Redox Biology (Jan 2020)

Multiple plasma metals, genetic risk and serum C-reactive protein: A metal-metal and gene-metal interaction study

  • Yu Yuan,
  • Pinpin Long,
  • Kang Liu,
  • Yang Xiao,
  • Shiqi He,
  • Jun Li,
  • Tingting Mo,
  • Yiyi Liu,
  • Yanqiu Yu,
  • Hao Wang,
  • Lue Zhou,
  • Xuezhen Liu,
  • Handong Yang,
  • Xiulou Li,
  • Xinwen Min,
  • Ce Zhang,
  • Xiaomin Zhang,
  • An Pan,
  • Meian He,
  • Frank B. Hu,
  • Ana Navas-Acien,
  • Tangchun Wu

Journal volume & issue
Vol. 29

Abstract

Read online

Background: C-reactive protein (CRP) is a well-recognized biomarker of inflammation, which can be used as a predictor of cardiovascular disease. Evidence have suggested exposure to multiple metals/metalloids may affect immune system and give rise to cardiovascular disease. However, it is lack of study to comprehensively evaluate the association of multiple metals and CRP, the interactions between metals, and the gene-metal interaction in relation to CRP levels. Aims: To explore the associations of multiple plasma metals with serum CRP, and to test the interactions between metals, and gene-metal interactions on the levels of serum CRP. Methods: We included 2882 participants from the Dongfeng-Tongji cohort, China, and measured 23 plasma metals and serum CRP concentrations. The genetic risk score (GRS) was calculated based on 7 established CRP-associated variants. For metals which were associated with the levels of CRP, we further tested the interactions between metals on CRP, and analyzed the gene-metal interactions on CRP. Results: The median level for CRP in the total population was 1.17 mg/L. After multivariable adjustment, plasma copper was positively associated with serum CRP (FDR < 0.001), whereas selenium was negatively associated with serum CRP (FDR = 0.01). Moreover, selenium and zinc attenuated the positive association between high plasma copper and CRP (P for interaction < 0.001). Participants with a higher GRS had a higher CRP level, with the increase in ln-transformed CRP per increment of 5 risk alleles were 0.64 for weighted GRS, and 0.54 for unweighted GRS (both P < 0.001). Furthermore, the genetic association with CRP was modified by copper concentration (P for interaction < 0.001). Conclusions: Our results suggest that serum CRP is positively associated with plasma concentration of copper, and inversely associated with selenium. Plasma zinc, selenium and CRP genetic predisposition would modify the associations between plasma copper and serum CRP. Keywords: Copper, Selenium, C-reactive protein, Metal-metal interaction, Gene-metal interaction, Genetic risk score