AUTHOR=Bidulescu Aurelian , Choudhry Shweta , Musani Solomon K., Buxbaum Sarah G., Liu Jiankang , Rotimi Charles N., Wilson James G., Taylor Herman A., Gibbons Gary H.

TITLE=Associations of adiponectin with individual European ancestry in African Americans: the Jackson Heart Study

JOURNAL=Frontiers in Genetics

VOLUME=5

YEAR=2014

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2014.00022

DOI=10.3389/fgene.2014.00022

ISSN=1664-8021

ABSTRACT=<p><bold>Background:</bold> Compared with European Americans, African Americans (AAs) exhibit lower levels of the cardio-metabolically protective adiponectin even after accounting for adiposity measures. Because few studies have examined in AA the association between adiponectin and genetic admixture, a dense panel of ancestry informative markers (AIMs) was used to estimate the individual proportions of European ancestry (PEA) for the AAs enrolled in a large community-based cohort, the Jackson Heart Study (JHS). We tested the hypothesis that plasma adiponectin and PEA are directly associated and assessed the interaction with a series of cardio-metabolic risk factors.</p><p><bold>Methods:</bold> Plasma specimens from 1439 JHS participants were analyzed by ELISA for adiponectin levels. Using pseudo-ancestral population genotype data from the HapMap Consortium, PEA was estimated with a panel of up to 1447 genome-wide preselected AIMs by a maximum likelihood approach. Interaction assessment, stepwise linear and cubic multivariable-adjusted regression models were used to analyze the cross-sectional association between adiponectin and PEA.</p><p><bold>Results:</bold> Among the study participants (62% women; mean age 48 ± 12 years), the median (interquartile range) of PEA was 15.8 (9.3)%. Body mass index (BMI) (<italic>p</italic> = 0.04) and insulin resistance (<italic>p</italic> = 0.0001) modified the association between adiponectin and PEA. Adiponectin was directly and linearly associated with PEA (β = 0.62 ± 0.28, <italic>p</italic> = 0.03) among non-obese (<italic>n</italic> = 673) and insulin sensitive participants (<italic>n</italic> = 1141; β = 0.74 ± 0.23, <italic>p</italic> = 0.001), but not among those obese or with insulin resistance. No threshold point effect was detected for non-obese participants.</p><p><bold>Conclusions:</bold> In a large AA population, the individual proportion of European ancestry was linearly and directly associated with plasma adiponectin among non-obese and non insulin-resistant participants, pointing to the interaction of genetic and metabolic factors influencing adiponectin levels.</p>