Genotype-phenotype correlations in hyperekplexia: apnoeas, learning difficulties and speech delay

Congenital hyperekplexia is a rare, potentially treatable neuromotor disorder. Three major genes of effect are known, and all three affect glycinergic neurotransmission. Two genes encode for subunits of the postsynaptic inhibitory glycine receptor, GLRA1 encoding the alpha(1) subunit and GLRB encoding the beta subunit. The third, SLC6A5, encodes the cognate presynaptic glycine transporter 2. Ninety-seven individuals had a clinical diagnosis of hyperekplexia confirmed by genetic testing: 61 cases had mutations in GLRA1, 24 cases in SLC6A5 and 12 in GLRB. Detailed retrospective clinical analysis ascertained that all gene-positive cases present in the neonatal period (occasionally prenatally) and that clonazepam is the treatment of choice (95% found it to be efficacious). We confirm that hyperekplexia is predominantly a recessive condition but dominant cases are seen (16%). We found no genetic evidence for 'major' or 'minor' forms of hyperekplexia on a population basis. Thirty-five gene-negative cases were studied for comparison, their cardinal feature was presentation after the first month of life (P < 0.001). In addition to the characteristic 'stiffness, startles and stumbles' of hyperekplexia, apnoea attacks (50 of 89) and delayed development (47 of 92) were frequently reported. Patients with SLC6A5 mutations were significantly more likely to have had recurrent infantile apnoeas (RR1.9; P < 0.005) than those with GLRA1 mutations. Patients with GLRB and SLC6A5 mutations were more likely to have developmental delay (RR1.5 P < 0.01; RR1.9 P < 0.03) than those with GLRA1 mutations; 92% of GLRB cases reported a mild to severe delay in speech acquisition. Molecular modelling of pathogenic mutations demonstrates specific patterns of protein disruption that can be used to predict phenotype severity. The developmental delay in hyperekplexia, and speech acquisition in particular, may represent failure of developmental neural networks or subtle neurogenic migration defects in the absence of presynaptic glycine release. We recommend early genetic testing for symptomatic neonates and possibly preconception counselling for those at risk for GLRB and SLC6A5 mutations, because of the more challenging phenotype.