The Cellular Response to Microenvironmental Stress in Barrett’s Oesophagus

The incidence of oesophageal adenocarcinoma (OAC) and its metaplastic precursor, Barrett’s oesophagus (BO), has increased markedly over recent decades. This is attributed to an increase in the prevalence of environmental risk factors such as gastro-oesophageal reflux disease, which exposes the lower oesophagus to frequent pulses of an acidic, bile-salt containing refluxate. The mechanisms linking stressors that are common to the lower oesophageal micro-environment to the dysplastic progression of BO are, however, unclear. Here, we integrate gene expression, chromatin accessibility and kinase activity profiles to characterise the response of normal squamous epithelial (NSE), non-dysplastic (NDBO) and high-grade dysplastic (HGD) BO cell lines to microenvironmental stress; simulated by a ten minute pulse of bile salts at acidic or neutral pH, acid alone, and a 48-hour period of serum-starvation. We demonstrate that acidic bile salts and serum-starvation both recapitulate gene-regulatory and phenotypic changes that are associated with BO dysplastic progression. In NDBO cells, acidic bile salts result in a gene signature enriched for pro-migratory and pro-inflammatory Th1 cytokine-mediated processes, whereas serum-starvation gives rise to a pro-autophagic, quiescent, transcriptional response. A number of transcription factors associated with BO dysplastic progression, such as KLF5 and NF-ĸB, were linked to these transcriptional changes through analyses of differences in chromatin accessibility. Upstream, kinase activity profiling and a functional genomic screen implicated a number of receptor tyrosine kinases (RTKs) in the BO stress response; most notably including activation of EGFR in response to cytosolic pH normalisation following acid exposure, in a process regulated by acidic sphingomyelinase. These data, and existing evidence for an association of RTK co-overexpression with BO dysplastic progression, together highlight a potential role for a subset of RTKs in mediating the BO dysplastic phenotype in response to microenvironmental stress. This requires further validation but may provide novel biomarkers and targets for risk stratification and chemopreventative strategies in early BO.