Chronic exposure to polychlorinated biphenyls (PCBs), a class of ubiquitous environmental toxicants, causes neurocognitive anomalies. The transcription factor repressor element 1-silencing transcription factor (REST) plays a critical role in neuronal phenotype elaboration in both neural progenitor cells and non-neuronal cells. Here, we investigated the possible relationship between PCBs and REST in neuroblastoma SH-SY5Y cells. In these cells, chronic exposure to the PCB mixture Aroclor 1254 (A-1254; 5–30 μg/ml) caused dose-dependent cell death via the induction of calpain but not caspase-3. Intriguingly, this effect was prevented by the calpain inhibitor calpeptin. Furthermore, A-1254 enhanced REST mRNA and protein expression levels after both 24 and 48 h. REST down-regulation by small interfering RNA prevented A-1254-induced cell death. In addition, A-1254 enhanced the binding of REST to the synapsin 1 gene promoter, and synapsin 1 knockdown potentiated A-1254-induced cell death. A-1254 (10 μg/ml) also increased the expression of the two REST cofactors, the REST corepressor and the mammalian SIN3 homolog A transcription regulator. Moreover, the PCB mixture decreased acetylation of the histone proteins H3 and H4. It is noteworthy that the histone deacetylase inhibitor trichostatin A prevented such decreases and reduced the A-1254-induced neurotoxic effect. Collectively, these results suggest that A-1254 exerts its toxic effect via REST by down-regulating synapsin 1 and decreasing H3 and H4 acetylation.
Polychlorinated biphenyls (PCBs) are a structurally related group of stable and highly lipophilic toxic chemicals with widespread distribution throughout the environment (Yang et al., 2009). Improper disposal and accidental release of these compounds has led to their diffusion in the environment, placing them on the list of widespread environmental contaminants. Their lipophilicity has favored the buildup of increasingly higher levels in various food chains. PCBs can now be detected globally in different environmental matrices, wildlife, food stuffs, and humans (Ross et al., 2000; Weintraub and Birnbaum, 2008; Tewari et al., 2009). Corroborating evidence exists for the toxicity of these compounds in the central nervous system of both humans and experimental animal models (Canzoniero et al., 2006; Carpenter, 2006). Indeed, alterations in the normal pattern of dendritic growth and plasticity have been associated with long-term exposure to PCBs (Yang et al., 2009).
Several lines of evidence indicate that the repressor element 1 (RE1)-silencing transcription factor (REST) is a key protein in the regulation of a large network of genes essential for neuronal functions. Indeed, being a transcriptional repressor, it participates in the regulation of gene expression throughout the body. REST was first identified in 1995 as a protein that binds to the sequences of the RE1, also known as neuron restrictive silencer element. These sequences are present in the rat Scn2a2 (also known as Nav1.2) and Stmn2 (also known as SCG10) genes (Mori et al., 1992). REST consists of three regions: 1) an N-terminal repressor domain, characterized by a cluster of eight zinc fingers and functioning as a DNA-binding domain, 2) a repeat region, and 3) a C-terminal repressor domain with a single zinc finger motif (Ballas and Mandel, 2005). Upon binding to RE-1, REST represses multiple target genes in non-neuronal tissues and undifferentiated neural precursors of the central nervous system. This mechanism, which ensures the proper control of neuronal gene expression timing during neurogenesis (Ballas et al., 2005), is involved in a number of different neurological disorders including brain ischemia (Formisano et al., 2007; Zukin, 2010), epilepsy (Spencer et al., 2006), Huntington's disease (Zuccato et al., 2003), Down syndrome (Lepagnol-Bestel et al., 2009), and neuropathic pain (Uchida et al., 2010).
Although the genotoxic effects of PCBs have been extensively investigated, and several epidemiological studies have highlighted their health risks (DeCaprio et al., 2005; Carpenter, 2006), the genomic mechanisms responsible for their neurodevelopmental effects are still unclear.
Thus, in an attempt to gain further insights into the involvement of REST in PCB-induced toxicity, we investigated the possible relationship between PCBs and REST in neuroblastoma SH-SY5Y cells. We identified REST and its target gene synapsin 1 as components of a novel molecular mechanism by which PCBs induce their toxic effects.
Materials and Methods
Drug and Chemicals.
Aroclor 1254 (A-1254) and trichostatin A (TSA) were obtained from Sigma-Aldrich (St. Louis, MO), and calpeptin was from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). All chemicals were diluted in cell culture medium. For those requiring dilution in DMSO, the final DMSO concentration was 1%. DMSO was added to the control cells at the same concentration as that used for treated cells. The addition of DMSO alone caused no cellular toxicity.
Human SH-SY5Y cells (LGC Standards S.r.l., Sesto San Giovanni, Italy) were cultured as monolayers in polystyrene dishes in Dulbecco's modified Eagle's medium containing 15% heat-inactivated fetal bovine serum, 1% l-glutamine (200 mM), 1% sodium pyruvate (100 mM), 100 IU/ml penicillin, and 100 μg/ml streptomycin. All of the above reagents were purchased from Invitrogen (Milan, Italy). Cells were grown in a humidified incubator at 37°C in a 5% CO2 atmosphere, and the medium was changed every 2 days. Each experiment was performed using cells (passages 15–30) plated on multiwell plates. After 24 and 48 h of cell seeding, cells were incubated with Aroclor 1254 (stock solution; 1 mg/ml) in Dulbecco's modified Eagle's medium containing 5% heat-inactivated fetal bovine serum.
Determination of Cell Viability, Evaluated as Mitochondrial Activity.
Cell death (apoptosis and/or necrosis) was analyzed by fluorescence-activated cell-sorting analysis (FACS) using annexin V-coupled fluorescein isothiocyanate (FITC) binding and propidium iodide (PI) staining (BD Pharmingen, Milan, Italy). PI stains the nuclei of dead and dying cells with damaged membranes. Annexin V-FITC-positive and PI-negative cells were defined as apoptotic. Cells were collected by centrifugation. After being washed twice with annexin binding buffer (10 mM HEPES, 140 mM NaCl, and 2.5 mM CaCl2, pH 7.4), the cells were resuspended in 100 μl of the same buffer and incubated with annexin V-FITC for 20 min on ice. Then, 400 μl of annexin binding buffer containing PI was added, and analysis was performed by FACS. These experiments were replicated four times on 20,000 cells/sample. Mitochondrial dysfunction was evaluated using the 3[4,5-dimethylthiazol-2-y1]-2,5-diphenyltetrazolium bromide (MTT) test. In this test, the dye MTT is metabolized by viable mitochondria to a colored product that can be detected using a spectrophotometer at a wavelength of 540 nm. Data obtained from three independent experimental sessions were expressed as a percentage of the mitochondrial viability of sham-treated cultures.
Western Blot Analysis and Immunoprecipitation.
After treatment, cells were collected and centrifuged. The pellet was lysed at 4°C in radioimmunoprecipitation assay buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Nonidet P-40, and 0.5% sodium deoxycholate, 0.1% SDS) containing the complete protease inhibitor cocktail (Calbiochem, San Diego, CA). Samples were cleared by centrifugation, and the supernatants were used for Western blot analysis. Protein concentration was determined by Bradford's method (Bradford, 1976). Samples (100 or 70 μg) were first separated on 8% (for REST, CoREST, mSin3A, and calpain) or 12% (procaspase 3, H3 acetyl, and H4 acetyl) SDS-polyacrylamide gel electrophoresis and then transferred onto Hybond ECL nitrocellulose membranes (GE Healthcare, Milan, Italy). Membranes were blocked with 5% nonfat dry milk in 0.1% Tween 20 (2 mM Tris-HCl and 50 mM NaCl, pH 7.5) for 2 h at room temperature. Next, they were incubated overnight at 4°C in blocking buffer with one of the following antibodies: 1:1000 antibody for REST (polyclonal rabbit antibody; Millipore Corporation, Milan, Italy), 1:1000 antibody for CoREST (polyclonal rabbit antibody; Millipore Corporation), 1:250 antibody for mSin3A (polyclonal rabbit antibody; Santa Cruz Biotechnology Inc.), 1:250 antibody for calpain (polyclonal rabbit antibody; Santa Cruz Biotechnology Inc.), 1:1000 antibody for H3 acetyl (polyclonal rabbit antibody; Millipore Corporation), 1:1000 antibody for H4 acetyl (polyclonal rabbit antibody; Millipore Corporation), or 1:500 antibody for procaspase 3 (monoclonal mouse antibody; Santa Cruz Biotechnology Inc.), in each case together with 1:3000 antibody for β-actin (monoclonal mouse antibody; Sigma-Aldrich). Immunoreactive bands were detected using enhanced chemiluminescence reagent (GE Healthcare), and their optical density was determined using a Chemi-Doc Imaging System (Bio-Rad Laboratories, Hercules, CA). The experiments were replicated as follows: four times for REST, three times for CoREST and mSin3A, three times for calpain and procaspase, and four times for H3 and H4 acetyl proteins. For immunoprecipitation experiments, nuclear extracts (1000 μg) were incubated with 1:100 antibody for REST (polyclonal goat antibody; Santa Cruz Biotechnology Inc.) and IgG (as the negative control) for 16 h at 4°C in 300 μl of immunoprecipitation buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 1 mM Na3VO4, and 50 mM NaF). Protein A/G-agarose beads (25 μl) were used to precipitate the associated protein. The protein-agarose beads were pelleted, washed three times with cold lysis buffer, and boiled with SDS loading buffer. Immunoprecipitated proteins were resolved using 8% SDS-polyacrylamide gel electrophoresis and then detected by Western blot analysis using 1:1000 antibody for CoREST (polyclonal rabbit antibody; Millipore Corporation) or 1:250 antibody for mSin3A (polyclonal rabbit antibody; Santa Cruz Biotechnology Inc.).
RNA Extraction, RT-PCR, and Real-Time RT-PCR.
Total RNA was extracted from human SH-SY5Y cells using TRIzol by following the supplier's instructions (Invitrogen). Total RNA was treated with DNase I (1 U/μl; Sigma-Aldrich) for 15 min at room temperature. The first-strand cDNA was synthesized using 5 μg of the total RNA and 500 ng of random primers by means of the SuperScript first-strand synthesized system for the reverse transcriptase-polymerase chain reaction (high-capacity cDNA RT kit; Applied Biosystems, Monza, Italy). Using 1/10 of the cDNAs as a template, quantitative real-time PCR was carried out in a 7500 fast real-time PCR system (Applied Biosystems) with Fast SYBR Green Master Mix (Applied Biosystems). Samples were amplified simultaneously in triplicate in a one-assay run, and the threshold cycle (cT) value for each experimental group was determined. Normalization of the data were performed using β-actin. To evaluate differences in mRNA content between groups, normalized values were entered into the formula 2−ΔΔct. The oligonucleotide sequences for synapsin 1 were 5′-GCACGTCCTGGCTGGGTTTCTGGG-3′ and 5′-AGGCTACCCGTCAGACATCCGTCTC-3′, whereas those for β-actin were 5′-AGACCTCTAGCCAACACAGT-3′ and 5′-GACACACCTAACCACCGAGAT-3′. The PCR experiments were replicated four times for each gene.
Transfection with Small Interfering RNA or Antisense Oligodeoxynucleotides.
Transfection of the human SH-SY5Y cell-line was carried out using HiPerFect Transfection Reagent (QIAGEN, Milan, Italy). Scrambled control (siCTL) or small interfering RNA against REST (siREST; 20 nM) was added to HiPerFect Transfection Reagent for 15 min at room temperature, and then the mixture was added to cells plated on 35-mm-diameter plastic dishes in Optimem medium (Invitrogen). After 2 h of incubation, the medium was replaced with fresh medium. Synapsin 1 antisense oligodeoxynucleotides were prepared as described previously (Iwakuma et al., 2003).
Chromatin Immunoprecipitation Assay.
Cells were cross-linked with 1% formaldehyde, collected, and then lysed in a buffer containing 50 mM Tris, pH 8.1, 1% SDS, 10 mM EDTA, and antiprotease. DNA was sheared by sonication. After removal of cellular debris by centrifugation, equal amounts of DNA were incubated with 3 μg of antibody for REST, previously bound to Protein A-agarose/salmon sperm DNA (Millipore Corporation), with IgG as negative control. After rotating for 2 h at 4°C on a spinning wheel, the beads were washed once with each of the following buffers: high-salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, and 500 mM NaCl), low-salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, and 150 mM NaCl), LiCl buffer (0.25 M LiCl, 1% Nonidet P-40, 1% deoxycholate, 1 mM EDTA, and 10 mM Tris-HCl, pH 8.1), and Tris/EDTA buffer (10 mM Tris, pH 8.1 and 1 mM EDTA). The precipitated fragments were eluted with a buffer containing 1% SDS and 0.1 M NaHCO3. DNA was analyzed by quantitative real-time PCR using Fast SYBR Green Master Mix (Applied Biosystems). The data of the mean of three different real-time PCR experiments were expressed as a percentage of the relevant controls, all normalized for the DNA input. The following oligonucleotides were used for the amplification of immunoprecipitated DNA: RE-1 synapsin 1, forward-5′AAGGTGGCCGGGAAGGGGAGT-3′; reverse-5′-GGGGAAACAGGATGCGGAG-3′. The chromatin immunoprecipitation (ChIP) experiment, which evaluated the binding of REST to the synapsin 1 promoter region, was replicated four times.
The data were evaluated as means ± S.E.M. Statistically significant differences among means were determined by analysis of variance followed by the Student-Newman-Keuls test (p < 0.05).
A-1254 Exposure Induced Cell Death by Increasing REST Expression in SH-SY5Y Neuroblastoma Cells.
A 48-h exposure to A-1254 (3–30 μg/ml) induced SH-SY5Y cell death in a concentration-dependent manner (Fig. 1A). More specifically, when the cells were exposed to 10 μg/ml of A-1254, which produced an intermediate degree of toxicity, a significant increase in the number of PI-positive cells was detected by FACS (Fig. 1A, inset). Under the same conditions, calpain protein expression was significantly increased after 24 and 48 h of exposure (versus control) (Fig. 1B), whereas the apoptotic marker procaspase-3 was unaffected (Fig. 1C).
To confirm the involvement of calpain in A-1254-induced cell death, we tested the effect of its inhibitor calpeptin on SH-SY5Y cell survival (Fig. 1D). This inhibitor (30 μM) was able to prevent the toxic effects of a 48-h exposure to A-1254 (10 μg/ml) (Fig. 1D).
Furthermore, 24 and 48 h of exposure to A-1254 (10 μg/ml) significantly increased the expression of REST mRNA and protein in SH-SY5Y cells, which constitutively express this transcription factor (Fig. 2, A and B). Then, to evaluate the role of REST in A-1254-induced toxicity, we knocked down its expression by almost 50% using small interfering RNA (Fig. 2C). It is noteworthy that the siREST counteracted the detrimental effect of A-1254, whereas its scrambled control sequence did not (Fig. 2D).
A-1254 Increased the Expression of Both CoRest and mSin3A Transcription Regulator Cofactors and Decreased Synapsin 1 and Histone Protein Acetylation.
Prolonged PCB exposure, generally associated with specific neurocognitve alterations (Carpenter, 2006), alters expression of several neuronal genes. For instance, evidence has shown that synapsin 1, one of the REST target genes regulated by histone deacetylase (HDAC), is involved in the pathogenesis of some neurocognitive disorders (Wu et al., 2007) and neurodegenerative diseases (Jung et al., 2004). Accordingly, we hypothesized that this gene could serve as a possible mediator of PCB-induced neurocognitve alterations.
Thus, to verify the involvement of synapsin 1 in neuronal impairment caused by PCB exposure, we subjected SH-SY5Y cells to the ChIP assay after 24- or 48-h exposure to A-1254 (10 μg/ml). A significant increase in the binding of REST to the synapsin 1 gene promoter was detected at both time points (Fig. 3A). Moreover, A-1254 (10 μg/ml) significantly reduced synapsin 1 gene expression in SH-SY5Y cells (Fig. 3B). Then to further confirm that synapsin 1 was involved in A-1254-induced cell death, we treated the cells with antisense oligonucleotides (ODN-AS) for 48 h (Fig. 3C) to reduce its expression. ODN-AS (1 μM) significantly decreased synapsin 1 expression by approximately 70%, whereas its missense sequence was ineffective (Fig. 3C). Furthermore, when the same concentration of ODN-AS was administered concomitantly with A-1254 (10 μg/ml/48 h), it significantly potentiated the toxic effect of A-1254 on SH-SY5Y cells (Fig. 3D).
It is well known that REST requires two cofactors (mSin3A at the N terminus and CoREST at the C terminus) to repress its target genes (Ballas et al., 2005). As shown in Fig. 4, SH-SY5Y neuroblastoma cells constitutively express both CoREST and mSin3A. Exposure to A-1254 (10 μg/ml) for 24 or 48 h up-regulated both cofactors (Fig. 4, A and B). To confirm that the increased proteins were associated with REST, we immunoprecipitated REST with REST antibody and, after that, we immunoblotted the samples with specific antibodies for CoREST or mSin3A. After 48 h of exposure to A-1254, REST bound to both CoREST and mSin3A, whereas no binding was detected in the preimmune group (Fig. 4C). It has been reported that REST exerts its effects via epigenetic mechanisms such as histone acetylation (Formisano et al., 2007). A significant reduction in H3 and H4 acetylation was observed after incubation with A-1254 (10 μg/ml for 24 or 48 h) (Fig. 4, D and E). These reductions were prevented by the HDAC inhibitor TSA (50 nM) (Kazantsev and Thompson, 2008) (Fig. 5, A and B). It is noteworthy that cell death induced by A-1254 exposure (10 μg/ml, 48 h) was inhibited in a concentration-related manner by the coapplication of TSA (Fig. 5C).
The present study has demonstrated that prolonged exposure to A-1254 induced a concentration-dependent reduction in cell viability in SH-SY5Y neuroblastoma cells through a calpain-dependent pathway. PCB exposure enhanced the expression of the transcription factor REST at both the transcript and protein levels. Consistently, the two REST cofactors mSin3A and CoREST were also up-regulated and bound to REST after A-1254 exposure. It is noteworthy that REST knockdown prevented A-1254-induced cell death, which occurred through calpain but not caspase-3 activation. In this context, it should be mentioned, however, that the involvement of this pathway in PCB-induced cell death is still controversial. Although some authors support the hypothesis that caspase-3-dependent apoptosis may mediate PCB-induced cell death (Lee et al., 2001), others (Cannavò et al., 2005) instead maintain that additional factors may indeed be involved. In particular, PCBs may cause necrosis or apoptosis depending on the type of cell line adopted, the concentrations used, and the duration of exposure. On the other hand, calpain activation is consistent with the increase in intracellular Ca2+ concentration occurring in SH-SY5Y cells after A-1254 exposure (Canzoniero et al., 2006).
It has been reported that REST dysregulation is implicated in some pathological disorders of the nervous system, such as global ischemia and epileptogenesis (Spencer et al., 2006; Yang et al., 2009). Likewise, it has been shown that REST is overexpressed in experimental models of Parkinson's disease (Yu et al., 2009). To our knowledge, we provide here the first evidence that REST is involved in PCB-induced neurotoxicity. Furthermore, we identified the possible mechanism by which REST might mediate A-1254-induced toxicity in neuroblastoma cells. It is noteworthy that Ballas et al. (2005) reported that REST acts through several epigenetic mechanisms. Specifically, they pointed out that it recruits its corepressors CoREST and mSin3, which, in turn, interact directly with the histone deacetylases. Our data revealed that A-1254 exposure not only up-regulated CoREST and mSin3A, but also significantly decreased H3 and H4 acetylation. Furthermore, the HDAC inhibitor trichostatin A significantly reduced not only the decreases in H3 and H4 acetylation but also the neurotoxicity of A-1254 mixture in neuroblastoma cells. This suggests that the detrimental effects of this PCB mixture are exerted at least in part via an inhibition of the epigenetic mechanism involved in acetylation. Indeed, histone protein acetylation is also modulated by other environmental toxicants, including the polyaromatic hydrocarbon benzopyrene that is able to change the acetylation status of specific genes (Sadikovic et al., 2008). However, on the basis of the present findings, we cannot exclude the likelihood that changes in the methylation of histone proteins might take part in the detrimental REST-mediated effects of A-1254 in neuroblastoma cells.
It is noteworthy that inhibition of HDAC by trichostatin A can prevent REST-mediated down-regulation of synapsin 1 in human neural stem cells (Ekici et al., 2008). In our study, synapsin 1 was significantly down-regulated after prolonged exposure to A-1254, an effect that was associated with an increase in REST binding to the synapsin 1 promoter. In addition, synapsin 1 knockdown by antisense oligonucleotides potentiated the toxic effect of A-1254 on SH-SY5Y cells. These results are consistent with the findings of Yu et al. (2009). In their study, conducted on a model of Parkinson's disease, they demonstrated that REST down-regulated synapsin 1. Other authors have demonstrated that synapsin 1 is activated in hippocampus exposed to ischemic injury as part of a plastic adaptation mechanism (Jung et al., 2004). All of this evidence suggests that synapsin 1 could represent a marker of neurodegeneration because it is involved in the modulation of neurotransmitter release and synaptic vesicle reclustering (Nemani et al., 2010). Consistently, transgenic mice overexpressing α-synuclein and, thus, more prone to develop Parkinson's disease, show a 20 to 45% reduction in the amount of synapsin 1 and an impairment of neurotransmitter release (Nemani et al., 2010). Furthermore, the hypoxia-inducible factor-1-dependent increase in dopamine release prevents 6-hydroxydopamine-dependent cell death in a dopaminergic cell model similar to SH-SY5Y cells, thereby unmasking the neurothrophic role of the neurotransmitter (Johansen et al., 2010).
Accordingly, the REST-mediated decrease in synapsin 1 expression observed in this study could indeed underlie the damaging effect of PCBs on neuronal differentiation possibly via a dysfunctional neurotransmitter release. Nonetheless, activation of other REST target genes by A-1254 cannot be ruled out. Collectively, the present results indicate that the calpain-dependent cell death induced by A-1254 in SH-SY5Y cells depends on REST overexpression. Furthermore, we demonstrated that this activation elicits an interwoven chain of events entailing the recruitment of its two cofactors, CoREST and mSin3A, and the subsequent decreases in H3 and H4 acetylation and synapsin 1 expression. These results may pave the way toward a better understanding of those human neurocognitive anomalies generally associated with prolonged exposure to PCBs. Moreover REST-mediated epigenetic modifications highlighted in this study may be a new research area that can be developed to provide new therapeutic agents.
Participated in research design: Formisano, Secondo, Di Renzo, and Canzoniero.
Conducted experiments: Formisano, Guida, Cocco, Secondo, Sirabella, Ulianich, and Paturzo.
Performed data analysis: Formisano, Guida, Cocco, Secondo, Sirabella, Ulianich, and Paturzo.
Wrote or contributed to the writing of the manuscript: Formisano, Secondo, Di Renzo, and Canzoniero.
We thank Dr. Paola Merolla for editorial revisions.
This work was supported by COFIN 2008; the Ministero della Salute, Ricerca Sanitaria [Grant RF-FSL352059]; Ricerca Finalizzata 2006; the Ministero della Salute, Ricerca Oncologica 2006; the Ministero della Salute, Progetto Strategico 2007; and the Ministero della Salute, Progetto Ordinario 2007.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- polychlorinated biphenyl
- Aroclor 1254
- repressor element 1
- RE1-silencing transcription factor
- REST corepressor
- small interfering RNA against REST
- scrambled control
- histone deacetylase
- trichostatin A
- dimethyl sulfoxide
- 3[4,5-dimethylthiazol-2-y1]-2,5-diphenyltetrazolium bromide
- chromatin immunoprecipitation
- reverse transcriptase
- polymerase chain reaction
- RNA interference
- fluorescence-activated cell-sorting analysis
- fluorescein isothiocyanate
- propidium iodide
- antisense oligonucleotides
- mammalian Sin3 homolog A.
- Received March 3, 2011.
- Accepted June 20, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics