A Gene-environment Study of Metallothionein Single Nucleotide Polymorphisms, Mercury Biomarker Levels and Peripheral Nerve Function.

Abstract

Mercury (Hg) is a potent neurotoxicant. Recent studies have suggested that several genes that mediate mercury metabolism are polymorphic in humans. We hypothesized that single nucleotide polymorphisms (SNPs) in metallothionein (MT) genes may underlie inter-individual differences in mercury biomarker levels and peripheral nerve function. We also investigated associations of Hg exposure with nerve function and the agreement between screening procedures for neuropathy in the feet. Dental professionals (n=515) were recruited in 2009 and 2010. Samples and measurements included: hair and urine for biomarker measurements; sensory nerve conduction measurements (onset latency, peak latency and amplitude) of the median, ulnar and sural nerves; monofilament testing of the right great toe; and buccal swabs for DNA. Questionnaires were completed for demographics, exposures to dental amalgam and dietary fish intake, neuropathic symptoms, and confounders. The mean Hg levels in urine (1.06µg/L) and hair (0.51µg/g) were not significantly different from the US population (0.95 µg/L and 0.47µg/g, respectively) (NHANES 1999-2000, 2003-2004). Multivariate regression analysis found subjects with MT1M (rs2270836) AA genotype (n=10) had lower urinary Hg levels than GG after controlling for exposure and potential cofounders. After controlling for methylmercury intake from fish, subjects with MT1A (rs8052394) GA and GG (n=24), MT1E (rs708274) GT and TT (n=51), MT1M (rs9936741) TT had lower hair Hg levels compared to AA, GG, and TC and CC (n=15), respectively. After accounting for hair Hg levels, increased polyunsaturated fatty acid intake from fish was found to improve nerve function in all nerve measurements except median peak and onset latencies. No consistent relationship was observed for urine Hg levels and nerve function. Only 3 out of a total of 504 multivariate models that investigated effect modification of SNPs on nerve-function-biomarker relationships had stable and statistically significant interaction terms. Overall, the findings suggested that some MT genetic polymorphisms may influence mercury biomarker levels. There might be a beneficial effect of fish consumption on nerve function but little evidence was shown for effect modification of the studied SNPs on nerve-function-biomarker relationships. Poor agreement (Kappa: -0.14 ~ 0.44) between procedures demonstrated the challenges for further development and evaluation of methods in a non-clinical population.