Patterns of Fish and Whale Consumption and Concentrations of Methylmercury in Hair Among Residents of Western Canadian Arctic Communities

Background. Methylmercury contamination of the environment represents a substantial environmental health concern. Human exposure to methylmercury occurs primarily through consumption of fish and marine mammals. Heavily exposed subgroups include sport or subsistence fishers residing in Arctic communities. We aimed to estimate the association of fish/whale consumption patterns of Canadian Arctic subsistence fishers with the internal dose of methylmercury as measured in hair. Methods. This research was conducted within ongoing community projects led by the CANHelp Working Group in Aklavik and Fort McPherson, Northwest Territories and Old Crow, YT. We interviewed each participant using a fish-focused food-frequency questionnaire during September-November 2016 and collected hair samples concurrently. Methylmercury was measured in the full-length of each hair sample using gas chromatography inductively-coupled plasma-mass spectrometry. Multivariable random-effects linear regression estimated beta-coefficients and 95% confidence intervals (CIs) for the effect of fish/whale consumption on hair-methylmercury concentrations. Results. In total, 101 participants provided hair samples and diet data. The mean number of fish/whale species eaten by participants was 3.5 (SD:1.9). The mean hair-methylmercury concentration was 0.60μg/g (SD:0.47). Fish/whale consumption was positively associated with hair-methylmercury concentration, after adjusting for sex, hair length and use of permanent hair treatments. Hair-methylmercury concentrations among participants who consumed the most fish/whale in each season ranged from 0.30- 0.50μg/g higher than those who consumed <1 meal/week. Conclusions. Hair-methylmercury concentrations were below the 6.0μg/g threshold for safe exposure defined by Health Canada, suggesting that fish/whale consumption patterns among participants are not increasing their risk of known serious health effects of methylmercury exposure.


Introduction
Mercury is a chemical element with three valence states: elemental mercury (Hg 0 ); divalent inorganic mercury compounds (Hg 2+ ); and organic compounds. Given their capacity to induce potent toxicological effects in humans, contamination of the environment with mercury compounds represents a substantial environmental health concern. For this reason, mercury has been the focus of a large body of research aimed at identifying the mechanisms through which it enters the environment, as well as pathways for human exposure and subsequent toxicological effects.
Mercury is stored in geological reservoirs within the earth's crust in its elemental form (1)(2)(3). Release from these reservoirs into surface soil, water and the atmosphere occurs through geological weathering, defined as the alteration or breakdown of rocks and minerals by mechanical and chemical processes (1,2,4,5). Weathering is a natural process resulting from changes in temperature or pressure, exposure to wind and water, or volcanic events (1,5). Anthropogenic activities of an industrial nature can accelerate weathering and the consequent release of elemental mercury from geological reservoirs (1)(2)(3)(5)(6)(7)(8)(9)(10). Additionally, human-caused global climate change increases the release of mercury by inducing changes to the carbon cycle that are conducive to chemical weathering (3,4,11). Other human activities, including burning fossil fuels, specific industrial processes and waste incineration, lead to direct release of mercury into the environment(1-3,6-10). Following release from geological reservoirs, mercury undergoes biogeochemical cycling that results in the formation of inorganic and organic mercury compounds (1,2,4).
When elemental mercury finds its way into aquatic systems, some of it transforms into methylmercury (MeHg), a process mediated by anaerobic organisms that involves the accumulate mercury. Residents of inland communities in the western Canadian Arctic, however, are target audiences for public health messages about fish consumption, without concurrent exposure assessments (14,15). Our preliminary ethnographic research in western Canadian Arctic communities revealed residents' concerns about mercury accumulation in their bodies and how it relates to their fish and marine mammal consumption habits. In response, we conducted this research to analyze data collected from residents of inland communities in the Canadian Arctic to: characterize fish and identifying as Gwich'in [Athabascan First Nation] or Inuvialuit [Inuit]) (18,19). Projects began in 2010 in Old Crow, Yukon (YT) (2011 census population=245, ~85% identifying as Vuntut Gwich'in(20,21), and in 2012 in Fort McPherson, NT (2011 census population=844, 90% Tetlit Gwich'in) (22). Participation in the mercury exposure project was open to all residents of these three communities during September-November 2016. Recruitment activities involved radio announcements, social media posts, flyers on community message boards, and directly contacting participants of CANHelp Working Group projects for which current contact information was available. Informed consent was obtained from all participants by providing them with an information sheet that outlined the study objectives, methods, information to be collected, benefits and potential risks of participation and confidentiality. Following review of this document, each participant was asked to fill out a consent form, confirming they had received enough information about the project and that they agree to participate. Project information sheets and consent forms were reviewed and approved by the Research Ethics Board at the University of Alberta and have been published previously (23).

Exposure Time Window.
Among healthy individuals, estimates of the scalp hair growth rate range from 0.6 to 3.36 cm/month, with an average of 1 cm/month (37)(38)(39). The concentration of mercury measured in hair reflects exposure over the growth period of the sampled hair, typically, the past few months depending on hair length. According to input from local planning committees, residents of participating communities consume the greatest amount of aquatic species, on average, during the spring and summer seasons.
For this reason, hair sample collection took place during the fall season (September-November).
Hair Sample Collection. The procedures for collecting hair samples were adapted from protocols outlined by the United States Centers for Disease Control (CDC) for use in the National Health and Nutrition Examination Survey (NHANES) (41). We collected all hair samples from the occipital region of the scalp using stainless steel shears, obtaining a minimum of 120 mg of hair from each participant to allow for duplication of the laboratory measurements for quality assurance/quality control (QA/QC) purposes. To ensure enough hair was obtained from each participant, we used a high precision digital scale to weigh the sample immediately following collection. Given that hair length determines the exposure period represented in the strand, we also used a ruler to measured hair length (in cm) before transferring samples into a zip closable plastic bag and applying a label specifying the sample ID number, collection date, sample weight and hair length.
Additionally, we recorded information on use of permanent hair treatments, including hair dye or permanent waves, and time since the most recent treatment.
Laboratory Analysis of Samples. The collected hair samples were analyzed by the University of Alberta Biogeochemical Analytical Service Laboratory (BASL). This lab has been accredited by the Canadian Association for Laboratory Accreditation (CALA) as meeting ISO/IEC 17025 standards for the performance of specific tests. MeHg was measured in the full-length of each hair sample using gas chromatography inductively coupled plasma-mass spectrometry (GC-ICP-MS) (42,43). Quality control methods employed by the lab included the use of reference material 1AEA-085 for MeHg, total mercury and other trace elements in hair. Single point calibration was applied, and the calibration standard was analyzed in 4 replicates. The relative standard deviation for the ratio of Hg isotope 201:202 was considered acceptable if the value was less than 5%. If the value was greater than 5%, the calibration was repeated. Instrument and method blanks and a second source reference material were also used to monitor contamination with MeHg, accuracy and instrumental drift during analysis. These were incorporated into the analysis at of the year. Given the potential for preparation methods to alter the bioavailability of mercury in consumed fish, the FFQ also asked participants to specify how they typically prepare each type of fish/whale for eating and the parts they consume (45). Most participants were able to identify the specific species they consumed; pictures were available for those who were unsure. The potential for the overall composition of an individual's diet and intake of specific nutrients to directly or indirectly influence the toxicokinetic properties of MeHg has been described in the scientific literature (46).   To avoid assuming that the relationships between increasing consumption of food items and hair mercury levels were linear, each variable was converted to a categorical format.
When possible, category boundaries were defined so that there was no more than a twofold increase in number of servings within a category (44). The purpose of this was to generate categories within which the effect of interest does not vary substantially (44,48).
If data were too sparse to permit the use of optimal category boundaries, adjacent categories were collapsed to improve statistical precision. To confirm whether these variables could be modeled as continuous, the linearity of the relationship of hair MeHg concentration with the continuous form of each variable was assessed using a lowess plot (bandwidth: 0.80). The presence of a trend in the relationship between fish/whale consumption frequency and hair MeHg concentrations (μg/g) was detected using an extension of the Wilcoxon rank-sum test for testing trends over ordered groups that incorporates a correction for ties (49).
We used purposeful selection, as proposed by Hosmer and Lemeshow (2000), to identify the best set of adjustment variables for each of the season-specific exposure variables (50). This method follows a change-in-estimate approach, with variable selection decisions based on the extent to which each potential covariate influences the magnitude of exposure effects of interest (48,50): All potential covariates were included in a multivariable random effects model and subsequently removed one at a time. If the coefficient of any independent variable changed by ≥10% with the removal of a given covariate, the removed variable was included in the final model (48,50). Variables considered for inclusion in the model were: age, sex, use of permanent hair treatments, the proportion of consumed fish/whale species harvested from the ocean or local rivers, the proportion of consumed species usually prepared by cooking (versus eaten raw, dried or smoked), and other dietary frequencies, including fruits and vegetables, dairy products or regular use of dietary supplements. Hair length was automatically included in the final model to account for variation in the exposure period represented in hair strands of different lengths.
Community Effect. Although data were insufficient for estimating species-specific effects, some of this effect was likely picked up by the random effect, given the considerable variation in fish/whale species consumed across communities. A sensitivity analysis explored the extent to which variation in species consumed by participants from different communities explained the residual variation. We inspected community-specific patterns of fish/whale consumption, for each season separately, to identify species most likely to discriminate between communities based on the relative frequencies of their intake. We included species-specific and total fish/whale consumption frequencies in the same model if the correlation coefficients were <0.7 (48,50). The variables representing intake of the selected species were then added to the model for each season, to quantify changes in the residual variation across communities as measured by the SD. Given that data were limited, the linearity of the relationship between consumption of each speciesspecific consumption variable and hair mercury concentration was assessed to see whether they could be modeled as continuous.
Bias Analysis. Given the potential for MeHg measurement error to produce outcome misclassification, we conducted a quantitative bias analysis using the measured hair mercury concentrations among duplicated samples as parameters. The percent change between analyses of the same participant's hair was calculated. To achieve this, the value obtained during the repeat analysis of an individual's sample was subtracted from the originally measured value and the difference was divided by the originally measured value. For participants with more than 2 measurements, the largest difference between measured concentrations was used. To quantify the extent to which measurement error influenced inferences drawn from this analysis, the originally measured MeHg value was adjusted in two ways. First the overall mean percent change and the proportion of the repeated measurements that increased or decreased in value were used to estimate the magnitude of measurement error and frequency of change in either direction in the entire study population. Second, the mean percent change between repeated measurements and the proportion that increased or decreased were stratified by participant characteristics to apply stratum-specific estimates of the magnitude and direction of measurement error to corresponding subsets of participants, selecting at random the participants assigned increasing or decreasing MeHg concentrations. All analyses were repeated using the adjusted MeHg concentrations as outcome variables.

Results
In the three communities combined, 101 participants provided hair samples and diet data Almost all participants (96%; 97/101) reported eating fish or marine mammals in the past 12 months. The data obtained from the fish-focussed FFQ was consistent with input from community planning committees, which identified the summer as the main season during which community members consume fish/whale. However, there was considerable variation by species and community ( Figure 1).   Table 2 shows the five most frequent species consumed ≥ 1 time/week by community and season. The mean number of different species eaten by participants was 3.5 (SD: 1.9; Range: 0-9). The main waterways and sites from which participants reported harvesting fish and whale are shown in Figure 2. On average, participants reported harvesting most of the species they consume from local rivers, followed by the ocean and nearby lakes.  Among participants from all communities combined, the mean concentration of MeHg in hair samples was 0. Mean hair mercury levels (μg/g) ± SD stratified by population characteristics are shown in Tables 3 and 4. No participants had hair mercury levels that exceeded the exposure maximum defined by Health Canada.  Because, data were insufficient for estimation of species-specific effects on hair mercury levels, this analysis was limited to effects of total fish/whale consumption. Table 5 shows  Conversely, the magnitude of this effect was lowest for intake during the winter (β: 0.28; 95%CI: 0.07, 0.50). Tables 6-9 also show the intercepts from these models, representing the expected mean concentration of mercury (μg/g) if all covariates are at their reference level, and corresponding SDs, representing the variation in these values associated with clustering in communities, which reflect residual clustering in each model, suggesting that variation in baseline hair-mercury concentrations across communities is not fully explained by the variables in the model. ∞ Each model included a random intercept for the effect of clustering in communities Φ Adjusted for sex, hair length, use of hair dyes or permanent treatments, the proportion of fish meals usually prepared by c fish/whale consumption frequency in the spring  ∞ Each model included a random intercept for the effect of clustering in communities Φ Model covariates: sex, hair length, use of hair dyes or permanent treatments, the proportion of fish meals usually prepared and fish/whale consumption frequency in the fall  Table 10 shows the distribution of ascertained diet components overall and by community.  Hair length was inversely correlated with MeHg concentration, with 1 cm increases in length corresponding to slight reductions in μg/g of mercury after adjusting for sex, permanent hair treatment use, fish/whale consumption and the proportion of fish/whale meals prepared by cooking (tables 6-9). Multivariable random effects regression yielded evidence of reduced hair mercury concentration among those who reported recent use of permanent hair treatments relative to those with untreated hair (tables 6-9). Figure 4 shows the adjusted effects of consuming different quantities of fish/whale in each season on hair mercury concentration, stratified by use of permanent hair treatments. These graphs show that within categories of fish/whale intake, participants who used permanent hair treatments had lower hair mercury concentrations relative to those who did not use such treatments. Tables 6-9 show SDs and 95% CIs representing the random effect for clustering in communities. Visual comparison of community-specific patterns of fish/whale intake (table 2) revealed the following species as having the most divergent consumption patterns across communities: Beluga Whale (D.leucas), Arctic Grayling (T.arcticas), Chinook Salmon (O.tshawytscha) and Burbot (L.lota). Assessment of the linearity of the relationships between intake frequency and hair MeHg concentration indicated that each of these variables could be modeled as continuous. Table 11 shows the SDs and 95% CIs  provided key input on fish/whale consumption practices in the study population, the development of the fish-focussed FFQ, and the timing of hair sample collection. This led to collection of hair samples following the season of greatest exposure and incorporation of commonly used names for aquatic species, which likely improved participants' ability to provide accurate consumption data. The accuracy of consumption data is evident in the consistency between our results and the existing body of evidence on fish/whale consumption and hair mercury concentration; for example, the strong association between Beluga Whale (D.leucas) consumption and MeHg is expected due to reported MeHg levels in large marine mammals (13,92).

Conclusions
This mercury exposure project revealed that a large proportion of western Arctic Canadians regularly consume a wide range of fish species, as well as Beluga Whale.
Increased fish/whale consumption in each season was associated with increased hair MeHg concentration. Overall, however, hair MeHg concentrations were low, indicating that fish consumption practices among participating residents of western Arctic Canadian communities are not placing them at elevated risk of serious health outcomes associated with mercury exposure.

Declarations
Ethics Approval and Consent to Participate. Informed consent was obtained from all participants by providing them with an information sheet that outlined the study objectives, methods, information to be collected, benefits and potential risks of participation and confidentiality. Following review of this document, each participant was asked to fill out a consent form, confirming they had received enough information about the project and that they agree to participate. Project information sheets and consent forms were reviewed and approved by the Research Ethics Board at the University of Alberta and have been published previously (23).

Availability of Data and Materials. All data collected and created in partnership with
Indigenous communities is considered confidential, sensitive and vulnerable to misappropriation. Data cannot be shared without the expressed permission of the communities who participated in the research. Researchers who are interested in the data can send a proposal to the corresponding author for consideration by community planning committees.    Hair MeHg levels (μg/g) for different categories of fish/whale consumption frequency stratified by use of permanent hair treatments, adjusted for sex, proportion of fish/whale meals usually prepared by cooking and hair length