Date of Award

Spring 5-15-2019

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Neurosciences)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



The enteric nervous system (ENS) is a complex interconnected network of neurons and glia in the bowel wall that regulates intestinal motility, blood flow, and epithelial function. The ENS also controls aspects of inflammatory signaling within the bowel. To perform these tasks, there are at least 20 types of enteric neuron and four types of enteric glia. Although much is known about early events in ENS development, signals governing the development of specific neuronal subtypes and communication with neighboring cell types within the bowel remain poorly understood. One fundamental hypothesis is that diverse trophic factors support distinct neuronal populations in the bowel.

Based on our observations that the hepatocyte growth factor (HGF) receptor Met is expressed in all ENS precursors and in a subset of adult enteric neurons, we investigated the role of HGF and MET in the ENS. We found that mice lacking functional MET receptor in the ENS (Met cKO mice) exhibit defects in a subset of CGRP-expressing sensory neurons in the myenteric plexus. These sensory cells, known as intrinsic primary afferent neurons (IPANs), are responsible for transmitting signals such as bowel stretch and villus deformation from the bowel lumen to other myenteric neurons. Met cKO mice have altered CGRP-IR neurite patterning with a corresponding failure to trigger peristalsis in response to villus stroking, but a normal response to bowel stretch. This work suggests that MET is important for the development and function of a subset of IPANs the bowel.

HGF has also been known for more than a decade to protect the bowel from injury in colitis models. It had been assumed that HGF’s protective effects were mediated by MET present on bowel epithelial cells. Using Met cKO mice that are missing MET in the ENS, but have normal MET in gut epithelium, we showed that neuronal HGF signaling contributes to the protective effects of HGF in the bowel that have been previously reported. Met cKO mice have significantly more bowel injury following treatment with dextran sodium sulfate (DSS) to induce colitis. Furthermore, these animals show reduced proliferation of epithelial cells in the context of injury. Together these findings suggest that that HGF/MET signaling is important for development and function of a subset IPANs and that these cells regulate intestinal motility and epithelial cell proliferation in response to bowel injury.

Our studies of Met cKO mice highlighted how defects in neurite patterning of enteric neurons can profoundly affect bowel function. Surprisingly little is known about factors that govern appropriate formation of neuronal connections in the bowel. To identify other cues important for enteric neuron axon patterning, we used broad microarray gene expression profiling. Gene expression in embryonic day 17.5 (E17.5) ENS cells was compared to gene expression in surrounding cells to identify genes encoding cell surface receptors or adhesion molecules enriched in the ENS that have known roles in axon pathfinding. We found that the semaphorin receptor Plexin-A4 was highly enriched in the ENS and confirmed this by in situ hybridization (ISH) at E17.5. Immunohistochemistry using a commercial antibody to Plexin-A4 suggested that Plexin-A4 was found in a subset of calretinin-IR neurons. However, staining of Plexin A4 knockout (Plexin A4 KO) bowel revealed this antibody to be non-specific. Despite testing two other Plexin-A4 antibodies, we were unable to determine which enteric neurons produce Plexin A4. We analyzed the ENS of Plexin A4 knockout (Plexin A4 KO) mice but could not identify any gross defects in neurite patterning. Additionally, functional analysis of Plexin A4 KO mice did not reveal any defects in the peristaltic reflex of these animals. As is true for other parts of the nervous system, it is likely that Plexin A4 acts redundantly with Plexin A2 (also enriched in the ENS) and any effects on axon pathfinding in the ENS would only be revealed in Plexin A4/Plexin A2 double knockout mice.

Remarkably, the non-specific Plexin A4 antibody we had been using also labeled a population of poorly characterized muscularis macrophages within the bowel muscularis externa, and we decided to study these cells. Bowel macrophages integrate a variety of environmental stimuli to assume either a pro-inflammatory or tissue-protective phenotype. A growing body of evidence suggests that neuronal cholinergic and noradrenergic signaling dampens the inflammatory phenotype of muscularis macrophages found in close contact with enteric neurons. Additionally, it’s been suggested that enteric neurons produce CSF1, the main survival factor for muscularis macrophages. This would imply that these macrophages would be abnormal when the ENS is missing. Surprisingly, we found that muscularis macrophage colonization of the bowel precedes colonization by enteric neurons and that the main source of CSF1 during development is non-neuronal. Furthermore, Ret knockout mice that are completely missing enteric neurons in the small bowel and colon contain normal numbers of well patterned macrophages. Additionally, these macrophages do not differ in their expression of activating cell surface markers, or in their ex-vivo response to lipopolysaccharide (LPS) stimulation. These studies help clarify the role of ENS in the homeostasis and activation of muscularis macrophages, suggesting that, at least developmentally, enteric neurons are dispensable for muscularis macrophage survival and do not alter the baseline inflammatory status of muscularis macrophages.


English (en)

Chair and Committee

Robert O. Heuckeroth

Committee Members

Aaron DiAntonio, Paul Bridgman, Joseph Dougherty, Kelly Monk,


Permanent URL: https://doi.org/10.7936/e7gg-8082