v
Search
Advanced

Publications > Journals > Neurosurgical Subspecialties> Article Full Text

  • Original Article
  • OPEN ACCESS

Immune Cell Phenotypes, Inflammatory Proteins and Epilepsy: A Mendelian Randomization Study to Investigate Causal Pathways

  • Guizhi Wu1,2,#,
  • Mengen Liu1,2,#,
  • Hongdong Wang1,2,
  • Xu Shao2,
  • Jiancong Weng2,*  and
  • Hong Tian2,* 
 Author information 

Abstract

Background and objectives

Emerging evidence implicates immune dysregulation and neuroinflammation in the pathogenesis of epilepsy, yet the causal mechanisms remain unclear. This study aimed to investigate the causal effects of immune cells and inflammatory proteins on epilepsy and evaluate the mediating role of inflammatory proteins.

Methods

This study utilized the largest available genome-wide association study data on immune cell phenotypes and inflammatory proteins as exposures, and epilepsy genome-wide association study data from the FinnGen dataset as outcomes. Five Mendelian randomization (MR) methods were applied within a two-sample MR framework to assess causal effects. Furthermore, a two-step MR analysis was conducted to quantify the proportion of epilepsy and its subtypes influenced by immune cells through inflammatory proteins.

Results

The two-sample MR analysis identified 32 immune cell phenotypes associated with epilepsy risk (19 risk-increasing, e.g., CD19+ B cells; 13 protective, e.g., regulatory T cell subsets). Subtype analyses revealed 30 immune phenotypes associated with generalized epilepsy and 26 with focal epilepsy. Eight inflammatory proteins showed suggestive causal effects on epilepsy: C-C chemokine ligand 23, C-X-C motif chemokine ligand 6, C-X-C motif chemokine ligand 11, and vascular endothelial growth factor A increased epilepsy risk, while interleukin-13 (IL-13), leukemia inhibitory factor receptor, tumor necrosis factor, and osteoprotegerin conferred protection. Mediation analysis indicated that inflammatory proteins mediated 6.3–13.5% of the immune effects on epilepsy. Specifically, CD14+CD16+ monocytes increased epilepsy risk through elevated C-C chemokine ligand 23 levels (8.5% mediation), while effector memory double-negative (CD4−CD8−) T cells reduced epilepsy risk via upregulation of IL-13 (6.3%). Sensitivity analyses confirmed the robustness of these findings (P heterogeneity/pleiotropy > 0.05). Although no associations reached Bonferroni-corrected significance, the findings implicate B cells, monocytes, regulatory T cells, and cytokines (e.g., IL-13, leukemia inhibitory factor receptor) in the pathogenesis of epilepsy, with inflammatory proteins acting as partial mediators.

Conclusions

These results enhance our understanding of immune-inflammatory pathways in epilepsy and highlight potential therapeutic targets. Future studies should validate these findings across diverse populations and further elucidate the molecular mechanisms underlying the identified associations.

Keywords

Immune cell, Inflammatory protein, Epilepsy, Pathogenesis, Mendelian randomization, Causal relationship

Introduction

Epilepsy represents a chronic and debilitating neurological condition marked by recurrent seizures caused by abnormal synchronized neuronal discharges.1 Seizures, the hallmark of epilepsy, present with diverse and complex clinical presentations, although they often occur in a consistent or stereotypical manner in individual patients. The condition can affect individuals of any age, with the highest incidence observed in infants under one year of age and adults over 50 years.2 Epilepsy significantly impairs both the physiological and psychosocial well-being of patients and poses direct or indirect risks to their lives. Currently, it represents a major global public health challenge, affecting approximately 70 million people worldwide.2

The underlying mechanisms of epilepsy remain incompletely understood; however, increasing evidence highlights the critical roles of immune dysregulation and neuroinflammation in its pathogenesis.3,4 Aberrant immune responses and neuroinflammatory activation have been consistently observed in both animal models of epilepsy and in epileptogenic brain regions of human patients.5,6 Several inflammatory mediators released by immune cells are known to lower the neuronal excitability threshold, thereby facilitating seizure activity.7 Therefore, elucidating the roles of specific immune phenotypes and inflammatory mediators in epilepsy is critical for advancing our comprehension of its pathophysiology and guiding the development of targeted therapeutic strategies.

Mendelian randomization (MR) is a robust analytical approach that utilizes genetic variants as instrumental variables (IVs) to infer causal relationships between exposures and outcomes.8 This methodology is based on Mendel’s second law, which states that genetic variants are randomly assigned during gamete formation.8 This randomization helps minimize the influence of reverse causation and confounding, common limitations in conventional observational studies.9 MR has become an increasingly utilized tool in recent epidemiological research for exploring causal associations between epilepsy and various risk factors.10,11 Notably, accumulating evidence from MR analyses has suggested potential causal links between epilepsy and immune cell phenotypes or inflammatory mediators.12–14

In this study, we leveraged novel genetic datasets to perform MR analyses aimed at elucidating the causal effects of circulating immune cells and inflammatory proteins on epilepsy. Given that inflammatory proteins are primarily produced through immune cell activity, we also conducted mediation analyses to assess the potential intermediary role of these proteins in the pathogenic pathways linking immune cells to epileptogenesis. These analyses are intended to advance our comprehension of the immunological and inflammatory mechanisms underlying epilepsy and to identify promising targets for therapeutic intervention.

Materials and methods

Study design

This study was designed around three key components, as depicted in Figure 1. The first component investigates the causal effects of 731 immune cell phenotypes on epilepsy and its subtypes. The second component examines the causal associations between epilepsy and 91 inflammatory proteins. The third component utilizes the two-step Mendelian randomization approach to conduct mediation analysis, estimating the extent to which inflammatory proteins mediate the effects of immune cell phenotypes on epilepsy. In the MR analysis, single nucleotide polymorphisms (SNPs) were employed as IVs, which must satisfy three key assumptions: (1) SNPs are strongly associated with the exposure; (2) SNPs are independent of confounding factors; and (3) SNPs influence the outcome solely through the exposure, without any direct effect on the outcome itself.

Study flow chart.
Fig. 1  Study flow chart.

SNP, single-nucleotide polymorphism.

This study complied with the STROBE-MR reporting guidelines. Ethical approval was not required, as the analysis was based on publicly available genome-wide association studies (GWAS) summary statistics and did not involve access to individual-level data.

Data sources

The GWAS summary statistics for immune-related traits are accessible through the GWAS Catalog, with identifiers ranging from GCST90001391 to GCST90002121.15 These datasets were generated from a Sardinian population cohort comprising 3,757 individuals, covering 731 distinct immunophenotypes. In parallel, GWAS summary statistics for inflammatory protein markers are also available in the GWAS Catalog, with reference numbers GCST90274758 to GCST90274848. These datasets originate from a large-scale protein quantitative trait locus analysis, which investigated 91 inflammation-associated proteins in a European-ancestry cohort of 14,824 participants.16

Regarding epilepsy-related genetic data, the GWAS summary statistics were extracted from the FinnGen project’s tenth data release. FinnGen represents a comprehensive genomic initiative that has analyzed more than 500,000 biological samples from the Finnish population, correlating genetic variants with clinical outcomes to advance the understanding of disease etiology and genetic risk factors.17 In this study, the exposure and outcome data were sourced from two distinct, non-overlapping cohorts.

Instrument selection

In accordance with established methodologies,15 IVs for each immune trait and inflammatory protein were selected at a significance threshold of P < 5 × 10−5. To eliminate the effects of linkage disequilibrium, all IVs were required to meet the criteria of r2 < 0.001 and a physical distance > 10,000 kb. Furthermore, the F-statistic was calculated for each IV to assess its strength of association with the exposure. Variants with an F-statistic below 10, indicating weak correlation or limited explanatory power, were excluded to ensure IV robustness.18

MR analysis

We employed five distinct methods—inverse variance weighted (IVW), weighted median, MR-Egger, weighted mode, and simple mode—to perform two-sample MR analysis. These analyses evaluated the causal effects of 731 immune cell phenotypes and 91 inflammatory proteins on epilepsy. The IVW method served as the primary analytical approach. Statistical significance required an IVW P-value < 0.05 and consistent effect direction across the other four MR methods. MR results were quantified using odds ratios (ORs) accompanied by 95% confidence intervals (CIs).

Additionally, a two-step Mendelian randomization approach was employed for mediation analysis to explore whether immune cells influence epilepsy risk by modulating inflammatory protein levels. First, the total causal effect (β_total) of immunophenotypes on epilepsy risk was established using primary MR analysis. Subsequently, a two-step mediation framework was constructed: (1) MR analysis evaluating immunophenotypes as exposures associated with inflammatory protein levels (β1, P < 0.05), followed by (2) MR assessment of inflammatory proteins as exposures influencing epilepsy risk (β2, P < 0.05). The mediation effect was quantified using the product-of-coefficients approach: β_mediation = β1 × β2. The direct effect was calculated as: β_direct = β_total - β_mediation. Directional consistency was confirmed between the mediation and total effects. The proportion mediated was computed as (β_mediation/β_total) × 100%.

To address multiple testing, Bonferroni correction was applied to adjust the significance thresholds. The corrected thresholds were set at 0.0000684 (0.05/731) for immune cells and 0.0005495 (0.05/91) for inflammatory proteins. P-values below these corrected thresholds were interpreted as indicating definitive causal effects, while those above the corrected thresholds but below 0.05 were considered suggestive of potential causal effects. This rigorous approach ensures robust and reliable identification of causal relationships while minimizing the risk of false positives.

Sensitivity analysis

To ensure the robustness and validity of our findings, we implemented a comprehensive sensitivity analysis framework. First, between-instrument heterogeneity was assessed using Cochrane’s Q test. When significant heterogeneity (P < 0.05) was detected, a random-effects IVW model was employed to account for potential variability across genetic instruments. To address biases from horizontal pleiotropy, we employed two complementary analytical approaches: (1) MR-Egger regression, where a non-zero intercept (P < 0.05) indicated directional pleiotropy; and (2) Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) analysis, which systematically identified and removed outlier variants with disproportionate influence on the causal estimates. Additionally, leave-one-out analyses were conducted by sequentially excluding individual IVs and recalculating the IVW estimates to detect influential variants and assess the consistency of the results. All analyses were performed using R software (version 4.3.1). The TwoSampleMR and MR-PRESSO packages were employed for data analysis and visualization.

Results

Causal effects of immune traits on epilepsy

Our study revealed potential correlations between 32 immune cell types and epilepsy, including seven B cell types, three myeloid cell types, four monocyte types, four maturation stages of T cell types, four TBNK types, eight regulatory T cell (Treg) types, and two conventional dendritic cells types (Fig. 2). IVW results showed that 19 immune cell types, including CD14+ CD16+ monocyte absolute count (AC) (OR = 1.038, 95% CI: 1.004–1.073, P = 0.027), CD28 on CD4 Treg (OR = 1.022, 95% CI: 1.001–1.044, P = 0.044), natural killer (NK) AC (OR = 1.047, 95% CI: 1.007–1.088, P = 0.020), and human leukocyte antigen-DR isotype on plasmacytoid dendritic cell (OR = 1.021, 95% CI: 1.003–1.040, P = 0.026) were positively associated with the risk of epilepsy. Conversely, 13 immune cell types, including IgD+ CD38− AC (OR = 0.935, 95% CI: 0.888–0.983, P = 0.009), effector memory double-negative (EM DN CD4−CD8−) %DN (OR = 0.971, 95% CI: 0.946–0.997, P = 0.031), CD14− CD16− AC (OR = 0.982, 95% CI: 0.966–0.999, P = 0.038), and CD28− CD25++ CD8br AC (OR = 0.969, 95% CI: 0.944–0.995, P = 0.021) were negatively associated with the risk of epilepsy. Moreover, MR studies on the two major subtypes of epilepsy indicated that 30 immune cell types were causally associated with generalized epilepsy (File S1: Fig. S2), and 26 immune cell types were causally associated with focal epilepsy (File S1: Fig. S1). It is noteworthy that certain immune cell types may have promotive or inhibitory effects on multiple types of epilepsy. After adjusting for multiple testing using the Bonferroni correction method, no immunophenotypes were detected at the corrected threshold for significance.

Causal effects between immunophenotypes and epilepsy.
Fig. 2  Causal effects between immunophenotypes and epilepsy.

AC, absolute count; CD, cluster of differentiation; cDC, conventional dendritic cells; CI, confidence interval; CM, central memory; DC, dendritic cell; DN, double-negative; EM, effector memory; HLA-DR, human leukocyte antigen-DR isotype; Ig, immunoglobulin; NK, natural killer; nSNP, number of single-nucleotide polymorphisms; OR, odds ratio; TBNK, T-cell, B-cell, and natural killer cell panel; Treg, regulatory T cells.

Causal effects of inflammatory proteins on epilepsy

A total of 13 circulating inflammatory proteins were identified as potentially causally linked to epilepsy and its subtypes (Fig. 3). C-C chemokine ligand 23 (CCL23, OR = 1.069, 95% CI: 1.011–1.131, P = 0.019), C-X-C motif chemokine ligand 11 (CXCL11, OR = 1.076, 95% CI: 1.006–1.150, P = 0.032), CXCL6 (OR = 1.048, 95% CI: 1.003–1.095, P = 0.038), and vascular endothelial growth factor A (VEGF_A, OR = 1.073, 95% CI: 1.025–1.123, P = 0.003) may promote the onset of epilepsy, while interleukin-13 (IL-13, OR = 0.948, 95% CI: 0.901–0.998, P = 0.040), leukemia inhibitory factor receptor (LIF-R, OR = 0.916, 95% CI: 0.852–0.984, P = 0.017), osteoprotegerin (OPG, OR = 0.923, 95% CI: 0.859–0.992, P = 0.030), and tumor necrosis factor (TNF, OR = 0.929, 95% CI: 0.874–0.987, P = 0.018) have the potential to inhibit the occurrence of epilepsy. Furthermore, CXCL6 (OR = 1.060, 95% CI: 1.001–1.122, P = 0.045) exhibited a promotive effect, while IL-13 (OR = 0.902, 95% CI: 0.846–0.962, P = 0.002), LIF-R (OR = 0.897, 95% CI: 0.830–0.969, P = 0.006), and TNF (OR = 0.922, 95% CI: 0.856–0.994, P = 0.034) exerted inhibitory effects in focal epilepsy. Additionally, eukaryotic initiation factor 4E binding protein 1 (OR = 0.695, 95% CI: 0.533–0.906, P = 0.007) and IL-12B (OR = 0.815, 95% CI: 0.671–0.990, P = 0.039) were negatively correlated with the risk of generalized epilepsy, while CD5 (OR = 1.280, 95% CI: 1.031–1.591, P = 0.026) and interleukin 10 receptor subunit beta (OR = 1.172, 95% CI: 1.045–1.315, P = 0.007) showed positive correlations. After applying the Bonferroni correction for multiple testing, no circulating inflammatory proteins were found to have a definitive causal effect on epilepsy.

Causal effects between inflammatory proteins and epilepsy.
Fig. 3  Causal effects between inflammatory proteins and epilepsy.

CD, cluster of differentiation; CI, confidence interval, CCL, C-C chemokine ligand; CXCL, C-X-C motif chemokine ligand; IL, interleukin; IL10RB, interleukin 10 receptor subunit beta; LIF-R, leukemia inhibitory factor receptor; nSNP, number of single-nucleotide polymorphisms; OPG, osteoprotegerin; OR, odds ratio; TNF, tumor necrosis factor; TNFSF, tumor necrosis factor superfamily protein; VEGF_A, vascular endothelial growth factor A; 4EBP1, eukaryotic initiation factor 4E binding protein 1.

Mediation analysis of inflammatory factors

We employed a two-step Mendelian randomization approach to conduct a mediation analysis to further explore the mediating role of inflammatory proteins. Six inflammatory proteins (VEGF_A, IL-13, CCL23, LIF-R, CXCL11, OPG) were identified as mediators between immune cells and epilepsy (Table 1). CD14+ CD16+ monocyte AC and human leukocyte antigen-DR isotype on plasmacytoid dendritic cell were found to promote the occurrence of epilepsy through increased CCL23, with mediation effects accounting for 8.5% and 6.5%, respectively. NK AC was found to promote epilepsy, with 13.5% of the effect mediated through CXCL11. On the other hand, EM DN (CD4−CD8−) %DN exerted a protective effect on epilepsy through promoting IL-13 (6.3%); CD14− CD16− AC inhibited the occurrence of epilepsy through promoting LIF-R (12.4%); CD28− CD25++ CD8br AC exerted a protective effect on epilepsy through promoting OPG (7.3%).

Table 1

The mediating effect of inflammatory proteins between immune cell phenotypes and epilepsy

OutcomeExposureMediatorTotal effect (β)Indirect effect (β1*β2)Direct effect (β-β1*β2)Proportion (β1*β2/β)
EpilepsyIgD+ CD38− ACVEGF_A−0.06760.0507 × 0.0704−0.0712–
EM DN (CD4−CD8−) %DNIL-13−0.02920.0344 × −0.0534−0.02736.29%
CD14+ CD16+ monocyte ACCCL230.03470.0472 × 0.06700.03428.46%
CD14− CD16− ACLIF-R–0.01780.0252 × −0.0877−0.015612.43%
NK ACCXCL110.04570.0845 × 0.07310.039513.52%
CD28− CD25++ CD8br ACOPG−0.03130.0285 × −0.0800−0.02907.29%
CD28 on CD4 TregLIF-R0.02160.0207 × −0.08770.0234–
HLA DR on plasmacytoid DCCCL230.02100.0205 × 0.06700.01976.52%
HLA DR on DCLIF-R0.03200.0221 × −0.08770.0339–
HLA DR on CD33br HLA DR+ CD14−OPG0.02140.0162 × −0.08000.0227–
Focal epilepsyCD127 on CD8brIL-130.0523−0.0691 × −0.10310.045113.63%
CD8 on NKTIL-13−0.0527−0.0396 × −0.1031−0.0567–
CD14− CD16− ACLIF-R−0.02650.0252 × −0.1090−0.023810.36%
CD8 on NKTTNF−0.0527−0.0325 × −0.0809−0.0553–
Generalized epilepsyCD4 on CD45RA+ CD4+4EBP1−0.0995−0.0295 × −0.3646−0.1102–
Naive-mature B cell %lymphocyteCD50.17980.0445 × 0.24720.16886.12%
CD16-CD56 on NKCD50.10440.0253 × 0.24720.09825.98%
CD4 on CD45RA+ CD4+CD5−0.0995−0.0309 × 0.2472−0.09187.67%
CD14+ CD16+ monocyte %monocyteIL-12B0.1076−0.0456 × −0.02050.09838.69%
TCRgd %T cellTNFSF14−0.08160.0463 × 0.2020−0.0909–
CD4 on CD45RA+ CD4+TNFSF14−0.0995−0.0392 × 0.2020−0.09167.97%

AC, absolute count; CCL, C-C chemokine ligand; CD, cluster of differentiation; DC, dendritic cell; DN, double-negative; EM, effector memory; HLA-DR, human leukocyte antigen-DR isotype; Ig, immunoglobulin; IL, interleukin; LIF-R, leukemia inhibitory factor receptor; NK, natural killer; OPG, osteoprotegerin; TCR, T cell receptor; TNF, tumor necrosis factor; TNFSF, tumor necrosis factor superfamily protein; VEGF_A, vascular endothelial growth factor A; 4EBP1, eukaryotic initiation factor 4E binding protein 1.

The mediating effects of inflammatory proteins were also observed in the analysis of different subtypes of epilepsy. For focal epilepsy, CD127 on CD8br exerted a detrimental effect by decreasing IL-13 (13.6%), while CD14− CD16− AC exerted a protective effect by increasing LIF-R (10.4%). For generalized epilepsy, naive-mature B cell %lymphocyte and CD16−CD56 on NK exerted a detrimental effect by increasing CD5, whereas CD4 on CD45RA+ CD4+ exerted a protective effect by decreasing CD5. Furthermore, CD14+ CD16+ monocyte %monocyte promoted the occurrence of generalized epilepsy by decreasing IL-12B (8.7%), while CD4 on CD45RA+ CD4+ inhibited its occurrence by decreasing tumor necrosis factor superfamily protein 14 (8.0%).

Sensitivity analysis

We excluded immune cells and inflammatory proteins exhibiting horizontal pleiotropy as identified by MR-Egger regression and MR-PRESSO analysis, ensuring that our results were not influenced by horizontal pleiotropy (P > 0.05; File S2: Tables S4, S6, S8, S12). Assessment of between-variant heterogeneity using Cochran’s Q test indicated no substantial heterogeneity across our MR analyses (P > 0.05; File S2: Tables S4, S6, S8, S12). Furthermore, scatter plots, forest plots, funnel plots, and leave-one-out plots were also employed to further demonstrate the reliability of our results (File S1). Concordant causal directions were observed across five complementary MR methods (IVW, weighted median, MR-Egger, weighted mode, simple mode), with detailed results shown in scatter plots. Funnel plots demonstrated no evidence of influential outliers, with all scatter points exhibiting symmetrical distribution patterns around the IVW estimate axis. Sensitivity analyses revealed no outlier SNPs influencing results in leave-one-out testing. Considering the potentially overly stringent nature of Bonferroni correction, we additionally performed a false discovery rate adjustment for multiple testing. Unfortunately, no significant findings remained after correction.

Discussion

In this study, we employed MR analysis to investigate the causal relationship between peripheral immune cells and epilepsy, and to determine whether this relationship is mediated through inflammatory proteins. The results revealed suggestive causal associations between 32 immune phenotypes and epilepsy. Among these immune phenotypes, 13 were associated with reduced epilepsy risk, while 19 were linked to increased risk. Furthermore, we identified potential causal relationships between eight out of 91 investigated inflammatory proteins and epilepsy incidence — four demonstrating positive associations and four showing negative correlations. Mediation analysis demonstrated that five inflammatory proteins mediated the effects of six immune cells on epilepsy, accounting for 6.3% to 13.5% of the total effect.

We found that several phenotypes of CD19-positive B lymphocytes are causally associated with epilepsy and its subtypes, contributing to increased susceptibility to epilepsy. Hansen et al.19,20 also found increased expression of CD19+ B cells in temporal lobe epilepsy caused by autoantibody-negative limbic encephalitis, which correlates with patients’ memory impairments. Furthermore, our mediation analysis identified that within the B lymphocyte phenotype, naive-mature B cell % lymphocyte can promote generalized epilepsy by upregulating CD5. CD5 is a cysteine-rich scavenger receptor family glycoprotein that exhibits constitutive expression across mature T lymphocytes and specific B cell subsets, with the capability to modulate intracellular signaling strength induced by antigen receptors in T and B cells.21 Prayson et al.22 reported the presence of CD5+ T lymphocyte infiltration in the brain tissue of patients with refractory epilepsy caused by Rasmussen’s encephalitis. Our study unveiled that CD5 increases susceptibility to generalized epilepsy, with three types of immune cells (naive-mature B cell % lymphocyte, CD16-CD56 on NK cells, and CD4 on CD45RA+ CD4+ cells) capable of modulating CD5 levels to impact epilepsy risk.

Previous studies have indicated monocyte infiltration in the brain tissue of epilepsy patients, which can promote inflammatory responses and exacerbate neuronal damage.23 Our study found that various monocyte phenotypes can increase the risk of epilepsy, while certain monocyte phenotypes can also decrease this risk. This suggests that monocytes may have dual effects on epilepsy, with the impact depending on monocyte phenotype. Yamanaka et al.24 discovered that levels of intracellular cytokines in peripheral blood monocytes of patients with epilepsy were higher than those in normal control groups, suggesting that inflammatory factors play an intermediary role between monocytes and epilepsy. CCL23 is a chemokine that inhibits the production and release of polymorphonuclear leukocytes and monocytes,25 and participates in inflammatory responses by stimulating the production of adhesion molecules and pro-inflammatory cytokines.26 Several studies have reported the role of CCL23 in the inflammatory response associated with Alzheimer’s disease, cerebral hemorrhage, and acute brain injury.27–29 We found that CCL23 promotes epilepsy, and that CD14+ CD16+ monocyte AC increases epilepsy risk by enhancing CCL23 levels. Leukemia inhibitory factor and ciliary neurotrophic factor are neurotrophic factors that exert their effects via receptor complexes involving the LIF-R, playing crucial roles in neuron survival and maintenance.30 We discovered that LIF-R can decrease susceptibility to epilepsy, and CD14− CD16− AC reduces epilepsy risk by enhancing LIF-R. Rosell et al.31 found a rapid, intense, and transient upregulation of leukemia inhibitory factor in the brain after epileptic seizures, accompanied by high expression of LIF-R in the hippocampus. These changes may relate to neuronal damage repair after seizures. Additionally, within the monocyte phenotype, we also discovered that CD14+ CD16+ monocyte % monocyte promotes the occurrence of generalized epilepsy by reducing IL-12B.

Treg cells can suppress immune responses, and their protective role in epilepsy has been demonstrated.32,33 Similarly, in our study, we found that various Treg cell phenotypes confer protective effects against epilepsy. CD39+ CD4+ % T cells and CD28− CD25++ CD8br AC protect against focal epilepsy; CD39+ CD8br % T cells, CD28− CD8br AC, and CD4 on resting Treg cells protect against generalized epilepsy. Moreover, through mediation analysis, we discovered that OPG can reduce susceptibility to epilepsy and that CD28− CD25++ CD8br AC can upregulate OPG. OPG, together with receptor activator of nuclear factor kappa B (RANK) and RANK ligand (RANKL), constitutes the RANKL/RANK/OPG axis, which is primarily involved in regulating bone metabolism.34 However, research has also reported the neuroprotective effects of OPG after brain injury.35 Additionally, we observed that certain Treg cell subtypes may promote epilepsy, suggesting a dual role of Treg cells in epilepsy. IL-13 is an anti-inflammatory cytokine that can ameliorate neuroinflammation and promote functional recovery after brain injury.36,37 Qi et al.38 found elevated levels of IL-13 in the hippocampus of epileptic rats after vagus nerve stimulation therapy, indicating a protective role of IL-13 in epilepsy. Our findings are consistent with this, and we further discovered that CD127 on CD8br promotes epilepsy by downregulating IL-13. Furthermore, several studies have reported alterations in NK cells, CD4+ T cells, and CD8+ T cells in the peripheral blood of epilepsy patients.39–41 In our study, we also found that the phenotype of NK cells, maturation stages of T cells, and conventional dendritic cells can influence susceptibility to epilepsy.

The correlation between inflammatory factors and immune responses suggests that certain inflammatory factors may play critical roles in the pathophysiological processes through which immune reactions induce epilepsy. In our study, we identified six types of immune cells that influence epileptogenesis by modulating the levels of five inflammatory proteins. However, the mediating effects of these inflammatory proteins were limited (proportion of effect: 6.3–13.5%), indicating that additional pathways may mediate this pathological process. Consistent with our findings, previous studies have reported other mediators bridging immune responses and epileptogenesis. Research by Chen et al.42 suggests that the paraxanthine/linoleate (18:2n6) ratio might act as a mediator in the causal association between CD64+ immune cells (CD14−CD16−) and epilepsy, accounting for 5.05% of the observed mediation effect.

In our research, distinct phenotypes of immune cells, such as monocytes and Tregs, exhibit differential impacts on epilepsy, with certain subsets increasing seizure susceptibility while others confer neuroprotection. This dual functionality of immune cells appears intrinsically linked to their subtype differentiation patterns. For example, classical monocytes (CD14++CD16−) predominantly mediate innate immune responses through expression of chemokines, scavenger receptors, and pro-inflammatory cytokines, facilitating phagocytosis and tissue migration; intermediate monocytes (CD14++CD16+) demonstrate enhanced capacity for antigen processing/presentation and participate in monocyte activation and inflammatory processes; non-classical monocytes (CD14+CD16++) primarily execute vascular surveillance and patrolling functions.43 Furthermore, microenvironmental cues dynamically modulate immune cell functional polarization, with monocyte subsets demonstrating divergent operational plasticity in homeostatic versus pathological states.44 Macrophages exhibit functional plasticity through microenvironment-driven polarization into pro-inflammatory M1 or anti-inflammatory M2 phenotypes.45 Factors including activation states, environmental influences, or the functional differentiation of cell subtypes could play significant roles in modulating their effects on epilepsy. Further investigation into the conditions or contexts under which these immune cell phenotypes might promote or suppress epilepsy will advance the mechanistic understanding of epilepsy immunopathology.

Our study employed distinct immune cell profiles and inflammatory proteins as research targets, revealing their potential causal relationships with epileptogenesis and offering novel perspectives and methodological approaches to elucidate the pathological axis of “immune-inflammation-epilepsy”. Regrettably, current evidence remains predominantly correlational and insufficient to confirm these immune phenotypes and inflammatory proteins as direct etiological factors in epilepsy. Future validation through large-scale clinical trials and longitudinal follow-up studies will be required to better assess the potential clinical utility of these findings in epilepsy diagnosis and therapeutic strategies.

There are certain limitations in this study that should be noted. Firstly, despite incorporating 731 immune cell phenotypes and 91 inflammatory proteins, numerous immune cell phenotypes and inflammatory proteins could not be analyzed due to data insufficiency. Therefore, future studies will need to include additional immune cell phenotypes and inflammatory proteins for a more systematic and comprehensive analysis. Secondly, the number of significant SNPs in GWAS data for immune cells and inflammatory proteins is limited, so we employed a relatively lenient criterion (P < 5 × 10−5) for selecting IVs. Due to the potential reduction in statistical power caused by weak IVs in MR analysis, subsequent studies will need to utilize more robust IVs to validate our conclusions. Thirdly, the data predominantly originate from individuals of European ancestry, thereby precluding direct extrapolation of the conclusions to other racial or ethnic populations. Subsequent studies are needed to conduct MR analyses using GWAS data from other ethnicities to validate the generalizability of our findings. Fourthly, the immunophenotype data (Sardinian cohort) and inflammatory protein data (European cohort) differ in population structure and measurement protocols. Despite allele alignment and linkage disequilibrium pruning to reduce errors, residual heterogeneity may affect causal estimates, requiring validation in homogeneous populations. Finally, while this study utilized MR analysis to infer causal relationships, the conclusions remain largely correlative in nature and do not explore specific molecular mechanisms. Future investigations will employ multi-omics approaches, including snRNA-seq and RNA-seq data, to systematically characterize key molecular pathways and cellular subpopulations.

Conclusions

By integrating conventional and mediation Mendelian randomization approaches, we identified suggestive causal relationships between immune signatures and epilepsy risk, while uncovering mediating roles of specific inflammatory proteins in this pathological cascade. These findings offer novel perspectives and approaches for delineating the “immune-inflammation-epilepsy” pathogenic axis. Future studies should validate these findings across diverse populations and elucidate the molecular mechanisms underlying the identified associations.

Supporting information

Supplementary material for this article is available at https://doi.org/10.14218/NSSS.2025.00010 .

File S1

Supplementary Figures.

(PDF)

Click here for additional data file.

File S2

Supplementary Tables.

(XLSX)

Click here for additional data file.

Declarations

Acknowledgement

We express our gratitude to all individual patients who contributed samples that made the data available, and to all investigators who provided these data to support this study. During the preparation of this manuscript, the authors utilized ChatGPT (GPT-4 architecture) exclusively for linguistic refinement and grammatical optimization of the original text. Crucially, all analytical processes and scientific conclusions were independently formulated and rigorously validated by the research team through conventional academic methodologies.

Ethical statement

Ethical approval was not required, as the analysis was based on publicly available GWAS summary statistics and did not involve access to individual-level data.

Data sharing statement

The data and data processing utilized in this study can be accessed through the first author or corresponding author upon reasonable request.

Funding

This work was supported by the National High Level Hospital Clinical Research Funding (2023-NHLHCRF-YYPP-TS-05); the Elite Medical Professionals Project of China-Japan Friendship Hospital (NO. ZRJY2023-QM25); and the Youth Fund of the National Natural Science Foundation of China (82301470).

Conflict of interest

The authors declare no conflicts of interest.

Authors’ contributions

Study concept and design (HT, JW), acquisition of data (GW, ML), analysis and interpretation of data (GW, ML, XS, HW), drafting of the manuscript (GW, ML, HW), critical revision of the manuscript for important intellectual content (HT, JW, XS), administrative, technical, or material support (HT, JW), and study supervision (HT). All authors have made significant contributions to this study and have approved the final manuscript.

References

  1. Fisher RS, van Emde Boas W, Blume W, Elger C, Genton P, Lee P, et al. Epileptic seizures and epilepsy: definitions proposed by the International League Against Epilepsy (ILAE) and the International Bureau for Epilepsy (IBE). Epilepsia 2005;46(4):470-472 View Article PubMed/NCBI
  2. Thijs RD, Surges R, O’Brien TJ, Sander JW. Epilepsy in adults. Lancet 2019;393(10172):689-701 View Article PubMed/NCBI
  3. Flammer J, Neziraj T, Rüegg S, Pröbstel AK. Immune Mechanisms in Epileptogenesis: Update on Diagnosis and Treatment of Autoimmune Epilepsy Syndromes. Drugs 2023;83(2):135-158 View Article PubMed/NCBI
  4. Husari KS, Dubey D. Autoimmune Epilepsy. Neurotherapeutics 2019;16(3):685-702 View Article PubMed/NCBI
  5. Crespel A, Coubes P, Rousset MC, Brana C, Rougier A, Rondouin G, et al. Inflammatory reactions in human medial temporal lobe epilepsy with hippocampal sclerosis. Brain Res 2002;952(2):159-169 View Article PubMed/NCBI
  6. Arisi GM, Foresti ML, Katki K, Shapiro LA. Increased CCL2, CCL3, CCL5, and IL-1β cytokine concentration in piriform cortex, hippocampus, and neocortex after pilocarpine-induced seizures. J Neuroinflammation 2015;12:129 View Article PubMed/NCBI
  7. Vezzani A, Balosso S, Ravizza T. Neuroinflammatory pathways as treatment targets and biomarkers in epilepsy. Nat Rev Neurol 2019;15(8):459-472 View Article PubMed/NCBI
  8. Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ 2018;362:k601 View Article PubMed/NCBI
  9. Burgess S, Butterworth A, Malarstig A, Thompson SG. Use of Mendelian randomisation to assess potential benefit of clinical intervention. BMJ 2012;345:e7325 View Article PubMed/NCBI
  10. Fang Y, Si X, Wang J, Wang Z, Chen Y, Liu Y, et al. Alzheimer Disease and Epilepsy: A Mendelian Randomization Study. Neurology 2023;101(4):e399-e409 View Article PubMed/NCBI
  11. Guo X, Zhu Y, Ying C, Xu K, Hong Y. The impact of serum magnesium and calcium on the risk of epilepsy: A mendelian randomization study. CNS Neurosci Ther 2023;29(10):3062-3067 View Article PubMed/NCBI
  12. Chen Z, Sun H, Zhang W, Hou S, Yang X, Lin J, et al. Exploring correlations between immune cell phenotypes and the risk of epilepsy: A bidirectional Mendelian randomization study. Epilepsy Behav 2024;157:109896 View Article PubMed/NCBI
  13. Wang K, Yang J, Xu W, Wang L, Wang Y. Interplay between immune cells and metabolites in epilepsy: insights from a Mendelian randomization analysis. Front Aging Neurosci 2024;16:1400426 View Article PubMed/NCBI
  14. Sun H, Ma D, Hou S, Zhang W, Li J, Zhao W, et al. Exploring causal correlations between systemic inflammatory cytokines and epilepsy: A bidirectional Mendelian randomization study. Seizure 2024;114:44-49 View Article PubMed/NCBI
  15. Orrù V, Steri M, Sidore C, Marongiu M, Serra V, Olla S, et al. Complex genetic signatures in immune cells underlie autoimmunity and inform therapy. Nat Genet 2020;52(10):1036-1045 View Article PubMed/NCBI
  16. Zhao JH, Stacey D, Eriksson N, Macdonald-Dunlop E, Hedman ÅK, Kalnapenkis A, et al. Genetics of circulating inflammatory proteins identifies drivers of immune-mediated disease risk and therapeutic targets. Nat Immunol 2023;24(9):1540-1551 View Article PubMed/NCBI
  17. Kurki MI, Karjalainen J, Palta P, Sipilä TP, Kristiansson K, Donner KM, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 2023;613(7944):508-518 View Article PubMed/NCBI
  18. Burgess S, Small DS, Thompson SG. A review of instrumental variable estimators for Mendelian randomization. Stat Methods Med Res 2017;26(5):2333-2355 View Article PubMed/NCBI
  19. Hansen N, Widman G, Önder D, Schwing K, Leelaarporn P, Prusseit I, et al. Increased T- and B-cells associated with the phenotype of autoimmune limbic encephalitis with mainly memory dysfunction. J Transl Autoimmun 2022;5:100167 View Article PubMed/NCBI
  20. Hansen N, Önder D, Schwing K, Widman G, Leelaarporn P, Prusseit I, et al. CD19+ B-cells in autoantibody-negative limbic encephalitis. Epilepsy Behav 2020;106:107016 View Article PubMed/NCBI
  21. Raman C. CD5, an important regulator of lymphocyte selection and immune tolerance. Immunol Res 2002;26(1-3):255-263 View Article PubMed/NCBI
  22. Prayson RA, Frater JL. Rasmussen encephalitis: a clinicopathologic and immunohistochemical study of seven patients. Am J Clin Pathol 2002;117(5):776-782 View Article PubMed/NCBI
  23. Varvel NH, Neher JJ, Bosch A, Wang W, Ransohoff RM, Miller RJ, et al. Infiltrating monocytes promote brain inflammation and exacerbate neuronal damage after status epilepticus. Proc Natl Acad Sci U S A 2016;113(38):E5665-E5674 View Article PubMed/NCBI
  24. Yamanaka G, Takamatsu T, Morichi S, Yamazaki T, Mizoguchi I, Ohno K, et al. Interleukin-1β in peripheral monocytes is associated with seizure frequency in pediatric drug-resistant epilepsy. J Neuroimmunol 2021;352:577475 View Article PubMed/NCBI
  25. Shih CH, van Eeden SF, Goto Y, Hogg JC. CCL23/myeloid progenitor inhibitory factor-1 inhibits production and release of polymorphonuclear leukocytes and monocytes from the bone marrow. Exp Hematol 2005;33(10):1101-1108 View Article PubMed/NCBI
  26. Kim J, Kim YS, Ko J. CK beta 8/CCL23 induces cell migration via the Gi/Go protein/PLC/PKC delta/NF-kappa B and is involved in inflammatory responses. Life Sci 2010;86(9-10):300-308 View Article PubMed/NCBI
  27. Faura J, Bustamante A, Penalba A, Giralt D, Simats A, Martínez-Sáez E, et al. CCL23: A Chemokine Associated with Progression from Mild Cognitive Impairment to Alzheimer’s Disease. J Alzheimers Dis 2020;73(4):1585-1595 View Article PubMed/NCBI
  28. Simats A, García-Berrocoso T, Penalba A, Giralt D, Llovera G, Jiang Y, et al. CCL23: a new CC chemokine involved in human brain damage. J Intern Med 2018;283(5):461-475 View Article PubMed/NCBI
  29. Bonaventura A, Montecucco F. CCL23 is a promising biomarker of injury in patients with ischaemic stroke. J Intern Med 2018;283(5):476-478 View Article PubMed/NCBI
  30. Li M, Sendtner M, Smith A. Essential function of LIF receptor in motor neurons. Nature 1995;378(6558):724-727 View Article PubMed/NCBI
  31. Rosell DR, Nacher J, Akama KT, McEwen BS. Spatiotemporal distribution of gp130 cytokines and their receptors after status epilepticus: comparison with neuronal degeneration and microglial activation. Neuroscience 2003;122(2):329-348 View Article PubMed/NCBI
  32. Xu D, Robinson AP, Ishii T, Duncan DS, Alden TD, Goings GE, et al. Peripherally derived T regulatory and γδ T cells have opposing roles in the pathogenesis of intractable pediatric epilepsy. J Exp Med 2018;215(4):1169-1186 View Article PubMed/NCBI
  33. Yue J, Xu R, Yin C, Yang H, Zhang C, Zhao D. Negative effects of brain regulatory T cells depletion on epilepsy. Prog Neurobiol 2022;217:102335 View Article PubMed/NCBI
  34. Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, et al. Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 1998;93(2):165-176 View Article PubMed/NCBI
  35. Glasnović A, O’Mara N, Kovačić N, Grčević D, Gajović S. RANK/RANKL/OPG Signaling in the Brain: A Systematic Review of the Literature. Front Neurol 2020;11:590480 View Article PubMed/NCBI
  36. Miao W, Zhao Y, Huang Y, Chen D, Luo C, Su W, et al. IL-13 Ameliorates Neuroinflammation and Promotes Functional Recovery after Traumatic Brain Injury. J Immunol 2020;204(6):1486-1498 View Article PubMed/NCBI
  37. Li S, Olde Heuvel F, Rehman R, Aousji O, Froehlich A, Li Z, et al. Interleukin-13 and its receptor are synaptic proteins involved in plasticity and neuroprotection. Nat Commun 2023;14(1):200 View Article PubMed/NCBI
  38. Qi R, Wang M, Zhong Q, Wang L, Yang X, Huang B, et al. Chronic vagus nerve stimulation (VNS) altered IL-6, IL-1β, CXCL-1 and IL-13 levels in the hippocampus of rats with LiCl-pilocarpine-induced epilepsy. Brain Res 2022;1780:147800 View Article PubMed/NCBI
  39. Bauer S, Köller M, Cepok S, Todorova-Rudolph A, Nowak M, Nockher WA, et al. NK and CD4+ T cell changes in blood after seizures in temporal lobe epilepsy. Exp Neurol 2008;211(2):370-377 View Article PubMed/NCBI
  40. Nowak M, Bauer S, Haag A, Cepok S, Todorova-Rudolph A, Tackenberg B, et al. Interictal alterations of cytokines and leukocytes in patients with active epilepsy. Brain Behav Immun 2011;25(3):423-428 View Article PubMed/NCBI
  41. Rosa DV, Rezende VB, Costa BS, Mudado F, Schütze M, Torres KC, et al. Circulating CD4 and CD8 T cells expressing pro-inflammatory cytokines in a cohort of mesial temporal lobe epilepsy patients with hippocampal sclerosis. Epilepsy Res 2016;120:1-6 View Article PubMed/NCBI
  42. Chen J, Yu H, Liu H, Yu H, Liang S, Wu Q, et al. Causal relationship between immune cells and epilepsy mediated by metabolites analyzed through Mendelian randomization. Sci Rep 2024;14(1):19644 View Article PubMed/NCBI
  43. Williams H, Mack C, Baraz R, Marimuthu R, Naralashetty S, Li S, et al. Monocyte Differentiation and Heterogeneity: Inter-Subset and Interindividual Differences. Int J Mol Sci 2023;24(10):8757 View Article PubMed/NCBI
  44. Wong KL, Yeap WH, Tai JJ, Ong SM, Dang TM, Wong SC. The three human monocyte subsets: implications for health and disease. Immunol Res 2012;53(1-3):41-57 View Article PubMed/NCBI
  45. Wculek SK, Dunphy G, Heras-Murillo I, Mastrangelo A, Sancho D. Metabolism of tissue macrophages in homeostasis and pathology. Cell Mol Immunol 2022;19(3):384-408 View Article PubMed/NCBI
Copyright © 2025 Authors. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial 4.0 License (CC BY-NC 4.0), permitting all non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

About this Article

Cite this article
Wu G, Liu M, Wang H, Shao X, Weng J, Tian H. Immune Cell Phenotypes, Inflammatory Proteins and Epilepsy: A Mendelian Randomization Study to Investigate Causal Pathways. Neurosurgical Subspecialties. Published online: Jun 17, 2025. doi: 10.14218/NSSS.2025.00010.
Copy        Export to RIS        Export to EndNote
Article History
Received Revised Accepted Published
March 2, 2025 May 8, 2025 May 20, 2025 June 17, 2025
DOI http://dx.doi.org/10.14218/NSSS.2025.00010
  • Neurosurgical Subspecialties
  • pISSN 0000-0000
  • eISSN 3067-6150
  • © 2025 Xia & He Publishing Inc.
  • Total visits : 3660422
  • Visits today : 1848
Back to Top

Immune Cell Phenotypes, Inflammatory Proteins and Epilepsy: A Mendelian Randomization Study to Investigate Causal Pathways

Guizhi Wu, Mengen Liu, Hongdong Wang, Xu Shao, Jiancong Weng, Hong Tian
  • Reset Zoom
  • Download TIFF