Causal relationship between oral microbiota and epilepsy risk: Evidence from Mendelian randomization analysis in East Asians

Abstract

Objective

Gut microbiota can traverse into the brain, activate the vagus nerve, and modulate immune responses and inflammatory processes, thereby influencing the onset of epileptic seizures. However, research on oral microbiota and epilepsy remains limited, and observational studies have been inconsistent. We aim to estimate the potential links between oral microbiota and epilepsy and elucidate which specific oral microbes may directly influence the pathogenesis of epilepsy.

Methods

A two‐sample MR analysis was conducted using genome‐wide association study (GWAS) data specific to OM and epilepsy in East Asian individuals. Single nucleotide polymorphisms (SNPs) independent of confounders served as instrumental variables (IVs) to deduce causality. MR methodologies, including inverse variance weighted (IVW), MR‐Egger, weighted median, and weighed mode methods, were utilized. Sensitivity analysis, including Cochrane's Q test, MR‐Egger intercept test, and leave‐one‐out analysis, was applied to confirm the robustness of results.

Results

Among the 3117 bacterial taxa examined, we observed that 14 OM, like s_Streptococcus_mitis, s_Streptococcus_pneumoniae, and s_Haemophilus, were positively associated with epilepsy, while 7 OM, like g_Fusobacterium and g_Aggregatibacter, were negatively related to epilepsy. The MR‐Egger intercept suggested that no evidence of horizontal pleiotropy was observed (p > 0.05). The leave‐one‐out analysis validated the robustness of the results.

Significance

This study underscores the effect of OM on epilepsy, suggesting potential mechanisms between the OM and epilepsy. Further investigation into the potential role of the OM is needed to enhance our in‐depth understanding of the pathogenesis of epilepsy.

Plain Language Summary

Previous research has demonstrated that the microbiota may influence the onset of epileptic seizures. We applied 3117 oral microbiota from the newest publicly available database of East Asian populations. Mendelian randomization analysis was utilized to estimate the causal relationship between oral microbiota and epilepsy. Our results showed that a causal effect exists between 21 oral microbiota and epilepsy. We provided genetic evidence for risk assessment and early intervention in epilepsy.

Key points.

• We applied 3117 oral microbiotas from the newest GWAS of East Asian populations.

• Causal effect exists between 21 oral microbiotas and epilepsy.

• Genetic evidence for risk assessment and early intervention in epilepsy.

1. INTRODUCTION

Epilepsy is a chronic neurological disorder characterized by spontaneous, recurrent, and transient central nervous system dysfunction. According to the World Health Organization, 5 million new epilepsy cases are diagnosed worldwide each year. ¹ It is noteworthy that 80% of individuals with epilepsy reside in low‐ and middle‐income countries, placing a significant economic burden on these regions. ² Given its status as a major public health concern, intensive research is urgently needed to develop innovative therapies for epilepsy. Recent studies have indicated that gut microbes can traverse into the brain, activate the vagus nerve, and modulate immune responses and inflammatory processes, thereby influencing the onset of epileptic seizures. ³ Furthermore, probiotic supplementation has been shown to have potential positive effects on epilepsy onset and therapy. ⁴ , ⁵ The ketogenic diet, through modulation of microbial composition, exhibits antiepileptic properties. ⁶ These findings not only underscore the significance of the microbiota‐gut‐brain axis in neurological disorders but also furnish scientific grounds for exploring nontraditional approaches such as probiotic supplementation and the ketogenic diet.

Despite the critical role of gut microbiota in epilepsy research, prior studies have often relied on fecal samples rather than gut microbiota directly, due to the poor accessibility of gut microbiota. However, fecal microbiota may differ from gut microbiota in terms of quantity and composition. Consequently, oral microbiota may represent an advantageous solution, as it is more easily accessible than the intestinal one. The oral microbiota, the second‐largest bacterial community in the body after the gut, exhibits a comparable level of diversity to that of the gut. ⁷ Reduced microbial diversity can lead to a low‐grade, chronic inflammatory response characterized by the production of anti‐inflammatory cytokines rather than pro‐inflammatory ones. Given the intimate connection between microbes and disease and the ease of sampling oral microbiota, exploring the relationship between oral microbes and epilepsy has emerged as a promising research direction. A 16S rRNA gene sequencing study involving children with cerebral palsy and epilepsy demonstrated that alterations in the microbial structure of the gut and oral cavity could potentially impact the development of clinical diseases, such as periodontitis, dental caries, and nutritional deficiencies, with children with epilepsy experiencing significant changes in the structure of both the oral and gut flora. ⁸ Moreover, a recent study comparing the oral microbiota of epileptic patients with healthy subjects revealed higher diversity in the oral microbiome of epileptic patients. Specifically, the oral cavity of epileptic patients exhibited an increased presence of 26 genera of Streptococcus and Bacillus, while showing a decreased presence of 14 genera such as Streptococcus gastricus, Neisseria, and Serratia. Importantly, these differences were diminished after epilepsy control. ⁹ Additionally, microbiota is characterized by a high capacity to communicate with each other and continuous exchange with environmental factors. All organs, including the brain, host a microbial population that is closely related to each other. This close relationship between the microbiota of different organs allows us to consider them, in some way, mirrors of the intestinal one, which is the most important of all and which has been well documented to have a role in epilepsy. Nonetheless, researches on oral microbes and epilepsy remain limited, and observational studies are constrained by factors such as sample size, observation time, and reverse causality. Further exploration is necessary to determine whether there is a causal relationship between the oral flora and epilepsy and whether it is linked to immune and inflammatory mechanisms.

The present study proposes the use of Mendelian randomization (MR), a robust strategy that utilizes genetic variation as a natural “randomized trial” to assess the causal relationship between exposures and disease outcomes. ¹⁰ In MR studies, single nucleotide polymorphisms (SNPs) are regarded as instrumental variables (IVs) for estimating the causal relationship between exposure and the outcome of interest. SNPs adhere to the principle of random assignment of genetic variation during meiosis and are less susceptible to confounding and reverse causation, as allelic transmission is inherently fixed and independent of the disease. ¹¹ Consequently, the MR approach mitigates the effects of confounding factors and reverse causation compared to traditional observational studies. By integrating data from genome‐wide association studies (GWAS) of oral microbiota and epilepsy in East Asian populations, the aim is to delve deeper into the potential causal links between oral microbes and epilepsy and elucidate which specific oral microbes may directly influence the pathogenesis of epilepsy. This study not only anticipates opening new avenues for early identification and prevention of epilepsy but also holds the promise of microbial intervention as a novel target for epilepsy treatment, offering transformative therapeutic perspectives for this age‐old disease.

2. METHODS

2.1. Study design and data source

The purpose of this study was to investigate the potential relationship between oral microbiota and epilepsy using a two‐sample MR analysis. The study framework is illustrated in Figure 1, and the combined data were obtained from the Genome‐Wide Association Study (GWAS) of East Asian populations, encompassing 3117 oral microbiota traits and epilepsy conditions, respectively. The study ensured that the genetic variants met the three fundamental prerequisites for MR analysis: (i) the genetic variants were significantly associated with oral microbiota, (ii) the genetic variants were not directly associated with epilepsy, and (iii) the genetic variants only affected epilepsy by influencing the microbiota. ¹²

FIGURE 1.

Study design of the mendelian randomization between oral microbiota and epilepsy. The solid lines represent the association between the instrumental variables (SNPs) and exposure as well as the association between exposure and outcome. Dash lines with cross means that the association meets two basic assumptions of mendelian randomization: (i) genetic variants do not show associations with the outcome (epilepsy); and (ii) genetic variants solely influence outcomes through the exposure. IVs, Instrumental variable; SNPs, Single nucleotide polymorphisms.

Data on oral microbiota was extracted from a recently published large‐scale GWAS (PMID 34873157), which included 2017 dorsal tongue and 1915 saliva samples from 2984 healthy Chinese individuals. ¹³ This study represents the first large‐scale GWAS of an East Asian population. Data on epilepsy was sourced from a substantial cohort from the BioBank of Japan, consisting of 2466 epileptic patients and 175 788 controls. ¹⁴ All cases were strictly categorized following the G40 codes in the ICD‐10 criteria. The authors of the original study provided detailed clinical information on the diagnosis of epilepsy, age of onset, seizure frequency, history of surgery, symptoms during seizures, and seizure types.

2.2. IV selection

To ensure the accuracy and validity of our MR analyses, we implemented a rigorous process for IVs' selection. Initially, the GWAS significance threshold was set at p < 5 × 10⁻⁸, and due to the limited number of eligible single nucleotide polymorphisms (SNPs), this threshold was moderately relaxed to p < 5 × 10⁻⁵. Additionally, we utilized the European Thousand Genomes Project reference panel and applied a strict linkage disequilibrium (LD) independence criterion (r ² < 0.001, cluster spacing 10 000 kb) to filter out closely interlinked SNPs. Subsequently, SNPs with a minor allele frequency (MAF) < 0.01 were removed, while SNPs showing significant correlation by the Steiger's test were retained. Finally, SNPs with F > 10 were filtered based on the F statistic (F = R ² × (N−2)/(1−R ²)), where R ² represents the proportion of exposed variance explained by the instrumental variable to reduce weak instrumental variable bias.

2.3. MR analysis

For the MR analysis, we employed the inverse variance weighting (IVW) method as the primary analytical approach to estimate the causal effect of the exposure on the outcome. Following the IVW analysis, we set a p‐value significance threshold of 0.0125, ¹⁵ and relationships meeting this threshold were considered as potential relationships. Recognizing that the IVW method provides reliable causal estimates only in the absence of horizontal pleiotropy, ¹⁶ we also utilized MR‐Egger, weighted median, and weighted mode as auxiliary methods to ensure the robustness of our findings. MR‐Egger corrects for and detects horizontal pleiotropy, providing valuable estimates, although its assessment is not as effective as the IVW method. ¹⁷ The weighted median method uses a median‐based approach combined with rate estimation from genetic instruments, providing reliable estimates even when up to 50% of the instruments are invalid. ¹⁸ The weighted mode method, on the other hand, is used for MR analysis when multiple instrumental variables are included to determine the direction and strength of causality. In cases where the IVW method yields a p‐value of less than 0.0125 and the direction of the β‐values from the other four MR methods is consistent with IVW, a possible causal relationship between the exposure and outcome variables is considered to exist, even if the p‐value of the other methods exceeds 0.0125.

2.4. Sensitivity analysis

To further validate our MR results, sensitivity analyses were conducted. For MR results showing consistent effect estimates across all methods, we utilized the Cochran's Q statistic for the MR‐Egger intercept and IVW to test for heterogeneity of oral microbial genetic instrumental variables in epilepsy. A p‐value greater than 0.05 in the Cochran's Q statistic test indicated no significant heterogeneity in the analysis. Additionally, the MR‐Egger intercept was employed to test for the presence of horizontal pleiotropy in the analyses, with a p‐value greater than 0.05 suggesting the absence of significant horizontal pleiotropy in the study. Lastly, the leave‐one‐out method was used to assess the potential effect of specific SNPs on the observed causal effects by removing individual SNPs one by one.

2.5. Statistical analysis

In the statistical analysis, the joint odds ratio (OR) and 95% confidence interval (CI) were applied to evaluate the relationship between oral microbiota levels and epilepsy risk. A causal relationship was considered statistically significant if the p‐value was less than 0.0125. The MR analyses were performed using the TwoSampleMR software package and R Foundation version 4.3.0.

3. RESULTS

3.1. Selection of instrumental variables

In this study, we investigated 1549 salivary microbiota and 1568 dorsal tongue microbiota using GWAS data to conduct a comprehensive taxonomic cluster analysis across six levels: phylum, order, family, genus, and species. Following stringent screening to eliminate SNPs significantly associated with the outcome and data harmonization, 60–91 high‐quality SNPs were retained for each group of microbiota. These SNPs passed the steiger test and got an F‐statistic more than 10, ensuring the absence of bias in the analysis due to weak instrumental variables.

3.2. Causal role of oral bacterial flora in epilepsy

The preliminary results of the correlation analysis between genetically predicted oral microbiota and epilepsy risk are presented in Table S1. Our study uncovers a potential relationship between oral microbiota and epilepsy. Among the 3117 bacterial taxa, we identified 21 oral microbiota with a statistically significant association with epilepsy (Figure 2, Figure 3). The findings from the IVW method revealed that g_Centipeda, g_UBA2866, g_Eikenella, s_Streptococcus_mitis, g_F0040‐1, s_Streptococcus_pneumoniae, s_Haemophilus‐1, g_Haemophilus‐2, g_Pauljensenia‐1, g_F0422, g_Streptococcus‐1, s_Haemophilus‐3, g_TM7x, g_Streptococcus‐2 clusters were positively correlated with epilepsy, indicating that their increase was associated with a higher risk of epilepsy (OR >1). Conversely, s_Veillonella_parvula, s_Granulicatella, g_F0040‐2, g_Pauljensenia‐2, g_Fusobacterium, g_Aggregatibacter, and g_Pseudopropionibacterium clusters were negatively associated with epilepsy, suggesting that their presence may reduce the risk of epilepsy (OR <1). It is important to note that the IVW method was employed in addition to three other methods, including MR‐Egger, weighted median, and weighted modal, nearly all of which yielded consistent conclusions, further affirming the reliability of the identified relationships (Table 1).

FIGURE 2.

The estimates of 21 oral microbiota on epilepsy via IVW method. CI, Confidence interval; IVW, Inverse variance weighting; OR, Odds ratio; SNP, Single nucleotide polymorphism.

FIGURE 3.

Circular Heatmap of mendelian randomization results of oral microbiota at species level and epilepsy.

TABLE 1.

Estimates of 21 oral microbita on epilepsy via four mendelian randomization methods.

Outcome	SNP	IVW		MR‐egger		Weighted median		Weighted mode
		OR (95%CI)	p	OR (95%CI)	p	OR (95%CI)	p	OR (95%CI)	p
g_Centipeda	74	1.28 (1.13,1.44)	0.00	4.58 (0.25,84.72)	0.31	1.40 (1.18,1.66)	0.00	1.44 (1.14,1.83)	0.00
g_UBA2866	74	1.27 (1.10,1.48)	0.00	5.19 (0.18,153.4)	0.34	1.23 (0.98,1.56)	0.08	1.28 (0.97,1.70)	0.09
g_Eikenella	63	1.25 (1.07,1.47)	0.01	2.87 (0.02,350.6)	0.67	1.20 (0.95,1.51)	0.12	1.22 (0.99,1.52)	0.07
s_Streptococcus_mitis	73	1.22 (1.08,1.39)	0.00	2.49 (0.38,16.37)	0.35	1.30 (1.08,1.58)	0.01	1.41 (1.11,1.78)	0.01
g_F0040‐1	80	1.22 (1.08,1.38)	0.00	5.46 (0.94,31.69)	0.06	1.22 (1.02,1.46)	0.03	1.23 (1.03,1.47)	0.03
s_Streptococcus_pneumoniae	72	1.22 (1.07,1.39)	0.00	1.03 (0.19,5.52)	0.98	1.24 (1.02,1.50)	0.03	1.20 (0.96,1.49)	0.11
s_Haemophilus‐1	82	1.21 (1.08,1.36)	0.00	1.55 (0.07,34.78)	0.78	1.22 (1.03,1.45)	0.02	1.19 (0.92,1.53)	0.19
g_Haemophilus‐2	91	1.20 (1.07,1.34)	0.00	1.92 (0.42,8.83)	0.40	1.21 (1.01,1.44)	0.04	1.19 (0.95,1.48)	0.14
g_Pauljensenia‐1	78	1.20 (1.07,1.34)	0.00	1.01 (0.12,8.30)	0.99	1.21 (1.04,1.41)	0.01	1.18 (0.94,1.49)	0.16
g_F0422	68	1.18 (1.04,1.34)	0.01	4.43 (0.38,51.17)	0.24	1.08 (0.89,1.29)	0.44	1.08 (0.86,1.37)	0.50
g_Streptococcus‐1	83	1.18 (1.05,1.32)	0.01	7.85 (0.73,84.12)	0.09	1.18 (0.99,1.40)	0.06	1.22 (0.98,1.51)	0.08
s_Haemophilus‐3	72	1.17 (1.04,1.33)	0.01	3.55 (0.75,16.70)	0.11	1.19 (1.00,1.43)	0.05	1.20 (0.99,1.47)	0.07
g_TM7x	85	1.16 (1.04,1.30)	0.01	1.40 (0.09,21.56)	0.81	1.08 (0.92,1.26)	0.34	1.13 (0.93,1.37)	0.22
g_Streptococcus‐2	79	1.16 (0.04,1.29)	0.01	2.17 (0.35,13.42)	0.41	1.25 (1.07,1.46)	0.01	1.35 (1.10,1.66)	0.01
s_Veillonella_parvula	73	0.85 (0.76,0.96)	0.01	0.68 (0.07,6.30)	0.73	0.82 (0.69,0.96)	0.02	0.83 (0.68,1.01)	0.07
s_Granulicatella	73	0.84 (0.74,0.95)	0.01	0.16 (0.02,1.35)	0.10	0.85 (0.71,1.00)	0.05	0.84 (0.68,1.03)	0.10
g_F0040‐2	78	0.84 (0.75,0.95)	0.00	0.48 (0.02,9.72)	0.63	0.88 (0.75,1.05)	0.16	0.88 (0.72,1.09)	0.26
g_Pauljensenia‐2	84	0.84 (0.74,0.95)	0.01	0.88 (0.07,10.83)	0.92	0.77 (0.65,0.91)	0.00	0.74 (0.61,0.91)	0.01
g_Fusobacterium	60	0.84 (0.73,0.96)	0.01	0.53 (0.03,11.04)	0.69	0.79 (0.65,0.97)	0.02	0.76 (0.58,0.99)	0.05
g_Aggregatibacter	66	0.83 (0.73,0.95)	0.01	0.51 (0.04,6.09)	0.60	0.83 (0.68,1.01)	0.06	0.87 (0.68,1.10)	0.24
g_Pseudopropionibacterium	67	0.80 (0.72,0.90)	0.00	0.43 (0.05,4.03)	0.46	0.78 (0.66,0.93)	0.00	0.75 (0.59,0.94)	0.01

Abbreviations: 95% CI, Lower and upper 95% confidence intervals; IVW, inverse variance weighting; MR, Mendelian randomization; OR, Odds ratio; p, p‐value; SNP, single‐nucleotide polymorphism.

3.3. Sensitivity analysis

In the sensitivity analysis, we conducted several tests to ensure the robustness of the results. First, the MR‐Egger intercept test indicated no evidence of horizontal pleiotropy (p > 0.05), suggesting that the 21 identified relationships were not influenced by this bias (Figure 3, Table 2, Table S2). Leave‐one‐out analysis further substantiated the stability of the results by demonstrating no significant alteration in the overall MR estimates after removing each SNP individually. Additionally, although there was no substantial evidence of heterogeneity among most microbiota (p > 0.05), a slight heterogeneity was observed in the association between g_Pauljensenia‐2 and epilepsy (p = 0.032) (Figure 3, Table 2, Table S3). In conclusion, the findings from our MR study demonstrated robustness and reliability.

TABLE 2.

Sensitivity analysis of 21 oral microbiota on epilepsy.

Outcome	Heterogeneity		Pleiotropy
	Cochrane Q test P‐MR egger	Cochrane Q test P‐IVW	P‐MR egger intercept
	0.133493635	0.136978662	0.39306704
g_UBA2866	0.194632875	0.201339543	0.418703749
g_Eikenella	0.984798607	0.649704738	0.736038807
s_Streptococcus_mitis	0.371713824	0.385849949	0.463455989
g_F0040‐1	0.693946049	0.638006595	0.09843381
s_Streptococcus_pneumoniae	0.632536253	0.662935182	0.838795691
s_Haemophilus‐1	0.656157813	0.684423163	0.876449209
g_Haemophilus‐2	0.683609432	0.700083838	0.544775619
g_Pauljensenia‐1	0.67776425	0.706006778	0.876915442
g_F0422	0.651069616	0.645836331	0.293484452
g_Streptococcus‐1	0.856942049	0.824049212	0.11996848
s_Haemophilus‐3	0.289372323	0.260423873	0.164377786
g_TM7x	0.5473385	0.577538199	0.89436576
g_Streptococcus‐2	0.952573214	0.956363471	0.500112524
s_Veillonella_parvula	0.567551025	0.599177741	0.840738176
s_Granulicatella	0.178563479	0.148606197	0.130954843
g_F0040‐2	0.576616988	0.604085132	0.716432932
g_Pauljensenia‐2	0.031889911	0.037608007	0.970209542
g_Fusobacterium	0.53920431	0.572917407	0.771019031
g_Aggregatibacter	0.622818939	0.651229582	0.700880147
g_Pseudopropionibacterium	0.747494063	0.766425279	0.58221693

Abbreviations: IVW, inverse variance weighting; MR, Mendelian randomization; p, p‐value.

4. DISCUSSION

In this study, we conducted a systematic evaluation of the relationship between 3117 oral microbiota and epilepsy. Ultimately, we identified a total of 21 OM that are associated with epilepsy, with 14 OM being identified as risk factors for epilepsy and 7 OM as protective factors. The findings of this study underscore the significance of OM in the context of epilepsy.

A mounting body of evidence suggests a robust connection between microbiota and epilepsy. Several microorganisms from the oral cavity, gut, or bacteremia exert influence on immune and inflammatory processes through interactions with the host. These interactions impact neuronal excitability modulate brain function and behavior. ³ Our study demonstrated a positive effect on epilepsy (OR >1) in 11 bacterial species, with more pronounced effects observed in Streptococcus spp. and Haemophilus spp. Streptococcus is a prevalent pathogen and an early dominant species in the human oral microbiome, exerting substantial influence on oral health. ¹⁹ Studies have indicated that specific oral streptococcal subtypes, such as Streptococcus mitis, can trigger bacteremia and endocarditis, ²⁰ whereas streptococcus agalactiae (Group B Streptococcus, or GBS) and streptococcus pneumoniae are leading causes of meningitis, ²¹ emphasizing their pathogenic significance. The connection between streptococci and epilepsy has been validated in prior studies. An observational study reported a heightened risk of epilepsy in neonates with GBS infections. ²² Recently, it was found that individuals with epilepsy exhibit a higher oral streptococcal relative abundance compared to healthy controls, with this elevation persisting even after epilepsy management. ⁹ Our investigation concurs with these precedents, affirming the correlation. The potential mechanism implicating streptococci in epilepsy pathogenesis might involve neuroinflammation. Research has elucidated that gut‐residing streptococci modulate levels of interleukin‐6 (IL‐6) and tumor necrosis factor‐alpha (TNF‐α). ²³ TNF‐α induces a substantial calcium influx, augmenting neuronal excitability, a key feature in epileptic episodes. ²⁴ Furthermore, evidence supports a role for IL‐6, a pro‐inflammatory cytokine, in the etiology of epilepsy. ²⁵ Haemophilus species are among the normal inhabitants of the human oral microbiota. They are also principal pathogens implicated in bacterial meningitis, with epilepsy being a frequent complication of the latter. Studies have observed a marked increase in Haemophilus colonization in the gut of epilepsy patients. ²⁶ Individuals with Haemophilus‐induced meningitis exhibit a higher frequency of epileptic seizures compared to those with meningitis caused by streptococcus pneumoniae or other bacteria. ²⁷ Another study revealed a doubled risk of developing epilepsy after vaccination with a type b Haemophilus‐containing Haemophilus influenzae B vaccine. ²⁸ Our findings similarly indicate an elevated risk associated with Haemophilus, suggesting a potential link between Haemophilus and epilepsy. We speculate that Haemophilus might influence oral immune responses, migrate to the gut, provoking chronic inflammation, or translocate bacterial toxins across the blood–brain barrier, affecting neuron excitability and thus triggering epilepsy. However, due to the paucity of direct research addressing the Haemophilus‐epilepsy connection, the mechanisms underlying the involvement of Haemophilus in epilepsy remain unclear.

Our findings also suggest that certain oral microbial communities may have a protective effect on the occurrence of epilepsy. We identify the genus Bifidobacterium as a protective factor, consistent with previous research. Many studies have demonstrated that certain Bifidobacterium species mitigate epilepsy incidence by producing beneficial metabolites. For instance, a decline in beneficial Bifidobacteria, like Bifidobacterium and Prevotella, is observed in the gut of epileptic patients, possibly linked to their production of short‐chain fatty acids (SCFAs). ²⁹ SCFAs are critical regulators of microglial development and function; microglia, as key immune cells in neural tissue, play a pivotal role in maintaining central nervous system homeostasis, with pathology activating them to secrete inflammatory factors such as IL‐1, IL‐8, and TNFα, increasing neuronal excitability contributing to epileptic mechanisms. ³⁰ Other studies reveal that Bacteroides and Propionibacterium modulate serum and gut amino acid levels, influencing neurotransmitter levels (like GABA and glutamate) implicated in seizures, exerting protective roles in murine epilepsy. ³¹ Fusobacterium, a prevalent anaerobic gram‐negative bacterium in the oral cavity and one of the most abundant, is also identified as a protective factor in our study, contrasting previous observational findings. Past research noted an increase in Fusobacterium in epileptic patients' mouths. ⁸ One study revealed a significant distinction between epileptic patients and healthy volunteers, with 0.6% of epileptic patients harboring Fusobacterium in their oral cavities compared to near absence in healthy subjects. It further implicates autoimmunity and inflammation in epilepsy etiology. ²⁶ Research shows Fusobacterium selectively recruits immune cells, fostering a tumor‐supportive inflammatory microenvironment, promoting colorectal tumorigenesis. ³² Fusobacterium is known to generate reactive oxygen species (ROS) and cytokines, ³³ creating a pro‐inflammatory milieu crucial in many diseases. ³⁴ Yet, Fusobacterium uniquely secretes adhesin Fap2, clustering pathogens on oral biofilms, ³⁵ such as co‐colonizing with candida, potentially promoting candida colonization and oral carcinogenesis via its metabolic products. ³⁶ The interspecies aggregation challenges understanding of a bacterium's singular or combined roles in disease. Moreover, with various species exhibiting different biological activities, they may exert disparate effects on the host, as Granulicatella and Pauljensenia in our study showed contrasting impacts on epilepsy. These discrepancies warrant deeper exploration into the differential species‐specific mechanisms of Fusobacterium in disease.

The results of this study have significant implications for public health and clinical practice. The detection of alterations in the oral microbiome could potentially be used as a basis for assessing the risk of epilepsy. Moreover, early interventions targeting dysbiosis in the oral microbiota, such as dietary modifications or probiotic supplementation, may emerge as key strategies for epilepsy prevention. Previous research has already demonstrated antiepileptic effects of the ketogenic diet through modulation of microbiota composition, and probiotics/prebiotics show promise in improving health through regulatory mechanisms.

This study has several notable strengths. Most notably, it leverages the latest GWAS data on the oral microbiome and employs a two‐sample MR approach to confirm relationships between oral microbiota and epilepsy in East Asian populations. This provides new insights for future research on the oral microbiome and potential therapeutic avenues for epilepsy. Furthermore, the investigation of as many as 3117 oral bacterial taxa and the identification of differential outcomes in 21 taxa underscore the breadth of coverage and large sample size, thereby providing robust results. However, there are also several limitations to acknowledge. First, the diversity of oral microbiota, with multiple species within each genus, may exert varied influences on disease, leading to discrepancies with prior research. Second, epilepsy encompasses numerous subtypes, such as idiopathic, refractory, and specific syndromic forms, each with unique characteristics, which our study did not explore in detail due to lack of subtype differentiation. Third, we acknowledge that the oral microbiome is not the sole factor influencing epilepsy. Some postnatal influences, such as environmental factors and dietary style, can also affect the onset of epilepsy. Additionally, our study did not stratify by age, gender, or type of epilepsy, so further research is needed to elucidate the relationship between these factors and epilepsy. Lastly, since both the oral microbiome and epilepsy GWAS data were derived from Asian populations, the generalizability of our findings to other ethnic groups remains uncertain.

5. CONCLUSION

In summary, this study represents a pioneering exploration of the relationship between the oral microbiome and epilepsy. It highlights the relationship of the microbiota with epilepsy, investigates potential inflammatory pathways impacting seizure onset, and suggests genetic evidence for risk assessment, early intervention, and inflammatory mechanism studies in epilepsy. Collectively, our work advances the understanding of the intricate interplay between oral microorganisms and neurological health, highlighting the microbiome as a promising frontier for novel diagnostic and therapeutic strategies in epilepsy management.

AUTHOR CONTRIBUTIONS

Chenyang Zhao: Conceptualization, data curation, investigation, methodology, software, validation, visualization, writing—original draft, writing—review and editing. Fei Chen and Qiong Li: Conceptualization, data curation, validation, writing—review and editing. Wei Zhang: writing—original draft, writing—review and editing. Lixiu Peng: Conceptualization, data curation, supervision, writing—review and editing. Chaoyan Yue: Conceptualization, data curation, formal analysis, supervision, writing—review and editing. All authors contributed to the article and approved the submitted version.

FUNDING INFORMATION

Technical projects within the First People's Hospital of Chenzhou, 2021B048.

CONFLICT OF INTEREST STATEMENT

None of the authors has any conflict of interest to disclose.

ETHICS STATEMENT

We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

CONSENT TO PARTICIPATE

Only publicly available GWAS data were used in this study, and the ethics approval and consent to participate could be available in the original GWAS study.

CONSENT FOR PUBLICATION

Not applicable.

Supporting information

Data S1:

Zhao C, Chen F, Li Q, Zhang W, Peng L, Yue C. Causal relationship between oral microbiota and epilepsy risk: Evidence from Mendelian randomization analysis in East Asians. Epilepsia Open. 2024;9:2419–2428. 10.1002/epi4.13074

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.