Niche regulation of gastric epithelial cell differentiation

About This Grant

Project Summary Distinct stomach epithelial cell types, all arising from a common multipotential progenitor (stem cell), serve essential digestive or barrier functions. Defects in the fractions or properties of these cells are associated with various prevalent human disorders, including pre-neoplastic conditions. Pit (surface mucous), isthmus (epithelial self-renewal), neck (deep mucous), and chief (zymogenic) cells –each with distinctive morphology, functions and gene products– are precisely zonated along the long axis of gastric corpus glands. Informed by other self-renewing tissues –such as blood and intestine– and by growing appreciation of epithelial plasticity, each zonated cell type is regarded as a “committed” terminal product of a unidirectional cellular hierarchy that originates in proliferative isthmus cells and retains some capacity to “dedifferentiate”. However, we detect modest transcriptional or epigenetic distinction among the zonated cells and their phenotypes inter-convert readily in vitro and in vivo in response to specific cues: BMP signaling for pit cells and canonical Wnt signaling for neck/chief cells. Our preliminary data therefore suggest the radically different view that pit, neck, and chief cell properties are not hard-wired in the conventional sense of cell “determination”, but represent reversible, signal-responsive phenotypes within a continuum of native cell states directed by finely graded signals from specific sub-epithelial cells. Of mechanistic note, PDGFRA+ sub-epithelial mesenchymal cells positioned near pits are enriched for BMP expression, while Wnt/Rspo mRNAs localize in similar cells positioned near chief cells at the gland base; niche-derived NRG1 concentrates near the isthmus zone. We propose that isthmus cell replication and overt phenotypes of surface and deep corpus gland cells reflect the summation of these sub-epithelial signals. Challenging classical views of a gastric cell hierarchy, this project examines epithelial cell properties with respect to specific niche signals. Mechanistically, two Specific Aims address each side of this paracrine dialogue to decipher the epigenetic basis (Aim 1) and regulatory logic (Aim 2) of the mouse gastric corpus. In Aim 1, five discrete sub-Aims will collectively delineate determinants of zonated corpus epithelial cell properties. We will define how extracellular signals modulate groups of phenotype-restricted cis- regulatory enhancers to generate distinct phenotypes and test whether pit-biased (KLF4) and neck-biased (CREB3L4) transcription factors are required for these effects. Three sub-Aims in Aim 2 will define how key signaling components in the mesenchymal niche act individually and in combination to effect precise zonation. We will determine how individual BMPs, Wnt/Rspo signaling, and NRG1 together influence epithelial zonation and restrict cell replication to the isthmus. Significant gaps in knowledge currently limit actionable insights into gastric mucosal homeostasis, metaplasias, cancer, and other disorders. This project aims to narrow that gap via mechanistic investigation of a dynamic epithelium that continuously renews itself while generating specialized cellular phenotypes in response to an exquisitely graded mesenchymal niche.