Abstract
BackgroundHyperuricemia is associated with multiple diseases, including gout, cardiovascular disease, and renal disease. Serum urate is highly heritable, yet association studies of single nucleotide polymorphisms (SNPs) and serum uric acid explain a small fraction of the heritability. Whether copy number polymorphisms (CNPs) contribute to uric acid levels is unknown.ResultsWe assessed copy number on a genome-wide scale among 8,411 individuals of European ancestry (EA) who participated in the Atherosclerosis Risk in Communities (ARIC) study. CNPs upstream of the urate transporter SLC2A9 on chromosome 4p16.1 are associated with uric acid (, p=3.19×10-23). Effect sizes, expressed as the percentage change in uric acid per deleted copy, are most pronounced among women (3.974.935.87 [ 2.55097.5 denoting percentiles], p=4.57×10-23) and independent of previously reported SNPs in SLC2A9 as assessed by SNP and CNP regression models and the phasing SNP and CNP haplotypes (). Our finding is replicated in the Framingham Heart Study (FHS), where the effect size estimated from 4,089 women is comparable to ARIC in direction and magnitude (1.414.707.88, p=5.46×10-03).ConclusionsThis is the first study to characterize CNPs in ARIC and the first genome-wide analysis of CNPs and uric acid. Our findings suggests a novel, non-coding regulatory mechanism for SLC2A9-mediated modulation of serum uric acid, and detail a bioinformatic approach for assessing the contribution of CNPs to heritable traits in large population-based studies where technical sources of variation are substantial.
Highlights
Hyperuricemia is associated with multiple diseases, including gout, cardiovascular disease, and renal disease
These genomic intervals capture 317 non-contiguous loci constituting the Copy number polymorphism (CNP) ascertained by the hidden Markov model (HMM) among European ancestry (EA) Atherosclerosis Risk in Communities (ARIC) participants, and most span known regions of copy number variation reported in the Database of Genomic Variants [25]
To assess whether the CNP associations are independent of SLC2A9 Single nucleotide polymorphism (SNP) among EA participants, we evaluated a series of models for uric acid concentrations that include SNPs and/or the gender-specific CNP slopes
Summary
Hyperuricemia is associated with multiple diseases, including gout, cardiovascular disease, and renal disease. Serum urate is highly heritable, yet association studies of single nucleotide polymorphisms (SNPs) and serum uric acid explain a small fraction of the heritability. Serum uric acid levels are highly heritable and associated with several diseases, including gout, hypertension, and cardiovascular disease [1,2,3,4]. Among the most common artifacts are genomic waves [20,21], an autocorrelation of the marker-level estimates when plotted against physical position, and batch effects, differences between groups of samples arising from technical sources of variation such as sample preparation, reagents, and laboratory personnel [22,23,24]. Approaches to reduce wave and batch artifacts include models for adjusting log R ratios by the GC composition of the local sequence as in [21] and surrogates of batch such as chemistry plate in association models when confounding between batch and phenotype is incomplete
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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
