Low doses of a compound called romidepsin might help those on the autism spectrum overcome the social challenges that define their condition.
So far it has only been shown to be effective in mice, but the mechanisms behind the drug's activity make it a promising candidate for an autism treatment in humans – the first of its kind.
Autism spectrum disorder (ASD) is an umbrella term for traits that interfere with the brain's ability to process stimuli and negotiate social cues, often making communication difficult.
While there is no single gene that's responsible for these characteristics, past research has found a number of genes do influence the disorder's appearance. Some are inherited, while others seem to be the product of mutations early in life.
Researchers from the University at Buffalo in New York focussed their studies on a third way that genes can be changed – epigenetics.
Environmental factors can often trigger enzymes in the body to wrap up parts of the chromosome containing certain genes a little too tightly, effectively hiding them away. There is strong evidence suggesting a number of genes commonly associated with ASD are silenced by this overenthusiastic wrapping process.
What's more, high levels of an enzyme called histone deacetylase, or HDAC2, observed in lab studies involving mice could be responsible for this process.
"In the autism model, HDAC2 is abnormally high, which makes the chromatin in the nucleus very tight, preventing genetic material from accessing the transcriptional machinery it needs to be expressed," says senior author and molecular biologist Zhen Yan.
In other words, pumped up levels of HDAC2 could be locking away key genes, giving rise to some of ASD's signature traits.
Inhibiting this enzyme could help the strands of DNA relax and let the genes do their job again, potentially restoring the brain's ability to navigate social situations.
We already have a US government-approved drug that does this; a compound called romidepsin, currently used as a treatment for certain kinds of lymphoma.
Turning to a cancer drug wasn't a mere stab in the dark, either.
"The extensive overlap in risk genes for autism and cancer, many of which are chromatin remodelling factors, supports the idea of repurposing epigenetic drugs used in cancer treatment as targeted treatments for autism," says Yan.
To test whether the drug would indeed help alleviate ASD symptoms, the researchers used mice engineered with mutated versions of the Shank3 gene.
Previous research had already shown how changes to Shank3 were behind major differences in areas of the brain associated with socialisation.
More importantly, around 1 percent of ASD diagnoses are currently thought to be associated with Shank3 mutations. That might not sound like much, but given a condition as complex as autism it's significant enough to be worth pursuing.
The team speculated that there had to be a link between the loss of Shank3 and the ramping up of HDAC2 in the nucleus.
Through a series of experiments designed to study the links between elevated HDAC2 levels and Shank3 mutations, the scientists teased out the biochemical steps to show how epigenetic changes were ultimately responsible for ASD's social challenges in a number of individuals.
The best part is romidepsin successfully impeded HDAC2 and improved social skills in their mice models, suggesting there was a way to ease these challenges in humans.
"Autism involves the loss of so many genes," says Yan. "To rescue the social deficits, a compound has to affect a number of genes that are involved in neuronal communication."
Of course mice aren't people, but since the mechanisms appear to be the same, there's hope that the drug's effects will be as well.
The fact it's already FDA approved also suggests a treatment is tantalisingly on the horizon.
It's important to keep in mind that there's still plenty of research between this study and a publically available treatment.
Still, given there are currently no other treatments that come close to covering this ground, it's hard not to get excited.
This research was published in Nature Neuroscience.