A single species of bacteria reverses autism-like features in mice exposed to a high-fat diet in utero1. The finding suggests that this microbe, Lactobacillus reuteri, is the missing link between diet during pregnancy and autism-like behaviors in the pups. But some researchers are sharply critical of the findings, saying they have limited relevance to the condition in people.
Mouse pups exposed to a high-fat diet in utero are born with a deficient microbiome — the sum total of microbes in the gut. They also have unusual brain activity and little interest in social interaction. Replenishing the pups’ microbiomes with repeated doses of L. reuteri reverses all of these features. Researchers reported the findings 16 June in Cell.
The new study may help explain the elevated autism risk in children born to obese women. The findings also hint at a possible connection between a pregnant woman’s diet and the gastrointestinal problems seen in many children with autism.
“We were surprised to find that animals that came from moms with high-fat diets had a profound deficit in social behaviors,” says senior researcher Mauro Costa-Mattioli, associate professor of neuroscience at Baylor College of Medicine in Houston, Texas. “And it was very surprising to find that one [species of] bacteria was able to improve those social behaviors.”
“This work connects the dots between maternal diet and behavioral changes in offspring,” says Sarkis Mazmanian, professor of biology at the California Institute of Technology in Pasadena, who was not involved in the work.
Still, Mazmanian and others say L. reuteri — a microbe found in yogurt, probiotic supplements and in the human gut — may not have the same beneficial effects in children with autism.
“There are already too many hucksters selling high-priced probiotics to vulnerable families of children with autism; we don’t need this study to be cited in support of their commercial exploitation,” says Helen Tager-Flusberg, director of the Center for Autism Research Excellence at Boston University. Most mothers of children with autism are neither obese nor ate a high-fat diet during pregnancy, she says.
Of mice and microbes:
The researchers started with two groups of female mice: The control group ate standard rodent chow for eight weeks prior to mating, and the other group ate chow with more than four times the amount of fat and became obese. The mice remained on these diets while pregnant. Their pups were all fed a standard diet after weaning.
The obese mice gave birth to pups with smaller populations of eight bacterial species in their guts compared with the controls, the researchers found. The most dramatic decrease was in L. reuteri, which promotes the production of oxytocin in neurons. Oxytocin is a hormone linked to autism.
Pups with low levels of L. reuteri also have decreased levels of oxytocin. They have weak communication among dopamine neurons in the brain’s reward center. They also show little interest in greeting new pups, bury marbles for long periods, and are anxious. All of these are proxies for autism behaviors in mice.
The researchers supplemented the mice’s drinking water over a four-week period with an L. reuteri strain isolated from the human gut. They found that this normalizes social behavior, oxytocin levels and neuronal activity in the pups exposed to a high-fat diet. They found the same results with a nasal spray of oxytocin. Neither treatment has any effect on the controls.
“You put [L. reuteri] in the water and the animal becomes better,” says Costa-Mattioli. “What can I say? This is it.”
Boosting levels of this microbe may shift the rest of the gut microbiome — and the brain — toward its optimal balance, he says. Exactly how that happens is unclear, however.
Some support for the new theory comes from work in primates. Microbes also appear to link maternal obesity with social behavior in young Japanese macaques, says Elinor Sullivan, assistant professor of biology at the University of Portland in Oregon.
A high-fat diet in a pregnant macaque alters her baby’s microbiome and social development, Sullivan says2. Like the rodents in the new study, macaques born to obese mothers show increased anxiety and repetitive behaviors, and impaired communication among dopamine neurons3.
Still, the relationship might be infinitely more complicated in complex conditions such as autism, Sullivan says.
“It is unlikely that disruption of a single microbial strain is the sole contributor to this disorder,” Sullivan says.
The new study “provides proof of concept for the potential of probiotic treatments for social disorders,” says Mazmanian. In 2013, his team showed that the microbe Bacteroides fragilis reverses obsessive behaviors and anxiety in a mouse model of autism. But many autism treatments have shown promise in mouse models and failed in people, Mazmanian says.
The gut’s structure and microbiome differ dramatically between people and mice. The mouse used in the new study, known as C57Bl6/J, is also prone to obesity.
The scientific community should “be very careful about how we present and interpret such work,” says Jonathan Eisen, professor of medical microbiology and immunology at the University of California, Davis. Eisen has been openly critical of the study’s media coverage.
Costa-Mattioli himself is cautiously enthusiastic. “It’s tremendously fascinating just to conceive the idea that these bacteria could potentially be used as a treatment,” he says. But “we need to understand how the bacteria is able to improve the behavior in animals before this can be leveraged in humans.”