Integrated multi-omics signatures to decipher the bioavailability of dietary (poly)phenols and identify phenolic metabotypes to promote cardiometabolic health

Background: Modest long-term intake of (poly)phenols has been associated with reduced risk of cardiovascular diseases and type 2 diabetes, although findings remain inconsistent. Inter-individual variability in (poly)phenol absorption, distribution, metabolism, and excretion (ADME), influenced by genetics, microbiome, sex, age, lifestyle and health status, may contribute to these differences. Aims: Metabolic phenotypes (metabotypes) may explain distinct patterns of phenolic catabolite production. Methods: The Oral (Poly)phenol Challenge Test (OPCT) study examined (poly)phenol bioavailability by identifying metabotypes and related determinants. A standardised acute intervention, consisting of 3 tablets containing dietary-relevant amounts of 15 (poly)phenol classes, was administered to 300 healthy adults (18-74 y). Participants provided dietary and lifestyle information, underwent anthropometric assessments, and provided clinical data and biological samples (blood, urine, faeces). Urine samples collected 24h after the OPCT were analysed by UPLC-ESI-QqQ-MS/MS to quantify phenolic metabolite excretion, enabling metabotype clustering. Blood was analysed for cardiometabolic markers and genome-wide genotyping. Faecal samples underwent microbial profiling via ITS sequencing. Cardiometabolic risk scores were also calculated. Statistical analyses included several clustering models and other multivariate approaches. Results: Up to 298 volunteers completed the study. A targeted approach identified over 250 (poly)phenol metabolites, enabling clustering into two metabotypes: low and high-producers of phenolic metabolites. Clusters were primarily driven by gut microbiota composition, with colonic-derived phenolic metabolites being the most discriminant. Main differences between groups emerged in age, sex, anthropometrics, and dietary habits, while the probability of being a high-producer was associated with age and sex. Genome-wide association analysis revealed strong associations between metabotypes and genetic variants, likely affecting (poly)phenol metabolism and bioavailability. The two metabotypes also displayed distinct microbiome profiles at the species level. Associations with cardiometabolic outcomes were further observed. Conclusions: Individuals metabolise differently due to multiple factors, which may influence their cardiometabolic responses. Metabotyping by phenolic profile in precision nutrition could help tailoring dietary recommendations to prevent cardiometabolic diseases.