A metabolomic study of biomarkers of meat and fish intake

William Cheung, Pekka Keski-rahkonen, Nada Assi, Pietro Ferrari, Heinz Freisling, Sabina Rinaldi, Nadia Slimani, Raul Zamora-ros, Milena Rundle, Gary Frost, Helena Gibbons, Eibhlin Carr, Lorraine Brennan, Amanda J Cross, Valeria Pala, Salvatore Panico, Carlotta Sacerdote, Domenico Palli, Rosario Tumino, Tilman KühnRudolf Kaaks, Heiner Boeing, Anna Floegel, Francesca Mancini, Marie-christine Boutron-ruault, Laura Baglietto, Antonia Trichopoulou, Androniki Naska, Philippos Orfanos, Augustin Scalbert

Research output: Contribution to journalArticlepeer-review

149 Citations (Scopus)


Background: Meat and fish intakes have been associated with various chronic diseases. The use of specific biomarkers may help to assess meat and fish intake and improve subject classification according to the amount and type of meat or fish consumed.

Objective: A metabolomic approach was applied to search for biomarkers of meat and fish intake in a dietary intervention study and in free-living subjects from the European Prospective Investigation into Cancer and Nutrition (EPIC) study.

Design: In the dietary intervention study, 4 groups of 10 subjects consumed increasing quantities of chicken, red meat, processed meat, and fish over 3 successive weeks. Twenty-four-hour urine samples were collected during each period and analyzed by high-resolution liquid chromatography–mass spectrometry. Signals characteristic of meat or fish intake were replicated in 50 EPIC subjects for whom a 24-h urine sample and 24-h dietary recall were available and who were selected for their exclusive intake or no intake of any of the 4 same foods.

Results: A total of 249 mass spectrometric features showed a positive dose-dependent response to meat or fish intake in the intervention study. Eighteen of these features best predicted intake of the 4 food groups in the EPIC urine samples on the basis of partial receiver operator curve analyses with permutation testing (areas under the curve ranging between 0.61 and 1.0). Of these signals, 8 metabolites were identified. Anserine was found to be specific for chicken intake, whereas trimethylamine-N-oxide showed good specificity for fish. Carnosine and 3 acylcarnitines (acetylcarnitine, propionylcarnitine, and 2-methylbutyrylcarnitine) appeared to be more generic indicators of meat and meat and fish intake, respectively.

Conclusion: The meat and fish biomarkers identified in this work may be used to study associations between meat and fish intake and disease risk in epidemiologic studies. This trial was registered at clinicaltrials.gov as NCT01684917.
Original languageEnglish
Pages (from-to)600-608
JournalThe American Journal of Clinical Nutrition
Issue number3
Early online date25 Jan 2017
Publication statusPublished - 1 Mar 2017
Externally publishedYes


Dive into the research topics of 'A metabolomic study of biomarkers of meat and fish intake'. Together they form a unique fingerprint.

Cite this