compound 6 (109.4 s at 2 mg/kg body weight p. o.) were able to significantly restore memory deficits induced by scopolamine in mice.

Table 2 The cognitive–enhancing effect of 80 % methanolic extract, EtOAc fraction, and compound 6 of S. pseudo-lasiogyne twigs on scopolamine-induced amnesic mice in the passive avoidance test^a.

Experimental treatment	Step-through latency (s) (% of control)
The values shown are the mean latency ± SEM. Results differ significantly from the value in the scopolamine-treated group (* p < 0.05 and ** p < 0.01). a Control indicates the 0.5 % CMC and saline-treated group (10 mL/kg body weight, p. o.). b Scopolamine (Sco) indicates the 0.5 % CMC and scopolamine-treated group (1 mg/kg body weight, s. c.). c b. w.: body weight
Controla	176.1 ± 3.7 (100 %)
Scob	28.1 ± 5.0 (16.0 %)
Sco + 80 % methanolic extract (50 mg/kg b. w.c)	63.1 ± 13.1 (35.8 %)
Sco + 80 % methanolic extract (100 mg/kg b. w.)	109.3 ± 23.1 (62.1 %)*
Sco + EtOAc fraction (50 mg/kg b. w.)	31.9 ± 3.8 (18.1 %)
Sco + EtOAc fraction (100 mg/kg b. w.)	68.9 ± 14.9 (39.1 %)
Sco + compound 6 (1 mg/kg b. w.)	39.8 ± 6.8 (22.6 %)
Sco + compound 6 (2 mg/kg b. w.)	109.4 ± 19.9 (62.1 %)*
Sco + donepezil (2 mg/kg b. w.)	119.7 ± 16.9 (68.0 %)**

There has been a controversy over whether oxidative stress plays a primary role in, or is only a consequence of, the process of neurodegenerative diseases such as Alzheimerʼs, Parkinsonʼs, and Huntingtonʼs diseases [16]. Nevertheless, the suppression of elevated oxidative stress can be a therapeutic target when attempting to attenuate neurodegenerative disorders [17]. Thus, we studied whether the 80 % methanolic extract, the EtOAc fraction, and compound 6 from S. pseudo-lasiogyne twigs, which exhibited anti-inflammatory activities in BV2 microglial cells, have the potential to repress oxidative stress on scopolamine-induced memory deficit in mice. In the passive avoidance test, treatments with the 80 % methanolic extract (50 mg/kg body weight, p. o.) or the EtOAc fraction (50 mg/kg body weight p. o.) did not result in remarkable changes in step-though latency ([Table 2]). However, the mice treated with 80 % methanolic extract or the EtOAc fraction at 100 mg/kg body weight p. o. significantly recovered their memory deficit. After the passive avoidance test, the cortex and hippocampus of the mice were removed for an antioxidant enzyme assay. The amnesic mice treated with the 80 % methanolic extract or the EtOAc fraction had significantly restored the reduced levels of antioxidant enzymes, such as superoxide dismutase (SOD), glutathione peroxidase (GPx), and glutathione reductase (GR), as well as restored cellular glutathione content ([Table 3]). In addition, treatment with compound 6 resulted in a significant reversal in GR activity and glutathione content, which had been lowered by scopolamine in the mouse cortex and hippocampus; however the activities of SOD and GPx were virtually unchanged. In a redox cycle, GR reduces oxidized glutathione (GSSG) to reduced glutathione (GSH) by using NADPH [18]. We suggest that compound 6 may selectively restore the activity of GR, after which there is an elevation in GSH content. In contrast to the results from the compound 6-treated mice, 80 % methanolic extract- and EtOAc fraction-treated mice substantially reversed all of the antioxidant markers, including SOD and GPx, as well as increased the activity of GR and GSH content. Several studies have reported that the synergistic effects of combined components in a natural products extract are more effective than the effects of isolated single compounds or of the sum of some of them [19], [20], [21]. Although a profound mechanism study of the synergic effect between compound 6 derivatives and other components in S. pseudo-lasiogyne twigs would be needed, it seems that processed extractor compounds from this plant might have potential as therapeutic agents in cognitive disorders.

Table 3 Antioxidative effects of 80 % methanolic extract, EtOAc fraction, and compound 6 of S. pseudo-lasiogyne twigs on the activities of antioxidant enzymes and glutathione level within the cortex and hippocampus of scopolamine-induced amnesic mice in the passive avoidance test.

Each value represents the mean ± SD. Sco: scopolamine. Results differ significantly from the value in the scopolamine-treated group (* p < 0.05 and ** p < 0.01) and from the value of normal control group: # p < 0.05; ## p < 0.01
Groups	SOD (U mg-1 protein)		GPx (µmol NADPH oxidized/min/mg protein)		GR (µmol NADPH oxidized/min/mg protein)
	Cortex	Hippocampus	Cortex	Hippocampus	Cortex	Hippocampus
Control	15.411 ± 0.605	28.402 ± 2.194	0.105 ± 0.005	0.173 ± 0.023	46.857 ± 5.939	52.019 ± 3.086
Sco	10.381 ± 0.171#	17.122 ± 0.117##	0.084 ± 0.009##	0.106 ± 0.012##	32.273 ± 0.823##	40.207 ± 2.757##
Total	14.907 ± 0.761*	27.122 ± 0.778**	0.102 ± 0.007	0.125 ± 0.007	43.157 ± 1.834**	43.157 ± 1.834
EtOAc	15.875 ± 0.688**	24.867 ± 1.503**	0.100 ± 0.010*	0.178 ± 0.008**	45.786 ± 2.377**	49.443 ± 6.685**
Compound 6 (1 mg/kg)	12.592 ± 0.213*	16.825 ± 0.948	0.083 ± 0.007	0.114 ± 0.002	39.675 ± 1.713*	37.413 ± 1.687
Compound 6 (2 mg/kg)	11.407 ± 0.975	17.567 ± 2.759	0.086 ± 0.008	0.143 ± 0.006**	41.851 ± 0.554*	41.737 ± 3.830
Donepezil	15.093 ± 0.358**	24.704 ± 3.343**	0.094 ± 0.004**	0.129 ± 0.019*	45.193 ± 0.387**	50.122 ± 3.848**
Groups	Total GSH (nmol/mg protein)		GSH (nmol/mg protein)		GSSG/total GSH
	Cortex	Hippocampus	Cortex	Hippocampus	Cortex	Hippocampus
Control	19.958 ± 0.578	22.616 ± 0.545	13.165 ± 0.341	14.926 ± 1.127	0.346 ± 0.047	0.328 ± 0.030
Sco	17.669 ± 1.008	14.615 ± 0.533##	10.451 ± 0.289	9.240 ± 1.013##	0.381 ± 0.027	0.364 ± 0.037
Total	25.779 ± 0.650**	22.171 ± 1.414**	19.262 ± 0.485	15.001 ± 0.592**	0.272 ± 0.015	0.291 ± 0.034
EtOAc	24.559 ± 0.550**	25.871 ± 2.268**	19.351 ± 0.441**	18.654 ± 0.545**	0.227 ± 0.022	0.280 ± 0.031
Compound 6 (1 mg/kg)	16.909 ± 0.581	16.473 ± 0.958	12.411 ± 0.578*	11.750 ± 0.470*	0.285 ± 0.071	0.262 ± 0.025
Compound 6 (2 mg/kg)	20.657 ± 0.858*	20.420 ± 0.914**	15.596 ± 0.552**	14.671 ± 0.829*	0.256 ± 0.024	0.241 ± 0.039
Donepezil	20.049 ± 1.282	18.609 ± 1.115*	14.426 ± 0.515**	13.362 ± 0.347*	0.271 ± 0.046	0.290 ± 0.024

Materials and Methods

CC was carried out on Kiesgel 60 silica gel (40–60 µm, 230–400 mesh; Merck), YMC-GEL ODS-A (5–150 µm; YMC), and Sephadex LH-20 (25–100 µM; Pharmacia). Thin-layer chromatography was carried out on Kiesgel 60 F₂₅₄ coated normal silica gel and RP-18 F₂₅₄ coated C₁₈ silica gel. The1D and 2D spectral data were measured on a Bruker AMX 400 or 500 spectrometer. Solvent signals were used as internal standards. High-resolution and low-resolution FABMS results were obtained on a JEOL JMS-AX505WA. The FT-IR spectra were measured with a JASCO FT/IR-300 spectrophotometer. The HPLC system consisted of a G-321 pump (Gilson), a G-151 UV detector (Gilson), and an YMC-Pack Pro C₁₈ column (250 mm × 10 mm i. d.; 5 µm). HPLC grade solvents (Fisher Scientific) were used in the MeOH-H₂O system.

The S. pseudo-lasiogyne twigs were collected at the Medicinal Plant Garden, Seoul National University, Goyang, Korea, in July 2009. Air-dried S. pseudo-lasiogyne twigs were identified by Dr. Jong Hee Park, a professor of the College of Pharmacy, Pusan National University, Korea. A voucher specimen (SNUPH-1105) has been stored in the Herbarium of the Medicinal Plant Garden, Seoul National University, Korea.

The obtained S. pseudo-lasiogyne twigs (17 kg) were extracted with 80 % MeOH (15 L × 3) three times in an ultrasonic apparatus (3 h × 3). The solvent was removed in vacuo, and an 80 % MeOH extract (1.2 kg) was suspended in H₂O and successively partitioned into n-hexane (48 g), EtOAc (121 g), and n-BuOH (160 g) fractions. The EtOAc fraction was subjected to silica gel CC (20 × 60 cm) eluted with CHCl₃:MeOH of increasing polarity (50 : 1, 30 : 1, 10 : 1, 5 : 1, 3 : 1, 0 : 1; 20 l) to give eight fractions (SXE1–8). SXE7 was subjected to normal silica gel CC (2 × 60 cm) to afford three fractions (SXE7A–C). By reverse-phase (RP) C₁₈HPLC with MeOH : H₂O (7 : 3, 2 mL/min), compound 2 (28 mg) was obtained from SXE7A3 which was separated from ODS silica gel CC (MeOH : H₂O 2 : 8, 8 : 2, 2 mL/min). SXE7B was separated into four fractions (SXE7B1–4) by normal silica gel CC (2 × 60 cm) with CHCl₃ : MeOH (15 : 1, 0 : 1; 1 l). SXE7B4 was chromatographed by RP C₁₈ HPLC (MeOH : H₂O = 8 : 2, 2 mL/min) to give compound 1 (9 mg).

3′-O-Acetylsalicin (1): yellowish needles; [α]_D ²⁵ − 60.1 (c 0.80, EtOH); ¹HNMR (500 MHz, pyridine-d ₅): δ 7.76 (1H, d, J = 6.9 Hz, H-3), 7.54 (1H, d, J = 10.9 Hz, H-6), 7.22 (1H, td, J = 8.5, 1.5 Hz, H-5), 7.09 (1H, t, J = 7.3 Hz, H-4), 5.89 (1H, t, J = 9.5 Hz, H-3′), 5.52 (1H, d, J = 7.9 Hz, H-1′), 5.25 (1H, d, J = 13.8 Hz, H-7a), 5.08 (1H, d, J = 13.7 Hz, H-7b), 4.48 (1H, dd, J = 9.7, 2.0 Hz, H-6′a), 4.38 (2H, m, H-4′ and H-6′b), 4.28 (1H, t, J = 8.0 Hz, H-2′), 4.06 (1H, m, H-5′), 1.98 (3H, s, H-2′′); δ 170.7 (C-1′′), 156.3 (C-1), 133.0 (C-2), 128.6 (C-3 and C-5), 122.9 (C-4), 116.3 (C-6), 103.3 (C-1′), 79.3 (C-3′), 78.6 (C-5′), 73.0 (C-2′), 68.9 (C-4′), 61.8 (C-6′), 60.1 (C-7), 21.1 (C-2′′); FABMS m/z 329 [M + H]⁺; HRFABMS m/z 329.1234 [M + H]⁺ (calcd. for 329.1236).

2′,6′-O-Acetylsalicortin (2): colorless oil; [α]_D ²⁵ − 130.1 (c 1.12, EtOH); ¹HNMR (500 MHz, CD₃OD): δ 7.29 (1H, m, H-3), 7.28 (1H, m, H-5), 7.13 (1H, d, J = 8.1 Hz, H-6), 7.04 (1H, t, J = 7.5 Hz, H-4), 6.15 (1H, m, H-11), 5.77 (1H, d, J = 9.8 Hz, H-10), 5.13 (2H, m, H-7), 5.07 (1H, m, H-1′), 4.99 (1H, t, J = 9.6 Hz, H-2′), 4.40 (2H, m, H-6′a), 4.27 (1H, m, H-6′b), 3.68 (1H, m, H-5′), 3.63 (1H, m, H-3′), 3.47 (1H, m, H-4′), 2.88 (1H, m, H-13a), 2.66 (1H, m, H-12a), 2.53 (1H, m, H-13b), 2.48 (1H, m, H-12b), 2.12 (3H, s, H-2′′), 2.02 (3H, s, H-2′′′); ¹³CNMR (125 MHz, CD₃OD): δ 208.1 (C-14), 173.4 (C-1′′′), 172.7 (C-1′′), 172.2 (C-8), 157.1 (C-1), 134.1 (C-11), 131.7 (C-3), 131.3 (C-5), 130.2 (C-10), 127.1 (C-2), 124.8 (C-4), 117.7 (C-6), 101.2 (C-1′), 80.0 (C-9), 76.6 (C-3′), 76.3 (C-5′), 75.7 (C-2′), 72.4 (C-4′), 64.8 (C-7), 65.3 (C-6′), 37.6 (C-13), 28.0 (C-12), 21.8 (C-2′′), 21.5 (C-2′′′); FABMS m/z 509 [M + H]⁺; HRFABMS m/z 531.1481 [M + Na]⁺ (calcd. for 531.1478).

The BV2 microglial cells were maintained in DMEM (Sigma) supplemented with 10 % FBS, 100 IU/mL penicillin (Sigma), and 100 µg/mL streptomycin (Sigma) at 37 °C in a humidified incubator containing 5 % CO₂ gas. Compounds 1–12 were dissolved in DMSO (final concentration, < 0.1 %). The purity of the tested compounds was verified to be above 95 % by using an HPLC-UV system. For those assays, the cells were seeded in 48-well plates at a density of 4 × 10⁵ cells/mL and incubated overnight. BV2 cells were treated with vehicle or compounds 1–12 at concentrations from 1 µM to 100 µM for 24 h. Inhibitory activity of each compound on LPS-induced NO production in BV2 cells was assessed by using the Griess assay [22]. L-NIL [L–N6-(1-iminoethyl)lysine, > 97 % purity] used as a positive control was purchased from Sigma. Male ICR (Harlan Sprague–Dawley; 4 weeks old) mice, weighing 25–30 g each, were used after a 1-week adaptation period at room temperature under a 12-h light cycle and fed ad libitum with free access to water. Ten mice were used per group. All experiments and the method used for euthanasia were according to the guidelines of the Institutional Animal Care and Use Committee at Seoul National University (SNU-120430–1, 16-Nov-2010). Mice were orally treated with the sample. Amnesia was subcutaneously induced in mice with scopolamine (Sigma, 1 mg/kg body weight s. c.). All of the samples for the in vivo test were dissolved in 0.5 % CMC (carboxylmethyl cellulose; Sigma). Donepezil (purity > 98 %) was from Sigma. The passive avoidance test was performed as described in our previous report [23], and the method details were included in the Supporting Information. After the passive avoidance test, the mice were immediately  euthanized with urethane (1.5 g/kg) to allow measurement of antioxidant enzyme activity levels. The cerebral cortex and hippocampus of the mice were rapidly dissected and homogenized. The homogenates were centrifuged and the supernatant used for measurement of antioxidant enzyme activity and GSH content. The SOD activity was determined by using the xanthine–xanthine oxidase reaction method [24]. The GR activity was measured by using a method based on the reduction of GSSG by GR in the presence of NADPH [25]. The activity of GPx was determined by quantifying the rate of oxidation of GSH to GSSG by cumene hydroperoxide [26]. Total GSH content in the supernatant was determined spectrophotometrically by using an enzymatic cycling method [27]. Protein concentration was determined by using a bicinchoninic acid (BCA) kit (Sigma) with bovine serum albumin as a standard. Data from the passive avoidance tests were expressed as mean ± SEM, while data for the level of the antioxidant activity were expressed as mean ± SD. Passive avoidance latencies and antioxidant activity values were analyzed by one-way ANOVA. The data were considered to be statistically significant when the probability (p) value was 0.05 or less.

Supporting information

Original spectral data of 1 and 2 as well as detailed descriptions of bioassay protocols are available as Supporting Information.

Acknowledgements

This research was supported by a grant (2011K000290) from the Brain Research Center of the 21st Century Frontier Research Program funded by the Ministry of Science and Technology, Korea.

Conflict of Interest

There is no conflict of interest.

• References

• 1 Mahdi JG, Mahdi AJ, Mahdi AJ, Bowen ID. The historical analysis of aspirin discovery, its relation to the willow tree and antiproliferative and anticancer potential. Cell Prolif 2006; 39: 147-155
• 2 Koh KS, Jeon ES. Ferns, fern-allies and seed-bearing plants of Korea. Seoul: Iljin-sa; 2003
• 3 Jung WK, Lee DY, Park C, Choi YH, Choi I, Park SG, Seo SK, Lee SW, Yea SS, Ahn SC, Lee CM, Park WS, Ko JH, Choi IW. Cilostazol is anti-inflammatory in BV2 microglial cells by inactivating nuclear factor-kappaB and inhibiting mitogen-activated protein kinases. Br J Pharmacol 2010; 159: 1274-1285
• 4 Henn A, Lund S, Hedtjarn M, Schrattenholz A, Porzgen P, Leist M. The suitability of BV2 cells as alternative model system for primary microglia cultures or for animal experiments examining brain inflammation. ALTEX 2009; 26: 83-94
• 5 Aquilano K, Baldelli S, Rotilio G, Ciriolo MR. Role of nitric oxide synthases in Parkinsonʼs disease: a review on the antioxidant and anti-inflammatory activity of polyphenols. Neurochem Res 2008; 33: 2416-2426
• 6 Shimomura H, Sashida Y, Yoshinari K. Phenolic glucosides from the heartwood of Prunus grayana . Phytochemistry 1989; 28: 1499-1502
• 7 Zapesochnaya GG, Kurkin VA, Braslavskii VB, Filatova NV. Phenolic compounds of Salix acutifolia bark. Chem Nat Compd 2002; 38: 314-318
• 8 Reichardt PB, Merken HM, Clausen TP, Wu J. Phenolic glycosides from Salix lasiandra . J Nat Prod 1992; 55: 970-973
• 9 Thieme H. The phenol glycosides in Salicaceae. 2. Isolation and demonstration. Pharmazie 1964; 19: 471-475
• 10 Ruuhola T, Julkunen-Tiitto R, Vainiotalo P. In vitro degradation of willow salicylates. J Chem Ecol 2003; 29: 1083-1097
• 11 Dommisse RA, Van Hoof L, Vlietinck AJ. Structural analysis of phenolic glucosides from Salicaceae by NMR spectroscopy. Phytochemistry 1986; 25: 1201-1204
• 12 Ikonen A, Tahvanainen J, Roininen H. Phenolic secondary compounds as determinants of the host plant preferences of the leaf beetle Agelastica alni . Chemoecology 2002; 12: 125-131
• 13 Dagvadorj E, Shaker KH, Windsor D, Schneider B, Boland W. Phenolic glucosides from Hasseltia floribunda . Phytochemistry 2010; 71: 1900-1907
• 14 Shen CC, Chang YS, Hott LK. Nuclear magnetic resonance studies of 5,7-dihydroxyflavonoids. Phytochemistry 1993; 34: 843-845
• 15 Fisher TH, Chao P, Upton CG, Day AJ. One- and two-dimensional NMR study of resol phenol–formaldehyde prepolymer resins. Magn Reson Chem 1995; 33: 717-723
• 16 Andersen JK. Oxidative stress in neurodegeneration: cause or consequence?. Nat Med 2004; 10 (Suppl.) S18-S25
• 17 Ienco EC, Logerfo A, Carlesi C, Orsucci D, Ricci G, Mancuso M, Siciliano G. Oxidative stress treatment for clinical trials in neurodegenerative diseases. J Alzheimers Dis 2011; 24 (Suppl. 02) 111-126
• 18 Lu SC. Regulation of hepatic glutathione synthesis: current concepts and controversies. Faseb J 1999; 13: 1169-1183
• 19 Junio HA, Sy-Cordero AA, Ettefagh KA, Burns JT, Micko KT, Graf TN, Richter SJ, Cannon RE, Oberlies NH, Cech NB. Synergy-directed fractionation of botanical medicines: a case study with goldenseal (Hydrastis canadensis). J Nat Prod 2011; 74: 1621-1629
• 20 Wagner H, Ulrich-Merzenich G. Synergy research: approaching a new generation of phytopharmaceuticals. Phytomedicine 2009; 16: 97-110
• 21 Spelman K, Duke JA, Bogenschutz-Godwin MJ. Natural products from plants. 2nd. edition. Boca Raton: CRC Taylor and Francis; 2006
• 22 Dawson VL, Brahmbhatt HP, Mong JA, Dawson TM. Expression of inducible nitric oxide synthase causes delayed neurotoxicity in primary mixed neuronal-glial cortical cultures. Neuropharmacology 1994; 33: 1425-1430
• 23 Jeong EJ, Ma CJ, Lee KY, Kim SH, Sung SH, Kim YC. KD-501, a standardized extract of Scrophularia buergeriana has both cognitive-enhancing and antioxidant activities in mice given scopolamine. J Ethnopharmacol 2009; 121: 98-105
• 24 McCord JM, Fridovich I. Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein). J Biol Chem 1969; 244: 6049-6055
• 25 Carlberg I, Mannervik B. Purification and characterization of the flavoenzyme glutathione reductase from rat liver. J Biol Chem 1975; 250: 5475-5480
• 26 Flohe L, Gunzler WA. Assays of glutathione peroxidase. Methods Enzymol 1984; 105: 114-121
• 27 Tietze F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues. Anal Biochem 1969; 27: 502-522