Fish Consumption and Mercury Exposures Among Sub-populations in Canada through Targeted Biomonitoring: Results from Dene/Métis Communities of the Northwest Territories and a Maternal-Infant Cohort Study

Abstract

Fish contain many important nutrients including omega-3 polyunsaturated fatty acids (PUFA) (such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)) and essential elements (such as selenium (Se)), and fish consumption has been associated with various health benefits. However, fish can also be a source of exposure to environmental contaminants, such as mercury (Hg). Fish consumption is considered to be a major source of Hg exposure for the Canadian general population, and this can pose a health risk for populations that consume significant amounts of fish or consume fish containing elevated levels of Hg. This thesis focused on two sub-populations in Canada that may be disproportionately impacted by exposure to Hg (either by being especially exposed to Hg or especially sensitive to the negative effects of Hg). The first was Dene/Métis communities in northern Canada, as these communities frequently consume locally harvested freshwater fish and elevated levels of Hg have been measured in some fish species from some local lakes in northern Canada. prompting concerns about Hg exposure as these populations may be especially exposed. The second sub-population examined was pregnant women and children in Canada, as early life stages are especially sensitive to the neurotoxic effects of Hg as their nervous systems are undergoing rapid development. To address concerns for both of these potentially disproportionately impacted sub-populations, focused biomonitoring projects can be used. In the Northwest Territories, to address concerns about exposure to environmental contaminants, including Hg from local fish, a community-based human biomonitoring project was implemented in First Nations communities in the Dehcho and Sahtú regions of the Northwest Territories. To understand chemical exposures, including Hg, in Canadian pregnant women and children, Health Canada is conducting the Maternal Infant Research on Environmental Chemicals (MIREC) pregnancy cohort study and follow-up studies, which include a biomonitoring component. This thesis draws on the results of both of these human biomonitoring projects, along with results and samples from on-going fish sampling projects and research in the Northwest Territories, to examine Hg exposures, in the context of fish consumption, among sub-populations in Canada. In order to do this, this thesis had five main objectives which were to: 1. measure Hg bioaccessibility in commonly consumed northern freshwater fish; 2. examine and describe Hg biomarker levels in members of northern Dene/Métis communities; 3. identify potential determinants of hair Hg, blood Hg, blood Se, and plasma EPA+DHA concentrations in members of northern Dene/Métis communities; 4. construct and refine a site-specific exposure assessment model for Hg, Se, and EPA+DHA, from the consumption of local freshwater fish, for the Dehcho and Sahtú regions of the NWT; and 5. determine if there are associations between early life exposures to low levels of Hg and child IQ, and if these associations are modified by prenatal fish consumption. The human biomonitoring project in the Northwest Territories (2016-2018) included the collection of biological samples (including hair and blood), dietary data (through the use a food frequency questionnaire (FFQ)), and demographic/anthropometric data. From the FFQ, the four most commonly consumed fish species were identified as Lake Whitefish (, Lake Trout, Northern Pike, and Walleye. Concentrations of Hg, and to a lesser extent Se and omega-3 PUFA, were available from waterbodies in the Northwest Territories for these fish species. However, to address the lack of Hg bioaccessibility data for northern freshwater fish species and fulfill objective 1 of this thesis, fish samples that were collected from waterbodies in northern Canada were analyzed for total Hg bioaccessibility and mean total Hg bioaccessibility in raw fish muscle tissue ranged between 56-96% percent for the species examined and was 39% in raw Burbot livers. Cooking increased total Hg concentrations (likely due to moisture loss) in the samples, but total Hg bioaccessibility was much lower in the cooked samples (on average 40% lower). Thus, bioaccessible concentrations of total Hg in the cooked samples were on average 32% lower. In order to fulfill objective 2 of this thesis, human biomarker concentrations of Hg (in hair and blood), were determined and hair to blood Hg concentration ratios were calculated. The majority of hair and blood Hg concentrations were below health-based guidance values and blood Hg concentrations were similar to or below those measured in the Canadian general population. Hair Hg concentrations, especially for the Sahtú region, were higher on average than those measured in the Canadian general population, however, hair:blood Hg concentration ratios in these regions (619:1 for the Dehcho region and 1220:1 for the Sahtú region) were significantly higher than ratios typically reported in the literature for other populations. For long hair samples, hair Hg concentrations varied over time and total reported fish consumption seemed to follow a similar seasonal pattern. Human biomarker concentrations of Se (in blood) and EPA and DHA (in blood plasma) were also measured and potential determinants of these and the Hg biomarker levels were examined to fulfill objective 3 of this thesis. Blood Se concentrations were similar to those measured in the Canadian general population and plasma EPA and DHA concentrations were similar to levels observed in other Canadian studies. In models to explain hair Hg, blood Hg, and plasma EPA and DHA, participant age, the consumption of at least one species of fish, and the consumption of waterfowl (those that are omnivorous or piscivorous) were included as explanatory variables in the final models. The final model explaining variability in blood Se concentrations included region, and the consumption of several species of fish, burbot livers, and caribou livers and kidneys.. Results generated in addressing objectives 1-3, were used to construct and refine a probabilistic exposure assessment model for Hg, Se, and EPA and DHA, from the consumption of the four most commonly consumed fish species (to fulfill objective 4 of this thesis). The majority of Hg estimates were below tolerable daily intakes (TDIs) and the inclusion of waterbody-specific Hg concentration data and species-specific Hg bioaccessibility data resulted in lower estimated Hg intakes which was more consistent with biomarker concentrations. Model estimates for Se and EPA and DHA, indicated that the consumption of these four fish species was an important contributor to the dietary intakes of these nutrients. Overall, model estimates and biomarker levels indicate that although Hg concentrations may be elevated in some fish species, from some waterbodies in these regions, human Hg exposures in the Dehcho and Sahtú regions are relatively low. Finally, since early life stages are particularly susceptible to the neurotoxic effects of Hg, and as it is unclear if these effects are also seen at low Hg concentrations (like those seen in the Canadian general population and in the Dehcho and Sahtú regions), associations between prenatal and childhood Hg exposures and child intelligence quotient (IQ), were examined using data from the MIREC cohort study (2013-2015) to address objective 5 of this thesis. Higher cord blood Hg concentrations were associated with lower Performance IQ (PIQ) (ꞵ=-3.27; 95%CI: -6.44, -0.09) in male children with the lowest prenatal fish consumption. Progressively stronger positive associations were observed with PIQ in male children for moderate (ꞵ=1.08; 95%CI: -0.10, 2.26) and high (ꞵ=3.07; 95%CI: 1.95, 4.19) prenatal fish consumption. Cord blood Hg concentrations were positively associated with female children’s Full-Scale IQ (FSIQ) (ꞵ = 1.29; 95% CI: 0.77, 1.81) and PIQ (ꞵ = 2.01; 95% CI: 1.19, 2.83); however, when stratified only in the highest fish consumption subgroup. Among female children, higher child blood Hg concentrations were associated with an approximately 1-point increase in FSIQ, VIQ, and a General Language Composite (GLC). Prenatal exposure to low levels of Hg was associated with lower PIQ scores in male children with low prenatal fish intake. Findings from this analysis suggest that children exposed to low levels of Hg may be vulnerable to adverse cognitive effects especially when exposure occurs through non-fish sources of Hg. This reinforces the benefits of fish consumption (of species low in Hg) during pregnancy. This work emphasizes that although fish consumption can be a source of Hg exposure for the Canadian general population and northern Indigenous communities, it is also a source of important nutrients. This thesis fills an important gap in the literature on the levels, determinants, and dietary intakes of Hg, Se, and omega-3 PUFA in First Nations communities in the Dehcho and Sahtú regions of the Northwest Territories. It also provides evidence of the importance of fish consumption as a modifier of the associations between early life exposures to low levels of Hg and child IQ. Future work in northern regions should include the collection of additional waterbody-specific fish Se and omega-3 fatty acid concentration and bioaccessibility data. Future exposure assessments and epidemiological studies on Hg exposure should take into account the important nutritional benefits of fish consumption. The consumption of fish, of those species that are considered low in Hg, should continue to be encouraged for populations in Canada.