Rizatriptan benzoate tablets were purchased from OuLi Pharmaceutical Co., Ltd. Fructus Viticis samples, which is the dry fruit of Vitex trifolia L, were obtained from the Hehuachi tradition herbal medicine market (Chengdu, China) and authenticated by Professor Yuntong Ma, Chengdu university of Traditional Chinese Medicine (Chengdu, China). The sample of the plant material was deposited into a Specimen bank of the Institute of Materia Medica Integration and Transformation for Brain Disorders (No. 510106180514WWen001). VME and rizatriptan benzoate tablets were respectively suspended in a 0.5% carboxymethyl cellulose (CMC)-Na solution for gastrointestinal administration. Nitroglycerin (5 mg/ml) was obtained from YiMin Pharmaceutical Co., Ltd. 5-HT, GABA, NE, Asp and Glu standards were purchased from Sigma-Aldrich (Merck KGaA). Rabbit polyclonal c-fos antibody (cat. no. ab209794) was provided by Abcam. Horseradish peroxidase-conjugated goat anti-rabbit IgG (cat. no. sc-2004) was obtained from Santa Cruz Biotechnology, Inc. Aspirin enteric-coated tablets were purchased from Bayer Healthcare Co., Ltd. Morphine hydrochloride injection was obtained from Shenyang First Pharmaceutical Co., Ltd. Normal goat serum was purchased from Zhejiang Tianhang Biotechnology Co., LTD. Acetic acid was purchased from Chengdu Kelong Chemical Reagent Factory. Chengdu Alfa Biotechnology Co., Ltd. provided casticin, luteolin and 3,6,7-trimethylquercetagetin. A SP ELISA kit (cat. no. ERC112-96) was purchased from ExCell Bio Biotechnology, Inc. A CGRP ELISA kit (cat. no. E-EL-R0135c) was obtained from Elabscience Biotechnology, Inc.

The whole Fructus Viticis samples were shredded and soaked for 30 min at room temperature each in volumes of methanol that were five times the weight of Fructus Viticis and underwent reflux extraction at 65°C for 30 min. The extract was then collected, following which the aforementioned reflux extraction process was repeated again. For administration of the extract to the animals, the methanol was removed from the VME extract in advance in a vacuum using rotary evaporator where the residue was dissolved in 0.5% CMC-Na (used as the vehicle in the control and administration groups) at a concentration of 20 g/ml. The Agilent 1260 Infinity High-Performance Liquid Chromatography instrument (Agilent Technologies, Inc.) was used to analyze the extract solution and chromatographic separation was performed using an Ultimate AQ-C18 Column (4.6×250 mm, 5 µm; Agilent Technologies, Inc.) with a column temperature of 25°C. A gradient-elution was provided with water as solvent A and acetonitrile as solvent B, using a flow rate of 1.0 ml/min, the elution started at 70:30 A:B (v/v) with the gradient gradually increased to 50:50 (A:B, v/v) for 15 min, and maintained 15 min, where 20 µl sample was injected and the effluent absorbance was measured at 254 nm. The concentration of luteolin, 3,6,7-trimethylquercetagetin and casticin was determined, which was performed using luteolin, 3,6,7-trimethylquercetagetin and casticin as reference compounds to verify the authenticity and quality of Fructus Viticis.

All animal care and experimental procedures (including the analgesic experiments and migraine experiment) were approved by the Ethics Committee for Animal Experiments of the Institute of Materia Medica Integration and Transformation for Brain Disorders (Chengdu, China). A total of 60 male Sprague Dawley rats (weight, 250–300 g; age, 8 weeks) were purchased from the Experimental Animal Center of Sichuan Academy of Medical Sciences (certificate no. SCXK2013-15). Furthermore, 18 male Sprague Dawley rats (weight, 180–220 g; age, 6 weeks) were purchased from Chongqing Enswell Biotechnology Co., Ltd. (certificate no. SCXK20170003). A total of 48 female and 16 male Kunming mice (weight, 16–22 g; age, 5 weeks) were obtained from Dashuo Biological Technology Co., Ltd. (certificate no. SCXK2013-24). All of the animals were housed at a suitable temperature (23±2°C) and humidity (40–70%) with ad libitum access to water and food and a standard 12-h light/dark cycle, where the male animals were kept separately from the females. They were acclimatized to their housing environment for seven days prior to experimentation. All animal experiments complied with the Animal Research: Reporting in vivo Experiments guidelines and were performed in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (NIH Publication no. 8023, revised 1978).

Two classic analgesic experiments were used to verify the analgesic effect of Fructus Viticis. One was the hot plate test and the other was the acetic acid-induced writhing test. Subsequently, a rat model was used to test the effect of Fructus Viticis on migraine. All products that were injected into the animals had been sterilized.

The procedures performed in the present study was performed according to a previously published protocol (25). First, all female mice were placed on a hot plate at a temperature of 55±0.5°C to test the pain threshold; the pain reaction of the mice was paw licking (the time from placing on the hot plate to the occurrence of paw licking was recorded as the paw licking time). The mice were tested twice, with no less than 5 min between each test. Onset of paw licking was considered as the basic pain threshold. Mice with paw licking times of >30 sec and <5 sec were removed and the remaining female mice (with a paw licking time of 5–30 sec) were randomly divided into four groups. These included a control group, a morphine group, a VME high-dose group and a VME low-dose group (n=8 for each group). The VME high-dose, VME low-dose, control and morphine groups were gastrointestinally administered with corresponding concentrations of VME or 0.5% CMC-Na. The oral gavage administration was given once a day for five days. At 30 min prior to the hot plate test, the morphine group received an intraperitoneal injection of 0.1% morphine hydrochloride (0.01 ml/g). The pain threshold of all groups was determined at 30, 60, 90 and 120 min after administration. If the paw licking time was >60 sec, the pain threshold was recorded as 60 sec.

The procedures performed for this test were adopted from that reported previously (26). A total of 16 male and 16 female mice were randomly divided into four groups: A control group, aspirin group, VME high-dose group and VME extract low-dose group. The VME high-dose group, the VME low-dose group and the control group received gastrointestinal administration of corresponding concentrations of VME or 0.5% CMC-Na. The aspirin group was administered with 0.6 g/kg enteric-coated aspirin by oral gavage. The administration continued for 5 days, given once a day (27). Each mouse received 0.7% acetic acid solution in normal saline by intraperitoneal injection at a dosage of 0.2 ml each hour after the last drug administration. The pain latency and number of times of writhing were observed and recorded within 15 min.

The method used in the present study was according to a previously published protocol (21). A total of 60 rats were randomly divided into five groups: Control, model, rizatriptan, VME high-dose (7 g/kg) and low-dose (1.75 g/kg) groups (n=12). The dosage of Fructus Viticis for clinical use in human patients is 30 g/day, which may be converted to dose for rats of 3.5 g/kg/day based on a weight change algorithm (28). The high-dose group received twice the clinical dose whereas the low-dose group received half the clinical dose (29). The clinical dose of rizatriptan for human patients is 10 mg/d, which may be converted to 1 mg/kg for rats. The control group, model group, VME high-dose group and VME low-dose group received the corresponding concentrations of VME or 0.5% CMC-Na only by oral gavage administration. The administration continued for 15 days, given once a day.

At 30 min after the last administration, rats in the model, rizatriptan, VME high-dose and VME low-dose groups were injected subcutaneously with nitroglycerin at a dose of 10 mg/kg. The migraine animal model used in the present study was based on that established in previous studies (30). The rats in the control group were injected subcutaneously with physiological saline at a dose of 2 ml/kg. A few minutes later, the rats in the model group appeared restless, with characteristic symptoms of red ears, head scratching, cage climbing and body shaking. These symptoms lasted for ~2 h. Red ears, head scratching and cage climbing signs were more typical than other symptom characteristics. These three symptoms were therefore used to evaluate whether the migraine model was successfully established. For each rat, the latency of ear redness was recorded. If latency was >60 min, then the latency was recorded as 60 min. The numbers of times of head scratching and cage climbing within 150 min were recorded.

The method used was described previously (31–33). At 150 min after modeling, six randomly selected rats in each group were euthanized. First, 3% pentobarbital sodium was intraperitoneally injected at a dosage of 60 mg/kg and when the righting reflex disappeared, rats were decapitated. The brain was taken out and rapidly frozen at −80°C.

For testing for NE and 5-HT, a standard curve was recorded and analysis was performed using an Acquity Ultra-Performance Liquid Chromatography/Synapt G2 Quadrupole Time-of-Flight mass spectrometer (Waters China, Ltd.). The entire brain was collected for amino acid detection. Analysis of Asp, GABA and Glu was performed using the Agilent 1260 Infinity system (Agilent Technologies, Inc.) coupled with a Nova-Pak^® C18 (5 µm) column (Waters China, Ltd). The chromatographic analysis method used was that described in the AccQ-Tag amino acid chemical reagent package (Waters China, Ltd.). The Waters AccQ-Tag amino acid chemical reagent package contains Amino acid derivatization reagent, Nova-Pak^® C18 (5 µm) column, amino acid standards and an instruction of operation and analysis.

At 150 min after modeling, six rats from each group were anesthetized with 1.5% isoflurane. Blood samples were collected from the abdominal aorta. Subsequently, CGRP and SP were detected using the ELISA kits according to the manufacturer's protocols.

Immunohistochemical staining of the brain stem was performed using a previously described method (34–36). The brain stem was fixed with fresh 4% paraformaldehyde solution. Paraffin-embedded sections with a thickness of 4 µm were then prepared, where the slices were deparaffinized and rehydrated using a decreasing ethanol gradient before being blocked with 10% normal goat serum at room temperature for 1 h. The tissue sections were incubated with rabbit anti-c-fos primary antibody (1:50) at 4°C for 48 h before incubation with goat anti-rabbit secondary antibody (1:200) at 25°C for 1 h. Then nuclei were counterstained with hematoxylin at room temperature for 2 min. The number of c-fos-immunoreactive cells (c-fos IR cells) was determined.

Values are expressed as the mean ± standard deviation. Each group of data was analyzed using SPSS 25 software (IBM Corp.). Kolmogorov-Smirnov was used to verify whether the data conformed to the normal distribution. Data with a normal distribution were compared using one-way analysis of variance followed by Tukey's post-hoc test, P<0.05 or P<0.01 were considered to indicate statistically significant differences. Statistical analyses of the results were performed out in a double-blinded fashion.

Luteolin, 3,6,7-trimethylquercetagetin and casticin are characteristic components of Fructus Viticis and were detected by HPLC. A chromatogram of the extracts and reference compounds is provided in Fig. 1. All three compounds were detected in the VME and the content of casticin met the Chinese Pharmacopoeia standard of casticin >0.03% (37).

The results of the hot plate test are provided in Fig. 2, indicating that the morphine significantly decreased the pain threshold at 30, 60 and 90 min (P<0.05), and the high dose of VME decreased the pain threshold slightly at the same time points, but no significance was observed (P>0.05). As depicted in Fig. 3, the acetic acid-induced writhing test indicated that the aspirin obviously reduced the pain latency and the number of times of writhing (P<0.01 for each). In addition, the high dose VME reduced the pain latency (P<0.05) and the number of times of writhing (P<0.05).

The rats of the model group had migraine-like responses after 5 min of nitroglycerin injection, including the use of fore- or hind paws to scratch the head and development of ear redness. There was no ear redness in the control group after subcutaneous injection of normal saline and within 120 min, only one or two episodes of head scratching and a few incidents of cage climbing were observed (Fig. 4), which may have been at random. Compared with the model group, there were significant differences regarding these parameters (P<0.01 for each). The rats in the model group exhibited shorter latencies of ear redness, as well as frequent head scratching and cage climbing, with an increased amount of discomfort-associated behaviour and a modeling success rate of 100% (P<0.01). The latency of ear redness in the rizatriptan and VME high-dose group was significantly higher compared with that in the model group (P<0.01; Fig. 4A). The amount of head scratching and cage climbing was also significantly lower than that of the model group (P<0.01; Fig. 4B and C). The effect in the VME low-dose group compared with the trend in the model group was not statistically significant.

As presented in Fig. 5, the effect of the test drugs on various neurotransmitters in the brain of rats with migraine was assessed. The 5-HT levels in the model group were lower than those in the control group (P<0.05), while pre-treatment with rizatriptan and VME at the high dose inhibited this phenomenon (P<0.01; P<0.05, Fig. 5A). Furthermore, subcutaneous injection of nitroglycerin reduced the level of NE, while this was prevented by pre-treatment with high-dose rizatriptan and VME (P<0.01, for each; Fig. 5B). The low dose of VME had no significant effect on rats with migraine (P>0.05).

The amino acid contents are provided in Fig. 5C-E. The Asp and Glu levels were increased following subcutaneous injection of nitroglycerin (P<0.01), which was significantly inhibited by pre-treatment with high-dose VME (P<0.05 and P<0.01, respectively), which was similar to the effect of rizatriptan (P<0.01 and P<0.05, respectively). The levels of GABA were decreased in the model group (P<0.05), and in comparison, they were significantly increased in the rizatriptan and high-dose VME groups (P<0.01).

The levels of SP and CGRP are provided in Fig. 6A and B, respectively. The levels of SP and CGRP were increased following subcutaneous injection of nitroglycerin (P<0.01 and P<0.05, respectively), which was significantly inhibited by pre-treatment high-dose VME (P<0.05). There was no significant difference in the SP and CGRP content between the high-dose VME and control groups (P>0.05).

C-fos-positive staining in the brain stem neurons was assessed as a central pain sensory marker. The number of c-fos IR cells in the rats of different groups was determined (Fig. 7). In the migraine model group, the number of c-fos IR cells was significantly higher than that of the control group (P<0.01). The number of c-fos IR cells in the brain stem of the VME high-dose group, the rizatriptan and VME low-dose groups were significantly lower than that in the model group (P<0.01 or P<0.05), indicating that VME reduced the migraine-associated expression of c-fos in the brain stem.