| Literature DB >> 25187575 |
Ursula M Schick1, Paul L Auer2, Joshua C Bis3, Honghuang Lin4, Peng Wei5, Nathan Pankratz6, Leslie A Lange7, Jennifer Brody3, Nathan O Stitziel8, Daniel S Kim9, Christopher S Carlson1, Myriam Fornage10, Jeffery Haessler1, Li Hsu11, Rebecca D Jackson12, Charles Kooperberg1, Suzanne M Leal13, Bruce M Psaty14, Eric Boerwinkle15, Russell Tracy16, Diego Ardissino17, Svati Shah18, Cristen Willer19, Ruth Loos20, Olle Melander21, Ruth Mcpherson22, Kees Hovingh23, Muredach Reilly24, Hugh Watkins25, Domenico Girelli26, Pierre Fontanillas27, Daniel I Chasman28, Stacey B Gabriel27, Richard Gibbs29, Deborah A Nickerson9, Sekar Kathiresan30, Ulrike Peters1, Josée Dupuis31, James G Wilson32, Stephen S Rich33, Alanna C Morrison5, Emelia J Benjamin34, Myron D Gross6, Alex P Reiner.
Abstract
C-reactive protein (CRP) concentration is a heritable systemic marker of inflammation that is associated with cardiovascular disease risk. Genome-wide association studies have identified CRP-associated common variants associated in ∼25 genes. Our aims were to apply exome sequencing to (1) assess whether the candidate loci contain rare coding variants associated with CRP levels and (2) perform an exome-wide search for rare variants in novel genes associated with CRP levels. We exome-sequenced 6050 European-Americans (EAs) and 3109 African-Americans (AAs) from the NHLBI-ESP and the CHARGE consortia, and performed association tests of sequence data with measured CRP levels. In single-variant tests across candidate loci, a novel rare (minor allele frequency = 0.16%) CRP-coding variant (rs77832441-A; p.Thr59Met) was associated with 53% lower mean CRP levels (P = 2.9 × 10(-6)). We replicated the association of rs77832441 in an exome array analysis of 11 414 EAs (P = 3.0 × 10(-15)). Despite a strong effect on CRP levels, rs77832441 was not associated with inflammation-related phenotypes including coronary heart disease. We also found evidence for an AA-specific association of APOE-ε2 rs7214 with higher CRP levels. At the exome-wide significance level (P < 5.0 × 10(-8)), we confirmed associations for reported common variants of HNF1A, CRP, IL6R and TOMM40-APOE. In gene-based tests, a burden of rare/lower frequency variation in CRP in EAs (P ≤ 6.8 × 10(-4)) and in retinoic acid receptor-related orphan receptor α (RORA) in AAs (P = 1.7 × 10(-3)) were associated with CRP levels at the candidate gene level (P < 2.0 × 10(-3)). This inquiry did not elucidate novel genes, but instead demonstrated that variants distributed across the allele frequency spectrum within candidate genes contribute to CRP levels.Entities:
Mesh:
Substances:
Year: 2014 PMID: 25187575 PMCID: PMC4334838 DOI: 10.1093/hmg/ddu450
Source DB: PubMed Journal: Hum Mol Genet ISSN: 0964-6906 Impact factor: 6.150