Network pharmacology strategy to investigate the pharmacological effects of Suanzaoren Decoction on insomnia

Abstract Suanzaoren Decoction (SZRD) is an ancient prescription used in the treatment of insomnia. This study aimed to investigate the components and targets of SZRD in treating insomnia. First, the compounds of five herbs in SZRD were collected from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP), and the putative targets for treating insomnia were obtained from DrugBank to construct the herb-compound-target- disease network. A protein-protein interaction (PPI) network was constructed in the STRING database, and then Gene Ontology functional enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were performed to predict the mechanism of action of intersection target. Finally, 30 mice were divided into five groups: control, model, and quercetin groups (100, 50, 25 mg/kg). The sleep latency and duration of pentobarbital-induced sleeping were measured. The production of interleukin-6 (IL-6) and γ-aminobutyric acid (γ-GABA) was detected by using an enzyme-linked immunosorbent assay kit (ELISA), and Gamma-aminobutyric acid type a receptor subunit alpha1 (GABRA1) was tested by Reverse Transcription-Polymerase Chain Reaction (RT-PCR). A total of 152 active ingredients, including 80 putative targets of SZRD, were obtained. The main active compounds included quercetin and kaempferol, and the key targets involved IL-6 and nitric oxide synthase 3 (NOS3). The results of pathway enrichment analysis indicated that the putative targets of SZRD mainly participated in Neuroactive ligand-receptor interaction. The experiment of P-chlorophenylalanine (PCPA)-induced insomnia model showed that quercetin obviously shortened the sleep latency and prolonged the sleep duration of the insomnia model. The production of IL-6, γ-GABA, and GABRA1 mRNA was significantly increased in mice treated with quercetin. This study predicted the active ingredients and potential targets of SZRD on insomnia on the basis of a systematic network pharmacology approach and illustrated that SZRD might exert hypnotic effects via regulating IL-6, γ-GABA, and GABRA1.

Suanzaoren Decoction (SZRD) is an ancient prescription and classic clinical decoction comprising five herbs, including Ziziphi Spinosae Semen (ZSS), Anemarrhenae Rhizoma (AR), Poria (P), Chuanxiong Rhizoma (CR) and Glycyrrhizae Radix Et Rhizoma (GRR); moreover, it has anticonvulsant, blood-nourishing, cardioprotective, and immunity-enhancing effects (Song et al., 2020).In a previous study, SZRD was found to shorten sleep latency and prolong sleep time to improve sleep quality in a sleep-deprivation rat model (Zhan et al., 2020).Despite the popular use of SZRD, few studies have explored the potential molecular mechanism for insomnia treatment because of the multiple pathways, target points, and elements of prescription (Lee et al., 2018).
Network pharmacology is a new technology that integrates chemoinformatics, bioinformatics, network biology, network analysis, and traditional pharmacology and meets the key ideas of the holistic philosophy of TCM (Berger, Iyengar, 2009;Li, Zhang, 2013).For example, Yu et al. (2018) investigated the mechanism of Zuojinwan for the treatment of gastritis by network pharmacologybased strategy.
Consequently, the present study aimed to investigate the components and targets of SZRD to explore its mechanism in the treatment of insomnia.In addition, on the basis of the results of network pharmacology, in vivo experiment was conducted to explore the potential compounds and mechanism of treating insomnia by SZRD.

herb-compound-target-disease network constructed
To further determine the molecular mechanism of SZRD on insomnia, the common targets of SZRD and insomnia were determined, and the herb-compoundtarget-disease network was constructed using Cytoscape 3.6.1,which is a software package for visualizing network analysis.In these graphical networks, the compounds, proteins, or pathways were expressed as nodes, whereas the compound-target or target-pathway interactions were expressed as edges.

PPI network constructed
PPI network data were derived from the STRING database (http://string-db.org/).The common targets of SZRD and insomnia were imported into the STRING database with the limited condition of "Homo sapiens" to generate the PPI network.The graphical networks were constructed using Cytoscape 3.6.1.

Pathway enrichment performance
To determine the biological process (BP), molecular function (MF), and cellular component (CC) about the putative targets, Gene ontology (GO) and KEGG analysis was performed in the DAVID database (https://david.ncifcrf.gov/).The top 20 terms were sorted to draw a histogram using GraphPad Prism 8. KEGG database (http://www.genome.jp/kegg/)analysis was visualization by ggplot2 database 2.2.0.

Measurement of sleep latency and duration
Each mouse was intraperitoneally injected with a threshold dose of pentobarbital sodium (45 mg/kg) after the final 30 min of administration quercetin or saline.The sleep latency was recorded from the time of injection to the time when the righting reflex of mice disappeared for 1 min, and the sleeping time was recorded from the time when the righting reflex disappeared until recovery.

Determination of IL-6
The activity of IL-6 in mouse serum, which was collected after mice sacrificed the last administration, was determined using an ELISA (09/2019, Shanghai Enzyme-linked Biotechnology Co., Ltd.) in accordance with the manufacturer's instructions.

Detection of γ-GABA levels in the hypothalamus
After the mice were sacrificed, hypothalamus tissues were taken for the detection of γ-GABA production by ELISA (09/2019, Shanghai Enzyme-linked Biotechnology Co., Ltd.).First, hypothalamus tissues were placed in 1.5 mL eppendorf (EP) tubes and added 9 times the volume of Phosphate Buffered Saline (PBS) buffer solution.Then, the EP tubes were transferred to a high-speed homogenizer for sample homogenization at 65 Hz for S. FIGURE 1 -The simplified timeline for control, model and quercetin group.

PCPA-induced insomnia mice model
The insomnia mouse model was established via intraperitoneal injection of PCPA.The PCPA-induced mice were received PCPA in the first and second day and the dose was 300 mg/kg each time.Meanwhile, the control group was intraperitoneally injected with physiological saline.On the third day, the quercetin group mice were orally given different concentrations of quercetin (100, 50, 25 mg/kg), whereas the control and model group mice were orally administered with the same volume of saline for another 7 days.The mice were sacrificed after the final 30 min of administration quercetin or saline (S.

GABARA1 mRNA expression detected by RT-PCR
The total RNA of mouse hypothalamus was extracted using the TRIzol method.RNA integrity was detected by 1% agarose gel electrophoresis, and the total RNA concentration and OD 260 /OD 280 value were measured using a microspectrophotometer.And the satisfactory RNA samples synthesis templates of cDNA.The PCR primers (Table I) were designed on the basis of the NCBI reference sequence database for each gene and using Primer3 software (Rozen, Skaletsky, 2000).β-actin was used as a reference gene, and the expression between groups was compared using the 2 -ΔΔCT method.

DATA AND STATISTICAL ANALYSIS
Statistical analysis was processed with IBM SPSS Statistics 19.0 (SPSS Inc., NY, USA).Data were expressed as the mean ± SD and analyzed using Student's t-test.Differences between groups were considered to be statistically significant if values of P< 0.05.

Active ingredient of SZRD screening
The five herbs in SZRD had 152 active ingredients, including 21 compounds of ZSS, 16 compounds of P, 7 compounds of CR, 15 compounds of AR, and 93 compounds of GRR, after screening with the limited condition: "Oral bioavailability (OB)≥30%" and "Druglikeness (DL)≥0.18"(S.Table I).

Gene ontology (GO) functional enrichment analysis
GO functional enrichment analysis were composed with three aspects of BP, CC and MF, and the top 5 items were selected to generate a graph in GraphPad Prism 8 software.The top 3 items of BP were response to drug, positive regulation of nitric oxide biosynthetic process, and adenylate cyclase-activating adrenergic receptor signaling pathway (Figure 3).

KEGG pathway enrichment analysis
To determine the underlying mechanism of SZRD in treating insomnia, all of the pathways interacting with the candidate targets were extracted from the KEGG pathway database using DAVID.Then, the "target-pathway" network was constructed, which was composed of the top 10 related pathways, including Neuroactive ligandreceptor interaction, Calcium signaling pathway, Fluid shear stress, and atherosclerosis (Figure 4).

The effects of quercetin on insomnia mice model
The results of pentobarbital-induced sleeping showed that the sleep latency and sleep duration of model group mice were obviously increased and decreased compared with those of control group mice, respectively (P < 0.05, 0.01).The sleep latency and sleep duration of Que-H mice were significantly reduced and increased compared with those of model group mice, respectively (P < 0.05, 0.01).Although 50 mg/ kg of quercetin (Que-M) had little significant effect on the sleep latency, the sleep duration was remarkably increased compared with the model group (Figure 5A  and B).

Quercetin inhibited the production of IL-6
According to the results of the PPI network, IL-6 might be the main target of SZRD in treating insomnia.Therefore, the production of IL-6 in the serum of mice with insomnia was determined to validate the assumption.As shown in Figure 6, PCPA treatment obviously induced the secretion of IL-6 in the model group, whereas treatment with 100, 50, and 25 mg/kg of quercetin in insomnia model mice attenuated the secretion of IL-6, indicating that quercetin could inhibit inflammatory cytokine secretion.
FIGURE 6 -The effects of quercetin on the production of IL-6.The concentration of IL-6 in mice serum were measured by Elisa kit.Data are expressed as the mean ± SD of three independent experiments.# P < 0.05, ## P < 0.01 compare with the control group; * P < 0.05 and ** P < 0.01 compare with the model group.

Quercetin increased the levels of γ-GABA and GABRA1 in mice hypothalamus
The KEGG results showed that Neuroactive ligandreceptor interaction was the main pathway in SZRD treatment of insomnia, and it involved 28 genes such as GABRA1.Thus, the expression of GABRA1 mRNA and γ-GABA ligand in the hypothalamus was detected.The results showed that γ-GABA production and relative expression of GABRA1 mRNA were obviously decreased in the model group compared with the control group (P < 0.01, 0.05).As expected, high and middle doses of quercetin significantly improved the γ-GABA and GABRA1 mRNA levels of the insomnia model (P < 0.01, 0.05) (Figure 7A and B).The sleep latency (A) and sleep duration (B).Data are expressed as the mean ± SD of three independent experiments.# P < 0.05, ## P < 0.01 compare with the control group; * P < 0.05 and ** P < 0.01 compare with the model group.

DISCUSSION
In this study, the potential mechanism of SZRD on insomnia was determined on the basis of network pharmacology, and quercetin was found to be the main active ingredient of SZRD, which might regulate IL-6 and the Neuroactive ligand-receptor interaction pathway for the treatment of insomnia.Then, an insomnia model was established to validate the hypothesis according to the results of TCM network pharmacology.
First, the results of network pharmacology indicated that a total of 152 ingredients, including quercetin and kaempferol, could be the potential compounds of SZRD in treating insomnia.Quercetin, a kind of polyphenol, widely exists in nature and is closely related to neuronal injury and neurodegenerative diseases (Costa et al., 2016;Ossola, Kaariainen, Mannisto, 2009).Several studies had evaluated the antinociception, antidepression, and learning and memory of quercetin and exhibited good treatment for those neurological disorder (Bigelman et al., 2011;Kanazawa et al., 2016).It was found that quercetin of Flos albiziae significantly increased the rate of sleep onset, decreased sleeping time, and prolonged sleep latency, which were possibly mediated by the serotonergic system (Ye et al., 2015).Quercetin was also reported to enhance non-rapid eyes movement (REM) sleep and reduce REM sleep, and the GABAergic pathway was found to be partly involved in the latter effect (Kambe et al., 2010).Furthermore, a recent clinically relevant survey indicated that quercetin can be used as complementary treatment for patients with insomnia (Afrasiabian et al., 2019).As the main compound of SZRD, quercetin was presumed to be the active substance that exerts a hypnotic effect.In this study, high dose (100 mg/kg) of quercetin significantly prolonged sleep duration and shortened sleep latency in the insomnia mouse model.Moreover, Qing Cao and Weikai (2009) first separated quercetin from SZRD and identified its structure by spectral analysis.Although the research had been reported that quercetin could be the key active ingredient of SZRD by treating with insomnia, but it also needed to demonstrated quercetin was the main compound of SZRD by chemical analysis in our future research (Zijun et al., 2018).Therefore, quercetin was considered one of the important active ingredients of SZRD in the treatment with insomnia.
The PPI network showed that the degrees of IL-6, NOS3, VEGFA, TNF, CAT, PTGS2, TP53, IL-1β, and ACHE were obviously higher than those of other nodes, and IL-6 played an essential role for its close association with chemical ingredients of SZRD.Accumulating evidence suggested that sleep disturbance is associated with inflammation manifesting in circulating levels of inflammatory markers such as IL-6 (Shearer et al., 2001).IL-6, a pleiotropic cytokine with multiple functions, could be an inexpensive marker to assess sleep loss in humans under various experimental settings (Halota et al. 1990).An animal experiment also found increased plasma IL-6 levels in rats 8 days after chronic total sleep deprivation (Thimgan et al., 2013).In the current study, we also found that the PCPA-induced insomnia mouse model had higher levels of IL-6 compared with the control group, but quercetin could decrease the production of IL-6.The results suggested that quercetin in SZRD improves sleep quality by regulating the secretion of IL-6.
To further explore the potential mechanism of SZRD on insomnia, KEGG pathway enrichment was performed, and the results showed that the Neuroactive ligandreceptor interaction was the most important pathway.Neuroactive ligand-receptor interaction mediated the initiation and progression of insomnia, and it was involved with 28 targets such as GABRA1 in our study (Xu et al., 2020).GABRA1, the receptor of γ-GABA, is the neuronal inhibition in the brain and primary target for benzodiazepines, which are widely used to treat insomnia (Dixon et al., 2015).γ-GABA could bind to GABRA1 causing sedation, muscle relaxion, memory impairment, and anticonvulsant effects (McKernan, Whiting 1996;Mohler, Fritschy, Rudolph, 2002).Enhancing GABRA1 activity inducing the hypnosis was the important treatment for sleep disorders (Krystal et al., 2003).Suanzaoren, Fuling, and Gancao in SZRD were identified as the most frequently used sedative and hypnotic herbs, and their pharmacological actions are related to the stimulation of GABA and GABRA1 (Singh, Zhao, 2017).For example, it was demonstrated that SZRD upregulated the expression of GABRA1 to exert soporific effects (Zhan et al., 2020).Quercetin has also been found to influence the sleep-wake cycle through the activation of GABA(A) receptors (Kambe et al., 2010).On the basis of the results of pharmacology network, GABA and GABRA1 mRNA expressions were detected, which revealed that quercetin increased the levels of GABA and GABRA1 mRNA to improve sleep quality in the insomnia mouse model.

CONCLUSION
The pharmacology network of this study determined the potential targets and novel signaling pathways of SZRD in treating insomnia, which revealed the characteristics of TCM treatment with multi component, multiple target and multiple pathways.The present study also demonstrated that quercetin, the main compound of SZRD, could increase the expression levels of IL-6, GABA, and GABRA1 mRNA to improve sleep quality in the insomnia model.Although this study preliminarily explored the underlying mechanism of SZRD on insomnia, the components and targets of SZRD in the treatment of insomnia should be further investigated.

FIGURE 2 -
FIGURE 2 -The PPI network of SZRD.The larger the node, the deeper the color, represented the greater the degree of the node.

FIGURE 3 -
FIGURE 3 -Gene Ontology (GO) enrichment analysis of common targets of compound and insomnia.The analysis items were included with biological process (BP), cellular component (CC) and molecular function (MF).

FIGURE 4 -
FIGURE 4 -Gene KEGG pathway enrichment.The color of circles represented the Q-value and the size of circles was represented the Count.The vertical axis represents the pathway name; the horizontal axis represents the Gene Ratio.

FIGURE 5 -
FIGURE 5 -The effects of quercetin (100, 50, 25 mg/kg) on the sleep time and sleep duration by sodium pentobarbital (n=6).The sleep latency (A) and sleep duration (B).Data are expressed as the mean ± SD of three independent experiments.# P < 0.05, ## P < 0.01 compare with the control group; * P < 0.05 and ** P < 0.01 compare with the model group.

FIGURE 7 -
FIGURE 7-Quercetin improved the γ-GABA production and relative expression of GABRA1 mRNA.γ-GABA production was detected by Elisa kit (A) and GABRA1 mRNA was measured by RT-PCR.Data are expressed as the mean ± SD of three independent experiments.# P < 0.05, ## P < 0.01 compare with the control group; * P < 0.05 and ** P < 0.01 compare with the model group.