When brain development gets off to a bad start, the consequences are lifelong.
One example is a condition called SCN2A haploinsufficiency, in which children are born with just one functioning copy of the SCN2A gene — instead of the normal two. They develop defects in the connections, or synapses, between some of their brain cells that affect how well they can send signals. As a result, they do not learn to speak, and many of them experience seizures.
Now, scientists at UC San Francisco have used a version of the gene-editing technology CRISPR to ameliorate some of these problems in mice, which can be engineered to carry the same mutation that humans do. But rather than trying to edit the defective copy of the gene, the scientists just turned up the volume on the healthy one.
The procedure worked in mice that were roughly equivalent in age to 10-year-old children, a clue that the brain may still be amenable to treatment well after much of its development has been completed. This is likely because SCN2A haploinsufficiency compromises how the brain fine-tunes its signals, but it does not affect other aspects of brain development.
The study, which was supported by the National Institutes of Health (NIH), appears in Nature on Sept. 17.
“We were surprised to see that the anatomy of the brain is totally intact — the synapses are there, but they fail to mature when there isn’t enough SCN2A,” said Kevin Bender, PhD, a professor in the UCSF Weill Institute for Neurosciences and co-senior author of the study. “By ramping up SCN2A levels in the brain, we brought those synapses online and restored signaling that prevented seizures.”
Two copies of SCN2A
Nerve cells (pink) with two functioning copies of the SCN2A gene easily connect with other nerve cells and produce normal-sized, short wires (green).
One copy of SCN2A
Nerve cells with only one functioning copy of the SCN2A gene grew long wires that were unable to make good connections.
Restored levels of of SCN2A
When treated with CRISPRa, these cells produced normal levels of the SCN2A protein, despite having only one functioning copy of the gene. These cells produced normal-length wires that could easily connect with other nerve cells.
Images by Tamura et al, Nature.
A lighter touch from CRISPR
More than a decade ago at UCSF, a new type of CRISPR was developed by Jonathan Weissman, PhD. It could find a gene and trigger its expression without creating DNA edits, and was dubbed CRISPRa, for “activation.”
“This can compensate for the shortage of a gene that results from having just one good copy,” said Nadav Ahituv, PhD, director of the UCSF Institute for Human Genetics, professor in the Department of Bioengineering and Therapeutic Sciences, and co-senior author of the paper.
In 2018, a team led by Ahituv used CRISPRa to treat a model of severe obesity in mice that was caused by the loss of one copy of an obesity-related gene.
Bender heard about Ahituv’s success with CRISPRa and obesity, and realized it might also resolve the neurodevelopmental issues that come with insufficient SCN2A. The two UCSF scientists teamed up to see if it could work.
Too little SCN2A, too little learning
SCN2A haploinsufficiency, which results in half the normal level of SCN2A, can lead to epilepsy, neurodevelopmental delay, and autism. Less SCN2A protein, in turn, means that brain cells cannot adjust their signals, like turning the dial of a radio to find the right station. Such adjustment is the basis for how the brain learns.
“This persistent function of SCN2A is critical for your ability to be human and experience and learn about the world,” Bender said.
Making a model of disease
Last year, Bender’s team discovered a quirk in the eye movements of mice engineered to have SCN2A haploinsufficiency — something they later observed in children with the disorder. The quirk could serve as the basis for an easy clinical test for SCN2A haploinsufficiency in humans. The finding also meant they could study the disorder more readily in mice, since, according to Bender, it’s rare for mutations that are found in people with autism to result in similar problems in mice.
The Bender laboratory meticulously tested what this lack of SCN2A did to mice. Having less SCN2A made the mice prone to seizures and altered signals in their brain.
“Thanks to the beautiful characterization of physiology and behavior in the mice done by Kevin and his lab, we knew what needed to be restored to show that our CRISPRa approach was working,” Ahituv said.
CRISPRa restores SCN2A levels and healthier brain function
A therapy like this in the clinic could improve their ability to talk and even live independently,” Bender said. “We hope our work can help make these dreams a reality.”
Ahituv’s lab designed the CRISPRa to find the healthy copy of SCN2A and dial it up so it produced the same amount of protein that two copies of the gene would. And Bender’s lab tested it in mice with SCN2A haploinsufficiency.
The CRISPRa intervention increased levels of SCN2A throughout the brain. The scientists saw normal amounts of SCN2A in nerve cells. But these animals had grown up with too little SCN2A from birth, and the mice were now a few weeks old, which is akin to a pre-adolescent human.
Would the additional protein be enough to treat the condition?
Remarkably, the additional SCN2A protein gave new life to the existing neural connections in these mice. Their brain signals looked normal, and they no longer were prone to seizures.
The intervention worked both when the CRISPRa was introduced directly to the brain and also when it was injected into the blood. Regel Therapeutics has licensed this technology from UCSF to treat patients with SCN2A haploinsufficiency disorders.
A key consideration will be how the brain responds to adding more SCN2A protein. The team was encouraged to find that it didn’t harm healthy mice, those with two functioning copies of the gene.
“Too much of any protein might cause a lot of trouble,” Ahituv said. “We found that there is a natural limit to levels of this protein in SCN2A mice, but future therapies will need to confirm the safety of the approach in humans.”
For such a severe disorder, the prospect of a treatment that works well into childhood would be miraculous.
“A therapy like this in the clinic could improve their ability to talk and even live independently,” Bender said. “We hope our work can help make these dreams a reality.”
Authors: Other UCSF authors are Serena Tamura, PhD, Andrew D. Nelson, PhD, Perry W.E. Spratt, PhD, Elizabeth C. Hamada, Xujia Zhou, PhD, Henry Kyoung, Zizheng Li, Coline Arnould, PhD, Vladyslav Barskyi, Beniamin Krupkin, MS, Kiana Young, Jingjing Zhao, PhD, Stephanie S. Holden, PhD, Atehsa Sahagun, Caroline M. Keeshen, PhD, Roy Ben-Shalom, PhD, Sunrae E. Taloma, Selin Schamiloglu, MS, Ying C. Li, MD, PhD, Jeanne T. Paz, PhD, Stephan J. Sanders, PhD, and Navneet Matharu, PhD. For all authors, see the paper.
Funding: This work was supported by the National Institutes of Health (P30 DK063720, S10 1S10OD021822-01, R01 MH125978: KJB; F32 MH125536 and K99 MH135209: ADN; R01 NS078118 and R01 NS121287: JTP; R01 MH115045, R01 NS 108874, and R01 MH118298: JQP; T32 GM007449: SSH); SFARI (629287, 513133), the Broad Institute Target Practice Initiative, the Autism 955 Science Foundation, the Weill Neurohub Investigator Program, the Natural Sciences and Engineering Research Council of Canada, the Ford Foundation Dissertation Fellowship, and the Weill Foundation.
Disclosures: Ahituv and Matharu are co-founders of Regel Therapeutics. Ahituv, Matharu, and Bender are scientific advisors to Regel Therapeutics. For all disclosures see the paper.