Epilepsy is a chronic neurological disorder characterized by recurrent and unpredictable seizures. Seizures are caused by abnormal electrical activity in the brain, which can result in sudden and uncontrolled bursts of abnormal brain cell firing, leading to various symptoms (1, 2). These symptoms can be mild, such as brief periods of impaired consciousness, confusion, or altered sensations. They can also be severe, including muscle convulsions, fainting, loss of consciousness, generalized convulsions, and foaming at the mouth (3). In recent years, our understanding of the pathogenesis of epilepsy has deepened, although the exact mechanisms behind its onset are still unclear (4). To provide more precise guidance for preventing and treating epilepsy, it is necessary to unravel its pathogenesis at the molecular level. In this study, we utilized a dataset of epilepsy from the Gene Expression Omnibus (GEO) database and conducted bioinformatics analysis on the genes associated with epilepsy.

Ferroptosis is a similar form of cell death with distinctive biological features to other common cell death mechanisms. A prominent feature of ferroptosis is the accumulation of free iron within cells resulting in an irreversible lipid peroxidation reaction (5). Excessive accumulation of intracellular ions can lead to lipid oxidation, membrane damage, and functional impairment, ultimately resulting in cell death. Factors involved in the regulation of ferroptosis include lipid peroxidation products, antioxidants, and iron-related proteins, as well as molecules and pathways associated with cell survival and death (6). Significant progress has been made in the study of ferroptosis, particularly in understanding its regulatory mechanisms, related signaling pathways, and its association with the development of epilepsy (7, 8). These studies contribute to elucidating the deeper mechanisms underlying cell death and provide theoretical and practical foundations for developing novel therapeutic approaches for epilepsy.

Gene set variation analysis (GSVA) is a non-parametric unsupervised analysis method primarily used to transform gene expression matrices into gene set expression matrices for the evaluation of transcriptional enrichment of different metabolic pathways among different samples (9, 10). To investigate the biological process variations between epilepsy patients and normal controls, the R package “GSVA” can be employed for gene set variation analysis based on gene expression profiling datasets from different epilepsy patients and normal controls.

Ferroptosis plays an important role in epilepsy, and this research aims to identify key genes associated with epilepsy and ferroptosis, which may serve as novel biomarkers or possible therapeutic targets for epilepsy. In this study, the CIBERSORT algorithm was performed to evaluate the immune infiltration characteristics of epilepsy and normal samples in the integrated dataset (GSE32534 and GSE143272), assessing the composition of 22 immune cell types in each group. The selected hub genes were then associated with the levels of immune cell infiltration using Pearson correlation coefficients and significance levels. Finally, a compound network, drug network, mRNA–miRNA network, and transcription factor network of the three hub genes were constructed. This study provides a research foundation for exploring potential regulatory targets and possible mechanisms of epilepsy, offering new insights into the treatment of this disease.

Epilepsy is one of the most common chronic neurological disorders. Moreover, it is a complex disease primarily caused by genetic and environmental factors, such as a family history of epilepsy, neurological comorbidities, premature birth, maternal alcohol abuse, and smoking during pregnancy. The identification of candidate biomarkers and the understanding of the pathogenic mechanisms are crucial for early diagnosis and effective treatment of epilepsy (14). Moreover, ferroptosis is associated with various diseases, such as tumors and neurological disorders. Ferroptosis is a form of cell death that can cause damage to cells and tissues when there is an excessive amount of iron ions (15). In epilepsy, excessive ferroptosis may be involved in the occurrence and progression of the disease (16). However, the specific mechanisms and role of ferroptosis in these diseases and their treatments require further research and exploration.

In our research, we identified the differentially expressed genes related to ferroptosis in epilepsy patients. A total of 33 differentially expressed genes were identified in epilepsy, and these findings suggest the involvement of ferroptosis-related genes in the pathogenesis of epilepsy and indicate potential pharmacological targets (17, 18). To understand the role of these ferroptosis-related genes in epilepsy, GO enrichment analyses were conducted. The results revealed that the biological process is primarily involved in pathways associated with response to reactive oxygen species or oxidative stress and regulation of reactive oxygen species metabolic process. The cellular component analysis predominantly involves the secondary lysosome, endocytic vesicle lumen, and endocytic vesicle. In terms of molecular function, these genes are mainly enriched in ubiquitin protein ligase binding, ubiquitin-like protein ligase blinding, and peroxidase activity, which is consistent with previous studies (19). In previous studies, some antioxidant drugs have already been used in medical use, which provided potential treatment for reducing the seizure burden (20). A study indicated that (21) abnormal synaptic transmission regulation is crucial in the development of various brain disorders, including epilepsy. In epilepsy, several mechanisms are involved in the modulation of synaptic transmission. These mechanisms encompass presynaptic regulators responsible for synaptic vesicle formation and release, as well as postsynaptic receptors and neuromodulators that influence the frequency and intensity of epileptic seizures. Moreover, through KEGG pathway enrichment we found the 33 FDEGS were mostly associated with legionellosis, HIF-1 signaling pathway, and autophagy-animal. A case report showed a patient with legionella longbeachae pneumonia developed epileptic seizure after using moxifloxacin and provided a possible connection between epileptic and legionellosis (22). Li et al. (23) indicated that the HIF-1α–Notch signaling pathway plays a crucial role in enhancing neurogenesis during acute epilepsy. Furthermore, they observed a reduction in neurogenesis during epileptogenesis when this pathway was blocked. Previous research found electroacupuncture treatment promotes autophagy during epilepsy onset by significantly downregulating the AKT/mTOR signaling pathway, which is consistent with our research (24). Then, we conducted the GSEA, and the result showed that the top 4 pathways were T-cell receptor signaling pathway, cytokine receptor interaction, ribosome, and natural killer cell-mediated cytotoxicity. These results may provide ideas for our future research. GSVA based on known gene sets (25), including KEGG pathways and GO functional terms, evaluated the enrichment level of these gene sets by calculating the overall pattern variation of gene expression in two groups, which helps us understand the relationship between gene sets, diseases, and biological processes.

Furthermore, the PPI network revealed three key FDEGs, namely, HIF1A, TLR4, and CASP8. These hub genes have high diagnostic efficiency in epilepsy. Studies have demonstrated that the activation of hypoxia-inducible factor 1 (HIF-1) can enhance the transcription of multidrug transporters, such as P-glycoprotein (P-gp), in astrocytes. As a result, this process reduces the accumulation of antiepileptic drugs in the brain (26). The former study found that the status epilepticus could induce P-gp and EPO-R expression in cortical pyramidal neurons. In vitro, excitotoxic stress can also induce the expression of EPO-R and P-gp simultaneously with both HIF-1α and NFkB nuclear translocation in primary cortical neurons (27). P-gp overexpression in the brain is associated with changes in membrane depolarization in refractory epilepsy (28). Moreover, a previous study found that the expression levels of HIF-1α and P-gp were coordinately increased in the hippocampus and temporal lobes of patients with mesial temporal lobe epilepsy (29, 30). Li et al. (31) showed HIF-1α as a direct target of miRNA-153 and may serve as a potential diagnostic biomarker and therapeutic target for refractory epilepsy. Some studies showed that the TLR4 signaling pathway produces an antiepileptic effect. Yang et al. (32) successfully established an epileptic model and built a pharmacological network, which found that the hippocampus experienced in the epilepsy group contained an upregulation of TLR4, MYD88, and Caspase-3 compared with the control group counterparts. Zhu et al. (33) showed that blocking TLR4/MYD88 signaling attenuated KA-induced neuroinflammation and neuronal damage in the hippocampus.

Recent research demonstrated that after the termination of seizures, the cleavage of (p18) caspase-8 can be detected, along with an increase in the cleavage of the substrate Ile-Glu-Thr-Asp (IETD)-nitroanilide and the appearance of cleaved (p15) Bid. In vivo experiments, intraventricular administration of z-iETD-fluoromethylketone significantly reduces the activity of caspase 8, 9, and 3 induced by seizures as well as decreases the cleavage of Bid and caspase-9 and neuronal death. The study showed that intervention targeting caspase-8 and/or the death receptor signaling pathway may protect the brain from injury caused by seizures (34). According to our study, these hub genes were possibly associated with ferroptosis-dependent epilepsy, which still needs more research.

Neuroinflammation is thought to be a contributing factor to epilepsy, and ferroptosis is also reported to be associated with immune disease (35). In our study, we investigated the presence of immune cells and immune infiltration in epilepsy, which revealed significant differences in certain immune cell populations between the epilepsy group and the control group. Compared with the normal group, the CD4-naïve T cells, gamma delta T cells, M1 macrophages, and neutrophils showed low expression in the epilepsy group. It is worth noting that recent research carried out by Xu et al. showed that γδ T lymphocytes producing the pro-inflammatory cytokine IL-17 are concentrated in the epileptogenic zone and positively correlated with the severity of epilepsy. In contrast, the number of infiltrating regulatory T cells (Tregs) in the brain is negatively correlated with disease severity (36). A systematic review showed that the neutrophil-to-lymphocyte ratio (NLR) in epilepsy was higher than in healthy controls. Moreover, elevated NLR value was a good biomarker of inflammation (37). The results were different from our study, possibly because this study is about the acute phase of epilepsy. We also found a significant difference between mast cell-activated cells and 20 other immune cells. Previous research indicated that mast cells activated cells may provide protective activity against seizure mediated by serotonin (38). In addition, Francesco Girolamo et al. proved mast cells influenced by microbes have been shown to promote neuroprotection in neurodegenerative disorders (39). Our study proved the three hub genes were significantly correlated with monocytes (40), neutrophils (41), CD4-naïve T cells, and follicular helper T cells (42). Based on their association, hub genes may profoundly affect epilepsy by regulating immune cell infiltration and modulating inflammation (43, 44).

Previous studies found that ferroptosis impacts immune cells in two fundamentally different ways. On the one hand, ferroptosis affects the number and function of the immune cells themselves. On the other hand, ferroptotic cells can be recognized by immune cells, and then, they trigger a range of inflammatory or specific responses. In our study, we also explored the connection between immune cells and ferroptosis-related genes. An imbalance of macrophage M1/M2 polarization contributes to various diseases or inflammatory conditions (45). Mounting evidence suggests that macrophage polarization and ferroptosis can influence each other at the cell-autonomous level or by communication with other cells in a context-dependent manner. Infiltrating monocytes have also been suggested to promote inflammation after status epilepticus (40). It is possible that alternative monocyte-derived populations, like differentiated macrophages, may affect seizure severity and hippocampal damage in different ways. In our research, M1 macrophage in the epilepsy group were higher than in the normal group, which may be the result of increased TLR4 leading to decreased M1 macrophages and affecting their ability to clear ferroptosis cells, ultimately exacerbating epilepsy (46). Previous several in vitro and in vivo observations suggest that the activity and function of cytotoxic T cells (CD8) and helper T cells (CD4) are regulated by lipid peroxidation and ferroptosis. Moreover, the study identified that CD4 T cells in the peripheral blood are present in a higher proportion of epilepsy patients, which means CD4 T cells may play an important role in the occurrence and development of epilepsy. Neutrophil granulocytes are the first immune cells recruited to sites of inflammation, and the current evidence suggests that neutrophils participate in sustaining inflammation caused by ferroptotic tissue damage (47). Moreover, a study reveals neutrophils and TNFα as central regulators of neuronal hyperexcitability of diverse etiology in human epilepsy. However, the specific roles of these immune cells in epilepsy and ferroptosis remain unclear. Therefore, further research is urgently needed to elucidate the underlying mechanisms of these immune cells in epilepsy. This research may contribute to identifying new therapeutic or adjunctive treatment strategies for patients with epilepsy.

Despite the remarkable sense, there are also many limitations in our study. First, the data were from the GEO database, including patients with tumors and patients in the restore stage, and our results take the intersection of the different databases. As the original dataset did not provide complete details of other comorbidities or diseases, it is difficult to guarantee that the epilepsy patients included in our analysis were free from other illnesses. Therefore, it poses a challenge to accurately assess and isolate the influence of other diseases on our results. Second, despite integrating multiple epilepsy microarray datasets, the sample size remained limited; increasing the sample size can enhance the statistical power to detect hub genes and immune cells. In addition, although three target genes were identified, it is indeed necessary to conduct wet lab experiments to validate them. By validating the target genes through wet lab experiments, more specific and reliable results can further support our research findings.

In this study, we successfully identified and validated differentially expressed genes (DEGs) specific to epilepsy that are associated with ferroptosis. This finding enhances our understanding of the molecular mechanisms underlying epilepsy and their relationship to the process of ferroptosis. Additionally, our analysis revealed significant differences in immune infiltration between the epilepsy group and the normal control group, highlighting the potential role of the immune system in epilepsy pathogenesis. Furthermore, our study employed bioinformatics approaches to predict target compounds, drugs, miRNAs, and transcription factors that may be implicated in epilepsy. These predictions offer valuable insights for future research aiming to elucidate the exact pathogenic mechanisms underlying epilepsy and provide potential targets for the development of innovative treatment strategies.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

Ethical approval was not required for the study involving humans in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and the institutional requirements.

DX: Data curation, Formal analysis, Methodology, Writing – original draft, Writing – review & editing, Software. MC: Writing – review & editing, Investigation, Methodology, Validation. YC: Writing – review & editing, Data curation, Formal analysis. YF: Writing – review & editing, Methodology, Software, Visualization. JW: Investigation, Supervision, Writing – review & editing. XZ: Writing – review & editing, Formal analysis, Supervision. FX: Writing – review & editing, Conceptualization, Project administration.