Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jan 1.
Published in final edited form as: Neurotoxicology. 2017 Jul 17;64:134–141. doi: 10.1016/j.neuro.2017.07.007

Prenatal co-exposure to manganese and depression and 24-months neurodevelopment

Teresa Verenice Muñoz Rocha 1, Marcela Tamayo y Ortiz 1,2, Martín Romero 1, Ivan Pantic 3, Lourdes Schnaas 3, David Bellinger 4, Birgit Claus Henn 5, Rosalind Wright 6, Robert O Wright 7, Mara Téllez-Rojo 1
PMCID: PMC5771973  NIHMSID: NIHMS896385  PMID: 28728787

Abstract

Background

Normal prenatal neurodevelopment follows stages that are potentially influenced by both chemical and psychosocial environments. Exposure to elevated manganese during this critically vulnerable period has been found to be neurotoxic. Independently, maternal prenatal depression has been associated with subsequent neurodevelopmental decrements in children. The association between child neurodevelopment and prenatal co-exposure to manganese and maternal depression has not been sufficiently studied.

Methods

During pregnancy and at birth, we measured maternal blood and cord blood manganese levels respectively. Maternal depression was assessed in the 3rd trimester of pregnancy using the Edinburgh Depression Scale. Neurodevelopment was evaluated at 24 months of age with the Bayley Scales of Infant Development. A multivariate multiple regression model was used to analyze cognitive, language and motor scores simultaneously for 473 children from the PROGRESS birth cohort in Mexico City.

Results

Over 25% of our study participants reported having depressive symptoms. 3rd trimester blood manganese as well as depressive symptoms were independently negatively associated with all neurodevelopment scores in adjusted models. In stratified analyses, the negative association between manganese (maternal as well as cord blood) and 24-month language scores was stronger among women with depressive symptoms. Receptive language was mostly affected. Inverted U-shaped curves were seen for the association between with cord blood manganese and neurodevelopment scores.

Conclusions

Our findings are in line with previous studies of manganese and depression neurotoxicity. The prenatal period may be particularly sensitive to manganese and depression co-exposures and should be of interest for public health interventions to promote healthy emotional and nutritional pregnancies.

Keywords: prenatal, neurodevelopment, depression, manganese, co-exposure

1. Background

The prenatal period is critical to neurodevelopment. During this time, the nervous system matures towards the achievement of the cognitive, motor and language functions. Independently, exposure to elevated manganese (Mn) and the presence of depression in pregnant women are factors that can influence the neurodevelopment of their offspring (Chung et al., 2015; Coetzee et al., 2016; Glover, 2014; Nulman et al., 2015).

Manganese (Mn) is an essential nutrient that contributes substantially to a variety of body functions including the breakdown of cholesterol, carbohydrates and proteins and also plays a key role in brain development and bone formation (Freeland-Graves and Turnlund, 1996). Mn is a core component of various metalloenzymes that ensure proper functioning of the central nervous system (Aschner and Aschner, 1991). The general population is exposed to Mn through food, water and air. Mn absorbed from ambient air might occur in areas with heavy traffic (Bueno-Brito et al., 2005). However, levels in ambient air are considered negligible when compared to the natural abundance of Mn in soil (Gulson et al., 2006). In food, its concentration is highest in string beans, nuts, legumes, seeds, tea, whole grains and green-leaf vegetables. Guidelines have been established for different exposure routes: the WHO establishes 0.15 μg/m3 for air and 0.4 mg/L in water; the U.S. FDA establishes a limit of 0.05 mg/L in bottled drinking water and as for food the EPA has a reference dose of 0.14 mg/kg/day (ATSDR, 2012). According to the Food and Nutrition Board of the Institute of Medicine, 2 mg of Mn per day constitute an adequate intake for pregnant women, with the tolerable upper intake level of 11 mg (Food and Nutrition Board, Institute of Medicine, 2001). Notwithstanding its protective effect against oxidative damage, in larger quantities Mn becomes an oxidizer itself (ATSDR, 2012) and is a known neurotoxicant. Mn accumulates in mitochondria-rich membranes and penetrates the blood-brain and placental barriers. Evidence of the neurotoxic effects of Mn is increasing, specifically with regard to cognition, memory, behavior and motor function (Claus Henn et al., 2012, 2010; Hernández-Bonilla et al., 2011; Menezes-Filho et al., 2011, 2009; Rodríguez-Barranco et al., 2013). Several studies have examined the neurodevelopmental effects of intrauterine Mn exposure, and results suggest that environmental exposure to Mn in utero may affect the psychomotor development of children at an early age (Gunier et al., 2015; Lin et al., 2013a; Mora et al., 2015a).

Perinatal depression has also been identified as a risk factor for inadequate infant neurodevelopment in the long term (Dossett, 2008). According to the World Health Organization, depression is a disease that alters the mood, thoughts, appetite, sleep, perception of self-esteem and overall lifestyle of individuals (WHO, 2016). Feeling sad, apathetic and hurt, those who suffer from this emotional disorder find it difficult to interact with the environment (Lara et al., 2006). Prenatal depression is differentiated by its adverse effects on the baby, whose neuronal development during this period is directly affected by context (Olhaberry et al., 2013).

Research performed during outpatient consultations at the National Institute of Perinatology indicated that as many as 21.7% of pregnant women in Mexico may experience probable depression episodes (Ortega et al., 2001). Another study revealed a 23.3% prevalence of depression among pregnant users of primary healthcare services at Family Medicine Unit 171 of the Mexican Social Security Institute (Instituto Mexicano del Seguro Social, IMSS) (Delgado-Quiñones et al., 2015). A recent study in the US estimated that the prevalence of minor depression was higher among pregnant women (16.6%) compared to non-pregnant women (11.4%) (adjusted PR=1.5, [95% confidence interval (CI): 1.2, 1.9]). Studies have found that depression occurs particularly during weeks 6 to 10 and in the third trimester of pregnancy, when the body prepares for labor and delivery (Carter and Kostara, 2005).

Depression can affect the central nervous system through anomalous levels of aminergic neurotransmitters (e.g., serotonin, norepinephrine and dopamine) acting upon the neurons of the central nervous system (Escobar Izquierdo et al., 2009). Mn is essential to the proper functioning of the hypothalamus, a region of the brain containing a large number of neurons. One of the primary functions of the hypothalamus is to link the nervous and endocrine systems by secreting neurohormones through the pituitary gland. The molecular mechanisms by which Mn affects hypothalamic processes are not fully understood, therefore determining exactly which mechanisms are at play in the interaction between Mn and depression and consequently child neurodevelopment is not completely clear (Tellerias and Paris, 2008). However, both could be acting on the same dopamine pathway and epidemiologic evidence is scarce.

Because both Mn and depression are common exposures in pregnancy, we explored their joint associations with children’s neurodevelopment at 24 months of age. We hypothesized that depression could be an effect modifier of the association between Mn and neurodevelopment.

2. Methods

2.1 Study population

The Programming Research in Obesity, Growth, Environment and Social Stressors (PROGRESS) birth cohort recruited women during their prenatal visit at four IMSS clinics in Mexico City, between 2007 and 2011 and followed them since their 2nd trimester. Inclusion criteria have been previously described in more detail (Braun et al., 2014). Briefly, women had to be healthy, at least 18 years old, between 12-24 weeks pregnant, and plan to live in Mexico City. All protocols were approved by the institutional review boards of the Icahn School of Medicine at Mount Sinai, the Harvard T.H. Chan School of Public Health, the National Institute of Public Health Mexico, the National Institute of Perinatology Mexico, and the Mexican Social Security System. Upon consent, women visited our research centers during their 2nd and 3rd trimesters, we collected samples and data at birth and 760 mother-infant pairs visited our research centers at 6, 12, 18 and 24 months postpartum. Our study analyzed data from 541 dyads for which a neurodevelopmental assessment was administered to the child at the 24 month visit. We excluded from analyses children who were born at ≤32 weeks and/or weighed ≤1,500 g (n=4).

2.2 Child neurodevelopment

Neurodevelopment was assessed at 24 months of age using the Bayley Scales of Infant and Toddler Development 3rd edition (Albers and Grieve, 2007). Trained psychologists blind to the prenatal Mn and depression levels evaluated the cognitive, language and motor development of children.

2.3 Maternal gestational depression

The Edinburgh Depression Scale (EDS) was administered during the 3rd trimester of pregnancy. The questionnaire consists of ten polytomous (four-option) response items exploring last-week symptoms of a major depression episode. Each question is rated on a scale of 0 to 3 points. Studies have found that a cut-off of 12 indicates a probable depressive disorder (Evans et al., 2001; Murray and Cox, 1990). The scale was applied in a Chilean cohort (Hispanic population, closer to our study population) and found that a cut-off between 12 and 13 had an 87.4% correct classification of cases and non-cases (Alvarado et al., 2012). For our main analyses we had a conservative approach and used a dichotomous variable with a cutoff of ≥13 for depressive symptoms. We also explored using quartiles of the EDS for a secondary analysis.

2.4 Prenatal manganese

Mn was measured in maternal venous blood samples in the 3rd trimester (between the 30th and 34th week) of pregnancy, from here on referred to as BMn and during birth in umbilical cord blood (CMn). Royal blue trace metal Vacutainer tubes (Becton-Dickinson and Company, Franklin Lakes, New Jersey) containing EDTA were used to collect the samples which were stored at 4 °C until shipped. They were hence stored at −20 °C until analyzed with a mass spectrometer (Elan 6100; PerkinElmer, Norwalk, CT) at the Trace Element Lab of the Harvard T.H. Chan School of Public Health. All of the samples exceeded the 0.09 μg/dL limit of detection.

2.5 Covariates

We selected the following covariates for inclusion in models based on previous literature: infant sex, birth weight (kg), gestational age (weeks) and maternal education (years of schooling), age at delivery (years), marital status (with/without a partner), 3rd trimester blood lead (μg/dL collected using the same methods described above for Mn) and the HOME (Home Observation Measurement of the Environment) score at 24 months of age (Chung et al., 2015; Lin et al., 2013b; Mora et al., 2015b; Nulman et al., 2012).

Without considering the HOME evaluation, complete covariate information were obtained from 473 and 308 mother-child pairs for the BMn and CMn analysis respectively. Adjusting for the HOME score reduced the sample size to 341 and 214 respectively and did not change the main effect estimates by more than 10%, therefore our final models did not include this covariate.

2.6 Statistical analysis

We first examined the distribution and summary statistics for all variables using univariate and bivariate regression models. We ran separate analyses for BMn and CMn concentrations. The distributions for both BMn and CMn were slightly skewed to the right but were included without any transformation in the models. We included both Mn variables as continuous in analyses. Neurodevelopment scores were normally distributed and used as continuous variables, as were the rest of the covariates except for depression symptoms, child’s sex and maternal marital status which were analyzed as dichotomous variables.

To assess non-linearity of the association between each variable and neurodevelopment, we first used lowess graphs for visual examination. All covariates showed a linear relationship, however both BMn and CMn suggested a quadratic relation. We then generated models including a quadratic term for BMn and CMn, separately. The effect coefficients were only statistically significant for CMn. We therefore decided to assume the association between BMn and neurodevelopment scores as linear and as quadratic for CMn.

We used multivariate linear regression models, an extension of the multiple regression model that allows quantifying relations between the multiple response and explanatory variables while controlling for covariates, taking into account residual correlations. Therefore, a single model with three response variables (cognitive, language and motor scores) rather than three independent models was preferred. All models included either BMn or CMn, depressive symptoms and were adjusted for covariates. We applied t-tests to detect inter-group variance, specifically in terms of significant neurodevelopmental differences by maternal depression.

In order to examine a possible interaction between Mn and depression, we included an interaction term in the adjusted models. If the interaction coefficients were statistically significant we also stratified our models by depression in order to assess its possible role as an effect modifier of the Mn-neurodevelopment association. We verified that model assumptions had been met. Lastly, we constructed graphs illustrating the association between Mn, depression and neurodevelopment scores by saving the predicted values of the adjusted models and graphed by depressive symptoms (yes/no). As a secondary analysis in order to verify the cut-off point for depression, we included the EDS as quartiles in our models for each Mn predictor. All statistical analyses were performed with STATA 13 software.

3. Results

The mean (SD) BMn and CMn concentrations were 27.7 μg/L (8.7) and 50.1 (16.5) μg/L respectively. The Spearman correlation between them was −0.02 with a p=0.68. On a scale of 0 to 28, our study participants averaged 8.6 SD (5.8) (median=8, IQR=9) points on the EDS questionnaire. Based on the EDS cutoff point (≥13), 26.5% of pregnant women were found to suffer from depressive symptoms, and quartiles were distributed as follows: 1st quartile from 0–4 points, 2nd quartiles 5–8 points, 3rd quartile 9–13 points and 4th quartile 14–28 points on the EDS scale. Characteristics of our analytic sample participants and non-participants can be found in Table 1. None of the differences between groups were statistically significant. Spearman correlations with a p<0.05 level of significance were observed between neurodevelopment scores: 0.57 between the language and cognitive functions, 0.45 between the language and motor functions, and 0.50 between the cognitive and motor functions. BMn and depression scores had a Spearman correlation of 0.05 (p=0.26). Mn (maternal or cord) and depression symptoms in pregnancy had independently, statistically significant negative associations with neurodevelopment scores in unadjusted models.

Table 1.

Characteristics of analytic sample participants and non-participants

Participants
n=473		 Non-participants
n=287
Children mean (SD) mean (SD)
Cord blood manganese (μg/L)a 50.1 (16.5) 51.9 (19.4)
Neurodevelopment – Bayley IIIb
 Cognitive 92.5 (8.7) 90.8 (6.8)
 Language 90.0 (9.3) 87.3 (7.4)
 Motor 94.1 (9.5) 92.5 (8.5)
Sex: Male (%) 51.7 55.6
Birth weight (kg) 3.1 (0.4) 3.0 (0.4)
HOMEc 31.7 (5.5) 30.9 (5.1)
Gestational age (weeks) 38.8 (1.3) 40.2 (1.2)
Mother
3T blood manganese (μg/L) d 27.7 (8.7) 26.4 (8.7)
Symptoms of depression: Yes (%) e 26 28
Age at delivery (years) 27.0 (5.5) 27.4 (5.2)
Education (total years of schooling) 11.9 (2.7) 11.8 (2.9)
Marital status: With partner (%) 80.7 80.5
3rd trimester blood lead (μg/dL) 4.1 (2.8) 3.9 (2.6)
Open in a new tab
a

Cord Blood Mn: participants =308 non-participants = 125

b

Bayley III: non-participants = 64

c

HOME: participants =341 non-participants = 106

d

3T Mn non-participants n=201

e

Depression non-participants: n=183

Results from our model using BMn and depression are shown in table 2. On average, 1 μg/L increase in maternal blood manganese (BMn) and the presence of depression symptoms in pregnancy were associated respectively, with a decrease of 0.12 (p<0.01) and 2.40 (p<0.01) points on cognitive scores, 0.20 (p<0.01) and 2.47 (p=0.01) points on language scores, and 0.10 (p=0.04) and 1.01(p=0.19) points on motor scores adjusting for covariates (infant sex, birth weight, gestational age and maternal education, age at delivery, marital status, 3rd trimester blood lead). In the model including an interaction term between BMn and depressive symptoms, the coefficients were statistically significant for language scores (cognitive β=−0.08 p=0.25, language β=−0.25 p<0.01, motor β=−0.10 p=0.89). The results of the model stratified by depression suggest an effect modification of the association between maternal blood BMn and neurodevelopment (Table 3). On average, language scores for children whose mothers were depressed during pregnancy were lower than for children whose mothers were not depressed during pregnancy (Figure 1).

Table 2.

Multivariate regression model for 24 months of age neurodevelopmental scores and maternal 3rd trimester blood Mn (μg/L) (n=473).

Cognitive Language Motor
β (SE) p β (SE) p β (SE) p
3rd trimester blood Mn −0.12 (0.0) <0.01 −0.20 (0.0) <0.01 −0.10 (0.0) 0.04
Depression: Yes −2.40 (0.8) <0.01 −2.47 (0.9) 0.01 −1.01 (0.8) 0.19
Sex: Male −2.34 (0.7) <0.01 −3.43 (0.8) <0.01 −2.47 (0.8) < 0.01
Birth weight   1.31 (1.0) 0.09   2.34 (1.0) 0.03   1.01 (1.1) 0.38
Gestational age (weeks)   0.00 (0.3) 0.98   0.21 (0.3) 0.53   0.51 (0.3) 0.17
Marital status (with partner) −0.73 (0.9) 0.38   0.37 (1.0) 0.65 −1.05 (1.1) 0.30
Maternal age   0.03 (0.0) 0.76 −0.07 (0.1) 0.64 −0.01 (0.0) 0.86
Education (yrs)   0.55 (0.1) <0.01   0.52 (0.2) <0.01   0.15 (0.1) 0.36
Blood lead (μg/dL) −0.97(0.1) 0.48 −0.13 (0.1) 0.26 −0.01 (0.1) 0.83
Open in a new tab

Table 3.

Adjusted multivariate regression model coefficients for neurodevelopmental scores and maternal 3rd trimester blood Mn stratified by depression (n=473, with depression n=124, 26.3%)

Cognitive Language Motor
Depression β [95% CI] β [95% CI] β [95% CI]
No −0.15 (−0.25 to −0.04) −0.18 (−0.29 to −0.07) −0.14 (−0.25 to −0.04)
Yes −0.13 (−0.32 to −0.06) −0.27 (−0.46 to −0.08) −0.13 (−0.22 to 0.10)
p interaction 0.25 <0.01 0.89
Open in a new tab

Figure 1.

Figure 1

Open in a new tab

Graphical depiction of depression (women with depression n=124) as an effect modifier of the association between 3rd trimester manganese blood (n=473) and 24 month neurodevelopmental Bayley scores.

Results from the analysis using CMn are shown in Table 4. Mean curves for the adjusted association between CMn and neurodevelopment scores had an inverted U shape. The interaction term between CMn and language scores was statistically significant (not for cognitive or motor scores). In stratified analyses by depression, CMn was most strongly associated with language scores (Figure 2). Cognitive scores also differed depending on depression; however, associations were not statistically significant.

Table 4.

Multivariate regression model for 24 months of age neurodevelopment and cord blood manganese (μg/L) (n=307).

Cognitive Language Motor
β (SE) p β (SE) p β (SE) p
Cord manganese   0.38 (.15) 0.01   .533 (.16) <0.01   .345 (.16) 0.03
Cord manganese2 −.003 (.0) 0.01 −.0046 (.00) <0.01 −.0028 (0.0) 0.05
Depression: Yes −2.2 (1.1) 0.06 −2.17 (1.2) 0.08 −0.92 (1.1) 0.35
Sex: Male −2.6 (0.9) 0.01 −3.93 (1.0) <0.01 −3.16 (1.0) <0.01
Birth weight   1.6 (1.2) 0.01   3.38 (1.4) 0.01   0.76 (1.5) 0.56
Gestational age −0.1 (0.3) 0.85 −0.15 (0.4) 0.73   0.08 (0.4) 0.83
 Marital status (w/partner) −0.0 (1.2) 0.99   0.26 (1.4) 0.84 −0.42 (1.4) 0.73
Maternal age   0.0 (0.0) 0.95 −0.04 (0.1) 0.61 −0.11 (0.1) 0.23
Education (total years)   0.6 (0.1) 0.04   0.61 (0.2) 0.05   0.14 (0.2) 0.46
Blood Lead (μg/dL) −0.3 (0.1) 0.06 −0.24 (0.1) 0.17 −0.25 (0.1) 0.15
Open in a new tab

Figure 2.

Figure 2

Open in a new tab

Graphical depiction of depression (women with depression n=124) as an effect modifier of the association between cord blood manganese (n=307) and 24 month neurodevelopmental Bayley scores.

Since our models showed a consistently strong association with language scores and both BMn and CMn, as well as an effect modification by depression, we generated separate models looking at the receptive (comprehension) and expressive (ability to use vocabulary and words to sentences) components of the language score. We found that receptive language score ranged from 2 to 16 with a mean of 8.9 (95% CI: 8.76, 9.07). On average, 1 μg/L increase of BMn was associated with a decrease of 0.02 receptive language points for children whose mothers did not have depressive symptoms during pregnancy (p=0.02), compared to a decrease of 0.05 receptive language points for children whose mothers had depressive symptoms during pregnancy (p=0.02). The interaction term between BMn and depression for receptive language had a p<0.01. We also saw a negative association between BMn and expressive language, however there was no difference between children born to women with or without depressive symptoms during pregnancy (p for interaction=0.09).

Results from the secondary analysis using EDS quartiles in adjusted models for BMn comparing the 4th EDS quartile to the 1st showed: a decrease of 3.8 cognitive points (p<0.01), a decrease of 2.6 language points (p=0.03) and a decrease of 2.34 motor points (p=0.07). Associations for the 2nd and 3rd quartiles compared to the 1st were not statistically significant. For the CMn analysis, comparing the 4th EDS quartile to the 1st we saw: a decrease of 4.4 cognitive points (p<0.01), a decrease of 2.2 language points (p=0.15) and a decrease of 2.9 motor points (p=0.06). Negative associations were also seen comparing the 2nd quartile to the 1st for cognitive scores (decrease of 2.9 points, p=0.03) and for motor scores (decrease of 3.2 points, p=0.02. All other associations were not statistically significant but in the same negative direction.

4. Discussion

Our findings support studies showing a negative association between prenatal (in-utero) exposure to Mn and 24 month neurodevelopment (Lin et al., 2013b). Likewise, they are in line with the literature of maternal depressive symptoms during this period and reduced cognitive, language and motor development in their children (Field, 2011; Nulman et al., 2012). After adjusting for covariates, we observed the strongest negative associations for Mn with language scores, compared to associations with cognitive and motor scores. Furthermore, only the association between BMn and language was modified by depression. Together our results suggest that the prenatal mechanism by which language is developed is affected by the co-exposure to Mn and depression and should be further explored.

Although manganese is an essential metal for our bodily functions, there is evidence that both insufficiently low and excessively high levels are harmful to our health (O’Neal and Zheng, 2015; Shan et al., 2016). We explored the effects of maternal Mn deficiency (through 3rd trimester blood) on infant neurodevelopment; however we did not find a negative quadratic relationship, but rather a negative linear association for higher maternal Mn concentrations. These findings are similar to those seen in the Tar Creek superfund site where authors found a negative linear association for maternal blood Mn but null association for cord blood Mn (Claus Henn et al., 2017). By contrast, we found a negative quadratic relationship between CMn and neurodevelopment scores, consistent with findings from other studies of early childhood blood Mn (Claus Henn et al., 2010). In their study of infant neurodevelopment, Chung et al. collected maternal blood just before delivery and found the same inverted U shape association (Chung et al., 2015). The difference in the shape of the functions may be reflecting the rapid growth of the infant’s brain in the last gestational stage, with a higher nutritional need for manganese. In our study, CMn might be illustrating this exposure time window better than BMn.

Mean BMn levels found in our study are higher, than those reported by Oulhoute et al from the NHANES 2001–2012 (27.6 vs 12.5 μg/L respectively) (Oulhote et al., 2014). This could be due to the timing during pregnancy in the collection of the sample, which is unspecified in the article. Compared to concentrations found by other studies, our results for BMn and are closer to those found by by Zota et al in Tar Creek, BMn: 24 μg/L or by Mora et al in Costa Rica banana plantation workers: BMn 24.4 μg/L, and lower than those found by Guan in China BMn: 54.98 μg/L. With respect to CMn, our results (mean 50.1 μg/L) were lower than those found by Guan: 78.75 μg/L, closer to those found by Zota: 42 μg/L, but higher than those found by Tasker at el in Southwest Quebec of 34.4 μg/L (Guan et al., 2014; Mora et al., 2014; Takser et al., 2004; Zota et al., 2009). These differences could be due to a variety of reasons including exposure to Mn in the different populations, the timing in collection of the sample, or even genetic/epigenetic characteristics of the specific populations.

We found a weak and non-significant correlation between BMn and CMn, these finding are in line with those reported by Rudge et al who studied the placental permeability of Mn and found twice as high BMn concentrations than CMn, as in our study (Rudge et al., 2009). However many other studies have found positive correlations (Guan et al., 2014; Takser et al., 2004; Zota et al., 2009). One important difference is the timing of the sample collection, in our study BMn was collected in the third trimester, whereas the mentioned studies collected BMn at the moment of birth. Maternal blood Mn during pregnancy may not adequately represent levels in the fetus, which may be due to immature fetal liver function and Mn tending to accumulate (Yoon et al., 2011). In line with the above, we saw a positive correlation (Spearman 0.16, p<0.01) between CMn and Mn in samples of maternal blood collected within the 24 hours after delivery, which we did not consider in our analysis to respect the timing of exposure (postnatal Mn levels may not be an adequate reflection of gestational exposure, even if only 24 hours had elapsed).

Our finding of significant associations between maternal blood Mn and child neurodevelopment may indicate a role for the placenta in influencing child neurodevelopment. The different ranges of concentrations for maternal vs. cord blood might also be influencing the ability to find an inverted U shaped association since BMn ranged from 9.8–50.3 μg/L and CMn ranged from 11.5–106.6 μg/L.

Higher Mn blood levels might be a reflection of women’s exposure to higher concentrations of Mn in air, water or food, however all the women in our cohort lived in Mexico City throughout the period of this study, therefore, although we do not have a manganese exposure assessment, the potential for bias due to difference in exposure is unlikely. The correlation between BMn and depression scores was very small (Spearman 0.05) and not statistically significant, so it is unlikely that depression could have a direct association with maternal levels of Mn. Another possible explanation for the difference in Mn blood concentrations might be due to Mn regulation resulting from microbiome composition (Dunlop et al., 2015)or other inherent characteristics (genetic or epigenetic (Maccani et al., 2015)), which could result in a different biliary excretion and GI absorption; these were unmeasured in our study and if these mechanisms were also related to the neurodevelopmental processes, this could be a potential source of confounding in our study.

Gestational depression often persists after childbirth, impairing mother-child bonding and interfering with the role that mothers normally perform in raising their children (Field, 2010). Depressive symptoms in mothers can obstruct the establishment of an affective connection with their newborns, which is a major contributing factor to child neurodevelopment (Koutra et al., 2013). Although we did not explore postnatal depression in this study, when looking at the receptive and expressive language scores separately, we saw a difference in the association depending on depressive symptoms only for receptive scores. A depressed mother likely speaks less to her child, providing less support for the child’s development of language skills. Less verbal input and interaction might have less impact on a child’s motor and cognitive development (Stein et al., 2008; Tse et al., 2010).

The results from our secondary analysis using quartiles of EDS in adjusted models, supported the associations found with the 13 point cut-off. In fact, when comparing the 4th to the 1st quartile of EDS, the associations with the different cognitive, language and motor scores the associations increased in magnitude and improved statistical significance. All other associations were not statistically significant with exception of the CMn associations for cognitive and motor scores comparing the 2nd to the 1st EDS quartiles which also showed a decrease in the scores.

Our study has several strengths: it is longitudinal in design, with data from the PROGRESS birth cohort which includes biomarkers as well as psychosocial measures from prenatal stages and follow-up of mother-infant pairs. Another strength of our work was the use of multivariate multiple regression models, which allowed analysis of the three neurodevelopment scores simultaneously.

The reduced analytic sample due to missing data on cord blood is a limitation of our study. Missing information on the HOME evaluation also limited our ability to adjust for this covariate, however our results remained without its inclusion in analyses.

5. Conclusions

More in-depth research is now required to determine whether Mn toxicity increases among children of depressed mothers. Our findings highlight the importance of understanding the effects of co-exposure to these conditions, particularly during sensitive development stages.

Prenatal care is important because of its basic contribution to adequate neurodevelopment in the offspring. Over a quarter of our study participants had depressive symptoms in pregnancy; this highlights a need for special attention must be paid to the emotional state of the expectant mother, alongside the nutrients she requires to ensure a stable pregnancy without complications or deficiencies.

It is important for decision makers to promote strategies aimed specifically at preventing, detecting and treating maternal depression across perinatal public health services in Mexico. Public policy also needs to address the growing evidence of environmental Mn toxicity and the irreversible short- and long-term effects of intrauterine exposure, particularly on neurodevelopment.

Table 5.

Adjusted multivariate regression model coefficients for 24-month neurodevelopment scores and cord manganese (μg/L) stratified by depression (n=307)

Cognitive Language Motor
Depression: No (n=235) β (SE) p β (SE) p β (SE) p
Cord manganese 0.32 (.17) 0.05 0.48 (.20) 0.01 0.36 (.19) 0.06
Cord manganese2 −.003 (0.0) 0.05 −.0042 (.00) 0.01 −.0029(0.0) 0.08
Depression: Yes (n=72)
Cord manganese .501 (.33) 0.14 .781 (.28) <0.01 .28 (.31) 0.38
Cord manganese2 −.004 (.00) 0.17 −.0072 (.00) <0.01 −.0027 (0.0) 0.45
p interaction Mn 0.14 <0.01 0.38
p interaction Mn2 0.17 <0.01 0.45
Open in a new tab

Highlights.

  • Maternal blood Mn was negatively linearly associated with neurodevelopment

  • Cord blood Mn had an inverted U-shape association with neurodevelopment

  • Interaction between Mn and maternal gestational depressive symptoms was found

  • Language was more strongly negatively associated by Mn if depressive symptoms were present

  • Negative associations were significant for receptive vs expressive language

Acknowledgments

We thank the Centro Médico ABC and the National Institute of Perinatology, México for their support with this research. Authors from INSP are members of the Mexican Network for Children’s Environmental Health.

Funding Sources

This work was supported by NIH grants R01ES013744, R01ES014930, R01ES021357 and P30ES023515. It was also supported and partially funded by the National Institute of Public Health/Ministry of Health of Mexico.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Competing financial interest’s declaration: All authors have no actual or potential competing interests regarding the submitted article and the nature of those interests.

References

  1. Albers CA, Grieve AJ. Test Review: Bayley, N. (2006). Bayley Scales of Infant and Toddler Development- Third Edition San Antonio, TX: Harcourt Assessment. J Psychoeduc Assess. 2007;25:180–190. doi: 10.1177/0734282906297199. [DOI] [Google Scholar]
  2. Alvarado R, Guajardo V, Rojas G, Jadresic E. Validación de la escala de Edmiburgo para embarazadas. 2012 http://web.minsal.cl/sites/default/files/files/InformeFinalproyectovalidacionEPDSembarazadasV2.pdf.
  3. Aschner M, Aschner JL. Manganese neurotoxicity: cellular effects and blood-brain barrier transport. Neurosci Biobehav Rev. 1991;15:333–40. doi: 10.1016/s0149-7634(05)80026-0. [DOI] [PubMed] [Google Scholar]
  4. ATSDR. Toxicological profile for manganese :2012. [Google Scholar]
  5. Braun JM, Wright RJ, Just AC, Power MC, Tamayo Y Ortiz M, Schnaas L, Hu H, Wright RO, Tellez-Rojo MM. Relationships between lead biomarkers and diurnal salivary cortisol indices in pregnant women from Mexico City: a cross-sectional study. Environ Health. 2014;13:50. doi: 10.1186/1476-069X-13-50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bueno-Brito C, Sánchez-Ramos A, Armenta-Solís A, González-Vera E. Contenido de plomo y manganeso en despachadores de gasolina. Bioquimia. 2005;30:41–46. [Google Scholar]
  7. Carter D, Kostara X. Psychiatric disorders in pregnancy. BC Med J. 2005;47 [Google Scholar]
  8. Chung SE, Cheong HK, Ha EH, Kim BN, Ha M, Kim Y, Hong YC, Park H, Oh SY. Maternal Blood Manganese and Early Neurodevelopment: The Mothers and Children’s Environmental Health (MOCEH) Study. Environ Health Perspect. 2015 doi: 10.1289/ehp.1307865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Claus Henn B, Bellinger D, H M, Coull B. Prenatal manganese exposure and early childhood neurodevelopment among residents near a mining-impacted Superfund site. Env Heal Perspect. 2017 doi: 10.1289/EHP925. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Claus Henn B, Ettinger AS, Schwartz J, Téllez-Rojo MM, Lamadrid-Figueroa H, Hernández-Avila M, Schnaas L, Amarasiriwardena C, Bellinger DC, Hu H, Wright RO. Early postnatal blood manganese levels and children’s neurodevelopment. Epidemiology. 2010;21:433–9. doi: 10.1097/ede.0b013e3181df8e52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Claus Henn B, Schnaas L, Ettinger AS, Schwartz J, Lamadrid-Figueroa H, Hernández-Avila M, Amarasiriwardena C, Hu H, Bellinger DC, Wright RO, Téllez-Rojo MM. Associations of early childhood manganese and lead coexposure with neurodevelopment. Environ Health Perspect. 2012;120:126–31. doi: 10.1289/ehp.1003300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Coetzee DJ, McGovern PM, Rao R, Harnack LJ, Georgieff MK, Stepanov I. Measuring the impact of manganese exposure on children’s neurodevelopment: advances and research gaps in biomarker-based approaches. Environ Health. 2016;15:91. doi: 10.1186/s12940-016-0174-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Delgado-Quiñones EG, López-Trejo LA, Mariscal-Rivera CE, HernándezRivera LN, Orozco-Sandoval R. Prevalencia de depresión en embarazadas en primer nivel de atención. Rev Médica MD. 2015;6:237. [Google Scholar]
  14. Dossett EC. Perinatal Depression. Obstet Gynecol Clin North Am. 2008;35:419–434. doi: 10.1016/j.ogc.2008.04.004. [DOI] [PubMed] [Google Scholar]
  15. Dunlop AL, Mulle JG, Ferranti EP, Edwards S, Dunn AB, Corwin EJ. Maternal Microbiome and Pregnancy Outcomes That Impact Infant Health: A Review. Adv Neonatal Care. 2015;15:377–85. doi: 10.1097/ANC.0000000000000218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Escobar Izquierdo A, Guadarrama L, Zhang L. Bases neuroquímicas y neuroanatómicas de la depresión. Rev la Fac Med. 2009;49 [Google Scholar]
  17. Evans J, Heron J, Francomb H, Oke S, Golding J. Cohort study of depressed mood during pregnancy and after childbirth. BMJ. 2001;323 doi: 10.1136/bmj.323.7307.257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Field T. Prenatal depression effects on early development: A review. Infant Behav Dev. 2011;34:1–14. doi: 10.1016/j.infbeh.2010.09.008. [DOI] [PubMed] [Google Scholar]
  19. Field T. Postpartum depression effects on early interactions, parenting, and safety practices: A review. Infant Behav Dev. 2010;33:1–6. doi: 10.1016/j.infbeh.2009.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Food and Nutrition Board, Institute of Medicine, N.A. Dietary Reference Intakes (DRIs): Recommended Dietary Allowances and Adequate Intakes, Elements [WWW Document] 2001 URL http://www.nationalacademies.org/hmd/~/media/Files/ActivityFiles/Nutrition/DRI-Tables/2_RDAandAIValues_VitaminandElements.pdf?la=en (accessed 11.10.16)
  21. Freeland-Graves JH, Turnlund JR. Deliberations and evaluations of the approaches, endpoints and paradigms for manganese and molybdenum dietary recommendations. J Nutr. 1996;126:2435S–2440S. doi: 10.1093/jn/126.suppl_9.2435S. [DOI] [PubMed] [Google Scholar]
  22. Glover V. Maternal depression, anxiety and stress during pregnancy and child outcome; what needs to be done. Best Pract Res Clin Obstet Gynaecol. 2014;28:25–35. doi: 10.1016/j.bpobgyn.2013.08.017. [DOI] [PubMed] [Google Scholar]
  23. Guan H, Wang M, Li X, Piao F, Li Q, Xu L, Kitamura F, Yokoyama K. Manganese concentrations in maternal and umbilical cord blood: related to birth size and environmental factors. Eur J Public Health. 2014;24:150–157. doi: 10.1093/eurpub/ckt033. [DOI] [PubMed] [Google Scholar]
  24. Gulson B, Mizon K, Taylor A, Korsch M, Stauber J, Davis JM, Louie H, Wu M, Swan H. Changes in manganese and lead in the environment and young children associated with the introduction of methylcyclopentadienyl manganese tricarbonyl in gasoline–preliminary results. Environ Res. 2006;100:100–114. doi: 10.1016/j.envres.2005.03.013. [DOI] [PubMed] [Google Scholar]
  25. Gunier RB, Arora M, Jerrett M, Bradman A, Harley KG, Mora AM, Kogut K, Hubbard A, Austin C, Holland N, Eskenazi B. Manganese in teeth and neurodevelopment in young Mexican–American children. Environ Res. 2015;142:688–695. doi: 10.1016/j.envres.2015.09.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hernández-Bonilla D, Schilmann A, Montes S, Rodríguez-Agudelo Y, Rodríguez-Dozal S, Solís-Vivanco R, Ríos C, Riojas-Rodríguez H. Environmental exposure to manganese and motor function of children in Mexico. Neurotoxicology. 2011;32:615–621. doi: 10.1016/j.neuro.2011.07.010. [DOI] [PubMed] [Google Scholar]
  27. Koutra K, Chatzi L, Bagkeris M, Vassilaki M, Bitsios P, Kogevinas M. Antenatal and postnatal maternal mental health as determinants of infant neurodevelopment at 18 months of age in a mother–child cohort (Rhea Study) in Crete, Greece. Soc Psychiatry Psychiatr Epidemiol. 2013;48:1335–1345. doi: 10.1007/s00127-012-0636-0. [DOI] [PubMed] [Google Scholar]
  28. Lara MA, Navarro C, Navarrete L, Cabrera A, Almanza J, Morales F, Juárez F. Síntomas depresivos en el embarazo y factores asociados en pacientes de tres instituciones de salud de la ciudad de México. Salud Ment. 2006;29:55–62. [Google Scholar]
  29. Lin CC, Chen YC, Su FC, Lin CM, Liao HF, Hwang YH, Hsieh WS, Jeng SF, Su YN, Chen PC. In utero exposure to environmental lead and manganese and neurodevelopment at 2 years of age. Environ Res. 2013a;123:52–7. doi: 10.1016/j.envres.2013.03.003. [DOI] [PubMed] [Google Scholar]
  30. Lin CC, Chen YC, Su FC, Lin CM, Liao HF, Hwang YH, Hsieh WS, Jeng SF, Su YN, Chen PC. In utero exposure to environmental lead and manganese and neurodevelopment at 2 years of age. Environ Res. 2013b;123:52–57. doi: 10.1016/j.envres.2013.03.003. [DOI] [PubMed] [Google Scholar]
  31. Maccani JZJ, Koestler DC, Houseman EA, Armstrong DA, Marsit CJ, Kelsey KT. DNA methylation changes in the placenta are associated with fetal manganese exposure. Reprod Toxicol. 2015;57:43–9. doi: 10.1016/j.reprotox.2015.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Menezes-Filho JA, Bouchard M, Sarcinelli PdeN, Moreira JC. Manganese exposure and the neuropsychological effect on children and adolescents: a review. Rev Panam Salud Publica. 2009;26:541–8. doi: 10.1590/s1020-49892009001200010. [DOI] [PubMed] [Google Scholar]
  33. Menezes-Filho JA, Novaes CdeO, Moreira JC, Sarcinelli PN, Mergler D. Elevated manganese and cognitive performance in school-aged children and their mothers. Environ Res. 2011;111:156–163. doi: 10.1016/j.envres.2010.09.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Mora AM, Arora M, Harley KG, Kogut K, Parra K, Hernández-Bonilla D, Gunier RB, Bradman A, Smith DR, Eskenazi B. Prenatal and postnatal manganese teeth levels and neurodevelopment at 7, 9, and 10.5years in the CHAMACOS cohort. Environ Int. 2015a;84 doi: 10.1016/j.envint.2015.07.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Mora AM, Arora M, Harley KG, Kogut K, Parra K, Hernández-Bonilla D, Gunier RB, Bradman A, Smith DR, Eskenazi B. Prenatal and postnatal manganese teeth levels and neurodevelopment at 7, 9, and 10.5years in the CHAMACOS cohort. Environ Int. 2015b;84:39–54. doi: 10.1016/j.envint.2015.07.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Mora AM, van Wendel de Joode B, Mergler D, Córdoba L, Cano C, Quesada R, Smith DR, Menezes-Filho JA, Lundh T, Lindh CH, Bradman A, Eskenazi B. Blood and Hair Manganese Concentrations in Pregnant Women from the Infants’ Environmental Health Study (ISA) in Costa Rica. Environ Sci Technol. 2014;48:3467–3476. doi: 10.1021/es404279r. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Murray D, Cox JL. Screening for depression during pregnancy with the edinburgh depression scale (EDDS) J Reprod Infant Psychol. 1990;8:99–107. doi: 10.1080/02646839008403615. [DOI] [Google Scholar]
  38. Nulman I, Koren G, Rovet J, Barrera M, Pulver A, Streiner D, Feldman B. Neurodevelopment of Children Following Prenatal Exposure to Venlafaxine, Selective Serotonin Reuptake Inhibitors, or Untreated Maternal Depression. Am J Psychiatry. 2012;169:1165–1174. doi: 10.1176/appi.ajp.2012.11111721. [DOI] [PubMed] [Google Scholar]
  39. Nulman I, Koren G, Rovet J, Barrera M, Streiner DL, Feldman BM. Neurodevelopment of Children Prenatally Exposed to Selective Reuptake Inhibitor Antidepressants. J Clin Psychiatry. 2015;76:e842–e847. doi: 10.4088/JCP.14m09240. [DOI] [PubMed] [Google Scholar]
  40. O’Neal SL, Zheng W. Manganese Toxicity Upon Overexposure: a Decade in Review. Curr Environ Heal Reports. 2015;2:315–328. doi: 10.1007/s40572-015-0056-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Olhaberry M, Escobar M, San Cristóbal P, Santelices MP, Farkas C, Rojas G, Martínez V. Intervenciones psicológicas perinatales en depresión materna y vínculo madre-bebé: una revisión sistemática. Ter psicológica. 2013;31:249–261. doi: 10.4067/S0718-48082013000200011. [DOI] [Google Scholar]
  42. Ortega L, Lartigue T, Figueroa ME. Prevalencia de depresión, a través de la Escala de Depresión Perinatal de Edinburgh (EPDS), en una muestra de mujeres mexicanas embarazadas. Perinatol y Reprod Humana. 2001;15:11–20. [Google Scholar]
  43. Oulhote Y, Mergler D, Bouchard MF. Sex- and age-differences in blood manganese levels in the U.S. general population: national health and nutrition examination survey 2011–2012. Environ Health. 2014;13:87. doi: 10.1186/1476-069X-13-87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Rodríguez-Barranco M, Lacasaña M, Aguilar-Garduño C, Alguacil J, Gil F, González-Alzaga B, Rojas-García A. Association of arsenic, cadmium and manganese exposure with neurodevelopment and behavioural disorders in children: A systematic review and meta-analysis. Sci Total Environ. 2013;454:562–577. doi: 10.1016/j.scitotenv.2013.03.047. [DOI] [PubMed] [Google Scholar]
  45. Rudge CV, Röllin HB, Nogueira CM, Thomassen Y, Rudge MC, Odland JØ, Oskarsson A, Dietrich KN, Bornschein R, Greene T, Rothenberg SJ, Needleman HL, Schnaas L, Wasserman G, Graziano J, Roberts R. The placenta as a barrier for toxic and essential elements in paired maternal and cord blood samples of South African delivering women. J Environ Monit. 2009;11:1322. doi: 10.1039/b903805a. [DOI] [PubMed] [Google Scholar]
  46. Shan Z, Chen S, Sun T, Luo C, Guo Y, Yu X, Yang W, Hu FB, Liu L. U-Shaped Association between Plasma Manganese Levels and Type 2 Diabetes. Environ Health Perspect. 2016;124:1876–1881. doi: 10.1289/EHP176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Stein A, Malmberg LE, Sylva K, Barnes J, Leach P. The influence of maternal depression, caregiving, and socioeconomic status in the post-natal year on children’s language development. Child Care Health Dev. 2008;34:603–612. doi: 10.1111/j.1365-2214.2008.00837.x. [DOI] [PubMed] [Google Scholar]
  48. Takser L, Mergler D, de Grosbois S, Smargiassi A, Lafond J. Blood manganese content at birth and cord serum prolactin levels. Neurotoxicol Teratol. 2004;26:811–815. doi: 10.1016/j.ntt.2004.07.001. [DOI] [PubMed] [Google Scholar]
  49. Tellerias L, Paris E. Impacto de los tóxicos en el neurodesarrollo. Rev Chil pediatría. 2008;79:55–63. doi: 10.4067/S0370-41062008000700010. [DOI] [Google Scholar]
  50. Tse AC, Rich-Edwards JW, Rifas-Shiman SL, Gillman MW, Oken E. Association of maternal prenatal depressive symptoms with child cognition at age 3 years. Paediatr Perinat Epidemiol. 2010;24:232–40. doi: 10.1111/j.1365-3016.2010.01113.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. WHO, W.H.O. Depression [WWW Document] WHO; 2016. URL http://www.who.int/topics/depression/en/ (accessed 11.10.16) [Google Scholar]
  52. Yoon M, Schroeter JD, Nong A, Taylor MD, Dorman DC, Andersen ME, Clewell HJ. Physiologically Based Pharmacokinetic Modeling of Fetal and Neonatal Manganese Exposure in Humans: Describing Manganese Homeostasis during Development. Toxicol Sci. 2011;122:297–316. doi: 10.1093/toxsci/kfr141. [DOI] [PubMed] [Google Scholar]
  53. Zota AR, Ettinger AS, Bouchard M, Amarasiriwardena CJ, Schwartz J, Hu H, Wright RO. Maternal blood manganese levels and infant birth weight. Epidemiology. 2009;20:367–73. doi: 10.1097/EDE.0b013e31819b93c0. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES