Mercury, polychlorinated biphenyls, and immune response indicators in the National Health and Nutrition Examination Survey (NHANES)
Gallagher, Carolyn Michele
MetadataShow full item record
Autoimmune diseases are among the leading causes of death among young and middle-aged women, and prevalence is increasing; however, autoimmune etiology is poorly understood. The need to investigate relationships between environmental factors and human autoimmune responses has been prioritized by the World Health Organization and the National Institutes of Health (NIH), yet epidemiologic research is sparse. Mercury (Hg) and polychlorinated biphenyls (PCBs) are ubiquitous and persistent environmental contaminants with early evidence of immunotoxic associations. Immunotoxic responses fall on a continuum, with the normal range of immune responses at the center, immunostimulation and autoimmunity at one end, and immunosuppression and infection at the other. Common immune response indicators include antibodies against "self", that is, the body's own normal cells, and antibodies against pathogens such as wild-type or vaccine-type viral antigens. Serum vaccine antibody concentration is an increasingly recognized parameter of children's immune system response to environmental contaminants in population studies. The NIH and the Centers for Disease Control and Prevention have prioritized the need to evaluate differential immune responses to vaccination; and the U.S. Department of Health and Human Services (DHHS) identified the need to evaluate the relationships between environmental exposures and immunologic outcomes in susceptible subpopulations. Further, sex, nutritional deficiencies and related metabolic cofactors may represent important immune susceptibility cofactors. The objective of this dissertation research is to evaluate the relationships between total blood Hg and serum PCBs with human immune response indicators across the continuum of immunotoxicology among susceptible subpopulations. My dissertation research uses cross-sectional data obtained from U.S. probability samples of the National Health and Nutrition Examination Survey to address gaps in epidemiologic research of the relationships between PCBs and Hg and immune response indicators in each of four separate studies A, B, C and D. Study A evaluates the relationships between total blood Hg, serum PCBs, and antinuclear antibody (ANA) positivity, an indicator of systemic autoimmune response, stratified by sex. Study B evaluates the relationship between Hg and indicators of organ-specific immune response, specifically, thyroid autoantibody positivity, as well as a risk factor for hypothyroidism, stratified by sex and iodine status in order to evaluate these recognized autoimmune thyroid disease susceptibility factors. Studies C and D evaluate the relationships between Hg exposure and serum concentrations of measles antibodies, and between Hg exposure and serum concentrations of rubella antibodies, in nutritionally-susceptible subpopulations of children. Key findings are: (A) There were no associations between Hg and ANA, or between nondioxin-like PCBs and ANA, among males or females; however, among females, dioxin-like PCBs were significantly and positively related to ANA positivity. (B) Hg was positively associated with the hypothyroidism risk factor among females with lower iodine and iodine deficient females, and in these same subsamples, positive associations were also observed between Hg and thyroid peroxidase autoantibodies. In contrast to females, who showed an overall pattern of elevated odds ratios for thyroid measures that included autoantibodies, most consistently among females with lower and deficient iodine levels, an overall inverse pattern was evident among males with higher and excessive iodine levels. (C) Positive associations were observed between Hg and serum measles antibodies in boys with lower folate and B-12, and higher homocysteine levels, but inverse associations were observed in all other children. (D) Positive associations were observed between Hg and serum rubella antibodies in the same nutritionally susceptible subset as in study C, with inverse associations among the remaining subset; however, findings were observed in both boys and girls, combined. In this thesis, I present novel findings of positive associations between (A) dioxin-like PCBS and ANA positivity in females; (B) Hg exposure and a risk factor for hypothyroidism in women with lower or deficient iodine levels; and (C) Hg exposure and elevated serum measles IgG antibodies in boys with lower folate and vitamin B-12 and higher homocysteine levels, as well as elevated rubella IgG antibodies in boys and girls with this same nutritional susceptibility. Moreover, these are the first studies to show these relationships in a general population with lower immunotoxicant exposure levels to these ubiquitous contaminants as compared with studies in populations with known higher exposures due to industrial pollution and/or contaminated dietary sources. Given the divergent findings by sex and nutritional cofactors, an overarching recommendation is for environmental epidemiology to identify susceptible subpopulations and to conduct epidemiological research stratified by these vulnerability subsets. Taken together, these findings shine the spotlight on the need for prospective studies that take an integrative approach to evaluate these relationships across the immunotoxicological continuum, and have the potential to guide public health decision-making for the protection of the most vulnerable.