Integrative Network Pharmacology and Glucose Uptake Studies Identify Potential Phytochemicals From Syzygium cumini as Promising Antidiabetic Agents

ABSTRACT

Diabetes mellitus (DM) is a multifaceted metabolic illness with significant detrimental effects during the conventional treatment approach. Therefore, this study aims to explore novel therapeutic targets and potential phytoconstituents using an edge‐weighted network pharmacology approach. The bioactive compounds from Syzygium cumini were selected based on the literature evidence. Potential gene targets corresponding to these phytoconstituents were identified using gene‐disease association, enrichment calculations, and KEGG pathway mapping analysis. These results reveal that PTPN1 is closely associated with the disease progression of DM. Binding interactions between phytochemicals and the candidate gene (PTPN1) were determined using molecular docking combined with molecular dynamics simulation and MM/PBSA calculations. Notably, chrysanthemin derived from S. cumini exhibited substantial inhibitory action in glucose uptake assays using L6 cell lines, PTP1B inhibition, and enzyme kinetics analyses. These findings suggest that this compound might be a promising candidate for PTPN1 inhibition, warranting further extensive research for its potential therapeutic application in DM.

1. Introduction

One of the most common metabolic diseases in the world is diabetes mellitus (DM). It is mainly characterized by insulin resistance, decreased insulin production, or a mix of both. Therefore, a multi‐targeted therapy approach is necessary for the optimal management of DM, a complicated, multifaceted disease that affects several organs. From 108 million cases in 1980 to an estimated 764 million cases by 2030, the worldwide burden of diabetes has increased at an alarming rate in recent decades. With 78 million cases in 2015 alone, Southeast Asia has emerged as a key hotspot for the disease, with India contributing about 70 million cases at that time [1, 2].

Type II diabetes mellitus (T2DM) is usually treated by using artificial drugs such as sulfonylureas and repaglinide, which stimulate insulin discharge; troglitazone and metformin, which help in enhancing insulin function in adipose tissue and liver; and α‐glucosidase inhibitors like acarbose and miglitol, that delay carbohydrate absorption [3]. While these pharmacological interventions help in glycemic control, their prolonged use could lead to common side effects such as hypoglycemia, gastrointestinal discomfort, hepatotoxicity, and weight gain.

Furthermore, combination treatments, like insulin administered along with metformin or sulfonylureas, though they help in lessening insulin dependence, can also cause problems associated with patient adherence and tolerability [4, 5, 6].

These limitations associated with conventional therapeutic strategies have intensified the search for more reliable, more effective alternatives, leading to increased attention in natural products and plant‐derived compounds. Traditional herbal compounds have served as a pillar in global healthcare for centuries. About 80% of the world's population still depends on natural remedies for the treatment of chronic ailments such as diabetes [7, 8, 9]. The increasing concerns about the pharmacological potential of medicinal plants have directed researchers to identify novel bioactive constituents that could act as effective alternative or complementary therapeutic agents [10, 11, 12].

Syzygium cumini (L.) Skeels, commonly known as Java plum or Indian blackberry, is one of the promising options used for the treatment of diabetes [13]. It has long been utilized in traditional medicine for its antidiabetic qualities. Phytochemical studies have revealed that S. cumini is abundant in flavonoids, anthocyanins, tannins, and other medicinally active compounds [14, 15]. These compounds, especially flavonoids, are proven to exert antidiabetic activity by modulating key enzymes involved in glucose metabolism and insulin sensitivity.

Although literature has revealed the antidiabetic properties of S. cumini, a thorough evaluation of its bioactive components and their exact mechanisms of action is still unexplored. To study the effect of phytochemicals and their target proteins in disease networks, network pharmacology, an integrative method that blends computer modeling with experimental validation, is now used as a potential tool.

The current study attempts to explore the molecular mechanisms behind the influence of phytoconstituents from S. cumini on the treatment of diabetes. Advanced computational approaches combined with free energy calculations help to discover the potential impact of phytoconstituents from S. cumini. Furthermore, glucose uptake assay and enzyme inhibition assays were performed to confirm the efficiency of the compounds in enhancing glucose uptake and improving insulin responsiveness.

2. Results and Discussion

2.1. Target Identification

This study aims to highlight S. cumini as a potent natural source for developing alternative therapeutic agents for T2DM. Initially, a total of 44 phytoconstituents were identified and retrieved from the selected sour plant S. cumini. All these phytoconstituents were taken for target mapping analysis. Two hundred two gene targets resulted from the screening of the chemicals from the selected plant source. For instance, 160 targets from the BindingDB database for 29 phytoconstituents, 38 from CHEMBL for 13 phytoconstituents, and 4 from Therapeutic Target Database (TTD) for 2 phytoconstituents (Table 1). Among the targets retrieved, 114 target genes were filtered as diabetes associated genes when searched in disease databases such as DisGeNET with a GDA score ranging from 0.01 to 0.7, corresponding to 44 phytoconstituents [16]. Consolidating the overlapped and unique genes led to the recovery of our 114 target genes.

TABLE 1.

List of protein targets identified based on structural similarity across databases.

Target databases	Number of targets
Binding database	160
CHEMBL	38
TTD	4
Total number of targets retrieved	202

It is evident from the table that S. cumini possessed more phytoconstituents corresponding to the target with GDA ≥ 0.5 for T2DM. Phytoconstituents such as chrysanthemin, delphinidin‐3‐gentiobioside, and rutin are associated with the target genes having the GDA score of ≥ 0.5. The complete details of targets that showed gene‐disease association (GDA), along with their corresponding phytoconstituents for T2DM, are listed in Table S1. Although screening was carried out using the GDA score, the potency of target genes corresponding to 44 phytoconstituents with even a modest GDA score was taken into consideration for further investigation, rather than depending solely on a single cut‐off filter.

2.2. Network Pharmacology Analysis

A pharmacological interaction network comprising 44 phytoconstituents and 114 target genes was constructed, resulting in 178 nodes (proteins) and 307 edges (interactions), which was used for network parameter analysis. The visual representation of this network is shown in Figure 1, and detailed network information is provided in Table S2. Figure S1A displays the scatter plot of node degree distribution, where the node degree refers to the number of connections a node has within the network. Typically, nodes with higher degrees exert greater influence over the network [17]. Based on an average node degree threshold of ≥ 10, 10 phytoconstituents met the screening criteria. Figure S1B illustrates the scatter plot of neighbor numbers relative to closeness centrality. Closeness centrality measures how central a node is within the network based on its distance to other nodes—nodes with higher closeness centrality are, on average, closer to all others. In this analysis, approximately 50% of the nodes exhibited the shortest distance to all other nodes. Figure S1C indicates that the network's average shortest path length falls between 3 and 4, suggesting small‐world properties. In such networks, moving from one node to another typically requires only a few steps [18]. Figure S1D presents the betweenness centrality distribution, highlighting that the majority of nodes are connected to up to 20 neighbors. While various topological parameters were assessed to characterize the network, ligand selection for further analysis primarily considered nodes with a degree ≥ 10. In addition, ligands were filtered based on GDA scores. Only those phytoconstituents that satisfied both criteria, node degree ≥ 10 and GDA score ≥ 0.5, were shortlisted, as detailed in Table 2.

FIGURE 1.

Phytoconstituent–Target Gene–GDA network constructed for visualizing the interaction of target genes with corresponding phytoconstituents, which exhibited GDA score for Type II diabetes.

TABLE 2.

Selected phytoconstituents based on two screening filters (degree ≥ 10 and GDA ≥ 0.5).

Phytoconstituents	Degree	GDA ≥ 0.5	Target exhibited GDA ≥ 0.5
Rutin	12	0.5	PTPN1
Chyrysanthemin	11	0.5	PTPN1
Delphinidin‐3‐gentiobioside	10	0.5	PTPN1

2.3. Gene Enrichment Calculations

To validate the potency of 114 targets identified in the DisGeNET database, gene enrichment and KEGG analysis were performed to interpret the pharmacological mechanistic involvement in Type II diabetes. The GenBank IDs were used as an input to study the role of 114 genes in Homo sapiens during enrichment analysis. Biological pathways connected to diabetes were carefully selected and filtered with a significant p‐value of less than 0.05 by using the “DAVID” database [19]. On visualizing the output in a heat map by Morpheus [20], PTPN1, INSR, and NOS3 were identified as the most enriched genes among the 114 genes related to T2D, which has been depicted in Figure S2. The peer interpretation of protein interaction under KEGG pathway analysis was performed using KEGG mapper [21], which highlighted PTPN1, GSK‐3β, and INSR as the highest potent genes involved in Figure S3. PTPN1 was selected as a potent target based on the results obtained from enrichment calculation and KEGG pathway mapping analysis.

2.4. ADME/T and Molecular Docking Analysis

ADME/T study is a crucial step in identifying and forecasting potential side effects of substances before their delivery to patients. The molecular pharmacokinetic properties and drug likeness were projected in Table S3. The results shown that chrysanthemin had the lowest Lipinski's violation among the three ligands. Based on the molecular weight chrysanthemin only passed the violation rule and had better skin permeability. On analyzing hydrogen bond donors and acceptors of three ligands, chrysanthemin had better binding stability with the target receptor.

Molecular docking calculations were performed to understand the interaction behaviors of the plant phytochemicals with the identified target protein PTPN1. Initially, the docking protocol was validated by re‐docking the co‐crystallized ligand into the active site of the target protein, and the resulting pose was compared with the experimental conformation (Figure S4). The low root mean square deviation (RMSD) value (1.92 Å) confirmed the reliability and accuracy of the docking procedure used in this study. Later, the binding affinity calculations were performed for the top‐ranked compounds. The binding affinity values of the compounds and their interacting residues were mentioned in Table 3. The interaction images were presented in Figure 2. It is to note that osunprotafib standard inhibitor of PTPN1 was taken as the reference compound for this analysis.

TABLE 3.

Binding affinity and interacting amino acid residues of potential phytoconstituents.

S. no.	Compound name	Hydrogen bond interactions	Docking score (kcal/mol)
1.	Osunprotafib (reference)	LYS116, ARG221 (2), GLN266, GLY183 (2), THR263, VAL49, TYR46	−7.6
2.	Rutin	PHE196, LYS292(2), ILE281	−8.2
3.	Chrysanthemin	ASP236, PHE196 (2), ARG199, ASN193, LYS292, ARG79, LEU204	−8.0
4.	Delphinidin‐3‐gentiobioside	GLY183, ASP191, ARG221, CYS215, ALA217, LYS116, TYR46, VAL49, LYS120, GLN262, GLN266, GLY183	−7.9

FIGURE 2.

Ligand interaction diagram of the top‐ranked phytoconstituents along with the reference compound. The top images represent the interactions in 2D and the bottom image represents the 3D binding position. (a) Osunprotafib; (b) rutin; (c) chrysanthemin; (d) delphinidin‐3‐gentiobioside.

It is to observe that all the three ranked compounds exhibited better binding affinity value than the reference compound. Notably, rutin has displayed an affinity value of −8.2 kcal/mol. Followed by chrysanthemin and delphinidin‐3‐gentiobioside, which showed binding energy values of −8.0 and −7.9 kcal/mol with PTPN1. Although rutin has resulted in a higher binding affinity value, it has displayed less number of hydrogen bond interactions than the other two compounds, as well as the reference compound. Osunprotafib and delphinidin‐3‐gentiobioside exhibited seven conventional hydrogen bond interactions with the reference compound. On the other hand, chrysanthemin exhibited five hydrogen bonds, four pi–alkyl, and one pi–pi interactions with the protein binding site residues [22]. All three compounds exhibited considerable binding affinity and interactions with binding site residues, rendering them appropriate for further computations.

2.5. Molecular Dynamics Simulation Studies

Structural stability of the filtered ligands was analyzed during molecular dynamics (MD) simulation studies using 100 ns trajectories. RMSD analysis was performed on the carbon alpha atoms of the protein to analyze the structural integrity of the complexes. Figure 3A represents the RMSD graph of all the systems investigated in this study. The black color graph represents the protein without a ligand in all the MD simulation analyses. It is worth noting that the three compounds showed an initial deviation of up to 1 nm. Although the compound chrysanthemin displayed higher deviation (up to ∼2 nm) at the initial stage of simulation, the compound attained its equilibration state after 40 ns of simulation time. It is worth mentioning that the compounds rutin and delphinidin‐3‐gentiobioside showed less deviation than the native protein. This represents that these compounds are perfectly packed into the active site portion and have tight structural stability.

FIGURE 3.

(A) Structural integrity of the compounds and native protein calculated using root mean square deviation. (B) Fluctuation of amino acids during the binding event measured using root mean square fluctuation. (C) Structural compactness of the protein structure analyzed using radius of gyration (color code: black: chrysanthemin; red: delphinidin‐3‐gentiobioside; blue: rutin; magenta: protein).

During the ligand binding event, we have monitored the amino acid fluctuation using root mean square fluctuations (RMSFs). The results were presented in Figure 3B. The fluctuation of all the compounds aligns with the native protein and it displayed less than 2 nm fluctuations. Especially the compounds displayed stable fluctuations around the binding site locations.

Furthermore, structural compactness of the complexes were investigated through radius of gyration (RoG) analysis (Figure 3C). It measures the protein folding during the complex binding in a dynamic environment. It is to observe that in and delphinidin‐3‐gentiobioside has better compactness than the other compound, since it shown less than 3 nm deviation. In particular, the compound rutin has much more compactness with binding site residues as it shows stable equilibration from the initial simulation period. Pairwise distance analysis was executed to understand the molecular interaction, conformational changes and molecular structural comparison. The complex which shows lowest distance has higher chance of interacting with target site. The binding conformations of the compounds obtained during various simulation time frames are depicted in the Figure 4. It is to observe that no much deviations were observed for delphinidin‐3‐gentiobioside and rutin complexes. Although a slight deviation was observed during the 20 ns time frame of chrysanthemin, it maintained its binding position throughout the simulation time. Chrysanthemin‐protein complex showed better results throughout the simulation in the mean value of 0.1689 nm, as shown in Figure S5A. In addition, hydrogen bond interaction was also examined to evaluate the interacting potential of the compounds. The hydrogen bond results were shown in Figure S5B. It is worth noting that the compound displayed a maximum of eight hydrogen bond interactions. The other two compounds resulted in a maximum of five hydrogen bond interactions. It is worth mentioning that all the compounds displayed consistent hydrogen bond interactions during the MD simulation time. Overall, the MD simulation studies reveal that all three compounds displayed strong structural stability with the target protein PTPN1.

FIGURE 4.

Binding conformation of ligand molecules in the binding pockets of target protein with 20 ns time interval. The drug molecules are depicted in different colors inside the binding pocket based on the time frame: red—0th ns, green—20th ns, blue—40th ns, magenta—60th ns, cyan—80th ns, orange—100th ns.

2.6. Molecular Mechanics Poisson–Boltzmann Surface Area Calculations

Binding free energy estimates carried out using molecular mechanics Poisson–Boltzmann surface area (MM/PBSA) calculations. It is important to note that these free energy estimates are directly proportional to the experimentally determined binding affinity values [23]. The binding energy values and individual contributions of various energy terms were supplied in Table 4. In all the complexes major contribution was supplied by Van der Waals interaction energy. It is worth noting that all the compounds have resulted in a binding free energy value in the range of −130 to −140 kJ/mol. Among the investigated compounds chrysanthemin has better binding stability with the target protein, since it shown higher binding free energy value of −140.369 kJ/mol. Collectively, chrysanthemin has a better ability to bind with the binding site of PTPN1 when compared to the other investigated compounds.

TABLE 4.

MM/PBSA binding free energy analysis of protein–ligand complex.

Compounds	Van der Waal energy (kJ/mol)	Electrostatic energy (kJ/mol)	Polar solvation energy (kJ/mol)	SASA energy (kJ/mol)	Binding energy (kJ/mol)
Chrysanthemin	−199.659	−13.308	91.897	−19.298	−140.369
Delphinidin‐3‐gentiobioside	−247.940	−37.273	178.150	−25.419	−132.481
Rutin	−217.412	−77.129	180.200	−21.352	−135.694

2.7. Protein Tyrosine Phosphatase 1B Inhibitory Activity

An in vitro antidiabetic study was conducted to evaluate the dose‐dependent protein tyrosine phosphatase 1B (PTP1B) inhibitory activity of the plant‐derived compound, which is also essential to validate the computational findings [24]. With increasing concentrations, the chrysanthemin gradually inhibited PTP1B activity, as seen in Figure 5A. Chrysanthemin exhibited 5.55% inhibition at the lowest tested dose of 6.25 µg/mL. With an increase in concentration, the inhibition effect increased steadily, with a notable increase in activity, with 37.71% inhibition; however, at the highest dosage of 100 µg/mL, there was 55.53% inhibition. The compound's IC₅₀ value was determined to be 83.46 ± 0.04 µg/mL. This suggests that the molecule may have a significant effect in controlling insulin signaling pathways in addition to its strength in blocking PTP1B [25]. PTP1B strongly influences insulin signaling by dephosphorylating phosphoinositide 3‐kinase (PI3K) and insulin receptor substrates (IRS1), and its overexpression is closely linked to metabolic diseases associated with obesity and insulin resistance. As a promising therapeutic candidate for the treatment of Type II diabetes, the compound's effective inhibition of PTP1B raises the possibility of improving insulin sensitivity and encouraging glucose absorption [26, 27].

FIGURE 5.

(A) Percentage inhibition of PTP1B activity at varying concentrations (µg/mL) of the tested compound chrysanthemin. (B) Lineweaver–Burk plot representing the double reciprocal of enzyme kinetics of the compound chrysanthemin.

The results demonstrate the plant‐derived chrysanthemin's strong inhibitory action and its potential as a treatment for insulin resistance. Since PTP1B plays a crucial part in metabolic dysregulation, blocking it presents a useful approach to the creation of antidiabetic medications.

The kinetic parameters of enzyme inhibition by chrysanthemin were evaluated using the Lineweaver–Burk plot (Figure 5B). The calculated slope (m) was 1.096, and the y‐intercept (b) was 0.006, from which the maximum reaction velocity (V _max) was determined to be 166.67, and the Michaelis–Menten constant (K _m) was calculated as 182.67. These values offer important insights into the mode of inhibition exhibited by chrysanthemin, indicative of a competitive inhibition mechanism. In this form of inhibition, the compound competes with the substrate for binding at the enzyme's active site, thereby increasing the K _m without affecting the V _max [28].

The elevated K _m value suggests a reduced affinity between the enzyme and its substrate in the presence of chrysanthemin, necessitating higher substrate concentrations to achieve half‐maximal enzymatic activity. The unchanged V _max supports the concept that, despite competition, the enzyme retains its catalytic efficiency at saturating substrate levels. Furthermore, the inhibition profile showed a concentration‐dependent trend, with the highest inhibition (55.53%) recorded at 100 µg/mL. This reinforces the evidence that chrysanthemin directly interferes with substrate binding by occupying the active site [29].

These results are consistent with classical competitive inhibition, characterized by an increased K _m and unchanged V _max, suggesting that chrysanthemin may serve as an effective natural modulator of enzyme activity. Such a mechanism is particularly valuable in therapeutic contexts, where fine‐tuning rather than complete inhibition of enzyme function is desirable [30].

2.8. Glucose Uptake Measurements in L6 Myotubes

To investigate the effect of chrysanthemin on glucose transport, L6 myotubes were treated with different concentrations of the extract (25, 50, and 100 µg/mL), and glucose uptake was measured at a single time point of 24 h (Figure 6). A clear, concentration‐dependent increase in glucose uptake was observed upon treatment with chrysanthemin, indicating its potential to enhance glucose absorption in muscle cells. The maximum glucose uptake activity was achieved at the highest concentration of 100 µg/mL, showing a 71.7% glucose uptake. The results demonstrated a significant glucose uptake higher than that of the control group. Notably, at all concentrations tested, chrysanthemin exhibited approximately 75% of the glucose uptake activity of insulin, suggesting an insulin analog‐like function.

FIGURE 6.

Glucose uptake patterns of chrysanthemin at different concentrations (25, 50, and 100 µg/mL) compared to insulin. L6 myotubes were incubated with the extract and insulin, followed by a glucose uptake assay.

Chrysanthemin's ability to increase glucose uptake highlights the possibility that it plays a role in insulin‐mediated glucose transport processes. Through the translocation of the GLUT4 transporter to the plasma membrane, skeletal muscle tissue promotes glucose absorption, which is essential for preserving glucose homeostasis [31]. Because of their strong reactivity to insulin and glucose uptake assays, L6 myotubes are used as a model system to investigate insulin resistance and glucose uptake. The results of this investigation demonstrate that chrysanthemin increases glucose absorption in a concentration‐dependent fashion, indicating a noteworthy influence on the insulin signaling system. Chrysanthemin demonstrated its potential as an efficient regulator of glucose metabolism by achieving a maximum glucose absorption of 71.7% at a concentration of 100 µg/mL. Notably, previous research has shown that chrysanthemin exhibits similar glucose absorption activity to other natural substances such quercetin, berberine, and rutin [32, 33]. These findings further validate chrysanthemin's potential as an antidiabetic drug and offer compelling justification for more research to elucidate its underlying mode of action and evaluate its therapeutic effectiveness in addressing insulin resistance and hyperglycemia.

3. Conclusions

These results demonstrate S. cumini's medicinal potential, with its flavonoid‐rich profile exhibiting great promise for targeting PTPN1 in the treatment of Type II diabetes. The potential of important phytoconstituents to modify insulin signaling pathways is highlighted by their noteworthy interactions with PTPN1, including rutin, delphinidin‐3‐gentiobioside, and chrysanthemin. The strong binding affinity of chrysanthemin, along with the stability of its complex with PTP1B observed in molecular docking and dynamics studies, highlights its potential as a promising candidate for further investigation. The ability of chrysanthemin's to improve the cellular glucose absorption was validated using glucose uptake studies in L6 cell lines. In addition, enzyme kinetics and PTP1B inhibition assays demonstrate that the compound can effectively inhibit negative regulators of insulin signaling. Together, these tests confirm chrysanthemin's dual mode of action and strengthen its ability to enhance insulin sensitivity and glucose metabolism. The study has certain limitations, including the moderate potency of the candidate compound compared to standard inhibitors and the absence of in vivo validation. Furthermore, studies will be focused on the in vivo evaluation and structural optimization to enhance the potency against PTP1B. In addition, we have proposed to include mechanistic studies to understand the therapeutic potential in metabolic disorders. This work lays the foundation for future preclinical and clinical studies on PTPN1‐targeted small‐molecule therapies with improved potency and selectivity, offering a potentially safe and effective alternative for diabetes treatment.

4. Experimental Section

4.1. Dataset Preparation and Preprocessing

4.1.1. Retrieval of Phytoconstituents

The phytoconstituents were extracted from published scientific literature using appropriate keywords such as “Syzygium cumini phytochemicals,” “Syzygium cumini bioactive compounds,” “Syzygium cumini secondary metabolites,” and “Syzygium cumini chemical constituents” [34, 35, 36, 37]. Only distinct and relevant bioactive molecules were taken into consideration for additional investigation. Superfluous phytochemicals were eliminated from the dataset to prevent false positive outcomes. Later, the three‐dimensional structures of the phytochemicals were retrieved from the PubChem database. To proceed further for computational investigations, such as docking, MD simulations, and ADMET profiling, the structures were downloaded in Structure Data File (SDF) format [38].

4.1.2. Networking Databases

Three well‐known databases, the BindingDB [39], ChEMBL [40], and the TTD [41], were used to find possible molecular targets for the phytoconstituents that were recovered. To identify target proteins, the BindingDB was first searched to find compounds that were structurally related to or identical to the phytochemicals that were obtained. A similarity search is conducted in this database, and scores are assigned according to structural resemblance; a score of 1 denotes 100% similarity.

Data on the bioactivity of small compounds with drug‐like characteristics can be found in the manually curated ChEMBL library. It gives phytochemicals distinct ChEMBL IDs and links them to putative target proteins. The search was expanded to TTD, which connects chemical compounds to recognized therapeutic targets based on experimental and clinical validation, if neither BindingDB nor ChEMBL could yield any targets.

The DisGeNET platform was used to identify genes directly associated with T2DM to further refine the selection. The condition ID C0011860 was used in this investigation to obtain genes relevant to the condition. The GDA score, which goes from 0 to 1, was used to quantify the connections; higher scores suggest stronger evidence. To guarantee the inclusion of highly significant and firmly associated genes connected to Type II diabetes, a cut‐off criterion of ≥ 0.5 was used [42].

4.1.3. Edge Weighted Network Analysis

The Cytoscape (v3.9.0) platform was used to obtain a network to visualize the relationship between selected genes and targets in a preferred layout, which was used to prioritize the target proteins. Phytoconstituents and targets were given as key nodes, and the respective GDA scores were fed as weight parameters to construct the phytoconstituent‐target network [43]. Along with general network properties, node degree, betweenness and closeness centralities were analyzed and node degree parameter was taken to filter the potential candidates from the whole network setting, with a filter limit cut‐off of node degree ≥ 10 [44].

4.1.4. Gene Enrichment and Screening of Target Genes

DAVID database and KEGG mapper were used for gene enrichment analysis and to find the importance of collected target genes in pathways related to Type II diabetes [45]. GenBank ID of resulted genes were given as input to investigate the role of genes in H. sapiens in enrichment analysis and biological pathways related to diabetes were carefully chosen and are further filtered with a significant p ≤ 0.05. The interactive heatmap tool—Morpheus is employed to visualize the result of gene enrichment and expression level as diabetes are compared [46]. Finally, potent target gene and ligand candidates were shortlisted from those selected candidates that crossed the cut‐off level of both filters, GDA ≥ 0.5 and node degree ≥ 10.

4.1.5. Preliminary Screening of Database Compounds

The canonical form of phytoconstituents that qualified the screening filters of node degree ≥ 10 and GDA ≥ 0.5 ligands were screened and evaluated for pharmacokinetic and pharmacodynamics properties. We have utilized Swiss ADME to predict the properties of the compounds [47]. In particular, Lipinski's rule of five was examined for verifying the adverse effect of phytoconstituents.

4.1.6. Molecular Docking Analysis

PTPN1 was screened as a potent target from preliminary studies. The x‐ray crystallographic structure of PTPN1 catalytic domain (PDB ID: 1T48) was retrieved from Protein Data Bank [48, 49]. AutoDock Vina in the PyRx virtual screening tool was used to perform the molecular docking calculations. Initially, the protein molecules underwent several preprocessing steps, including the removal of water molecules, addition of missing hydrogen atoms, and assignment of appropriate charges. The structures were then converted into PDBQT format for molecular docking calculations and loaded into the PyRx virtual screening tool. Similarly, the phytoconstituents were prepared using standard protocols such as bond order corrections, optimization, and assigning torsions. The X, Y, Z coordinate center of the grid box, such as 51.0782, 30.5373, and 23.5171, with box sizes of 30.7615, 30.6323, and 27.9528, respectively. The top 20 docked complexes based on binding energy were downloaded to study the optimized complex. All the complex files were then visualized in BIOVIA Discovery Studio Visualizer to obtain interaction images and binding poses [50, 51, 52].

4.1.7. MD Simulation and MM/PBSA Calculations

The GROMACS 2018 package was used to perform MD simulation of the top‐ranked compounds with the target protein. Protein and ligand topologies were generated using g_pdb2gmx and the CGENFF tool [53]. A dodecahedron‐shaped simulation box with 1 Å and solvated using the TIP3 water model. Sodium and Chlorine ions were added in place of water atoms to neutralize the system to zero charge. Furthermore, the steepest descent energy minimization algorithm was implemented up to a maximum of 1000 steps. We have also carried out two ensemble algorithms, such as NVT and NPT, to maintain the system equilibration under the temperature of 300 K and pressure of 1 bar, respectively. The final production run was subjected to a 100 ns simulation period. The trajectories of all the MD systems were retrieved as an .xtc file and used for essential dynamic analysis.

The MM/PBSA approach was used to assess the strength of the binding forces of protein–drug interactions in terms of free energy. Stable MD trajectories were selected and 100 snapshots with 1 ns time intervals were selected for MM/PBSA calculations [54]. The calculated average free energy (ΔG _bind) of each complex was calculated using the following equation:

$$\mathrm{\Delta }{G}_{\mathrm{bind}}=\mathrm{\Delta }\mathrm{EMM}+\mathrm{\Delta }\mathrm{GBPSA}-\mathrm{T}\mathrm{\Delta }\mathrm{SMM}$$

where ΔG _bind is the calculated average free energy, ΔEMM is the average molecular mechanics' energy, ΔGPBSA represents the solvation free energy, and TΔSMM denotes solute configuration entropy.

4.1.8. PTP1B Inhibition Studies and Enzyme Kinetics

PTP1B inhibition studies were performed to evaluate the inhibitory potential of chrysanthemin on this enzyme, a negative regulator of insulin signaling [55]. The assay was done using a colorimetric method based on the detection of free phosphate released from a synthetic substrate. The PTP1B inhibition assay was done by using the Colorimetric Assay Kit from BPS Bioscience, USA.

Briefly, the PTP1B enzyme was incubated with different concentrations of the chrysanthemin (6.25, 12.5, 25, 50, and 100 µg/mL) in reaction buffer with 50 mM Tris‐HCl (pH 7.4), 0.1 mM EDTA, and 2 mM dithiothreitol (DTT). The mixture was pre‐incubated at 37°C for 10 min before the addition of the synthetic phosphopeptide substrate. The reaction was allowed to proceed for 30 min at 37°C, after which it was stopped by adding a stop solution.

The amount of free phosphate released during the reaction was evaluated using the malachite green reagent, which binds with phosphate to form a colored complex. The absorbance of this complex was studied at 415 nm using a microplate reader [56]. The inhibition of PTP1B activity was expressed as a percentage and calculated using the following formula:

$$\begin{array}{ccc}& & \mathrm{Percentage}\mathrm{inhibition}\hfill \\ & =& \left(\mathrm{Absorbance}\mathrm{of}\mathrm{control}-\mathrm{Absorbance}\mathrm{of}\mathrm{treated}\mathrm{sample}\right)/\hfill \\ & & \mathrm{Absorbance}\mathrm{of}\mathrm{control}\times 100\%\hfill \end{array}$$

All experiments were performed in triplicate to ensure statistical reliability.

4.1.9. Kinetics of PTP1B Inhibiton

To determine the inhibition effect of chrysanthemin against PTP1B enzyme, increasing concentrations of p‐nitrophenyl phosphate (pNPP) were used as the substrate in the absence and presence of chrysanthemin at different concentrations. The data were determined by Lineweaver–Burk plot analysis, calculated from the results according to Michaelis–Menten kinetics.

4.1.10. Glucose Uptake Assay in L6 Cell Line

The glucose uptake assay was conducted using L6 myotube cells to evaluate the effect of chrysanthemin and insulin on glucose absorption. L6 cell line (rat skeletal muscle cell line) was procured from the National Centre for Cell Sciences (NCCS), Pune, India. L6 cells were cultured in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin and incubated at 37°C in a humidified atmosphere with 5% CO₂. Once the cells reached 80%–90% confluency, they were differentiated into myotubes by replacing the growth medium with a low‐serum medium (2% FBS) for 5–7 days. Differentiated myotubes were then seeded into 96‐well plates at an appropriate density and incubated overnight [56].

For the glucose uptake experiment, the cells were treated with varying concentrations (25, 50, and 100 µg/mL) of the chrysanthemin and insulin, while untreated cells served as the control group. The cells were incubated with the treatments for 24 h. After the incubation period, the glucose concentration in the culture medium was measured. The absorbance was recorded at 630 nm using a microplate reader [56, 57]. A standard glucose curve was generated, and the glucose concentration in the medium was determined based on the equation derived from the graph (y = 0.001x). The percentage of glucose uptake was calculated using the formula:

$$\begin{array}{ccc}& & \mathrm{Percentage}\mathrm{glucose}\mathrm{uptake}\hfill \\ & & =\left(\mathrm{Initial}\mathrm{glucose}\mathrm{concentration}-\mathrm{Final}\mathrm{glucose}\mathrm{concentration}\right)/\hfill \\ & & \mathrm{Initial}\mathrm{glucose}\mathrm{concentration}\times 100\%\hfill \end{array}$$

Statistical comparisons were conducted using one‐way analysis of variance (ANOVA). A p < 0.05 was considered statistically significant. These methods were applied consistently across all experiments to ensure reliable interpretation of the results.

Author Contributions

All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by S.S., M.K.T., and V.M. The first draft of the manuscript was written by S.S. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: cbdv70665‐sup‐0001‐SuppMat.pdf

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.