Lipidomic and Fatty Acid Biomarkers in Whole Blood Can Predict the Dietary Intake of Eicosapentaenoic and Docosahexaenoic Acids in a Danish Population

Juan J Aristizabal-Henao; Anja P Biltoft-Jensen; Tue Christensen; Ken D Stark

doi:10.1016/j.tjnut.2024.04.038

Lipidomic and Fatty Acid Biomarkers in Whole Blood Can Predict the Dietary Intake of Eicosapentaenoic and Docosahexaenoic Acids in a Danish Population

J Nutr. 2024 May 4:S0022-3166(24)00269-4. doi: 10.1016/j.tjnut.2024.04.038. Online ahead of print.

Authors

Juan J Aristizabal-Henao¹, Anja P Biltoft-Jensen², Tue Christensen², Ken D Stark³

Affiliations

¹ Department of Kinesiology and Health Sciences, University of Waterloo, Waterloo, ON, Canada; Platforms and Translational Sciences, BPGbio Inc., Framingham, MA, United States.
² National Food Institute, Technical University of Denmark, Kongens Lyngby, Denmark.
³ Department of Kinesiology and Health Sciences, University of Waterloo, Waterloo, ON, Canada. Electronic address: kstark@uwaterloo.ca.

PMID: 38710305
DOI: 10.1016/j.tjnut.2024.04.038

Abstract

Background: The intake of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) have been associated with health benefits. Blood levels of these fatty acids, measured by gas chromatography (GC), are associated with their dietary intake, but the relationships with lipidomic measurements are not well defined.

Objectives: This study aimed to determine the lipidomic biomarkers in whole blood that predict intakes of EPA + DHA and examine the relationship between lipidomic and GC-based n-3 polyunsaturated fatty acid (n-3 PUFA) biomarkers.

Methods: Lipidomic and fatty acid analyses were completed on 120 whole blood samples collected from Danish participants. Dietary intakes were completed using a web-based 7-d food diary. Stepwise multiple linear regression was used to identify the fatty acid and lipidomic variables that predict intakes of EPA + DHA and to determine lipidomic species that predict commonly used fatty acid biomarkers.

Results: Stepwise regression selected lipidomic variables with an R² = 0.52 for predicting EPA + DHA intake compared to R² = 0.40 for the selected fatty acid GC-based variables. More predictive models were generated when the lipidomic variables were selected for females only (R² = 0.62, n = 68) and males only (R² = 0.72, n = 52). Phosphatidylethanolamine plasmalogen species containing EPA or DHA tended to be the most predictive lipidomic variables. Stepwise regression also indicated that selected lipidomic variables can predict commonly used fatty acid GC-based n-3 PUFA biomarkers as the R² values ranged from 0.84 to 0.91.

Conclusions: Both fatty acid and lipidomic data can be used to predict EPA + DHA intakes, and fatty acid GC-based biomarkers can be emulated by lipidomic species. Lipidomic-based biomarkers appear to be influenced by sex differences, probably in n-3 PUFA and lipoprotein metabolism. These results improve our ability to understand the relationship between novel lipidomic data and GC fatty acid data and will increase our ability to apply lipidomic methods to fatty acid and lipid nutritional research.

Keywords: blood biomarkers; dietary assessment; dietary intake; docosahexaenoic acid; eicosapentaenoic acid; gas chromatography; lipidomics; omega-3 polyunsaturated fatty acids; tandem mass spectrometry.