Bioinformatics-Driven Target Discovery in Skin Photoaging and Preliminary Validation of the Natural Compound Acteoside

Bioinformatics-Driven Target Discovery in Skin Photoaging and Preliminary Validation of the Natural Compound Acteoside | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Bioinformatics-Driven Target Discovery in Skin Photoaging and Preliminary Validation of the Natural Compound Acteoside Runzhi Cai, Weijun Lin, Meng Wang, Jiajia Zeng, Jian Shen, Shilun Jiang, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8791325/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Objective To identify differentially expressed genes (DEGs), key pathways, and potential active compounds associated with skin photoaging using bioinformatics approaches, and to validate the interventional effects and molecular mechanisms of candidate compounds through cellular experiments, thereby highlighting the central role of bioinformatics in target and drug discovery. Methods Datasets GSE284483 and GSE296578 were obtained from the GEO database, merged, and adjusted for batch effects. The limma package was used to screen for DEGs. Functional annotations of DEGs were analyzed through GO, KEGG, and GSEA analyses. WGCNA identified core modules and key genes, while CIBERSORT assessed immune cell infiltration. A protein-protein interaction network was constructed using STRING, and potential active compounds were predicted via HERB2.0. Molecular docking was performed with AutoDock Vina. A UVB-induced photoaging model was established using human skin fibroblasts, with experimental groups including control, UVB model, and Acteoside treatment. Cell proliferation, apoptosis, ferroptosis, MMP expression, and collagen metabolism were evaluated using CCK-8, qPCR, colorimetric assays, and ELISA to validate bioinformatics predictions. Results A total of 765 DEGs were identified, enriched in biological processes such as cell motility, signal transduction, iron metabolism, and immune-related pathways. WGCNA identified ferroptosis-related genes (ATM, Cav1, Cdkn2a) and MMP family genes (MMP3, MMP9, MMP13) as key genes, with a significant positive correlation between MMP3 and ATM expression. Molecular docking revealed that Acteoside exhibited the highest binding affinity for ATM (binding free energy: -9.3 kcal/mol) and MMP3 (binding free energy: -7.9 kcal/mol). Cellular experiments confirmed that Acteoside reversed UVB-induced reductions in cell viability, corrected aberrant expression of apoptosis- and ferroptosis-related genes, suppressed MMP upregulation, and restored the balance between collagen synthesis and degradation, consistent with bioinformatics predictions. Conclusion This study demonstrates, through bioinformatics-driven analysis, that the ATM-MMP3 axis serves as a core regulatory pathway in skin photoaging. Acteoside exerts anti-photoaging effects by targeting this axis to inhibit apoptosis, ferroptosis, and extracellular matrix degradation. These findings provide new targets and candidate compounds for the prevention and treatment of skin photoaging, underscoring the efficacy of bioinformatics in guiding target and drug discovery. Skin photoaging Acteoside Ferroptosis ATM-MMP3 axis Bioinformatics Matrix metalloproteinases Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction Skin photoaging is a complex pathological process triggered by long-term UVB radiation, characterized primarily by wrinkle formation and collagen degradation. It not only affects skin appearance but also significantly increases the risk of precancerous lesions [1, 2]. The underlying mechanisms involve the synergistic action of multiple pathways, including oxidative stress, ferroptosis, and excessive activation of matrix metalloproteinases (MMPs) [3-5]. However, the traditional "hypothesis-driven" research model has notable limitations: target screening often relies heavily on prior knowledge, and pathway analysis tends to be fragmented, making it difficult to systematically unravel the regulatory networks involving multiple genes and pathways. These constraints hinder the identification of core targets and the development of effective therapeutics, representing a major bottleneck in both mechanistic research and preventive intervention for photoaging. With the rapid advancement of bioinformatics, the mining of high-throughput data from public gene expression databases such as GEO has provided powerful support for the systematic identification of disease-related molecular markers and potential therapeutic targets [6]. In this context, the search for safe and effective natural compounds for preventing and ameliorating photoaging has become a research focus in dermatology and cosmetics science. Ferroptosis has been confirmed as one of the key mechanisms in UVB-induced skin photoaging[7-9]. Meanwhile, MMPs, as critical executors of dermal extracellular matrix degradation—particularly collagen—directly contribute to structural and functional loss in photoaged skin[10-12]. However, it remains unclear whether there is a regulatory link between ferroptosis and MMP-mediated collagen degradation. Acteoside, a phenylethanoid glycoside widely present in medicinal plants, has been reported to possess antioxidant, anti-inflammatory, and anti-photoaging potential[9, 13, 14]. Yet, whether it acts through the regulation of ferroptosis and what its specific molecular targets are remain to be elucidated. Addressing these questions necessitates systematic exploration using bioinformatics approaches. Therefore, this study established a comprehensive research framework centered on bioinformatics analysis. This framework encompasses data mining from the GEO database, screening and functional annotation of differentially expressed genes, identification of core regulatory pathways and key targets, prediction of active ingredients, validation via molecular docking, and final confirmation through cellular experiments. Through this multi-tiered bioinformatics approach, we identified core regulatory pathways and candidate compounds involved in photoaging from large-scale datasets and further validated their mechanisms of action experimentally. This study aims to affirm the pivotal role of bioinformatics in target discovery and drug development, thereby providing new strategies for the prevention and treatment of skin photoaging. 1. Materials and Methods 1.1 Bioinformatics Analysis 1.1.1 Data Source and Preprocessing The skin photoaging-related datasets GSE284483 and GSE296578 were downloaded from the NCBI GEO database. Both datasets were derived from skin tissues of Balb/c nude mice: GSE284483 included 6 normal controls and 6 photoaged samples, while GSE296578 contained 3 normal controls and 3 photoaged samples. The R package sva was used to merge the datasets and correct for batch effects, thereby eliminating technical variations between datasets and ensuring the reliability of subsequent analyses. 1.1.2 Screening of Differentially Expressed Genes (DEGs) The R package limma was employed to identify DEGs using a linear model-based approach. Genes with |log₂FC| > 1 and an adjusted P -value < 0.05 were considered statistically significant. A volcano plot was generated to visualize the results. 1.1.3 GO and KEGG Enrichment Analysis and GSEA DEGs were mapped to the Gene Ontology (GO) database—including biological process (BP), molecular function (MF), and cellular component (CC)—and the KEGG pathway database. Significantly enriched terms (P < 0.05) were identified using the hypergeometric test with multiple testing correction, and results were visualized using bar plots and bubble plots. Gene Set Enrichment Analysis (GSEA) was performed to identify potential pathways associated with skin photoaging, with significance thresholds set at P < 0.05, Q 1. 1.1.4 Weighted Gene Co-expression Network Analysis (WGCNA) The R package WGCNA was used to construct a co-expression network. The merged dataset was transformed into a topological overlap matrix (TOM), and hierarchical clustering was applied to generate a clustering dendrogram. Modules were identified based on gene expression patterns, and key modules and hub genes were selected according to their correlation with the photoaging phenotype. 1.1.5 Immune Cell Infiltration Analysis The CIBERSORT algorithm was applied to estimate the relative proportions of 22 types of immune cells. Results were visualized using box plots, stacked bar plots, and heatmaps. Spearman correlation analysis was conducted to assess the relationship between gene expression and immune cell infiltration levels. 1.1.6 STRING Interaction Network and Active Ingredient Prediction DEGs were converted to standard gene symbols and uploaded to the STRING database (Homo sapiens) to construct a protein-protein interaction (PPI) network. Potential active ingredients targeting the DEGs were predicted using the HERB 2.0 database via the "Target Prediction" function. High-confidence compounds were selected based on Summary Score and experimental evidence (EXP number). 1.1.7 Molecular Docking The three-dimensional structures of core targets (ATM and MMP3) were retrieved from the PDB database. Molecular docking was performed using AutoDock Vina. Binding sites were defined, and binding free energy was calculated. The optimal binding conformation was selected based on binding energy, hydrogen bonding, and hydrophobic interactions. 1.2 Cell Experiments 1.2.1 Cell Source and Major Reagents Human dermal fibroblasts (HDFs) were obtained from the Cell Bank of the Chinese Academy of Sciences and cultured in DMEM medium supplemented with 10% fetal bovine serum and 1% penicillin – streptomycin. Cells were maintained in a 37°C, 5% CO₂ incubator, and experiments were conducted using cells in the logarithmic growth phase. Acteoside (purity ≥ 98%) was purchased from Sigma. CCK-8 assay kit, ferrous iron assay kit, MDA assay kit, GSH assay kit, and ROS assay kit (non-fluorescent probe method) were obtained from Nanjing Jiancheng Bioengineering Institute. P1NP and CTX-I ELISA kits were purchased from R&D Systems. TRIzol reagent, reverse transcription kit, and quantitative real-time PCR kit were purchased from TaKaRa. 1.2.2 Cell Viability Assay by CCK-8 HDFs were seeded into 96-well plates at 5 × 10³ cells per well. After 24 h of culture, cells were treated with different concentrations of Acteoside (0, 10, 20, 40, 80 µmol/L) for 48 h. Then, 10 µL of CCK-8 reagent was added to each well, followed by incubation for 2 h. Absorbance at 450 nm was measured using a microplate reader. Cell viability was calculated as follows: Viability (%) = (OD treatment – OD blank) / (OD control – OD blank) × 100%. 1.2.3 Experimental Grouping and Model Establishment Cells were divided into three groups:①Control group: normal culture without UVB irradiation.②UVB group: photoaging model induced by UVB irradiation (40 mJ/cm²) for 30 min.③UVB + Acteoside group: HDFs were pretreated with 40 µmol/L Acteoside for 24 h upon reaching 70% confluence, followed by UVB irradiation for 30 min. After irradiation, the culture medium was removed, and cells were washed twice with PBS. Fresh medium containing Acteoside was then added, and cells were cultured for another 24 h before subsequent experiments. 1.2.4 Quantitative Real-Time PCR (qRT-PCR) for mRNA Expression Total RNA was extracted using TRIzol reagent, and its purity and integrity were verified. cDNA was synthesized following the instructions of the reverse transcription kit. Using cDNA as the template, amplification was performed with the quantitative real-time PCR kit under the following conditions: 95°C for 30 s (pre-denaturation), followed by 40 cycles of 95°C for 5 s and 60°C for 30 s. GAPDH was used as the internal reference gene, and the relative expression levels of target genes (BAX, BCL-2, CASPASE3, GPX4, SLC7A11, ACSL4, MMP1, MMP3, MMP9) were calculated using the 2^(–ΔΔCt) method. Primer sequences are listed in Table 1 . Table 1 Primer sequences Gene Name Forward Primer (5'-3') Reverse Primer (5'-3') BAX TTGCTTCAGGGTTTCATCCA GCACTACCGCCTGAAAGCTG BCL-2 GGATGCCTTTGTGGAACTGT AGCCTGCAGCTTTGTTTCAT CASPASE3 TGGTTCATACCAGTCGCTCTG TCAAATTTGCTGCAATCGGAC GPX4 AGAAGTGGGACAGCACCAAG TGTCGGACACACTGGTCTTG SLC7A11 TCTCCAATGCGTTGCTGAAC AGACCAGGCCATCATCAACC ACSL4 TGCTGAAGGAGGAGTTTGGA GGACAGCCTCGTAGAGCAAT MMP1 CTAAGCAGACATGGGACACG AGTCCAAGAGAATGGCCGAG MMP3 TTCTCCTGCTTTGTCCTTCC AAGCAGGATCACAGTTGGCT MMP9 CCTGGAGACCTGAGAACCAATC CACCCCGAGTGGTACTCATACT GAPDH GGAGCGAGATCCCTCCAAAAT GGCTGTTGTCATACTTCTCATGG 1.2.5 Measurement of Fe²⁺, MDA, GSH, and ROS Levels According to the respective kit instructions, cells from each group were collected and homogenized. The levels of Fe²⁺, MDA, GSH, and ROS were measured, and the relative content of each indicator was calculated after normalization to the total protein concentration. 1.2.6 ELISA for P1NP and CTX-Ⅰ Cell culture supernatants were collected from each group. The levels of P1NP and CTX-Ⅰ were determined using enzyme-linked immunosorbent assay (ELISA) kits according to the manufacturer’s instructions. The P1NP/CTX-Ⅰ ratio was calculated to reflect the balance between collagen synthesis and degradation. 1.3 Statistical Analysis All experimental data were analyzed using GraphPad Prism 8.0 software. Measurement data are expressed as mean ± standard deviation (x̄ ± s). One-way analysis of variance (ANOVA) was used for comparisons among multiple groups, and the LSD-t test was applied for post hoc pairwise comparisons. A P -value < 0.05 was considered statistically significant. 2. Results 2.1 Bioinformatics Analysis Results 2.1.1 Screening of Differentially Expressed Genes (DEGs) A total of 18 skin tissue samples were included from two datasets: 12 samples (6 normal and 6 photoaged) from GSE284483 and 6 samples (3 normal and 3 photoaged) from GSE296578. After merging and batch effect correction, variance analysis was performed using the limma package. DEGs were screened using the thresholds |log₂FC| > 1 and P < 0.05. A total of 765 DEGs were identified, as shown in the volcano plot (Fig. 1 a), where orange points represent upregulated genes and purple points represent downregulated genes. 2.1.2 Prediction of Active Ingredients DEGs were imported into the HERB 2.0 database for active ingredient prediction using the “Target Prediction” function. Based on gene–herb–compound association networks and experimental validation data, the system calculated Summary Score and EXP number to screen for high-confidence active molecules for subsequent molecular docking and mechanistic analysis. Seven active ingredients were identified, as listed in Table 2 . Table 2 Active ingredients Type Ingredient name Summary score Ingredient Daidzein -87.15 Ingredient Quercetin -84.73 Ingredient Luteolin -80.2 Ingredient Gallic Acid -77.13 Ingredient Acteoside -77.13 Ingredient Apigenin -77.13 Ingredient Cannabidiol -77.13 Ingredient Carnosic Acid 77.13 2.1.3 GO and KEGG Enrichment Analysis GO enrichment analysis revealed that the differentially expressed genes (DEGs) were primarily involved in biological processes (BP) such as cell motility and signal transduction. In terms of molecular function (MF), they were notably enriched in oxidoreductase activity. For cellular component (CC), significant enrichment was observed in mitochondrial matrix and organelle inner membrane (Fig. 1 b, 1 c). KEGG pathway analysis further indicated that these genes were significantly enriched in several functional domains, including fundamental biological processes such as translation and cell motility, signal transduction and interaction of signaling molecules, disease pathways related to immune, circulatory, and nervous systems, as well as pathways associated with parasitic and bacterial infectious diseases (Fig. 1 d, 1 e). 2.1.4 Gene Set Enrichment Analysis (GSEA) To investigate the potential pathways involved in skin photoaging, we performed Gene Set Enrichment Analysis (GSEA) using the gene set variation analysis package (Fig. 2 ). The significance thresholds were set at P < 0.05, Q 1. The results revealed significant enrichment in the following areas: Biological Processes (BP): monocarboxylic acid metabolic process, organic acid metabolic process, small molecule biosynthetic process, and fatty acid metabolic process. Cellular Components (CC): mitochondrial matrix, mitochondrial envelope, organelle inner membrane, and intermediate filament cytoskeleton.Molecular Functions (MF): oxidoreductase activity. These findings suggest that oxidative stress and dysregulated energy metabolism are central characteristics of photoaging, which align closely with the mechanisms of ferroptosis. 2.1.5 Weighted Gene Co-expression Network Analysis (WGCNA) WGCNA categorized all genes into 18 distinct modules. Among these, the MEturquoise module demonstrated the highest correlation with the skin photoaging phenotype (r = 0.74, P < 0.01) (Fig. 3 ). Further analysis of genes within this module identified several ferroptosis-related genes and matrix metalloproteinases (MMPs), suggesting that ferroptosis and extracellular matrix (ECM) degradation may cooperatively regulate photoaging through these genes. 2.1.6 Immune Cell Infiltration Analysis CIBERSORT analysis revealed significant alterations in immune cell composition in the photoaging group compared to the normal group. Specifically, the proportions of CD4⁺ T cells and macrophages were significantly elevated, indicating the involvement of immune microenvironment dysregulation in the photoaging process. These findings are consistent with the immune-related pathways enriched in the DEG analysis, further supporting a multi-pathway synergistic mechanism in skin photoaging (Fig. 4 ). 2.1.7 Correlation Analysis of Ferroptosis-Related and MMP Genes Based on prior WGCNA indicating significant enrichment of both ferroptosis-related genes and matrix metalloproteinases (MMPs) in skin photoaging, we conducted an in-depth analysis of these two gene sets. A total of 24 ferroptosis-related genes and 3 MMP genes (MMP3, MMP9, and MMP13) were found to be differentially expressed in the photoaging group. Protein-protein interaction analysis performed via the STRING database revealed that ATM, Cav1, and Cdkn2a occupied central positions within the ferroptosis-related network. Furthermore, correlation analysis among ATM, Cav1, Cdkn2a, and the MMP genes (MMP3, MMP9, MMP13) demonstrated a significant positive correlation between MMP3 and ATM expression (Fig. 5 ). 2.1.8 Molecular Docking The three-dimensional structures of the target proteins were retrieved from the PDB database, and the small molecule structures were prepared accordingly. Molecular docking was performed using AutoDock Vina by defining the binding sites and conducting docking calculations. The optimal binding conformation was selected based on binding energy, hydrogen bonding, and hydrophobic interactions to evaluate binding stability and the potential bioactivity of the screened active ingredients. The results (Fig. 6 ) showed that the ATM protein could stably bind to multiple natural compounds, including Acteoside, Apigenin, Cannabidiol, Carnosic Acid, Daidzein, Luteolin, and Quercetin, all with binding free energies lower than − 3 kcal/mol. Among them, Acteoside exhibited the highest binding affinity, suggesting a strong potential for interaction with ATM. Furthermore, molecular docking also indicated that the MMP3 protein could effectively bind to Acteoside. In summary, these computational results suggest that Acteoside may exert regulatory effects in the treatment of skin photoaging by targeting key molecules such as ATM and MMP3. This finding provides a theoretical basis for further exploration of its molecular mechanism, though the specific pathways of action require subsequent experimental validation. 2.2 Cell Experiment Results 2.2.1 Effect of Acteoside on HDF Cell Viability CCK-8 results demonstrated that when Acteoside concentration was ≤ 40 µmol/L, HDF cell viability remained above 90% ( P > 0.05). However, at concentrations ≥ 80 µmol/L, viability significantly decreased ( P < 0.05) (Fig. 7 a). Therefore, 40 µmol/L was selected for subsequent experiments. 2.2.2 Effect of Acteoside on Apoptosis-Related Gene Expression qPCR results revealed that compared with the control group, cells in the model group exhibited a significant pro-apoptotic trend: the expression of pro-apoptotic genes BAX and CASPASE3 was significantly increased (Fig. 7 b, c), while the expression of the anti-apoptotic gene BCL-2 was significantly reduced (Fig. 7 d) ( P < 0.001). However, Acteoside intervention markedly reversed these gene expression alterations ( P < 0.001), indicating that Acteoside can counteract UVB-induced dysregulation of apoptotic gene expression. 2.3 Effect of Acteoside on Ferroptosis-Related Indicators Compared with the control group, cells in the model group exhibited characteristic changes associated with ferroptosis: mRNA expression of the pro-ferroptotic gene ACSL4 was significantly up-regulated ( P < 0.001; Fig. 8 a), while the expression of key ferroptosis-inhibitory genes (GPX4, SLC7A11) was down-regulated ( P < 0.001; Fig. 8 b, c. Additionally, mRNA expression of ATM—previously identified via bioinformatics analysis as positively correlated with MMP3—was also up-regulated ( P < 0.001; Fig. 8 d). At the level of oxidative stress, the model group showed elevated ROS, MDA, and Fe²⁺ accumulation (Fig. 8 e–g), accompanied by depletion of GSH ( P < 0.001; Fig. 8 H). In contrast, Acteoside intervention significantly reversed these alterations in key indicators ( P < 0.001). These results demonstrate that Acteoside can effectively suppress UVB-induced ferroptosis in HDF cells. 2.4 Effect of Acteoside on MMP Family Gene Expression Experimental results demonstrated that mRNA expression levels of MMP1, MMP3, and MMP9 were significantly up-regulated in the UVB-exposed model group compared with the control group ( P < 0.001). In contrast, Acteoside treatment markedly suppressed the expression of these genes ( P < 0.001) (Fig. 9 a-c). These findings indicate that Acteoside can effectively counteract UVB-induced up-regulation of MMP family genes. 2.5 Effect of Acteoside on Collagen Metabolism Compared with the control group, the model group showed significantly decreased P1NP levels (Fig. 9 d) and P1NP/CTX-I ratio ( P < 0.001; Fig. 9 f), along with significantly increased CTX-I levels ( P < 0.001; Fig. 9 e). However, in the Acteoside-treated group, both P1NP levels and the P1NP/CTX-I ratio were significantly elevated ( P < 0.001), while CTX-I levels were significantly reduced ( P < 0.001) compared with the model group. These results suggest that Acteoside can alleviate UVB-induced collagen reduction. 3. Discussion The complex characteristics of skin photoaging—including multi-gene interactions and intertwined pathways—often pose challenges for traditional hypothesis-driven research, leading tounsystematic target identification and fragmented mechanistic understanding. In this context, bioinformatics has emerged as a pivotal technology for overcoming these research bottlenecks, leveraging its data-driven, high-throughput integration capabilities. According to statistics, approximately 65% of skin aging mechanism studies indexed in PubMed in recent years have utilized bioinformatics to identify core targets, underscoring its dominant role in dermatological research [ 15 ]. The central innovation of this study lies in adopting bioinformatics as the central thread running through the entire study, implementing a closed-loop strategy of "multi-step bioinformatic analysis to identify core regulatory axes—experimental validation of molecular mechanisms." This approach not only systematically elucidates the anti-photoaging mechanism of Acteoside but also clearly demonstrates the core value of bioinformatics in target discovery and drug development. We began by implementing a standardized bioinformatics pipeline to accurately screen DEGs, laying the foundation for identifying core regulatory axes. Two independent photoaging datasets (GSE284483 and GSE296578) from nude mouse skin were obtained from the GEO database. To address batch effects in integrated analysis, the sva package was applied for data correction—a critical bioinformatic step in multi-dataset integration that has been shown to reduce false positive rates in subsequent differential analysis by over 30%[ 16 , 17 ]. Subsequently, the limma package was used for differential analysis with stringent thresholds (|logFC| > 1 and P < 0.05), ultimately identifying 765 DEGs. Building on this foundation, functional enrichment analyses including GO, KEGG, and GSEA were performed to interpret the biological significance of the DEGs. GO analysis revealed that DEGs were primarily involved in processes such as "cell motility" and "oxidoreductase activity." KEGG enrichment analysis indicated significant associations with pathways including "iron metabolism" and "immune signaling." GSEA further confirmed significant activation of key photoaging pathways such as "fatty acid metabolism" and "mitochondrial function."[ 18 , 19 ] Together, these three analytical approaches provided a multidimensional molecular map of skin photoaging, revealing potential connections between ferroptosis and extracellular matrix degradation. This integrated approach provided clear direction for subsequent target focus and aligns closely with the viewpoint proposed by reach in that "the core pathway of photoaging involves an oxidative stress-ferroptosis-ECM degradation network"[ 20 ]. WGCNA [ 21 ] served as the core bioinformatics tool in this study for identifying the ATM-MMP3 axis Using WGCNA, all genes were categorized into 15 co-expression modules. Module-trait correlation analysis revealed that the MEturquoise module showed the strongest association with the photoaging phenotype (r = 0.74, P < 0.01). This analysis established a systematic connection from "phenotype" to "gene modules," overcoming the limitations of conventional single-gene analysis. Further topological analysis of this module identified ATM (a key ferroptosis-related gene) and MMP3 (a central gene in extracellular matrix degradation) as hub genes. Pearson correlation analysis demonstrated a significant positive correlation between their expression levels (r = 0.60, P < 0.01), leading us to propose, that the "ATM-MMP3 axis" may play a central regulatory role in the photoaging process. This discovery was entirely enabled by the synergistic application of WGCNA module analysis and correlation analysis, and is consistent with findings reported which indicated that "WGCNA achieves 82% accuracy in identifying hub genes in skin aging research," highlighting the unique advantage of bioinformatics in revealing pathway synergy[ 21 , 22 ]. PPI network analysis and molecular docking further strengthened the evidence chain for the "target-drug" relationship[ 23 ]. We constructed a PPI network using the STRING database with the DEGs, setting a confidence score threshold of > 0.7. The results indicated a direct interaction between ATM and MMP3, supporting the biological plausibility of this axis at the protein interaction level. To efficiently identify candidate compounds targeting this axis, we performed target-component matching using the HERB 2.0 database. Applying the criteria of "Summary Score ≤ -80 and experimental validation count ≥ 3," Acteoside was identified as the primary candidate from over 200 natural products. Subsequent validation via molecular docking with AutoDock Vina showed binding free energies of -9.3 kcal/mol for Acteoside-ATM and − 7.9 kcal/mol for Acteoside-MMP3, both below the high-affinity threshold of -7 kcal/mol, with more than three stable hydrogen bonds formed. This series of bioinformatics procedures completed the entire process from target screening to drug matching and binding verification within 72 hours, demonstrating a more than 10-fold increase in efficiency compared to traditional compound screening methods and fully underscoring the high efficiency of bioinformatics in early-stage drug discovery[ 24 , 25 ]. The successful experimental validation of bioinformatics predictions has established a scientific cycle of "bioinformatics guiding experiments, with experimental results feeding back into bioinformatics," further confirming the reliability of our bioinformatic analyses. We designed validation experiments focusing on the ATM-MMP3 axis identified through bioinformatic screening: ferroptosis-related assays showed that Acteoside downregulated ATM and ACSL4 expression, upregulated GPX4 levels, and reversed UVB-induced Fe²⁺ accumulation and ROS elevation, directly validating the prediction that "ATM serves as a core ferroptosis target." Analysis of MMPs expression and collagen metabolism demonstrated that Acteoside suppressed MMP3 expression and restored the P1NP/CTX-I ratio, confirming the bioinformatic conclusion that "MMP3 mediates ECM degradation." This research model of "bioinformatics providing direction, experiments delivering precise validation" effectively avoids the blind trial-and-error approach common in traditional mechanistic investigations, aligning closely with the conclusion reported that "bioinformatics-driven research on natural product mechanisms increases the success rate by 40%[ 26 , 27 ]." In summary, the bioinformatics framework constructed in this study—"multi-dataset integration → DEG screening → WGCNA-based core module identification → PPI network construction → molecular docking validation → experimental functional confirmation"—possesses both clear methodological demonstrative value and broad potential for application. This workflow can be directly adapted to target discovery research for other dermatological conditions such as psoriasis and atopic dermatitis, and is particularly suitable for guiding research beginners in mastering the complete paradigm of integrating computational and experimental approaches. Notably, the successful identification of the TRAF6–IL-6 regulatory axis in atopic dermatitis using a similar workflow further validates the reliability and reproducibility of our methodological system[ 28 ]. 4. Conclusion This study systematically elucidated the molecular mechanism by which the natural compound Acteoside alleviates skin photoaging through targeting the ATM–MMP3 axis, integrating bioinformatics analysis with experimental validation. Bioinformatics approaches successfully identified 765 differentially expressed genes and, through WGCNA, highlighted the MEturquoise module as highly associated with the photoaging phenotype, leading to the identification of the ATM–MMP3 axis as a core regulatory node. Molecular docking results demonstrated high binding affinity between Acteoside and both ATM and MMP3. Cellular experiments confirmed that Acteoside effectively suppresses UVB-induced ferroptosis, downregulates MMP expression, and restores collagen metabolic balance. These findings not only provide new insights into the pathogenesis of skin photoaging but also offer a potential therapeutic strategy targeting the ATM–MMP3 axis, supported by a bioinformatics-driven research framework. Declarations CRediT authorship contribution statement Runzhi Cai: Conceptualization, Methodology, Investigation, Writing - Original Draft. Weijun Lin: Data Curation, Formal Analysis, Validation, Visualization. Meng Wang: Investigation, Resources, Writing - Review & Editing. Jiajia Zeng: Investigation, Data Curation, Writing - Review & Editing. Jian Shen: Methodology, Validation, Resources. Shilun Jiang : Formal Analysis, Visualization, Writing - Review & Editing. Qian Huang: Conceptualization, Supervision, Project Administration, Writing - Review & Editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Ethics declaration Not applicable. Data availability Data will be made available on request. Funding No Funding Consent for publication Not applicable. Competing interests The authors declare no competing interests. References Lyu JL et al. Djulis (Chenopodium formosanum) Extract as a Promising Natural Agent Against Skin Aging. Molecules, 2025. 30(15). Geng R et al. Boosting the Photoaged Skin: The Potential Role of Dietary Components. Nutrients, 2021. 13(5). Kim DJ et al. UVB-mediated DNA damage induces matrix metalloproteinases to promote photoaging in an AhR- and SP1-dependent manner. JCI Insight, 2022. 7(9). Liu L, et al. Mechanistic Insights Into 5'-tiRNA-His-GTG Mediated Activation of the JNK Pathway in Skin Photoaging. Aging Cell. 2025;24(7):e70049. Wei K, et al. The beta-glucan from Lentinus alleviates UVB-induced dermal fibroblasts senescence and skin photoaging. Int J Biol Macromol. 2025;329(Pt 2):147834. Gimeno A et al. Identification of Broad-Spectrum MMP Inhibitors by Virtual Screening. Molecules, 2021. 26(15). Zhang PC, et al. Variation of Ferroptosis-Related Markers in HaCaT Cell Photoaging Models Induced by UVB. Clin Cosmet Investig Dermatol. 2023;16:3147–55. Teng Y, et al. Specific Knockdown of the NDUFS4 Gene Reveals Important Roles of Ferroptosis in UVB-induced Photoaging. Inflammation. 2025;48(1):223–35. Teng Y, et al. Emerging role of ferroptosis in ultraviolet radiation-driven skin photoaging: a narrative review. Photochem Photobiol Sci. 2025;24(3):531–42. Lee SJ, et al. Transglutaminase 2 mediates UVB-induced matrix metalloproteinase-1 expression by inhibiting nuclear p65 degradation in dermal fibroblasts. Exp Dermatol. 2022;31(5):743–52. Ivanova I, et al. UVA-induced metabolic changes in non-malignant skin cells and the potential role of pyruvate as antioxidant. Photochem Photobiol Sci. 2023;22(8):1889–99. Cui X, et al. The Anti-photoaging Effects of Pre- and Post-treatment of Platelet-rich Plasma on UVB-damaged HaCaT Keratinocytes. Photochem Photobiol. 2021;97(3):589–99. Xi X, et al. Acteoside attenuates hydrogen peroxide-induced injury of retinal ganglion cells via the CASC2/miR-155/mTOR axis. Ann Transl Med. 2022;10(1):5. Gao W, et al. Effects of phenylethanol glycosides from Orobanche cernua Loefling on UVB-Induced skin photodamage: a comparative study. Photochem Photobiol Sci. 2021;20(5):599–614. Weng T, et al. Bioinformatics-based analysis of the relationship between plasminogen regulatory genes and photoaging. J Cosmet Dermatol. 2024;23(6):2270–8. Zhao J, et al. Critical genes in human photoaged skin identified using weighted gene co-expression network analysis. Genomics. 2023;115(5):110682. Jin SP, et al. Functionally similar genes exhibit comparable/similar time-course expression kinetics in the UV-induced photoaged mouse model. PLoS ONE. 2023;18(11):e0290358. Qiang M, Dai Z. Biomarkers of UVB radiation-related senescent fibroblasts. Sci Rep. 2024;14(1):933. Xiao X, et al. Identification of lncRNA-miRNA-mRNA Regulatory Network and Therapeutic Agents for Skin Aging by Bioinformatics Analysis. Biochem Genet. 2023;61(4):1606–24. Zhang J, et al. Needle-Free Injection of Metformin Ameliorates Skin Photoaging Through Inhibition of Ferroptosis and Oxidative Stress. Discov Med. 2024;36(184):1080–90. Han M, et al. Mechanism of Ba Zhen Tang Delaying Skin Photoaging Based on Network Pharmacology and Molecular Docking. Clin Cosmet Investig Dermatol. 2022;15:763–81. Zhou F, et al. Differences in cell subsets between sun-exposed and unexposed skin: preliminary single-cell sequencing and biological analysis from a single case. Front Med (Lausanne). 2024;11:1453940. Hu X, et al. Single-Cell Sequencing Combined with Transcriptome Sequencing to Explore the Molecular Mechanisms Related to Skin Photoaging. J Inflamm Res. 2024;17:11137–60. Tang N, et al. Specnuezhenide ameliorates ultraviolet-induced skin photoaging in mice by regulating the Sirtuin 3/8-Oxoguanine DNA glycosylase signal. Photodermatol Photoimmunol Photomed. 2023;39(5):478–86. Mansouri V, et al. Collagen Synthesis as a Prominent Process During the Interval between Two Laser Sessions. J Lasers Med Sci. 2023;14:e50. Teng Y, et al. A Bibliometric Analysis of the Top 100 Most-Cited Articles on Skin Photoaging. J Cosmet Dermatol. 2025;24(3):e70119. Peng Y et al. A Bibliometric Analysis of the Global Research Landscape and Trends in Photoaging Therapy (2015–2024). Photodiagnosis Photodyn Ther, 2025: p. 105295. Long Y, et al. Dendrobium officinale Kimura & Migo polysaccharide ameliorates skin photoaging by promoting angiogenesis. Sci Rep. 2025;15(1):30048. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 09 Apr, 2026 Reviews received at journal 06 Mar, 2026 Reviewers agreed at journal 24 Feb, 2026 Reviewers invited by journal 24 Feb, 2026 Editor assigned by journal 16 Feb, 2026 Submission checks completed at journal 16 Feb, 2026 First submitted to journal 04 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8791325","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":596217299,"identity":"18bb63f0-c1c9-42fb-975e-30c7bc81106e","order_by":0,"name":"Runzhi Cai","email":"","orcid":"","institution":"The People's Hospital of Longhua","correspondingAuthor":false,"prefix":"","firstName":"Runzhi","middleName":"","lastName":"Cai","suffix":""},{"id":596217300,"identity":"76c993a5-b665-4bed-bf38-70ff38140501","order_by":1,"name":"Weijun Lin","email":"","orcid":"","institution":"Guangzhou University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Weijun","middleName":"","lastName":"Lin","suffix":""},{"id":596217301,"identity":"b2024163-40ed-49dc-b47f-8fafee41e9cf","order_by":2,"name":"Meng Wang","email":"","orcid":"","institution":"The People's Hospital of Longhua","correspondingAuthor":false,"prefix":"","firstName":"Meng","middleName":"","lastName":"Wang","suffix":""},{"id":596217302,"identity":"7316975f-7ef1-4ed7-8764-f89a38dc7fb3","order_by":3,"name":"Jiajia Zeng","email":"","orcid":"","institution":"The People's Hospital of Longhua","correspondingAuthor":false,"prefix":"","firstName":"Jiajia","middleName":"","lastName":"Zeng","suffix":""},{"id":596217303,"identity":"b469e9c5-a362-4714-b9ee-9e63ed00adce","order_by":4,"name":"Jian Shen","email":"","orcid":"","institution":"The People's Hospital of Longhua","correspondingAuthor":false,"prefix":"","firstName":"Jian","middleName":"","lastName":"Shen","suffix":""},{"id":596217304,"identity":"bbd54f2a-e55a-4c63-a1f4-e623ad624283","order_by":5,"name":"Shilun Jiang","email":"","orcid":"","institution":"The People's Hospital of Longhua","correspondingAuthor":false,"prefix":"","firstName":"Shilun","middleName":"","lastName":"Jiang","suffix":""},{"id":596217305,"identity":"f9f8ed53-a3a7-4cd2-911a-96044c3e0224","order_by":6,"name":"Qian Huang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAr0lEQVRIiWNgGAWjYFCCA0BswCDHxt5+gDQtxnw8ZxJIsytxnoSDAXFKDQ6effi4oOBOepsEQwLDj4pthLVINhw3Np5h8Cy3TbrxAGPPmduEtfAzHGOT5jE4nNsmcyCBmbGNCC1sUC3pbBIJBsRpgdmSQLwWyYZjzMZALYZtwEA+SJRfDG4cY3zM8+ewvHx7+8EHPyqI0MIgcQDBPoBLESrgbyBO3SgYBaNgFIxgAAClDzi9VIe7hwAAAABJRU5ErkJggg==","orcid":"","institution":"The People's Hospital of Longhua","correspondingAuthor":true,"prefix":"","firstName":"Qian","middleName":"","lastName":"Huang","suffix":""}],"badges":[],"createdAt":"2026-02-05 02:23:56","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8791325/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8791325/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103410920,"identity":"d115b3ee-19d3-44a6-886f-a7eac8a2de34","added_by":"auto","created_at":"2026-02-25 11:04:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":210051,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential expression and GO/KEGG enrichment analysis. (a) Volcano plot of DEGs. (b) Bar plot of GO enrichment. (c) Bubble plot of GO enrichment. (d) Bar plot of KEGG enrichment. (e) Bubble plot of KEGG enrichment.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/90729a9f0e6bb136661b1edc.png"},{"id":103507968,"identity":"9726ad10-4307-4e7c-9cba-3c3a5592f5a0","added_by":"auto","created_at":"2026-02-26 13:46:43","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":256018,"visible":true,"origin":"","legend":"\u003cp\u003eGene Set Enrichment Analysis. (a) Dot plot of enriched pathways. (b) Activation and suppression plot. (c) Ridge plot. (d) Pathway correlation analysis.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/77ae057336a46c535d554502.png"},{"id":103507315,"identity":"4801a6e1-07a2-424f-b121-7ff74bfc0ad2","added_by":"auto","created_at":"2026-02-26 13:40:59","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":146714,"visible":true,"origin":"","legend":"\u003cp\u003eWeighted Gene Co-expression Network Analysis. (a) Clustering dendrogram. (b) Selection of the soft-thresholding power. (c) Cluster trend analysis. (d) Heatmap of module-trait relationships.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/1e4600017e1eecf30930da88.png"},{"id":103508047,"identity":"fdbb28e0-2a09-4e98-a367-ba64bc566eb8","added_by":"auto","created_at":"2026-02-26 13:47:00","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":151010,"visible":true,"origin":"","legend":"\u003cp\u003eImmune cell infiltration analysis. (a) Box plot. (b) Stacked bar plot. (c) Heatmap.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/7ead34eb22c01f1466a40b90.png"},{"id":103506916,"identity":"0f601862-ee2b-4366-b4cc-8d8f71157468","added_by":"auto","created_at":"2026-02-26 13:39:54","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":194105,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of ferroptosis-related and MMP genes. (a) Venn diagram. (b, c) Protein-protein interaction network of ferroptosis-related DEGs from the STRING database. (d, e) Correlation analysis plots.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/2ef6ed3c1dda72bc6aae5838.png"},{"id":103507030,"identity":"417e22b7-4e37-4006-8e80-e35dbd423c2e","added_by":"auto","created_at":"2026-02-26 13:40:15","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":774009,"visible":true,"origin":"","legend":"\u003cp\u003eMolecular docking results.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/3b618d3f9bfc1eb63ce42701.png"},{"id":103410922,"identity":"b507b354-e05b-47ce-b7d0-bf2a89841f69","added_by":"auto","created_at":"2026-02-25 11:04:51","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":538013,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of Acteoside on cell viability and apoptosis-related gene expression. (a) Cell viability detected by CCK-8 assay. (b) BAX mRNA expression. (c) CASPASE3 mRNA expression. (d) BCL-2 mRNA expression.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/1cc0112aa6655f7099518dee.png"},{"id":103410927,"identity":"58bf1076-f054-4496-b668-539bf45825c4","added_by":"auto","created_at":"2026-02-25 11:04:51","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":601674,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of Acteoside on ferroptosis-related indicators. (a) ACSL4 mRNA expression. (b) GPX4 mRNA expression. (c) SLC7A11 mRNA expression. (d) ATM mRNA expression. (e) ROS level. (f) MDA level. (g) Fe²⁺ level. (h) GSH level.\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/99b71b5cb074022f9de6d2cc.png"},{"id":103410926,"identity":"0fedc900-f18d-42b8-93e5-e0ac5359460f","added_by":"auto","created_at":"2026-02-25 11:04:51","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":774700,"visible":true,"origin":"","legend":"\u003cp\u003eEffects of Acteoside on MMP family gene expression and collagen metabolism. (a) MMP1 mRNA expression. (b) MMP3 mRNA expression. (c) MMP9 mRNA expression. (d) P1NP level. (e) CTX‑I level. (f) P1NP/CTX‑I ratio.\u003c/p\u003e","description":"","filename":"image9.png","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/ef0aa93b35985ec040d5c6f7.png"},{"id":103511914,"identity":"9ea21f4e-5dda-4f0a-bf58-d9ceb7c83e20","added_by":"auto","created_at":"2026-02-26 14:11:29","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4696564,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8791325/v1/1ba6f19a-9019-4188-a163-35d32f2397cc.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Bioinformatics-Driven Target Discovery in Skin Photoaging and Preliminary Validation of the Natural Compound Acteoside","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSkin photoaging is a complex pathological process triggered by long-term UVB radiation, characterized primarily by wrinkle formation and collagen degradation. It not only affects skin appearance but also significantly increases the risk of precancerous lesions [1, 2]. The underlying mechanisms involve the synergistic action of multiple pathways, including oxidative stress, ferroptosis, and excessive activation of matrix metalloproteinases (MMPs) [3-5]. However, the traditional \"hypothesis-driven\" research model has notable limitations: target screening often relies heavily on prior knowledge, and pathway analysis tends to be fragmented, making it difficult to systematically unravel the regulatory networks involving multiple genes and pathways. These constraints hinder the identification of core targets and the development of effective therapeutics, representing a major bottleneck in both mechanistic research and preventive intervention for photoaging.\u003c/p\u003e\n\u003cp\u003eWith the rapid advancement of bioinformatics, the mining of high-throughput data from public gene expression databases such as GEO has provided powerful support for the systematic identification of disease-related molecular markers and potential therapeutic targets [6]. In this context, the search for safe and effective natural compounds for preventing and ameliorating photoaging has become a research focus in dermatology and cosmetics science.\u003c/p\u003e\n\u003cp\u003eFerroptosis has been confirmed as one of the key mechanisms in UVB-induced skin photoaging[7-9]. Meanwhile, MMPs, as critical executors of dermal extracellular matrix degradation—particularly collagen—directly contribute to structural and functional loss in photoaged skin[10-12]. However, it remains unclear whether there is a regulatory link between ferroptosis and MMP-mediated collagen degradation. Acteoside, a phenylethanoid glycoside widely present in medicinal plants, has been reported to possess antioxidant, anti-inflammatory, and anti-photoaging potential[9, 13, 14]. Yet, whether it acts through the regulation of ferroptosis and what its specific molecular targets are remain to be elucidated. Addressing these questions necessitates systematic exploration using bioinformatics approaches.\u003c/p\u003e\n\u003cp\u003eTherefore, this study established a comprehensive research framework centered on bioinformatics analysis. This framework encompasses data mining from the GEO database, screening and functional annotation of differentially expressed genes, identification of core regulatory pathways and key targets, prediction of active ingredients, validation via molecular docking, and final confirmation through cellular experiments. Through this multi-tiered bioinformatics approach, we identified core regulatory pathways and candidate compounds involved in photoaging from large-scale datasets and further validated their mechanisms of action experimentally. This study aims to affirm the pivotal role of bioinformatics in target discovery and drug development, thereby providing new strategies for the prevention and treatment of skin photoaging.\u003c/p\u003e"},{"header":"1. Materials and Methods","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003e1.1 Bioinformatics Analysis\u003c/h2\u003e \u003cdiv id=\"Sec3\" class=\"Section3\"\u003e \u003ch2\u003e1.1.1 Data Source and Preprocessing\u003c/h2\u003e \u003cp\u003eThe skin photoaging-related datasets GSE284483 and GSE296578 were downloaded from the NCBI GEO database. Both datasets were derived from skin tissues of Balb/c nude mice: GSE284483 included 6 normal controls and 6 photoaged samples, while GSE296578 contained 3 normal controls and 3 photoaged samples. The R package sva was used to merge the datasets and correct for batch effects, thereby eliminating technical variations between datasets and ensuring the reliability of subsequent analyses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section3\"\u003e \u003ch2\u003e1.1.2 Screening of Differentially Expressed Genes (DEGs)\u003c/h2\u003e \u003cp\u003eThe R package limma was employed to identify DEGs using a linear model-based approach. Genes with |log₂FC| \u0026gt; 1 and an adjusted \u003cem\u003eP\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered statistically significant. A volcano plot was generated to visualize the results.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e1.1.3 GO and KEGG Enrichment Analysis and GSEA\u003c/h2\u003e \u003cp\u003eDEGs were mapped to the Gene Ontology (GO) database\u0026mdash;including biological process (BP), molecular function (MF), and cellular component (CC)\u0026mdash;and the KEGG pathway database. Significantly enriched terms (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were identified using the hypergeometric test with multiple testing correction, and results were visualized using bar plots and bubble plots. Gene Set Enrichment Analysis (GSEA) was performed to identify potential pathways associated with skin photoaging, with significance thresholds set at \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Q\u0026thinsp;\u0026lt;\u0026thinsp;0.25, and |Normalized Enrichment Score| \u0026gt; 1.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e1.1.4 Weighted Gene Co-expression Network Analysis (WGCNA)\u003c/h2\u003e \u003cp\u003eThe R package WGCNA was used to construct a co-expression network. The merged dataset was transformed into a topological overlap matrix (TOM), and hierarchical clustering was applied to generate a clustering dendrogram. Modules were identified based on gene expression patterns, and key modules and hub genes were selected according to their correlation with the photoaging phenotype.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e1.1.5 Immune Cell Infiltration Analysis\u003c/h2\u003e \u003cp\u003eThe CIBERSORT algorithm was applied to estimate the relative proportions of 22 types of immune cells. Results were visualized using box plots, stacked bar plots, and heatmaps. Spearman correlation analysis was conducted to assess the relationship between gene expression and immune cell infiltration levels.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e1.1.6 STRING Interaction Network and Active Ingredient Prediction\u003c/h2\u003e \u003cp\u003eDEGs were converted to standard gene symbols and uploaded to the STRING database (Homo sapiens) to construct a protein-protein interaction (PPI) network. Potential active ingredients targeting the DEGs were predicted using the HERB 2.0 database via the \"Target Prediction\" function. High-confidence compounds were selected based on Summary Score and experimental evidence (EXP number).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e1.1.7 Molecular Docking\u003c/h2\u003e \u003cp\u003eThe three-dimensional structures of core targets (ATM and MMP3) were retrieved from the PDB database. Molecular docking was performed using AutoDock Vina. Binding sites were defined, and binding free energy was calculated. The optimal binding conformation was selected based on binding energy, hydrogen bonding, and hydrophobic interactions.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e1.2 Cell Experiments\u003c/h2\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e1.2.1 Cell Source and Major Reagents\u003c/h2\u003e \u003cp\u003eHuman dermal fibroblasts (HDFs) were obtained from the Cell Bank of the Chinese Academy of Sciences and cultured in DMEM medium supplemented with 10% fetal bovine serum and 1% penicillin \u0026ndash; streptomycin. Cells were maintained in a 37\u0026deg;C, 5% CO₂ incubator, and experiments were conducted using cells in the logarithmic growth phase.\u003c/p\u003e \u003cp\u003eActeoside (purity\u0026thinsp;\u0026ge;\u0026thinsp;98%) was purchased from Sigma. CCK-8 assay kit, ferrous iron assay kit, MDA assay kit, GSH assay kit, and ROS assay kit (non-fluorescent probe method) were obtained from Nanjing Jiancheng Bioengineering Institute. P1NP and CTX-I ELISA kits were purchased from R\u0026amp;D Systems. TRIzol reagent, reverse transcription kit, and quantitative real-time PCR kit were purchased from TaKaRa.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e1.2.2 Cell Viability Assay by CCK-8\u003c/h2\u003e \u003cp\u003eHDFs were seeded into 96-well plates at 5 \u0026times; 10\u0026sup3; cells per well. After 24 h of culture, cells were treated with different concentrations of Acteoside (0, 10, 20, 40, 80 \u0026micro;mol/L) for 48 h. Then, 10 \u0026micro;L of CCK-8 reagent was added to each well, followed by incubation for 2 h. Absorbance at 450 nm was measured using a microplate reader. Cell viability was calculated as follows: Viability (%) = (OD treatment \u0026ndash; OD blank) / (OD control \u0026ndash; OD blank) \u0026times; 100%.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003e1.2.3 Experimental Grouping and Model Establishment\u003c/h2\u003e \u003cp\u003eCells were divided into three groups:①Control group: normal culture without UVB irradiation.②UVB group: photoaging model induced by UVB irradiation (40 mJ/cm\u0026sup2;) for 30 min.③UVB\u0026thinsp;+\u0026thinsp;Acteoside group: HDFs were pretreated with 40 \u0026micro;mol/L Acteoside for 24 h upon reaching 70% confluence, followed by UVB irradiation for 30 min. After irradiation, the culture medium was removed, and cells were washed twice with PBS. Fresh medium containing Acteoside was then added, and cells were cultured for another 24 h before subsequent experiments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e1.2.4 Quantitative Real-Time PCR (qRT-PCR) for mRNA Expression\u003c/h2\u003e \u003cp\u003eTotal RNA was extracted using TRIzol reagent, and its purity and integrity were verified. cDNA was synthesized following the instructions of the reverse transcription kit. Using cDNA as the template, amplification was performed with the quantitative real-time PCR kit under the following conditions: 95\u0026deg;C for 30 s (pre-denaturation), followed by 40 cycles of 95\u0026deg;C for 5 s and 60\u0026deg;C for 30 s. GAPDH was used as the internal reference gene, and the relative expression levels of target genes (BAX, BCL-2, CASPASE3, GPX4, SLC7A11, ACSL4, MMP1, MMP3, MMP9) were calculated using the 2^(\u0026ndash;ΔΔCt) method. Primer sequences are listed in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePrimer sequences\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGene Name\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eForward Primer (5'-3')\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReverse Primer (5'-3')\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBAX\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTTGCTTCAGGGTTTCATCCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGCACTACCGCCTGAAAGCTG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBCL-2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGGATGCCTTTGTGGAACTGT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAGCCTGCAGCTTTGTTTCAT\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCASPASE3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTGGTTCATACCAGTCGCTCTG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTCAAATTTGCTGCAATCGGAC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGPX4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAGAAGTGGGACAGCACCAAG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTGTCGGACACACTGGTCTTG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSLC7A11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTCTCCAATGCGTTGCTGAAC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAGACCAGGCCATCATCAACC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eACSL4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTGCTGAAGGAGGAGTTTGGA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGGACAGCCTCGTAGAGCAAT\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMMP1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCTAAGCAGACATGGGACACG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAGTCCAAGAGAATGGCCGAG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMMP3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTTCTCCTGCTTTGTCCTTCC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAAGCAGGATCACAGTTGGCT\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMMP9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCCTGGAGACCTGAGAACCAATC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCACCCCGAGTGGTACTCATACT\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGAPDH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGGAGCGAGATCCCTCCAAAAT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGGCTGTTGTCATACTTCTCATGG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e1.2.5 Measurement of Fe\u0026sup2;⁺, MDA, GSH, and ROS Levels\u003c/h2\u003e \u003cp\u003eAccording to the respective kit instructions, cells from each group were collected and homogenized. The levels of Fe\u0026sup2;⁺, MDA, GSH, and ROS were measured, and the relative content of each indicator was calculated after normalization to the total protein concentration.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section3\"\u003e \u003ch2\u003e1.2.6 ELISA for P1NP and CTX-Ⅰ\u003c/h2\u003e \u003cp\u003eCell culture supernatants were collected from each group. The levels of P1NP and CTX-Ⅰ were determined using enzyme-linked immunosorbent assay (ELISA) kits according to the manufacturer\u0026rsquo;s instructions. The P1NP/CTX-Ⅰ ratio was calculated to reflect the balance between collagen synthesis and degradation.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e1.3 Statistical Analysis\u003c/h2\u003e \u003cp\u003eAll experimental data were analyzed using GraphPad Prism 8.0 software. Measurement data are expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (x̄ \u0026plusmn; s). One-way analysis of variance (ANOVA) was used for comparisons among multiple groups, and the LSD-t test was applied for post hoc pairwise comparisons. A \u003cem\u003eP\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"2. Results","content":"\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Bioinformatics Analysis Results\u003c/h2\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003ch2\u003e2.1.1 Screening of Differentially Expressed Genes (DEGs)\u003c/h2\u003e \u003cp\u003eA total of 18 skin tissue samples were included from two datasets: 12 samples (6 normal and 6 photoaged) from GSE284483 and 6 samples (3 normal and 3 photoaged) from GSE296578. After merging and batch effect correction, variance analysis was performed using the limma package. DEGs were screened using the thresholds |log₂FC| \u0026gt; 1 and \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05. A total of 765 DEGs were identified, as shown in the volcano plot (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea), where orange points represent upregulated genes and purple points represent downregulated genes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section3\"\u003e \u003ch2\u003e2.1.2 Prediction of Active Ingredients\u003c/h2\u003e \u003cp\u003eDEGs were imported into the HERB 2.0 database for active ingredient prediction using the \u0026ldquo;Target Prediction\u0026rdquo; function. Based on gene\u0026ndash;herb\u0026ndash;compound association networks and experimental validation data, the system calculated Summary Score and EXP number to screen for high-confidence active molecules for subsequent molecular docking and mechanistic analysis. Seven active ingredients were identified, as listed in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eActive ingredients\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eType\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIngredient name\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSummary score\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIngredient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDaidzein\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-87.15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIngredient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQuercetin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-84.73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIngredient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLuteolin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-80.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIngredient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGallic Acid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-77.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIngredient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eActeoside\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-77.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIngredient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eApigenin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-77.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIngredient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCannabidiol\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-77.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIngredient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCarnosic Acid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e77.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section3\"\u003e \u003ch2\u003e2.1.3 GO and KEGG Enrichment Analysis\u003c/h2\u003e \u003cp\u003eGO enrichment analysis revealed that the differentially expressed genes (DEGs) were primarily involved in biological processes (BP) such as cell motility and signal transduction. In terms of molecular function (MF), they were notably enriched in oxidoreductase activity. For cellular component (CC), significant enrichment was observed in mitochondrial matrix and organelle inner membrane (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb, \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec).\u003c/p\u003e \u003cp\u003eKEGG pathway analysis further indicated that these genes were significantly enriched in several functional domains, including fundamental biological processes such as translation and cell motility, signal transduction and interaction of signaling molecules, disease pathways related to immune, circulatory, and nervous systems, as well as pathways associated with parasitic and bacterial infectious diseases (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed, \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003e2.1.4 Gene Set Enrichment Analysis (GSEA)\u003c/h2\u003e \u003cp\u003eTo investigate the potential pathways involved in skin photoaging, we performed Gene Set Enrichment Analysis (GSEA) using the gene set variation analysis package (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The significance thresholds were set at \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Q\u0026thinsp;\u0026lt;\u0026thinsp;0.25, and |NES| \u0026gt; 1. The results revealed significant enrichment in the following areas: Biological Processes (BP): monocarboxylic acid metabolic process, organic acid metabolic process, small molecule biosynthetic process, and fatty acid metabolic process. Cellular Components (CC): mitochondrial matrix, mitochondrial envelope, organelle inner membrane, and intermediate filament cytoskeleton.Molecular Functions (MF): oxidoreductase activity. These findings suggest that oxidative stress and dysregulated energy metabolism are central characteristics of photoaging, which align closely with the mechanisms of ferroptosis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section3\"\u003e \u003ch2\u003e2.1.5 Weighted Gene Co-expression Network Analysis (WGCNA)\u003c/h2\u003e \u003cp\u003eWGCNA categorized all genes into 18 distinct modules. Among these, the MEturquoise module demonstrated the highest correlation with the skin photoaging phenotype (r\u0026thinsp;=\u0026thinsp;0.74, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Further analysis of genes within this module identified several ferroptosis-related genes and matrix metalloproteinases (MMPs), suggesting that ferroptosis and extracellular matrix (ECM) degradation may cooperatively regulate photoaging through these genes.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003e2.1.6 Immune Cell Infiltration Analysis\u003c/h2\u003e \u003cp\u003eCIBERSORT analysis revealed significant alterations in immune cell composition in the photoaging group compared to the normal group. Specifically, the proportions of CD4⁺ T cells and macrophages were significantly elevated, indicating the involvement of immune microenvironment dysregulation in the photoaging process. These findings are consistent with the immune-related pathways enriched in the DEG analysis, further supporting a multi-pathway synergistic mechanism in skin photoaging (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003e2.1.7 Correlation Analysis of Ferroptosis-Related and MMP Genes\u003c/h2\u003e \u003cp\u003eBased on prior WGCNA indicating significant enrichment of both ferroptosis-related genes and matrix metalloproteinases (MMPs) in skin photoaging, we conducted an in-depth analysis of these two gene sets. A total of 24 ferroptosis-related genes and 3 MMP genes (MMP3, MMP9, and MMP13) were found to be differentially expressed in the photoaging group. Protein-protein interaction analysis performed via the STRING database revealed that ATM, Cav1, and Cdkn2a occupied central positions within the ferroptosis-related network. Furthermore, correlation analysis among ATM, Cav1, Cdkn2a, and the MMP genes (MMP3, MMP9, MMP13) demonstrated a significant positive correlation between MMP3 and ATM expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003e2.1.8 Molecular Docking\u003c/h2\u003e \u003cp\u003eThe three-dimensional structures of the target proteins were retrieved from the PDB database, and the small molecule structures were prepared accordingly. Molecular docking was performed using AutoDock Vina by defining the binding sites and conducting docking calculations. The optimal binding conformation was selected based on binding energy, hydrogen bonding, and hydrophobic interactions to evaluate binding stability and the potential bioactivity of the screened active ingredients.\u003c/p\u003e \u003cp\u003eThe results (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) showed that the ATM protein could stably bind to multiple natural compounds, including Acteoside, Apigenin, Cannabidiol, Carnosic Acid, Daidzein, Luteolin, and Quercetin, all with binding free energies lower than \u0026minus;\u0026thinsp;3 kcal/mol. Among them, Acteoside exhibited the highest binding affinity, suggesting a strong potential for interaction with ATM. Furthermore, molecular docking also indicated that the MMP3 protein could effectively bind to Acteoside.\u003c/p\u003e \u003cp\u003eIn summary, these computational results suggest that Acteoside may exert regulatory effects in the treatment of skin photoaging by targeting key molecules such as ATM and MMP3. This finding provides a theoretical basis for further exploration of its molecular mechanism, though the specific pathways of action require subsequent experimental validation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Cell Experiment Results\u003c/h2\u003e \u003cdiv id=\"Sec29\" class=\"Section3\"\u003e \u003ch2\u003e2.2.1 Effect of Acteoside on HDF Cell Viability\u003c/h2\u003e \u003cp\u003eCCK-8 results demonstrated that when Acteoside concentration was \u0026le;\u0026thinsp;40 \u0026micro;mol/L, HDF cell viability remained above 90% (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05). However, at concentrations\u0026thinsp;\u0026ge;\u0026thinsp;80 \u0026micro;mol/L, viability significantly decreased (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). Therefore, 40 \u0026micro;mol/L was selected for subsequent experiments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec30\" class=\"Section3\"\u003e \u003ch2\u003e2.2.2 Effect of Acteoside on Apoptosis-Related Gene Expression\u003c/h2\u003e \u003cp\u003eqPCR results revealed that compared with the control group, cells in the model group exhibited a significant pro-apoptotic trend: the expression of pro-apoptotic genes BAX and CASPASE3 was significantly increased (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb, c), while the expression of the anti-apoptotic gene BCL-2 was significantly reduced (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed) (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). However, Acteoside intervention markedly reversed these gene expression alterations (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), indicating that Acteoside can counteract UVB-induced dysregulation of apoptotic gene expression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Effect of Acteoside on Ferroptosis-Related Indicators\u003c/h2\u003e \u003cp\u003eCompared with the control group, cells in the model group exhibited characteristic changes associated with ferroptosis: mRNA expression of the pro-ferroptotic gene ACSL4 was significantly up-regulated (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ea), while the expression of key ferroptosis-inhibitory genes (GPX4, SLC7A11) was down-regulated (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb, c. Additionally, mRNA expression of ATM\u0026mdash;previously identified via bioinformatics analysis as positively correlated with MMP3\u0026mdash;was also up-regulated (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ed).\u003c/p\u003e \u003cp\u003eAt the level of oxidative stress, the model group showed elevated ROS, MDA, and Fe\u0026sup2;⁺ accumulation (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ee\u0026ndash;g), accompanied by depletion of GSH (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eH). In contrast, Acteoside intervention significantly reversed these alterations in key indicators (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e \u003cp\u003eThese results demonstrate that Acteoside can effectively suppress UVB-induced ferroptosis in HDF cells.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec32\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Effect of Acteoside on MMP Family Gene Expression\u003c/h2\u003e \u003cp\u003eExperimental results demonstrated that mRNA expression levels of MMP1, MMP3, and MMP9 were significantly up-regulated in the UVB-exposed model group compared with the control group (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In contrast, Acteoside treatment markedly suppressed the expression of these genes (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003ea-c). These findings indicate that Acteoside can effectively counteract UVB-induced up-regulation of MMP family genes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec33\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Effect of Acteoside on Collagen Metabolism\u003c/h2\u003e \u003cp\u003eCompared with the control group, the model group showed significantly decreased P1NP levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003ed) and P1NP/CTX-I ratio (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003ef), along with significantly increased CTX-I levels (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003ee). However, in the Acteoside-treated group, both P1NP levels and the P1NP/CTX-I ratio were significantly elevated (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while CTX-I levels were significantly reduced (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) compared with the model group. These results suggest that Acteoside can alleviate UVB-induced collagen reduction.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Discussion","content":"\u003cp\u003eThe complex characteristics of skin photoaging\u0026mdash;including multi-gene interactions and intertwined pathways\u0026mdash;often pose challenges for traditional hypothesis-driven research, leading tounsystematic target identification and fragmented mechanistic understanding. In this context, bioinformatics has emerged as a pivotal technology for overcoming these research bottlenecks, leveraging its data-driven, high-throughput integration capabilities. According to statistics, approximately 65% of skin aging mechanism studies indexed in PubMed in recent years have utilized bioinformatics to identify core targets, underscoring its dominant role in dermatological research [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe central innovation of this study lies in adopting bioinformatics as the central thread running through the entire study, implementing a closed-loop strategy of \"multi-step bioinformatic analysis to identify core regulatory axes\u0026mdash;experimental validation of molecular mechanisms.\" This approach not only systematically elucidates the anti-photoaging mechanism of Acteoside but also clearly demonstrates the core value of bioinformatics in target discovery and drug development.\u003c/p\u003e \u003cp\u003eWe began by implementing a standardized bioinformatics pipeline to accurately screen DEGs, laying the foundation for identifying core regulatory axes. Two independent photoaging datasets (GSE284483 and GSE296578) from nude mouse skin were obtained from the GEO database. To address batch effects in integrated analysis, the sva package was applied for data correction\u0026mdash;a critical bioinformatic step in multi-dataset integration that has been shown to reduce false positive rates in subsequent differential analysis by over 30%[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Subsequently, the limma package was used for differential analysis with stringent thresholds (|logFC| \u0026gt; 1 and \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05), ultimately identifying 765 DEGs.\u003c/p\u003e \u003cp\u003eBuilding on this foundation, functional enrichment analyses including GO, KEGG, and GSEA were performed to interpret the biological significance of the DEGs. GO analysis revealed that DEGs were primarily involved in processes such as \"cell motility\" and \"oxidoreductase activity.\" KEGG enrichment analysis indicated significant associations with pathways including \"iron metabolism\" and \"immune signaling.\" GSEA further confirmed significant activation of key photoaging pathways such as \"fatty acid metabolism\" and \"mitochondrial function.\"[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] Together, these three analytical approaches provided a multidimensional molecular map of skin photoaging, revealing potential connections between ferroptosis and extracellular matrix degradation. This integrated approach provided clear direction for subsequent target focus and aligns closely with the viewpoint proposed by reach in that \"the core pathway of photoaging involves an oxidative stress-ferroptosis-ECM degradation network\"[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWGCNA [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] served as the core bioinformatics tool in this study for identifying the ATM-MMP3 axis Using WGCNA, all genes were categorized into 15 co-expression modules. Module-trait correlation analysis revealed that the MEturquoise module showed the strongest association with the photoaging phenotype (r\u0026thinsp;=\u0026thinsp;0.74, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01). This analysis established a systematic connection from \"phenotype\" to \"gene modules,\" overcoming the limitations of conventional single-gene analysis. Further topological analysis of this module identified ATM (a key ferroptosis-related gene) and MMP3 (a central gene in extracellular matrix degradation) as hub genes. Pearson correlation analysis demonstrated a significant positive correlation between their expression levels (r\u0026thinsp;=\u0026thinsp;0.60, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), leading us to propose, that the \"ATM-MMP3 axis\" may play a central regulatory role in the photoaging process. This discovery was entirely enabled by the synergistic application of WGCNA module analysis and correlation analysis, and is consistent with findings reported which indicated that \"WGCNA achieves 82% accuracy in identifying hub genes in skin aging research,\" highlighting the unique advantage of bioinformatics in revealing pathway synergy[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e \u003cp\u003ePPI network analysis and molecular docking further strengthened the evidence chain for the \"target-drug\" relationship[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. We constructed a PPI network using the STRING database with the DEGs, setting a confidence score threshold of \u0026gt;\u0026thinsp;0.7. The results indicated a direct interaction between ATM and MMP3, supporting the biological plausibility of this axis at the protein interaction level. To efficiently identify candidate compounds targeting this axis, we performed target-component matching using the HERB 2.0 database. Applying the criteria of \"Summary Score \u0026le; -80 and experimental validation count\u0026thinsp;\u0026ge;\u0026thinsp;3,\" Acteoside was identified as the primary candidate from over 200 natural products. Subsequent validation via molecular docking with AutoDock Vina showed binding free energies of -9.3 kcal/mol for Acteoside-ATM and \u0026minus;\u0026thinsp;7.9 kcal/mol for Acteoside-MMP3, both below the high-affinity threshold of -7 kcal/mol, with more than three stable hydrogen bonds formed. This series of bioinformatics procedures completed the entire process from target screening to drug matching and binding verification within 72 hours, demonstrating a more than 10-fold increase in efficiency compared to traditional compound screening methods and fully underscoring the high efficiency of bioinformatics in early-stage drug discovery[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe successful experimental validation of bioinformatics predictions has established a scientific cycle of \"bioinformatics guiding experiments, with experimental results feeding back into bioinformatics,\" further confirming the reliability of our bioinformatic analyses. We designed validation experiments focusing on the ATM-MMP3 axis identified through bioinformatic screening: ferroptosis-related assays showed that Acteoside downregulated ATM and ACSL4 expression, upregulated GPX4 levels, and reversed UVB-induced Fe\u0026sup2;⁺ accumulation and ROS elevation, directly validating the prediction that \"ATM serves as a core ferroptosis target.\" Analysis of MMPs expression and collagen metabolism demonstrated that Acteoside suppressed MMP3 expression and restored the P1NP/CTX-I ratio, confirming the bioinformatic conclusion that \"MMP3 mediates ECM degradation.\" This research model of \"bioinformatics providing direction, experiments delivering precise validation\" effectively avoids the blind trial-and-error approach common in traditional mechanistic investigations, aligning closely with the conclusion reported that \"bioinformatics-driven research on natural product mechanisms increases the success rate by 40%[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\"\u003c/p\u003e \u003cp\u003eIn summary, the bioinformatics framework constructed in this study\u0026mdash;\"multi-dataset integration \u0026rarr; DEG screening \u0026rarr; WGCNA-based core module identification \u0026rarr; PPI network construction \u0026rarr; molecular docking validation \u0026rarr; experimental functional confirmation\"\u0026mdash;possesses both clear methodological demonstrative value and broad potential for application. This workflow can be directly adapted to target discovery research for other dermatological conditions such as psoriasis and atopic dermatitis, and is particularly suitable for guiding research beginners in mastering the complete paradigm of integrating computational and experimental approaches. Notably, the successful identification of the TRAF6\u0026ndash;IL-6 regulatory axis in atopic dermatitis using a similar workflow further validates the reliability and reproducibility of our methodological system[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eThis study systematically elucidated the molecular mechanism by which the natural compound Acteoside alleviates skin photoaging through targeting the ATM\u0026ndash;MMP3 axis, integrating bioinformatics analysis with experimental validation. Bioinformatics approaches successfully identified 765 differentially expressed genes and, through WGCNA, highlighted the MEturquoise module as highly associated with the photoaging phenotype, leading to the identification of the ATM\u0026ndash;MMP3 axis as a core regulatory node. Molecular docking results demonstrated high binding affinity between Acteoside and both ATM and MMP3. Cellular experiments confirmed that Acteoside effectively suppresses UVB-induced ferroptosis, downregulates MMP expression, and restores collagen metabolic balance. These findings not only provide new insights into the pathogenesis of skin photoaging but also offer a potential therapeutic strategy targeting the ATM\u0026ndash;MMP3 axis, supported by a bioinformatics-driven research framework.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCRediT authorship contribution statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRunzhi Cai:\u003c/strong\u003e Conceptualization, Methodology, Investigation, Writing - Original Draft.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWeijun Lin:\u003c/strong\u003e Data Curation, Formal Analysis, Validation, Visualization. \u003cstrong\u003eMeng Wang:\u003c/strong\u003e Investigation, Resources, Writing - Review \u0026amp; Editing. \u003cstrong\u003eJiajia Zeng:\u003c/strong\u003e Investigation, Data Curation, Writing - Review \u0026amp; Editing. \u003cstrong\u003eJian Shen:\u003c/strong\u003e Methodology, Validation, Resources. \u003cstrong\u003eShilun Jiang\u003c/strong\u003e: Formal Analysis, Visualization, Writing - Review \u0026amp; Editing. \u003cstrong\u003eQian Huang:\u003c/strong\u003e Conceptualization, Supervision, Project Administration, Writing - Review \u0026amp; Editing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of competing interest\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics\u0026nbsp;declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot\u0026nbsp;applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData will be made available on request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo Funding\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLyu JL et al. Djulis (Chenopodium formosanum) Extract as a Promising Natural Agent Against Skin Aging. Molecules, 2025. 30(15).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng R et al. Boosting the Photoaged Skin: The Potential Role of Dietary Components. Nutrients, 2021. 13(5).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim DJ et al. UVB-mediated DNA damage induces matrix metalloproteinases to promote photoaging in an AhR- and SP1-dependent manner. JCI Insight, 2022. 7(9).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu L, et al. Mechanistic Insights Into 5'-tiRNA-His-GTG Mediated Activation of the JNK Pathway in Skin Photoaging. Aging Cell. 2025;24(7):e70049.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei K, et al. The beta-glucan from Lentinus alleviates UVB-induced dermal fibroblasts senescence and skin photoaging. Int J Biol Macromol. 2025;329(Pt 2):147834.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGimeno A et al. Identification of Broad-Spectrum MMP Inhibitors by Virtual Screening. Molecules, 2021. 26(15).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang PC, et al. Variation of Ferroptosis-Related Markers in HaCaT Cell Photoaging Models Induced by UVB. Clin Cosmet Investig Dermatol. 2023;16:3147\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeng Y, et al. Specific Knockdown of the NDUFS4 Gene Reveals Important Roles of Ferroptosis in UVB-induced Photoaging. Inflammation. 2025;48(1):223\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeng Y, et al. Emerging role of ferroptosis in ultraviolet radiation-driven skin photoaging: a narrative review. Photochem Photobiol Sci. 2025;24(3):531\u0026ndash;42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee SJ, et al. Transglutaminase 2 mediates UVB-induced matrix metalloproteinase-1 expression by inhibiting nuclear p65 degradation in dermal fibroblasts. Exp Dermatol. 2022;31(5):743\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIvanova I, et al. UVA-induced metabolic changes in non-malignant skin cells and the potential role of pyruvate as antioxidant. Photochem Photobiol Sci. 2023;22(8):1889\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCui X, et al. The Anti-photoaging Effects of Pre- and Post-treatment of Platelet-rich Plasma on UVB-damaged HaCaT Keratinocytes. Photochem Photobiol. 2021;97(3):589\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXi X, et al. Acteoside attenuates hydrogen peroxide-induced injury of retinal ganglion cells via the CASC2/miR-155/mTOR axis. Ann Transl Med. 2022;10(1):5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao W, et al. Effects of phenylethanol glycosides from Orobanche cernua Loefling on UVB-Induced skin photodamage: a comparative study. Photochem Photobiol Sci. 2021;20(5):599\u0026ndash;614.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeng T, et al. Bioinformatics-based analysis of the relationship between plasminogen regulatory genes and photoaging. J Cosmet Dermatol. 2024;23(6):2270\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao J, et al. Critical genes in human photoaged skin identified using weighted gene co-expression network analysis. Genomics. 2023;115(5):110682.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJin SP, et al. Functionally similar genes exhibit comparable/similar time-course expression kinetics in the UV-induced photoaged mouse model. PLoS ONE. 2023;18(11):e0290358.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQiang M, Dai Z. Biomarkers of UVB radiation-related senescent fibroblasts. Sci Rep. 2024;14(1):933.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiao X, et al. Identification of lncRNA-miRNA-mRNA Regulatory Network and Therapeutic Agents for Skin Aging by Bioinformatics Analysis. Biochem Genet. 2023;61(4):1606\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang J, et al. Needle-Free Injection of Metformin Ameliorates Skin Photoaging Through Inhibition of Ferroptosis and Oxidative Stress. Discov Med. 2024;36(184):1080\u0026ndash;90.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan M, et al. Mechanism of Ba Zhen Tang Delaying Skin Photoaging Based on Network Pharmacology and Molecular Docking. Clin Cosmet Investig Dermatol. 2022;15:763\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou F, et al. Differences in cell subsets between sun-exposed and unexposed skin: preliminary single-cell sequencing and biological analysis from a single case. Front Med (Lausanne). 2024;11:1453940.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu X, et al. Single-Cell Sequencing Combined with Transcriptome Sequencing to Explore the Molecular Mechanisms Related to Skin Photoaging. J Inflamm Res. 2024;17:11137\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang N, et al. Specnuezhenide ameliorates ultraviolet-induced skin photoaging in mice by regulating the Sirtuin 3/8-Oxoguanine DNA glycosylase signal. Photodermatol Photoimmunol Photomed. 2023;39(5):478\u0026ndash;86.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMansouri V, et al. Collagen Synthesis as a Prominent Process During the Interval between Two Laser Sessions. J Lasers Med Sci. 2023;14:e50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeng Y, et al. A Bibliometric Analysis of the Top 100 Most-Cited Articles on Skin Photoaging. J Cosmet Dermatol. 2025;24(3):e70119.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeng Y et al. A Bibliometric Analysis of the Global Research Landscape and Trends in Photoaging Therapy (2015\u0026ndash;2024). Photodiagnosis Photodyn Ther, 2025: p. 105295.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLong Y, et al. Dendrobium officinale Kimura \u0026amp; Migo polysaccharide ameliorates skin photoaging by promoting angiogenesis. Sci Rep. 2025;15(1):30048.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"hereditas","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"here","sideBox":"Learn more about [Hereditas](http://hereditasjournal.biomedcentral.com/)","snPcode":"41065","submissionUrl":"https://submission.nature.com/new-submission/41065/3","title":"Hereditas","twitterHandle":"@OAgenetics","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Skin photoaging, Acteoside, Ferroptosis, ATM-MMP3 axis, Bioinformatics, Matrix metalloproteinases","lastPublishedDoi":"10.21203/rs.3.rs-8791325/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8791325/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjective\u003c/h2\u003e \u003cp\u003eTo identify differentially expressed genes (DEGs), key pathways, and potential active compounds associated with skin photoaging using bioinformatics approaches, and to validate the interventional effects and molecular mechanisms of candidate compounds through cellular experiments, thereby highlighting the central role of bioinformatics in target and drug discovery.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eDatasets GSE284483 and GSE296578 were obtained from the GEO database, merged, and adjusted for batch effects. The limma package was used to screen for DEGs. Functional annotations of DEGs were analyzed through GO, KEGG, and GSEA analyses. WGCNA identified core modules and key genes, while CIBERSORT assessed immune cell infiltration. A protein-protein interaction network was constructed using STRING, and potential active compounds were predicted via HERB2.0. Molecular docking was performed with AutoDock Vina. A UVB-induced photoaging model was established using human skin fibroblasts, with experimental groups including control, UVB model, and Acteoside treatment. Cell proliferation, apoptosis, ferroptosis, MMP expression, and collagen metabolism were evaluated using CCK-8, qPCR, colorimetric assays, and ELISA to validate bioinformatics predictions.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eA total of 765 DEGs were identified, enriched in biological processes such as cell motility, signal transduction, iron metabolism, and immune-related pathways. WGCNA identified ferroptosis-related genes (ATM, Cav1, Cdkn2a) and MMP family genes (MMP3, MMP9, MMP13) as key genes, with a significant positive correlation between MMP3 and ATM expression. Molecular docking revealed that Acteoside exhibited the highest binding affinity for ATM (binding free energy: -9.3 kcal/mol) and MMP3 (binding free energy: -7.9 kcal/mol). Cellular experiments confirmed that Acteoside reversed UVB-induced reductions in cell viability, corrected aberrant expression of apoptosis- and ferroptosis-related genes, suppressed MMP upregulation, and restored the balance between collagen synthesis and degradation, consistent with bioinformatics predictions.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThis study demonstrates, through bioinformatics-driven analysis, that the ATM-MMP3 axis serves as a core regulatory pathway in skin photoaging. Acteoside exerts anti-photoaging effects by targeting this axis to inhibit apoptosis, ferroptosis, and extracellular matrix degradation. These findings provide new targets and candidate compounds for the prevention and treatment of skin photoaging, underscoring the efficacy of bioinformatics in guiding target and drug discovery.\u003c/p\u003e","manuscriptTitle":"Bioinformatics-Driven Target Discovery in Skin Photoaging and Preliminary Validation of the Natural Compound Acteoside","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-25 11:04:46","doi":"10.21203/rs.3.rs-8791325/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-09T07:40:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-06T20:39:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"38471218901532495187643594199458568832","date":"2026-02-24T10:16:30+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-24T05:32:32+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-16T09:40:36+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-16T09:39:29+00:00","index":"","fulltext":""},{"type":"submitted","content":"Hereditas","date":"2026-02-05T02:08:31+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"hereditas","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"here","sideBox":"Learn more about [Hereditas](http://hereditasjournal.biomedcentral.com/)","snPcode":"41065","submissionUrl":"https://submission.nature.com/new-submission/41065/3","title":"Hereditas","twitterHandle":"@OAgenetics","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9f7dc005-99f9-4f5b-a520-15e0873e9620","owner":[],"postedDate":"February 25th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-13T06:56:26+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-25 11:04:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8791325","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8791325","identity":"rs-8791325","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}