Bioactive Benzofuran Derivatives from Cortex Mori Radicis, and Their Neuroprotective and Analgesic Activities Mediated by mGluR1

Four new benzofuran-type stilbene glycosides and 14 known compounds including 8 benzofuran-type stilbenes and 6 flavonoids were isolated from the traditional Chinese medicine, Cortex Mori Radicis. The new compounds were identified as (9R)-moracin P 3′-O-α-l-arabinopyranoside (1), (9R)-moracin P 9-O-β-d-glucopyranoside (2), (9R)-moracin P 3′-O-β-d-glucopyranoside (3), and (9R)-moracin O 10-O-β-d-glucopyranoside (4) based on the spectroscopic interpretation and chemical analysis. Three benzofuran-type stilbenes, moracin O (5), R (7), and P (8) showed significant neuroprotective activity against glutamate-induced cell death in SK-N-SH cells. In addition, moracin O (5) and P (8) also demonstrated a remarkable inhibition of the acetic acid-induced pain. The molecular docking with metabotropic glutamate receptor 1 (mGluR1) results indicated that these neuroprotective benzofuran-type stilbenes might be the active analgesic components of the genus Morus, and acted by mediating the mGluR1 pathway.

L-Glutamate is the major excitatory neurotransmitter in the central nervous system (CNS) and glutamate-mediated excitotoxicity plays a crucial role in neurodegenerative disorder, particularly in Parkinson's disease, Alzheimer's disease, epilepsy, spinal cord trauma, and ischemic stroke. As glutamate receptors are also found in peripheral tissue, they may be implicated in persistent or chronic pain, including inflammatory or joint-related pain (e.g., rheumatoid arthritis, osteoarthritis) and neuropathic pain resulting from injury and/or diseases of central (e.g., spinal cord injury) or peripheral nerves (e.g., diabetic neuropathy, radiculopathy) [21,22].
Neuroprotective effect of stilbenes, including piceid and resveratrol, has been widely reported [23], but the applications of benzofuran stilbene on CNS diseases [24] and analgesia were rarely studied. In this paper, we report four new benzofuran stilbene glycosides (1)(2)(3)(4) and the neuroprotective and analgesic activities of their aglycones. Three benzofuran stilbenes showed significant protective activities against glutamate-induced neurotoxicity. Correspondingly, they also exhibited remarkable analgesic activities by inhibition of the acetic acid-induced pain. Structural determinations of compounds were carried out on the basis of spectroscopic data analyses. The primary action mechanism of neuroprotection and analgesia was also discussed with the help of the molecular docking with metabotropic glutamate receptor 1 (mGluR 1 ).
Neuroprotective effect of stilbenes, including piceid and resveratrol, has been widely reported [23], but the applications of benzofuran stilbene on CNS diseases [24] and analgesia were rarely studied. In this paper, we report four new benzofuran stilbene glycosides (1)(2)(3)(4) and the neuroprotective and analgesic activities of their aglycones. Three benzofuran stilbenes showed significant protective activities against glutamate-induced neurotoxicity. Correspondingly, they also exhibited remarkable analgesic activities by inhibition of the acetic acid-induced pain. Structural determinations of compounds were carried out on the basis of spectroscopic data analyses. The primary action mechanism of neuroprotection and analgesia was also discussed with the help of the molecular docking with metabotropic glutamate receptor 1 (mGluR1).
There was only one chiral carbon C-9 in aglycones of compounds 1-4. In order to determine their absolute configuration, compounds 1-4 were hydrolyzed by HCl, and the aglycones were partitioned by EtOAc. After drying the EtOAc-soluble fraction in vacuo, the residue was dissolved in MeOH to determine the optical rotation. All aglycones showed negative optical rotation values (see Supplementary Materials), which confirmed that aglycones 1a-3a were (9R)-moracin P [38] and aglycone 4a was (9R)-moracin O [38,39].
Although the sugars were substituted at different position, the ECD (electronic circular dichroism) spectra of 1-4 and their aglycones 1a and 4a were quite similar ( Figure 2). There were two Cotton effects in their ECD spectra. The broad negative Cotton effect over 280-340 nm was regarded as π→π* transition from MO86 to MO87 of 2-benzylbenzofuran. The negative Cotton effect at 240 nm was generated by π→π* transition from MO86 to MO90 of benzofuran moiety and π→π* transition from MO86 to MO89 of benzyl moiety, according to the MO analysis in the calculated ECD of 1a (Supplementary Materials). It seemed that the chiral carbon on the rigid skeleton attached to the benzofuran affected the Cotton effect at 240 nm more than the chiral center on the flexible benzyl moiety, deduced from the similar ECD spectra of 1-4, 1a, and 4a.
in their ECD spectra. The broad negative Cotton effect over 280-340 nm was regarded as π→π* transition from MO86 to MO87 of 2-benzylbenzofuran. The negative Cotton effect at 240 nm was generated by π→π* transition from MO86 to MO90 of benzofuran moiety and π→π* transition from MO86 to MO89 of benzyl moiety, according to the MO analysis in the calculated ECD of 1a (Supplementary Materials). It seemed that the chiral carbon on the rigid skeleton attached to the benzofuran affected the Cotton effect at 240 nm more than the chiral center on the flexible benzyl moiety, deduced from the similar ECD spectra of 1-4, 1a, and 4a.

Postulating the Neuroprotective Pathway by Molecular Docking
In order to further understand the possible mechanisms of the benzofuran stilbenes against glutamate-induced neuronal death, 10 glutamate receptors, including various ionotropic and metabotropic types, were selected from 64 Homo sapiens GluRs in the Protein Data Bank (PDB) website to study the ligand binding to these receptors in a molecular-docking method.
Among the selected glutamate receptors, the mGluR 1 (3KS9) was the only one that could be docked with all active compounds (5-8, 13, and 15) as well as glutamic acid. In contrast, all of the inactive compounds (1-4, 9) could not be docked into the binding site of the mGluR 1 (3KS9). These results indicated that the neuroprotective activities of moracin O (5), R (7), and P (8), might be mediated by the mGluR 1 pathway.
Molecular-docking results ( Figure 3) showed each of the three ligands, moracin O (5), R (7), and P (8), had π-π interaction between the aromatic plane and Trp 110, and three to four H-bonds between hydroxyl groups of the ligands and the residues of Arg 71, Arg 78, Tyr 236 (or Asp 208, Ser165, Thr188) in mGluR 1 , which indicated that the aromatic group and the hydroxyl group play an important role in the bindings. Molecular-docking results ( Figure 3) showed each of the three ligands, moracin O (5), R (7), and P (8), had π-π interaction between the aromatic plane and Trp 110, and three to four H-bonds between hydroxyl groups of the ligands and the residues of Arg 71, Arg 78, Tyr 236 (or Asp 208, Ser165, Thr188) in mGluR1, which indicated that the aromatic group and the hydroxyl group play an important role in the bindings.

Analgesic Activities
Although some compounds from the genus Morus exhibited promising neuroprotective activity, Morus species have been rarely used to treat neurodegenerative diseases or other CNS diseases. As mGlu receptors had been considered to be valid targets for chronic pain control due to their ability of modulating rather than mediating excitatory synaptic activity [41], it was reasonable to hypothesize that these neuroprotective isoprenylated stilbenes might be the active analgesic components of Cortex Mori Radicis.
The three most effective neuroprotective stilbenes-moracin O (5), R (7), P (8)-and reported [20] flavonoid morusin (15) were assayed with the analgesic test and paracetamol was a positive control. As predicted, both moracin O (5) and P (8), with cyclic structural moieties, demonstrated more than 95% of inhibitions of acetic acid-induced pain at 80 mg/kg, which were as effective as morusin, much better than paracetamol. These results were consistent with the reported inhibition of morusin, which was about 73% at 60 mg/kg [20]. However, moracin R (7), which had an acyclic moiety, showed moderate inhibition (Table 4). As compounds 1-4 and 9 were the glycosides of moracin O (5) and P (8), they might act as predrugs

Analgesic Activities
Although some compounds from the genus Morus exhibited promising neuroprotective activity, Morus species have been rarely used to treat neurodegenerative diseases or other CNS diseases. As mGlu receptors had been considered to be valid targets for chronic pain control due to their ability of modulating rather than mediating excitatory synaptic activity [41], it was reasonable to hypothesize that these neuroprotective isoprenylated stilbenes might be the active analgesic components of Cortex Mori Radicis.
The three most effective neuroprotective stilbenes-moracin O (5), R (7), P (8)-and reported [20] flavonoid morusin (15) were assayed with the analgesic test and paracetamol was a positive control. As predicted, both moracin O (5) and P (8), with cyclic structural moieties, demonstrated more than 95% of inhibitions of acetic acid-induced pain at 80 mg/kg, which were as effective as morusin, much better than paracetamol. These results were consistent with the reported inhibition of morusin, which was about 73% at 60 mg/kg [20]. However, moracin R (7), which had an acyclic moiety, showed moderate inhibition (Table 4). As compounds 1-4 and 9 were the glycosides of moracin O (5) and P (8), they might act as predrugs through oral administration, although they did not show neuroprotective effects in vitro.

Plant Material
The Cortex Mori Radicis were bought from Anguo herb market, Hebei, China, which were collected from Hunan Province, China, in 2012.

Animals
Male ICR (Institute of Cancer Research) mice (18-22 g) from Vital River Laboratories (Beijing, China) were used in these experiments. The animals were housed in a soundproof room at an appropriate temperature (25 ± 1 • C) and humidity (50% ± 10%) for three days before the experiment. They were kept under a 12 h light/12 h dark cycle (light from 7:00 a.m. to 7:00 p.m.) with free access to food and water. Every effort was made to minimize the number of experimental animals and the discomfort that they might experience. All animals were treated humanely in compliance with the "Principles of Laboratory Animal Care" and the "Guide for the Care and Use of Laboratory Animals of Peking Union Medical College and Chinese Academy of Medical Sciences". All experimental protocols were approved by the Animal Care and Use Committee of the College (Approval NO. 00000251).

Acid Hydrolysis of Compounds 1-4
Compounds 1-4 (ca. 0.5 mg) were dissolved in 0.5 N HCl (0.1 mL) and heated at 90 • C for 2 h. The solution was neutralized with aqueous ammonia and partitioned with EtOAc (0.2 mL × 3). After drying the H 2 O-soluble fraction in vacuo, the residue was dissolved in pyridine (0.1 mL) containing L-cysteine methyl ester hydrochloride (0.5 mg) and heated at 60 • C for 1 h. A 0.1 mL solution of o-torylisothiocyanate (0.5 mg) in pyridine was added to the mixture, which was heated at 60 • C for 1 h. The reaction mixture was directly analyzed by analytical HPLC under the following conditions: column was Grace Prevail C 18 (4.6 × 250 mm, 5 µm, Grace, Deerfield, IL, USA); the mobile phase was acetonitrile and 50 mM H 3 PO 4 (25:75); flow rate was 0.8 mL/min; wavelength was 250 nm; column temperature was 35 • C. Under the above conditions, authentic D/L-glucose, D/L-arabinose gave HPLC peaks at t R 20.6/18.8 and 24.0/22.3 min, respectively. The t R of the sugars of the isolates obtained by acid hydrolysis gave similar results as those of standard sugars.

Neuroprotective Assay
Human neuroblastoma SH-SY5Y cells were grown in Dulbecco's Modified Eagle's Medium (DMEM) (Sigma-Aldrich, St. Louis, MO, USA) containing 10% fetal bovine serum (FBS), 100 U/mL penicillin/streptomycin. Cell cultures were incubated at 37 • C in a humid 5% CO 2 /95% air environment. SK-N-SH cells were cultured in 96-well microplates at a density of 1 × 10 5 cells/well. Compounds were prepared in DMSO as 100 mM stock solution. Glutamate and 3-NP were freshly prepared prior to each experiment. Na 2 S 2 O 4 was prepared in phosphate buffer solution (PBS) as 50 mM stock solution, stored in a refrigerator at −20 • C and kept away from light. The cells were preincubated with compounds for 4 h, and then the cells were separately incubated with 30 mM glutamate for 4 h, 5 mM Na 2 S 2 O 4 with DMEM lacking D-glucose solution for 24 h, and 10 µM 3-NP solution for 24 h. After the scheduled time, MTT solution (5 mg/mL) was added for another 4 h at 37 • C. MTT formazan crystals were solubilized by DMSO and spectrophotometrically measured at 570 nm (the max. emission wavelength of detected samples was at 410 nm). All data presented in our study were obtained from at least three independent experiments and expressed as the mean ± SEM (standard error of the mean). Significant differences between groups were compared using the one-way ANOVA procedure followed by a least significant difference (LSD) post hoc test using SPSS ver. 10.0 software (SPSS Inc., Chicago, IL, USA). The differences were considered statistically significant at p < 0.05.

Acetic Acid-Induced Abdominal Constrictions
ICR mice (n = 10 each group) were pretreated intraperitoneally (i.p.) with the compounds (80 mg/kg) in a separate set or standard drug paracetamol (200 mg/kg) i.p. 30 min before acetic acid injection (0.7%). Control animals received a similar volume of the appropriate vehicle (10 mL/kg) used to dilute the compounds.
The mice were placed individually into glass beakers and 5 min were allowed to elapse. The mice were then observed for a period of 10 min and the number of writhes was recorded for each animal. For scoring purposes, a writhe was indicated by stretching of the abdomen with simultaneous stretching of at least one hind limb. The formula for computing percent inhibition was: average writhes in the control group minus writhes in the drug group divided by writhes in the control group times 100%.

Molecular Modeling
Computational molecular modeling studies were carried out using Discovery Studio 4.1.0 (Accelrys, San Diego, CA, USA). All structures of glutamate receptors (2NZS, 1S50, 3UA8, 4F39, 4NF4, 3KS9, 5CNI, 5CNK, 5C5C, and 3LMK) were downloaded from the PDB website. The binding site was defined from the PDB site records. The screening of glutamate receptors was carried out by positioning the ligand in the binding site using CDocker [42], and evaluated by -CDocker energy. The interaction of ligand and receptor was further calculated using Flexible Docking [43] and the conformation of ligand and protein-binding site was evaluated by -CDocker interaction energy.

Conclusions
In summary, 4 new benzofuran-type stilbene glycosides (1-4) along with 14 known compounds were isolated from the traditional Chinese medicines, Cortex Mori Radicis. The new compounds were identified as moracin P glycosides and moracin O glycosides. Three isoprenylated benzofuran-type stilbenes, moracin O (5), R (7), and P (8), showed significant neuroprotective activity against glutamate-induced cell death in SK-N-SH cells. The molecular-docking results indicated that the active stilbenes could be mediated by the mGluR 1 pathway for their neuroprotective activities. As glutamate receptors may be implicated in persistent or chronic pain, we hypothesize that these neuroprotective isoprenylated stilbenes might be the active analgesic components of the genus Morus, and this opinion was confirmed by the inhibition of the acetic acid-induced pain by three stilbenes. Although the neuroprotective effect on glutamate-induced neuron death of benzofuran-type stilbenes such as moracin E was reported [44], this paper firstly postulated their mGluR 1 pathway and correlated the neuroprotective activity with analgesic activity, which was related with the traditional antirheumatic usage of Morus species. As compounds 1-4 and 9 were the glycosides of moracin O (5) and P (8), they may act as predrugs through oral administration, although they did not show neuroprotective effect in vitro.
As only two analgesic compounds were reported in Morus plants, we believe that our findings gave a new alternative method to discover and evaluate the analgesic components in Morus species.