Chinese Journal of Natural Medicines  2018, Vol. 16Issue (12): 916-925  DOI: 10.1016/S1875-5364(18)30133-X
0

Cite this article as: 

WANG Jia-Wei, LIANG Feng-Yin, OUYANG Xiang-Shuo, LI Pei-Bo, PEI Zhong, SU Wei-Wei. Evaluation of neuroactive effects of ethanol extract of Schisandra chinensis, Schisandrin, and Schisandrin B and determination of underlying mechanisms by zebrafish behavioral profiling[J]. Chinese Journal of Natural Medicines, 2018, 16(12): 916-925.
[Copy]

Research funding

This work was supported by Guangzhou Clinical Research and Translational Centre for Major Neurological Diseases (No. 201604020010) and Natural Science Foundation of Guangdong Province (No. 2015A030310255)

Corresponding author

PEI Zhong, Tel:86-020-87755766-8282, E-mail:peizhong@mail.sysu.edu.cn
SU Wei-Wei, Tel:86-020-84112398, E-mail:lssswwhk@gmail.com

Article history

Received on: 01-Oct-2018
Available online: 20 December, 2018
Evaluation of neuroactive effects of ethanol extract of Schisandra chinensis, Schisandrin, and Schisandrin B and determination of underlying mechanisms by zebrafish behavioral profiling
WANG Jia-Wei1 , LIANG Feng-Yin2 , OUYANG Xiang-Shuo1 , LI Pei-Bo1 , PEI Zhong2 , SU Wei-Wei1     
1 Guangdong Engineering and Technology Research Centre for Quality and Efficacy Re-evaluation of Post-marketed TCM, Guangdong Key Laboratory of Plant Resources, School of Life Sciences, Sun Yat-sen University, Guangzhou 510275, China;
2 Department of Neurology, National Key Clinical Department and Key Discipline of Neurology, Guangdong Provincial Key La-boratory for Diagnosis and Treatment of Major Neurological Diseases, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510080, China
[Abstract]: Schisandra chinensis, a traditional Chinese medicine (TCM), has been used to treat sleep disorders. Zebrafish sleep/wake behavioral profiling provides a high-throughput platform to screen chemicals, but has never been used to study extracts and components from TCM. In the present study, the ethanol extract of Schisandra chinensis and its two main lignin components, schisandrin and schisandrin B, were studied in zebrafish. We found that the ethanol extract had bidirectional improvement in rest and activity in zebrafish. Schisandrin and schisandrin B were both sedative and active components. We predicted that schisandrin was related to serotonin pathway and the enthanol extract of Schisandra chinensis was related to seoronin and domapine pathways using a database of zebrafish behaviors. These predictions were confirmed in experiments using Caenorhabditis elegans. In conclusion, zebrafish behavior profiling could be used as a high-throughput platform to screen neuroactive effects and predict molecular pathways of extracts and components from TCM.
[Key words]: Traditional Chinese medicine     Schisandra chinensis     Zebrafish     Behavior profiling     Neurotransmitter    
Introduction

Sleep disorder is a major health issue worldwide and can have a great negative impact on individuals [1-2]. Pharmacological agents are the most popular treatments for insomnia, because of easy accessibility and time-saving. Western hypnotic drugs have many side effects, including addiction or mental disorders [3]. In contrast, traditional Chinese medicine (TCM) has been employed to treat insomnia for centuries [4]. The popularity of TCM therapy for insomnia is increasing worldwide, because of its stable curative effect and minimal adverse reactions [5].

Schisandra chinensis is widely used in many TCM prescriptions to cure various neurobehavioral disorders, including sleep disorder [6]. The ethanol extract of Schisandra chinensis (EESC) is commonly used in pharmacological research of this plant [7-8]. Lignins are active components of Schisandra chinensis [9-10]. Schisandrin and schisandrin B are the major lignins of Schisandra chinensis, accounting for up to 28%-50% of EESC [11]. Thus, we used EESC, schisandrin and schisandrin B to evaluate the effects of Schisandra Chinensis.and explore the underlying mechanisms of action.

Considering the complexity of brain activity, using appropriate in vivo model for screening novel neuroactive drugs, instead of experiments in vitro, is a rational approach. Compared with the high-cost but low-throughput screening in mice and rats, zebrafish are small, easy to breed, and inexpensive to maintain and grow rapidly. Zebrafish larvae and adults have anatomical, behavioral, pharmacological and molecular correlates of mammalian sleep [12-13]. Particularly, drugs that modulate sleep/wake in mammals, including melatonin, barbiturates, and benzodiazepines, have similar effects in zebrafish larvae [14-16]. Furthermore, a large-scale zebrafish larvae behavioral profiling database has been established [17]. It includes the behavior characteristics of nearly 4000 small molecules of known structures and functions. However, the behavior profiling system has not been applied to studying neuromodulative effects or mechanism of actions of extracts or small molecules from TCM.

In the present study, Caenorhabditis elegans (C. elegans) was used to confirm the results obtained in zebrafish. It is considered as an ideal model of neurological disability [18], because the neural pathways of nematodes and mammals are highly conserved, such as dopamine and serotonin pathways [19]. Its well-characterized behavioral phenotypes, including egg laying and pharyngeal pumping, and diverse strains are advantageous for studying mechanistic pathways [20-22].

Materials and Methods Drug preparation

Fructus Schisandrae was purchased from Guangzhou Daxiang herb shop and identified according to the guidelines of the Chinese Pharmacopoeia (2015) by Prof. LIAO Wen-Bo, and conserved in Sun Yer-Sen herbarium of type specimens (187013). The raw material was extracted with 95% aqueous EtOH thrice by reflux. The combined extracts were concentrated under reduced pressure to dryness. Schisandrin (CAS#61281- 38-7), schisandrin B (CAS#61281-37-6), fluoxetine (CAS#54910- 89-3) and serotonin (CAS#50-67-9) standard substances were obtained from Sigma–Aldrich (Saint Louis, MO, USA).

Zebrafish preparation

Parental zebrafish (AB strain, Zebrafish Model Animal Facility, Institute of Clinical and Translational Research, Sun Yat-Sen University) were cultivated in tanks containing circulating water at a constant temperature of 28.5 ℃. The light/dark cycle of the room was 14 h/10 h per day. Two hours after fertilization, the eggs were collected and transferred into a flat dish with Holt buffer (KCl 0.05 g·L-1; NaHCO3 0.025 g·L-1; NaCl 3.5 g·L-1; CaCl2 0.1 g·L-1; 1 mg·L-1 of methylene blue, PH 7.0). Dead embryos were removed and Holt buffer replaced daily. All experimental protocols and procedures involving zebrafish were undertaken in accordance with the Guide For The Care And Use Of Laboratory Animals (number 8023, revised in 1996; National Institutes of Health, Bethesda, MD, USA).

Exposure to drug solutions

At 4 days post-fertilization (4 dpf), the zebrafish larvae were transferred into 96-well plates with 300 μL Holt buffer separately. Then, 60 µL of mother solution (six times concentration) was added to each well. Each row of the 96-well plate was a group containing 12 larvae. The first row was the dimethyl sulfoxide (DMSO) group, containing 3‰ DMSO. The second row contained 10 μg·mL-1 of fluoxetine. The other rows contained a range of final concentrations of drugs (40, 20, 10, 5, 1, and 0.2 μg·mL-1). EESC and lignins were dissolved in DMSO to obtain stock solutions of 100 mg·mL-1.

Zebrafish quantization

The 96-well plate was placed in a plastic groove containing homothermal flowing water at 28.5 ℃. The water was circulated between the groove and water-flow machine (ViewPoint, Nice, France). The behaviors of zebrafish larvae were observed by a video tracking system (VideoTrack, ViewPoint, France). The program parameters for detection were set as follows: detection sensitivity, 15; detection threshold, 40; burst, 25; freeze, 4; and bin size, 60s. The light/dark parameters were set as follows: light between 09: 00 and 23: 00; light intensity, 50%; and light changing duration, 10s.

Zebrafish behavior data analysis

The raw data for each zebrafish larva were time series recording the amount of time it moved in each minute during 24 h. Six parameters were calculated for each larva as reported previously [17]. One minute with less than 0.1 second of movement was defined as a "rest minute". A continuous string of rest minutes was defined as a "rest bout". The amount of moving time in each minute was defined as "activity". The six parameters were defined as follows:

Rest total: the total amount of rest minutes;

Rest bout: the number of rest bouts;

Rest bout length: the average length of rest bouts;

Rest latency: the duration between a light transition and the first rest bout;

Activity total: average amount of detected activity in seconds, including all rest minutes; and

Waking activity: the total amount of detected activity, excluding all one-minute periods of rest.

These parameters were generated automatically by R software (The Comprehensive R Archive Network, TUAN Team, Tsinghua University). The program was written in advance.

The parameters of each group for each experimental day and night were the mean of three independent experiments (36 larvae), and expressed as means ± standard deviation (SD) from the control group for fingerprints. Color gradation was undertaken to present the degrees of standard deviation. Positive values were shown in yellow, and negative value in blue. The standard deviation approaching zero was shown in black.

Histogram

Histograms were created to reflect the change of mean rest total and waking activity for each concentration (0.2, 1, 5, 10, 20, and 40 μg·mL-1), and separated to day and night. Each value represents the average of 36 larvae. Bars are the means ± SEM (n = 36). Statistical analysis were done using one way ANOVA.

Time series graph

Rest and waking activity were averaged in 10-min intervals and then normalized to line graphs. Each line represents the mean of 36 larvae of each group. Groups of EESC (10 μg·mL-1), schisandrin (20 μg·mL-1) and schisandrin B (20 μg·mL-1) were selected to compare with the respective controls groups. Instead of a time axis, white and black bars were placed on the horizontal axis to represent the day and night measurements, respectively. Differences in each 10-min interval between treatments and controls were calculated by one-way ANOVA. Significantly different time series > 1 h are shown by a horizontal line.

Dose correlation

All correlation analysis was undertaken using SPSS v19.0 (IBM, Armonk, NY, USA). Uncentered Pearson's correlation coefficients were calculated for different concentration pairs according to mean ± SD values from the control group. Correlation coefficients were also calculated using color gradation with positive values shown in red.

Target correlation

Rihel et al. [17] have reported a high-through put method about zebrafish larvae and applied to predict the target of hundreds of compounds. We generated the database from fingerprints as reported previously [23]. Representative chemicals were sorted by neural pathways, including the agonist or antagonist of adrenaline, serotonin, dopamine, gamma-aminobutyric acid (GABA), histamine, adenosine, melatonin, and glutamate receptors. A random chemical was selected for each target to form a brief database. Based on this database, the correlation coefficients between our treatments and representative chemicals were calculated using color gradation.

C. elegans preparation

N2 and tph-1 mutation strains of C. elegans were generated from the Department of Neurology, First Affiliated Hospital, Sun Yat-Sen University, Guangdong, China). Adult worms who had not suffered starvation in past three generation were used. They were fed E. coli OP 50 on nematode growth medium (NGM) agar plates.

Egg laying

Egg laying behavior was evaluated following a well-established assay protocol [24]. Thirty L4-stage larvae were removed and placed on fresh plates seeded with bacteria for each group. About 20 h later, the larvae were transferred onto a 96-well plate containing 50 μL of solution of a particular drug at 20 ℃. Ten worms from each group were selected randomly and placed into wells. The concentrations of drugs in M9 buffer (μg·mL-1) were 10 and 100 for EESC, 10 and 100 for schisandrin, 100 for 5-hydroxytryptamine (5-HT), and 100 for fluoxetine, respectively. In the control group, 3% DMSO was added in M9 buffer. The plate was left for 60 min at room temperature. Egg-laying during this period was counted using a dissecting microscope with a 12× objective lens.

Pharyngeal pumping

After synchronization, the eggs were cultured on NGM agar plates containing different content and placed into an incubator with constant temperature (20 ℃) for 2 days. The drugs were added to E. coli OP 50 bacterial solution at the same concentrations used in the egg-laying experiment. Thirty L4-stage larvae were removed and added to new plates containing drugs. About 20 h later, C. elegans were prepared for use for observation of pharyngeal pumping [25]. The latter was recorded using a stereo microscope (Nikon, Tokyo, Japan) and videos were generated by a software (UV 200; Nikon). Ten worms were selected randomly from 30 worms of each group. The rate of pharyngeal pumping was counted by three assistants by replaying the videos at one-third of the original speed. The experiments were repeated thrice independently.

Results Establishment of screening of zebrafish-larvae behavior

High-throughput screening based on larval locomotor activity was utilized to explore the neural activity of EESC, schisandrin, and schisandrin B. The 96-well plates were prepared by adding 4-dpf larvae to treatment solution and 12 larvae in each row formed a group. The first row was the control group (containing 3‰ DMSO). The positive control group (containing 10 μg·mL-1 of fluoxetine) was employed for each plate to compare the change in larvae activity between generations for different plates. Other rows were treatment groups (six concentrations of drugs) (Fig. 1A). The effects of different treatments were assessed using the total amount of rest, the number and length of rest bouts, the rest latency, and activity (Fig. 1B).

Figure 1 Quantification of the behavior of zebrafish larvae. (A) Each well of a 96-well plate contained a single 4-dpf larva and a content of treatment. Twelve larvae were prepared for each group and parallel experiments were undertaken for at least thrice. The first group was control with 3‰ DMSO, and the second group was positive control with 10 μg·mL-1 fluoxetine. The concentrations of treatments (EESC, Schisandrin and Schisandrin B) were 40, 20, 10, 5, 1 and 0.2 μg·mL-1. (B) Locomotor activity of a represent larvae. The rest and wake behavior were recorded, including the number of inactivity minutes (rest total), a continuous minute of rest (rest bout), the duration between a light transition and the first rest bout (rest latency), average active time per minute (activity total), and the average activity excluding rest periods (waking activity)
Changes in behavior of zebrafish larvae induced by different treatments

Fingerprinting was applied to show the behavioral characters induced by treatments. We found that 40 and 20 μg·mL-1 of EESC caused massive mortality of larvae and difficult to be calculated, leaving four concentrations of EESC to be analyzed. Fingerprinting revealed that different concentrations of EESC increased the rest parameters (rest total, rest bout, and rest bout length) slightly and waking activity during the observation. Activity total was decreased slightly. Rest latency was increased during daytime and decreased at night. Interestingly, schisandrin presented similar effects to EESC with regard to rest parameters. Rest total and rest bout length were improved during day or night, and rest bout length was changed remarkably. Rest latency changed in opposite ways during light or dark measurements. Schisandrin had few effects on activity total or waking activity. In contrast, schisandrin B enhanced activity in the daytime but had few effects on rest.

Quantitative analysis of rest total and waking activity induced by treatments

The rest total and waking activity were used for quantitative analysis as representative parameters for rest/wake behavior in larvae [17, 23]. EESC (1-10 μg·mL-1) showed a neuroactive effect during day and night. EESC (< 5 μg·mL-1) acted in a dose-dependent manner (Fig. 3A). Schisandrin increased rest total in light and dark measurement, and the efficacy was dose-dependent between 1-20 μg·mL-1. Waking activity was changed slightly by schisandrin (except for 40 μg·mL-1) (Fig. 3B). Schisandrin B (2-20 μg·mL-1) improved waking activity during the daytime, but had no effect on rest total (Fig. 3C).

Figure 2 Drug-induced characteristics of zebrafish larvae behavior. Each row represents a unique concentration of a certain treatment for 36 larvae. Six parameters (rest total, rest bout, rest bout length, rest latency, activity total and waking activity) were calculate to describe behavior characters of zebrafish larvae behavior for each group. White and black bars under the parameter indicate day and night measurements, respectively. The standard deviations of parameters between the control group and each group were calculated and presented by color gradation. Yellow: this parameter is higher than controls. Blue: this parameter is higher than controls
Figure 3 Quantitative analysis of rest total and waking activity. (A-C)Each value represents the mean of 36 larvae. Error bars represent the standard error of the means (SEM). *P < 0.05 and **P < 0.01 vs control by one way ANOVA. The white and black bars represent the day and night measurements respectively
Changes in behavior characteristics in time series

We selected representative concentrations of treatments in fingerprint and created graphs of rest and waking activity time series to compare behavior during the day and night more intuitionally. The significant effects (P < 0.05) on rest and activity of EESC (5 μg·mL-1) were separated to several periods (Fig. 4A). Schisandrin had significant effect (P < 0.05) on rest total at all times. Schisandrin B advanced wakefulness during most of the daytime (Figs. 4B and C).

Figure 4 Normalized time series of behavioral profiles. (A-C) Rest and waking activity were averaged in 10-minute intervals of 36 larvae, then normalized to line graphs. The solid line indicates the average of controls. The dotted lines indicate the average of 10 μg·mL-1 of EESC, 20 μg·mL-1 schisandrin and 20 μg·mL-1 schisandrin B groups, respectively. The white and black bars on horizontal axis represent the day and night measurements respectively. *P < 0.05 and **P < 0.01 vs control significantly different by one way ANOVA. Significant different time series > 1 h was shown by a horizontal line
Prediction of the neural signaling pathways

Correlation matrices of different concentrations were established first to exclude uncorrelated concentrations for target prediction. Schisandrin (40 and 0.2 μg·mL-1), and schisandrin B (0.2 μg·mL-1) were uncorrelated concentrations (Fig. 5A). If the correlation coefficient between concentrations and known chemicals was > 0.6, the treatment was considered to be related to the target corresponding to the chemicals (Fig. 5B). Heat maps revealed that, EESC was related to different targets, including serotonin and dopamine. Schisandrin was related to a serotonin pathway. Schisandrin B did not show a correlation to any of the targets in the database.

Figure 5 Target correlation analysis. This analysis based on the parameters of each group of 36 larvae, and correlation coefficient was calculated in uncentred Pearson's correlation coefficient. (A) Each color square of matrix represents the correlation coefficient of different concentrations. Red indicates high correlation and blue indicates a low correlation. (B) Each color square of matrix represents the correlation coefficient between treatments and known chemicals. Neuropathways were separated in the first row, including agonist or antagonist of adrenaline, serotonin, dopamine, GABA, histamine, melatonin, adenosine and the glutamate receptor. A representative chemical for each target was selected randomly. Red indicates high correlation and blue indicates a low correlation
Confirmation of neuroactive action on C. elegans

To test our prediction, egg laying and pharyngeal pumping assays were conducted. Pharyngeal pumping is a serotonin dependent behavior [26]. Egg laying is modulated by serotonin and dopamine system [27]. A serotonin synthesis defective tph-1 mutant strain was introduced to study behavior in the absence of serotonin [28]. Serotonin and fluoxetine, a typical medicine of selective serotonin reuptake inhibitor (SSRI), were chosen as positive controls [29].

EESC, schisandrin, serotonin, and fluoxetine improved the rate of pharyngeal pumping and egg laying on the N2 strain comparing with the control (Figs. 6A and B). Schisandrin (100 μg·mL-1) induced a higher egg laying rate than serotonin or fluoxetine. The pharyngeal pumping rate (Figs. 6B and D) was decreased in the serotonin synthesis defective tph-1 mutant strain than in N2 strain (tph-1 vs N2: 30-40/10s vs 50-60/10s). The egg laying rate (Figs. 6A and C) was also inhibited on tph-1 mutant (tph-1 vs N2: 0.58/h vs 0.83/h). Serotonin addition induced a significant improvement in the pharyngeal pumping and egg laying rates in the tph-1 strain (Figs. 6C and D). Interestingly, the improvement effect on egg laying and pharyngeal pumping of schisandrin and fluoxetine in the N2 strain disappeared in the tph-1 mutant. Moreover, 10 and 100 μg·mL-1. EESC also promote the egg laying rate in the tph-1 mutant.

Figure 6 Drug-induced behaviors in C. elegans. The control group was treated by 3‰ DMSO (grey), positive control groups were treated by 100 μg·mL-1 serotonin and fluoxetine respectively. White/ black bars represent 10 μg·mL-1 and 100 μg·mL-1 treatments. Bars represent mean ± SEM, n = 30, *P < 0.05 and **P < 0.01 vs control, #P < 0.05 vs fluoxetine, calculated by one way ANOVA. (A-B) Different effects of treatments on pharyngeal pumping and egg laying in the N2 strain. (C-D) TPH-1 mutant was used to reduce the production of 5-HT remarkably. The graphs represent pumping and egg laying in an absence of 5-HT
Discussion

In the present study, we showed for the first time that EESC had hypnotic and neuroactive effects in zebrafish, probably through serotonin and dopamine pathways. Also schisandrin was one of the components responsible for the sedative effect of Schisandra chinensis and this effect was serotonin- dependent. In contrast, schisandrin B was an active component. The present study indicated the efficacy and mechanism of Schisandra chinensis, and the serotonin-like action of schisandrin could be used to alleviate sleep disorders and depression.

Rest total is the most direct parameter showing the sedative effects of treatments. Rest bout is another important criterion for behavior defined sleep-like states, and the duration of rest bouts reflects the circadian rhythms in zebrafish [30]. Rihel et al. have shown that hypnotic drugs, including melatonin and GABA-A agonists, promote the two parameters [17]. In our study, EESC and schisandrin improved rest length and rest bout length, indicating a hypnotic effect on zebrafish. Moreover, EESC and schisandrin had considerable similar sedative effects, suggesting that schisandrin is one of the components responsible for the sedative effects of Schisandra chinensis. Under the maximum non-toxic concentrations, schisandrin induced significant promotion of rest total compare with EESC, suggesting its potential application for alleviating sleep disorders. Conversely, EESC and schisandrin B improved on waking activity, suggesting that schisandrin B wasa potential active component of EESC.

Although EESC had neuromodulative actions, they were relatively weak both in zebrafish or C. elegans because Schisandra chinensis is a mutiple-function herb with several non-neuromodulative active sites. Identification and evaluation of its neuro modulatory active sites could improve specific and intensive effects [31]. In the present study, schisandrin was identified as an active chemical of the sedative effect of EESC. In contrast, schisandrin B had no significantly rest change, suggesting that it could be removed in sedative site acquisition of EESC.

Neuroactive molecular pathways were predicted based on correlations with known chemicals. This analysis suggested that EESC was related to a serotonin and dopamine system. These two pathways induced improvement on rest and activity in zebrafish larvae [17]. Nevertheless, the behavioral characteristics of these two pathways were distinguishable in the database, suggesting that the correlations of EESC differed. Schisandrin was related to a serotonin system. Slight behavior changes of schisandrin B did not match the matrix, because quantitative values from the known compounds in database are large.

This prediction was confirmed by experiments in C. elegans. The pharyngeal pumping rate is a serotonin-reliant behavior. Mutations bearing the tph-1 deletion do not synthesize serotonin [28]. Thus, the pharyngeal pumping rate decreased in the tph-1 mutant in our study. EESC and schisandrin increased the pharyngeal pumping rate significantly in the N2 strain, but this promotion disappeared in the absence of serotonin. These results demonstrated that EESC and schisandrin were related to the serotonin system. The egg-laying rate is modulated by a serotonin and dopamine system. The increase in the egg-laying rate induced by schisandrin disappeared in the tph-1 strain. In contrast, EESC could promote the egg-laying rate in both strains. Hence, EESC not only induced behaviors through a serotonin system, but also had effects on a dopamine system. Schisandrin has serotonin-dependent effects similar to selective serotonin reuptake inhibitors (SSRIs). Some scholars have reported that schisandrin is an active component related to serotonin, and that EESC has a dopaminergic effect [32-33]. Our research suggested that the molecular pathways of extracts or components from Schisandra Chinensis could be found efficiently using profiling of zebrafish behavior.

Serotonin controls synchronization of the biological clock. It has been proposed to be sedative [34-36] and related to sleep problems under pathological conditions [37]. Some hypnotic medicines elicit their effects through a serotonin pathway [38]. Dopamine also participates in the control of sleep and wakefulness [39]. These hypotheses suggest the sedative effect of EESC and schisandrin. Given that schisandrin can pass through the blood-brain barrier [40], identification of the serotonin-like effects of EESC and schisandrin may provide information to help alleviate depression, anxiety and sleep disorders [41-42]. Nevertheless, the biological activities of other major chemicals in EESC, the relations between active chemicals, as well as a more accurate target verification of schisandrin to SSRIs, will be studied further by our research team in the future.

In conclusion, we demonstrated that EESC had sedative and neuroactive effects on zebrafish larvae. Schisandrin and schisandrin B were two of the sedative and active components of Schisandrin chinensis. We also predicted that schisandrin was related to serotonin pathway and EESC was related to serotonin and dopamine pathways by using a database of zebrafish behavior. These predictions were also confirmed with experiments of C.elegans. We further demonstrated that zebrafish behavior profiling could be used as a high-through platform to screen efficacy and predict neuroactive molecular pathways of extracts and components of TCM.

Acknowledgements

We thank the Zebrafish Model Animal Facility, Institute of Clinical and Translational Research, Sun Yat-Sen University for providing AB strain zebrafish eggs.

References
[1]
Neikrug AB, Maglione JE, Liu L, et al. Effects of sleep disorders on the non-motor symptoms of parkinson disease[J]. J Clin Sleep Med, 2013, 9(11): 1119-1129.
[2]
Ohayon MM. The effects of breathing-related sleep disorders on mood disturbances in the general population[J]. J Clin Psychiat, 2003, 64(10): 1195-1200. DOI:10.4088/JCP.v64n1009
[3]
Kuriyama A, Honda M, Hayashino Y. Ramelteon for the treatment of insomnia in adults: a systematic review and meta- analysis[J]. Sleep Med, 2014, 15(4): 385-392. DOI:10.1016/j.sleep.2013.11.788
[4]
Shi MM, Piao JH, Xu XL, et al. Chinese medicines with sedative-hypnotic effects and their active components[J]. Sleep Med Rev, 2016, 29(3): 108-118.
[5]
MM SO, LB HI, et al. The use of natural products for sleep: A common practice?[J]. Sleep Med, 2009, 10(9): 982-987. DOI:10.1016/j.sleep.2008.10.009
[6]
Zhang FY, LI JJ, Zhou Y, et al. Review for sedative and hypnotic mechanism of sedative traditional Chinese medicine and relative active components on neurotransmitters[J]. Chin J Chin Mater Med, 2016, 41(23): 4320.
[7]
Chun JN, Cho M, So I, et al. The protective effects of Schisandra chinensis fruit extract and its lignans against cardiovascular disease: a review of the molecular mechanisms[J]. Fitoterapia, 2014, 97(17): 224.
[8]
Li Z, He X, Liu F, et al. A review of polysaccharides from Schisandra chinensis and Schisandra sphenanthera: Properties, functions and applications[J]. Carbohyd Polym, 2018, 184: 178-190. DOI:10.1016/j.carbpol.2017.12.058
[9]
Lin HM, Han HX, Li YH, et al. Correlation study on lignin contents of Schisandra chinensis and ecological factors[J]. Chin J Chin Mater Med, 2013, 38(38): 4281-4286.
[10]
Wang BL, Hu JP, Sheng L, et al. Chemical-pharmacokinetic- pharmacodynamic fingerprints of Schisandra chinensis alcoholic extract[J]. Acta Pharm Sin, 2013, 48(5): 734-740.
[11]
Lee HJ, Kim CY. Simultaneous determination of nine lignans using pressurized liquid extraction and HPLC-DAD in the fruits of Schisandra chinensis[J]. Food Chem, 2010, 120(4): 1224-1228. DOI:10.1016/j.foodchem.2009.11.068
[12]
Rihel J, Schier AF. Behavioral screening for neuroactive drugs in zebrafish[J]. Dev Neurobiol, 2012, 72(3): 373-385. DOI:10.1002/dneu.20910
[13]
Richendrfer H, Creton R. Cluster analysis profiling of behaviors in zebrafish larvae treated with antidepressants and pesticides[J]. Neurotoxicol Teratol, 2018, 69: 54-61. DOI:10.1016/j.ntt.2017.10.009
[14]
Zhdanova IV, Wurtman RJ, Regan MM, et al. Melatonin treatment for age-related insomnia[J]. J Clin Endocr Metab, 2001, 86(10): 4727. DOI:10.1210/jcem.86.10.7901
[15]
Ruuskanen JO, Peitsaro N, Kaslin JV, et al. Expression and function of alpha-adrenoceptors in zebrafish: drug effects, mRNA and receptor distributions[J]. J Neurochem, 2005, 94(6): 1559-1569. DOI:10.1111/jnc.2005.94.issue-6
[16]
Renier C, Faraco JP, Motley T, et al. Genomic and functional conservation of sedative-hypnotic targets in the zebrafish[J]. Pharmacogenet genom, 2007, 17(4): 237-253. DOI:10.1097/FPC.0b013e3280119d62
[17]
Rihel J, Prober DA, Arvanites A, et al. Zebrafish behavioral profiling links drugs to biological targets and rest/wake regulation[J]. Science, 2010, 327(5963): 348-351. DOI:10.1126/science.1183090
[18]
Calahorro F, Ruiz-Rubio M. Caenorhabditis elegans as an experimental tool for the study of complex neurological diseases: Parkinson's disease, Alzheimer's disease and autism spectrum disorder[J]. Invert Neurosci, 2011, 11(2): 73-83. DOI:10.1007/s10158-011-0126-1
[19]
Schmid T, Snoek LB, Fröhli E, et al. Systemic regulation of RAS/ MAPK signaling by the serotonin metabolite 5-HIAA[J]. Plos Genet, 2015, 11(5): e1005236. DOI:10.1371/journal.pgen.1005236
[20]
Brown MK, Luo Y. Bilobalide modulates serotonin-controlled behaviors in the nematode Caenorhabditis elegans[J]. BMC Neurosci, 2009, 10(1): 62. DOI:10.1186/1471-2202-10-62
[21]
Dempsey CM. Serotonin (5HT), Fluoxetine, Imipramine and Dopamine Target Distinct 5HT Receptor Signaling to Modulate Caenorhabditis elegans Egg-Laying Behavior[J]. Genetics, 2004, 169(3): 1425-1436.
[22]
Hobson RJ, Hapiak VM, Xiao H, et al. SER-7, a caenorhabditis elegans 5-HT7-like receptor, is essential for the 5-HT stimulation of pharyngeal pumping and egg laying[J]. Genetics, 2006, 172(1): 159-169.
[23]
Wang YN, Hou YY, Sun MZ, et al. Behavioural screening of zebrafish using neuroactive traditional Chinese medicine prescriptions and biological targets[J]. Sci Rep, 2014, 4(24): 5311.
[24]
Dempsey CM, Mackenzie SM, Gargus A, et al. Serotonin (5HT), fluoxetine, imipramine and dopamine target distinct 5HT receptor signaling to modulate caenorhabditis elegans egg-laying behavior[J]. Genetics, 2005, 169(3): 1425-1436. DOI:10.1534/genetics.104.032540
[25]
Hart A. Behavior[M]. WormBook, 2006.
[26]
Horvitz HR, Chalfie M, Trent C, et al. Serotonin and octopamine in the nematode Caenorhabditis elegans[J]. Science, 1982, 216(4549): 1012. DOI:10.1126/science.6805073
[27]
Weinshenker D, Garriga G, Thomas JH. Genetic and pharmacological analysis of neurotransmitters controlling egg laying in C. elegans[J]. J Neurosci, 1995, 15(10): 6975-6985. DOI:10.1523/JNEUROSCI.15-10-06975.1995
[28]
Sze JY, Victor M, Loer C, et al. Food and metabolic signalling defects in a Caenorhabditis elegans serotonin-synthesis mutant[J]. Nature, 2000, 403(6769): 560-564. DOI:10.1038/35000609
[29]
Avery L, Horvitz HR. Effects of starvation and neuroactive drugs on feeding in Caenorhabditis elegans[J]. J Exp Zool, 1990, 253(3): 263-270. DOI:10.1002/(ISSN)1097-010X
[30]
Hirayama J, Kaneko M, Cardone L, et al. Analysis of circadian rhythms in zebrafish[J]. Method Enzymol, 2005, 393(18-19): 186-204.
[31]
Guo Z. Artemisinin anti-malarial drugs in China[J]. Acta Pharm Sin B, 2016, 6(2): 115-124. DOI:10.1016/j.apsb.2016.01.008
[32]
Yan T, He B, Wan S, et al. Antidepressant-like effects and cognitive enhancement of Schisandra chinensis in chronic unpredictable mild stress mice and its related mechanism[J]. Sci REP-UK, 2017, 7(1): 6903. DOI:10.1038/s41598-017-07407-1
[33]
Zhang C, Zhao X, Mao X, et al. Pharmacological evaluation of sedative and hypnotic effects of schizandrin through the modification of pentobarbital-induced sleep behaviors in mice[J]. EUR J Pharmacol, 2014, 744: 157-163. DOI:10.1016/j.ejphar.2014.09.012
[34]
Jouvet M. Neuromediators and hypnogenic factors[J]. Revue Neurol, 1984, 140(6-7): 389.
[35]
Moriya S, Tahara Y, Sasaki H, et al. Phase-delay in the light- dark cycle impairs clock gene expression and levels of serotonin, norepinephrine, and their metabolites in the mouse hippocampus and amygdala[J]. Sleep Med, 2015, 16(11): 1352-1359. DOI:10.1016/j.sleep.2015.06.020
[36]
Zhang X, Yan H, Luo Y, et al. Thermoregulation-independent regulation of sleep by serotonin revealed in mice defective in serotonin synthesis[J]. Mol Pharmacol, 2018, 93(6): 657-664. DOI:10.1124/mol.117.111229
[37]
Wilson H, Giordano B, Turkheimer FE, et al. Serotonergic dysregulation is linked to sleep problems in Parkinson's disease[J]. Neuroimage-Clin, 2018, 18: 630-637. DOI:10.1016/j.nicl.2018.03.001
[38]
Zhao X, Cui XY, Chen BQ, et al. Tetrandrine, a bisbenzylisoquinoline alkaloid from Chinese herb Radix, augmented the hypnotic effect of pentobarbital through serotonergic system[J]. EUR J Pharmacol, 2004, 506(2): 101-105. DOI:10.1016/j.ejphar.2004.10.046
[39]
Oishi Y, Lazarus M. The control of sleep and wakefulness by mesolimbic dopamine systems[J]. Neurosci Res, 2017, 118: 66-73. DOI:10.1016/j.neures.2017.04.008
[40]
Wei B, Li Q, Fan R, et al. UFLC–MS/MS method for simultaneous determination of six lignans of Schisandra chinensis (Turcz.) Baill. in normal and insomniac rats brain microdialysates and homogenate samples: towards an in‐depth study for its sedative-hypnotic activity[J]. J Mass Spectrom, 2013, 48(4): 448. DOI:10.1002/jms.3176
[41]
Tchekalarova J, Stoynova T, Ilieva K, et al. Agomelatine treatment corrects symptoms of depression and anxiety by restoring the disrupted melatonin circadian rhythms of rats exposed to chronic constant light[J]. Pharmacol Biochem Be, 2018, 171: 1-9. DOI:10.1016/j.pbb.2018.05.016
[42]
Aleksovski B, Novotni A, Vujović V, et al. Evaluation of peripheral serotonin content and α2-adrenergic receptor function as potential markers for life-long recurrent depressive disorder by using methodological improvements[J]. INT J Geriatr Psych, 2017, 22(3): 215-224.