Back to search

The University of Edinburgh

Understanding colorectal cancer risk loci that alter transcriptional dynamics

Abstract

dc:description.abstract

Germline variation at 11q23.1 is associated with colorectal cancer (CRC) risk and expression quantitative trait locus (eQTL) effects on three cis- (POU2AF2, COLCA1, POU2AF3) and 23 trans-genes. Unfortunately, because of a lack of knowledge of the function of 11q23.1 eQTL targets, high linkage disequilibrium at this region and correlated expression at the bulk level, uncovering mechanistic insights into 11q23.1 variation-associated CRC risk has proven difficult. In this work, colocalisation analysis in GTEx transverse colon (n=367) shows rs3087967 to be the lead variant of cis-eQTL effects at 11q23.1 that are associated with CRC risk, with conditional analysis confirming this to be representative of a single genetic signal. While the relevance of distant transcriptional dynamics is often overlooked, use of GTEx and SOCCS colorectal RNAseq (total n=580) showed rs3087967 variation to exhibit CRC risk-associated trans-eQTL effects on 21 targets by summarised mendelian randomisation. Mapping of expression across epithelial cell-types of the human colon showed a striking specificity of trans-eQTL target expression to tuft cells across independent single cell RNAseq (scRNAseq) datasets (n=32,361 and n=11,651), with agnostic gene network analysis showing these genes to likely comprise a POU2AF2-correlated, tuft cell-specific transcriptional network. POU2AF2 was found to bind several trans-eQTL targets at sites enriched for the motif of known transcriptionally regulator of tuft cells and novel trans-eQTL target, POU2F3, by ChIPseq, with the majority of these targets showing exclusive genomic accessibility in colonic tuft cells by scATACseq (n=21,620). Importantly, quantification of marker abundance by immunofluorescence confirmed CRC risk genotype at 11q23.1 to be associated with reduced abundance of colonic tuft cells across samples from 15 individuals. Using orthogonal genetically engineered mouse models, the expression of trans-eQTL target homologs and colorectal tuft cell abundance was experimentally decoupled to Pou2af2 expression specifically. Concordantly, increased mortality and exacerbated tumour burden of ApcMin/+ mice was observed in association with Pou2af2, but not Pou2af3 genotype, further reinforcing both POU2AF2 expression and reduced tuft cell abundance to be associated with CRC risk. Preserved cis-eQTL effects were also observed on POU2AF2, but not COLCA1 or POU2AF3 in human CRC tumour RNAseq (n=380), and while this was not significantly associated with changes in immune cell infiltration, consensus molecular subtype or microsatellite stability status, tuft-like cells demarcated by POU2AF2 and trans-eQTL target expression were identified in tumour epithelial scRNAseq (n=43,474), highlighting their relevance in this context also. While preliminary, using a pairwise module preservation analysis of gene modules identified across tumour epithelial cells, the lower abundance of healthy tuft cells was found to be associated with potential changes in stem-related signalling of a transcriptionally distinct, epithelial cell-type, which may be regulated by HSPA1A expression that is also correlated with trans-eQTL target expression in tumour tuft-like cells. In summary, this evidence shows the abundance of tuft cells to be correlated with and likely relevant to CRC-risk governed by genetic variation at 11q23.1. Additionally, this work also provides a potential mechanism for how POU2AF2 is specifically associated with these effects.

Degree

thesis:*
Grantor dc:publisher
The University of Edinburgh
Year dc:date.issued
2024

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Harris, Bradley
Advisors dc:contributor.advisor
  • Farrington, Susan
  • Dunlop, Malcolm

Subjects

dc:subject × 4

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
University of Edinburgh
Base URL
era.ed.ac.uk/server/oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Harris, Bradley. Understanding colorectal cancer risk loci that alter transcriptional dynamics. The University of Edinburgh, 2024. https://hdl.handle.net/1842/41584