Abstract
Colorectal cancer (CRC) is one of the leading causes of death by cancer in New Zealand, largely due to metastasis of the primary tumour to secondary sites around the body. The initiation, progression and eventual metastasis has been shown to be largely influenced by epimutations. DNA methylation is a stable, heritable epigenetic mark which has been heavily implicated in disease. In cancer contexts, global hypomethylation and focal hypermethylation act to grossly dysregulate the genome, while simultaneously acting as potential biomarkers for initial detection and therapeutic response.
CRC is a molecularly heterogeneous disease, including between patients, tumours and even within the tumours. As well as this, tumours are comprised of immune cells, healthy tissue cells and blood cells in tandem with the neoplastic cells. Traditionally, these populations are sequenced as a whole. Consequently, the methylomes of each cell are coalesced, giving rise to averaged methylation profiles. The emergence of single cell technologies allows us deconvolute heterogeneous populations of cells and identify different cell states and types.
Obtaining high quality data from a single cell is difficult due to the minimal amount of starting DNA, particularly in single-cell bisulfite sequencing as the bisulfite conversion is harsh on the DNA. Hence, the main aim of this project was to optimise and implement single-cell bisulfite sequencing on a sorted human colorectal cancer cell line. Following this, my second aim was to compare the methylation profiles of the single-cell methylation libraries to uncover heterogeneity in the population. In this project, I use a post-bisulfite adaptor tagging (PBAT) method to perform single cell bisulfite sequencing on the CRC human cell line HT29. Following this, I use a publicly available data set to investigate intra- and inter- tumour heterogeneity.
With a few optimization steps the PBAT method was able to successfully amplify very small numbers of cells, including 100, 10, 5 and two single cell samples. Following this, publicly available data showed even in a small population of single cells, there was evident heterogeneity regarding global, chromosomal and focal promoter methylation.
These results highlighted the heterogeneity which can be unmasked using single cell technologies, even on a small scale. While also confirming renowned biological models such as the global hypomethylation undertaken by cancer cells.