Gut bacteria influence autism-like behaviors in mice

Autism spectrum disorder (ASD) affects an estimated 1 in 59 people in the United States, causing a variety of difficulties with social communication and repetitive behavior. Many factors, including genetic and environmental effects, are believed to influence symptoms, and there are no approved treatments. Now, using mouse models, Caltech researchers have discovered that gut bacteria directly contribute to autism-like behaviors in mice.

An article published Friday in The Economist, “More evidence that autism is linked to gut bacteria,” details how studies in humans and mice are coming together to help us gain a better understanding of causes and treatments for autism.

“This is a recognition that the zillions of apparently nonpathogenic bacteria on and in human bodies, hitherto largely ignored, are actually important for people’s health. They may even help to explain the development of some mysterious conditions. One such condition is autism,” reports The Economist.

Rosa Krajmalnik-Brown, a researcher in the Biodesign Swette Center for Environmental Technology and an associate professor in the Ira A. Fulton Schools of Engineering, has been working for nearly a decade with James Adams, a President’s Professor in the ASU School of Engineering, Matter, Transport and Energy, to explore the gut-brain connection as it relates to autism. Most recently Krajmalnik-Brown collaborated with researchers at Caltech who are exploring similar dynamics in mice. Their work was published online Friday in the journal Cell.

Krajmalnik-Brown and Adams’ study was published a few weeks ago in Scientific Reports.

“In recent years, numerous studies have revealed differences in the bacterial composition of the gut microbiome between individuals with ASD and neurotypical subjects,” said Caltech's Sarkis Mazmanian, Luis B. and Nelly Soux Professor of Microbiology and Heritage Medical Research Institute investigator, whose labarotory conducted much of the work on the mouse model.

“However, while this previous research identifies potentially important associations, it is unable to resolve whether observed microbiome changes are a consequence of having ASD or if they contribute to symptoms.”

"Our study shows that the gut microbiota is sufficient to promote autism-like behaviors in mice. However, these findings do not indicate that the gut microbes cause autism," emphasizes Gil Sharon, senior postdoctoral scholar in the Mazmanian lab and the study's first author. “Additional studies are needed to address the impact of gut bacteria in humans.”

Repetitive and social behavioral abnormalities in mice with microbiomes from patients with autism spectrum disorder can be corrected by the administration of specific metabolites.

The communities of microorganisms that inhabit the human gut are called the microbiota, and their collective genomes are known as the microbiome. These organisms live in a symbiotic state with humans. In exchange for a warm and nutrient-rich environment, bacteria help us digest food, affect metabolism and educate our immune system.

To examine the microbiota's role in autism-like behavior in mice, the team used "germ-free" mice — laboratory animals that are grown in the absence of microorganisms. Gut microorganisms from children with autism were transferred into these mice via fecal transplantation, and samples from people without autism were transplanted into other groups of animals.

The mice with microbiota from individuals with ASD exhibited autism-like behaviors. Specifically, they spent less time socially interacting with other mice, vocalized less, and exhibited repetitive behaviors. These symptoms are analogous to behavioral characteristics of people with ASD. The mice harboring microbiota from typically developed individuals did not show these symptoms.

In addition to the behavioral differences, mice colonized with human ASD microbiota also showed altered gene expression in their brains and differences in the types of metabolites present (metabolites are the molecules produced as byproducts of digestion and microbial metabolism). Two metabolites in particular were found in lower amounts in these mice: 5-aminovaleric acid (5AV) and taurine. ASD is sometimes characterized by an imbalance in the ratio of excitation and inhibition in the brain, so the researchers were intrigued by the lower amounts of 5AV and taurine, as both affect certain inhibitory neural receptors called GABA receptors.

"We were surprised to see how profound the effects were," said Sharon.

"What I find truly fascinating is that by just transferring fecal samples from human children with autism to the mice moms, the offspring (mice kids) had behavioral deficits,” said Krajmalnik-Brown.

The team then turned to mice that offer a different model of autism. This strain of mice, called BTBR mice, naturally exhibits autism-like behaviors. The researchers wanted to see if treating these mice with 5AV or taurine would reduce these behavioral symptoms. Notably, treated mice did indeed show decreases in autism-like behaviors. Further, brain examinations showed that 5AV, in particular, decreased neural excitability.

"There are many factors that make autism more complicated in humans than in mice. In mice, we can model the symptoms of the disorder but not reproduce it," said Mazmanian. "However, this research provides clues into the role that the gut microbiota plays in neural changes that are associated with ASD. It suggests that ASD symptoms may one day be remedied with bacterial metabolites or a probiotic drug. Further, it opens the possibility that ASD, and perhaps other classical neurologic conditions, may be treated by therapies that target the gut rather than the brain, a seemingly more tractable approach."

Dae-Wook Kang, formerly a research scientist in Krajmalnik-Brown’s team at the Biodesign Swette Center for Environmental Technology, acquired the human fecal samples used to colonize the guts of mice and associated data and performed the microbiome sequencing and analysis. Kang is now an assistant professor at the University of Toledo.

“Dr. Mazmanian’s compelling observation provides us with more insight on the gut and brain axis, and strengthens the knowledge base in our evolving work with autism,” said Kang.  

“Multidisciplinary collaborations are critical to gaining a better understanding of the root causes of autism,” said Krajmalnik-Brown. “Scientists who focus on animal models and those who focus on humans can learn a lot from one another, advancing knowledge in ways that bring us closer to solutions.” 

The paper is titled "Human Gut Microbiota from Autism Spectrum Disorder Promote Behavioral Symptoms in Mice." In addition to Sharon and Mazmanian, Caltech-affiliated co-authors are Nikki Jamie Cruz, research technician assistant; Bo Wang, postdoctoral scholar; Michael Sweredoski, senior bioinformatician; Annie Moradian, senior lab manager and research scientist; and Carlos Lois, research professor. Additional co-authors are Dae-Wook Kang of the University of Toledo; Michael Gandal and Daniel Geschwind of UCLA; Young-Mo Kim, Erika Zink, Cameron Casey, Lisa Bramer, Nancy Isern, David Hoyt, Janet Jansson and Thomas Metz of Pacific Northwest National Laboratory; Bryn Taylor and Rob Knight of UC San Diego; Christianne Lane of USC; Cecilia Noecker of the University of Washington; Elhanan Borenstein of the University of Washington, Tel Aviv University, and the Santa Fe Institute; and Krajmalnik-Brown.

Funding was provided by Pacific Northwest National Laboratory, an Autism Speaks Meixner Postdoctoral Fellowship in Translation Research, a Human Frontiers Science Program Long-Term Fellowship, a SFARI Bridge to Independence Award, the San Diego Diversity Fellowship, the National Biomedical Computation Resource, the National Institutes of Health, the Autism Research Institute, the Emch Foundation, the Brenen Hornstein Autism Research and Education Foundation, Lynda and Blaine Fetter, the Simons Foundation, and the Heritage Medical Research Institute. Mazmanian is an affiliated faculty member of the Tianqiao and Chrissy Chen Institute for Neuroscience at Caltech.

Adapted from Lori Dajose, California Institute of Technology