The Genotypic and Phenotypic Spectrum of GOSR2 Mutations: Clinical and Pathophysiological Insights

North Sea-Progressive Myoclonus Epilepsy (NS-PME) is a progressive neurological disorder, initially only associated with the homozygous GOSR2 founder mutation (c.430G>T; p.Gly144Trp). Clinical symptoms include untreatable early-onset ataxia, cortical myoclonus and epilepsy. Recently, the spectrum of GOSR2 mutations and associated phenotypic variability has expanded. To improve care and to facilitate genotype-phenotype predictions for NS-PME patients, we systematically reviewed all reported GOSR2 mutations, clinical phenotypes, and pathophysiological findings. A narrative review literature search was conducted in PubMed, EMBASE, and Web of Science (1985-August 2024) using the keywords "GOSR2", "GS27 protein", "Bos1", and "Membrin". Only studies in English and specifically studies on GOSR2 function, pathogenic variants, clinical manifestations, and potential therapies were included. A total of 42 patients with 11 different GOSR2 mutations were identified. Three main phenotypes were observed: progressive myoclonus ataxia/epilepsy (PMA/PME), congenital muscular dystrophy, and hearing loss. Orthopedic abnormalities were frequently reported. Intercurrent infections or fever often led to a worsening of symptoms. Glycosylation defects were reported in several compound heterozygous GOSR2 variants. Molecularly, GOSR2 mutations result in (partial) loss of function of the GOSR2/SNARE complex, with mutation severity and the involvement of specific isoforms contributing to phenotypic variability. GOSR2 mutations lead to progressive neurological disorders, primarily characterized by myoclonus ataxia/epilepsy, muscular dystrophy, and hearing loss. The genotypic background of NS-PME has expanded with pathogenic biallelic GOSR2 variants beyond the original homozygous founder mutation. Understanding the clinical spectrum and molecular mechanisms of GOSR2-related diseases may facilitate more targeted treatment strategies as well as better-informed phenotype predictions.