DNM1 encephalopathy: A new disease of vesicle fission

Abstract

Objective: To evaluate the phenotypic spectrum caused by mutations in dynamin 1 (DNM1), encoding the presynaptic protein DNM1, and to investigate possible genotype-phenotype correlations and predicted functional consequences based on structural modeling.

Methods: We reviewed phenotypic data of 21 patients (7 previously published) with DNM1 mutations. We compared mutation data to known functional data and undertook biomolecular modeling to assess the effect of the mutations on protein function.

Results: We identified 19 patients with de novo mutations in DNM1 and a sibling pair who had an inherited mutation from a mosaic parent. Seven patients (33.3%) carried the recurrent p.Arg237Trp mutation. A common phenotype emerged that included severe to profound intellectual disability and muscular hypotonia in all patients and an epilepsy characterized by infantile spasms in 16 of 21 patients, frequently evolving into Lennox-Gastaut syndrome. Two patients had profound global developmental delay without seizures. In addition, we describe a single patient with normal development before the onset of a catastrophic epilepsy, consistent with febrile infection-related epilepsy syndrome at 4 years. All mutations cluster within the GTPase or middle domains, and structural modeling and existing functional data suggest a dominant-negative effect on DMN1 function.

Conclusions: The phenotypic spectrum of DNM1-related encephalopathy is relatively homogeneous, in contrast to many other genetic epilepsies. Up to one-third of patients carry the recurrent p.Arg237Trp variant, which is now one of the most common recurrent variants in epileptic encephalopathies identified to date. Given the predicted dominant-negative mechanism of this mutation, this variant presents a prime target for therapeutic intervention.

Figure 1. Locations of identified variants in DNM1 protein

DNM1 protein (NP_004399) domain structure with locations of variants identified in patients (listed above the image) and controls from population databases such as ESP, ExAC, and 1000Genomes (listed below the image). All members of the dynamin family contain the GTPase domain involved in GTP binding and hydrolysis, the middle domain and GED required for oligomerization and stimulation of the GTPase activity, the PH domain for lipid binding, and the PRD, which interacts with Src-homology-3 domain–containing proteins (figure generated with IBS). DNM1 = dynamin 1; ESP = Exome Sequencing Project; ExAC = Exome Aggregation Consortium; GED = GTPase effector domain; PH = pleckstrin-homology; PRD = proline-rich domain.

Figure 2. Functional consequences of DNM1 mutations

(A) The entire DNM1 monomer is shown as a cartoon and colored by the GTPase domain (peach), middle domain (white), GED (teal), and PH domain (gold). ExAC missense alteration (blue spheres) and variants discussed in this study (magenta spheres) are shown. (B) Close-up view of the GTPase domain GTP-binding surface with detailed descriptions of the mode of anticipated protein disruption from select observed variants (p.Thr65Asn, p.Ala177Pro, p.Lys206Asn, p.Arg237Trp). (C) Detailed description of the mode of anticipated protein disruption of observed middle domain variants (p.Gly359Ala and p.Gly397Asp). DNM1 = dynamin 1; ExAC = Exome Aggregation Consortium; GED = GTPase effector domain; PH = pleckstrin-homology.