Stringent feedback regulation of the aryl hydrocarbon receptor (AHR) helps maintain gut health, signalling to both the colonic mucosa and the enteric nervous system
A diet rich in AHR ligands can affect intestinal immunity and the barrier function of the gut wall, protecting from infection and cancer
Via the enteric nervous system, AHR signalling has a direct effect on regulation of peristalsis in response to changes in the gut flora
The lining of the intestines must act both as a barrier and a sensor of the contents of the gut lumen. Work from the laboratories of Brigitta Stockinger and Vassilis Pachnis has identified the aryl hydrocarbon receptor (AHR) as a crucial component of intestinal health, with roles in regulation of mucosal immunity, regeneration and repair of the intestinal epithelium, and most recently, in the control of gut peristalsis via the enteric nervous system.
One of a family of evolutionarily conserved environment sensors, AHR is a ligand-dependent transcription factor, first identified as a molecular villain, as it facilitates the effects of environmental pollutants such as dioxin. However, it is now clear AHR also has many beneficial effects. Notably, AHR-/- mice are fatally susceptible to the pathogen Citrobacter rodentium, as they are unable to maintain the barrier function of the gut wall in the face of infection.
The gut lumen contains many AHR ligands, derived from dietary components, xenobiotics and the microbiota. Due to this abundance, AHR’s activity is unlikely to be limited by ligand availability. Rather, there is a stringent feedback loop: AHR activation leads to upregulation of negative regulators such as the cytochrome P4501 (CYP1) enzymes Cyp1A1 and Cyp1B1, which oxygenate AHR ligands, leading to their metabolic clearance, and shutdown of the AHR pathway. The existence of this feedback loop suggests that curtailing AHR signalling is important.
Some years ago, the Stockinger lab showed that AHR is expressed in a specialised subset of T helper cells, TH17, which play an important role in maintaining mucosal barriers, and contribute to pathogen clearance at mucosal surfaces. TH17 cells depend on AHR for the induction of IL-22, whereas other cells of the intestinal immune system depend on AHR for their survival (reviewed in Stockinger et al, 2020).
Given AHR’s important role in intestinal immunity, Stockinger and colleagues decided to look more closely at the regulation of AHR signalling in the gut, and in 2017, they established the physiological importance of the CYP1-mediated feedback loop, and also discovered that intestinal epithelial cells are the gatekeepers for the supply of AHR ligands to the host (Schiering et al, 2017).
Schiering and coworkers showed that in mice in which the Cyp1a1 gene was overexpressed either constitutively, or specifically in the intestinal epithelial cells, the reservoir of natural AHR ligands was depleted due to their being more rapidly metabolised, which resulted in a quasi AHR-deficient phenotype: development of TH17 cells and other AHR-dependent immune cell types was compromised, and mice were highly susceptible to fatal C. rodentium infection. Feeding the Cyp1a1 overexpressing mice a diet rich in high affinity AHR ligands was able to partially correct the phenotype, demonstrating the sensitivity and importance of the feedback loop, and the critical role of intestinal epithelial cells in controlling ligand availability.
The following year, Stockinger’s group further defined the role of AHR in protecting the gut from C. rodentium infection (Metidji et al, 2018), when they showed that targeted knockout of AHR in intestinal epithelial cells resulted in loss of barrier integrity. This was due to a failure of intestinal stem cells to differentiate and replenish pathogen-damaged intestinal epithelial cells. In a further twist, stem cell over-proliferation, coupled to inflammation caused by barrier disruption, led to malignant transformation.
Working with colon organoids in collaboration with Vivian Li’s laboratory, Stockinger’s group went on to show that either loss of AHR or overexpression of Cyp1a1 resulted in dysregulation of the Wnt-β-catenin signalling pathway, leading to excess proliferation in stem cell crypts. Proliferation in response to Cyp1a1 overexpression could be rescued by adding high affinity AHR ligands to the cultures.
These data suggested that not only is stimulation of the AHR pathway required for epithelial cell differentiation from crypt stem cells, but that it might also be a guardian of the stem cell niche, protecting against colon cancer. To test this idea, mice overexpressing Cyp1a1 in their intestinal epithelial cells were fed a diet supplemented by the high affinity AHR ligand indole-3-carbinol (I3C) either prior to or during experimental induction of tumours. In line with the colon organoid work, I3C ingestion could reduce or halt tumour formation relative to controls.
With this paper, Metidji and coworkers may have solved a long standing mystery: much evidence suggests that eating a diet rich in green vegetables reduces the risk of colorectal carcinoma, but the mechanism for this was unknown. Strikingly, some of the highest affinity dietary AHR ligands are derived from phytochemicals, strongly suggesting that the AHR pathway is intimately involved.
Stockinger’s work looking at how the gut is protected from injury and inflammation by the AHR pathway has now dovetailed neatly with work from the laboratory of Vassilis Pachnis. Pachnis is interested in the enteric nervous system (ENS), which regulates most aspects of gastrointestinal physiology — for example, peristalsis, the blood supply to the gut wall, and secretion by epithelial cells — and is a relay station in the bidirectional neuroendocrine pathway that connects the digestive system with the brain. He and Stockinger are collaborating in an exploration of AHR’s role in the ENS.
The ENS is composed of a vast number of neurons and glial cells which are organised into two concentric networks of interconnected ganglia within the gut wall, known as the myenteric plexus and the submucusal plexus. Often called the “second brain” because of its sheer number of cells and complexity, the ENS is experimentally very challenging, as the ganglia are tightly packed and layered in close apposition to other tissues, such as the contractile sheets of smooth muscle. Due to this organisation and the extensive branching of enteric neurons and glial cells, it is notoriously difficult to isolate ENS cells for molecular analysis without damaging them and thereby triggering expression of genes involved in cell stress and injury.
To circumvent these problems, Yuuki Obata, a postdoc in the Pachnis lab, discovered that he could specifically label enteric neuronal nuclei by tail-vein injection of a viral vector (derived from adeno-associated virus) expressing nuclear-localised GFP under the control of a neuron-specific promoter. This enabled him to isolate pure populations of neuronal nuclei from the gut by flow sorting and subject them to molecular and transcriptional analysis.
Using these techniques, enteric neuron nuclei were purified from different parts of the mouse gut, and their transcriptional differences were characterised by bulk RNA-seq. Comparison of the transcriptional profiles of neuronal nuclei isolated from the colon and small intestine of specific pathogen free (SPF) and germ-free mice, showed that there were segment-specific gene expression programmes, but this regionalisation was generally similar between the two types of experimental animals, suggesting that many of the recorded differences were independent of the presence or absence of gut microbiota. However, overlaid on this pattern was another, microbiota-specific one: looking in neurons from the colon (which contains the highest load of microbiota) for genes upregulated specifically in response to microbial colonisation yielded three hits, one of which was AHR. Immunostaining confirmed that in conventional mice, colonic neurons expressed the highest levels of AHR, and that this expression was dramatically reduced in microbiota-depleted animals (either germ-free, or treated with antibiotics). Interestingly, expression of AHR in colonic neurons of germ-free mice could be restored by re-colonisation of the gut with microbiota.
Since AHR functions as a transcription factor, its activation by gut microbes and their products is expected to alter the transcriptional programmes and functionality of neural circuits in the colon. To find potential AHR effector genes, the nuclear transcriptomes of enteric neurons from the colons of control mice and mice treated with the AHR ligand 3-methylcholanthrene were examined. Pleasingly, the 30 ligand-induced genes identified included Cyp1a1, but most intriguing was the presence of Kcnj12, a gene whose product, Kir2.2, encodes a potassium channel known to regulate the excitability of cardiac muscle and neuronal cells. Previous electrophysiological studies have detected an electrical current driven by this type of potassium channel in mammalian enteric neurons, and addition of a pharmacological blocker specific for the subfamily of channels to which Kir2.2 belongs altered the electrical firing of enteric neurons. Analysis of AhrEN-KO mice, which lack AHR specifically in enteric neurons, showed they had significantly lower Kcnj12 transcription, confirming this gene as a regulatory target. Together, these studies pointed to a molecular link between AHR and the functional output of intestinal neural circuits, such as peristalsis.
Measurement of the intestinal transit time (ITT) of AhrEN-KO animals demonstrated that this was the case: their total ITT was increased relative to controls, and the frequency and organisation of colonic migrating motor complexes (a peristaltic programme of the colon that depends on the activity of intrinsic neural circuits) was also reduced. The increased ITT time was also replicated in mice engineered to overexpress Cyp1a1 in their enteric neurons, and could be largely rescued by supplementing their diet with I3C. Finally, to show that AHR signalling regulates intestinal mobility in response to microbiota, neuronal-specific overexpression of AHR could partially rescue the marked increase in ITT of antibiotic-treated mice.
Cell-autonomous activity of AHR in enteric neurons is therefore the cornerstone of a mechanism that allows the ENS to constantly monitor the lumen of the gut in terms of diet and microbial products, and to respond to physiological or potentially harmful changes in contents by adjusting intestinal motility as a means to maintain or restore normality.
These papers shed light on an important and hitherto unrecognised control system, but also have significant clinical implications for gut health. Pharmacological or dietary interventions to regulate AHR activity are likely to be useful in treating defects of intestinal homeostasis on a number of levels - the repair and health of intestinal epithelial cells, modulation of the mucosal immune system, and regulation of peristalsis by the nervous system.