Introduction Single nucleotide polymorphisms (SNPs) on chromosome 9p21 are associated with coronary artery disease, stroke, diabetes and several cancers. The mechanisms mediating these associations are unknown, but risk variants are mainly non-coding and may act through regulation of gene expression in cis. We investigated whether 9p21 SNPs were associated with total and allelic expression of the three nearest genes; the cell-cycle inhibitors CDKN2A/CDKN2B, and a non-coding RNA of unknown function, ANRIL.
Methods We genotyped 56 SNPs, selected to tag the common variation in the region and include SNPs with reported functional effects. We studied peripheral blood expression in two populations of healthy volunteers: 177 British Caucasians and 310 mixed-ancestry South Africans. Total expression was measured using TaqMan real-time PCR. Allelic expression was quantified by mass spectrometry (Sequenom) and analysed by novel methodology using data from two transcribed SNPs per gene. Association between mapping SNPs and expression was assessed using likelihood ratio tests.
Results Total expression of CDKN2A, CDKN2B and ANRIL was correlated (p<0.05), suggesting that they are co-regulated. Allelic expression was also correlated (p<0.05), suggesting that transcription is influenced by shared cis acting elements. SNP associations mapped by total and allelic expression were similar (r=0.96, p=2×10−92), but the power to detect effects was greater for allelic expression, indicating the presence of substantial trans-acting effects on expression. The proportion of expression variance attributable to cis-acting effects was 8% for CDKN2A, 5% for CDKN2B, and 20% for ANRIL. SNP associations were similar in the two populations (r=0.94, p=1×10−72) which permitted analysis of the combined dataset (Abstract C figure 1). Multiple SNPs were independently associated with expression of each gene (p<0.01), suggesting that several sites may modulate disease susceptibility. Risk SNPs for coronary disease and stroke were all highly associated with allelic expression of ANRIL (all p<1×10−7), whilst association with the other two genes was only detectable for some. Risk alleles all correlated with ANRIL under-expression (up to 1.9-fold). Individual SNPs were associated with opposite effects on ANRIL and CDKN2B expression, suggesting a possible role of ANRIL in CDKN2B regulation.
Conclusions Multiple independent cis-acting SNPs influence CDKN2A, CDKN2B and ANRIL expression. Risk SNPs correlate with reduced ANRIL expression, suggesting that ANRIL down-regulation mediates susceptibility to atherosclerosis; this pathway may be an important target for future therapies. Allelic expression has greater power than total expression for mapping cis-acting effects; this has implications for future studies investigating SNP effects on gene expression at other loci.
- Coronary artery disease