Mendelian Randomization Analysis of the Causal Relationship Between Immune Cells and Epilepsy: The Mediating Role of Metabolites

doi:10.21203/rs.3.rs-4336289/v1

Our study investigated the causal relationship between immune cells, metabolites, and epilepsy using two-sample Mendelian Randomization (MR) and mediation MR analysis of 731 immune cell traits and 1,400 metabolites. Our core methodology centered on inverse-variance weighted MR, supplemented by other methods. This approach was crucial in clarifying the potential intermediary functions of metabolites in the genetic links between traits of immune cells and epilepsy. We found a causal relationship between immune cells and epilepsy. Specifically, the genetically predicted levels of CD64 on CD14-CD16 are positively correlated with the risk of epilepsy (p < 0.001, OR = 1.0826, 95% CI 1.0361–1.1312). Similarly, metabolites also exhibit a causal relationship with both immune cells (OR = 1.0438, 95% CI:1.0087–1.0801, p = 0.0140) and epilepsy (p = 0.0334, OR = 1.0897, 95% CI: 1.0068–1.1795), and sensitivity analysis was conducted to further validate these relationships. Importantly, our intermediate MR results suggest that the metabolite Paraxanthine to linoleate (18:2n6) ratio may mediate the causal relationship between immune cell CD64 on CD14-CD16 and epilepsy, with a mediation effect of 5.05%. The results suggest the importance of specific immune cell levels and metabolites in understanding epilepsy's pathogenesis. This is significant for understanding the pathogenesis of epilepsy and its prevention and treatment.

Biological sciences/Genetics

Biological sciences/Immunology

Biological sciences/Neuroscience

Epilepsy

Mendelian Randomization

Immune cells

Metabolites

Epilepsy, a chronic neurologic condition, is marked by atypical electrical brain activity that results in recurrent seizures, known as spontaneous recurrent seizures (SRS)¹. Clinical manifestations of epilepsy include convulsions, loss of consciousness, myoclonus (sudden muscle jerks), aura, automatisms, hypotonia (decreased muscle tone), altered sensations, and prolonged muscle contractions^2–4. Per the 2017 definition from the International League Against Epilepsy (ILAE), epilepsy is categorized as generalized epilepsy, focal epilepsy, combined generalized and focal epilepsy, and unknown epilepsy⁵. Epilepsy affects people of all ages, regions, and races, although it is more commonly diagnosed in children and adolescents. Its prevalence worldwide ranges from 0.5–1.0%⁵. Various factors can contribute to the development of epilepsy, including genetic predisposition, acquired brain injuries, infections, and metabolic or immune disorders². Despite these known factors, the precise cause of epilepsy remains unclear.

In the pathogenesis of epilepsy, inflammation, and immune responses are vital factors^6–8. During epileptic seizures, the secretion of inflammatory mediators and the activation of immune cells can lead to excessive neuronal excitation and abnormal synaptic connectivity^9–11. A complex interaction exists between immune cells and neurons, known as neuro-immune modulation¹². Immune cells can release various cytokines and chemokines, which may affect neuronal excitability and stability, thereby influencing epileptic seizures^8,13. However, the existence of a direct causal relationship between epilepsy and immune cells remains uncertain. Thus, a study design that minimizes biases is necessary to elucidate this potential causality. Moreover, given the intricate relationship between immune cells and metabolites^14–16, and findings that certain metabolites, such as circulating metabolites¹⁷, can serve as biomarkers for epilepsy, it is conceivable that metabolites may act as potential mediators between immune cells and epilepsy.

Mendelian Randomization (MR) is a research method grounded in Mendel's Second Law^18–20. It utilizes the random allocation of alleles from parents to offspring during the creation of gametes, facilitating the establishment of links between traits of exposure, such as intermediate phenotypes, and outcomes of diseases^18–20. This method employs genetic variation as a key variable, serving as an instrument to deduce the impact of exposure factors on outcomes based on observational data¹⁸. The inherent randomness of allele distribution in Mendelian Randomization naturally minimizes non-random errors or misleading factors, and prevents causal inversion through the principles of Mendelian inheritance¹⁸. Therefore, MR provides a reliable mechanism for drawing causal inferences from observational data. Thus, our objectives were to (i) ascertain if there is a causal link between immune cells and epilepsy, and (ii) evaluate the degree to which metabolites mediate the effects of immune cells on epilepsy.

Study design

Our analysis was founded on publicly accessible data that have been previously approved by the institutional review boards of the corresponding studies. This meant that no further approvals were necessary for our work. Detailed descriptions of the results from our analysis can be found in both the main article and the supplementary materials accompanying it.

To ensure dependable outcomes, a two-sample MR study must satisfy three fundamental assumptions: (1) The instrumental variables (IVs) ought to exhibit a notable correlation with immune cells; (2) IVs should remain uncorrelated with any confounding variables capable of impacting the connection between exposure and outcome factors; (3) IVs should solely influence outcomes via the pathway of immune cells.

Our study initially utilized a two-sample MR to ascertain the causal association between immune cells, metabolites, and epilepsy. Subsequently, to further investigate the potential impact of immune cells on epilepsy mediated through metabolites, we employed a mediating MR approach, also known as a two-step MR.

Data sources

The immune cell exposure data is sourced from OpenGWAS, which includes 126 billion genetic associations from 14,582 complete GWAS datasets²¹. The intermediary (metabolite) data comes from GWAS Catalog²². The outcome (epilepsy) data is derived from FinnGen GWAS²³ R10, including 12,891 cases and 312,803 controls, with all individuals being of European descent.

Selection for genetic variation

For immune cells to outcome, immune cells to metabolites, and metabolites to outcome, we selected instrumental variables (IVs) with a p-value threshold of < 1.0×10^-5. To ensure independent loci for the IVs, we used the "TwoSampleMR" package, setting the linkage disequilibrium (LD) threshold for the 1000 Genomes EUR data to R^2 < 0.001, with a clumping distance of 10,000 kb.

For the reverse direction (epilepsy to immune cells), IVs were chosen with a p-value threshold of < 5.0×10^-6. Similarly, to obtain independent loci for the IVs, we applied the "TwoSampleMR" package with the LD threshold for the 1000 Genomes EUR data set to R^2 < 0.001, and a clumping distance of 10,000 kb. The results revealed no evidence of reverse causation. For MR analysis, palindromic and ambiguous SNPs were omitted from IVs ²⁴. Furthermore, we computed the F statistic (i.e., the proportion of variability attributed to Single Nucleotide Polymorphisms in each exposure), represented as \(F=\left(\frac{N-K-1}{K}\right)\left(\frac{{R}^{2}}{1-{R}^{2}}\right)\), where K stands for the count of genetic variants and N denotes the sample size²⁵. We excluded instrumental variables with low strength (F-statistics < 10) ²⁵.

Statistical analysis

Although the primary emphasis of our research was on outcomes using the inverse variance weighting (IVW) method²⁶, it was crucial to confirm that the directions of the results were consistent across all methods used, including IVW. For a thorough secondary analysis, we applied various methods like MR-Egger regression, weighted median, simple mode, weighted mode, and MR-Pleiotropy Residual Sum and Outlier (MR-PRESSO)^20,27–29, to conduct a sensitivity check of our IVW results.

Pleiotropy and heterogeneity analysis

Our study began with employing the MR-PRESSO²⁹ method to pinpoint and remove outliers. We then reanalyzed the data without these outliers. Next, to investigate the influence of each Single Nucleotide Polymorphism (SNP) on the link between exposure and outcome, we executed a leave-one-out sensitivity analysis, systematically omitting one SNP at a time. Additionally, the MR-Egger regression test was applied to detect horizontal pleiotropy in the Mendelian randomization analysis, particularly focusing on the importance of the intercept term's statistical significance ³⁰. To explore heterogeneity within our findings, we calculated the Cochran Q statistic, considering P = 0.05 as the significance threshold³¹. All these statistical procedures were performed using R (version 4.2.1), utilizing the MR and MR-PRESSO packages.

1.Association of immune cells with epilepsy

We first screened for exposure tools related to the outcome (epilepsy) from 731 immune cells. After excluding palindromic SNPs and those identified by MR Steiger filtering as having incorrect causal directions, there are a total of 34 qualified immune cells (Table 1). We then conducted analyses using MR Egger, Weighted median, IVW, Simple mode, and Weighted mode, and illustrated these in forest plots(Supplementary Fig. 1). Here, we selected the immune cell subgroup CD64 on CD14-CD16 (id ebi-a-GCST90002001) as our exposure tool variable. In analyses such as MR Egger, Weighted median, IVW, Simple mode, and Weighted mode, we looked for results with p-values below the commonly used threshold for statistical significance (e.g., 0.05). In this case, the IVW method had the lowest p-value at 0.0004, indicating statistical significance in this analysis.

We examined the odds ratio (OR) and the 95% confidence interval (CI). In the IVW method, the OR was 1.0826, with a 95% CI of 1.0361 to 1.1312, indicating not only a positive direction of effect but also that the confidence interval does not include 1, suggesting an association between CD64 on CD14-CD16 and the outcome variable. Additionally, the OR remained above 1 across different estimation methods, indicating consistent result directions across methods, i.e., the exposure increases the risk of epilepsy. To validate this, we also conducted sensitivity analysis (Fig. 1a). Moreover, our MR analysis results showed no evidence of reverse causation between CD64 on CD14-CD16 and the outcome (epilepsy).

2.Association of metabolites with epilepsy

From 1,400 metabolites, we identified exposure tools related to the outcome (epilepsy). After excluding palindromic SNPs and those identified by MR Steiger filtering as having incorrect causal directions, there are a total of 75 qualified metabolites(Table 2). We then performed analyses using MR Egger, Weighted median, IVW, Simple mode, and Weighted mode, and presented these in forest plots༈Supplementary Fig. 2༉. We chose the ratio of Paraxanthine to linoleate (18:2n6) (id GCST90200982) as the exposure tool for the outcome of epilepsy. In all methods, the IVW's p-value was the lowest (0.0334), indicating statistical significance. Across different estimation methods, the OR was consistently above 1, indicating a consistent association between this ratio and an increased risk of epilepsy. Although some methods had confidence intervals close to 1, the IVW method provided a sufficiently wide confidence interval (1.0068–1.1795), along with a narrower interval, suggesting higher precision and adding credibility to the result. We further performed a sensitivity analysis to verify this result (Fig. 1b).

3.Selected relationship between immune cells and positive metabolites.

The results show that the selected immune cell subgroup CD64 on CD14-CD16 (OR = 1.0438, 95% CI:1.0087–1.0801, p = 0.0140) influences the aforementioned positive metabolite Paraxanthine to linoleate (18:2n6) ratio, as demonstrated by the IVW method (Fig. 2a). Since we have verified the connections between "immune cells → epilepsy" and "immune cells → metabolites," we hypothesize that certain metabolites, such as the Paraxanthine to linoleate (18:2n6) ratio, may play a modulating role in the relationship between immune cells and epilepsy. We additionally performed a sensitivity analysis to reinforce this finding (Fig. 1c).

4. The intermediary role of metabolites in the causal relationship between immune cells and epilepsy.

A two-step method was used to analyze the mediating role of the metabolite Paraxanthine to linoleate (18:2n6) ratio in the causal relationship between the immune cell subgroup CD64 on CD14-CD16 and epilepsy. As shown in Fig. 2b, the metabolite accounts for 5.05% of the effect. This indicates that the metabolite may play a certain mediating role in the connection between immune cells and epilepsy, but this role is limited.

In this research, we employed MR analysis to investigate the causal link between immune cells and epilepsy, and to investigate whether this relationship is mediated through metabolites. Our findings suggest a causative link between genetically forecasted levels of immune cells (specifically CD64 on CD14-CD16) and increased risk of epilepsy, indicating that an increase of one standard deviation in CD64 on CD14-CD16 levels is associated with an approximate 8.26% increase in epilepsy risk under constant conditions. Notably, about 5.05% of this effect appears to be mediated through the metabolite Paraxanthine to linoleate (18:2n6) ratio.

Studies show that inflammation is a major factor in epilepsy^32,33. For instance, a study discussed the role of P2X7 receptors in epilepsy and neuroinflammation, particularly their expression on various cells in the brain and their involvement in the inflammatory process during epileptic seizures¹³. Other studies³⁴ have demonstrated that tumor necrosis factor-alpha inhibits epileptic seizures through specific receptors, while interleukin-1 receptor antagonists have been found to exhibit antiepileptic effects in mice. Overexpression of cytokines in the brain, such as interleukin-6, is associated with the induction of neurological disorders in animal models. This suggests a close link between immune responses and neural health. The migration of immune cells to the central nervous system and their role in neurological disorders is also a significant area of research. Additionally, studies³⁵ have explored the role of the blood-brain barrier in immune functions and dysfunctions, which is crucial for understanding how peripheral immune responses affect the central nervous system. These studies collectively emphasize the intricate interplay between the immune system and neurological disorders, especially epilepsy.

When immune cell activity changes, different metabolites may be released, which can influence the risk of epilepsy by affecting the function of neural cells. For example, certain metabolites might alter neuronal excitability or affect neural conduction pathways³⁶. A study utilized ¹H-NMR and DI/LC-MS/MS technologies to identify and quantify 212 metabolites³⁷. The results showed that among these 212 detected metabolites, 14 exhibited significant concentration changes between epilepsy patients and the control group (p < 0.05, q < 0.05) ³⁷.Therefore, by studying these changes in metabolites, scientists can better understand the connection between the immune system and the onset of epilepsy, and how modulating these metabolic pathways might reduce the risk of epilepsy. This study found that the Paraxanthine to linoleate (18:2n6) ratio may play a certain mediating role in the relationship between immune cells and epilepsy. This suggests that this metabolite might be a key factor linking changes in immune activity and epilepsy risk. The variation in this ratio may reflect the activation state of immune cells, impacting the function of the nervous system. For instance, if changes in this ratio lead to increased excitability of neural cells or altered neural communication pathways, this might increase the risk of epilepsy. Therefore, changes in the Paraxanthine to linoleate (18:2n6) ratio not only reflect the state of the immune system but may also be an important biomarker for predicting or intervening in epilepsy risk. This discovery underscores the potential importance of metabolites in neuroimmunology and disease prevention.

In our research, we were the first to employ mediation MR technology to explore the causal association among metabolites, immune cells, and epilepsy. We employed multiple typical sensitivity analyses and excluded the likelihood of reverse causation. Our initial results indicate a causal link between immune cells and epilepsy and its intermediaries, further supporting the theoretical basis for treating and preventing epilepsy and proposing new methods for its management. For example, the risk of epilepsy onset can be managed by initially regulating specific metabolites through diet, exercise, or other methods.

While our study includes several advantages, such as a substantial sample size and the utilization of various sensitivity analysis techniques to ensure the credibility of our research and mitigate confounding variables, it is essential to acknowledge the presence of unavoidable limitations. Firstly, although the sample size is substantial, participants are entirely of European descent, potentially limiting considerations of genetic diversity and environmental influences. Secondly, despite our efforts to identify and remove outlier variables, we cannot completely exclude the possibility of horizontal pleiotropy influencing our findings. Thirdly, while our results suggest a potential mediation effect, the fact that the p-value is greater than 0.05 and the confidence interval includes negative numbers may be due to the presence of unknown moderating variables. Perhaps future research could focus on including participants from a more diverse range of ethnicities and regions, expanding the sample size, and conducting experiments at the cellular level, with animal models, as well as clinical trials. Advanced analytical methods or stricter statistical controls could be used to reduce the impact of potential biases. Additionally, the robustness of research results could be further validated by using different statistical models or tools.

Our study has established a causal link between immune cells and epilepsy, with specific levels of immune cells (such as CD64 on CD14-CD16) being associated with an increased risk of epilepsy. Additionally, the ratio of Paraxanthine to linoleate (18:2n6) appears to exert a moderating influence on this relationship, although its impact is limited. Subsequent investigations should prioritize larger sample sizes, deeper exploration of mechanisms, controlling potential confounding factors, and conducting comparative studies across different populations to validate and extend the current findings. This is significant for understanding the pathogenesis of epilepsy and its prevention and treatment.

Abbreviations	Full form
MR	Mendelian Randomization
SRS	Spontaneous Recurrent Seizures
ILAE	International League Against Epilepsy
IVs	Instrumental Variables
LD	Linkage Disequilibrium
IVW	Inverse Variance Weighting
SNP	Single Nucleotide Polymorphism
OR	Odds Ratio
CI	Confidence Interval

Acknowledgments

The authors thank all the researchers for sharing their data.

Authors' contributions

L.D. conceived, initiated, and supervised the project. J.C. wrote a draft of the manuscript. H.Y., H.L.,H.Y.,S.L., Q.W., X.Z.and R.Z.collected and analyzed the data. The authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests

Availability of data and materials

The data supporting the findings of this study are openly available on the following platforms: the IEU OpenGWAS Project (https://gwas.mrcieu.ac.uk), the GWAS Catalog (https://www.ebi.ac.uk/gwas/), and the FinnGen GWAS R10 version (https://www.finngen.fi/en/access_results). Additional data can be found in the article and its supplementary materials. For further inquiries, please contact the corresponding authors.

Funding

This work was supported by the China National Natural Sciences Foundation (Grant No. 82360875), China National Natural Sciences Foundation (Grant No. 81960858) and Guangxi Science and Technology Base and talent Project (Grant No. 2021AC18028).

Fisher R S, Boas W van E, 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)[J]. Epilepsia, 2005, 46(4): 470–472.
Scheffer I E, Berkovic S, Capovilla G, Connolly M B, French J, Guilhoto L, et al. ILAE classification of the epilepsies: Position paper of the ILAE Commission for Classification and Terminology[J]. Epilepsia, 2017, 58(4): 512–521.
Yin F, Ni D, Xu C, Yan X, Ma K, Zhang X, et al. Auras in intractable frontal lobe epilepsy: Clinical characteristics, values, and limitations[J]. Epilepsy & Behavior, 2021, 115: 107724.
Ahmedt-Aristizabal D, Sarfraz M S, Denman S, Nguyen K, Fookes C, Dionisio S, et al. Motion Signatures for the Analysis of Seizure Evolution in Epilepsy[A]. 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)[C]. Berlin, Germany: IEEE, 2019: 2099–2105.
Fiest K M, Sauro K M, Wiebe S, Patten S B, Kwon C-S, Dykeman J, et al. Prevalence and incidence of epilepsy[J]. Neurology, Wolters Kluwer, 2017, 88(3): 296–303.
Kumar P, Shih D C W, Lim A, Paleja B, Ling S, Lai L Y, et al. Proinflammatory IL-17 pathways dominate the architecture of the immunome in pediatric refractory epilepsy[J]. JCI Insight, 2019, 4(8): e126337.
Langenbruch L, Bleß L, Schulte-Mecklenbeck A, Sundermann B, Brix T, Elger C E, et al. Blood and cerebrospinal fluid immune cell profiles in patients with temporal lobe epilepsy of different etiologies[J]. Epilepsia, 2020, 61(10).
Huang Y, Wang Q, Liu X, Du W, Hao Z, Wang Y. Transcriptional Signatures of a Dynamic Epilepsy Process Reveal Potential Immune Regulation[J]. Molecular Neurobiology, 2023. doi: 10.1007/s12035-023-03786-x
Vieira É L M, de Oliveira G N M, Lessa J M K, Gonçalves A P, Oliveira A C P, Bauer M E, et al. Peripheral leukocyte profile in people with temporal lobe epilepsy reflects the associated proinflammatory state[J]. Brain, Behavior, and Immunity, 2016, 53: 123–130.
Vezzani A, Maroso M, Balosso S, Sanchez M-A, Bartfai T. IL-1 receptor/Toll-like receptor signaling in infection, inflammation, stress and neurodegeneration couples hyperexcitability and seizures[J]. Brain, Behavior, and Immunity, 2011, 25(7): 1281–1289.
Vezzani A, Friedman A, Dingledine R J. The role of inflammation in epileptogenesis[J]. Neuropharmacology, 2013, 69: 16–24.
Forsythe P. Mast Cells in Neuroimmune Interactions[J]. Trends in Neurosciences, 2019, 42(1): 43–55.
Meng F, Yao L. The role of inflammation in epileptogenesis[J]. Acta Epileptologica, 2020, 2(1): 15.
Wu J, Meng F, Ran D, Song Y, Dang Y, Lai F, et al. The Metabolism and Immune Environment in Diffuse Large B-Cell Lymphoma[J]. Metabolites, 2023, 13(6): 734.
Weiss H J, Angiari S. Metabolite Transporters as Regulators of Immunity[J]. Metabolites, 2020, 10(10): 418.
Choi C, Finlay D K. Diverse Immunoregulatory Roles of Oxysterols—The Oxidized Cholesterol Metabolites[J]. Metabolites, 2020, 10(10): 384.
Godoi A B, do Canto A M, Donatti A, Rosa D C, Bruno D C F, Alvim M K, et al. Circulating Metabolites as Biomarkers of Disease in Patients with Mesial Temporal Lobe Epilepsy[J]. Metabolites, 2022, 12(5): 446.
Birney E. Mendelian Randomization[J]. Cold Spring Harbor Perspectives in Medicine, 2021: a041302.
Davies N M, Holmes M V, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians[J]. BMJ, 2018: k601.
Pierce B L, Burgess S. Efficient Design for Mendelian Randomization Studies: Subsample and 2-Sample Instrumental Variable Estimators[J]. American Journal of Epidemiology, 2013, 178(7): 1177–1184.
Elsworth B, Lyon M, Alexander T, Liu Y, Matthews P, Hallett J, et al. The MRC IEU OpenGWAS data infrastructure[R]. Genetics, 2020. doi: 10.1101/2020.08.10.244293.
Buniello A, MacArthur J A L, Cerezo M, Harris L W, Hayhurst J, Malangone C, et al. The NHGRI-EBI GWAS Catalog of published genome-wide association studies, targeted arrays and summary statistics 2019[J]. Nucleic Acids Research, 2019, 47(Database issue): D1005–D1012.
Kurki M I, Karjalainen J, Palta P, Sipilä T P, Kristiansson K, Donner K M, et al. FinnGen provides genetic insights from a well-phenotyped isolated population[J]. Nature, 2023, 613(7944): 508–518.
Hemani G, Zheng J, Elsworth B, Wade K H, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome[J]. eLife, 7: e34408.
Burgess S, Thompson S G, CRP CHD Genetics Collaboration. Avoiding bias from weak instruments in Mendelian randomization studies[J]. International Journal of Epidemiology, 2011, 40(3): 755–764.
Palmer T M, Lawlor D A, Harbord R M, Sheehan N A, Tobias G H, Timpson N J, et al. Using multiple genetic variants as instrumental variables for modifiable risk factors[J]. Statistical Methods in Medical Research, 2012, 21(3): 223–242.
Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression[J]. International Journal of Epidemiology, 2015, 44(2): 512–525.
Bowden J, Davey Smith G, Haycock P C, Burgess S. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator[J]. Genetic Epidemiology, 2016, 40(4): 304–314.
Verbanck M, Chen C-Y, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases[J]. Nature genetics, 2018, 50(5): 693–698.
Burgess S, Thompson S G. Erratum to: Interpreting findings from Mendelian randomization using the MR-Egger method[J]. European Journal of Epidemiology, 2017, 32(5): 391–392.
Egger M, Smith G D, Phillips A N. Meta-analysis: principles and procedures.[J]. BMJ : British Medical Journal, 1997, 315(7121): 1533–1537.
Vezzani A, Balosso S, Ravizza T. Neuroinflammatory pathways as treatment targets and biomarkers in epilepsy[J]. Nature Reviews Neurology, 2019, 15(8): 459–472.
Wang L, Duan C, Wang R, Chen L, Wang Y. Inflammation-related genes and immune infiltration landscape identified in kainite-induced temporal lobe epilepsy based on integrated bioinformatics analysis[J]. Frontiers in Neuroscience, 2022, 16: 996368.
Brebner K, Hayley S, Zacharko R, Merali Z, Anisman H. Synergistic Effects of Interleukin-1β, Interleukin-6, and Tumor Necrosis Factor-α Central Monoamine, Corticosterone, and Behavioral Variations[J]. Neuropsychopharmacology, 2000, 22(6): 566–580.
Passaro A P, Lebos A L, Yao Y, Stice S L. Immune Response in Neurological Pathology: Emerging Role of Central and Peripheral Immune Crosstalk[J]. Frontiers in Immunology, 2021, 12: 676621.
Suleymanova E M. Behavioral comorbidities of epilepsy and neuroinflammation: Evidence from experimental and clinical studies[J]. Epilepsy & Behavior, 2021, 117: 107869.
Lalwani A M, Yilmaz A, Bisgin H, Ugur Z, Akyol S, Graham S F. The Biochemical Profile of Post-Mortem Brain from People Who Suffered from Epilepsy Reveals Novel Insights into the Etiopathogenesis of the Disease[J]. Metabolites, 2020, 10(6): 261.

Table 1 Genetic Association Analysis of Epilepsy with Immune Cell Types and Markers.

id	pvalue	Types or Markers of Immune Cells
ebi-a-GCST90001396	0.008878046	IgD+ CD38- AC
ebi-a-GCST90001531	0.00517134	CD33dim HLA DR- AC
ebi-a-GCST90001532	0.006606124	Basophil AC
ebi-a-GCST90001541	0.044654038	Naive CD4+ %CD4+
ebi-a-GCST90001551	0.004667347	Naive CD8br AC
ebi-a-GCST90001563	0.00216603	CM DN (CD4-CD8-) AC
ebi-a-GCST90001570	0.031477456	EM DN (CD4-CD8-) %DN
ebi-a-GCST90001580	0.027067671	CD14+ CD16+ monocyte AC
ebi-a-GCST90001581	0.038268065	CD14- CD16- AC
ebi-a-GCST90001587	0.031598029	CD16+ monocyte %monocyte
ebi-a-GCST90001626	0.008344311	HLA DR+ CD4+ %lymphocyte
ebi-a-GCST90001644	0.021711251	B cell %lymphocyte
ebi-a-GCST90001645	0.020119915	NK AC
ebi-a-GCST90001651	0.043506543	Granulocyte AC
ebi-a-GCST90001658	0.011535717	CD39+ CD4+ %T cell
ebi-a-GCST90001669	0.005021942	CD28+ CD45RA- CD8dim AC
ebi-a-GCST90001678	0.020511711	CD28- CD25++ CD8br AC
ebi-a-GCST90001726	0.040757543	CD19 on IgD+ CD38-
ebi-a-GCST90001730	0.030657637	CD19 on IgD+ CD38dim
ebi-a-GCST90001741	0.042287047	CD19 on IgD+
ebi-a-GCST90001783	0.017714312	CD25 on IgD+ CD38br
ebi-a-GCST90001787	0.026205839	CD25 on IgD- CD38-
ebi-a-GCST90001788	0.002511127	CD25 on IgD- CD38br
ebi-a-GCST90001899	0.043891698	CD28 on CD4 Treg
ebi-a-GCST90001926	0.03128246	CD127 on granulocyte
ebi-a-GCST90001934	0.005542075	CD25 on CD45RA+ CD4 not Treg
ebi-a-GCST90001976	0.014318799	FSC-A on NKT
ebi-a-GCST90002001	0.00039017	CD64 on CD14- CD16-
ebi-a-GCST90002064	0.020624656	CD4 on resting Treg
ebi-a-GCST90002069	0.049771099	CD4 on CD39+ secreting Treg
ebi-a-GCST90002105	0.025971632	HLA DR on plasmacytoid DC
ebi-a-GCST90002106	0.004962448	HLA DR on DC
ebi-a-GCST90002108	0.003396284	HLA DR on CD33br HLA DR+ CD14-
ebi-a-GCST90002109	0.017140626	HLA DR on CD33br HLA DR+ CD14dim

Table 2 Genetic Association Analysis of Epilepsy and Metabolic Levels.

id	pvalue	Reported Trait
GCST90199637	0.035004621	N-acetylglutamate levels
GCST90199644	0.000457545	Theobromine levels
GCST90199653	0.04408322	1-myristoyl-2-palmitoyl-gpc (14:0/16:0) levels
GCST90199681	0.021563403	2-hydroxyoctanoate levels
GCST90199683	0.009643237	Indoleacetate levels
GCST90199700	0.006991842	3-hydroxylaurate levels
GCST90199711	0.03945197	Docosadienoate (22:2n6) levels
GCST90199723	0.011948843	5-dodecenoate (12:1n7) levels
GCST90199744	0.010790532	Eicosenoate (20:1) levels
GCST90199747	0.023626463	Isovalerylcarnitine (C5) levels
GCST90199753	0.029679753	Gamma-glutamyltryptophan levels
GCST90199763	0.004365508	1-methylxanthine levels
GCST90199823	0.004535746	O-cresol sulfate levels
GCST90199832	0.019507094	Dimethylarginine (sdma + adma) levels
GCST90199843	0.045989215	Chiro-inositol levels
GCST90199847	0.01794526	4-allylphenol sulfate levels
GCST90199848	0.021558422	Succinylcarnitine levels
GCST90199859	0.003985094	"5alpha-androstan-3alpha,17beta-diol disulfate levels"
GCST90199960	0.006826377	N-formylanthranilic acid levels
GCST90199972	0.043217479	"Sphingomyelin (d18:1/24:1, d18:2/24:0) levels"
GCST90200014	0.048344808	3beta-hydroxy-5-cholestenoate levels
GCST90200032	0.011255736	2-hydroxybutyrate/2-hydroxyisobutyrate levels
GCST90200053	0.024210085	"Sphingomyelin (d18:1/21:0, d17:1/22:0, d16:1/23:0) levels"
GCST90200072	0.049970989	Thioproline levels
GCST90200092	0.045776627	Glycocholate glucuronide (1) levels
GCST90200094	0.036420663	Ascorbic acid 2-sulfate levels
GCST90200099	0.034174041	N-oleoylserine levels
GCST90200107	0.028807959	Gamma-glutamyl-alpha-lysine levels
GCST90200115	0.000513124	N-palmitoyl-heptadecasphingosine (d17:1/16:0) levels
GCST90200131	0.048118224	"Sphingomyelin (d18:1/19:0, d19:1/18:0) levels"
GCST90200175	0.020203189	Glycine conjugate of C10H14O2 (1) levels
GCST90200203	0.031194186	Indoleacetoylcarnitine levels
GCST90200212	0.048093597	Glucuronide of piperine metabolite C17H21NO3 (5) levels
GCST90200244	0.012916584	5-hydroxymethyl-2-furoylcarnitine levels
GCST90200311	0.036126849	4-acetamidobutanoate levels
GCST90200319	0.032188976	"Cys-gly, oxidized levels"
GCST90200337	0.048487501	Taurochenodeoxycholate levels
GCST90200338	0.044597138	4-hydroxyphenylacetate levels
GCST90200367	0.006025932	3-Hydroxybutyrate levels
GCST90200369	0.026651347	2-hydroxyphenylacetate levels
GCST90200377	0.00511502	Histidine levels
GCST90200402 GCST90200420	0.043643867 0.010843509	Lysine levels Taurine levels
GCST90200431	0.00895426	Alanine levels
GCST90200433	0.019690958	Cytosine levels
GCST90200473	0.015151642	X-11795 levels
GCST90200506	0.002750803	X-12812 levels
GCST90200539	0.040536944	X-14939 levels
GCST90200558	0.02635407	X-18888 levels
GCST90200648	0.037850225	X-24585 levels
GCST90200655	0.025381338	X-25828 levels
GCST90200663	0.038075027	X-25420 levels
GCST90200669	0.049614027	X-25519 levels
GCST90200673	0.014707482	Carnitine C4 levels
GCST90200676	0.039475199	N-acetyl-L-glutamine levels
GCST90200680	0.002303833	5-acetylamino-6-formylamino-3-methyluracil levels
GCST90200717	0.036735217	3-phosphoglycerate to adenosine 5'-diphosphate (ADP) ratio
GCST90200724	0.021637868	Cholate to taurocholate ratio
GCST90200737	0.014854875	Adenosine 5'-diphosphate (ADP) to 2'-deoxyuridine ratio
GCST90200746	0.025084551	Adenosine 5'-diphosphate (ADP) to gluconate ratio
GCST90200760	0.046036017	Methionine to methionine sulfoxide ratio
GCST90200763	0.042195841	Phosphate to N-acetylneuraminate ratio
GCST90200765	0.044466635	Phosphate to alanine ratio
GCST90200780	0.02403292	Alanine to pyruvate ratio
GCST90200782	0.02219789	Mannose to trans-4-hydroxyproline ratio
GCST90200807	0.011330952	Carnitine to ergothioneine ratio
GCST90200903	0.049106155	Phosphate to glycerol ratio
GCST90200914	0.031096816	Tyrosine to pyruvate ratio
GCST90200919	0.012454706	Caffeine to theophylline ratio
GCST90200929	0.039771623	Glycerol to carnitine ratio
GCST90200982	0.033377915	Paraxanthine to linoleate (18:2n6) ratio
GCST90200984	0.016856574	Cholesterol to benzoate ratio
GCST90200987	0.039659483	Benzoate to oleoyl-linoleoyl-glycerol (18:1 to 18:2) [2] ratio
GCST90200997	0.039762413	Histidine to glutamine ratio
GCST90201002	0.012976684	Maltose to sucrose ratio

No competing interests reported.

Mendelian Randomization Analysis of the Causal Relationship Between Immune Cells and Epilepsy: The Mediating Role of Metabolites

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Methods

Study design

Data sources

Selection for genetic variation

Statistical analysis

Pleiotropy and heterogeneity analysis

Results

1.Association of immune cells with epilepsy

2.Association of metabolites with epilepsy

Discussion

Conclusion

Abbreviations

Declarations

References

Tables

Additional Declarations

Status:

Journal Publication

Version 1