Genetic susceptibility to methylmercury developmental neurotoxicity matters

Epidemiological studies have demonstrated the developmental neurotoxicity associated with prenatal methylmercury exposure (Grandjean and Landrigan, 2006); However, susceptibility to methylmercury toxicity may be increased by genetic factors. This observation raises the question of possible dependence of developmental neurotoxicity on genetic predisposition. A few years ago, a National Research Council (NRC) evaluation of the scientific background for risk assessment concluded that “attention should be directed to vulnerable individuals and subpopulations that may be particularly susceptible or more highly exposed” (National Research Council, 2009). The panel also noted that “variability in susceptibility and vulnerability has received less detailed evaluation in most EPA health effects assessments.” A previous NRC review estimated that, under certain circumstances, individual susceptibility could range up to 50,000-fold, and as much as 5% of the population could well be at least 25-fold more susceptible than the average (National Research Council, 2000). In risk assessment, an uncertainty factor of 10 is commonly used to take into account intraspecies susceptibility. In regard to methylmercury, at the recommendation of the NRC (National Research Council, 2009), EPA used the default 10-fold intraspecies uncertainty factor. In contrast, the European Food Safety Authority recently argued that, for methylmercury, a partial uncertainty factor of 2 would be sufficient when a benchmark dose level (BMDL) had been obtained from a birth cohort that would represent the most vulnerable population (European Food Safety Authority, 2012). The magnitude of the intraspecies uncertainty factor therefore appears controversial, and better scientific documentation has been recommended (Dorne, 2010). Given that gene-environment interaction (GxE) may also play a role in regard to disease pathogenesis in a more general sense, as highlighted by a recent NIH workshop (Bookman et al., 2011), the variability in susceptibility to neurotoxicity between population groups appears to be an important research priority. The recent findings on GxE for methylmercury neurotoxicity are supported by several previous studies. Thus, mutations in certain genes seem to convey a greater risk of elemental mercuryassociated neurobehavioral deficits or symptoms in adults working in dental clinics (Echeverria et al., 2006, 2010; Heyer et al., 2008, 2009). This evidence was recently extended to children exposed to inorganic mercury from amalgam fillings (Woods et al., 2012, 2013). The reasons for such interactions are only partially understood, but some gene variants may predict a greater retention of mercury compounds in the body. Thus, gene mutations seem to affect the retention of inorganic mercury and methylmercury in the body, e.g., genes that affect glutathione (GSH) and metallothionein metabolism (Gundacker et al., 2007; Schlawicke Engstrom et al., 2008; Wang et al., 2012). Other studies have also considered absorption-distributionmetabolism-elimination (ADME) genes that may be of importance. Thus, methylmercury is eliminated from the liver as GSH conjugates, and the ratelimiting enzyme for GSH synthesis is glutamyl-cysteine ligase (GCL), which is composed of a catalytic subunit (GCLC) and a modifier subunit (GCLM). Further, the glutathione-S-transferases (GST) catalyze the conjugation of GSH (Gundacker et al., 2007). A recent study in Sweden indicates that a GCLC polymorphism affects methylmercury retention, and that glutathione S-transferase pi 1 (GSTP1) may play a role in conjugating methylmercury with GSH (Schlawicke Engstrom et al., 2008). The GCLC SNP (Single Nucleotide Polymorphism) rs1555903 also showed a highly significant (p = 0.007) main effect on mercury retention in the umbilical cord in a UK birth cohort (Julvez et al., 2013). Mutations in glutathione Stransferase mu 1 (GSTM1) and GCLM also seem to affect the retention of methylmercury from fish and seafood (Barcelos et al., 2013). Another possibility is that certain genotypes are less resistant to particular toxic effects. For example, in a study of dentists and dental assistants, deficits in neuropsychological performance occurred most frequently if the subject had a mutation in the coproporphyrinogen oxidase gene (CPOX4) (Echeverria et al., 2006); the same SNP also showed a significant main effect on general cognitive function in a UK birth cohort (Julvez et al., 2013). Recently, Woods