Circulating plasma metabolites and risk of rheumatoid arthritis in the Nurses’ Health Study
Open Access
- 20 April 2020
- journal article
- research article
- Published by Oxford University Press (OUP) in Rheumatology
- Vol. 59 (11), 3369-3379
- https://doi.org/10.1093/rheumatology/keaa125
Abstract
RA develops slowly over years. We tested for metabolic changes prior to RA onset using a large non-targeted metabolomics platform to identify novel pathways and advance understanding of RA development. Two hundred and fifty-four incident RA cases with plasma samples drawn pre-RA onset in the Nurses’ Health Study (NHS) cohorts were matched 1:2 to 501 controls on age, race, menopause/post-menopausal hormone use and blood collection features. Relative abundances of 360 unique, known metabolites were measured. Conditional logistic regression analyses assessed associations between metabolites and incidence of RA, adjusted for age, smoking and BMI, accounting for multiple comparisons. Subgroup analyses investigated seropositive (sero+) RA and RA within 5 years of sample collection. Significant metabolites were then tested in a female military pre-RA case–control study (n = 290). In the NHS, metabolites associated with RA and sero+RA in multivariable models included 4-acetamidobutanoate (odds ratio (OR) = 0.80/S.d., 95% CI: 0.66, 0.95), N-acetylputrescine (OR = 0.82, 95% CI: 0.69, 0.96), C5 carnitine (OR = 0.84, 95% CI: 0.71, 0.99) and C5:1 carnitine (OR = 0.81, 95% CI: 0.68, 0.95). These were involved primarily in polyamine and leucine, isoleucine and valine metabolism. Several metabolites associated with sero+RA within 5 years of diagnosis were replicated in the independent military cohort: C5 carnitine (OR = 0.55, 95% CI: 0.33, 0.92), C5:1 carnitine (OR = 0.62, 95% CI: 0.39, 0.99) and C3 carnitine (OR = 0.57, 95% CI: 0.36, 0.91). Several metabolites were inversely associated with incidence of RA among women. Three short-chain acylcarnitines replicated in a smaller dataset and may reflect inflammation in the 5-year period prior to sero+RA diagnosis.Funding Information
- National Institute of Arthritis and Musculoskeletal and Skin Diseases (R01 AR049880, R01 AR057327, K24 AR066109, K23 AR075070, K23 AR069688, L30 AR066953, P30 AR070253, P30 AR072577, R01 AR071326)
- National Human Genome Research Institute (U01 HG008685)
- National Heart, Lung, and Blood Institute (R01 HL119718)
- National Institutes of Health
- Congressionally Directed Medical Research Program (PR120839, W81XWH-13-1-0408)
- NIH (UM1 CA186107, UM1 CA17626, R01 CA049449, R01 CA067262)
This publication has 46 references indexed in Scilit:
- Differences between Human Plasma and Serum Metabolite ProfilesPLOS ONE, 2011
- Diagnostic properties of metabolic perturbations in rheumatoid arthritisArthritis Research & Therapy, 2011
- β-Elemene Piperazine Derivatives Induce Apoptosis in Human Leukemia Cells through Downregulation of c-FLIP and Generation of ROSPLOS ONE, 2011
- Total cholesterol and LDL levels decrease before rheumatoid arthritisAnnals Of The Rheumatic Diseases, 2009
- Comparison of Threshold Cutpoints and Continuous Measures of Anti-Cyclic Citrullinated Peptide Antibodies in Predicting Future Rheumatoid ArthritisThe Journal of Rheumatology, 2009
- Fatty acids as modulators of the cellular production of reactive oxygen speciesFree Radical Biology & Medicine, 2008
- Spermine and spermidine mediate protection against oxidative damage caused by hydrogen peroxideAmino Acids, 2007
- Identification of an urinary metabolite profile associated with osteoarthritisOsteoarthritis and Cartilage, 2005
- Carnitine metabolism during exerciseLife Sciences, 1994
- The american rheumatism association 1987 revised criteria for the classification of rheumatoid arthritisArthritis & Rheumatism, 1988