Authors :
Ashley Lin, BS – Duke University
Vijaykumar Pillalamarri, PhD – Rosalind Franklin Medical School
Praveen Ramani, M.D. – Duke University School of Medicine
Jad Kourany, HSD – Duke University
David Mueller, PhD – Rosalind Franklin Medical School
Presenting Author: Mohamad Mikati, MD – Duke University
Rationale:
Most patients with ATP1A3 related epilepsy have missense (nonsynonymous) mutations. Many have Developmental and epileptic encephalopathy (DEE). Few have polymicrogyria. Missense mutations cause disease through loss or gain of function, or dominant negative effects. Deletion and splice site, labeled here as non-missense, mutations have other effects: generation of truncated mutant protein leading to non-sense mediated decay (NMD) and haploinsufficiency. Thus, their resultant phenotypes may differ from those of missense mutations. Phenotypes from non-missense ATP1A3 mutations, which occur very rarely, have not been described and need to be characterized for prognostic and physiological implications.Methods:
1) Clinical profile analysis of patients from our ATP1A3-Related-Disease-Registry who have non-missense mutations. 2) Analysis of these mutations’ pathogenicity using the American College of Medical Genetics (ACMG) criteria, ClinVar database and Franklin AI-powered Mutation Interpretation Platform. 3) Molecular sequence alignment and visualization using Clustal W, Aline, RCSB Protein Data Bank at RCSB.org and PyMOL platforms.Results:
5/145 registry patients qualified (5-16 y.o., 3 males). Two had severe DEE with cerebral atrophy without polymicrogyria: one Localization Related and the other Lennox-Gastaut syndrome. Neither had dystonia or hemiplegia. Mutations were c.2562_2565delTGGCinsG (p.G855del) and c.2570_2572delTCT (p.F857del) respectively. Three had a mild Alternating Hemiplegia of Childhood (AHC) phenotype with no epileptic seizures (2 had slow EEG background) and had c.2755_2757del (p.V919del), c.2542+1G >A and c.388_390delGTG (p.V130del) respectively. All variants were classified as pathogenic or likely pathogenic using the Franklin platform and ACMG criteria. Except for c.2542+1G >A, all result in deleting residues (corresponding to Val130, Gly855, Phe857, and Val919) that are highly conserved and within one of 10 transmembrane helices (M1-M10). Val130 is in M2 and its hydrophobic interaction with M4 is important in converting from E1 to E2 conformation necessary for ion transport and the mutation is expected to alter this transition. Gly855 and Phe857 are in M7 and Val 919 in M8 and each of these mutations results in structural distortions that affect the Na+ binding site. The c.2542+1G >A is a splice site mutation in intron 18-exon 19 junction predicted to disrupt the sequence after residue 848 and consequently to NMD and haploinsufficiency.Conclusions:
Non-missense ATP1A3 Mutations cause a phenotype spectrum ranging from severe DEE to mild AHC. This is not too different from phenotypes previously reported with missense mutations. The phenotype of the splice site mutation, being so severe, suggests that haploinsufficiency can cause severe phenotype of DEE without the need of a dominant negative effect. These findings indicate that prognostication for patients with non-missense mutations should depend on the clinical presentation and on the specific mutation and not merely on the class of mutation.Funding: Duke University Funds and CureAHC Donation.