Bi-allelic PRRT2 variants may predispose to Self-limited Familial Infantile Epilepsy

Abstract

Heterozygous PRRT2 variants are frequently implicated in Self-limited Infantile Epilepsy, whereas homozygous variants are so far linked to severe presentations including developmental and epileptic encephalopathy, movement disorders, and intellectual disability. In a study aiming to explore the genetics of epilepsy in the Sudanese population, we investigated several families including a consanguineous family with three siblings diagnosed with self-limited infantile epilepsy. We evaluated both dominant and recessive inheritance using whole exome sequencing and genomic arrays. We identified a pathogenic homozygous splice-site variant in the first intron of PRRT2 [NC_000016.10(NM_145239.3):c.-65-1G > A] that segregated with the phenotype in this family. This work taps into the genetics of epilepsy in an underrepresented African population and suggests that the phenotypes of homozygous PRRT2 variants may include milder epilepsy presentations without movement disorders.

Introduction

Contrary to severe developmental and epileptic encephalopathies (DEE) and neurodevelopmental disorders with epilepsy that are largely monogenic disorders with a major contribution from de novo mutations, less severe epilepsies are genetically complex, with small fraction of diagnosed individuals harboring pathogenic variants in single genes [1]. Family-based studies facilitated the identification of genes mediating high disease risk [2, 3]. Genetic testing can achieve a diagnostic rate between 45 and 50% in early life epilepsies [4]. As most of these studies and estimates were based on European cohorts, it is conceivable that genetic studies in other populations, particularly populations in which consanguinity is rather prevalent, could help decipher a wider phenotype-genotype spectrum of epilepsy genes. The Sudanese population shows considerable diversity and has a particularly high rate of consanguinity [5, 6]. Prior epidemiological studies showed a considerable burden of epilepsy in Sudan [7]. Aiming to study the genetics of epilepsy in this underrepresented population, we performed sequencing and genotyping in a small cohort of fourteen families with a range of seizure phenotypes and consanguinity. Here, we highlight one consanguineous family that links homozygous PRRT2 variants to self-limited rather than (the known) severe epilepsy presentations.

Subjects and methods

We collected preliminary data on sixty Sudanese families and selected fourteen families for a genetic workup (Table S1). In each family, one proband was tested using whole exome sequencing (WES) whereas several siblings (those diagnosed with epilepsy, and healthy sibling) and their parents were tested using genome-wide arrays. We evaluated short variants from WES, considering both dominant and recessive inheritance in 125 epilepsy genes from OMIM epilepsy-related phenotypic series (PS600669, PS254800 and PS308350), 425 genes from the Genomics England Genetic epilepsy syndromes panel v2.2 [8], and 977 epilepsy genes collected by Wang and colleagues [9] who reviewed known and suspected epilepsy genes. These lists included DEE-related loci and were mostly overlapping (approximately totaling thousand genes). We also considered 3,570 genes that are highly intolerant to Loss-of-Function (LoF) variants (genes with pLI score > 0.9). We used the array data to perform copy number variants analysis (CNVs), homozygosity mapping, and haplotype analysis. Further details on the study methods are provided in the Supplementary.

Results

In this report, we describe a family with neonatal onset epilepsy and a novel PRRT2 presentation. The presenting symptom in all three siblings in this family was tonic seizures at the age of three to four months, described by their mother as episodes of generalized symmetrical body stiffness with abrupt onset, lasting for seconds up to a minute, followed by full recovery. Additionally, the two elder siblings had atonic seizures described as episodes of floppiness involving the whole body, of abrupt onset, lasting maximally for a few minutes. In all three siblings, these seizures were infrequent, associated with confusion and followed by full recovery.

There was no history of abnormal movements or an association with fever or episodes of infection. The elder siblings at school age had satisfactory performance, and there was no impression of an intellectual disability (formal testing was not performed). Pregnancy and birth history were uneventful, and all siblings achieved developmental milestones appropriate for their age. Sleep EEG at five months of age in the eldest sibling showed generalized brain slowing without spike activity, whereas sleep EEG in the other siblings was normal. The two elder siblings were responsive to valproate (VPA) and were off anti-seizure medication (ASM) at the time of evaluation; the youngest was started on phenobarbital shortly before testing. The clinical presentation is summarized in Table 1. Further details are presented in the Supplementary.

Table 1.

Summary of presentations and genetic findings in siblings diagnosed with epilepsy from a Sudanese consanguineous family (E11).

IID	Age, Sex	Onset	Presenting symptom	Related symptoms	EEG	ASM	Response	Genetic test	Genes with a candidate variant
76	8 y, F	3 m	Tonic seizures	Atonic seizures	Generalized slowing	VPA	Responder (off ASM)	Segregation	PRRT2, JMJD1C
77	5 y, M	3 m	Tonic seizures	Atonic seizures	Normal	VPA	Responder (off ASM)	WES	PRRT2, JMJD1C
78	6 m, M	3 m	Tonic seizures	–	NA	PB	Unclassified (just started on ASM)	Segregation	PRRT2

This table presents a summary of the seizure types and candidate variants detected in epilepsy-related genes in three siblings from family E11.

ASM Anti-Seizure Medications, CBZ Carbamazepine, EEG Electroencephalography, d day(s), F female, FID Family Identifier, IID Individual Identifier, M Male, m month(s), NA Not Available, PB Phenobarbitone, VPA Valproate, WES Whole Exome Sequencing, y year(s).

The mother of these siblings did not recall having experienced any episodes of seizures or abnormal movements, whereas the father reported an episode of brief loss of body tone and consciousness around the age of 30 years (falling on a table in front of him while sitting) which was not preceded with any prodromal symptoms. It is thought to have lasted a few minutes, with no previous history or further occurrence of similar episodes. No EEG was performed, and cardiac evaluation was normal. There was a family history of seizure disorders (Fig. 1). This was described as having a comparatively severe nature and was associated with an intellectual disability. It was not possible to obtain an accurate description of the phenotypes as these individuals were not available for evaluation.

Fig. 1. Variants identified in two epilepsy genes in family E11.

The figure shows the pedigree of family E11 and the segregation of two variants in PRRT2 (homozygous) and JMJD1C (heterozygous). PRRT2 variant IVS1 (classified as pathogenic) is in an exon-intron junction, whereas JMJD1C missense variant (classified as benign) changes a conserved amino acid. The individual identifiers of investigated individuals are shown (see Table 1 for the phenotypes summary). The proband investigated using whole exome sequencing (WES) is indicated with a red arrow. Sanger sequencing electropherograms of the IVS1 splice variant are shown for the three siblings diagnosed with epilepsy (E11-76, E11-77, E11-78), their mother (E11-75), father (E11-80), and a sibling not diagnosed with epilepsy (E11-79).

We identified a homozygous canonical splice-acceptor site variation in PRRT2 (NC_000016.10(NM_145239):c.-65-1G > A). This variant was not seen in public population databases as well as in 130 jointly called exomes from Sudanese individuals (Supplementary Methods). The segregation of this variant in the two remaining siblings with epilepsy was confirmed using Sanger sequencing. An unaffected sibling did not carry the variant, whereas their parents were heterozygous carriers (Fig. 1). This variant was not previously reported and was classified as pathogenic based on these criteria: canonical splice acceptor variant in a gene where LOF is a known mechanism of disease (PVS1), absent from controls, or at extremely low frequency and absent in homozygous state if recessive (PM2), and co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease, considered moderate evidence (PP1).

A second, heterozygous missense variant in JMJD1C (NM_032776:p.(Pro506Ala)) was paternally inherited in two affected siblings; it was classified as likely benign based on the following criteria: lack of segregation in affected members of a family (BS4), variant found in a case with an alternate molecular basis for disease (PB5), absent from controls, or at extremely low frequency and absent in homozygous state if recessive (PM2). No relevant CNVs were identified. The remaining individuals in other families are summarized in Table S1 and the candidate variants found in these families are summarized in Table S2.

Discussion

We found a novel pathogenic PRRT2 variant in three Sudanese siblings with a homogenous phenotype of generalized (tonic, atonic) seizures with infantile onset. PRRT2 predisposes to one of the commonest “single-gene” epilepsies [10]. Pathogenic variants in PRRT2 are frequently identified in individuals with familial infantile epilepsy, paroxysmal movement disorders with and without epilepsy, and less frequently in those with sporadic infantile epilepsies, hemiplegic migraine and episodic ataxia [11–15]. The phenotype in this report was consistent with Self-limited Familial Infantile Epilepsy and thus concordant with known PRRT2-related presentations. Bilateral tonic seizures are a known disease presentations, reported for instance in individuals with the recurrent heterozygous frameshift variant c.649dupC [14]. Atonic seizures, seen here in two of three patients, are also a known – though relatively infrequent – presentation [11].

Whereas heterozygous PRRT2 variants are detectable in around 80% of individuals with Self-limited Familial Infantile Epilepsy [10, 11], with penetrance estimated around 75% [15], homozygous variants typically underlie a severe presentation with prolonged episodes of paroxysmal movement disorders, learning difficulties or developmental delay in addition to mild or severe epilepsy [13–16]. However, the presentation in the reported siblings, who carried a bi-allelic splice-site variant, was rather mild in comparison. This could be explained by an arguably weak effect of IVS1-1G > A on splicing or by the existence of other variants not detected through exome sequencing (e.g., other intronic and regulatory variants) that may have additional effect on the splicing [17].

Recent guidelines recommend incorporating the strength of in silico predictions, predicted rescue, proportion of expressed transcripts containing the affected exon, and sequence quality in the evaluation framework of splice alterations [18]. The sequence quality (Sanger validation) and several in silico observations support the variant reported here (Table 2). All brain-expressed isoforms of PRRT2 contain this exon, which contains the start codon and codes for most of the protein sequence. The splice acceptor variant IVS1-1G > A has a low-to-moderate SpliceAI score (0.44) which may reflect a hypomorphic allele with low-to-moderate effect. Splice variants often cause a mixture of multiple functional and non-functional aberrant transcripts [19], which also may explain a partial LoF. The main limitation of this work is the lack of functional validation of these prediction. RNA samples were not available from our patients to examine these effects. It was also not feasible to apply a mini-gene assay to evaluate this variant further.

Table 2.

Evaluation of candidate variants in two epilepsy-related genes in family E11.

Gene Symbol	PRRT2	JMJDC1
Known phenotypes	SeLFIE, DEE, PKD	NDD, ID, ASD
Variant
	GRCh38(chr16):g.29812989 G > A	GRCh38(chr10):g.63214651 G > C
Zygosity	homozygous	heterozygous
Coding change	NC_000016.10(NM_145239.3):c.-65-1G>	NC_000010.11(NM_032776.3):c.1516 C > G
Protein change	pLOF (IVS1)	p.(P506A)
Frequency
gnomAD MAF	0	0
TOPMed MAF	0	0
DiscovEHR MAF	0	0
Sudanese exomes (n = 130)	0	0
Evaluation
GERP + +	4.6	5.1
CADD	31	23.3
Other relevant in silico predictions	rf: 0.94 (D) ada: 1.0 (D) MaxEnt: 8.75 (D) TraP: 0.96 (D) HSF: No significant impact	MTR: 1.08 (T) REVEL: 0.23 (T) ClinPred: 0.99 (D) M-CAP: 0.01 (B)
Other considerations	Sequence quality: high (Sanger validated) pext score: 1 SpliceAI score: 0.44 SpliceAI rescue: 0	–
Segregation	Yes	No
ACMG	Pathogenic, PVS1, PM2, PP1M	Likely Benign, PM2, PB5, BS4
ClinVar accession	SCV002058115	SCV002059214

This table shows the allele frequency in silico evaluation and classification of two variants detected in PRRT2 and JMJDC1. GERP + + RS is a conservation score. MTR is a regional mutational intolerance score. CADD, ClinPred, M-CAP and REVEL are prediction tools for deleteriousness or pathogenicity of nonsynonymous variants. SpliceAI, HSF, TraP, MaxEnt, rf and ada are prediction tools for the deleteriousness or pathogenicity of splice site variants. “D” indicates a score in the pathogenic/deleterious/damaging range whereas “T” indicates score in the tolerated/benign range. SpliceAI cut-off for high sensitivity is 0.2. SpliceAI predicted an acceptor loss- and gain of a rescue site, each with low probability, and with a net probability of 0 indicating no rescue. Pext is the proportion of expressed transcripts, calculated for neuronal tissues. Pext score of 1 indicates that the exon is present in all neuronally expressed transcripts. MAF in Sudanese individuals were estimated in 130 joinlty called exome samples.

ASD Autism Spectrum Disorder, DEE Developmental and Epileptic Encephalopathy, ID Intellectual Disability, MAF Minor Allele Frequency, NDD Neurodevelopmental disorder, PKD Paroxysmal Kinesigenic Dyskinesia, pLOF predicted loss-of-function variant, SeLFIE Self-limited Familial Infantile Epilepsy.

We identified one likely pathogenic variant in PCDH19 and other rare VUS in the other families we studied, and these variants were mostly absent from population databases (Table S2). However, proper variant evaluation requires reliable estimates of population-level allele frequency. The Sudanese population is underrepresented in genomic databases. We harmonized the exomes produced here with exomes of other Sudanese individuals collected through our collaborations (totaling 131 WES samples) and used it to rule out common population polymorphisms. Our efforts continue to increase the size of this dataset to detect rarer population variants. M. E. Ibrahim leads a wider call to the Sudanese investigators to establish a Sudanese Genome Project.

In summary, the current observations from an understudied African populations with known complex genetic ancestry and prevalent consanguinity offered insights into the complexity of genetic inheritance of less severe and self-limited epilepsies. The findings suggest that the phenotypes of homozygous PRRT2 variants may include Self-limited Infantile Epilepsy without paroxysmal movement disorders or developmental aberrations.

Author contributions

Conception and Design: HL, LE, JS, AY, MK, MT, ME, IM, AH, MEI, MS. Primary clinical evaluation: ME, IM, AH, MS, AY. Recruitment and sampling of families: AY, RA, MK, AA, MA, IO, EA, EZ, FE, WA, EOA, EE, MOI, NH, HM, AB. Control datasets: LE, AY, MEI, MA, YB, MOI, MK. Exome Sequencing: JA, JS, MK, LE, AY, MA, YB, MEI, PN, HL. Array genotyping: MT, JS, MK, PN, HL. Data analysis and interpretation: MK, JS, HL. Writing - first draft: MK. Writing - critical revisions: MK, HL, ME, LE, AY. All authors revised and approved the final version of the manuscript. See the Supplementary for a detailed account of individual contributions.

Funding

This work was supported by grants from the ‘Deutsche Forschungsgemeinschaft’ (DFG Research Unit FOR2715; grants Le1030/16-1 and Le1030/16-2) and the ‘Stiftung no epilep’ to Holger Lerche. Mahmoud Koko was supported by a doctoral grant from the ‘Deutscher Akademischer Austauschdienst’ (DAAD program 57214224). LEOE was supported by Princess Nourah bint Abdulrahman University Researchers supporting project number (PNURSP2022R172), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia. MAS was supported by King Saud University Researchers supporting project number (RSP-2020/38), King Saud University, Riyadh, Saudi Arabia. The funding source(s) had no involvement in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

Data availability

The reported variants in family E11 were submitted to ClinVar (SCV002058115; SCV002059214). Exome and genomic array data are not publicly available. Additional details on the study that led to this manuscript, the clinical presentation of the individuals with PRRT2 variants, and genetic testing are available as Supplementary Methods.

Competing interests

The authors declare no competing interests.

Ethics and informed consent

Written informed consent for publication of their clinical details and/or clinical images was obtained from the participants/patients or their parents. All procedures in this study were performed following the standards the 1975 Helsinki declaration and its later amendments as well as the standards of the research committees of the Faculty of Medicine, University of Khartoum (Khartoum, Sudan).

Supplementary information

The online version contains supplementary material available at 10.1038/s41431-024-01541-x.