0% READ
FULL TEXT
INTRODUCTION
Diabetes Mellitus (DM) is a group of chronic metabolic disorders characterized by persistent hyperglycemia due to impaired insulin secretion, insulin action, or both. It has become a major global health concern, with its prevalence rising sharply in both developed and developing nations. In 2017, approximately 424.9 million people were affected worldwide. By 2019, this number had increased to over 463 million adults, and projections estimate it may reach 578 million by 2030 and 700 million by 2045 (Federation, 2019). The burden is significantly higher in urban areas (10.8%) compared to rural settings (7.2%). DM is broadly categorized into Type 1 diabetes, Type 2 diabetes, and Gestational Diabetes Mellitus (GDM). Type 1 diabetes is an autoimmune disorder resulting in the destruction of pancreatic β-cells, leading to insulin deficiency. Type 2 diabetes, the most prevalent form, is associated with insulin resistance and progressive β-cell dysfunction. GDM occurs during pregnancy, typically in the second or third trimester, affecting 2-10% of pregnancies, and is linked to hormonal changes that reduce insulin sensitivity. The pathogenesis of diabetes involves a complex network of interrelated biochemical pathways. Key mechanisms include oxidative stress, protein kinase C activation, hexosamine biosynthetic pathway flux, formation of Advanced Glycation End-products (AGEs), and activation of the polyol pathway. These overlapping mechanisms contribute not only to the onset of hyperglycemia but also to the development of long-term complications such as nephropathy, neuropathy, and retinopathy (Bailey and Day, 1989; DeFronzo, 1999). Current therapeutic agents target specific aspects of this complex network; however, they often exhibit limited efficacy in halting disease progression and are associated with adverse effects or drug resistance (DeFronzo, 1999). Consequently, there is an urgent need for novel, multi-target therapeutic approaches that can modulate various biochemical pathways involved in diabetes pathophysiology. Medicinal plants represent a promising source of such therapeutic agents. Numerous phytochemicals derived from traditional herbs have shown potential in regulating glucose metabolism and improving insulin sensitivity (Bailey and Day, 1989; He et al., 2017). Among these, Gentiana kurroo, an endangered medicinal herb native to the Himalayas, has garnered attention for its wide spectrum of pharmacological activities. Traditionally used for treating fever, inflammation, and liver disorders, G. kurroo contains several bioactive compounds such as amarogentin, gentiopicroside, gentianine, gentianidine, and isogentisin, many of which exhibit glucose-lowering and antioxidant properties (Manner and Humpf, 2014). Given the potential of these phytoconstituents to interact with multiple molecular targets, the present study aims to explore their efficacy using both in silico molecular docking and in vitro enzyme inhibition assays. The targets selected include PPAR-γ (3DZY), Glucokinase (4IXC), SIRT6 (3K35), GSK-3 (3F7Z), and Tyrosine kinase (1IR3), key regulators of glucose homeostasis, insulin signaling, and lipid metabolism. This dual approach seeks to identify natural compounds capable of modulating multiple diabetic targets simultaneously, thereby offering a holistic and effective strategy for diabetes management.
MATERIALS AND METHODS
In silico Study
Compound Selection and Preparation
In this molecular docking investigation, key phytoconstituents of Gentiana kurroo were selected based on literature evidence supporting their antidiabetic and antioxidant potential. Compounds such as amarogentin, gentiopicroside, gentianine, gentianidine, and isogentisin have been reported to act on various diabetes-related biochemical pathways, including oxidative stress, polyol pathway flux, protein kinase C activation, intracellular AGE formation, and the hexosamine biosynthetic pathway (Manner and Humpf, 2014; Bailey and Day, 1989; Oboh and Rocha, 2008). The 2D structures of the selected ligands were drawn using ChemDraw Ultra 7.0 and their corresponding 3D structures were generated and energy minimized. These molecular structures were then retrieved or cross-verified from the PubChem database. The ligand files were converted into PDBQT format after the addition of Gasteiger charges using AutoDock Tools (version 1.5.6).
Target Protein Preparation
Five diabetes-associated targets - PPAR-γ (3DZY), glucokinase (4IXC), SIRT6 (3K35), GSK-3 (3F7Z), and tyrosine kinase (1IR3) - were obtained from the RCSB Protein Data Bank. Proteins were prepared by removing water and native ligands, adding hydrogens and Kollman charges, energy minimization, and saving in PDBQT format using AutoDock Tools following Trott and Olson (2010) (Trott and Olson, 2010).
Molecular Docking Protocol
Docking simulations were performed using AutoDock Tools v1.5.6. Active sites were defined using optimized grid parameters generated by AutoGrid. Docking was executed with the Lamarckian Genetic Algorithm against five protein targets, using acarbose as reference. Binding energies and interaction details were analyzed from .dlg files (Trott and Olson, 2010).
Visualization and Interaction Analysis
Protein-ligand interactions were visualized using PyMOL (Pettersen et al., 2004), and further analyzed with the Protein-Ligand Interaction Profiler (PLIP) to characterize hydrogen bonds, π-π stacking, hydrophobic contacts, and salt bridges (Salentin et al., 2015). Docking results were ranked by binding energy and interaction stability to identify promising compounds for in vitro evaluation.
In vitro Anti-Diabetic Assays
α-Amylase and α-glucosidase inhibition assays were performed in vitro to evaluate the antidiabetic potential of amarogentin, gentiopicroside, gentianine, gentianidine, and isogentisin at different concentrations, using acarbose as the standard drug. Based on molecular docking results, Gentiopicroside and Isogentisin were selected for detailed enzyme inhibition studies targeting key carbohydrate-metabolizing enzymes. The assays were conducted in 96-well microplates following established protocols (Bernfeld, 1955; Kim et al., 2005). Briefly, 30 µL of test compounds (25-100 µg/mL) was incubated with 60 µL of enzyme solution (1 U/mL, PBS pH 6.9) at 37ºC for 10 min. The reaction was terminated with 1 M HCl and 5% KI, and absorbance was measured at 630 nm. All experiments were performed in triplicate, and percentage inhibition was calculated using standard equations.
ADME and Drug-likeness Evaluation
Swiss ADME and pkCSM online tools were used for the prediction of Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) profiles (Daina et al., 2017; Pires et al., 2015). Drug-likeness was evaluated using Lipinski’s Rule of Five.
Statistical Analysis
Statistical analysis was carried out using GraphPad Prism 6 software. Data are presented as mean±Standard Deviation (SD). Statistical significance was evaluated by one-way ANOVA followed by Dunnett’s multiple comparison post hoc test (*p<0.05, **p<0.01, ***p<0.001). Duncan’s multiple range test was also used where applicable (p<0.05).
RESULTS
Molecular Docking Analysis
Molecular docking was performed to evaluate the binding affinities of Amarogentin, Gentiopicroside, Gentianine, Gentianidine, and Isogentisin against five diabetes-related protein targets: PPAR-γ (3DZY), Glucokinase (4IXC), SIRT6 (3K35), GSK-3 (3F7Z), and Tyrosine kinase (1IR3). Binding energies ranged from -6.0 to -9.0 kcal/mol across all ligand-protein complexes. Isogentisin exhibited the strongest binding affinity toward Glucokinase (-9.0 kcal/mol) and showed consistently favorable binding energies across all targets (-7.9 to -9.0 kcal/mol). Gentiopicroside also demonstrated strong binding, particularly with Glucokinase (-8.9 kcal/mol), along with stable interactions with other targets. Key molecular interactions included hydrogen bonding, π-π stacking, and hydrophobic contacts. Isogentisin formed hydrogen bonds with THR65, CYS220, and TYR473, π-π stacking interactions with PHE282, and van der Waals interactions with LEU330. Gentiopicroside formed hydrogen bonds with HIS435, CYS220, and THR65, supported by van der Waals interactions with ALA292 and LEU330. Amarogentin, Gentianine, and Gentianidine exhibited moderate binding affinities stabilized by hydrogen bonds and hydrophobic interactions. Detailed binding energies and interaction profiles are presented in Table 1, while representative 2D and 3D interaction diagrams are shown in Figure 1.
| Compound | PPAR-γ (3DZY) | Glucokinase (4IXC) | SIRT6 (3K35) | GSK-3 (3F7Z) | Tyrosine Kinase (1IR3) | Key Interactions |
|---|---|---|---|---|---|---|
| Amarogentin | -6.8 kcal/mol | -6.7 kcal/mol | -8.9 kcal/mol | -8.6 kcal/mol | -8.7 kcal/mol | H-bond: CYS220, HIS323; Hydrophobic: PHE282, LEU330 |
| Gentiopicroside | -7.2 kcal/mol | -8.9 kcal/mol | -8.6 kcal/mol | -7.3 kcal/mol | -7.3 kcal/mol | H-bond: HIS435, CYS220, THR65; van der Waals: ALA292, LEU330 |
| Gentianine | -6.4 kcal/mol | -7.1 kcal/mol | -6.1 kcal/mol | -6.4 kcal/mol | -6.0 kcal/mol | H-bond: THR65; Hydrophobic: VAL135, PHE282 |
| Gentianidine | -6.0 kcal/mol | -7.0 kcal/mol | -6.1 kcal/mol | -6.5 kcal/mol | -6.0 kcal/mol | H-bond: SER342; van der Waals: VAL135, ALA292 |
| Isogentisin | -7.5 kcal/mol | -9.0 kcal/mol | -7.9 kcal/mol | -8.1 kcal/mol | -8.0 kcal/mol | H-bond: THR65, CYS220, TYR473; π-π stacking with PHE282; van der Waals: LEU330 |
In vitro Anti-diabetic Activity
Enzyme Inhibition by Selected Compounds at 100 µg/mL
At 100 µg/mL, Gentiopicroside and Isogentisin showed strong α-amylase (87.3%, 85.9%) and α-glucosidase (89.1%, 88.5%) inhibition, comparable to acarbose. Amarogentin showed moderate, and Gentianine and Gentianidine lower inhibition. IC₅₀ values reflected these trends (Table 2, Figure 2).
| Compound | α-Amylase Inhibition (%) (Mean±SD) | α-Glucosidase Inhibition (%) (Mean±SD) | α-Amylase IC₅₀ (µg/mL) | α-Glucosidase IC₅₀ (µg/mL) |
|---|---|---|---|---|
| Acarbose | 89.2±1.3 | 91.7±1.1 | 56.3 | 48.9 |
| Amarogentin | 73.5±2.2 *** | 75.4±2.0 *** | 71.4 | 68.2 |
| Gentiopicroside | 87.3±1.6 ns | 89.1±1.3 ns | 58.7 | 50.1 |
| Gentianine | 65.7±2.8 *** | 68.2±2.5 *** | 80.5 | 76.3 |
| Gentianidine | 61.4±3.1 *** | 64.5±2.6 *** | 85.1 | 78.9 |
| Isogentisin | 85.9±1.5 ns | 88.5±1.2 ns | 59.2 | 51.3 |
Concentration-Dependent Enzyme Inhibition by Gentiopicroside and Isogentisin
Gentiopicroside and Isogentisin were evaluated for dose-dependent inhibition at 25-100 µg/mL, showing increasing α-amylase and α-glucosidase inhibition with concentration. Maximum inhibition occurred at 100 µg/mL, with lower activity at reduced doses (Table 3).
| Concentration (µg/mL) | % α-Amylase Inhibition (Mean±SD), Gentiopicroside | % α-Amylase Inhibition (Mean±SD), Isogentisin | % α-Glucosidase Inhibition (Mean±SD), Gentiopicroside | % α-Glucosidase Inhibition (Mean±SD), Isogentisin |
|---|---|---|---|---|
| 100 | 87.3±1.6 (ns) | 85.9±1.5 (ns) | 87.3±1.6 (ns) | 85.9±1.5 (ns) |
| 75 | 70.8±1.4 (***) | 72.3±1.3 (***) | 70.8±1.4 (***) | 72.3±1.3 (***) |
| 50 | 53.6±1.1 (***) | 55.8±1.0 (***) | 53.6±1.1 (***) | 55.8±1.0 (***) |
| 25 | 32.7±0.8 (***) | 34.1±0.7 (***) | 32.7±0.8 (***) | 34.1±0.7 (***) |
Table 3 showing the Concentration-Dependent Enzyme Inhibition by Gentiopicroside Isogentisin.
ADME Profiling
ADME and toxicity predictions using SwissADME and pkCSM indicated Amarogentin violated Lipinski’s rules and had low gastrointestinal absorption. Gentiopicroside, Gentianine, Gentianidine, and Isogentisin complied with Lipinski’s criteria and showed high absorption. No compounds were predicted hepatotoxic or mutagenic; BBB permeability was observed only for Gentianine and Gentianidine (Table 4).
| Property | Amarogentin | Gentiopicroside | Gentianine | Gentianidine | Isogentisin |
|---|---|---|---|---|---|
| Molecular Weight (g/mol) | 524.47 | 402.37 | 223.24 | 237.28 | 298.27 |
| H-bond Donors | 6 | 5 | 1 | 1 | 2 |
| H-bond Acceptors | 13 | 9 | 3 | 4 | 5 |
| LogP (Lipophilicity) | -0.42 | -0.3 | 1.12 | 1.3 | 1.05 |
| Gastrointestinal (GI) Absorption | Low | High | High | High | High |
| Lipinski’s Rule of Five Violations | 2 Violations | No Violation | No Violation | No Violation | No Violation |
| Blood-Brain Barrier (BBB) Permeability | No | No | Yes | Yes | No |
| Hepatotoxicity | No | No | No | No | No |
| AMES Toxicity (Mutagenicity) | No | No | No | No | No |
DISCUSSION
The present study employed an integrated in silico and in vitro approach to comprehensively evaluate the antidiabetic potential of selected phytoconstituents. This multi-level strategy generated converging evidence on molecular interactions, functional enzyme inhibition, and pharmacokinetic suitability, thereby strengthening the biological relevance of the findings. Molecular docking analysis identified Isogentisin and Gentiopicroside as the most promising candidates, exhibiting strong and consistent binding affinities toward multiple diabetes-related protein targets, supporting a multitarget mechanism of action. Such multitarget activity is particularly advantageous in diabetes management, given the complex dysregulation of glucose metabolism, insulin signaling, and associated metabolic pathways. Among the evaluated targets, glucokinase plays a pivotal role in glucose sensing and insulin secretion, while PPAR-γ, GSK-3, SIRT6, and tyrosine kinase are critically involved in insulin sensitivity, glycogen synthesis, lipid metabolism, and intracellular signaling. The favorable binding energies observed for Isogentisin and Gentiopicroside across these targets suggest their potential to modulate interconnected pathways governing glycemic control. Increasing evidence indicates that natural compounds capable of interacting with multiple molecular targets may offer therapeutic advantages over single-target synthetic drugs, particularly in chronic metabolic disorders such as diabetes. The stability and specificity of ligand-protein complexes were supported by extensive hydrogen bonding, hydrophobic interactions, and π-π stacking. Hydrogen bonds with catalytically important residues, including THR65 and CYS220, contribute significantly to binding specificity and complex stability, while π-π stacking interactions enhance affinity by stabilizing aromatic moieties within enzyme active sites. These interaction patterns indicate effective molecular recognition and favorable ligand orientation. Comparable interaction profiles have been reported for established antidiabetic phytochemicals such as quercetin and epicatechin, which demonstrate multitarget binding and improve glycemic control through enzyme inhibition and insulin-sensitizing mechanisms (Ali et al., 2024; Zu et al., 2021). The consistency of these observations supports the mechanistic plausibility of the present findings. The in vitro enzyme inhibition assays further validated the molecular docking results. Gentiopicroside and Isogentisin exhibited potent inhibitory activity against both α-amylase and α-glucosidase, with inhibition percentages and IC₅₀ values comparable to acarbose, a clinically established antidiabetic drug. Inhibition of these digestive enzymes delays carbohydrate digestion and reduces intestinal glucose absorption, thereby lowering postprandial hyperglycemia. The strong agreement between computational docking scores and experimental enzyme inhibition data reinforces the reliability of the in silico predictions and confirms the functional relevance of the observed molecular interactions. In contrast, Amarogentin exhibited moderate enzyme inhibition, while Gentianine and Gentianidine showed comparatively lower activity. These differences may be attributed to weaker binding affinities, fewer stabilizing interactions, or suboptimal molecular orientation within enzyme active sites. Such structure-activity relationships highlight the importance of specific functional groups and interaction patterns in determining biological efficacy. These findings are consistent with previous reports indicating that phenolic and flavonoid compounds with optimal hydrogen-bonding capacity and aromatic features tend to exhibit enhanced antidiabetic activity (Patel et al., 2019). The concentration-dependent inhibition observed for Gentiopicroside and Isogentisin further supports their pharmacological relevance. A clear dose-response relationship was evident, with increasing concentrations leading to enhanced inhibition of both α-amylase and α-glucosidase. Dose dependency is a desirable characteristic for therapeutic agents, as it reflects predictable biological responses and facilitates rational dose optimization. Similar concentration-dependent enzyme inhibition profiles have been reported for other natural antidiabetic compounds, reinforcing the translational potential of these findings (Anshika et al., 2022). ADME and drug-likeness analysis indicated that Gentiopicroside and Isogentisin comply with Lipinski’s criteria, show good gastrointestinal absorption, and lack predicted hepatotoxicity or mutagenicity. In contrast, Amarogentin showed Lipinski violations and lower absorption. Limited blood-brain barrier permeability suggests reduced CNS-related adverse effects. Overall, the combined in silico and in vitro findings identify Gentiopicroside and Isogentisin as promising multitarget antidiabetic candidates with effective enzyme inhibition and favorable pharmacokinetic profiles. However, the absence of in vivo and clinical data warrants further pharmacodynamic, bioavailability, and long-term safety studies in appropriate animal models to support clinical translation.
CONCLUSION
The present study demonstrates the antidiabetic potential of selected phytoconstituents through an integrated in silico and in vitro approach. Molecular docking analysis identified Gentiopicroside and Isogentisin as the most promising compounds, exhibiting strong and consistent binding affinities toward multiple diabetes-related targets, including glucokinase, PPAR-γ, GSK-3, SIRT6, and tyrosine kinase. These interactions were supported by favorable hydrogen bonding, hydrophobic contacts, and π-π stacking interactions, suggesting stable ligand-protein complexes. In vitro enzyme inhibition assays further confirmed the computational findings, with both compounds showing potent α-amylase and α-glucosidase inhibitory activities comparable to acarbose, along with clear concentration-dependent responses. ADME and drug-likeness predictions indicated favorable pharmacokinetic and safety profiles, particularly for Gentiopicroside and Isogentisin. Overall, the results highlight these compounds as promising multitarget antidiabetic candidates. However, further in vivo studies and clinical investigations are required to validate their efficacy, bioavailability, and long-term safety before therapeutic application.