Immunotoxic effects of mercury in humans: In vivo and in vitro evidence.

摘要

Experimental evidence, using both in vivo and in vitro models, supports a role for mercury (Hg) exposure in the modulation of immune response, though the impact of Hg can be both qualitatively and quantitatively different depending upon the species of Hg used and the genetic and environmental background against which exposure takes place. Epidemiological studies also suggest that Hg can have immunotoxic effects in humans, though this literature is far smaller compared to experimental evidence of Hg neurotoxicity. The goal of this dissertation is to enhance understanding of the effects of low, physiologically- and environmentally-relevant concentrations of Hg on the human immune system. Two approaches were used.;First, I examined evidence from a cross-sectional epidemiological study in terms of biomarkers of immune dysfunction in an Hg-exposed population as compared to a non-exposed referent population. In Chapter 2, I report that Hg-exposed gold miners from Amazonian Brazil are more likely to have elevated titers of antinuclear and antinucleolar autoantibodies in serum, a biomarker for autoimmune dysfunction, compared to non-exposed Brazilian emerald and diamond miners. I also measured serum cytokine levels, and I report here that pro-inflammatory cytokines are elevated in the serum of Hg-exposed gold miners compared diamond and emerald miners.;Second, I extensively characterized an in vitro model using low concentrations of Hg in human peripheral blood mononuclear cells (PBMCs) in culture. I first examined changes in cellular subset phenotypes and cytokine release as a result of low-concentration mercuric chloride (HgCl 2) treatment, both in the presence and absence of the antigenic stimulus lipopolysaccharide (LPS). In Chapter 3, I report that pro-inflammatory cytokine production is the most sensitive endpoint to measure Hg response in human PBMCs stimulated by LPS. I also report that CD86 is down-regulated on CD11b+TLR4+ macrophages in a concentration-response manner with HgCl2 treatment in the presence of LPS. I observed no changes in cellular phenotypes or cytokine release as a result of HgCl2 treatment in the absence of LPS.;In order to more precisely estimate the effects of iHg on cytokine release and to understand the underlying variability in cytokine production by human PBMCs, I employed a multilevel study design to collect cytokine data on PBMCs from donating volunteers over multiple visits. I recruited 20 volunteers to donate blood six separate times. Seven cytokines were measured, representing pro-inflammatory signals, anti-inflammatory signals, and markers of T cell subset activation. Data collected on each cytokine were modeled using hierarchical, multi-level models. In Chapter 4, I report that the pro-inflammatory cytokines IL-1beta and TNF-alpha increase in a concentration-dependent fashion in response to Hg treatment, while the anti-inflammatory cytokines IL-10 and IL-1Ra are decreased.;Finally, the effects of methyl mercury (MeHg) and ethyl mercury (EtHg) in this in vitro system were compared to the effects of HgCl 2 (inorganic Hg, iHg,). In Chapter 5, I report that low concentrations of iHg and MeHg increase pro-inflammatory cytokine release in LPS-stimulated PBMCs from six individuals, though the trend for TNF-alpha release in response to MeHg was not statistically significant. EtHg treatment decreases pro-inflammatory cytokine release. Release of anti-inflammatory cytokines was not significantly affected by any Hg treatment in the absence of LPS.