Serum metabolomic profiling and its association with 25-hydroxyvitamin D

• Published in: Clinical nutrition (Edinburgh, Scotland) Scotland), 2020-04, Vol.39 (4), p.1179-1187
• Main Authors: Leung, Raymond Y.H., Li, Gloria H.Y., Cheung, Bernard M.Y., Tan, Kathryn C.B., Kung, Annie W.C., Cheung, Ching-Lung
• Format: Article
• Language: English
• Subjects: 25-Hydroxyvitamin D; Epidemiological studies; Metabolomics; Serum metabolites; Vitamin D
• ISSN: 0261-5614; 1532-1983
• DOI: 10.1016/j.clnu.2019.04.035

The beneficial effect of vitamin D on the risk of non-musculoskeletal diseases has been investigated in observational studies and randomized clinical trials, but the findings were inconsistent. Identification of the metabolomic profile associated with vitamin D helps to identify novel biomarkers and increase the understanding of the biochemical and physiological role of vitamin D in different health conditions. Serum metabolomic profiling was performed using liquid chromatography/tandem mass spectrometry [LC/MS] and their association with serum 25(OH)D was evaluated using multivariable linear regression in the baseline cohort of 316 participants (aged 20 or above; 92 men, 224 women; mean age±SD: 48.1 ± 15.8 years) and in the follow-up cohort of 275 participants (aged 20 or above; 12 men, 263 women; mean age: 56.2 ± 9.6) of the Hong Kong Osteoporosis Study. We discovered and validated potential metabolites; and by meta-analysis of these associations in two cohorts, we identified metabolites that were significantly associated with serum 25(OH)D levels. Among 835 known metabolites, 102 metabolites showed significant correlation with 25(OH)D levels at baseline visit. Of these metabolites, 27 were validated in the follow-up visit. In meta-analysis of data from these two visits, 13 metabolites were highly correlated with 25(OH)D. The majority of metabolites identified were lipid in nature. Docosahexaenoylcarnitine and eicosapentaenoylcholine had the highest correlations, with effect estimates 0.2554 (p = 9.60 × 10−9) and 0.1682 (p = 4.94 × 10−7) respectively. In Hong Kong Chinese at least, serum vitamin D level is closely related to lipid metabolism. Our finding highlights an important new direction in the study of vitamin D in different health conditions.