학술논문
Genetic and clinical determinants of abdominal aortic diameter: genome-wide association studies, exome array data and Mendelian randomization study
Document Type
article
Author
Portilla-Fernandez, Eliana; Klarin, Derek; Hwang, Shih-Jen; Biggs, Mary L; Bis, Joshua C; Weiss, Stefan; Rospleszcz, Susanne; Natarajan, Pradeep; Hoffmann, Udo; Rogers, Ian S; Truong, Quynh A; Völker, Uwe; Dörr, Marcus; Bülow, Robin; Criqui, Michael H; Allison, Matthew; Ganesh, Santhi K; Yao, Jie; Waldenberger, Melanie; Bamberg, Fabian; Rice, Kenneth M; Essers, Jeroen; Kapteijn, Daniek MC; van der Laan, Sander W; de Knegt, Rob J; Ghanbari, Mohsen; Felix, Janine F; Ikram, M Arfan; Kavousi, Maryam; Uitterlinden, Andre G; Roks, Anton JM; Danser, AH Jan; Tsao, Philip S; Damrauer, Scott M; Guo, Xiuqing; Rotter, Jerome I; Psaty, Bruce M; Kathiresan, Sekar; Völzke, Henry; Peters, Annette; Johnson, Craig; Strauch, Konstantin; Meitinger, Thomas; O’Donnell, Christopher J; Dehghan, Abbas; Program, VA Million Veteran
Source
Human Molecular Genetics. 31(20)
Subject
Language
Abstract
Progressive dilation of the infrarenal aortic diameter is a consequence of the ageing process and is considered the main determinant of abdominal aortic aneurysm (AAA). We aimed to investigate the genetic and clinical determinants of abdominal aortic diameter (AAD). We conducted a meta-analysis of genome-wide association studies in 10 cohorts (n = 13 542) imputed to the 1000 Genome Project reference panel including 12 815 subjects in the discovery phase and 727 subjects [Partners Biobank cohort 1 (PBIO)] as replication. Maximum anterior-posterior diameter of the infrarenal aorta was used as AAD. We also included exome array data (n = 14 480) from seven epidemiologic studies. Single-variant and gene-based associations were done using SeqMeta package. A Mendelian randomization analysis was applied to investigate the causal effect of a number of clinical risk factors on AAD. In genome-wide association study (GWAS) on AAD, rs74448815 in the intronic region of LDLRAD4 reached genome-wide significance (beta = -0.02, SE = 0.004, P-value = 2.10 × 10-8). The association replicated in the PBIO1 cohort (P-value = 8.19 × 10-4). In exome-array single-variant analysis (P-value threshold = 9 × 10-7), the lowest P-value was found for rs239259 located in SLC22A20 (beta = 0.007, P-value = 1.2 × 10-5). In the gene-based analysis (P-value threshold = 1.85 × 10-6), PCSK5 showed an association with AAD (P-value = 8.03 × 10-7). Furthermore, in Mendelian randomization analyses, we found evidence for genetic association of pulse pressure (beta = -0.003, P-value = 0.02), triglycerides (beta = -0.16, P-value = 0.008) and height (beta = 0.03, P-value