Urinary metabolomics of phenolic compounds reveals biomarkers of type-2 diabetes within the PREDIMED trial

Domínguez-López, Inés; Lozano-Castellón, Julián; Vallverdú-Queralt, Anna; Jáuregui, Olga; Martínez-González, Miguel Ángel, 1957-; Hu, Frank B.; Fitó Colomer, Montserrat; Ros, Emilio; Estruch, Ramón; Lamuela-Raventós, Rosa M.

doi:http://dx.doi.org/10.1016/j.biopha.2023.114703

Urinary metabolomics of phenolic compounds reveals biomarkers of type-2 diabetes within the PREDIMED trial

License

Citation

Domínguez-López I, Lozano-Castellón J, Vallverdú-Queralt A, Jáuregui O, Martínez-González MÁ, Hu FB, Fitó M, Ros E, Estruch R, Lamuela-Raventós RM. Urinary metabolomics of phenolic compounds reveals biomarkers of type-2 diabetes within the PREDIMED trial. Biomed Pharmacother. 2023 Jun;162:114703. DOI: 10.1016/j.biopha.2023.114703

Abstract

Background: Phenolic compounds have been associated with protective effects against type-2 diabetes (T2D). We used a metabolomics approach to determine urinary phenolic metabolites associated with T2D and fasting plasma glucose. Methods: This case-control study within the PREDIMED trial included 200 participants at high cardiovascular risk, 102 of whom were diagnosed with T2D. A panel of urinary phenolic compounds were analysed using a novel method based on liquid chromatography coupled to mass spectrometry. Multivariate statistics and adjusted logistic regressions were applied to determine the most discriminant compounds and their association with T2D. The relationship between the discriminant phenolic compounds and plasma glucose was assessed using multivariable linear regressions. Results: A total of 41 phenolic compounds were modeled in the orthogonal projection to latent structures discriminant analysis, and after applying adjusted logistic regressions two were selected as discriminant: dihydrocaffeic acid (OR = 0.22 (CI 95 %: 0.09; 0.52) per 1-SD, p-value = 0.021) and genistein diglucuronide (OR = 0.72 (CI 95%: 0.59; 0.88) per 1-SD, p-value = 0.021). Both metabolites were associated with a lower risk of suffering from T2D, but only dihydrocaffeic acid was inversely associated with plasma glucose (β = -17.12 (95 % CI: -29.92; -4.32) mg/dL per 1-SD, p-value = 0.009). Conclusions: A novel method using a metabolomics approach was developed to analyse a panel of urinary phenolic compounds for potential associations with T2D, and two metabolites, dihydrocaffeic acid and genistein diglucuronide, were found to be associated with a lower risk of this condition.