More Epilepsy Cases Linked to the Mosaic Brain

Mutations that occur only in a subset of brain cells may account for a significant fraction of specific types of epilepsy, according to a new study led by researchers in the Institute for Genomic Medicine (IGM) and Department of Neurology at Columbia University Vagelos College of Physicians and Surgeons.

The study adds to a growing body of evidence that certain genetic mutations that arise during embryonic brain development–and are therefore present only in a subset of brain cells–increase the risk of developing epilepsy. The research will be published in the Annals of Neurology.

Biology textbooks tell us that every cell in our body contains the same DNA. But in recent years, genomic analyses have begun to paint a different picture. Instead of containing cells that are genetically identical, our bodies are a mosaic of cells. Mosaicism arises when errors occur as our cells divide during early development and throughout our lifetime, creating groups of cells with slightly different genetic identities. If a mutation occurs very early in embryonic development, it may appear in most cells in our body. But if a mutation occurs later, it can be restricted to certain tissue types, a small number of cells, or, in some cases, just one single cell.

The brain is no different. Most scientists agree that most, if not all people, will have some sets of genetically distinct brain cells, but we are still learning about the extent of mosaicism in the human brain and how these mutations may contribute to disease.

For neurologists, the reality of mosaicism in the brain has made identifying the genetic cause of a patient’s epilepsy more challenging. “We generally rely on DNA testing of blood cells to find genetic mutations in epilepsy patients,” says Columbia neurologist Melodie Winawer, MD, lead author of this work. “But if the mutation exists only in brain cells, or some brain cells, we cannot find it with genomic analysis of blood cells.”

In this study, geneticists sequenced DNA from brain tissue and blood samples of 56 individuals with intractable neocortical epilepsy, a type of epilepsy that doesn’t respond well to any anti-seizure medication. Nearly 9 percent of the study participants were found to have a subset of their brain cells with a small genetic difference in the gene SCL35A2 that was not detectable in the blood and very likely responsible for the seizures.

SLC35A2 encodes a protein that is partly responsible for glycosylation of proteins and lipids that are important for proper brain functioning. This gene was previously reported to cause a very rare and severe type of epilepsy when the mutation is present in all or most of the cells of the body.

“This study tells us that mutations in the same gene and likely the same biological alterations give rise to very different types of epilepsy depending on when the mutation occurs during development,” says Erin Heinzen, PhD, a human geneticist at the IGM and a senior author of this study. “This suggests that if we can find an effective therapy for the rare epilepsy associated with SLC35A2 mutations, it may also benefit other individuals with difficult-to-treat epilepsies.

Identifying the responsible gene is a crucial first step toward developing new treatments for patients with intractable neocortical epilepsy. The next phase of research will focus on studying the biologic effects of these mutations and how they give rise to seizures.


Melodie Winawer is an associate professor of neurology (in the Gertrude H. Sergievsky Center) at the Columbia University College of Physicians and Surgeons.

Erin Heinzen is the deputy director of the Institute for Genomic Medicine and assistant professor of pathology & cell biology at the Columbia University Vagelos College of Physicians and Surgeons.

Other senior authors of this work included Ann Poduri at Children’s Hospital of Boston and Peter Crino at the University of Maryland.

Other Columbia authors: Nicole G. Griffin, Jorge Samanamud, Evan H. Baugh, Dinesh Rathakrishnan, Catherine Schevon, Sameer A. Sheth, Guy M, McKhann, Colin D. Malone, Danielle K. McBrian, Alison M. Pack, Cigdem I. Akman, and Peter D. Canoll.

This work was funded by the National Institutes of Health (R01NS089552, R01NS094596, NS053998, NS077274, NS077303, NS077276, P30HD018655), the Institute for Genomic Medicine, and Boston Children's Hospital. See paper for full details.

Sameer Sheth reports personal fees from Boston Scientific and personal fees from Medtronic.