Skip to main content
Download PDF

Factors associated with blood mercury concentrations and their interactions with three glutathione S-transferase genes (GSTT1, GSTM1, and GSTP1): an exposure assessment study of typically developing Jamaican children

BMC Pediatrics volume 24, Article number: 14 (2024) Cite this article

Abstract

Background

Jamaican soil is abundant in heavy metals including mercury (Hg). Due to availability and ease of access, fish is a traditional dietary component in Jamaica and a significant source of Hg exposure. Mercury is a xenobiotic and known neuro-toxicant that affects children's neurodevelopment. Human glutathione S-transferase (GST) genes, including GSTT1, GSTM1, and GSTP1, affect Hg conjugation and elimination mechanisms.

Methods

In this exposure assessment study we used data from 375 typically developing (TD) 2–8-year-old Jamaican children to explore the association between environmental Hg exposure, GST genes, and their interaction effects on blood Hg concentrations (BHgCs). We used multivariable general linear models (GLMs).

Results

We identified the child’s age, consumption of saltwater fish, canned fish (sardine, mackerel), string beans, grain, and starches (pasta, macaroni, noodles) as the environmental factors significantly associated with BHgCs (all P < 0.05). A significant interaction between consumption of canned fish (sardine, mackerel) and GSTP1 in relation to BHgC using either a co-dominant or recessive genetic model (overall interaction P = 0.01 and P < 0.01, respectively) indicated that consumption of canned fish (sardine, mackerel) was significantly associated with higher mean BHgC only among children with the GSTP1 Ile105Val, Ile/Ile [Ratio of mean Hg (95% CI) = 1.59 (1.09, 2.32), P = 0.02] and Ile/Val [Ratio of mean Hg (95% CI) = 1.46 (1.12, 1.91), P = 0.01] genotypes.

Conclusions

Since this is the first study from Jamaica to report these findings, replication in other populations is recommended.

Peer Review reports

Background

Mercury is a highly toxic omnipresent heavy metal, xenobiotic, and known neuro-toxicant [1,2,3,4,5,6]. There is no known safe limit for human Hg exposure [7]. In the environment, Hg is available in three chemical forms: elemental Hg (eHg), organic Hg (oHg), and inorganic Hg (iHg). Additionally, oHg has three forms methyl Hg (mHg), ethyl Hg (eHg), and phenyl Hg (pHg) [8, 9]. Each of these forms has a different level of toxicity and bioavailability [9]. For example, oHg is more hazardous and toxic to humans than iHg [10]. In the marine environment, iHg is transformed into highly toxic mHg by the bio-methylation process. Methyl Hg then biomagnifies up the food chain and bioaccumulates in fish [10,11,12,13]. The total blood Hg concentration (BHgC) is a valid biomarker for recent mHg exposure [14]. Hg exposure may affect the health of people of all age groups [11], however, children are at specific risk [7] as it may cause harm to their growing central nervous system (CNS) [3]. Moreover, human renal, reproductive, cardiovascular, respiratory, and immune systems may be affected by Hg [7, 11]. Adverse effects on the CNS are particularly crucial since 5–10% of infants born worldwide are affected by neurodevelopmental disorders [15, 16]. A low to moderate early childhood oral mHg exposure may cause subtle nervous system (NS) impairments, [17] and may be associated with reduced attention span, memory, language, and visual-motor skill development [18].

Heavy metals (e.g. Hg) are present in all parts of the biotic and abiotic environment [19]. Most important environmental Hg exposure sources are food of animal (fish) and plant (rice and vegetables) origin, pharmaceuticals and utility products (Hg fever thermometers, skin lightening lotion and creams) [3, 7, 11, 13]. Humans may be exposed to Hg through ingestion, inhalation, transdermal and trans-placental absorptions [7, 20]. Skin lightening or bleaching products use is a public health concern in many countries, including Jamaica, as these products may contain Hg [21]. Therefore, pregnant women and women of childbearing age (18–44 years old) are of special concern regarding Hg exposure through the consumption of fish as well as through the use of Hg containing skin products as they can transfer Hg to their fetuses through the placenta [21]. Although perinatal and pediatric mHg exposure are worrisome, prenatal Hg exposure is also a concern as the developing brain is very sensitive to mHg poisoning [22]. Therefore, both prenatal and early childhood Hg exposure is a critical public health concern for many countries where significantly higher amounts of fish are usually consumed.

Fish is one of the most essential and traditional dietary components for people in Jamaica because of its availability and easy accessibility [23]. According to the Food and Agricultural Organization (FAO) of the United Nations (UN) 2016 data, in Jamaica, the average per capita fish consumption was 27.1 kg/year and significantly higher than the world’s per capita supply of fish for consumption (19.7 kg/year) [24]. Ricketts et al. (2020) found the Hg levels in commonly consumed fish samples from Jamaica were between 0.09–0.32 μg/g, which were much lower than the acceptable limit of Hg level (1 μg/g) in fish set by the Food and Drug Administration (FDA) [23]. Therefore, the risk of high level Hg exposure through fish consumption was low in Jamaica [23]. Hence, there is a possibility that a significantly higher average fish consumption is responsible for the high BHgC of Jamaican children who consume specific types of fish. Additionally, geochemical investigations showed Jamaican soil was enriched with several heavy metals, including Hg (with a mean of 221 µg/kg to a maximum of 830 µg/kg), and in some areas it was higher than the world average soil Hg levels of 60 µg/kg [25,26,27]. Hence, there is a possibility that crops, vegetables, and fruits grown in Jamaica may contain high levels of Hg. Exposure to Hg through high per capita fish consumption and other Hg contaminated foods grown in Jamaica may pose adverse health outcomes in Jamaican children.

The glutathione S-transferase (GST) superfamily includes six genes of which GSTT1, GSTM1, and GSTP1 are well known for their vital roles in excretory process and detoxification of xenobiotics (e.g. Hg) [28]. Several candidate genes are linked to the glutathione (GSH) detoxification system, including glutathione peroxidase and GST enzymes [4]. Glutathione has a crucial role in Hg metabolism [29]. Methyl Hg is excreted into the bile as GSH conjugates, and it is removed from the body with feces [29]. Reduced GSH production results in less excretion of Hg in the bile, which results in Hg retention [29]. GSTs are actively engaged in Hg conjugation and protect cells from oxidative stress and toxic chemicals [30, 31]. Many GST genes, including GSTT1, GSTM1, and GSTP1, are highly polymorphic [31]. GSTT1 and GSTM1 deletion polymorphisms have epistatic effects, which is a cause for the enhanced susceptibility to Hg exposure [4]. Moreover, null alleles of GSTT1 and GSTM1 result in a lack of enzyme activity [32], which causes slower Hg detoxification [33]. All this evidence suggests that genetic variations may provide an explanation for differences in BHgC in among populations.

In collaboration with faculty members at the University of the West Indies (UWI), Mona campus, in Jamaica, our research team at the University of Texas Health Science Center at Houston (UTHealth) has been examining the additive and interactive associations of six heavy metals, including Hg and GST genes (GSTT1, GSTM1 and GSTP1), among children with and without autism spectrum disorder (ASD) in Jamaica since 2009. Using data from 266, 1:1 age- and sex-matched (2–8 years) ASD cases and typically developing (TD) controls, we previously investigated interactive roles of three GST genes (GSTT1, GSTM1, and GSTP1) and ASD status in relation to BHgCs of Jamaican children. In GLMs after adjusting for child’s age, socioeconomic status (SES), consumption of leafy vegetables, fried plantain, canned fish, and the interaction between GSTP1 and GSTT1, we found significant interactions between GSTP1 and children’s ASD status in relation to BHgC using either a co-dominant or dominant genetic model for GSTP1 (P < 0.001, P = 0.007, respectively). This study found that in TD control children with the GSTP1 Ile/Val genotype compared with ASD children the geometric mean (GM) BHgCs were significantly higher than in children with the Ile/Ile genotype (0.72 vs. 0.49 µg/L, P = 0.03) or the Val/Val genotype (0.72 vs. 0.51 µg/L, P = 0.04). These findings suggest a possible role of GSTP1 in the detoxification of Hg that may affect the adverse health outcomes in relation to Hg exposure among the Jamaican TD children [34]. In another ERAJ study, we found a statistically significant relationship (all P < 0.05) between specific types of fish and seafood (i.e., saltwater fish, sardine, or mackerel fish) consumption and higher GM BHgCs in Jamaican children regardless of their ASD status; for example children who consumed seafood > 6 times per week had GM BHgC among ASD cases and controls of 1.57 µg/L and 1.31 µg/L, respectively and the GM BHgC of Jamaican children with ASD was significantly higher (0.75 μg/L) among those who ate specific types of fish than those who did not eat those fish (0.15 μg/L) [35]. However, the association between Hg exposure and its elimination and excretory mechanism influenced by the polymorphisms of GST genes and the effect of their interaction with environmental factors on BHgC are unknown. In this present study, we assessed the relationship of the most common environmental sources of Hg exposure and sociodemographic factors, as well as the possible interactions of these factors with the genotypes for the three GST genes (GSTT1, GSTM1, GSTP1), in relation to BHgCs in Jamaican TD children.

Materials and methods

General description and the study population

In this exposure assessment study, we used data from 375 TD Jamaican children (2–8 years old), who were enrolled as controls between December 2009 and September 2017, in the Epidemiological Research on Autism in Jamaica (ERAJ) and ERAJ Phase-2 (ERAJ-2), that are age-and sex-matched case control studies of ASD. Inclusion criteria for TD children in ERAJ studies included 1) the child must be born in Jamaica and 2) should be between 2–8 years old at the time of enrollment. Age and sex-matched TD controls were recruited and enrolled from schools, well-child clinics, and community churches, mostly from Kingston Parish of Jamaica [36]. The matching criteria for this study required the control children to be the same sex and to be within six months of the age of their matched ASD cases. The procedure for enrolling TD controls has been explained in detail elsewhere [37, 38]. To exclude the possibility of any developmental disorders in TD children, the life-time form of the Social Communication Questionnaire (SCQ) was administered to the parents or guardians [39, 40]. Children with SCQ scores ≤ 6 were considered for enrollment as TD controls in this study. The cut-off value of 6 was one standard deviation higher than the mean SCQ score of TD school children [41]. We also administered a SES questionnaire to the parents or guardians to collect socio-demographic and economic status information regarding children and their parents [34, 42]. A food frequency questionnaire was used to assess children’s possible Hg and other heavy metals exposure from food consumption [37]. Questions were asked about how often the child ate specific food items per week, focusing on the types and frequency of fruits, vegetables, grains, and fish and seafood consumption. For example, types of fish and seafood were classified into the following categories: saltwater fish, freshwater fish (pond fish, tilapia), canned fish (sardine or mackerel), canned tuna, salted fish (pickled mackerel), shell-fish (lobsters, crabs), and shrimp [42]. This food frequency data represented the typical current consumption of food items by children at the time of data collection [34, 35, 39]. The different types of food consumption were considered as binary categorical variables (consumed vs. never consumed) during data analysis. Blood samples (about 4–5 mL of whole blood) were collected from every child at the end of the interview and other study procedures to evaluate the exposure to heavy metals, including Hg, and to determine children’s GST genotypes.

Assessment of Hg exposures

The BHgC analyses were conducted at the Trace Metals Lab, a Centers for Disease Control and Prevention (CDC) certified lab at the MDHHS in Lansing, Michigan, United States. The technology for metal concentrations detection in blood samples has altered over the past decade. For this reason, the MDHHS reported different limits of detection (LoD) for Hg in phases 1 and 2 of the ERAJ studies, which were 0.3 µg/L and 0.25 µg/L, respectively [34]. For quality control (QC), bovine blood spiked with known quantities of Hg was used as a control for Hg levels in the blood samples. All blood samples were diluted and analyzed on a PerkinElmer Elan DRC II inductively coupled plasma mass spectrometer (PerkinElmer, Waltham, MA, USA).

Genetic analysis

For genetic analysis, regions of the GSTM1 and GSTT1 genes were amplified in two independent TaqMan Copy Number Assay reactions, GSTM1 Assay ID: Hs02575461_cn and GSTT1 Assay ID: Hs00010004_cn (www.thermofisher.com), to identify insertion or deletion polymorphisms. GSTM1 and GSTT1 homozygous deletions were coded as DD, and the presence of an insertion was coded as I* (I/I or I/D). Assessment of the GSTP1 Ile105Val polymorphism (rs1695) was carried out using the TaqMan Drug Metabolism SNP genotyping assay C_3217198_20. All three assays have been previously described in detail [34, 39].

Statistical analysis

Descriptive analyses were conducted to describe the distributions of demographic and SES characteristics of TD children and their parents. The BHgCs were not normally distributed in this dataset so we transformed this data using the natural logarithm (ln). Then the mean of the log transformed BHgC was changed to the original scale (μg/L) using an exponential function, herein called the GM of BHgC, that is (Exp [Mean (ln BHgC) = GM BHgC]). This study found 9.6% of BHgCs were below LoD or censored. We imputed BHgC data by dividing the LoD by the square root of two (LoD/√2) to include all the BHgCs below LoD in this data analysis [34, 43, 44].

The genotyping test for GSTT1 and GSTM1 does not differentiate between homozygotes (I/I) and heterozygotes (I/D); hence, only the recessive genetic model was taken into consideration for these two genes using binary variables: I* (I/I or I/D) and DD (null). For the GSTP1 Ile105Val polymorphism, there are three genotypes: Ile/Val, Val/Val, and Ile/Ile. Analysis is possible for all three GSTP1 genotypes. To assess the GSTP1 genotypes, we used three genetic models, including the dominant model [Val* (Ile/Val or Val/Val) vs. Ile/Ile], co-dominant model [Ile/Ile, Ile/Val, and Val/Val], and recessive model [Val/Val vs. Ile* (Ile/Ile or Ile/Val)]. We used the Chi-square test based on available data for Jamaican TD children to determine whether the GSTP1 polymorphism satisfies the Hardy–Weinberg equilibrium expectations. This was done due to the presence of three genotypes (Ile/Ile, Ile/Val, and Val/Val) for GSTP1.

We used univariable GLMs to investigate the possible additive association of various environmental exposures, three GST genes, socio-demographic characteristics, and the consumption of different types of food with log transformed BHgC as the dependent variable.

We used multivariable GLMs (both additive and interactive) to evaluate the potential gene-environment interactions between the three GST genes (GSTT1, GSTM1, and GSTP1) and various environmental factors with log transformed BHgC. In the multivariable GLMs, we only included variables that were significant in the univariable GLMs. Consequently, additive and interactive multivariable models were fitted to evaluate the relationship of the environmental factors and GST genes with log transformed BHgC, and in the final multivariable interactive GLM we only kept variables that were statistically significant. Additionally, to decrease the potential effects of multi-collinearity, we looked at pairwise correlations between all pairs of individual environmental factors that were statistically significant in the additive models. When the model became unstable by adding both correlated variables, we chose one of the variables to remain in the model. The ratios of mean Hg and 95% confidence interval (CI) for the association between the children's exposure to environmental factors and BHgC by the children's genotypes for GST genes were calculated when we identified significant gene-environment interactions using the SAS ESTIMATE statement [45]. The same method was used to calculate the ratios of the mean Hg and 95% CI for the association between the children's GST genotypes and BHgC by exposure to environmental factors. Ratios of the mean Hg were calculated as shown below using the ratios of mean Ln_Hg in children who consumed saltwater fish vs. those who did not consume saltwater fish as an example,

$$\begin{array}{c}Exp (Ln Hg) = {\beta }_{0} + {\beta }_{1} * saltwater\, fish \,consumption\\ {\beta }_{1}\hat{}= Ln\_Hg (consumed \,saltwater \,fish) - Ln\_Hg (did \,not \,consume \,saltwater \,fish)\\ = Ln [Hg (consumed \,saltwater \,fish) / Hg (did \,not \,consume \,saltwater \,fish)]\\ = EXP [Ln \{Hg (consumed \,saltwater \,fish) / Hg (did \,not \,consume \,saltwater \,fish)\}] \\ = Ratio \,of \,mean \,Hg \left(children \,consumed \,saltwater \,fish \,vs. \,those \,who \,did \,not \,consume \,saltwater \,fish\right).\end{array}$$

The significance level for all statistical tests was set at 0.05, and without corrections for multiple comparisons. All data analyses were performed using SAS 9.4 software [46].

Results

A description of the socio-demographic features of 375 TD children and their parents, as well as children’s GST genotypes are provided in Table 1. As expected, 81.9% of TD children were male as they were matched by sex of the participating children with ASD in the ERAJ studies. About 25% of children were 72 months or older at the time of enrollment. Most children (97.3%) were Afro-Caribbean, and nearly 62% were born in Kingston parish of Jamaica. The null (DD) genotype frequency for GSTT1 and GSTM1 was 25.8% and 24.9% among TD children, respectively. The frequencies of the three genotypes for GSTP1 (Ile/Ile, Ile/Val, and Val/Val) in TD children were 26.7%, 50.7%, and 22.6%, respectively, which were in agreement with Hardy–Weinberg equilibrium expectations (P = 0.77). At the time of the child’s birth, 11.7% of mothers were aged 35 years or more, and 44.1% of children had at least one parent with an education level beyond high school. Higher levels of SES were measured by the ownership of a car by the parents or the family of the children. At the time of enrollment 40.3% of the families had higher SES.

Table 1 Sociodemographic characteristics of Jamaican TD children and their parents and children’s GST genotypes (n = 375)
Full size table

In the univariable GLM that assessed the association of the children’s sociodemographic characteristics, GST genotypes and exposure to various dietary and environmental factors with BHgC, we found a significant association between the children’s age and BHgC. Specifically, children aged 72 months or more had higher mean BHgC than children aged less than 72 months [Ratio of mean Hg (95% CI) = 1.25 (1.05, 1.48), P = 0.01]. When we assessed the association between seafood consumption and BHgCs, we found children who consumed saltwater fish and canned fish (sardine, mackerel) had higher mean BHgCs than children who did not consume those fish [Ratio of mean Hg (95% CI) = 1.40 (1.19, 1.64), P < 0.01; and 1.37 (1.12, 1.68), P < 0.01, respectively]. Similar results were found in relation to consumption of canned tuna and shrimp. Specifically, children who consumed canned tuna and shrimp had higher mean BHgCs than children who did not consume canned tuna and shrimp [Ratio of mean Hg (95% CI) = 1.21 (1.03, 1.41), P = 0.02; and 1.22 (1.01, 1.49), P < 0.05, respectively]. For beans consumption, we found children who ate peas (red peas and gungo peas) and broad beans had higher mean BHgCs compared to those who did not eat these types of food [Ratio of mean Hg (95% CI) = 1.28 (1.06, 1.54), P = 0.01; and 1.17 (1.01, 1.36), P = 0.04, respectively]. For fruits and vegetable consumption, we found significantly higher mean BHgCs in TD children who consumed leafy vegetables (lettuce) [Ratio of mean Hg (95% CI) = 1.25 (1.07, 1.45), P = 0.01], legumes (string beans) [Ratio of mean Hg (95% CI) = 1.31 (1.13, 1.52), P < 0.01], and fruit (avocado]) [Ratio of mean Hg (95% CI) = 1.21 (1.04, 1.41), P = 0.02] as compared to children who did not consume these fruits and vegetables. There were no significant associations between the BHgC and children’s GST (GSTT1, GSTM1, and GSTP1) genotypes (all P > 0.05) (see Table 2).

Table 2 Associations of environmental factors and children’s GST genotypes with blood Hg concentrations based on univariable General Linear Models (GLMs) (n = 375)
Full size table

Additional details regarding the descriptive analysis of distribution of BHgC before any log transformation (e.g., median and the interquartile range) are also shown as part of the supplementary materials (see Table S1).

In Table 3, we also assessed the relationship between environmental exposures and BHgC by children’s genotype for GST genes using unadjusted interactive multivariable GLMs that included the interactions between GST genes and environmental exposures in relation to BHgC. Using a co-dominant genetic model for GSTP1, we identified a significant interaction between the children’s birth parish and the children’s genotypes for GSTP1 in relation to BHgCs (overall interaction P = 0.04). Specifically, in children with the Ile/Val genotype, TD children who were born in Kingston parish had higher mean BHgC compared to those children who were born in other parishes [Ratio of mean Hg (95% CI) = 1.28 (1.03, 1.58), P = 0.02]. There was no statistically significant association present between children’s birth parish and BHgC among children with Ile/Ile genotype [Ratio of mean Hg in children who were born in Kingston parish vs. those born in other parishes (95% CI) = 0.94 (0.69 -1.28), P = 0.69] and Val/Val genotype [Ratio of mean Hg in children who were born in Kingston parish vs. those born in other parishes (95% CI) = 0.81 (0.59, 1.12), P = 0.20].

Table 3 Associations between children’s exposure to environmental factors and BHgC by children’s genotypes for GST genes based on the multivariable General Linear Models that include the interaction between GST genes and the primary environmental exposure (n = 375)
Full size table

In a recessive genetic model for GSTT1, we identified a significant interaction between exposure to traffic and children’s genotype for GSTT1 in relation to BHgCs (overall interaction.

P = 0.046). Specifically, among the TD children with I* (I/I or I/D) genotype, those who lived near a high traffic road had lower mean BHgC than those who did not live near a high traffic road [Ratio of mean Hg (95% CI) = 0.83 (0.69, 0.99), P = 0.04]. Whereas among children with DD (null) genotype, no statistically significant association was found between the exposure to traffic and BHgC [Ratio of mean Hg in children who lived near a high traffic road vs. those who did not (95% CI) = 1.19 (0.87, 1.63), P = 0.27].

Additionally, using a dominant genetic model for GSTP1, we identified a significant interaction between exposure to traffic and the children’s genotypes for GSTP1 in relation to BHgCs (overall interaction P = 0.01). Specifically, TD children with Val* (Val/Val or Ile/Val) genotypes and lived near a high traffic road had lower mean BHgC than those who did not live near a high traffic road [Ratio of mean Hg (95% CI) = 0.79 (0.67, 0.95) P = 0.01]. On the other hand, in children with the Ile/Ile genotype, no statistically significant association was found between exposure to traffic and BHgCs [Ratio of mean Hg in children who lived near a high traffic road vs. those who did not (95% CI) = 1.31 (0.96, 1.77), P = 0.08]. Similarly, under a co-dominant genetic model for GSTP1, we identified a significant interaction between exposure to traffic and the children’s genotypes for GSTP1 in relation to BHgC (overall interaction P = 0.02). Specifically, among children with Ile/Val genotype, the mean BHgC was lower for children who lived near a high traffic road than for children who did not live near a high traffic road [Ratio of mean Hg (95% CI) = 0.80 (0.65, 0.98) P = 0.03]. However, there was no statistically significant association between exposure to traffic and BHgCs in children with Ile/Ile [Ratio of mean Hg in children who lived near a high traffic road vs. those who did not (95% CI) = 1.31 (0.97, 1.76), P = 0.08] and Val/Val [Ratio of mean Hg in children who lived near a high traffic road vs. those who did not (95% CI) = 0.75 (0.54, 1.04), P = 0.08] genotypes.

Additionally, using a dominant GSTP1 genetic model, we found a significant interaction between parental education level and the children’s genotype for GSTP1 in relation to BHgC (overall interaction P = 0.03). Specifically, children with Val* (Ile/Val or Val/Val) genotypes who had at least one parent with education beyond high school had lower mean BHgC than children who had both parents with up to high school education [Ratio of mean Hg (95% CI) = 0.79 (0.66, 0.95), P = 0.01]. However, in children with Ile/Ile genotypes, no statistically significant association was found between parental education level and BHgC [Ratio of mean Hg in children who had at least one parent with education beyond high school vs. children who had both parents with up to high school education (95% CI) = 1.19 (0.88, 1.60), P = 0.27].

For ‘grain and starches consumption’, using a recessive genetic model for GSTT1, we found a significant interaction between consumption of white bread and the children’s genotype for GSTT1 in relation to BHgC (overall interaction P = 0.02). Specifically, in TD children with DD (null) genotypes, those who ate white bread had lower mean BHgC than children who did not eat white bread [Ratio of mean Hg (95% CI) = 0.63 (0.45, 0.88), P = 0.01]. However, in children with I* (I/I or I/D) genotypes, there was no significant association between white bread consumption and BHgC [Ratio of mean Hg in children who ate white bread vs. those who did not (95% CI) = 0.98 (0.81, 1.18), P = 0.82].

Moreover, using a recessive genetic model for GSTM1, we identified a significant interaction between the consumption of fried dumplings and the children’s genotype for GSTM1 in relation to BHgC (overall interaction P = 0.01). Specifically, in children with DD (null) genotypes, those who consumed fried dumplings had lower mean BHgC than children who did not consume fried dumplings [Ratio of mean Hg (95% CI) = 0.62 (0.43, 0.91), P = 0.01]. However, in children with I* (I/I or I/D) genotypes, there was no statistically significant association between fried dumpling consumption and BHgC [Ratio of mean Hg in children who consumed fried dumplings vs. those who did not (95% CI) = 1.09 (0.87, 1.37), P = 0.43].

For ‘fruits and vegetable’ consumption, in the recessive genetic model for GSTT1, we found a significant interaction between consumption of green banana and children’s GSTT1 genotypes in relation to BHgCs (overall interaction P = 0.03). Moreover, using a recessive genetic model for GSTM1, we found a significant interaction between consumption of carrots and a marginally significant interaction between the consumption of string beans and the children’s GSTM1 genotype in relation to BHgCs (overall interaction, P = 0.04 and P = 0.08, respectively). Specifically, among children with I* (I/I or I/D) genotypes for GSTM1, children who consumed string beans had higher mean BHgC compared to children who did not consume string beans [Ratio of mean Hg (95% CI) = 1.36 (1.15, 1.63), P < 0.01]. However, in children with the DD (null) GSTM1 genotypes there was no significant association between string bean consumption and BHgC [Ratio of mean Hg in children who consumed string beans vs. those who did not [Ratio of mean Hg (95% CI) = 1.00 (0.74, 1.35), P = 1.00].

For ‘seafood consumption’ using a recessive and co-dominant genetic model for GSTP1, we identified significant interactions between children’s genotypes for GSTP1 and consumption of canned fish (sardine, mackerel) in relation to BHgC (overall interaction P = 0.01 and P = 0.02, respectively). Specifically, in children with Ile* (Ile/Ile or Ile/Val) genotypes for the recessive genetic model of GSTP1, those who consumed canned fish (sardine, mackerel) had higher mean BHgC than those who did not consume those canned fish [Ratio of mean Hg (95% CI) = 1.56 (1.24, 1.95), P < 0.01]. However, in children with Val/Val genotype under the GSTP1 recessive genetic model, there was no significant association found between the consumption of canned fish (sardines, mackerel) and BHgC [Ratio of mean Hg in children who consumed canned fish (sardines, mackerel) vs. those who did not (95% CI) = 0.76 (0.48, 1.19) P = 0.22]. Moreover, using the co-dominant genetic model for GSTP1, in children with Ile/Ile and Ile/Val genotypes, those who consumed canned fish (sardine, mackerel) had higher mean BHgC than children who did not consume these canned fish [Ratio of mean Hg (95% CI) = 1.68 (1.14, 2.48), P = 0.01, and 1.51 (1.15, 1.98), P < 0.01, respectively]. However, in children with the Val/Val genotype for the co-dominant genetic model for GSTP1, there was no significant association between consumption of canned fish (sardine, mackerel) and BHgC [Ratio of mean Hg in children who consumed canned fish (sardine, mackerel) vs. those who did not (95% CI) = 0.76 (0.48, 1.19), P = 0.22].

Additional details regarding the unadjusted association between the children’s genotype for GST genes and BHgC by environmental factors are also shown as part of the supplementary materials (see Table S2).

In Table 4, the adjusted associations between TD children’s exposure to environmental factors and their interaction with GST genes in relation to BHgC are described. In the final interactive multivariable models, we found a significant interaction between GSTP1 genotypes and the consumption of canned fish (sardine, mackerel) in relation to BHgC using either a co-dominant or recessive genetic model (overall interaction P = 0.01 and P < 0.01, respectively). In addition, we identified the child’s age, as well as the consumption of saltwater fish, grains, and starches (pasta, macaroni, and noodles), and string beans as the other environmental factors significantly associated with BHgC (all P ≤ 0.04 for both adjusted models). Specifically, using the codominant genetic model for GSTP1, we found that children with Ile/Ile and Ile/Val genotypes who consumed canned fish (sardine, mackerel) had higher mean BHgCs compared to children who did not consume these canned fish [Ratio of mean Hg (95% CI) = 1.59 (1.09, 2.32), P = 0.02, and 1.46 (1.12, 1.91), P = 0.01, respectively].

Table 4 Adjusted associations between children’s exposure to environmental factors and BHgC by children’s genotypes for GST genes based on General Linear Models that include the interactions between GST genes and the main environmental exposures (n = 375)
Full size table

However, in children with the Val/Val genotype under the co-dominant genetic model for GSTP1, there was no significant association between the consumption of canned fish (sardine, mackerel) and BHgC [Ratio of mean Hg in children who consumed canned fish (sardine, mackerel) vs. those who did not (95% CI) = 0.72 (0.47, 1.11), P = 0.14]. In the co-dominant genetic model, we also found that in children aged 72 months or older, the mean BHgC was higher than in children aged less than 72 months [Ratio of mean Hg (95% CI) = 1.21 (1.03, 1.42), P = 0.02]. Moreover, TD children who consumed saltwater fish and string beans had higher mean BHgCs compared to those who did not consume those foods [Ratio of mean Hg (95% CI) = 1.35 (1.15, 1.59), P < 0.01, and 1.17 (0.58, 0.89), P = 0.04, respectively]. On the other hand, children who consumed grain and starches (pasta, macaroni, and noodles) had lower mean BHgC than children who did not consume those foods, [Ratio of mean Hg (95% CI) = 0.72 (0.58, 0.89), P < 0.01].

We also identified similar findings using the recessive genetic model for GSTP1. We found that children with Ile* (Ile/Ile or Ile/Val) genotypes who consumed canned fish (sardine, mackerel) had higher mean BHgC than children who did not consume those canned fish [Ratio of mean Hg (95% CI) = 1.50 (1.20, 1.87), P < 0.01]. However, in children with the Val/Val genotype under the recessive genetic model for GSTP1, there was no statistically significant association between consumption of canned fish (sardine and mackerel) and BHgCs [Ratio of mean Hg in children who consumed canned fish (sardine, mackerel) vs. those who did not (95% CI) = 0.72 (0.46, 1.12), P = 0.14].

More details about the adjusted associations between the children’s GST genotypes and BHgC are shown as part of the supplementary materials (see Table S3).

Discussion

The findings from this study suggest that the consumption of saltwater fish and string beans has resulted in a 35% and 17% higher mean BHgC among Jamaican children who consumed those foods compared to those who did not consume those foods, respectively. On the other hand, the consumption of grain and starches (pasta, macaroni, and noodles) was associated with a 28% lower mean BHgC in Jamaican children who consumed those foods compared to those who did not consume those foods. In addition, we found that children aged 72 months and older had a 21% higher mean BHgC than the group of younger children. The association between the consumption of canned fish (sardine, mackerel) and BHgCs varied by children’s GSTP1 genotypes, using either a recessive or a co-dominant genetic model (overall interaction.

P < 0.01 and P = 0.01, respectively). The consumption of canned fish (sardine, mackerel) was significantly associated with 59% and 46% higher mean BHgCs only among children with GSTP1 Ile/Ile, and Ile/Val genotypes respectively. We discuss the main study findings separately in the following sections.

Role of child’s age in BHgC

When comparing our findings with the GM BHgC of 7–11-year-old children in other countries, we observed that the GM BHgC of these TD Jamaican children is near the level reported in children in some European countries, but compared with Ecuador and China is 3–4 times lower (see Table S4).

Our study also showed that children’s age is significantly associated with GM BHgC among Jamaican TD children, which is consistent with a study conducted in the United States that showed the GM BHgC among children aged 1–5 years was 0.34 µg/L [95% CI, (0.30, 0.39) µg/L] and in women aged 16–49 years was 1.02 µg/L [95% CI, (0.85, 1.20) µg/L] [14]. Another study in New Zealand provided similar evidence of increased BHgC with increasing age. This study showed that the GM BHgCs among children and adults were 0.86 µg/L and 1.65 µg/L respectively [47].

All of this evidence supports the present study findings of 1.21 times higher mean BHgC among TD children aged 72 months and older compared to children in the younger age group (P = 0.02). The reason behind this phenomenon may be the slow clearance of Hg from the human body, which causes the BHgC to gradually increase as children get older [48]. Confirmation and dissemination of this study finding that mean BHgC may gradually increase with child’s age is warranted for the parents in countries where large amounts of fish and seafood are consumed like Jamaica in order to help protect their children from neurodevelopmental disorders by reducing and limiting their children’s Hg exposure through more suitable food choices.

Association of seafood consumption and BHgCs

Our study finding of 1.35 times higher mean BHgC in children who consumed saltwater fish compared with children who did not consume saltwater fish is consistent with a study conducted in New Zealand that demonstrated fish consumption three or more times per week was associated with 2.7 times higher total BHgC in children when compared to eating fish less than once per week [47]. Moreover, another study in China showed children who consumed saltwater fish 4 to 6 times per week had a significantly higher odds ratio for BHgCs compared to children who ate saltwater fish less than or equal to 1–3 times per month in both adjusted and unadjusted models (P < 0.01) [49]. A previous study in Jamaica demonstrated the median of BHgC for TD children (266 children) when adjusted for the frequency of seafood consumption, maternal age, and parental education was 0.55 [39].

Fish and seafood are good sources of low-fat protein [50], and rich in various necessary vitamins and minerals [50]. For example, selenium found in fish has a crucial role in fetal brain development, growth, thyroid hormone metabolism, calcium regulation, and prevention or reversal of oxidative damage [23, 51]. Fish and seafood are also vital sources of nutrients such as essential fatty acids, including omega-3 and omega-6 fatty acids [50], eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) [52, 53]. These essential fatty acids have been associated with reduced risk of cardiovascular disease, heart attack [23], and coronary deaths [52]. Additionally, these essential fatty acids are beneficial for early neurodevelopment [52, 54], and also have been implicated in preventing some specific types of cancers [55]. However, in addition to the enormous health benefits of fish and seafood consumption, there are also risks of exposure to chemical contaminants such as heavy metals [23]. Fish is the primary source of Hg exposure, especially mHg for many populations [35, 53, 56,57,58]. For example, a study conducted in Taiwan showed that saltwater fish was the primary dietary exposure for mHg among women of childbearing age [53]. A report titled “Seafood Choices: Balancing Risks and Benefits” by the US Institute of Medicine recommends that although fish and seafood can replace other protein sources with greater levels of saturated fat, it is still important for consumers to choose food carefully that will reduce or limit their exposure to pollutants like mHg that are present in fish and seafood [59].

It was mentioned previously that fish is the traditional and most essential dietary component for people of the Caribbean island nation, Jamaica, where the average per capita fish consumption (27.1 kg/year) [60] was higher than the world’s average per capita food fish supply (19.7 kg/year) [24]. As a result, information about exposure data for toxic heavy metals (e.g., Hg) through the higher average amount of fish consumption would be helpful for the Jamaican public health authorities to generate to increase awareness about Hg exposure in their country. Based on the previous study findings in Jamaica, the policies of Jamaica's Ministry of Health and Wellness and Ministry of Education and Youth, in consultation with the Planning Institute of Jamaica, were altered, and children are no longer served canned fish (sardine or mackerel) in Jamaican public schools [34].

These present study findings will contribute additional knowledge on fish and seafood consumption and their effects on BHgC of Jamaican children who are one of the most vulnerable populations for developing adverse health outcomes, specifically neurodevelopmental disorders from Hg exposure.

Consumption of fruit and vegetables and BHgCs

Mercury concentrations are high in the surface soil in Jamaica compared to other countries, with a mean of 221 µg/kg to a maximum of 830 µg/kg [25, 27, 61]. For example, the guideline limits for agricultural land use with Hg levels in the soil of Canada and Denmark are 6.6 µg/kg and 3 µg/kg, respectively [25]. Moreover, there is a possibility of Hg contamination of surface water in Jamaica due to anthropogenic activities and other industrial activities such as bauxite processing [61]. Previous studies provided evidence that plant roots absorb Hg from the soil and plant leaves absorb Hg from the air [62]. Therefore, consuming vegetables grown in Jamaican soil polluted with organic Hg (mainly mHg) may cause significant risks to human health. Haixin Yu et al. (2018) have reported in their meta-analysis that beans, such as string beans or long beans, absorb Hg from the soil, although the bio-concentration factor (BCF) is lower than for other vegetables such as green peppers and spinach [62]. These findings are consistent with this present study findings that children who consumed string beans had 1.17 times higher mean BHgC than children who did not consume string beans (P = 0.04). This finding suggests the possibility that consumption of string beans and other crops, fruits and vegetables grown in highly Hg-contaminated Jamaican soil may pose an increased risk of higher mean BHgC in Jamaican children. So, further study of Hg concentrations of different crops grown in Jamaica and their possible effects on BHgC of Jamaican children is warranted to make parents aware about proper food choices for their children to lower their Hg exposure through food.

Role of GST genes in BHgCs of Jamaican children with and without consumption of canned fish (sardine, mackerel)

One of the distinctive features of this present study is the availability of genetic data on Jamaican children. Data are available for children’s genotypes for the GSTT1, GSTM1, and GSTP1 genes. Detailed description of the distribution of these three GST genotypes for Jamaica and a few other countries are provided as part of the supplementary materials (see Table S5). While investigating the possible interactive effects of GSTP1 genotypes and the consumption of canned fish (sardine, mackerel) in relation to BHgCs in Jamaican children after controlling for environmental factors such as consumption of saltwater fish, string beans, and consumption of grain and starches (pasta, macaroni, and noodles), as well as the children’s age we found that the GSTP1 gene could be an effect modifier for the association between the consumption of canned fish (sardine, mackerel) and BHgCs when using either a co-dominant or recessive genetic model.

Though the detoxification efficiency or capacity has not been examined or tested in this study, the association we found between the GSTP1 Ile105Val polymorphism and a lower BHgC suggests a better Hg detoxification capacity among children having two Val alleles (Val/Val) for the GSTP1 Ile105Val polymorphism who consumed canned fish (sardine, mackerel). This finding is consistent with the finding of the previous study by Rahbar et al., (2021) in Jamaica. That study showed that under the co-dominant genetic model for GSTP1, TD children with Val/Val genotypes had a lower GM BHgC than children with Ile/Val genotypes (0.51 µg/L vs. 0.72 µg/L P = 0.04) [34].

Currently, there is a knowledge gap in available information on the effect of genetic predisposition on metal toxicokinetics (i.e. how the body metabolizes and eliminates a toxic metal) in the human body [63, 64]. However, the evidence that suggests specific genetic variants affect Hg toxicokinetics is growing [63, 64]. For example, cytochrome P450 monooxygenases (CYP450) catalyze the oxidation and metabolism of a large number of xenobiotics including Hg. CYP450 enzymes evolved as the primary defense against xenobiotics (Hg) and during this process are also responsible for the bioactivation of toxicants to a more reactive intermediate [65]. Furthermore, GSH plays a crucial role in Hg metabolism [29]. Experiments have shown that mHg exposure could cause depleted concentrations of antioxidant molecules such as tripeptide glutathione (GSH) and reduced activities of important antioxidant enzymes, such as catalase (CAT), glutathione-peroxidase (GPx), and superoxide dismutase (SOD) that might lead to oxidative damage in lipids, proteins, and DNA [64]. Some genes are related to the GSH detoxification systems, namely genes coding for enzymes involved in glutathione metabolism as well as GPx and GST enzymes. The intracellular binding reaction with GSH is catalyzed by the various GST isoenzymes and leads to the formation of stable GSH-metal (e.g., Hg) conjugates, which are transported out of the cells and excreted from the human body via feces and urine [66, 67]. Increased Hg retention in blood and plasma, as well as increased urinary excretion, may be linked to GSTP1 polymorphisms with reduced or no conjugation capability [29]. A study conducted in Sweden showed that study participants who carried at least one Val allele for the GSTP1 Ile105Val polymorphism had lower Hg concentrations in erythrocytes [68]. To the best of our knowledge, we are the first to report an interactive association between GSTP1 Ile105Val and the consumption of canned fish (sardine, mackerel) in relation to BHgC among Jamaican, 2–8 years old TD children. Therefore, the validation of this finding in other populations is warranted.

Limitations

We acknowledge several limitations to this study. The first limitation is that this study’s participants were the TD control children from the ERAJ studies selected to match the ASD cases by age and sex and were more likely to be enrolled from the Kingston parish of Jamaica. As a result, around 82% and 62% of the study participants (TD children) were male and born in Kingston parish of Jamaica, respectively. However, one of the most consistent findings from previous studies of ASD is the 4:1 male to female ratio [69]. Therefore, participants in this study may not represent a random sample from the general population of all children in Jamaica. As a result, the findings regarding BHgCs described here may not be generalizable to all Jamaican children.

Secondly, the BHgCs data from the ERAJ studies did not distinguish between oHg and iHg exposures. Therefore, this study does not provide a complete overview of different sources of Hg exposure.

Moreover, although maternal exposure to Hg during pregnancy through fish consumption and using Hg containing skin bleaching and whitening products are also important sources of Hg exposure for their children, we did not have any data regarding these kinds of exposures.

In addition, since data on the quantity of food items consumed was not available in this study, we used a binary (consumed vs. never consumed) approach for food exposures.

Furthermore, fish consumption is a crucial source of Hg exposure in Jamaica because of the presence of significant amounts of fish in the traditional Jamaican diet. Although we used a standard and culturally appropriate food frequency questionnaire to measure the frequency of fish and seafood consumption, the questionnaire did not separate the frequency of different types of fish and seafood consumption. This limits the assessment of individual associations of different kinds of fish or seafood consumption with BHgC in Jamaican TD children. For example, we have information on how many sardines or mackerel fish servings children consumed per week, but this study did not assess the frequency of sardine and mackerel consumption individually. This is crucial because, according to Silbernagel et al. (2011), mackerel is one of the highest mHg contaminated fish while sardines are considered to have lower mHg concentrations [70].

In univariable models, we found that the consumption of peas (red peas, gungo peas), broad beans, leafy vegetables (lettuce), and fruit (avocado) were positively associated with BHgCs. However, due to the presence of associated independent variables in the regression models, these dietary items were eliminated from the multivariable analysis to reduce the possibility of multicollinearity. Fruits and vegetables are significant sources of nutritional factors, including vitamins, minerals, and fibers. However, many of these fruits and vegetables are also contaminated with heavy metals such as Hg, arsenic, and others. Moreover, fruits and vegetables should make up one-half of the plate at each meal according to the 2015–2020 US Dietary Guidelines for Americans [71]. We would encourage more research on exploring the Hg levels in different food items and the risk–benefit assessment of each type of fruit and vegetable consumed in Jamaica.

Although several studies have shown rice is one of the most crucial sources of Hg exposure in many regions, we did not find any association between rice consumption and BHgC among Jamaican TD children, possibly because nearly all of our study participants (98%) consumed rice. It is possible that some of the participants ate imported food from other countries. However, this option was not included in the food frequency questionnaire used for this study, which means that the notion that most of the food items consumed by the study participants were also locally cultivated and caught remains in doubt. Furthermore, without examining the health outcomes that are associated with these exposures, the results from this study should be interpreted with great caution especially when being used for food safety considerations and advisories. Data's delay owing to the date of sample collection is another weakness for this study. Moreover, considering the half-life of Hg, using a single sample for BHgC evaluation is another limitation of this study.

Moreover, based on n = 375, our calculations indicate that we have 83% power to detect small effect-sizes of 0.3 or higher at 5% level of significance. However, since this study is a secondary exploratory analysis, we didn’t adjust for multiple comparisons. Accordingly, we recommend that further investigation of these reported relationships be conducted in other country and population settings with larger sample sizes, adjusted for multiple comparisons, and using data for Hg levels in commonly consumed food categories and fish in Jamaica.

Conclusion

In this study, we identified the child’s age, as well as consumption of saltwater fish, canned fish (sardine, mackerel), string beans, grain, and starches (pasta, macaroni, noodles), as the environmental factors significantly associated with BHgCs in Jamaican TD children.

Based on the adjusted multivariable analysis, which included gene-environment interaction, the findings suggest that only children with Ile/Ile and Ile/Val genotypes for the GSTP1 Ile105Val polymorphism who consumed canned fish (sardine, mackerel) had a higher mean BHgC compared to children who did not consume those canned fish. This finding implies that having two Val alleles (Val/Val) for the GSTP1 Ile105Val polymorphism is associated with more effective detoxification of Hg in Jamaican children who ate canned fish (sardine, mackerel). Therefore, increasing the understanding among parents regarding how these dietary and environmental factors could affect BHgCs will potentially lead to lower Hg exposure among Jamaican children, especially among those who are more susceptible to adverse health outcomes related to Hg exposures due to their genetic predispositions.

Availability of data and materials

The data analyzed in this study are from two grants (i.e., R21 and R01). The data from R01 are or will be publicly available through the National Database for Autism Research (NDAR). Data from R21 will also be available upon request from the corresponding author.

Abbreviations

Hg:

Mercury

BHgC:

Blood Hg concentration

GST:

Glutathione S-transferase

TD:

Typically developing

eHg:

Elemental Hg

oHg:

Organic Hg

iHg:

Inorganic Hg

mHg:

Methyl Hg

eHg:

Ethyl Hg

pHg:

Phenyl Hg

NS:

Nervous system

FAO:

Food and Agricultural Organization

FDA:

Food and Drug Administration

GM:

Geometric mean

GSH:

Glutathione

ASD:

Autism spectrum disorder

UWI:

University of the West Indies

UTHealth:

The University of Texas Health Science Center at Houston

ERAJ:

Epidemiological Research on Autism in Jamaica

SCQ:

Social Communication Questionnaire

SES:

Socioeconomic status

CDC:

Centers for Disease Control and Prevention

LoD:

Limits of Detection

QC:

Quality control

GLMs:

General linear models

CI:

Confidence interval

EPA:

Eicosapentaenoic acid

DHA:

Docosahexaenoic acid

CAT:

Catalase

GPx:

Glutathione-peroxidase

SOD:

Superoxide dismutase

References

  1. Zhou H, Lin Y, Zhao W, Teng Y, Cui Y, Wang T, et al. The role of Hipk2-p53 pathways in arsenic-induced autistic behaviors: A translational study from rats to humans. Environ Pollut. 2020;267:115568.

    Article  CAS  PubMed  Google Scholar 

  2. Puklová V, Krsková A, Cerná M, Cejchanová M, Rehůrková I, Ruprich J, et al. The mercury burden of the Czech population: an integrated approach. Int J Hyg Environ Health. 2010;213:243–51.

    Article  PubMed  Google Scholar 

  3. Counter SA, Buchanan LH. Mercury exposure in children: a review. Toxicol Appl Pharmacol. 2004;198:209–30.

    Article  CAS  PubMed  Google Scholar 

  4. Gundacker C, Komarnicki G, Jagiello P, Gencikova A, Dahmen N, Wittmann KJ, et al. Glutathione-S-transferase polymorphism, metallothionein expression, and mercury levels among students in Austria. Sci Total Environ. 2007;385:37–47.

    Article  CAS  PubMed  Google Scholar 

  5. Karagas MR, Choi AL, Oken E, Horvat M, Schoeny R, Kamai E, et al. Evidence on the human health effects of low-level methylmercury exposure. Environ Health Perspect. 2012;120:799–806.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Driscoll CT, Mason RP, Chan HM, Jacob DJ, Pirrone N. Mercury as a global pollutant: sources, pathways, and effects. Environ Sci Technol. 2013;47:4967–83.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Bose-O’Reilly S, McCarty KM, Steckling N, Lettmeier B. Mercury exposure and children’s health. Curr Probl Pediatr Adolesc Health Care. 2010;40:186–215.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Clarkson TW, Magos L, Myers GJ. The toxicology of mercury–current exposures and clinical manifestations. N Engl J Med. 2003;349:1731–7.

    Article  CAS  PubMed  Google Scholar 

  9. Trasande L, Landrigan PJ, Schechter C. Public health and economic consequences of methyl mercury toxicity to the developing brain. Environ Health Perspect. 2005;113:590–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. So SCA, Tsoi MF, Cheung AJ, Cheung TT, Cheung BMY. Blood and urine inorganic and organic mercury levels in the United States from 1999 to 2016. Am J Med. 2021;134:e20–30.

    Article  CAS  PubMed  Google Scholar 

  11. Zahir F, Rizwi SJ, Haq SK, Khan RH. Low dose mercury toxicity and human health. Environ Toxicol Pharmacol. 2005;20:351–60.

    Article  CAS  PubMed  Google Scholar 

  12. Gilmour CC, Podar M, Bullock AL, Graham AM, Brown SD, Somenahally AC, et al. Mercury methylation by novel microorganisms from new environments. Environ Sci Technol. 2013;47:11810–20.

    Article  CAS  PubMed  Google Scholar 

  13. Weil M, Bressler J, Parsons P, Bolla K, Glass T, Schwartz B. Blood mercury levels and neurobehavioral function. JAMA. 2005;293:1875–82.

    Article  CAS  PubMed  Google Scholar 

  14. Schober SE, Sinks TH, Jones RL, Bolger PM, McDowell M, Osterloh J, et al. Blood mercury levels in US children and women of childbearing age, 1999–2000. JAMA. 2003;289:1667–74.

    Article  CAS  PubMed  Google Scholar 

  15. Grandjean P. Methylmercury toxicity and functional programming. Reprod Toxicol. 2007;23:414–20.

    Article  CAS  PubMed  Google Scholar 

  16. Boyle CA, Decouflé P, Yeargin-Allsopp M. Prevalence and health impact of developmental disabilities in US children. Pediatrics. 1994;93:399–403.

    Article  CAS  PubMed  Google Scholar 

  17. Tollefson L, Cordle F. Methylmercury in fish: a review of residue levels, fish consumption and regulatory action in the United States. Environ Health Perspect. 1986;68:203–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Wang Y, Habibullah-Al-Mamun M, Han J, Wang L, Zhu Y, Xu X, et al. Total mercury and methylmercury in rice: Exposure and health implications in Bangladesh. Environ Pollut. 2020;265:114991.

    Article  CAS  PubMed  Google Scholar 

  19. Hapke H-J. Heavy metal transfer in the food chain to humans. In: Fertilizers and Environment. 1996;431–6.

  20. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological profile for mercury. Atlanta: Agency for Toxic Substances and Disease Registry; 1999. https://www.atsdr.cdc.gov/ToxProfiles/tp46.pdf. Accessed 20 Oct 2023.

  21. Ricketts P, Knight C, Gordon A, Boischio A, Voutchkov M. Mercury Exposure Associated with Use of Skin Lightening Products in Jamaica. J Heal Pollut. 2020;10:200601.

    Article  Google Scholar 

  22. Antunes Dos Santos A, Appel HM, Culbreth M, López-Granero C, Farina M, Rocha JBT, et al. Methylmercury and brain development: a review of recent literature. J trace Elem Med Biol organ Soc Miner Trace Elem. 2016;38:99–107.

    Article  CAS  Google Scholar 

  23. Ricketts P, Voutchkov M, Chan HM. Risk-benefit assessment for total mercury, arsenic, selenium, and omega-3 fatty acids exposure from fish consumption in Jamaica. Biol Trace Elem Res. 2020;197:262–70.

    Article  CAS  PubMed  Google Scholar 

  24. Food and Agriculture Organization of the United Nations (FAO). The State of World Fisheries and Aquaculture 2016. In Contributing to Food Security and Nutrition for All; 2016. https://www.fao.org/3/i5555e/i5555e.pdf. Accessed 20 Oct 2023.

  25. Howe A, Fung LH, Lalor G, Rattray R, Vutchkov M. Elemental composition of Jamaican foods 1: a survey of five food crop categories. Environ Geochem Health. 2005;27:19–30.

    Article  CAS  PubMed  Google Scholar 

  26. Gworek B, Dmuchowski W, Baczewska-Dąbrowska AH. Mercury in the terrestrial environment: a review. Environ Sci Eur. 2020;32:128.

    Article  CAS  Google Scholar 

  27. Antoine JMR, Fung LAH, Grant CN. Assessment of the potential health risks associated with the aluminium, arsenic, cadmium and lead content in selected fruits and vegetables grown in Jamaica. Toxicol reports. 2017;4:181–7.

    Article  CAS  Google Scholar 

  28. Hayes JD, Pulford DJ. The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol. 1995;30:445–600.

    Article  CAS  PubMed  Google Scholar 

  29. Custodio HM, Harari R, Gerhardsson L, Skerfving S, Broberg K. Genetic influences on the retention of inorganic mercury. Arch Environ Occup Health. 2005;60:17–23.

    Article  CAS  PubMed  Google Scholar 

  30. McGuire S, Daggett DA, Bostad E, Nelson S, Wright LS, Siegel FL, et al. Increased levels of glutathione S transferases and appearance of novel alpha class isoenzymes in kidneys of mice exposed to mercuric chloride. I Biochemical and immunohistochemical studies Nephron. 1997;77:452–60.

    CAS  PubMed  Google Scholar 

  31. Hayes JD, Strange RC. Glutathione S-transferase polymorphisms and their biological consequences. Pharmacology. 2000;61:154–66.

    Article  CAS  PubMed  Google Scholar 

  32. Bruhn C, Brockmöller J, Kerb R, Roots I, Borchert HH. Concordance between enzyme activity and genotype of glutathione S-transferase theta (GSTT1). Biochem Pharmacol. 1998;56:1189–93.

    Article  CAS  PubMed  Google Scholar 

  33. Westphal GA, Schnuch A, Schulz TG, Reich K, Aberer W, Brasch J, et al. Homozygous gene deletions of the glutathione S-transferases M1 and T1 are associated with thimerosal sensitization. Int Arch Occup Environ Health. 2000;73:384–8.

    Article  CAS  PubMed  Google Scholar 

  34. Rahbar MH, Samms-Vaughan M, Saroukhani S, Bressler J, Hessabi M, Grove ML, et al. Associations of metabolic genes (GSTT1, GSTP1, GSTM1) and blood mercury concentrations differ in Jamaican children with and without autism spectrum disorder. Int J Environ Res Public Health. 2021;18:1377.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Rahbar MH, Samms-Vaughan M, Loveland KA, Ardjomand-Hessabi M, Chen Z, Bressler J, et al. Seafood consumption and blood mercury concentrations in Jamaican children with and without autism spectrum disorders. Neurotox Res. 2013;23:22–38.

    Article  CAS  PubMed  Google Scholar 

  36. Rahbar MH, Samms-Vaughan M, Dickerson AS, Loveland KA, Ardjomand-Hessabi M, Bressler J, et al. Blood lead concentrations in Jamaican children with and without autism spectrum disorder. Int J Environ Res Public Health. 2014;12:83–105.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Rahbar MH, Samms-Vaughan M, Loveland KA, Pearson DA, Bressler J, Chen Z, et al. Maternal and paternal age are jointly associated with childhood autism in Jamaica. J Autism Dev Disord. 2012;42:1928–38.

    Article  PubMed  Google Scholar 

  38. Rahbar MH, Samms-Vaughan M, Dickerson AS, Loveland KA, Ardjomand-Hessabi M, Bressler J, et al. Factors associated with blood lead concentrations of children in Jamaica. J Environ Sci Heal Part A, Toxic/hazardous Subst Environ Eng. 2015;50:529–39.

    CAS  Google Scholar 

  39. Rahbar MH, Samms-Vaughan M, Lee M, Zhang J, Hessabi M, Bressler J, et al. Interaction between a Mixture of Heavy Metals (Lead, Mercury, Arsenic, Cadmium, Manganese, Aluminum) and GSTP1, GSTT1, and GSTM1 in Relation to Autism Spectrum Disorder. Res Autism Spectr Disord. 2020;79:101681.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Marvin AR, Marvin DJ, Lipkin PH, Law JK. Analysis of Social Communication Questionnaire (SCQ) screening for children less than age 4. Curr Dev Disord reports. 2017;4:137–44.

    Article  Google Scholar 

  41. Mulligan A, Richardson T, Anney RJL, Gill M. The Social Communication Questionnaire in a sample of the general population of school-going children. Ir J Med Sci. 2009;178:193–9.

    Article  CAS  PubMed  Google Scholar 

  42. Rahbar MH, Samms-Vaughan M, Zhao Y, Saroukhani S, Zaman SF, Bressler J, et al. Additive and interactive associations of environmental and sociodemographic factors with the genotypes of three glutathione s-transferase genes in relation to the blood arsenic concentrations of children in Jamaica. Int J Environ Res Public Health. 2022;19:466.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Hornung R, Reed L. Estimation of Average Concentration in the Presence of Nondetectable Values. Appl Occup Environ Hyg. 1990;5(1):46–51.

    Article  CAS  Google Scholar 

  44. Cusack LK, Smit E, Kile ML, Harding AK. Regional and temporal trends in blood mercury concentrations and fish consumption in women of child bearing Age in the united states using NHANES data from 1999–2010. Environ Health. 2017;16:10.

    Article  PubMed  PubMed Central  Google Scholar 

  45. SAS Institute. SAS/STAT® 13.1 User’s Guide: The GLM Procedure. SAS Institute. 2103. https://support.sas.com/documentation/onlinedoc/stat/131/glm.pdf.

  46. SAS Institute Inc. SAS®, 9.4. Cary: SAS Institute Inc.; 2013.

  47. Mannetje t’ A, Coakley J, Douwes J. Total blood mercury and its determinants in New Zealand children and adults. J Expo Sci Environ Epidemiol. 2021;31:289–98.

    Article  Google Scholar 

  48. Minnesota Department of Health. Mercury. [https://data.web.health.state.mn.us/biomonitoring_mercury#About.]. Accessed on date: 10/20/2023

  49. Gao Z-Y, Li M-M, Wang J, Yan J, Zhou C-C, Yan C-H. Blood mercury concentration, fish consumption and anthropometry in Chinese children: a national study. Environ Int. 2018;110:14–21.

    Article  CAS  PubMed  Google Scholar 

  50. Bernstein AS, Oken E, de Ferranti S. Fish, Shellfish, and Children’s Health: An Assessment of Benefits, Risks, and Sustainability. Pediatrics. 2019;e.20190999. [Online ahead of print.].

  51. Ralston NVC, Kaneko JJ, Raymond LJ. Selenium health benefit values provide a reliable index of seafood benefits vs risks. Journal Trace Elements Medicine Biology. 2019;55:50–7.

    Article  CAS  Google Scholar 

  52. Mozaffarian D, Rimm EB. Fish intake, contaminants, and human health: evaluating the risks and the benefits. JAMA. 2006;296:1885–99.

    Article  CAS  PubMed  Google Scholar 

  53. Chang C-H, Chiang C-F, Liao J-W, Yen G-C, Huang Y-C, Ni S-P, et al. Dietary exposure assessment of methylmercury and polyunsaturated fatty acids in saltwater fish and processed foods among Taiwanese women of child-bearing age and children: A novel core food-matching approach. Chemosphere. 2021;262:128249.

    Article  CAS  PubMed  Google Scholar 

  54. Huffman SL, Harika RK, Eilander A, Osendarp SJM. Essential fats: how do they affect growth and development of infants and young children in developing countries? A Literature Review Matern Child Nutr. 2011;7(3):44–65.

    Article  PubMed  Google Scholar 

  55. D’Angelo S, Motti ML, Meccariello R. ω-3 and ω-6 Polyunsaturated fatty acids. Obesity Cancer Nutrients. 2020;12:2751.

    PubMed  Google Scholar 

  56. Petrova MV, Ourgaud M, Boavida JRH, Dufour A, Tesán Onrubia JA, Lozingot A, et al. Human mercury exposure levels and fish consumption at the French Riviera. Chemosphere. 2020;258:127232.

    Article  CAS  PubMed  Google Scholar 

  57. Canham R, González-Prieto AM, Elliott JE. Mercury exposure and toxicological consequences in fish and fish-eating wildlife from anthropogenic activity in Latin America. Integr Environ Assess Manag. 2021;17:13–26.

    Article  CAS  PubMed  Google Scholar 

  58. Stratakis N, Conti DV, Borras E, Sabido E, Roumeliotaki T, Papadopoulou E, et al. Association of fish consumption and mercury exposure during pregnancy with metabolic health and inflammatory biomarkers in children. JAMA Netw open. 2020;3:e201007.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Oken E, Bellinger DC. Fish consumption, methylmercury and child neurodevelopment. Curr Opin Pediatr. 2008;20:178–83.

    Article  PubMed  PubMed Central  Google Scholar 

  60. Food and Agriculture Organization (FAO). Fishery and Aquaculture Country Profiles Angola. United Nations; 2022. https://www.fao.org/fishery/en/facp/ago?lang=en. Accessed 20 Oct 2023.

  61. United Nations Environment Programme. Mercury Release Inventory, Waste Storage and Disposal in Jamaica. 2016.https://wedocs.unep.org/handle/20.500.11822/31215. Accessed 20 Oct 2023.

  62. Yu H, Li J, Luan Y. Meta-analysis of soil mercury accumulation by vegetables. Sci Rep. 2018;8:1261.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Gundacker C, Wittmann KJ, Kukuckova M, Komarnicki G, Hikkel I, Gencik M. Genetic background of lead and mercury metabolism in a group of medical students in Austria. Environ Res. 2009;109:786–96.

    Article  CAS  PubMed  Google Scholar 

  64. Barcelos GRM, de Souza MF, de Oliveira AÁS, van Lengert AH, de Oliveira MT, de Camargo RB, et al. Effects of genetic polymorphisms on antioxidant status and concentrations of the metals in the blood of riverside Amazonian communities co-exposed to Hg and Pb. Environ Res. 2015;138:224–32.

    Article  CAS  PubMed  Google Scholar 

  65. Shankar K, Mehendale HM. Cytochrome P450. P. Wexler (Ed.), Encyclopedia of Toxicology (Third edition). Oxford: Academic Press; 2014. p. 1125-7.

  66. Ballatori N. Transport of toxic metals by molecular mimicry. Environ Health Perspect. 2002;110(5):689–94.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  67. Custodio HM, Broberg K, Wennberg M, Jansson J-H, Vessby B, Hallmans G, et al. Polymorphisms in glutathione-related genes affect methylmercury retention. Arch Environ Health. 2004;59:588–95.

    Article  CAS  PubMed  Google Scholar 

  68. Schläwicke Engström K, Strömberg U, Lundh T, Johansson I, Vessby B, Hallmans G, et al. Genetic variation in glutathione-related genes and body burden of methylmercury. Environ Health Perspect. 2008;116:734–9.

    Article  PubMed  Google Scholar 

  69. Nag HE, Nordgren A, Anderlid B-M, Nærland T. Reversed gender ratio of autism spectrum disorder in Smith-Magenis syndrome. Mol Autism. 2018;9:1.

    Article  PubMed  PubMed Central  Google Scholar 

  70. Silbernagel SM, Carpenter DO, Gilbert SG, Gochfeld M, Groth E 3rd, Hightower JM, et al. Recognizing and preventing overexposure to methylmercury from fish and seafood consumption: information for physicians. J Toxicol. 2011;2011:983072.

    Article  PubMed  PubMed Central  Google Scholar 

  71. Wallace TC, Bailey RL, Blumberg JB, Burton-Freeman B, Chen C-YO, Crowe-White KM, et al. Fruits, vegetables, and health: A comprehensive narrative, umbrella review of the science and recommendations for enhanced public policy to improve intake. Crit Rev Food Sci Nutr. 2020;60:2174–211.

    Article  CAS  PubMed  Google Scholar 

  72. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)–a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–81.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We acknowledge that the collection and management of survey data were done using REDCap [72]. We also acknowledge contributions by colleagues in the Analytical Chemistry Lab at MDHHS for analyzing and storing the whole blood samples for the assessments of heavy metal concentrations, under a service contract.

Informed consent

All parents provided written informed consent, and when applicable, an assent was collected from 7-8 years old children before participating in this ERAJ study. 

Funding

This research is funded by the National Institute of Environmental Health Sciences (NIEHS) by a grant (R01ES022165), as well as the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) and the National Institutes of Health Fogarty International Center (NIH-FIC) by a grant (R21HD057808) awarded to University of Texas Health Science Center at Houston. We also acknowledge the support provided by the Biostatistics/Epidemiology/Research Design (BERD) component of the Center for Clinical and Translational Sciences (CCTS) for this project. CCTS is mainly funded by the NIH Centers for Translational Science Award (NIH CTSA) grant (UL1 RR024148), awarded to University of Texas Health Science Center at Houston in 2006 by the National Center for Research Resources (NCRR), and its 2012 renewal (UL1 TR000371) as well as another 2019 grant (UL1TR003167) by the National Center for Advancing Translational Sciences (NCATS). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIEHS, NICHD, NIH-FIC, NCRR, or NCATS.

Author information

Authors and Affiliations

  1. Department of Epidemiology, Human Genetics, and Environmental Sciences (EHGES), School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX, 77030, USA

    Sheikh Farzana Zaman, Jan Bressler, Megan L. Grove & Mohammad H. Rahbar

  2. Biostatistics/Epidemiology/Research Design (BERD) Component, Center for Clinical and Translational Sciences (CCTS), The University of Texas Health Science Center at Houston, Houston, TX, 77030, USA

    Sheikh Farzana Zaman, Sepideh Saroukhani, Manouchehr Hessabi & Mohammad H. Rahbar

  3. Department of Child & Adolescent Health, The University of the West Indies (UWI), Mona Campus, Kingston 7, Kingston, Jamaica

    Maureen Samms-Vaughan & Sydonnie Shakespeare Pellington

  4. Department of Internal Medicine, Division of Clinical and Translational Sciences, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX, 77030, USA

    Sepideh Saroukhani & Mohammad H. Rahbar

  5. Human Genetics Center, School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX, 77030, USA

    Jan Bressler & Megan L. Grove

  6. Louis A Faillace, MD, Department of Psychiatry and Behavioral Sciences, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX, 77054, USA

    Katherine A. Loveland

Authors
  1. Sheikh Farzana Zaman
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Maureen Samms-Vaughan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Sepideh Saroukhani
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Jan Bressler
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Manouchehr Hessabi
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Megan L. Grove
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Sydonnie Shakespeare Pellington
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Katherine A. Loveland
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Mohammad H. Rahbar
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Conceptualization, M.H.R., M.S.-V., S.F.Z, and J.B.; methodology, M.H.R., M.S.-V., S.F.Z., J.B, and M.L.G.; validation, M.H.R.; formal analysis, M.H.R. S.F.Z, and S.S.; investigation, M.H.R., S.S.-P., M.S.-V.; resources, M.H.R. and M.S.-V.; data curation, M.S.-V., M.L.G., and S.S.-P.; visualization, M.H. and M.H.R.; writing-original draft preparation, S.F.Z., S.S., and M.H.R.; writing-review and editing, M.H.R., S.F.Z., S.S., J.B., M.H., and K.A.L.; supervision, M.H.R.; project administration, M.H.R and M.S.-V.; funding acquisition, M.H.R. All authors have read and agreed to the published version of the manuscript.

Corresponding author

Correspondence to Mohammad H. Rahbar.

Ethics declarations

Ethics approval and consent to participate

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The ERAJ study protocol has been approved by the Institutional Review Boards (IRBs) of both University of Texas Health Science Center at Houston (UTHealth) (IRB Protocol number: HSC-SPH-09–0059) and The University of the West Indies (UWI), in Jamaica.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary file 1:

Table S1. Distribution of Blood Hg Concentration (Median, IQR) with Different Environmental Exposures and GST Genotypes among Jamaican Children. Table S2. Associations between children’s GST genotypes and BHgC by exposure to environmental factors based on the multivariable General Linear Models that include the interactions between GST genes and the main environmental exposures (n = 375). Table S3. Association between children’s genotype for GSTP1 and BHgC by consumption of canned fish (sardine, mackerel) based on General Linear Models that adjust for child’s age, consumption of saltwater fish, grain and starches (pasta, macaroni, and noodles), and string beans. Table S4. Geometric Mean (GM) Blood Mercury Concentration (BHgC) among 7-11 years old children in Different Countries*. Table S5. GST Gene (GSTT1, GSTM1, GSTP1) Genotypes Distribution in Different countries.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zaman, S.F., Samms-Vaughan, M., Saroukhani, S. et al. Factors associated with blood mercury concentrations and their interactions with three glutathione S-transferase genes (GSTT1, GSTM1, and GSTP1): an exposure assessment study of typically developing Jamaican children. BMC Pediatr 24, 14 (2024). https://doi.org/10.1186/s12887-023-04452-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12887-023-04452-w

Keywords

BMC Pediatrics

ISSN: 1471-2431

Contact us