A nod to gut–brain signalling: Nod‐like receptors are critical for gut–brain axis signalling in mice

Exposure to highly stressful events in early life can impact gastrointestinal (GI) function and increase an individual's probability of developing an inflammatory bowel disease (IBD) as an adult. For example, exposure to trauma early in life induces visceral hypersensitivity and increases gut permeability in adulthood. The mechanism for this change likely involves the hypothalamic–pituitary–adrenal axis, as individuals exposed to stress show an enduring hyper-responsiveness, and GI changes have been found to be closely mediated by corticotropin releasing factor, as well as nerve growth factor and mast cells (Theodorou, 2013). Gut–brain axis signalling in inflammatory diseases likely involves serotonergic signalling, as levels of 5-HT are elevated in inflamed mucosa from Crohn's disease patients and in animal models of other IBDs. A mouse model deficient for the expression of tryptophan hydroxylase 1 – the enzyme responsible for the production of 5-HT in intestinal epithelia – shows a protective effect from developing colitis in two different chemical models. Mice deficient for serotonin reuptake transporter (SERT) develop more severe colitis in both chemical and transgenic animal models of the disease (Levin & van den Brink, 2014). Nucleotide oligomerization domain 1 (Nod1) is a cytosolic receptor responsible for triggering an innate immune response to antigens present within the cytoplasm. An exacerbated defence response to commensal microbiota is observed in IBDs. A negative feedback pathway appears to exist between Nod1 and SERT; Nod1 activation inhibits SERT activity and expression in human enterocyte-like Caco-2/TC7 cells (Layunta et al. 2018). Therefore, Nod1 likely plays a role in gut–brain axis signalling involved in IBD pathologies. Building on this prior knowledge, in a recent article published in The Journal of Physiology, Pusceddu et al. (2019) used Nod knockout animals to elucidate the role of Nod receptors in regulating gut–brain signalling. Deletion of Nod1/2 resulted in increased stress-like behaviour and levels of corticosterone, with decreased levels of serotonin. Together their data suggest a role for intestinal Nod1 in regulating stress-induced anxiety-like behaviour and cognition, and as a modulator of gut–brain communication. Pusceddu et al. (2019) utilized a Nod1/Nod2 double systemic knockout mouse model (NodDKO) and an intestine-specific single knockout model of either Nod1 or Nod2. Intestine-specific single knockout was achieved using loxP sites flanking either gene crossbred with IEC-specific Cre-Expressing (VilCre) mice: VilCre+Nod1f/f and VilCre+Nod2f/f mice. Animals were bred and maintained under specific-pathogen-free conditions. The water avoidance stress (WAS) task was used as a model of induced stress. The light/dark box test (L/D box test) and open field task (OFT) were used to assess anxiety-like behaviour. The novel object recognition (NOR) task was used to assess recognition memory. To investigate the role of serotonin, some NodDKO animals underwent chronic treatment with the selective serotonin reuptake inhibitor Fluoxetine for 28 days. NodDKO and VilCre+Nod1f/f animals but not VilCre+Nod2f/f animals showed higher levels of anxiety and decreased recognition memory compared to wild-type (WT) animals following WAS. This was accompanied by higher levels of serum corticosterone. Both findings were reversed in NodDKO mice treated with fluoxetine. In NodDKO mice, decreased recognition memory was accompanied by decreased neural activation in the dentate gyrus and cornus ammonium 3 (CA3) regions of the hippocampus, as measured by the expression of Arc and c-Fos – immediate early genes indicative of neural activation – and quantification of cell proliferation using Ki67 and DCX staining. Because serotonin signalling is able to influence adult hippocampal neurogenesis, Pusceddu et al. (2019) investigated hippocampal levels of expression of enzymes within the serotonin signalling pathway: Sert, Tph2 and 5HTr1a. Expression of all three were reduced in the hippocampus of NodDKO animals compared to WT in baseline animals and following WAS exposure. This suggests Nod1 or Nod2 may play a role in regulating 5-HT levels in the hippocampus even in the absence of stressful stimuli. Analysis of expression of enzymes of the serotonin signalling pathway in intestinal tissues showed mixed results. Levels of gene expression were analysed in colon and ileum samples from NodDKO and VilCre+Nod1f/f mice; data were not reported for VilCre+Nod2f/f mice, likely because they showed no behavioural deficits, as reported above. Intestinal tissue generally showed an increase in 5HTr1a in both WAS exposed and non-exposed animals with only one exception: WAS exposure appeared to normalize 5HTr1a expression in the colon of NodDKO mice. Also worth noting is a colon-specific decrease in expression of Tph1 in VilCre+Nod1f/f WAS-exposed mice, not observed in NodDKO mice. To assess changes in intestinal physiology researchers used a Ussing chamber to measure permeability in the ileum and colon. Samples showed a significant increase in permeability in WAS-exposed NodDKO mice. Fluoxetine treatment did not affect these findings. Mice with intestine-specific Nod1 or Nod2 knockout showed largely similar trends to those of double knockout mice with the exception of colon preparations from VilCre+Nod1f/f mice, which showed no difference in permeability from WT control animals. This suggests that Nod1 may preferentially influence the permeability of the ileum over the colon. As a means of explaining differences in serum 5-HT levels, the level of tryptophan (Trp) was quantified in serum, brain stem and hippocampal tissue. Trp is the sole precursor of 5-HT and reduced serum levels are known to correlate with alterations in mood and cognition. NodDKO showed significantly lower levels of Trp in serum and brain stem tissue following WAS exposure, though no significant difference from WT control mice in hippocampal tissue. Fluoxetine treatment was sufficient to reverse this effect. Overall, this paper presents a strong case for intestinal Nod1 affecting anxiety-like behaviour, although the exact mechanism through which it exerts this effect requires further elucidation. The data presented indicate a link between Nod1 intestinal expression and cognitive or mood defects. In the case of anxiety-like behaviour, changes in the expression of Nod1 – but not Nod2 – in the intestines is clearly sufficient to impact performance in the L/D box and OFT. Use of fluoxetine treatment in the double knockout model provides convincing evidence that serotonin plays a critical role in this process. Likewise, performance on the NOR task, a measure of recognition memory, is impacted by Nod1 expression and recovered by fluoxetine treatment in the double knockout model. The authors provide strong evidence to suggest that these differences are due to deficits in neural activation and cell proliferation. Evidence for the role of intestinal Nod1 would be improved if the authors assessed neural activation and cell proliferation in the VilCre+Nod1f/f mouse model as well, though given that hippocampal concentrations of serotonin signalling biomarkers were consistently reduced between NodDKO and VilCre+Nod1f/f mice, and that the authors cite past research implicating serotonin signalling in neural activation and proliferation, it is reasonable to infer that the differences would have been consistent in the single transgenic animal. In terms of the molecular mechanism behind these cognitive and mood defects, it is clear that Nod1/2 knockouts influence serotonergic signalling, though we look forward to future work elucidating this interaction more thoroughly. The technique used here, qPCR, provides a snapshot of gene expression at the time the animal was killed but does necessarily give definitive evidence for the effects of serotonin signalling at a functional level. Protein concentration, post-translational modification, protein sequestration, and other factors contribute to serotonergic signalling. Though some of these factors may seem outside of the scope of this paper, we would be interested to see tissue protein concentrations from naive animals as it would provide insight into the long term effects of Nod1/2 knockouts at the protein level and could be readily obtained from TRIzol-homogenized tissue, as used here. The use of Ussing chambers for the analysis of intestinal permeability clearly indicates a role for Nod1/2 in intestinal physiology. In the absence of Nod1/2, permeability was generally decreased compared to WT animals, indicating that Nod1/2 are necessary for maintaining normal intestinal physiology. Interestingly, intestine-specific knockout of either Nod1 or Nod2 did not result in a significant change in colon permeability, despite the fact that double knockout systemic mice consistently did show a change. This suggests that permeability in the colon may be regulated by Nod1 or Nod2 from elsewhere in the body. Finally, we found the analysis of serum Trp levels to provide a reasonable, intuitive explanation for intestinal Nod1/2 regulation of serotonergic signalling. However, this evidence would be strongly improved if Pusceddu et al. (2019) had repeated this analysis in mice expressing intestine-specific knockout of Nod1/2. We will be interested to see future work focusing on the mechanism by which Nod1/2 affects the absorption of Trp. Overall this paper clearly illustrates a link between intestinal Nod1 and cognitive and behavioural defects in mice. Nod1/2 affects serotonergic signalling, likely by affecting Trp absorption in the intestines. Nod1/2 also appear to affect intestinal physiology, though whether these effects are due exclusively to their expression in enterocytes or if they exert this influence via another region of the brain is still unclear. This study offers an excellent starting point for future research to explore the role of Nod1/2 in the gut–brain signalling axis. None declared. All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. None.