Background/Purpose: Trace elements may be differentially associated with osteoarthritis (OA) risk. Single-nucleotide polymorphisms (SNPs) have been found to correlate to cumulative exposure level, free of temporal effects. The purpose of this study was to identify whether SNPs known to be associated with the body concentration of certain trace elements were associated with OA development in a prospective cohort and then explore functional consequences (Figure 1).

Methods: This was a cross-sectional analysis from the second follow-up visit (2006-11) of the Johnston County Osteoarthritis Project, a prospective population-based cohort designed to evaluate risk factors for development of OA. This analysis was limited to White participants with toenail metals data, as exposure SNPs were selected from those with robust (p < 5x10-8) associations with trace element body concentration in a Scandinavian population (PMID: 38594418). Radiographic knee (rKOA) and hip (rHOA) OA were defined by Kellgren-Lawrence (K/L) grade ≥ 2 on bilateral posteroanterior knee or hip radiograph, respectively, at any point throughout follow-up. Symptomatic OA (sOA) consisted of radiographic OA with pain in the same joint. Linear regression of each SNP variant with natural log-transformed toenail metal concentrations was assessed before performing logistic regression for OA outcomes and each SNP, metal, and their interaction as predictors, providing odds ratio [OR]; 95% confidence interval [CI]) if interaction p-value< 0.1. Each model was adjusted for age, sex, body mass index (BMI), smoking, and alcohol use. A functional study was then conducted for a novel impacted gene (FAM210B) in a previously described chondrocyte OA model (PMID: 33248223).

Results: Among 955 White adults (mean age 66 ± 11, mean BMI 30 ± 6 kg/m2, 38% male), 42% had rKOA and 30% had rHOA. There were no associations between toenail metal concentrations of selenium, mercury, chromium, scandium, zinc, iron, or cadmium with selected SNPs. However, SNPs previously linked to decreased manganese concentration including rs6099115 and rs855791 were associated with rOA, but not sOA. In multivariable regression, both heterozygous and homozygous variants of rs6099115 were associated with increased odds of rKOA (OR 2.66; 95% CI 1.01, 7.05 and 3.05; 1.18, 7.90, respectively, Table 1). Homozygous affected variant rs855791 was associated with increased odds of rKOA only when toenail levels of chromium or iron were increased (OR 1.62; 1.13, 2.31, and 1.53; 1.07, 2.18, respectively). Variants in rs855791 result in missense mutations in transmembrane protease serine 6 (TMPRSS6), involved in extracellular matrix degradation and hepcidin regulation. Variants in rs6099115 result in missense mutations of FAM210B, a mitochondrial membrane transporter involved in cellular response to iron, estradiol, and cell differentiation. In a chondrocyte OA model, FAM210B expression was decreased by 69.2%, p < 5x10-63 (Figure 2).

Conclusion: Two novel associations were identified, linking risk of rOA with SNPs known to be associated with decreased manganese concentrations, with functional consequences in FAM210B. Further investigation in the roles of FAM210B and TMPRSS6 in the pathogenesis of OA is needed.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/manganese-associated-single-nucleotide-polymorphisms-are-associated-with-radiographic-knee-osteoarthritis-the-johnston-county-osteoarthritis-project/