Abstract
Prediabetes is one of the main health concerns in public health, and various etiological factors contribute to its onset. This study aimed to evaluate genetic associations and gene-macronutrient interaction with prediabetes-related metabolites to understand how genetic variation and dietary intake contribute to dysglycemia. We analyzed a total of 482 self-identified Mexican American participants recruited from Starr County, Texas in 2018-2019. Untargeted metabolomic profiling was performed using LC-MS. Nutrient densities of six macronutrients were derived from a 106-item food frequency questionnaire. Genetic associations for each metabolite were tested using Generalized linear Mixed Model Association Tests (GMMAT). Gene-macronutrient interactions on prediabetes-associated metabolites were assessed with the Mixed Model Association Test for GEne-Environment Interaction (MAGEE). Age, gender, and BMI were included as covariates in all association tests. Among 308 named and 2,471 unnamed metabolites, 17 novel variant-metabolite pairs were discovered, including rs10947898 in LRFN2 associated with diacylglycerol DG32:1(p-value: 8.95E-09). Among 145 named and 687 unnamed metabolites after filtering, gene-macronutrient interaction analyses identified seven named metabolites, including a variant(rs111251222) in MXD3 that interacted with monounsaturated fat to influence eicosadienoic acid levels (Interaction p-value: 9.88E-09). Prediabetes and nutrient-related metabolites in Mexican Americans showed significant genetic associations and gene-nutrient interactions.