Introduction: Several studies have tested metabolic risk factors following dietary intervention with Nordic diets. The SYSDIET study tested a healthy Nordic diet according to the Nordic Nutrition Recommendation (NNR) in five different countries while the SHOPUS study in Denmark tested the “New Nordic Diet” designed to meet NNR while also being sustainable and palatable. Objectives: To investigate whether metabolic profiles 1) reflect Nordic diets, 2) are improved by data fusion and 3) reflect dietary compliance. Methods: Plasma and urine samples from both studies were profiled by one or several metabolomics platforms (LC-MS, GC-MS, NMR) and the data analysed by PLS-DA. Results: Metabolic profiles of both urine and plasma reflected the Nordic or control diets with varying degree of performance, depending on the analytical platform used. The best ROC-curves for the SHOPUS study had AUC’s above 0.8. Data fusion across platforms or sample types did not improve these models. The metabolic profiles from SYSDIET also discriminated between the countries and centres where the samples had been collected. There was only about 10% overlap between the volunteers identified as potentially non-compliant based on their most discriminating urine or plasma profiles. This may reflect that relatively many volunteers are only occasionally non-compliant while only few are more consistently non-compliant. For this latter group the marker patterns included markers of several foods that were clearly not part of the diet they were supposed to follow, supporting the interpretation that these subjects were in fact non-compliant and not just having individual characteristics of metabolism placing them outside the main pattern. Conclusion: Dietary patterns with Nordic foods are reflected with good accuracy by metabolomics at the group level, but patterns of several samples from each volunteer may be needed to identify the more consistently non-compliant participants. Data fusion did not improve the models in this study.

Metabolomic response to Nordic foods / Dragsted, L; Acar, E; Gurdeniz, G; Andersen, Mb; Poulsen, S; Astrup, A; Bro, R; Engelsen, Sb; Savorani, F; Brader, L; Hermansen, K; Schwab, U; Kolehmainen, M; Paananen, J; Poutanen, Ks; Cloetens, L; Akesson, B; Siloaho, M; Savolainen, Mj; Gunnarsdottir, I; Thorsdottir, I; Ulven, Sm; Rosqvist, F; Riserus, U; Uusitupa, M; Larsen, Tm. - In: ANNALS OF NUTRITION AND METABOLISM. - ISSN 0250-6807. - STAMPA. - 67:(2015), pp. 55-55. (Intervento presentato al convegno 12th European Nutrition Conference (FENS) tenutosi a Berlin, Germany nel October 20-23, 2015).

Metabolomic response to Nordic foods

Savorani, F;
2015

Abstract

Introduction: Several studies have tested metabolic risk factors following dietary intervention with Nordic diets. The SYSDIET study tested a healthy Nordic diet according to the Nordic Nutrition Recommendation (NNR) in five different countries while the SHOPUS study in Denmark tested the “New Nordic Diet” designed to meet NNR while also being sustainable and palatable. Objectives: To investigate whether metabolic profiles 1) reflect Nordic diets, 2) are improved by data fusion and 3) reflect dietary compliance. Methods: Plasma and urine samples from both studies were profiled by one or several metabolomics platforms (LC-MS, GC-MS, NMR) and the data analysed by PLS-DA. Results: Metabolic profiles of both urine and plasma reflected the Nordic or control diets with varying degree of performance, depending on the analytical platform used. The best ROC-curves for the SHOPUS study had AUC’s above 0.8. Data fusion across platforms or sample types did not improve these models. The metabolic profiles from SYSDIET also discriminated between the countries and centres where the samples had been collected. There was only about 10% overlap between the volunteers identified as potentially non-compliant based on their most discriminating urine or plasma profiles. This may reflect that relatively many volunteers are only occasionally non-compliant while only few are more consistently non-compliant. For this latter group the marker patterns included markers of several foods that were clearly not part of the diet they were supposed to follow, supporting the interpretation that these subjects were in fact non-compliant and not just having individual characteristics of metabolism placing them outside the main pattern. Conclusion: Dietary patterns with Nordic foods are reflected with good accuracy by metabolomics at the group level, but patterns of several samples from each volunteer may be needed to identify the more consistently non-compliant participants. Data fusion did not improve the models in this study.
File in questo prodotto:
File Dimensione Formato  
pdfresizer.com-pdf-split.pdf

accesso aperto

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Creative commons
Dimensione 168.64 kB
Formato Adobe PDF
168.64 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2815392