Autism symptoms replicated in mice after faecal transplants

Study aims to discover whether gut microbes play a part in development of the condition
Mice that had transplants from children with autism did not wrestle, push and sniff other mice as much as the control group of mice. Photograph: Redmond O Durrell/Alamy

Ian SampleScience editor @iansample-Thu 30 May 2019

Scientists have induced the hallmarks of autism in mice by giving them faecal transplants from humans with the condition.

The experiments were designed to test whether the communities of gut microbes found in people with autism have a role in their symptoms, an ideathat is gaining ground among researchers.

In the study, animals that had faecal transplants from children with autism became less sociable, less vocal and developed repetitive behaviours. In contrast, genetically identical mice that had transplants from people without autism were unaffected by the procedure.

Sarkis Mazmanian, a microbiologist who led the project at the California Institute of Technology, said that while gut microbes did not appear to cause autism, the findings raised the prospect of new treatments for some of the most common symptoms of the condition.

“Potentially this opens up the possibility that microbiome-based interventions may be effective in autism,” he said. “We’ve identified particular organisms and the products of those organisms that are drivers of symptoms in mice, but we don’t know if they drive symptoms in humans too.”

To perform the faecal transplants, the researchers fed a known amount of human stool down a tube into the recipient mouse’s stomach. The animals were then tested to ensure the microbes in the stool had colonised the gut.

The human gut is home to trillions of bugs. In return for moisture, warmth and nutrition, the microbes help to digest food, train the immune system, and keep metabolism in check. People have different populations of gut microbes depending on their diet, lifestyle and genetics, but studies have found particular differences in people with autism.

Writing in the journal Cell, the scientists describe how they used a tracking system to monitor how well mice socialised after they had received faecal transplants. They also recorded how often the animals buried marbles placed on the wood chips in their cages. Finally, they used ultrasonic microphones to eavesdrop on the creatures’ communications.

Mice that had transplants from children with autism did not wrestle, push and sniff other mice as much as the control group of mice, which had transplants from people without autism. Nor did they produce as many ultrasonic squeaks. And while control mice might bury one or two marbles and then move on, mice with autism-related microbiomes kept going, an indication of repetitive behaviour.

Autism affects about one in 60 people and tends to be diagnosed when children show deficits in social and verbal interactions and display repetitive behaviours such as hand-flapping, spinning and finger-flicking. “We were able to see all three of the core features of autism replicated in the mice,” said Mazmanian.

Tests on the mice with autism-related microbiomes revealed changes to gene expression in their brains and low levels of certain metabolites in their bodies, notably two substances called taurine and 5-aminovaleric acid, or 5AV. In follow-up experiments, the researchers fed taurine and 5AV to strains of mice that naturally exhibit autism-like behaviour, and found their repetitive behaviour and social skills improved.

Gil Sharon, the first author of the study, said that while human trials might one day test whether specific bacteria or their products can help people with autism, they may not happen soon. “A lot more work needs to be done before we can say we’re ready for human trials,” he said.

“We don’t want to give parents, children and loved ones false hope,” Mazmanian added. “We have not solved the problem. All we’ve done is introduce a potential new strategy that needs to be tested in people. This is the start of what may ultimately be a therapeutic for people with autism, but we are certainly not there yet.”

Last month, researchers at Arizona State University announced that faecal transplants had almost halved symptoms of autism in 18 children. Two years after the procedure, the number of children rated as having “severe” autism had fallen from 83% to 17%.

Jeremy Nicholson, the pro-vice chancellor for health at Murdoch University in Perth, Australia, said the California study was impressive, but he was doubtful about treating autism symptoms with microbes or pills based on the substances they produce. “Humans have very diverse microbiomes, so there is no one-size-fits-all [treatment]. It’s why probiotics work on some people and not others. We need to understand the mechanism behind this, and then work out if there are parts of that network that are druggable.”