Authors :
Presenting Author: Kristen Park, MD – University of Colorado School of Medicine Departments of Pediatrics and Neurology
Ilona Krey, MD – Leipzig University
Scott Myers, PhD – Emory University School of Medicine
Sukhan Kim, BS – Emory University
Riley Perszyk, PhD – Emory University
Tim Benke, MD – Children’s Hospital Colorado
Johannes Lemke, MD – University of Leipzig
Dennis Lal, PhD – Department of Neurology, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX, USA
James Allen, BS – Emory University
Ellington McDaniels, BS – Emory University
John Wiedenhoeft, PhD – University of Leipzig
Steve Traynelis, PhD – Emory University School of Medicine
Rationale: Missense, in-frame insertions and deletions, and frameshift variants in the N-methyl-D-aspartate receptor subunit GluN1, encoded by GRIN1, have been associated with a broad spectrum of developmental disorders, neurological deficits, and epileptic encephalopathy (DEE). However, we currently lack an explanation of this broad phenotypic heterogeneity.
Methods: We evaluated phenotypic data of 224 individuals with GRIN1 DEE due to heterozygous (likely) pathogenic missense variants (corresponding to 100 different variants) recruited from our GRIN Registry (https://grin-portal.broadinstitute.org) and the literature. All variants were functionally tested and classified by six electrophysiological/biochemical experiments. Additionally, we developed and applied an independent classification approach, excluding hidden dependency structures. We analyzed clinical, genetic, and functional data of an additional 17 cases with bi-allelic variants in GRIN1, 9 cases with non-missense variants, and 26 cases with heterozygous missense variants of unknown significance.
Results: We observed a clustering of pathogenic missense variants in the ligand-binding domains S1 and S2, the transmembrane domains M2, M3, and M4, as well as the linker domains in between (80/100; 80 %). By contrast, the amino-terminal and carboxy-terminal domains are largely spared of pathogenic missense variants (5/100; 5 %). Functional analysis of the NMDA receptor revealed GoF (gain-of-function) in 53 and LoF (loss-of-function) in 47 variants, confirmed by our independent second classification approach. Individuals with GoF variants are prone to more severe developmental phenotypes, including developmental stagnation or regression, cortical blindness, significantly more seizures, and earlier seizure onset. Polymicrogyria occurs almost exclusively in the context of GoF. LoF variants were often associated with developmental delay and intellectual disability.
Conclusions: This is the largest cohort to date of individuals with GRIN1-related disorder. We show clear phenotypic differences depending on the underlying LoF, GoF, or biallelic null, with increasing phenotypic severity among these three sub-groups. Thus, determining the functional consequence of a variant is essential to suggesting potential treatment strategies and predicting phenotypic severity. This stratification may be exemplary for other DEEs, particularly in the context of channelopathies.
Funding:
This work was supported by a grant from SFARI (732132 to JRL, TAB, SFT) and the NIH (NS111619 to SFT and HD082373 to HY).