Abstract Background The Wenpitongluo Decoction (WPTLD) was a classical herbal formula composed of medicinal herbs with both edible and therapeutic properties. It demonstrated clinical efficacy in treating Cardiorenal Syndrome (CRS), though its mechanism of action remained unclear. Although inflammatory and oxidative stress pathways in CRS have been intensively studied, the roles of ferroptosis and anoikis, which may be activated by these pathways, have received little attention. Methods First, the active components of WPTLD were obtained through the TCMSP and Herb databases, and then identified using UHPLC-HRMS. Subsequently, target prediction of the identified components was carried out via the SwissTargetPrediction platform. While CRS-related targets were retrieved from GEO, GeneCards, and PharmGKB. A gene library of ferroptosis- and anoikis-associated targets was established. Tissue-specific mRNA expression profiles were analyzed via BioGPS. Subsequently, protein-protein interaction (PPI) networks were constructed to identify core targets, followed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses using Metascape. Finally, molecular docking assessed binding affinities between active components and core targets, with top-ranked complexes undergoing molecular dynamics (MD) simulations. Results Fifteen bioactive components and 39 component-disease interaction targets were identified, predominantly localized in kidney, thymus, lung, adipocytes, adrenal gland, and heart tissues. Topological analysis of PPI networks revealed eight core targets, including ferroptosis-/anoikis-associated SIRT1, PTGS2, and PRKCA. KEGG analysis highlighted critical pathways such as AMPK and PI3K-Akt signaling. Notably, molecular docking and MD simulations demonstrated stable binding between active compounds and core targets. Conclusion This study systematically deciphers WPTLD’s anti-CRS mechanisms via targeting ferroptosis- and anoikis-related genes through multi-pathway modulation. These findings not only clarify the pathological roles of ferroptosis and anoikis in CRS but also provide a computational framework for developing therapeutic strategies. Keywords: cardiorenal syndrome, ferroptosis, anoikis, Wenpitongluo decoction, oxidative stress, computational biology 1 Introduction Cardiorenal syndrome (CRS) is a clinical syndrome characterized by acute or chronic injury to one organ (heart or kidney) resulting from acute or chronic dysfunction of the other ([37]Ronco et al., 2010). Over 60% of patients with acute decompensated heart failure exhibit coexisting chronic kidney disease (CKD) ([38]McCallum and Sarnak, 2023). Globally, the prevalence of CKD continues to rise, serving as an independent risk factor for cardiovascular diseases (CVDs) such as coronary artery disease and congestive heart failure. Notably, CKD patients demonstrate higher CVD prevalence, with CVD incidence and mortality rates positively correlating with CKD severity ([39]Bikbov et al., 2020; [40]Bagshaw et al., 2010). Despite these associations, the pathophysiological mechanisms underlying CRS remain incompletely understood. Current evidence suggests that CRS pathogenesis may involve hemodynamic alterations, neurohumoral dysregulation, inflammation, oxidative stress, endothelial dysfunction, and iron metabolism abnormalities ([41]Kim et al., 2023; [42]Obi et al., 2016). While inflammatory and oxidative stress mechanisms in CRS are well-established ([43]Rangaswami et al., 2019) and known to trigger ferroptosis ([44]Chen et al., 2023), research specifically investigating ferroptosis in CRS remains limited. Ferroptosis, an iron-dependent regulated cell death driven by lethal lipid peroxidation accumulation, has emerged as a critical player in cardiovascular and renal pathophysiology. Inhibition of ferroptosis reduces cardiomyocyte death, alleviates heart failure (HF) symptoms, and delays HF progression ([45]Chen et al., 2024). Moreover, ferroptosis significantly contributes to HF and CKD progression, representing a promising therapeutic target ([46]Wang J. et al., 2022; [47]Wang K. et al., 2022). Another form of cell death, anoikis, which is a type of programmed cell death occurring when cells lose contact with the extracellular matrix (ECM), has garnered increasing attention and is widely applied in cancer research. Although initially studied in cancer, anoikis was implicated in cardiovascular diseases (CVDs) as early as 2003 ([48]Michel, 2003). Recent cellular experiments further demonstrate that modulating anoikis suppresses renal fibrosis ([49]Liu et al., 2022). Previous studies have shown that during ferroptosis, lipid peroxidation can directly disrupt cell membrane structure, oxidize ECM components, and promote matrix degradation, thereby affecting cell-ECM interactions and potentially inducing anoikis ([50]Dixon et al., 2012). Thus, this study focuses on the mechanisms of ferroptosis and anoikis, aiming to uncover their unknown roles in CRS and establish new directions for researching the pathogenesis of CRS. Clinically, diuretics remain the primary strategy for managing volume overload in CRS. However, diuretic resistance frequently develops in advanced CRS ([51]Freda et al., 2011), complicating treatment. Concurrent cardiac and renal dysfunction substantially increases mortality, complication rates, and healthcare costs, leading to poor prognoses ([52]Forman et al., 2004). These challenges underscore the urgent need for novel therapeutic approaches. Traditional Chinese Medicine (TCM), guided by the holistic concept of systemic organ regulation and balance, aligns well with the multifactorial pathogenesis of CRS. The Wenpitongluo Decoction (WPTLD), derived from two classical TCM formulas, Linggui Zhugan Decoction (LGZGD) and Huangqi Chifeng Decoction (HQCFD, incorporates additional components (Shenqu and Yimucao). These medicines come from various parts of different plants and have edible and medicinal properties. LGZGD exhibits anti-inflammatory, antioxidant, and cardioprotective properties, demonstrating therapeutic potential in HF ([53]Wang and Huang, 2024; [54]Sun S. et al., 2022). Pharmacological studies reveal that LGZGD contains multiple bioactive components capable of treating nephrotic syndrome via multi-target mechanisms ([55]Shi et al., 2023). Modified HQCFD formulations regulate apoptosis, suppress mesangial cell inflammatory proliferation, and mitigate chronic glomerulonephritis progression ([56]Ma et al., 2023). Furthermore, HQCFD derivatives ameliorate podocyte injury, reduce proteinuria, and alleviate renal fibrosis and glomerulosclerosis ([57]Zhao et al., 2022; [58]Zhao et al., 2024). Clinical observations indicate promising efficacy of WPTLD in CRS management. Building on this evidence, our study employs computational biology strategies, including network pharmacology, molecular docking, and molecular dynamics simulations, to investigate the molecular mechanisms of WPTLD in CRS treatment, with a focus on ferroptosis and anoikis pathways. This integrative approach aims to elucidate novel pathological mechanisms and therapeutic targets for CRS. The research workflow is illustrated in [59]Figure 1. FIGURE 1. [60]Flowchart illustrating the research framework for WPTLD and its targets related to cardiorenal syndrome and programmed cell death. It includes databases TCSMP, HERB, CNKI, UHPLC-HRMS, and others for data collection. Components targets, disease targets, interaction targets, core targets, and active ingredients are analyzed using various tools like Swiss Target Prediction, KEGG pathway enrichment, and PPI network. The diagram also shows molecular docking and molecular dynamics simulations to explore active ingredients' interactions. [61]Open in a new tab Flow chart of the study. 2 Materials and methods 2.1 Screening of WPTLD active components and targets WPTLD comprises Fuling (FL; Poria), Guizhi (GZ; Cinnamomi Ramulus), Baizhu (BZ; Atractylodis Macrocephalae Rhizoma), Gancao (GC; Glycyrrhizae Radix), Huangqi (HQ; Astragali Radix), Chishao (CS; Paeoniae Radix Rubra), Fangfeng (FF; Saposhnikoviae Radix), Yimucao (YMC; Leonuri Herba), and Chaoshenqu (SQ; Massa Medicata Fermentata). Chemical constituents of these herbs were retrieved from the Traditional Chinese Medicine Systems Pharmacology Database (TCMSP). Components unavailable in TCMSP were supplemented via the Herb Database ([62]http://herb.ac.cn/) and literature mining. Compounds were screened based on oral bioavailability (OB) of ≥30% and drug-likeness (DL) of ≥0.18. Those without predicted target proteins were excluded. 2.2 UHPLC-HRMS analysis All herbal materials were obtained from Guang’anmen Hospital, China Academy of Chinese Medical Sciences. After 30-min soaking, herbs were decocted twice for 1 h each in 8 volumes of water at 100°C under atmospheric pressure. The resulting decoctions were combined and concentrated to a density of 2.5 g/mL. WPTLD components were analyzed using ultra-performance liquid chromatography coupled with high-resolution mass spectrometry (UHPLC-HRMS; ACQUITY UPLC I-Class HF system, Waters Corporation) equipped with an ACQUITY UPLC HSS T3 column (100 mm × 2.1 mm, 1.8 µm; Waters Corporation) and a Thermo Orbitrap QE mass spectrometer (Thermo Fisher Scientific). For detailed information, please refer to the [63]Supplementary Material. 2.3 Construction of target libraries Target prediction for mass spectrometry-identified active components was performed using the SwissTargetPrediction platform to obtain WPTLD targets. Two datasets ([64]GSE66494 and [65]GSE21610) were retrieved from the GEO database using the keyword “cardiorenal syndrome.” mRNA expression profiles were analyzed via multi-chip joint analysis with the R limma package, identifying differentially expressed genes (DEGs) using thresholds of |log2FC| > 1 and P < 0.05. Additionally, CRS-related disease targets were collected from five databases: GeneCards ([66]http://www.genecards.org/), OMIM ([67]http://www.omim.org/), Therapeutic Target Database (TTD; [68]http://db.idrblab.net/ttd/), DrugBank ([69]https://go.drugbank.com/), and PharmGKB ([70]https://www.pharmgkb.org/). The retrieved targets were merged and deduplicated to establish a CRS target library. Ferroptosis-related genes were downloaded from the FerrDb database ([71]http://www.zhounan.org/ferrdb/current/), while anoikis-associated genes were obtained by searching “anoikis” in GeneCards. Corresponding target libraries were constructed for ferroptosis and anoikis. 2.4 Venn diagram visualization The intersection of WPTLD component targets (identified via mass spectrometry), CRS targets, ferroptosis genes, and anoikis genes was analyzed using Venny 2.1.0 ([72]http://bioinfo.cnb.csic.es/tools/venny/). This generated component-disease interaction targets and overlapping targets between ferroptosis/anoikis genes, visualized through Venn diagrams. 2.5 GO and KEGG enrichment analyses Interaction targets were uploaded to the Metascape platform with species set to Homo sapiens for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses. GO analysis included biological processes (BP), molecular functions (MF), and cellular components (CC), with significance thresholds of false discovery rate (FDR) < 0.05 and P < 0.05. Results were visualized using R software (version 3.4.1). 2.6 PPI network and “component-target-pathway” network construction Interaction targets were imported into the STRING database ([73]https://www.string-db.org/) to construct a protein-protein interaction (PPI) network model, with species restricted to Homo sapiens and interaction confidence set to “highest confidence” (>0.9). The PPI network was visualized using Cytoscape 3.9.0. Network topology parameters—including degree, betweenness, and closeness centrality—were analyzed via the built-in Network Analyzer tool to identify core therapeutic targets of WPTLD for CRS. A “component-target-pathway” network was further constructed by mapping WPTLD components, their targets, and enriched KEGG pathways in Cytoscape 3.9.0. Key bioactive components were screened based on topological parameters. 2.7 Tissue-organ network analysis To explore potential metabolic sites and target organs of WPTLD, mRNA expression levels of interaction targets across tissues were retrieved from the BioGPS database ([74]https://biogps.org). A “target-tissue/organ” network was generated using Cytoscape 3.7.1. 2.8 Molecular docking and molecular dynamics (MD) simulations Molecular docking was performed using AutoDock v4.2.6 and CB-Dock2. Ligand 3D structures were obtained from PubChem ([75]https://pubchem.ncbi.nlm.nih.gov/), while core target protein structures were downloaded from the RCSB PDB ([76]https://www.rcsb.org/). Target proteins were prepared in AutoDock 4.2.6, with docking grids generated via AutoGrid. Docking simulations were executed using AutoDock Vina, and results were visualized in PyMol v2.6 and LigPlot + v2.2.8. Subsequently, 100 ns molecular dynamics (MD) simulations were performed on the protein-ligand complexes obtained from molecular docking using GROMACS v2022.03 ([77]Van Der Spoel et al., 2005; [78]Abraham et al., 2015). The CHARMM36 force field ([79]Klauda et al., 2010) was employed for the protein system, while the GAFF force field ([80]Özpınar et al., 2010) was assigned to the ligand using AmberTools22. The ligand was hydrogenated and subjected to RESP charge calculation using Gaussian 16 W. The protein-ligand complex was solvated in a TIP3P water model ([81]Nayar et al., 2011) with a minimum distance of ≥1.2 nm between protein atoms and the edge of the cubic water box. System charge neutralization was achieved by adding appropriate numbers of Na^+ and Cl^− ions (concentration: 0.154 M). Energy minimization (EM) was conducted using the steepest descent algorithm ([82]Donnelly et al., 2021). The system was then gradually heated from 0 K to 300 K under the isothermal-isochoric (NVT) ensemble with solute position restraints, followed by equilibration at 300 K and 1 bar pressure in the isothermal-isobaric (NPT) ensemble. Finally, production MD simulations were conducted for 100 ns with trajectory recording. The MD trajectories were analyzed for root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), solution accessible surface area (SASA), and hydrogen bond (H-bond) formation. Gibbs free energy was calculated using the built-in GROMACS utilities “g_sham” and “xpm2txt.py” based on RMSD and Rg values. Additionally, binding free energy was estimated using the “MMPBSA.py v.16.0” script ([83]Genheden and Ryde, 2015) through MM/PBSA calculations. 3 Results 3.1 Active components from TCMSP database By searching the TCMSP database for the active ingredients of WPTLD, a total of 166 active ingredients were retrieved. Among them, there were 6 Poria, 7 Gui Zhi, 4 Fried Atractylodes, 88 Glycyrrhiza, 20 Raw Astragalus, 15 Paeonia lactiflora, 18 Saposhnikovia divaricata, and 8 Yi Mu Cao; and 36 active ingredients were screened for the active ingredients of the Fried Shen Qu through the Herb database as well as literature search. After screening, 144 active ingredients were retained ([84]Supplementary Table S6). 3.2 UHPLC-HRMS analysis UHPLC-HRMS, covering both mass spectrometry and chromatography, was used to analyze WPTLD and detected 960 compounds ([85]Supplementary Table S5). An intersection was taken with the 144 active components from “[86]Section 3.1”, and 15 effective bioactive components were identified by ensuring OB ≥ 30% and DL ≥ 0.18 ([87]Table 1). Specifically, the categories of compounds detected by UHPLC-HRMS are shown in [88]Figure 2A; the chromatograms of the compounds are in [89]Figure 2B; and the mass spectra are presented in [90]Figures 2C,D. TABLE 1. Active ingredient information table. PubChem CID Metabolites OB% DL SMILES 5281605 Baicalein 33.52 0.21 C1(O)=CC2OC(C3C=CC=CC=3)=CC(=O)C=2C(O)=C1O 5281855 Ellagic acid 43.06 0.43 OC1=C(O)C2=C3C(=C1)C(=O)OC1=C3C(=CC(O)=C1O)C(=O)O2 124052 Glabridin 53.25 0.47 C12OC(C)(C)C=CC1=C1OC[C@@]([H])(C3C(O)=CC(O)=CC=3)CC1=CC=2 5281619 Glepidotin A 44.72 0.35 C1C=C(C2=C(O)C(=O)C3C(O)=CC(O)=C(C/C=C(\C)/C)C=3O2)C=CC=1 480859 Glyasperin C 45.56 0.40 CC(=CCC1=C(C2=C(C=C1O)OCC(C2)C3=C(C=C(C=C3)O)O)OC)C 480787 Glycyrin 52.61 0.47 C1(OC)C(C/C=C(\C)/C)=C(OC)C2C=C(C3=CC=C(O)C=C3O)C(=O)OC=2C=1 5318585 Isolicoflavonol 45.17 0.42 C1(O)=CC2OC(C3C=CC(O)=C(C/C=C(\C)/C)C=3)=C(O)C(=O)C=2C(O)=C1 5281654 Isorhamnetin 49.6 0.31 C1(O)C=C2OC(C3=CC=C(O)C(OC)=C3)=C(O)C(=O)C2=C(O)C=1 5318679 Isotrifoliol 31.94 0.42 COC1=C2C3=C(C4=C(O3)C=C(O)C=C4)C(=O)OC2=CC(O)=C1 5280863 Kaempferol 41.88 0.24 C1(O)C=C2OC(C3=CC=C(O)C=C3)=C(O)C(=O)C2=C(O)C=1 5318999 Licochalcone B 76.76 0.19 C1(O)=CC=C(C(=O)/C=C/C2C=CC(O)=C(O)C=2OC)C=C1 5319013 Licoricone 63.58 0.47 C1(O)=CC2OC=C(C3C(O)=CC(OC)=C(C/C=C(/C)\C)C=3OC)C(=O)C=2C=C1 336327 Medicarpin 49.22 0.34 C1(O)=CC2OC[C@@]3([H])C4C=CC(OC)=CC=4O[C@@]3([H])C=2C=C1 442534 Paeoniflorin 53.87 0.79 CC12CC3(O)OC(O1)C1(COC(=O)C4=CC=CC=C4)C3CC21OC1OC(CO)C(O)C(O)C1O 5481948 Semilicoisoflavone B 48.78 0.55 C1(O)C=C(O)C2C(=O)C(C3=CC(O)=C4OC(C)(C)C=CC4=C3)=COC=2C=1 [91]Open in a new tab FIGURE 2. [92]Four-panelled scientific image:A: Pie chart showing chemical composition, including categories like carbohydrates (20.38%), terpenes (24.80%), and others.B: Two charts showing ellagic acid's abundance and relative intensity across certain mass-to-charge ratios.C: Chromatogram labeled BPC (POS) illustrating abundance over time.D: Chromatogram labeled BPC (NEG), similar to panel C, showing abundance over time. [93]Open in a new tab UHPLC-HRMS analysis results. (A) Content distribution and classification of herbal components; (B) EIC (Extracted Ion Chromatogram) of ellagic acid and its MS/MS spectrum compared with the LuMet-TCM standard library; (C) BPC (Base Peak Chromatogram) diagram in the positive ion mode; (D) BPC diagram in the negative ion mode. 3.3 Network pharmacology visualization results Target prediction for the identified components via SwissTargetPrediction yielded 509 targets after deduplication. Integrated analysis of disease-related genes from multiple sources identified 1568 CRS targets, 484 ferroptosis-related genes, and 919 anoikis-associated genes. Venn diagram analysis ([94]Figure 3A) revealed 39 component-disease interaction targets ([95]Table 2). FIGURE 3. [96]A composite image showing various data analyses: A) A Venn diagram displaying overlaps between four lists: WPTTD, CRS, Ferroptosis, and Anoikis, with a bar graph beneath indicating the size of each list.B) A network diagram illustrating connections between genes, highlighted for Closeness, Betweenness, and Degree - criteria for visual analysis.C) A circular network chart showing genes connected by lines, centralizing genes TNF and BCL2.D) A network diagram correlating genes with cell types, indicating gene expressions across different cells. [97]Open in a new tab WPTLD target screening and tissue network. (A) Component - disease target Venn diagram. (B,C) PPI network diagrams. (D) Interaction target - tissue/organ network diagram. TABLE 2. Venn diagram component - disease interaction targets. No. Gene Class 1 AKR1B1 2 PTGS2 *△ 3 CD38 4 AHR 5 IGF1R △ 6 SIRT1 *△ 7 GSR 8 ALDH2 9 PON1 10 ODC1 11 CYP11B2 12 HDAC10 13 HDAC2 14 BCL2 △ 15 PIK3CD 16 NAAA 17 PIK3CA *△ 18 MTOR *△ 19 MDM4 * 20 PDE4C 21 IGFBP2 22 CXCR4 △ 23 CLK1 24 GLI2 △ 25 YWHAG 26 PTGS1 27 GRK2 28 SLC9A1 29 AGTR1 30 PDK3 31 ERCC5 32 PRKCA *△ 33 FGFR1 △ 34 PLAA 35 LGALS3 △ 36 PABPC1 37 OGA 38 MME 39 TNF △ [98]Open in a new tab ^* , Ferroptosis - related genes. △, Anoikis - related genes. The PPI network constructed using STRING ([99]Figure 3C) comprised 33 nodes and 105 edges. Topological analysis based on median degree centrality (DC: 6.364), closeness centrality (CC: 0.015), and betweenness centrality (BC: 34.909) identified 8 core targets ([100]Figure 3B), including ferroptosis-/anoikis-associated SIRT1, PTGS2, and PRKCA. YWHAG was excluded due to unavailable BioGPS data, leaving 38 interaction targets for tissue mapping. The “target-tissue/organ” network demonstrated predominant localization in kidney, thymus, lung, adipocytes, adrenal gland, and heart tissues ([101]Figure 3D). Enrichment analysis was conducted through the Metascape platform, and the graphs were drawn using the adjusted P value (q-value). KEGG pathway was enriched in PI3K-Akt signaling pathway, AMPK signaling pathway, Apoptosis and other signaling pathways ([102]Figure 4A); GO enrichment analysis showed that Biological Processes were enriched in response to oxygen levels, apoptocic signaling pathway, etc., Cellular Component was enriched in membrane raft, intercalated disc, and Molecular Functions was enriched in protein kinase activity, etc. ([103]Figure 4B). The “component-target-pathway” network ([104]Figure 4C) prioritized seven bioactive components: paeoniflorin, kaempferol, isorhamnetin, licoricone, glabridin, ellagic acid, and baicalein. KEGG pathway mapping further annotated core targets within signaling cascades ([105]Figure 4D). FIGURE 4. [106]A composite image consists of four panels. Panel A shows a Sankey diagram on the left, illustrating gene relationships with pathways, and a dot plot on the right displaying gene ratios and significance. Panel B presents bar charts for biological processes, cellular components, and molecular functions with annotated count and significance levels. Panel C is a network diagram with nodes and connections illustrating interactions involving genes. Panel D is a detailed KEGG pathway map highlighting the PI3K-Akt signaling pathway with various gene interactions and connections. [107]Open in a new tab Enrichment analysis and WPTLD component screening. (A) KEGG pathway enrichment analysis. (B) GO enrichment analysis. (C) Component - target - pathway network. (D) PI3K - Akt signaling pathway. 3.4 Molecular docking and molecular dynamics simulation The major bioactive components selected were paeoniflorin, kaempferol, isorhamnetin, licoricone, glabridin, ellagic acid, and baicalein. Molecular docking verification was performed using the intersection targets of ferroptosis and anoikis, namely, SIRT1 (PDB ID: 4ZZI), PTGS2 (PDB ID: 5F19), and PRKCA (PDB ID: 4DNL). The results are shown in [108]Figure 5. A binding energy of less than 0 kcal/mol indicates that the combination can occur spontaneously, and the lower the binding energy, the stronger the binding ability. In this study, the binding energy between the predicted major bioactive components and the core targets ranged from −5.5 kcal/mol to −10.9 kcal/mol, indicating good binding ability. FIGURE 5. [109]Heatmap showing binding affinities of compounds to four proteins. Glabridin has the lowest affinity for PTGS2 at negative ten point nine. Colors range from light orange to dark red, indicating higher affinities. [110]Open in a new tab Molecular docking result heatmap. The molecular docking model of SIRT1-ellagic acid ([111]Figure 6) shows that the binding site of SIRT1-ellagic acid on SIRT1 is on the original activator. Therefore, in this study, 100 ns of MD simulation analysis was carried out on the SIRT1-ellagic acid and SIRT1-original activator complex systems, including root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), solution accessible surface area (SASA), and statistical analysis of hydrogen bond changes throughout the process, to study the dynamic properties of the molecular docking. As shown in the molecular docking model of ellagic acid and SIRT1 in [112]Figure 6A, the optimal docking conformation of ellagic acid and SIRT1 overlaps with that of the original activator and is closer to the protein’s β - sheet region. From the 2D structure of ellagic acid ([113]Figure 6B), it can be found that ellagic acid has 4 hydroxyl groups and 2 lactone ring structures. We speculate that these groups enable ellagic acid to form more stable interactions with the β - sheet region. Subsequently, a visual analysis of the two complex systems was carried out. From the 3D and 2D diagrams, it can be seen that the original activator only forms hydrogen bonds with the Asn226 residue of SIRT1 and has hydrophobic interactions with surrounding amino acids ([114]Figure 6C). In contrast, ellagic acid can form 1 hydrogen bond with Thr209 of SIRT1, 1 hydrogen bond with Asn226, and 2 hydrogen bonds with Glu230, as well as hydrophobic interactions with surrounding amino acids. Therefore, we consider that ellagic acid can form a more stable complex with SIRT1 than the original activator. FIGURE 6. [115]Molecular structures and interactions are depicted in diagrams of a protein-ligand complex. Panel A shows the binding site with original activator and ellagic acid highlighted. Panel B presents the chemical structure of ellagic acid. Panel C illustrates molecular interactions with amino acids like ILE-223 and ASN-226. Panel D provides a closer view of interactions involving GLU-230 and ILE-223. The right-side diagrams detail further interactions with bond distances. [116]Open in a new tab SIRT1 - Ellagic acid molecular docking model. (A) Docking model of Ellagic acid and SIRT1. (B) 2D structure of Ellagic acid. (C,D) Schematic diagrams of the docking results. MD results show that the RMSD curves ([117]Figure 7A) of the two complex systems gradually tend towards equilibrium after 40 ns, and the RMSD fluctuation range of both is within 1 nm, indicating that both the original activator and ellagic acid can form stable complex systems with SIRT1. In addition, to study the effect of small molecule binding on the flexibility of protein amino acid residues, the RMSF values of the amino acids of SIRT1 were calculated. As shown in the RMSF results ([118]Figure 7B), during the 100 ns simulation, the original agonist had 3 severe fluctuations, while ellagic acid maintained stability throughout the MD process, indicating that the small molecule ellagic acid can bind continuously and stably to SIRT1. Notably, compared with the original activator system, the binding of ellagic acid significantly reduces the flexibility of amino acid residues in this region. This result further indicates that ellagic acid can form more stable interactions with amino acid residues in this region than the original activator, thereby reducing the fluctuation of amino acids in this region and helping ellagic acid maintain continuous stability ([119]Figure 7E). Rg analysis shows that the Rg curves of the two complex systems have the same trend and overlap, indicating that the changes in the tightness of the SIRT1 protein during the simulation are the same ([120]Figure 7C). SASA analysis shows that the binding of ellagic acid results in a lower SASA value of the protein than the original activator system after 30 ns, suggesting that ellagic acid forms more hydrophobic interactions with SIRT1, reducing the hydrophilicity of the protein ([121]Figure 7D). As shown in [122]Figure 7F, the number of hydrogen bonds in the original activator system ranges from 1 to 5, while that in the ellagic acid system ranges from 4 to 7. Ellagic acid can form more hydrogen bonds with SIRT1 than the original activator, which is beneficial for the formation of a more stable complex system between ellagic acid and SIRT1. The Gibbs free energy landscape describes the stability of the receptor - ligand complex. As shown in [123]Figure 8A, the 3D topography of the Gibbs free energy of the SIRT1 - original activator complex is relatively rough, and the 2D diagram shows the presence of 2 minimum energy zones. In contrast, the 3D topography of the Gibbs free energy of the SIRT1 - ellagic acid complex forms a nearly single and smooth energy cluster, consistent with the 2D diagram ([124]Figure 8B). This result indicates that among the selected compounds, ellagic acid has the most stable dynamic conformation during the MD simulation, which is consistent with the MD results of the RMSD analysis. According to the MD calculation results, the total binding free energy of the SIRT1 - original activator complex is −7.71 kcal/mol, while that of the SIRT1 - ellagic acid complex is −10.89 kcal/mol ([125]Table 3). FIGURE 7. [126]Six graphs showing simulations comparing two compounds, SIRT1-original activator (blue) and SIRT1-Ellagic acid (red), over time. Graph A: RMSD, Graph B: RMSD, Graph C: Radius of gyration, Graph D: Solvent accessible surface area, Graph E: RMSF by residue, Graph F: Number of hydrogen bonds. Each graph illustrates different parameter fluctuations over 100 nanoseconds. [127]Open in a new tab 100 ns MD simulation analysis of the SIRT1 - Ellagic acid complex. (A) RMSD curve of the complex. (B) RMSD curve of the small molecule. (C) Rg curve of the complex. (D) SASA curve of the complex. (E) RMSF curve of SIRT1. (F) Hydrogen bond change curve of the complex. FIGURE 8. [128]Panel A shows Gibbs Energy Landscapes for SIRT1 with an original activator, featuring a 3D plot and contour map with energy values ranging from zero to fourteen kilojoules per mole. Panel B displays similar plots for SIRT1 with ellagic acid, with energy values between zero and sixteen kilojoules per mole. [129]Open in a new tab Complex Gibbs free energy analysis. (A) 3D and 2D free energy landscapes of the SIRT1 - original activator complex. (B) 3D and 2D free energy landscapes of the SIRT1 - Ellagic acid complex. Blue - and purple - shaded areas indicate that the stable conformation of the complex can be mapped at lower energy within the minimum free energy zone. Weak or unstable interactions lead to multiple, rough clusters in the free energy landscape, while strong, stable interactions form single, smooth clusters. TABLE 3. Average binding free energy (kcal/mol) of the two complexes calculated by the MM/PBSA method. Energy contributions SIRT1-original activator SIRT1-ellagic acid ΔVDWAALS −8.31 −6.16 ΔE[elec] −3.23 −10.92 ΔE[GB] 4.94 9.27 ΔE[surf] −1.11 −0.69 ΔG[gas] −11.54 −19.47 ΔG[solvation] 3.83 8.58 ΔTotal −7.71 −10.89 [130]Open in a new tab 4 Discussion CRS involves crosstalk between the heart and kidneys, with both pathological links and causal relationships. Its treatment remains exploratory. Many heart failure (HF) medications may negatively impact kidney function, and vice-versa. For instance, diuretics can reduce cardiac load but may cause hypovolemia, decreased renal perfusion, and pre - renal AKI if over-used. Similarly, high-dose intravenous fluid to correct pre-renal AKI (e.g., from hypovolemia or shock) can worsen HF by increasing cardiac preload. These situations complicate CRS treatment and management. In contrast, traditional Chinese medicine (TCM) formulas can simultaneously target the heart and kidneys. With growing research and development in TCM formulas, more evidence-based studies have confirmed their efficacy and safety, marking a new direction for clinical research. Our study identified the major bioactive components of WPTLD via UHPLC-HRMS as paeoniflorin, kaempferol, isorhamnetin, licoricone, glabridin, ellagic acid, and baicalein. The core targets associated with ferroptosis and anoikis were SIRT1, PTGS2, and PRKCA. BioGPS database analysis showed these targets are highly expressed in multiple organs. Molecular docking and MD simulations validated the good binding ability between active molecules and targets. In the study of cardiorenal disease treatment, flavonoids have been widely studied for their diverse biological activities. Previous research has shown that some flavonoids, such as kaempferol and isorhamnetin, have significant antioxidant, anti-inflammatory and cardioprotective effects ([131]Sun S. et al., 2022; [132]Dong et al., 2015). Moreover, recent studies have found that flavonoids can also inhibit ferroptosis and regulate apoptosis through multiple pathways, thereby exerting cardioprotective properties ([133]Xu et al., 2024). This study not only reveals the potential of these compounds in CRS treatment but also explores their specific mechanisms in regulating ferroptosis and apoptosis, offering key insights for their precise application in disease intervention. Cell experiments have confirmed kaempferol as a key component of LGZGD, effective in treating HF ([134]Sun S. et al., 2022). WPTLD, derived from LGZGD, has been shown by UHPLC-HRMS to retain the active component kaempferol, indirectly proving its therapeutic effect on HF. Furthermore, this study has demonstrated that the active components function by targeting core targets such as SIRT1. This is consistent with prior research findings. Wang A discovered that kaempferol can effectively reduce the expression of pro-inflammatory factors and inhibit the occurrence of oxidative stress and ferroptosis by activating SIRT1, including the HMGB1/TLR4/NF-κB and NRF2/SLC7A11/GPX4 pathways ([135]Wang et al., 2025). Previous study have shown that baicalein can enhance the ability of cells to resist ferroptosis, making it a potential therapeutic agent for ferroptosis - related tissue damage ([136]Xie et al., 2016). For example, Wang IC ([137]Wang IC. et al., 2023) and Fan ZY ([138]Fan et al., 2021) found that baicalein protected cardiomyocytes from ferroptosis induced by ferroptosis inducers and ischemia/reperfusion (I/R). Liang GQ discovered that baicalein improved renal function, inhibited renal ferroptosis, and slowed renal fibrosis ([139]Qiang et al., 2024). These findings highlight baicalein’s promise in cardiorenal disease therapy. Animal and cell experiments in an AKI model showed baicalein downregulated Fe^2+, MDA, and PTGS2, while upregulating SCL7A11, GPX4, and GSH, exerting ferroptosis - inhibiting effects. This was achieved by enhancing SIRT1 expression to promote p53 deacetylation ([140]Yu et al., 2023). This was consistent with the core target of action identified by this study through computational biology. Moreover, previous studies have shown that baicalein can exert cardioprotective effects by inhibiting apoptosis and inflammation ([141]Qu et al., 2016). Anoikis was a subtype of apoptosis, which suggests that baicalein might suppress anoikis. In this study, we discovered that baicalein exerts therapeutic effects on CRS by targeting anoikis - related targets, advancing its research in anoikis. Ellagic acid was a natural polyphenolic compound. Besides its strong antioxidant activity, in an experiment on a rat model of kidney injury, it was found that after intervention with ellagic acid, the levels of PTGS2 and MDA in rat serum decreased, while the activities of SOD and GSH significantly increased. The renal tissue structure was improved, indicating that ellagic acid also has an iron - death - improving effect. Meanwhile, it was observed that ellagic acid can inhibit apoptosis and autophagy pathways by reducing the expression of LC3B ([142]Bhattacharjee et al., 2021). This finding has been confirmed by other scholars. Research showed that ellagic acid can regulate cellular iron metabolism, boost free-iron excretion, activate the Nrf2/keap1 pathway, and increase GPX4 synthesis, thereby alleviating oxidative stress and ferroptosis ([143]Yang et al., 2024). This expands the research on ellagic acid, a natural polyphenol, in cardioprotection and offers new directions for its use in treating CRS and other diseases. Our study indicates that in CRS, ellagic acid may protect cardiac and renal functions via antiferroptotic and anoikis - inhibiting effects. Studies have shown that SIRT1 was a core target in ferroptosis, anoikis, and CRS. As a member of the Sirtuins (SIRTs) family, SIRT1 was widely distributed in cells and has strong deacetylase activity. Research has indicated that the activation or overexpression of SIRT1 can deacetylate p53. Since p53 can directly inhibit the GPX4 pathway and, under ROS, indirectly enhance the lipoxygenase (LOX) family’s function by inhibiting SLC7A11 ([144]Jiang et al., 2015), SIRT1’s deacetylation of p53 blocks this process, thereby inhibiting ferroptosis ([145]De Angelis et al., 2015; [146]Chen et al., 2022). These studies have revealed another function of SIRT1 and a novel therapeutic strategy: inhibiting ferroptosis by activating SIRT1. In this study, molecular docking and MD simulations showed that ellagic acid, an active component of WPTLD, can stably bind to the activation site of SIRT1. Notably, previous research has confirmed that ellagic acid can form a stable bond with the SIRT1 activation site. It has also been shown that SIRT1 activation leads to the deacetylation and activation of Nuclear factor E2-related factor 2 (NRF2), thereby regulating the cellular antioxidant response ([147]Bai et al., 2025). The activation of NRF2 upregulates antioxidants like GPX4 and GCLC. These enzymes scavenge lipid peroxides, prevent the build-up of lipid peroxidation products like MDA, and inhibit ferroptosis ([148]Shi and Ning, 2025). In a myocardial infarction study, the activation of the SIRT1/AMPK signaling pathway following traditional Chinese medicine intervention promoted mitophagy, which in turn suppressed oxidative stress and inflammatory responses in cardiomyocytes, thus improving cardiac function ([149]Sun X. et al., 2022). AMPK is a key cellular energy sensor that plays an important role in maintaining energy homeostasis and regulating cellular metabolism. Meanwhile, the AMPK signaling pathway has a dual - edged regulation on ferroptosis. On the one hand, AMPK activation can protect cells from ferroptosis by inhibiting lipid peroxidation and enhancing antioxidant capacity. On the other hand, activated AMPK may also promote ferroptosis by regulating specific signaling pathways such as mTOR and SLC7A11 ([150]Wang F. et al., 2023). In conclusion, in the CRS model, SIRT1 may exert therapeutic effects by regulating oxidative stress and ferroptosis through multiple pathways. Anoikis, a physiological process and special programmed cell death, contributes to tissue and cell homeostasis. In 2003, Michel JB linked Anoikis to CVD, suggesting it may drive cell loss in the cardiovascular system. If the balance between Anoikis and cell healing is disrupted, it can result in abnormal tissue remodeling, such as cardiomyocyte loss in early overloaded left ventricles, progressing to HF. In the ECM, Anoikis, along with inhibited cell adhesion and growth, may be a primary obstacle to cell healing, presenting potential new therapeutic targets ([151]Michel, 2003). This finding expands the study of anoikis and ECM in CVD, but so far, it has not been sufficiently emphasized or explored in mechanism - related research. However, Hong Liu analyzed CKD - related genes and anoikis - associated genes and identified common targets. Six hub genes (LAMC2, NRP1, CDH3, NDRG1, CLDN1, and LAMB3) were found through bioinformatics. Then, four CKD mouse models were set up to verify these targets. RT - qPCR tests showed changes in the expression of hub genes in CKD. It turned out that these anoikis - related genes might serve as potential diagnostic markers for CKD ([152]Liu et al., 2025). This undoubtedly represents a further advancement in the research on anoikis. The above results indicate the association between anoikis and heart and kidney diseases. After heart and kidney damage, anoikis occurs. This then activates macrophages and Myofibroblast (MyoFb) fibrosis, further destroying the ECM structure, exacerbating the cell adhesion disorder, and promoting more anoikis, forming a vicious cycle. The pathological changes of the kidneys in CRS included renal fibrosis, which was characterized by excessive deposition of ECM. MyoFbs were the main cells that generated ECM. When activated, MyoFbs produced large amounts of ECM, leading to fibrosis. Therefore, inhibiting MyoFb activation was a potential therapeutic strategy for controlling disease progression. Studies have demonstrated that acquired anoikis resistance constitutes a hallmark feature of MyoFb activation. Li XH established an anoikis model in MyoFb via TGF-β1 induction and observed significantly lower anoikis rates in MyoFb compared to human proximal tubular epithelial cells (HK-2). Western blot analysis revealed markedly elevated P-PI3K and P-AKT expression alongside reduced cleaved-caspase3 levels in MyoFb. Subsequent administration of PI3K/AKT pathway inhibitors partially reversed the increased proliferation and decreased apoptosis rates in MyoFb, confirming their anoikis resistance mechanism mediated through the PI3K/Akt signaling pathway. Furthermore, TSSC3 intervention effectively reduced PI3K-P85 and AKT levels in MyoFb, demonstrating that downregulating the PI3K/Akt pathway suppresses both anoikis resistance and pro-fibrotic capacity of renal-derived MyoFb, thereby attenuating renal fibrosis ([153]Liu et al., 2022). These results provide detailed explanations of the pathological role of Anoikis in heart and kidney diseases, as well as the significance of the PI3K/Akt signaling pathway. Therefore, in the early treatment of CRS, the occurrence of Anoikis should be avoided as early as possible, the activation of MyoFb should be inhibited, and the development of renal fibrosis should be blocked; while after the kidneys have already developed fibrosis, the PI3K/Akt pathway can be used to inhibit the resistance to anoikis, promote MyoFb apoptosis, thereby delaying renal fibrosis. In our prior bioinformatics analysis, we found that WPTLD’s targets were associated with anoikis. Thus, we hypothesized that WPTLD might combat CRS by targeting both ferroptosis and anoikis. Anoikis might have contributed to cardiovascular tissue remodeling, such as myocyte detachment in HF, endothelial denudation, and plaque rupture in atherosclerosis. In these contexts, the intracellular mechanisms of anoikis involve PI3K/Akt-mediated regulation of focal adhesions and integrin-linked kinase activity ([154]Aoudjit and Vuori, 2001; [155]Stupack et al., 2001). The PI3K/Akt signaling pathway was a crucial signaling pathway for cell survival. However, when cells lose contact with the ECM, the activity of the PI3K/Akt signaling pathway was inhibited, leading to cell apoptosis. Therefore, controlling this pathway can regulate apoptosis. Studies showed that activating the PI3K/Akt signaling pathway through stimulation or intervention inhibited apoptosis - related proteins like the Bcl-2 family by phosphorylating downstream targets, thus regulating apoptosis ([156]Taddei et al., 2012). 5 Conclusion In summary, this study employed mass spectrometry to identify active components of WPTLD and integrated network pharmacology, molecular docking, and molecular dynamics (MD) simulations to reveal, for the first time, that WPTLD contains bioactive compounds such as ellagic acid and baicalein. These components synergistically regulate the AMPK and PI3K/Akt signaling pathways via SIRT1/PTGS2/PKCA cross-talk, thereby targeting both ferroptosis and anoikis in cardiac and renal cells to combat CRS. While this work uncovers the potential roles of ferroptosis and anoikis in CRS pathogenesis, offering novel insights into its molecular mechanisms and scientific validation for the modernization of traditional Chinese medicine formulations, it remains limited by the absence of in vivo and in vitro experimental confirmation. Further investigations are warranted to elucidate the interplay between ferroptosis and anoikis, as well as their spatiotemporal dynamics in CRS progression. Funding Statement The author(s) declare that financial support was received for the research and/or publication of this article. This research was funded by the High Level Chinese Medical Hospital Promotion Project (Grant no. HLCMHPP2023040), Special Fund Project for the Construction of the Clinical Medical Research Center at Guang’anmen Hospital, China Academy of Chinese Medical Sciences (Key Research Project) (Grant no. 2022LYJSZX05). Data availability statement The original contributions presented in the study are included in the article/[157]Supplementary Material, further inquiries can be directed to the corresponding authors. Author contributions XM: Writing – original draft, Methodology. SS: Conceptualization, Writing – review and editing, Validation. CC: Conceptualization, Writing – review and editing. YL: Methodology, Writing – review and editing. BZ: Methodology, Writing – review and editing. QS: Funding acquisition, Writing – review and editing, Conceptualization. Conflict of interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Generative AI statement The author(s) declare that no Generative AI was used in the creation of this manuscript. Publisher’s note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Supplementary material The Supplementary Material for this article can be found online at: [158]https://www.frontiersin.org/articles/10.3389/fchem.2025.1617676/fu ll#supplementary-material [159]Table2.xlsx^ (10.1KB, xlsx) [160]Table3.xlsx^ (9.7KB, xlsx) [161]Table6.xlsx^ (16.8KB, xlsx) [162]Table4.xlsx^ (10.2KB, xlsx) [163]Table1.xlsx^ (10KB, xlsx) [164]Table5.xlsx^ (310.4KB, xlsx) Abbreviations AKI, Acute kidney injury; BP, Biological processes; CC, Cellular components; CKD, Chronic kidney disease; CRS, Cardiorenal syndrome; CVDs, Cardiovascular diseases; ECM, Extracellular matrix; GO, Gene Ontology; HF, Heart failure; HQCFD, Huangqichifeng Decoction; I/R, Sschemia/reperfusion; KEGG, Kyoto Encyclopedia of Genes and Genomes; LGZGD, Lingguizhugan Decoction; MD, Molecular dynamics; MF, Molecular functions; PPI, Protein-protein interaction; PRKCA, Protein kinase C alpha; PTGS2, Prostaglandin-endoperoxide synthase 2; Rg, Radius of gyration; RMSD, Root mean square deviation; RMSF, Root mean square fluctuation; SASA, Solution accessible surface area; SIRT1, NAD-dependent histone deacetylase sirtuin-1; TCM, Traditional chinese medicine; UHPLC-HRMS, Ultra-performance liquid chromatography coupled with high-resolution mass pectrometry; WPTLD, Wenpitongluo Decoction. References