From splitting GLUT1 deficiency syndromes to overlapping phenotypes.

INTRODUCTION: Glucose transporter type 1 deficiency syndrome (GLUT1DS) is a rare genetic disorder due to mutations or deletions in SLC2A1, resulting in impaired glucose uptake through the blood brain barrier. The classic phenotype includes pharmacoresistant epilepsy, intellectual deficiency, microcephaly and complex movement disorders, with hypoglycorrhachia, but milder phenotypes have been described (carbohydrate-responsive phenotype, dystonia and ataxia without epilepsy, paroxysmal exertion-induced dystonia). The aim of our study was to provide a comprehensive overview of GLUT1DS in a French cohort. METHODS: 265 patients were referred to the French national laboratory for molecular screening between July 2006 and January 2012. Mutations in SLC2A1 were detected in 58 patients, with detailed clinical data available in 24, including clinical features with a focus on their epileptic pattern and electroencephalographic findings, biochemical findings and neuroimaging findings. RESULTS: 53 point mutations and 5 deletions in SLC2A1 were identified. Most patients (87.5%) exhibited classic phenotype with intellectual deficiency (41.7%), epilepsy (75%) or movement disorder (29%) as initial symptoms at a medium age of 7.5 months, but diagnostic was delayed in most cases (median age at diagnostic 8 years 5 months). Sensitivity to fasting or exertion in combination with those 3 main symptoms were the main differences between mutated and negative patients (p < 0.001). Patients with myoclonic seizures (52%) evolved with more severe intellectual deficiency and movement disorders compared with those with Early Onset Absence Epilepsy (38%). Three patients evolved from a classic phenotype during early childhood to a movement disorder predominant phenotype at a late childhood/adulthood. CONCLUSIONS: Our data confirm that the classic phenotype is the most frequent in GLUT1DS. Myoclonic seizures are a distinctive feature of severe forms. However a great variability among patients and overlapping through life from milder classic phenotype to paroxysmal-prominent- movement-disorder phenotype are possible, thus making it difficult to identify definite genotype-phenotype correlations.

GRID2 mutations span from congenital to mild adult-onset cerebellar ataxia.

OBJECTIVES: In a large family of Algerian origin, we aimed to identify the genetic mutation segregating with simultaneous presence of adult-onset, paucisymptomatic, slowly progressive, cerebellar ataxia in 7 adults and congenital ataxia in 1 child, and then to assess the involvement of GRID2 mutations in 144 patients with congenital cerebellar ataxia. METHODS: We used a combined approach of linkage analysis and whole-exome sequencing in one family, and a targeted gene panel sequencing approach in 144 congenital ataxias. RESULTS: In the large family with spinocerebellar ataxia, we identified a missense mutation (c.1966C>G/p.Leu656Val) in the GRID2 gene, in a heterozygous state in adults, and in a homozygous state in one child with congenital ataxia, compatible with a semidominant transmission pattern. In 144 patients affected with congenital ataxia, we identified 2 missense de novo GRID2 mutations in 2 children (c.1960G>A/p.Ala654Thr, c.1961C>A/p.Ala654Asp). They affect the same amino acid as the previously described Lurcher mutation in mice; the variant in the large family concerns a nearby amino acid. CONCLUSIONS: In humans, GRID2 had only been involved in ataxia through complete loss-of-function mutations due to exon deletions. We report the first point mutations in this gene, with putative gain-of-function mechanisms, and a semidominant transmission as was observed in the Lurcher mice model. Of note, cerebellar ataxia is the core phenotype, but with variable severity ranging from very mild adult-onset to congenital-onset ataxias linked to both the heterozygous and homozygous state of the variant, and the position of the mutation.