Microbiota are recognized as key regulators of the gut-brain axis, but whether brain neurons can directly sense bacterial components, and conversely, if bacteria are involved in modulating physiological processes via the brain, has not been demonstrated.

Researchers at the Institute Pasteur in Paris have now shown that muropeptides – shed fragments of peptidoglycan, which is a major component of the bacterial cell wall – directly inhibit the activity of neurons in the hypothalamus to regulate appetite, nesting behavior and body temperature in mice.

This effect is mediated by nucleotide-binding oligomerization domain-containing protein (Nod2), a widely expressed pattern recognition receptor that signals the presence of viruses, bacteria and fungi on mucosal surfaces. Specifically, Nod2 recognizes muramyl dipeptide, a motif found in every type of bacterial peptidoglycan.

Although it is known that microbial metabolites, such as short-chain fatty acids and tryptophan derivatives, work through a variety of receptors to regulate many processes, this is said to be the first explicit demonstration that structural components of the bacterial microbiota can be directly sensed by neurons in the hypothalamus.

The finding could open up new approaches to treating metabolic disorders, but also has implications for cancer drugs targeting Nod2.

"One could think that muropeptides could be used as satiety drugs," said Ilana Gabanyi, neuroimmunologist and lead author of a paper in Science, April 15, describing the research. "Moreover, Nod2 ligands are already being tested in the clinic for cancer therapies for example, so our work shows that their effects in the brain must be taken in account when developing these drugs," Gabanyi told BioWorld Science.

Working from the understanding that mice lacking neuronal expression of Nod2 develop changes in food intake, nesting behavior and body temperature control, Gabanyi and colleagues investigated expression patterns of Nod2 in the CNS, using knockin mice that simultaneously expressed a functional Nod2 receptor and a green fluorescent protein.

Neurons that expressed the fluorescent protein were found in the striatum, thalamus and particularly in the hypothalamus.

To determine if muropeptides from the intestine can reach the brain, mice were fed radiolabeled muropeptides. Tissues collected 4 hours later showed the muropeptides had crossed the gut barrier to reach the blood circulation and had accumulated in the brain.

The same accumulation of muropeptides in the brain occurred when mice were colonized with Escherichia coli containing radiolabeled peptidoglycan. In this case, the tissues were examined after 24 hours, allowing time for fragments of peptidoglycan to be released.

Having shown muropeptides can reach the brain from the gut, the researchers moved on to assess if restricting neuronal expression of Nod2 in the hypothalamus – the control center for hormone production, homeostasis, appetite and temperature – affects brain-controlled metabolism and behavior in mice.

Mice that were genetically engineered to be deficient in Nod2 gained more weight than controls, a difference that became significant at 6 months and continued thereafter. In addition, temperature regulation was impaired and life span decreased.

This phenotype was far more marked in female than in male Nod2-deficient mice. As well as eating more at each meal and gaining more weight, female mice deficient in Nod2 in the hypothalamus had a tendency to develop diabetes.

As a further demonstration of muropeptide signaling from the gut to the brain, mice given antibiotics to ablate the gut microbiome had the same deficits as Nod2-deficient mice.

Taken together, the research shows there is a gut-brain communication pathway in which expression of the Nod2 receptor in hypothalamic neurons regulates appetite and body temperature in response to bacteria-derived muropeptides.

The transient increase in the gut microbial population on feeding prompts an increase in bacterial cell wall-derived muropeptides, generating a bacterial signal to stop eating.

But "it is very hard to know" if the Nod2 mechanism evolved to control appetite, or as a route by which the bacterial microbiota modulate the host's feeding behavior to stabilize its intestinal niche, said Gabanyi.

"Every time you eat, there is this increase in muropeptide release into circulation that also reaches the brain, at the same time as anorexic hormones are produced. It is possible that the brain understands this signal because it comes after eating [and] together with other anorexic signals," she said.

The particularly strong effects on the appetites of Nod2-deficient female mice over 6 months of age mirrors sex- and age-dependent phenotypes that occur in a number of brain and metabolism-related diseases.

Given this, understanding the mechanisms behind such biases may lead to more specific and efficient therapeutic approaches. "I want to understand the sex bias in the context of the gut-brain axis and uncover other bacterial compounds that affect brain neurons," Gabanyi said.

Evaluating different parameters showed that only increased accumulation of muropeptides in the female brain relative to the male brain correlated with sex-specific hypothalamic neuronal activation in response to Nod2.

That led Gabanyi to propose that muropeptide control of appetite and temperature takes over as the production of the female sex hormone estradiol declines with age.

Given similar changes in estradiol production are associated with weight gain and hot flashes in menopausal women, it is suggested muropeptides could be a nonhormonal target for drugs to treat menopausal symptoms.

The researchers also caution that the effects of Nod2 activation on brain neuronal activity must be considered in the development of peptidoglycan metabolites as immunotherapeutics.