Abstract
Background
Cervicitis is a common gynecological disorder with high recurrence and poorly understood mechanisms. Ainsliaea fragrans Champ. (AF) is used in traditional medicine for treating cervicitis, yet its active components and mechanism remain unclear.
Aim of the Study
This study aims to identify the anti-cervicitis active components of AF, and to evaluate its possible mechanism of action.
Methods
The chemical constituents absorbed into serum were identified using UPLC-Q-Exactive-Orbitrap-MS. Network pharmacology was applied to predict targets and pathways, and a rat cervicitis model was established for pharmacological evaluation, including histopathological examination, inflammatory cytokine assays, and cervical transcriptome sequencing. Key signaling pathways were further validated by western blotting.
Results
31 AF-derived components were detected in the serum. AF treatment ameliorated weight loss, reduced the vaginal index, decreased serum levels of IL-1β, IL-6, and TNF-α, and alleviated histopathological damage in cervicitis rats. Integrated analysis of network pharmacology and transcriptomics suggested the involvement of the AMPK-mTOR-HIF-1α signaling axis. Molecular docking indicated favorable binding affinities between key candidate components (e.g., glycyrrhizin, sanggenone H) and core targets (STAT3, MTOR, HIF-1α, AMPK). Western blot analysis confirmed that AF treatment upregulated p-AMPK expression while downregulating p-mTOR and HIF-1α protein levels.
Conclusion
The findings suggest that the improvement of cervicitis inflammation and injury by AF is associated with the modulation of the AMPK-mTOR-HIF-1α signaling axis, and this integrated strategy provides a mechanistic basis for the anti-cervicitis effect of AF, supporting its further investigation as a potential therapeutic agent.
1. Introduction
Cervicitis is a prevalent gynecological inflammatory disorder that poses a major clinical challenge and exerts adverse impacts on women’s reproductive health globally. 1 It is characterized by localized inflammation, epithelial damage, and immune cell infiltration within the cervical mucosa. If left untreated, chronic cervicitis may progress to severe complications, including pelvic inflammatory disease, infertility, ectopic pregnancy, and an elevated risk of contracting sexually transmitted infections such as HPV. 2 Current first-line therapeutic regimens for cervicitis are mainly based on broad-spectrum antibiotics or antiviral agents. However, the emergence of antibiotic resistance, high recurrence rate, and the failure of these therapies to effectively ameliorate the underlying dysregulated inflammation and immune responses underscore the urgent demand for novel therapeutic strategies with alternative mechanisms of action.3,4
Traditional Chinese Medicine (TCM) harbors a wealth of potential therapeutic for inflammatory and immune disorders. 5 Ainsliaea fragrans Champ. (AF, Chinese name: XingXiangTuErFeng, XXTRF), a medicinal herb traditionally used for clearing heat, resolving stagnation, dissipating stagnant nodules, cooling blood, and detoxifying, has a well-documented for treating gynecological inflammation, hemorrhage, and infections, has emerged as a promising agent in Jiangxi Province of China. 6 Modern pharmacological studies have confirmed its traditional medicinal applications, demonstrating that AF extracts exhibit potent anti-inflammatory, antibacterial, and hemostatic activities.6-8 Our previous studies have verified the anti-inflammatory and dysmenorrhea-alleviating effects of AF, and preparations containing AF are included in the Chinese Pharmacopoeia for the treatment of cervical erosion and colitis.6,9 However, the underlying molecular mechanisms by which AF exerts its therapeutic effects against cervicitis remain poorly elucidated, which restricts its clinical acceptance and application worldwide.
Transcriptomics, a powerful technique for the comprehensive analysis of cellular RNA transcripts, has become an indispensable tool for dissecting gene expression patterns and regulatory networks in complex biological processes. Network pharmacology, by contrast, provides a systems-level approach to deciphering the multi-target mechanisms of drug action, and clarifies the intricate interactions among drugs, targets, genes, and diseases. In the present study, we adopted an integrated strategy combining transcriptomics and network pharmacology to decipher the multi-target mechanisms underlying AF’s therapeutic effects against cervicitis.10,11
We hypothesized that the therapeutic effects of AF against cervicitis may be mediated by the crosstalk of inflammatory and metabolic pathways. By integrating network pharmacology with cervical tissue transcriptomics, we aimed to identify a pivotal signaling node that could integrate multi-omics data and serve as a mechanistic target for experimental validation. The present study was designed to investigate the therapeutic efficacy of AF on a well-characterized rodent model of cervicitis and to delineate its mechanism of action by examining its impact on the potential pathway and its downstream inflammatory mediators. Our findings are intended to provide a mechanistic foundation for the clinical application of AF and position it as a multi-target therapeutic agent for inflammatory gynecological diseases.
2. Materials and Methods
2.1. Chemicals and Reagents
AF herb was obtained from the School of Pharmacy, Jiangxi University of Traditional Chinese Medicine. Anti-Cervicitis Tablets (ACT, Jiangxi Hailesi Pharmaceutical Co., LTD., 10250201). General-purpose tissue fixative (neutral) (Servicebio, G1101). Rat interleukin 1β (IL-1β) ELISA kit (Batch No. FXs00283), rat interleukin 6 (IL-6) ELISA kit (Batch No. FXs00271), and rat tumor necrosis factor α (TNF-α) ELISA kit (Batch No. FXs01560) were all purchased from Shanghai Fenxi Biotechnology Co., Ltd.
2.2. The Preparation of AF
AF extract: The AF was collected and placed in a double-layer decoction bag. Ten times the volume of water was added for boiling for 40 min, after which the liquid was filtered. Fresh water was then added to repeat the boiling process for an additional 15 min. The mixture was filtered through gauze, and both batches were combined. The concentrated solution was adjusted to the required concentration (0.9 g/mL) for oral administration in high-dose group animals. Low-dose group samples were diluted proportionally using the high-dose solution.
Phenol Gel Solution: Prepare a 25% phenol gel solution by combining 5 mL of phenol, 1 g of Euphorbia pekinensis gum powder, 4 mL of glycerol, and 20 mL of distilled water.
ACT Solution: The dosage is 0.4212 g/kg. Prepare the solution using a gavage volume of 10 mL/kg to achieve a concentration of 0.04212 g/mL.
2.3. The Establishment of Rat Cervicitis Model and Drug Intervention
A total of 40 SPF-grade SD rats (female, 180-220 g, 6-8 weeks old) were housed at the Laboratory Animal Center under controlled conditions: temperature 22-24 °C, relative humidity 40%-70%, ammonia concentration <1.0 mg/m3, noise level <60 dB, and 12/12-hour circadian rhythm. The animals received standard feed and free access to water. All efforts were made to minimize animal stress during handling and procedures. All procedures were conducted strictly in accordance with the laboratory animal center’s guidelines. To mitigate the influence of uncontrolled variables on group comparisons, all rats were of similar age and weight, and were co-housed under identical environmental conditions throughout the study, ensuring that potential confounding factors (e.g., baseline stress, ambient microbiome) were distributed equally across groups. Animal qualification certificate: 20220004075104; Shanghai SLEAK Laboratory Animal Co., Ltd.; Animal license number: SCXK (Shanghai) 2022-0004. This experiment was approved by the Animal Welfare and Ethics Committee of Zhejiang Huitong Testing Technology (Group) Co., Ltd. (Animal Ethics Approval No.: HTDW-202502009). The reporting of this animal study conforms to the ARRIVE 2.0 guidelines. 12
After 1 week of adaptive feeding, 40 rats were randomly divided into 8 groups each: the blank group, model group, high-dose AF group, low-dose AF group, and positive drug group (Anti-cervicitis tablets). Except for the blank group, all other groups underwent modeling as previously study. 13 A 1.0-mL syringe was gently inserted into the deep part of the rat vagina about 1 cm, and 0.2 mL of 25% phenol paste was injected. The procedure was repeated once a day for 5 consecutive days. Vaginal changes were observed during the procedure. Successful modeling was manifested by white vaginal secretions, redness and swelling of the vagina, and congestion in the vaginal cavity.
On the 3rd day of modeling, drug intervention was carried out. High dose group of AF (AF-H, 9 g/kg), low dose group of AF (AF-L, 2.5 g/kg) and positive drug group (ACT, 0.4212 g/kg) were administered by gavage every day for 14 consecutive days.
2.4. Identification of AF-Derived Components Absorbed Into the Blood
UPLC-Q-Exactive-Orbitrap-MS was used to identify the blood components of AF, and the main active ingredients were used for network pharmacological analysis to predict possible targets against cervicitis.
Chromatographic conditions: Chromatographic column: Waters HSS T3 (1002.1 mm, 1.8μm); mobile phases A (0.1% formic acid aqueous solution) and B (0.1% formic acid acetonate solution); flow rate 0.3 mL/min, column temperature 40°C, injection volume 2μL. Elution gradient: 0 min A/B (100:0, v/v); 1 min A/B (1000:0, v/v). Mass spectrometry conditions: A Thermo Q Exactive HFX high-resolution mass spectrometer (Thermo, USA) was used to collect primary and secondary spectra. The system is equipped with an electrospray ionization source (ESI), 40 arb of sheath gas, 10 arb of auxiliary gas, ion spray voltage +3000 V/-2800 V, temperature 350°C, and ion transmission tube temperature 320°C. The scanning mode is Full ms ddMS 2, scanning in both positive and negative ions. The primary mass spectrum scan range is 70-1050 Da, with primary resolution 70000 and secondary resolution 17500, collision energies of 10/30/60V.
2.5. Network Pharmacology Analysis14,15
2.5.1. Collection of Component and Cervicitis Disease Targets
31 components absorbed into the blood of AF, SMILES structures, and targets were collected from the PubChem (https://pubchem.ncbi.nlm.nih.gov/) database. Target collection was sourced from both PubChem and SwissTarget Prediction (https://www.swisstargetprediction.ch/), with a screening threshold set at Probability > 0.
Genes associated with Cervicitis were obtained from the GeneCards database (https://www.genecards.org/) using a search term of Cervicitis, downloaded the spreadsheet of disease targets, and screened for genes with scores above average.
2.5.2. Construction of Drug-Component-Target Network
A drug-component-target table was created and imported into Cytoscape 3.7.2 software (https://cytoscape.org/). Click File in the menu bar, then select Import> From File to import the spreadsheet. Adjust node properties in the left control panel: Fill color controls the background color, Label Front Size determines font size, and Shape determines node shape.
2.5.3. Construction of PPI Network
Using Jvenn (https://jvenn.toulouse.inrae) tools, we generated an Overlapping Gene (OGE) map of compound-target and disease-target interactions. All OGEs were input into the STRING database (https://stringdb.org/) by pasting the list of overlapping targets into the “Name List” dialog box. After selecting Rattus norvegicus under “Organisms” and clicking “Continue”, we selected “Exports” in the toolbar to download concise tabular texts of protein-protein interaction (PPI) networks in both PNG and TSV formats.
2.5.4. Screening of Core Proteins
Cytoscape 3.7.2(https://cytoscape.org/) was opened, and the TSV format file from step 2.5.3 was imported. Optimize network node colors, fonts, and edges through the style bar in the control panel. Conduct network topology analysis using the “Network Analysis” feature. The Molecular Complex Detection (MCODE) application calculates hub networks to identify dense regions. Genes within these dense regions identified by MCODE are termed hub targets. The CytoNCA plugin is then used to analyze three preliminary network topology parameters: Degree, closeness centrality, and betweenness centrality. The calculation equations and definitions of these parameters demonstrate the topological significance of nodes within hub networks.
2.5.5. GO Enrichment and KEGG Pathway Analysis
Gene IDs were obtained using the Omicshare platform (https://www.omicshare.com/). Subsequently, KEGG pathway enrichment analysis and GO functional annotation—including biological process (BP), molecular function (MF), and cellular component (CC) categories—were performed via the Metascape website (https://metascape.org/gp/index.html), with the species set to Rattus norvegicus. Results were filtered using a threshold of p < 0.05 and sorted in descending order based on the number of enriched genes.
2.6. General Observation Indicators Detection
During the administration, the condition of rats was observed every day, and the weight of rats before and after modeling and administration was recorded. The basic vaginal condition, including dryness, redness, congestion, exudate, and secretion, was also scored. The vaginal condition was assessed daily using a semi-quantitative scoring system (0-4) based on the following parameters: normal appearance (0); mild redness and slight swelling (1); moderate redness and swelling with visible secretions (2); severe redness, marked swelling, and purulent or bloody discharge (3-4). Scoring was performed independently by two researchers blinded to the treatment groups.
2.7. Hematoxylin & Eosin Staining
Following drug intervention, sodium pentobarbital was administered via intraperitoneal injection. Blood was collected from the heart into EP tubes, left at room temperature for 1 h, then centrifuged at 4°C, 3000 rpm for 15 min. Serum was collected and stored at -80 °C for later testing. After blood collection, the abdominal cavity was incised with ophthalmic scissors, ovarian tissue isolated, and the uterus cervix visually inspected. The tissue was formalin-fixed for paraffin embedding and H&E staining, with histopathological evaluation performed. One tissue sample was preserved in an ultra-low temperature freezer or liquid nitrogen for future use.
After formaldehyde fixation, the tissues were removed from the fixative, rinsed under running water, and subjected to ethanol gradient dehydration. The specimen underwent xylene clearing and paraffin embedding. A microtome was used to slice the paraffin-embedded specimens into 3 μm-thick sections. The prepared sections were mounted on a glass slide and baked at 60°C for 2 hours. Dehydration and hydration: The sections were first dehydrated in xylene I for 15 min. Subsequently, they were soaked in ethanol solutions (100%,90%,80%,70%) for 5 min each, followed by 3 min rinses with distilled water. The sections were then placed in hematoxylin stain for 5-10 min, rinsed under running water for 5 min, and treated with 1% ethyl alcohol hydrochloride (20 s) for 10 min of dehydration. The sections were stained in eosin solution for 3 min, rinsed under running water for 1 min, and underwent gradient dehydration using 80%, 90%, 95%, and 100% ethanol. The sections were finally cleared in xylene I for 15 min and xylene II for 15 min. Neutral resin was applied to the tissue, followed by mounting with coverslips. The specimens were observed under an optical microscope after drying to observe histological changes.
2.8. Detection the Levels of IL-1β, IL-6 and TNF-α In Serum by ELISA
The test was carried out according to the ELISA/biochemical kit instructions, and the contents of serum IL-1β, IL-6, TNF-α, and other factors were detected. Then, the levels of inflammatory cytokines were compared between groups.
2.9. Transcriptomic Analysis
Total RNA was extracted from cervical tissues treated with AF or control using TRIzol reagent. RNA integrity was verified by Nanodrop ND-2000 (Thermo Scientific, USA), and the RIN value of RNA was determined by Agilent Bioanalyzer 4150 (Agilent Technologies, CA, USA). The PE library was prepared in accordance with the instructions of ABclonal mRNA-seq Lib Prep Kit (ABclonal, China). The Agilent Bioanalyzer 4150 was used to assess library quality. Finally, RNA-seq was performed using the Illumina Novaseq 6000/MGISEQ-T7 platform. Data generated from the Illumina/BGI platform were used for bioinformatics analysis.
The DESeq2 software package (https://bioconductor.org/packages/release/bioc/html/DESeq2.html) was employed for inter-group gene differential expression analysis, with default screening thresholds set at |log2FC|> 1 and P <0.05 for differential expression genes. GO and KEGG enrichment analyses were conducted to elucidate functional enrichment patterns, revealing inter-sample differences at the gene function level. Using the clusterProfiler R package, GO functional enrichment and KEGG pathway enrichment analyses were performed, with significance (P <0.05) indicating substantial functional enrichment. For species included in the AnimalTFDB (https://bioinfo.life.hust.edu.cn/AnimalTFDB/) and PlantTFDB (https://planttfdb.cbi.pku.edu.cn/) databases, differential genes are annotated with transcription factors through gene ID-based screening. For species not covered by these databases, differential gene annotations are performed using Pfam (https://pfam.xfam.org/) annotation data combined with transcription factor family information from DBD (https://dblp.uni-trier.de/rec/journals/nar/KummerfeldT06.html). Protein-protein interaction (PPI) analysis investigates gene interactions by mapping protein associations to corresponding genes using STRING (https://www.string-db.org/). For database-registered species, network construction involves extracting target gene lists. For unregistered species, the process begins with blastX alignment of target gene sequences against reference species proteins in STRING’s interaction database, then establishing interaction networks based on these reference species’ biological relationships.
2.10. Molecular Docking
2D structural formula of core components absorbed into the serum of AF were downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov/) the website. In the PDB (https://www.rcsb.org/) database, select PDB format files of protein structures with high resolution and crystalline structures of the complexed original ligand. Use PyMol-2.5.2 software to remove water molecules and optimize the receptor protein structure. Import the component and target structure files into the AutoDockTools software (1.5.7). Click Edit> Remove Water to eliminate water molecules, then click Edit> Hydrogen> Add to add hydrogen atoms. Next, set the “Receptor” column to the name “receptor.pdbqt” and the “Ligand” column to “ligand.pdbqt”, converting both the protein and compound into pdbqt format. Configure molecular docking parameters, upload the receptor protein and ligand compound, set the docking box, and zoom in until they fully encapsulate the proteins. This process identifies the optimal binding position between the ligand and receptor, with the binding energy recorded at this optimal configuration. To validate the reliability of the docking protocol, the native ligand from each target protein’s crystal structure was extracted, re-docked into its respective binding site, and the root-mean-square deviation (RMSD) between the docked pose and the original crystallographic conformation was calculated. An RMSD value of < 2.0 Å was considered acceptable, confirming the robustness of our docking settings. All docking parameters use default settings. Finally, export the results using PyMol-2.5.2 software to visualize the docking outcomes and map interaction patterns.
2.11. Signal Pathway Key Target Validation Using Western Blotting
The signaling pathway was determined by combining the results of network pharmacology and transcriptomics detection and prediction, and the expression of key cervicitis targets was detected by western blotting. Western blotting analysis was performed on tissue lysates. Briefly, key target proteins were separated by SDS-PAGE, transferred to PVDF membranes, and probed with specific primary antibodies against p-AMPK, AMPK, p-mTOR, mTOR, HIF-1α (Affinity, Cat. AF3423, AF6423, AF3308, 81670-1-RR, AF1009, respectively) and Loading Control, GAPDH (Affinity, Cat. AF7021), followed by incubation with HRP-conjugated secondary antibodies. Gels were visualized using ECL reagent and quantified by densitometry.
2.12. Statistical Analysis
Experimental results are presented as mean ± standard deviation (SD). Data were processed using GraphPad Prism 10 (GraphPadSoftware LLC. USA). One-way ANOVA was employed to compare mean values across treatments, with Tukey’s test used for inter-group significance analysis. */#P <0.05; **/##P <0.01; ***/###P <0.001 indicates statistically significant results.
3. Results
3.1. Ingredients of AF Absorbed Into the Blood
UPLC-Q-Exactive-Orbitrap-MS analysis revealed that AF was predominantly composed of isoprene lipids (22.92%), organic oxides (12.4%), flavonoids (9.42%), and fatty acyl compounds (8.79%) (Figure 1A and B). Additionally, a total of 31 AF-derived compounds were identified in the serum samples (Figure 1C and Table 1), These serum-absorbed components, including gallic acid, glycyrrhizin, 7-hydroxycoumarin, hispanolone, curcumadionol, gamma-linolenic acid, stearidonic acid, aeruginolactone, and piceoside, were designated as candidate bioactive constituents and subsequently subjected to network pharmacology analysis. Ingredients AF extract and its absorption into the blood through UPLC-Q-Exactive-Orbitrap-MS. (A) Total ion chromatograms of AF, Medicated serum, and Blank serum (Bserum) in positive and negative ionization modes. (B) The relative content of compounds. (C) Venn diagram of components between different groups Characterization of Ingredients of AF Absorbed Into the Blood
3.2. Evaluation of the Efficacy of AF In Improving Cervicitis
A rat model of cervical inflammation was successfully established, and AF intervention was administered for 4 weeks (the animal experimental protocol is presented in Figure 2A). After treatment, the body weight loss induced by the modeling was significantly ameliorated (Figure 2B), and the vaginal clinical score was markedly reduced (Figure 2C). Further analysis of inflammatory cytokines IL-1β, IL-6, and TNF-α demonstrated that both AF-H and AF-L effectively inhibited the serum expression of these cytokines, which indicated that AF could attenuate the inflammatory response in cervical tissue (Figure 2D–F). Cervical pathological examination demonstrated that AF improved both inflammatory infiltration and tissue damage in the cervix (Figure 2G). AF ameliorates cervical inflammation in the SD rat model of cervicitis. (A) Flow chart illustrating the phenol paste induction method used to establish a cervicitis rat model. (B) Changes in rat weight. (C) Changes in rat vaginal score. (D–F) Levels of inflammatory factors IL-1β, IL-6, and TNF-α in rat serum. (G) H&E staining pathological rat cervix. The results were expressed as mean ± SD (n = 8), with statistical significance indicated as **P < 0.01
3.3. Network Pharmacology Prediction of Cervicitis of Blood-Entering Ingredients
To construct the compound–putative target network of AF, we first identified the 31 serum-absorbed active compounds of AF and their corresponding 644 putative targets, which were collected and visualized as an herb–compound–target network using Cytoscape 3.7.2. The resulting network consisted of 675 nodes and 1795 edges (Figure 3A). A total of 122,598 potential targets associated with cervicitis were retrieved from the GeneCards database. Considering the excessive number of retrieved targets, we further screened out the targets with a relevance score higher than 3, yielding 1670 valid targets. The intersection between these cervicitis-related targets and AF-related targets was then identified using a Venn diagram tool, yielding 188 overlapping targets, designated as overlapping genes (OGEs) (Figure 3B). These 188 OGEs were submitted to the STRING database for protein–protein interaction (PPI) analysis. The resulting PPI network contained 188 nodes and 4095 edges. Subsequent analysis via Cytoscape 3.7.2 software revealed the interaction network among the OGEs. Module analysis using the MCODE plugin identified the cluster module with the highest score (MCODE score = 48.034), which comprised 59 hub targets. This cluster was considered the key target module of AF for the treatment of cervicitis (Figure 3C). Using the CytoNCA plugin, we screened the top 5 nodes with the highest connectivity (i.e., Degree ≥ 58.0; Betweenness Centrality ≥ 20.615221; Closeness Centrality ≥ 1.0) as the core network, which included IL6, GAPDH, MTOR, STAT3, and CCND1 (Figure 3D). Subsequently, GO functional annotation and KEGG pathway analysis were performed on the 59 key genes using the Metscape database to further elucidate their biological roles. The top 20 enriched terms/pathways were visualized in a bubble plot (Figure 3E), which suggested that AF may alleviate cervicitis through the ErbB, FoxO, HIF-1, and MAPK signaling pathways. In terms of BP, AF was primarily associated with the positive regulation of cell migration and motility, cellular response to growth factor stimulus, and response to peptide hormone. Regarding MF, the active components of AF were mainly linked to protein kinase activity and protein kinase binding. For CC, the targets were closely related to the receptor complex, focal adhesion, and cell-substrate junction (Figure 3F). Network pharmacology analysis elucidating the mechanism of AF against cervicitis. (A) The herb-compound-target interaction network of AF. (B) Venn diagram identifying overlapping targets between AF and cervicitis. (C) PPI network of overlapping targets with the key module highlighted. (D) The PPI network of the selected core target. (E) KEGG pathway enrichment analysis. (F) GO enrichment analysis of core targets
3.4. Transcriptomics Analysis
Differential expression analysis among the Control, Model, and AF groups identified 123 differentially expressed genes (DEGs) (Figure 4A). Further pairwise comparisons were conducted between Control vs. Model and Model vs. AF groups, and the results were visualized as volcano plots and heatmaps (Figure 4B and C). Subsequent GO functional annotation and KEGG pathway analyses of these DEGs revealed that, after excluding pathways directly associated with specific diseases, the PPAR and AMPK signaling pathways were among the most significantly enriched pathways (Figure 4D and E). Transcriptomic analysis identifies AF-regulated DEGs and enriched pathways. (A) Identification of DEGs across groups. (B) Volcano plots of DEGs. (C) Heatmaps of DEGs. (D) GO enrichment analysis of DEGs. (E) KEGG pathway enrichment analysis of DEGs, highlighting the PPAR and AMPK signaling pathways (comparison groups: Model vs control, medel vs AF)
3.5. Molecular Docking of Key Targets and Components
The Molecular Docking Results
The molecular docking of 10 pairs of key-target and components
3.6. Verification of Key Targets of AMPK/mTOR/HIF-1α Signaling Axis
Western blot analysis demonstrated that in the model group, the phosphorylation level of AMPK was significantly downregulated, while phosphorylation level of mTOR and the protein expression of HIF-1α were markedly upregulated (Figure 6A). AF treatment effectively reversed these molecular alterations, significantly upregulating the protein level of p-AMPK (Figure 6B) and markedly downregulating the protein levels of both p-mTOR and HIF-1α (Figure 6C and D). These findings suggest that the amelioration of cervicitis by AF is associated with the modulation of the AMPK–mTOR–HIF-1α signaling axis. The detection of key proteins of the AMPK/mTOR/HIF-1α signal axis using western blotting. (A) Protein expression levels of p-AMPK, p-mTOR, and HIF-1α. (B and C) The intensity analysis of p-AMPK, p-mTOR, and HIF-1α was detected by western blot. The results were expressed as mean ± SD (n = 3). ##p < 0.01 vs control group; *p < 0.05, **p < 0.01 vs model group; ns, not significant between the indicated groups
4. Discussion
The present study confirms that AF exerts a significant ameliorative effect on experimental cervicitis, and we propose that the AMPK-mTOR-HIF-1α signaling axis acts as a core mechanistic node mediating its therapeutic effects. This discovery was enabled by our integrated research strategy that combined serum pharmacochemistry, network pharmacology, and tissue transcriptomics.
Initial network pharmacology analysis, based on serum-absorbed constituents of AF, predicted that its therapeutic action against cervicitis may involve key molecular targets including IL-6, mTOR, and STAT3, as well as critical signaling pathways such as HIF-1 and MAPK—both of which are closely associated with inflammatory responses and hypoxic stress. Concurrently, transcriptomic profiling of cervical tissue revealed that AF treatment significantly modulated genes enriched in the PPAR and AMPK signaling pathways, suggesting a role in metabolic-inflammatory crosstalk.
To reconcile these findings, we systematically investigated the well-characterized crosstalk between these signaling pathways. Published evidence indicates that: (1) PPARγ activation can upregulate AMPK activity, and AMPK can in turn enhance PPARγ function, thereby forming a reciprocal regulatory loop that modulates metabolic and inflammatory processes16,17; (2) AMPK is a well-recognized upstream negative regulator of mTOR; (3) both AMPK activation and mTOR inhibition synergistically converge to suppress the stabilization and transcriptional activity of HIF-1α18,19; (4) HIF-1α itself acts as a master transcriptional regulator of inflammatory responses and was identified as a core predicted target in our network pharmacology analysis; (5) cervicitis, as a chronic inflammatory gynecological disease, is typically associated with a local tissue microenvironment characterized by concurrent hypoxia and persistent inflammation20,21; (6) the activity of the HIF-1 signaling pathway is closely and positively correlated with the severity of cervical pathological lesions, 22 and inflammatory cytokines commonly transduce signals through the MAPK signaling pathway. 23 On the basis of this comprehensive integrative bioinformatics analysis, the AMPK-mTOR-HIF-1α signaling axis was identified as a putative convergent pathway that bridges the metabolic regulatory effects (PPAR/AMPK axis) revealed by transcriptomic profiling with the inflammatory and hypoxic response pathways (HIF-1/MAPK) predicted by network pharmacology. This axis was therefore selected as the key pathway for subsequent experimental validation.
The AMPK-mTOR-HIF-1α signaling axis represents a core regulatory pathway governing inflammatory responses, cellular energy homeostasis, and hypoxic stress adaptation. As a master cellular energy sensor. As a master cellular energy sensor, AMPK exerts an inhibitory effect on mTOR — a key modulator of cell growth, proliferation, and protein synthesis that is frequently dysregulated with hyperactivation in pro-inflammatory microenvironments. 24 Notably, activation of AMPK and inhibition of mTOR have both been demonstrated to attenuate the stability and transcriptional activity of HIF-1α.24,25 In inflammatory contexts, HIF-1α is stabilized synergistically by hypoxia, inflammatory stimuli, and reactive oxygen species (ROS); this stabilized HIF-1α then drives the transcription of various pro-inflammatory genes (e.g., TNF-α, IL-6, IL-1β), promotes glycolytic metabolism (the Warburg effect) in immune cells, and enhances vascular permeability, thereby exacerbating the inflammatory cascade. 26 In cervicitis, the inflamed and edematous tissue typically exhibits a localized hypoxic and nutrient-deprived microenvironment, which supports the notion that the AMPK-mTOR-HIF-1α axis serves as a key regulatory node in the pathogenesis of cervicitis. Our histological and cytokine profiling analyses consistent with this regulatory mechanism, demonstrating that AF treatment markedly reduced inflammatory cell infiltration and downregulated serum levels of TNF-α, IL-6, and IL-1β, findings that are consistent with previous research on anti-cervicitis therapeutic agents. 27
It is critical to interpret these findings in the context of the experimental model employed in the present study. The phenol-induced cervicitis model is a well-recognized standardized chemical irritant model that robustly recapitulates key features of inflammation and tissue damage, providing a controllable system for studying anti-inflammatory mechanisms. Although this model does not mimic the infectious etiologies (e.g., bacterial, viral) underlying most cases of clinical cervical inflammation in humans, the core inflammatory signaling pathways involved (e.g., pro-inflammatory cytokine secretion, immune cell infiltration) are highly conserved across different pathological contexts. The AMPK-mTOR-HIF-1α axis identified in our study is a central regulatory node of cellular metabolism and inflammatory response, which can be activated by a variety of stimuli including infection, hypoxia, and sterile injury. Thus, AF-mediated regulation of this axis may have broad relevance for alleviating the inflammatory component in different subtypes of cervicitis, although its therapeutic efficacy in pathogen-specific cervical inflammatory contexts warrants further in-depth investigation.
We acknowledge several limitations of the present study. First, the identification of AF-derived serum-absorbed components was solely qualitative; while these components are likely to contribute to the observed therapeutic effects, definitive confirmation of their individual bioactivities and mechanistic roles requires future quantitative pharmacokinetic and pharmacodynamic studies, as well as in vitro and in vivo validation using pure phytochemical compounds. Second, the molecular docking results, while generating valuable ligand-target interaction hypotheses, require further experimental validation via techniques including competitive binding assays and cellular target engagement assays to confirm their biological relevance and functional significance in cervical tissue. Third, the proposed causal role of the AMPK-mTOR-HIF-1α axis, while supported by correlative protein expression data and pathway prediction, awaits functional validation through rescue experiments (e.g., using AMPK inhibitors or mTOR agonists). Such mechanistic perturbation assays are fundamental to definitively establishing this signaling axis as the primary therapeutic target mediating AF’s effects in cervicitis and will represent a key focus of our subsequent research. Accordingly, future work should prioritize: (1) identifying the specific bioactive compound(s) within AF responsible for modulating the axis; (2) performing functional validation of the axis using genetic or pharmacological tools; (3) evaluating efficacy in infection-associated cervicitis models; and (4) conducting systematic preclinical pharmacokinetic and safety assessments.
5. Conclusion
In conclusion, the present study definitively demonstrates that AF extract exerts a potent therapeutic effect in ameliorating experimental cervicitis in rats. By integrating serum pharmacochemistry, network pharmacology, transcriptomics, and experimental validation, we have constructed a multi-faceted understanding of its action. Our findings identify the AMPK-mTOR-HIF-1α signaling axis as a core convergent regulatory pathway that mediates the potent anti-inflammatory and cervical tissue-protective effects of AF. This work bridges traditional knowledge with modern molecular pharmacology, providing a robust mechanistic foundation for the application of AF in inflammatory gynecological conditions and identifying a promising target axis for future therapeutic development.
Footnotes
Ethical Considerations
This study was approved by the Zhejiang Huitong Testing and Evaluation Technology (Group) Co., LTD (Approval Number: HTDW-202502009).
Author Contributions
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was partially supported by the Jiangxi Provincial Department of Education Science and Technology Research Project (Project No: 191640); National Inheritance Studio for Veteran Chinese Pharmacy Experts (Grant No. 2024-255 from the National Administration of Traditional Chinese Medicine).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Statement of Human and Animal Rights
All institutional and national guidelines for the care and use of laboratory animals were followed. This article does not contain any studies with human participants performed by any of the authors.