Visual abstract

Introduction

Type 2 diabetes mellitus (T2DM) is a chronic metabolic disorder characterized by persistent hyperglycemia arising from impaired insulin secretion, insulin resistance, and dysregulated glucose absorption [1]. Although current therapeutic agents (such as α-glucosidase inhibitors, insulin sensitizers, and compounds that enhance peripheral glucose uptake) target specific aspects of glucose regulation, their clinical efficacy is often limited by adverse effects or insufficient multi-pathway modulation [2,3]. This has led to increased interest in identifying novel compounds capable of acting on multiple metabolic checkpoints to provide a safer and more effective antidiabetic intervention [4].

Postprandial hyperglycemia is a major driver of glucotoxicity in T2DM, making intestinal α-glucosidase inhibition a well-established strategy to reduce the rate of carbohydrate digestion and glucose absorption [5]. In parallel, insulin resistance is partially mediated by the overactivity of protein tyrosine phosphatase 1B (PTP1B) and remains a key pathological hallmark of T2DM [6]. PTP1B negatively regulates insulin receptor signaling and therefore its inhibition has emerged as a validated therapeutic target for restoring insulin sensitivity [7,8].

Skeletal muscle (represented in this study by a mouse cell line differentiated from C2C12 myoblasts to form C2C12 myotubes) plays a central role in whole-body glucose homeostasis and accounts for 70%–80% of insulin-stimulated glucose disposal [9,10]. Enhancing glucose uptake in muscle cells, primarily through Glucose Transporter Type 4 translocation and Akt-dependent insulin signaling, serves as a crucial functional indicator when evaluating the metabolic efficacy of candidate antidiabetic agents [11]. Despite extensive research on these mechanisms, few studies have integrated them into a cohesive, multilevel screening framework that mirrors the physiological sequence of glucose metabolism [12].

In this study, 5 materials [β-glucan, indigestible maltodextrin, Helianthus tuberosus (Jerusalem artichoke), Momordica charantia (bitter melon), and Lagerstroemia speciosa (banaba leaf)] were selected for evaluation based on their relevance to glucose metabolism or metabolic effects. Indigestible maltodextrin was included as a functional carbohydrate known to modulate postprandial glycemic response [13–17].

To address this gap, a screening strategy was established that included: (1) α-glucosidase inhibition to evaluate the suppression of glucose entry; (2) PTP1B inhibition to assess the enhancement of insulin signaling; (3) glucose uptake in skeletal muscle cells to confirm functional metabolic efficacy (based on these assays, the compound exhibiting the strongest overall antidiabetic activity was selected for subsequent network pharmacology analysis to elucidate its putative molecular targets and mechanistic pathways); and (4) network pharmacology-based prediction of active compounds and their molecular targets (performed only for the topper-forming candidate identified through experimental screening).

By combining biochemical assays, cell-based evaluations, and computational analysis, this integrated pipeline offers a comprehensive approach for identifying natural or synthetic compounds with multitarget antidiabetic potential. Such a systematic strategy enhances screening accuracy, reduces false-positive outcomes, and facilitates the discovery of compounds with synergistic activities relevant to T2DM management.

Materials and Methods

1. Plant materials and extraction

The plant-derived materials including Lagerstroemia speciosa (banaba leaf), Helianthus tuberosus (Jerusalem artichoke), and Momordica charantia (bitter melon) were purchased from Gyeongdong Herbal Market (Seoul, Republic of Korea). Maltodextrin and β-glucan (≥ 95% purity) were obtained from Sigma-Aldrich (St. Louis, MO, USA).

Each dried material was pulverized and sieved to obtain a uniform fine powder. Subsequently, 2g of each powdered sample was mixed with distilled water to make a 20-fold (w/v) volume of mixture which was subjected to ultrasonic-assisted extraction for 60 minutes at room temperature. After extraction, the mixtures were centrifuged at 4,000 × g for 20 minutes, and the supernatants were carefully collected.

The harvested supernatants were filtered through a 0.2 μm membrane filter (Corning Inc., Corning, NY, USA) and subsequently freeze-dried using a laboratory lyophilizer. The freeze-drying process was performed under the following conditions: freezing at −80°C for at least 12 hours, followed by primary drying at −40°C under a vacuum (< 0.1 mbar), and secondary drying at 20°C to remove residual moisture. The resulting lyophilized powders were sealed in airtight containers and stored at −20°C.

2. α-Glucosidase inhibition assay

The α-glucosidase inhibitory activity was determined using Saccharomyces cerevisiae α-glucosidase. Briefly, the enzyme solution was preincubated with various concentrations of the test samples at 37°C for 10 minutes. The reaction was initiated by the addition of the substrate 5 mM p-nitrophenyl-α-D-glucopyranoside and incubated for another 20 minutes under the same conditions. The reaction was terminated by adding 0.1 M sodium carbonate, and the absorbance was measured at 405 nm using a microplate reader. The percentage inhibition was calculated relative to the control.

3. PTP1B inhibition assay

The inhibitory activity of the samples against PTP1B was evaluated using a commercial PTP1B inhibitor screening kit (Abcam, Cambridge, UK), following the manufacturer’s instructions. Briefly, the assay was performed by adding the enzyme working solution and the pNPP substrate solution to each well, followed by treatment with the test samples at the indicated concentrations. After incubation at 37°C for 30 minutes, the reaction was terminated with the provided stop solution. The absorbance was measured at 405 nm using a microplate reader, and the inhibitory rate was calculated relative to the kit control wells.

4. Cell culture

The cell line C2C12 was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and maintained in Dulbecco’s Modified Eagle Medium (Corning Inc., Corning, NY, USA; DMEM) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (Gibco, Grand Island, NY, USA) at 37°C in a humidified atmosphere containing 5% CO₂. For differentiation purposes, C2C12 cells were cultured to 80%–90% confluence and a differentiation medium (Corning Inc., Corning, NY, USA) containing 2% horse serum and 1% penicillin-streptomycin used for 4 days to induce myotube formation.

5. Cell viability

Cell viability was assessed using the EZ-CytoX assay (Daeil Lab Service Co., Ltd., Seoul, Republic of Korea) and performed according to the manufacturer’s protocol. Differentiated C2C12 myotubes were seeded onto 96-well plates and treated with the sample extracts for 24 hours in DMEM containing 2% horse serum and 1% penicillin-streptomycin. After treatment, EZCytoX reagent was added to each well and incubated for 1 hour at 37°C. Absorbance was measured at 450 nm using a microplate reader, and cell viability was calculated as a percentage relative to the untreated control group.

6. Glucose uptake assay

Following differentiation, C2C12 myotubes were treated with the sample extracts for 16 h in DMEM containing 1% penicillin-streptomycin, 2% horse serum, 10% fetal bovine serum, and 2% bovine serum albumin. Glucose uptake activity was measured using the 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl) amino)-2-deoxyglucose (2-NBDG) uptake assay kit (Sigma-Aldrich, St. Louis, MO, USA) following the manufacturer’s instructions. Fluorescence intensity was detected at excitation/emission wavelengths of 485/535 nm using a microplate reader, and the uptake was expressed as a percentage relative to untreated control cells.

7. Statistical analysis

All experimental data were obtained from 3 independent replicates and are presented as the mean ± SE of the mean. Differences between the treatment and control groups were evaluated using one-way analysis of variance. Statistical analyses were performed using GraphPad Prism 10 software (Version 10.6.1, GraphPad Software Inc., San Diego, CA, USA). A value of p < 0.05 was considered to indicate statistical significance.

8. Network pharmacology analysis

8.1. Scope of analysis

This study aimed to investigate the potential molecular mechanisms underlying the antidiabetic effects of Lagerstroemia speciosa (banaba leaf) using a network pharmacology approach. The analysis encompassed the identification of bioactive compounds, prediction of their putative protein targets, exploration of disease-associated targets, and subsequent pathway enrichment analyses (GO and KEGG). Through integrative computational analysis, the relationship between active phytochemicals and diabetes-related targets was systematically evaluated.

8.2. Collection and preprocessing of bioactive compounds

Information on the bioactive constituents of Lagerstroemia speciosa was retrieved from natural product databases, including TM-MC [18] and COCONUT [19]. The chemical structures of all compounds were standardized in SMILES (Simplified Molecular Input Line Entry System) format and verified via PubChem [20]. The biological targets associated with each compound were obtained from multiple resources such as STITCH [21], TM-MC, and ChEMBL [22]. All compound-target associations were curated and integrated into a unified dataset for further analysis.

8.3. Identification of disease-associated targets

Diabetes-related gene targets were collected from the Comparative Toxicogenomics Database [23]. Two categories of disease associations were considered: (1) Marker/Mechanism genes directly involved in disease pathogenesis or serving as biomarkers, and (2) Therapeutic genes with established pharmacological relevance. Only well-documented genes supported by literature evidence were prioritized for analysis. Overlapping targets between compound-derived and disease-related gene sets were identified as potential therapeutic targets. Disease classifications were annotated according to the MeSH (Medical Subject Headings) hierarchy to define relevant disease categories.

8.4. Network construction and pathway analysis

Common targets derived from compound-disease intersections were used to construct a protein-protein interaction (PPI) network using STRING [24]. Interactions with a combined confidence score ≥ 0.7 were retained for network construction. Network topology was analyzed using Cytoscape.js [25], calculating degree, betweenness, and closeness centralities to identify key hub targets. This web-based approach was designed as a lightweight alternative to traditional desktop-based tools such as Cytoscape MCODE.

8.5. GO and KEGG pathway enrichment analysis

Functional enrichment analysis was performed for the core target genes using the Enrichr platform [26]. Both Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were evaluated. Pathways with p < 0.05 were considered significantly enriched, and biological functions associated with the core targets were interpreted in the context of glucose metabolism, insulin signaling, and inflammation-related processes relevant to diabetes pathophysiology.

Results

1. α-Glucosidase inhibition

The inhibitory activity of α-Glucosidase was assessed by using sample concentrations of 6.25, 12.5, 25, 50, 100, and 200 μg/mL (Figure 1). The positive control, acarbose, exhibited 56.9% inhibition relative to the control, while banaba leaf extract showed a dose-dependent decrease in α-glucosidase activity by 1.4%, 14%, 33%, 40.4%, 46%, and 48.1% compared with the control, and demonstrated statistically significant differences (*p < 0.05, **p < 0.01, ***p < 0.001). However, Jerusalem artichoke, bitter melon, maltodextrin, and β-glucan showed little to no inhibitory activity against the α-glucosidase enzyme.

2. PTP1B inhibition

PTP1B inhibitory activity was evaluated by treating samples at concentrations of 0, 50, 200, and 500 μg/mL (Figure 2). The positive control exhibited inhibitory activities of 58.7% and 82.4% as the concentration increased. Jerusalem artichoke showed less than 10% PTP1B inhibition at all concentrations, indicating minimal enzyme inhibition. Banaba leaf extract showed, approximately, a 49%–50% PTP1B inhibition at 50 μg/mL, but had relatively low activity at 200 and 500 μg/mL indicating that inhibition was not dose-dependent. Bitter melon exhibited less than 10% PTP1B inhibition at all tested concentrations. Maltodextrin showed about 20% PTP1B inhibition at the highest concentration (500 μg/mL), but overall inhibition was minimal. β-glucan demonstrated low PTP1B inhibition ranging from about 8% to 18% at 50 to 500 μg/mL, with IC₅₀ values above 500 μg/mL for all samples except banaba leaf. Overall, Jerusalem artichoke, bitter melon, maltodextrin, and β-glucan demonstrated minimal PTP1B inhibition, while banaba leaf showed relatively high inhibition at low concentrations.

3. Cell viability

The cell viability assessment revealed no significant difference, compared with the control, in the banaba leaf, Jerusalem artichoke, bitter melon, and maltodextrin treatment groups at all tested concentrations (6.25 to 200 μg/mL; Figure 3). β-glucan showed a slight decrease in cell viability at 200 μg/mL which was significant (p < 0.01). However, cell viability remained above 90% at 200 μg/mL, indicating no cytotoxicity. Consequently, experiments proceeded using all tested concentrations.

4. Glucose uptake

The glucose uptake assay was used to evaluate samples at concentrations of 6.25 to 200 μg/mL (Figure 4). The positive control, insulin, increased glucose uptake by 58% relative to the control. Treatment with Jerusalem artichoke, bitter melon, maltodextrin, and β-glucan showed a slight increase, but did not reach statistical significance at any concentration. In contrast, banaba leaf treatment resulted in glucose uptake increases of 15.8%, 25.3%, and 42.2% at 50, 100, and 200 μg/mL, respectively. These increases were statistically significant compared with the control at all tested concentrations, demonstrating a dose-dependent enhancement of glucose uptake within the examined concentration range (*p < 0.05, **p < 0.01, ***p < 0.001).

5. Scope of analysis

This analysis aimed to identify the pharmacological targets of bioactive constituents from banaba leaf by compiling 20 compounds from the TM-MC and COCONUT databases, standardizing their chemical structures using PubChem SMILES data, and categorizing them based on target availability in resources such as STITCH and ChEMBL. Amongst these, 7 compounds possessed validated target information, while thirteen lacked clear associations. The 7 annotated compounds were included in the network pharmacology workflow, yielding a total of 397 unique target genes, which served as the basis for subsequent disease-association analysis and PPI network construction.

6. Collection and preprocessing of bioactive compounds

To evaluate the antidiabetic potential of banaba leaf, insulin resistance was selected as the primary disease model for network pharmacology analysis. Using the MeSH classification system, this condition is categorized within the following hierarchical structure; Nutritional and Metabolic Diseases > Metabolic Diseases > Glucose Metabolism Disorders > Hyperinsulinism > Insulin Resistance (MeSH ID: D007333).

7. Identification of disease-associated targets

A total of 397 compound-associated targets and 83 experimentally validated insulin-resistance-related targets were retrieved from public databases. Intersection analysis identified 13 overlapping genes shared between the 2 datasets (Figure 5). These overlapping targets were classified as putative core targets of banaba leaf in the context of insulin resistance and were subsequently used for downstream network and pathway analyses (Table 1).

8. Network construction and pathway analysis

A PPI network was constructed for the 13 potential targets, in which darker red nodes represented proteins with higher degree values, indicating stronger connectivity amongst interacting nodes (Figure 6A). Core target selection criteria were established based on network centrality measures, and the following thresholds were applied to identify highly central nodes: degree ≥ 14, betweenness ≥ 0.015, and Closeness ≥ 0.5. These values correspond to hub proteins mediating multiple interactions within the network, key intermediates in signal-transduction pathways, and nodes capable of influencing the overall network with high efficiency, respectively, and were determined using the average centrality values of all nodes to ensure functional relevance and biological reliability. Diabetes involves complex mechanisms including insulin resistance, inflammatory responses, and energy metabolism dysregulation. Therefore, the analysis parameters were optimized so that genes associated with metabolic control and cell survival (such as insulin, Insulin-like Growth Factor 1, Mechanistic Target of Rapamycin, Adenosine Monophosphat -Activated Protein Kinase (AMPK), Forkhead Box (FOX) O1, FOXO3, and Tumor Necrosis Factor) met the selection thresholds. This resulted in the identification of 19 core targets (Figure 6B; Table 2).

9. GO and KEGG pathway enrichment analysis

GO enrichment analysis of the 19 key targets revealed significant terms across the 3 major GO categories: Biological Process, Cellular Component, and Molecular Function. The most enriched Biological Process terms included pathways related to mitogen-activated protein kinase (MAPK) regulation, phosphatidylinositol 3-kinase-protein kinase (PI3K-Akt) signaling regulation, and insulin receptor signaling. Cellular Component enrichment identified membrane associated structures such as plasma membrane rafts, caveolae, and protein kinase complexes. Molecular Function enrichment highlighted kinase binding, cytokine activity, and insulin like growth factor receptor binding.

KEGG pathway analysis identified 124 significantly enriched pathways (p < 0.05). Amongst the top 20 pathways, insulin signaling, PI3K-Akt signaling, MAPK signaling, FOXO signaling, AMPK signaling, autophagy, and diabetes related pathways (Type I diabetes mellitus, Type II diabetes mellitus, insulin resistance) were prominently enriched (Figure 7). These results collectively indicate that the predicted targets are closely associated with key metabolic and insulin regulatory processes.

To further contextualize these enriched pathways with compound-target interactions, the Compound-Target-Pathway (CTP) network was constructed (Figure 8). In this network, compounds positioned earlier in the sequence, such as ellagic acid, exhibited higher node degrees and stronger connections to hub targets including insulin, insulin receptor, Insulin-like Growth Factor 1, Epidermal Growth Factor Receptor, and Tumor Necrosis Factor. These hubs correspond directly to the major pathways highlighted in the KEGG analysis, and illustrate a coherent relationship between computational predictions and network level biological relevance.

Conversely, compounds with fewer target interactions that appeared later in the layout and were rendered with lower opacity, reflected weaker but still meaningful associations within the metabolic landscape. The dense interconnectivity observed amongst targets and KEGG pathways (e.g., PI3K-Akt signaling, MAPK signaling, insulin resistance, AMPK signaling) demonstrated how individual compounds converge onto shared metabolic pathways.

KEGG enrichment highlights the critical signaling pathways involved in glucose regulation and insulin sensitivity, and the CTP network maps how the selected compounds interface with these pathways through key molecular targets (Figures 7 and 8). This integrated interpretation supports the multi target antidiabetic potential of the compounds and strengthens the mechanistic justification for their biological activity.

Discussion

A comprehensive evaluation using a multistage screening approach, established in this study, demonstrated that Lagerstroemia speciosa (banaba leaf) exhibits coherent and biologically meaningful antidiabetic activity (Figure 9). Whilst 5 materials were initially selected, only banaba showed consistent activity in α-glucosidase inhibition, PTP1B suppression, and enhancement of glucose uptake in C2C12 myotubes. This indicated banaba leaf capacity to modulate glucose metabolism at multiple physiological checkpoints. Notably, the PTP1B assay did not reveal a dose dependent response of banaba which may be attributed to interference caused by the intrinsic color of banaba which can affect absorbance-based measurements and compromise the accuracy of optical detection. This limitation is not unique to banaba extract, as other plant-derived samples used in this study may also introduce similar interference, potentially leading to variability or underestimation of inhibitory activity. It has been reported that in such scenarios the observed results should be interpreted with caution when assessing the true inhibitory potential of the tested samples [27]. When using absorbance-based enzyme assays reliable quantitative evaluation in the presence of strongly pigmented compounds may not always be provided. In this context, the use of alternative analytical approaches such as fluorescence-based PTP1B inhibition assays may help to reduce background interference and improve detection sensitivity. In addition, complementary validation using cell-based assays, including the analysis of insulin signaling pathways (e.g., phosphorylation of Akt or Glucose Transporter Type 4 translocation), could provide indirect, yet biologically relevant, evidence supporting PTP1B inhibition. While banaba showed the most coherent activity across all assays, the remaining samples exhibited more limited or assay-specific effects under the conditions in this study. The other samples including Jerusalem artichoke, bitter melon, maltodextrin, and β-glucan displayed partial or negligible effects in the assays used and did not exhibit a consistent functional profile within the tested concentration ranges. However, this does not necessarily indicate a lack of intrinsic antidiabetic potential, because the activity of natural materials can vary substantially depending on extraction conditions (such as the use of aqueous versus ethanol-based solvents) as well as the selected dose range. In this regard, the platform used in this study may be more appropriately interpreted as a strategy for identifying candidates with reproducible multitarget activity under standardized conditions, rather than for excluding all other materials as false positives. Network pharmacology analysis supported the experimental results and revealed 13 overlapping targets between banaba-derived compounds and insulin resistance, along with 19 key hub genes associated with insulin signaling, inflammatory regulation, and metabolic homeostasis. These targets were enriched in pathways closely linked to T2DM pathophysiology, including PI3K-Akt, AMPK, MAPK, FOXO, and insulin signaling, which may demonstrate a strong mechanistic alignment between predicted molecular interactions and observed biological activities. Furthermore, amongst the predicted bioactive compounds, ellagic acid was identified as a key component with high connectivity to multiple hub targets. Previous studies have reported that aqueous extracts of Lagerstroemia speciosa contain bioactive constituents, such as ellagic acid and corosolic acid, and this supports their relevance to glucose metabolism and insulin signaling [28,29]. Moreover, it has been documented that strongly pigmented plant extracts can interfere with absorbance-based enzymatic assays, potentially leading to variability or underestimation of inhibitory activity [30]. Taken together, these findings suggest that ellagic acid and related compounds may contribute to the observed antidiabetic effects. Further analytical approaches will be required to more precisely define their roles within the extract. Collectively, these findings highlight the value of integrating biochemical assays, cellular functional tests, and in silico modeling to identify natural products with genuine multitarget antidiabetic potential. The results of this study reinforce the strength of the 4-stage systematic framework spanning α-glucosidase inhibition, PTP1B inhibition, glucose uptake enhancement, and network pharmacology which enabled a rigorous and sequential evaluation of metabolic efficacy whilst facilitating the prioritization of candidates with consistent multitarget activity. Despite these strengths, the absence of in vivo validation remains a notable limitation, as the physiological relevance, bioavailability, and systemic metabolic effects of banaba-derived compounds cannot be fully inferred from in vitro and in silico analyses alone. Several limitations must be acknowledged, including the exclusive use of aqueous extracts, reliance on a single muscle cell model, and the absence of in vivo validation, all of which warrant further investigation.

Conclusion

This study established an integrated and multilevel antidiabetic screening strategy that combines biochemical assays, cellular functional evaluation, and network pharmacology to more accurately identify natural products with multitarget therapeutic potential. Amongst the 5 tested samples, Lagerstroemia speciosa (banaba leaf) consistently demonstrated superior activity by inhibiting α-glucosidase, suppressing PTP1B, and enhancing glucose uptake in skeletal muscle cells, thereby indicating its capacity to modulate key metabolic pathways relevant to T2DM. Although the PTP1B assay did not show a dose-dependent response (maybe due to color interference that limited accurate absorbance detection). Network pharmacology findings supported the in vitro results by identifying core molecular targets and enriched pathways including insulin signaling, PI3K-Akt, MAPK, AMPK, and FOXO signaling that closely align with diabetes pathophysiology. Together, these results underscore the methodological strength of the 4-stage screening framework and position banaba leaf as a promising natural candidate for further preclinical antidiabetic investigations. Future studies incorporating diverse extraction methods, additional metabolic cell models, and comprehensive in vivo validation will be essential to fully elucidate its therapeutic potential, bioavailability, and mechanistic relevance of banaba in T2DM.

Article information

Author Contributions

All authors conceptualized and designed the study, performed the literature review, and were responsible for data analysis and interpretation.

Conflicts of Interest

The authors have no conflicts of interest to declare.

Author Use of AI Tools Statement

During the preparation of this manuscript, the authors used ChatGPT, a generative artificial intelligence tool developed by OpenAI, for language editing, grammar correction, and improvement of clarity and readability.

Funding

This work was supported by the Technology Innovation Program (no.: RS-2025-13642970) funded by the Ministry of Trade, Industry and Energy, Republic of Korea. This research was financially also supported by the Ministry of Trade, Industry and Energy, Republic of Korea, under the Bio-Industry Open Ecosystem Creation Promotion Project (no.: P0027490).

Ethics Statement

This research did not involve any human or animal experiments.

Data Availability

Figure 1

Effects of test samples on α-glucosidase inhibitory activity: (A) Jerusalem artichoke; (B) banaba leaf; (C) bitter melon; (D) maltodextrin; and (E) β-glucan. ** p < 0.01, and *** p < 0.001.

α-Glucosidase inhibition was measured after treatment with each sample at concentrations ranging from 6.25 to 200 μg/mL.

Figure 2

Effects of test samples PTP1B inhibitory activity: (A) Jerusalem artichoke; (B) banaba leaf; (C) bitter melon; (D) maltodextrin; (E) β-glucan; and (F) positive control.

PTP1B inhibitory activity was measured after treatment with each sample at concentrations ranging from 50 to 500 μg/mL, with the positive control used at 50 to 100 μM.

Figure 3

Effects of test samples on the viability of C2C12 myoblasts: (A) Jerusalem artichoke; (B) banaba leaf; (C) bitter melon; (D) maltodextrin; and (E) β-glucan. ** p < 0.01. Cell viability was measured after 24 hours of treatment with each sample at concentrations ranging from 6.25 to 200 μg/mL.

Figure 4

Effects of test samples on glucose uptake: (A) Jerusalem artichoke; (B) banaba leaf; (C) bitter melon; (D) maltodextrin; and (E) β-glucan. ** p < 0.01, and *** p < 0.001.

Glucose uptake assay was measured after treatment with each sample at concentrations ranging from 6.25 to 200 μg/mL, with 100 nM insulin as an inducer.

Figure 5

Predicted targets of the selected compound(s): (A) the total number of compound-associated targets; (B) experimentally validated insulin-resistance-related target genes; and (C) resistance-related targets.

Figure 6

A PPI network was constructed for the 13 potential targets: (A) is based on the criteria of degree ≥ 14, betweenness ≥ 0.015, and closeness ≥ 0.5; and (B) shows a total of 19 key targets were identified.

Each node represents a protein, and each edge indicates a protein-protein interaction. The redder the node, the higher its centrality, meaning it is more highly connected to other nodes. The network contains 49 nodes and 339 edges.

PPI = protein-protein interaction.

Figure 7

KEGG enrichment analysis of the top 20 significantly enriched pathways.

Bubble size represents the number of associated genes (Gene Count), while bubble color indicates statistical significance based on -log10 (p value). Gene ratio denotes the proportion of input genes mapped to each pathway.

Figure 8

CTP network visualization. The CTP network illustrates interactions among selected compounds, hub targets, and enriched KEGG pathways. Compounds, targets, and pathways are represented as diamonds, circles, and hexagons, respectively. Node size reflects degree connectivity.

CTP = Compound-Target-Pathway.

Figure 9

Schematic illustration of the systematic antidiabetic screening approach.

References

• [1] Galicia-Garcia U, Benito-Vicente A, Jebari S, Larrea-Sebal A, Siddiqi H, Uribe KB, et al. Pathophysiology of type 2 diabetes mellitus. Int J Mol Sci 2020;21(17):6275. ArticlePubMedPMC
• [2] Alssema M, Ruijgrok C, Blaak EE, Egli L, Dussort P, Vinoy S, et al. Effects of alpha-glucosidase-inhibiting drugs on acute postprandial glucose and insulin responses: a systematic review and meta-analysis. Nutr Diabetes 2021;11(1):11. ArticlePubMedPMCPDF
• [3] Singh A, Singh K, Sharma A, Kaur K, Kaur K, Chadha R, et al. Recent developments in synthetic α-glucosidase inhibitors: A comprehensive review with structural and molecular insight. J Mol Structures 2023;1281:135115. Article
• [4] Pan L, Li Z, Wang Y, Zhang B, Liu G, Liu J. Network pharmacology and metabolomics study on the intervention of traditional Chinese medicine Huanglian Decoction in rats with type 2 diabetes mellitus. J Ethnopharmacol 2020;258:112842. ArticlePubMed
• [5] Hogan S, Zhang L, Li J, Sun S, Canning C, Zhou K. Antioxidant rich grape pomace extract suppresses postprandial hyperglycemia in diabetic mice by specifically inhibiting alpha-glucosidase. Nutr Metab (Lond) 2010;7:71. ArticlePubMedPMCPDF
• [6] He RJ, Yu ZH, Zhang RY, Zhang ZY. Protein tyrosine phosphatases as potential therapeutic targets. Acta Pharmacol Sin 2014;35(10):1227−46.ArticlePubMedPMCPDF
• [7] Liu G, Trevillyan JM. Protein tyrosine phosphatase 1B as a target for the treatment of impaired glucose tolerance and type II diabetes. Curr Opin Investig Drugs 2002;3(11):1608−16.PubMed
• [8] Hirata AE, Alvarez-Rojas F, Carvalheira JBC, de Oliveira Carvalho CR, Dolnikoff MS, Saad MJA. Modulation of IR/PTP1B interaction and downstream signaling in insulin sensitive tissues of MSG-rats. Life Sci 2003;73(11):1369−81.ArticlePubMed
• [9] Merz KE, Thurmond DC. Role of skeletal muscle in insulin resistance and glucose uptake. Compr Physiol 2020;10(3):785−809.ArticlePubMedPMCPDF
• [10] Wong CY, Al-Salami H, Dass CR. C2C12 cell model: its role in understanding of insulin resistance at the molecular level and pharmaceutical development at the preclinical stage. J Pharm Pharmacol 2020;72(12):1667−93.ArticlePubMedPDF
• [11] Deng D, Yan N. GLUT, SGLT, and SWEET: structural and mechanistic investigations of the glucose transporters. Protein Sci 2016;25(3):546−58.PubMedPMC
• [12] Petersen MC, Shulman GI. Mechanisms of insulin action and insulin resistance. Physiol Rev 2018;98(4):2133−223.ArticlePubMedPMC
• [13] El Khoury D, Cuda C, Luhovyy BL, Anderson GH. Beta glucan: health benefits in obesity and metabolic syndrome. J Nutr Metab 2012;2012:851362. PubMed
• [14] Leung L, Birtwhistle R, Kotecha J, Hannah S, Cuthbertson S. Anti-diabetic and hypoglycaemic effects of Momordica charantia(bitter melon): a mini review. Br J Nutr 2009;102(12):1703−8.ArticlePubMed
• [15] Stohs SJ, Miller H, Kaats GR. A review of the efficacy and safety of banaba (Lagerstroemia speciosa L.) and corosolic acid. Phytother Res 2012;26(3):317−24.ArticlePubMedPDF
• [16] Martin KR, Narang P, Franco JLM, Meurice N, MacKeiga JP. Integrating virtual and biochemical screening for protein tyrosine phosphatase inhibitor discovery. Methods 2014;65(2):219−28.ArticlePubMedPMC
• [17] EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific Opinion on the substantiation of a health claim related to Nutriose® 06 and a reduction of post-prandial glycaemic responses pursuant to Article 13 (5) of Regulation (EC) No 1924/2006. EFSA J 2014;12(10):3839. ArticlePubMedPMCPDF
• [18] Korea Institute of Oriental Medicine [Internet]. TM-MC 2.0: an enhanced chemical database of medicinal materials in Northeast Asian traditional medicine: [cited 2026 Jun 2]. Available from: https://tm-mc.kr/.
• [19] COCONUT [Internet]. COCONUT: COlleCtion of Open Natural products: [cited 2026 Jun 2]. Available from: https://coconut.naturalproducts.net/.
• [20] National Library of Medicine [Internet]. PubChem: [cited 2026 Jun 2]. Available from: https://pubchem.ncbi.nlm.nih.gov/.
• [21] STITCH [Internet]. STITCH: search tool for interactions of chemicals: [cited 2026 Jun 2]. Available from: http://stitch.embl.de/.
• [22] EMBL-EBI [Internet]: ChEMBL: [cited 2026 Jun 2]. Available from: https://www.ebi.ac.uk/chembl.
• [23] MDI Biological Laboratory, NC State University [Internet]. Comparative toxicogenomics database: [cited 2026 Jun 2]. Available from: https://ctdbase.org/.
• [24] Global Biodata Coalition, ELIXIR [Internet]. STRING: [cited 2026 Jun 2]. Available from: https://string-db.org/.
• [25] Franz M, Lopes C, Fong D, Kucera M, Bader G. [Internet] Cytoscape.js: [cited 2026 Jun 2]. Available from: https://js.cytoscape.org/.
• [26] Enrichr [Internet]: [cited 2026 Jun 2]. Available from: https://maayanlab.cloud/Enrichr/.
• [27] Baell JB, Holloway GA. New substructure filters for removal of pan assay interference compounds (PAINS) from screening libraries and for their exclusion in bioassays. J Med Chem 2010;53(7):2719−40.ArticlePubMed
• [28] Judy WV, Hari SP, Stogsdill WW, Judy JS, Naguib YMA, Passwater R. Antidiabetic activity of a standardized extract (Glucosol™) from Lagerstroemia speciosa leaves in type II diabetics: a dose-dependence study. J Ethnopharmacol 2003;87(1):115−7.PubMed
• [29] Miura T, Ueda N, Yamada K, Fukushima M, Ishida T, Kaneko T, et al. Antidiabetic effects of corosolic acid in KK-Ay diabetic mice. Biol Pharm Bull 2006;29(3):585−7.ArticlePubMed
• [30] Lankatillake C, Luo S, Flavel M, Lenon GB, Gill H, Huynh T, et al. Screening natural product extracts for potential enzyme inhibitors: protocols, and the standardisation of the usage of blanks in α-amylase, α-glucosidase and lipase assays. Plant Methods 2021;17(1):3. ArticlePubMedPMCPDF

Figure & Data

References

Citations

Citations to this article as recorded by