The Association Between Inflammatory and Oxidative Stress Biomarkers and Plasma Metabolites in a Longitudinal Study of Healthy Male Welders

Shangzhi Gao,1 Corbin Quick,2 Marta Guasch-Ferre,3 Zhu Zhuo,2 John M Hutchinson,2 Li Su,1 Frank Hu,3 Xihong Lin,2 David Christiani1,4 1Environmental Health, Harvard University T H Chan School of Public Health, Boston, MA, USA; 2Biostatistics, Harvard University T H Chan School of Public Health, Boston, MA, USA; 3Nutrition, Harvard University T H Chan School of Public Health, Boston, MA, USA; 4Pulmonary and Critical Care Division, Department of Medicine, MA General Hospital, Boston, Massachusetts, USACorrespondence: David ChristianiEnvironmental Health, Harvard University T H Chan School of Public Health, Boston, MA, 02115, USAEmail dchris@hsph.harvard.eduIntroduction: Human metabolism and inflammation are closely related modulators of homeostasis and immunity. Metabolic profiling is a useful tool to understand the association between metabolism and inflammation at a systemic level.Objective: To investigate the longitudinal associations between the concentration of plasma metabolites and biomarkers related to inflammation and oxidative stress.Methods: We conducted a repeated cross-sectional analysis consisting of 8 short-term panels that included 88 healthy adult male welders in Massachusetts, USA. In each panel, we collected 1– 6 repeated measurements of blood and urine. We used a human vascular injury panel assay and custom cytokine/chemokine assay to quantify inflammatory biomarker plasma levels, liquid chromatography-mass spectrometry to quantify the concentrations of 665 plasma metabolites, and a competitive enzyme-linked immunoassay to quantify urinary 8-OHdG and 8-isoprostane levels. We used linear mixed effects models to estimate the longitudinal association between each inflammatory and oxidative stress biomarker and each metabolite.Results: At a 5% FDR threshold, we detected ≥ 1metabolite association for 8 unique inflammatory and oxidative stress biomarkers: urinary 8-isoprostane, plasma C-reactive protein (CRP), serum amyloid A (SAA), intercellular adhesion molecule 1, circulating vascular cell adhesion molecule-1, interleukin 8 (IL-8), interleukin 10 (IL-10) and vascular endothelial growth factor. Specifically, 3 metabolites in the androgenic steroids pathway were negatively associated with SAA; 3 dihydrosphingomyelins metabolites were positively associated with 1 or more of CRP, SAA, IL-8 and IL-10; 4 metabolites in acyl choline metabolism pathways were negatively associated with IL-8; 7 lysophospholipid metabolites were negatively associated with 1 or more of CRP, SAA and IL-8; 4 sphingomyelins were positively associated with CRP and/or SAA; and 10 metabolites in the xanthine pathway were positively associated with urinary 8-isoprostane.Conclusion: We found that metabolites in phospholipid groups had strong associations with multiple inflammatory biomarkers, especially CRP, SAA and IL-8. The mechanism of these associations warrants further investigation.Keywords: metabolism, inflammation, phospholipids, metals, particulates, metabolomics, occupational health