|
|
Chretien JP,
Coresh J, Fink NE, Klag MJ, Johns Hopkins University; Marcovina
SM, University of Washington; Smith MW, National Cancer Institute
Higher Lp(a) concentrations, which are associated with cardiovascular
disease and kidney failure, are more frequent in Africans and African-Americans
(AAs) than in European-Americans (EAs); this ethnic difference is
suspected to be genetic but its etiology is unknown. This study
aims to demonstrate the utility of Mapping by Admixture Linkage
Disequilibrium (MALD) in mapping the genetic basis for high Lp(a)
levels in AAs compared to EAs. The study population consists of
200 AA and 450 EA kidney failure patients in the CHOICE (Choices
for Healthy Outcomes in Caring for ESRD) Study. Lp(a) concentration
was measured at baseline with an apo(a) size-independent ELISA,
and 14 short tandem repeats (STRs) were typed within 21 cM of the
apo(a) gene. The genotype agreement rate among 38 quality-control
pair duplicates was >99%. Median Lp(a) concentrations in the AAs
and EAs were 88.5 mg/dL and 36.3 mg/dL, respectively. Linear regression
of log([Lp(a)+1) on the average MALD score (difference in allele
frequency between AAs and EAs) should result in a positive beta
if ALD exists. We found significant positive betas (1-tailed p<0.01)
for a pentanucleotide repeat polymorphism in the apo(a) promoter
(p=0.0002) and for D6S437 at 4.3 cM from apo(a) (p=0.004) in AAs
but for none of the markers in EAs. After Bonferroni correction,
individual alleles were significantly associated with log([Lp(a)]+1)
in both AAs and EAs for markers at 0 cM from apo(a), but only in
AAs for markers at distances >0 cM. This new analytic strategy for
MALD suggests that genetic variation at or near the apo(a) locus
contributes to higher Lp(a) concentrations in AAs compared to EAs.
View
Slides
|
