According to the US Centers for Disease Control and Prevention, one in 68 U.S. children have autism spectrum disorder (ASD). Recent evidence suggests that the risk of ASD is increased for the offspring of mothers with obesity. In both humans and non-human primates, the offspring of obese mothers have also been shown to have abnormal microbiomes. And some people with ASD have imbalanced gut microbes, or dysbiosis. Baylor College of Medicine’s Mauro Costa-Mattioli and colleagues sought to better understand how maternal obesity, the microbiome, and ASD are interconnected.

The team turned to mice for answers. The researchers gave female animals high-fat diets before setting up matings, later finding that a “large proportion” of the offspring exhibited ASD-like behaviors, including reduced social interaction, repetitive behaviors, and anxiety. The team analyzed the microbiomes of these offspring, finding that they differed from those of control animals.

Mice are coprophagic—“they eat each other’s poop,” said Costa-Mattoli. Consequently, when housed with control mice, those animals whose mothers had eaten a high-fat diet (“MHFD” mice) developed microbiomes more similar to their non-MHFD cagemates. They also resumed normal social interactions.

To confirm that the microbes rather than co-housing were responsible for restoring social function, the team performed fecal transplants. Mice born and maintained in a sterile environment have similar social deficits to those of the MHFD-mice. Fecal transplants from control mice—but not from MHFD mice—could restore sociability to these germ-free mice, the team showed.

A closer inspection of the microbiomes of the MHFD animals revealed that the most markedly diminished bacterial species was Lactobacillus reuteri, which, coincidentally, had been shown to boost levels of oxytocin—a hormone involved in social bonding—in an earlier mouse study. Remarkably, when the team gave the MHFD mice this single species of bacteria—via their drinking water—the animals’ social deficits once again disappeared. Neither coprophagia nor L. reuteri treatment corrected the anxiety or repetitive behaviors of the mice, however.

There were fewer oxytocin-expressing neurons in the brains of the MHFD mice, the team showed, and the activity of the reward circuitry during social interactions was impaired. Both of these neural abnormalities were also reversed by L. reuteri treatment or by oxytocin administration.

This study is not the first to report that a single species of bacteria can affect mouse behavior. Dosing animals with L. rhamnosus has been shown to reduce anxiety- and depression-associated behaviors, while treatment with Bacteroides fragilis has been shown to reverse certain ASD-like behaviors, including anxiety and repetitive behavior.

“The direct testing of specific bacteria for their behavioral benefits provides proof of concept for the potential of probiotic treatments for social disorders,” Caltech’s Sarkis Mazmanian, who was not involved in the work, wrote in an email to The Scientist. However, “much more work needs to be done before this perspective can be tested in humans, let alone validated,” he added.

It will be important to determine the mechanisms by which these various microbes affect brain function and behavior, said Cryan. Nonetheless, he added, together these studies show the importance of the gut-brain connection. “We spend so long trying to kill bacteria, but what we really need to appreciate now is that, to have a healthy brain and brain development, we need to keep an eye on our bacteria, especially during pregnancy and early life.”

S.A. Buffington et al., “Microbial reconstitution reverses maternal diet-induced social and synaptic deficits in offspring,” Cell, 165:1–14, 2016.