Introduction
Febrile seizure (FS) is the most common type of seizure during childhood period, and defined as seizures provoked by fever without central nervous system (CNS) infection [1]. Febrile states are induced by pyrogenic response to various infections, and the magnitude of pyrogenic response influence the body temperature of each child. Interleukin (IL)-1β, IL-6, and tumor necrosis factor-α (TNF-α) are major pro-inflammatory cytokines controlling pyrogenic actions [2]. Overproduction of pro-inflammatory cytokines can boost pyrogenic action and in some children, therefore body temperatures may overwhelm the seizure threshold, provoking to develop FS.
The association between cytokine genetic variants and susceptibility to FS and epilepsy are still controversial. IL-1β-511 promoter variants were reported to have an association with FS [3,4]. TNF-α-308 genotype showed no significant association with FS in meta-analysis study [5-8]. In other studies, GG genotypes of TNF-α-238 were more prevalent than GA genotype among FS compared to controls in Iranian children [9]. Japanese FS study showed significant lower frequencies of the IL-10-592C/-819C/-1082A haplotype than controls [10]. In contrast, IL-10-592, -819, and -1082 showed no significant allelic association in Iranian FS study [11].
Genetic epilepsy with febrile seizure plus (GEFS+) is a familial disorder with association of FS and epilepsy and shows autosomal dominance inheritance with variable penetrance [12]. And febrile seizure plus (FS+) is a same phenotypic disorder to GEFS+ without family history. To date, sodium voltage-gated channel alpha subunit 1 (SCN1A), sodium voltage-gated channel beta subunit 1 (SCN1B), and gamma-aminobutyric acid type A receptor gamma2 subunit (GABRG2) are known to be disease-causing genes of GEFS+ [13]. Inheritance in GEFS+ is typically autosomal dominant with incomplete penetrance, although other complex inheritance patterns may also occur. However, whether genetic susceptibility to inflammation may be one of the genetic causes for FS or GEFS+ is still unclear.
To determine whether promotor variants of TNF-α and IL-10 influence the susceptibility to FS, FS+, and GEFS+, we analysed genetic variants in the promotor region of TNF-α and IL-10 among children with FS and GEFS+ patients and compared to controls.
Materials and Methods
1. Patient information
Children with FS, FS+, and GEFS+ patients were enrolled in this study from June 2008 to May 2013, visiting the emergency room of Seoul Metropolitan Government Seoul National University Boramae Medical Center with acute seizure attacks. Inclusion criteria for FS were children with seizures associated with fever above 38℃ between 6-month-old to 5-year-old, without CNS infection, neurologic deficits and previous afebrile seizures [14]. Diagnosis of GEFS+ followed the criteria established in the 2017 International Classification of Epileptic Syndromes [15]. GEFS+ is usually diagnosed in families whose members have FSs that may continue past the usual age where these are expected to resolve and/or be accompanied by afebrile seizures that may be generalized seizures or focal seizures. FS+ are distinguished from the GEFS+ on the basis of family history. Controls were children matched for age without history of FS nor epilepsy. This study was approved by the Institutional Review Board at the Seoul Metropolitan Government Seoul National University Boramae Medical Center (20080918/06-2008-74/76). Informed consent was obtained from the parent of each child.
2. Variants selection
A total of four variants located in the promotor region of TNF-α, -238 (rs361525), -308 (rs1800629), -857 (rs1799724), -863 (rs1800630), and also four variants located in the promotor region of IL-10, -592 (rs1800872), -819 (rs1800871), -1082 (rs1800896), -1352 (rs1800893), were selected from the dbSNP database (www.ncbi.nlm.nih.gov/SNP) and the HapMap human SNP database (www.hapmap.org). For selecting variants, variants with a minor allele frequency above 0.05 were included. To estimate pairwise linkage disequilibrium of variant marker, we used Haploview v4.0 (http://www.broadinstitute.org/haploview/haploview). All variants did not show the results of the chi-square test to reject the Hardy-Weinberg equilibrium. The default confidence interval algorithm of the Haploview program revealed 1 haplotype block (Fig. 1A) of TNF-α-857, -863 and 1 haplotype block (Fig. 1B) of IL-10-592, -819, -1082, -1352, from patient group data.
3. Variant sequencing and genotyping
Probes and primers were designed with genomic sequence information. After amplifying the variant spanning fragments by polymerase chain reaction, genotyping was performed with SNaPshot (Sequenom, San Diego, CA, USA). The person analysing the genotype result was blinded to the clinical data.
4. Statistical analysis
The trend test, chi-square test, Fisher exact test and the logistic regression test were the statistical approaches used analysing the genotype distributions of patient group including FS, FS+, and GEFS+ and then comparing with controls, depending on mode of inheritance [16], such as additive, dominant and recessive, based on the minor allele of each variants. IBM SPSS statistics version 20 (IBM Co., Armonk, NY, USA) and R version 3.2.5 (http://www.r-project.org) were used to analyse the tests. The statistical significance of differences was set as P<0.05 for all tests.
Results
1. Patient characteristics
Fifty-seven children with FS, 32 FS+, and 12 GEFS+ patients and 108 controls were enrolled. Semiology of FS were 46 (81%) simple types and 11 (19%) complex types. Three FS children had a history of febrile status epilepticus. All patients with FS+ and GEFS+ developed epilepsy after previous FS attacks. The children with FS, FS+, and GEFS+ patients did not show significant differences by sex, age, and laboratory findings with controls.
2. TNF-α allele and genotype variants
AA genotypes at TNF-α-863 were present only in controls. AA genotype at TNF-α-863 showed significant negative association with FS, FS+, and GEFS+ (P=0.029) (Table 1). TNF-α-238, TNF-α-308, and TNF-α-857 showed no significant allelic and genotypic differences (Table 2).
Table 1.
Comparison of genotypic frequencies of 4 TNF-α SNPs between the patients with FS, FS+, and GEFS+ versus controls
| Gene variants | Genotype | No. (%) | Genetic mode | P value | |
|---|---|---|---|---|---|
| FS, FS+, GEFS+ (n=100) | Control (n=106) | ||||
| TNF-α-238 rs361525 | G/G | 84 (84.0) | 94 (88.7) | Additive | 0.329 |
| G/A | 16 (16.0) | 12 (11.3) | Dominant | 0.327 | |
| A/A | Recessive | ||||
| TNF-α-308 rs1800629 | G/G | 85 (85.9) | 93 (87.7) | Additive | 0.852 |
| G/A | 14 (14.1) | 12 (11.3) | Dominant | 0.691 | |
| A/A | 0 (0.0) | 1 (0.9) | Recessive | 1.000 | |
| TNF-α-857 rs1799724 | C/C | 71 (71.7) | 73 (68.9) | Additive | 0.828 |
| C/T | 24 (24.2) | 30 (28.3) | Dominant | 0.656 | |
| T/T | 4 (4.0) | 3 (2.8) | Recessive | 0.714 | |
| TNF-α-863 rs1800630 | C/C | 69 (69.0) | 69 (65.1) | Additive | 0.203 |
| C/A | 31 (31.0) | 31 (29.2) | Dominant | 0.551 | |
| A/A | 0 (0.0) | 6 (5.7) | Recessive | 0.029a | |
Table 2.
Comparison of allelic frequencies of 4 TNF-α SNPs between the patients with FS, FS+, and GEFS+ versus controls
3. IL-10 allele and genotype variants
IL-10-592, -819, -1082, and -1352 failed to show significant allelic or genotypic association with FS, FS+, and GEFS+ compared to controls (Table 3 and 4).
Table 3.
Comparison of genotypic frequencies of 4 IL-10 SNPs between the patients with FS, FS+, and GEFS+ versus controls
Table 4.
Comparison of allelic frequencies of 4 IL-10 SNPs between the patients with FS, FS+, and GEFS+ versus controls
4. Haplotype analysis: TNF-α-857 and TNF-α-863
Haplotype frequencies of block 1 consisted with TNF-α-863 and TNF-α-857 showed no significant association with patients group of FS, FS+, and GEFS+ compared to controls.
5. Haplotype analysis: IL-10-592, -819, -1082, and -1352
Haplotype frequencies of block 1 consisted with IL-10-592, -819, -1082, and -1352 showed no significant association with patients group of FS, FS+, and GEFS+ compared to controls (Table 5).
Table 5.
Haplotype frequency analysis between FS, FS+, and GEFS+ versus controls
Discussion
This study demonstrates that allelic and genotypic frequencies of TNF-α and IL-10 promotor variants showed no significant differences between FS, FS+, and GEFS+ versus controls in Korean children. However, AA genotypes at TNF-α-863 were present only in controls, therefore AA genotype at TNF-α-863 showed significant negative association with FS, FS+, and GEFS+ compared to controls. Thus, this results may suggest that AA genotypes at TNF-α-863 show protective effects against FS, FS+, and GEFS+. However, our study population is small, so further study is needed with larger number of patients.
Pro-inflammatory cytokines play major actions in seizure generation and exacerbation [17]. IL-1β and TNF-α showed elevated levels in brains of experimental animals after electrical stimulation of the amygdala [18]. TNF-α is mostly released by microglia in the brain [19] and induces astrocytes to release glutamate [20]. An increase in extracellular glutamate may stimulate glutamatergic neurons, leading neuronal hyper-excitability. TNF-α upregulates α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors, increasing glutamatergic transmission [21]. TNF-α also upregulates endocytosis of gamma-Aminobutyric acid, or γ-aminobutyric acid (GABA) receptors, and eventually suppresses effects of the inhibition [22]. Altogether TNF-α leads to increase seizure susceptibility [22,23].
TNF-α is a potent pro-inflammatory cytokine showing implications with a large number of human diseases including many autoimmune diseases [7]. TNF-α shows alternate roles depending on variants of TNF-α gene regulating its effect and production [24]. Therefore, genetic variants that upregulate cytokine production may increase susceptibility to inflammation; subsequently, an exaggerated pro-inflammatory cytokine responses during infection may predispose in certain children to develop FS and subsequent epilepsy, especially FS+ and GEFS+. The postictal serum levels of IL-1β, IL-6, TNF-α, and high mobility group box 1 (HMGB1) showed significant elevation among children with FS attacks and children with epilepsy in afebrile seizure attacks, shown in our previous study [25].
TNF-α-238, -308, and -863, among promoter regions, demonstrated to alter transcriptional activity [26]. TNF-α-863 genotype showed a negative association with FS, FS+, and GEFS+ among Korean children in our study. AA genotype at TNF-α-863 correlates with a lower risk to develop FS, FS+, and GEFS+. CC genotype at TNF-α-863 was correlated with an increased risk of juvenile rheumatoid arthritis in Turkish patients [27]. Also, CC genotypes at TNF-α-863 showed higher serum levels of TNF-α, both pre-operatively and post-operatively after cardiac surgery in 122 German patients [28]. Taken together, CC genotype at TNF-α-863 is correlated with more pro-inflammatory action than AA genotype.
Korean Reference Genome DB KRGDB (http://coda.nih.go.kr/coda/KRGDB/index.jsp) are the free database of 1,100 Korean genomes. A allele frequency of TNF-α-863 is 15% and genotype frequencies are no available. In our study, A allele frequency is 15% in patients group of FS, FS+, and GEFS+ and 20% in controls; therefore, we can assume that our study population is not deviant to the general Korean population. The GG genotypes of TNF-α-238 were more prevalent than GA genotype among FS compared to controls in an Iranian study [9]. However, in our Korean population, there were no significant genotypic differences at TNF-α-238 in patient group of FS, FS+, and GEFS+ compared to controls.
TNF-α-308 is reported to have an association with higher susceptibility to asthma, atopic dermatitis, increased fatality in meningococcemia and ankylosing spondylitis [29-32]. However, in FS meta-analysis study, TNF-α-308 genotype showed no significant association [5-8]. Our study also showed no significant association at TNF-α-308 with FS, FS+, and GEFS+.
IL-10 is a major cytokine having anti-inflammatory action in immune system. IL-10 injected animals showed significantly higher threshold for provoking FS attacks than that in the controls, suggesting a protective effect to FS development [10]. IL-10 serum levels are controversial in several FS studies with some reporting increased [33] or others not increased levels [3,34].
IL-10 transmits negative feedback signals to decrease the immune system activation upon various inflammatory stimuli [35]. The IL-10-592, -819, and -1082 are placed in the IL-10 promotor regions having putative regulatory actions [36]. In a study of Japanese FS patients, the frequencies of the IL-10-592C/-819C/-1082A haplotype were significantly lower than controls [10]. In our study, the haplotype frequencies of IL-10-592C/-819C/-1082A/-1352G were also decreased in patient group with FS, FS+, and GEFS+ compared to controls, although statistically insignificant (18.5% vs. 24.5%, P=0.497). In contrast, Iranian FS study reported that IL-10-592, -819, and -1082 showed no significant allelic association [11].
The limitation of our study is relatively small number of patients enrolled. Therefore, further studies with larger number of patients with different ethnicities are needed to reveal the exact association of TNF-α gene variants with FS, FS+, and GEFS+ in children.
In summary, allelic and genotypic frequencies of TNF-α and IL-10 promotor variants showed no significant differences between FS, FS+, and GEFS+ versus controls. However, AA genotypes at TNF-α-863 were present only in controls; therefore, TNF-α-863 (rs1800630) promoter variants may be negatively associated with FS, FS+, and GEFS+. Our results support that the promotor genetic variant linked to lesser production of pro-inflammatory cytokine TNF-α may be implicated in the protection to fever-provoked seizures in Korean children.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Cited
