Aggregate metabolic phenotypes related to the bioavailability of dietary (poly)phenols and their association with cardiometabolic health

Increasing evidence suggests that modest long-term intakes of (poly)phenols can reduce the risk of chronic diseases, especially cardiovascular diseases and type 2 diabetes (T2D). Nevertheless, the role of (poly)phenols in cardiometabolic protection has not been consistently demonstrated yet. Inter-individual variability plays an important role in the physiological response, mainly influenced by differences in the absorption, distribution, metabolism, and excretion (ADME) of (poly)phenols, along with other factors, including genetic background, gut microbiota, sex, age, ethnicity, lifestyle (diet, smoking, and physical activity), (patho)physiological status and medication. After ingestion, (poly)phenols reach the colon, where they undergo modifications by the gut microbiota, being converted to smaller catabolites, principally as conjugated phase II metabolites, which can act as mediators of diet-induced effects on health. The inter-individual differences in gut microbial composition and functionality can lead to quantitative and qualitative differences in the production of specific metabolites, influencing the bioactivity of (poly)phenols in the host. The different catabolite production patterns may be related to metabolic phenotypes (the so called metabotypes). The application of metabotyping in precision nutrition could help tailoring dietary advice to prevent health problems, especially concerning cardiometabolic status. The Oral (Poly)phenol Challenge Test (OPCT) study is directed to understand the association between aggregate metabolic phenotypes for the main dietary (poly)phenols and the factors determining their formation, as well as with the cardiometabolic health. An acute intervention study was carried out on 300 healthy adults (18-74 y) who met specific inclusion and exclusion criteria: a BMI ranging between 18.5 and 35.0 kg/m2, free from cardiometabolic diseases and impairments mostly related to the gastrointestinal tract, renal and liver functionality. During Visit 1, after signing the informed consent, they were asked to provide dietary and lifestyle information and to undergo anthropometric measurements. Clinical data and biological samples (blood, urine, and faeces) were delivered at Visit 2, when subjects underwent a standardised oral (poly)phenol challenge test consisting in an acute supplementation of up to 15 classes of dietary (poly)phenols in the form of 3 tablets. Urine samples collected during the following 24h were analysed through UHPLC-ESI-IMS-qToF-HRMS and UPLC-ESI-QqQ-MS/MS to assess the individual urinary excretion of phenolic metabolites, allowing clustering according to aggregate metabotypes. Blood samples were analysed to determine common cardiometabolic health biomarkers (total cholesterol, HDL-cholesterol, triglycerides, glucose, insulin) and for whole-genome genotyping focused on genetic polymorphisms (SNPs). Faeces were subjected to microbial profiling to determine gut microbiota composition at species level by using ITS profiling method. Cardiometabolic risk scores as well as clinical measures were also assessed. Several statistical methods were performed, considering both univariate and multivariate analyses. Different clustering models were carried out and partial least squares-discriminant analysis (PLS-DA) was selected to model data from clustering analysis. Logistic and multiple regression analyses were performed to examine the relationship between clusters and cardiometabolic biomarkers. Up to 298 volunteers finished the study. Our analyses showed that the cohort was made up of 57% of women, having an average age of 40.7 y (SD ± 16.3); regarding anthropometric measures, 73% of the sample had a normal weight, 22% was overweight and 5% obese. The mean values of the clinical data concerning cardiometabolic health ranged within the reference values. A targeted approach was performed on 297 subjects for the identification of more than 250 (poly)phenol metabolites and to allow population clustering according to different metabotypes. Several clustering methods identified two main metabotypes, the so called low and high-producers of phenolic metabolites, respectively. Clusters were mainly defined by differences related likely to gut microbiota composition; indeed, among all the phenolic metabolites identified, the most discriminating ones were those of colonic origin. Main statistically significant differences between the two groups are mostly based on age, sex, anthropometric measures, and previous dietary habits. In addition, results on genetic polymorphisms (SNPs), following genome-wide association study on >6 million variants, evidenced strong associations between metabotypes and several genetic variants. Variations in these alleles likely affected the metabolism of (poly)phenols as they could not be transformed in more hydrophilic molecules, resulting less bioavailable. To provide a deeper knowledge on the association between metabotypes and cardiometabolic health, different regression models were performed. Logistic regression analysis showed that the probability to be in the cluster 2 of high-producers is mainly linked to age and sex. At the same time, multiple regression model proved that subjects in the cluster 1 were more likely to report a slightly higher BMI than those in the cluster 2. Individuals metabolise dietary (poly)phenols in different ways due to several factors, showing, at the same time, the interlink among different families of (poly)phenols. Thus, metabotyping according to the metabolism of the whole set of dietary (poly)phenols may represent a promising attempt for cardiometabolic health promotion through personalised nutrition initiatives.