Caulis spatholobi was procured from Beijing Tongrentang Herbal Pieces Co.,Ltd. (20160274) and authenticated by professor Tingting Liu of the Qiqihar Medical University. Approximately 1000 g of raw Caulis spatholobi herb was weighed and ground into powder. This powder was soaked in 1,000 mL of distilled water for 30 min, then heated to maintain a gentle boil for 45 min. The mixture was filtered, and the residue was extracted with additional distilled water for another 30 min. The two filtrates were combined. The combined filtrates were concentrated to 1 g/mL using a 100 °C water bath, and the resultant solution was stored at 0–5 °C prior to subsequent use.

The AECS was precisely measured to 1 mL using an electronic analytical balance (Shanghai, China). The extract was mixed with 10 mL of chromatographic-grade methanol via vortexing, followed by ultrasonic-assisted extraction for 15 min in an ultrasonic cleaning machine (Yunnan, China; 400 W, 40 kHz). After cooling to room temperature, the mixture was centrifuged at 8,000 rpm for 10 min using a 1–14 high-speed microcentrifuge (Sigma, Germany). The resulting supernatant was collected, filtered through a 0.22 μm microporous membrane, and transferred to an autosampler vial for subsequent analysis using a Thermo UHPLC-Q-Exactive Orbitrap-MS spectrometer (UltiMate 3,000, MA, United States).

Chromatographic separation was performed on a Waters ACQUITY UPLC BEH C18 Column (100 mm × 2.1 mm, 1.7 μm). The mobile phase was composed of phase A (0.1% formic acid aqueous solution) and phase B (0.1% formic acid–acetonitrile solution). The gradient elution protocol used for the analysis was as follows: 0–12.0 min, 5% B→95% B; 12.0–12.9 min, 95% B; 12.9–13.0 min, 95% B→5% B; 13.0–15.0 min, 5% B.

The electrospray ionization source (ESI source) was operated in both positive and negative ion monitoring modes (ESI⁺, ESI⁻) using full MS/ddMS2 scanning. Key parameters were set as follows: sheath gas flow rate at 9 arb, spray voltage at 3 kV, capillary temperature at 320 °C, S-lens voltage at 55 kV, and auxiliary gas heating temperature at 30 °C. Full MS scanning was performed over a range of m/z 100-1,200 with a resolution of 70,000, while maintaining 1e6 ions in the C-Trap and a maximum injection time of 100 ms. For ddMS2 mode, the resolution was set to 35,000 with an apex trigger time of 3–9 s, collision energies of 20, 40, and 60 kV, and a dynamic exclusion time of 8 s.

Based on the precise molecular weight information obtained from mass spectrometry, molecules were selected when the deviation between the measured and theoretical mass-to-charge ratios of their primary quasi-molecular ion peaks was less than 3 ppm. These molecules were then matched against a self-constructed database of drug chemical components. By integrating the characteristic fragmentation patterns of secondary mass spectrometry and neutral loss rules, the chemical identities were inferred.

This animal study was approved by the Animal Ethical Care Committee of Qiqihar Medical University (Approval No.: QMU-AECC-2024-174). Thirty 6-week-old SPF-grade male SD rats, weighing (220 ± 20) g, were purchased from Liaoning Changsheng Biotechnology Co., Ltd. (Animal License No.: SYXK(Hei) 2021-013). The rats were housed in the Animal Experiment Center of Qiqihar Medical University under a temperature of 25 °C, humidity of 45%, and a 12-h light-dark cycle, with free access to water and food.

After 1 week of acclimatization, 24 male Sprague-Dawley rats (n = 6) were randomly selected and administered 250 mg/kg adenine (Yuanye Bio-Technology Co., Ltd., Shanghai, China) suspension via oral gavage once daily for three consecutive weeks to establish the RIF model (Luo et al., 2022). Throughout the experimental period, animals were provided ad libitum access to standard rodent chow and water. The RIF rats were randomly divided into 4 groups: model group, Losartan group (LST) (Yuanye Bio-Technology Co., Ltd., Shanghai, China),low-dose AECS group (AECSL) and high-dose AECS group (AECSH), with 6 rats in each group. Body weight was measured every 2 weeks for each group. After successful modeling was confirmed by histopathological staining, the AECSL and AECSH groups were administered 150 and 300 mg/kg/day, respectively, the LST group was administered 50 mg/kg/day, and the sham and model groups were administered an equal amount of 0.9% NaCl, with gavage intervention for 4 weeks.

Following the 4-week treatment, 24-h urine volume were collected using metabolic cages. The rats were restrained and anesthetized under sterile conditions. Blood was drawn from the abdominal aorta of the rats in each group. Renal function indicators serum creatinine (Scr), urine creatinine (Ucr), blood urea nitrogen (BUN) were assessed with an enzyme marker apparatus (SpectraMax iD3, CA, United States), according to the methods outlined in the creatinine and blood urea nitrogen assay kits (Jiancheng Bioengineering Institute, Nanjing, China).

Upon completion of blood collection, the abdominal cavity was incised layer by layer, with the costovertebral angle serving as a landmark. The tissues surrounding the kidney were meticulously separated, and the kidney was extracted through the incision. The renal pedicle was clamped and ligated to prevent bleeding. Subsequently, the kidney was removed, and the fat and fascia were swiftly excised. Surface body fluids were wiped with filter paper. The corresponding renal index was then calculated. The calculation formula was: Renal index (%) = Kidney weight(g)/Body weight(g)×100.

Kidney tissues were fixed in 10% paraformaldehyde solution and then washed with running water overnight. Standard procedures for dehydration, clearing, infiltration, embedding, sectioning, and mounting were performed using a paraffin slicer (Leica HistoCore MULTICUT, Germany) and tissue embedding machine (Leica HistoCore Arcadia H, Germany). The resulting sections were stained with hematoxylin and eosin (HE) and Masson’s trichrome (Masson) stain. HE-stained sections were examined under a light microscope to assess morphological changes in the renal tissue, while Masson-stained sections were analyzed to evaluate the extent of renal fibrosis. Fibrotic area analysis avoids vascular and glomerular areas. Semiquantitative analysis of renal damage was performed using Image Pro Plus 6.0 software.

Male Sprague-Dawley rats (n = 6) were randomly allocated into sham control (equivalent volume of 0.9% NaCl), low-dose AECS (150 mg/kg/day), and high-dose AECS (300 mg/kg/day) groups, with all treatments administered via oral gavage for 7 consecutive days. Hepatic, pulmonary, and cardiac tissues were collected and subjected to H&E staining to assess dose-dependent effects of CS extract administration.

Immunohistochemical techniques were utilized to assess the expression of α-SMA and fibronectin proteins (diluted 1:100, Proteintech, China) in rat kidney cortex sections. Initially, the sections were deparaffinized in xylene and subjected to gradient ethanol dehydration. Antigen retrieval was subsequently performed using citrate buffer, followed by washing with phosphate-buffered saline (PBS). Endogenous peroxidase activity was then inhibited using a 3% hydrogen peroxide solution, and the sections were blocked with bovine serum albumin. Primary antibodies against α-SMA and fibronectin were applied and incubated overnight at 4 °C, followed by the addition of corresponding secondary antibodies (diluted 1:100) with incubation at room temperature for 1 h. Following DAB staining, hematoxylin counterstaining, dehydration in ethanol, and clearing in xylene, the sections were mounted with neutral resin. Finally, the samples were examined under a microscope, and the brown-positive protein expression was quantitatively analyzed.

Paraffin-embedded kidney sections were dewaxed, rehydrated through an ethanol gradient, and antigen-retrieved, then washed with PBS and blocked at room temperature for 15 min. Sections were incubated overnight at 4 °C with diluted primary antibodies CD68 or CD206 (1:200, Boster, United States), washed with PBS, and incubated with fluorescent-labeled secondary antibodies for 1 h at room temperature. After PBS washing, blocking was performed with normal goat serum, and the α-SMA primary antibody was applied overnight at 4 °C. Following PBST washing, HRP-conjugated secondary antibody was added for 30 min at room temperature. The sections were washed with PBS, counterstained for nuclei, rinsed with PBST, and mounted with anti-fade mountant. Images were acquired using an upright microscope (Nikon, Germany).

Total RNA was extracted from kidney tissues, reverse transcribed into cDNA, and amplified by RT-qPCR using SYBR Green Supermix. Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH) served as the internal reference, with three biological replicates. Relative quantification was performed using the 2^−ΔΔCt method. Primer sequences were listed in Table 1.

Each experiment was independently repeated at least three times. Data were expressed as mean ± SEM and analyzed statistically using GraphPad Prism 10.1.2 software, which was also used for image processing. Normally distributed data were assessed by one-way ANOVA, while non-normally distributed data were analyzed using the nonparametric Wilcoxon rank-sum test. A P value of less than 0.05 was considered statistically significant.

Based on predictions from the UHPLC-Q-Exactive Orbitrap-MS and TCMSP database and supplemented by literature, 20 valid active ingredients were identified after deduplication. Using the Swiss Target Prediction database, 484 potential target proteins for CS were predicted. These targets were subsequently entered into OMIM, STRING, GENE CARDS, DisGeNET, and DrugBank databases. After integration and removal of duplicates, a total of 905 RIF disease targets were obtained. A Venn diagram was generated using the EVenn platform, and 97 common targets were identified (Figure 3A).

The data for drugs, active ingredients, and key targets were matched to construct the in teraction network. In the diagram, bluish violet diamonds represent the key targets of CS affecting RIF, while pink circles denote the active ingredients of the drugs. The results (Figure 3D) illustrate the complex interrelationships between“drug-active ingredient-target”, reflecting the intricate biological mechanisms of drug action in the body. The main active ingredients were identified as licochalcone A, 3,7-dihydroxy-6-methoxy-dihydroflavonol, petunidin, augelicin, and 3-Hydroxystigmast-5-en-7-one. These ingredients are suggested to be the key active ingredients for CS in the treatment of RIF.

The 97 key targets were imported into the STRING database to obtain a PPI network rela tionships (Figure 3B). These relationships were subsequently visualized in Cytoscape, resulting in the construction of a protein interaction network diagram (Figure 3C). The network topology parameters were analyzed based on the target protein interaction network. TNF, AKT1, EGFR, and SRC were identified as the top four core targets ranked by node degree in the PPI network, suggesting that they may be critical targets for CS in the treatment of RIF.

The DAVID database was utilized to perform GO and KEGG enrichment analysis on the intersected drug-disease targets obtained. The top 10 terms for BP, CC, and MF, along with 18 KEGG pathways, were selected for visualization (P < 0.05). The GO analysis indicated (Figures 4A–C) that CS primarily participated in biological processes such as positive regulation of gene expression and protein phosphorylation; it involved cellular components like the plasma membrane and nucleus; and influenced molecular functions including ATP binding and identical protein binding. The KEGG analysis results showed (Figures 4D,E) that CS likely exerted its anti-RIF effects through signaling pathways such as PI3K-Akt, Proteoglycans in cancer, AGE-RAGE, and EGFR.

To further investigate the mechanism by which CS treats RIF, molecular docking was conducted between the active ingredients of CS and the top-ranked targets in the PPI network based on Degree value. The binding energies between the core ingredients and their targets were predicted, and binding affinities were analyzed (Figure 5A). The results demonstrated (Figure 5B) that all binding energies were below −5.0 kcal/mol, indicating a strong binding affinity between CS and the targets, which was consistent with predictions from the network pharmacology platform. It was further predicted that CS primarily acts through these targets and active ingredients to exert its effects. The findings were deemed reliable.

Renal function is pivotal in assessing the efficiency of drug metabolism and excretion. The preparation of SD rat models for RIF disease and the administration plan for AECS were depicted in Figure 6A. Compared to the sham group, the model group exhibited significantly elevated levels of SCr, UCr, BUN, and 24-h urine volume (Figures 6E–H, P < 0.01). Conversely, the LST group and each subgroup of AECS demonstrated significantly reduced levels of SCr, UCr, BUN and 24-h urine volume (P < 0.01) compared to the modelgroup. Among the AECS dose groups, the high-dose group exhibited the most effective results, indicating that AECS effectively ameliorated the occurrence and progression of RIF.

Compared to the sham group, the body weight of the model group was significantly reduced (Figures 6B p < 0.01), accompanied by late-stage symptoms such as mental lethargy, decreased food intake, pale auricles, and dry body hair. The renal index (Figure 6D) was significantly elevated in the model group compared to the sham group (P < 0.01). In contrast, the LST group and each subgroup of AECS exhibited a significantly reduced renal index (P < 0.01) compared to the model group. These findings suggest that AECS may ameliorate the slow weight gain induced by adenine in rats and partially delay the deterioration of renal tissue.

The evolution of renal tissue pathology is a crucial component in verifying the pharmacological mechanisms underlying drug therapeutic effects. As shown in Figure 6C, the kidneys in the model group were visibly enlarged and appeared pale with an irregular surface and nodular or granular changes, in contrast to the deep red kidneys of the sham group. In comparison, the aforementioned symptoms were improved to varying degrees in the treatment groups relative to the model group.

HE staining results (Figure 7A) revealed that renal units in the sham group maintained a normal structure without significant pathological damage, whereas the model group exhibited urate and hemosiderin deposition. The number of glomeruli was reduced, with glomerular atrophy and sclerosis observed. Renal tubules were dilated, and epithelial cells showed degeneration and necrosis. Extensive mononuclear and lymphocytic inflammatory infiltration was present in the renal interstitium, accompanied by fibrotic changes. In contrast, improvements in glomerular, tubular, and interstitial inflammatory infiltration and lesions were observed in the LST group and the AECS groups compared to the model group.

Masson staining results (Figure 7B) demonstrated minimal blue staining and collagen fiber deposition in the kidneys of the sham group. Conversely, extensive collagen fiber deposition and widespread fibrosis were found in the model group. Compared to the model group, varying degrees of reduction in collagen fiber deposition and fibrosis were observed in the LST group and the AECS groups. The positive fibrotic area in the kidneys of the model group was significantly increased compared to the sham group (P < 0.01). Conversely, the positive fibrotic area was significantly decreased in the LST group and the AECS groups compared to the model group (P < 0.01), as shown in Figure 7C. These findings further indicate that CS has an ameliorative effect on RIF.

H&E staining was performed on hepatic, pulmonary, and cardiac tissues, and no significant inflammation or histopathological changes were detected across all groups through histological examination (Figure 7D).

The pathological hallmarks of RIF include the upregulation of the interstitial marker α-SMA and the excessive deposition of the epithelial-mesenchymal transition (EMT)-related marker FN (Djudjaj and Boor, 2019). In the sham group, α-SMA and FN proteins were minimally expressed in the glomerular mesangial areas and renal interstitium of rats. Compared to the sham group, In comparison with the control group, the expression of α-SMA and FN in the renal tissues of the model group showed a significant increase (P < 0.01). However, both LST and AECS treatments markedly downregulated the expression of α-SMA and FN compared to the model group (P < 0.01). As shown in Figures 8A–D. These findings indicate that CS protects against RIF by concurrently suppressing EMT-driven fibrogenesis and pathological ECM deposition.

MMT was closely associated with renal pathologies, with M2 macrophages playing a pivotal role in RIF (Li G. et al., 2024). Rat kidney tissues were co-stained for general macrophage markers CD68+α-SMA⁺(Figures 9A,C) and M2-specific markers CD206+α-SMA⁺(Figures 9B,E) using double immunofluorescence. Compared to the sham group, the model group exhibited significantly enhanced fluorescence intensity of MMT cells, particularly within M2 macrophages (P < 0.01). LST and AECS treatment markedly reduced MMT cells fluorescence intensity in both total (p < 0.05) and M2 (P < 0.01) macrophage populations as shown in Figures 9D,F.

The network pharmacology-predicted core targets AKT1 and EGFR, along with inflammatory cytokines IL-6 and IL-10, were experimentally validated. Compared to the sham group, the model group exhibited significantly elevated mRNA levels of Akt1, Egfr, Il-6 and Il-10 in renal tissues (P < 0.01). AECS treatment significantly decreased the expression of Akt1, Egfr, Il-6 and Il-10 compared to the model group (P < 0.01), as shown in Figure 9G.