Molecular mechanisms of Tibetan medicine Sug-Mel-sum-thang in treating insomnia disorder: network pharmacology, molecular docking, and surface plasmon resonance

Background: The Tibetan medicine Sug-Mel-sum-thang (SMST) is often used to treat insomnia disorder (ID); however, the underlying mechanism remains unclear. In this study, network pharmacology prediction was conducted to analyze the molecular mechanisms of SMST in treating ID. Methods: The active ingredients of the three herbs used in SMST were obtained from the Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences. Chemistry Database [DB/OL]. The chemical formulae of the active ingredients were downloaded from PubChem in canonical SMILES format. These were then imported into the SwissTargetPrediction platform to identify drug targets. ID targets were obtained from databases such as DrugBank, GeneCards, and DisGeNET. Potential targets of SMST for the treatment of ID were identified using the online Venny mapping platform. The potential target protein-protein interaction was constructed using the STRING platform, and the obtained protein-protein interaction information was imported into Cytoscape 3.7.1 for image optimization and core gene extraction and ClueGO in Cytoscape 3.7.1 was used for the enrichment analysis of potential targets. Cytoscape 3.7.1 was also used to construct the regulatory network of the Tibetan medicine compound target pathway and conduct a topological analysis. Results: The mechanism of action of SMST in the treatment of ID involved 72 compounds, including nerolidol, apigenin, luteolin, and piperine. The key targets identified were COMT, CNR1, AKT1, SLC6A4, TNF, CTNNB1, and CHRNA4. The enrichment analysis obtained 69 KEGG pathways, mainly related to Alzheimer’s disease, cAMP signaling pathway, serotonergic synapse, and other pathways. The core active ingredient molecule docked with the target to a higher degree. The results showed that the active ingredient exhibited good binding activity with the related targets. Some parts with good docking fractions were selected for surface plasmon resonance analysis. Conclusion: Various active components in SMST play a role in the treatment of ID by acting on key targets, such as COMT and CNR1, to regulate multiple signaling pathways.


Background
Insomnia disorder (ID) is a common clinical condition. The prevalence of ID is approximately 10-20%, of which approximately 50% of cases have a chronic course [1]. ID is mainly characterized by dissatisfaction with sleep duration or quality, difficulty in initiating or maintaining sleep, severe anxiety, and daytime dysfunction. Moreover, it frequently co-occurs with other medical or psychiatric conditions, such as pain and depression. Persistent insomnia impairs physical health, leading to reduced quality of life and physical or mental illness. Despite its high incidence, little is known about the etiology, pathology, or physiology of ID. The classification and diagnosis of ID has undergone significant changes over the last decade, and the perception of ID has shifted from being a symptom to a disease [2,3].
The most widely prescribed drugs for ID treatment are benzodiazepines and benzodiazepine receptor agonists, which bind to the γ-aminobutyric acid site of the benzodiazepine receptor and increase the intrinsic activity of the inhibitory neurotransmitter γ-aminobutyric acid to exert a therapeutic effect. However, approximately 40% of patients with chronic ID fail to achieve sustained remission with these treatments, and their safety and efficacy are limited by the tolerance and dependence that develop with long-term use [4,5]. The Tibetan medicine Sug-Mel-sum-thang (SMST) has been used to treat ID for more than 1,400 years, and is considered a classic formula based on the theory of Tibetan medicine. It comprises three herbs commonly used in Tibetan medicine: Amomum kravanh Pierre ex Gagnep. (Bai dou kou), Cuminum cyminum L. (Zi ran), and Piper longum L. (Bi ba). SMST was first recorded in the Four Medical Tantras, one of the most important collections of Tibetan medical texts with a history of 1,400 years of clinical application [6]. Because Mongolian medicine and Tibetan medicine are similar, SMST is used in the clinic as "Sugmuler-3 flavor soup", and its formula differs slightly from that of the Tibetan medicine SMST [7]. Although SMST is frequently used in clinical practice and has a significant curative effect [7,8], the mechanism whereby it ameliorates ID is unclear, thus restricting its development and application.
With the rapid development of bioinformatics, molecular biology, medicinal chemistry, cell biology, and other disciplines, various biological databases have emerged for querying active ingredients in single and compound medicines and predict the targets of drugs and diseases. Predicting the action targets of drugs and diseases and analyzing the mechanism of drug action aid in the rapid development of traditional Chinese medicine [9][10][11]. Bioinformatics can also be used to conduct modern research on Tibetan medicine and analyze its mechanism of action of Tibetan medicine. In this study, the mechanism of action of SMST in the treatment of ID was predicted using network pharmacology and molecular docking technology and a scientific basis for the treatment of ID by SMST was proposed. The development of SMST provides novel insights and can serve as reference for future studies.

Materials and methods
The study methodology is illustrated in Figure 1.

Ingredients and drug targets of SMST
The components and potential targets of SMST were obtained from the Shanghai Institute of Organic Chemistry of the Chinese Academy of Sciences. Chemistry Database [DB/OL]. Collection of A. kravanh, P. longum, and C. cyminum in SMST A total of 72 compounds were obtained from Tibetan medicines. The chemical formula of canonical SMILES was obtained from PubChem (https://pubchem. ncbi. nlm. nih. gov/). The canonical SMILES file was imported into the SwissTargetPrediction platform (http://www.swisstargetprediction.ch/), and its attribute was set to "homosapiens" to predict the targets of the chemical components.

Acquisition of disease target and construction of Venn diagram
Disease targets of insomnia were obtained from DrugBank (https://go. drugbank. com/), GeneCards (https://genecards. weizmann. ac. il/v3/), and DisGeNET (http://www. disgenet. org/). Disease targets from the three databases were combined to remove duplicates. The intersection of the drug and disease targets was determined and the potential target was represented as a Venn diagram.

Protein-protein interaction (PPI) network construction
The potential targets of item "2.2" were imported into the STRING database (https://cn.string-db.org/), and the biological species was set to "Homosapiens". The protein interaction network map was obtained and saved in tab separated values format. The PPI data file was imported into Cytoscape 3.7.1 for image optimization, and network analysis was performed to analyze the network topology parameters of the effective components.

Enrichment analysis
The potential targets of item "2.2" were imported into the plug-in ClueGO in Cytoscape 3.7.1, "Selectspecies" was set to "Homosapiens", and a P value of < 0.05 was set for KEGG pathway enrichment analysis.

Construction of "Tibetan medicine-compound-target-pathway" regulatory network
The potential targets of item "2.2" and their corresponding compounds were combined with the KEGG pathway enrichment analysis results of item "2.4" to construct a "Tibetan medicine-compound-target-pathway" network using Cytoscape 3.7.1 software. Network analysis was performed to analyze the network topology parameters of the active ingredients and determine the core active ingredients.

Molecular docking
The core active ingredients in item "2.5" was compared with the 14 targets of AKT1, SLC6A3, SLC6A4, COMT, TNF, CHRNA4, CTNNB1, GRM5, DRD2, ESR1, CNR1, CYP3A4, GRIN2A, and HTR2A in item "2.3" molecular docking was done. The 3D molecular structure of the active ingredient was downloaded from the PubChem database and the target protein structure was downloaded from the protein data bank database (https://www. rcsb. org/) based on human origin, resolution, protein length, and related reports. The target protein was imported into Maestro software for hydrogen bonding optimization, dehydration, and energy minimization. Subsequently, the processed proteins and small molecules were imported into Maestro software for molecular docking.

Surface plasmon resonance (SPR) analysis
A phosphate-buffered saline solution containing 1% dimethyl sulfoxide as the driving buffer was prepared. The CNR1 and COMT proteins (Proteintech) were coupled to a CM5 chip (Cytiva) using a biomolecular interaction detector. Apigenin and piperine were dissolved in dimethyl sulfoxide and diluted to 8 concentrations (16, 8, 4, 2, 1, 0.5, 0.25, and 0.125 mM). Each solution was diluted 100 times with phosphate-buffered saline to obtain the test solutions. The ligand dissociation was recorded using a biomolecular interaction detector. Affinity and kinetic fitting analyses of the ligands and small molecules were performed using the software provided with the instrument.

Active ingredients and targets of SMST
A total of 72 active compounds from A. kravanh, P. longum, and C. cyminum were obtained from the Chemical Professional Database of the Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences. Among the 72 active ingredients, 25 were from A. kravanh, 14 were from P. longum, 33 were from C. cyminum, and seven were common components of C. cyminum and A. kravanh, as shown in Table  1. Drug targets were obtained using the SwissTargetPrediction platform. A total of 207 targets were obtained for A. kravanh, 445 for P. longum, and 257 for C. cyminum. Target 623 was obtained by combining the above targets and removing duplicates.

Potential targets of SMST in the treatment of ID
The ID targets were screened using the DrugBank, Gencards, and DisGeNET databases. After de-duplication, 813 target genes were identified. Using the Venny mapping platform, the active ingredients of SMST were intersected with ID-related targets, and 107 potential targets were obtained, as shown in Figure 2.

Construction of PPI network of potential targets of SMST in the treatment of ID
The potential targets of SMST treatment ID were uploaded to the STRING database to obtain the PPI network and related information, and the PPI information was imported into Cytoscape 3.7.1 to obtain the PPI network diagram (Figure 3). Network analysis was performed to analyze the network topology parameters of the active ingredients. We found that the AKT1, SLC6A3, SLC6A4, COMT, TNF, CHRNA4, CTNNB1, GRM5, DRD2, ESR1, CNR1, CYP3A4, GRIN2A, and HTR2A nodes had high degree values. These targets are important in the network and are core targets for SMST in ID therapy.

Enrichment analysis
KEGG pathway enrichment analysis was performed on the potential targets of SMST in the treatment of ID using the ClueGO plug-in in Cytoscape. The results showed that 69 pathways may be associated with ID, including Alzheimer's disease, cAMP signaling pathway, serotonergic synapse, calcium signaling pathway, and dopaminergic synapse, as shown in Figure 4. Cytoscape 3.7.1 was used to construct a   network diagram of the "Tibetan medicine-compound-target-pathway". The graph nodes represent active ingredients, squares represent targets, inverted triangles represent compounds, quadrilaterals represent pathways, and hexagons are traditional Chinese medicines. The area and color transparency of the node represents the degree values. The larger the area and darker the color, the more important the node, as shown in Figure 5.

Molecular docking
Molecular docking was conducted to explore the binding activity between the target protein and its corresponding components. The active ingredients with node degrees greater than twice the median in the "Tibetan medicine-compound-target-pathway" network diagram constructed herein were screened as the core ingredients of SMST for the treatment of ID (Table 2). Node degree values greater than twice the median in the PPI network were selected as the core targets, which could be considered the core targets of ID treatment. These 13 genes were selected as receptors, as shown in Table 3.
It is generally believed that the smaller the docking score, the lower the binding energy to the receptor and the stronger the potential activity of the ligand. The results obtained by molecular docking showed that Van der Waals forces, Pi-alkyl, H-bond, Pi-Pi stacking, and other forces were present, indicating that the core active components and the target have good binding activity. The molecular docking results are shown in Figure 6, and part of the molecular docking is shown in Figure 7.

Discussion
Sleep is an essential physiological process in humans, as it is necessary for individual survival and has several functions, such as improving memory, promoting growth and development, delaying aging, and enhancing immunity. Research shows that approximately one-third of adults have ID, which is currently treated with benzodiazepines and other sleep-helping drugs (including anti-anxiety drugs, antidepressants, melatonin steroids, and hormone drugs). However, these drugs generally have disadvantages such as strong drug dependence, easy addiction, and easy development of drug resistance [12].
Traditional Tibetan medicine has unique sleep-improving properties and has accumulated a wealth of clinical prescriptions. SMST is the first choice of Tibetan medicine in ID treatment. In Tibetan medicine, ID is considered a disorder of "rLung" in the "three humors", and the combination of sweet and pungent herbs and "hot" and "greasy" effects will have a good therapeutic effect. SMST is a classic Tibetan medicinal formula composed of three herbs: A. kravanh, P. longum, and C. cyminum. A. kravanh has an acrid taste, whereas P. longum and C. cyminum are sweet and acrid. These three herbs have the effect of being "hot" and "greasy", and can thus inhibit the "rLung" disease and soothe the nerves, thus aiding in sleep.
Nine core components in SMST were obtained by combining the PPI network and the "Tibetan medicine-compound-target-pathway" network constructed in this study and screening the active ingredients with a node degree greater than twice the median. Apigenin has been used to improve the symptoms of chronic ID and has shown preliminary efficacy [13,14]. Both insomnia and depression are central nervous system diseases. The blood-brain barrier (BBB) facilitates the supply of nutrients to the brain, regulates the brain microenvironment, and influences the biological activities of various compounds [15]. The BBB also protects the central nervous system from compounds that negatively affect its function. Studies have shown that flavonoids can easily penetrate the BBB in the following order: genistein > isoliquiritigenin > apigenin > puerarin > kaempferol > hesperidin > rutin > quercetin [16,17]. Therefore, apigenin may cross the BBB and play a role in ID treatment [13].
The KEGG pathway enrichment results showed that the treatment of ID by SMST mainly involved the Alzheimer's disease, cAMP signaling, serotonergic synapse, and calcium signaling pathways. Alzheimer's disease, the most common type of dementia, is a neurodegenerative disorder characterized by memory deficits, particularly forgetting recent information, impaired recall, loss of time tracking, problem-solving, language, and difficulty in identification. Research suggests that ID is involved in the pathogenesis of Alzheimer's disease and may affect its symptoms and development [18][19][20][21][22][23][24].
Molecular docking results showed that the core ingredients, piperine, apigenin, and piperlonguminine, had good docking activity with the target proteins COMT, CNR1, and SLC6A4. BIAcore technology can monitor the entire reaction process of bonded substances on the surface of metal films by tracking and detecting the SPR angle of incident light in the medium in real-time. For example, the entire process of the binding and dissociation of biomolecules can be detected in real-time. Through the BIAcore experiment, it was verified that CNR1 combined well with piperine and apigenin, and COMT combined well with piperine [25][26][27][28]. Therefore, it can be concluded that the active ingredients of SMST combine well with the target protein and can play a role in ID treatment. No animal experiments were performed in this study. The sedative and hypnotic effects of the predicted core compound on mice were observed, and core compounds with good effects were selected to further explore the mechanism of action of ID therapeutics and provide support for clinical application.

Conclusion
In this study, a comprehensive analytical method combining active ingredient screening, target prediction, molecular docking, and SPR technology was used for the first time to identify the main active ingredients in SMST. Additionally, the possible molecular mechanisms underlying ID treatment were elucidated. This study provides theoretical and experimental bases for the development and application of SMST in ID therapy.