Alternating Hemiplegia of Childhood: Retrospective Genetic Study and Genotype-Phenotype Correlations in 187 Subjects from the US AHCF Registry.

Mutations in ATP1A3 cause Alternating Hemiplegia of Childhood (AHC) by disrupting function of the neuronal Na+/K+ ATPase. Published studies to date indicate 2 recurrent mutations, D801N and E815K, and a more severe phenotype in the E815K cohort. We performed mutation analysis and retrospective genotype-phenotype correlations in all eligible patients with AHC enrolled in the US AHC Foundation registry from 1997-2012. Clinical data were abstracted from standardized caregivers' questionnaires and medical records and confirmed by expert clinicians. We identified ATP1A3 mutations by Sanger and whole genome sequencing, and compared phenotypes within and between 4 groups of subjects, those with D801N, E815K, other ATP1A3 or no ATP1A3 mutations. We identified heterozygous ATP1A3 mutations in 154 of 187 (82%) AHC patients. Of 34 unique mutations, 31 (91%) are missense, and 16 (47%) had not been previously reported. Concordant with prior studies, more than 2/3 of all mutations are clusteredin exons 17 and 18. Of 143 simplex occurrences, 58 had D801N (40%), 38 had E815K(26%) and 11 had G947R (8%) mutations [corrected].Patients with an E815K mutation demonstrate an earlier age of onset, more severe motor impairment and a higher prevalence of status epilepticus. This study further expands the number and spectrum of ATP1A3 mutations associated with AHC and confirms a more deleterious effect of the E815K mutation on selected neurologic outcomes. However, the complexity of the disorder and the extensive phenotypic variability among subgroups merits caution and emphasizes the need for further studies.

Agenesis of the mesencephalon and metencephalon with cerebellar hypoplasia: putative mutation in the EN2 gene--report of 2 cases in early infancy.

Congenital absence of the midbrain and upper pons is a rare human malformation. We describe two unrelated infants with this anomaly and cerebellar hypoplasia who were born at term but died in early infancy from lack of central respiratory drive. MRI in both cases disclosed the lesions during life. Neuropathological examination, performed in one, included immunocytochemical studies of NeuN, synaptophysin, vimentin, and glial fibrillary acidic protein (GFAP). Autopsy revealed a thin midline cord passing through the clivus, in place of the midbrain; it corresponded to hypoplastic and fused corticospinal tracts with ectopic neural tissue in the surrounding leptomeninges. Some ectopia were immunoreactive for synaptophysin and NeuN and others were nonreactive. The neural surfaces facing the subarachnoid fluid-filled space left by the absent midbrain and upper pons were lined by an abnormal villous ependyma. The architecture of the cerebellar cortex was imperfect but generally normal, and Bergmann glial cells had normal radial processes shown by vimentin and GFAP. Structures of the telencephalon, diencephalon, lower brainstem, and spinal cord were generally well formed, but inferior olivary and dentate nuclei were rudimentary and the spinal central canal was dilated at lumbar levels. The cerebral cortex was normally laminated, but pyramidal neurons of layer 5 were sparse in the frontal lobes. The hippocampus, olfactory system, and corpus callosum were formed. An ectopic lingual thyroid was found and had been associated with hypothyroidism during life. A murine model resembling this dysgenesis is demonstrated by homozygous mutations of the organizer genes Wnt1 or En1, also resulting in cerebellar aplasia, and En2, associated with cerebellar hypoplasia. These genes are essential to the formation of the mesencephalic neuromere and rhombomere 1 (metencephalon or upper pons and cerebellum). Pax8 has binding sites in the promoter for En2 and is essential for thyroid development. We speculate that in the human, the failure to form a mesencephalon and metencephalon, with cerebellar hypoplasia, results from a mutation or deletion in the EN2 (Engrailed-2) gene.