Inflammation Rewires the Enteric Nervous System Through Neurogenic Monocyte Recruitment

The proper organization of the enteric nervous system (ENS) is critical for normal gastrointestinal (GI) physiology. Inflammatory bowel disease (IBD) dysregulates GI physiology including bowel movements (motility), but in many IBD patients, GI motility disorders persist in remission through a poorly understood pathological process. Here we uncover that post-inflammatory GI dysmotility (PI-GID) stems from structural ENS remodeling driven by a combination of neuronal loss and neurogenesis. Enteric neurons respond to mucosal inflammation by upregulating CCL2 expression and facilitating the recruitment of CCR2+ monocytes into the neural myenteric plexus of the intestinal muscle followed by the expansion of monocyte-derived macrophages, their migration into the myenteric ganglia and phagocytosis of neurons. However, excessive recruitment of monocytes promotes disproportionate ENS remodeling and PI-GID. Expansion of immune cells in the tissue also promotes tissue hypoxia. We find that enteric neurons are hypoxic upon colitis; hypoxia-induced signaling via HIF1alpha induces an adaptation program in enteric neurons to suppress CCL2 expression and limit monocyte recruitment. We demonstrate that reinforcing HIF1alpha signaling in enteric neurons prevents PI-GID by reducing colitis-associated monocyte recruitment in the myenteric plexus and mitigating ENS remodeling. In summary, our findings unveil PI-GID pathogenesis and identify a regulatory axis for its prevention. Overall design: 8-9 week-old female Cx3Cr1+/GFP Ccr2+/RFP littermates were used to identify and sort different macrophage subsets present in the colonic muscularis during colitis. Mice received sterile 3% dextran sodium sulfate (DSS) in drinking water for 5 consecutive days followed by sterile water till they reached peak of inflammation. Control mice received sterile water throughout the experiment. DSS treated mice were euthanized at the peak of inflammation i.e. D7-8 along with control mice (n=3 to 4, per group). Single cell suspensions prepared from isolated colonic muscularis tissues were pooled to generate 1 replicate per group. In total, 5 such replicates were prepared thus representing 5 independent sorting experiments. BD FACSAria™ Fusion Flow cytometer was used to sort each replicate into the following subsets: CCR2+, CCR2+Ly6c+MHCII+, CCR2+MHCII+, MHCII+ and Ly6c-MHCII- (DN, double negative). Total RNA was isolated from sorted cell populations using Trizol protocol.