Abstract: The intestinal epithelium is maintained by intestinal stem cells that replace each other stochastically over time. Stem cells that are excluded from the stem cell pool will differentiate into the absorptive or secretory cell lineages. However, intestinal cell-fate specification is not always restrictive and early secretory progenitors can revert back to the stem cell pool. Human colon cancer develops by acquisition of oncogenic mutations in the colonic epithelium throughout life. As secretory progenitors have unexpectedly been shown to be a significant source of colonic stem cells, they could serve a cell of origin for tumorigenesis. Furthermore, it has been speculated that tumours could arise in fields that encode pro-oncogenic mutations prior to overt tumour development. Mutations in KRAS can be found in 10% of healthy human colons in fields of up to 100 crypts that may be the cause of a subset of colorectal tumours. The extent to which such fields are predisposed to colorectal cancer development is unknown.
The aims of this PhD project were to characterise the cell of origin and stem cell behaviour of KrasG12D fields and to develop new tools to recapitulate sequential mutations in vivo. Cre mediated lineage tracing of Atoh1 expressing early secretory progenitors carrying KrasG12D mutations demonstrated that KrasG12D expression does not change cell fate choices in homeostasis but appears to increase Atoh1+ stem cell contribution in the small intestine after Lgr5 depletion. In addition, Atoh1 derived KrasG12D stem cells have a competitive bias and Atoh1 derived KrasG12D crypts can multiply over time to create fields and polyps in the colonic epithelium. In addition, a mouse model that utilises Flp and Cre to temporally separate intestinal KrasG12D recombination from lineage-tracing was developed to study stem cell behaviour in KrasG12D fields. Lineage tracing in this model shows that crypts in KrasG12D epithelium have a markedly higher monoclonal conversion rate and accumulate an increased mutation load over time compared to WT crypts. Furthermore, the faster monoclonal conversion rate is shown to be dependent on Mek signalling downstream of Kras. Thus, KrasG12D fields may fix secondary mutations at an accelerated rate and so represent pro-oncogenic areas. Lastly, in this thesis a Rosa26 DrePr mouse model, that allows for recombination independently of other recombinases, is developed and used to initiate lineage tracing in a sequential model of Cre activated intestinal tumorigenesis.
Collectively, the data presented in this thesis allows in-dept investigation of the cell of origin and stem cell behaviour in pro-oncogenic KrasG12D fields and contributes to the understanding of how such fields might lead to colon cancer.