As common as it is, autism is a puzzling disorder. Scientists have found more than 500 genetic variants that increase the risk of autism, but most of these only raise the risk a tiny bit. And for the vast majority of them, it’s anybody’s guess how they contribute to repetitive behaviors, social difficulties, language impairments, and other problems. Now, some scientists see promise in a new approach to unraveling the biology of autism: collecting cells from individual autistic children and turning them into neurons they can study in the lab.
“If we sequence two people with very similar symptoms, what we see is they don’t necessarily have mutations in the same genes,” said Alysson Muotri, a neuroscientist at the University of California, San Diego. “This is not one disease, there are probably several diseases under the umbrella of autism.”
In hopes of getting a better grip on this variability, Muotri and a handful of other scientists have turned to a more individualized strategy that’s become possible only in the last few years. These scientists are collecting cells from the skin, blood - or in Muotri’s case, teeth - of autistic children, and turning them into neurons in their labs. By examining those neurons under a microscope and studying their electrical properties, they hope to figure out what’s wrong on a patient-by-patient basis. And, ideally, how to fix it.
The strategy is based on the Nobel-prize winning discovery that it’s possible to turn back the clock on mature cells, returning them to an immature state in which they have the potential to grow into many different types of cells - including neurons. These intermediate cells are called induced pluripotent stem cells, or iPS cells for short.
The first attempts to use iPS cells to study autism involved Rett syndrome and Timothy syndrome, two forms of autism that are caused by a known genetic mutation.
In a study published today in Molecular Psychiatry, Muotri and colleagues extend the approach to a far more common situation - a case with no known genetic cause. The subject was an 8-year old boy with autism. His parents sent Muotri one of his baby teeth when it fell out, and Muotri’s lab isolated cells from the dental pulp, turned these into iPS cells, and turned the iPS cells into neurons.
Under the microscope, these neurons didn’t look right. They had fewer branches and fewer synapses than neurons made the same way from people without autism. They also fired less. The researchers saw what they thought might be a clue to these abnormalities in the boy’s genome: He has a mutation that disrupts a gene called TRPC6, which makes a protein that regulates the flux of calcium ions into cells.
Next, the researchers treated the neurons from the autistic boy with a drug called hyperforin, which boosts TRPC6 activity. The results were encouraging: The neurons’ appearance and firing activity became more normal.
Based on these and other lab experiments, Muotri thinks the TRPC6 mutation is a likely culprit in this boy’s autism. It’s not a gene that’s been linked to autism before. But that’s not to say it’s the sole cause. “TRPC6 is one of the genes that’s affected,” Muotri said. “I think it’s not the only one.”
These uncertainties highlight the difficulty of getting to the bottom of so-called idiopathic cases of autism, the vast majority of cases with no known genetic cause, says Ricardo Dolmetsch, global head of neuroscience at Novartis Institutes for Biomedical Research. “There’s the issue of are you absolutely sure that a mutation is causative,” Dolmetsch said. “It’s hard to know unless you find it multiple times.”
Dolmetsch was among the first researchers to use iPS cells to study autism, and he believes the approach will pay off, especially for understanding forms of autism that are caused by a handful of gene mutations rather than a single devastating mutation. “iPS cells will be important to understand how these mutations interact,” he said.
The ultimate goal, of course, is better treatments. One optimistic scenario is personalized medicine for autism, in which doctors use a patient’s genome and neurons derived from iPS cells to make a diagnosis and select the most effective drugs for that particular patient. Drugs could even be tested on the patient’s own neurons before being prescribed.
It’s hardly a definitive test, but the parents of the boy in Muotri’s study tried giving him hyperforin, the drug that reversed the anatomical and physiological abnormalities in his lab-grown neurons. Hyperforin is an ingredient in St. John’s wort, and the boy took the herb for about a month, Muotri says. His father, therapists, and school all reported an improvement in the boy’s focus and social behavior.
“We have videos before and after,” Muotri said. “Before, someone will ask him to sit down and draw something, and you see that his mind is all over the place, he can’t sit for a minute, he doesn’t pay attention. Then after one month, he will sit there he will look at the person and understand what they want and start playing with the paper.”
But this wasn’t meant to be a rigorous trial, and the boy’s mother said she saw no change in his behavior. And there’s no reason to think St. John’s wort would be a useful treatment for autism in anyone without this specific mutation, Muotri adds.
Even if the iPS cell strategy could be refined into an accurate diagnostic tool, it wouldn’t come cheap. Muotri estimates that creating and characterizing neurons from a single patient would cost about $100,000.
Another way that iPS cells could lead to better treatments - and probably a more likely way in the short-to-medium term - is by helping scientists identify different categories of autism with different underlying causes. Neurons and other cells derived from iPS cells could also be used in high-throughput drug screens to identify promising new drug candidates - or old drugs that have been approved for other disorders and could be prescribed “off label” for autism. Muotri is doing this in collaboration with the National Center for Advancing Translational Sciences at the National Institutes of Health, and Dolmetsch says Novartis has made a big investment in iPS cells for autism and other brain disorders.
One limitation to this approach is that a relatively small number of lab-grown neurons can’t compare to the complex networks of neurons in a living human brain. If faulty networks turn out to be the core deficit in autism, iPS cells might not capture that. On the other hand, if problems at the level of individual cells are the key, iPS cells could be an extremely valuable tool.
There may be hundreds of genetic variants that contribute to autism, but the number of biological processes that are affected is probably far less. Two new studies, among the largest autism genetics studies to date, suggest that many of the gene mutations tied to autism converge on just two biological processes: regulating gene activity and synaptic communication between neurons (TRPC6 wasn’t one of the genes named, but it would fall in this second category).
“There are almost certainly more [mutations] than you could make drugs for,” Dolmetsch said. “The challenge is to put them into pathways, so that you don’t have to make 600 different drugs, you could make four or five drugs and use them in different combinations that would cover most kids with autism.”
Return to News Home
