Polychlorinated biphenyl (PCB) exposure has been associated with neurodegenerative diseases, such as Parkinson’s disease, amyotrophic lateral sclerosis, and dementia. Neuronal death elicited by the PCB mixture Aroclor 1254 (A1254) has been attributed to an increase in RE-1–silencing transcription factor (REST), which, in turn, correlates with a decrease in the synapsin-1 promoter gene. Although histone deacetylase (HDAC) inhibitors are known to be neuroprotective in several neurologic disorders, the core mechanisms governing this effect are not yet understood. Here, to examine how HDAC class I [N-(2-aminophenyl)-4-[N-(pyridin-3-yl-methoxycarbonyl)aminomethyl]-benzamide (MS-275)] and HDAC class II [3-[5-(3-(3-fluorophenyl)-3-oxopropen-1-yl)-1-methyl-1H-pyrrol-2-yl]-N-hydroxy-2-propenamide (MC-1568)] inhibitors prevent A1254-induced neuronal cell death, we exposed SH-SY5Y neuroblastoma cells to A1254. Exposure to A1254 (30.6 μM) for 24 and 48 hours resulted in a time-dependent cell death. Indeed, after 48 hours, MS-275, but not MC-1568, reverted A1254-induced cell death in a dose-dependent manner. Furthermore, A1254 significantly increased HDAC3, but not HDAC1 or HDAC2. Interestingly, REST physically interacted with HDAC3 after A1254 exposure. Chromatin immunoprecipitation assays revealed that MS-275 reverted the increased levels of HDAC3 binding and decreased acetylation of histone H3 within the synapsin-1 promoter region, thus reverting synapsin-1 mRNA reduction. Moreover, REST knockdown by small interfering RNA (siRNA) prevented HDAC3 from binding to the synapsin-1 promoter. Likewise, HDAC3 siRNA significantly reduced A1254-induced cell toxicity in SH-SY5Y cells and cortical neurons. Hence, this study demonstrates that inhibition of HDAC class I attenuates A1254-induced neuronal cell death by preventing HDAC3 binding and histone deacetylation within the synapsin-1 promoter region.
Polychlorinated biphenyls (PCBs) are developmental neurotoxins (Chen et al., 2011) that may damage a variety of neuropsychological functions in children, including general cognition, memory, attention, visual-spatial, executive, and motor functions (Humphrey et al., 2000; Boucher et al., 2012). Aroclor 1254 (A1254), a commercial mixture of PCBs (Webb and McCall, 1972), frequently found in various foods and in human specimens at contaminated sites (Tilson and Kodavanti, 1997), is widely used for studying PCB toxicity (Canzoniero et al., 2006; Adornetto et al., 2013). At the neuronal level, A1254 can induce mitochondrial dysfunction in dopaminergic neurons (Lee and Opanashuk, 2004; Lee et al., 2012) and neuronal death of cerebellar granule cells (Mariussen et al., 2002) and cortical neurons (Inglefield et al., 2001; Formisano et al., 2015). Neuronal death induced by A1254 is likely due to the activation of RE-1–silencing transcription factor (REST) via the ERK2\Sp1\Sp3 pathway (Formisano et al., 2015), a phenomenon that, in turn, downregulates synapsin-1 expression (Formisano et al., 2011). Because REST requires histone deacetylase (HDAC) activity to repress neuronal gene transcription in both non-neuronal (Iannotti et al., 2013) and neuronal cell lines (Formisano et al., 2013), it is likely that HDAC is also involved in the mechanisms leading to PCB-induced neuronal toxicity. Indeed, several lines of evidence indicate that alterations in the expression and function of HDAC are involved in neurodegeneration. Conversely, inhibition of HDAC can ameliorate stress-related behavior in a wide range of neurologic disorders such as Huntington's disease and Parkinson's disease, as well as psychiatric disorders such as anxiety, mood disorders, Rubinstein-Taybi syndrome, and Rett syndrome (Abel and Zukin, 2008).
The hypothesis that HDACs might play a role in the mechanisms leading to neuronal toxicity after PCB exposure is substantiated by several neurologic studies. One study shows that MS-275 [N-(2-aminophenyl)-4-[N-(pyridin-3-yl-methoxycarbonyl)aminomethyl]-benzamide], a class I histone deacetylase inhibitor (Beharry et al., 2014), reduces cell death following traumatic brain injury (Cao et al., 2013). Likewise, another study indicates that inhibition of class II histone deacetylases in the spinal cord attenuates inflammatory hyperalgesia (Bai et al., 2010). Still other studies demonstrate that MC-1568 (3-[5-(3-(3-fluorophenyl)-3-oxopropen-1-yl)-1-methyl-1H-pyrrol-2-yl]-N-hydroxy-2-propenamide), a class II HDAC inhibitor (Beharry et al., 2014), can prevent di(2-ethylhexyl)phthalate-induced neurotoxicity (Guida et al., 2014) and promotes neurite growth and arborization protecting neurite arbors against neurotoxic insult (Collins et al., 2014). Furthermore, in animal models of brain ischemia, injection of the pan-HDAC inhibitor trichostatin A, an inhibitor of the key components of the REST-corepressor complex of HDAC1 and HDAC2, ameliorates neuronal injury (Noh et al., 2012). Similarly, trichostatin A also prevents A1254-induced neurotoxic effects in a dose-dependent manner (Formisano et al., 2011).
Here, we investigated the effect of the class I HDAC inhibitor MS-275 and the class II HDAC inhibitor MC-1568 on A1254-induced cell death to verify which of these two classes is involved in the toxic effect of PCBs, and the involvement of the HDAC isoform with REST in A1254-induced synapsin-1 reduction and cell death.
Material and Methods
Drug and Chemicals.
A1254 (lot: LB794306Vstock solution 153 mM) was purchased from Supelco (Bellefonte, PA). Culture media and sera were obtained from Invitrogen (Milan, Italy). The HDAC inhibitors MS-275) (stock solution 100 µM) and MC-1568 (stock solution 1 M) were obtained from Sigma-Aldrich (St. Louis, MO). The calpain inhibitor calpeptin was from Santa Cruz Biotechnology (stock solution 1 mM; Santa Cruz, CA). All chemicals were diluted in cell culture medium as previously reported (Formisano et al., 2011; Guida et al., 2014).
Cell Lines and Culture Conditions.
Human neuroblastoma SH-SY5Y cells (IRCCS Azienda Ospedaliera Universitaria San Martino-IST-Istituto Nazionale per la Ricerca sul Cancro, Genoa, Italy) were grown as previously reported (Formisano et al., 2011; Guida et al., 2014). All experiments were conducted in cultures containing cells between the 10th and 25th passage. After 24 hours of cell seeding, A1254 (30.6 μM) was added to Dulbecco’s modified Eagle’s medium containing 5% fetal bovine serum. Thus, cells were exposed to A1254 for 24 and 48 hours. The experiments on cortical neurons were approved by the Animal Care Committee of “Federico II,” University of Naples (Naples, Italy). Cortical neurons were prepared and cultured from mixed-sex embryonic day 17 Sprague-Dawley rats (Charles River, Calco, Milan, Italy) as previously described (Guida et al., 2014; Vinciguerra et al., 2014; Formisano et al., 2015). To study the effects of HDAC and calpain inhibitors, cells were seeded and treated with 0.1% dimethylsulfoxide (vehicle) or pretreated with MS-275 (0.05, 0.5, and 5 µM in SH-SY5Y cells and 1 µM in cortical neurons) and MC-1568 (0.05, 0.5, and 5 µM in SH-SY5Y cells) for 2 hours, and with calpeptin (30 µM) for 6 hours. After pretreatments, cells were exposed to A1254 (30.6 μM) for 48 hours in SH-SY5Y cells and for 24 hours in cortical neurons. SH-SY5Y cells were plated at a density of 2 × 105 in 24-multiwell plates for MTT (3[4,5-dimethylthiazol-2-y1]-2,5-diphenyltetrazolium bromide) assay, lactate dehydrogenase (LDH) assay, and propidium iodide staining, and in 100-mm-well plates at a density of 10 × 105 for quantitative reverse-transcription polymerase chain reaction (PCR), Western blot, chromatin immunoprecipitation (ChIP), and immunoprecipitation analyses. Cortical neurons were plated at a density of 1 × 106 in 24-well plates for LDH and MTT assays and in 100-mm-well plates at a density of 15 × 105 for Western blot.
Gene silencing by small interfering RNA (siRNA) was carried out by transfecting cells with scrambled control, with siRNAs against human HDAC3 (siHDAC3), or against rat HDAC3 (siHDAC3) at 400 nM (Santa Cruz Biotechnology). Cells were also transfected with siRNAs against human REST (siREST) or against rat REST (siREST) at 20 nM (Qiagen, Milan, Italy). siRNAs for humans and rats, transfected into SH-SY5Y cells and cortical neurons, respectively, were performed as previously reported (Formisano et al., 2011; Guida et al., 2014). Transfection efficiency was ≅70% for SH-SY5Y cells and ≅60% for cortical neurons (data not shown). Finally, cells were exposed to A1254 (30.6 μM) in neurons for 24 hours and in SH-SY5Y cells for 48 hours.
Determination of Cell Viability.
Cell viability was determined as previously described (Guida et al., 2014) using the MTT (Sigma-Aldrich) staining. In particular, cells were incubated in 0.5 mg/ml MTT solution for 2 hours at 37°C, after treatment with A1254 in cortical neurons and in SH-SY5Y cells for 24 and 48 hours, respectively.
Cell injury was assessed by measuring LDH efflux into the medium after cortical neurons and SH-SY5Y cells had been treated with A1254 for 24 and 48 hours, respectively. LDH activity is correlated with the number of necrotic cells in the medium. Cytosolic levels of LDH in the extracellular medium were measured with the LDH Cytotoxicity Kitfrom Cayman, DBA (Milan, Italy). In brief, after A1254 exposure, the medium was removed and sampled for LDH content by measuring absorbance at 490 nm on a spectrophotometer (BioPhotometer; Eppendorf, Hamburg, Germany). Cell lysate, prepared with 1% Triton X-100 (Sigma-Aldrich), was used as a positive control, and its value was considered 100%.
Propidium Iodide Uptake.
After 24 hours of seeding, SH-SY5Y cells were pretreated for 2 hours with MS-275 (5 μM) and MC-1568 (5 μM), and for 6 hours with calpeptin (30 μM); they were then treated with A1254 for 48 hours. After A1254 exposure, they were washed with ice-cold phosphate-buffered saline and collected on ice. To evaluate necrosis, cells were stained with propidium iodide (PI) for 30 minutes in ice-cold phosphate-buffered saline containing PI (1.5 μM), and finally deposited in suspension on slides. A negative sample was acquired for control staining. The analysis was performed on 10 fields for each experimental condition (Guida et al., 2014). In addition, the cells were at a concentration of 300,000/ml, and analyses were performed on 75,000 cells for each experimental condition.
Western Blot Analysis and Immunoprecipitation.
Western blot analysis was performed as described elsewhere (Formisano et al., 2011; Sirabella et al., 2012; Guida et al., 2014; Formisano et al., 2015). Cells were lysed in lysis buffer. Samples (80 μg for REST, HDAC1, HDAC2, and HDAC3, and 60 μg for calpain) were subjected to SDS-PAGE. After electrophoresis, samples were transferred onto a polyvinylidene difluoride membrane (Amersham Biosciences/GE Healthcare, Pittsburgh, PA) and were immunoblotted with antibodies. Immunoprecipitation was performed as previously described (Formisano et al., 2011; Guida et al., 2014). In brief, cell lysates (1500 μg) were immunoprecipitated overnight at 4°C using antibodies with Protein A/G Plus agarose beads (Santa Cruz Biotechnology). The precipitated samples were then subjected to Western blot analysis. Specific antibodies were used against calpain (1:500, polyclonal goat antibody; Santa Cruz Biotechnology), HDAC1, HDAC2 (1:1000, monoclonal antibodies; Cell-Signaling, EuroClone, Milan, Italy), HDAC3 (1:1000, polyclonal antibody; Santa Cruz Biotechnology), REST (1:1000) (Formisano et al., 2011), and β-actin (1:1000) (Formisano et al., 2011).
ChIP was performed as previously described (Formisano et al., 2011, 2015; Guida et al., 2014). In brief, after 24 hours of seeding, cells were exposed to A1254 (30.6 μM) alone, after pretreatment with MS-275 and calpeptin, or after transfection with scrambled control and siHDAC3. SH-SY5Y cells were cross-linked with 1% formaldehyde (10 minutes), 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. Cell lysates were prepared and immunoprecipitated with 5 μl anti-HDAC3 (polyclonal antibody; Santa Cruz Biotechnology) and 5 μl anti–H3 acetyl (polyclonal rabbit antibody; Millipore, Milan, Italy). IgG rabbit antibody was used as a negative control. Data on real-time PCR were normalized for the DNA input. IgG controls were performed (not shown), and the values were subtracted from results for ChIP samples. The oligonucleotides used for amplification of the immunoprecipitated DNA of synapsin-1 promoter were the same as those reported elsewhere (Ekici et al., 2008).
Reverse-Transcription Real-Time PCR.
Total RNA isolation and quantitative real-time reverse-transcription PCR were performed as previously reported (Formisano et al., 2011). The primer pairs used for synapsin-1 and β-actin were the same as those reported elsewhere (Formisano et al., 2011). Samples were amplified simultaneously in triplicate in one assay run, and the threshold cycle value for each experimental group was determined. Normalization of data was performed using β-actin. To evaluate the differences in mRNA content between the groups, normalized values were entered into the Eq. 2−ΔΔct (Sirabella et al., 2012).
Data are expressed as the mean ± S.E.M. Statistical comparisons between the experimental groups were performed using one-way analysis of variance followed by the Newman-Keuls test. P < 0.05 was considered statistically significant.
A1254-Induced Necrotic Cell Death Is Inhibited by the Class I HDAC Inhibitor MS-275 in SH-SY5Y Cells.
Exposure of SH-SY5Y neuroblastoma cells to A1254 (30.6 μM) for 24 and 48 hours resulted in a time-specific reduction in mitochondrial activity, as revealed by MTT analysis (Fig. 1A). Because 48 hours of exposure damaged approximately 50% of cells, this time point was chosen for all experiments unless stated otherwise. To investigate the role of a specific class of HDACs involved in A1254-induced toxicity, cells were treated with the class I HDAC inhibitor MS-275 (Lanzillotta et al., 2013; Guida et al., 2014) and the class II HDAC inhibitor MC-1568 (Nebbioso et al., 2010; Spallotta et al., 2013; Guida et al., 2014) at 0.05, 0.5, and 5 µM. MTT assays showed that cell viability significantly improved in a dose-dependent manner when cells were pretreated with MS-275, as opposed to cells exposed to A1254 alone (Fig. 1, B and C). By contrast, no improvements were observed after MC-1568 pretreatment. HDAC specificity for MS-275 and MC-1568 in SH-SY5Y cells has previously been demonstrated by our group (Guida et al., 2014).
At 48 hours, A1254-induced calpain expression, an index of necrosis (Liu et al., 2004), was reverted by MS-275 treatment, but not by MC-1568 treatment (Fig. 1D). Necrosis, quantified with PI staining and analyzed with a Tali Image Cytometer (Life Technologies, Grand Island, NY), was determined by LDH efflux of damaged cells into the medium. After 48 hours of exposure to A1254 (30.6 μM), 45 and 55% of SH-SY5Y cells were positive for PI staining and LDH release, respectively, compared with control. By contrast, treatment with HDAC inhibitors alone had no effect on either PI staining or LDH efflux (Fig. 1, E and F). Noticeably, MS-275, but not MC-1568, reversed A1254-induced necrosis (Fig. 1, E and F). Further evidence for A1254-induced necrosis is that cell death was also prevented by the calpain inhibitor calpeptin (30 µM) (Fig. 1, E and F).
A1254-Induced REST and HDAC3 Complex in SH-SY5Y Cells.
MS-275 is a class I HDAC inhibitor of HDAC1, HDAC2, and HDAC3, but not HDAC8 (Beharry et al., 2014). Thus, to detect which of these isoforms was possibly involved in PCB-induced toxicity, we evaluated the expression of HDAC1, HDAC2, and HDAC3 protein levels using Western blot after treating SH-SY5Y cells with A1254 for 48 hours. A1254 (30.6 μM) significantly increased HDAC3, but not HDAC1 or HDAC2, compared with vehicle (Fig. 2, A–C). In addition, coimmunoprecipitation analysis revealed that A1254 induced REST binding to HDAC3 (Fig. 2F), but not to HDAC1 or HDAC2 (Fig. 2, D and E).
A1254 (via REST) Induced HDAC3 Binding and Deacetylation of Histone H3 within the Synapsin-1 Promoter Sequence, Determining Synapsin mRNA Reduction and Cell Death.
Because A1254-induced cell death is due to the binding of REST to the synapsin-1 promoter sequence, thus decreasing its expression (Formisano et al., 2011), we investigated whether HDAC3 might be involved in REST-induced synapsin-1 reduction in cells treated with A1254. The cells were treated with siREST and siHDAC3 for 48 hours. siREST decreased REST protein expression by 50% (Fig. 3A, upper panel). Similarly, siHDAC3 decreased HDAC3 by 60% (Fig. 3A, lower panel). As shown by ChIP analysis, HDAC3 binding to the synapsin-1 promoter increased after 48 hours of A1254 treatment; however, this binding was resolved following incubation with the class I HDAC inhibitor MS-275 and siHDAC3 (Fig. 3B). Cells exposed to A1254, MS-275, and siHDAC3 were able not only to revert H3 deacetylation within the synapsin-1 promoter (Fig. 3C) but also to block its mRNA reduction (Fig. 3D). Similarly, REST knockdown in A1254-treated cells was able to revert HDAC3 binding to synapsin-1 (Fig. 3B) and to block H3 deacetylation (Fig. 3C). These findings thus suggest that REST mediates the specific interaction between HDAC3 and the synapsin-1 promoter. Consistently, whereas siREST blocked the decrease in A1254-induced synapsin-1 mRNA (Fig. 3D), siHDAC3 counteracted A1254 neurotoxicity at 48 hours (Fig. 3E).
The HDAC Class I Inhibitor MS-275, REST, and HDAC3 Knockdown Reverted A1254-Induced Cell Death in Cortical Neurons.
Cortical neurons were treated with A1254 (30.6 μM) for 24 hours (Formisano et al., 2015). This time point was chosen because it reduced cell survival only by 50%, as opposed to longer periods which completely killed the neurons (data not shown). The PCB mixture increased both REST and HDAC3 protein expression (Fig. 4, A and B). By contrast, A1254-induced calpain expression was reverted at 24 hours by MS-275 and siHDAC3 treatments (Fig. 4, C and D). Notably, siHDAC3 significantly reduced HDAC3 expression by 70% (Fig. 4E). Then, to evaluate the roles of HDAC3 and REST in A1254-induced neurotoxicity, HDAC3 and REST were knocked down in cortical neurons (Fig. 4E, upper and lower panels, respectively).
Interestingly, both siREST and siHDAC3 transfection reverted A1254-induced neurotoxicity (Fig. 4, F and G). Moreover, cell death, induced after 24 hours of A1254 exposure (30.6 μM), was inhibited when cells were pretreated with MS-275 at 1 µM (Fig. 4, F and G), a concentration that is known to be neuroprotective in cortical neurons (Lanzillotta et al., 2013).
The present study demonstrates that exposure to A1254 induces a time-dependent reduction in cell viability in SH-SY5Y neuroblastoma cells. Our finding that PCB neurotoxicity is prevented by the class I HDAC inhibitor MS-275 suggests the involvement of class I HDACs in A1254-induced cell death. Indeed, of the three class I HDAC isoforms studied (HDAC1, HDAC2, and HDAC3), only HDAC3 was increased by A1254, whereas HDAC1 and HDAC2 remained unaffected. We thus hypothesized that HDAC3 expression was responsible for triggering a cell death mechanism. Indeed, we found that HDAC3 siRNA prevented A1254-induced cell death both in SH-SY5Y cells and in cortical neurons. Additional evidence for HDAC3-mediated cell death is that the HDAC3 inhibitor MS-275 blocked A1254-induced cell viability reduction in SH-SY5Y cells and neurons while decreasing calpain expression and LDH release. A remarkable finding of the present paper is that A1254 exposure caused an interaction between HDAC3 and REST. To our knowledge, this is the first evidence showing that (1) the transcription factor REST binds to HDAC3, and (2) this specific HDAC isoform is involved in A1254 neurotoxicity. These results fully echo the well documented role of HDAC in promoting cell death. In fact, recent evidence shows that small hairpin RNA–mediated suppression of HDAC3 expression protects against low-potassium–induced neuronal death (Bardai and D'Mello, 2011), and that expression of mutant huntingtin liberates HDAC3 from wild-type huntingtin, thus derepressing its neurotoxic activity (Bardai et al., 2013).
Another important aspect of this study is the characterization of a possible mechanism through which the REST-HDAC3 protein complex induces cell death after A1254 treatment. In particular, we found that exposure to A1254 caused a decrease in synapsin-1 gene expression that was thought to be inhibited by REST or HDAC3 knockdown and MS-275 treatment. These findings suggest the involvement of class I HDACs, specifically HDAC3, in A1254-induced synapsin-1 mRNA reduction.
Evidence for the role of HDAC3 in reducing synapsin-1 gene expression after A1254 exposure is further supported by the fact that MS-275 antagonized the suppressive effects of A1254 on synapsin-1 gene expression by reducing HDAC3 binding to and H3 deacetylation on the synapsin-1 promoter. Our results are in accordance with previous studies demonstrating that MS-275 reduces cell death in brain ischemia (Baltan et al., 2011; Lanzillotta et al., 2013; Murphy et al., 2014) and after traumatic brain injury (Cao et al., 2013). Recruitment of HDACs to a gene locus is achieved through the binding of specific transcription factors. Interestingly, after A1254 treatment, the transcription factor REST was identified as an important repressor for the expression of the synapsin-1 gene. Indeed, REST knockdown experiments in SH-SY5Y cells confirmed that after A1254 treatment, HDAC3 requires REST to bind to the synapsin-1 promoter. Thus, REST or HDAC3 knockdown and MS-275 treatment were all able to counteract the neurodetrimental effect of A1254 on cortical neurons.
Collectively, the present study suggests that A1254-induced necrotic cell death reflects accumulation of HDAC3 and formation of the REST/HDAC3 complex. Binding of the latter to the synapsin-1 promoter suppresses transcription due to deacetylation of promoter chromatin. Hence, these findings provide a novel mechanism for the neuroprotective effect of HDAC class I inhibitors on PCB-induced cell death.
The authors thank Dr. Mario Galgani from Laboratorio di Immunologia, Istituto di Endocrinologia e Oncologia Sperimentale, Consiglio Nazionale delle Ricerche (Napoli, Italy) for providing reagents for Tali Image-Based Cytometer. The authors also thank Carmine Capitale for technical support and Dr. Paola Merolla for stylistic editing.
Participated in research design: Formisano, Guida, Di Renzo.
Conducted experiments: Formisano, Guida, Laudati, Mascolo.
Contributed new reagents or analytic tools: Formisano.
Performed data analysis: Formisano, Guida, Laudati, Mascolo.
Wrote or contributed to the writing of the manuscript: Formisano, Guida, Di Renzo, Canzoniero.
- Received August 13, 2014.
- Accepted December 1, 2014.
This work was supported by grants from COFIN2008, Ricerca finalizzata , and Ricerca-Sanitaria progetto Ordinario  by Ministero della Salute.
- Aroclor 1254
- chromatin immunoprecipitation
- histone deacetylase
- lactate dehydrogenase
- 3[4,5-dimethylthiazol-2-y1]-2,5-diphenyltetrazolium bromide
- polychlorinated biphenyl
- polymerase chain reaction
- propidium iodide
- RE-1–silencing transcription factor
- small interfering RNA against HDAC3
- small interfering RNA against REST
- small interfering RNA
- Copyright © 2014 by The American Society for Pharmacology and Experimental Therapeutics