Utilizing Traditional Chinese Medicine Systems Pharmacology (TCMSP, https://www.tcmsp-e.com), Swiss Target Prediction (http://www.swisstargetprediction.ch), and the PubChem database (https://pubchem.ncbi.nlm.nih.gov), we identified protein targets associated with Centella asiatica. Additionally, we extracted DFU-related target proteins from the human genome database GeneCards (https://www.genecards.org) and the UniProt database (https://www.uniprot.org). Through Venn diagram analysis, we identified overlapping targets between DFU and Centella asiatica, which were subsequently considered potential therapeutic targets for the treatment of DFU using Centella asiatica.

GO and Reactome pathways enrichment analyses were performed to elucidate the core mechanisms and pathways underlying the anti-DFU properties of Centella asiatica using the STRING database (https://string-db.org) and Hiplot Website (https://hiplot.com.cn) (Li J. et al., 2022; Szklarczyk et al., 2023). A query was conducted using the STRING database to identify genes with shared targets, specifically restricting the analysis to the species “Homo sapiens.” This investigation encompassed GO categories, including biological processes, cellular components, and molecular functions, as well as Reactome pathways (Milacic et al., 2024).

We obtained mouse mononuclear macrophage cells (RAW 264.7) from Procell Company (Wuhan, China). These cells were cultured in RPMI 1640 medium with 11.1 mM glucose (Gibco, 12633012) at 37 °C in a humidified incubator with 5% CO2. The culture medium included 0.5% penicillin–streptomycin supplied by WelGENE Inc. (Korea), along with 10% fetal bovine serum from VivaCell (China).

Centella asiatica powder was supplied by Tianjiang Pharmaceutical Co., Ltd. (China). For processing, the samples should be kept at −80 °C. In addition, the active components were characterized by high-performance liquid chromatography (HPLC) analysis.

Raw 264.7 cells were tested for viability using Apexbio’s Cell Counting Kit-8 (CCK-8, K1018), as described in the protocol. The readings were made using a Synergy™ H1 microplate reader. Cells were seeded into a 96-well plate at a density of 2 × 10^4 cells per well and subsequently pre-treated with 200 μg/mL AGEs for 24 h (Khan et al., 2022), followed by the intervention with centella sinensis for 24 h. No release assay was conducted using the NO fluorescence probe kit (S0019S, Beyotime Biotechnology), and subsequent procedures were carried out strictly according to the kit instructions.

Proteins were extracted from cells using RIPA buffer with protease inhibitors (SC-364162, Santa Cruz, CA) and phosphatase inhibitors (B15001, Selleck). Protein concentration was measured using a BCA assay kit (ZJ102, Epizyme). Following this, the proteins were separated by polyacrylamide gel electrophoresis and transferred to a PVDF membrane (Millipore, Billerica). The membrane was blocked for 1 hour with TBST containing 5% milk, followed by overnight incubation at 4 °C with p-NF-κB (T55034, Abmart), p-AKT (T40067S, Abmart), AKT (T55561S, Abmart), p-AMPK (2531S, CST), AMPK (YT0216, ImmunoWay), iNOS (340668, Zen Bio), and β-Tubulin (WL01931, Wanleibio) antibodies. After three washes with TBST, an HRP-conjugated secondary antibody (Thermo Fisher) was applied. Antigen detection was performed using a chemiluminescence kit (NCM Biotech), visualized with a ChemiDoc Touch imaging system (Bio-Rad), and analyzed using ImageJ.

Total RNA was extracted using TRIzol™ reagent (Invitrogen) to obtain total RNA. The High-Capacity cDNA Reverse Transcription Kit (Invitrogen) was used to synthesize cDNA. iTaq™ Universal SYBR Green Supermix (Bio-Rad) was used for qRT-PCR, and the data were analyzed with CFX™96 Manager Software (Bio-Rad). To measure mRNA expression levels, we normalized them to actin.

After the tissues were fixed with paraformaldehyde for a duration of 16–24 h at 4 °C, 5-μm sections were subsequently prepared. The sections were subjected to deparaffinization and antigen retrieval utilizing sodium citrate, followed by a paraffin embedding step to facilitate the subsequent processing of the specimens. A blocking buffer was then applied to the sections. At 4 °C overnight incubated with antibodies specific to CD31 (R10021, Zen Bio), PCNA (27210, SAB), Cleaved-caspase3 (9661, CST), VEGF-A (WL00009b, WanLeiBio), α-SMA (55135-I-AP, Proteintech). The procedure was conducted utilizing secondary antibodies conjugated to Alexa Fluor 488 and Alexa Fluor 555 (Thermo Fisher). DAPI was used as a staining agent in the mounting medium (C1005, Beyotime). Additionally, Masson’s trichrome staining was conducted following the manufacturer’s protocol utilizing a kit (G1346, Solarbio). The quantification of collagen in the stained skin sections was performed using ImageJ software.

Male C57BL/6 J mice were obtained from Eye Biotech (Suzhou, China) and were maintained on a 12-h light/dark cycle with unrestricted access to standard chow and water. Over a period of 4 weeks, the mice were subjected to a high-fat diet (HFD, D12492; Research Diets) and received daily intraperitoneal injections of 35 mg/kg streptozotocin (STZ), which was dissolved in citrate buffer (pH 4.2–4.5), for a duration of five consecutive days. Blood samples collected from the tail were utilized to assess glucose levels. Mice exhibiting blood glucose concentrations exceeding 16.7 mmol/L after 1 week were classified as diabetic (Rui et al., 2024).

In accordance with the outlined methodology, an experimental model for wound healing was established (Xiao et al., 2024). Randomization: Male C57BL/6 J mice (n = 30) were randomly assigned to 3 groups (Centella asiatica-200 μg/cm² group, Centella asiatica-300 μg/cm² group, and Vehicle group) to ensure equal distribution of body weight and baseline glucose levels. Animals were anesthetized using isoflurane (RDW), and a 6 mm cylindrical biopsy tool (Health Link, Florida) was employed to create full-thickness wounds on both sides of the dorsum. Postoperatively, digital images of the wounds were captured every 2 days. Centella asiatica was formulated with Vaseline and administered to the wound edge at concentrations of 200ug/cm² and 300ug/cm², respectively, after each wound image acquisition. A standardized calibration protocol was employed to measure wound areas, and the wound closure rate was subsequently calculated using ImageJ software (National Institutes of Health, Bethesda, MD). To investigate the mechanisms of wound healing, mRNA and protein analyses were conducted on euthanized mice. Wound tissues were collected using 8 mm skin biopsy punches. The study is reported in accordance with ARRIVE guidelines (https://arriveguidelines.org).

In order to analyze the supplementary data, we used Prism 9.3 (GraphPad, San Diego, CA) as a statistical analysis program. Multiple groups were compared using one-way ANOVA with post hoc Tukey tests, while Student's t-tests were used to compare two groups. p < 0.05 was considered statistically significant.

We employed a systematic pharmacology approach to elucidate the potential mechanisms underlying the therapeutic effects of Centella asiatica on diabetic patients with DFU, as delineated in our workflow (Figure 1). Active phytochemicals from Centella asiatica were identified using the TCMSP database and extant literature, comprising three principal components: Cynarine, Quercetin, and Kaempferol. Target prediction utilized the PubChem and SwissTargetPrediction databases, yielding 106 unique targets following the removal of duplicates. From GeneCards and UniProt databases, we extracted 2,524 DFU-related targets. We constructed a Venn diagram to intersect the targets from both Centella asiatica and DFU, leading to the identification of 31 common targets posited as key mediators in Centella asiatica’s therapeutic strategy against DFU. More importantly, these prevalent targets encompass a diverse array of inflammation-associated regulators, including the matrix metalloproteinase family and estrogen receptors (Supplementary Table S1).

To further elucidate the mechanism by which Centella asiatica affects DFU, we conducted a GO enrichment analysis of 31 putative Centella asiatica targets. The analysis identified the terms across the biological process (BP), cellular component (CC), and molecular function (MF) categories associated with Centella asiatica. The BP category indicated that the active compounds predominantly influence cellular response to chemical stress, response to oxidative stress, protein autophosphorylation, protein kinase B signaling, and response to amyloid-beta (Figure 2A). As for MF, the inferred targets are chiefly implicated in metalloendopeptidase activity, protein serine/threonine kinase activity, protein serine/threonine/tyrosine kinase activity, protein tyrosine kinase activity, and serine-type endopeptidase activity (Figure 2B). The CC category comprises elements such as the caveola, cell leading edge, chromosome, telomeric region, membrane microdomain, membrane raft, and protein kinase complex (Figure 2C). Collectively, these results indicate a strong association with the regulation of kinase activities, oxidative stress, and inflammatory pathways.

A subsequent Reactome pathways enrichment analysis was conducted to delineate the role of Centella asiatica’s functional pathways in addressing DFU targets (Figure 2D). Pathways with the highest significance included those related to extra-nuclear estrogen signaling, PI3K/AKT signaling, tyrosine kinases, collagen degradation, activation of matrix metalloproteinases, and growth factor receptors signaling in diabetic complications. In line with the GO findings, these pathways are connected to inflammatory signaling. The GO and Reactome pathways analyses enrichment together suggest that Centella asiatica’s mechanism chiefly constitutes anti-inflammatory effects by modulating such signals.

To enhance the identification of the principal active compounds in Centella sinensis, we characterized key active substances Cynarine, Quercetin, and Kaempferol utilizing HPLC (Figure 3A). Subsequently, the anti-inflammatory effect on RAW264.7 macrophages was further studied in vitro. Initially, a cell viability assay established an optimal Centella asiatica concentration of 0.75 mg/mL for intervention (Figure 3B). Using AGEs to mimic the diabetic microenvironment in vitro (Xiao et al., 2024). The NO production by macrophages constitutes critical inflammatory events, which are principally driven by iNOS (Sumi and Ignarro, 2004; Yang et al., 2022). Through investigating AGEs-induced NO release, we observed a significant attenuation with Centella asiatica treatment (Figure 3C). Correspondingly, immunofluorescence analyses substantiated these findings (Figures 3D,E). Consequently, our preliminary findings have suggested that the intervention of Centella asiatica can significantly suppress macrophage-mediated inflammation, potentially serving as a critical factor in the diabetic wound healing process.

To further validate the regulatory effect of Centella asiatica on macrophage inflammation levels, Western blot analysis was subsequently employed. The results indicated that the AGEs-induced expression of iNOS had been downregulated following intervention with Centella asiatica (p < 0.05, Figures 4A,B). Moreover, qRT-PCR analysis demonstrated that Centella asiatica suppressed the expression of pro-inflammatory cytokines IL-1β, IL-6, and TNF-α, markedly inhibiting their mRNA levels (p < 0.05, Figures 4C–E), while upregulated anti-inflammatory cytokines IL-4 and IL-10 (p < 0.05, Figures 4F,G). In sum, our results support the anti-inflammatory attributes of Centella asiatica.

NF-κB played a pivotal role in mediating the inflammatory response by initiating the transcription of inflammatory cytokines, such as TNF-α, IL-6, and IL-1β. These responses could be notably augmented by AGEs (Dong et al., 2016). In this context, we investigated the impact of Centella asiatica on the phosphorylation of NF-κB, confirming its efficacy in mitigating NF-κB phosphorylation (p < 0.05, Figures 4H,I). Furthermore, to elucidate the mechanism underlying the suppression of AGEs-mediated inflammatory signaling, we assessed the influence of Centella asiatica on the phosphorylation of pivotal proteins within the AKT/MAPK signalling pathway, a regulator of the pro-inflammatory response in macrophages. The results indicated that Centella asiatica effectively attenuates the AGEs-induced activation of the AKT/MAPK pathway (p < 0.05, Figures 4J–M). Collectively, these outcomes have suggested that Centella asiatica intervention substantially diminishes the upregulation of inflammatory cytokines induced by AGEs, potentially due to the inhibition of the NF-κB pathway mediated by the AKT/AMPK signaling cascade.

The methodology for assessing the healing progression of diabetic wounds was depicted (Figure 5A). Both digital imaging and model photography were employed to illustrate the stages of wound healing in diabetic murine models after topical application of different concentrations of Centella asiatica on days 0,3,5,7 and 9 (Figures 5B,C). A comparative analysis of wound closure rates indicated that Centella asiatica significantly enhanced wound healing efficiency in diabetic wounds compared to the Vehicle group (p < 0.05, Figure 5D). Further quantitative evaluations of wound dimensions and the positioning of horizontal black lines in H&E-stained images highlighted that Centella asiatica’s influence on wound recovery and re-epithelialization markedly outperformed that of the Vehicle, with both dosages demonstrating therapeutic benefits, and optimal efficacy at 200 μg/cm² (p < 0.05, Figures 5 E–G). Additionally, Masson’s trichrome staining was utilized to measure collagen deposition, revealing that treatment with Centella asiatica significantly promoted collagen synthesis compared to the Vehicle group (p < 0.05, Figures 5H, I). These findings suggest that Centella asiatica effectively accelerates the healing of diabetic wounds.

Building on the results that Centella asiatica could enhance the healing of diabetic wounds, we examined its impact on apoptosis and granulation tissue growth in diabetic wound healing. Immunofluorescence analysis indicated that treatment with Centella asiatica remarkably reduced cellular apoptosis at the edges of diabetic wounds (p < 0.05, Figures 6A,B). Proliferation assays further demonstrated that exposure to Centella asiatica significantly enhanced cell proliferation, underlining its potential role as a therapeutic agent for promoting wound healing in diabetic patients (p < 0.05, Figures 6C,D). To enhance our understanding of the effects of Centella asiatica on angiogenesis, we utilized dual staining techniques for CD31 and α-SMA, which serve as markers for endothelial and smooth muscle cells, respectively, to assess angiogenic activity. Results from imaging and statistical analyses revealed that Centella asiatica treatment facilitated increased angiogenesis, indicating its potential in enhancing vascular growth and tissue repair (p < 0.05, Figures 6E,F). Additionally, we evaluated the secretion of VEGF-A following Centella asiatica treatment and observed a significant increase in its levels (p < 0.05, Figures 6G,H). These findings have suggested that Centella asiatica accelerated diabetic wound healing by enhancing granulation tissue growth.