The Contribution of Environmental Cues to Early Squamous Tumour Formation in the Upper Gastrointestinal Tract

Recent studies have revealed that normal epithelium in various tissue types harbour substantial populations of mutant clones similar to those found in cancer. This implies that the accumulation of somatic mutations alone may not be the sole cause of cancer and that other factors are required. Tumours are polyclonal in nature with the interplay between different mutant clones influencing tumour fate. Additionally, tumours are a complex ecosystem consisting not only of malignant cells but various stromal cells, which make up the tumour microenvironment and is integral to several stages of tumourigenesis. Thus, the risk and progression of cancer involves a complex interplay between mutant cells and their dynamic environment which is far more intricate than previously understood. However, our understanding of how the pre-cancer microenvironment contributes to early tumourigenesis remains limited. In this project, I made use of a unique early tumour mouse model based on the exposure to di-ethyl-nitrosamine (DEN), a carcinogen found in cigarette smoke. This model induces the formation of pre-neoplastic squamous tumours in the upper gastrointestinal tract that can be detected as early as 10 cells in size and sporadically progress towards invasive carcinoma. Importantly, the model recapitulates the mutational landscape of the normal human ageing oesophagus, making it an ideal system to explore early tumourigenesis. Using the DEN model in transgenic animals with a traceable DN-Maml1, a Notch deactivating mutation, I investigated whether stress from the physical collision between DN-Maml1 mutant clones and DEN-induced mutant clones influenced the formation of oesophageal tumours. Deep tissue imaging analysis revealed a reduction in the number of tumours and KRT17 clusters in the presence of DN-Maml1 clones compared to controls, confirming the protective phenotype of DN-Maml1. The stress markers, KRT17 and SOX9, were upregulated at the outside edges of competing DN-Maml1 and DEN-induced mutant clones. Interestingly, this was associated with a higher number of tumours and KRT17 clusters at the outside edge of DN-Maml1 clones compared to at the inside edge. These findings suggested that while DN-Maml1 clones reduce the number of tumours, the effects of stress at the boundaries of competing mutant clones may contribute to the formation of tumours. Next, I explored the contribution of stromal signals in promoting early tumour formation. Through secretomic, proteomic and scRNA-seq analysis, I identified key secretable ligands involved in early tumour epithelial-stromal crosstalk. Several ligands were involved in processes such as cell proliferation, ECM remodelling, and wound healing. Notably, in vitro experiments revealed that FINC and AREG promoted epithelial cell proliferation and migration, respectively, highlighting the potential role of tumour-associated ligands in altering epithelial cell behaviour and contributing to early tumourigenesis. In the final stage of my research, I generated a new scRNA-seq dataset to examine the early transcriptional landscape of tumours 10 days post-DEN treatment. The analysis revealed a highly pro-inflammatory environment characterised by interferon-stimulated genes and antigen-presenting genes expressed in both epithelial and stromal cells in response to the carcinogenic insult. Additionally, genes related to wound healing, ECM interactions and angiogenesis were enriched in epithelial tumours, highlighting early epithelial-stromal interactions and microenvironment modulation. Comparison with an 8-month scRNA-seq dataset, which captured more established tumours, suggested a shift from an early pro-inflammatory response to an immune-suppressive phenotype, potentially reflecting tumour adaptation mechanisms to evade host immune responses over time. Taken together, my findings reveal the involvement of environmental cues in the fate and survival of early oesophageal tumours, such as localised tissue stress, stromal-derived ligands, and the immune response. This study underscores the importance of the pre-cancer microenvironment as both a potential target for therapeutic intervention and a source of biomarkers for early detection, offering new avenues to intercept tumour progression at its inception.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2025

Author dc:creator

• Hill, Emily

Advisor dc:contributor.advisor

• Alcolea, Maria

• Diethylnitrosamine (DEN) mutagenesis mouse model
• Early tumourigenesis
• Environmental cues
• Epithelial-stromal crosstalk
• Mutant clones
• Oesophageal cancer
• Pre-neoplastic tumours
• Proteomics analysis
• scRNA-seq analysis
• Transgenic animals
• Confocal imaging
• In vitro studies

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/79aeaf71-b8bd-4eef-bfe6-114826bc1ba8/download
• http://purl.org/NET/rdflicense/allrightsreserved

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.120284

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/387539