Elsevier

NeuroToxicology

Volume 30, Issue 3, May 2009, Pages 338-349
NeuroToxicology

Postnatal exposure to methyl mercury from fish consumption: A review and new data from the Seychelles Child Development Study

https://doi.org/10.1016/j.neuro.2009.01.005Get rights and content

Abstract

Background

Fish is an important source of nutrition worldwide. Fish contain both the neurotoxin methyl mercury (MeHg) and nutrients important for brain development. The developing brain appears to be most sensitive to MeHg toxicity and mothers who consume fish during pregnancy expose their fetus prenatally. Although brain development is most dramatic during fetal life, it continues for years postnatally and additional exposure can occur when a mother breast feeds or the child consumes fish. This raises the possibility that MeHg might influence brain development after birth and thus adversely affect children's developmental outcomes. We reviewed postnatal MeHg exposure and the associations that have been published to determine the issues associated with it and then carried out a series of analyses involving alternative metrics of postnatal MeHg exposure in the Seychelles Child Development Study (SCDS) Main Cohort.

Methods

The SCDS is a prospective longitudinal evaluation of prenatal MeHg exposure from fish consumption. The Main Cohort includes 779 subjects on whom recent postnatal exposure data were collected at the 6-, 19-, 29-, 66-, and 107-month evaluations. We examined the association of recent postnatal MeHg exposure with multiple 66- and 107-month outcomes and then used three types of alternative postnatal exposure metrics to examine their association with the children's intelligence quotient (IQ) at 107 months of age.

Results

Recent postnatal exposure at 107 months of age was adversely associated with four endpoints, three in females only. One alternative postnatal metric was beneficially associated with 9-year IQ in males only.

Conclusions

We found several associations between postnatal MeHg biomarkers and children's developmental endpoints. However, as has been the case with prenatal MeHg exposure in the SCDS Main Cohort study, no consistent pattern of associations emerged to support a causal relationship.

Introduction

Mercury is naturally present in the earth's crust and is widespread in the environment (WHO, 1990). Common bacteria in aquatic environments methylate part of the environmental inorganic mercury to the organic form of MeHg. After entering the aquatic food web, MeHg is bioaccumulated and bioconcentrated and all fish acquire it to varying degrees. All fish consumption leads to some degree of MeHg exposure.

Several outbreaks of poisoning following exposure to high concentrations of MeHg took place during the twentieth century. Pathological studies of subjects poisoned during those episodes indicated that MeHg poisoning affects the brain differently depending upon the age at which exposure occurs (Takeuchi, 1968, Choi et al., 1978). Prenatal poisoning in Japan and Iraq was reported to cause diffuse brain damage while adult poisoning initially caused more focal damage affecting the visual cortex, motor area and cerebellum. Poisoning during childhood produced a pattern of damage with features of both prenatal and adult exposure. However, the pathological findings were closer to those of prenatal exposure. One study of prenatal MeHg exposure at the levels achieved by fish consumption found no pathological changes in the brain using routine histopathology (Lapham et al., 1995). Children prenatally poisoned by MeHg presented clinically with severe cognitive and motor deficits, seizures, and microcephaly (Harada, 1968). Adults who were poisoned presented initially with paresthesias, visual loss, and ataxia reflecting more focal impairment of the brain. Clinical outcomes when poisoning occurred during childhood were less well characterized but included motor and cognitive deficits (Amin-Zaki et al., 1976, Engleson and Herner, 1952). These poisoning outbreaks confirmed that MeHg appears to have its greatest neurotoxic effect on the developing brain.

Fish consumption and consequently exposure to MeHg is common. The Food and Agriculture Organization of the United Nations (FAO) estimates that one billion people worldwide depend upon fish for daily nutrition (FAO, 2000). Methyl mercury poisoning with clinical symptoms resulting from fish consumption has been reported on two occasions, both from Japan. Methyl mercury poisoning from other types of exposure has been reported on several occasions. The possibility of more subtle adverse associations that might be difficult to detect has led to widespread public concern (Myers et al., 2006). From a public health perspective it is important to know if prenatal or postnatal exposure to MeHg at the levels achieved by fish consumption causes neurotoxic consequences that can be detected in children. This question has been the focus of several epidemiologic studies. These studies have primarily focused on prenatal exposure to MeHg and examined its association with the children's developmental outcomes using epidemiological methods.

The rate of brain growth peaks during gestation, but it continues to progress rapidly during the first 2 years of life and at a slower pace through adolescence and beyond. There are no specific anatomical or physiologic events that occur in the brain at birth to mark the transition from fetal life to infancy. Exposure to MeHg can continue following birth if mothers consume fish and breast feed (Bakir et al., 1973, Amin-Zaki et al., 1974, Amin-Zaki et al., 1981, Grandjean et al., 1994, Grandjean et al., 1995, Chien et al., 2006). In addition, as children grow and fish is introduced into their diets, further exposure can occur.

In this paper we provide a background on brain development and review the literature on postnatal MeHg exposure in children and its association with children's development. We then describe the association of recent postnatal MeHg exposure and developmental testing present in the primary Seychelles Child Development Study (SCDS) analyses. Next we examine three alternative metrics for measuring postnatal exposure based on the recent postnatal MeHg measurements at multiple time points, and report on their association with the children's full scale IQ measured by the WISC III at 107 months of age. We present our results in two sections, first for the associations of postnatal exposure from the primary SCDS main cohort analyses that focused on prenatal MeHg exposure when the children were 66 (Davidson et al., 1998) and 107 months of age (Myers et al., 2003) and second for the alternative postnatal MeHg exposure metrics.

The normal neonatal brain weighs about 350–400 g at birth and triples in size during the first 18 years of life. This rapid brain development is most apparent in the enlarging postnatal head circumference (HC). The average HC of a child at birth is about 35 cm, but by age 2 years it increases to about 50 cm and by 18 years to 56 cm. The increase in HC and in corresponding brain weight is associated with a host of anatomic and physiological changes that occur continuously from birth through adolescence and beyond (Rice and Barone, 2000, Volpe, 2001). Nearly all neurons in the cortex develop in the germinal matrix near the ventricles prior to birth and then migrate to their final location in the cerebral cortex before establishing connections. Neurons in the cerebellum in contrast develop following birth. At birth some cortical neurons are still migrating and many have not started to mature or differentiate. Those that have reached their final destination are still in the process of extending axons and dendrites, developing dendritic spines, and forming synapses. The establishment of the myriad connections that characterize the mature brain is a continuing postnatal process that continues for many years following birth. Synapses begin to form in the third trimester and develop actively during the first 2 years of life. The maximum number of synapses is achieved at about age 2 years and they are subsequently pruned and reduced by approximately 40% by adolescence. Myelination is minimal at birth and continues well into adulthood. The numerous neurotransmitters present in the brain have individual patterns of development and some develop after birth (Rice and Barone, 2000). For example nicotinic receptors in the reticular formation develop during mid gestation while GABA receptors in the neocortex develop predominantly during the first year of life. The blood brain barrier does not completely form until after birth (Rodier, 2004). A significant number of the cortical neurons present at birth subsequently undergo apoptosis, but the factors that determine which neurons will die and which synapses will be retained are largely unknown. Neurotoxins might conceivably influence this process.

Several mechanisms have been proposed by which MeHg might damage the developing brain. Among them are MeHg induced alterations in microtubules, oxidative damage to neurons, impairment of neuronal and glial calcium homeostasis, and the potentiation of glutamatergic neurotransmission (Castoldi et al., 2003). The effect of disrupting microtubules and consequently mitosis, migration, and cortical organization of neurons is especially serious prenatally, although these processes still continue postnatally (Rodier, 2004, Vogel et al., 1985, Clarkson, 1987, Choi et al., 1978). Chemically, MeHg has a strong affinity for sulfhydryl groups that are present on proteins and glutathione. When MeHg complexes with these compounds it can adversely affect anabolic processes and protein synthesis (Bondy, 1994, Slikker, 1994, Syversen, 1982). By inactivating sulfhydryl enzymes MeHg can interfere with cellular metabolism and function. Methyl mercury is also known to catalyze the formation of excess reactive oxygen species and the regional distribution of this activity parallels the sites of known neuropathological changes (Bondy, 1994). Methyl mercury rapidly binds to reduced glutathione (GSH) which is present in most cells in millimolar concentrations (Clarkson and Magos, 2006). This binding may serve to protect intracellular proteins. There are also adverse effects of MeHg on the synthesis of fetal DNA in astrocytes, and on the growth cones of neurons (Marsh, 1994). Additionally, it can induce structural chromosomal aberrations in experimental animals (Ehrenstein et al., 2002). Most of these physiological processes are continuously active in the brain following birth.

Takeuchi (1968) noted a number of developmental brain deviations present in prenatal MeHg poisoning (fetal Minamata disease) that were not found with postnatal or infantile exposure. These included the presence of nerve cells in the cerebral medulla, columnar grouping of nerve cells in the cerebral cortex, abnormal cytoarchitecture of nerve cells in the cerebral and cerebellar cortices, and dysplasia of nerve cells with poor myelination (Choi, 1989).

In cell cultures, low dose exposure to MeHg has been shown to cause physiological disturbances such as cell cycle inhibition without cytotoxicity (Gribble et al., 2005). In experimental animals such as the rat, low dose exposure over long time periods has been reported to result in alteration of brain neurotransmitters (Slikker, 1994).

Reports of postnatal MeHg exposure fall into two general categories. Some are of children with overt clinical poisoning and others are from epidemiology studies looking for subtle population differences between normal children with varying levels of chronic MeHg exposure. In cases of overt poisoning such as occurred in Iraq, the Hg exposure level that correlated best with clinical outcomes was the peak hair value (Bakir et al., 1973). However, most human exposure is to small amounts of MeHg present in dietary sources such as fish and seafood where no significant peaks are usually present.

A small number of children have been reported in the literature with postnatal MeHg poisoning (Engleson and Herner, 1952, Harada, 1968, Amin-Zaki et al., 1976, Davis et al., 1994). Among these reports, individuals where exposure was from fish consumption were reported only from Minamata and Niigata Japan. No cases of poisoning from fish consumption have been reported elsewhere. The first child reported in the literature consumed MeHg treated grain and subsequently was diagnosed with a developmental delay (Engleson and Herner, 1952). In Japan the children poisoned at Minamata where exposed to MeHg along with several other neurotoxicants (Harada, 1968, Takeuchi and Eto, 1977). Several cases were reported from Iraq where exposure was to MeHg treated seed grain (Amin-Zaki et al., 1976, Amin-Zaki et al., 1978, Amin-Zaki et al., 1980, Amin-Zaki et al., 1981). Table 1 outlines the cases of postnatal MeHg poisoning that have been reported in children.

There is one report of children poisoned by MeHg in the United States (Snyder, 1972, Brenner and Snyder, 1980, Pierce et al., 1972, Davis et al., 1994). A family in New Mexico consumed a hog that had been fed MeHg treated seed grain and had a MeHg level of 35 ppm. The family included four children under the age of 18 years. Two of the children had severe neurological damage. An 8-year-old girl with a hair Hg level of 1398 ppm had cognitive impairment, choreoathetosis, seizures and quadriparesis. Her 13-year-old brother also had severe neurological impairment, but his exposure was not measured. Two sisters ages 9 and 16 years at exposure had no symptoms. The older girl had a hair Hg level of 329 ppm following exposure. These two girls were examined over 20 years later and reported to be neurologically normal (Davis et al., 1994).

Two longitudinal and three cross-sectional epidemiology studies have included an index of postnatal MeHg exposure. Both longitudinal studies obtained children's hair samples for postnatal exposure when they were undergoing clinical assessments to determine the children's development. Table 2 lists the reports and references from these studies. The Faeroe Islands study measured prenatal MeHg exposure in cord blood and maternal hair during pregnancy, and postnatal exposure in the children's blood at ages 7 and 14 years, and in children's hair at ages 1, 7, and 14 years. The SCDS measured prenatal MeHg exposure in maternal hair growing during pregnancy, and postnatal exposure in children's hair at ages 6, 19, 29, 66, and 107 months.

The Faeroe Islands investigators reported that at age 12 months longer periods of breastfeeding were associated with early achievement of developmental milestones assessed by history (Grandjean et al., 1995). At ages 7 and 14 years cohort children were administered an extensive battery of tests. The investigators reported significant adverse associations between children's hair Hg levels measured at 12 months of age and two endpoints measured at age 7 years (Finger Tapping with both hands and the Reaction Time from the Continuous Performance Test) (Grandjean et al., 1997). They also reported adverse associations between the children's hair Hg level measured at 7 years of age and several endpoints (the Continuous Performance Test, Reaction Time [CPT-RT], the Block Design subtest from the Wechsler Intelligence Scale for Children-Revised [WISC R], and the Bender Visual Motor Gestalt Copying Errors score [BG-ES]). None of the associations with postnatal MeHg exposure were significant when the analyses were adjusted for prenatal exposure measured in cord blood. At the 14-year evaluations they reported that “postnatal methylmercury exposure had no discernable effect” (Debes et al., 2006).

In the SCDS main cohort primary analyses, recent postnatal exposure was first included as a covariate in the 66-month evaluation since children were then actively consuming fish. At age 66 months in the primary analysis increasing postnatal total Hg (THg) exposure was associated with improving performance on the Preschool Language Scale-Total Score (PLS-TS), the Woodcock-Johnson Applied Problems (WJ-AP), and the BG-ES (Davidson et al., 1998). For the BG-ES the interaction of postnatal Hg with sex was significant (p = 0.004) (Davidson et al., 1998). For males their performance improved (regression coefficient = −0.16 ppm; p = 0.009). For females the slope was slightly positive indicating poorer performance, but it was not significant (p = 0.14). The association of THg exposure with improved performance on some tests was not expected since MeHg is toxic and has no known function in the human body. However, MeHg exposure is mainly from consuming fish, and they also contain nutrients. These results raised the intriguing possibility the nutrients present in fish might be having a significant beneficial influence on outcomes.

Section snippets

Methods

The SCDS is a longitudinal, prospective, double-blind epidemiological evaluation examining the association between prenatal MeHg exposure from maternal fish consumption and developmental outcomes in children. The main cohort consists of 779 maternal child pairs enrolled in 1989–1990 and followed longitudinally. Cohort children were administered a battery of developmental tests at ages 6, 19, 29, 66, and 107 months of age. The test batteries included global and domain specific tests including

Correlations between exposure measures

We first examined the association between prenatal and postnatal metrics measured at different time points. These correlations are presented in Table 4 and were all 0.30 or below. The correlation between prenatal and postnatal exposure was greatest at 6 months of age and decreased as the children matured. The correlation between postnatal exposures measured at different ages was greatest between the 66- and 107-month values. This suggests that by about age 5 years the children's hair THg values

Discussion

Using several different metrics for recent postnatal MeHg exposure we evaluated the SCDS main cohort for associations with children's developmental outcomes. We found a number of associations present at the 66- and 107-month evaluations in the primary linear analyses that examined the covariate adjusted association between prenatal exposure and outcomes. Some of the associations in the primary analyses were in the direction of declining performance as postnatal exposure increased and with

Acknowledgments

This study was supported by grants #PO1 ES01248, RO1 ES008442, 2 T 32 ES007271, and PO ES01247 from the National Institute of Environmental Health Sciences to the University of Rochester and by the Government of Seychelles.

References (60)

  • P. Grandjean et al.

    Cognitive deficit in 7-year-old children with prenatal exposure to methylmercury

    Neurotoxicol Teratol

    (1997)
  • L. Huang et al.

    Using measurement error models to assess effects of prenatal and postnatal methylmercury exposure in the Seychelles Child Development Study

    Environ Res

    (2003)
  • L. Huang et al.

    Exploring nonlinear association between prenatal methylmercury exposure from fish consumption and child development: evaluation of the Seychelles Child Development Study nine-year data using semiparametric additive models

    Environ Res

    (2005)
  • K. Murata et al.

    Delayed evoked potentials in children exposed to methylmercury from seafood

    Neurotoxicol Teratol

    (1999)
  • G.J. Myers et al.

    Secondary analysis from the Seychelles Child Development Study: the child behavior checklist

    Environ Res

    (2000)
  • G.J. Myers et al.

    Prenatal methylmercury exposure from ocean fish consumption in the Seychelles child development study

    Lancet

    (2003)
  • J.J. Strain et al.

    Associations of maternal long chain polyunsaturated fatty acids, methyl mercury, and infant development in the Seychelles Child Development Nutrition Study

    Neurotoxicology

    (2008)
  • D.G. Vogel et al.

    The effects of methyl mercury binding to microtubules

    Toxicol Appl Pharmacol

    (1985)
  • L. Amin-Zaki et al.

    Perinatal methylmercury poisoning in Iraq

    Am J Dis Child

    (1976)
  • L. Amin-Zaki et al.

    Methylmercury poisoning in Iraqi children: clinical observations over two years

    Br Med J

    (1978)
  • L. Amin-Zaki et al.

    Methylmercury poisoning in mothers and their suckling infants.

  • L. Amin-Zaki et al.

    Methylmercury poisoning in the Iraqi suckling infant: a longitudinal study over five years

    J Appl Toxicol

    (1981)
  • F. Bakir et al.

    Methylmercury poisoning in Iraq

    Science

    (1973)
  • D.C. Bellinger

    Very low lead exposures and children's neurodevelopment

    Curr Opin Pediatr

    (2008)
  • S.C. Bondy

    Chapter 20: induction of oxidative stress in the brain by neurotoxic agents

  • R.P. Brenner et al.

    Late EEG findings and clinical status after organic mercury poisoning

    Arch Neurol

    (1980)
  • A.F. Castoldi et al.

    Neurotoxic and molecular effects of methylmercury in humans

    Rev Environ Health

    (2003)
  • CDC (Center for Disease Control) growth charts accessed 4/25/08....
  • E. Cernichiari et al.

    Monitoring methylmercury during pregnancy: maternal hair predicts fetal brain exposure

    Neurotoxicology

    (1995)
  • B.H. Choi et al.

    Abnormal neuronal migration, deranged cerebral cortical organization, and diffuse white matter astrocytosis of human fetal brain: a majaor effect of methylmercury poisoning in utero

    J Neuropathol Exp Neurol

    (1978)
  • Cited by (0)

    View full text