Format

Send to

Choose Destination
See comment in PubMed Commons below
Am J Clin Nutr. 2014 Jul;100(1):208-17. doi: 10.3945/ajcn.113.078758. Epub 2014 Apr 16.

Metabolomics in nutritional epidemiology: identifying metabolites associated with diet and quantifying their potential to uncover diet-disease relations in populations.

Author information

1
From the Nutritional Epidemiology Branch (KAG, SCM, QX, RZS-S, RS, and AJC), the Biostatistics Branch (JNS), and the Occupational and Environmental Epidemiology Branch (W-YH), Division of Cancer Epidemiology and Genetics, National Cancer Institute, NIH, Department of Health and Human Services, Bethesda, MD.

Abstract

BACKGROUND:

Metabolomics is an emerging field with the potential to advance nutritional epidemiology; however, it has not yet been applied to large cohort studies.

OBJECTIVES:

Our first aim was to identify metabolites that are biomarkers of usual dietary intake. Second, among serum metabolites correlated with diet, we evaluated metabolite reproducibility and required sample sizes to determine the potential for metabolomics in epidemiologic studies.

DESIGN:

Baseline serum from 502 participants in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial was analyzed by using ultra-high-performance liquid-phase chromatography with tandem mass spectrometry and gas chromatography-mass spectrometry. Usual intakes of 36 dietary groups were estimated by using a food-frequency questionnaire. Dietary biomarkers were identified by using partial Pearson's correlations with Bonferroni correction for multiple comparisons. Intraclass correlation coefficients (ICCs) between samples collected 1 y apart in a subset of 30 individuals were calculated to evaluate intraindividual metabolite variability.

RESULTS:

We detected 412 known metabolites. Citrus, green vegetables, red meat, shellfish, fish, peanuts, rice, butter, coffee, beer, liquor, total alcohol, and multivitamins were each correlated with at least one metabolite (P < 1.093 × 10(-6); r = -0.312 to 0.398); in total, 39 dietary biomarkers were identified. Some correlations (citrus intake with stachydrine) replicated previous studies; others, such as peanuts and tryptophan betaine, were novel findings. Other strong associations included coffee (with trigonelline-N-methylnicotinate and quinate) and alcohol (with ethyl glucuronide). Intraindividual variability in metabolite levels (1-y ICCs) ranged from 0.27 to 0.89. Large, but attainable, sample sizes are required to detect associations between metabolites and disease in epidemiologic studies, further emphasizing the usefulness of metabolomics in nutritional epidemiology.

CONCLUSIONS:

We identified dietary biomarkers by using metabolomics in an epidemiologic data set. Given the strength of the associations observed, we expect that some of these metabolites will be validated in future studies and later used as biomarkers in large cohorts to study diet-disease associations. The PLCO trial was registered at clinicaltrials.gov as NCT00002540.

PMID:
24740205
PMCID:
PMC4144099
DOI:
10.3945/ajcn.113.078758
[Indexed for MEDLINE]
Free PMC Article
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for HighWire Icon for PubMed Central
    Loading ...
    Support Center