Wnt Signalling in Colorectal Cancer: Putting Gene Expression Data of Biopsies into the Biological Disease Context

Supervisors: Professor Stefan Hoppler, Dr Claus-Dieter Mayer & Dr Leslie Samuel (NHS)

Introduction:

Abnormally activated Wnt/beta-catenin signalling drives the initial stages of colorectal cancer. Wnt/beta-catenin signalling normally functions to regulate transcription of specific genes via DNA-binding TCF/LEF transcription factors. Laboratory experiments show that different TCF/LEF factor proteins can have distinct molecular activities and may regulate expression of different genes downstream of Wnt signalling, or of the same genes differently.

General Hypothesis:

Differences in TCF/LEF expression are therefore also likely to influence the expression of different specific genes downstream of abnormally activated Wnt signalling in colorectal cancer. These differently expressed genes can be expected to regulate particular cellular processes, which will affect the progression of tumours.

Research Strategy:

This project will carefully scrutinise gene expression data from human colorectal cancer tissue in order to develop specific hypotheses about the functional role of different TCF/LEF factors in colorectal cancer; which may provide insight for prognostic diagnosis and discover novel drug targets for future therapeutic strategies.

Project:

Gene expression data of human colorectal cancer tissue (with corresponding normal gut tissue) are already available, both microarray and next-generation sequencing data. We have already analysed some of this data with statistical methods and have discovered clear association between aberrant TCF/LEF factor expression on the one hand and altered transcriptomes on the other.

These raw results of our statistical analysis are now ready to be scrutinised from a scientific perspective, to be put into the context both of the known cancer biology and established Wnt signalling pathway mechanisms. Our existing statistical meta analysis of gene expression data of human colorectal cancer tissue already provides lists of genes whose expression is statistically correlated with expression of specific TCF/LEF genes. The statistical analysis also already contains information about differences between tumour and corresponding control tissue. Specific hypotheses developed from our existing analysis can then be further tested, statistically and subsequently in in vitro experiments.

A particularly exciting aspect of this studentship project will be the opportunity to access the new and comprehensive data from the "Grampian Cohort" of the MRC-funded ’Stratification in Colorectal Cancer’ (S:CORT) consortium. In addition to gene expression information, this remarkable resource includes genomic analysis, including DNA methylation, before and after treatment, as well as patient outcome information. The "Grampian Cohort" data currently encompasses this detailed information from 265 patients.

Access to this data will of course grant us the opportunity to refine our specific hypotheses, but additionally it will allow us to put Wnt pathway function and aberrant TCF/LEF factor expression in the context of currently established clinical treatments of Colorectal Cancer, and correlate this with patient outcome. We aim to contribute to a stratified medicine approach, for instance, aberrant TCF/LEF factor expression may be diagnostic, indicating likelihood of metastasis or linked to higher success with preoperative chemotherapy, etc.. Our analysis may eventually contribute to inform directly the treatment most likely to succeed for a specific patient with a specific type of tumour in order to maximise the patient’s survival and quality of life.

In practice, this project will involve handling large datasets, for instance by exploring different parameters in the existing statistical analysis in order to obtain biologically relevant and more manageable short lists of correlated genes. The molecular identity and known biological functions of correlated genes will be explored by scientific literature search and by consulting web resources, such as NCBI OMIM.

Outcome:

This project will allow us to formulate specific hypotheses for a detailed investigation of TCF/LEF isoform expression in normal and tumour tissue (using PCR) and correlation with an altered transcriptome (RNA-seq analysis).

Key Skills:

Dealing with large datasets, understanding of statistical analysis, literature search (eg with SCOPUS), online database searches, and presentation skills, such as accessible writing of results and review of literature searches.