Abstract
Genome-wide association studies of colorectal cancer (CRC) have identified a number of common variants associated with modest risk, including rs3802842 at chromosome 11q23.1. Several genes map to this region but rs3802842 does not map to any known transcribed or regulatory sequences. We reasoned, therefore, that rs3802842 is not the functional single-nucleotide polymorphism (SNP), but is in linkage disequilibrium (LD) with a functional SNP(s). We performed ChIP-seq for histone modifications in SW480 and HCT-116 CRC cells, and incorporated ChIP-seq and DNase I hypersensitivity data available through ENCODE within a 137-kb genomic region containing rs3802842 on 11q23.1. We identified SNP rs10891246 in LD with rs3802842 that mapped within a bidirectional promoter region of genes C11orf92 and C11orf93. Following mutagenesis to the risk allele, the promoter demonstrated lower levels of reporter gene expression. A second SNP rs7130173 was identified in LD with rs3802842 that mapped to a candidate enhancer region, which showed strong unidirectional activity in both HCT-116 and SW480 CRC cells. The risk allele of rs7130173 demonstrated reduced enhancer activity compared with the common allele, and reduced nuclear protein binding affinity in electromobility shift assays compared with the common allele suggesting differential transcription factor (TF) binding. SNPs rs10891246 and rs7130173 are on the same haplotype, and expression quantitative trait loci (eQTL) analyses of neighboring genes implicate C11orf53, C11orf92 and C11orf93 as candidate target genes. These data imply that rs10891246 and rs7130173 are functional SNPs mapping to 11q23.1 and that C11orf53, C11orf92 and C11orf93 represent novel candidate target genes involved in CRC etiology.
Highlights
Genome-wide association studies (GWAS) of colorectal cancer (CRC) have led to the identification of risk variants mapping to 20 chromosomal regions including 1q41, 3q26.2, 6p21, 8q23.3, 8q24.21, 10p14, 11q13.4, 11q23.1, 12q13.12, 14q22.2, 15q13.3, 16q22.1, 18q21.1, 19q13.11, 20p12.3, 20q13.33 and Xp22.2 in Caucasian populations (P-value ≤ 5 × 1028) (1 – 12)
Few CRC-associated tagSNPs are expected to be functional, and there is growing evidence that risk associated single-nucleotide polymorphism (SNP) are likely to be in linkage disequilibrium (LD) with functional SNPs that map within nearby genes, regulatory regions such as enhancers and promoters or long non-coding RNAs (16,18)
To investigate the functional basis of the chromosome 11q23.1 association with CRC etiology, we began by identifying potential functional variants using the LD between tagSNP rs3802842 and the other known SNPs within 1 Mb (Fig. 1)
Summary
Genome-wide association studies (GWAS) of colorectal cancer (CRC) have led to the identification of risk variants mapping to 20 chromosomal regions including 1q41, 3q26.2, 6p21, 8q23.3, 8q24.21, 10p14, 11q13.4, 11q23.1, 12q13.12, 14q22.2, 15q13.3, 16q22.1, 18q21.1, 19q13.11, 20p12.3, 20q13.33 and Xp22.2 in Caucasian populations (P-value ≤ 5 × 1028) (1 – 12). Few CRC-associated tagSNPs are expected to be functional, and there is growing evidence that risk associated SNPs are likely to be in LD with functional SNPs that map within nearby genes, regulatory regions such as enhancers and promoters or long non-coding RNAs (lncRNAs) (16,18). This implies that the identification of functional SNPs could be facilitated through incorporation of chromatin status such as histone modification and/or DNase I hypersensitivity sites into post-GWAS analyses, and that this may represent a powerful approach to identify tissue-specific regulatory elements (19– 21)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
