Skip to Main Content

In the spring of 2011, a handful of powerful new scientific tools came into existence — and each one had white tufts of facial hair and a long gray tail.

Now, five years later, those genetically modified monkeys and their offspring have been confirmed as the first ever nonhuman primates to show autism-like symptoms in the lab.


The findings, published Monday in Nature by a team at the Shanghai Institutes for Biological Sciences, could be a big step forward for scientists researching autism, which is notoriously hard to study in animals. Mice, which are often used as models for human disease, simply don’t have the complexity of brain wiring or social interaction to replicate how autism affects our biology and behavior.

Monkeys, on the other hand, are a lot more similar to us. And the behavior of the genetically modified long-tailed macaques was striking.

Just as human children with autism might flap their hands or rock back and forth, these monkeys obsessively ran in circles. And just as human autism patients can be stressed at the prospect of social interaction, the modified monkeys grunted and screamed when humans approached, and spent less time with their cage-mates than monkeys without the genetic modification.


“This paper is really important because it’s showing that you can have a genetic manipulation and see behavioral differences,” said Elinor Sullivan, a neuroscientist at Oregon Health and Science University who wasn’t involved with the study.

But Sullivan, who works often with primates, cautioned that it’s still tricky to extrapolate from monkeys to humans. Studying monkeys requires careful attention to a host of behaviors, she said, and so far the Shanghai scientists have focused on “very, very, very basic behavioral measures.”

Inducing autism in the macaques was no mean feat.

That’s because autism itself is so hard to pin down. The term covers a broad spectrum of developmental problems. And researchers have also identified hundreds of genes, as well as some environmental factors, that could be potentially responsible.

Because not all of these genetic differences are present in all autism patients, it isn’t always clear which genes scientists should be fiddling with to try to breed lab animals with the disorder.

The Shanghai team narrowed their focus to a single gene called MeCP2. “Some physicians discovered that in a group of boys who had severe autism … MeCP2 had been duplicated,” explained Zilong Qiu, the lead author of the study.

So the team inserted extra copies of the gene into monkey embryos using a hollowed-out virus as a vehicle. Then, they implanted the embryos into surrogate mothers and waited — five months for the baby monkeys to be born, and another year for them to wean off breast feeding. At that point, they could begin behavioral studies.

To insure a steady population of their modified macaques, the Shanghai researchers wanted to make sure the genetic trait would be passed to future generations. But that could mean waiting as many as four years until the monkeys reached sexual maturity.

Instead, Qiu and his team took a shortcut. One of their genetically modified male macaques had gotten sick — so sick, in fact, that they weren’t sure it would make it to reproductive age anyway — so they removed a slice of his testicular tissue and implanted it into a mouse, which hits puberty much more rapidly. They wanted to use the mouse’s hormones to coax that bit of the monkey’s gonad into growing as fast as possible.

“They were essentially using the mouse as a monkey sperm factory,” explained Jon Levine, director of the Wisconsin National Primate Research Center.

They used that sperm to fertilize new embryos. It worked: The macaque’s offspring had symptoms of autism as well.

Some researchers wonder to what extent monkeys with a single gene duplication will be able to stand in for human patients with autism, given that in humans, an entire mosaic of genes seem to be involved. But Qiu said he’s hopeful the model will be useful for developing therapies.

“Although there are many different genes associated, we believe there aren’t 100 types of autism,” he said.