Reprint of: Recessive APC2 missense variants associated with epilepsies without neurodevelopmental disorders.

OBJECTIVES: The APC2 gene, encoding adenomatous polyposis coli protein-2, is involved in cytoskeletal regulation in neurons responding to endogenous extracellular signals and plays an important role in brain development. Previously, the APC2 variants have been reported to be associated with cortical dysplasia and intellectual disability. This study aims to explore the association between APC2 variants and epilepsy. METHODS: Whole-exome sequencing (WES) was performed in cases (trios) with epilepsies of unknown causes. The damaging effects of variants were predicted by protein modeling and in silico tools. Previously reported APC2 variants were reviewed to analyze the genotype-phenotype correlations. RESULTS: Four pairs of compound heterozygous missense variants were identified in four unrelated patients with epilepsy without brain malformation/intellectual disability. All variants presented no or low allele frequencies in the controls. The missense variants were predicted to be damaging by silico tools, and affect hydrogen bonding with surrounding amino acids or decreased protein stability. Patients with variants that resulted in significant changes in protein stability exhibited more severe and intractable epilepsy, whereas patients with variants that had minor effect on protein stability exhibited relatively mild phenotypes. The previously reported APC2 variants in patients with complex cortical dysplasia with other brain malformations-10 (CDCBM10; MIM: 618677) were all truncating variants; in contrast, the variants identified in epilepsy in this study were all missense variants, suggesting a potential genotype-phenotype correlation. SIGNIFICANCE: This study suggests that APC2 is potentially associated with epilepsy without brain malformation/intellectual disability. The genotype-phenotype correlation helps to understand the underlying mechanisms of phenotypic heterogeneity.

Epilepsy-associated genes: an update.

PURPOSE: To provide an updated list of epilepsy-associated genes based on clinical-genetic evidence. METHODS: Epilepsy-associated genes were systematically searched and cross-checked from the OMIM, HGMD, and PubMed databases up to July 2023. To facilitate the reference for the epilepsy-associated genes that are potentially common in clinical practice, the epilepsy-associated genes were ranked by the mutation number in the HGMD database and by case number in the China Epilepsy Gene 1.0 project, which targeted common epilepsy. RESULTS: Based on the OMIM database, 1506 genes were identified to be associated with epilepsy and were classified into three categories according to their potential association with epilepsy or other abnormal phenotypes, including 168 epilepsy genes that were associated with epilepsies as pure or core symptoms, 364 genes that were associated with neurodevelopmental disorders as the main symptom and epilepsy, and 974 epilepsy-related genes that were associated with gross physical/systemic abnormalities accompanied by epilepsy/seizures. Among the epilepsy genes, 115 genes (68.5%) were associated with epileptic encephalopathy. After cross-checking with the HGMD and PubMed databases, an additional 1440 genes were listed as potential epilepsy-associated genes, of which 278 genes have been repeatedly identified variants in patients with epilepsy. The top 100 frequently reported/identified epilepsy-associated genes from the HGMD database and the China Epilepsy Gene 1.0 project were listed, among which 40 genes were identical in both sources. SIGNIFICANCE: Recognition of epilepsy-associated genes will facilitate genetic screening strategies and be helpful for precise molecular diagnosis and treatment of epilepsy in clinical practice.

NEXMIF variants are associated with epilepsy with or without intellectual disability.

OBJECTIVES: Variants in NEXMIF had been reported associated with intellectual disability (ID) without epilepsy or developmental epileptic encephalopathy (DEE). It is unkown whether NEXMIF variants are associated with epilepsy without ID. This study aims to explore the phenotypic spectrum of NEXMIF and the genotype-phenotype correlations. MATERIALS AND METHODS: Trio-based whole-exome sequencing was performed in patients with epilepsy. Previously reported NEXMIF variants were systematically reviewed to analyze the genotype-phenotype correlations. RESULTS: Six variants were identified in seven unrelated cases with epilepsy, including two de novo null variants and four hemizygous missense variants. The two de novo variants were absent in all populations of gnomAD and four hemizygous missense variants were absent in male controls of gnomAD. The two patients with de novo null variants exhibited severe developmental epileptic encephalopathy. While, the patients with hemizygous missense variants had mild focal epilepsy with favorable outcome. Analysis of previously reported cases revealed that males with missense variants presented significantly higher percentage of normal intellectual development and later onset age of seizure than those with null variants, indicating a genotype-phenotype correlation. CONCLUSION: This study suggested that NEXMIF variants were potentially associated with pure epilepsy with or without intellectual disability. The spectrum of epileptic phenotypes ranged from the mild epilepsy to severe developmental epileptic encephalopathy, where the epileptic phenotypes variability are potentially associated with patients' gender and variant type.

Recessive APC2 missense variants associated with epilepsies without neurodevelopmental disorders.

OBJECTIVES: The APC2 gene, encoding adenomatous polyposis coli protein-2, is involved in cytoskeletal regulation in neurons responding to endogenous extracellular signals and plays an important role in brain development. Previously, the APC2 variants have been reported to be associated with cortical dysplasia and intellectual disability. This study aims to explore the association between APC2 variants and epilepsy. METHODS: Whole-exome sequencing (WES) was performed in cases (trios) with epilepsies of unknown causes. The damaging effects of variants were predicted by protein modeling and in silico tools. Previously reported APC2 variants were reviewed to analyze the genotype-phenotype correlations. RESULTS: Four pairs of compound heterozygous missense variants were identified in four unrelated patients with epilepsy without brain malformation/intellectual disability. All variants presented no or low allele frequencies in the controls. The missense variants were predicted to be damaging by silico tools, and affect hydrogen bonding with surrounding amino acids or decreased protein stability. Patients with variants that resulted in significant changes in protein stability exhibited more severe and intractable epilepsy, whereas patients with variants that had minor effect on protein stability exhibited relatively mild phenotypes. The previously reported APC2 variants in patients with complex cortical dysplasia with other brain malformations-10 (CDCBM10; MIM: 618677) were all truncating variants; in contrast, the variants identified in epilepsy in this study were all missense variants, suggesting a potential genotype-phenotype correlation. SIGNIFICANCE: This study suggests that APC2 is potentially associated with epilepsy without brain malformation/intellectual disability. The genotype-phenotype correlation helps to understand the underlying mechanisms of phenotypic heterogeneity.

SZT2 variants associated with partial epilepsy or epileptic encephalopathy and the genotype-phenotype correlation.

Background: Recessive SZT2 variants are reported to be associated with developmental and epileptic encephalopathy 18 (DEE-18) and occasionally neurodevelopment abnormalities (NDD) without seizures. This study aims to explore the phenotypic spectrum of SZT2 and the genotype-phenotype correlation. Methods: Trios-based whole-exome sequencing was performed in patients with epilepsy. Previously reported SZT2 mutations were systematically reviewed to analyze the genotype-phenotype correlations. Results: SZT2 variants were identified in six unrelated cases with heterogeneous epilepsy, including one de novo null variant and five pairs of biallelic variants. These variants had no or low frequencies in controls. All missense variants were predicted to alter the hydrogen bonds with surrounding residues and/or protein stability. The three patients with null variants exhibited DEE. The patients with biallelic null mutations presented severe DEE featured by frequent spasms/tonic seizures and diffuse cortical dysplasia/periventricular nodular heterotopia. The three patients with biallelic missense variants presented mild partial epilepsy with favorable outcomes. Analysis of previously reported cases revealed that patients with biallelic null mutations presented significantly higher frequency of refractory seizures and earlier onset age of seizure than those with biallelic non-null mutations or with biallelic mutations containing one null variant. Significance: This study suggested that SZT2 variants were potentially associated with partial epilepsy with favorable outcomes without NDD, expanding the phenotypic spectrum of SZT2. The genotype-phenotype correlation helps in understanding the underlying mechanism of phenotypic variation.

Variants in BRWD3 associated with X-linked partial epilepsy without intellectual disability.

AIMS: Etiology of the majority patients with idiopathic partial epilepsy (IPE) remains elusive. We thus screened the potential disease-associated variants in the patients with IPE. METHODS: Trios-based whole exome sequencing was performed in a cohort of 320 patients with IPE. Frequency and molecular effects of variants were predicted. RESULTS: Three novel BRWD3 variants were identified in five unrelated cases with IPE, which were four male cases and one female case. The variants included two recurrent missense variants (c.836C>T/p.Thr279Ile and c.4234A>C/p.Ile1412Leu) and one intronic variant close to splice site (c.2475 + 6A>G). The two missense variants were located in WD40 repeat domain and bromodomain, respectively. They were predicted to be damaging by silico tools and change hydrogen bonds with surrounding amino acids. The frequency of mutant alleles in this cohort was significantly higher than that in the controls of East Asian and all population of gnomAD. All these variants were inherited from the asymptomatic mothers. Four male cases presented frequent seizures at onset, while the female case only had two fever-triggered seizures. They showed good responses to valproate and lamotrigine, then finally became seizure free. All the cases had no intellectual disability. Further analysis demonstrated that all previously reported destructive variants of BRWD3 caused intellectual disability, while missense variants located in WD40 repeat domains and bromodomains of BRWD3 were associated with epilepsy. CONCLUSION: BRWD3 gene is potentially associated with X-linked partial epilepsy without intellectual disability. The genotypes and locations of BRWD3 variants may explain for their phenotypic variation.

CELSR1 variants are associated with partial epilepsy of childhood.

CELSR1 gene, encoding cadherin EGF LAG seven-pass G-type receptor 1, is mainly expressed in neural stem cells during the embryonic period. It plays an important role in neurodevelopment. However, the relationship between CELSR1 and disease of the central nervous system has not been defined. In this study, we performed trios-based whole-exome sequencing in a cohort of 356 unrelated cases with partial epilepsy without acquired causes and identified CELSR1 variants in six unrelated cases. The variants included one de novo heterozygous nonsense variant, one de novo heterozygous missense variant, and four compound heterozygous missense variants that had one variant was located in the extracellular region and the other in the cytoplasm. The patients with biallelic variants presented severe epileptic phenotypes, whereas those with heterozygous variants were associated with a mild epileptic phenotype of benign epilepsy with centrotemporal spikes (BECTS). These variants had no or low allele frequency in the gnomAD database. The frequencies of the CELSR1 variants in this cohort were significantly higher than those in the control populations. The evidence from ClinGen Clinical-Validity Framework suggested a strong association between CELSR1 variants and epilepsy. These findings provide evidence that CELSR1 is potentially a candidate pathogenic gene of partial epilepsy of childhood.

BCOR variants are associated with X-linked recessive partial epilepsy.

OBJECTIVE: BCOR gene, encoding a corepressor of BCL6, plays an important role in fetal development. BCOR mutations were previously associated with oculofaciocardiodental syndrome (OFCD or MCOPS2, OMIM# 300166). The BCOR protein is ubiquitously expressed in multiple areas, including the brain. However, the role of BCOR in neurological disorder remains elusive. METHODS: Trios-based whole-exome sequencing was performed in a cohort of 323 cases with partial epilepsy without acquired causes. RESULTS: Seven hemizygous missense BCOR variants, including c 0.103 G>C/p.Asp35His, c.1079 A>G/p.His360Arg, c 0.1097 C>T/p.Thr366Ile, c 0.3301 C>T/p.Pro1101Ser, c 0.3391 C>T/p.Arg1131Trp, c 0.4199 G>A/p.Arg1400Gln, and c 0.5254 G>A/p.Asp1752Asn, were identified in seven cases with partial epilepsy. Two patients presented partial seizures with generalized seizures and/or generalized discharges. One case showed cortical dysplasia in the right temporal-occipital area on MRI. Two cases presented mild developmental delay. However, all patients achieved seizure-free. The frequency of BCOR variants in the present cohort was significantly higher than that in the controls of healthy Chinese volunteers and all populations of Genome Aggregation Database (gnomAD). Computational modeling, including hydrogen bond and prediction of protein stability, implied that the variants lead to structural impairment. Previously, OFCD associated BCOR mutations were mostly destructive mutations in an X-linked dominant (XLD) pattern; in contrast, the BCOR variants identified in this study were all missense variants, which were associated with partial epilepsy in an X-linked recessive (XLR) pattern. The proportion of missense mutations in epilepsy was significantly higher than that in OFCD. CONCLUSIONS: BCOR was potentially a candidate pathogenic gene of partial epilepsy with or without developmental delay. The genotype-phenotype correlation helps understanding the mechanism underlying phenotypic variation.

SHROOM4 Variants Are Associated With X-Linked Epilepsy With Features of Generalized Seizures or Generalized Discharges.

Objective: SHROOM4 gene encodes an actin-binding proteins, which plays an important role in cytoskeletal architecture, synaptogenesis, and maintaining gamma-aminobutyric acid receptors-mediated inhibition. SHROOM4 mutations were reported in patients with the Stocco dos Santos type of X-linked syndromic intellectual developmental disorder (SDSX; OMIM# 300434). In this study, we investigated the association between SHROOM4 and epilepsy. Methods: Trios-based whole-exome sequencing was performed in a cohort of 320 cases with idiopathic generalized epilepsy or idiopathic partial epilepsy. Protein modeling was used to assess the damaging effects of variations. Results: Six hemizygous missense SHROOM4 variants, including c.13C > A/p. Pro5Thr, c.3236C > T/p.Glu1079Ala, c.3581C > T/p.Ser1194Leu, c.4288C > T/p.Arg1430Cys, c.4303G > A/p.Val1435Met, c.4331C > T/p.Pro1444Leu, were identified in six cases with idiopathic epilepsy without intellectual disability. All patients presented with features of generalized seizures or generalized discharges. These hemizygous variants had no or extremely low allele frequencies in controls and showed statistically higher frequency in the case cohort than controls. All variants were predicted to alter hydrogen bond with surrounding amino acids or decreased protein stability. The SHROOM4 variants reported in patients with SDSX were mostly destructive or duplicative variants; in contrast, the SHROOM4 variants were all missense variants, suggesting a potential genotype-phenotype correlation. The two missense variants associated with SDSX were located in the middle of SHROOM4 protein, whereas variants associated with idiopathic epilepsy were located around the N-terminal PDZ domain and the C-terminal ASD2 domain. Significance: SHROOM4 was potentially a candidate pathogenic gene of idiopathic epilepsy without intellectual disability. The genotype-phenotype correlation and sub-regional effect helps understanding the mechanism underlying phenotypic variation.

Recessive LAMA5 Variants Associated With Partial Epilepsy and Spasms in Infancy.

Objective: The LAMA5 gene encodes the laminin subunit α5, the most abundant laminin α subunit in the human brain. It forms heterotrimers with the subunit ß1/ß2 and γ1/γ3 and regulates neurodevelopmental processes. Genes encoding subunits of the laminin heterotrimers containing subunit α5 have been reported to be associated with human diseases. Among LAMAs encoding the laminin α subunit, LAMA1-4 have also been reported to be associated with human disease. In this study, we investigated the association between LAMA5 and epilepsy. Methods: Trios-based whole-exome sequencing was performed in a cohort of 118 infants suffering from focal seizures with or without spasms. Protein modeling was used to assess the damaging effects of variations. The LAMAs expression was analyzed with data from the GTEX and VarCards databases. Results: Six pairs of compound heterozygous missense variants in LAMA5 were identified in six unrelated patients. All affected individuals suffered from focal seizures with mild developmental delay, and three patients presented also spasms. These variants had no or low allele frequencies in controls and presented statistically higher frequency in the case cohort than in controls. The recessive burden analysis showed that recessive LAMA5 variants identified in this cohort were significantly more than the expected number in the East Asian population. Protein modeling showed that at least one variant in each pair of biallelic variants affected hydrogen bonds with surrounding amino acids. Among the biallelic variants in cases with only focal seizures, two variants of each pair were located in different structural domains or domains/links, whereas in the cases with spasms, the biallelic variants were constituted by two variants in the identical functional domains or both with hydrogen bond changes. Conclusion: Recessive LAMA5 variants were potentially associated with infant epilepsy. The establishment of the association between LAMA5 and epilepsy will facilitate the genetic diagnosis and management in patients with infant epilepsy.

Recessive PKD1 Mutations Are Associated With Febrile Seizures and Epilepsy With Antecedent Febrile Seizures and the Genotype-Phenotype Correlation.

Objective: The PKD1 encodes polycystin-1, a large transmembrane protein that plays important roles in cell proliferation, apoptosis, and cation transport. Previous studies have identified PKD1 mutations in autosomal dominant polycystic kidney disease (ADPKD). However, the expression of PKD1 in the brain is much higher than that in the kidney. This study aimed to explore the association between PKD1 and epilepsy. Methods: Trios-based whole-exome sequencing was performed in a cohort of 314 patients with febrile seizures or epilepsy with antecedent febrile seizures. The damaging effects of variants was predicted by protein modeling and multiple in silico tools. The genotype-phenotype association of PKD1 mutations was systematically reviewed and analyzed. Results: Eight pairs of compound heterozygous missense variants in PKD1 were identified in eight unrelated patients. All patients suffered from febrile seizures or epilepsy with antecedent febrile seizures with favorable prognosis. All of the 16 heterozygous variants presented no or low allele frequencies in the gnomAD database, and presented statistically higher frequency in the case-cohort than that in controls. These missense variants were predicted to be damaging and/or affect hydrogen bonding or free energy stability of amino acids. Five patients showed generalized tonic-clonic seizures (GTCS), who all had one of the paired missense mutations located in the PKD repeat domain, suggesting that mutations in the PKD domains were possibly associated with GTCS. Further analysis demonstrated that monoallelic mutations with haploinsufficiency of PKD1 potentially caused kidney disease, compound heterozygotes with superimposed effects of two missense mutations were associated with epilepsy, whereas the homozygotes with complete loss of PKD1 would be embryonically lethal. Conclusion: PKD1 gene was potentially a novel causative gene of epilepsy. The genotype-phenotype relationship of PKD1 mutations suggested a quantitative correlation between genetic impairment and phenotypic variation, which will facilitate the genetic diagnosis and management in patients with PKD1 mutations.

CACNA1A Mutations Associated With Epilepsies and Their Molecular Sub-Regional Implications.

Purpose: Previously, mutations in the voltage-gated calcium channel subunit alpha1 A (CACNA1A) gene have been reported to be associated with paroxysmal disorders, typically as episodic ataxia type 2. To determine the relationship between CACNA1A and epilepsies and the role of molecular sub-regional on the phenotypic heterogeneity. Methods: Trio-based whole-exome sequencing was performed in 318 cases with partial epilepsy and 150 cases with generalized epilepsy. We then reviewed all previously reported CACNA1A mutations and analyzed the genotype-phenotype correlations with molecular sub-regional implications. Results: We identified 12 CACNA1A mutations in ten unrelated cases of epilepsy, including four de novo null mutations (c.2963_2964insG/p.Gly989Argfs*78, c.3089 + 1G > A, c.4755 + 1G > T, and c.6340-1G > A), four de novo missense mutations (c.203G > T/p.Arg68Leu, c.3965G > A/p.Gly1322Glu, c.5032C > T/p.Arg1678Cys, and c.5393C > T/p.Ser1798Leu), and two pairs of compound heterozygous missense mutations (c.4891A > G/p.Ile1631Val& c.5978C > T/p.Pro1993Leu and c.3233C > T/p.Ser1078Leu&c.6061G > A/p.Glu2021Lys). The eight de novo mutations were evaluated as pathogenic or likely pathogenic mutations according to the criteria of American College of Medical Genetics and Genomics (ACMG). The frequencies of the compound heterozygous CACNA1A mutations identified in this cohort were significantly higher than that in the controls of East Asian and all populations (P = 7.30 × 10-4, P = 2.53 × 10-4). All of the ten cases were ultimately seizure-free after antiepileptic treatment, although frequent epileptic seizures were observed in four cases. Further analysis revealed that episodic ataxia type 2 (EA2) had a tendency of higher frequency of null mutations than epilepsies. The missense mutations in severe epileptic phenotypes were more frequently located in the pore region than those in milder epileptic phenotypes (P = 1.67 × 10-4); de novo mutations in the epilepsy with intellectual disability (ID) had a higher percentage than those in the epilepsy without ID (P = 1.92 × 10-3). Conclusion: This study suggested that CACNA1A mutations were potentially associated with pure epilepsy and the spectrum of epileptic phenotypes potentially ranged from the mild form of epilepsies such as absence epilepsy or partial epilepsy, to the severe form of developmental epileptic encephalopathy. The clinical phenotypes variability is potentially associated with the molecular sub-regional of the mutations.

Optimization of in silico tools for predicting genetic variants: individualizing for genes with molecular sub-regional stratification.

Genes are unique in functional role and differ in their sensitivities to genetic defects, but with difficulties in pathogenicity prediction. This study attempted to improve the performance of existing in silico algorithms and find a common solution based on individualization strategy. We initiated the individualization with the epilepsy-related SCN1A variants by sub-regional stratification. SCN1A missense variants related to epilepsy were retrieved from mutation databases, and benign missense variants were collected from ExAC database. Predictions were performed by using 10 traditional tools with stepwise optimizations. Model predictive ability was evaluated using the five-fold cross-validations on variants of SCN1A, SCN2A, and KCNQ2. Additional validation was performed in SCN1A variants of damage-confirmed/familial epilepsy. The performance of commonly used predictors was less satisfactory for SCN1A with accuracy less than 80% and varied dramatically by functional domains of Nav1.1. Multistep individualized optimizations, including cutoff resetting, domain-based stratification, and combination of predicting algorithms, significantly increased predictive performance. Similar improvements were obtained for variants in SCN2A and KCNQ2. The predictive performance of the recently developed ensemble tools, such as Mendelian clinically applicable pathogenicity, combined annotation-dependent depletion and Eigen, was also improved dramatically by application of the strategy with molecular sub-regional stratification. The prediction scores of SCN1A variants showed linear correlations with the degree of functional defects and the severity of clinical phenotypes. This study highlights the need of individualized optimization with molecular sub-regional stratification for each gene in practice.