"Our research emphasizes the growing consensus that disruption of the microbiome during development can have a significant impact on physiology, including the brain." Philip Burnet
If we want to study the effect of factor A on factor B, we can do so in several ways. First by simple observation, then by some manipulation of factor A: its alteration, reinforcement, weakening, or complete blockade.
There is no longer any doubt about the influence of the gut microbiome on the human organism. We know that a newborn acquires an optimal composition of the gut microbiome during vaginal delivery, while a child born by cesarean section is far from having an ideal composition and is more prone to allergies, atopy, asthma, and other problems.
One way to study the effect of the gut microbiome on behavior in animal experiments is to breed a germ-free (GF) individual: a pup delivered by cesarean section is placed in a sterile environment and fed a sterilely treated diet. Germ-free mice and rats exhibit greater motor activity in a novel environment. While standard mice significantly reduce motor activity after half an hour, GF individuals continue exploring even after sixty minutes.
Germ-free young mice and young rats exhibit behavioral elements reminiscent of autism: they show less interest in social contact, in new stimuli, and have a tendency toward repetitive behavior. However, when placed after weaning in an enclosure with standard peers, this behavior can normalize.
Other research also supports the existence of a causal relationship between gut microbiome dysbiosis and social behavior. Offspring of mice fed a high-fat diet also display some features of autistic behavior: impaired sociability, increased anxiety, and repetitive behavior. They also have fewer oxytocin-expressing neurons in the hypothalamus and gut dysbiosis. This condition is repairable by administering Lactobacillus reuteri, after which both oxytocin levels and behavior normalize.
There is no need to use complex techniques to produce germ-free mice; antibiotics also wreak havoc on the gut microbiome and disrupt natural diversity. Studies on rodents show that antibiotic administration induces cognitive deficits, visceral pain, and behavioral changes including autism-like behavior. For example, administration of low doses of penicillin in early life stages has a long-term effect on microbiome composition and long-term behavioral changes characterized by reduced sociability and diminished interest in new social stimuli. In both studies conducted so far examining this, supplementation with the microbe Lactobacillus rhamnosus concurrently with antibiotics protected the pups from these changes.
Behavioral disruption through microbiome manipulation is not specific to the earliest age. An antibiotic cocktail administered during the key stage of adolescence disrupts the gut microbiome, cognitive function, and social behavior in a similar manner. Reduced expression of the neuropeptides vasopressin and oxytocin is also demonstrated; the latter only in situations where the animal is exposed to stress. And similarly, administration of Lactobacillus reuteri can lead to an increase in plasma oxytocin.
Katerina Johnson and Philip Burnet from Oxford examined the common characteristics and differences that appear in a germ-free young mouse and a mouse whose microbiome was devastated by a strong antibiotic cocktail. They focused not only on observable behavior but also on the behavior of neurotransmitter systems, particularly opioids, oxytocin, and vasopressin, since these are the main modulators of social behavior. In a methodically precise experiment, they observed the behavior of both groups of animals in vivo and subsequently their brains, noting that concise information on behavioral disorders already exists but their neurobiological mechanism is not clear.
The results were somewhat surprising, as there was a significant difference between the two groups in changes in the frontal cortex. Regarding gene expression of the monitored neurotransmitters, a decline was observed in all three after antibiotic administration. For opioids, significance was p=0.021, for oxytocin p=0.016, and for vasopressin p=0.079. These changes were not observed in the hippocampus or the hypothalamus, but opioid expression was significantly increased in the brainstem (p=0.046). There was considerable interindividual variability, but all told: the combined downregulation across all regions was significant at the one-percent significance level.
In contrast to this group, germ-free mice showed only one change, and that in a completely opposite direction. In the frontal cortex, opioid receptor expression was significantly increased (p=0.018). (This is consistent with Panksepp's opioid hypothesis of autism, which the authors do not mention.) Also unlike the antibiotic model, expression of this gene decreased in the hypothalamus (p=0.046). Expression of proopiomelanocortin genes also decreased in the frontal cortex and other regions.
The authors conclude that although in both cases there is an adverse effect on brain function induced by dysbiosis, these are two distinct models, which must be taken into account in further considerations. Johnson said: "The adverse effect of antibiotics on the endorphin system may have an impact not only on social behavior but also on pain regulation. In fact, we know that the gut microbiome influences pain response, so this could be one of the ways it does that.
A somewhat surprising observation from our research was the contrast in results between germ-free mice and those treated with antibiotics, because the neurogenetic changes were generally in the opposite direction. This is a relevant finding because the use of antibiotics to deplete the microbiome is often considered a more accessible alternative to germ-free animals. However, we emphasize that when investigating the effects of microbes on the brain and behavior, these two treatments should be regarded as distinct models of microbiome manipulation."
Author: MUDr. Radkin Honzák psychiatrist
Katerina V. A. Johnson, Philip W. J. Burnet. Opposing effects of antibiotics and germ-free status on neuropeptide systems involved in social behaviour and pain regulation. BMC Neuroscience, 2020; 21 (1) DOI: 10.1186/s12868-020-00583-3