Genome-wide association meta-analysis and Mendelian randomization analysis confirm the influence of ALDH2 on sleep durationin the Japanese population.

Takeshi Nishiyama,Masahiro Nakatochi,Atsushi Goto,Motoki Iwasaki,Tsuyoshi Hachiya,Yoichi Sutoh,Atsushi Shimizu,Chaochen Wang,Hideo Tanaka,Miki Watanabe,Akihiro Hosono,Yuya Tamai,Tamaki Yamada,Taiki Yamaji,Norie Sawada,Kentaro Fukumoto,Kotaro Otsuka,Kozo Tanno,Hiroaki Tomita,Kaname Kojima,Masao Nagasaki,Atsushi Hozawa,Asahi Hishida,Tae Sasakabe,Yuichiro Nishida,Megumi Hara,Hidemi Ito,Isao Oze,Yohko Nakamura,Haruo Mikami,Rie Ibusuki,Toshiro Takezaki,Teruhide Koyama,Nagato Kuriyama,Kaori Endoh,Kiyonori Kuriki,Tanvir C Turin,Takashima Naoyuki,Sakurako Katsuura-Kamano,Hirokazu Uemura,Rieko Okada,Sayo Kawai,Mariko Naito,Yukihide Momozawa,Michiaki Kubo,Makoto Sasaki,Masayuki Yamamoto,Shoichiro Tsugane,Kenji Wakai,Sadao Suzuki
2019-01-01
Abstract:Usual sleep duration has substantial heritability and is associated with various physical and psychiatric conditions as well as mortality. However, for its genetic locus, only PAX8 and VRK2 have been replicated in previous genome-wide association studies (GWAS). We conducted a GWAS meta-analysis of self-reported usual sleep duration using three population-based cohorts totaling 31 230 Japanese individuals. A genome-wide significant locus was identified at 12q24 (p-value < 5.0 × 10-8). Subsequently, a functional variant in the ALDH2 locus, rs671, was replicated in an independent sample of 5140 Japanese individuals (p-value = 0.004). The association signal, however, disappeared after adjusting for alcohol consumption, indicating the possibility that the rs671 genotype modifies sleep duration via alcohol consumption. This hypothesis explained a modest genetic correlation observed between sleep duration and alcohol consumption (rG = 0.23). A Mendelian randomization analysis using rs671 and other variants as instrumental variables confirmed this by showing a causal effect of alcohol consumption, but not of coffee consumption on sleep duration. Another genome-wide significant locus was identified at 5q33 after adjusting for drinking frequency. However, this locus was not replicated, nor was the PAX8 and VRK2. Our study has confirmed that a functional ALDH2 variant, rs671, most strongly influences on usual sleep duration possibly via alcohol consumption in the Japanese population, and presumably in East Asian populations. This highlights the importance of considering the involvement of alcohol consumption in future GWAS of usual sleep duration, even in non-East Asian populations, where rs671 is monomorphic.
What problem does this paper attempt to address?