Skip to main content
Premium Trial:

Request an Annual Quote

Children With Autism May Have Underdeveloped Gut Microbiomes, Metagenomic Analysis Finds

NEW YORK – Children with autism spectrum disorder appear to have underdeveloped gut microbiomes compared to typically developing children, a new study has found.

While genetic factors have been implicated in autism spectrum disorder risk, the gut microbiome is also suspected of playing a role through the gut-brain axis.

"Disruption of the gut-brain axis has been shown to be involved in the pathogenesis of a diverse range of diseases, including Parkinson's disease and irritable bowel syndrome," Siew Ng, a professor of medicine and therapeutics at the Chinese University of Hong Kong and senior author of the new study, said in an email. "Therefore, we believe that children with ASD harbor [an] altered gut microbiome and these bacteria might play a role in disease pathogenesis."

Ng and her colleagues compared bacteria in stool samples from 128 children, with and without ASD diagnoses. As they reported in the journal Gut, they found that age, weight, and whether the children had ASD had the greatest influence on the makeup of their gut microbiomes, with the results further suggesting that the development of the gut microbiomes of children with autism lags behind that of their typically developing peers.

Five species of bacteria in particular differ between the gut microbiomes of children with ASD and typically developing children, suggesting those bacteria could potentially serve as disease biomarkers, the researchers wrote.

For their study, they recruited 64 Chinese children between ages 3 and 6 who were diagnosed with autism spectrum disorder and 64 age- and sex-matched typically developing Chinese children as controls and conducted metagenomic sequencing on their fecal samples.

The researchers examined whether a range of host factors affected the microbiome composition, noting that age, body mass index, and autism had the largest, and independent, effects on the makeup of the children's gut microbiomes. Diet, they noted, did not correlate with microbiome composition.

Microbial richness was higher among children with ASD than typically developing age- and BMI-matched children. At the species level, the two microbiome populations were distinct, and the presence of five bacterial species — Alistipes indistinctus, candidate division_TM7_isolate_TM7c, Streptococcus cristatus, Eubacterium limosum, and Streptococcus oligofermentans — varied between the groups.

When the researchers analyzed the samples by age and composition, they identified about two dozen bacterial species that appeared to be markers of typical gut microbiome development in children without ASD. But the association between these bacteria and age was disrupted in children with autism spectrum disorder, pointing toward abnormal gut microbiome development early in life in these children.

There were also differences in microbiome functionality between children with and without autism spectrum disorder, as pathways associated with neurotransmitter biosynthesis were decreased in the gut microbiomes of children with autism. For instance, chorismate biosynthesis I and PWY-6163 were decreased in the gut microbiomes of children with the disorder.

This, Ng said, could have effects via the microbiota-gut-brain axis in autism. "Reduced neurotransmission-related pathways and metabolites might impair normal physiologic and psychiatric activities," she added.

She and her colleagues also developed a random forest classifier based on the five species that differed between the gut microbiomes of children with and without ASD. This tool, they found, could differentiate children with ASD and those without 82.6 percent of the time, and 76.2 percent of the time in a validation cohort of 18 children.

Ng noted that diagnosing autism can be difficult as there is no definitive test, and that these bacteria could serve as noninvasive biomarkers for autism spectrum disorder risk. She and her colleagues are now developing a PCR-based diagnostic tool, with which they plan to evaluate these markers in a larger, independent cohort and across different populations.