A problem with the way the brain develops may leave autistic children with too many connections among their brain cells, or neurons. This might make them vulnerable to overstimulation and contribute to their autism symptoms, new research suggests.
During infancy, the number of synapses — the connections that allow neurons to send and receive information — grows rapidly. Later on, during childhood and adolescence, these connections are pruned back in response to learning and interacting with the environment.
In the study, published yesterday in the journal Neuron, researchers from Columbia University Medical Center and colleagues at other institutions looked at the temporal lobe of the brain, an area involved in communication and social behavior. The brains of older children with autism had a greater number of connections in the temporal lobe than the brains of their peers without autism.
Less Pruning in Brains of Kids with Autism
Researchers examined tissue taken from the brains of 26 deceased children between the ages of 2 and 19, half with autism and half without. The researchers counted the dendritic spines, which are the protrusions at the ends of neurons that receive signals across the synapses.
At younger ages, the two groups of children had a similar number of spines. Among typically-developing children, the number of synapses dropped by 41 percent as they grew older. In children with autism, the number decreased by only 16 percent.
Because the younger children had a similar number of connections in their brains, researchers suggest that the problem is not an overproduction of synapses, but instead trouble pruning back the excess.
“While people usually think of learning as requiring formation of new synapses, the removal of inappropriate synapses may be just as important,” the study's senior investigator, David Sulzer, a professor of neurobiology at Columbia, said in a press statement.
The study suggests that decreased pruning is the result of a problem with autophagy, a process the brain uses to clear out old and degraded cells. Biomarkers and proteins in the brain tissue of the children with autism support the idea that a pruning deficiency is the root of the problem.
Drug Restores Synapse Pruning in Mice
While a treatment for autism remains some years away, the researchers were able to show that treatment is possible in mice. They used mice with a rare genetic disease that gives them certain social behaviors that are similar to autism in humans.
In these mice, a hyperactive protein called mTOR interfered with the brain’s ability to eliminate synapses through autophagy. Researchers were able to reverse this problem by giving the mice the drug rapamycin. This not only restored the pruning of synapses, but also reduced the autism-like social behaviors.
It is too soon to know whether a drug like rapamycin, an immunosuppressant with serious side effects, would work successfully in people with autism. Additional studies are needed to see if this research can be translated to people. Other experts in the field of autism research are cautiously optimistic.
“This is the kind of work that the field needs. This is in there at the level of mechanism — from molecules to mind,” said John Foxe, a professor of neuroscience and pediatrics and director of research at the Children’s Evaluation and Rehabilitation Center at Albert Einstein College of Medicine. “The number of areas that this opens up for now, in terms of potential targeted therapeutics, is really amazing. This is exactly what the field should be after.”
More Information Needed on Early Autism
It is difficult to know whether excess synapses cause autism directly or whether they result from the learning experiences of autistic children.
Learning and interacting with the environment plays a big role in reshaping the brain’s synapses. During childhood, typically-developing children go to school, learn many new things, and take in new information. All of this encourages pruning. Autistic children, however, may not have the same experiences, which could affect the level of pruning.
“Do they not prune these dendrites because of the autism, or is the lack of dendritic pruning causing the autism?” Foxe asked. “It’s very hard to know what’s going on there.”
Solving this problem will require better ways to diagnose autism at an earlier age, to see whether symptoms appear before or after the drop in pruning. In addition, because the only way to look at the number of connections between neurons is to examine tissue taken from a deceased child, researchers will need access to a larger number of brains, especially of 2- to 4-year-olds.
“This is work that’s really only been possible in recent times,” Foxe said. “It’s the kind of thing we need the community at large to be aware of, that people need to sign onto the brain banks ahead of time.”