Association of KCNJ1 variation with change in fasting glucose and new onset diabetes during HCTZ treatment

• Authors: J.H. Karnes, University of Florida; C.W. McDonough, University of Florida; Y. Gong, University of Florida; T.T. Vo, University of Florida; T.Y. Langaee, University of Florida; Arlene B Chapman, Emory University; J.G. Gums, University of Florida; A.L. Beitelshees, University of Maryland; K.R. Bailey, Mayo Clinic College of Medicine; J.L. Del-Aguila, University of Texas; E.A. Boerwinkle, University of Texas; C.J. Pepine, University of Florida; S.T. Turner, Emory University; J.A. Johnson, University of Florida; R.M. Cooper-DeHoff, University of Florida
• Journal Title: Pharmacogenomics Journal
• Volume: Volume 13, Number 5
• Publisher: Nature Publishing Group: Open Access Hybrid Model Option B | 2013-10-01, Pages 430-436
• Type of Work: Article | Post-print: After Peer Review
• Publisher DOI: http://doi.org/10.1038/tpj.2012.34
• Subjects: Health Sciences, Nutrition; Biology, Genetics; Health Sciences, Pharmacology
• Keywords: Science & Technology; Life Sciences & Biomedicine; Genetics & Heredity; Pharmacology & Pharmacy; pharmacogenetics; KCNJ1; ROMK1; hypertension; diabetes mellitus; hydrochlorothiazide; CORONARY-ARTERY-DISEASE; VERAPAMIL SR-TRANDOLAPRIL; CARDIOVASCULAR EVENTS; HYPERTENSIVE SUBJECTS; THIAZIDE DIURETICS; BLOOD-PRESSURE; POTASSIUM; MELLITUS; INVEST; TRIAL

Abstract

Thiazide-induced potassium loss may contribute to new onset diabetes (NOD). KCNJ1 encodes a potassium channel and one study observed that a KCNJ1 single-nucleotide polymorphism (SNP) was associated with changes in fasting glucose (FG) during hydrochlorothiazide (HCTZ) treatment. We used linear regression to test association of KCNJ1 SNPs and haplotypes with FG changes during HCTZ treatment in the Pharmacogenomic Evaluation of Antihypertensive Responses (PEAR) study. We used logistic regression to test association of KCNJ1 variation with NOD in HCTZ-treated patients from the International Verapamil SR Trandolapril Study (INVEST). Multivariate regression analyses were performed by race/ethnicity with false discovery rate (FDR) correction. In PEAR blacks, a KCNJ1 SNP was associated with increased FG during HCTZ treatment (beta=8.47, P FDR =0.009). KCNJ1 SNPs and haplotypes were associated with NOD risk in all INVEST race/ethnic groups (strongest association: odds ratio 2.14 (1.31-3.53), P FDR =0.03). Our findings support that KCNJ1 variation is associated with HCTZ-induced dysglycemia and NOD.