Identification and validation of senescence-associated secretory phenotype-related biomarkers in epilepsy

Identification and validation of senescence-associated secretory phenotype-related biomarkers in epilepsy | 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 Identification and validation of senescence-associated secretory phenotype-related biomarkers in epilepsy Difei Wang, Hui Gan, Yuxin Wu, Baohui Yang, Han Xiao, Haotian Tang, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9165812/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Epilepsy is increasingly linked to cellular senescence and the senescence-associated secretory phenotype (SASP), which propagates sterile neuroinflammation and tissue remodeling. However, the specific SASP-related molecular signatures driving epileptogenesis remain poorly characterized. Integrating human cortical transcriptomic profiles (GSE256068) with a curated SASP gene set, we identified 22 differentially expressed candidates. Machine-learning feature selection (LASSO and SVM-RFE) converged on a robust five-biomarker panel (IGFBP4, SERPINE1, CCL2, C3, and SPX), which demonstrated consistent differential expression in an independent, cross-regional hippocampal dataset (GSE134697). A diagnostic nomogram integrating these markers achieved excellent discrimination and clinical net benefit. Functional enrichment linked this panel to oxidative phosphorylation and translational control. Crucially, the up-regulation of SERPINE1 and CCL2 highlights a neurochemical mechanism involving extracellular matrix remodeling and inferred neuroimmune microenvironment shifts. In silico regulatory network analysis further predicted upstream transcription factors governing this axis. Finally, RT-qPCR validation in the peripheral blood of a kainic acid-induced epilepsy mouse model confirmed the significant dysregulation of four key biomarkers (IGFBP4, SERPINE1, CCL2, and C3). Collectively, our study defines a highly accurate SASP-associated biomarker panel and provides profound neurochemical insights into senescence-linked inflammatory networks, offering novel targets for the diagnosis and management of epilepsy. Epilepsy Senescence-associated secretory phenotype SERPINE1 Neuroinflammation Biomarkers Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction Epilepsy is a chronic neurological disorder arising from aberrant, hypersynchronous neuronal discharges, clinically defined by recurrent, transient cerebral dysfunction. Approximately 65 million people are affected worldwide. First-line management remains pharmacological, yet roughly one-third of patients develop drug-resistant epilepsy[ 1 , 3 ]. For these individuals, resective surgery is considered; however, it is burdened by post-operative morbidity, high costs, and unpredictable seizure freedom[ 2 , 4 ]; additionally, candidacy is restricted, further complicating seizure control. Current pathophysiological models depict epileptogenesis as a multi-layered network process. Ion-channel mutations—e.g., those in SCN1A or GABRG2—disrupt transmembrane ion homeostasis[ 5 ], while bidirectional gut–brain signalling modulates excitability through neuroimmune, neurotransmitter, and microbial metabolite pathways[ 6 , 8 ]. Over twenty antiseizure medications exist, yet all are purely symptomatic; none reverses disease progression[ 7 ]. Approximately 40% of patients require polytherapy, leading to an exponential increase in drug-drug interactions. Additionally, therapeutic options are severely limited in special populations such as pregnant individuals and the elderly. Moreover, conventional diagnostics suffer from delayed recognition of seizures and non-negligible misclassification of seizure types, underscoring an urgent need for novel biomarkers. The senescence-associated secretory phenotype (SASP) comprises cytokines, chemokines, and growth factors released by senescent cells. Via autocrine and paracrine circuits, SASP propagates sterile inflammation and transmits senescent cues to neighbouring cells; within the tissue microenvironment, reactive oxygen species (ROS)—often induced by SASP—amplify telomere attrition and accelerate cellular senescence[ 9 , 10 ]. Within the central nervous system, chronic SASP signaling fundamentally reshapes the neuroglial microenvironment. The continuous paracrine release of these inflammatory mediators impairs astrocyte glutamate buffering, triggers microglial hyperactivation, and chronically degrades blood-brain barrier (BBB) integrity, thereby establishing a pro-epileptogenic neurochemical loop[ 16 – 18 ]. Evidence linking SASP to epilepsy is now emerging. Single-cell sequencing by Ge et al. revealed up-regulation of P21, CCL2, and genes in the NF-κB pathway in pyramidal neurons within the epileptogenic cortex of drug-resistant patients; parallel murine models confirmed a robust molecular signature of neuronal senescence, establishing the first direct histopathological nexus between SASP and epilepsy[ 11 ]. Mechanistically, discrete SASP mediators drive epileptogenesis through convergent pathways. IL-6, a canonical SASP cytokine, exacerbates absence seizures in WAG/Rij rats via neuroinflammatory circuits; blockade of the IL-6 receptor with tocilizumab (TCZ) mitigates seizure aggravation induced by LPS or exogenous IL-6[ 12 ]. TNF-α and its receptor TNFR1 are markedly elevated across epileptic foci, engaging TRADD–FADD–caspase-8 signalling to precipitate neuronal apoptosis and hippocampal injury ; pharmacological interruption of this axis attenuates both ictal activity and neuronal loss[ 13 ]. Matrix metalloproteinase-9 (MMP-9), a SASP-driven extracellular matrix remodelling factor, degrades the tight-junction protein nectin-3 and compromises blood–brain barrier integrity; MMP-9 inhibition with marimastat significantly shortens seizure duration in mice, underscoring SASP-mediated barrier disruption[ 14 , 19 ]. Furthermore, SASP-driven extracellular matrix (ECM) remodeling is critically regulated by the delicate balance of proteolytic networks, including matrix metalloproteinases and the plasminogen activator system. Disruption in this balance—such as aberrant expression of endogenous protease inhibitors like SERPINE1 (PAI-1)—can exacerbate BBB leakage and facilitate neuroimmune cellular infiltration, yet the precise transcriptomic signatures of these ECM regulators in human epileptogenic foci remain obscure. Additionally, in a pre-clinical Mtor FCDII model, senolytic intervention reduced seizure frequency[ 15 ]. Collectively, these data position SASP as a tractable therapeutic node, and the development of precision agents directed at senescence effectors is emerging as a focal strategy to overcome drug-resistant epilepsy. Despite these crucial insights, the molecular circuitry linking SASP to epileptogenesis remains incompletely mapped. Deeper interrogation of SASP-mediated pathways is urgently required to identify robust non-invasive biomarkers, refine pre-surgical evaluation, and inform targeted interventions. Here, we leverage publicly available human brain transcriptomic datasets to identify a specialized SASP-associated biomarker panel in epilepsy. By applying integrative bioinformatic approaches and validating these signatures in a preclinical model, we aim to delineate the specific inflammatory and metabolic underpinnings of SASP in the epileptogenic network, providing novel mechanistic targets to ultimately improve diagnostic and therapeutic strategies for drug-resistant epilepsy. 2. Results 2.1 Enrichment analysis of 22 candidate genes The differential expression analysis of genes was conducted between EP and control samples in GSE256068. In total, 1,912 DEGs were detected. 1,280 up-regulated and 632 down-regulated genes were identified in EP ( Fig. 1 a-b ) . Subsequently, the intersection of the 1,912 DEGs and 153 SASP-RGs was taken, and 22 candidate genes were obtained ( Fig. 1 c ) . GO enrichment analysis revealed 1105 significantly enriched terms, including 1,050 BP terms, 7 CC terms, and 48 molecular MF terms. In the KEGG pathway analysis, 49 significantly enriched signaling pathways were obtained (Supplementary Table S1-2) . Specifically, in the GO-BP term, these genes showed significant enrichment in neutrophil migration and positive regulation of the MAPK cascade. In the GO-CC term, these genes were mainly concentrated in cellular structures such as collagen-containing extracellular matrix and microvilli. In terms of GO-MF, they were mainly enriched in molecular functions such as receptor ligand activity ( Fig. 1 d ) . In the KEGG enrichment analysis, they were mainly and significantly enriched in signal transduction pathways such as the receptor ligand activity and cytokine−cytokine receptor interaction ( Fig. 1 e ) . PPI network analysis showed that most proteins encoded by the genes exhibited extensive interactions ( Fig. 1 f ) . 2.2 Acquisition of 5 biomarkers Feature genes were selected from candidate genes using LASSO and SVM-RFE algorithms. LASSO regression analysis showed that when the optimal lambda was 0.006981826, seven genes with non-zero feature coefficients were obtained: IL1B, C3, IGFBP4, SERPINE1, CCL2, CCL3, and SPX ( Fig. 2 a ) . The SVM-RFE algorithm subsequently identified 16 genes (C3, IL1A, SEMA3F, CCL2, SPX, IGFBP4, ANGPT1, SERPINE1, FGF7, CCL8, MMP3, MMP1, CCL20, CCL5, CXCL8, and FOXA1) from the candidate genes ( Fig. 2 b ) . The intersection of genes from the two algorithms yielded five feature genes: SERPINE1, CCL2, IGFBP4, C3, and SPX ( Fig. 2 c ) . Subsequently, the feature gene expression levels between the EP and control groups were analyzed in the GSE256068 and GSE134697 datasets. In comparison with the control group, the EP group exhibited significantly down-regulated IGFBP4 expression and significantly up-regulated expressions of SERPINE1, CCL2, C3, and SPX (p < 0.05) ( Fig. 2 d ) . It was worth noting that this expression trend was highly consistent in the two datasets. Based on the above findings, these five genes were identified as biomarkers for EP. Notably, the consistent up-regulation of SERPINE1, CCL2, C3, and SPX across both cortical and hippocampal epileptogenic tissues suggests a pervasive, multi-focal SASP activation during disease progression. 2.3 Nomogram had good predictive performance A nomogram prediction model was developed for evaluating the diagnostic efficiency of EP-related biomarkers. The model assigned specific scores to each biomarker, with the cumulative total score positively correlated with the risk of EP ( Fig. 3 a ) . Calibration curve analysis showed that the Hosmer-Lemeshow test yielded a P-value of 0.198, suggesting good concordance between predicted and observed outcomes ( Fig. 3 b ) . DCA demonstrated that the nomogram's net benefit significantly exceeded that of the treat-all and treat-none strategies, confirming its clinical utility ( Fig. 3 c ) . 2.4 Inferred immune microenvironment remodeling associated with SASP biomarkers GSEA analysis results showed that IGFBP4, C3, CCL2, SERPINE1, and SPX were significantly enriched in 89, 102, 80, 115, and 79 functional pathways, respectively (Supplementary Table S3-7) . When sorted by significance, IGFBP4 was primarily enriched in pathways such as Alzheimer's Disease and oxidative phosphorylation; C3 mainly showed enrichment in pathways like ribosome and leishmania infection; CCL2 demonstrated notable enrichment in pathways including ribosome and viral myocarditis; SERPINE1 was predominantly enriched in pathways such as cytokine-cytokine receptor interaction and leishmania infection; SPX exhibited significant enrichment in pathways including oxidative phosphorylation and ribosome ( Fig. 4 a ) . Subsequently, the CIBERSORT algorithm was used to evaluate the infiltration of 22 types of immune cells during EP progression ( Fig. 4 b ) . Results revealed significant differences in the infiltration abundance of six immune cells between the EP and control groups. Specifically, the control group exhibited higher infiltration abundance of cells such as CD8 T cells[ 59 – 63 ] (p < 0.05), while the EP group had higher infiltration abundance of cells such as mast cells[ 64 – 66 ] (p < 0.05) ( Fig. 4 c ) . Additionally, significant correlations were observed: SERPINE1 positively correlated with activated mast cells (0.304, p < 0.001) and negatively with follicular helper T cells (-0.326, p < 0.001); CCL2 positively with activated mast cells (0.636, p < 0.0001) and negatively with resting mast cells (-0.415, p < 0.0001); IGFBP4 positively with follicular helper T cells (0.505, p < 0.0001); C3 positively with eosinophils[ 67 , 68 ] (0.318, p < 0.001) and negatively with follicular helper T cells (-0.363, p < 0.0001); SPX positively with activated mast cells (0.508, p < 0.0001) and negatively with CD8 + T cells (-0.560, p < 0.0001). ( Fig. 4 d ) . This indicated that these genes might participate in shaping the immune microenvironment of epilepsy by regulating the activities of different immune cells. 2.5 Molecular regulatory mechanisms of SERPINE1, CCL2, IGFBP4, C3, and SPX Using the miRNet database, 217 miRNAs related to IGFBP4, 233 miRNAs related to C3, 63 miRNAs related to CCL2, 195 miRNAs related to SERPINE1, and 10 miRNAs related to SPX were successfully predicted. A miRNA-mRNA regulatory network was further constructed based on the aforementioned predictions ( Fig. 5 a ) . Since TFs could specifically recognize downstream target genes, subsequent analysis via the miRNet database revealed that IGFBP4 interacted with 3 transcription factors, C3 interacted with 1 transcription factor, CCL2 interacted with 19 transcription factors, and SERPINE1 interacted with 20 transcription factors ( Fig. 5 b ) . Notably, TFs such as NFKB1, RELA, and SP1 were found to simultaneously target both SERPINE1 and CCL2, suggesting these transcription factors might act as hubs in the same signaling pathway or biological process by coordinately regulating the expression of SERPINE1 and CCL2. We also utilized physical interaction and co-expression analysis in the GeneMANIA database and identified 20 genes functionally related to SERPINE1, CCL2, IGFBP4, C3, and SPX, such as CCL15 and PAPPA. Among them, CCL2, CCL5, and CXCL8 collectively participated in pathways such as cytokine activity and response to chemokine ( Fig. 5 c ) . 2.6 In silico prediction of potential pharmacological modulators To explore potential drugs for EP treatment, the R package enrichR was utilized to predict drugs targeting the biomarkers. The analysis revealed a total of 30 drugs targeting CCL2, 30 drugs targeting SERPINE1, 30 drugs targeting IGFBP4, 16 drugs targeting C3, and 3 drugs targeting SPX ( Fig. 6 a ) . Subsequently, molecular docking was performed on the drug with the smallest p-value in the above prediction and its corresponding biomarker to evaluate the binding capacity between the small-molecule drug and the target protein. The docking combinations included IGFBP4 with the drug amantadine, SPX with trans-Lutein, CCL2 with fludrocortisone, C3 with propofol, and SERPINE1 with the compound 5630-53-5. Binding energies below − 5 kcal/mol were observed for all biomarker-drug pairs in the docking analysis, confirming their strong binding affinity ( Table 1 ) ( Fig. 6 b ) . This finding suggested these molecules might serve as promising candidates for EP therapy. Table 1 Predicted free binding energies between active pharmaceutical ingredients and key genes Targets Drug Binding Energy(KJ/mol) IGFBP4 amantadine -5.4 SPX trans-Lutein -7.5 CCL2 fludrocortisone -6.1 C3 propofol -5.7 SERPINE1 5630-53-5 -7.1 2.7 Validation of biomarkers RT-qPCR analysis revealed significant down-regulation of IGFBP4 and up-regulation of SERPINE1, CCL2, and C3 in EP compared to control. Notably, SPX expression did not differ significantly between groups, and these findings were consistent with bioinformatics results, confirming method validity ( Fig. 7 ) . 3. Discussion Cellular senescence and the senescence-associated secretory phenotype (SASP) are increasingly recognized as amplifiers of chronic, sterile inflammation in the central nervous system. In epilepsy, sustained neuroinflammation and metabolic stress contribute to network hyperexcitability, blood–brain barrier (BBB) dysfunction, and progressive tissue remodeling, all of which may be reinforced by senescent-cell secretomes. Although emerging evidence supports a relationship between senescence programs and epileptogenesis [ 11 – 15 ], SASP-oriented molecular signatures and their neurochemical implications remain incompletely defined. In the present study, we integrated human brain transcriptomic data with a curated SASP gene set and applied two machine-learning feature-selection strategies (LASSO and SVM-RFE) to derive a concise SASP-related biomarker panel for epilepsy. Five genes—IGFBP4, SERPINE1, CCL2, C3, and SPX—showed robust and reproducible differential expression across two independent GEO datasets, and a nomogram built on these markers demonstrated favorable discrimination and calibration. Beyond diagnostic modeling, functional enrichment and immune deconvolution analyses linked these biomarkers to translational control, oxidative phosphorylation, and a reshaped immune microenvironment. Finally, experimental RT–qPCR in a kainic acid (KA) mouse model provided initial biological support for four markers (IGFBP4, SERPINE1, CCL2, and C3), while SPX did not show a statistically significant change in peripheral blood, highlighting the need for further validation and tissue-context interpretation. 3.1 A SASP-centered biomarker axis in epilepsy The identified panel is neurochemically coherent in that it spans multiple, convergent processes that are repeatedly implicated in epilepsy and senescence-associated inflammation: (i) growth-factor signaling and metabolic regulation (IGFBP4), (ii) extracellular proteolysis and BBB homeostasis (SERPINE1), (iii) chemokine-driven neuroimmune recruitment and glial activation (CCL2), (iv) complement activation and neuroimmune synapse remodeling (C3), and (v) neuropeptide signaling potentially coupled to inflammatory and metabolic tone (SPX). Importantly, several of these molecules are either secreted or closely associated with secretory pathways, consistent with the conceptual framework of the SASP. Together, these markers suggest that epilepsy-related senescence programs may not be limited to a single pathway but may instead reflect a coordinated, multi-layer neurochemical remodeling involving inflammatory mediators, metabolic stress responses, and extracellular matrix/vascular interfaces. 3.2 IGFBP4: IGF signaling balance and metabolic vulnerability Insulin-like growth factor-binding protein 4 (IGFBP4) is a secreted glycoprotein that modulates IGF bioavailability, thereby shaping cell survival, differentiation, and stress responses[ 20 – 24 , 27 – 30 ]. In our datasets, IGFBP4 was downregulated in epilepsy, a pattern that may indicate altered IGF-axis homeostasis in epileptogenic tissue. IGF-1 signaling has context-dependent effects in epilepsy: it can exert neuroprotective and anti-inflammatory actions in some models, yet IGF-1 receptor activation may also increase excitability in specific circuit states [ 25 , 26 ]. Reduced IGFBP4 could theoretically increase free IGF-1 availability; however, the net consequence likely depends on region-specific receptor distribution, cell-type composition (neurons vs. astrocytes/microglia), and the inflammatory milieu. Notably, IGFBP4-associated enrichment in oxidative phosphorylation pathways aligns with the concept that epileptogenic networks operate under heightened energetic demand and oxidative stress[ 48 – 55 ]. Mitochondrial dysfunction and ROS accumulation can promote both neuronal injury and senescence-like phenotypes, potentially feeding into SASP-associated inflammation. Therefore, IGFBP4 may represent a bridge between growth-factor signaling and metabolic resilience in epilepsy, and its directionality and compartment specificity (brain vs. circulation) merit targeted experimental clarification. 3.3 SERPINE1 (PAI-1): extracellular proteolysis, BBB integrity, and inflammatory persistence SERPINE1 (plasminogen activator inhibitor-1, PAI-1) is the principal inhibitor of tissue-type and urokinase-type plasminogen activators, thereby regulating fibrinolysis and extracellular proteolysis [ 31 – 33 ]. Elevated SERPINE1 is widely associated with pro-thrombotic states and tissue fibrosis, whereas deficiency predisposes to bleeding due to hyperfibrinolysis, underscoring its importance in vascular biology[ 31 – 36 ]. SERPINE1 is also a well-recognized senescence-associated factor in multiple contexts, linking it conceptually to SASP-driven tissue remodeling. In epilepsy, BBB dysfunction is a key contributor to neuroinflammation and seizure propagation. Extracellular protease cascades (plasmin/MMPs) can influence tight-junction stability and basement membrane integrity[ 37 – 41 ]. Altered SERPINE1 may therefore participate in a feedback loop in which vascular and extracellular matrix remodeling facilitates immune-cell entry and glial activation, which in turn sustains inflammatory cytokine production and excitability. While our study does not establish causality, the consistent upregulation of SERPINE1 across datasets and its association with immune-infiltration features suggests that SERPINE1 may be an informative marker of a proteolysis–BBB–inflammation axis in epileptogenic tissue. 3.4 CCL2: chemokine amplification of neuroimmune circuits CCL2 is a central chemokine produced by astrocytes, microglia, endothelial cells, and neurons under inflammatory challenge. Through CCR2 signaling, CCL2 recruits peripheral monocytes/macrophages and reshapes local neuroimmune signaling, thereby amplifying cytokine cascades and potentially modulating inhibitory/excitatory balance [ 42 , 43 ]. Prior evidence supports CCL2 induction after seizures and its ability to promote pro-inflammatory cytokine release and neuronal injury [ 42 , 43 ]. In our analysis, CCL2 was among the most prominent upregulated SASP-related markers, consistent with the idea that a senescence-linked secretory environment may favor sustained chemokine signaling. From a neurochemical perspective, chronic CCL2 signaling can indirectly increase neuronal excitability by intensifying IL-1β/TNF-α–dependent modulation of synaptic receptor trafficking and by promoting microglial activation states that alter synaptic pruning and glutamate homeostasis. Thus, CCL2 likely represents both a biomarker and a plausible contributor to epileptogenic neuroinflammation. 3.5 C3: complement-driven neuroimmune remodeling and excitability Complement component C3 is a pivotal node of the complement cascade and has been increasingly linked to neuroinflammation, synapse remodeling, and microglial effector functions in multiple neurological disorders [ 44 – 46 ]. In epilepsy models, astrocytic upregulation of C3 and signaling via the C3a–C3aR axis can enhance microglial activation and inflammatory cytokine release, which may influence inhibitory tone and excitatory drive [ 46 ]. The consistent upregulation of C3 in our epilepsy datasets, together with its correlations with inferred immune-cell changes, supports complement activation as a relevant neurochemical feature of epileptogenic tissue. Importantly, complement activity can also intersect with senescence biology: senescent cells and SASP mediators may promote complement activation, while complement-driven inflammation can reinforce cellular stress responses. Therefore, C3 may be positioned at the intersection of SASP-linked inflammation and circuit remodeling, providing a mechanistically plausible bridge between molecular signatures and network-level epilepsy phenotypes. 3.6 SPX: neuropeptide signaling with tissue- and compartment-specific behavior Spexin (SPX) is a neuropeptide with broad tissue expression and reported roles in metabolic regulation, oxidative stress, and inflammatory tone via galanin receptor signaling [ 47 ]. In our transcriptomic analyses, SPX was upregulated in epilepsy, suggesting potential involvement in epileptogenic remodeling or compensatory neuropeptide responses. However, SPX did not show a statistically significant change in peripheral blood RT–qPCR in the KA mouse model. This discrepancy can be interpreted in several non-mutually exclusive ways: SPX regulation may be brain-region specific, time dependent (acute vs. chronic stages), or primarily localized to neural tissue without a strong peripheral signature. Additionally, neuropeptide transcripts can exhibit higher variability in blood and may require larger cohorts or protein-level assays for reliable detection. Accordingly, SPX should be considered a transcriptome-supported candidate whose translational utility (especially as a blood biomarker) requires additional validation in matched brain and plasma/serum samples, ideally including protein quantification. 3.7 Pathway-level interpretation: translation control and oxidative phosphorylation A recurring signal in our enrichment analyses was the involvement of ribosome/translation-related pathways and oxidative phosphorylation. These findings are neurochemically consistent with epilepsy as a disorder of sustained energetic strain and activity-dependent proteostasis. Hyperexcitability increases ATP demand, elevates ROS production, and can shift mitochondrial function, while chronic inflammation can further impair oxidative phosphorylation and redox homeostasis. In parallel, translational control is strongly regulated by mTOR signaling, which is frequently implicated in epilepsy and synaptic remodeling [ 56 – 58 ]. Increased ribosome biogenesis and dysregulated translation can modify receptor composition and synaptic protein abundance, influencing excitability thresholds and network synchronization. Within a SASP framework, these pathway signals are also plausible: oxidative stress and mitochondrial dysfunction can promote senescence-like states and reinforce SASP output, while sustained inflammatory signaling can alter translation programs in glia and neurons. Therefore, the pathway results complement the biomarker panel by suggesting that epilepsy-associated senescence signatures may be embedded in a broader metabolic–proteostatic remodeling landscape. 3.8 Upstream regulatory networks: NF-κB as a master SASP driver Beyond downstream effector functions, our in silico regulatory network analysis (Section 2.5 ) identified the transcription factors NFKB1 and RELA as crucial hubs simultaneously targeting SERPINE1 and CCL2. This convergence is highly mechanistically relevant, as the NF-κB complex (comprising NFKB1 and RELA subunits) is established as the primary transcriptional master regulator of the SASP [ 10 , 11 ]. In the context of epilepsy, chronic metabolic stress and recurrent seizures can induce DNA damage and persistent NF-κB activation within the neuroglial microenvironment [ 11 , 42 ]. Our network predictions suggest that this upstream NF-κB activation directly orchestrates the concurrent up-regulation of chemokines (CCL2) and extracellular matrix remodelers (SERPINE1). Consequently, identifying NFKB1 and RELA as shared upstream regulators not only corroborates the senescence-associated nature of our biomarker panel, but also highlights a potential therapeutic vulnerability: targeting the NF-κB/SASP axis might concurrently silence multiple epileptogenic effectors, offering a more comprehensive intervention than neutralizing single downstream cytokines. 3.9 Immune deconvolution: hypothesis-generating signals requiring orthogonal validation Immune-infiltration estimation suggested differences in the relative abundance of several immune-cell populations between epilepsy and control tissues and identified correlations between biomarkers (notably CCL2, SERPINE1, C3, and SPX) and immune-cell signatures. These results are consistent with the established role of neuroimmune interactions in epilepsy; however, they must be interpreted with distinct caution. Crucially, standard CIBERSORT-based deconvolution relies on reference signatures primarily derived from peripheral immune cells. The central nervous system possesses a highly specialized immune microenvironment dominated by resident microglia and astrocytes. Therefore, transcriptomic signals computationally mapped to peripheral subsets (such as mast cells or CD8 + T cells) might partially reflect the overlapping transcriptional states of activated resident glia, rather than definitive peripheral infiltration. Consequently, these inferred immune shifts should be viewed strictly as hypothesis-generating. Future investigations utilizing orthogonal approaches, such as spatial transcriptomics, single-cell RNA sequencing (scRNA-seq), or multi-parameter flow cytometry on matched resected epileptogenic tissues, are imperative to validate the precise cellular sources of these SASP mediators. 3.10 Translational implications and drug-repurposing signals The nomogram based on five SASP-related markers showed favorable diagnostic performance in the analyzed datasets, supporting the potential of a multi-marker strategy over single-gene assessment. Our KA mouse RT–qPCR results provide an initial step toward experimental validation and raise the possibility that brain-derived inflammatory remodeling may be reflected systemically, potentially through BBB perturbation and neuroimmune signaling. Furthermore, we performed in silico drug prediction and molecular docking to generate hypotheses for potential therapeutic targeting of the SASP network. While the calculated docking energies suggest structural binding feasibility between the identified compounds and our biomarker targets, these computational simulations do not establish pharmacodynamic efficacy, blood-brain barrier penetrability, in vivo target engagement, or actual anti-seizure activity. Thus, these structural findings are best regarded strictly as a computational prioritization tool to guide future in vitro and in vivo neuropharmacological testing, rather than a confirmation of therapeutic viability. 3.11 Limitations Several crucial limitations in this study should be acknowledged. First, our bioinformatics pipeline relies on retrospective public bulk transcriptomic datasets; differences in sample processing, platform effects, and clinical heterogeneity inevitably influence the results. Second, regarding experimental validation, although peripheral-blood RT–qPCR supports four biomarkers (IGFBP4, SERPINE1, CCL2, and C3) in a controlled KA mouse model, this cross-species and cross-compartment design (human brain transcriptomics vs. mouse peripheral blood validation) represents an indirect assessment. Peripheral blood mRNA fluctuations may serve as proxy indicators of systemic inflammatory responses linked to BBB disruption, but they do not directly equate to the local neurochemical alterations within the epileptogenic focus. The lack of direct validation at the protein level (e.g., via Western blotting or immunohistochemistry) within matched brain tissues is a notable constraint. Finally, mechanistic causality was not tested. While our study highlights the robust diagnostic potential of these SASP-related targets, functional experiments—such as targeted genetic or pharmacologic modulation of the SERPINE1-linked extracellular pathways or the CCL2/CCR2 axis in vivo—will be necessary to establish whether these biomarkers act as active drivers or secondary consequences of epileptogenic remodeling. 3.12 Conclusions In summary, we identified and validated a SASP-associated biomarker panel (IGFBP4, SERPINE1, CCL2, C3, and SPX) that discriminates epilepsy from controls and is linked to inflammatory, metabolic, and translational remodeling. These findings expand the neurochemical framework connecting senescence-associated secretory programs to epileptogenesis and provide candidate markers and pathways for future mechanistic and translational investigation. 4. Methods 4.1 Data collection Datasets GSE256068 and GSE134697 were retrieved from GEO (https://www.ncbi.nlm.nih.gov/geo/). Among them, GSE256068 was from the GPL24676 platform, which included cortical tissue samples from 121 epilepsy patients and 12 controls. Samples with the type of hippocampal tissue were excluded. GSE134697 was from the GPL16791 platform, in which 17 epilepsy samples with the type of hippocampus were regarded as the disease group, and 17 cortical samples served as the control group. Samples without epilepsy patients were excluded. By searching the references,[69, 70] a total of 153 senescence-associated secretory phenotype-related genes (SASP-RGs) were obtained ( Supplementary Table S 8) . While GSE134697 utilizes cortical tissues as controls for hippocampal epileptic samples, we deliberately leveraged this anatomically distinct dataset as a stringent cross-regional validation cohort. The fact that our 5-gene SASP signature (derived from cortical EP vs. control in GSE256068) maintained consistent differential expression in the hippocampal EP cohort underscores that this SASP signature is not a region-specific artifact, but rather a conserved neurochemical hallmark of the epileptogenic network across different brain regions. 4.2 Analysis of differential gene expression Using the "limma" package (v 3.54.0) [71], DEGs were screened between EP and controls in GSE256068 (p 1.5). For visualizing DEGs, a volcano plot and heatmap were created via "ggplot2" (v 3.4.1)[71]and "pheatmap" packages (v 1.0.12).[72] The top 10 DEGs were annotated in the figures, sorted by ascending order of |log 2 FC|. 4.3 Functional analyses of genes The "ggvenn" package (v 0.1.9)[73] was applied to intersect DEGs and SASP-RGs, and the resultant genes were then termed candidate genes. GO and KEGG analyses were carried out by leveraging the "clusterProfiler" package (v 4.2.2) [73] (p < 0.05). By sorting in descending order of significance based on p-values, the top 5 GO and top 10 KEGG results were visualized. A PPI network was built through STRING (https://string-db.org/) (confidence = 0.15). 4.4 Machine learning To further screen for feature genes, the "glmnet" package (v 4.1.18)[74] was utilized to conduct the LASSO regression analysis. The genes were screened from candidate genes by using 10-fold cross-validation and taking log(λ.min) as the screening threshold. The R package "e1071" (v 1.7.16)[75] was used to execute the SVM-RFE analysis, yielding the importance scores and corresponding ranks for each gene. Meanwhile, the error rate and accuracy of each iterative combination were calculated, and the combination with the highest accuracy was selected as the optimal one to screen out genes. Then, the "VennDiagram" package (v 1.7.1)[76] was used to obtain the intersection of genes derived from the two algorithms, yielding the final characteristic genes. Subsequently, feature gene expression in EP and control samples across GSE256068 and GSE134697 was analyzed via the Wilcoxon test. Genes that showed significantly differential expression and consistent expression patterns in both datasets were designated as biomarkers (p < 0.05). 4.5 Construction and validation of the nomogram To explore the role of biomarkers in the prediction of EP, the "RMS" package (v 6.8.1) [77] was employed to develop a nomogram model using the biomarkers as inputs. Subsequently, the "regplot" package (v 1.1) [77] was used to plot the calibration curve (Hosmer-Lemeshow, p > 0.05) and DCA (with the net benefit value greater than 0 as the criterion for good predictive performance). 4.6 GSEA Based on EP and control samples from GSE256068, as the reference gene set, "c2.cp.kegg.v7.5.1.symbols.gmt" was chosen from MSigDB (https://www.gsea-msigdb.org/gsea/msigdb). The "psych" package (v 2.1.6) [78] computed correlation coefficients between biomarkers and all other genes, with genes sorted in descending order of coefficients to derive related gene lists for each biomarker. Subsequently, the "clusterProfiler" package (v 4.2.2) was employed for GSEA (|NES| > 1, FDR < 0.25, and p < 0.05). 4.7 Immune cell infiltration To comprehensively investigate the dynamic immune infiltration during EP progression, based on EP and control samples in the GSE256068, the CIBERSORT algorithm was adopted to calculate relative abundances of 22 immune cells. Differences in immune cell infiltration between EP and control samples were analyzed by the Wilcoxon test (p 0.3, p < 0.05). 4.8 GENEMANIA To investigate the upstream and downstream regulatory interactions of biomarkers, using miRNet (https://www.mirnet.ca/), we predicted miRNAs and transcriptional regulators of the biomarkers. The miRNA-mRNA and TF-mRNA regulatory networks were then constructed using Cytoscape software. The GeneMANIA database ( https://genemania.org/ ) was applied to find genes that were functionally similar to the biomarkers. 4.9 Drug prediction and molecular docking Drugs targeting biomarkers were predicted via the "enrichR" package (v 3.2)[79]. To assess binding affinity, the top-ranked drug (based on the minimal p-value) was chosen for docking. 3D structures of biomarker-encoded proteins were retrieved from NCBI and UniProt (https://www.uniprot.org/), converted to PDB files, and designated as docking receptors. From PubChem ( https://pubchem.ncbi.nlm.nih.gov/ ), the 2D structures of predicted drugs in SDF format were obtained and converted to 3D mol3 format using ChemBio2D. Molecular docking was performed via the CB-Dock2 ( https://cadd.labshare.cn/cb-dock2/php/index.php ). 4.10 RT-qPCR Biomarker expression validation was performed via RT-qPCR. Five KA-induced epileptic mice and five saline-treated control mice (C57BL/6J, male, 8 weeks old) were used. Whole blood samples were collected from Experimental Animal Center of Chongqing Medical University. The ethical approval was provided by The Ethics Committee off Chongqing Medical University with the approval number IACUC-CQMU-2025-09030. The results were plotted using Graphpad Prism. First, RNA was isolated from 10 samples using TRIzol (Novozymes, Nanjing, China). Total RNA was converted to cDNA using HP All-in-one qRT Master Mix II RT203-Ver.1 kit (Yungene Bio, Kunming, China) following the provided protocols. RT-qPCR was performed with 2×Universal Blue SYBR Green qPCR Master Mix (Servicebio, Wuhan, China). The primer sequences are shown in Table 2 . Ethics declarations We certify that the research study titled Identification and validation of senescence-associated secretory phenotype-related biomarkers in epilepsy has been approved by the relevant ethics committee or institutional review board (IRB). The approval number is IACUC-CQMU-2025-09030. We confirm that all experiments were performed in accordance with ARRIVE guidelines and American Veterinary Medical Association (AVMA) guidelines. Patient Informed Consent As the human transcriptomic data analyzed in this study were obtained from the publicly available Gene Expression Omnibus (GEO) database, the requirement for ethical approval and patient informed consent was waived. Table 2 . The table of primer sequences. Gene number gene sequences Product length XM_036156346.1 IGFBP4 F CAGACACAAGAAGCGAACTCC 268 XM_036156346.1 IGFBP4 R CGCCCTGCTTAGATTCTGGA 268 NM_008871.2 SERPINE1 F CTCCAAGGGGCAACGGATAG 84 NM_008871.2 SERPINE1 R AAGCAAGCTGTGTCAAGGGA 84 NM_011333.3 CCL2 F CCACAACCACCTCAAGCACT 77 NM_011333.3 CCL2 R TAAGGCATCACAGTCCGAGTC 77 NM_009778.3 C3 F TTCCTTCACTATGGGACCAGC 127 NM_009778.3 C3 R CTCCAGCCGTAGGACATTGG 127 NM_001369015.1 SPX F TAGGGCGAGAGATTGTGGGG 159 NM_001369015.1 SPX R TTGGGGAGTCCAGTTCCTCT 159 NM_001411844.1 M-GAPDH F TGTGTCCGTCGTGGATCTGA 153 NM_001411844.1 M-GAPDH R GAGTTGCTGTTGAAGTCGCA 153 4.11 Statistical analysis R software (v 4.2.2) was employed. Intergroup differences were compared via the Wilcoxon test (p < 0.05). Declarations Acknowledgements We would like to express our sincere gratitude to all individuals and organizations who supported and assisted us throughout this research. Author contributions Difei Wang, Hui Gan and Yuxin Wu, Baohui Yang, Han Xiao wrote the main manuscript text and Haotian Tang, Hanli Qiu, Xinyu Dong, Kaiyi Kangprepared figures 1–3. All authors reviewed the manuscript. Competing Interests Statement All the authors declare no competing interests Data availability statement The datasets analysed during the current study are available in the Gene Expression Omnibus (GEO) repository (GSE256068 and GSE134697). Consent to publish Not applicable. Funding This research received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References Kanner AM, Bicchi MM. Antiseizure medications for adults with epilepsy: A review. JAMA. 2022;327:1269-1281. https://doi.org/10.1001/jama.2022.3880 Felix R, Ekman J, Bjellvi S, Ljunggren M, et al. Laser interstitial thermal therapy versus open surgery for mesial temporal lobe epilepsy: A systematic review and meta-analysis. World Neurosurg. 2024;192:224-235.e15. https://doi.org/10.1016/j.wneu.2024.09.090 Perucca E, Perucca P, White HS, et al. Drug resistance in epilepsy. Lancet Neurol. 2023;22:723-734. https://doi.org/10.1016/S1474-4422(23)00151-5 Kitschen A, Aleknonytė-Resch M, Sakalytė G, et al. Cost-effectiveness of surgical treatment compared to medical treatment in patients with drug-refractory epilepsy: A systematic review. Eur J Neurol. 2023;30:749-761. https://doi.org/10.1111/ene.15632 Li X, Guo S, Liu K, et al. Gabrg2 deletion linked to genetic epilepsy with febrile seizures plus affects the expression of GABA(A) receptor subunits and other genes at different temperatures. Neuroscience. 2020;438:116-136. https://doi.org/10.1016/j.neuroscience.2020.04.041 Socała K, Doboszewska U, Szopa A, et al. The role of microbiota-gut-brain axis in neuropsychiatric and neurological disorders. Pharmacol Res. 2021;172:105840. https://doi.org/10.1016/j.phrs.2021.105840 Perucca E, White HS, Bialer M. New GABA-targeting therapies for the treatment of seizures and epilepsy: II. Treatments in clinical development. CNS Drugs. 2023;37:781-795. https://doi.org/10.1007/s40263-023-01025-4 Yan R, Zhang L, Chen Y, et al. Therapeutic potential of gut microbiota modulation in epilepsy: A focus on short-chain fatty acids. Neurobiol Dis. 2025;209:106880. https://doi.org/10.1016/j.nbd.2025.106880 Thoppil H, Riabowol K. Senolytics: A translational bridge between cellular senescence and organismal aging. Front Cell Dev Biol. 2019;7:367. https://doi.org/10.3389/fcell.2019.00367 Ohtani N. The roles and mechanisms of senescence-associated secretory phenotype (SASP): Can it be controlled by senolysis? Inflamm Regen. 2022;42:11. https://doi.org/10.1186/s41232-022-00197-8 Ge Q, Yang J, Huang F, et al. Multimodal single-cell analyses reveal molecular markers of neuronal senescence in human drug-resistant epilepsy. J Clin Invest. 2025;135. https://doi.org/10.1172/JCI188942 Leo A, Nesci V, Tallarico M, et al. IL-6 receptor blockade by tocilizumab has anti-absence and anti-epileptogenic effects in the WAG/Rij rat model of absence epilepsy. Neurotherapeutics. 2020;17:2004-2014. https://doi.org/10.1007/s13311-020-00894-w Wang Q, Qin B, Yu H, et al. Dingxian pill alleviates hippocampal neuronal apoptosis in epileptic mice through TNF-α/TNFR1 signaling pathway inhibition. J Ethnopharmacol. 2024;334:118579. https://doi.org/10.1016/j.jep.2024.118579 Pijet B, Konopka A, Rejmak E, et al. The matrix metalloproteinase inhibitor marimastat inhibits seizures in a model of kainic acid-induced status epilepticus. Sci Rep. 2020;10:21314. https://doi.org/10.1038/s41598-020-78341-y Ribierre T, Bacq A, Donneger F, et al. Targeting pathological cells with senolytic drugs reduces seizures in neurodevelopmental mTOR-related epilepsy. Nat Neurosci. 2024;27:1125-1136. https://doi.org/10.1038/s41593-024-01634-2 Basnyat P, Peltola M, Raitanen J, et al. Elevated IL-6 plasma levels are associated with GAD antibodies-associated autoimmune epilepsy. Front Cell Neurosci. 2023;17:1129907. https://doi.org/10.3389/fncel.2023.1129907 Soltani Khaboushan A, Yazdanpanah N, Rezaei N. Neuroinflammation and proinflammatory cytokines in epileptogenesis. Mol Neurobiol. 2022;59:1724-1743. https://doi.org/10.1007/s12035-021-02685-z Mahawar S, Rakshit D, Patel I, et al. Fisetin-loaded chitosan nanoparticles ameliorate pilocarpine-induced temporal lobe epilepsy and associated neurobehavioral alterations in mice: Role of ROS/TNF-α-NLRP3 inflammasomes pathway. Nanomedicine. 2024;59:102752. https://doi.org/10.1016/j.nano.2024.102752 Bronisz E, Kurkowska-Jastrzębska I. Matrix metalloproteinase 9 in epilepsy: The role of neuroinflammation in seizure development. Mediators Inflamm. 2016;2016:7369020. https://doi.org/10.1155/2016/7369020 Wang S, Chi K, Wu D, et al. Insulin-like growth factor binding proteins in kidney disease. Front Pharmacol. 2021;12:807119. https://doi.org/10.3389/fphar.2021.807119 Yang B, Zhang L, Cao Y, et al. Overexpression of lncRNA IGFBP4-1 reprograms energy metabolism to promote lung cancer progression. Mol Cancer. 2017;16:154. https://doi.org/10.1186/s12943-017-0723-5 Bach LA. Insulin-like growth factor binding proteins 4-6. Best Pract Res Clin Endocrinol Metab. 2015;29:713-722. https://doi.org/10.1016/j.beem.2015.06.002 Lee YY, Mok MT, Kang W, et al. Loss of tumor suppressor IGFBP4 drives epigenetic reprogramming in hepatic carcinogenesis. Nucleic Acids Res. 2018;46:8832-8847. https://doi.org/10.1093/nar/gky589 Liang J, Huang Y, Peng D, et al. IGFBP2 and IGFBP4 interact to activate complement pathway in diabetic kidney disease. Ren Fail. 2025;47:2440528. https://doi.org/10.1080/0886022X.2024.2440528 Wu L, Qin Y, Yuan H, et al. Anti-inflammatory and neuroprotective effects of insulin-like growth factor-1 overexpression in pentylenetetrazole (PTZ)-induced mouse model of chronic epilepsy. Brain Res. 2022;1785:147881. https://doi.org/10.1016/j.brainres.2022.147881 Jiang G, Wang W, Cao Q, et al. Insulin growth factor-1 (IGF-1) enhances hippocampal excitatory and seizure activity through IGF-1 receptor-mediated mechanisms in the epileptic brain. Clin Sci (Lond). 2015;129:1047-1060. https://doi.org/10.1042/CS20150394 Dya GA, Lebedeva OS, Gushchevarov DA, et al. Specific cleavage of IGFBP-4 by PAPP-A in nervous tissue. Biochem Biophys Res Commun. 2024;733:150655. https://doi.org/10.1016/j.bbrc.2024.150655 Jiang X, Zhao J, Ju L, et al. Temporal expression patterns of insulin-like growth factor binding protein-4 in the embryonic and postnatal rat brain. BMC Neurosci. 2013;14:132. https://doi.org/10.1186/1471-2202-14-132 Niu H, Gou R, Xu Q, et al. Recombinant insulin-like growth factor binding protein-4 inhibits proliferation and promotes differentiation of neural progenitor cells. Neurosci Lett. 2017;642:71-76. https://doi.org/10.1016/j.neulet.2017.02.040 Zhang J, Moats-Staats BM, Ye P, et al. Expression of insulin-like growth factor system genes during the early postnatal neurogenesis in the mouse hippocampus. J Neurosci Res. 2007;85:1618-1627. https://doi.org/10.1002/jnr.21275 Catarinella G, Nicoletti C, Bracaglia A, et al. Serpine1 drives a cell-autonomous pathogenic signaling in Hutchinson-Gilford progeria syndrome. Cell Death Dis. 2022;13:737. https://doi.org/10.1038/s41419-022-05168-y Ghosh AK, Vaughan DE. PAI-1 in tissue fibrosis. J Cell Physiol. 2012;227:493-507. https://doi.org/10.1002/jcp.22783 Sillen M, Declerck PJ. A narrative review on plasminogen activator inhibitor-1 and its (patho)physiological role: To target or not to target? Int J Mol Sci. 2021;22. https://doi.org/10.3390/ijms22052721 Wang B, Gu B, Gao L, et al. Serpine1 facilitates metastasis in gastric cancer through anoikis resistance and tumor microenvironment remodeling. Small. 2025;21:e2500136. https://doi.org/10.1002/smll.202500136 Su YH, Wu YZ, Ann DK, et al. Obesity promotes radioresistance through serpine1-mediated aggressiveness and DNA repair of triple-negative breast cancer. Cell Death Dis. 2023;14:53. https://doi.org/10.1038/s41419-023-05576-8 Peng S, Xue G, Gong L, et al. A long-acting PAI-1 inhibitor reduces thrombus formation. Thromb Haemost. 2017;117:1338-1347. https://doi.org/10.1160/TH17-03-0185 Neuhaus W, Gaiser F, Mahringer A, et al. The pivotal role of astrocytes in an in vitro stroke model of the blood-brain barrier. Front Cell Neurosci. 2014;8:352. https://doi.org/10.3389/fncel.2014.00352 Dohgu S, Takata F, Matsumoto J, et al. Autocrine and paracrine up-regulation of blood-brain barrier function by plasminogen activator inhibitor-1. Microvasc Res. 2011;81:103-107. https://doi.org/10.1016/j.mvr.2010.10.007 Shan Y, Tan S, Lin Y, et al. The glucagon-like peptide-1 receptor agonist reduces inflammation and blood-brain barrier breakdown in an astrocyte-dependent manner in experimental stroke. J Neuroinflammation. 2019;16:242. https://doi.org/10.1186/s12974-019-1638-6 Cai Y, Liu J, Wang B, et al. Microglia in the neuroinflammatory pathogenesis of Alzheimer's disease and related therapeutic targets. Front Immunol. 2022;13:856376. https://doi.org/10.3389/fimmu.2022.856376 Yu W, He J, Cai X, et al. Neuroimmune crosstalk between the peripheral and the central immune system in amyotrophic lateral sclerosis. Front Aging Neurosci. 2022;14:890958. https://doi.org/10.3389/fnagi.2022.890958 Cerri C, Genovesi S, Allegra M, et al. The chemokine CCL2 mediates the seizure-enhancing effects of systemic inflammation. J Neurosci. 2016;36:3777-3788. https://doi.org/10.1523/JNEUROSCI.4513-15.2016 Tian DS, Peng J, Murugan M, et al. Chemokine CCL2-CCR2 signaling induces neuronal cell death via STAT3 activation and IL-1β production after status epilepticus. J Neurosci. 2017;37:7878-7892. https://doi.org/10.1523/JNEUROSCI.0173-17.2017 Zhang S, Gao Z, Feng L, et al. Prevention and treatment strategies for Alzheimer's disease: Focusing on microglia and astrocytes in neuroinflammation. J Inflamm Res. 2024;17:7235-7259. https://doi.org/10.2147/JIR.S483412 Tatomir A, Talpos-Caia A, Anselmo F, et al. The complement system as a biomarker of disease activity and response to treatment in multiple sclerosis. Immunol Res. 2017;65:1103-1109. https://doi.org/10.1007/s12026-017-8946-3 Wei Y, Chen T, Bosco DB, et al. The complement C3-C3aR pathway mediates microglia-astrocyte interaction following status epilepticus. Glia. 2021;69:1155-1169. https://doi.org/10.1002/glia.23955 Türkel İ, Memi G, Yazgan B. Impact of spexin on metabolic diseases and inflammation: An updated minireview. Exp Biol Med (Maywood). 2022;247:567-573. https://doi.org/10.1177/15353702211073145 Martín-Rivada Á, Guerra-Cantera S, Campillo-Calatayud A, et al. Pappalysins and stanniocalcins and their relationship with the peripheral IGF axis in newborns and during development. J Clin Endocrinol Metab. 2022;107:2912-2924. https://doi.org/10.1210/clinem/dgac109 Fluca AL, Pani B, Janjusevic M, et al. Unraveling the relationship among insulin resistance, IGF-1, and amyloid-beta 1-40: Is the definition of type 3 diabetes applicable in the cardiovascular field? Life Sci. 2024;352:122911. https://doi.org/10.1016/j.lfs.2024.122911 Opazo P, Viana da Silva S, Carta M, et al. CaMKII metaplasticity drives Aβ oligomer-mediated synaptotoxicity. Cell Rep. 2018;23:3137-3145. https://doi.org/10.1016/j.celrep.2018.05.022 Fang Y, Si X, Wang J, et al. Alzheimer disease and epilepsy: A mendelian randomization study. Neurology. 2023;101:e399-e409. https://doi.org/10.1212/WNL.0000000000207435 Vossel KA, Tartaglia MC, Nygaard HB, et al. Epileptic activity in Alzheimer's disease: Causes and clinical relevance. Lancet Neurol. 2017;16:311-322. https://doi.org/10.1016/S1474-4422(17)30044-2 Purnell BS, Alves M, Boison D. Astrocyte-neuron circuits in epilepsy. Neurobiol Dis. 2023;179:106058. https://doi.org/10.1016/j.nbd.2023.106058 Molina DP, Ariwodola OJ, Weiner JL, et al. Growth hormone and insulin-like growth factor-I alter hippocampal excitatory synaptic transmission in young and old rats. Age (Dordr). 2013;35:1575-1587. https://doi.org/10.1007/s11357-012-9458-1 Street ME, Miraki-Moud F, Sanderson IR, et al. Interleukin-1beta (IL-1beta) and IL-6 modulate insulin-like growth factor-binding protein (IGFBP) secretion in colon cancer epithelial (Caco-2) cells. J Endocrinol. 2003;179:405-415. https://doi.org/10.1677/joe.0.1790405 Gerasimenko A, Baldassari S, Baulac SM. mTOR pathway: Insights into an established pathway for brain mosaicism in epilepsy. Neurobiol Dis. 2023;182:106144. https://doi.org/10.1016/j.nbd.2023.106144 Zhao W, Xie C, Zhang X, et al. Advances in the mTOR signaling pathway and its inhibitor rapamycin in epilepsy. Brain Behav. 2023;13:e2995. https://doi.org/10.1002/brb3.2995 Mazumder AG, Sharma P, Patial V, et al. Ginkgo biloba L. Attenuates spontaneous recurrent seizures and associated neurological conditions in lithium-pilocarpine rat model of temporal lobe epilepsy through inhibition of mammalian target of rapamycin pathway hyperactivation. J Ethnopharmacol. 2017;204:8-17. https://doi.org/10.1016/j.jep.2017.03.032 Korn T, Kallies A. T cell responses in the central nervous system. Nat Rev Immunol. 2017;17:179-194. https://doi.org/10.1038/nri.2016.144 Garber C, Soung A, Vollmer LL, et al. T cells promote microglia-mediated synaptic elimination and cognitive dysfunction during recovery from neuropathogenic flaviviruses. Nat Neurosci. 2019;22:1276-1288. https://doi.org/10.1038/s41593-019-0427-y Suzuki Y. The immune system utilizes two distinct effector mechanisms of T cells depending on two different life cycle stages of a single pathogen, Toxoplasma gondii, to control its cerebral infection. Parasitol Int. 2020;76:102030. https://doi.org/10.1016/j.parint.2019.102030 Wu L, Ji NN, Wang H, et al. Domino effect of interleukin-15 and CD8 T-cell-mediated neuronal apoptosis in experimental traumatic brain injury. J Neurotrauma. 2021;38:1450-1463. https://doi.org/10.1089/neu.2020.7303 Gutiérrez-Melo N, Baumjohann D. T follicular helper cells in cancer. Trends Cancer. 2023;9:309-325. https://doi.org/10.1016/j.trecan.2023.01.002 Tanaka S, Furuta K. Roles of IgE and histamine in mast cell maturation. Cells. 2021;10. https://doi.org/10.3390/cells10082121 Yue J, Tan Y, Huan R, et al. Mast cell activation mediates blood-brain barrier impairment and cognitive dysfunction in septic mice in a histamine-dependent pathway. Front Immunol. 2023;14:1090288. https://doi.org/10.3389/fimmu.2023.1090288 Huang X, Lan Z, Hu Z. Role and mechanisms of mast cells in brain disorders. Front Immunol. 2024;15:1445867. https://doi.org/10.3389/fimmu.2024.1445867 Manohar M, Kandikattu HK, Upparahalli Venkateshaiah S, et al. Eosinophils in the pathogenesis of pancreatic disorders. Semin Immunopathol. 2021;43:411-422. https://doi.org/10.1007/s00281-021-00863-7 Gu Q, Wiggers ME, Gleich GJ, et al. Sensitization of isolated rat vagal pulmonary sensory neurons by eosinophil-derived cationic proteins. Am J Physiol Lung Cell Mol Physiol. 2008;294:L544-552. https://doi.org/10.1152/ajplung.00392.2007 Lee MK, Woo SR, Noh JK, et al. Prognostic significance of SASP-related gene signature of radiation therapy in head and neck squamous cell carcinoma. Mol Cancer Ther. 2024;23:1348-1359. https://doi.org/10.1158/1535-7163.MCT-23-0738 Saul D, Kosinsky RL, Atkinson EJ, et al. A new gene set identifies senescent cells and predicts senescence-associated pathways across tissues. Nat Commun. 2022;13:4827. https://doi.org/10.1038/s41467-022-32552-1 Ritchie ME, Phipson B, Wu D, et al. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. https://doi.org/10.1093/nar/gkv007 Zhang X, Chao P, Zhang L, et al. Single-cell RNA and transcriptome sequencing profiles identify immune-associated key genes in the development of diabetic kidney disease. Front Immunol. 2023;14:1030198. https://doi.org/10.3389/fimmu.2023.1030198 Zhou W, Li H, Zhang J, et al. Identification and mechanism analysis of biomarkers related to butyrate metabolism in COVID-19 patients. Ann Med. 2025;57:2477301. https://doi.org/10.1080/07853890.2025.2477301 Zhang H, Zheng C, Xu Y, et al. Comprehensive molecular and cellular characterization of endoplasmic reticulum stress-related key genes in renal ischemia/reperfusion injury. Front Immunol. 2024;15:1340997. https://doi.org/10.3389/fimmu.2024.1340997 Shi H, Yuan X, Liu G, et al. Identifying and validating GSTM5 as an immunogenic gene in diabetic foot ulcer using bioinformatics and machine learning. J Inflamm Res. 2023;16:6241-6256. https://doi.org/10.2147/JIR.S440529 Hou Q, Jiang J, Na K, et al. Potential therapeutic targets for COVID-19 complicated with pulmonary hypertension: A bioinformatics and early validation study. Sci Rep. 2024;14:9294. https://doi.org/10.1038/s41598-024-59891-7 Sui Z, Wu X, Du L, et al. Characterization of the immune cell infiltration landscape in esophageal squamous cell carcinoma. Front Oncol. 2022;12:879326. https://doi.org/10.3389/fonc.2022.879326 Harrison AM, Goozee R. Psych-related iPhone apps. J Ment Health. 2014;23:48-50. https://doi.org/10.3109/09638237.2013.869575 Qiang M, Dai Z. Biomarkers of UVB radiation-related senescent fibroblasts. Sci Rep. 2024;14:933. https://doi.org/10.1038/s41598-024-51543-y Supplementary Tables Supplementary Tables 1-8 are not available with this version. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 02 May, 2026 Reviewers invited by journal 29 Mar, 2026 Editor assigned by journal 23 Mar, 2026 Submission checks completed at journal 21 Mar, 2026 First submitted to journal 19 Mar, 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-9165812","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":615499382,"identity":"e2ea2f72-4551-431f-b899-76a5c6922540","order_by":0,"name":"Difei Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA90lEQVRIiWNgGAWjYLACHgM5BjZm5oMPP1RIyMkTqcWYgY2dLdlY4oyFsWEDUVoYjBkY+HnMJHjbKhIZDhBQbXD87OEXbwoM5PiYgVok50kkMDYwP3x0A5+WM3lplnMMDIzZmNmKLQq3SeSxM7AZG+fg0WJ2IMfMmMfgT2IbM/PGG5LbJIoZG3jYpPFqOf8GpMWgvo2ZwUCCd45EYsMBQlpu5Bg/BmpJYGNmMZLgbSBCi/2NN2aMQL8YtjGDAvmYhLFhMwG/SPbnGH9488dAXr7/MDAqa+rk5NmbHz7GpwUI2CRQ+cz4lYOVfCCsZhSMglEwCkY0AACSIURJsuE7aQAAAABJRU5ErkJggg==","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":true,"prefix":"","firstName":"Difei","middleName":"","lastName":"Wang","suffix":""},{"id":615499383,"identity":"91a9619d-15db-4dc9-864e-3d8d660341af","order_by":1,"name":"Hui Gan","email":"","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Hui","middleName":"","lastName":"Gan","suffix":""},{"id":615499384,"identity":"776f3e3e-b704-48bb-bc78-18703ee83c9a","order_by":2,"name":"Yuxin Wu","email":"","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yuxin","middleName":"","lastName":"Wu","suffix":""},{"id":615499385,"identity":"dab579e8-86f8-4369-bbd8-83c92842935b","order_by":3,"name":"Baohui Yang","email":"","orcid":"","institution":"Lanzhou University","correspondingAuthor":false,"prefix":"","firstName":"Baohui","middleName":"","lastName":"Yang","suffix":""},{"id":615499386,"identity":"8613a1e3-11b1-4138-93fb-2599eb4bbb67","order_by":4,"name":"Han Xiao","email":"","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Han","middleName":"","lastName":"Xiao","suffix":""},{"id":615499387,"identity":"74569240-6c49-4f82-a80c-cdebba9c2a45","order_by":5,"name":"Haotian Tang","email":"","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Haotian","middleName":"","lastName":"Tang","suffix":""},{"id":615499388,"identity":"c8e41870-e13a-42a5-9282-f230c2f1af1b","order_by":6,"name":"Yudong Zhou","email":"","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yudong","middleName":"","lastName":"Zhou","suffix":""},{"id":615499389,"identity":"9b594f6c-e31b-47e6-bfc9-5365e3f55079","order_by":7,"name":"Bin Zou","email":"","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Bin","middleName":"","lastName":"Zou","suffix":""},{"id":615499390,"identity":"6ec221f3-2938-4853-8289-736484c7e19f","order_by":8,"name":"Kaiyi Kang","email":"","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Kaiyi","middleName":"","lastName":"Kang","suffix":""},{"id":615499391,"identity":"4a3fb0a4-55ef-4234-a7bb-a00eef163213","order_by":9,"name":"Xuan Zhai","email":"","orcid":"","institution":"Children's Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xuan","middleName":"","lastName":"Zhai","suffix":""},{"id":615499392,"identity":"e1b7777b-3e5c-4962-b2a0-31188c9e960c","order_by":10,"name":"Wu Yan","email":"","orcid":"","institution":"The First People's Hospital of Urumqi (Urumqi Children's Hospital)","correspondingAuthor":false,"prefix":"","firstName":"Wu","middleName":"","lastName":"Yan","suffix":""}],"badges":[],"createdAt":"2026-03-19 06:40:38","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9165812/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9165812/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106042635,"identity":"b1aba76b-36e2-4673-bda3-f7c43bba0084","added_by":"auto","created_at":"2026-04-02 18:18:06","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2163817,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification and enrichment analysis of differentially expressed genes. (\u003cstrong\u003ea\u003c/strong\u003e) Volcano plot of DEGs. Note: The x-axis represents Log2FC, and the y-axis represents -Log10(P-value). Each dot stands for a gene. As divided by the reference lines, genes in the upper right corner are upregulated differentially expressed genes (indicated in red), and genes in the upper left corner are downregulated differentially expressed genes (indicated in blue). The remaining genes are those with no significant statistical significance (indicated in black). The labeled genes in the figure are the top 10 upregulated genes and top 10 downregulated genes with the largest |log2(FoldChange)| (i.e., the largest fold changes). (\u003cstrong\u003eb\u003c/strong\u003e) Heatmap of DEGs. (\u003cstrong\u003ec\u003c/strong\u003e) Screening of differential SASP-related genes (dSPRGs). Note: Red represents highly expressed genes, and blue represents lowly expressed genes. (\u003cstrong\u003ed\u003c/strong\u003e) GO analysis of candidate genes. (\u003cstrong\u003ee\u003c/strong\u003e) KEGG enrichment analysis of candidate genes. (\u003cstrong\u003ef\u003c/strong\u003e) Constructing a protein-protein interaction network for proteins encoded by differentially expressed SPRGs.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-9165812/v1/f6caafb2c4f3a119728f7ae5.png"},{"id":106094662,"identity":"b1c84850-333a-4f07-922e-ba0300bcafbf","added_by":"auto","created_at":"2026-04-03 11:43:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":9973210,"visible":true,"origin":"","legend":"\u003cp\u003eScreening of biomarkers. (\u003cstrong\u003ea\u003c/strong\u003e) The result of LASSO regression analysis. (\u003cstrong\u003eb\u003c/strong\u003e) The result of screening with the SVM-RFE algorithm. (\u003cstrong\u003ec\u003c/strong\u003e) Overlapping genes between the two algorithms. (\u003cstrong\u003ed\u003c/strong\u003e) Validation of the differential expression of the five selected feature genes (SERPINE1, CCL2, IGFBP4, C3, and SPX) between epilepsy and control cortical/hippocampal tissues across the GSE256068 (Training) and GSE134697 (Validation) datasets.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-9165812/v1/d3cd9916f5bf074a1d199b70.png"},{"id":106042637,"identity":"b386a236-3443-467c-9b74-86d9a31b70fc","added_by":"auto","created_at":"2026-04-02 18:18:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":5693488,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram construction and evaluation. (\u003cstrong\u003ea\u003c/strong\u003e) The Nomogram prediction model. (\u003cstrong\u003eb\u003c/strong\u003e) The Nomogram calibration curve. (\u003cstrong\u003ec\u003c/strong\u003e) The Decision Curve Analysis.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-9165812/v1/2b86f75bd19c4f40ba9b36cd.png"},{"id":106042638,"identity":"a986bc8b-d511-4fa7-bb49-250e3e374779","added_by":"auto","created_at":"2026-04-02 18:18:06","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2569227,"visible":true,"origin":"","legend":"\u003cp\u003eEnrichment analysis of biomarkers. (\u003cstrong\u003ea1-5\u003c/strong\u003e) GSEA-based enrichment analysis of the key genes (IGFBP4, C3, CCL2, SERPINE1 and SPX). (\u003cstrong\u003eb\u003c/strong\u003e) The distribution of 22 immune cell types. (\u003cstrong\u003ec\u003c/strong\u003e) Differential infiltration of 22 immune cell types. (\u003cstrong\u003ed\u003c/strong\u003e) Spearman correlation analysis between the expression of five key biomarkers and the abundance of differentially infiltrated immune cells. The color gradient (red to blue) represents the correlation coefficient (positive to negative), and circle size indicates the strength of the correlation. Statistical significance: *p \u0026lt; 0.05, **p \u0026lt; 0.01, ***p \u0026lt; 0.001, ****p \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-9165812/v1/c7d302af073600ef647d8724.png"},{"id":106094663,"identity":"efa4a86a-d701-4bac-9e98-9eba6ae20705","added_by":"auto","created_at":"2026-04-03 11:43:04","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":5881095,"visible":true,"origin":"","legend":"\u003cp\u003eMolecular regulatory mechanisms analysis of SERPINE1, CCL2, IGFBP4, C3, and SPX. (\u003cstrong\u003ea\u003c/strong\u003e) The miRNA-mRNA regulatory network. (\u003cstrong\u003eb\u003c/strong\u003e) The TF–mRNA interaction pairs. Note: Purple represents biomarkers, and yellow represents miRNA. (\u003cstrong\u003ec\u003c/strong\u003e) The Interaction network of biomarkers and their predicted regulators.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-9165812/v1/83f48c92be3e159b170f94b2.png"},{"id":106042640,"identity":"751dcdcf-f3da-4b8f-8b72-c1fef09c5637","added_by":"auto","created_at":"2026-04-02 18:18:06","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":2818418,"visible":true,"origin":"","legend":"\u003cp\u003eDrug prediction and molecular docking analysis. (\u003cstrong\u003ea\u003c/strong\u003e) Drug-biomarker interaction prediction. Note: In the figure, the blue coiled lines represent the 3D structure of the protein, and the gray lines represent the 3D structure of the compound. In the enlarged view, the blue dashed lines indicate hydrogen bonds, and the gray dashed lines indicate van der Waals forces.(\u003cstrong\u003eb1-5\u003c/strong\u003e) Residue-interaction maps for the drug complexes with IGFBP4, SPX, CCL2, C3, and SERPINE1.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-9165812/v1/14a562b308fd471af2a62bb2.png"},{"id":106042641,"identity":"46baee90-6c47-4eb8-9b82-3a601de1d48f","added_by":"auto","created_at":"2026-04-02 18:18:06","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":424056,"visible":true,"origin":"","legend":"\u003cp\u003eValidation of SASP-related biomarker expression in the peripheral blood of a kainic acid (KA)-induced epilepsy mouse model. RT-qPCR analysis showing the relative mRNA expression levels of C3, CCL2, IGFBP4, SERPINE1, and SPX. Data are presented as Mean±SD (n = 5 per group). Statistical significance was determined using the Wilcoxon test. *p \u0026lt; 0.05, **p \u0026lt; 0.01, ns = not significant compared to the saline-treated control group.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-9165812/v1/8400b44fb13b4e6d8e5d3aa1.png"},{"id":106402099,"identity":"8faacd62-93e8-44fe-80fd-47a7235f796d","added_by":"auto","created_at":"2026-04-08 09:11:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":26076182,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9165812/v1/e2ae7e95-3ce8-46ae-8428-e7039c07b76e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Identification and validation of senescence-associated secretory phenotype-related biomarkers in epilepsy","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eEpilepsy is a chronic neurological disorder arising from aberrant, hypersynchronous neuronal discharges, clinically defined by recurrent, transient cerebral dysfunction. Approximately 65\u0026nbsp;million people are affected worldwide. First-line management remains pharmacological, yet roughly one-third of patients develop drug-resistant epilepsy[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. For these individuals, resective surgery is considered; however, it is burdened by post-operative morbidity, high costs, and unpredictable seizure freedom[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]; additionally, candidacy is restricted, further complicating seizure control. Current pathophysiological models depict epileptogenesis as a multi-layered network process. Ion-channel mutations\u0026mdash;e.g., those in SCN1A or GABRG2\u0026mdash;disrupt transmembrane ion homeostasis[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], while bidirectional gut\u0026ndash;brain signalling modulates excitability through neuroimmune, neurotransmitter, and microbial metabolite pathways[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Over twenty antiseizure medications exist, yet all are purely symptomatic; none reverses disease progression[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Approximately 40% of patients require polytherapy, leading to an exponential increase in drug-drug interactions. Additionally, therapeutic options are severely limited in special populations such as pregnant individuals and the elderly. Moreover, conventional diagnostics suffer from delayed recognition of seizures and non-negligible misclassification of seizure types, underscoring an urgent need for novel biomarkers.\u003c/p\u003e \u003cp\u003eThe senescence-associated secretory phenotype (SASP) comprises cytokines, chemokines, and growth factors released by senescent cells. Via autocrine and paracrine circuits, SASP propagates sterile inflammation and transmits senescent cues to neighbouring cells; within the tissue microenvironment, reactive oxygen species (ROS)\u0026mdash;often induced by SASP\u0026mdash;amplify telomere attrition and accelerate cellular senescence[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Within the central nervous system, chronic SASP signaling fundamentally reshapes the neuroglial microenvironment. The continuous paracrine release of these inflammatory mediators impairs astrocyte glutamate buffering, triggers microglial hyperactivation, and chronically degrades blood-brain barrier (BBB) integrity, thereby establishing a pro-epileptogenic neurochemical loop[\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Evidence linking SASP to epilepsy is now emerging. Single-cell sequencing by Ge et al. revealed up-regulation of P21, CCL2, and genes in the NF-κB pathway in pyramidal neurons within the epileptogenic cortex of drug-resistant patients; parallel murine models confirmed a robust molecular signature of neuronal senescence, establishing the first direct histopathological nexus between SASP and epilepsy[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMechanistically, discrete SASP mediators drive epileptogenesis through convergent pathways. IL-6, a canonical SASP cytokine, exacerbates absence seizures in WAG/Rij rats via neuroinflammatory circuits; blockade of the IL-6 receptor with tocilizumab (TCZ) mitigates seizure aggravation induced by LPS or exogenous IL-6[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. TNF-α and its receptor TNFR1 are markedly elevated across epileptic foci, engaging TRADD\u0026ndash;FADD\u0026ndash;caspase-8 signalling to precipitate neuronal apoptosis and hippocampal injury ; pharmacological interruption of this axis attenuates both ictal activity and neuronal loss[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Matrix metalloproteinase-9 (MMP-9), a SASP-driven extracellular matrix remodelling factor, degrades the tight-junction protein nectin-3 and compromises blood\u0026ndash;brain barrier integrity; MMP-9 inhibition with marimastat significantly shortens seizure duration in mice, underscoring SASP-mediated barrier disruption[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Furthermore, SASP-driven extracellular matrix (ECM) remodeling is critically regulated by the delicate balance of proteolytic networks, including matrix metalloproteinases and the plasminogen activator system. Disruption in this balance\u0026mdash;such as aberrant expression of endogenous protease inhibitors like SERPINE1 (PAI-1)\u0026mdash;can exacerbate BBB leakage and facilitate neuroimmune cellular infiltration, yet the precise transcriptomic signatures of these ECM regulators in human epileptogenic foci remain obscure. Additionally, in a pre-clinical Mtor FCDII model, senolytic intervention reduced seizure frequency[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Collectively, these data position SASP as a tractable therapeutic node, and the development of precision agents directed at senescence effectors is emerging as a focal strategy to overcome drug-resistant epilepsy.\u003c/p\u003e \u003cp\u003eDespite these crucial insights, the molecular circuitry linking SASP to epileptogenesis remains incompletely mapped. Deeper interrogation of SASP-mediated pathways is urgently required to identify robust non-invasive biomarkers, refine pre-surgical evaluation, and inform targeted interventions. Here, we leverage publicly available human brain transcriptomic datasets to identify a specialized SASP-associated biomarker panel in epilepsy. By applying integrative bioinformatic approaches and validating these signatures in a preclinical model, we aim to delineate the specific inflammatory and metabolic underpinnings of SASP in the epileptogenic network, providing novel mechanistic targets to ultimately improve diagnostic and therapeutic strategies for drug-resistant epilepsy.\u003c/p\u003e"},{"header":"2. Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Enrichment analysis of 22 candidate genes\u003c/h2\u003e \u003cp\u003eThe differential expression analysis of genes was conducted between EP and control samples in GSE256068. In total, 1,912 DEGs were detected. 1,280 up-regulated and 632 down-regulated genes were identified in EP \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea-b\u003cb\u003e)\u003c/b\u003e. Subsequently, the intersection of the 1,912 DEGs and 153 SASP-RGs was taken, and 22 candidate genes were obtained \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec\u003cb\u003e)\u003c/b\u003e. GO enrichment analysis revealed 1105 significantly enriched terms, including 1,050 BP terms, 7 CC terms, and 48 molecular MF terms. In the KEGG pathway analysis, 49 significantly enriched signaling pathways were obtained \u003cb\u003e(Supplementary Table S1-2)\u003c/b\u003e. Specifically, in the GO-BP term, these genes showed significant enrichment in neutrophil migration and positive regulation of the MAPK cascade. In the GO-CC term, these genes were mainly concentrated in cellular structures such as collagen-containing extracellular matrix and microvilli. In terms of GO-MF, they were mainly enriched in molecular functions such as receptor ligand activity \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed\u003cb\u003e)\u003c/b\u003e. In the KEGG enrichment analysis, they were mainly and significantly enriched in signal transduction pathways such as the receptor ligand activity and cytokine\u0026minus;cytokine receptor interaction \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee\u003cb\u003e)\u003c/b\u003e. PPI network analysis showed that most proteins encoded by the genes exhibited extensive interactions \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ef\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Acquisition of 5 biomarkers\u003c/h2\u003e \u003cp\u003eFeature genes were selected from candidate genes using LASSO and SVM-RFE algorithms. LASSO regression analysis showed that when the optimal lambda was 0.006981826, seven genes with non-zero feature coefficients were obtained: IL1B, C3, IGFBP4, SERPINE1, CCL2, CCL3, and SPX \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea\u003cb\u003e)\u003c/b\u003e. The SVM-RFE algorithm subsequently identified 16 genes (C3, IL1A, SEMA3F, CCL2, SPX, IGFBP4, ANGPT1, SERPINE1, FGF7, CCL8, MMP3, MMP1, CCL20, CCL5, CXCL8, and FOXA1) from the candidate genes \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb\u003cb\u003e)\u003c/b\u003e. The intersection of genes from the two algorithms yielded five feature genes: SERPINE1, CCL2, IGFBP4, C3, and SPX \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec\u003cb\u003e)\u003c/b\u003e. Subsequently, the feature gene expression levels between the EP and control groups were analyzed in the GSE256068 and GSE134697 datasets. In comparison with the control group, the EP group exhibited significantly down-regulated IGFBP4 expression and significantly up-regulated expressions of SERPINE1, CCL2, C3, and SPX (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed\u003cb\u003e)\u003c/b\u003e. It was worth noting that this expression trend was highly consistent in the two datasets. Based on the above findings, these five genes were identified as biomarkers for EP. Notably, the consistent up-regulation of SERPINE1, CCL2, C3, and SPX across both cortical and hippocampal epileptogenic tissues suggests a pervasive, multi-focal SASP activation during disease progression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Nomogram had good predictive performance\u003c/h2\u003e \u003cp\u003eA nomogram prediction model was developed for evaluating the diagnostic efficiency of EP-related biomarkers. The model assigned specific scores to each biomarker, with the cumulative total score positively correlated with the risk of EP \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea\u003cb\u003e)\u003c/b\u003e. Calibration curve analysis showed that the Hosmer-Lemeshow test yielded a P-value of 0.198, suggesting good concordance between predicted and observed outcomes \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb\u003cb\u003e)\u003c/b\u003e. DCA demonstrated that the nomogram's net benefit significantly exceeded that of the treat-all and treat-none strategies, confirming its clinical utility \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Inferred immune microenvironment remodeling associated with SASP biomarkers\u003c/h2\u003e \u003cp\u003eGSEA analysis results showed that IGFBP4, C3, CCL2, SERPINE1, and SPX were significantly enriched in 89, 102, 80, 115, and 79 functional pathways, respectively \u003cb\u003e(Supplementary Table S3-7)\u003c/b\u003e. When sorted by significance, IGFBP4 was primarily enriched in pathways such as Alzheimer's Disease and oxidative phosphorylation; C3 mainly showed enrichment in pathways like ribosome and leishmania infection; CCL2 demonstrated notable enrichment in pathways including ribosome and viral myocarditis; SERPINE1 was predominantly enriched in pathways such as cytokine-cytokine receptor interaction and leishmania infection; SPX exhibited significant enrichment in pathways including oxidative phosphorylation and ribosome \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea\u003cb\u003e)\u003c/b\u003e. Subsequently, the CIBERSORT algorithm was used to evaluate the infiltration of 22 types of immune cells during EP progression \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb\u003cb\u003e)\u003c/b\u003e. Results revealed significant differences in the infiltration abundance of six immune cells between the EP and control groups. Specifically, the control group exhibited higher infiltration abundance of cells such as CD8 T cells[\u003cspan additionalcitationids=\"CR60 CR61 CR62\" citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e] (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), while the EP group had higher infiltration abundance of cells such as mast cells[\u003cspan additionalcitationids=\"CR65\" citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e] (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec\u003cb\u003e)\u003c/b\u003e. Additionally, significant correlations were observed: SERPINE1 positively correlated with activated mast cells (0.304, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and negatively with follicular helper T cells (-0.326, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001); CCL2 positively with activated mast cells (0.636, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) and negatively with resting mast cells (-0.415, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001); IGFBP4 positively with follicular helper T cells (0.505, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001); C3 positively with eosinophils[\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e, \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e] (0.318, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and negatively with follicular helper T cells (-0.363, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001); SPX positively with activated mast cells (0.508, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) and negatively with CD8\u0026thinsp;+\u0026thinsp;T cells (-0.560, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed\u003cb\u003e)\u003c/b\u003e. This indicated that these genes might participate in shaping the immune microenvironment of epilepsy by regulating the activities of different immune cells.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Molecular regulatory mechanisms of SERPINE1, CCL2, IGFBP4, C3, and SPX\u003c/h2\u003e \u003cp\u003eUsing the miRNet database, 217 miRNAs related to IGFBP4, 233 miRNAs related to C3, 63 miRNAs related to CCL2, 195 miRNAs related to SERPINE1, and 10 miRNAs related to SPX were successfully predicted. A miRNA-mRNA regulatory network was further constructed based on the aforementioned predictions \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea\u003cb\u003e)\u003c/b\u003e. Since TFs could specifically recognize downstream target genes, subsequent analysis via the miRNet database revealed that IGFBP4 interacted with 3 transcription factors, C3 interacted with 1 transcription factor, CCL2 interacted with 19 transcription factors, and SERPINE1 interacted with 20 transcription factors \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb\u003cb\u003e)\u003c/b\u003e. Notably, TFs such as NFKB1, RELA, and SP1 were found to simultaneously target both SERPINE1 and CCL2, suggesting these transcription factors might act as hubs in the same signaling pathway or biological process by coordinately regulating the expression of SERPINE1 and CCL2. We also utilized physical interaction and co-expression analysis in the GeneMANIA database and identified 20 genes functionally related to SERPINE1, CCL2, IGFBP4, C3, and SPX, such as CCL15 and PAPPA. Among them, CCL2, CCL5, and CXCL8 collectively participated in pathways such as cytokine activity and response to chemokine \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ec\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 In silico prediction of potential pharmacological modulators\u003c/h2\u003e \u003cp\u003eTo explore potential drugs for EP treatment, the R package enrichR was utilized to predict drugs targeting the biomarkers. The analysis revealed a total of 30 drugs targeting CCL2, 30 drugs targeting SERPINE1, 30 drugs targeting IGFBP4, 16 drugs targeting C3, and 3 drugs targeting SPX \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea\u003cb\u003e)\u003c/b\u003e. Subsequently, molecular docking was performed on the drug with the smallest p-value in the above prediction and its corresponding biomarker to evaluate the binding capacity between the small-molecule drug and the target protein. The docking combinations included IGFBP4 with the drug amantadine, SPX with trans-Lutein, CCL2 with fludrocortisone, C3 with propofol, and SERPINE1 with the compound 5630-53-5. Binding energies below \u0026minus;\u0026thinsp;5 kcal/mol were observed for all biomarker-drug pairs in the docking analysis, confirming their strong binding affinity \u003cb\u003e(\u003c/b\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e\u003cb\u003e) (\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb\u003cb\u003e)\u003c/b\u003e. This finding suggested these molecules might serve as promising candidates for EP therapy.\u003c/p\u003e \u003cp\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\u003ePredicted free binding energies between active pharmaceutical ingredients and key genes\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\u003eTargets\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDrug\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBinding Energy(KJ/mol)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIGFBP4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eamantadine\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-5.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSPX\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003etrans-Lutein\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-7.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCCL2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003efludrocortisone\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-6.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eC3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003epropofol\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-5.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSERPINE1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5630-53-5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-7.1\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=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Validation of biomarkers\u003c/h2\u003e \u003cp\u003eRT-qPCR analysis revealed significant down-regulation of IGFBP4 and up-regulation of SERPINE1, CCL2, and C3 in EP compared to control. Notably, SPX expression did not differ significantly between groups, and these findings were consistent with bioinformatics results, confirming method validity \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Discussion","content":"\u003cp\u003eCellular senescence and the senescence-associated secretory phenotype (SASP) are increasingly recognized as amplifiers of chronic, sterile inflammation in the central nervous system. In epilepsy, sustained neuroinflammation and metabolic stress contribute to network hyperexcitability, blood\u0026ndash;brain barrier (BBB) dysfunction, and progressive tissue remodeling, all of which may be reinforced by senescent-cell secretomes. Although emerging evidence supports a relationship between senescence programs and epileptogenesis [\u003cspan additionalcitationids=\"CR12 CR13 CR14\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], SASP-oriented molecular signatures and their neurochemical implications remain incompletely defined.\u003c/p\u003e \u003cp\u003eIn the present study, we integrated human brain transcriptomic data with a curated SASP gene set and applied two machine-learning feature-selection strategies (LASSO and SVM-RFE) to derive a concise SASP-related biomarker panel for epilepsy. Five genes\u0026mdash;IGFBP4, SERPINE1, CCL2, C3, and SPX\u0026mdash;showed robust and reproducible differential expression across two independent GEO datasets, and a nomogram built on these markers demonstrated favorable discrimination and calibration. Beyond diagnostic modeling, functional enrichment and immune deconvolution analyses linked these biomarkers to translational control, oxidative phosphorylation, and a reshaped immune microenvironment. Finally, experimental RT\u0026ndash;qPCR in a kainic acid (KA) mouse model provided initial biological support for four markers (IGFBP4, SERPINE1, CCL2, and C3), while SPX did not show a statistically significant change in peripheral blood, highlighting the need for further validation and tissue-context interpretation.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.1 A SASP-centered biomarker axis in epilepsy\u003c/h2\u003e \u003cp\u003eThe identified panel is neurochemically coherent in that it spans multiple, convergent processes that are repeatedly implicated in epilepsy and senescence-associated inflammation: (i) growth-factor signaling and metabolic regulation (IGFBP4), (ii) extracellular proteolysis and BBB homeostasis (SERPINE1), (iii) chemokine-driven neuroimmune recruitment and glial activation (CCL2), (iv) complement activation and neuroimmune synapse remodeling (C3), and (v) neuropeptide signaling potentially coupled to inflammatory and metabolic tone (SPX). Importantly, several of these molecules are either secreted or closely associated with secretory pathways, consistent with the conceptual framework of the SASP. Together, these markers suggest that epilepsy-related senescence programs may not be limited to a single pathway but may instead reflect a coordinated, multi-layer neurochemical remodeling involving inflammatory mediators, metabolic stress responses, and extracellular matrix/vascular interfaces.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.2 IGFBP4: IGF signaling balance and metabolic vulnerability\u003c/h2\u003e \u003cp\u003eInsulin-like growth factor-binding protein 4 (IGFBP4) is a secreted glycoprotein that modulates IGF bioavailability, thereby shaping cell survival, differentiation, and stress responses[\u003cspan additionalcitationids=\"CR21 CR22 CR23\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan additionalcitationids=\"CR28 CR29\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. In our datasets, IGFBP4 was downregulated in epilepsy, a pattern that may indicate altered IGF-axis homeostasis in epileptogenic tissue. IGF-1 signaling has context-dependent effects in epilepsy: it can exert neuroprotective and anti-inflammatory actions in some models, yet IGF-1 receptor activation may also increase excitability in specific circuit states [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Reduced IGFBP4 could theoretically increase free IGF-1 availability; however, the net consequence likely depends on region-specific receptor distribution, cell-type composition (neurons vs. astrocytes/microglia), and the inflammatory milieu. Notably, IGFBP4-associated enrichment in oxidative phosphorylation pathways aligns with the concept that epileptogenic networks operate under heightened energetic demand and oxidative stress[\u003cspan additionalcitationids=\"CR49 CR50 CR51 CR52 CR53 CR54\" citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]. Mitochondrial dysfunction and ROS accumulation can promote both neuronal injury and senescence-like phenotypes, potentially feeding into SASP-associated inflammation. Therefore, IGFBP4 may represent a bridge between growth-factor signaling and metabolic resilience in epilepsy, and its directionality and compartment specificity (brain vs. circulation) merit targeted experimental clarification.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.3 SERPINE1 (PAI-1): extracellular proteolysis, BBB integrity, and inflammatory persistence\u003c/h2\u003e \u003cp\u003eSERPINE1 (plasminogen activator inhibitor-1, PAI-1) is the principal inhibitor of tissue-type and urokinase-type plasminogen activators, thereby regulating fibrinolysis and extracellular proteolysis [\u003cspan additionalcitationids=\"CR32\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Elevated SERPINE1 is widely associated with pro-thrombotic states and tissue fibrosis, whereas deficiency predisposes to bleeding due to hyperfibrinolysis, underscoring its importance in vascular biology[\u003cspan additionalcitationids=\"CR32 CR33 CR34 CR35\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. SERPINE1 is also a well-recognized senescence-associated factor in multiple contexts, linking it conceptually to SASP-driven tissue remodeling.\u003c/p\u003e \u003cp\u003eIn epilepsy, BBB dysfunction is a key contributor to neuroinflammation and seizure propagation. Extracellular protease cascades (plasmin/MMPs) can influence tight-junction stability and basement membrane integrity[\u003cspan additionalcitationids=\"CR38 CR39 CR40\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Altered SERPINE1 may therefore participate in a feedback loop in which vascular and extracellular matrix remodeling facilitates immune-cell entry and glial activation, which in turn sustains inflammatory cytokine production and excitability. While our study does not establish causality, the consistent upregulation of SERPINE1 across datasets and its association with immune-infiltration features suggests that SERPINE1 may be an informative marker of a proteolysis\u0026ndash;BBB\u0026ndash;inflammation axis in epileptogenic tissue.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.4 CCL2: chemokine amplification of neuroimmune circuits\u003c/h2\u003e \u003cp\u003eCCL2 is a central chemokine produced by astrocytes, microglia, endothelial cells, and neurons under inflammatory challenge. Through CCR2 signaling, CCL2 recruits peripheral monocytes/macrophages and reshapes local neuroimmune signaling, thereby amplifying cytokine cascades and potentially modulating inhibitory/excitatory balance [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Prior evidence supports CCL2 induction after seizures and its ability to promote pro-inflammatory cytokine release and neuronal injury [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. In our analysis, CCL2 was among the most prominent upregulated SASP-related markers, consistent with the idea that a senescence-linked secretory environment may favor sustained chemokine signaling.\u003c/p\u003e \u003cp\u003eFrom a neurochemical perspective, chronic CCL2 signaling can indirectly increase neuronal excitability by intensifying IL-1β/TNF-α\u0026ndash;dependent modulation of synaptic receptor trafficking and by promoting microglial activation states that alter synaptic pruning and glutamate homeostasis. Thus, CCL2 likely represents both a biomarker and a plausible contributor to epileptogenic neuroinflammation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.5 C3: complement-driven neuroimmune remodeling and excitability\u003c/h2\u003e \u003cp\u003eComplement component C3 is a pivotal node of the complement cascade and has been increasingly linked to neuroinflammation, synapse remodeling, and microglial effector functions in multiple neurological disorders [\u003cspan additionalcitationids=\"CR45\" citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. In epilepsy models, astrocytic upregulation of C3 and signaling via the C3a\u0026ndash;C3aR axis can enhance microglial activation and inflammatory cytokine release, which may influence inhibitory tone and excitatory drive [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. The consistent upregulation of C3 in our epilepsy datasets, together with its correlations with inferred immune-cell changes, supports complement activation as a relevant neurochemical feature of epileptogenic tissue.\u003c/p\u003e \u003cp\u003eImportantly, complement activity can also intersect with senescence biology: senescent cells and SASP mediators may promote complement activation, while complement-driven inflammation can reinforce cellular stress responses. Therefore, C3 may be positioned at the intersection of SASP-linked inflammation and circuit remodeling, providing a mechanistically plausible bridge between molecular signatures and network-level epilepsy phenotypes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.6 SPX: neuropeptide signaling with tissue- and compartment-specific behavior\u003c/h2\u003e \u003cp\u003eSpexin (SPX) is a neuropeptide with broad tissue expression and reported roles in metabolic regulation, oxidative stress, and inflammatory tone via galanin receptor signaling [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. In our transcriptomic analyses, SPX was upregulated in epilepsy, suggesting potential involvement in epileptogenic remodeling or compensatory neuropeptide responses. However, SPX did not show a statistically significant change in peripheral blood RT\u0026ndash;qPCR in the KA mouse model. This discrepancy can be interpreted in several non-mutually exclusive ways: SPX regulation may be brain-region specific, time dependent (acute vs. chronic stages), or primarily localized to neural tissue without a strong peripheral signature. Additionally, neuropeptide transcripts can exhibit higher variability in blood and may require larger cohorts or protein-level assays for reliable detection. Accordingly, SPX should be considered a transcriptome-supported candidate whose translational utility (especially as a blood biomarker) requires additional validation in matched brain and plasma/serum samples, ideally including protein quantification.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.7 Pathway-level interpretation: translation control and oxidative phosphorylation\u003c/h2\u003e \u003cp\u003eA recurring signal in our enrichment analyses was the involvement of ribosome/translation-related pathways and oxidative phosphorylation. These findings are neurochemically consistent with epilepsy as a disorder of sustained energetic strain and activity-dependent proteostasis. Hyperexcitability increases ATP demand, elevates ROS production, and can shift mitochondrial function, while chronic inflammation can further impair oxidative phosphorylation and redox homeostasis. In parallel, translational control is strongly regulated by mTOR signaling, which is frequently implicated in epilepsy and synaptic remodeling [\u003cspan additionalcitationids=\"CR57\" citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. Increased ribosome biogenesis and dysregulated translation can modify receptor composition and synaptic protein abundance, influencing excitability thresholds and network synchronization.\u003c/p\u003e \u003cp\u003eWithin a SASP framework, these pathway signals are also plausible: oxidative stress and mitochondrial dysfunction can promote senescence-like states and reinforce SASP output, while sustained inflammatory signaling can alter translation programs in glia and neurons. Therefore, the pathway results complement the biomarker panel by suggesting that epilepsy-associated senescence signatures may be embedded in a broader metabolic\u0026ndash;proteostatic remodeling landscape.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.8 Upstream regulatory networks: NF-κB as a master SASP driver\u003c/h2\u003e \u003cp\u003eBeyond downstream effector functions, our in silico regulatory network analysis (Section \u003cspan refid=\"Sec7\" class=\"InternalRef\"\u003e2.5\u003c/span\u003e) identified the transcription factors NFKB1 and RELA as crucial hubs simultaneously targeting SERPINE1 and CCL2. This convergence is highly mechanistically relevant, as the NF-κB complex (comprising NFKB1 and RELA subunits) is established as the primary transcriptional master regulator of the SASP [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. In the context of epilepsy, chronic metabolic stress and recurrent seizures can induce DNA damage and persistent NF-κB activation within the neuroglial microenvironment [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Our network predictions suggest that this upstream NF-κB activation directly orchestrates the concurrent up-regulation of chemokines (CCL2) and extracellular matrix remodelers (SERPINE1). Consequently, identifying NFKB1 and RELA as shared upstream regulators not only corroborates the senescence-associated nature of our biomarker panel, but also highlights a potential therapeutic vulnerability: targeting the NF-κB/SASP axis might concurrently silence multiple epileptogenic effectors, offering a more comprehensive intervention than neutralizing single downstream cytokines.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.9 Immune deconvolution: hypothesis-generating signals requiring orthogonal validation\u003c/h2\u003e \u003cp\u003eImmune-infiltration estimation suggested differences in the relative abundance of several immune-cell populations between epilepsy and control tissues and identified correlations between biomarkers (notably CCL2, SERPINE1, C3, and SPX) and immune-cell signatures. These results are consistent with the established role of neuroimmune interactions in epilepsy; however, they must be interpreted with distinct caution. Crucially, standard CIBERSORT-based deconvolution relies on reference signatures primarily derived from peripheral immune cells. The central nervous system possesses a highly specialized immune microenvironment dominated by resident microglia and astrocytes. Therefore, transcriptomic signals computationally mapped to peripheral subsets (such as mast cells or CD8\u0026thinsp;+\u0026thinsp;T cells) might partially reflect the overlapping transcriptional states of activated resident glia, rather than definitive peripheral infiltration. Consequently, these inferred immune shifts should be viewed strictly as hypothesis-generating. Future investigations utilizing orthogonal approaches, such as spatial transcriptomics, single-cell RNA sequencing (scRNA-seq), or multi-parameter flow cytometry on matched resected epileptogenic tissues, are imperative to validate the precise cellular sources of these SASP mediators.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.10 Translational implications and drug-repurposing signals\u003c/h2\u003e \u003cp\u003eThe nomogram based on five SASP-related markers showed favorable diagnostic performance in the analyzed datasets, supporting the potential of a multi-marker strategy over single-gene assessment. Our KA mouse RT\u0026ndash;qPCR results provide an initial step toward experimental validation and raise the possibility that brain-derived inflammatory remodeling may be reflected systemically, potentially through BBB perturbation and neuroimmune signaling. Furthermore, we performed in silico drug prediction and molecular docking to generate hypotheses for potential therapeutic targeting of the SASP network. While the calculated docking energies suggest structural binding feasibility between the identified compounds and our biomarker targets, these computational simulations do not establish pharmacodynamic efficacy, blood-brain barrier penetrability, in vivo target engagement, or actual anti-seizure activity. Thus, these structural findings are best regarded strictly as a computational prioritization tool to guide future in vitro and in vivo neuropharmacological testing, rather than a confirmation of therapeutic viability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e3.11 Limitations\u003c/h2\u003e \u003cp\u003eSeveral crucial limitations in this study should be acknowledged. First, our bioinformatics pipeline relies on retrospective public bulk transcriptomic datasets; differences in sample processing, platform effects, and clinical heterogeneity inevitably influence the results. Second, regarding experimental validation, although peripheral-blood RT\u0026ndash;qPCR supports four biomarkers (IGFBP4, SERPINE1, CCL2, and C3) in a controlled KA mouse model, this cross-species and cross-compartment design (human brain transcriptomics vs. mouse peripheral blood validation) represents an indirect assessment. Peripheral blood mRNA fluctuations may serve as proxy indicators of systemic inflammatory responses linked to BBB disruption, but they do not directly equate to the local neurochemical alterations within the epileptogenic focus. The lack of direct validation at the protein level (e.g., via Western blotting or immunohistochemistry) within matched brain tissues is a notable constraint. Finally, mechanistic causality was not tested. While our study highlights the robust diagnostic potential of these SASP-related targets, functional experiments\u0026mdash;such as targeted genetic or pharmacologic modulation of the SERPINE1-linked extracellular pathways or the CCL2/CCR2 axis in vivo\u0026mdash;will be necessary to establish whether these biomarkers act as active drivers or secondary consequences of epileptogenic remodeling.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e3.12 Conclusions\u003c/h2\u003e \u003cp\u003eIn summary, we identified and validated a SASP-associated biomarker panel (IGFBP4, SERPINE1, CCL2, C3, and SPX) that discriminates epilepsy from controls and is linked to inflammatory, metabolic, and translational remodeling. These findings expand the neurochemical framework connecting senescence-associated secretory programs to epileptogenesis and provide candidate markers and pathways for future mechanistic and translational investigation.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Methods","content":"\u003cp\u003e\u003cstrong\u003e4.1 Data collection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDatasets GSE256068 and GSE134697 were retrieved from GEO (https://www.ncbi.nlm.nih.gov/geo/). Among them, GSE256068 was from the GPL24676 platform, which included cortical tissue samples from 121 epilepsy patients and 12 controls. Samples with the type of hippocampal tissue were excluded. GSE134697 was from the GPL16791 platform, in which 17 epilepsy samples with the type of hippocampus were regarded as the disease group, and 17 cortical samples served as the control group. Samples without epilepsy patients were excluded. By searching the references,[69, 70] a total of 153 senescence-associated secretory phenotype-related genes (SASP-RGs) were obtained \u003cstrong\u003e(\u003c/strong\u003e\u003cstrong\u003eSupplementary Table S\u003c/strong\u003e\u003cstrong\u003e8)\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003eWhile GSE134697 utilizes cortical tissues as controls for hippocampal epileptic samples, we deliberately leveraged this anatomically distinct dataset as a stringent cross-regional validation cohort. The fact that our 5-gene SASP signature (derived from cortical EP vs. control in GSE256068) maintained consistent differential expression in the hippocampal EP cohort underscores that this SASP signature is not a region-specific artifact, but rather a conserved neurochemical hallmark of the epileptogenic network across different brain regions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.2 Analysis of differential gene expression\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUsing the \u0026quot;limma\u0026quot; package (v 3.54.0) [71], DEGs were screened between EP and controls in GSE256068 (p \u0026lt; 0.05, |log\u003csub\u003e2\u003c/sub\u003eFC| \u0026gt; 1.5). For visualizing DEGs, a volcano plot and heatmap were created via \u0026quot;ggplot2\u0026quot; (v 3.4.1)[71]and \u0026quot;pheatmap\u0026quot; packages (v 1.0.12).[72] The top 10 DEGs were annotated in the figures, sorted by ascending order of |log\u003csub\u003e2\u003c/sub\u003eFC|.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.3 Functional analyses of genes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe \u0026quot;ggvenn\u0026quot; package (v 0.1.9)[73]\u003csup\u003e\u0026nbsp;\u003c/sup\u003ewas applied to intersect DEGs and SASP-RGs, and the resultant genes were then termed candidate genes. GO and KEGG analyses were carried out by leveraging the \u0026quot;clusterProfiler\u0026quot; package (v 4.2.2) [73] (p \u0026lt; 0.05). By sorting in descending order of significance based on p-values, the top 5 GO and top 10 KEGG results were visualized. A PPI network was built through STRING (https://string-db.org/) (confidence = 0.15).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.4 Machine learning\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo further screen for feature genes, the \u0026quot;glmnet\u0026quot; package (v 4.1.18)[74]\u003csup\u003e\u0026nbsp;\u003c/sup\u003ewas utilized to conduct the LASSO regression analysis. The genes were screened from candidate genes by using 10-fold cross-validation and taking log(\u0026lambda;.min) as the screening threshold. The R package \u0026quot;e1071\u0026quot; (v 1.7.16)[75]\u003csup\u003e\u0026nbsp;\u003c/sup\u003ewas used to execute the SVM-RFE analysis, yielding the importance scores and corresponding ranks for each gene. Meanwhile, the error rate and accuracy of each iterative combination were calculated, and the combination with the highest accuracy was selected as the optimal one to screen out genes. Then, the \u0026quot;VennDiagram\u0026quot; package (v 1.7.1)[76]\u003csup\u003e\u0026nbsp;\u003c/sup\u003ewas used to obtain the intersection of genes derived from the two algorithms, yielding the final characteristic genes. Subsequently, feature gene expression in EP and control samples across GSE256068 and GSE134697 was analyzed via the Wilcoxon test. Genes that showed significantly differential expression and consistent expression patterns in both datasets were designated as biomarkers (p \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.5 Construction and validation of the nomogram\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo explore the role of biomarkers in the prediction of EP, the \u0026quot;RMS\u0026quot; package (v 6.8.1) [77] was employed to develop a nomogram model using the biomarkers as inputs. Subsequently, the \u0026quot;regplot\u0026quot; package (v 1.1) [77] was used to plot the calibration curve (Hosmer-Lemeshow, p \u0026gt; 0.05) and DCA (with the net benefit value greater than 0 as the criterion for good predictive performance).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.6 GSEA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on EP and control samples from GSE256068, as the reference gene set, \u0026quot;c2.cp.kegg.v7.5.1.symbols.gmt\u0026quot; was chosen from MSigDB (https://www.gsea-msigdb.org/gsea/msigdb). The \u0026quot;psych\u0026quot; package (v 2.1.6) [78] computed correlation coefficients between biomarkers and all other genes, with genes sorted in descending order of coefficients to derive related gene lists for each biomarker. Subsequently, the \u0026quot;clusterProfiler\u0026quot; package (v 4.2.2) was employed for GSEA (|NES| \u0026gt; 1, FDR \u0026lt; 0.25, and p \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.7 Immune cell infiltration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo comprehensively investigate the dynamic immune infiltration during EP progression, based on EP and control samples in the GSE256068, the CIBERSORT algorithm was adopted to calculate relative abundances of 22 immune cells. Differences in immune cell infiltration between EP and control samples were analyzed by the Wilcoxon test (p \u0026lt; 0.05). Finally, the \u0026quot;psych\u0026quot; package (v 2.1.6) was utilized to perform Spearman correlation analysis between biomarkers and differentially infiltrated immune cells (|cor| \u0026gt; 0.3, p \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.8 GENEMANIA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate the upstream and downstream regulatory interactions of biomarkers, using miRNet (https://www.mirnet.ca/), we predicted miRNAs and transcriptional regulators of the biomarkers. The miRNA-mRNA and TF-mRNA regulatory networks were then constructed using Cytoscape software. The GeneMANIA database (\u003cu\u003ehttps://genemania.org/\u003c/u\u003e) was applied to find genes that were functionally similar to the biomarkers.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.9 Drug prediction and molecular docking\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDrugs targeting biomarkers were predicted via the \u0026quot;enrichR\u0026quot; package (v 3.2)[79]. To assess binding affinity, the top-ranked drug (based on the minimal p-value) was chosen for docking. 3D structures of biomarker-encoded proteins were retrieved from NCBI and UniProt (https://www.uniprot.org/), converted to PDB files, and designated as docking receptors. From PubChem (\u003cu\u003ehttps://pubchem.ncbi.nlm.nih.gov/\u003c/u\u003e), the 2D structures of predicted drugs in SDF format were obtained and converted to 3D mol3 format using ChemBio2D. Molecular docking was performed via the CB-Dock2 (\u003cu\u003ehttps://cadd.labshare.cn/cb-dock2/php/index.php\u003c/u\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.10 RT-qPCR\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBiomarker expression validation was performed via RT-qPCR. Five KA-induced epileptic mice and five saline-treated control mice (C57BL/6J, male, 8 weeks old) were used. Whole blood samples were collected from Experimental Animal Center of Chongqing Medical University. The ethical approval was provided by The Ethics Committee off Chongqing Medical University with the approval number IACUC-CQMU-2025-09030. The results were plotted using Graphpad Prism.\u003c/p\u003e\n\u003cp\u003eFirst, RNA was isolated from 10 samples using TRIzol (Novozymes, Nanjing, China). Total RNA was converted to cDNA using HP All-in-one qRT Master Mix II RT203-Ver.1 kit (Yungene Bio, Kunming, China) following the provided protocols. RT-qPCR was performed with 2\u0026times;Universal Blue SYBR Green qPCR Master Mix (Servicebio, Wuhan, China). The primer sequences are shown in \u003cstrong\u003eTable 2\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declarations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe certify that the research study titled \u003cstrong\u003eIdentification and validation of senescence-associated secretory phenotype-related biomarkers in epilepsy\u003c/strong\u003e has been approved by the relevant ethics committee or institutional review board (IRB). The approval number is IACUC-CQMU-2025-09030. We confirm that all experiments were performed in accordance with ARRIVE guidelines and American Veterinary Medical Association (AVMA) guidelines.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePatient Informed Consent\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs the human transcriptomic data analyzed in this study were obtained from the publicly available Gene Expression Omnibus (GEO) database, the requirement for ethical approval and patient informed consent was waived.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e. The table of primer sequences.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"419\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003eGene number\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003egene\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" colspan=\"2\" valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003esequences\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003eProduct length\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eXM_036156346.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eIGFBP4 \u0026nbsp; F\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" colspan=\"2\"\u003e\n \u003cp\u003eCAGACACAAGAAGCGAACTCC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e268\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eXM_036156346.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eIGFBP4 \u0026nbsp; R\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" colspan=\"2\"\u003e\n \u003cp\u003eCGCCCTGCTTAGATTCTGGA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e268\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_008871.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eSERPINE1 \u0026nbsp; F\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eCTCCAAGGGGCAACGGATAG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_008871.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eSERPINE1 \u0026nbsp; R\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eAAGCAAGCTGTGTCAAGGGA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_011333.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eCCL2 \u0026nbsp;F\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eCCACAACCACCTCAAGCACT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_011333.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eCCL2 \u0026nbsp;R\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eTAAGGCATCACAGTCCGAGTC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_009778.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eC3 \u0026nbsp; F\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eTTCCTTCACTATGGGACCAGC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e127\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_009778.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eC3 \u0026nbsp; R\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eCTCCAGCCGTAGGACATTGG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e127\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_001369015.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eSPX \u0026nbsp; F\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eTAGGGCGAGAGATTGTGGGG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e159\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_001369015.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eSPX \u0026nbsp; R\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eTTGGGGAGTCCAGTTCCTCT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e159\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_001411844.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eM-GAPDH \u0026nbsp;F\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eTGTGTCCGTCGTGGATCTGA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e153\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eNM_001411844.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eM-GAPDH \u0026nbsp;R\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003eGAGTTGCTGTTGAAGTCGCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\"\u003e\n \u003cp\u003e153\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e4.11\u003c/strong\u003e \u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eR software (v 4.2.2) was employed. Intergroup differences were compared via the Wilcoxon test (p \u0026lt; 0.05).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to express our sincere gratitude to all individuals and organizations who supported and assisted us throughout this research.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDifei Wang, Hui Gan\u0026nbsp;and\u0026nbsp;Yuxin Wu, Baohui Yang, Han Xiao\u0026nbsp;wrote the main manuscript text and\u0026nbsp;Haotian Tang, Hanli Qiu, Xinyu Dong, Kaiyi Kangprepared figures 1\u0026ndash;3. All authors reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll the authors declare no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets analysed during the current study are available in the Gene Expression Omnibus (GEO) repository (GSE256068 and GSE134697).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to publish\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eKanner AM, Bicchi MM. Antiseizure medications for adults with epilepsy: A review. JAMA. 2022;327:1269-1281. https://doi.org/10.1001/jama.2022.3880\u003c/li\u003e\n\u003cli\u003eFelix R, Ekman J, Bjellvi S, Ljunggren M, et al. Laser interstitial thermal therapy versus open surgery for mesial temporal lobe epilepsy: A systematic review and meta-analysis. World Neurosurg. 2024;192:224-235.e15. https://doi.org/10.1016/j.wneu.2024.09.090\u003c/li\u003e\n\u003cli\u003ePerucca E, Perucca P, White HS, et al. Drug resistance in epilepsy. Lancet Neurol. 2023;22:723-734. https://doi.org/10.1016/S1474-4422(23)00151-5\u003c/li\u003e\n\u003cli\u003eKitschen A, Aleknonytė-Resch M, Sakalytė G, et al. Cost-effectiveness of surgical treatment compared to medical treatment in patients with drug-refractory epilepsy: A systematic review. Eur J Neurol. 2023;30:749-761. https://doi.org/10.1111/ene.15632\u003c/li\u003e\n\u003cli\u003eLi X, Guo S, Liu K, et al. Gabrg2 deletion linked to genetic epilepsy with febrile seizures plus affects the expression of GABA(A) receptor subunits and other genes at different temperatures. Neuroscience. 2020;438:116-136. https://doi.org/10.1016/j.neuroscience.2020.04.041\u003c/li\u003e\n\u003cli\u003eSocała K, Doboszewska U, Szopa A, et al. The role of microbiota-gut-brain axis in neuropsychiatric and neurological disorders. Pharmacol Res. 2021;172:105840. https://doi.org/10.1016/j.phrs.2021.105840\u003c/li\u003e\n\u003cli\u003ePerucca E, White HS, Bialer M. New GABA-targeting therapies for the treatment of seizures and epilepsy: II. Treatments in clinical development. CNS Drugs. 2023;37:781-795. https://doi.org/10.1007/s40263-023-01025-4\u003c/li\u003e\n\u003cli\u003eYan R, Zhang L, Chen Y, et al. Therapeutic potential of gut microbiota modulation in epilepsy: A focus on short-chain fatty acids. Neurobiol Dis. 2025;209:106880. https://doi.org/10.1016/j.nbd.2025.106880\u003c/li\u003e\n\u003cli\u003eThoppil H, Riabowol K. Senolytics: A translational bridge between cellular senescence and organismal aging. Front Cell Dev Biol. 2019;7:367. https://doi.org/10.3389/fcell.2019.00367\u003c/li\u003e\n\u003cli\u003eOhtani N. The roles and mechanisms of senescence-associated secretory phenotype (SASP): Can it be controlled by senolysis? Inflamm Regen. 2022;42:11. https://doi.org/10.1186/s41232-022-00197-8\u003c/li\u003e\n\u003cli\u003eGe Q, Yang J, Huang F, et al. Multimodal single-cell analyses reveal molecular markers of neuronal senescence in human drug-resistant epilepsy. J Clin Invest. 2025;135. https://doi.org/10.1172/JCI188942\u003c/li\u003e\n\u003cli\u003eLeo A, Nesci V, Tallarico M, et al. IL-6 receptor blockade by tocilizumab has anti-absence and anti-epileptogenic effects in the WAG/Rij rat model of absence epilepsy. Neurotherapeutics. 2020;17:2004-2014. https://doi.org/10.1007/s13311-020-00894-w\u003c/li\u003e\n\u003cli\u003eWang Q, Qin B, Yu H, et al. Dingxian pill alleviates hippocampal neuronal apoptosis in epileptic mice through TNF-\u0026alpha;/TNFR1 signaling pathway inhibition. J Ethnopharmacol. 2024;334:118579. https://doi.org/10.1016/j.jep.2024.118579\u003c/li\u003e\n\u003cli\u003ePijet B, Konopka A, Rejmak E, et al. The matrix metalloproteinase inhibitor marimastat inhibits seizures in a model of kainic acid-induced status epilepticus. Sci Rep. 2020;10:21314. https://doi.org/10.1038/s41598-020-78341-y\u003c/li\u003e\n\u003cli\u003eRibierre T, Bacq A, Donneger F, et al. Targeting pathological cells with senolytic drugs reduces seizures in neurodevelopmental mTOR-related epilepsy. Nat Neurosci. 2024;27:1125-1136. https://doi.org/10.1038/s41593-024-01634-2\u003c/li\u003e\n\u003cli\u003eBasnyat P, Peltola M, Raitanen J, et al. Elevated IL-6 plasma levels are associated with GAD antibodies-associated autoimmune epilepsy. Front Cell Neurosci. 2023;17:1129907. https://doi.org/10.3389/fncel.2023.1129907\u003c/li\u003e\n\u003cli\u003eSoltani Khaboushan A, Yazdanpanah N, Rezaei N. Neuroinflammation and proinflammatory cytokines in epileptogenesis. Mol Neurobiol. 2022;59:1724-1743. https://doi.org/10.1007/s12035-021-02685-z\u003c/li\u003e\n\u003cli\u003eMahawar S, Rakshit D, Patel I, et al. Fisetin-loaded chitosan nanoparticles ameliorate pilocarpine-induced temporal lobe epilepsy and associated neurobehavioral alterations in mice: Role of ROS/TNF-\u0026alpha;-NLRP3 inflammasomes pathway. Nanomedicine. 2024;59:102752. https://doi.org/10.1016/j.nano.2024.102752\u003c/li\u003e\n\u003cli\u003eBronisz E, Kurkowska-Jastrzębska I. Matrix metalloproteinase 9 in epilepsy: The role of neuroinflammation in seizure development. Mediators Inflamm. 2016;2016:7369020. https://doi.org/10.1155/2016/7369020\u003c/li\u003e\n\u003cli\u003eWang S, Chi K, Wu D, et al. Insulin-like growth factor binding proteins in kidney disease. Front Pharmacol. 2021;12:807119. https://doi.org/10.3389/fphar.2021.807119\u003c/li\u003e\n\u003cli\u003eYang B, Zhang L, Cao Y, et al. Overexpression of lncRNA IGFBP4-1 reprograms energy metabolism to promote lung cancer progression. Mol Cancer. 2017;16:154. https://doi.org/10.1186/s12943-017-0723-5\u003c/li\u003e\n\u003cli\u003eBach LA. Insulin-like growth factor binding proteins 4-6. Best Pract Res Clin Endocrinol Metab. 2015;29:713-722. https://doi.org/10.1016/j.beem.2015.06.002\u003c/li\u003e\n\u003cli\u003eLee YY, Mok MT, Kang W, et al. Loss of tumor suppressor IGFBP4 drives epigenetic reprogramming in hepatic carcinogenesis. Nucleic Acids Res. 2018;46:8832-8847. https://doi.org/10.1093/nar/gky589\u003c/li\u003e\n\u003cli\u003eLiang J, Huang Y, Peng D, et al. IGFBP2 and IGFBP4 interact to activate complement pathway in diabetic kidney disease. Ren Fail. 2025;47:2440528. https://doi.org/10.1080/0886022X.2024.2440528\u003c/li\u003e\n\u003cli\u003eWu L, Qin Y, Yuan H, et al. Anti-inflammatory and neuroprotective effects of insulin-like growth factor-1 overexpression in pentylenetetrazole (PTZ)-induced mouse model of chronic epilepsy. Brain Res. 2022;1785:147881. https://doi.org/10.1016/j.brainres.2022.147881\u003c/li\u003e\n\u003cli\u003eJiang G, Wang W, Cao Q, et al. Insulin growth factor-1 (IGF-1) enhances hippocampal excitatory and seizure activity through IGF-1 receptor-mediated mechanisms in the epileptic brain. Clin Sci (Lond). 2015;129:1047-1060. https://doi.org/10.1042/CS20150394\u003c/li\u003e\n\u003cli\u003eDya GA, Lebedeva OS, Gushchevarov DA, et al. Specific cleavage of IGFBP-4 by PAPP-A in nervous tissue. Biochem Biophys Res Commun. 2024;733:150655. https://doi.org/10.1016/j.bbrc.2024.150655\u003c/li\u003e\n\u003cli\u003eJiang X, Zhao J, Ju L, et al. Temporal expression patterns of insulin-like growth factor binding protein-4 in the embryonic and postnatal rat brain. BMC Neurosci. 2013;14:132. https://doi.org/10.1186/1471-2202-14-132\u003c/li\u003e\n\u003cli\u003eNiu H, Gou R, Xu Q, et al. Recombinant insulin-like growth factor binding protein-4 inhibits proliferation and promotes differentiation of neural progenitor cells. Neurosci Lett. 2017;642:71-76. https://doi.org/10.1016/j.neulet.2017.02.040\u003c/li\u003e\n\u003cli\u003eZhang J, Moats-Staats BM, Ye P, et al. Expression of insulin-like growth factor system genes during the early postnatal neurogenesis in the mouse hippocampus. J Neurosci Res. 2007;85:1618-1627. https://doi.org/10.1002/jnr.21275\u003c/li\u003e\n\u003cli\u003eCatarinella G, Nicoletti C, Bracaglia A, et al. Serpine1 drives a cell-autonomous pathogenic signaling in Hutchinson-Gilford progeria syndrome. Cell Death Dis. 2022;13:737. https://doi.org/10.1038/s41419-022-05168-y\u003c/li\u003e\n\u003cli\u003eGhosh AK, Vaughan DE. PAI-1 in tissue fibrosis. J Cell Physiol. 2012;227:493-507. https://doi.org/10.1002/jcp.22783\u003c/li\u003e\n\u003cli\u003eSillen M, Declerck PJ. A narrative review on plasminogen activator inhibitor-1 and its (patho)physiological role: To target or not to target? Int J Mol Sci. 2021;22. https://doi.org/10.3390/ijms22052721\u003c/li\u003e\n\u003cli\u003eWang B, Gu B, Gao L, et al. Serpine1 facilitates metastasis in gastric cancer through anoikis resistance and tumor microenvironment remodeling. Small. 2025;21:e2500136. https://doi.org/10.1002/smll.202500136\u003c/li\u003e\n\u003cli\u003eSu YH, Wu YZ, Ann DK, et al. Obesity promotes radioresistance through serpine1-mediated aggressiveness and DNA repair of triple-negative breast cancer. Cell Death Dis. 2023;14:53. https://doi.org/10.1038/s41419-023-05576-8\u003c/li\u003e\n\u003cli\u003ePeng S, Xue G, Gong L, et al. A long-acting PAI-1 inhibitor reduces thrombus formation. Thromb Haemost. 2017;117:1338-1347. https://doi.org/10.1160/TH17-03-0185\u003c/li\u003e\n\u003cli\u003eNeuhaus W, Gaiser F, Mahringer A, et al. The pivotal role of astrocytes in an in vitro stroke model of the blood-brain barrier. Front Cell Neurosci. 2014;8:352. https://doi.org/10.3389/fncel.2014.00352\u003c/li\u003e\n\u003cli\u003eDohgu S, Takata F, Matsumoto J, et al. Autocrine and paracrine up-regulation of blood-brain barrier function by plasminogen activator inhibitor-1. Microvasc Res. 2011;81:103-107. https://doi.org/10.1016/j.mvr.2010.10.007\u003c/li\u003e\n\u003cli\u003eShan Y, Tan S, Lin Y, et al. The glucagon-like peptide-1 receptor agonist reduces inflammation and blood-brain barrier breakdown in an astrocyte-dependent manner in experimental stroke. J Neuroinflammation. 2019;16:242. https://doi.org/10.1186/s12974-019-1638-6\u003c/li\u003e\n\u003cli\u003eCai Y, Liu J, Wang B, et al. Microglia in the neuroinflammatory pathogenesis of Alzheimer\u0026apos;s disease and related therapeutic targets. Front Immunol. 2022;13:856376. https://doi.org/10.3389/fimmu.2022.856376\u003c/li\u003e\n\u003cli\u003eYu W, He J, Cai X, et al. Neuroimmune crosstalk between the peripheral and the central immune system in amyotrophic lateral sclerosis. Front Aging Neurosci. 2022;14:890958. https://doi.org/10.3389/fnagi.2022.890958\u003c/li\u003e\n\u003cli\u003eCerri C, Genovesi S, Allegra M, et al. The chemokine CCL2 mediates the seizure-enhancing effects of systemic inflammation. J Neurosci. 2016;36:3777-3788. https://doi.org/10.1523/JNEUROSCI.4513-15.2016\u003c/li\u003e\n\u003cli\u003eTian DS, Peng J, Murugan M, et al. Chemokine CCL2-CCR2 signaling induces neuronal cell death via STAT3 activation and IL-1\u0026beta; production after status epilepticus. J Neurosci. 2017;37:7878-7892. https://doi.org/10.1523/JNEUROSCI.0173-17.2017\u003c/li\u003e\n\u003cli\u003eZhang S, Gao Z, Feng L, et al. Prevention and treatment strategies for Alzheimer\u0026apos;s disease: Focusing on microglia and astrocytes in neuroinflammation. J Inflamm Res. 2024;17:7235-7259. https://doi.org/10.2147/JIR.S483412\u003c/li\u003e\n\u003cli\u003eTatomir A, Talpos-Caia A, Anselmo F, et al. The complement system as a biomarker of disease activity and response to treatment in multiple sclerosis. Immunol Res. 2017;65:1103-1109. https://doi.org/10.1007/s12026-017-8946-3\u003c/li\u003e\n\u003cli\u003eWei Y, Chen T, Bosco DB, et al. The complement C3-C3aR pathway mediates microglia-astrocyte interaction following status epilepticus. Glia. 2021;69:1155-1169. https://doi.org/10.1002/glia.23955\u003c/li\u003e\n\u003cli\u003eT\u0026uuml;rkel İ, Memi G, Yazgan B. Impact of spexin on metabolic diseases and inflammation: An updated minireview. Exp Biol Med (Maywood). 2022;247:567-573. https://doi.org/10.1177/15353702211073145\u003c/li\u003e\n\u003cli\u003eMart\u0026iacute;n-Rivada \u0026Aacute;, Guerra-Cantera S, Campillo-Calatayud A, et al. Pappalysins and stanniocalcins and their relationship with the peripheral IGF axis in newborns and during development. J Clin Endocrinol Metab. 2022;107:2912-2924. https://doi.org/10.1210/clinem/dgac109\u003c/li\u003e\n\u003cli\u003eFluca AL, Pani B, Janjusevic M, et al. Unraveling the relationship among insulin resistance, IGF-1, and amyloid-beta 1-40: Is the definition of type 3 diabetes applicable in the cardiovascular field? Life Sci. 2024;352:122911. https://doi.org/10.1016/j.lfs.2024.122911\u003c/li\u003e\n\u003cli\u003eOpazo P, Viana da Silva S, Carta M, et al. CaMKII metaplasticity drives A\u0026beta; oligomer-mediated synaptotoxicity. Cell Rep. 2018;23:3137-3145. https://doi.org/10.1016/j.celrep.2018.05.022\u003c/li\u003e\n\u003cli\u003eFang Y, Si X, Wang J, et al. Alzheimer disease and epilepsy: A mendelian randomization study. Neurology. 2023;101:e399-e409. https://doi.org/10.1212/WNL.0000000000207435\u003c/li\u003e\n\u003cli\u003eVossel KA, Tartaglia MC, Nygaard HB, et al. Epileptic activity in Alzheimer\u0026apos;s disease: Causes and clinical relevance. Lancet Neurol. 2017;16:311-322. https://doi.org/10.1016/S1474-4422(17)30044-2\u003c/li\u003e\n\u003cli\u003ePurnell BS, Alves M, Boison D. Astrocyte-neuron circuits in epilepsy. Neurobiol Dis. 2023;179:106058. https://doi.org/10.1016/j.nbd.2023.106058\u003c/li\u003e\n\u003cli\u003eMolina DP, Ariwodola OJ, Weiner JL, et al. Growth hormone and insulin-like growth factor-I alter hippocampal excitatory synaptic transmission in young and old rats. Age (Dordr). 2013;35:1575-1587. https://doi.org/10.1007/s11357-012-9458-1\u003c/li\u003e\n\u003cli\u003eStreet ME, Miraki-Moud F, Sanderson IR, et al. Interleukin-1beta (IL-1beta) and IL-6 modulate insulin-like growth factor-binding protein (IGFBP) secretion in colon cancer epithelial (Caco-2) cells. J Endocrinol. 2003;179:405-415. https://doi.org/10.1677/joe.0.1790405\u003c/li\u003e\n\u003cli\u003eGerasimenko A, Baldassari S, Baulac SM. mTOR pathway: Insights into an established pathway for brain mosaicism in epilepsy. Neurobiol Dis. 2023;182:106144. https://doi.org/10.1016/j.nbd.2023.106144\u003c/li\u003e\n\u003cli\u003eZhao W, Xie C, Zhang X, et al. Advances in the mTOR signaling pathway and its inhibitor rapamycin in epilepsy. Brain Behav. 2023;13:e2995. https://doi.org/10.1002/brb3.2995\u003c/li\u003e\n\u003cli\u003eMazumder AG, Sharma P, Patial V, et al. Ginkgo biloba L. Attenuates spontaneous recurrent seizures and associated neurological conditions in lithium-pilocarpine rat model of temporal lobe epilepsy through inhibition of mammalian target of rapamycin pathway hyperactivation. J Ethnopharmacol. 2017;204:8-17. https://doi.org/10.1016/j.jep.2017.03.032\u003c/li\u003e\n\u003cli\u003eKorn T, Kallies A. T cell responses in the central nervous system. Nat Rev Immunol. 2017;17:179-194. https://doi.org/10.1038/nri.2016.144\u003c/li\u003e\n\u003cli\u003eGarber C, Soung A, Vollmer LL, et al. T cells promote microglia-mediated synaptic elimination and cognitive dysfunction during recovery from neuropathogenic flaviviruses. Nat Neurosci. 2019;22:1276-1288. https://doi.org/10.1038/s41593-019-0427-y\u003c/li\u003e\n\u003cli\u003eSuzuki Y. The immune system utilizes two distinct effector mechanisms of T cells depending on two different life cycle stages of a single pathogen, Toxoplasma gondii, to control its cerebral infection. Parasitol Int. 2020;76:102030. https://doi.org/10.1016/j.parint.2019.102030\u003c/li\u003e\n\u003cli\u003eWu L, Ji NN, Wang H, et al. Domino effect of interleukin-15 and CD8 T-cell-mediated neuronal apoptosis in experimental traumatic brain injury. J Neurotrauma. 2021;38:1450-1463. https://doi.org/10.1089/neu.2020.7303\u003c/li\u003e\n\u003cli\u003eGuti\u0026eacute;rrez-Melo N, Baumjohann D. T follicular helper cells in cancer. Trends Cancer. 2023;9:309-325. https://doi.org/10.1016/j.trecan.2023.01.002\u003c/li\u003e\n\u003cli\u003eTanaka S, Furuta K. Roles of IgE and histamine in mast cell maturation. Cells. 2021;10. https://doi.org/10.3390/cells10082121\u003c/li\u003e\n\u003cli\u003eYue J, Tan Y, Huan R, et al. Mast cell activation mediates blood-brain barrier impairment and cognitive dysfunction in septic mice in a histamine-dependent pathway. Front Immunol. 2023;14:1090288. https://doi.org/10.3389/fimmu.2023.1090288\u003c/li\u003e\n\u003cli\u003eHuang X, Lan Z, Hu Z. Role and mechanisms of mast cells in brain disorders. Front Immunol. 2024;15:1445867. https://doi.org/10.3389/fimmu.2024.1445867\u003c/li\u003e\n\u003cli\u003eManohar M, Kandikattu HK, Upparahalli Venkateshaiah S, et al. Eosinophils in the pathogenesis of pancreatic disorders. Semin Immunopathol. 2021;43:411-422. https://doi.org/10.1007/s00281-021-00863-7\u003c/li\u003e\n\u003cli\u003eGu Q, Wiggers ME, Gleich GJ, et al. Sensitization of isolated rat vagal pulmonary sensory neurons by eosinophil-derived cationic proteins. Am J Physiol Lung Cell Mol Physiol. 2008;294:L544-552. https://doi.org/10.1152/ajplung.00392.2007\u003c/li\u003e\n\u003cli\u003eLee MK, Woo SR, Noh JK, et al. Prognostic significance of SASP-related gene signature of radiation therapy in head and neck squamous cell carcinoma. Mol Cancer Ther. 2024;23:1348-1359. https://doi.org/10.1158/1535-7163.MCT-23-0738\u003c/li\u003e\n\u003cli\u003eSaul D, Kosinsky RL, Atkinson EJ, et al. A new gene set identifies senescent cells and predicts senescence-associated pathways across tissues. Nat Commun. 2022;13:4827. https://doi.org/10.1038/s41467-022-32552-1\u003c/li\u003e\n\u003cli\u003eRitchie ME, Phipson B, Wu D, et al. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. https://doi.org/10.1093/nar/gkv007\u003c/li\u003e\n\u003cli\u003eZhang X, Chao P, Zhang L, et al. Single-cell RNA and transcriptome sequencing profiles identify immune-associated key genes in the development of diabetic kidney disease. Front Immunol. 2023;14:1030198. https://doi.org/10.3389/fimmu.2023.1030198\u003c/li\u003e\n\u003cli\u003eZhou W, Li H, Zhang J, et al. Identification and mechanism analysis of biomarkers related to butyrate metabolism in COVID-19 patients. Ann Med. 2025;57:2477301. https://doi.org/10.1080/07853890.2025.2477301\u003c/li\u003e\n\u003cli\u003eZhang H, Zheng C, Xu Y, et al. Comprehensive molecular and cellular characterization of endoplasmic reticulum stress-related key genes in renal ischemia/reperfusion injury. Front Immunol. 2024;15:1340997. https://doi.org/10.3389/fimmu.2024.1340997\u003c/li\u003e\n\u003cli\u003eShi H, Yuan X, Liu G, et al. Identifying and validating GSTM5 as an immunogenic gene in diabetic foot ulcer using bioinformatics and machine learning. J Inflamm Res. 2023;16:6241-6256. https://doi.org/10.2147/JIR.S440529\u003c/li\u003e\n\u003cli\u003eHou Q, Jiang J, Na K, et al. Potential therapeutic targets for COVID-19 complicated with pulmonary hypertension: A bioinformatics and early validation study. Sci Rep. 2024;14:9294. https://doi.org/10.1038/s41598-024-59891-7\u003c/li\u003e\n\u003cli\u003eSui Z, Wu X, Du L, et al. Characterization of the immune cell infiltration landscape in esophageal squamous cell carcinoma. Front Oncol. 2022;12:879326. https://doi.org/10.3389/fonc.2022.879326\u003c/li\u003e\n\u003cli\u003eHarrison AM, Goozee R. Psych-related iPhone apps. J Ment Health. 2014;23:48-50. https://doi.org/10.3109/09638237.2013.869575\u003c/li\u003e\n\u003cli\u003eQiang M, Dai Z. Biomarkers of UVB radiation-related senescent fibroblasts. Sci Rep. 2024;14:933. https://doi.org/10.1038/s41598-024-51543-y\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Supplementary Tables","content":"\u003cp\u003eSupplementary Tables 1-8 are not available with this version. \u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"neurochemical-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"nere","sideBox":"Learn more about [Neurochemical Research](https://www.springer.com/journal/11064)","snPcode":"11064","submissionUrl":"https://submission.nature.com/new-submission/11064/3","title":"Neurochemical Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Epilepsy, Senescence-associated secretory phenotype, SERPINE1, Neuroinflammation, Biomarkers","lastPublishedDoi":"10.21203/rs.3.rs-9165812/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9165812/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEpilepsy is increasingly linked to cellular senescence and the senescence-associated secretory phenotype (SASP), which propagates sterile neuroinflammation and tissue remodeling. However, the specific SASP-related molecular signatures driving epileptogenesis remain poorly characterized. Integrating human cortical transcriptomic profiles (GSE256068) with a curated SASP gene set, we identified 22 differentially expressed candidates. Machine-learning feature selection (LASSO and SVM-RFE) converged on a robust five-biomarker panel (IGFBP4, SERPINE1, CCL2, C3, and SPX), which demonstrated consistent differential expression in an independent, cross-regional hippocampal dataset (GSE134697). A diagnostic nomogram integrating these markers achieved excellent discrimination and clinical net benefit. Functional enrichment linked this panel to oxidative phosphorylation and translational control. Crucially, the up-regulation of SERPINE1 and CCL2 highlights a neurochemical mechanism involving extracellular matrix remodeling and inferred neuroimmune microenvironment shifts. In silico regulatory network analysis further predicted upstream transcription factors governing this axis. Finally, RT-qPCR validation in the peripheral blood of a kainic acid-induced epilepsy mouse model confirmed the significant dysregulation of four key biomarkers (IGFBP4, SERPINE1, CCL2, and C3). Collectively, our study defines a highly accurate SASP-associated biomarker panel and provides profound neurochemical insights into senescence-linked inflammatory networks, offering novel targets for the diagnosis and management of epilepsy.\u003c/p\u003e","manuscriptTitle":"Identification and validation of senescence-associated secretory phenotype-related biomarkers in epilepsy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-02 18:18:01","doi":"10.21203/rs.3.rs-9165812/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"37394301278523347264260006186688404061","date":"2026-05-02T13:18:36+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-29T19:54:38+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-23T17:21:40+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-21T04:47:11+00:00","index":"","fulltext":""},{"type":"submitted","content":"Neurochemical Research","date":"2026-03-19T06:24:39+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"neurochemical-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"nere","sideBox":"Learn more about [Neurochemical Research](https://www.springer.com/journal/11064)","snPcode":"11064","submissionUrl":"https://submission.nature.com/new-submission/11064/3","title":"Neurochemical Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"51f18d28-d660-4fce-a14b-b1c78f6413ab","owner":[],"postedDate":"April 2nd, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"37394301278523347264260006186688404061","date":"2026-05-02T13:18:36+00:00","index":35,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-02T18:18:01+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-02 18:18:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9165812","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9165812","identity":"rs-9165812","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}