Decoding cell lineages in human oesophagus, stomach and duodenal development for the Human Cell Atlas and to understand disease

The purpose of this project is to discover which genes are switched on or switched off in the developing upper intestine and stomach in individual cells and to compare this information to adult tissue. In parallel we will discover which bits of the surrounding genome are active in controlling how the genes are switched on or off. We need to do this in individual cells because these tissues are very complex and made up of many different types of cell. The approach is important because the first step from normal gullet to disease is a change to something called 'Barrett's oesophagus'. In turn Barrett's oesophagus is presumed to be a step on the way to oesophageal cancer, which all too frequently is very aggressive and incurable. Previously, we have thought that Barrett's oesophagus is a switch from one adult type of cell to another. Recently, we have shown that the switch looks more like a reversion to a primitive embryonic state. The same is true of oesophageal cancer. This makes detailed understanding of normal development very important.

We will generate datasets at different points in development so that we can understand the changes that occur over time at genome-scale using computational methods. We will ensure all the information is compatible with the requirements of the Human Cell Atlas so that we can share all our information freely and, in turn, benefit from data from other teams of researchers.

We will decode the transcriptomes of single cells from the human lower oesophagus, stomach and upper duodenum. We will do this at sequential embryonic and fetal stages compared with datasets from adult tissue. The work needs to be undertaken in single cells to resolve the heterogeneity within tissue structures. We will complement the transcriptomic data with regulatory information from single cell ATACseq, which delineates open chromatin. While the individual datasets will be an important contribution to the Human Cell Atlas (HCA), the more powerful HCA output comes from their computational assembly to define the different subpopulations, infer regulatory networks controlling transcription and to probe how these change over time as development proceeds. We will reconstruct key aspects of the data back into the 3D context using our long-standing collection of human developmental material. This work defining developmental trajectories and relationships between different cell populations needs the interdisciplinary team which we already have in place.

Importantly, our data will have immediate clinical application based on our recent discovery about Barrett's oesophagus (BO). BO has been considered a metaplastic switch of normal stratified epithelium to an adult intestinal columnar epithelium. While present in the lower oesophagus of millions of asymptomatic, otherwise healthy individuals, in a small minority, BO has an association with oesophageal adenocarcinoma (OAC). In contrast to the prevailing dogma, we have just shown that BO looks like reversion to a primitive developmental gut-like state. Thus, it is critical that we define this area of the human gastrointestinal tract, in its developing and adult state, in much greater detail. We could be on the cusp of an urgently needed mechanism that is ultimately responsible for OAC--a cancer, which, for the most part, is incurable with dreadful survival statistics.

The Human Cell Atlas is anticipated to be an impactful initiative and we look forward to playing a full role. Our work has major impact, evidenced by the journals in which we publish, their impact factors and our citation rates. Our work is inter-disciplinary which, by its nature, tends to produce more impactful outcomes. Focused research on post-implantation human development tends to produce impact outcomes in part because of its distinctiveness. Advances that ultimately relate to oesophageal cancer would be very impactful.

Directly related to this project:

1. We will provide unique data on the developing human intestine and stomach; datasets that would be unattainable to most researchers globally. Following dissemination of our work these data can be reapplied by other scientists in their projects. Next-to-nothing is known about the molecular phenotype of these native human embryonic cells currently. It is highly likely that we will discover unappreciated aspects of cell phenotype that will be informative for improving pluripotent stem cell differentiation and/or organoid/organotypic culture.

2. We will provide the first mechanistic detail on how the early human oesophagus, stomach and upper duodenum develop.

3. Our datasets will be freely available for download via our open access website.