Subscribe to eTOC

Synaptic Brain Disconnections Implicated as Cause for Autism in New Animal Studies


Investigators describe emerging data about mutations at synaptic connections and autism-related behaviors in mouse models.

It has long been theorized that autism is caused by problems with synaptic connections in the brain. Now researchers are providing hard proof that this is likely the case — based on emerging data about mutations at synaptic connections and autism-related behaviors in mouse models.

In one of the studies described at the annual meeting of the Society for Neuroscience this year, investigators at the University of Texas Southwestern Medical Center reported that deleting the neurexin1-alpha (NXN1-alpha) molecule in mice that helps neurons form brain synapses brings about some autism-related behavioral abnormalities, while some behaviors remain normal.

Mice in which NXN1-alpha was knocked out exhibited increased repetitive grooming, an increased response to auditory stimuli, impaired nest building, and enhanced motor learning. But social behavior, locomotor activity, anxiety-like behavior, and spatial memory were all essentially normal in the NXN1-alpha knockout mice, according the investigators.

The researchers targeted neurexin 1 because deletion or mutation of this gene has been identified as a possible cause of human autism.

Researchers say the changes in grooming activity offer hope that the mouse models can shed light specifically on repetitive behaviors, which, along with deficits in social interaction and language, is one of the three main groups of symptoms in autism and autism spectrum disorders.

University of Texas Southwestern Neurology and Psychiatry Assistant Professor Craig M. Powell, MD PhD, who is leading the research along with Thomas Südhof, MD, said his group was surprised that only certain autism-like behaviors were seen.

“We just hadn't expected them to be normal socially, or close to normal socially, and we hadn't really expected this repetitive behavior phenotype because we hadn't seen it before, but we found it interesting and relevant for autism,” Dr. Powell said.

The findings are another step in discerning which synaptic changes matter to autism and which don't.

“We can test whether those changes are relevant to the brain function or not relevant,” Dr. Powell said. “What we're trying to do is identify novel treatment targets for autism.”


DR. CRAIG M. POWELL and colleagues reported that deleting the neurexin1-alpha (NXN1-alpha) molecule in mice that helps neurons form brain synapses brings about some autism-related behavioral abnormalities, while some behaviors remain normal.


Last year, researchers at University of Texas Southwestern unveiled findings on the first true genetic mouse model of autism. The mice had classic signs of autism, showing enhanced spatial learning and wanting to spend less time with other mice than wild-type mice.

In those mice, the gene for neuroligin 3, which also helps create synapse connections, was substituted with a mutated version. Neuroligins join with neurexins to bind one neuron to another.

Now, researchers are trying to bore down even further in studying mouse models to understand what might be going on with cell-adhesion molecules in the brain of autism patients.

They have also examined the role of phosphatase and tensin homolog (PTEN), a protein that has been linked to many kinds of cancer as well as brain disorders such as enlarged brain, seizures and autism.

Led by Luis Parada, PhD, professor of developmental biology at University of Texas Southwestern, investigators developed a mouse model with PTEN knocked out in postmitotic cells in the cortex and hippocampus. The mice showed an increase in macrocephaly, neuronal hypertrophy, and behavioral changes similar to those in human autism.

They also found that rapamycin — which inhibits mammalian target of rapamycin complex 1 (mTORC1), a protein complex that acts as an energy sensor and controls protein synthesis — blocks the mTORC1 pathway in the CNS and can prevent or reverse oversized brain growth and autism-related behaviors in the PTEN knockout mice.

Dr. Powell said their team is planning to test mutations in specific subregions of the brain to determine the effects of cerebral functions.

“One key question is which part of the brain is most important,” he said. “We don't want to record from every synapse in the brain in every region.”

Then, researchers can focus on seeing whether reversing those brain abnormalities change the abnormal behavior, he said.

Dr. Parada's group will also be working with additional genetic mouse models, with the hope that different mutations might require the same treatment targets, making therapy easier.


Nancy Minshew, MD, professor of psychiatry and neurology at the University of Pittsburgh, said the findings are exciting for the autism research community. “I think it generates tremendous interest among the developmental neurobiologists, geneticists and neurologists,” she said.

“The identification in autism spectrum disorders of genes that are involved in cortical neurons and making connections confirmed the work that went before it that autism is a disorder of underdevelopment of cortical systems connectivity and overdevelopment of local fiber connections,” Dr. Minshew said by e-mail. “The whole picture from beginning to end goes together now.”

On the rapamycin work, she noted recent findings that the drug appears to be preventing seizures in infants and toddlers with tuberous sclerosis complex (TSC).

She added, though, that PTEN and TSC involves enlarged neurons but most autism cases do not.“So what works for PTEN and TSC is unlikely to work for all cases of autism,” she said. “The major point is that the molecular pathophysiology of all cases must be discerned and that will guide intervention to its successful end.”

Dr. Powell said that about 20 percent of patients with autism have enlarged brains, 20 percent have smaller brains and the rest are about normal size. So, he said, it isn't known how widely rapamycin might be able to be used.

“I don't think of rapamycin as applicable to all patients with autism in any way, shape, or form,” he said. Rather, it is more likely to become just one of several modes of treatment tailored by specific causes of autism.

“The most encouraging thing about these findings is the potential for reversing already established deficits — the very idea is revolutionary,” said Kathryn McFadden, MD, of the University of Pittsburgh School of Medicine Department of Pathology. “As a researcher and as a parent of an autistic child (17-year-old daughter) this gives me great hope.”

But she added, “As with anything, these findings should be approached with caution. That said, these studies must be done. Even if a handful of autism symptoms could be reversed, it would make a huge difference to families.”

She and Bernie Devlin, PhD, are looking at genes implicated in the Autism Genome Project genome-wide association scan. Those genes are involved in neuronal connectivity but at a stage before the synapse is established.

“We are all involved in examining the molecular substrates for brain connectivity, just at different stages,” she said. “The autisms are heterogeneous, but connectivity problems tie them together.”