Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Novel histone deacetylase (HDAC) inhibitors with improved selectivity for HDAC2 and 3 protect against neural cell death

Novel histone deacetylase (HDAC) inhibitors with improved selectivity for HDAC2 and 3 protect... BioscienceHorizons Volume 5 • 2012 10.1093/biohorizons/hzs003 Research article Novel histone deacetylase (HDAC) inhibitors with improved selectivity for HDAC2 and 3 protect against neural cell death Benjamin Durham* Faculty of Biological Sciences, Institute of Membrane and Systems Biology, University of Leeds, Leeds LS2 9JT, UK. *Corresponding author: Email: bs08bsd@leeds.ac.uk Neurodegenerative disorders, such as motor neurone disease, Alzheimer’s disease and responses to brain traumas such as stroke, involve the unwanted death of neural cells. Although the exact underlying mechanisms leading to neural cell death are not well defined, one contributory event in many situations is the over-excitation of cells caused by too much of the neu- rotransmitter glutamate. Drugs that inhibit enzymes called histone deacetylases (HDACs) can protect neural cells from gluta- mate excitotoxicity. However, current inhibitors lack specificity and although they function in vitro, they have a substantial potential for adverse side effects in vivo. HDAC2 and 3 have been implicated in neurotoxicity and here we investigated the neuroprotective potential of three novel HDAC inhibitors that show selectivity for these. The ability of these HDAC inhibitors to protect against glutamate excitotoxicity was tested using cultured organotypic cerebral slices from 7-day-old (P7) Wistar rats. Glutamate excitotoxicity was induced by 200 µM of the glutamate transporter blocker, DL-threo-β-benzyloxyaspartate (DL-TBOA). This was applied alone and alongside 1 µM of the novel HDAC2 and 3 selective inhibitors AH51, AH61 and AH62. Neural cell viability in slices was quantified from assays using the fluorescent stains, 4′,6-diamidino-2-phenylindole and ethid- ium homodimer-1. The induction of glutamate excitotoxicity by DL-TBOA resulted in 41.3 ± 6.1% (n = 7, P < 0.01) loss in cell viability as judged by ethidium homodimer-1 staining. All three novel HDAC inhibitors significantly prevented neural cell death in response to DL-TBOA (P < 0.01), with cell viabilities of 107.5 ± 6.01% (n = 4), 97.1 ± 16.5% (n = 3) and 106.7 ± 6.45% (n = 4) for AH51, AH61 and AH62, respectively. This study has shown that inhibitors selective for HDAC2 and 3 can protect neural cells from death and thus have potential as therapeutic agents against neurotoxicity. Key words: histone deacetylase inhibitors, HDAC2, HDAC3, neuroprotection, neurodegeneration, glutamate excitotoxicity Submitted November 2011; accepted February 2012 Introduction Glutamate is the major excitatory neurotransmitter in the CNS; it binds to and activates ionotropic and metabotropic The mechanisms that induce cell death in the central nervous glutamate receptors, expressed by neurones, astrocytes, system (CNS) are diverse, but many neurodegenerative diseases oligodendrocytes, reactive microglia and their precursors share some common features that can be exploited by thera- (Matute et al., 2002; Lee et al., 2010). Ionotropic receptors + 2+ + peutics. New drugs should aim to prevent cell death by acting are cation (Na , Ca and K ) channels and include the on specific targets, and this also reduces the risk of significant α-amino-3-hydroxy-5-methylisoxazole-4-propionate side effects. A major toxic insult implicated in the pathophysi- (AMPA), kainic acid and N-methyl-d-aspartic acid (NMDA) ology of neurodegenerative diseases, including motor neurone receptors (Matute et al., 2002). Metabotropic receptors disease, Alzheimer’s disease and stroke, is glutamate excitotox- (mGluRs) are coupled with G-proteins, intracellular cascades icity (reviewed by Dong, Wang and Qin, 2009) and targeting and include mGluR1 and mGluR5 linked to the inositol tri- 2+ this process is one potential therapeutic option. sphosphate (IP3)/Ca signalling pathway involving calcium © The Author 2012. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.5), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Research article Bioscience Horizons • Volume 5 2012 release from the endoplasmic reticulum (Pin and Duvoisin, distributed in the rat brain and are expressed by both 1995). During glutamate overload and excitotoxicity, over- neurones and glial cells (Broide et al., 2007). There are four activation of these glutamate receptors results in prolonged classes of HDACs: class I HDACs (1, 2, 3 and 8) are found and massive depolarizations, excessive calcium influx into within the cell nucleus where they can deacetylate histones; neural cells and calcium release from internal stores. class II HDACs (4–7, 9 and 10) shuttle between the nucleus Ultimately, this causes the inappropriate activation of cal- and the cytoplasm and as well as histones, they also deacety- cium-dependent processes such as proteases, caspases, late cytoplasmic proteins such as the microtubules; class III lipases, endonucleases, pro-apoptotic factors and the produc- HDACs, also known as the sirtuins, couple deacetylation to tion of free radicals from the mitochondria (Arundine and NAD+ hydrolysis and the single member of class IV HDACs, Tymianski, 2003). Mitochondria act as calcium sinks, but HDAC11, has features in common with both class I and II when these organelles are overloaded with large amounts of HDACs (reviewed by Gregoretti, Lee and Goodson, 2004). calcium, they produce high levels of free radicals (Carriedo Class I and II HDAC inhibitors, including valproic acid, et al., 1998), which oxidize and damage DNA, proteins and sodium phenylbutyrate and suberoylanilide hydroxamic acid lipids. One consequence of DNA damage is the accumulation (SAHA), are neuroprotective against glutamate excitotoxic- of p53 in the nucleus and this promotes the expression of ity. Using rat neurones in vitro, both valproic acid and sodium pro-apoptotic proteins that cause the mitochondrial release phenylbutyrate caused the up-regulation of pro-survival and of cytochrome c, a major apoptotic factor that promotes the anti-apoptotic genes, including heat shock protein-70 formation of the apoptosome and the activation of caspases (HSP70, Marinova et al., 2009) and Bcl-2 (Leng et al., 2010). (Jiang and Wang, 2004; Gogvadze and Orrenius, 2006). All SAHA has also been shown to be neuroprotective in a white these events lead to DNA, lipid and protein damage and matter ischaemic stroke model using mouse optic nerves; breakdown, degradation of cell integrity, organelles and ulti- SAHA preserved white matter structure and function, axonal mately triggering necrotic or apoptotic neural cell death survival and oligodendrocyte survival (Baltan et al., 2011). (Arundine and Tymianski, 2003). Although they show beneficial properties, both valproic Under normal conditions, over-activation of glutamate acid and SAHA lack specificity and are associated with receptors does not occur due to the regulation and termina- significant in vivo side effects, including cognitive dysfunction, tion of glutamatergic neurotransmission. Glutamate is headaches, sedation, nausea and vomiting, thrombosis and a removed from the extracellular space by Na -dependent glu- reduction in both blood cell count and blood electrolytes tamate transporters also known as excitatory amino-acid (Armon et al., 1996; MedlinePlus NIH, 2011; Merck, 2011). transporters (EAATs). These are located in the plasma mem- An attempt to reduce these side effects but still take advantage brane of neurones and glial cells (reviewed by Matute et al., of the promising neuroprotective ability of HDAC inhibitors 2002). Glutamate excitotoxicity can be induced experimen- would involve targeting and inhibiting specific HDACs. tally in neural cells by the application of an EAAT inhibitor such as DL-threo-β-benzyloxyaspartate (DL-TBOA, Shigeri, Activation and/or over-expression of specific HDACs has Seal and Shimamoto, 2004). Such application is a commonly been associated with neurodegenerative disease and neural cell used model of excitotoxicity and allows for the study of ther- toxicity. Levels of HDAC2 in the motor cortex and the spinal apeutics on neural cell death and survival. cord are higher in patients with amyotrophic lateral sclerosis compared with controls (Janssen et al., 2010). Also, activation Inhibition of histone deacetylases can of HDAC3 resulting from phosphorylation by glycogen syn- provide neuroprotection thase kinase 3 β (GSK3β), a kinase that is widely implicated in neurodegenerative disease, promotes neural cell death (Bhat, Histone acetyltransferases (HATs) add acetyl groups and Budd Haeberlein and Avila, 2004; Bardai and D’Mello, 2011). histone deacetylases (HDACs) remove acetyl groups from Furthermore, both HDAC2 and 3 negatively regulate memory lysine residues in proteins. These enzymes play pivotal roles (Guan et al., 2009; McQuown et al., 2011) and general class I in epigenetic regulation of gene transcription by remodelling HDAC inhibition was shown to restore some loss of memory chromatin structure. The fundamental unit of chromatin is function in an animal model of Alzheimer’s disease (Kilgore the nucleosome, composed of ~147 bp of DNA wrapped et al., 2010). Unlike the other class I HDACs, HDAC1 activity around an octamer structure of histone proteins; each histone is actually neuroprotective in stroke and Huntington’s disease protein has a protruding N-terminus with exposed lysine and inactivation or inhibition of HDAC1 leads to neurodegen- residues that are subjected to acetylation by HATs and eration (Bates et al., 2006; Kim et al., 2008). Therefore, a com- deacetylation by HDACs (reviewed by Kornberg and Lorch, pound that selectively inhibits HDAC2 and 3 but not HDAC1 1999). Histone acetylation causes a weaker association should be therapeutically better for treating disorders of the between DNA and histones, promoting a more open, more brain than the currently available non-selective general class I accessible chromatin structure, whereas deacetylation causes and II HDAC inhibitors. a tighter association between the DNA and histones, leading to a more compact, less accessible chromatin conformation The studies discussed reveal that there could be therapeutic for transcriptional machinery to initiate transcription and neuroprotective potential from selectively inhibiting (reviewed by Kornberg and Lorch, 1999). HDACs are widely HDAC2 and 3. The University of Leeds developed MI192, a 2 Bioscience Horizons • Volume 5 2012 Research article HDAC2 and 3 selective inhibitor (MI192 has low nM potency Assessment of cell viability against HDAC2 and 3. Fold selectivity for HDAC1, 2 and 3 vs. To assess cell viability in the cerebral slices, 500 µ l of 0.6 µ M HDAC3 is >250, 1.9 and 1, respectively, Cancer Research ethidium homodimer-1 (Invitrogen) dissolved in sterile Technology, 2011; Gillespie et al., 2011) and from this pro- phosphate-buffered saline (PBS, Oxoid) was applied to the duced a further three HDAC2 and 3 selective inhibitors AH51, slices. These were incubated for 30 min at 37°C in a humid AH61 and AH62. In this study, we have assessed the neuropro- atmosphere at 5% CO . The cerebral slices were cut from the tective potential of these daughter compounds, using an organ- inserts, placed on glass slides and the tissue was then fixed with otypic rat cerebral slice model of glutamate excitotoxicity. 100 µ l of 4% paraformaldehyde (PFA, Sigma) in PBS for 15 min, in the dark at room temperature. Excess PFA was Methods removed and the slices were washed three times for 5 min each with PBS. Excess PBS was removed and the slices were tissue- Organotypic cerebral slice culture dried. Slides were mounted using Vectashield with preparation 4′,6- diamidino-2-phenylindole (DAPI, Vector Laboratories) and stored at 4°C in the dark until analysis by confocal micros- Organotypic cerebral slices were produced as described previ- copy. Imaging of ethidium homodimer-1 and DAPI-stained ously (Stoppini, Buchs and Muller, 1991) with modifications. cerebral slices was performed using an upright-configured Seven-day-old (P7) Wistar rats underwent cervical dislocation Zeiss Observer Z1 confocal microscope with a 40× oil immer- followed by decapitation. The brain was rapidly removed sion objective lens. Images were taken at 405 and 555 nm under sterile conditions and placed in ice-cold, 0.22 µ m filter- wavelengths to visualize DAPI (blue) and ethidium homodi- sterilized (Millipore) dissection medium composed of mer-1 (red) fluorescence, respectively. Four randomly selected Minimum Essential Medium Eagle with Earle’s salts and non-overlapping images were taken from each slice and each NaHCO (MEM, Sigma) supplemented with 1% penicillin/ image was saved at 1024 × 1024 pixel resolution. streptomycin (Invitrogen). The olfactory bulb and the cerebel- lum were dissected and discarded; the remaining brain (cere- Non-biased quantification of the images was performed brum) was washed several times with ice-cold dissection independently by a blinded-observer using a 4 × 4 grid medium before being embedded in 4% molten high-strength (3000 µ m per grid-square) in ImageJ (NIH). The number of agar (Melford) dissolved in sterile water. Coronal cerebral DAPI-stained nuclei and ethidium homodimer-1-stained slices (250 µ m thick) were cut using a Leica VT1000S vibra- nuclei were counted within the 4 × 4 grid for each image. Data tome, collected and the agar around the slices removed. The are expressed as mean ± SEM percent cell viability taken as a slices were cultured on cell culture inserts (1 µ m pore size, percentage of control viability, n indicates the number of inde- Falcon), up to a maximum of two slices per insert. The inserts pendent cerebral slices per experimental condition. Statistical were placed in 6-well culture plates (Falcon) containing 1 ml analysis was performed with SPSS (IBM) using a one-way of filter sterilized, culture medium [MEM, 25% Hanks bal- analysis of variance (ANOVA) followed by either the Dunnett anced salt solution (HBSS, Sigma), 25% normal horse serum or Bonferroni post hoc tests at the 1% significance level. (Invitrogen), 11 mM NaHCO (Fisher Scientific), 4.6 mM L-glutamine (Sigma), 21 mM D-glucose (Fisher Scientific) and 4.2 µM L-ascorbic acid (Sigma)] pre-warmed to 37°C. Slice Results cultures were incubated at 37°C in a humid atmosphere at Selective HDAC2 and 3 inhibitors have potential in provid- 5% CO . After 3 days of incubation, the culture medium was ing neuroprotection against neurodegeneration. Here, we replaced with 1 ml of a filter-sterilized serum-free culture used an organotypic rat cerebral slice model of glutamate medium [MEM, 25% HBSS, 11 mM NaHCO , 4.6 mM excitotoxicity, a common pathophysiology involved in neu- L-glutamine, 21 mM D-glucose and 4.2 µM L-ascorbic acid rodegeneration and one that is often used to model this. and 0.3% B27 supplement (Invitrogen)] pre-warmed to 37°C. Glutamate excitotoxicity and the neuroprotective ability of our novel HDAC inhibitors against this, was assessed using Induction of glutamate excitotoxicity and cultured cerebral slices obtained from 7-day-old (P7) rats by histone deacetylase inhibitor application simultaneous incubation of the slices with the glutamate After a total of 5 days of slice culture, the culture medium transporter blocker, DL-TBOA and the HDAC inhibitors. was replaced with 1 ml of fresh, serum-free culture medium After 3 days post-exposure, slices were analysed for ethid- containing the experimental conditions, pre-warmed to ium-1 homodimer and DAPI staining (Fig. 1). Ethidium 37°C. Cerebral slices were treated with either control (culture homodimer-1 is a membrane-impermeable nucleic acid fluo- medium alone), 2 µM NaOH (Sigma) alone and 2 µM NaOH rescent stain, which only stains nuclei in cells when the mem- with 0.1% DMSO (Sigma) the drug vehicles, or 200 µM brane integrity is impaired, particularly when the cell is dead DL-TBOA (Tocris Bioscience, dissolved in 1 M NaOH) with or dying, the fluorescent stain DAPI, however, is cell or without 1 µM AH51, AH61 and AH62 (University of membrane permeable and labels all nuclei and nucleic acid Leeds, dissolved in DMSO). Slices were then cultured at 37°C of live, dead or dying cells. The amount of cell staining for in a humid atmosphere at 5% CO for 3 days before analysis. both stains, for all images per condition, was quantified. The 3 Research article Bioscience Horizons • Volume 5 2012 Figure 1. Novel HDAC2 and 3 selective inhibitors protect against DL-TBOA-induced death in cultured rat cerebral slices. Confocal images from one representative cerebral slice treated with (A) control (culture medium alone), (B) 200 µM DL-TBOA, (C) 200 µM DL-TBOA and 1 µM AH51, (D) 200 µM DL-TBOA and 1 µM AH61 or (E) 200 µM DL-TBOA and 1 µM AH62. Left panels show DAPI (blue), middle panels ethidium homodimer-1 (red) and right panels show a merged image. Arrows indicate condensed and fragmenting chromatin. Scale bar 10 µm. 4 Bioscience Horizons • Volume 5 2012 Research article control (culture medium alone) slice cell viability was 106.7 ± 6.45% (n = 4) for AH51, AH61 and AH62, respec- 81.7 ± 3.9% (n = 6) and this was expressed as 100%. For all tively (Figs 1C–E and 2). Our novel HDAC inhibitors pre- other experimental conditions, viability was quantified as vented neural cell in our model of glutamate excitotoxicity, mean ± SEM percent of cell viability, expressed as a percent- and these initial results show future promise, selective age of the control viability. HDAC2 and 3 inhibitors still retain the beneficial neuropro- tective effects associated with the general non-selective For slices treated with the vehicle conditions, cell viability HDAC inhibitors and may be useful as therapeutic agents was not significantly different to the control slices, incuba- against neural cell toxicity. tion with both 2 µM NaOH and 0.1% DMSO had a viability of 102 ± 12.4% (n = 3) and 2 µM NaOH alone the viability was 108 ± 3.9% (n = 3). Slices treated with 200 µM Discussion DL-TBOA showed a significant amount of cell death and the number of viable cells was reduced by 41.3 ± 6.1% (n = 7, Here, we have shown that three HDAC2 and 3 selective compare Figs 1A with B and 2). The dead cells showed indi- inhibitors, AH51, AH61 and AH62, are neuroprotective cations of DNA breakdown, fragmentation and condensing against glutamate excitotoxicity; a pathological process of the chromatin that occurs during apoptosis (Aoyama involved in many different neurodegenerative disorders. et al., 2005); this was demonstrated by the punctate ethidium AH51, AH61 and AH62 were effective neuroprotective homodimer-1 staining. These data show that our model of agents and because of their selectivity for HDAC2 and 3, glutamate excitotoxicity induces a significant amount of neu- these compounds are predicted to have better therapeutic ral cell death. We then tested our novel HDAC inhibitors to potential, due to their lack of efficacy to inhibit HDAC1 and see if this cell death could be prevented. other HDACs, which can contribute to the production of side effects associated with currently available general non-selec- To test the neuroprotective efficacy of the novel HDAC2 tive class I and II HDAC inhibitors. and 3 inhibitors, they were co-applied with 200 µM DL-TBOA and cell viability was assessed. AH51, AH61 and This study shows the possibility of replacing non-specific AH62 (1 µM) significantly reduced cell death brought upon HDAC inhibitors such as valproic acid and SAHA with more by the application of DL-TBOA (P < 0.01) and resulted in selective compounds as potential therapies for some neural a level of cell viability not significantly different from disorders. As discussed earlier and reviewed by Kazantsev control (culture medium alone) slice cultures or from each and Thompson (2008) and Chuang et al. (2009), general other; 107.5 ± 6.01% (n = 4), 97.1 ± 16.5% (n = 3) and non-selective HDAC inhibitors have been well reported as efficacious neuroprotective agents against glutamate excito- toxicity. Our novel selective HDAC2 and 3 inhibitors fully protected against neural cell death in our disease model; therefore, our study shows the potential of moving away from general inhibitors towards more selective ones, such as those against HDAC2 and 3 and there is no compromise in neuroprotective efficacy. The exact molecular mechanisms of neuroprotection by inhibiting HDAC2 and 3 are not yet known. However, it is likely that the mechanisms responsible involve the increased transcription of pro-survival and anti- apoptotic genes, but may also involve the prevention of neu- rotoxicity associated with increased HDAC2 and 3 activities. Cell culture models have been used to test HDAC inhibi- tors as neuroprotective agents against glutamate excitotoxic- ity (Marinova et al., 2009; Leng et al., 2010), but our model provides a more appropriate environment of the CNS with a mixture of cell types, which homogeneous cell cultures do not. Neurodegeneration in the CNS involves a complex inter- play between neurones and glial cells, so using intact cerebral slices maintains such a mix of different cell types and the interactions between them. We did not determine whether one Figure 2. Comparison of cell viability in cerebral slices between DL- specific cell type was more affected than others, what propor - TBOA and the novel HDAC2 and 3 selective inhibitors. Cerebral slices tion of cell death was derived from neurones or glial cells and were treated with control (culture medium alone, n = 6), DL-TBOA what possible changes in the interactions between glial cells (200 µM) alone (n = 7) or with 1 µM AH51 (n = 4), AH61 (n = 3) or AH62 and neurones took place. Nevertheless, HDAC inhibition pre- (n = 4). Data shown are mean ± SEM percent cell viability expressed as # vented DL-TBOA-induced cell death, suggesting that HDAC a percentage of control, P < 0.01 compared with control, *P < 0.01 inhibitors protect all cell types susceptible to DL-TBOA- compared with DL-TBOA. 5 Research article Bioscience Horizons • Volume 5 2012 mediated death. Using cell-specific immunohistochemical study. Furthermore, I thank Katy Burnage, Domenic Manfredi labelling and specific brain regions alongside the viability and Laura Anne Willis for their contributions with obtaining assay, one could directly correlate neural cell death with spe- the results. Finally, I thank Mohamed Al-Griw, for obtaining cific cell type and also assess any changes in glial and neurone the rat brains and providing the organotypic slice culture and interactions. By selecting specific brain regions such as the confocal microscopy training. cortex, hippocampus or the substantia nigra, our selective HDAC2 and 3 inhibitors can also be assessed to see whether Funding they better protect specific CNS regions and those predomi- nantly affected in different neurodegenerative diseases. This study was funded by the Faculty of Biological Sciences, University of Leeds, as part of the undergraduate final year General class I and II HDAC inhibitors are not only neu- project scheme. roprotective against glutamate excitotoxicity, but have been shown to be protective against other neurotoxic insults. Ryu et al. (2005) used organotypic mouse spinal cord slices from Author biography mice with mutant superoxide dismutase-1 (SOD1) amyo- trophic lateral sclerosis. By administering the general class I B.D. graduated in July 2011 with a First Class BSc (Hons) in and II HDAC inhibitor sodium phenylbutyrate, they saw a Medical Sciences from the Faculty of Biological Sciences, concentration-dependent increase in motor neurone survival. University of Leeds. In September 2011, he was shortlisted Coinciding with this finding was increased histone acetyla- for the Best European Biology Student of the Year at the tion and prevention of apoptosis through an increase in the Science, Engineering and Technology (SET) Awards. He has a expression of Bcl-2 and a reduction in the release of cyto- keen interest in uncovering the molecular mechanisms that chrome c from the mitochondria. Also, in a standard model control gene expression in human disease with a desire to of Parkinson’s disease, 1-methyl-4-phenylpyridinium (MPP ) develop and test therapies that reverse or compensate for induced, valproic acid prevented dopaminergic neurone loss this. B.D. started a PhD with Dr Ian C. Wood in October in cell culture (Kidd and Schneider, 2010) but also partially prevented the degeneration of the substantia nigra and striatum in in vivo rat models of the disease (Kidd and References Schneider, 2011). Chen et al. (2006) showed that valproic acid increases the expression of neurotrophins in dopaminer- Aoyama, K., Burns, D. M., Suh, S. W. et  al. (2005) Acidosis causes endo- gic neurones and glial cell cultures; including brain-derived plasmic reticulum stress and caspase-12-mediated astrocyte death, neurotrophic factor and glial cell line-derived neurotrophic Journal of Cerebral Blood Flow & Metabolism, 25 (3), 358–370. factor. These neurotrophins play prominent roles in neural cell development, neural cell survival and synaptic plasticity Armon, C., Shin, C., Miller, P. et  al. (1996) Reversible Parkinsonism and and therefore are ideal candidates to be increasingly expressed cognitive impairment with chronic valproate use, Neurology, 47 (3), to promote neuroprotection and neural cell function. The 626–635. novel HDAC2 and 3 selective inhibitors AH51, AH61 and Arundine, M. and Tymianski, M. (2003) Molecular mechanisms of cal- AH62, used in our study, fully protected neural cells from cium-dependent neurodegeneration in excitotoxicity, Cell Calcium, glutamate excitotoxicity like that observed by others with the 34 (4–5), 325–337. general class I and II HDAC inhibitors, and like the more general inhibitors, our selective ones may provide more wide- Baltan, S., Murphy, S. P., Danilov, C. A. et  al. (2011) Histone deacetylase spread protection and be effective in treating various neuro- inhibitors preserve white matter structure and function during isch- degenerative disorders. emia by conserving ATP and reducing excitotoxicity, Journal of Neuroscience, 31 (11), 3990–3999. Conclusion Bardai, F. H. and D’Mello, S. R. (2011) Selective toxicity by HDAC3 in neu- rons: regulation by Akt and GSK3β, Journal of Neuroscience, 31 (5), Selective HDAC inhibitors are a promising prospect for the 1746–1751. future treatment of neural cell death induced by glutamate excitotoxicity. This study has shown that novel HDAC2 and Bates, E. A., Victor, M., Jones, A. K. et al. (2006) Differential contributions 3 selective inhibitors can solely protect against this form of of Caenorhabditis elegans histone deacetylases to huntington poly- toxicity in an organotypic cerebral slice model. In addition, glutamate toxicity, Journal of Neuroscience, 26 (10), 2830–2838. these compounds could have further promise as therapeutic Bhat, R. V., Budd Haeberlein, S. L. and Avila, J. (2004) Glycogen synthase agents in other forms of neurodegeneration. kinase 3: a drug target for CNS therapies, Journal of Neurochemistry, 89 (6), 1313–1317. Acknowledgements Broide, R. S., Redwine, J. F., Aftahi, N. et al. (2007) Distribution of HIstone I would like to thank Dr Ian C. Wood for his insightful deacetylases 1–11 in the rat brain, Journal of Molecular Neuroscience, contribution and providing his laboratory resources for this 31 (1), 47–58. 6 Bioscience Horizons • Volume 5 2012 Research article Cancer Research Technology. Novel histone deacetylase (HDAC)-2 and 3 1,2,3,6-tetrahydropyridine mouse model of Parkinson’s disease, selective inhibitors (published online June 2011) accessed 14 Neuroscience, 194, 189–194. February 2012. Kilgore, M., Miller, C. A., Fass, D. M. et  al. (2010) Inhibitors of class I his- 2+ Carriedo, S. G., Yin, H. Z., Sensi, S. L. et al. (1998) Rapid Ca entry through tone deacetylase reverse contextual memory deficits in a mouse 2+ Ca permeable AMPA/kainite channels triggers marked intracellu- model of Alzheimer’s disease, Neuropyschopharmacology, 35 (4), 2+ lar Ca rises and consequent oxygen radical production, Journal of 870–880. Neuroscience, 18 (19), 7727–7738. Kim, D., Frank, C. L., Dobbin, M. M. et al. (2008) Deregulation of HDAC1 Chen, P. S., Peng, G. S., Li, G. et al. (2006) Valproate protects dopaminer- by p25/Cdk5 in Neurotoxicity, Neuron, 60 (5), 803–817. gic neurons in midbrain neuron/glia cultures by stimulating the Kornberg, R. D. and Lorch, Y. (1999) Twenty-five years of the nucleo - release of neurotrophic factors from astrocytes, Molecular Psychiatry, some, fundamental particle of the eukaryote chromosome, Cell, 98 11 (12), 1116–1125. (3), 285–294. Chuang, D.-M, Leng, Y., Marinova, Z. et al. (2009) Multiple roles of HDAC Lee, M.-C., Ting, K. K., Adams, S. et  al. (2010) Characterization of the inhibition in neurodegenerative conditions, Trends in Neurosciences, expression of NMDA receptors in human astrocytes, Public Library of 32 (11), 591–601. Science, 5 (11), e14123. Dong, X. X., Wang, Y. and Qin, Z. H. (2009) Molecular mechanisms of exci- Leng, Y., Marinova, Z., Reis-Fernandes, M. A. et  al. (2010) Potent neuro- totoxicity and their relevance to pathogenesis of neurodegenera- protective effects of novel structural derivatives of valproic acid: tive diseases, Acta Pharmacologica Sinica, 30 (4), 379–387. potential roles of HDAC inhibition and HSP70 induction, Gillespie, J., Savic, S., Wong, C. et al. (2011) Histone deacetylases are dys- Neuroscience Letters, 476 (3), 127–132. regulated in rheumatoid arthritis and a novel HDAC3-selective Marinova, Z., Ren, M., Wendland, J. R. et al. (2009) Valproic acid induces inhibitor reduces IL-6 production by PBMC of RA patients, Arthritis function heat-shock protein 70 via Class I histone deacetylase inhi- and Rheumatism, 64 (2), 418–422. bition in cortical neurons: a potential role of Sp1 acetylation, Journal Gogvadze, V. and Orrenius, S. (2006) Mitochondrial regulation of apop- of Neurochemistry, 111 (4), 976–987. totic cell death, Chemico-biological Interactions, 163 (1–2), 4–14. Matute, C., Alberdi, E., Ibarretxe, G. et  al. (2002) Excitotoxicity in glial Gregoretti, I. V., Lee, Y. M. and Goodson, H. V. (2004) Molecular evolu- cells, European Journal of Pharmacology, 447 (2–3), 239–246. tion of the histone deacetylase family: functional implications of McQuown, S. C., Barrett, R. M., Matheos, D. P. et  al. (2011) HDAC3 is a phylogenetic analysis, Journal of Molecular Biology, 338 (1), critical negative regulator of long-term memory formation, Journal 17–31. of Neuroscience, 31 (2), 764–774. Guan, J. S., Haggarty, S. J., Giacometti, E. et al. (2009) HDAC2 negatively MedlinePlus NIH. Valproic acid (published online 15 July 2011) accessed regulates memory formation and synaptic plasticity, Nature, 459 6 October 2011. (7243), 55–60. Merck. Patient Information ZOLINZA (vorinostat) Capsules (published Janssen, C., Schmalbach, S., Boeselt, S. et  al. (2010) Differential histone online November 2011) accessed 29 January 2012. deacetylase mRNA expression patterns in amyotrophic lateral scle- rosis, Journal of Neuropathology and Experimental Neurology, 69 (6), Pin, J. P. and Duvoisin, R. (1995) The metabotropic glutamate receptors: 573–581. structure and functions, Neuropharmacology, 34 (1), 1–26. Jiang, X. and Wang, X. (2004) Cytochrome C-mediated apoptosis, Annual Ryu, H., Smith, K., Camelo, S. I. et al. (2005) Sodium phenylbutyrate pro- Review of Biochemistry, 73 (1), 87–106. longs survival and regulates expression of anti-apoptotic genes in transgenic amyotrophic lateral sclerosis mice, Journal of Kazantsev, A. G. and Thompson, L. M. (2008) Therapeutic application of Neurochemistry, 93 (5), 1087–1098. histone deacetylase inhibitors for central nervous system disorders, Nature Reviews Drug Discovery, 7 (10), 854–868. Shigeri, Y., Seal, R. P. and Shimamoto, K. (2004) Molecular pharmacology Kidd, S. K. and Schneider, J. S. (2010) Protection of dopaminergic cells of glutamate transporters, EAATs and VGLUTs, Brain Research from MPP + -mediated toxicity by histone deacetylase inhibition, Reviews, 45 (3), 250–265. Brain Research, 1354 (1), 172–178. Stoppini, L., Buchs, P. A. and Muller, D. (1991) A simple method for Kidd, S. K. and Schneider, J. S. (2011) Protective effects of valproic acid organotypic cultures of nervous tissue, Journal of Neuroscience on the nigrostriatal dopamine system in a 1-methyl-4-phenyl- Methods, 37 (2), 173–182. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Bioscience Horizons Oxford University Press

Novel histone deacetylase (HDAC) inhibitors with improved selectivity for HDAC2 and 3 protect against neural cell death

Bioscience Horizons , Volume 5 – Apr 19, 2012

Loading next page...
 
/lp/oxford-university-press/novel-histone-deacetylase-hdac-inhibitors-with-improved-selectivity-rsGyr5Q650
Publisher
Oxford University Press
Copyright
The Author 2012. Published by Oxford University Press.
Subject
Research articles
eISSN
1754-7431
DOI
10.1093/biohorizons/hzs003
Publisher site
See Article on Publisher Site

Abstract

BioscienceHorizons Volume 5 • 2012 10.1093/biohorizons/hzs003 Research article Novel histone deacetylase (HDAC) inhibitors with improved selectivity for HDAC2 and 3 protect against neural cell death Benjamin Durham* Faculty of Biological Sciences, Institute of Membrane and Systems Biology, University of Leeds, Leeds LS2 9JT, UK. *Corresponding author: Email: bs08bsd@leeds.ac.uk Neurodegenerative disorders, such as motor neurone disease, Alzheimer’s disease and responses to brain traumas such as stroke, involve the unwanted death of neural cells. Although the exact underlying mechanisms leading to neural cell death are not well defined, one contributory event in many situations is the over-excitation of cells caused by too much of the neu- rotransmitter glutamate. Drugs that inhibit enzymes called histone deacetylases (HDACs) can protect neural cells from gluta- mate excitotoxicity. However, current inhibitors lack specificity and although they function in vitro, they have a substantial potential for adverse side effects in vivo. HDAC2 and 3 have been implicated in neurotoxicity and here we investigated the neuroprotective potential of three novel HDAC inhibitors that show selectivity for these. The ability of these HDAC inhibitors to protect against glutamate excitotoxicity was tested using cultured organotypic cerebral slices from 7-day-old (P7) Wistar rats. Glutamate excitotoxicity was induced by 200 µM of the glutamate transporter blocker, DL-threo-β-benzyloxyaspartate (DL-TBOA). This was applied alone and alongside 1 µM of the novel HDAC2 and 3 selective inhibitors AH51, AH61 and AH62. Neural cell viability in slices was quantified from assays using the fluorescent stains, 4′,6-diamidino-2-phenylindole and ethid- ium homodimer-1. The induction of glutamate excitotoxicity by DL-TBOA resulted in 41.3 ± 6.1% (n = 7, P < 0.01) loss in cell viability as judged by ethidium homodimer-1 staining. All three novel HDAC inhibitors significantly prevented neural cell death in response to DL-TBOA (P < 0.01), with cell viabilities of 107.5 ± 6.01% (n = 4), 97.1 ± 16.5% (n = 3) and 106.7 ± 6.45% (n = 4) for AH51, AH61 and AH62, respectively. This study has shown that inhibitors selective for HDAC2 and 3 can protect neural cells from death and thus have potential as therapeutic agents against neurotoxicity. Key words: histone deacetylase inhibitors, HDAC2, HDAC3, neuroprotection, neurodegeneration, glutamate excitotoxicity Submitted November 2011; accepted February 2012 Introduction Glutamate is the major excitatory neurotransmitter in the CNS; it binds to and activates ionotropic and metabotropic The mechanisms that induce cell death in the central nervous glutamate receptors, expressed by neurones, astrocytes, system (CNS) are diverse, but many neurodegenerative diseases oligodendrocytes, reactive microglia and their precursors share some common features that can be exploited by thera- (Matute et al., 2002; Lee et al., 2010). Ionotropic receptors + 2+ + peutics. New drugs should aim to prevent cell death by acting are cation (Na , Ca and K ) channels and include the on specific targets, and this also reduces the risk of significant α-amino-3-hydroxy-5-methylisoxazole-4-propionate side effects. A major toxic insult implicated in the pathophysi- (AMPA), kainic acid and N-methyl-d-aspartic acid (NMDA) ology of neurodegenerative diseases, including motor neurone receptors (Matute et al., 2002). Metabotropic receptors disease, Alzheimer’s disease and stroke, is glutamate excitotox- (mGluRs) are coupled with G-proteins, intracellular cascades icity (reviewed by Dong, Wang and Qin, 2009) and targeting and include mGluR1 and mGluR5 linked to the inositol tri- 2+ this process is one potential therapeutic option. sphosphate (IP3)/Ca signalling pathway involving calcium © The Author 2012. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.5), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Research article Bioscience Horizons • Volume 5 2012 release from the endoplasmic reticulum (Pin and Duvoisin, distributed in the rat brain and are expressed by both 1995). During glutamate overload and excitotoxicity, over- neurones and glial cells (Broide et al., 2007). There are four activation of these glutamate receptors results in prolonged classes of HDACs: class I HDACs (1, 2, 3 and 8) are found and massive depolarizations, excessive calcium influx into within the cell nucleus where they can deacetylate histones; neural cells and calcium release from internal stores. class II HDACs (4–7, 9 and 10) shuttle between the nucleus Ultimately, this causes the inappropriate activation of cal- and the cytoplasm and as well as histones, they also deacety- cium-dependent processes such as proteases, caspases, late cytoplasmic proteins such as the microtubules; class III lipases, endonucleases, pro-apoptotic factors and the produc- HDACs, also known as the sirtuins, couple deacetylation to tion of free radicals from the mitochondria (Arundine and NAD+ hydrolysis and the single member of class IV HDACs, Tymianski, 2003). Mitochondria act as calcium sinks, but HDAC11, has features in common with both class I and II when these organelles are overloaded with large amounts of HDACs (reviewed by Gregoretti, Lee and Goodson, 2004). calcium, they produce high levels of free radicals (Carriedo Class I and II HDAC inhibitors, including valproic acid, et al., 1998), which oxidize and damage DNA, proteins and sodium phenylbutyrate and suberoylanilide hydroxamic acid lipids. One consequence of DNA damage is the accumulation (SAHA), are neuroprotective against glutamate excitotoxic- of p53 in the nucleus and this promotes the expression of ity. Using rat neurones in vitro, both valproic acid and sodium pro-apoptotic proteins that cause the mitochondrial release phenylbutyrate caused the up-regulation of pro-survival and of cytochrome c, a major apoptotic factor that promotes the anti-apoptotic genes, including heat shock protein-70 formation of the apoptosome and the activation of caspases (HSP70, Marinova et al., 2009) and Bcl-2 (Leng et al., 2010). (Jiang and Wang, 2004; Gogvadze and Orrenius, 2006). All SAHA has also been shown to be neuroprotective in a white these events lead to DNA, lipid and protein damage and matter ischaemic stroke model using mouse optic nerves; breakdown, degradation of cell integrity, organelles and ulti- SAHA preserved white matter structure and function, axonal mately triggering necrotic or apoptotic neural cell death survival and oligodendrocyte survival (Baltan et al., 2011). (Arundine and Tymianski, 2003). Although they show beneficial properties, both valproic Under normal conditions, over-activation of glutamate acid and SAHA lack specificity and are associated with receptors does not occur due to the regulation and termina- significant in vivo side effects, including cognitive dysfunction, tion of glutamatergic neurotransmission. Glutamate is headaches, sedation, nausea and vomiting, thrombosis and a removed from the extracellular space by Na -dependent glu- reduction in both blood cell count and blood electrolytes tamate transporters also known as excitatory amino-acid (Armon et al., 1996; MedlinePlus NIH, 2011; Merck, 2011). transporters (EAATs). These are located in the plasma mem- An attempt to reduce these side effects but still take advantage brane of neurones and glial cells (reviewed by Matute et al., of the promising neuroprotective ability of HDAC inhibitors 2002). Glutamate excitotoxicity can be induced experimen- would involve targeting and inhibiting specific HDACs. tally in neural cells by the application of an EAAT inhibitor such as DL-threo-β-benzyloxyaspartate (DL-TBOA, Shigeri, Activation and/or over-expression of specific HDACs has Seal and Shimamoto, 2004). Such application is a commonly been associated with neurodegenerative disease and neural cell used model of excitotoxicity and allows for the study of ther- toxicity. Levels of HDAC2 in the motor cortex and the spinal apeutics on neural cell death and survival. cord are higher in patients with amyotrophic lateral sclerosis compared with controls (Janssen et al., 2010). Also, activation Inhibition of histone deacetylases can of HDAC3 resulting from phosphorylation by glycogen syn- provide neuroprotection thase kinase 3 β (GSK3β), a kinase that is widely implicated in neurodegenerative disease, promotes neural cell death (Bhat, Histone acetyltransferases (HATs) add acetyl groups and Budd Haeberlein and Avila, 2004; Bardai and D’Mello, 2011). histone deacetylases (HDACs) remove acetyl groups from Furthermore, both HDAC2 and 3 negatively regulate memory lysine residues in proteins. These enzymes play pivotal roles (Guan et al., 2009; McQuown et al., 2011) and general class I in epigenetic regulation of gene transcription by remodelling HDAC inhibition was shown to restore some loss of memory chromatin structure. The fundamental unit of chromatin is function in an animal model of Alzheimer’s disease (Kilgore the nucleosome, composed of ~147 bp of DNA wrapped et al., 2010). Unlike the other class I HDACs, HDAC1 activity around an octamer structure of histone proteins; each histone is actually neuroprotective in stroke and Huntington’s disease protein has a protruding N-terminus with exposed lysine and inactivation or inhibition of HDAC1 leads to neurodegen- residues that are subjected to acetylation by HATs and eration (Bates et al., 2006; Kim et al., 2008). Therefore, a com- deacetylation by HDACs (reviewed by Kornberg and Lorch, pound that selectively inhibits HDAC2 and 3 but not HDAC1 1999). Histone acetylation causes a weaker association should be therapeutically better for treating disorders of the between DNA and histones, promoting a more open, more brain than the currently available non-selective general class I accessible chromatin structure, whereas deacetylation causes and II HDAC inhibitors. a tighter association between the DNA and histones, leading to a more compact, less accessible chromatin conformation The studies discussed reveal that there could be therapeutic for transcriptional machinery to initiate transcription and neuroprotective potential from selectively inhibiting (reviewed by Kornberg and Lorch, 1999). HDACs are widely HDAC2 and 3. The University of Leeds developed MI192, a 2 Bioscience Horizons • Volume 5 2012 Research article HDAC2 and 3 selective inhibitor (MI192 has low nM potency Assessment of cell viability against HDAC2 and 3. Fold selectivity for HDAC1, 2 and 3 vs. To assess cell viability in the cerebral slices, 500 µ l of 0.6 µ M HDAC3 is >250, 1.9 and 1, respectively, Cancer Research ethidium homodimer-1 (Invitrogen) dissolved in sterile Technology, 2011; Gillespie et al., 2011) and from this pro- phosphate-buffered saline (PBS, Oxoid) was applied to the duced a further three HDAC2 and 3 selective inhibitors AH51, slices. These were incubated for 30 min at 37°C in a humid AH61 and AH62. In this study, we have assessed the neuropro- atmosphere at 5% CO . The cerebral slices were cut from the tective potential of these daughter compounds, using an organ- inserts, placed on glass slides and the tissue was then fixed with otypic rat cerebral slice model of glutamate excitotoxicity. 100 µ l of 4% paraformaldehyde (PFA, Sigma) in PBS for 15 min, in the dark at room temperature. Excess PFA was Methods removed and the slices were washed three times for 5 min each with PBS. Excess PBS was removed and the slices were tissue- Organotypic cerebral slice culture dried. Slides were mounted using Vectashield with preparation 4′,6- diamidino-2-phenylindole (DAPI, Vector Laboratories) and stored at 4°C in the dark until analysis by confocal micros- Organotypic cerebral slices were produced as described previ- copy. Imaging of ethidium homodimer-1 and DAPI-stained ously (Stoppini, Buchs and Muller, 1991) with modifications. cerebral slices was performed using an upright-configured Seven-day-old (P7) Wistar rats underwent cervical dislocation Zeiss Observer Z1 confocal microscope with a 40× oil immer- followed by decapitation. The brain was rapidly removed sion objective lens. Images were taken at 405 and 555 nm under sterile conditions and placed in ice-cold, 0.22 µ m filter- wavelengths to visualize DAPI (blue) and ethidium homodi- sterilized (Millipore) dissection medium composed of mer-1 (red) fluorescence, respectively. Four randomly selected Minimum Essential Medium Eagle with Earle’s salts and non-overlapping images were taken from each slice and each NaHCO (MEM, Sigma) supplemented with 1% penicillin/ image was saved at 1024 × 1024 pixel resolution. streptomycin (Invitrogen). The olfactory bulb and the cerebel- lum were dissected and discarded; the remaining brain (cere- Non-biased quantification of the images was performed brum) was washed several times with ice-cold dissection independently by a blinded-observer using a 4 × 4 grid medium before being embedded in 4% molten high-strength (3000 µ m per grid-square) in ImageJ (NIH). The number of agar (Melford) dissolved in sterile water. Coronal cerebral DAPI-stained nuclei and ethidium homodimer-1-stained slices (250 µ m thick) were cut using a Leica VT1000S vibra- nuclei were counted within the 4 × 4 grid for each image. Data tome, collected and the agar around the slices removed. The are expressed as mean ± SEM percent cell viability taken as a slices were cultured on cell culture inserts (1 µ m pore size, percentage of control viability, n indicates the number of inde- Falcon), up to a maximum of two slices per insert. The inserts pendent cerebral slices per experimental condition. Statistical were placed in 6-well culture plates (Falcon) containing 1 ml analysis was performed with SPSS (IBM) using a one-way of filter sterilized, culture medium [MEM, 25% Hanks bal- analysis of variance (ANOVA) followed by either the Dunnett anced salt solution (HBSS, Sigma), 25% normal horse serum or Bonferroni post hoc tests at the 1% significance level. (Invitrogen), 11 mM NaHCO (Fisher Scientific), 4.6 mM L-glutamine (Sigma), 21 mM D-glucose (Fisher Scientific) and 4.2 µM L-ascorbic acid (Sigma)] pre-warmed to 37°C. Slice Results cultures were incubated at 37°C in a humid atmosphere at Selective HDAC2 and 3 inhibitors have potential in provid- 5% CO . After 3 days of incubation, the culture medium was ing neuroprotection against neurodegeneration. Here, we replaced with 1 ml of a filter-sterilized serum-free culture used an organotypic rat cerebral slice model of glutamate medium [MEM, 25% HBSS, 11 mM NaHCO , 4.6 mM excitotoxicity, a common pathophysiology involved in neu- L-glutamine, 21 mM D-glucose and 4.2 µM L-ascorbic acid rodegeneration and one that is often used to model this. and 0.3% B27 supplement (Invitrogen)] pre-warmed to 37°C. Glutamate excitotoxicity and the neuroprotective ability of our novel HDAC inhibitors against this, was assessed using Induction of glutamate excitotoxicity and cultured cerebral slices obtained from 7-day-old (P7) rats by histone deacetylase inhibitor application simultaneous incubation of the slices with the glutamate After a total of 5 days of slice culture, the culture medium transporter blocker, DL-TBOA and the HDAC inhibitors. was replaced with 1 ml of fresh, serum-free culture medium After 3 days post-exposure, slices were analysed for ethid- containing the experimental conditions, pre-warmed to ium-1 homodimer and DAPI staining (Fig. 1). Ethidium 37°C. Cerebral slices were treated with either control (culture homodimer-1 is a membrane-impermeable nucleic acid fluo- medium alone), 2 µM NaOH (Sigma) alone and 2 µM NaOH rescent stain, which only stains nuclei in cells when the mem- with 0.1% DMSO (Sigma) the drug vehicles, or 200 µM brane integrity is impaired, particularly when the cell is dead DL-TBOA (Tocris Bioscience, dissolved in 1 M NaOH) with or dying, the fluorescent stain DAPI, however, is cell or without 1 µM AH51, AH61 and AH62 (University of membrane permeable and labels all nuclei and nucleic acid Leeds, dissolved in DMSO). Slices were then cultured at 37°C of live, dead or dying cells. The amount of cell staining for in a humid atmosphere at 5% CO for 3 days before analysis. both stains, for all images per condition, was quantified. The 3 Research article Bioscience Horizons • Volume 5 2012 Figure 1. Novel HDAC2 and 3 selective inhibitors protect against DL-TBOA-induced death in cultured rat cerebral slices. Confocal images from one representative cerebral slice treated with (A) control (culture medium alone), (B) 200 µM DL-TBOA, (C) 200 µM DL-TBOA and 1 µM AH51, (D) 200 µM DL-TBOA and 1 µM AH61 or (E) 200 µM DL-TBOA and 1 µM AH62. Left panels show DAPI (blue), middle panels ethidium homodimer-1 (red) and right panels show a merged image. Arrows indicate condensed and fragmenting chromatin. Scale bar 10 µm. 4 Bioscience Horizons • Volume 5 2012 Research article control (culture medium alone) slice cell viability was 106.7 ± 6.45% (n = 4) for AH51, AH61 and AH62, respec- 81.7 ± 3.9% (n = 6) and this was expressed as 100%. For all tively (Figs 1C–E and 2). Our novel HDAC inhibitors pre- other experimental conditions, viability was quantified as vented neural cell in our model of glutamate excitotoxicity, mean ± SEM percent of cell viability, expressed as a percent- and these initial results show future promise, selective age of the control viability. HDAC2 and 3 inhibitors still retain the beneficial neuropro- tective effects associated with the general non-selective For slices treated with the vehicle conditions, cell viability HDAC inhibitors and may be useful as therapeutic agents was not significantly different to the control slices, incuba- against neural cell toxicity. tion with both 2 µM NaOH and 0.1% DMSO had a viability of 102 ± 12.4% (n = 3) and 2 µM NaOH alone the viability was 108 ± 3.9% (n = 3). Slices treated with 200 µM Discussion DL-TBOA showed a significant amount of cell death and the number of viable cells was reduced by 41.3 ± 6.1% (n = 7, Here, we have shown that three HDAC2 and 3 selective compare Figs 1A with B and 2). The dead cells showed indi- inhibitors, AH51, AH61 and AH62, are neuroprotective cations of DNA breakdown, fragmentation and condensing against glutamate excitotoxicity; a pathological process of the chromatin that occurs during apoptosis (Aoyama involved in many different neurodegenerative disorders. et al., 2005); this was demonstrated by the punctate ethidium AH51, AH61 and AH62 were effective neuroprotective homodimer-1 staining. These data show that our model of agents and because of their selectivity for HDAC2 and 3, glutamate excitotoxicity induces a significant amount of neu- these compounds are predicted to have better therapeutic ral cell death. We then tested our novel HDAC inhibitors to potential, due to their lack of efficacy to inhibit HDAC1 and see if this cell death could be prevented. other HDACs, which can contribute to the production of side effects associated with currently available general non-selec- To test the neuroprotective efficacy of the novel HDAC2 tive class I and II HDAC inhibitors. and 3 inhibitors, they were co-applied with 200 µM DL-TBOA and cell viability was assessed. AH51, AH61 and This study shows the possibility of replacing non-specific AH62 (1 µM) significantly reduced cell death brought upon HDAC inhibitors such as valproic acid and SAHA with more by the application of DL-TBOA (P < 0.01) and resulted in selective compounds as potential therapies for some neural a level of cell viability not significantly different from disorders. As discussed earlier and reviewed by Kazantsev control (culture medium alone) slice cultures or from each and Thompson (2008) and Chuang et al. (2009), general other; 107.5 ± 6.01% (n = 4), 97.1 ± 16.5% (n = 3) and non-selective HDAC inhibitors have been well reported as efficacious neuroprotective agents against glutamate excito- toxicity. Our novel selective HDAC2 and 3 inhibitors fully protected against neural cell death in our disease model; therefore, our study shows the potential of moving away from general inhibitors towards more selective ones, such as those against HDAC2 and 3 and there is no compromise in neuroprotective efficacy. The exact molecular mechanisms of neuroprotection by inhibiting HDAC2 and 3 are not yet known. However, it is likely that the mechanisms responsible involve the increased transcription of pro-survival and anti- apoptotic genes, but may also involve the prevention of neu- rotoxicity associated with increased HDAC2 and 3 activities. Cell culture models have been used to test HDAC inhibi- tors as neuroprotective agents against glutamate excitotoxic- ity (Marinova et al., 2009; Leng et al., 2010), but our model provides a more appropriate environment of the CNS with a mixture of cell types, which homogeneous cell cultures do not. Neurodegeneration in the CNS involves a complex inter- play between neurones and glial cells, so using intact cerebral slices maintains such a mix of different cell types and the interactions between them. We did not determine whether one Figure 2. Comparison of cell viability in cerebral slices between DL- specific cell type was more affected than others, what propor - TBOA and the novel HDAC2 and 3 selective inhibitors. Cerebral slices tion of cell death was derived from neurones or glial cells and were treated with control (culture medium alone, n = 6), DL-TBOA what possible changes in the interactions between glial cells (200 µM) alone (n = 7) or with 1 µM AH51 (n = 4), AH61 (n = 3) or AH62 and neurones took place. Nevertheless, HDAC inhibition pre- (n = 4). Data shown are mean ± SEM percent cell viability expressed as # vented DL-TBOA-induced cell death, suggesting that HDAC a percentage of control, P < 0.01 compared with control, *P < 0.01 inhibitors protect all cell types susceptible to DL-TBOA- compared with DL-TBOA. 5 Research article Bioscience Horizons • Volume 5 2012 mediated death. Using cell-specific immunohistochemical study. Furthermore, I thank Katy Burnage, Domenic Manfredi labelling and specific brain regions alongside the viability and Laura Anne Willis for their contributions with obtaining assay, one could directly correlate neural cell death with spe- the results. Finally, I thank Mohamed Al-Griw, for obtaining cific cell type and also assess any changes in glial and neurone the rat brains and providing the organotypic slice culture and interactions. By selecting specific brain regions such as the confocal microscopy training. cortex, hippocampus or the substantia nigra, our selective HDAC2 and 3 inhibitors can also be assessed to see whether Funding they better protect specific CNS regions and those predomi- nantly affected in different neurodegenerative diseases. This study was funded by the Faculty of Biological Sciences, University of Leeds, as part of the undergraduate final year General class I and II HDAC inhibitors are not only neu- project scheme. roprotective against glutamate excitotoxicity, but have been shown to be protective against other neurotoxic insults. Ryu et al. (2005) used organotypic mouse spinal cord slices from Author biography mice with mutant superoxide dismutase-1 (SOD1) amyo- trophic lateral sclerosis. By administering the general class I B.D. graduated in July 2011 with a First Class BSc (Hons) in and II HDAC inhibitor sodium phenylbutyrate, they saw a Medical Sciences from the Faculty of Biological Sciences, concentration-dependent increase in motor neurone survival. University of Leeds. In September 2011, he was shortlisted Coinciding with this finding was increased histone acetyla- for the Best European Biology Student of the Year at the tion and prevention of apoptosis through an increase in the Science, Engineering and Technology (SET) Awards. He has a expression of Bcl-2 and a reduction in the release of cyto- keen interest in uncovering the molecular mechanisms that chrome c from the mitochondria. Also, in a standard model control gene expression in human disease with a desire to of Parkinson’s disease, 1-methyl-4-phenylpyridinium (MPP ) develop and test therapies that reverse or compensate for induced, valproic acid prevented dopaminergic neurone loss this. B.D. started a PhD with Dr Ian C. Wood in October in cell culture (Kidd and Schneider, 2010) but also partially prevented the degeneration of the substantia nigra and striatum in in vivo rat models of the disease (Kidd and References Schneider, 2011). Chen et al. (2006) showed that valproic acid increases the expression of neurotrophins in dopaminer- Aoyama, K., Burns, D. M., Suh, S. W. et  al. (2005) Acidosis causes endo- gic neurones and glial cell cultures; including brain-derived plasmic reticulum stress and caspase-12-mediated astrocyte death, neurotrophic factor and glial cell line-derived neurotrophic Journal of Cerebral Blood Flow & Metabolism, 25 (3), 358–370. factor. These neurotrophins play prominent roles in neural cell development, neural cell survival and synaptic plasticity Armon, C., Shin, C., Miller, P. et  al. (1996) Reversible Parkinsonism and and therefore are ideal candidates to be increasingly expressed cognitive impairment with chronic valproate use, Neurology, 47 (3), to promote neuroprotection and neural cell function. The 626–635. novel HDAC2 and 3 selective inhibitors AH51, AH61 and Arundine, M. and Tymianski, M. (2003) Molecular mechanisms of cal- AH62, used in our study, fully protected neural cells from cium-dependent neurodegeneration in excitotoxicity, Cell Calcium, glutamate excitotoxicity like that observed by others with the 34 (4–5), 325–337. general class I and II HDAC inhibitors, and like the more general inhibitors, our selective ones may provide more wide- Baltan, S., Murphy, S. P., Danilov, C. A. et  al. (2011) Histone deacetylase spread protection and be effective in treating various neuro- inhibitors preserve white matter structure and function during isch- degenerative disorders. emia by conserving ATP and reducing excitotoxicity, Journal of Neuroscience, 31 (11), 3990–3999. Conclusion Bardai, F. H. and D’Mello, S. R. (2011) Selective toxicity by HDAC3 in neu- rons: regulation by Akt and GSK3β, Journal of Neuroscience, 31 (5), Selective HDAC inhibitors are a promising prospect for the 1746–1751. future treatment of neural cell death induced by glutamate excitotoxicity. This study has shown that novel HDAC2 and Bates, E. A., Victor, M., Jones, A. K. et al. (2006) Differential contributions 3 selective inhibitors can solely protect against this form of of Caenorhabditis elegans histone deacetylases to huntington poly- toxicity in an organotypic cerebral slice model. In addition, glutamate toxicity, Journal of Neuroscience, 26 (10), 2830–2838. these compounds could have further promise as therapeutic Bhat, R. V., Budd Haeberlein, S. L. and Avila, J. (2004) Glycogen synthase agents in other forms of neurodegeneration. kinase 3: a drug target for CNS therapies, Journal of Neurochemistry, 89 (6), 1313–1317. Acknowledgements Broide, R. S., Redwine, J. F., Aftahi, N. et al. (2007) Distribution of HIstone I would like to thank Dr Ian C. Wood for his insightful deacetylases 1–11 in the rat brain, Journal of Molecular Neuroscience, contribution and providing his laboratory resources for this 31 (1), 47–58. 6 Bioscience Horizons • Volume 5 2012 Research article Cancer Research Technology. Novel histone deacetylase (HDAC)-2 and 3 1,2,3,6-tetrahydropyridine mouse model of Parkinson’s disease, selective inhibitors (published online June 2011) accessed 14 Neuroscience, 194, 189–194. February 2012. Kilgore, M., Miller, C. A., Fass, D. M. et  al. (2010) Inhibitors of class I his- 2+ Carriedo, S. G., Yin, H. Z., Sensi, S. L. et al. (1998) Rapid Ca entry through tone deacetylase reverse contextual memory deficits in a mouse 2+ Ca permeable AMPA/kainite channels triggers marked intracellu- model of Alzheimer’s disease, Neuropyschopharmacology, 35 (4), 2+ lar Ca rises and consequent oxygen radical production, Journal of 870–880. Neuroscience, 18 (19), 7727–7738. Kim, D., Frank, C. L., Dobbin, M. M. et al. (2008) Deregulation of HDAC1 Chen, P. S., Peng, G. S., Li, G. et al. (2006) Valproate protects dopaminer- by p25/Cdk5 in Neurotoxicity, Neuron, 60 (5), 803–817. gic neurons in midbrain neuron/glia cultures by stimulating the Kornberg, R. D. and Lorch, Y. (1999) Twenty-five years of the nucleo - release of neurotrophic factors from astrocytes, Molecular Psychiatry, some, fundamental particle of the eukaryote chromosome, Cell, 98 11 (12), 1116–1125. (3), 285–294. Chuang, D.-M, Leng, Y., Marinova, Z. et al. (2009) Multiple roles of HDAC Lee, M.-C., Ting, K. K., Adams, S. et  al. (2010) Characterization of the inhibition in neurodegenerative conditions, Trends in Neurosciences, expression of NMDA receptors in human astrocytes, Public Library of 32 (11), 591–601. Science, 5 (11), e14123. Dong, X. X., Wang, Y. and Qin, Z. H. (2009) Molecular mechanisms of exci- Leng, Y., Marinova, Z., Reis-Fernandes, M. A. et  al. (2010) Potent neuro- totoxicity and their relevance to pathogenesis of neurodegenera- protective effects of novel structural derivatives of valproic acid: tive diseases, Acta Pharmacologica Sinica, 30 (4), 379–387. potential roles of HDAC inhibition and HSP70 induction, Gillespie, J., Savic, S., Wong, C. et al. (2011) Histone deacetylases are dys- Neuroscience Letters, 476 (3), 127–132. regulated in rheumatoid arthritis and a novel HDAC3-selective Marinova, Z., Ren, M., Wendland, J. R. et al. (2009) Valproic acid induces inhibitor reduces IL-6 production by PBMC of RA patients, Arthritis function heat-shock protein 70 via Class I histone deacetylase inhi- and Rheumatism, 64 (2), 418–422. bition in cortical neurons: a potential role of Sp1 acetylation, Journal Gogvadze, V. and Orrenius, S. (2006) Mitochondrial regulation of apop- of Neurochemistry, 111 (4), 976–987. totic cell death, Chemico-biological Interactions, 163 (1–2), 4–14. Matute, C., Alberdi, E., Ibarretxe, G. et  al. (2002) Excitotoxicity in glial Gregoretti, I. V., Lee, Y. M. and Goodson, H. V. (2004) Molecular evolu- cells, European Journal of Pharmacology, 447 (2–3), 239–246. tion of the histone deacetylase family: functional implications of McQuown, S. C., Barrett, R. M., Matheos, D. P. et  al. (2011) HDAC3 is a phylogenetic analysis, Journal of Molecular Biology, 338 (1), critical negative regulator of long-term memory formation, Journal 17–31. of Neuroscience, 31 (2), 764–774. Guan, J. S., Haggarty, S. J., Giacometti, E. et al. (2009) HDAC2 negatively MedlinePlus NIH. Valproic acid (published online 15 July 2011) accessed regulates memory formation and synaptic plasticity, Nature, 459 6 October 2011. (7243), 55–60. Merck. Patient Information ZOLINZA (vorinostat) Capsules (published Janssen, C., Schmalbach, S., Boeselt, S. et  al. (2010) Differential histone online November 2011) accessed 29 January 2012. deacetylase mRNA expression patterns in amyotrophic lateral scle- rosis, Journal of Neuropathology and Experimental Neurology, 69 (6), Pin, J. P. and Duvoisin, R. (1995) The metabotropic glutamate receptors: 573–581. structure and functions, Neuropharmacology, 34 (1), 1–26. Jiang, X. and Wang, X. (2004) Cytochrome C-mediated apoptosis, Annual Ryu, H., Smith, K., Camelo, S. I. et al. (2005) Sodium phenylbutyrate pro- Review of Biochemistry, 73 (1), 87–106. longs survival and regulates expression of anti-apoptotic genes in transgenic amyotrophic lateral sclerosis mice, Journal of Kazantsev, A. G. and Thompson, L. M. (2008) Therapeutic application of Neurochemistry, 93 (5), 1087–1098. histone deacetylase inhibitors for central nervous system disorders, Nature Reviews Drug Discovery, 7 (10), 854–868. Shigeri, Y., Seal, R. P. and Shimamoto, K. (2004) Molecular pharmacology Kidd, S. K. and Schneider, J. S. (2010) Protection of dopaminergic cells of glutamate transporters, EAATs and VGLUTs, Brain Research from MPP + -mediated toxicity by histone deacetylase inhibition, Reviews, 45 (3), 250–265. Brain Research, 1354 (1), 172–178. Stoppini, L., Buchs, P. A. and Muller, D. (1991) A simple method for Kidd, S. K. and Schneider, J. S. (2011) Protective effects of valproic acid organotypic cultures of nervous tissue, Journal of Neuroscience on the nigrostriatal dopamine system in a 1-methyl-4-phenyl- Methods, 37 (2), 173–182.

Journal

Bioscience HorizonsOxford University Press

Published: Apr 19, 2012

Keywords: histone deacetylase inhibitors HDAC2 HDAC3 neuroprotection, neurodegeneration glutamate excitotoxicity

There are no references for this article.