Salidroside alleviates TNF-α-induced endothelial inflammatory injury by modulating NF-κB/NLRP3 inflammasome-related signaling: an integrated network pharmacology and experimental study

Introduction

Kawasaki disease (KD) is an acute febrile vasculitis primarily affecting infants and young children (1). It is now recognized as one of the leading causes of acquired heart disease in children, especially in developed countries and many regions of Asia (2). Coronary artery lesions (CALs) are the most clinically important complication of KD and are closely related to long-term cardiovascular risk (1). Although standard treatment with intravenous immunoglobulin (IVIG) and aspirin has greatly improved outcomes, approximately 10–20% of patients remain resistant to IVIG and still develop coronary artery injury (3). Therefore, identifying new therapeutic targets and adjunctive treatment strategies remains important (4).

Progress in understanding the pathogenesis of KD has provided a theoretical basis for exploring new intervention strategies. Although the exact cause of KD remains unclear, current evidence suggests that KD is an inflammatory disease characterized by dysregulation of both innate and adaptive immunity (5). Endothelial injury is a key event in KD-associated vascular pathology (6). Several studies have explored various aspects of vascular endothelial cell injury and dysfunction in KD, with potential mechanisms including apoptosis, autophagy, pyroptosis, anti-endothelial cell autoantibodies, and oxidative stress (7-10). However, definitive conclusions and optimal treatment strategies remain limited.

Increasing evidence indicates that the NOD-like receptor family pyrin domain containing 3 (NLRP3) inflammasome plays an important role in KD vasculitis (11). The NLRP3 inflammasome is an intracellular multiprotein complex composed of NLRP3, apoptosis-associated speck-like protein containing a CARD (ASC), and Caspase-1 (6). Activated Caspase-1 cleaves gasdermin D (GSDMD) to generate the pore-forming N-terminal fragment and also processes pro-interleukin (IL)-1β and pro-IL-18 into their mature forms, thereby amplifying systemic inflammatory responses (12). In KD patients and experimental models, excessive activation of NLRP3 has been associated with endothelial injury, leukocyte infiltration, and progression of cardiovascular inflammation, and is considered an important driver of KD and CALs (13).

Salidroside (SAL) is a natural compound extracted from Rhodiola rosea (14). It has multiple biological effects, including anti-inflammatory, anti-apoptotic, anti-hypoxic, antioxidant, and immunomodulatory effects in a variety of cardiovascular and inflammatory diseases (14). Previous studies have shown that SAL exerts protective effects in a variety of cardiovascular and inflammatory disease models (15). It has also been reported that SAL can inhibit NLRP3 inflammasome assembly and suppress the release of IL-1β and IL-18 (16,17). In different disease settings, SAL has been shown to regulate NLRP3 inflammasome activation through pathways involving AMP-activated protein kinase (AMPK), Sirt1/ protein kinase B (Akt)/nuclear factor erythroid 2-related factor 2 (Nrf2), and Toll-like receptor 4 (TLR4)/nuclear factor kappa B (NF-κB) signaling (18-20). However, whether SAL can protect vascular endothelial cells under KD-related inflammatory conditions and whether this effect is associated with inhibition of NLRP3 inflammasome activation remain unclear.

Network pharmacology provides a systems-level approach for predicting drug-disease interactions and identifying potential multi-target mechanisms of natural compounds (21). Molecular docking is a drug design method based on receptor characteristics and receptor-drug molecule interactions, simulating the binding mode between drugs and target proteins (22). RNA sequencing (RNA-seq) is a high-throughput sequencing technique used to analyze transcript abundance, revealing genome-wide transcriptional changes after drug intervention (23). This integrated design is therefore well suited to exploring how SAL regulates endothelial inflammatory injury relevant to KD.

In this study, we employed an integrated research strategy combining network pharmacology, RNA-seq, molecular docking, and in vitro experiments to investigate the protective effects of SAL in a tumor necrosis factor-α (TNF-α)-induced human coronary artery endothelial cells (HCAECs) inflammatory injury model. We aimed to determine whether SAL alleviates endothelial inflammatory injury and provide a theoretical basis for new therapeutic strategies relevant to KD-related vascular inflammation. We present this article in accordance with the MDAR reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0388/rc).

Methods

Network pharmacology analysis

Compound target collection

Potential targets of SAL were predicted using reverse virtual screening and structure-based target prediction via the SwissTargetPrediction platform (http://www.swisstargetprediction.ch). No additional high-probability cutoff was applied; all predicted targets with a probability score >0 were retained for exploratory screening, whereas targets with a probability score of 0 were excluded.

Disease target collection

KD-related targets were retrieved from the GeneCards (http://www.genecards.org/) and DisGeNET (https://www.disgenet.org/) databases using the search term “Kawasaki disease” with the species limited to Homo sapiens. For GeneCards, targets with a relevance score ≥1.0 were retained. For DisGeNET, targets with a gene-disease association score >0 were included. Additional searches were performed in the Online Mendelian Inheritance in Man (OMIM) and Therapeutic Target Database (TTD) databases, but no additional KD targets were found. The retrieved targets from the two databases were combined and duplicates were removed. The overlapping targets of SAL and KD were identified using Venny 2.1 and imported into Cytoscape to construct a compound-target-disease network. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Functional enrichment analysis

The common targets of SAL and Kawasaki disease were imported into DAVID Bioinformatics Resources 6.8 for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. GO analysis included biological process (BP), cellular component (CC), and molecular function (MF). A threshold of P<0.05 was applied for screening. The top 10 enriched terms were displayed using bar graphs and bubble charts. KEGG Mapper was used to plot signaling pathway diagrams.

Protein-protein interaction (PPI) network construction and key target screening

The common targets were input into the STRING database to construct a PPI network, with the species set to Homo sapiens and default parameters as the threshold. The TSV file obtained from STRING was imported into Cytoscape software. Topological analysis of the PPI network was performed using three centrality algorithms: degree centrality, betweenness centrality, and closeness centrality.

Molecular docking analysis

The crystal structures of target proteins were obtained from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB) database. The Schrödinger software was used for protein preparation (Protein Preparation Wizard module). The SiteMap module and Receptor Grid Generation module in Schrödinger were used to identify the active sites of the proteins. The 2D structure-data file (SDF) structure file of SAL was downloaded from the PubChem database and preprocessed using the LigPrep module. Extra Precision (XP) molecular docking and Molecular Mechanics/Generalized Born Surface Area (MM-GBSA) analysis were performed between the processed ligand compound and the active sites of the ten proteins. An XP Gscore <−6 was considered as stable binding, and an MM-GBSA dG Bind <−30 kcal/mol was considered as low binding free energy indicating stable binding.

Cell culture and stimulation

HCAECs (Shanghai Fuheng Biotechnology, Shanghai, China) were cultured in RPMI-1640 medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37 ℃ in a 5% CO₂ incubator. HCAECs were stimulated with TNF-α to establish an endothelial inflammatory injury model relevant to KD-related endothelial inflammation. The experimental groups were set: control group, TNF-α group, and SAL + TNF-α group. HCAECs were pretreated with SAL and then treated with TNF-α for corresponding experiments.

Cell Counting Kit-8 (CCK-8) assay

Cell viability of HCAECs under different treatments was detected using the CCK-8 kit (DOJINDO, Kumamoto, Japan). HCAECs were seeded in 96-well plates at a density of 1×10⁴ cells/well. HCAECs were treated with different concentrations of TNF-α (MCE, Monmouth Junction, NJ, USA), such as 0, 25, 50, 100, and 500 ng/mL, to establish a TNF-α-induced endothelial inflammatory injury model (24). Using this model, HCAECs were treated with different concentrations of SAL (Yuanye, Shanghai, China), such as 0, 50, 100, 500, and 1,000 µmol/L, to explore its optimal working concentration. After incubation with SAL for 24 hours, TNF-α was added and the cells were cultured for another 48 hours. Subsequently, 10 µL of CCK-8 solution was added to each well, followed by incubation at 37 ℃ for 1 hour. Finally, the absorbance at 450 nm was measured using a microplate reader.

Wound healing assay

HCAECs were seeded in 6-well plates and cultured to 90% confluence. A sterile 200 µL pipette tip was used to create a linear scratch across the cell monolayer. After washing with phosphate-buffered saline (PBS) to remove detached cells, the cells were incubated in serum-free medium with different treatments: control, TNF-α, and SAL + TNF-α. Images of the scratch area were captured at 0 and 24 hours using an inverted microscope.

Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) assay

Apoptosis was detected using the Annexin V-FITC/PI Apoptosis Detection Kit (Yeasen, Shanghai, China). After treatment, HCAECs were harvested, washed twice with cold PBS, and resuspended in 1× binding buffer. Cells were then stained with 5 µL Annexin V-FITC and 10 µL PI for 15 min at room temperature in the dark. Apoptosis was analyzed within 1 hour using a flow cytometer. Total apoptosis was calculated as the sum of early and late apoptosis.

RNA-seq

RNA-seq samples were prepared from HCAECs stimulated with 100 ng/mL TNF-α for 24 hours, with or without pretreatment with 100 µmol/L SAL for 12 hours. RNA-seq was performed with three biological replicates per group. RNA extraction, library construction, sequencing, and upstream data processing were performed by Novogene (Beijing, China). Differentially expressed genes (DEGs) were analyzed using the DESeq2 package in R software with raw counts. Because the RNA-seq analysis was used as an exploratory screening step to identify candidate genes and pathways for subsequent validation, genes with P≤0.05 and |log₂(fold change)| ≥0.5 were selected as candidate DEGs. GO and KEGG enrichment analyses of the DEGs were performed using Metascape database, and bubble charts were drawn to visualize the enriched signaling pathways.

Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

Total RNA was extracted from HCAECs using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and reverse-transcribed into complementary DNA (cDNA) using the PrimeScript RT Reagent Kit (Takara, Shiga, Japan). qPCR reactions were performed using the TB Green Premix Ex Taq kit (Takara, Shiga, Japan). GAPDH was used as the internal reference gene. The relative expression levels were calculated using the 2^−ΔΔCt method, and all data were normalized to the control group. Primer sequences used in this study are listed in Table 1.

Western blot (WB)

Cells were lysed using radioimmunoprecipitation assay (RIPA) lysis buffer containing protease inhibitors. Protein concentration was determined using the BCA protein assay kit (Beyotime, Shanghai, China), and proteins were denatured at 95 ℃ for 10 minutes. Protein samples were separated using 12.5% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) gels and then transferred onto polyvinylidene fluoride (PVDF) membranes (Thermo Fisher Scientific, Waltham, MA, USA). The membranes were blocked with 5% skim milk or 5% bovine serum albumin (BSA) at room temperature for 2 hours. Subsequently, the membranes were incubated with primary antibodies at 4 ℃ overnight. Primary antibodies included NF-kB p65 (1:1,000 dilution, 8242S, CST, Danvers, MA, USA), phospho-NF-kB p65 (1:1,000 dilution, 3031SS, CST), IL-6 (1:1,000 dilution, 12153S, CST), NLRP3 (1:1,000 dilution, 15101S, CST), GSDMD (1:1,000 dilution, 39754T, CST), Caspase-1 (1:2,000 dilution, 22915-1-AP, Proteintech, Wuhan, China), and α-tubulin (1:1,000 dilution, 5873S, CST). The membranes were then incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit immunoglobulin G (IgG) antibody (1:10,000 dilution, SA00001-2, Proteintech, China) and HRP-conjugated goat anti-mouse IgG antibody (1:10,000 dilution, SA00001-1, Proteintech) at room temperature for 1 hour. Finally, specific protein bands were detected using an enhanced chemiluminescence immunoblotting substrate kit. Semi-quantitative analysis was performed using ImageJ software.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 9.0 software. Data are presented as mean ± standard deviation from at least three independent experiments. For data following a normal distribution, comparisons between two groups were analyzed using independent Student’s t-test, and comparisons among multiple groups were assessed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. For non-normally distributed data, the Kruskal-Wallis test was applied. A value of P<0.05 was considered statistically significant (*P<0.05, **P<0.01, ***P<0.001). All quantitative experiments were independently repeated at least three times as biological replicates. For assays performed in multi-well plates, technical replicates were included within each experiment where applicable.

Results

Information and potential targets of SAL and KD

SAL [2-(4-hydroxyphenyl)ethyl, CAS 10338-51-9] has the molecular structure shown in Figure 1A. A total of 226 targets of SAL were identified and 2,686 KD targets were obtained (Figure 1B). The intersection between SAL targets and KD targets yielded 56 overlapping candidate targets, which were used for subsequent exploratory analysis (Figure 1C). A compound-target-disease network was constructed to visualize the complex relationships among SAL, its targets, and KD (Figure 1D).

Figure 1 Identification of potential therapeutic targets of SAL in KD. (A) Chemical structure of SAL. (B) Predicted SAL-related targets. (C) Venn diagram showing the overlap between SAL-related targets and KD-related targets. (D) Compound-target-disease network illustrating the relationships among SAL, the overlapping targets, and KD. KD, Kawasaki disease; SAL, salidroside.

GO and KEGG enrichment analyses of common targets

To clarify the functional characteristics of the 56 common targets, GO and KEGG enrichment analyses were performed (P<0.05). The top 10 enriched terms in each GO category are displayed in Figure 2A. In the BP category, the main enriched terms included “positive regulation of cell population proliferation”, “response to hypoxia”, “positive regulation of mitogen-activated protein kinase (MAPK) cascade”, and “positive regulation of cell migration”. CC terms were mainly enriched in “cytoplasm”, “nucleus”, “cytosol”, and “extracellular space”. MF terms were primarily concentrated in “zinc ion binding”, “endopeptidase activity”, “metalloendopeptidase activity”, and “protein-containing complex binding”. KEGG pathway enrichment identified 76 significant pathways (P<0.05). The immune- and inflammation-related pathways are shown in Figure 2B, including the phosphatidylinositol 3-kinase​(PI3K)-protein kinase B (Akt) signaling pathway, MAPK signaling pathway, IL-17 signaling pathway, TNF signaling pathway, necroptosis, and apoptosis.

Figure 2 Functional enrichment analysis of overlapping targets between SAL and KD. (A) GO enrichment analysis of the 56 overlapping targets, including BP, CC, and MF categories. (B) KEGG enrichment analysis showing the major pathways associated with the overlapping targets. Only significantly enriched terms are shown (P<0.05). AGE-RAGE, advanced glycation end products-receptor for advanced glycation end products; Akt, protein kinase B; ATP, adenosine triphosphate; BP, biological process; CC, cellular component; GO, Gene Ontology; IL, interleukin; KD, Kawasaki disease; KEGG, Kyoto Encyclopedia of Genes and Genomes; MAPK, mitogen-activated protein kinase; MF, molecular function; PI3K, phosphoinositide 3-kinase; SAL, salidroside; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor.

PPI network and key targets

The 56 common targets were uploaded to the STRING database to construct a PPI network (Figure 3A). Topological analysis was performed using degree centrality, betweenness centrality, and closeness centrality. The top 20 targets with their main topological parameters are listed in Table 2. Among these, matrix metalloproteinase (MMP)-9, heat shock protein (HSP) 90AA1, MAPK1, MMP-3, and fibroblast growth factor 2 (FGF2) were used as preliminary candidates for subsequent molecular docking analysis.

Figure 3 PPI network and molecular docking of SAL with key targets. (A) PPI network constructed from the 56 overlapping targets. (B) Representative molecular docking models of SAL with MMP3, HSP90AA1, MAPK1, MMP9, and FGF2. MMP, matrix metalloproteinase; PPI, protein-protein interaction; SAL, salidroside.

Molecular docking

Molecular docking was performed between SAL and the core target proteins. Comprehensive analysis of the XP docking and MM-GBSA results showed that the docking scores of SAL with MMP3, HSP90AA1, MAPK1, MMP9, and FGF2 were −8.301, −7.956, −7.936, −7.775, and −5.576, respectively, and the corresponding MM-GBSA results were −48.47, −43.90, −30.11, −30.95, and −31.39 kcal/mol, respectively. These low docking scores and binding free energies indicate sufficiently stable binding between SAL and these five proteins. The molecular docking structures and energies are shown in Figure 3B and Table 3.

Optimal concentrations of TNF-α and SAL in HCAECs

TNF-α is one of the most important pro-inflammatory factors in KD. To establish a TNF-α-induced HCAEC inflammatory injury model, HCAECs were treated with different concentrations of TNF-α (0, 25, 50, 100, 500 ng/mL) for 48 hours. CCK-8 assay showed that 100 and 500 ng/mL TNF-α significantly inhibited cell viability compared with the control group (Figure 4A). Therefore, 100 ng/mL was selected as the optimal working concentration for TNF-α.

Figure 4 SAL improves cell viability and attenuates excessive migration and inflammatory cell death in HCAECs. (A) CCK-8 assay showing the effects of different concentrations of TNF-α on the viability of HCAECs after 48 hours of stimulation. (B) CCK-8 assay showing the effects of SAL pretreatment followed by TNF-α stimulation on HCAECs viability. (C) Representative wound-healing images and quantitative analysis of wound closure in control, TNF-α, and SAL + TNF-α groups. Images were captured at 0 and 24 hours. Scale bar =100 µm. (D) Representative Annexin V-FITC/PI flow cytometry plots and quantitative analysis of apoptotic/necrotic cells. Data are presented as mean ± SD, n=3. *, P<0.05; **, P<0.01. CCK-8, Cell Counting Kit-8; FITC/PI, fluorescein isothiocyanate/propidium iodide; HCAECs, human coronary artery endothelial cells; ns, not significant; SAL, salidroside; SD, standard deviation; TNF, tumor necrosis factor.

To determine the appropriate intervention concentration of SAL, HCAECs were pretreated with various concentrations of SAL (0, 50, 100, 500, 1,000 µmol/L) for 12 hours followed by TNF-α (100 ng/mL) stimulation for 48 hours. CCK-8 results showed that 100, 500, and 1,000 µmol/L SAL significantly restored cell viability compared with TNF-α alone (Figure 4B). Considering both cell viability and drug safety, 100 µmol/L SAL was selected for subsequent experiments because it was the lowest concentration that significantly improved HCAECs viability after TNF-α stimulation.

SAL inhibits TNF-α-induced endothelial excessive migration and apoptosis

A wound healing assay was performed to evaluate the effect of SAL on endothelial cell migration under inflammatory conditions. As shown in Figure 4C, TNF-α treatment significantly increased the wound closure rate compared with the control group, indicating that TNF-α promotes excessive migration of HCAECs. Pretreatment with SAL significantly reduced this effect, demonstrating that SAL inhibits TNF-α-induced excessive migration.

Annexin V-FITC/PI flow cytometry was used to detect apoptosis. TNF-α treatment significantly increased the proportions of early apoptotic (Annexin V+/PI−) and late apoptotic (Annexin V+/PI+) cells, while SAL pretreatment markedly attenuated total apoptosis and necrosis (Figure 4D). These results indicate that SAL protects HCAECs from TNF-α-induced apoptosis.

DEGs and enriched pathways after SAL treatment

To further explore the mechanism and targets of the protective effect of SAL on HCAECs, RNA-seq was performed on HCAECs from the TNF-α group and the SAL + TNF-α group. A total of 480 candidate DEGs were identified, including 113 upregulated and 367 downregulated genes (Figure 5A). GO enrichment analysis showed that the main BP terms included “cell activation”, “regulation of cell activation”, “vasculature development”, and “extracellular matrix organization”. CC terms were mainly “extracellular matrix”, “collagen-containing extracellular matrix”, and “basement membrane”. MF terms included “integrin binding”, “cytokine receptor binding”, and “metallopeptidase activity” (Figure 5B).

Figure 5 RNA-seq reveals that SAL regulates a broad inflammatory network in HCAECs. (A) Volcano plot of DEGs identified by RNA-seq in TNF-α-stimulated HCAECs with or without SAL pretreatment. (B) GO enrichment analysis of DEGs. (C) KEGG pathway enrichment analysis of DEGs. DEGs were defined as genes with P≤0.05 and |log₂(fold change)| ≥0.5. n=3. DEGs, differentially expressed genes; GO, Gene Ontology; HCAECs, human coronary artery endothelial cells; IL-17, interleukin-17; KEGG, Kyoto Encyclopedia of Genes and Genomes; RNA-seq, RNA sequencing; SAL, salidroside; TNF, tumor necrosis factor.

KEGG pathway analysis further showed that DEGs were mainly enriched in immune and inflammation-related pathways, including the IL-17 signaling pathway, TNF signaling pathway, cytokine-cytokine receptor interaction, and focal adhesion (Figure 5C). Taken together, these findings suggest that SAL may regulate a broad inflammatory network involved in endothelial injury. Because both network pharmacology and RNA-seq pointed to inflammatory pathway regulation, and RNA-seq showed changes in NLRP3 and IL-1β, we next focused on validating inflammatory mediators and NF-κB/NLRP3-related molecules at the cellular level.

SAL alleviates inflammation and reduces NF-κB/NLRP3 inflammasome-related molecular changes in HCAECs

Next, we tested whether SAL plays a direct role in protecting HCAECs. Compared to the control group, RT-qPCR results showed that TNF-α significantly increased the messenger RNA (mRNA) levels of IL-6, interleukin-17 receptor A (IL-17RA), MMP-9, and intercellular adhesion molecule-1 (ICAM-1) compared with the control group, while SAL pretreatment significantly reduced these levels (P<0.05) (Figure 6A). Similarly, western blot results showed that TNF-α increased the phospho-NF-κB p65, IL-6 protein expression, which were reversed by SAL pretreatment (P<0.05) (Figure 6B). Similarly, compared with the control group, TNF-α stimulation significantly increased the mRNA levels of IL-1β, NLRP3 (Figure 6C), as well as the protein expression of NLRP3, Caspase-1, and GSDMD, whereas SAL pretreatment markedly reduced these changes (P<0.05) (Figure 6D). The full western blot membranes and the corresponding regions selected for quantitative analysis are shown in Figure S1. These findings suggest that SAL suppresses TNF-α-induced inflammatory signaling and reduces NF-κB/NLRP3 inflammasome-related molecular changes in HCAECs.

Figure 6 SAL suppresses inflammatory mediator expression and reduces NF-κB/NLRP3 inflammasome-related molecular changes in TNF-α-stimulated HCAECs. (A) RT-qPCR analysis of IL-6, IL-17RA, MMP-9, and ICAM-1 mRNA expression in control, TNF-α, and SAL + TNF-α groups. (B) WB analysis and quantification of phospho-NF-κB p65 and IL-6 protein expression. (C) RT-qPCR analysis of IL-1β and NLRP3 mRNA expression. (D) WB analysis and quantification of NLRP3, GSDMD, and Caspase-1 protein expression. HCAECs were treated with TNF-α in the presence or absence of SAL under the conditions described in Methods. Data are presented as mean ± SD, n=3. *, P<0.05; **, P<0.01; ***, P<0.001. ns, not significant. GSDMD, gasdermin D; HCAECs, human coronary artery endothelial cells; ICAM-1, intercellular adhesion molecule-1; ICAM-1, intercellular adhesion molecule-1; IL, interleukin; IL-17RA, interleukin-17 receptor A; IL-17RA, interleukin-17 receptor A; MMP, matrix metalloproteinase; NF-κB, nuclear factor kappa B; NLRP3, NOD-like receptor family pyrin domain containing 3; ns, not significant; RT-qPCR, Reverse transcription quantitative polymerase chain reaction; SAL, salidroside; SD, standard deviation; TNF, tumor necrosis factor; WB, western blot.

Discussion

KD is an acute febrile illness and a major cause of acquired heart disease in children. Without timely treatment, approximately 25% of patients develop CALs (1). Therefore, exploring effective intervention targets and safe, efficient treatment strategies is crucial for improving the prognosis of children with KD (3). Currently, traditional Chinese medicines containing SAL as a main component are widely used in clinical practice for treating coronary heart disease. For example, Rhodiola capsules are used to treat angina pectoris and improve cardiac function, with a good safety profile (25).

In this study, we integrated network pharmacology predictions, transcriptomic analysis, and in vitro experiments to investigate the protective effects of SAL in TNF-α-induced endothelial inflammatory injury relevant to KD. The results showed that SAL restored endothelial viability, reduced inflammatory mediator expression, inhibited excessive migration and inflammatory cell death, and reduced NF-κB/NLRP3 inflammasome-related molecular changes n in HCAECs. Together, these findings support SAL as a multi-target anti-inflammatory compound with potential relevance to KD-related vascular injury.

Through network pharmacology, we predicted 56 potential targets of SAL against KD, constructed a PPI network, and identified core targets such as MMP-9, HSP90AA1, MAPK1, and FGF2. And molecular docking suggested possible binding interactions between SAL and selected candidate proteins. In addition, we found that pathway enrichment results highlighted several inflammation-related pathways, especially IL-17 signaling, TNF signaling. These results suggested that SAL may affect multiple inflammatory and vascular-remodeling pathways. However, the network pharmacology analysis was used mainly as an exploratory pathway-level screening tool. Therefore, we integrated the network enrichment results with RNA-seq findings and the biological context of TNF-α-induced endothelial inflammation. Both network pharmacology and RNA-seq highlighted inflammatory pathways, especially TNF and IL-17 signaling, while RNA-seq also showed changes in NLRP3 and IL-1β. However, these results are used as exploratory support. Therefore, we further performed in vitro validation in TNF-α-stimulated HCAECs. TNF-α is one of the major pro-inflammatory cytokines implicated in KD and plays an important role in endothelial activation and vascular injury. Under TNF-α stimulation, endothelial cells produce multiple inflammatory mediators and undergo functional changes that contribute directly to vascular injury (6). This model has been widely used in studies of KD endothelial injury. This model allowed us to examine the direct response of coronary endothelial cells to inflammatory stress. However, this model represents only one aspect of KD-related vascular injury and cannot recapitulate the full disease process, including systemic immune activation, inflammatory-cell infiltration, coronary arteritis, vascular remodeling, and CAL formation. Therefore, the findings should be interpreted as endothelial cell-based mechanistic evidence rather than direct evidence of therapeutic efficacy in KD (10). In our study, the concentration of TNF-α was selected based on the dose-response CCK-8 results and its ability to induce a reproducible endothelial inflammatory injury phenotype. The concentration of 100 µmol/L SAL was selected as the lowest concentration that significantly improved cell viability in the dose-response assay.

Our study found that TNF-α stimulation enhanced the migration ability of HCAECs, and SAL inhibited this change. Under inflammatory conditions, inflammatory mediators and cytokines activate multiple signaling pathways within endothelial cells, such as NF-κB and MAPK, inducing cytoskeletal rearrangement and extracellular matrix remodeling, thereby enhancing endothelial cell migration (11). However, excessive migration can lead to structural disorganization of the endothelial layer and pathological vascular remodeling. In addition, flow cytometry results showed that SAL reduced TNF-α-induced apoptosis of HCAECs. These results suggest that SAL helps maintain endothelial cell number and barrier function by inhibiting TNF-α-induced abnormal migration and apoptosis.

Our cell experiments showed that SAL significantly decreased the transcription and protein expression of downstream effector molecules such as IL-1β, IL-6, IL-17RA, ICAM-1 and MMP-9. These findings are consistent with the bioinformatics prediction that SAL regulates a broader inflammatory network. TNF-α increased phosphorylation of NF-κB p65, whereas SAL significantly suppressed this change, suggesting that SAL inhibits upstream inflammatory priming. NF-κB is a central transcription factor in inflammation, immunity, and stress responses, and plays a critical role in the pathogenesis of KD (11). Similarly, Lei et al discovered that SAL may reduce apoptosis in pulmonary artery endothelial cells by inhibiting NF-κB and activating the Nrf2/heme oxygenase (HO)-1 pathway (20). Thus, inhibition of the NF-κB pathway appears to be a core upstream mechanism by which SAL exerts its anti-inflammatory and endothelial-protective effects, providing a key rationale for its potential as a therapeutic agent targeting inflammatory pathways.

Activation of the NF-κB pathway is not only a key driver of inflammatory mediator expression, but also the classical priming signal required for full activation of the NLRP3 inflammasome (26). In our study, TNF-α stimulation significantly increased NLRP3 and IL-1β transcription and upregulated the protein expression of NLRP3, Caspase-1, and GSDMD in HCAECs, whereas SAL markedly reversed these changes. The NLRP3 inflammasome has gained increasing attention in vascular inflammatory diseases because of its role in cytokine maturation, membrane damage, and inflammatory cell death (11). Previous studies have shown that SAL promoted AMPK activation and suppressed NLRP3/GSDMD-related pyroptosis in pancreatic β-cells under diabetic conditions (27). It also alleviates acetaminophen-induced liver injury by upregulating Sirt1 and inhibiting the Akt/Nrf2 and NF-κB/NLRP3 axis, and attenuates microglial inflammasome activation in cerebral ischemia-reperfusion injury through inhibition of TLR4/NF-κB signaling (28). Therefore, our results further suggest that SAL reduces NF-κB/NLRP3 inflammasome-related molecular changes in KD-related vascular inflammation.

However, the study has several limitations. First, the available databases are limited and it is impossible to predict all potential targets of SAL. Second, molecular docking results cannot meet the needs of some specific in vivo BP and high-precision calculations. Third, although the TNF-α-stimulated HCAECs model captures a critical effector phase of KD vasculitis, it does not fully recapitulate the complex immunological pathology of KD, which involves multiple immune cell types. Whether SAL is pharmacologically achievable or safe in pediatric KD patients remains unknown, and the present findings should be interpreted as mechanistic in vitro evidence. Therefore, our findings mainly reflect the direct effects of SAL on endothelial cells under inflammatory conditions. It is necessary to conduct further validation in KD-like vasculitis models and clinical trials to validate the therapeutic potential of SAL in KD.

Conclusions

In summary, this study systematically integrated network pharmacology, transcriptomics, molecular docking, and in vitro experiments to investigate the protective mechanism of SAL in TNF-α-induced endothelial inflammatory injury. SAL alleviates TNF-α-induced endothelial inflammatory injury in HCAECs by reducing inflammatory mediator expression, abnormal migration and inflammatory cell death. These effects were associated with modulation of NF-κB/NLRP3 inflammasome-related molecular changes. The present findings provide preliminary in vitro evidence for further investigation of SAL in KD-related vascular inflammation.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the MDAR reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0388/rc

Data Sharing Statement: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0388/dss

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0388/prf

Funding: This study was supported by the National Natural Science Foundation of China (Nos. 82070513 and 82100524), the Shanghai Natural Science Foundation Project (No.25ZR1401030), and the Young Clinical Professional Research Construction of Outstanding Medical Innovation Talents (No. DGF828030-4/035).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0388/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. McCrindle BW, Rowley AH, Newburger JW, et al. Diagnosis, Treatment, and Long-Term Management of Kawasaki Disease: A Scientific Statement for Health Professionals From the American Heart Association. Circulation 2017;135:e927-99. [Crossref] [PubMed]
2. Yu J, Cheng L, Zhan H, et al. Potential Mechanisms and Hypotheses for Pathogenic Microorganisms Triggering Kawasaki Disease. Clin Rev Allergy Immunol 2025;68:110. [Crossref] [PubMed]
3. Jone PN, Tremoulet A, Choueiter N, et al. Update on Diagnosis and Management of Kawasaki Disease: A Scientific Statement From the American Heart Association. Circulation 2024;150:e481-500. [Crossref] [PubMed]
4. Campbell AJ, Burns JC. Adjunctive therapies for Kawasaki disease. Journal of Infection 2016;72:S1-S5.
5. Chen MR, Chang TY, Chiu NC, et al. Validation of genome-wide associated variants for Kawasaki disease in a Taiwanese case-control sample. Sci Rep 2020;10:11756. [Crossref] [PubMed]
6. Atici AE, Noval Rivas M, Arditi M. The Central Role of Interleukin-1 Signalling in the Pathogenesis of Kawasaki Disease Vasculitis: Path to Translation. Can J Cardiol 2024;40:2305-20. [Crossref] [PubMed]
7. Jia C, Zhang J, Chen H, et al. Endothelial cell pyroptosis plays an important role in Kawasaki disease via HMGB1/RAGE/cathespin B signaling pathway and NLRP3 inflammasome activation. Cell Death Dis 2019;10:778. [Crossref] [PubMed]
8. Tsuge M, Uda K, Eitoku T, et al. Roles of Oxidative Injury and Nitric Oxide System Derangements in Kawasaki Disease Pathogenesis: A Systematic Review. Int J Mol Sci 2023;24:15450. [Crossref] [PubMed]
9. Sakurai Y. Autoimmune Aspects of Kawasaki Disease. J Investig Allergol Clin Immunol 2019;29:251-61. [Crossref] [PubMed]
10. Zheng Y, Huang S, Zhang J, et al. Melatonin alleviates vascular endothelial cell damage by regulating an autophagy-apoptosis axis in Kawasaki disease. Cell Prolif 2022;55:e13251. [Crossref] [PubMed]
11. Porritt RA, Zemmour D, Abe M, et al. NLRP3 Inflammasome Mediates Immune-Stromal Interactions in Vasculitis. Circ Res 2021;129:e183-200. [Crossref] [PubMed]
12. Paik S, Kim JK, Silwal P, et al. An update on the regulatory mechanisms of NLRP3 inflammasome activation. Cell Mol Immunol 2021;18:1141-60. [Crossref] [PubMed]
13. Shahi A, Afzali S, Firoozi Z, et al. Potential roles of NLRP3 inflammasome in the pathogenesis of Kawasaki disease. J Cell Physiol 2023;238:513-32. [Crossref] [PubMed]
14. Pu WL, Zhang MY, Bai RY, et al. Anti-inflammatory effects of Rhodiola rosea L.: A review. Biomed Pharmacother 2020;121:109552. [Crossref] [PubMed]
15. Fan F, Yang L, Li R, et al. Salidroside as a potential neuroprotective agent for ischemic stroke: a review of sources, pharmacokinetics, mechanism and safety. Biomed Pharmacother 2020;129:110458. [Crossref] [PubMed]
16. Li QY, Wang CB, Liu WJ, et al. Advances in natural medicinal plant-based interventions against hypoxia-related neuroinflammation. J Ethnopharmacol 2026;360:121161. [Crossref] [PubMed]
17. Cai Y, Chai Y, Fu Y, et al. Salidroside Ameliorates Alzheimer's Disease by Targeting NLRP3 Inflammasome-Mediated Pyroptosis. Front Aging Neurosci 2021;13:809433. [Crossref] [PubMed]
18. Zhao M, Wu N, Piao Y, et al. Research and development of natural product Salidroside: Pharmacology, total synthesis and structural modifications. Eur J Med Chem 2025;300:118186. [Crossref] [PubMed]
19. Zhu Z, Li J, Zhang X. Salidroside protects against ox-LDL-induced endothelial injury by enhancing autophagy mediated by SIRT1-FoxO1 pathway. BMC Complement Altern Med 2019;19:111. [Crossref] [PubMed]
20. Lei W, Chen MH, Huang ZF, et al. Salidroside protects pulmonary artery endothelial cells against hypoxia-induced apoptosis via the AhR/NF-κB and Nrf2/HO-1 pathways. Phytomedicine 2024;128:155376. [Crossref] [PubMed]
21. He Y, Wu S, Li J, et al. Artificial Intelligence in Traditional Chinese Medicine: Unraveling Herbal Medicine's Mechanisms. Research (Wash D C) 2026;9:1224. [Crossref] [PubMed]
22. Pinzi L, Rastelli G. Molecular Docking: Shifting Paradigms in Drug Discovery. Int J Mol Sci 2019;20:4331. [Crossref] [PubMed]
23. Hu W, Wu Y, Shi Q, et al. Systematic characterization of cancer transcriptome at transcript resolution. Nat Commun 2022;13:6803. [Crossref] [PubMed]
24. Ding Y, Peng Y, Wu H, et al. The protective roles of liraglutide on Kawasaki disease via AMPK/mTOR/NF-κB pathway. Int Immunopharmacol 2023;117:110028. [Crossref] [PubMed]
25. Noval Rivas M, Arditi M. Kawasaki Disease Vasculitis: From Diagnosis to New Concepts in Pathophysiology and Therapeutic Approaches. Annu Rev Med 2026;77:87-102. [Crossref] [PubMed]
26. Liu F, Zhao Y, Pei Y, et al. Role of the NF-kB signalling pathway in heterotopic ossification: biological and therapeutic significance. Cell Commun Signal 2024;22:159. [Crossref] [PubMed]
27. Zhou J, Yan S, Guo X, et al. Salidroside protects pancreatic β-cells against pyroptosis by regulating the NLRP3/GSDMD pathway in diabetic conditions. Int Immunopharmacol 2023;114:109543. [Crossref] [PubMed]
28. Yang K, Zeng L, He Q, et al. Advancements in research on the immune-inflammatory mechanisms mediated by NLRP3 inflammasome in ischemic stroke and the regulatory role of natural plant products. Front Pharmacol 2024;15:1250918. [Crossref] [PubMed]