Published on 11 October 2018

Laure Tabouy, researcher at the Neuroscience Institute (Université Paris Sud – Université Paris-Saclay), presents the results of research carried out on mouse models, showing the links that exist between the host’s genome and gut flora, and opening the door to new possibilities in the treatment of autism, using probiotics.

Neurodevelopmental conditions, including autism, have a strong genetic etiology: certain gene mutations influence the severity of the disorder. However, the part played by these genes is not restricted to our bodies.

Our study shows that some of the micro-organisms which make up gut flora are sensitive to the mutations of genes associated with autism. Their number decreases when the latter are inactivated. And we now know that our gut flora plays an important part not only in our digestion, but also in other body functions, including the nervous system.

By restoring the balance of the gut flora in mouse models for autism, we have successfully attenuated their symptoms. This study suggests that we may, in the future, consider treating patients suffering from certain autism spectrum disorders with probiotics.

Here is the fascinating story of the strong interaction between gut bacteria and their autistic host.

Genes influencing the severity of the disorder

According to the DSM-5, autism spectrum disorders (ASD) cover a range of neurodevelopmental conditions characterized by persistent deficits in communication and social interaction, restricted and repetitive behaviors, interests and/or activities, hyper or hypoactivity, with or without intellectual disabilities, with or without language delay.

These disorders, which generally appear before the age of 3, bear the mark of each individual’s personality and depend on their day-to-day environment as well as on their genetic predisposition. The current estimate is that ASDs affect 1 person in 68. A ratio of 4 boys diagnosed for every girl is generally observed, but it seems that the number of girls diagnosed is underestimated.


A 2014 study established a correlation between the mutations of the SHANK gene family (SHANK1, SHANK2 and SHANK3, which come into play in the development and function of neuronal circuits), the severity of the disorder and patients’ physical characteristics. Among the mutations affecting the three SHANK genes, patients with a SHANK3 gene mutation have a lower intellectual quotient than patients from the other two sub-groups.

The microbiome plays a key part in the proper functioning of the body

Recent studies have shown that certain neurodevelopment disorders could be linked to microbiome imbalance. The word “microbiome” describes all the non-pathogenic microorganisms (bacteria, viruses, parasites and fungi) which can be found on the skin, in the mouth and also in the respiratory, uro-genital and digestive systems.

In our digestive system, the gut microbiome (or “gut flora”) alone has up to 100 000 billion microorganisms, representing around 2 kg of total body weight. We now know that it plays an important part not only in digestion and metabolism, but also in immune and neurological processes. A change in the composition or function of gut flora (also called “dysbiosis”) is associated with a number of pathologies, including Alzheimer’s and Parkinson’s, as well as addiction, depression, schizophrenia and autism.

The relationship between the gut microbiome and the brain is a direct one: microorganisms produce compounds that act on several levels. Some of these molecules are similar to those used by neurons to communicate among each other, others can activate the immune system or act on hormonal regulation.

Can genes associated with neurodevelopmental disorders such as autism influence the gut flora? Conversely, could an altered flora make these conditions worse?

Autistic people’s gut flora is different to other people’s

To find out, our team first analyzed the profile of the gut microbiome in mice with a SHANK3 mutation and compared it to that of “normal” (control) mice. The idea was to establish what species of bacteria lived in the gut of both types of mouse, in what quantities and so on. The composition of our microbiome depends on where we were born, the places we live in, the way we eat, our habits, etc. It is unique to us, just like fingerprints are.

The results show that the microbiome of mice carrying a gene mutation is different to that of “normal” mice, both in terms of its composition and in the range of bacteria present: certain phyla, types and species of bacteria are over or under-represented.

Looking at the two main bacteria types in the human gut flora, Firmicutes and Bacteroidetes, the first group is over-represented, whereas the second is under-represented, with the decrease being sharper in females with SHANK3 mutations compared to males.

The following types of bacteria are under-represented in mutant mice: Prevotella, (oral, vaginal and intestinal bacteria, associated with obesity, gut and upper respiratory tract inflammation, etc.) Christensenella (gut bacteria, associated with obesity, le cholesterol, diabetes, etc.), Streptococcus (naturally present in mucous membranes, mouth, upper respiratory tract, digestive system), Coprococcus (fecal microbiome bacteria) and especially Lactobacillus (vaginal, oral and intestinal lactic bacteria, present in breast milk and used in the manufacture of fermented products).

On the other hand, Veillonella (bacteria in the gut and mouth, associated with tooth decay, chronic fatigue syndrome, etc.) and Akkermansia (gut bacteria, associated with obesity, diabetes, gut inflammation, etc. are over-represented.

We focused our research on the Lactobacillus type, specifically on the Lactobacillus reuteri (L. reuteri) species. These probiotic lactic bacteria are sold in pharmacies to help prevent colic in infants. They produce reuterin, an antibiotic compound which inhibits the growth of potentially pathogenic bacteria such as Escherichia coli.

Gut bacteria modifying brain activity

Recent studies show that L. reuteri produces and secretes GABA (or a very similar molecule), a compound which is used in neural communication. It acts directly on the brain by binding itself to other molecules on the cells (called “receptors”).

As a reminder, we note that SHANK3 mutant mice display two of the most common characteristics of autism: repetitive and compulsive behavior and avoiding social interaction. There is a difference in the number of receptors in several molecules, including GABA, between mutant and normal mice. For instance, there are fewer GABA receptors in the hippocampus (an area of the brain associated with memory), whereas there are more of them in the prefrontal cortex (an area that plays an important part in emotions and mood swings). There also differences in other receptors such as those for oxytocin (a hormone secreted by the pituitary gland) and glutamate (an amino acid associated with neural transmission).

We also discovered that these differences correlated with the quantity and expression of L. reuteri bacteria in the gut. The correlations are either positive or negative but they are significant. Finally, the analysis of six cytokines (immune system hormones) in the plasma of mutant and control mice showed that the latter’s expression is no longer regulated in mutant mice; it also showed a significant a significant correlation between them and L. reuteri bacteria.

Our various studies and analyses led us to conclude that there was a real difference between SHANK3 mutant mice and control mice, both in the brain and in the immune system.

As these differences were correlated with differences in the composition of the gut microbiome as regards L. reuteri bacteria, we thought we would investigate this further.

L. reuteri to the rescue

L. reuteri is sold as a probiotic and recent studies have described it as having an impact on social behavior and stress. We therefore decided to force-feed these bacteria, which we grew ourselves, to four-week old mutant mice. We set up two groups: a control group consisting of those mutant mice which had not been given the bacteria and one consisting of those which had been given the bacteria. Following the L. reuteri-based feeding regime, social behavior tests were carried out and we then sacrificed the mice so as to examine the various receptors and hormones again.

Autism is a human pathology and this study, which uses animal models, does not examine all the characteristics of autism. However, the results of the behavioral studies we carried out were interesting. Firstly, males and females did not have the same response to the intake of L. reuteri. For the former, there was a decrease in social deficit and repetitive behavior. In males, L. reuteri stimulated their interest in mice they don’t know. On the other hand, there was very little effect on their anxious behavior.

Females reacted differently to L. reuteri bacteria. They proved to be less attracted to new social interactions. In addition, their repetitive behavior decreased significantly, as did their anxiety, although in that case not significantly. Molecular analysis showed that force-feeding with L. reuteri had a direct impact on the various areas of the brain and on the composition of the mice’s plasma as regards cytokines.  

The results tend to indicate that the gut microbiome could have had played a part in the development of the brain. Given that the SHANK3 gene is also expressed in intestinal neurons, it could be involved in the deterioration and modification of the microbiome. This would explain the differences in the gut flora of individuals suffering from an autism spectrum disorder because of a mutation of the SHANK3 gene and those who are unaffected.

Continuing to decipher how we interact with our microbiome

This study provides a first indication of the interaction that takes place between the host’s genetic profile and their gut flora. Further investigation is now needed to find out how an autistic individuals’ genome interacts with and modifies their gut microbiome.

These results open the door to new possibilities in the treatment of autism, using probiotics. Despite their transitory presence in the gut, they are active, and taking them produces an effect.

The interaction between the genome of autistic individuals and their microbiome is increasingly clear. We now need to understand it.

Laure Tabouy, doctor in neurosciences, CNRS contract researcher, Neuroscience Institute, Université Paris Sud – Université Paris-Saclay

The original version of this article was published on The Conversation.