Impact of early developmental arsenic exposure on promotor CpG-island methylation of genes involved in neuronal plasticity

Neurochemistry International 58 (2011) 574–581

Contents lists available at ScienceDirect

Neurochemistry International journal homepage: www.elsevier.com/locate/neuint

Impact of early developmental arsenic exposure on promotor CpG-island methylation of genes involved in neuronal plasticity Liborio Martı´nez a,1, Vero´nica Jime´nez a,1, Christian Garcı´a-Sepu´lveda b, Fa´tima Ceballos a, ˜ o-Moreno d, Lesly Doniz c, Vı´ctor Saavedra-Alanı´s a, Claudia G. Castillo a, Juan Manuel Delgado a, Perla Nin a Martha E. Santoyo , Roberto Gonza´lez-Amaro c, Marı´a E. Jime´nez-Capdeville a,* a

Departamento de Bioquı´mica, Facultad de Medicina, Universidad Auto´noma de San Luis Potosı´, Av. V. Carranza 2405, C.P. 78210 San Luis Potosı´, Mexico Laboratorio de Geno´mica Viral y Humana, Facultad de Medicina, Universidad Auto´noma de San Luis Potosı´, Av. V. Carranza 2405, C.P. 78210 San Luis Potosı´, Mexico c Departamento de Inmunologı´a, Facultad de Medicina, Universidad Auto´noma de San Luis Potosı´, Av. V. Carranza 2405, Col. Los Filtros, C.P. 78210 San Luis Potosı´, Mexico d Laboratorio de Inmunologı´a Celular y Molecular, Facultad de Ciencias Quı´micas, Universidad Auto´noma de San Luis Potosı´, Av. Manuel Nava 6, C.P. 78210, Mexico b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 25 May 2010 Received in revised form 23 December 2010 Accepted 19 January 2011 Available online 15 February 2011

Epigenetic mechanisms are crucial to regulate the expression of different genes required for neuronal plasticity. Neurotoxic substances such as arsenic, which induces cognitive deficits in exposed children before any other manifestation of toxicity, could interfere with the epigenetic modulation of neuronal gene expression required for learning and memory. This study assessed in Wistar rats the effects that developmental arsenic exposure had on DNA methylation patterns in hippocampus and frontal cortex. Animals were exposed to arsenic in drinking water (3 and 36 ppm) from gestation until 4 months of age, and DNA methylation in brain cells was determined by flow cytometry, immunohistochemistry and methylation-specific polymerase chain reaction (PCR) of the promoter regions of reelin (RELN) and protein phosphatase 1 (PP1) at 1, 2, 3 and 4 months of age. Immunoreactivity to 5 methyl-cytosine was significantly higher in the cortex and hippocampus of exposed animals compared to controls at 1 month, and DNA hypomethylation was observed the following months in the cortex at high arsenic exposure. Furthermore, we observed a significant increase in the non-methylated form of PP1 gene promoter at 2 and 3 months of age, either in cortex or hippocampus. In order to determine whether this exposure level is associated with memory deficits, a behavioral test was performed at the same age points, revealing progressive and dose-dependent deficits of fear memory. Our results demonstrate alterations of the methylation pattern of genes involved in neuronal plasticity in an animal model of memory deficit associated with arsenic exposure. ß 2011 Elsevier Ltd. All rights reserved.

Keywords: Arsenic Epigenetic regulation Neuronal plasticity Wistar rat

1. Introduction Numerous recent reports are gradually revealing that environmental and dietary signals reach the deepest levels of cellular organization by modifying epigenetic programs leading to the silencing or over expression of certain genes. This issue is becoming of critical importance for the function of the central nervous system since not only during development but also in fully differentiated neurons, epigenetic mechanisms play a crucial role in neuronal plasticity (Borrelli et al., 2008; Feng et al., 2010). The continuous creation, reinforcing and elimination of synapses that are characteristic of neuronal plasticity involve

* Corresponding author. Tel.: +33 52 444 826 2349x530; fax: +33 52 444 826 2352. E-mail address: [email protected] (M.E. Jime´nez-Capdeville). 1 These authors contributed equally to this work. 0197-0186/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2011.01.020

changes in gene expression. Levenson et al. (2006) demonstrated in vitro that the methylation of the promoter regions of genes involved in neuronal plasticity, such as reelin (RELN) and brain derived neurotrophic factor (BDNF), is a dynamic process, showing also that the inhibition of DNA methylation blocks the induction of long term potentiation (LTP). In vivo, employing a mice model of contextual fear conditioning (CFC), Miller and Sweatt (2007) found that the inhibition of DNA methylation blocks memory formation. The process of CFC was associated with decreased methylation of the RELN promoter with a significant increase of protein expression. In contrast, the methylation of protein phosphatase 1 gene (PP1) promoter was associated to a diminution of protein expression. Therefore, the concerted expression of different genes required for synapse reinforcement that underlies memory formation is intimately linked to dynamic changes in DNA methylation. Recently, the inhibition of histone deacetylase 1 (HDAC1) has been reported to ameliorate memory formation in a mice model of Alzheimer disease (Kilgore et al., 2010). Thus, in vitro

L. Martı´nez et al. / Neurochemistry International 58 (2011) 574–581

and in vivo evidence supports the notion that dynamic chromatin remodeling takes place during the process of adult neuronal plasticity. Nowadays, the study of changes of histone acetylation and methylation as well as DNA methylation, among other epigenetic mechanisms, is providing molecular basis for the alterations in gene expression observed in individuals with addiction, depression and neuropsychiatric disorders (Murgatroyd et al., 2009; Gra¨ff and Mansuy, 2009; Tsankova et al., 2007). Epigenetic mechanisms are potential targets for the neurotoxic action of diverse substances that induce cognitive dysfunction in the human population, mainly when such exposure takes place during prenatal and early postnatal development (Vahter, 2007; Bellinger, 2008). An archetypical example of these substances is arsenic, whose presence in drinking water results in millions of exposed individuals around the world (IARC, 2004; NRC, 2001; WHO, 2001), with a growing number of reports demonstrating cognitive deficits in exposed children (Calderon et al., 2001; Tsai et al., 2003; Wasserman et al., 2004; Vahter, 2007; Rosado et al., 2007). One of the plausible links between arsenic exposure and epigenetic alterations is its biotransformation, which has an effect on DNA methylation by sharing the same methyl donor (Sadenosyl methionine) required for chromatin remodeling (Lin et al., 2002). In this regard, it has been reported that arsenic exposure alters the pattern of DNA methylation in keratinocyte cultures (Reichard et al., 2007). In this context, we hypothesized that developmental arsenic exposure in rats induces learning deficits through the disruption of hippocampal and cortical DNA methylation patterns. For this purpose, we assessed the effect of intrauterine and postnatal arsenic exposure on contextual learning processes and brain DNA methylation. Our data corroborate that arsenic exposure is associated with memory deficit, and suggest that alterations of the methylation pattern of genes involved in neuronal plasticity may contribute to this condition. 2. Experimental procedures 2.1. Reagents and antibodies Most reagents were molecular biology or analytical grade reagents and were prepared in molecular grade water (18.0 V/cm). Solvents, salts and enzymes were obtained from Sigma–Aldrich (St. Louis, MO, USA), Bio-Rad Lab (Hercules, CA, USA) and J.T Baker (Phillisburg, NJ, USA). The rest of the reagents and antibodies were CpG methyltransferase from New England BioLabs (Ipswich, Ma, USA), the wizard DNA clean up system from Promega (Madison, WI, USA), Taq DNA polymerase from Vivantis (Vivantis Technologies Sdn. Bhd. Singapore), anti-5 methyl cytosine antibody (Calbiochem, Darmstadt, Germany), biotinylated rabbit anti-mouse secondary antibody (DAKO, Carpinteria, CA, USA), FITC-coupled secondary antibody (eBioscience, San Diego, CA, USA), RPMI cell culture medium and fetal bovine serum (GIBCO, BRL). 2.2. Animal model and sample collection The experiments were performed according to the Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral Research (2003) and the protocol was approved by the ethics committee of Universidad Autonoma de San Luis Potosi. The experiments started by assigning 24 female and 24 male Wistar rats weighting between 200 and 250 g to three experimental groups of eight couples each. One group received arsenic-free drinking water, the second group had access to water containing 3 ppm of sodium arsenite, a concentration that resulted in an approximate ingestion of 0.3–0.4 mg/kg/day of arsenic (an exposure level that has already been reported in human population (Hsieh et al., 2008). At this concentration only morphological alterations in myelinated central tracts have been demonstrated but no other manifestation of toxicity (Rios et al., 2009). Finally a third group received 36 ppm of arsenic (3–4 mg/kg/day), a dosage that has been associated with structural alterations of myelinated tracts in the striatum, smaller and heterochromatic cell nuclei as well as with demyelination (Zarazua et al., 2010). Each mating pair was placed in a cage and maintained under a 12 h light–dark cycle with food (Prolab RMH 2500, PMI Nutrition International, Brentwood, MO) and water ad libitum. After 10 days the males were removed and arsenic exposure continued in the pregnant rats throughout gestation and lactation. Offspring were weaned at 4 weeks, separated by sex and subjected to continued arsenic exposure for 1, 2, 3 or 4 months. As the animals aged, the experiments were always performed in groups of rats belonging to different litters for each treatment as follows: (a) Groups of three female rats for immunohistochemical studies, (b) Groups of three

575

females for flow cytometry determinations in brain and blood cells, (c) Groups of five females for the methylation analysis of specific genes, and, (d) Groups of four to five male rats for behavioral test. 2.3. Behavioral test The training chamber (acrylic, 20 cm  25 cm  25 cm) for the contextual fear conditioning (CFC) test was located inside a larger wooden box so as to isolate the chamber from the surrounding environment and to allow only one person to observe the rats. This person was unaware of the treatment to which each animal had been subjected to. Training started by placing the animal inside the chamber and allowing an exploration period of 2 min after which the rat received the conditioning stimulus (CS) consisting of a four second burst of a 4 dB tone followed by 2 second electric shock (US) of 0.75 mA. Three pairs of these CS/US were applied in 30 s intervals after which animals were given 30 additional seconds in the chamber before returning them to their cage. The evaluations were performed in each animal after 1, 6, 24 and 72 h of the conditioning phase. After placing the animal in the cage and allowing a 30 s exploration period, the CS (tone) was applied resulting in a characteristic immobility behavior (Colon et al., 2006). The animal was observed during 5 min in order to obtain the total frozen time and then returned to its cage. The comparison between total freezing times indicated the difference between treatments at each test point, while the analysis of the response over time for each group allowed the evaluation of the extinction of the conditioned response. 2.4. Flow cytometry analysis For cell isolation from blood and brain samples, the animals were decapitated, blood was collected in heparinized tubes and the brain was extracted from the skull. The rostral third of each hemisphere was manually fragmented with a surgical blade in 1 mL of Hanks balanced solution and incubated for 1 h with 0.1% collagenase at 37 8C. The resulting suspension was then filtered through a 1 mm2 mesh, centrifuged and the cells obtained by decantation. Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll-Hypaque gradient centrifugation. The number of 5-methyl cytosine positive cells in brain and PBMC was quantified by using a specific monoclonal antibody and flow cytometry analysis. For this purpose, cells were fixed in 4% paraformaldehyde for 10 min at room temperature and washed with phosphate buffered saline (PBS). After permeabilization with methanol at 20 8C for 20 min, the cells were treated with 2 N HCl at 37 8C for 30 min in order to expose the CpG islands, followed by one step neutralization with sodium borate (0.1 M, pH 8.5) for 1 min. The incubation with the primary antibody (mouse monoclonal anti 5-methylcitosine) was performed at 37 8C during 45 min, followed by the addition of the secondary rabbit anti-mouse IgG secondary antibody coupled to FITC during 20 min. After washing (1% albumin in PBS), cells were fixed with 1% paraformaldehyde, and analyzed in a FACSCalibur flow cytometer (Becton-Dickinson, San Jose, CA). Results were expressed as the percent of positive cells and negative controls were run in all experiments. 2.5. Immunohistochemistry At the time of sacrifice, three rats from each experimental group were deeply anesthetized with 50 mg/kg intraperitoneal sodium pentobarbital. Rats were perfused through the heart with a blunted 18-gauge needle that passed through a small incision in the left ventricle to the proximal ascending aorta. Three hundred mL of 0.1 M PBS; pH 7.4 followed by 300 mL of 4% paraformaldehyde and 0.6% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) were delivered by gravity flow through the needle. Brains were removed, the hemispheres separated from the cerebellum and blocked into rostral and caudal sections, through a coronal division approximately 1 mm posterior to bregma (Paxinos and Watson, 2005), and placed in 4% paraformaldehyde and 0.6% glutaraldehyde in 0.1 M phosphate buffer for 24 h at 4 8C and embedded in paraffin. Coronal brain sections (7 mm) were collected in silanized slides from the frontal cortex region between 3.2 and 4 mm anterior to bregma, and the region of the dorsal hippocampus located approximately at 2.1 of bregma. Next, sections were dewaxed, rehydrated and subjected to a sequence of incubation steps in a humidity chamber, starting with sodium citrate at 100 8C (pH 6) during 30 min for epitope recovery (Lorincz and Nusser, 2008), followed by HCl 3.5 N during 15 min at room temperature with the purpose of exposing CpG islands, and finally with 3% hydrogen peroxide in methanol during 15 for blocking endogenous peroxidase. Sections were incubated overnight at 4 8C with 1:200 mouse monoclonal anti-methylcytocine antibody (which is suited for use in paraffin sections, according to the manufacturer instructions) followed by the streptavidin–biotin marked secondary antibody for 15 min at room temperature. Peroxidase activity was visualized by incubating the sections with diaminobenzidine and the sections were counterstained with Harris hematoxiline. All rinses between the incubation steps were performed with TBST (Tris buffer saline–tween, 0.05 M, pH 7.4). The sections were processed in batches that included 2 slides from each treatment (control, low and high arsenic) with 3 sections each. A negative control was included which consisted of tissue sections treated solely without the primary antibody for each batch. The sections were viewed with a Nikon microscope (Nikon, Labophot – 2, Japan) equipped with a digital camera. Three

576

L. Martı´nez et al. / Neurochemistry International 58 (2011) 574–581

Table 1 Primers for methylated and non-methylated promoter regions of Reelin and PP1 genes. Gene

Type

Sense

Sequence

Reelin

Non-methylated

Forward Reverse Forward Reverse Forward Reverse Forward Reverse

50 -TGT-TAA-ATT-TTT-GTA-GTA-TTG-GGG-ATG-T-30 50 -TCC-TTA-AAA-TAA-TCC-AAC-AAC-ACA-CC-30 50 -GGT-GTT-AAA-TTT-TTG-TAG-TAT-TGG-GGA-C-30 50 -TCC-TTA-AAA-TAA-TCC-AAC-AAC-ACG-C-30 50 -GAG-GAG-AGT-TTG-GTG-TTT-ATA-AGA-TGG-T-30 50 -TCC-TCC-AAA-AAC-TCA-ACT-CAA-ACA-A-30 50 -GGA-GAG-TTT-GGT-GTT-TAT-AAG-ATG-GC-30 50 -CGA-AAA-CTC-GAC-TCG-AAC-GA-30

Methylated PP1

Non-methylated Methylated

photographs per section (2 slides containing 3 sections for each region, cortex and hippocampus, Fig. 3), were imaged and analyzed for staining intensity using the program Image J 1.44k (National Institutes of Health, USA). All color images obtained at the same magnification power were digitally transformed to grey scale, and the intensity of each image was assessed by applying a uniform digital filter to all of them. Approximately 18 values obtained from each rat were averaged to yield a final measurement of immunostaining intensity per animal. The quantitative results are given in pixels, and they arise from 3 animals for each time point (1, 2, 3 and 4 months of age), and 3 treatments (control, low and high arsenic).

ANOVA, where the factors were the age of the animals (1, 2, 3 and 4 months) and the treatment (control, low As and high As) followed by the pos hoc test of Fischer LSD. In the case of flow cytometry data, two-way ANOVA was followed by the Bonferroni test. These statistical analyses were performed with a commercial software package (Statistica 6.0). The results obtained in the immunohistochemical experiments were subjected to x2-test followed by Mann–Whitney comparison between groups. Values of p below 0.05 were considered statistically significant in all cases.

2.6. DNA extraction and purification When the animals reached the appropriated age, 5 rats from each group were sacrificed by decapitation. After brain extraction, the cortex from the frontal pole and the hippocampus were dissected from both hemispheres in a cold plate and immediately frozen in liquid nitrogen. The tissues were stored at 70 8C until processing and DNA extraction was performed by the salting out method (Miller et al., 1988). Contaminating proteins were removed by proteolytic digestion incubating tissue homogenates during 8 h at 55 8C with 500 mL of lysis buffer (50 mM Tris–HCl, 50 mM EDTA, 50 mM NaCl, 1% SDS) and 10 mL proteinase K (10 mg/mL), followed by centrifugation at 3000  g for 15 min. Proteins were precipitated from 500 mL of the supernatant with the addition of 5 M NaCl and, after a second spin, 600 mL of the supernatant withdrawn again for DNA precipitation with isopropanol at 20 8C. The DNA was washed twice with 70% ethanol and resuspended in TAE 1X pH 8.0 and stored at 20 8C. The yield, quality and integrity of DNA samples were evaluated by spectrophotometry and agarose gel electrophoresis, as previously described (Garcia-Sepulveda et al., 2010). 2.7. Bisulfite assay This modification is based on the conversion of non-methylated cytosine to uracil through bisulfite action, while methylated cytosine remains unaltered. The procedure was performed according to Frommer et al. (1992). Thirty mL of DNA (containing 3 mg of DNA) were mixed with 20 mL DEPC water, 5.5 mL NaOH 2 M and incubated for 10 min at 37 8C. After the addition of 30 mL of 10 nM hydroquinone and 520 mL of 3 M sodium bisulfite, pH 5, the mixture was covered with 200 mL of mineral oil and incubated for 16 h at 50 8C. Following bisulfite conversion, DNA was purified using the Wizard DNA Clean up System, according to the manufacturer’s instructions. After incubating at room temperature for 5 min with 5.5 mL of 3 M NaOH, 33 mL of 10 M ammonium acetate were added and DNA was precipitated with 270 mL of absolute ethanol overnight at 20 8C. After washing with 70% ethanol DNA was finally resuspended in 20 ml TE 1 buffer (pH 8) and stored at 20 8C until processing. 2.8. PCR Methylation specific PCR was carried out using primers specific for the methylated and non-methylated promoter forms of RELN and PP1 genes (Table 1). Reactions were performed on 50 ng of genomic DNA using 800 nM of each primer, 200 mM of each deoxynucleotide triphoshate, 2 mM MgCl2 and 0.1 IU of Taq DNA polymerase in a final volume of 12.5 mL with a Maxygen thermocycler (Axygen, Union City, CA, USA). Genomic DNA methylated with CpG methyltransferase was included as a positive control. DNA samples extracted from rat mononuclear cells subjected to the bisulfite procedure were included as negative controls in each PCR batch. The amplified products were run in 8% acrylamide gels followed by ethidium bromide staining (solution 10 mg/ml, Sigma–Aldrich, St. Louis, MO, USA) and visualization under UV light. Gel images were analyzed for median intensity, background fluorescence and area fluorescence with the Kodak 1D image analysis software V 3.6. The optical density values for each amplified product corresponded to the subtraction of background values from medium band intensity. 2.9. Statistics Variance homogeneity and normality of data were tested by means of Levene and Shapiro–Wilk tests, respectively. The statistical significance of the CFC test and of the methylation level of the genes was determined by means of two-way

Fig. 1. Contextual fear conditioning. Freezing time (s) presented by each group of animals 1, 6, 12 and 72 h after conditioning. Points represent means  SEM of four to five independent experiments. (A) Two months, *significant difference of the high arsenic group as compared to control and low arsenic groups; ysignificant difference for both arsenic exposed groups between freezing time at 72 h as compared with 1 h (extinction). (B) Three months, *significant difference of both arsenic exposed groups as compared to controls; ysignificant extinction at 6 and 24 h for the high arsenic group; yysignificant extinction of all 3 groups at 72 h. (C) *Significant difference for both arsenic exposed groups as compared to controls, **only the high arsenic group was significantly different from the control group.

L. Martı´nez et al. / Neurochemistry International 58 (2011) 574–581

3. Results 3.1. Contextual memory deficits induced by arsenic exposure First, the animals were tested from 2 months of age in order to detect how early the developmental arsenic exposure induced contextual memory deficits. A significant effect of the treatment was present in the fear conditioning test from this first assessment [F(2,44) = 9.3, p < 0.001]. Although 1 h after training all three groups showed similar freezing behavior (Fig. 1A), significantly shorter freezing periods were observed 6, 24 and 72 h after training in the highly exposed group, as compared to controls and low level arsenic exposures (p < 0.05, in all cases). At 3 months of age, no differences were observed in the first hour, but both exposed groups showed significantly different freezing behavior from controls 6, 24 and 72 h later [F(2,44) = 7.1, p < 0.01, Fig. 1B]. Finally, 4 month old exposed animals were at all times significantly different from controls [F(2,44) = 27, p < 0.001, Fig. 1C]. Extinction was a second parameter in which control and exposed animals were different. Control animals showed a significantly lower freezing time only at 3 months, 72 h after training, while exposed animals presented this effect since 2 months of age. Moreover, animals exposed to high levels of arsenic had shorter freezing responses at the 6 h test. At 4 months, no extinction was observed in any experimental group [F(3,44) = 1.6, p = 0.2]. 3.2. Global DNA methylation in the rat brain The percentage of 5-methylcytosine positive cells in whole brain was quantified by flow cytometry as a first approach to detect changes associated with either postnatal age and/or arsenic exposure. A significant increase of 5-methylcytosine+

577

cells was associated with the age of the animals in all groups (Fig. 2B), but no differences were attributable to arsenic exposure. In order to explore whether this DNA methylation increase related to age was present in other cell types, PBMC were also analyzed. Although monocytes also showed an increased percentage of positive cells according to age, this change reached statistical significance only at 3 months as compared with the first month (Fig. 2C). In contrast, lymphocytes showed a very high percentage of positive cells all along the 4 months examined (close to 100%) with no significant differences related to age (Fig. 2D). As in the case of brain cells, no differences in the percent of 5-methylcytosine+ PBMC were observed between control and exposed animals. 3.3. Immunohistochemistry Next, the immunoreactivity to 5-methylcytosine was assessed in situ in sections of hippocampus and frontal cortex. Significant effects of the treatment were found in both regions. The control group presented low levels of immunostaining in the first month followed by an important increase of DNA methylation in the second month of age in both regions (Fig. 3C and D). In contrast, both arsenic exposed groups showed a significantly higher immunoreactivity in the first month indicating DNA hypermethylation as compared with the control group in the hippocampus, while only the high arsenic group showed this effect in the cortex (p < 0.05). In the cortex (Fig. 3C), levels of immunoreactivity remained significantly lower in the high exposed group than in the controls all along the third and fourth months tested (p < 0.05). In the hippocampus (Fig. 3D), the levels of immunostaining were not significantly different from controls at the second, third and fourth month, respectively.

Fig. 2. Results of flow cytometry assays. (A) Representative histogram showing the population of central nervous system 5-methylcytosine positive cells in control, high and low arsenic exposed animals. (B–D) Percentages of 5-methylcytosine positive cell in three populations isolated from brain and blood of control and arsenic exposed rats. Bars represent means  SEM of three independent experiments for each age. No significant treatment effects were observed, only age/time of exposure effects. Ap < 0.01 as compared with 1 month values, ap < 0.05 as compared with 1 month values, Bp < 0.01 as compared with 2 month values, and bp < 0.05 as compared with 2 month values. Two way ANOVA followed by Bonferroni post test.

578

L. Martı´nez et al. / Neurochemistry International 58 (2011) 574–581

Fig. 3. Diagrams of coronal rat brain sections representing sampling zones in frontal cortex and hippocampus, modified from Paxinos and Watson (2005). Numbers represent approximate positions relative to bregma. (A), (B) Photomicrographs of 7 mm coronal sections showing 5-methylcytosine positive cell nuclei (arrows). Scale bar 10 mM. (C), (D) Graphic representation of the temporal course of immunoreactivity to 5-methylcytosine in rat brain sections (a total of 18 per group and age). Asterisks represent significant differences between groups analyzed by means of x2 followed by Mann–Whitney comparison between groups,*p < 0.05.

3.4. Methylation of PP1 and RELN The effects of arsenic exposure on the methylation levels of the promoter regions of two genes involved in memory formation, PP1 and RELN were examined in hippocampal and cortical neurons. The main effects associated with arsenic exposure were observed in PP1 in both regions. In the hippocampus, arsenic exposure was associated with a significant change in the non methylated form of PP1 gen [F(2,46) = 8.51, p < 0.001, Fig. 4C], while in the methylated form no changes were observed [F(2,44) = 1.19, p = 0.314]. Post hoc analysis showed a significant increase in the animals exposed to low arsenic levels compared to both, the control group at 2 months (p = 0.04) and the group exposed to high levels of arsenic at 3 and 4 months (p = 0.01 and p = 0.04, respectively). The non methylated form of PP1 gene also showed a significant increase in the cortical region of brains from rats exposed to arsenic [F(2,46) = 5.5, p < 0.01, Fig. 4E]. At the second month, the group of rats exposed to high levels of arsenic had significantly higher PCR product yields than the control group (p = 0.04), and at the third month both exposed groups showed increased PCR yields of non methylated PP1 as compared to controls (p < 0.001 in both cases). As in the case of hippocampal region, the methylated form of PP1 gene in cortex did not show significant changes related to arsenic exposure (Fig. 4D). Both in hippocampal and cortical regions, the methylated form of PP1 exhibited a significant effect of the age of the animals in all experimental groups [F(3,44) = 12.2, p < 0.001, and F(3,46) = 35.3, p < 0.001, for hippocampus and cortex, respective-

ly]. The higher levels were found at the third month in the cortex and at the fourth month in the hippocampus (Fig. 4 B and D). Finally, the methylated form of RELN did not show significant changes associated to arsenic exposure in the hippocampus [F(2,47) = 0.07, p = 0.9], and only a significant effect of the age was observed in the exposed groups (p < 0.001 for low arsenic, p < 0.05 for high arsenic, Fig. 5B). In cortex, the high arsenic group showed a significant increase at the second month (p < 0.05, Fig. 5C). Since the PCR product yields of the non methylated form of RELN were too low at months 3 and 4, these data were not analyzed. A summary of all significant differences between arsenic exposed groups and the control group is displayed in Table 2. 4. Discussion Our results provide valuable and original information about changes of the DNA methylation state of brain cells in cerebral regions involved in memory formation of arsenic exposed animals. Although other epigenetic processes such as the regulation of histone acetylation in the hippocampus (Levenson et al., 2004) have been demonstrated to participate in memory formation, this study focused in DNA methylation due to the potential link between arsenic biotransformation and methyl availability for different cellular processes. More than a decade ago, Zhao et al. (1997) proposed that a DNA hypomethylation state eventually leading to defective genetic expression could be a consequence of cellular methyl group

L. Martı´nez et al. / Neurochemistry International 58 (2011) 574–581

579

Fig. 4. (A) Representative acrylamide electrophoresis gel of methylated and unmethylated forms of the PP1 promoter in brain regions. MWM, molecular weight marker, 200 bp (top) and 100 bp (bottom), and extreme right M and UM, Positive (methylated) and negative (unmethylated) controls for the bisulfite reaction (B) and (C). Mean optic density of the methylated PCR product along the four experimental months. yp < 0.05 for control and low arsenic values as compared with control at 1 month; yyp < 0.01 of for all three groups with respect to values at 2 and 4 months. (D), (E) Mean optic density of the unmethylated PCR product. Hippocampus: *p < 0.05 as compared with high arsenic and controls, and **p < 0.01 for low arsenic group compared with high arsenic group. Cortex: *p < 0.05, high arsenic vs. control group, **p < 0.01 both exposed groups vs. controls and yp < 0.05 for both exposed groups at 3 months compared with values at 1,2 and 4 months. Bars represent SEM values (n = 5).

depletion induced by the mono- and dimethylation that arsenic undergoes in different tissues, especially in the liver. Since then, both DNA hypo- and hypermethylation associated with arsenic exposure has been reported in vitro and in vivo (Mass and Liangjun, 1997; Goering et al., 1999; Sciandrello et al., 2004; Reichard et al., 2007), which is accompanied by an altered expression of oncogenes, tumor suppression genes (Chanda et al., 2006; Xie et al., 2004) and other genes (Chen et al., 2004). These findings support that epigenetic changes could account for the carcinogenic effect of arsenic as well as its ability to induce apoptosis and inflammatory gene expression (Bourdonnay et al., 2009; SalgadoBustamante et al., 2010). The results of the present study also point in the direction of a participation of DNA methylation in the neurotoxicity of arsenic. 4.1. Epigenetic mechanism in the context of other arsenic effects in the CNS In light of the above mentioned epigenetic alterations associated with arsenic exposure in non-neural cells, it is plausible to hypothesize that the low exposure levels at which cognitive deficits are observed in arsenic-exposed children involves an

epigenetic dysregulation (Fagliolini et al., 2009). Our data demonstrate a high global DNA methylation at the first month in both brain regions, followed by hypomethyation at the third and fourth months but only in the cortex. This early increase of DNA methylation could be related to the silencing of certain genes. On the other hand, the observed increase of the non-methylated form of PP1 could also be related to higher expression of this protein, considered a critical regulator of memory (Munton et al., 2004; Koshibu et al., 2009), and hence to defective synaptic plasticity. Nevertheless, the pattern of methylation of the chosen genes was irregular and does not show association neither with the dose nor with the time of exposure. It is important to notice that the immunohistochemical findings reported here represent the total methylated/unmethylated cytosines, non-neuronal specific, found either in promoter or non promoter regions of DNA, and that 50% of CpG islands are located outside gene coding regions (Sharma et al., 2010). Likewise, gene-specific methylation of promoter regions analyzed through PCR is not restricted to certain cellular type, and enhanced levels of the unmethylated form of PP1 where not accompanied by a decrease in the methylated fraction of PP1 promoter. Our explanation for this observation is that since we have not performed quantitative PCR, it is very feasible that under

L. Martı´nez et al. / Neurochemistry International 58 (2011) 574–581

580

Fig. 5. (A) Representative acrylamide electrophoresis gel of methylated and unmethylated forms of the RELN promoter in brain regions. MWM, molecular weight marker, 200 bp (top) and 100 bp (bottom), extreme right M and UM, Positive (methylated) and negative (unmethylated) controls for the bisulfite reaction. (B), (C) Mean optic density of the methylated PCR product along the four experimental months. yp < 0.05 and yyp < 0.01 for time of exposure related differences. Cortex: *p < 0.05 for high arsenic group as compared to low arsenic and control groups. Bars represent SEM values (n = 5).

Table 2 Summary of significant differences with respect to the control group found through the enlisted assays. Treatment/age

Low arsenic 1 2 3 4 High arsenic 1 2 3 4

Behavioral test

Flor cytometry

Immuno histochemistry

PP1 (cortex or hippocampus)

Cortex

Hippocampus

Methylated

Non-methylated

Reelin methylated (cortex or hippocampus)

– U U

– – – –

– – – –

U – – –

– – – –

– U U –

– – – –

U U U

– – – –

U – U U

U – – –

– – – –

– U U –

– U – –

our experimental conditions only a small fraction of the sequences is unmethylated, thus, a modest but significant increase in this fraction does not apparently affect the levels of the methylated fraction. Therefore, further studies are needed to elucidate how changes of DNA methylation are later reflected in functional deficits that are dose and time of exposure dependent. In this regard, quantitative changes of global DNA methylation were not detected by flow cytometry, neither in brain nor in blood cells from exposed rats. In exposed human populations, however, genomic DNA has been found hypermethylated in lymphocytes (Pilsner et al., 2007; Majumdar et al., 2009). In contrast with these data of DNA methylation in exposed animals, arsenic accumulation in the CNS, oxidative stress induction (Garcı´a Cha´vez et al., 2003), changes of enzymatic activities (Zarazua et al., 2006) and morphological alteration (Zarazua et al., 2010) in axons and myelin tracts follow more closely a pattern of association with time and dose of exposure. This is in accordance with the notion that epigenetic mechanisms associated with synaptic plasticity are dynamic changes that take place in response to specific signals in order to precisely tune protein expression underlying synaptic plasticity. Therefore, we consider that the present approach provides only a picture of general methylation patterns in a brain region and in a given

promoter region, without allowing to make inferences about its participation in defined synaptic processes. 4.2. Perspectives After this initial approach, the next step is to test the capability of the system of exposed animals to respond to behavioral training through epigenetic modifications, not only throughout DNA methylation but also involving other mechanisms in the course of learning and memory processes, which means to test gene activation and/or silencing as well as protein expression during behavioral tasks. In this context, changes of DNA methylation can provide more important information, as it has been the case when the dynamic response of neurotransmitter systems was tested in vivo, in freely moving animals exposed to arsenic (Rodriguez et al., 1998; Zarazua et al., 2006). Considering these responses, dynamic DNA methylation might be substantially different in exposed animals when tested in the course of neuronal plasticity processes. 4.3. Summary By employing an animal model of chronic developmental arsenic exposure, this work demonstrates alterations of DNA

L. Martı´nez et al. / Neurochemistry International 58 (2011) 574–581

methylation in the rat brain, both through immunohistochemical staining of methylated DNA and methylation specific PCR, at exposure levels that induce memory deficits. These findings open the possibility to further analyze the epigenetic regulation of neuronal plasticity as a key target or arsenic and other neurotoxic substances. Acknowledgements This work was supported by the grants PIFI 2007 (UASLP), CONACYT 105937 and PROMEP 103.5/09/571. CONACYT provided the fellowships 208981 to V. Jimenez and 208828 to L. Martinez. The collaboration of D.U. Campos-Delgado and R. Cisneros is gratefully acknowledged. The authors declare they have no financial or employment interest that could have influenced this research. References Bellinger, D.C., 2008. Very low lead exposures and children’s neurodevelopment. Curr. Opin. Pediatr. 20, 172–177. Borrelli, E., Nestler, E.J., Allis, C.D., Sassone-Corsi, P., 2008. Decoding the epigenetic language of neuronal plasticity. Neuron 60, 961–974. Bourdonnay, E., Morzadec, C., Sparfel, L., Galibert, M.-D., Jouneau, S., Martin-Chouly, C., Fardel, O., Vernhet, L., 2009. Global effects of inorganic arsenic on gene expression profile in human macrophages. Mol. Immunol. 46, 649–656. Calderon, J., Navarro, M.E., Jimenez-Capdeville, M.E., Santos Diaz, M.A., Golden, A., Rodriguez, L.I., 2001. Exposure to arsenic and lead and neuropsychological development in Mexican children. Environ. Res. 85, 69–76. Chanda, S., Dasgupta, U.B., GuhaMazumder, D., Gupta, M., Chaudhuri, U., Lahiri, S., Das, S., Ghosh, N., Chatterjee, D., 2006. DNA hypermethylation of promoter of gene p53 and p16 in arsenic-exposed people with and without malignancy. Toxicol. Sci. 89, 431–437. Chen, H., Li, S., Liu, J., Diwan, B.A., Barret, J.C., Waalkes, M.P., 2004. Chronic inorganic arsenic exposure induces hepatic global and individual gene hypomethylation: implications for arsenic hepatocarcinogenesis. Carcinogenesis 25, 1779–1786. Colon, C.M., Jianpeng, W., Ramos, X., Garcia, H.G., Da´vila, J.J., Laguna, J., Rosado, C., ˜ a, O.S., 2006. An inhibitor of DNA recombination blocks memory consolidaPen tion, but not reconsolidation in context fear conditioning. J. Neurosci. 26, 5524– 5533. Fagliolini, M., Jensen, C.L., Champagne, F., 2009. Epigenetic influences on brain development and plasticity. Curr. Opin. Neurobiol. 19, 207–212. Feng, J., Zhou, Y., Campbell, S.L., Le, T., Li, E., Sweatt, J.D., Silva, A.J., Fan, G., 2010. Dnmt1 and Dnmt3a maintain DNA methylation and regulate synaptic function in adult forebrain neurons. Nat. Neurosci. doi:10.1038/nn.2514. Frommer, M., McDonald, L.E., Millar, D.S., Collis, C.M., Watt, F., Grigg, G.W., Molloy, P.L., 1992. A genomic sequencing protocol that yields a positive display of 5methyl-cytosine residues in individual DNA strand. Proc. Natl. Acad. Sci. U.S.A. 80, 1579–1583. Garcı´a Cha´vez, E., Santamarı´a, A., Dı´az-Barriga, F., Mandeville, P., Jua´rez, B.I., Jime´nez-Capdeville, M.E., 2003. Arsenite-induced formation of hydroxyl radical in the striatum of awake rats. Brain Res. 976, 82–89. ˜ a, E., Guerra-Palomares, S.E., Barriga-Moreno, Garcia-Sepulveda, C.A., Carrillo-Acun M., 2010. Maxiprep genomic DNA extractions for molecular epidemiology studies and biorepositories. Mol. Biol. Rep. 37, 1883–1887. Goering, P.L., Aposhian, H.V., Mass, M.J., Cebria´n, M., Beck, B.D., Waalkes, M.P., 1999. The enigma of arsenic carcinogenesis: role of metabolism. Toxicol. Sci. 49, 5–14. Gra¨ff, J., Mansuy, I.M., 2009. Epigenetic dysregulation in cognitive disorders. Eur. J. Neurosci. 30, 1–8. Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral Research, 2003. National Research Council of the Academies. The National Academies Press. Washington, DC. Hsieh, F.I., Hwang, T.S., Hsieh, Y.C., Lo, H.C., Su, C.T., Hsu, H.S., Chiou, H.Y., Chen, C.J., 2008. Risk of erectile dysfunction induced by arsenic exposure through well water consumption in Taiwan. Environ. Health Perspect. 116, 532–536. IARC (International Agency for Research on Cancer), 2004. Monograph on the evaluation of carcinogenic risks to humans: some drinking water disinfectants and contaminants, including arsenic. IARC 84, 185–270. Kilgore, M., Miller, C.A., Fass, D.M., Hennig, K.M., Haggarty, S.J., Sweatt, J.D., Rumbaugh, G., 2010. Inhibitors of class 1 histone deacetylases reverse contextual memory deficits in a mouse model of Alzheimer’s disease. Neuropsychopharmacology 35, 870–880. Koshibu, K., Gra¨ff, J., Beullens, M., Heitz, F.D., Berchtold, D., Russig, H., Farinelli, M., Bollen, M., Mansuy, I.M., 2009. Protein phosphatase 1 regulates the histone code for long-term memory. J. Neurosci. 29, 13079–13089. Levenson, J.M., O’Riordan, K.J., Brown, K.D., Trinh, M.A., Molfese, D.L., Sweatt, J.D., 2004. Regulation of histone acetylation during memory formation in the hippocampus. J. Biol. Chem. 279, 40545–40559.

581

Levenson, J.M., Roth, T.L., Lubin, F.D., Miller, C.A., Huang, I.C., Desai, P., Malone, L.M., Sweatt, J.D., 2006. Evidence that DNA (cytosine-5 methyltransferase regulates synaptic plasticity in the hippocampus. J. Biol. Chem. 281, 15763– 15773. Lin, S., Shi, Q., Nix, B., Styblo, M., Melinda, A.B., Herbin-David, K.M., Hall, L.L., Simeonsson, J.B., Thomas, D.J., 2002. A novel S-adenosyl-L-methionine: arsenic (III) methyltransferase from rat liver cytosol. J. Biol. Chem. 277, 10795–10803. Lorincz, A., Nusser, Z., 2008. Specificity of immunoreactions: the importance of testing specificity in each method. J. Neurosci. 28, 9083–9086. Majumdar, S., Chanda, S., Ganguli, B., Mazumder, D.N., Lahiri, S., Dasgupta, U.B., 2009. Arsenic exposure induces genomic hypermethylation. Environ. Toxicol. doi:10.1002/tox.20497. Mass, M.J., Liangjun, W., 1997. Arsenic alters cytosine methylation patterns of the promoter of the tumor suppressor gene p53 in human lung cells: a model for a mechanism of carcinogenesis. Mutat. Res. 386, 263–277. Miller, C.A., Sweatt, J.D., 2007. Covalent modification of DNA regulates memory formation. Neuron 53, 857–869. Miller, S.A., Dykes, D.D., Polesky, H.F., 1988. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res. 16, 1215–1218. Munton, R.P., Vizi, S., Mansuy, I.M., 2004. The role of protein phosphatase-1 in the modulation of synaptic and structural plasticity. FEBS Lett. 567, 121–128. Murgatroyd, C., Patchev, A.V., Wu, Y., Micale, V., Bockmu¨hl, Y., Fischer, D., Holsboer, F., Wotjak, C.T., Almeida, O.F., Spengler, D., 2009. Dynamic DNA methylation programs persistent adverse effects of early-life stress. Nat. Neurosci. 12, 1559– 1566. NRC (National Research Council), 2001. Arsenic in drinking water: 2001 Update. National Academy Press, Washington, DC. Paxinos, G., Watson, C., 2005. The Rat Brain in Stereotaxic Coordinates. Academic Press, New York. Pilsner, J.R., Liu, X., Ahsan, H., Ilievski, V., Slavkovich, V., Levy, D., Factor-Litvak, P., Graziano, J.H., Gamble, M.V., 2007. Genomic methylation of peripheral blood leukocyte DNA: influences of arsenic and folate in Bangladeshi adults. Am. J. Clin. Nutr. 86, 1179–1186. Reichard, J.F., Schnekenburger, M., Puga, A., 2007. Long term low-dose arsenic exposure induces loss of DNA methylation. Biochem. Biophys. Res. Comm. 352, 188–192. Rios, R., Zarazua, S., Santoyo, M.E., Sepulveda-Saavedra, J., Romero-Diaz, V., Jimenez, V., Perez-Severiano, F., Vidal-Cantu, G., Delgado, J.M., Jimenez-Capdeville, M.E., 2009. Decreased nitric oxide markers and morphological changes in the brain arsenic-exposed rats. Toxicology 261, 68–75. Rodriguez, V.M., Dufour, L., Carrizales, L., Diaz-Barriga, F., Jimenez-Capdeville, M.E., 1998. Effects of oral exposure to a mining waste on in vivo striatal dopamine release. Environ. Health Perspect. 106, 487–491. Rosado, J.L., Ronquillo, D., Kordas, K., Rojas, O., Alatorre, J., Lopez, P., Garcia-Vargas, ˜ o, M., Cebria´n, M.E., Stolzfus, R.J., 2007. Arsenic exposure G., Del Carmen Caaman and cognitive performance in Mexican schoolchildren. Environ. Health Perspect. 115, 1371–1375. Salgado-Bustamante, M., Ortiz-Perez, M.D., Calderon-Aranda, E., Estrada-Capetillo, ˜ o-Moreno, P., Gonzalez-Amaro, R., Portales-Perez, D., 2010. Pattern of L., Nin expression of apoptosis and inflammatory genes in humans exposed to arsenic and/or fluoride. Sci. Total Environ. 408, 760–767. Sciandrello, G., Caradonna, F., Mauro, M., Barbata, G., 2004. Arsenic-induced DNA hypomethylation affects chromosomal instability in mammalian cells. Carcinogenesis 25, 413–417. Sharma, R.P., Gavin, D.P., Grayson, D.R., 2010. CpG methylation in neurons: message, memory or mask? Neuropsychopharmacology 35, 2009–2020. Tsai, S.Y., Chou, H.Y., The, H.M., Chen, C.M., Chen, C.J., 2003. The effects of chronic arsenic exposure from drinking water on the neurobehavioral development in adolescence. Neurotoxicology 24, 747–753. Tsankova, N., Renthal, W., Kumar, A., Nestler, E.J., 2007. Epigenetic regulation in psychiatric disorders. Nat. Rev. 8, 355–367. Vahter, M., 2007. Health effects of early life exposure to arsenic. Basic Clin. Pharmocol. Toxicol. 102, 204–211. Wasserman, G.A., Liu, X., Parvez, F., Ahsan, H., Facto-Litvak, P., Van-Geen, A., Slavkovich, V., Lolacono, N.J., Cheng, Z., Hussain, I., Momotaj, H., Graziano, J.H., 2004. Water arsenic exposure and children’s intellectual function in Araihazar, Bangladesh. Environ Health Perspect. 112, 1329–1334. WHO (World Health Organization), 2001. Arsenic and Arsenic Compounds. International Program on Chemical Safety. World Health Organization, Geneva, Switzerland. Xie, Y., Trouba, K.J., Liu, J., Waalkes, M.P., Germolec, D.R., 2004. Biokinetics and subchronic toxic effects of oral arsenite, arsenate, monomethylarsonic acid, and dimethylarsinic acid in v-Ha-ras transgenic, Tg.AC) mice. Environ. Health Perspect. 112, 1255–1263. Zarazua, S., Perez-Severiano, F., Delgado, J.M., Martinez, L.M., Ortiz-Perez, D., Jimenez-Capdeville, M.E., 2006. Decreased nitric oxide production in the rat brain after chronic arsenic exposure. Neurochem. Res. 31, 1069–1077. Zarazua, S., Rios, R., Delgado, J.M., Santoyo, M.E., Ortiz-Perez, D., Jimenez-Capdeville, M.E., 2010. Decreased arginine methylation and myelin alterations in arsenic exposed rats. Neurotoxicology 31, 94–100. Zhao, Q.C., Young, R.M., Diwan, A.B., Coogan, P.T., Waalkes, P.M., 1997. Association of arsenic-induced malignant transformation with DNA hypomethylation and aberrant gene expression. Proc. Natl. Acad. Sci. U.S.A. 94, 10907–10912.