JPI HDHL: The Food Biomarkers Alliance

Food frequency questionnaires, food diaries, and 24 h recall represent the most commonly used dietary assessment tools in human studies on nutrition and health. Because food intake biomarkers represent a more objective reflection of intake compared to selfreported data, food intake biomarkers may provide valuable complementary data. However, very few of these biomarkers are sufficiently validated. In a three year research program, the Food Biomarkers Alliance (FoodBAll) have developed clear strategies for food intake biomarker discovery and validation, and to identify and validate biomarkers for a range of foods consumed across Europe. The methods included extensive literature reviews, acute intervention studies, as well as analyses of existing intervention and observational studies. Importantly, metabolomics techniques were used for the discovery of biomarkers as main -omics technique. In addition, new biomarker sampling techniques were investigated. Moreover, the FoodBAll consortium developed a new platform for sharing knowledge and resources with the scientific community (i.e. open access databases on food metabolites, public web portal on useful resources for annotation of food metabolome profiles and food intake biomarker discovery, and a food-derived chemical library), and have provided input on the value of biomarkers for the evaluation of health claims of foods. Ultimately, the FoodBAll consortium contributed to: 1) validation of existing dietary assessment tools; 2) provide markers of compliance for dietary intervention studies; and 3) improve the reliability of observational studies on the role of diet in human health. The project is organized in 6 work-pages (WP) the following way: WP1. Defining strategies for the discovery of novel biomarkers of food intake WP2. Validation of alternative nutrient biomarker sampling techniques WP3. Food intake biomarker classification and validation WP4. Sharing knowledge and resources on food intake biomarkers WP5. Bridging dietary biomarkers to health pathways WP6. Policy, aiming to provide input for national authorities on the value of food intake biomarkers for evaluation of nutrition- or health-claims of foods. WP2 is coordinated by the Norwegian representative professor emeritus Christian A. Drevon and wrote an extensive, comprehensive and representative review of validated and potentially new food intake biomarkers in different matrices. To date, venous blood sampling is the most commonly used way to collect biological samples from human volunteers. This is a time-consuming process that requires well-trained personnel. Recent technological developments allow performing food intake biomarker analyses with very high sensitivity requiring very small blood volumes or tissue samples. Combining these novel and sensitive techniques with non-professional biological sampling (e.g. blood collection by finger-pricking or filters/sticks) offers the opportunity to collect biological data without the employment of health professionals. In preparation of the experimental work within this WP, an extensive literature review has been performed summarizing the current evidence with respect to food intake biomarkers from different body fluids, e.g. blood, white blood cells, plasma, serum, urine, stool, saliva, sweat, hair, nails, skin, adipose tissue, and skeletal muscle. WP2 has a particular focus on the use of alternative sampling techniques such as dried blood spots (DBS). To a large extent researchers within this WP2 have conducted validation studies on the use of DBS and Mitra sticks for analyses of lipid biomarkers (lipidomics) as well as different other nutrients including sodium (Na), magnesium (Mg), potassium (K), calcium (Ca), iron (Fe), cupper (Cu), zinc (Zn), selenium (Se), and iodine (I).

Many nutrients can be measured accurately after proper non-professional collection of small amounts of blood, drying, and mailing the sample to a professional laboratory. Most small molecules, and many proteins, can be measured after proper standardization and preparation. Especially fatty acids, some vitamins and metals can be measured in large populations. Many hormonal proteins and glycated hemoglobin (HbA1c) may also be measured. We measured successfully several metals, amino acids, fat-soluble vitamins and carotenoids on alternative matrices. Our lipidomics data show that the HPLC-ELSD can be applied for analyses of very small biological samples. By a sophisticated chromatographic method, UHPLC-TOF-MS we performed untargeted and targeted lipidomics analyses of murine livers, and observed clear differences in hepatic ceramides between sexes. We can measure hundreds of lipid species on DBS with good stability but we have some more work to do on stability of some complex lipids.

FOODBALL will carry out systematic exploration and validation of biomarkers to get good coverage of the food intake in different population groups within Europe. FOODBALL will cover many relevant foods in Europe, and apply metabolomics to discover biomarkers. The use or combination of other media than blood (matrices) and easier sampling techniques will be explored. The current dietary biomarker classification will be revised with classes defined primarily by marker kinetics. A validation scoring system will be developed to enhance and harmonise validation methods, and applied to the new metabolites. The biological effects of intake will be explored combining metabolomics and other -omics technologies, and case studies will exhibit the added value of exploring biological effects using intake markers. Smart data integration as developed in ENPADASI will be used. The initial untargeted and subsequent targeted analyses will be facilitated by infrastructures like biobanks of well characterised populations and studies. The FOODBALL consortium has wide access to samples and data from large cohorts, multi-centre dietary interventions, and short-term interventions with specific foods or food groups. It has also access to state-of-the-art analytical platforms allowing measurement of thousands of biomarkers. The ENPADASI data infrastructure will be used for data sharing, analysis and as a tool for biomarker discovery. There will be a close link with databases covering food consumption and the expertise on dietary assessment from DEDIPAC, as well as food composition from EuroFIR. Finally, FOODBALL will provide a unique platform for sharing knowledge and resources within and beyond the project with the scientific community, in particular: 1) appropriate databases on food metabolites; 2) software tools for annotation of food metabolites; and 3) chemical libraries. Consolidation of these recourses and knowledge may be accomplished via ENPADASI.