Lab techniques and bioinformatic tools are essential for helping physicians and scientists distinguish between the benign and disease-causing SCN1A gene variants associated with Dravet syndrome and other disorders, according to a recent review study.
The review, “SCN1A variants from bench to bedside — improved clinical prediction from functional characterization,” was published in Human Mutation.
The SCN1A gene provides instructions for making a sub-unit of a sodium channel that is essential for the generation and transmission of electrical signals in the brain.
“Variants in the SCN1A gene are associated with a wide spectrum of diseases ranging from genetic epilepsy with febrile seizures plus (GEFS+) and familial hemiplegic migraine (FHM) to the severe childhood epilepsy Dravet syndrome (DS),” the researchers said.
However, not all SCN1A variants are considered pathogenic, or able to trigger the onset of a disease. Indeed, some may even be considered benign — when their effects are either non-apparent or very mild — or of unknown significance, meaning that their ability to induce disease is unknown.
Over the last years, scientists and clinicians have struggled to find ways to quickly tell if a given SCN1A variant is pathogenic or not. On top of that, current bioinformatic prediction tools — which may be useful for making that first determination — then fall short in predicting disease severity associated with a given variant.
“Functional data are often not available to complement variant interpretation as these are labor intensive and costly to obtain,” the researchers said.
“Although thousands of SCN1A variants are reported, only a small fraction have been functionally assessed,” they said.
In this review study, the investigators focused on describing the functional assays — or investigative procedures — that have been used thus far to characterize SCN1A gene variants.
First, the team used Pubmed — the database of the U.S. National Library of Medicine at the National Institutes of Health — to look for studies published through February 2019 that described the characterization of SCN1A variants. Next, the researchers decided to focus on 58 different missense variants that had been functionally characterized by patch-clamp experiments between 2002 and 2019.
Missense variants are those arising from a modification of a single nucleotide — the building blocks of DNA — that is enough to change a protein’s composition; patch-clamp is a technique that allows scientists to study the electrical properties of neurons.
Almost half of these variants (45%) were directly associated with Dravet, followed by a third (36%) that were associated with a range of disorders — DS, GEFS+ and FS+ — and 9% that were linked to FHM. Functional analyses also showed that more than half of the variants (71%) were the loss-of-function type, whereas only 8% were gain-of-function variants.
Loss-of-function variants are those associated with a reduction in the activity of a protein, while gain-of-function variants are those linked with an increase in protein activity.
Although bioinformatic in silico analyses — computer-assisted reviews of pharmacologic processes — failed to distinguish variants that were associated with specific disorders, it predicted with 89% accuracy if these variants would be pathogenic or not.
Bioinformatic in silico biological research involves bringing mathematical, statistical and/or computational tools to the study of biological data.
In contrast, patch-clamp experiments done in cells cultured in a lab dish highlighted functional differences between these variants. Those differences, in turn, were directly associated with disease severity.
“Those presenting with milder phenotypes [disease manifestations] retained a degree of [sodium] channel function measured as residual whole‐cell [electric] current, whereas those without any whole‐cell [electric] current were often associated with DS,” which is typically more severe, the investigators said.
“Both, in silico prediction and in vitro functional data are important for variant classification,” the team said. “Electrophysiological testing of [sodium] channel function contributes to the prediction of disease severity, whereas in silico tools are useful in distinguishing benign from pathogenic variants.”
“Being able to predict the disease phenotype [symptoms] from functional data may facilitate swift appropriate treatment of epilepsies that are predicted to be severe. This may involve the use of medications that have been shown beneficial in DS and avoiding those that are known to lead to exacerbation of seizures, such as sodium channel blockers,” they concluded.