Estimation and implications of the genetic architecture of fasting and non-fasting blood glucose

• Published in: Nature communications 2023-01, Vol.14 (1), p.451-11, Article 451
• Main Authors: Qiao, Zhen, Sidorenko, Julia, Revez, Joana A., Xue, Angli, Lu, Xueling, Pärna, Katri, Snieder, Harold, Visscher, Peter M., Wray, Naomi R., Yengo, Loic
• Format: Article
• Language: English
• Subjects: 45/43; 631/208/205/2138; 692/53/2423; 692/699/317; Blood; Blood Glucose - analysis; Diabetes mellitus (non-insulin dependent); Diabetes Mellitus, Type 2 - genetics; Fasting; Genetic control; Genetic diversity; Genetic variance; Genome-wide association studies; Genome-Wide Association Study; Genomes; Glucose; Heritability; Homeostasis; Humanities and Social Sciences; Humans; multidisciplinary; Plateaus; Polymorphism, Single Nucleotide; Science; Science (multidisciplinary); Single-nucleotide polymorphism; Standard error
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-023-36013-1

The genetic regulation of post-prandial glucose levels is poorly understood. Here, we characterise the genetic architecture of blood glucose variably measured within 0 and 24 h of fasting in 368,000 European ancestry participants of the UK Biobank. We found a near-linear increase in the heritability of non-fasting glucose levels over time, which plateaus to its fasting state value after 5 h post meal ( h 2  = 11%; standard error: 1%). The genetic correlation between different fasting times is > 0.77, suggesting that the genetic control of glucose is largely constant across fasting durations. Accounting for heritability differences between fasting times leads to a ~16% improvement in the discovery of genetic variants associated with glucose. Newly detected variants improve the prediction of fasting glucose and type 2 diabetes in independent samples. Finally, we meta-analysed summary statistics from genome-wide association studies of random and fasting glucose ( N  = 518,615) and identified 156 independent SNPs explaining 3% of fasting glucose variance. Altogether, our study demonstrates the utility of random glucose measures to improve the discovery of genetic variants associated with glucose homeostasis, even in fasting conditions. Most genetic studies of glucose levels have been done on fasting samples, which can be difficult to obtain. Here, the authors identify 156 genetic loci controlling the physiological variation of glucose levels in healthy non-fasting individuals, demonstrating that the results non-fasting samples can be used to predict fasting glucose levels.