Histone Hyperacetylation Up-regulates Protein Kinase Cδ in Dopaminergic Neurons to Induce Cell Death

Background: Dysregulation of neuronal acetylation homeostasis promotes neurodegeneration. Results: Histone hyperacetylation up-regulates PKCδ in dopaminergic neurons and augments susceptibility to oxidative damage. Conclusion: Epigenetic regulation of PKCδ plays a proapoptotic role in neuronal cell death. Significance: The up-regulation of PKCδ expression by hyperacetylation provides an epigenetic molecular basis of neurodegenerative disease. The oxidative stress-sensitive protein kinase Cδ (PKCδ) has been implicated in dopaminergic neuronal cell death. However, little is known about the epigenetic mechanisms regulating PKCδ expression in neurons. Here, we report a novel mechanism by which the PKCδ gene can be regulated by histone acetylation. Treatment with histone deacetylase (HDAC) inhibitor sodium butyrate (NaBu) induced PKCδ expression in cultured neurons, brain slices, and animal models. Several other HDAC inhibitors also mimicked NaBu. The chromatin immunoprecipitation analysis revealed that hyperacetylation of histone H4 by NaBu is associated with the PKCδ promoter. Deletion analysis of the PKCδ promoter mapped the NaBu-responsive element to an 81-bp minimal promoter region. Detailed mutagenesis studies within this region revealed that four GC boxes conferred hyperacetylation-induced PKCδ promoter activation. Cotransfection experiments and Sp inhibitor studies demonstrated that Sp1, Sp3, and Sp4 regulated NaBu-induced PKCδ up-regulation. However, NaBu did not alter the DNA binding activities of Sp proteins or their expression. Interestingly, a one-hybrid analysis revealed that NaBu enhanced transcriptional activity of Sp1/Sp3. Overexpression of the p300/cAMP-response element-binding protein-binding protein (CBP) potentiated the NaBu-mediated transactivation potential of Sp1/Sp3, but expressing several HDACs attenuated this effect, suggesting that p300/CBP and HDACs act as coactivators or corepressors in histone acetylation-induced PKCδ up-regulation. Finally, using genetic and pharmacological approaches, we showed that NaBu up-regulation of PKCδ sensitizes neurons to cell death in a human dopaminergic cell model and brain slice cultures. Together, these results indicate that histone acetylation regulates PKCδ expression to augment nigrostriatal dopaminergic cell death, which could contribute to the progressive neuropathogenesis of Parkinson disease.

The oxidative stress-sensitive protein kinase C␦ (PKC␦) has been implicated in dopaminergic neuronal cell death. However, little is known about the epigenetic mechanisms regulating PKC␦ expression in neurons. Here, we report a novel mechanism by which the PKC␦ gene can be regulated by histone acetylation. Treatment with histone deacetylase (HDAC) inhibitor sodium butyrate (NaBu) induced PKC␦ expression in cultured neurons, brain slices, and animal models. Several other HDAC inhibitors also mimicked NaBu. The chromatin immunoprecipitation analysis revealed that hyperacetylation of histone H4 by NaBu is associated with the PKC␦ promoter. Deletion analysis of the PKC␦ promoter mapped the NaBu-responsive element to an 81-bp minimal promoter region. Detailed mutagenesis studies within this region revealed that four GC boxes conferred hyperacetylation-induced PKC␦ promoter activation. Cotransfection experiments and Sp inhibitor studies demonstrated that Sp1, Sp3, and Sp4 regulated NaBu-induced PKC␦ up-regulation. However, NaBu did not alter the DNA binding activities of Sp proteins or their expression. Interestingly, a one-hybrid analysis revealed that NaBu enhanced transcriptional activity of Sp1/Sp3. Overexpression of the p300/cAMP-response elementbinding protein-binding protein (CBP) potentiated the NaBumediated transactivation potential of Sp1/Sp3, but expressing several HDACs attenuated this effect, suggesting that p300/CBP and HDACs act as coactivators or corepressors in histone acetylation-induced PKC␦ up-regulation. Finally, using genetic and pharmacological approaches, we showed that NaBu up-regula-tion of PKC␦ sensitizes neurons to cell death in a human dopaminergic cell model and brain slice cultures. Together, these results indicate that histone acetylation regulates PKC␦ expression to augment nigrostriatal dopaminergic cell death, which could contribute to the progressive neuropathogenesis of Parkinson disease.
Protein kinase C (PKC) isozymes, a family of at least 12 serine/threonine kinases, are key regulators of a broad spectrum of cellular functions, including proliferation, differentiation, cell cycle progression, gene transcription and translation, altered cell shape and migration, and cell death (1). According to their structures and sensitivities to calcium and diacylglycerol (DAG), 2 these kinases have been categorized into three subfamilies, namely conventional PKCs (␣, ␤I, ␤II, and ␥), novel PKCs (␦, ⑀, , and ), and atypical PKCs ( and /). Conventional PKCs require calcium and DAG for complete activation; novel PKCs are fully activated by DAG only, and atypical PKCs are calcium-and DAG-insensitive and regulated in a different manner. Among the various PKC isoforms, PKC␦ is known to play a critical role as a mediator of apoptotic responses in various cell types (2). The role of PKC␦ signaling in the neurodegenerative processes of Parkinson disease (PD) is now better understood. We previously showed that this kinase is preferentially expressed in dopaminergic neurons in the substantia nigra (3). We and others also have established that caspase-3dependent proteolytic cleavage and the ensuing activation of PKC␦ mediate the dopaminergic cell death triggered by a variety of dopaminergic neurotoxic insults in neuronal culture models as well as in animal models (4 -12). PKC␦ inhibition by genetic manipulations (e.g. siRNA knockdown and overexpression of dominant-negative, caspase-3 cleavage-resistant, or catalytically dead form of PKC␦) or by using a peptide inhibitor directed against PKC␦ proteolytic cleavage effectively attenuated neurotoxin-induced dopaminergic apoptosis (13)(14)(15). Furthermore, primary dopaminergic neurons isolated from PKC␦-deficient mice were resistant to TNF␣ toxicity (9). Consistent with these studies, we demonstrated in an in vivo study that administering rottlerin, a PKC␦-specific inhibitor, protected against MPTP-induced loss of tyrosine hydroxylase-positive neurons in the nigrostriatum and the depletion of striatal dopamine and its metabolites, as well as MPTP-induced locomotor deficits (16). In a methamphetamine-intoxicated animal model, PKC␦ deficiency and rottlerin effectively attenuated methamphetamine-induced dopaminergic damage and behavioral deficits, further supporting that PKC␦ could represent a valid pharmacological target for the treatment of dopaminergic neuronal degeneration (17,18). Interestingly, we also showed that PKC␦ negatively modulates dopamine synthesis by inhibiting the rate-limiting enzyme, tyrosine hydroxylase (3). In addition to PD, deregulation of PKC␦ activity has been linked to various diseases, including cancer, stroke, diabetic complications, autoimmune diseases, atherosclerosis, and myocardial infarction (19 -25). Although regulation of PKC␦ activity is generally through post-translational modifications (e.g. phosphorylation and proteolysis), there are also reports of changes in the expression of PKC␦ in several pathophysiological conditions (26 -33). Thus, it is of both physiological and pathological interest to study the molecular basis of PKC␦ induction and expression.
PKC␦ is ubiquitously expressed in most tissues as well as in many cell types. The PKC␦ gene, which is located on human chromosome 3, includes 18 exons and 17 introns and spans ϳ32 kb. A gap of larger than 17 kb between the transcription start and translation start sites of the human PKC␦ gene, coupled with an unusually long 5Ј-untranslated region, underscores the complicated structure of this gene (34). The 5Ј-flanking region of the PKC␦ gene lacks a typical TATA box but has a GC-rich region in the proximal region. Recently, we characterized the regulatory mechanisms underlying basal PKC␦ expression in neurons (35). Serial deletion analysis of the ϳ2-kb mouse PKC␦ promoter revealed the presence of multiple positive and negative regulatory elements contributing to PKC␦ transcriptional activity. Notably, basal transcription of the PKC␦ gene is directed from a core promoter located between nucleotides Ϫ147 and ϩ289, relative to transcription initiation. We also revealed that four GC boxes residing in this core promoter specifically bind the Sp transcription factors (e.g. Sp1, Sp3, and Sp4) and play a pivotal role in mediating constitutive PKC␦ gene expression in neuronal cells. Additionally, it has been reported that PKC␦ can be regulated by the transcription factors NFB, c-Jun/ATF2, and p53 family proteins (34, 36 -39). To the best of our knowledge, the involvement of epigenetic mechanisms in controlling PKC␦ expression has thus far not been explored.
Post-translational modification of histone proteins, such as acetylation, methylation, and phosphorylation, has been recognized as an integral mechanism in the epigenetic regulation of gene transcription and other important cellular functions. Acetylation of histone tails, catalyzed by histone acetyltransferases (HATs), promotes a more relaxed chromatin structure, which facilitates the recruitment of transcription factors and increases gene transcription. Conversely, turnover of acetylated histones by histone deacetylase (HDACs) leads to chromatin condensation and correlates with transcriptional repression. An optimal balance between HAT and HDAC activity is required for cell survival, and disruption of this balance has been implicated in neurodegenerative diseases (40). We previously reported that neurotoxic pesticide exposure induced hyperacetylation of histones H3 and H4, which contributed to cell death in cell culture models of PD (41)(42)(43). Similarly, overexpression of a HAT, Tip60, was found to induce neuronal apoptosis in Drosophila (44). In line with these findings, experiments in a variety of cellular and rodent models of neurodegeneration have described a beneficial role for many individual HDACs (45)(46)(47)(48)(49). Paradoxically, other groups have reported that either loss of HAT activity or increased HDAC activity is associated with several neurodegenerative conditions (50 -52). Hence, the role that specific HATs or HDACs play in neurodegenerative diseases remains equivocal. A similar debate exists about the functional response of pharmacological manipulation with HDAC inhibitors, whereas the effects of the HDAC inhibitor appear to be mainly neuroprotective. A growing body of evidence has also suggested that increased acetylation levels by the HDAC inhibitor can be detrimental for neurons (53)(54)(55)(56).
This study was designed to determine whether the proapoptotic PKC␦ gene is regulated by altered histone acetylation homeostasis and, if so, to decipher the molecular mechanisms responsible for this epigenetic regulation. Our results indicate that HDAC inhibition markedly induced the PKC␦ gene expression in the striatum and substantia nigra of animals, in mouse corticostriatal slices, in primary nigral and striatal neuron cultures, and in human dopaminergic LUHMES and mouse NIE115 and MN9D cells. In vitro experiments reveal that butyrate induced hyperacetylation of histone H4 in association with the PKC␦ promoter. The minimal region of the PKC␦ promoter mediating butyrate induction was mapped to an 81-bp GC-rich region, and four functioning GC boxes within this region regulated the butyrate effect. Furthermore, we present evidence indicating that butyrate increased the transactivating capacity of Sp proteins to activate the PKC␦ promoter without changes in their DNA binding activities or protein expression levels. Most notably, we found that increased levels of PKC␦ by HDAC inhibition increased the sensitivity to oxidative stress in human dopaminergic neuronal cells and mouse slice cultures. These data represent a novel molecular basis of the proapoptotic PKC␦ gene up-regulation in neurodegenerative processes through acetylation-mediated epigenetic dysregulation during neurotoxic stress. lentiviral packaging mix, mithramycin A, NaBu, trichostatin A (TSA), dibutyryl cAMP, tetracycline, poly-L-ornithine, fibronectin, mouse ␤-actin antibody, and tolfenamic acid were purchased from Sigma. Valproic acid (VPA), scriptaid, and apicidin were obtained from Alexis Biochemicals (Plymouth Meeting, PA). The Bradford protein assay kit was purchased from Bio-Rad. Lipofectamine 2000 reagent, Alexa 680-conjugated anti-mouse secondary antibody, penicillin, streptomycin, fetal bovine serum, L-glutamine, Hanks' balanced salt solution, Neurobasal medium, B27 supplement, N-2 supplement, DMEM, minimal essential medium, and Advanced DMEM/ F-12 were obtained from Invitrogen. Antibodies against PKC isoforms (␦, ⑀, , ␣, and ), Sp1, Sp3, Sp4, c-Myc, and HA tag were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The pan-acetyl histone H4 antibody was obtained from Active Motif (Carlsbad, CA), and the Fluoro-Jade stain, rabbit polyclonal antibody for acetyl-lysine, mouse p300, and histone H3 antibodies were obtained from Millipore (Billerica, MA). IRDye800-conjugated anti-rabbit secondary antibody was obtained from Rockland Labs (Gilbertsville, PA). Recombinant human glial cell line-derived neurotrophic factor and basic FGF were purchased from R&D Systems (Minneapolis, MN). Cell Titer 96 Aqueous One Solution proliferation assay kit was obtained from Promega (Madison, WI). [ 3 H]Dopamine was obtained from PerkinElmer Life Sciences. The HDAC antibody sampler kit, which was used to evaluate the levels of HDACs, was obtained from Cell Signaling Technology (Danvers, MA).
Animal Experiments-Six-to 8-week-old C57BL/6 male mice were housed in a temperature-controlled, 12-h light/dark room, and were allowed free access to food and water. NaBu was dissolved in sterile saline and administered to C57BL/6 mice by i.p. injection at a dose of 1.2 g/kg for 6 -24 h. An equal volume of saline was given to control animals. Mice were then sacrificed, and the brain areas of interest were immediately and carefully dissected out and stored at Ϫ80°C. PKC␦ Ϫ/Ϫ C57BL/6 mice were originally obtained from Dr. Keiichi Nakayama at the Medical Institute of Bioregulation, Fukuoka, Japan (22,63). Animal care procedures strictly followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Iowa State University IACUC.
Mouse Striatal and Nigral Neurons in Primary Cultures and Treatments-Plates (6-well for striatal neurons and 12-well for nigral neurons) were coated overnight with 0.1 mg/ml poly-Dlysine. Striatal or substantia nigral tissue was dissected from gestational 14-day-old mice embryos and kept in ice-cold Ca 2ϩ -free Hanks' balanced salt solution. Cells were then dissociated in Hanks' balanced salt solution containing trypsin, 0.25% EDTA for 30 min at 37°C. After enzyme inhibition with 10% heat-inactivated fetal bovine serum in Dulbecco's modified Eagle's medium, the cells were suspended in Neurobasal medium supplemented with 2% B27 supplement, 500 M L-glutamine, 100 IU/ml penicillin, and 100 g/ml streptomycin, plated at 2 ϫ 10 6 cells in 2 ml/well, and incubated in a humidified CO 2 incubator (5% CO 2 and 37°C). Half of the culture medium was replaced every 2 days, and experiments were conducted between culture days 5 and 6. After exposure to various doses of HDAC inhibitors (NaBu, VPA, Scriptaid, TSA, or apicidin) for 24 -48 h, the primary cultures were collected for Western blot or real time RT-PCR analysis.
Cell Lines, siRNAs, Transient Transfections, and Reporter Gene Assays-All cells were grown at 37°C in a humidified 95% air, 5% CO 2 atmosphere. The mouse dopaminergic MN9D cell line was a kind gift from Dr. Syed Ali (National Center for Toxicological Research/Food and Drug Administration, Jefferson, AR), and the mouse neuroblastoma NIE115 cell line was a generous gift from Dr. Debomoy Lahiri (Indiana University School of Medicine, Indianapolis, IN). MN9D and NIE115 cells were cultured in DMEM supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 50 IU penicillin, and 50 IU streptomycin, as described previously (35). The Lund human mesencephalic (LUHMES) cell line, derived from female human embryonic ventral mesencephalic cells by conditional immortalization (tetracycline-controlled v-Myc overexpression) and subsequent clonal selection, was obtained from the American Type Culture Collection. This cell line can be differentiated into post-mitotic neurons with a clear dopaminergic phenotype that was described previously in detail (64 -66). Undifferentiated LUHMES cells were propagated in Advanced DMEM/F-12 supplemented with 1ϫ N-2 supplement, 2 mM L-glutamine, and 40 ng/ml recombinant basic FGF on plastic flasks or multiwell plates precoated with 50 g/ml poly-L-ornithine and 1 g/ml fibronectin. Differentiation of LUHMES cells was initiated by the addition of differentiation medium containing advanced DMEM/F-12, 1ϫ N-2 supplement, 2 mM L-glutamine, 1 mM dibutyryl cAMP, 1 g/ml tetracycline, and 2 ng/ml recombinant human glial cell line-derived neurotrophic factor. After 2 days, cells were trypsinized and seeded onto multiwell plates at a cell density of 1.5 ϫ 10 5 cells/cm 2 . LUHMES cells differentiate into a dopaminergic phenotype after an additional 3 days culture in differentiation medium. For 6-OHDA treatment studies, at day 5 of differentiation, differentiated LUHMES cells were pretreated in the presence or absence of NaBu (1 mM) or rottlerin (0.3 M) for 1 h and then coincubated with 6-OHDA (30 M) for another 24 h before experiments.
The dicer-substrate RNA (DsiRNA) and scrambled negative control siRNA (NC1) were purchased from Integrated DNA Technologies (Coralville, IA). The sense strands of the individual siRNA sequences were as follows: HDAC1, 5Ј-AGAGGAA-AGUCUGUUACUACUACGA-3Ј; and HDAC2, 5Ј-CGAUC-AAUAAGACCAGAUAAUAUGT-3Ј. We carried out siRNA transfections in MN9D cells with Lipofectamine 2000 reagent according to the manufacturer's protocol. In brief, cells were plated in 6-well plates at 7 ϫ 10 4 cells/well 1 day before transfection. For each well, 100 pmol of siRNA duplex mixed with 5 l of Lipofectamine 2000 were added to the cells. We also performed a second identical transfection 24 h later. Cell lysates collected 96 h after the initial transfection were analyzed using Western blotting to confirm the extent of HDAC1, HDAC2, and PKC␦ expression.
For luciferase reporter experiments, transient transfections of NIE115 and MN9D cells were performed using Lipofectamine 2000 reagent according to the manufacturers' instructions. Cells were plated at 0.3 ϫ 10 6 cells/well in 6-well plates 1 day before transfection. Each transfection was performed with 4 g of reporter constructs. Cells were harvested at 24 h post-transfection, lysed in 200 l of Reporter Lysis Buffer (Promega), and assayed for luciferase activity. For cotransfection assays, various amounts of expression plasmids as indicated in the figures were added to the reporter plasmids. The total amount of DNA was adjusted by adding empty vector. In HDAC inhibitor treatment experiments, indicated doses of HDAC inhibitors were added 24 h after DNA transfection, and cells were collected at designated time points and assayed for luciferase activity.
Luciferase activity was measured on a Synergy 2 Multi-Mode Microplate Reader (BioTek, Winooski, VT) using the luciferase assay system (Promega). The ratio of luciferase activity to total amount of proteins was used as a measure of normalized luciferase activity.
Lentivirus-based shRNA Transduction-For the production of shRNA lentiviruses, the 293FT cell line (Invitrogen) was transfected with individual clones from the PKC␦-targeting shRNA plasmid set or the scrambled control plasmid along with MISSION lentiviral packaging mix (Sigma) using Lipofectamine 2000 reagent according to the manufacturers' recommendations. The lentiviral particles in the medium were collected by centrifuging at 48 h after transfection. Lentiviral transduction of LUHMES cells was performed as described previously (67) with slight modifications. Briefly, LUHMES cells were pre-differentiated for 2 days and infected with lentiviral particles at a multiplicity of infection of 2.5 in the presence of Polybrene (8 g/ml). Two or 4 days post-infection, cells were subjected to quantitative real time RT-PCR, Western blotting, or 6-OHDA treatment studies.
Organotypic Brain Slice Cultures and Treatments-Organotypic brain slices were prepared as described previously (68,69) with slight modifications. Briefly, adolescent C57BL/6 wildtype or PKC␦ Ϫ/Ϫ mice were anesthetized and decapitated. For Fluoro-Jade staining experiments, we used 9 -12-day-old C57BL/6 wild-type and PKC␦ Ϫ/Ϫ pups. The brains were quickly removed and embedded in 2% low melting point agarose (Invitrogen, prepared freshly in GBSS) in a specimen syringe. The Compresstome TM Vf-300 (Precisionary Instruments Inc., Greenville, NC) was used to obtain 250-m-thick coronal sections of corticostriatal brain regions in ice-cold Gey's balanced salt solution (137 mM NaCl, 5  Millicell-CM Biopore PTFE membrane inserts (Millipore) in 6-well culture plates. Each culture well contained 1 ml of slice culture medium (50% minimal essential medium, 25% basal Eagle's medium, 25% horse serum, 0.65% glucose, supplemented with penicillin/streptomycin and GlutaMAX). Organotypic slice cultures were maintained in a humidified 37°C incubator with 5% CO 2 and 95% air. Half of the media was exchanged every day, and experiments were conducted after 7-14 days in culture. For MPP ϩ treatment studies, corticostriatal slices were pretreated in the presence or absence of NaBu (1 mM) or rottlerin (5 M) for 3 h and then cotreated with 300 M MPP ϩ for another 24-h period.
Propidium Iodide Uptake and Fluoro-Jade Staining Assays-The propidium iodide uptake and Fluoro-Jade staining methods were used to determine the extent of neuronal damage after MPP ϩ treatment in brain slice cultures as described previously (69,70). PI is a fluorescent molecule that is excluded from cells with intact membranes, but it labels nucleic acids in cells that have damaged cell membranes to produce red fluorescence (71,72). Fluoro-Jade is an anionic fluorochrome, which selectively and specifically stains degenerating neurons in brain slices (69,70,72). In brief, after treatment, the slice cultures were incubated with 5 g/ml PI in culture medium for 20 min at 37°C. Fluoro-Jade staining was performed according to the method of Noraberg et al. (69). PI and Fluoro-Jade fluorescent images were viewed using a Nikon TE2000 microscope (Tokyo, Japan) with ϫ2 or ϫ20 magnification and captured with a SPOT color digital camera (Diagnostic Instruments, Sterling Heights, MI). For quantitative analysis of PI and Fluoro-Jade fluorescence, we measured average pixel intensities from randomly selected regions using ImageJ software (National Institutes of Health).
Immunostaining and Microscopy-For immunohistochemistry, the corticostriatal organotypic slices were washed with PBS and fixed in 4% paraformaldehyde for 1-2 h. After washing, the membrane inserts containing slices were blocked with the blocking agent (2% goat serum and 0.1% Triton X-100 in PBS) for 1 h. Membrane inserts were then incubated with the antibodies against PKC␦ (1:1000, Santa Cruz Biotechnology) and ␤-III tubulin (1:1000, Millipore) for 2-5 days at 4°C. Fluorescently conjugated secondary antibodies (Alexa Fluor 555-conjugated anti-mouse antibody, 1:2000, and Alexa Fluor 488-conjugated anti-rabbit antibody, 1:2000) were used to visualize the proteins. Hoechst 33342 (10 g/ml) was used as a nuclear stain. The membranes were then removed from the inserts and mounted directly on microscope slides. Finally, images were viewed using a Nikon TE2000 microscope at ϫ20 magnification. The images were captured with a SPOT color digital camera and processed using ImageJ software.
For immunostaining of PKC␦ and ␤-III tubulin in differentiated LUHMES cells, cells were fixed with 4% paraformaldehyde for 30 min at room temperature. After washing, the cells were permeabilized with 0.2% Triton X-100 in PBS, washed, and blocked with blocking agent (2% bovine serum albumin, 0.5% Triton X-100, and 0.05% Tween 20 in PBS) for 1 h. Cells were then incubated with the antibody against PKC␦ (1:1000, Santa Cruz Biotechnology) and ␤-III tubulin (1:1000, Millipore) overnight at 4°C, followed by incubation with fluorescently conjugated secondary antibodies (Alexa Fluor 555-conjugated anti-mouse antibody, 1:2000, and Alexa Fluor 488-conjugated antirabbit antibody, 1:2000) for 1 h at room temperature. After this, Hoechst 33342 (10 g/ml) was added for 3 min at room temperature to stain the nucleus. Images were viewed using a Nikon TE2000 microscope, captured with a SPOT color digital camera, and processed using the ImageJ software.
Quantitative Real Time RT-PCR-Total RNA was isolated from fresh cell pellets using the Absolutely RNA Miniprep kit (Stratagene, La Jolla, CA). First strand cDNA synthesis was synthesized using an AffinityScript quantitative PCR cDNA synthesis kit (Stratagene). Real time PCR was performed in an Mx3000P quantitative PCR system (Stratagene) using the Brilliant SYBR Green quantitative PCR Master Mix kit (Stratagene), with cDNAs corresponding to 150 ng of total RNA, 12.5 l of 2ϫ master mix, 0.375 l of reference dye, and 0.2 M of each primer in a 25-l final reaction volume. All reactions were performed in triplicate. Mouse ␤-actin or human 18 S rRNA was used as an internal standard for normalization. The sequences for mouse PKC␦ primers are as follows: forward, 5Ј-TCTGGGAGTGACATCCTAGACAACAACGGG-3Ј, and reverse, 5Ј-CAGATGATCTCAGCTGCATAAAACGTAGCC-3Ј. Validated QuantiTect primer sets for human PKC␦, mouse ␤-actin, and human 18 S rRNA (Qiagen) were also used. The PCR cycling conditions contained an initial denaturation at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 30 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s. Fluorescence was detected during the annealing step of each cycle. Dissociation curves were run to verify the singularity of the PCR product. The data were analyzed using the comparative threshold cycle (C t ) method (73).
Acid Extraction of Histones-Acid extraction of histones was performed as described previously (74) with modifications. Briefly, fresh cell pellets were suspended with 5 volumes of lysis buffer (10 mM HEPES, pH 7.9, 1.5 mM MgCl 2 , 10 mM KCl, 1ϫ halt protease inhibitor mixtures) and hydrochloric acid at a final concentration of 0.2 M and subsequently lysed on ice for 30 min. After centrifugation at 11,000 ϫ g for 10 min at 4°C, the histone mixtures were collected from the supernatant.
Immunoblotting-Organotypic corticostriatal slices were lysed using the tissue extraction reagent (Invitrogen) containing protease and phosphatase inhibitor mixture (Thermo Scientific). Cell lysates and brain homogenates were prepared as described previously (38). Immunoblotting was performed as described previously (15). Briefly, the samples containing equal amounts of protein were fractionated through a 7.5-15% SDS-PAGE and transferred onto a nitrocellulose membrane (Bio-Rad). Membranes were blotted with the appropriate primary antibody and developed with either IRDye 800 anti-rabbit or Alexa Fluor 680 anti-mouse secondary antibodies. The immunoblot imaging was performed with an Odyssey Infrared Imaging system (LI-COR, Lincoln, NE).
Chromatin Immunoprecipitation (ChIP)-ChIP assays were conducted with chromatin isolated from NIE115 cells using the ChIP-IT Express enzymatic kit from Active Motif according to the manufacturer's instructions with slight modifications. Briefly, after cross-linking, the nuclei were prepared and subjected to enzymatic digestion to generate chromatin fragments between 200 and 1500 bp. The sheared chromatin was collected by centrifuge, and a 10-l aliquot was removed to serve as a positive input sample. Aliquots of 70 l of sheared chromatin were immunoprecipitated with protein-G magnetic beads and 3 g of pan-acetyl histone H4 antibody (Active Motif) or 3 g of normal mouse IgG. Equal aliquots of each chromatin sample were saved for no-antibody controls. The immunoprecipitated DNA was analyzed by PCR to amplify a region (ϩ2 to ϩ289) within the PKC␦ promoter. Primers for amplification are as follows: forward, 5Ј-ATATATCTCGAGTCCTGGGCTCCA-TTGTGTGTG-3Ј, and reverse, 5Ј-GTCTATAAGCTTACCT-CACCCAGGTGCCGG-3Ј. Conditions of linear amplification were determined empirically for these primers. PCR conditions are as follows: 94°C for 3 min; 94°C for 30 s, 59°C for 30 s, and 68°C for 30 s for 35 cycles. PCR products were resolved by electrophoresis in a 1.2% agarose gel and visualized after ethidium bromide staining.
DNA Affinity Precipitation Assay-Nuclear and cytoplasmic proteins were prepared using the NE-PER nuclear and cytoplasmic extraction kit (Thermo Scientific, Waltham, MA). 5Ј-Biotinylated oligonucleotides corresponding to the sequence between ϩ204 and ϩ238 of the PKC␦ promoter were synthesized by Integrated DNA Technologies (Coralville, IA) and annealed. Twenty pmol of oligonucleotides were incubated with 100 g of Dynabeads M-280 (Dynal Biotech, Oslo, Norway) in B&W buffer at room temperature for 10 min. Unconjugated DNA was washed off with a magnetic particle concentrator (Dynal Biotech). After blocking with 0.5% BSA in TGED buffer (20 mM HEPES, pH 7.9, 1 mM EDTA, 10% glycerol, 0.01% Triton X-100) at 4°C for 2 h, the DNA-conjugated beads were incubated with 350 g of nuclear extracts from NIE115 cells treated with or without 1 mM NaBu for 4 h at 4°C. After an extensive wash by TGED buffer, the beads were eluted with 50 l of 2ϫ SDS loading buffer. Complexing proteins were resolved on a 7.5% SDS-polyacrylamide gel and examined by immunoblotting with polyclonal anti-Sp3 and -Sp4 antibodies.
Determination of Cell Viability-Cell viability assay was performed using the Cell Titer 96 Aqueous One Solution proliferation assay kit from Promega, according to the manufacturer's instructions. This colorimetric assay is based on the cellular conversion of an MTS tetrazolium compound into a colored formazan product that is soluble in culture medium. Briefly, differentiated LUHMES cells were seeded onto 96-well plates at a cell density of 3.6 ϫ 10 4 per well. Treatment of differentiated LUHMES cells was performed at day 5 of differentiation. After treatment, Cell Titer 96 Aqueous One Solution Reagent was added to each well, and the plates were incubated in a humidified incubator (5% CO 2 and 37°C) for 2-3 h. Measurements were made at 490 nm with a 96-well plate reader (SpectraMax 190 spectrophotometer, Molecular Devices).

High Affinity [ 3 H]Dopamine
Uptake Assays-Dopamine uptake measurements were measured as described previously (75) with some modifications. Briefly, differentiated LUHMES cells grown in 6-well plates were washed twice with Krebs-Ringer buffer (16 mM NaH 2 PO 4 , 16 mM Na 2 HPO 4 , 120 mM NaCl, 4.7 mM KCl, 1.8 mM CaCl 2 , 1.2 mM MgSO 4 , 1.3 mM EDTA, and 5.6 mM glucose, pH 7.4), followed by incubation with 10 nM [ 3 H]dopamine in Krebs-Ringer buffer for 30 min at 37°C. Nonspecific uptake was determined by adding 10 M dopamine uptake blocker mazindol. Afterward, the cultures were washed three times with fresh ice-cold Krebs-Ringer buffer and then lysed with 1 N NaOH. The radioactivity was measured with a liquid scintillation counter (Tri-Crab 4000, Packard Instrument Co.) after the addition of a 5-ml scintillation mixture to each vial. The specific dopamine uptake was expressed as mean values of counts, subtracted from nonspecific counts measured in the presence of mazindol.
Statistical Analysis-Unless otherwise stated, all data were determined from three independent experiments, each done in triplicate, and expressed as means Ϯ S.E. All statistical analyses were performed using the GraphPad Prism 4.0 software (GraphPad Software, San Diego). One-way analysis of variance test followed by Tukey's multiple comparison test were used for statistical comparisons, and differences were considered significant if p Ͻ 0.05.

PKC␦ mRNA and Protein Levels Are Stimulated by Exposure to HDAC Inhibitors in Vivo and in
Vitro-Previously, we reported that activation of PKC␦ mediates dopaminergic neurodegeneration in multiple models of PD (4 -9, 12). Given the importance of histone acetylation in modulating signaling events (40), we asked whether histone acetylation could influence PKC␦ signaling in neurons. In our first set of experiments, we assessed the effect of increased histone acetylation through HDAC inhibition on the PKC␦ protein levels in a variety of primary and cultured neuronal cells. NaBu, a short-chain fatty acid, potently inhibits HDAC activity at millimolar concentrations (76). Treatment with 1 mM NaBu markedly increased the levels of total PKC␦ protein in primary mouse nigral (Fig. 1A, left panel) and striatal (Fig. 1A, right panel) neurons following 24 -48 h of drug exposure. Because butyrate has multiple pharmacological actions that may not be due to inhibition of HDAC (77), we further analyzed whether other HDAC inhibitors had similar effects on PKC␦ protein expression. For this, we exposed striatal neurons to increasing concentrations of multiple HDAC inhibitors including TSA, valproic acid, scriptaid, and apicidin for 48 h, and PKC␦ protein levels were determined by Western blot analysis. As seen with NaBu, Fig. 1B shows a dose-dependent induction of PKC␦ protein in cells treated with VPA, another short-chain fatty acid. Induction of native PKC␦ by VPA was accompanied by a dose-dependent increase in PKC␦ proteolytic cleavage. Scriptaid, which is structurally unrelated to NaBu, also enhanced total PKC␦ levels in the dose range tested (Fig. 1C), whereas induction of PKC␦ proteolytic cleavage was only observed at the lower dose (1.23 M). PKC␦ levels were also dramatically up-regulated after treatment of striatal neurons with nanomolar levels of TSA or apicidin, two other structurally unrelated HDAC inhibitors (Fig. 1, D and E). Next, analysis of mouse neuroblastoma NIE115 cells demonstrates that 48 h of NaBu (1 mM) treatment elevated PKC␦ protein levels up to ϳ2-fold compared with untreated cells (Fig.  1F). Additional analysis of various PKC isoforms in mouse dopaminergic MN9D cells shows that NaBu increased PKC␦ protein levels, whereas PKC␣, -⑀, and -were not affected (Fig.  1G). Interestingly, PKC, a typical PKC isoform, was elevated after 24 h NaBu treatment, suggesting that the effect of HDAC inhibitors on PKC␦ gene expression is somewhat isoformspecific.
Following the characterization of PKC␦ protein up-regulation during HDAC inhibition in multiple neuronal models, we then determined whether the effect of HDAC inhibitors on PKC␦ up-regulation occurs at the level of gene transcription. PKC␦ mRNA expression was measured using real time RT-PCR at various doses and time points. As shown in Fig. 2, exposure of primary nigral ( Fig. 2A, left panel) and striatal ( Fig. 2A, right panel) cultures to 1 mM NaBu for 24 or 48 h significantly increased PKC␦ mRNA expression. The magnitude of the inductions varied from 4-to 6-fold relative to untreated neurons. Furthermore, when nigral and striatal cells were administered increasing concentrations of NaBu  (0.2-5 mM) for 24 h, we found a dose-dependent increase in PKC␦ mRNA, peaking at 1 mM (Fig. 2B). In addition, similar inductions of endogenous PKC␦ mRNA by NaBu were observed in both NIE115 and MN9D neuronal cells (Fig. 2C). These increases peaked at ϳ3-fold after the 48-h sodium butyrate treatment.
To further address whether the effect of HDAC inhibition on PKC␦ expression observed above reflects the regulation of PKC␦ expression in vivo, we administered NaBu (1.2 g/kg, a single i.p. injection) to male C57BL/6 mice and quantified the brain levels of PKC␦ at various time points (6 -24 h) after injection. Interestingly, we found that PKC␦ had increased 3-fold in the mouse substantia nigra by 24 h post-injection, whereas tyrosine hydroxylase and actin remained unchanged under these conditions (Fig. 3A). Moreover, the striatal regions exhibited a similar NaBu-induced up-regulation of PKC␦ (Fig.  3B). Notably, PKC␦ showed little change in cerebrum regions (Fig. 3C), indicating a brain region-specific mechanism for HDAC inhibition-induced PKC␦ up-regulation. Collectively, these data clearly demonstrate that HDAC inhibition induces PKC␦ gene expression in nigrostriatal dopaminergic neuronal cells in both cell culture and animal models.

Butyrate Induces Hyperacetylation of PKC␦ Promoter Histones-
The next set of experiments was designed to ascertain the molecular mechanisms whereby histone acetylation stimulates PKC␦ gene expression. For mechanistic studies, we used NIE115 or MN9D neuronal cells because these neuronal cells were highly amenable to transfection of gene constructs. Because butyrate inhibits the activity of many HDAC isoforms (class I and IIa), we first confirmed the influence of HDAC inhibition on histone acetylation in NIE115 cells. As expected, the systemic acetylation of histone H3 and H4 in NIE115 cells was upregulated following exposure to 1 mM NaBu, whereas total histone H3 levels were not changed (Fig. 4A). The acetylation of histone proteins promotes an open chromatin structure and thereby leads to transcriptional activation. We then sought to determine whether change in PKC␦ expression occurs through a chromatinspecific regulation. To this end, we performed ChIP assays using chromatin isolated from NIE115 neuronal cells and an antibody specific for histone H4 acetylation. As depicted in Fig. 4B, exposure of NIE115 cells to 1 mM NaBu resulted in a dramatic enrichment of histone H4 acetylation at the PKC␦ promoter (Fig. 4B). Together, these data indicate that chromatin remodeling is at least in part responsible for the transcription activation of PKC␦ gene brought about by NaBu treatment.
HDAC Inhibition Activates PKC␦ Promoter Transcription, Delineation of the Histone Acetylation-responsive Elements on the PKC␦ Promoter-We further explored whether HDAC inhibition stimulated PKC␦ transcription by directly activating the PKC␦ promoter. To address this, the effect of HDAC inhibitors on PKC␦ promoter activity was assessed in a luciferase reporter construct-based transient transfection assay. Our previously cloned mouse PKC␦ promoter/luciferase reporter construct pGL3Ϫ1694/ϩ289 (35), which contains 1694 bp of the 5Ј-flanking sequences and 289 bp of noncoding exon 1 (access number GU182370), or pGL3-Basic empty vector was transfected into NIE115 and MN9D cells. Transfected cells were incubated with increasing concentrations of NaBu (0.2 to 1 mM) for 24 h. The incubation of NaBu enhanced transcription of the luciferase reporter pGL3Ϫ1694/ϩ289 in a dose-dependent manner up to ϳ5-fold in both MN9D and NIE115 neuronal cells (Fig. 5A). No changes in luciferase activity were noted in cells transfected with pGL3-Basic control, implying the stimulatory effect of NaBu on PKC␦ promoter is specific. Interestingly, exposure of MN9D cells to varying doses of VPA, TSA, scriptaid, or apicidin for 24 h revealed a more robust activation of PKC␦ promoter activity than that achieved with NaBu (Fig. 5,  B-E). The extent of maximum activation for those HDAC inhibitors varied from 8-to 14-fold compared with untreated cells. Taken together, these results indicate that histone acetylation-induced PKC␦ gene up-regulation involves transcriptional regulation of PKC␦ promoter.
The regulation of PKC␦ promoter activity by HDAC inhibition was further confirmed by cotransfection with the pGL3Ϫ1694/ϩ289 promoter construct and expression vectors for several class I and IIa HDAC isoforms (HDAC1, -4, -5, and -7) under either basal or butyrate-stimulated conditions. Efficient overexpression of these HDACs was verified by Western blot (data not shown). Consistent with the HDAC inhibitor data, exogenous expression of all four HDAC proteins led to a significant inhibition of basal PKC␦ promoter activity in both NIE115 and MN9D cells, with HDAC4 and HDAC5 being the most potent repressors (ϳ60 and 80% repression for HDAC4 and HDAC5, respectively) (Fig. 5F). Furthermore, butyrate-induced activation of PKC␦ promoter was reduced by expressing various amounts of HDAC1, HDAC4, or HDAC5 in MN9D cells (Fig. 5G, left panel) and NIE115 cells (Fig. 5G, right panel), although only marginal changes were observed for HDAC7. We also performed siRNA silencing experiments to examine the role that class I HDACs (HDAC1 and -2) play in regulating PKC␦ protein expression. As shown in Fig. 5, H and I, silencing of HDAC1 and HDAC2 induced a significant up-regulation of PKC␦ protein levels in MN9D cells. These results collectively indicate that PKC␦ up-regulation is mediated by multiple HDACs. In support of this view, we found that multiple class I and IIa HDACs, including HDAC1-4, express at various levels in NIE115 and MN9D cells (Fig. 5J). By comparison, the expression of two class IIa isoforms, HDAC5 and HDAC7, was too low to be detected in both cells, demonstrating a distinct HDACs expression profile in these cells.
We were next interested in understanding the mechanism underlying histone acetylation-driven PKC␦ promoter activation. First, we delineated the regions at the PKC␦ promoter that respond to butyrate-induced acetylation. A series of truncated promoter constructs in the Ϫ1694/ϩ289 region were analyzed by transient transfection for their response to NaBu in MN9D and NIE115 cells. As shown in Fig. 6, A and B, in MN9D cells, NaBu strongly increased luciferase activities from the promoter reporter construct pGL3Ϫ147/ϩ289 as well as the pGL3ϩ2/ ϩ289 plasmid up to 3.9-and 4.2-fold, respectively, which is comparable with that obtained from the full size promoter (pGL3Ϫ1694/ϩ289, 4.9-fold). However, the lack of the sequence from ϩ2 to ϩ289 led to a significant loss of butyrate responsiveness. Furthermore, similar results were obtained FIGURE 4. Sodium butyrate increases levels of total histone acetylation and histone acetylation of PKC␦ promoter-associated chromatin. A, NIE115 cells were exposed to 1 mM NaBu for 24 h. Total histones were prepared for blotting with specific anti-acetyl-lysine and anti-H3 antibodies. A representative immunoblot is shown. B, ChIP analysis of hyperacetylated histone H4 on PKC␦ promoter. NIE115 cells were treated with 1 mM NaBu for 24 h, after which chromatin was prepared and sheared by enzymatic digestion. The sheared DNA was then immunoprecipitated with antibody against pan-acetylated histone H4, normal mouse IgG, or without antibody (No Ab). After reversal of cross-linking, immunoprecipitated DNA fragments were analyzed by PCR amplification with primers specific for the PKC␦ promoter region generating a 312-bp fragment. A representative gel electrophoresis is shown. con, control. DECEMBER 12, 2014 • VOLUME 289 • NUMBER 50 JOURNAL OF BIOLOGICAL CHEMISTRY 34751 using NIE115 cells (Fig. 6C). Thus, our mapping suggests that the major histone acetylation-responsive elements are located within the ϩ2/ϩ289 region. We therefore focused our characterization of the PKC␦ promoter on this region.

HDAC Inhibition Induces PKC␦ in Neurons
Sodium Butyrate Stimulates PKC␦ Promoter Activity through Four GC Box Elements-In our previous study (35), we found that the PKC␦ promoter region between ϩ2 and ϩ289 is GCrich and contains multiple Sp-binding sites, including four consecutive GC boxes designated GC(1) to GC(4) within ϳ250 bp downstream of the transcription start site, as well as a CACCC box located at position ϩ35 bp downstream of the transcription start site (Fig. 7A). We also reported that those Sp sites act as crucial cis-elements regulating constitutive PKC␦ transcription in neuronal cells (35). To address whether these Sp sites have any functional role in the acetylation-induced activation of PKC␦, we performed site-directed mutagenesis of pGL3Ϫ147/ ϩ209 and pGL3ϩ165/ϩ289 constructs. The former possesses the proximal CACCC site, whereas in the latter, only the four GC boxes are present (Fig. 7A). Those mutated and wild-type reporter plasmids were used and assayed for luciferase activity following NaBu treatment. Exposure to NaBu did not activate luciferase activity of the wild-type pGL3Ϫ147/ϩ209, and it even reduced its activity in MN9D cells, suggesting that the CACCC site is not involved in the activation by butyrate (Fig.  7B). Indeed, mutation of the CACCC site (mCACCC) did not diminish the NaBu responsiveness (Fig. 7B). Moreover, NaBu significantly activated the luciferase activity of wild-type pGL3ϩ165/ϩ289 up to 3.4-and 4.7-fold in MN9D and NIE115 cells, respectively (Fig. 7, C and D). These findings also indicate a minimal 81-bp NaBu-responsive promoter region from ϩ209 to ϩ289. Alteration of the most distal GC(4) or GC(3) site reduced the NaBu responsiveness by 15 and 24%, respectively, compared with that of wild-type pGL3ϩ165/ϩ289 in MN9D cells (Fig. 7C). In contrast, mutation of either the proximal GC(2) box or GC(1) box caused major decrements in response to NaBu, resulting in about 35 and 33% elimination, respec-   DECEMBER 12, 2014 • VOLUME 289 • NUMBER 50

JOURNAL OF BIOLOGICAL CHEMISTRY 34753
tively, compared with that of wild type. Furthermore, triple mutated reporter constructs, mGC123, mGC124, mGC134, or mGC134, in which only site GC(4), GC(3), GC(2), or GC(1) remains active, respectively, all resulted in a complete loss of NaBu-induced promoter activity in both cell types (Fig. 7D), suggesting that cooperative interactions among the different GC sites are required for NaBu-mediated transactivation of the PKC␦ promoter. Taken together, these data suggest that the GC(1) and GC(2) sites, and less significantly the GC(3) and GC(4) sites rather than the CACCC site, are the main NaBuinduced histone acetylation-responsive elements and that these GC boxes cooperate in an additive manner in the transmission of the NaBu response.
To confirm further that Sp sites indeed mediate transcriptional activation by NaBu, we generated an Sp1 reporter construct (Sp1-luc), composed of three SV40 promoter-derived consensus Sp1-binding elements inserted into the promoter-less luciferase reporter vector (pGL3-Basic). The effects of NaBu on its transcriptional activity were subsequently examined in transient transfection studies performed in MN9D and NIE115 cells. The luciferase activities of Sp1-luc were significantly elevated following NaBu exposure (up to ϳ4.0and 5.0fold activation in MN9D and NIE115 cells, respectively), whereas mutations of all Sp1 consensus binding sites (mSp1-luc) completely abolished the NaBu-induced transcriptional activation (Fig. 7E). Consistently, expression of HDAC isoforms inhibited both basal and NaBu-induced promoter activity of Sp1-luc (data not shown).
Sp Family Proteins Are Required for Mediating Histone Acetylation Induction of PKC␦ Expression-Our recent data implicated that Sp families of transcription factors (Sp1, Sp3, and Sp4) play a crucial role in transcriptional regulation of PKC␦ through specific interaction with those multiple GC sites, with Sp3 being the most robust activator (35). These observa- FIGURE 7. Sodium butyrate activates PKC␦ promoter through the GC box elements. MN9D and NIE115 cells were transfected with either the wild-type or mutated PKC␦ promoter and Sp1 site-driven promoter constructs for 24 h. Cells were then incubated with or without NaBu (1 mM) for 24 h, and the luciferase activities were measured and normalized by total cellular protein. Luciferase activity following transfection of the wild-type construct (pGL3Ϫ147/ϩ209, pGL3ϩ165/ϩ289, or Sp1-luc) was arbitrarily set to 100, and all other data are expressed as a percentage thereof. The results are represented as means Ϯ S.E. of three independent experiments performed in triplicate. Above each blue bar is the fold change in activation following NaBu exposure in cells transfected with the individual promoter construct. A, schematic representation of the wild-type PKC␦ promoter reporter constructs pGL3Ϫ147/ϩ209 and pGL3ϩ165/ϩ289. The multiple Sp sites are depicted by either a circle or square. B, MN9D (left panel) and NIE115 (right panel) cells were transfected with 4 g of either wild-type (pGL3Ϫ147/ϩ209) or mCACCC-mutated luciferase reporter constructs. C, MN9D cells were transfected with either the wild-type (pGL3ϩ165/ϩ289) or single mutated luciferase reporter constructs. D, wild-type (pGL3ϩ165/ϩ289) or triple mutated luciferase reporter constructs, as indicated, were transfected into MN9D (left panel) and NIE115 (right panel) cells. E, Sp1 consensus site-driven luciferase reporter plasmid (Sp1-luc) or its mutant construct (mSp1-luc) was individually transfected into MN9D (left panel) and NIE115 (right panel) cells.
tions led to a hypothesis that the NaBu-induced transcriptional activation of PKC␦ might be mediated by Sp transcriptional factors. To test this possibility, we analyzed the functional impact of ectopically expressed Sp proteins on NaBu-induced transcriptional activation in transient transfections. The PKC␦ promoter reporter construct pGL3Ϫ147/ϩ289, as illustrated in Figs. 6 and 7A, was cotransfected into NIE115 cells along with 4 g of expression vectors for Sp family proteins (pN3-Sp1, pN3-Sp3, and pN3-Sp4), or alternatively a control empty vector (pN3), in the presence or absence of 1 mM NaBu for 24 h. All of these Sp expression plasmids have been shown to express stable proteins in both NIE115 and MN9D cells (35). In accordance with butyrate induction of PKC␦ promoter activity as shown in Figs. 5 and 6, exposing the empty vector-transfected cells to NaBu induced an ϳ4.5-fold activation of the pGL3Ϫ147/ϩ289 reporter, whereas in the absence of NaBu, overexpression of Sp3 alone led to ϳ2.5-fold activation (Fig. 8A). Importantly, elevated synergistic activation of promoter activity up to ϳ11.5-fold was seen when cells overexpressing Sp3 protein were treated with NaBu. The synergism was also evident in Sp1and Sp4-transfected cells after 24 h of incubation with NaBu (Fig. 8A). These findings clearly indicate that activation of the PKC␦ promoter by NaBu is mediated by the Sp family of transcription factors. In addition, parallel transfection studies of NIE115 cells with two different amounts of expression vector for wild-type or dominant-negative mutant Sp1/Sp3 were done to confirm the effects of Sp proteins on NaBu transactivation (Fig. 8B). In these experiments, expression of wild-type Sp1 or Sp3 caused a dose-dependent increase in the NaBu-induced enhancement of PKC␦ promoter activity. In contrast, expression of a dominant-negative construct pN3-DN-Sp1 or pN3-DN-Sp3, which has an intact DNA binding domain but lacks the complete transactivation domains of Sp1/3, had no effect on the NaBu-mediated induction of PKC␦ promoter activity. Interestingly, even the highest dose of these mutant constructs did not affect the basal PKC␦ promoter activity (Fig. 8B).
To further corroborate our observation that forced expression of Sp family proteins affects NaBu transactivation of the ; right panel, pN3-DN-Sp3) lacking the transactivation domains did not enhance the NaBu induction of PKC␦ promoter activity in NIE115 cells. NIE115 cells were cotransfected with pGL3Ϫ147/ϩ289 and varying concentrations (4 -8 g) of pN3-Sp1, pN3-DN-Sp1, pN3-Sp3, or pN3-DN-Sp3 for 24 h. Cells were then exposed to 1 mM NaBu for 24 h, and luciferase activities were determined and normalized. The results are represented as means Ϯ S.E. of three independent experiments performed in triplicate. Variations in total DNA were compensated with the corresponding empty vector pN3. C and D, mithramycin A (C) and tolfenamic acid (D) inhibited the response to NaBu. NIE115 cells were transfected with the PKC␦ promoter reporter construct pGL3Ϫ147/ϩ289 for 24 h. After pretreatment with different doses of mithramycin A and tolfenamic acid for 1 h, the cells were incubated with or without NaBu (1 mM) for 24 h. Cells were then harvested, and luciferase activities were determined and normalized by total cellular protein. Values are expressed as a percentage of the activity obtained from control samples without NaBu and mithramycin A or tolfenamic acid treatment and are represented as means Ϯ S.E. of three independent experiments performed in triplicate. **, p Ͻ 0.01; ***, p Ͻ 0.001; mithramycin A or tolfenamic acid-treated versus untreated samples. DECEMBER 12, 2014 • VOLUME 289 • NUMBER 50 PKC␦ promoter, we employed different types of known Spspecific inhibitors to test whether they block NaBu-induced PKC␦ promoter activity. Pretreatment with mithramycin A, an aureolic antibiotic that is known to bind to the GC-rich motif and selectively inhibit Sp transcription factor binding (35,78,79), significantly compromised NaBu-induced transactivation of the PKC␦ promoter in a concentration-dependent manner (Fig. 8C). Furthermore, tolfenamic acid, which has been shown to induce Sp protein degradation (80), also inhibited NaBu transactivation (Fig. 8D).

Sodium Butyrate Enhances the Transactivational Activity of
Sp Proteins-To further investigate the mechanisms underlying the stimulation of PKC␦ promoter activity by NaBu, we determined whether NaBu affects the protein levels of Sp effectors. Previously, we showed that Sp3 and Sp4 are endogenously expressed at appreciable levels in both MN9D and NIE115 cells, but the expression of endogenous Sp1 was not detected in these cells (35); therefore, in this study, the effect of NaBu on the expression of Sp3 and Sp4 was examined. Western blot analyses indicate that NaBu did not change the protein levels of Sp3 or FIGURE 9. NaBu increases Sp1/3 transcriptional activity. A, Sp3 and Sp4 expression was unaffected by NaBu treatment. NIE115 cells were incubated with or without 1 mM NaBu for 24 h. Whole cell lysates were prepared and immunoblotted for Sp3, Sp4, or ␤-actin (loading control). Both short Sp3 (sSp3) and long Sp3 (lSp3) isoforms are shown. B, NaBu treatment did not lead to increased Sp3 DNA binding. NIE115 cells were treated with or without 1 mM NaBu for 24 h, and nuclear extracts from harvested cells were incubated with biotinylated PKC␦ promoter probe spanning the GC(1) and GC(2) sites. The presence of Sp3 was detected by immunoblotting analysis. A representative immunoblot is shown. C, stimulation by NaBu of the Sp1/3 transactivational potential. The reporter plasmid pG5-luc, which contains five Gal4-binding sites upstream of a minimal TATA box, and the effector plasmids for Gal4 (pM), Gal4-Sp3 (pM-Sp3), Gal4-Sp3DBD (pM-Sp3DBD), Gal4-Sp1 (pM-Sp1), Gal4-Sp1DBD (pM-Sp1DBD) were cotransfected into NIE115 (left panel) and MN9D (right panel) cells and incubated with or without 1 mM NaBu for 24 h. Luciferase activities were then determined and normalized by cellular protein.
Values are expressed as the fold induction of luciferase activity following transfection of the pG5-luc alone without NaBu treatment and are represented as means Ϯ S.E. of three independent experiments performed in triplicate. Above each blue bar is the fold change in activity in the presence of NaBu over that observed in the absence of NaBu. D, effects of overexpression of HDAC isoforms on the NaBu-induced Gal4-Sp1 (left panel) and Gal4-Sp3 (right panel) transactivation. NIE115 cells were cotransfected with reporter plasmid pG5-luc and 8 g of Gal4-Sp1 or Gal4-Sp3 in combination with 4 g of HDAC1, HDAC4, HDAC5, or HDAC7 expression plasmids or empty vector control (pcDNA3.1). The cells were then treated with or without NaBu (1 mM) for 24 h, and luciferase activities were determined. Values are expressed as fold induction over the activity obtained following transfection of the Gal4 without NaBu treatment and are represented as means Ϯ S.E. of three independent experiments performed in triplicate. ***, p Ͻ 0.001; pCDNA3.1-transfected versus HDACs-transfected samples. Sp4 (Fig. 9A). We next examined the possibility that NaBu might stimulate PKC␦ transcription by elevating the recruitment of Sp proteins to the PKC␦ promoter. DNA affinity protein binding assays were performed using a biotin-labeled oligonucleotide spanning the GC(1) and GC(2) elements, between positions ϩ204 and ϩ238 on the PKC␦ promoter, and nuclear extracts from NIE115 cells. Surprisingly, the association of Sp3 (Fig. 9B) or Sp4 (data not shown) with this oligonucleotide was unaltered after incubation with NaBu. These findings indicate that stimulation by NaBu resulted from a mechanism other than alteration of Sp protein levels or DNA binding. We then asked whether NaBu could directly increase the transactivating potential of Sp proteins. To address this issue, we utilized a Gal4-based one-hybrid system, in which Sp1 or Sp3 is fused to the DNA-binding domain of the yeast transcription factor Gal4, and the effects of NaBu on the activity of these chimeric proteins were assayed in MN9D and NIE115 cells using a luciferase reporter plasmid pG5-luc containing five Gal4 DNA-binding sites. As shown in Fig. 9C, NaBu had a negligible effect on either the pG5-luc reporter alone or pG5-luc cotransfected with the empty control vector Gal4. In contrast, a vast stimulation of transactivation of Gal4-Sp1 or Gal4-Sp3 upon NaBu treatment was observed (12-and 19-fold in MN9D cells; 32-and 31-fold in NIE115 cells for Gal4-Sp3 and Gal4-Sp1, respectively). However, the transactivation by NaBu was almost abolished when the chimeric proteins Gal4-Sp1DBD or Gal4-Sp3DBD lacking the Sp transactivation domains were used, suggesting the specificity of NaBu on Sp1/3 transactivational ability. It should be noted that under the basal condition, however, Gal4-Sp1 is a stronger activator than Gal4-Sp3. In addition, overexpressing HDAC4 or HDAC5 resulted in a significant reduction in butyrate-induced transactivation of Gal4-Sp1 or Gal4-Sp3, whereas we observed only minimal effects of HDAC1 or HDAC7 overexpression (Fig. 9D). We conclude that NaBu specifically increases the transactivational capacity of Sp1/3 proteins and that multiple HDACs might be involved in regulating Sp transcriptional activity by NaBu.
Characterizing Domains of Sp1 and Sp3 for Mediating Responsiveness of PKC␦ to Sodium Butyrate-Sp transcription factors (Sp1, Sp3, and Sp4) contain several conserved regions constituting two transactivation domains (A and B boxes) close to the C terminus with regions rich in serine/threonine and glutamine residues, an extreme N-terminal transactivation domain (D box), an N-terminal DNA binding domain (zinc finger), and a domain of highly charged amino acids (C box) immediately adjacent to the zinc finger on its N-terminal side. Additionally, Sp1 and Sp3 each possess an inhibitory domain located in the extreme N terminus of Sp1 and near the C terminus of Sp3, respectively (81,82). To characterize the regions of Sp1/3 required for NaBu responsiveness, a series of truncated Gal4-Sp1 or Gal4-Sp3 expression constructs was generated and is depicted schematically in Fig. 10, A and C. Similar to the above experiments, the ability of these chimeric proteins to transactivate pG5-luc activity in both the presence and absence of NaBu was assayed in NIE115 cells. As shown in Fig. 10, B and D, the chimeras that retain the entire N-terminal part (AϩBϩC boxes) or AϩB boxes of Sp1 (Gal4-Sp1N and Gal4-Sp1AB) or Sp3 (Gal4-Sp3AB) displayed comparable capacities to activate transcription in response to NaBu in comparison with Gal4-Sp1 or Gal4-Sp3 full-length fusions. Interestingly, the Gal4-Sp3AB chimera lacking the inhibitory domain located adjacent to zinc fingers even confers a higher NaBu responsiveness to the G5-luc reporter construct than that obtained following overexpression of the Gal4-Sp3 full-length protein, suggesting that the inhibitory domain of Sp3 may have a negative regulatory action in mediating NaBu induction of PKC␦ promoter activity. Further analysis of the N-terminal region (AϩB boxes) revealed that sequences within the three subdomains A Gln , B Ser/Thr , and B Gln , corresponding to amino acids Sp1(146 -494) and Sp3 , are essential to the transactivation actions of NaBu, as removal of any one of the three subdomains showed a significant decrease in their capacity to mediate the butyrateinduced transactivation. Furthermore, each of these subdomains was unable to render the G5-luc reporter construct NaBu responsiveness. Interestingly, the A Ser/Thr subdomain of Sp1  had no influence on the ability of NaBu to enhance the transcriptional activity.
Ectopic Expression of p300/CBP Stimulates Sodium Butyratemediated Transactivation of Sp1 and Sp3-Given that the HATs p300 and CBP can function as coactivators for Sp transcription factors, we investigated whether there is a cooperative function of p300 or CBP and Sp1/3 in the NaBu-stimulated PKC␦ promoter activation. We addressed the question by performing cotransfection assays with expression vectors for p300 or CBP and measuring the transcriptional activity of Sp1/3 with the aforementioned Gal4 luciferase assay one-hybrid system. The coexpression of CBP or p300 significantly enhanced the NaBu-induced transactivation of Gal4-Sp1 and Gal4-Sp3 in a dose-dependent manner (Fig. 11, A and B). Interestingly, we found that coexpression of a p300 mutant lacking HAT activity did not affect the NaBu-stimulated transcriptional activity of Gal4-Sp1 and Gal4-Sp3. These findings indicate that p300/CBP synergistically participates in the Sp-dependent transcription by NaBu.

Functional Studies in Human Dopaminergic Neuronal Cells and Organotypic Brain Slice Culture Models Show Acetylationinduced PKC␦ Up-regulation Enhances Sensitivity to Oxidative
Stress-Our biochemical experiments described above suggest that HDAC inhibition promotes the expression of PKC␦ protein and mRNA through Sp-dependent transcription activation in neuronal cells. The ultimate step in our study was to explore the functional relevance of PKC␦ up-regulation due to hyperacetylation. Evidence to date has implicated PKC␦ as a key regulator of oxidative stress-induced neuronal cell death (10,12). We therefore reasoned that increased level of PKC␦ seen after hyperacetylation enhances the sensitivity property of dopaminergic neurons in response to oxidative stress. To test this hypothesis, we utilized the human dopaminergic cell cultures, namely LUHMES, and analyzed the role of NaBu-induced PKC␦ up-regulation on oxidative stress-induced neuronal cell death. Differentiated LUHMES cells display unique properties of human dopaminergic neurons and offer a suitable model to replace primary neuron cultures (64,65). Consistent with this, our own analysis revealed that LUHMES cells express the dopamine transporter, vesicular monoamine transporter 2 (VMAT-2), tyrosine hydroxylase, and the neuronal marker ␤-III-tubulin upon 5-6 days of differentiation (data not shown). The dopaminergic specific toxin 6-OHDA, which is widely used as an oxidative stress inducer (83,84), was chosen as a test compound. First, we confirmed that treatment with 1 mM butyrate for 24 h increased PKC␦ protein levels in differentiated LUHMES cells by Western blot (Fig. 12A) and immunocytochemical analysis (Fig. 12B). We then proceeded to analyze the 6-OHDA-induced oxidative toxicity by MTS viability and dopamine reuptake assays. To determine the effects of PKC␦, we applied the PKC␦ inhibitor rottlerin, which has been shown to protect against MPTP and methamphetamine dopaminergic neurotoxicity and alleviate LPS-induced neuroinflammation both in vitro and in vivo (16,18,85). Our preliminary analyses indicate 0.3 M rottlerin for a 24-h treatment would be optimal for differentiated LUHMES cells (data not shown). As shown in Fig. 12, C and D, 30 M 6-OHDA treatment led to significant neurotoxicity in the LUHMES dopaminergic neuronal model, as evidenced from reduced dopamine reuptake and cell viability. In contrast, pretreatment and cotreatment with 1 mM NaBu substantially potentiated 6-OHDA-induced cellular toxicity in these assays. However, exposure to 1 mM NaBu or 0.3 M rottlerin had only marginal effects in unstressed LUHMES cells when added alone. Rottlerin effectively inhibited the LUHMES cell death brought about by 6-OHDA treatment in the presence of NaBu. Concerning the specificity of rottlerin (86), we further evaluated the effects of PKC␦ depletion by lentiviral transduction of short hairpin RNA (shRNA) in LUHMES cells. We used an shRNA set with five individual clones and tested their efficacy to knock down PKC␦ mRNA. As shown in Fig. 12E, four of the five tested shRNA clones knocked down PKC␦ mRNA by Ͼ50%, with sh-PKC␦ #1 being the most effective (ϳ70% silencing). By MTS viability analysis, we found that PKC␦ depletion remarkably blocked the 6-OHDA-induced cell death in the absence or in the presence of NaBu. More importantly, we showed that knockdown of PKC␦ could also suppress the NaBu-potentiated 6-OHDA cytotoxicity. As depicted in Fig.  12F (left panel), in the presence of NaBu, a significantly increased (p Ͻ 0.05) 6-OHDA-induced cell death was again observed in scrambled shRNA-infected cells. However, NaBu had no significant effect (p Ͼ 0.05) on cell death induced by 6-OHDA when cells were transduced with sh-PKC␦ #1. The effect of sh-PKC␦ #1 on PKC␦ protein expression was also confirmed by Western blotting. As shown in Fig. 12F (right panel), infection with sh-PKC␦ #1 lentivirus for 48 and 96 h almost completely diminished PKC␦. Taken together, these results demonstrate that PKC␦ up-regulation resulting from histone hyperacetylation sensitizes dopaminergic neurons to oxidative damage.
Finally, we extended our findings from a human dopaminergic cell culture model to an ex vivo organotypic brain slice culture model. Brain slices preserve the tissue architecture of the brain regions and can replicate many aspects of the in vivo context. The organotypic slice cultures of corticostriatal tissue have been used extensively to study the mechanistic basis of neurodegeneration (72). We prepared organotypic slice cultures from adult C57BL6 mice and assayed neuronal cell damage after dopaminergic neurotoxicant MPP ϩ treatment in the presence and absence of butyrate and/or rottlerin. Cell damage was determined by PI uptake assay. As illustrated in Fig. 13, A and B, control slices showed minimal PI uptake. Likewise, neither NaBu (1 mM) nor rottlerin (5 M) on its own caused a noticeable increase in PI-labeled cells relative to control slices. However, slices exposed to 300 M MPP ϩ displayed great PI labeling neurodegeneration. Consistent with data obtained in LUHMES cells, in the presence of 1 mM NaBu, there was a significant increase in MPP ϩ -induced neurodegeneration. Most notably, rottlerin at 5 M greatly suppressed the PI-labeling degenerating neurons induced by MPP ϩ treatment in the presence and absence of NaBu, confirming a critical role for NaBu-mediated PKC␦ up-regulation in promoting oxidative stress-induced toxicity. To further establish the role of increased PKC␦ levels during oxidative stress, we examined the organotypic striatal slice cultures prepared from PKC␦ knock-out (PKC␦ Ϫ/Ϫ ) mice. Similarly, PKC␦ Ϫ/Ϫ and wild-type organotypic brain slice cultures were treated with 300 M MPP ϩ for 24 h in the presence and absence of 1 mM NaBu and analyzed cell death by PI uptake analysis. In wild-type slice cultures, NaBu treatment consistently augmented MPP ϩ -induced neuronal cell death. In contrast, NaBu-induced enhancement of MPP ϩ neurotoxicity was significantly attenuated in PKC␦ Ϫ/Ϫ slice cultures (Fig. 13, C  and D). To further substantiate the PI staining results, we used Fluoro-Jade, which is capable of selectively staining degenerating neurons in brain tissues and organotypic slices (69,70,72,87). The organotypic brain slice cultures from wild-type and PKC␦ knock-out pups were subjected to NaBu and MPP ϩ as described above, and neurotoxicity was measured by Fluoro-Jade staining. Again, PKC␦ deficiency significantly diminished the NaBu-induced increase in MPP ϩ neurotoxicity (Fig. 13, E and F). The effect of NaBu treatment (1 mM for 24 h) on PKC␦ protein expression in wild-type organotypic striatal slice cultures was also confirmed by Western blot and immunohistochemistry analyses (Fig. 13, G and H). These data are consistent with the analyses of LUHMES cells above and establish a functional correlation between PKC␦ up-regulation induced by hyperacetylation and altered sensitivity to oxidative stress in neuronal cells.

DISCUSSION
In this study, we present evidence for a novel model of mouse PKC␦ transcriptional regulation by an epigenetic control mechanism involving HDAC inhibition and histone hyperacetylation in vitro and in vivo. These findings are particularly interesting because PKC␦ is a major kinase critically involved in apoptotic signaling in various cell types. Indeed, there is considerable evidence supporting the notion that activation of PKC␦ via caspase-dependent proteolysis plays an essential role in oxidative stress-induced dopaminergic cell death in PD (11,12,14,15). Lines of evidence have also demonstrated that the PKC␦ modulators exhibit a neuroprotective effect in the MPTP mouse model (16). The prominent role PKC␦ plays in regulating multiple biological events would suggest that its expression is tightly regulated. Although a number of studies have docu- mented changes in PKC␦ levels in response to multiple stimuli, knowledge of the regulation of PKC␦ gene expression at the transcriptional level is rather limited. The PKC␦ promoter lacks a TATA or TATA-like box and contains GC-rich sequences in the proximal promoter region (34,88). We have recently char-acterized the PKC␦ promoter in neuronal cells (35,38). We showed that a proximal 400-bp genomic fragment surrounding the transcription start site functions as a basal PKC␦ promoter to sustain the basal expression of PKC␦ in neurons, and we identified multiple functional transcription factor-binding sites that contribute to basal PKC␦ expression, including two for NFB and five for Sp family transcription factors. Nevertheless, the epigenetic regulation of PKC␦ expression mediated by histone acetylation in neuronal cells has never been reported. Thus, we carried out a comprehensive study to determine whether altered histone acetylation homeostasis has a regulatory role in PKC␦ expression in neurons.
We initiated our study by evaluating the possible alterations in PKC␦ levels after exposing cultured nigrostriatal neurons to the HDAC inhibitor NaBu, and our result showed that PKC␦ protein levels are dramatically increased in NaBu-exposed cells (Fig. 1A). Importantly, this induction of PKC␦ also occurred in human dopaminergic LUHMES neurons (Fig. 12, A and B), in mouse organotypic slice cultures (Fig. 13, G and H), and in a mouse model following acute NaBu treatment (Fig. 3, A and B), suggesting that histone acetylation-mediated PKC␦ expression occurs both in vitro and in vivo. Interestingly, our in vivo findings suggest a brain region-specific regulation of PKC␦ protein expression following NaBu treatment. The precise mechanism underlying regional specific up-regulation of PKC␦ gene expression and its relationship to histone acetylation remain unclear. NaBu has been shown to induce hyperacetylation of nuclear histones in other brain regions, such as the cortex and hippocampus (89 -91), suggesting that global histone acetylation alone would not explain the region-specific PKC␦ gene up-regulation. One possibility is that NaBu induces a regionspecific histone acetylation status on the PKC␦ gene promoter. Another likely explanation would be that NaBu alters the acetylation of key regulators governing PKC␦ expression, such as NFB and Sp proteins, in a region-specific manner. We also demonstrate that the up-regulation of PKC␦ protein levels by NaBu correlates with increased PKC␦ mRNA levels. Other structurally unrelated HDAC inhibitors, including apicidin, scriptaid, and TSA, also induced PKC␦ mRNA, implying that the NaBu induction of robust PKC␦ expression appears not to be due to nonspecific action of NaBu. Interestingly, these findings, combined with previous evidence, indicate a cell typespecific regulation of PKC␦ gene expression in response to HDAC inhibition. For instance, in nasopharyngeal carcinoma and AA/C1 adenoma cells, HDAC inhibitors, such as TSA and NaBu, did not alter the PKC␦ expression levels but rather led to a distinct cellular sub-localization of the PKC␦ protein (92,93). By contrast, in human hepatoma Hep3B cells, TSA was found to specifically down-regulate PKC␦ levels (94). Of note, how-ever, this study did not address the specific mechanisms responsible for TSA-induced PKC␦ repression or its potential functional relevance.
We next investigated the molecular mechanism underlying NaBu-mediated PKC␦ gene activation. Analysis of systemic histone acetylation levels indicated that NaBu significantly increased cellular histone acetylation levels. Importantly, an increase in PKC␦ promoter histone acetylation was observed after NaBu treatment (Fig. 4B). Thus, these results suggest that the HDAC inhibitor mediated chromatin remodeling by enhancing histone acetylation, which constitutes a basic mechanism underlying the NaBu induction of PKC␦. To clarify whether the up-regulation of PKC␦ mRNA is accompanied by activation of the PKC␦ promoter, we analyzed PKC␦ promoter activity using a promoter region (Ϫ1494/ϩ289) that we cloned recently (35). Our results indicate that NaBu and other HDAC inhibitors significantly increased the luciferase activity of this reporter (Fig. 5, A-E). Furthermore, luciferase reporter assays using serial deletion PKC␦ reporter constructs revealed that the major histone acetylation response elements resided in the 289-bp noncoding exon 1 region (Fig. 6). The proximal region of the PKC␦ promoter conferring NaBu responsiveness is GCrich and contains multiple Sp-binding sites, including one proximal CACCC box and four distal consecutive GC boxes. Previously, we have shown that the CACCC box and the GC boxes act differentially in mediating promoter activation by ectopic expression of Sp transcription factors (35). Therefore, we performed experiments to determine the possible involvement of these Sp sites in the NaBu-mediated induction (Fig. 7). Unexpectedly, a smaller construct, namely pGL3Ϫ147/ϩ209, which possesses the upstream CACCC box but lacks the downstream GC boxes, was not activated by NaBu treatment. Moreover, NaBu strongly reduced promoter activity in this promoter context (Ϫ147 to ϩ209) in MN9D cells, suggesting that the CACCC box is not required for NaBu induction. Indeed, mutation of the CACCC box had no effect on NaBu-mediated activation of the PKC␦ promoter. However, using another small construct, pGL3ϩ165/ϩ289, we found that all GC boxes are required for the full response to NaBu. Moreover, cooperative actions of different GC boxes are required for mediating the NaBu response, because triple mutation of any three GC boxes completely diminished it. Analysis using a luciferase reporter (Sp1-luc) containing three Sp1 consensus sequences further FIGURE 13. Increased PKC␦ sensitizes organotypic corticostriatal slice cultures to MPP ؉ -induced neurodegeneration. A, organotypic corticostriatal slice cultures from wild-type adult C57 black mice were preincubated in the presence or absence of 1 mM NaBu and/or 5 M rottlerin for 3 h followed by a 24-h incubation with 300 M MPP ϩ . Cell death was measured using PI uptake assay. Representative PI uptake fluorescent images are shown. Images were obtained using a Nikon TE2000 fluorescence microscope. Magnification, ϫ2. Scale bar, 500 m. B, quantification of PI fluorescence intensity. PI fluorescence was measured in each group using ImageJ software. C, organotypic corticostriatal slice cultures from wild-type and PKC␦ knock-out (PKC␦ Ϫ/Ϫ ) C57 black mice were pretreated with or without 1 mM NaBu for 3 h and then cotreated with 300 M MPP ϩ for 24 h, and cell death was measured using PI uptake assay. Representative PI uptake fluorescent images are shown. Magnification, ϫ2. Scale bar, 500 m. D, quantification of PI fluorescence intensity. E, organotypic corticostriatal slice cultures from 9-to 12-day-old wild-type and PKC␦ Ϫ/Ϫ pups were pretreated with or without 1 mM NaBu for 3 h and then cotreated with 300 M MPP ϩ for 24 h, and cell death was measured using Fluoro-Jade staining assay. Representative Fluoro-Jade fluorescent images are shown. Magnification, ϫ20. Scale bar, 100 m. F, quantification of Fluoro-Jade fluorescence intensity. G, organotypic corticostriatal slice cultures from adult C57 black mice were treated with 1 mM NaBu for 24 h, and levels of PKC␦ were analyzed by immunoblot. H, exposure of organotypic corticostriatal slices to 1 mM NaBu increased PKC␦ immunoreactivity. After treatment, cultures were immunostained for PKC␦ (green) and neuron-specific marker ␤-III tubulin (red), and the nuclei were counterstained by Hoechst 33342 (blue). Images were taken using a Nikon TE2000 fluorescence microscope. Magnification, ϫ20. Scale bar, 50 m. Representative immunofluorescence images are shown. Quantification of normalized PKC␦ (PKC␦/TuJ1) immunoreactivity is shown in the right panel. Fluorescence immunoreactivity of PKC␦ and TuJ1 was measured in each group using ImageJ software. All data expressed as percent of wild-type control (con) group are mean Ϯ S.E. and representative for results obtained from three separate experiments. ns, not significant. *, p Ͻ 0.05; **, p Ͻ 0.01; ***, p Ͻ 0.001.
implicates the cluster of four GC boxes in NaBu-induced transcriptional control of the PKC␦ promoter.
The Sp family of transcription factors, Sp1, Sp3, and Sp4, are all structurally similar and bind to GC and GT/CACCC boxes found in a variety of promoter and enhancers through three characteristic zinc fingers located at the C terminus of the proteins. Sp1 and Sp3 are ubiquitously expressed, whereas the expression of Sp4 is limited to the brain (95,96). Growing studies have implicated GC-rich Sp1-binding sites in the regulation of a number of HDAC inhibitor-regulated genes, including the IN4K gene (97), WAF1/Cip gene (62,98), HMG-CoA synthase gene (99), HSP70 gene (100), EGFR gene (101), and MMP11 gene (102). Dissecting the mechanism of NaBu induction of the PKC␦ gene revealed a dependence on Sp proteins. First, histone acetylation-induced PKC␦ promoter activity was dramatically enhanced by overexpressing the Sp1, Sp3, or Sp4 protein (Fig.  8A). The most potent effect was provided by exogenous Sp3, whereas Sp1 and Sp4 displayed weaker activation, consistent with our observation that Sp3 is the strongest transactivator of basal PKC␦ promoter activity (35). Next, we cloned a dominant-negative isoform of the Sp1 or Sp3 protein, which has an intact DNA binding domain but lacks the full transactivation domains. Its ectopic expression was effective in abolishing the synergistic activation of PKC␦ promoter activity by wild-type Sp1/3 and NaBu (Fig. 8B). Finally, we used pharmacological inhibitors to block the Sp signaling pathways and assessed their effects on NaBu-stimulated PKC␦ promoter activity. Fig. 8C shows that mithramycin A, an inhibitor of Sp-mediated transcriptional activation (35,78,79), directly blocks histone hyperacetylation-induced PKC␦ promoter activity. In addition, tolfenamic acid, another Sp inhibitor known to induce degradation of Sp proteins (80), also significantly diminishes the NaBu response (Fig. 8D). We therefore conclude that NaBu activates PKC␦ transcription via Sp3, Sp1, and Sp4.
Although neither Sp3 and Sp4 levels nor direct association of Sp3 and Sp4 with the PKC␦ promoter was affected by NaBu, treatment with NaBu significantly stimulated Sp1-and Sp3medated luciferase activity of the Gal4-luc reporter construct in one-hybrid assays (Fig. 9C). Sp1 and Sp3 contain multiple domains, including a zinc finger DNA-binding domain and a bipartite transactivation domain composed of glutamine-rich and serine/threonine-rich regions. Using a serial Gal4-Sp1 or Gal4-Sp3 fusion chimeric, we were able to show that the increased transactivational potency of Sp1 and Sp3 induced by NaBu is specific to the transactivation domains of Sp1 and Sp3. The three subdomains A Gln , B Ser/Thr , and B Gln (amino acids from 146 to 494 in Sp1; amino acids from 81 to 499 for Sp3) are all required for histone acetylation-induced transcription from the PKC␦ promoter. It remains unclear how the transcriptional capacities of Sp1 and Sp3 are up-regulated by histone acetylation. Regulation of Sp1 and Sp3 activity could be achieved by protein-protein interactions. Previous reports demonstrated that both Sp1 and Sp3 functionally interact with HDAC1, HDAC2, and HDAC4 (103)(104)(105)(106). HDACs act as transcriptional repressors, repressing gene expression by forming complexes with several corepressors, including mSin3A, SMRT, and N-CoR (107). In our experimental conditions, the consistent expression pattern of the HDAC isoforms 1-4, but not 5 and 7 ( Fig. 5H), was intriguing and suggests a role for HDACs 1-4 in mediating histone acetylation-induced PKC␦ up-regulation. In line with this, we found that PKC␦ protein levels were significantly up-regulated after the individual knockdown of the class I HDACs (HDAC1 and -2) (Fig. 5, H and I), as seen following NaBu treatment. At present, however, it is unclear whether HDAC3/4 deacetylase isoforms contribute to NaBu's regulation of the PKC␦ promoter. However, we found that overexpression of HDAC4 dramatically reduced NaBu's enhancement of the transcriptional activity of Gal4-Sp1 and Gal4-Sp3 (Fig.  9D). It is also noteworthy that HDAC4 alone is enzymatically inactive, and it acts as a scaffold protein that recruits catalytically active HDACs such as HDAC1 and HDAC3 into the transcriptional corepressor complexes (106,108). Thus, further investigations are needed to fully elucidate whether HDAC3/4 participates in the PKC␦ up-regulation upon HDAC inhibition and, if so, which catalytically active HDAC isoform is required.
Additionally, Sp1 and Sp3 also bind directly to coactivators p300 and its homolog CBP (109,110). Our results indicate ectopic expression of p300/CBP-stimulated Gal4-Sp1-and Gal4-Sp3-dependent transcription in the presence of NaBu. Interestingly, the p300 stimulation is independent of HAT activity. These data suggest that the cooperative, possibly physical, interactions between Sp proteins and p300/CBP may represent a secondary mechanism responsible for the NaBu-stimulated transactivating activity of Sp1/3. The recruitment of p300/CBP into the transcription complex is also supported by our previous observation that transcription from the PKC␦ promoter is significantly activated by overexpression of p300/CBP (35). Taken together, it seems likely that NaBu-induced acetylation alters the transcriptional activities of Sp1 and Sp3 by inhibiting HDAC activity, disrupting the repressor complex containing HDACs, and allowing the recruitment of the coactivators p300/CBP to the transcription complex bound to the GC boxes on the PKC␦ promoter.
In addition to protein-protein interactions, regulating the activities of Sp proteins also includes post-translational modifications. For example, acetylation of Sp1/Sp3 stimulates their activity (111,112), whereas sumoylation of Sp1/Sp3 inactivates them (113). Moreover, phosphorylation of Sp1 also mediates the activation of Sp-dependent transcription (114). Our preliminary in vitro results suggest that NaBu does not cause gross change in the amount of acetylation of Sp1/3, which is supported by our observation that transcriptional activation by NaBu is HAT-independent. However, it remains to be examined whether it is true in the in vivo situation. It is also unclear whether HDAC inhibition modulates specific intracellular signaling pathways to affect the amount of phosphorylation or other modifications of Sp or Sp-interacting proteins.
Thus far, the specific role of HDAC activity in neuronal survival and cell death has been a matter of debate. Experimental evidence has suggested that activation of HDACs or their pharmacological inhibition can have both protective and toxic effects, depending on cell type and proliferation state (46,56,115,116). Different mechanisms have been postulated to explain how HDAC activity promotes or protects against neurodegeneration. One postulated mechanism is through an altered balance between the expression of proapoptotic and anti-apoptotic genes. Accordingly, a key goal for studies on HDAC enzymes and their pharmacological inhibitors is to identify their in vivo downstream effectors. In this regard, we provide evidence here that the proapoptotic gene PKC␦ can be up-regulated by HDAC inhibition, and more importantly, we show that increased levels of PKC␦ kinase exacerbate neuronal injury and degeneration, thereby providing a new molecular insight into the enhanced vulnerability of dopaminergic neurons during histone acetylation dysregulation. Also, we believe that our study will promote others in the field to take a closer look at the role deacetylase inhibition plays in neurons, which may facilitate optimization of the clinical applicability of deacetylase inhibitors. Notably, although HDAC inhibitors invariably induce hyperacetylated histones and facilitate gene expression, only a small number of genes (2-17%) have been found to alter their expression upon HDAC inhibition (117). Our studies herein add the PKC␦ gene to this small list.
Besides transcriptional regulation of PKC␦ production, PKC␦ activity can be regulated at multiple integrated levels, including caspase-mediated proteolysis and coordinated phosphorylation. Our published (5,35) and preliminary results indicate that both the neurotoxin MPP ϩ and the oxidative stressinducer hydrogen peroxide do not alter the transcriptional machinery to increase PKC␦ activity but may activate the kinase by activation loop phosphorylation or caspase-mediated proteolysis. It is likely that the nature of neurotoxic agents, as well as their severity and duration of exposure, may dictate PKC␦ gene up-regulation. In this regard, recent evidence has linked exposure to environmental chemicals and toxicants, such as pesticides and metals, with altered acetylation homeostasis in cells (43, 118 -120). Particularly, we reported that dieldrin and paraquat, two neurotoxic pesticides implicated in the etiopathogenesis of PD, increase cellular histone acetylation by modulating HDAC and HAT activity (41,42). Future studies should clarify whether chronic exposure to environmental neurotoxic pesticides increases PKC␦ expression in nigral dopaminergic neurons due to hyperacetylation. Nevertheless, our present results show that imbalanced histone acetylation/ deacetylation can result in abnormal expression of the apoptotic PKC␦, providing mechanistic insights into hyperacetylationmediated epigenetic changes and the subsequent degenerative fate of neurons.
In summary, we demonstrate here for the first time that modulation of the HAT/HDAC balance by inhibiting HDAC activity induces proapoptotic PKC␦ transcription in neurons. Moreover, PKC␦ induction is triggered by the acetylation of histone proteins associated with the PKC␦ promoter and the subsequent enhancement of the transcriptional capacities of Sp transcription factors. Our results further demonstrate that up-regulated PKC␦ levels produce a proapoptotic state, thereby increasing susceptibility to neurotoxic insults. We propose that induction of proapoptotic PKC␦ by HDAC inhibition in dopaminergic neurons may represent a novel epigenetic molecular basis for the neurodegenerative processes in etiology of Parkinson disease.