Scientists have uncovered a wealth of clues about how genetic mutations lead to severe forms of autism, offering opportunities for testing drugs that could treat the condition, according to a study published Thursday.
The researchers charted how mutations change the way proteins work together in cells, altering the development of the brain.
“This is making maps of unknown territories that’s really necessary to move the biology forward,” said Dan Geschwind, a neurogeneticist at the University of California, Los Angeles, who was not involved in the study.
Researchers have been studying autism for over 80 years, but it has only been in recent years that they have been able to explore its molecular biology. One reason that progress has been so slow is that autism is not just one condition, but a broad constellation of them.
People with autism can have difficulty with language and making social connections, and often display restricted or repetitive behaviors. While many children with an autism diagnosis can grow up to lead independent adult lives, others may be nonspeaking, have intellectual disabilities, and require round-the-clock care.
But in around 30 percent of people diagnosed with autism, typically those with the most severe disabilities, scientists have identified single gene mutations that are almost guaranteed to cause the disorder.
When those genes started coming to light 20 years ago, scientists eagerly hoped they could find precise treatments for severe autism. In some studies, they engineered mice with the mutations and then tried different drugs to reverse their autism-like symptoms. But those efforts have not led to any effective treatment targeting a specific gene.
Matthew State, a psychiatrist at the University of California, San Francisco, and an author of the new study, said that it had become painfully clear that scientists needed a much deeper understanding of what mutations actually do to the brain before they can hope to create drugs to treat autism.
“There’s a huge missing step,” Dr. State said. “What the genes are doing is they’re building proteins that are driving the biology of the cell.”
A number of autism researchers are now trying to fill in that gap. Dr. Geschwind and his colleagues, for example, are looking at how genes turn on and off in neurons that carry mutations linked to autism. Others, including Dr. State, have looked at what the proteins encoded by those genes do.
When a gene makes a protein inside a cell, that protein almost never works alone. To know what a protein does, you have to get to know the other proteins it interacts with. In the new study, Dr. State and his colleagues picked out 100 proteins strongly linked to severe forms of autism and searched for their molecular partners.
The researchers injected each protein into a batch of cells and allowed them to grab onto other proteins. Then they fished the autism proteins out, along with any partner proteins stuck to them.
This fishing expedition revealed more than 1,000 proteins that are partners to the 100 autism proteins.
Many of these partnerships had never been documented before. “There is a huge amount here for people to follow up,” said Jonathan Sebat, a geneticist at the University of California San Diego, who was not involved in the study.
Further experiments revealed that mutations linked to autism change how these proteins join together. In some cases, the mutation caused a pair of proteins to become sticky, so that they held on more tightly to each other. In other cases, the mutation made them slippery, so that the proteins tended to fall away.
The scientists suspect that these changes affect how neurons develop in the brain. “I believe these are drivers of the disease,” said Nevan Krogan, a molecular biologist at the University of California, San Francisco, and an author of the new study.
The researchers then picked a few mutations to study in detail, in order to see if that hunch was true. In one experiment, they focused on a pair of genes called FOXP1 and FOXP4 that become active in developing brains. Normally, the two proteins stick together and then grab onto certain genes, turning them on so that the cell can divide, producing new brain cells.
A mutation linked with autism alters the shape of FOXP1. Dr. Krogan and his colleagues found that the mutation takes away the molecular glue that keeps FOXP1 and FOXP4 stuck together.
On its own, FOXP1 keeps on switching on the right genes for normal brain development. But the now-liberated FOXP4 “goes rogue,” Dr. Krogan said.
The rogue FOXP4 protein grabs onto other genes that should be silent and turns them on. The result is devastating: The cell can no longer divide properly. Dr. Krogan and his colleagues argue that this change to the way brain cells divide could be how some mutations help give rise to autism.
“That’s exactly the kind of mechanistic connection we want to see,” Dr. Sebat said.
This intimate picture of how autism mutations work offers scientists new ideas for how to undo their effects. A drug might serve as a molecular glue to stick FOXP1 and FOXP4 back together again, for example. Another strategy might be to block FOXP4 proteins from attaching to DNA, so that it can’t wreak havoc.
If the scientists can find a drug that can reverse the FOXP1 mutation, it’s possible that it will be effective only for people with that particular mutation. If that’s true, then finding treatments for severe autism may become a long slog, requiring scientists to search for a different drug for each molecular disruption. Similar clinical trials are underway to identify potential genetic treatments for individual mutations.
But Dr. State argued that the search might not be so difficult. He and his colleagues found that many of the 100 autism proteins shared partners in common — suggesting that they belong to the same networks.
“We’re not pulling out random things,” Dr. State said. “They’re biologically coherent.”
Far less is known about the genetic changes contributing to milder autism, and some autistic people view the condition as a natural variation that does not need a cure. But, for people with severe disabilities and their families, the convergence identified in the study could mean it will be possible to use the same treatments to help people with severe autism caused by different genetic mutations.
“To me, the most exciting thing that the study showed is that these different mutations converge on the same molecular pathways,” said Alison Singer, director of the Autism Science Foundation and the mother of a 29-year-old daughter with profound autism.
Ms. Singer said that the new finding could bring together families of people who have different genetic mutations driving their autism, as well as those, like her daughter, who do not yet have a known genetic cause. “What it says is that you may not need a separate therapeutic strategy for each gene,” she said.
But some other researchers said the new study didn’t provide enough evidence to close the case for common networks in autism. “Those are weak claims,” Dr. Sebat said.
Yet Joseph Buxbaum, a neuroscientist at the Icahn School of Medicine at Mount Sinai, said that even some overlap between autism mutations could help push forward treatments. “If it works at all, we’ll be very happy,” he said.
