| Summary: | <p>Oesophageal adenocarcinoma originates in the gastroesophageal junction and has a
poor prognosis. Barrett’s oesophagus is a common, innocuous condition occurring
prior to most adenocarcinoma cases. This lesion is frequent in patients suffering
from gastroduodenal reflux and involves columnar cells replacing a segment of the
squamous lining of the distal oesophagus. High variability in the length of the
segment exists between individuals and typically patients with longer lesions have
an increased risk of developing adenocarcinoma. However, the pathways promoting
this replacement and determining its final length are unknown. In this thesis, I
investigate models to study this phenomenon in vitro.</p>
<p>In the first part of this thesis, I describe a system that I developed to study
interactions between epithelial cell populations and apply it to combinations of cells
which are crucial for Barrett’s oesophagus progression. My findings indicate that
squamous and early Barrett’s oesophagus derived cells form stable junctions and
that reflux-like epidermal growth factor concentrations can disrupt this stability.</p>
<p>In part two, I adapt existing 3D culture methods to study heterotypic junctions
in vitro. Combining oil bath and plastic printing techniques, I have constructed
rings with organoid epithelial lining. I demonstrate as well that it is possible to
control the ring’s size via the addition of fibroblasts into the collagen. Through
addition of a divider, I was able to recapitulate heterotypic junctions between
organoid populations and, by making the moulds higher, I show that it is possible
to generate tube structures using the same technique. Additionally, I demonstrate
that using fibroblast gradients in the tubes can control their shape. Building upon
the contractibility of the fibroblasts, I also show that adaptors can be fixed at
the ends of the tubes to control flow through it. Using this system, I am able
to coat the tube’s lumen with epithelia and infect the cells with pathogens to
recapitulate a more natural infection.</p>
<p>Lastly, I generated an endoscopic mucosal resection biobank for microdissection
with the samples being annotated with the assistance of pathologists. I then
validated this system by micro-dissecting three candidate patients.</p>
|
|---|