Novel CSNK2B Mutation Sheds Light on Epilepsy Severity in Rare Syndrome

Researchers in China have identified a previously unrecorded mutation in the CSNK2B gene in three members of a single family with early-onset seizures, and their broader analysis of more than one hundred cases suggests that where a mutation sits on the protein may matter more for epilepsy severity than what kind of mutation it is. The study, published in Acta Epileptologica, combined the family’s detailed clinical workup with a synthesis of nearly all previously reported patients carrying pathogenic CSNK2B mutations, offering one of the most complete pictures yet of the epileptic spectrum tied to this gene.

CSNK2B encodes the beta subunit of casein kinase 2, a highly conserved serine/threonine kinase involved in signal transduction, metabolism, replication, transcription and translation. The kinase typically operates as a tetramer built from two catalytic subunits and two regulatory beta subunits. The human CSNK2B protein spans 215 amino acids and contains five annotated functional regions: a KEN box at residues 32 to 40, a destruction box motif at residues 47 to 55, an acidic region at residues 55 to 77, a zinc finger domain at residues 105 to 140, and a CK2alpha-interaction region at residues 181 to 203. Mutations in the gene are known to cause Poirier-Bienvenu neurodevelopmental syndrome, an autosomal dominant disorder also called CSNK2B encephalopathy, characterized by early-onset seizures and variably impaired intellectual development with highly heterogeneous presentation.

The new family came from a screening of 200 Chinese families with unexplained epilepsy recruited at the Affiliated Hospital of Zunyi Medical University between May 2015 and August 2025. Using trio-based whole-exome sequencing of the proband and both parents, followed by Sanger validation, the team identified a nonsense mutation, c.240T>G, p.Tyr80*, in three affected relatives: a two-year-old girl, her thirteen-year-old sister, and their father. The mother and an unaffected brother did not carry the variant. The mutation was absent from the gnomAD v4.1.0 population database, the affected amino acid is highly conserved across species, and under American College of Medical Genetics and Genomics criteria it was classified as pathogenic.

The three family members showed markedly different courses despite sharing the same variant. The proband developed focal seizures evolving to bilateral tonic-clonic seizures at three months of age, initially striking five to six times a day, but her development remained age-appropriate and her seizures have been fully controlled for more than two years on a combination of oxcarbazepine and levetiracetam. Her older sister had neonatal-onset focal seizures that persisted into adolescence, with mild intellectual impairment and learning difficulties despite normal motor and language development. The father’s seizures began at six months, remitted after roughly four years, and he developed normally. Electroencephalography in the proband revealed diffuse slowing and multifocal epileptiform discharges with seizures originating frontally, while her brain MRI was normal.

To place this family in context, the researchers reviewed published reports of patients carrying pathogenic or likely pathogenic CSNK2B mutations, using the Human Gene Mutation Database as a starting point and verifying mutation types against the original papers. After merging duplicate records and excluding cases with insufficient phenotypic detail, 127 individuals with CSNK2B mutations were identified, including the three new patients. Of these, 103 had sufficiently detailed epilepsy-related data for analysis; 93 carried de novo mutations, with the remainder inherited or of unknown origin.

The pooled data confirmed epilepsy as the dominant clinical feature. Seizures occurred in 94.2 percent of patients, epilepsy was diagnosed in 87.4 percent, and 87.2 percent of those with known onset dates developed seizures before 24 months of age, with a median onset of 7.5 months. Among the 85 cases with clearly recorded seizure types, generalized tonic-clonic seizures were most common at 49 cases, followed by focal seizures in 26, myoclonic seizures in 17, and absence seizures in 12; most patients, 72.9 percent, had a single seizure type. Electroencephalographic abnormalities appeared in 82.1 percent of patients with complete records, typically generalized or bilateral multifocal spike-and-wave discharges with frontal predominance, while brain MRI abnormalities were less frequent at 35.2 percent, with delayed myelination the most common finding. Neither test proved disease-specific.

When the team grouped mutations by type, the results were largely uninformative. Missense and non-missense mutations showed no significant differences in seizure, epilepsy or drug-resistant epilepsy rates, and comparisons across nonsense, frameshift and splice-site categories were likewise not statistically significant. Splitting missense mutations by Grantham score into conservative and radical substitution groups also revealed no meaningful differences in epilepsy or drug-resistant epilepsy rates. These findings align with earlier suggestions that broad mutation grouping alone has limited value in predicting epilepsy severity in CSNK2B-related disorders, although the authors note that Grantham scores capture only static physicochemical differences and cannot fully predict complex outcomes in living cells.

Protein location told a different story. Mutations outside the five annotated canonical domains were associated with significantly higher epilepsy penetrance than mutations within domains, 95.7 percent versus 80.4 percent, a difference that reached statistical significance. Outside-domain mutations clustered mainly in the N-terminal region and the linker region, which may contain a conserved helical groove contributing to protein-protein interactions within the CK2beta dimer. Within domains, epilepsy penetrance was complete in the D-box and CK2alpha-interaction regions, 85.0 percent in the zinc finger domain, 69.2 percent in the KEN box, and 66.7 percent in the acidic region. Drug-resistant epilepsy frequencies were highest in the KEN box and CK2alpha-interaction domains, both at 66.7 percent, followed by the acidic region at 62.5 percent, and were lowest in the D-box at 25.0 percent and the zinc finger domain at 11.8 percent. Notably, most nonsense mutations sat outside the annotated domains, while seven of the eight nonsense mutations within domains clustered in the D-box, a pattern the authors speculate may explain why D-box mutations, though always accompanied by epilepsy, more often act through truncation and haploinsufficiency, whereas KEN box mutations, more often missense or other alterations, may produce aberrant proteins that exert dominant-negative effects on the holoenzyme and raise drug-resistance risk.

The authors stress that these interpretations remain hypotheses. No functional studies were performed on the new p.Tyr80* mutation, the dataset is literature-derived and retrospective, and several subgroups such as start-loss, copy-number deletion and in-frame mutations were too small for robust inference. No adjustment for multiple comparisons was made, so all inferential results should be read cautiously. Questions about nonsense-mediated mRNA decay also remain open: premature termination codons in the last exon or within 55 base pairs of the last exon-exon junction can escape degradation, and among the summarized cases only one mutation, p.Lys208Glnfs*38, fell in that escape region and presented with drug-resistant epilepsy, but confirming the transcript-level consequences will require patient-sample experiments. Even patients sharing the identical p.Asp32Asn mutation showed different phenotypes, pointing to possible modifier genes or genetic background effects on penetrance.

For clinicians, the practical message is that a CSNK2B diagnosis does not by itself predict outcome. The spectrum runs from seizure-free individuals and febrile seizures, reported in 6.8 percent of the analyzed cohort, through drug-responsive epilepsy to severe drug-resistant disease, and the three new family members, with their favorable prognoses despite early onset, illustrate that variability vividly. The authors call for larger cohorts and functional assays to validate the domain-level patterns and the proposed mechanisms linking mutation position, protein truncation, dominant-negative effects and epilepsy severity in this rare neurodevelopmental syndrome.

Subject of Research: Genotype-phenotype correlations in epilepsy associated with CSNK2B mutations and Poirier-Bienvenu neurodevelopmental syndrome

Article Title: CSNK2B-associated epilepsy: family report and literature review

Article References: Zhang, J., Li, Y., Xiong, Y., Pan, S., Yang, L., Li, R., & Tian, M. (2026). CSNK2B-associated epilepsy: family report and literature review. Acta Epileptologica, 8(1), Article 35. https://doi.org/10.1186/s42494-026-00278-y

Image Credits: AI Generated

DOI: 10.1186/s42494-026-00278-y

Keywords: CSNK2B, Poirier-Bienvenu neurodevelopmental syndrome, epilepsy, genetics, whole-exome sequencing, drug-resistant epilepsy, nonsense mutation, casein kinase 2, genotype-phenotype correlation, neurodevelopmental disorder, seizure onset, protein domains