Monoallelic variants in BRSK1 are associated with a neurodevelopmental disorder with or without epilepsy.

BRSK1

Partial loss of function of BRSK1, an AMPK-related kinase: the three patient-derived alleles tested in Drosophila only partially rescue the phenotypes of the null mutant of the ortholog sff

BRSK1 encodes an AMPK-related serine/threonine kinase (AMP-activated protein kinase), required for neuronal polarization and synaptic function. A first individual carrying a BRSK1 variant was identified using AI-MARRVEL in the Texome Project, which provides genomic diagnosis to financially disadvantaged individuals in Texas, and nine individuals harbouring rare heterozygous variants were then found through GeneMatcher; affected individuals present with developmental delay and variable phenotypes including anxiety, attention-deficit hyperactivity disorder (ADHD), autism and seizures, and all variants (missense and nonsense) are predicted to be deleterious by pathogenicity prediction tools. In Drosophila, loss of sff, the ortholog of BRSK1, gives viable flies with severe locomotor impairment, bang and heat sensitivity and a short lifespan, together with overgrowth of the neuromuscular junctions and elevated levels of Futsch (MAP1B), a microtubule-associated protein. Expression of the human reference cDNA rescues these phenotypes, whereas the three patient-derived alleles (p.Ile202Val, p.Arg237Cys and p.Thr406Ile) provide only partial rescue and fail to normalise neuromuscular junction morphology or Futsch levels, suggesting partial loss-of-function alleles. The authors conclude that heterozygous loss of BRSK1 leads to a neurodevelopmental syndrome with variable phenotypes.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The strength of the paper lies in convergence: a non-specific but recurrent human phenotype (developmental delay, anxiety, ADHD, autism, seizures) and a graded rescue in Drosophila, complete with the human reference cDNA and only partial with the three patient-derived alleles. It is, however, a second description of the association: a series published in Epilepsia in 2025 (six novel variants in seven probands among 394 epilepsy cases explored by trio sequencing, with a knockout mouse model) had already proposed BRSK1 haploinsufficiency, and this one supports it with another model organism. Since the gene is already covered by any exome or genome, the issue is interpretation: this functional assay could serve as an argument for classifying BRSK1 missense variants that have remained of uncertain significance, provided that only three were modelled and that the abstract specifies neither the de novo status nor familial segregation.

Analysis by Dr Thibaut Benquey