Inter-individual variability in the production of flavan-3-ol colonic metabolites: preliminary elucidation of urinary metabotypes

Abstract

Purpose. There is much information on the bioavailability of (poly)phenolic compounds following acute intake of various foods. However, there are only limited data on the effects of repeated and combined exposure to specific (poly)phenol food sources and the inter-individual variability in their bioavailability. This study evaluated the combined urinary excretion of (poly)phenols from green tea and coffee following daily consumption by healthy subjects in free-living conditions. The inter-individual variability in the production of phenolic metabolites was also investigated. Methods. Eleven participants consumed both tablets of green tea and green coffee bean extracts daily for 8 weeks and 24-h urine was collected on five different occasions. The urinary profile of phenolic metabolites and a set of multivariate statistical tests were used to investigate the putative existence of characteristic metabotypes in the production of flavan-3-ol microbial metabolites. Results. (Poly)phenolic compounds in the green tea and green coffee bean extracts were absorbed and excreted after simultaneous consumption, with green tea resulting in more inter-individual variability in urinary excretion of phenolic metabolites. Three metabotypes in the production of flavan-3-ol microbial metabolites were tentatively defined, characterized by the excretion of different amounts of trihydroxyphenyl-γ-valerolactones, dihydroxyphenyl-γ-valerolactones, and hydroxyphenylpropionic acids. Conclusions. The selective production of microbiota-derived metabolites from flavan-3-ols and the putative existence of characteristic metabotypes in their production represent an important development in the study of the bioavailability of plant bioactives. These observations will contribute to better understand the health effects and individual differences associated with consumption of flavan-3-ols, arguably the main class of flavonoids in the human diet.

This work was partially funded by MRC core funding (Physiological Modelling of Metabolic Risk, MC_UP_A090_1005, and Nutrition, Surveys and Studies, MC_U105960384) and University of Parma core funding (FIL 2014-2017). P.M. was partially funded by a Grant of the Postdoctoral Fellowship Program from Fundación Séneca (Murcia Region, Spain). I.A.L. was supported by a postdoctoral fellowship funded by the Spanish Ministry of Economy and Competitiveness (IJCI-2014-20689).