Combined neurodevelopmental exposure to deltamethrin and corticosterone is associated with Nr3c1 hypermethylation in the midbrain of male mice

Abstract

Attention-Deficit Hyperactivity Disorder (ADHD) is one of the most common neurodevelopmental disorders and manifests inattention, hyperactivity, and impulsivity symptoms in childhood that can last throughout life. Genetic and environmental studies implicate the dopamine system in ADHD pathogenesis. Work from our group and that of others indicates that deltamethrin insecticide and stress exposure during neurodevelopment leads to alterations in dopamine function, and we hypothesized that exposure to both of these factors together would lead to synergistic effects on DNA methylation of key genes within the midbrain, a highly dopaminergic region, that could contribute to these findings. Through targeted next-generation sequencing of a panel of cortisol and dopamine pathway genes, we observed hypermethylation of the glucocorticoid receptor gene, Nr3c1, in the midbrain of C57/BL6N males in response to dual deltamethrin and corticosterone exposures during development. This is the first description of DNA methylation studies of Nr3c1 and key dopaminergic genes within the midbrain in response to a pyrethroid insecticide, corticosterone, and these two exposures together. Our results provide possible connections between environmental exposures that impact the dopamine system and the hypothalamic-pituitary-adrenal axis via changes in DNA methylation and provides new information about the presence of epigenetic effects in adulthood after exposure during neurodevelopment.

1. Introduction

Attention-Deficit Hyperactivity Disorder (ADHD) affects 7–10% of children, and is defined by inattention, hyperactivity, and impulsivity symptoms that appear by age 12 (Thomas et al., 2015; American Psychiatric, 2013). While the exact pathophysiology is unknown, studies reveal risk factors and neurotransmitter systems associated with ADHD pathogenesis and severity. Several genetic studies implicate key dopaminergic components in ADHD pathogenesis (Gizer et al., 2009). Gene variants in the dopamine receptor 5 gene (DRD5) predict age of ADHD onset, while variants in the dopamine transporter (DAT1) gene are associated with increased severity of hyperactivity and impulsivity symptoms (Akutagava-Martins et al., 2016; Maitra et al., 2016). Changes in DAT expression are also associated with ADHD (Volkow et al., 2002; Madras et al., 2005; Sakrikar et al., 2012), and first-line treatments of ADHD, such as methylphenidate, target dopamine and norepinephrine reuptake. Treatment efficacy is associated with DAT1 genotype in humans (Kasparbauer et al., 2015) as well. In mice, a Dat1 knockout exhibits increased locomotor activity and a Dat1 overexpressing model displays increased impulsivity (Gainetdinov et al., 1999; Salahpour et al., 2008; Leo and Gainetdinov, 2013; Gainetdinov and Caron, 2000; Efimova et al., 2016).

Chemical and psychosocial exposures such as maternal smoking, psychosocial stress, and low socioeconomic status are associated with ADHD (Thapar et al., 2013). We illustrated that neurodevelopmental exposure to deltamethrin, a pyrethroid insecticide, increases DAT and dopamine receptor 1 (DRD1) expression in the striatum and produced ADHD-like hyperactivity, inattention, and impulsivity behaviors (Richardson et al., 2015). Recently, we found that neurodevelopmental deltamethrin exposure is associated with decreased midbrain Pitx3 expression, decreased cortical tyrosine hydroxylase (TH), and increased dopamine uptake in the dorsal striatum (Vester et al., 2019). In children, we implicated elevated levels of pyrethroid metabolites in urine with ADHD (Richardson et al., 2015). Others found that this association was higher in boys and in children with hyperactive-impulsive symptoms (Wagner-Schuman et al., 2015). Pyrethroid insecticides are synthetic analogues of pyrethrins and are a derivative of the Chrysanthemum cinerariaefolium flower. Pyrethroids bind the α subunit of voltage-gated Na⁺ channels, holding them open to induce neuronal hyperexcitability (Klaassen, 2013). Type II pyrethroids, such as deltamethrin, also inhibit GABA_A receptors and voltage-gated chloride channels to potentiate excitability (Burr and Ray, 2004). Off-target effects on dopamine dynamics and DAT expression have been described by our group and others (Bloomquist et al., 2002; Gillette and Bloomquist, 2003; Elwan et al., 2006; Mubarak Hossain et al., 2006), and may be particularly important to study in understanding mechanisms of ADHD development.

Socioeconomic status is a significant psychosocial risk factor in ADHD (Russell et al., 2014; Russell et al., 2016). Potentially, the effect of a lower socioeconomic status on ADHD could be mediated by higher levels of psychosocial stress associated with factors affecting low socioeconomic status groups (Steptoe and Feldman, 2001; Baum et al., 1999; McEwen and Gianaros, 2010). Psychosocial stressors activate the hypothalamic-pituitary-adrenal (HPA) axis, which induces release of the major stress hormone cortisol (corticosterone in rodents) (McEwen and Tucker, 2011). Interestingly, chronic stress in rodent models leads to increased dopamine release in the medial prefrontal cortex via glucocorticoid effects in the ventral tegmental area (Butts and Phillips, 2013) and decreased DAT and DRD2 expression in the striatum (Lucas et al., 2004). Separation of rodent pups from their mothers induces stress, and these offspring later exhibit hyperactivity and inattention behaviors reversed by methylphenidate (Bock et al., 2016).

One important mechanism through which environmental exposures could impact long term neurobehavioral risk and neurodevelopment is DNA methylation (Everson et al., 2017; Stroud et al., 2014; Marsit, 2016; Zhang et al., 2016a). Our work illustrates the role of differential DNA methylation in human neurodevelopment (Paquette et al., 2015; Paquette et al., 2013; Paquette et al., 2016; Marsit et al., 2012; Maccani et al., 2013; Lesseur et al., 2014a; Lesseur et al., 2014b; Green et al., 2015; Bromer et al., 2013). We identified changes in transcription factors that regulate expression of key dopaminergic genes after neurodevelopmental exposure to pesticides, suggesting protein alterations may be mediated by changes at the level of the gene (Richardson et al., 2008; Richardson et al., 2006). Furthermore, exposure to chronic stress is associated with altered DNA methylation of genes involved in development of the dopamine circuit (Peter et al., 2016; Cunliffe, 2016; Sasagawa et al., 2017) and those related to cortisol response (Parent et al., 2017). In addition to stress exposures, others have demonstrated that exposure to persistent organic pollutants alters DNA methylation in humans and animal models (Rusiecki et al., 2008; Kim et al., 2010; Howard et al., 2016; Shutoh et al., 2009; Song et al., 2010; Song et al., 2011; Rusiecki et al., 2016). Lastly, lower DNA methylation of DRD4 is associated with an increase in ADHD symptoms in children at age 6 (van Mil et al., 2014), further suggesting DNA methylation is relevant to ADHD pathogenesis.

Neurodevelopment is vulnerable to environmental perturbations that have lasting effects on neuropsychiatric outcomes (Shen et al., 2007; Landrigan et al., 2010; Landrigan and Goldman, 2011; Grandjean and Landrigan, 2014). Given the importance of proper neurodevelopment, we sought to model real-world exposures that affect particularly vulnerable groups of children and contribute to ADHD risk. Low socioeconomic groups are exposed to increased levels of environmental contaminants (Brown, 1995), including pyrethroids (Lu et al., 2013; Julien et al., 2008; Wason et al., 2013). They also experience higher levels of stress (Steptoe and Feldman, 2001; Baum et al., 1999; McEwen and Gianaros, 2010) and are at increased risk of developing ADHD (Russell et al., 2014; Russell et al., 2016). DNA methylation could potentially occur in response to pyrethroid and stress exposure during critical periods of neurodevelopment. Our approach provides a novel platform to assess environmental factors involved in ADHD. We evaluated DNA methylation within genes that encode key dopaminergic components: catechol-o-methyltransferase (Comt), an enzyme responsible for dopamine metabolism; the dopamine transporter (Dat1), which is the primary mode of dopamine removal from the synapse; dopamine receptor 4 (Drd4); tyrosine hydroxylase (Th), the enzyme essential for dopamine synthesis; and the vesicular monoamine transporter (Vmat2), which packages dopamine into synaptic vesicles. We also assessed DNA methylation of two transcription factors necessary for proper dopamine neuron development: Nurr1 (nuclear-receptor related 1 protein; transcription factor) and Pitx3 (pituitary homeobox 3; transcription factor). Finally, we measured DNA methylation of Nr3c1 (nuclear receptor subfamily 3 group C member 1), which encodes the corticosterone-responsive glucocorticoid receptor. We hypothesized these exposures could be related to ADHD pathogenesis via their effects on the dopamine system, and that combined neurodevelopmental exposure to deltamethrin and chronic stress would be associated with synergistic changes in DNA methylation.

2. Materials & methods

2.1. Animals

Eight-week-old C57BL/6NCrl wild-type mice (Charles River Labs) received food and water ad libitum and were maintained on a 12:12 dark/light cycle. Females were dosed with 3 mg/kg deltamethrin (Sigma-Aldrich) every three days for two weeks prior to breeding in triplicate. Deltamethrin dosing continued during breeding and gestation. One male and one female mouse per litter were utilized for each experiment to reduce confounding due to litter effects. Throughout, an independent litter represents one sample (n = 1). All experiments were approved by the Institutional Animal Care and Use Committee at Emory University and were conducted in accordance with the National Institutes of Health Guide of Care and Use of Laboratory Animals.

2.2. Exposure paradigm

We utilized a deltamethrin dose of 3 mg/kg based on our previous study showing dopaminergic effects upon neurodevelopmental exposure to 3 mg/kg deltamethrin in male offspring, but not 0.3 mg/kg or 1 mg/kg (Richardson et al., 2015). Additionally, this 3 mg/kg dose falls between the developmental no observable adverse effect limit (NOAEL) of 12 mg/kg and no observable effect limit (NOEL) of 1 mg/kg determined by the United States Environmental Protection Agency (Kavlock R Fau - Chernoff et al., n.d.). Adult C57BL/6NCrl females were exposed to deltamethrin (N = 17) or vehicle (N = 20) every 3 days during gestation, lactation, and weaning at postnatal day (PND) 21. Deltamethrin was administered orally via corn oil dissolved in peanut butter every 3 days to minimize trauma to pregnant mice. A dose was placed in each mouse cage and the mouse was observed until they had completely consumed either the dose in its entirety. Corticosterone (CORT) hemisuccinate (Steraloids, Inc.) was dissolved in the drinking water to minimize handling stress and reduce variation in CORT levels seen in behavioral chronic stress paradigms (Gong et al., 2015; Yin et al., 2016). A CORT dose of 25 μg/mL was used because it was previously validated in a rodent neurobehavioral model (Gourley and Taylor, 2009). Offspring litters had continuous access to CORT (N = 15) dissolved in their drinking water, or drinking water vehicle (N = 22) from adolescence through adulthood (PND21–60) (Fig. 1). Animals in each exposure group had access to only one of two types of water at a time. Water bottles were weighed daily to assess intake and mice were weighed at weaning and at least once a week to assess proper weight gain. At PND60 offspring were sacrificed and midbrains dissected for DNA methylation and RNA expression studies.

Fig. 1.

Exposure paradigm timeline.

2.3. DNA and RNA sample isolation

Mice underwent rapid decapitation at 8–10 weeks of age and the midbrain was isolated and immediately flash frozen. Total RNA and DNA were extracted with a Qiagen Allprep DNA/RNA Mini Kit (Qiagen, Germantown, MD, USA) following the manufacturer’s recommended protocol, and each were stored at −80 °C until assayed.

2.4. DNA isolation, bisulfite conversion, library preparation and sequencing

DNA, isolated as described above, was quantified via a Qubit fluorometer dsDNA high sensitivity kit (Thermo Fisher Scientific, Waltham, MA) and 2 μg DNA was bisulfite-converted using a Zymo EZ DNA Methylation Lightning Kit according to manufacturer guidelines (Zymo Research, Irvine, CA). We evaluated DNA methylation around CpG islands residing within the following key dopaminergic genes: Comt (catechol-o-methyltransferase), Dat1 (dopamine transporter), Drd4 (dopamine receptor 4), Nurr1 (nuclear-receptor related 1 protein; transcription factor), Pitx3 (pituitary homeobox 3; transcription factor), Th (tyrosine hydroxylase), and Vmat2 (vesicular monoamine transporter 2). We also evaluated DNA methylation of Nr3c1 (nuclear receptor subfamily 3 group C member 1), which encodes the corticosterone-responsive glucocorticoid receptor. For genomic coordinates, see Supplemental Table 1.

DNA methylation was assessed utilizing targeted next-generation sequencing BisPCR2 methodology described previously (Bernstein et al., 2015). Briefly, primers were designed to amplify bisulfite-modified regions of interest in intervals of approximately 300 base pairs (for primer sequences, see Supplemental Table 1) and included partial adapter overhangs: PCR#1 Left Primer Overhang: 5′-ACACTCTTTCCC TACACGA CGCTCTTCCGATCT-3′; PCR#1 Right Primer Overhang: 5′-GTGACTGGAGTTCAGACGTGTGCTCTTCC GATCT-3′. Up to 5 primer pairs were multiplexed per PCR reaction. Primers were diluted to a final concentration of 5 μM and approximately 30 ng genomic DNA was amplified utilizing the following PCR cycle conditions: 95 °C – 2:00; 35 cycles; 95 °C – 0:15, 56 °C – 1:30, 72 °C – 0:30; 72 °C – 2:00; 4 °C hold. PCR reactions were size-restricted and purified using AMPure XP beads (Beckman Coulter, Brea, CA, USA) at a 1:1 volume ratio. For each sample, 5 μL of each multiplex reaction was pooled and the subsequent pool was concentrated using AMPure XP beads at a 1:1 volume ratio. A subsequent PCR reaction incorporated Illumina DNA sequencing barcodes unique to each sample (see Supplemental Table 2). Barcode primers were diluted to a final concentration of 5 μM and 1 ng of pooled DNA from PCR #1 was amplified using the following PCR cycle conditions: 95 °C – 1:00; 8 cycles: 95 °C – 0:30, 56 °C – 0:30, 72 °C – 1:00; 72 °C – 1:00; 4 °C hold. PCR products were again purified using AMPure XP beads at a 1:1 volume ratio and concentrations measured using the Qubit fluorometer dsDNA high sensitivity kit prior to sequencing. Targeted next-generation sequencing was performed on an Illumina MiSeq instrument using the reagent V2 kit according to manufacturer instructions. Sequencing was conducted by the Emory Integrated Genomics Core at Emory University (Atlanta, GA, USA).

2.5. mRNA expression assays

After isolation, we converted 10 ng total RNA to cDNA using SuperScript IV Reverse Transcriptase (Thermo Fisher Scientific, Waltham, MA, USA) and the PCR cycle conditions recommended by the manufacturer: 94 °C – 2:00; 35 cycles: 94 °C – 0:15, 55 °C – 0:30, 68 °C – 1:00; 4 °C – hold. Taqman gene expression assays were utilized for Nr3c1 and Actb (beta actin). Each assay plate contained a non-template control, positive control derived from pooled adult mouse brain tissue, and a beta actin internal standard. ^ΔΔCt values were calculated for each animal at every gene tested, and results are expressed relative to gene expression of the vehicle/vehicle control group.

2.6. Data analysis pipeline

Sequences were aligned to the bisulfite-converted NCBI GRCm38 reference genome using the BS-Seeker2 pipeline and Bowtie2 alignment tool (Guo et al., 2013; Guo et al., 2018; Langmead and Salzberg, 2012). Briefly, Illumina adapter sequences were trimmed, paired-end reads for each sample were aligned, and then the unaligned sense and antisense sequences were aligned and both sets of files merged via SAMtools (Li et al., 2009). Methylation was called on the merged file via BSseeker2, and total number of reads, number of methylated reads, and percentage methylation were reported. CpG sites with coverage below 250× were removed. To compare methylation at individual loci across exposure groups we used the R/Bioconductor package limma (Ritchie et al., 2015). This package fits a linear model for each CpG site and exposure group, and then develops an empirical Bayes estimation to assess whether there is a significant difference between groups. All analyses were stratified by offspring sex, since previous studies of deltamethrin and stress exposure observed sex-specific effects (Richardson et al., 2015; Lucas et al., 2004). The Benjamini-Hochberg correction was used to adjust for multiple comparisons and an adjusted p-value was calculated for methylation at each CpG site.

3. Results

3.1. Dual exposure to deltamethrin and CORT is associated with increased methylation of Nr3c1 in males, but not females

After quality control, 24 CpG sites met our criteria and represented 5 of our target genes: Nr3c1, Nurr1, Pitx3, Dat1, Th, and Vmat2. There was a significant increase in average percent methylation of a CpG site at chr18: 39489427 (Nr3c1) in males exposed to deltamethrin and CORT compared to unexposed males (1.57% ± 1.51% vs. 3.00% ± 2.19%, adjusted p-value = 1.53 × 10⁻¹⁰) (Fig. 2). There was no significant difference in methylation at this site in any of the other groups or in females (Supplemental Table 3). There were also no significant differences in average methylation across any of the other CpG sites studied. The Nr3c1 CpG site falls within the NR3C1_6 amplicon and resides within a CpG island containing 190 CpGs present in the promoter region of Nr3c1 (Fig. 2A). A scatterplot representation of individual sample methylation values is shown in Fig. 2B, and also highlights that, interestingly, a proximal CpG site we also tested did not express any observable methylation.

Fig. 2.

A) Location of regions of interest and CpG islands assessed in Nr3c1. Image obtained from http://genome.ucsc.edu, GRCm38/mm10. NR3C1 = Nuclear Receptor Subfamily 3 Group C Member 1 (Kent et al., 2002) B) Scatterplot representation of percent methylation of each sample. Red lines indicate average percent methylation for group. Empirical Bayes testing followed by Benjamini-Hochberg correction were conducted and stratified by sex. For males, n = 5–7. For females, n = 6. DM = deltamethrin. Circles represent individual samples with sequencing coverage of at least 250×. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.2. Developmental exposure to deltamethrin and CORT does not significantly alter midbrain Nr3c1 RNA expression

Given the observed changes in DNA methylation in the midbrain we then examined whether mRNA expression would concomitantly be affected. Interestingly, we did not observe any significant differences in midbrain Nr3c1 RNA expression in males or females in any of the exposure groups via two-way ANOVA (Fig. 3). There were also no significant differences between expression values in males versus females. A Spearman correlation analysis was also conducted to determine whether DNA methylation and RNA expression of Nr3c1 of each midbrain sample were correlated. We did not observe any significant correlation between the two (data not shown).

Fig. 3.

Gene expression in adult midbrain. There is no significant difference in midbrain Nr3c1 expression across all groups in males and females. Data is expressed as 2 ^ΔΔCt analysis of qPCR data and compared to expression in vehicle/vehicle group. Exposure group differences were assessed via two-way ANOVA. * = p < 0.05, n = 5–8, and error bars represent SEM.

4. Discussion

4.1. Summary of results

We utilized a novel combined neurodevelopmental exposure paradigm to model how environmental exposures associated with ADHD might impact DNA methylation of ADHD-relevant genes. We hypothesized that, due to the known dopaminergic effects of both deltamethrin and CORT, we would observe synergistic effects in genes important for proper dopaminergic development. We assessed DNA methylation in exposed offspring utilizing a targeted next-generation sequencing protocol and analyzed midbrain tissue because this is where the majority of the brain’s dopamine cell bodies reside. We did not observe differential DNA methylation in any of our targeted CpG sites in the Dat1, Th, Vmat2, Drd4 genes, nor in the Nurr1 and Pitx3 transcription factors. There was a significant increase in DNA methylation within a CpG island in the 5′ region of the Nr3c1 gene, which encodes the CORT-responsive glucocorticoid receptor. This increase in DNA methylation was only observed in males exposed to both deltamethrin and CORT, indicating that this effect is sex specific. Interestingly, we did not observe a concomitant change in Nr3c1 mRNA expression in the midbrain. We also did not observe a significant correlation between DNA methylation and mRNA expression levels from each midbrain sample in any of the exposure groups in males or females.

4.2. Possible explanations

We assessed DNA methylation in adulthood to investigate whether any persistent DNA methylation changes occur after neurodevelopmental exposure. Perhaps, we did not see significant differences in methylation in any of the studied dopaminergic genes or in Nr3c1 in females because alterations in methylation did not persist into adulthood. It is also possible that while increased DNA methylation of Nr3c1 in males exposed to deltamethrin and CORT was persistent, corresponding Nr3c1 RNA transcript expression changes were not. We analyzed midbrain mRNA expression of DAT1, VMAT2, TH, and COMT and did not observe any significant differences in midbrain mRNA expression in males or females in any of the exposure groups (data not shown). Alternatively, while most studies of environmental epigenetic mechanisms have focused on DNA methylation, additional epigenetic regulatory mechanisms such as miRNA expression or histone modification could also be altered (Marsit, 2015).

In addition, it is possible that the doses of deltamethrin and CORT did not accumulate at high enough levels to induce persistent changes in DNA methylation. The expected plasma and brain levels of deltamethrin and CORT in pregnant mice and their offspring, or the respective levels in humans, have not yet been described. However, some references may provide context. One group orally exposed 12 male rats to 0.93 mg/kg of deltamethrin and measured whole blood and brain concentrations 3.5 h after administration. Mean blood level was 59 ng/mL ± 28.5 ng/mL and mean brain level was 12 ng/g ± 5.6 ng/g (Kamstra et al., 2018). In humans, pyrethroid insecticide metabolite levels have been primarily measured in urine because this an easily accessible matrix for pregnant women and children. Barr et al. (2010) found that in 1999–2000, mean urinary pyrethroid metabolite levels were 0.417 ng/mL (0.292–0.595 95% CI) in children 6–11 years in the US, and 0.292 ng/mL (0.247–0.345 95% CI) in the general population. One very small study by Li et al. measured deltamethrin levels in 100 plasma samples. Only two male samples had a deltamethrin level above the limit of detection – the mean was 0.63 ng/mL ± 3.07 ng/mL (Li et al., n.d.). As far as corticosterone, mice that received the same dose as this current manuscript (25 μg/mL) in their drinking water were measured to have a mean serum CORT level of 87.0 ng/mL ± 14.1 ng/mL (Gourley and Taylor, 2009). Thus, plasma and brain levels of deltamethrin and CORT are much lower than the exogenous exposure.

Lastly, cell-to-cell heterogeneity poses a unique challenge in studies of DNA methylation (Huan et al., 2018), and particularly within solid tissues such as the brain where cell composition differences greatly affect detectable epigenetic marks (Florio et al., 2017). We extracted DNA and mRNA from the entire midbrain instead of cutting it in half to reduce some of this variability. However, while the midbrain is primarily dopaminergic, additional cell types such as GABAergic populations (Li et al., 2014) and various glial populations (La Manno et al., 2016) are present. This cell-to-cell heterogeneity could have contributed to the variability we observed in our DNA methylation results.

4.3. Potential mechanisms

Effects of CORT and stress exposure on Nr3c1 methylation have been studied in humans extensively (Bromer et al., 2013; Parent et al., 2017; Tyrka et al., 2015; Stroud et al., 2016a; Sheinkopf et al., 2016; Parade et al., 2016; Palma-Gudiel et al., 2015a; Palma-Gudiel et al., 2015b; Kember et al., 2012) and a mouse model of early life stress shows hypermethylation of a CpG island shore proximal to the Nr3c1 promoter in the hypothalamus (Bockmuhl et al., 2015). However, this is the first description of Nr3c1 hypermethylation within the dopaminergic midbrain in response to deltamethrin and CORT. Interestingly, there was no hypermethylation in the CORT-only exposed males or females, nor in the deltamethrin-only exposed males or females. Perhaps, combined effects of deltamethrin and CORT on Nr3c1 in this region were necessary to achieve a detectable methylation difference. Glucocorticoid receptors are present in the midbrain’s substantia nigra and ventral tegmental area, and are expressed in dopaminergic neurons in these areas (Hensleigh and Pritchard, 2013). In vitro studies also show that pyrethroid insecticides antagonize the glucocorticoid receptor (Zhang et al., 2016b). Thus, Nr3c1 hypermethylation could occur due to direct effects of deltamethrin and CORT on glucocorticoid receptor expression, or through indirect effects of deltamethrin and CORT on dopaminergic neurons that express the glucocorticoid receptor.

We only observed Nr3c1 hypermethylation in males exposed to deltamethrin and CORT. In our previous study of developmental deltamethrin, we did not observe dopaminergic effects of deltamethrin in females either (Richardson et al., 2015; Vester et al., 2019). Moreover, this same sex-specific effect is seen in an epidemiologic study of children with ADHD who were exposed to pyrethroid insecticides (Wagner-Schuman et al., 2015). ADHD affects men significantly more than women. Studies posit this could be to variability in the dopamine system (Rhee and Waldman, 2004; Andersen and Teicher, 2000) or even due to protective effects that estrogen has on dopaminergic neurons undergoing toxicant insults (Bains and Roberts, 2016; Bains et al., 2007; Sawada and Shimohama, 2000; Jourdain et al., 2005). This could perhaps explain why deltamethrin exposure in females did not sufficiently alter DNA methylation of Nr3c1.

4.4. The HPA axis, dopamine, and ADHD

While glucocorticoid function has not been studied as extensively in ADHD development as dopamine function, one rat ADHD model showed that administration of a glucocorticoid receptor agonist (dexamethasone) led to increased dopamine in the striatum and prefrontal cortex. Dexamethasone also ameliorated hyperactivity behavior and inattention measured via a Y-maze assay (Chen et al., 2017). Lastly, plasma corticosterone in an ADHD rat model and plasma cortisol in boys with ADHD was significantly lower (Wu et al., 2017). In humans, we previously illustrated that increased NR3C1 methylation is associated with early neurodevelopmental indicators in infants. NR3C1 hypermethylation was associated with altered infant cry acoustics, a measure of stress responsiveness and neurological status (Sheinkopf et al., 2016). Additionally, we saw a significant positive association between placental NR3C1 promoter methylation and infant movement and attention measured via the NICU Network Behavioral Scales (Stroud et al., 2016b). Furthermore, functional NR3C1 polymorphisms affect symptom severity (Schote et al., 2016) and mediate methylphenidate treatment response (Fortier et al., 2013) in children with ADHD.

4.5. Future directions, impact

To measure DNA methylation we used a targeted next-generation sequencing approach first described by Bernstein et al. (Bernstein et al., 2015). While this BisPCR² method has been utilized in human tissue samples and cell lines (Zhao et al., 2017; Roeh et al., 2018; Franzen et al., 2017; Bernstein et al., 2017), zebrafish and sea bass (Kamstra et al., 2018; Kamstra et al., 2017; Anastasiadi et al., 2018), and a mouse model of intestinal cancer (Sheaffer et al., 2016; Elliott et al., 2016), this is the first described use of the method in a mouse model of brain development. Mouse models of neurodevelopment are particularly tractable for studies of environmental exposure and neuropsychiatric disease. Epigenetic mechanisms are important intermediates for environment x gene interactions and this targeted next-generation sequencing approach provides a cost- and time-efficient method of study. In addition, this exposure paradigm establishes a novel exposure approach that is more relevant to environmental exposure in vulnerable populations than individual exposure models utilized to date. Results of our study provide possible connections between environmental exposures that impact the dopamine system and the HPA axis and contribute to our current understanding of the complex etiopathogenesis of ADHD.