Signs of autism—such as impaired social skills and repetitive, ritualistic movements—usually begin to appear when a child is about 18 months old. Autism is thought to result from miswired connections in the developing brain, and many experts believe that therapies must begin during a "critical window," before the faulty circuits become fixed in place. But a new study online today in Science shows that at least one malfunctioning circuit can be repaired after that window closes, holding out hope that in some forms of autism, abnormal circuits in the brain can be corrected even after their development is complete.
According to developmental neurobiologist Peter Scheiffele of the University of Basel in Switzerland, autism doesn't result from a handful of "culprit" genes that point to a treatable flaw. Instead, patients appear to carry mutations in one out of dozens, even hundreds of risk genes. "This genetic complexity is a huge issue with respect to developing treatments [for autism]," Scheiffele says. To complicate the picture further, autism is not always an isolated disorder; it's often a common feature in syndromes that otherwise differ drastically. For example, in fragile X syndrome, a form of mental retardation, about 25% of patients are also autistic.
Scheiffele and colleagues were studying a gene called neuroligin-3 (Nlgn3), involved in building the contact points, called synapses, between neurons. Many researchers believe that autism begins at the synapse, and mutations in Nlgn3 have appeared in some forms of the disorder. Sheiffele's team was focusing on synapses in the cerebellum, a part of the brain that controls movement, but, according to recent research, may also be involved in social behavior. Abnormalities in this region may contribute to both the unusual movements and the social problems seen in autistic patients.
To get a better handle on the role of Nlgn3, the scientists studied mice whose Nlgn3 genes were engineered with an on-off switch, called a promoter region, that is controlled by the antibiotic doxycycline. The animals were raised with the drug in their drinking water, which kept the switch in the off position. With the Nlgn3 gene disabled in the mice, neurons in their cerebellum made the abnormal connections seen in the autistic brain.
Specifically, and much to the researchers' surprise, the lack of Nlgn3 led to the overactivation of a receptor abbreviated as mGluR1α. This receptor is a component of a pathway that is also disrupted in fragile X syndrome, though it results from mutations in an entirely different gene. In the mice, the overabundance of these receptors led the neurons to make synaptic connections in the wrong places.
To see if turning Nlgn3 gene back on would correct these problems, the researchers withdrew the doxycycline. It worked: With Nlgn3 functioning once more, levels of the extraneous receptor receded back to normal, and the misplaced synapses began to disappear.
"Our finding demonstrates that there is still flexibility after the 'critical window' of brain development," Scheiffele says. "It raises the question: To what extent can a miswired brain be corrected?" The next step, he says, is to see whether motor abnormalities, such as ladder-climbing difficulties, and social interactions can be corrected with similar treatment in the engineered mice. His team is also studying whether drugs that block the mGluR1α receptor can have the same effect as genetically controlling the Nlgn3 gene, which isn't a treatment option for humans.
"This study holds out hope for children and even adults with developmental disorders. Maybe their conditions aren't set in stone and can be treated," says neuroscientist Kimberly Huber of the University of Texas Southwestern Medical Center in Dallas. Huber adds that drugs that block a similar receptor, mGluR5, are in clinical trials to treat fragile X syndrome.