A Mercaptoacetamide-Based Class II Histone Deacetylase Inhibitor Increases Dendritic Spine Density via RasGRF1/ERK Pathway

Abstract

Background:

The accumulation of amyloid-β (Aβ) leads to the loss of dendritic spines and synapses, which is hypothesized to cause cognitive impairments in Alzheimer’s disease (AD) patients. In our previous study, we demonstrated that a novel mercaptoacetamide-based class II histone deacetylase inhibitor (HDACI), known as W2, decreased Aβ levels and improved learning and memory in mice. However, the underlying mechanism of this effect is unknown.

Objective:

Because dendritic spine formation is associated with cognitive performance, here we investigated whether HDACI W2 regulates dendritic spine density and its molecular mechanism of action.

Methods:

To examine the effect of HDACI W2 on dendritic spine density, we conducted morphological analysis of dendritic spines using GFP transfection and Golgi staining. In addition, to determine the molecular mechanism of W2 effects on spines, we measured the levels of mRNAs and proteins involved in the Ras signaling pathway using quantitative real-time PCR, immunocytochemistry, and western analysis.

Results:

We found that HDACI W2 altered dendritic spine density and morphology in vitro and in vivo. Additionally, W2 increased the mRNA or protein levels of Ras GRF1 and phospho-ERK. Moreover, knockdown of RasGRF1 and inhibition of ERK activity prevented the W2-mediated spinogenesis in primary hippocampal neurons.

Conclusion:

Our Class II-selective HDACI W2 promotes the formation and growth of dendritic spines in a RasGRF1 and ERK dependent manner in primary hippocampal neurons.

Keywords

Alzheimer’s disease dendritic spine HDAC inhibitor ras signaling

INTRODUCTION

Two enzymes competitively regulate the post-translational modification of histone acetylation: histone acetyltransferases (HATs) and histone deacetylases (HDACs). HDACs are responsible for catalyzing the acetyl group removal from N-terminal histone tails, thereby repressing transcription. Alternatively, histone acetyltransferases promote transcription by catalysis of histone acetylation, which causes chromatinrelaxation. Many disorders in the nervous system, including Alzheimer’s disease (AD), have been associated with irregular HDAC activity [1]. Importantly, transcriptional regulation mediated by epigenetic mechanisms is a reversible modification, as opposed to that regulated by genetic modification. Histone deacetylase inhibitors (HDACIs) have, therefore, gained interest as a promising treatment alternative for many pathological conditions, including neurological disorders.

Excitatory neurotransmission in the central nervous system occurs primarily at dendritic spines, and remodeling of these spines is correlated with cognitive function [2]. Specifically, a recent study has demonstrated that dendritic spine density is correlated with learning and memory [3]. Fischer et al. showed that HDACIs rescued deficits in learning and memory by inducing dendrite growth and synapse functioning [4]. Guan et al. then demonstrated that overexpression of HDAC2 led to decreased dendritic spine density [5]. Because histone acetylation status is altered during memory formation, HDACIs have recently gained attention in the field of neurodegenerative diseases [6]. Transcriptionally active chromatin structure has been shown to be crucial for functional learning and memory in mouse neurons, and the activity of chromatin structure depends on epigenetic mechanisms such as histone acetylation [7]. Therefore, the dysregulation of histone acetylation by HDACs has been studied as a potential mechanism of memory loss.

In our previous study, we demonstrated that a novel mercaptoacetamide-based class II HDACI, called W2, decreased the level of amyloid-β peptides (Aβ) and improved learning and memory deficits in a mouse model of AD [8]. In the present study, we investigated the biological effects of W2 on dendritic spine density. We found that W2 significantly increased dendritic spine density and size in primary hippocampal neurons and also increased the levels of Ras protein-specific guanine nucleotide-releasing factor 1 (RasGRF1) as well as phosphorylated extracellular signal-regulated kinase (phospho-ERK). In addition, W2 failed to increase dendritic spine density upon knockdown of RasGRF1 and inhibition of ERK signaling. These data suggest that HDACI W2 may promote spinogenesis by upregulating RasGRF1 and p-ERK levels and/or activity.

MATERIALS AND METHODS

Primary neuron culture and immunostaining

Primary hippocampal and cortical neurons from E19 Sprague–Dawley rats were cultured as previously described [9]. Primary hippocampal neurons were transfected with pEGFP, RasGRF1 shRNA vector, or pLL3.7 (control empty vector for RasGRF1 shRNA) using Lipofectamine 2000 (Invitrogen) and treated with HDAC inhibitor (10μM) or PD98059 (MEK inhibitor, Calbiochem) vehicle (0.05% DMSO). After transfection, cells were incubated with the following antibodies: mouse anti-GFP (Green Fluorescent Protein; Novus Biologicals, NB600-597), rabbit anti-GFP (Invitrogen, A11122), rabbit anti-GluA1 (Glutamate Receptor 1; Calbiochem, PC246), mouse anti-GluA2 (Glutamate Receptor 2; BD Pharmingen, 556341), mouse anti-PSD-95 (postsynaptic density-95; NeuroMabs, Davis, CA, USA), mouse anti-Synaptophysin (Sigma Aldrich, S5768), rabbit anti-RasGRF1 (Ras protein-specific guanine nucleotide-releasing factor 1; Santa Cruz, sc-224), rabbit anti-p-ERK1/2 (Thr202/Tyr204) (phosphorylated-extracellular signal-related protein kinase; Invitrogen, 368800), rabbit anti-p-CREB (cAMP response element-binding protein; Millipore, 06-519), GluN1, GluN2A, GluN2B (NeuroMabs). The immunostained cultured hippocampal neuron images were obtained with LSM 510 laser scanning confocal microscope (Zeiss).

Live cell surface immunostaining

To measure surface GluA1 and GluA2, primary hippocampal neurons were transfected with GFP and treated with HDACI (10μM) or vehicle. After 24 h, live primary hippocampal neurons were incubated with N-terminal GluA1 or GluA2 antibodies (10μg/mL in conditioned medium) for 10 min, and then fixed in 4% paraformaldehyde under non-permeabilizing conditions for 5 min. Surface-labeled GluA1 or GluA2 was detected with Alexa fluor-555 secondary antibodies. Cells were then permeabilized in methanol (–20°C, 90 s), and incubated with anti-GFP antibody to identify transfected neurons.

Animals

Wild-type C57BL/6J were obtained from Jackson Laboratories. All animal experiments were approved by the Institutional Animal Care and Use Committees at Georgetown University. All animals were maintained according to protocols approved by the Animal Welfare and Use Committee Georgetown University.

Golgi staining of dendritic spines

To analyze dendritic spine density in the brain, wild-type mice were injected with vehicle (DMSO, i.p.) or HDACI (50 mg/kg, i.p) daily for 4 weeks. After 4 weeks, the mice were sacrificed, and Golgi staining was conducted using the FD Rapid GolgiStain Kit (FD NeuroTechnologies). Briefly, dissected mouse brains were immersed in Solution A and B for 2 weeks in dark conditions at room temperature and transferred into Solution C for 24 h at 4°C. A VT1000 S Vibratome (Leica) was used for brain slicing at 150μm thickness. Dendritic images were acquired by Axioplan 2 (Zeiss) under brightfield microscopy. Spine density of cortical layers II/III and CA1 of hippocampus was measured using Scion image software (Scion Corporation). Images were coded, and dendritic spines were counted in a blinded manner. Spines from 0.2 to 2μm in length were included for analysis.

Quantitative real time PCR (qPCR)

Total RNAs from rat primary neuronal cells were extracted using RNAzol® RT (Molecular Research) and transcribed using High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). qPCR was performed with ABI 7500 instrument (Applied Biosystems) using Power SYBR Green PCR Master Mix (Applied Biosystems) under the default thermal cycling program with the following primers: glyceraldehyde-3-phosphate dehydrogenase (GAPDH) forward primer: AGGTCGGTGTGAACGGATTTG and reverse primer: TGTAGACCATGTAGTTGAGGTCA; RAS protein-specific guanine nucleotide-releasing factor 1 (Rasgrf1) forward primer: AGATCTGTACTGTTTCTCCAG and reverse primer: CCACAGATGGAAAGCTAAATG; The level of GAPDH was used as normalization reference. We used GenEx 5.3.2 (Multid Analyses) to analyze the relative mRNA levels from the acquired Ct values.

Image analysis

To measure dendritic spine density, images were acquired by LSM 510 confocal laser microscope (Zeiss), and dendritic spines were measured using ImageJ software. Dendritic spines were defined as protrusions 0.2 –2μm in length with or without a head and neck and were counted on secondary or tertiary dendritic segments. Dendritic spine density was quantified as dendritic spine number per dendrite length (dendritic spine density = dendritic spine number/ 10μm dendrite length). To measure puncta number of PSD-95 and Synaptophysin, we performed automated puncta analysis using MetaMorph software (Molecular Devices) [10].

Western blot

Cells were homogenized with lysis buffer, containing protease inhibitor and phosphatase inhibitor (Thermo Scientific). Lysate from cell were centrifuged at 15,000 rpm for 20 min at 4°C. After supernatant were collected and measured protein concentration using DC protein assay kit (Bio-Rad). Same protein amount were loaded 8% or 12% SDS gel and transferred PVDF membrane (Millipore) using Mini-PROTEIN Tetra cell electrophoresis system (Bio-Rad). The following primary antibodies and dilutions were used: rabbit phospho-ERK (1 : 1000, Invitrogen, 368800), rabbit ERK (1 : 1000, Cell signaling, 4377), mouse phospho-CREB (1 : 1000 Millipore, 06–519), rabbit CREB (1 : 1000, Cell signaling, 9197), mouse Histone H3 (1 : 1000, Cell signaling, 14269), rabbit acetylated histone H3 (1 : 1000, Cell signaling, 9677), rabbit alpha-tubulin (1 : 4000, Abcam, ab18251), mouse acetylated alpha-tubulin (1 : 1000, Abcam, ab24610), RasGRF1 (1 : 1000, Santa Cruz, sc-224) and mouse β-actin (1 : 7500, Santa Cruz, sc-47778). The following secondary antibodies and dilutions were used; horseradish peroxidase (HRP)-conjugated anti-rabbit or mouse IgG (1 : 7000, Santa Cruz). Immunoblot bands were visualized by chemiluminescent detection reagent (GE healthcare Life Science) and quantified using Image Gauge V4.0(Fuji Film).

Statistical analyses

All data were analyzed with Graphpad Prism 4 software using either a t-test with Welch’s correction or 1-way ANOVA with Tukey’s post hoc test for multiple comparisons, with significance determined at p < 0.05. Descriptive statistics were calculated with StatView 4.1 and expressed as mean ± SEM.

RESULTS

HDACI W2 increases spinogenesis in vitro

Before testing whether HDACI W2 can alter spinogenesis, we first confirmed that W2 could inhibit HDAC activity in neurons. Primary cortical neurons were treated with vehicle control (dimethyl sulfoxide, 0.05% DMSO) or W2 (1, 5, and 10μM) (Fig. 1A–D).After 4 h, we conducted western blotting with anti-histone H3, acetylated histone H3 (a substrate for class I and II HDACs), alpha tubulin, acetylated-alpha tubulin (a substrate of HDAC6, a class II histone deacetylase), and β-actin as loading control. Consistent with our previous finding [11], we found that W2 did not changes the levels of acetylated histone H3 but did increase the levels of acetylated-alpha tubulin (Fig. 1A–D).

To determine whether W2 has HDACI activity after a longer treatment in our culture system, primary cortical neurons were treated with vehicle control (0.05% DMSO) or W2 (10μM, a dose that enhances acetylation of alpha tubulin) for 24 h (Fig. 1E, F) and conducted western blotting for acetylated-alpha tubulin, alpha tubulin and β-actin. We found that W2 significantly increased the levels of acetylated-alpha tubulin, but did not alter the levels of total alpha tubulin (Fig. 1E, F). These data suggested that HDACI W2 selectively inhibited HDAC class 2 activity.

We then examined the effect of W2 on dendritic spine formation. Because cognitive performance correlates with dendritic spine density [12, 13] and W2-injected mice display improved cognitive behavior [8], we hypothesized that this HDACI may improve cognitive performance by increasing spine density. To test this idea, we transfected primary hippocampal neurons at day in vitro (DIV) 19 with green fluorescent protein (GFP) and treated with vehicle control (0.05% DMSO) or W2 (1, 5, and 10μM) (Fig. 1G). After 24 h, spine number was counted using ImageJ software. Treatment with 5 and 10μM of W2 significantly increased dendritic spine density (see Methods for calculation), whereas 1μM of W2 did not (Fig. 1H).

We then tested whether W2 can affect dendritic spine number in younger neurons, during a developmental period corresponding to the peak of synaptogenesis. Primary hippocampal neurons at DIV 12 were transfected with GFP and treated with control or W2 (10μM) for 24 h. We found that W2 increased dendritic spine density during peak synaptogenesis (Fig. 1I, J). These data suggest that W2 can affect dendritic spine density during neuronal development.

HDACI W2 increases dendritic spine density in wild-type mice

To test whether W2 increases dendritic spine density in vivo, we injected wild-type mice with W2 (50 mg/kg, i.p.) or vehicle (DMSO, i.p) daily for 4 weeks. This dosage was selected because it improved learning and memory in our previous study [8]. After 4 weeks, we conducted Golgi staining to measure dendritic spine density in wild-type mice. Consistent with our in vitro data, we found that W2-injected wild-type mice exhibited significantly increased dendritic spine density on apical oblique (AO), basal (BS), and total (AO+BS) dendrites in the hippocampal cornu ammonis 1 (CA1) region (Fig. 2A, B). In cortical layers II and III, we found that HDACI W2-injected wild-type mice had significantly increased dendritic spine density in BS and AO+BS (total) dendrites, but not in AO dendrites (Fig. 2C, D). These results suggest that W2 alters dendritic spine number in vivo, potentially in an input- and region-specific manner.

HDACI W2 significantly increased the puncta number of synaptophysin and PSD-95

Next, we investigated whether W2 can affect the number of excitatory synapses. Since we observed that 10μM W2 was effective in altering dendritic spine density, we selected this dose for subsequent experiments. Primary hippocampal neurons at DIV 19 were transfected with GFP and treated with vehicle control (0.05% DMSO) or W2 (10μM). After 24 h, cells were fixed and immunostained for synaptophysin (pre-synaptic marker) and postsynaptic density protein-95 (PSD-95; postsynaptic marker). We found that compared to vehicle, W2-treated neurons showed an increase in the number of both synaptophysin puncta (Fig. 3A, B) and PSD-95 puncta (Fig. 3C, D). We also examined whether increases in PSD-95 correlates with spine size or spine morphology. To test this, we measured spine area, spine head width, and spine head length in the presence and absence of W2. Interestingly, we found that W2-treated cells had longer and wider spines compared to vehicle-treated cells, accompanied by increases in PSD-95 integrated intensity (Fig. 3E–H). These data suggest that W2 may increase functional synaptic properties as well as size of spines.

HDACI W2 selectively increases the level of the NMDA receptor subunit GluN1

After determining that W2 increased the number of dendritic spines and excitatory synapses in hippocampal neuronal cultures, we then examined whether W2 altered the levels of N-methyl-D-aspartate receptor (NMDA) subunits, which are associated with hippocampal synaptic plasticity and learning and memory [14]. For this experiment, we transfected primary hippocampal neurons at DIV 19 with GFP and treated these neurons with vehicle control (0.05% DMSO) or W2 (10μM). After 24 h, the cells were fixed and immunostained for the NMDA receptor subunits GluN1, GluN2A, and GluN2B. We observed that W2-treated primary hippocampal neurons showed significantly increased levels of GluN1 (Fig. 4A, B), but not GluN2A (Fig. 4C, D) or GluN2B (Fig. 4E, F). These data suggest that the HDACI W2 differentially affects the expression levels of the NMDA receptor subunits in vitro.

HDACI W2 increases cell surface levels of the AMPA receptor subunits GluA1 and GluA2

Because AMPA receptors are important factors regulating memory function in the hippocampus [15], we investigated the effect of W2 on AMPA receptor trafficking. To address this issue, primary hippocampal neurons at DIV 19 were transfected with GFP and treated with vehicle control (0.05% DMSO) or W2 (10μM). After 24 h, we conducted immunostaining to measure the levels of cell surface and total AMPA receptors. We found that W2 significantly increased the cell surface levels of AMPA receptor subunits GluA1 (Fig. 5A, B) and GluA2 (Fig. 5E, F), without altering their total levels (Fig. 5C, D, G, H). These data suggest that W2 promotes cell surface expression of GluA1 and GluA2.

HDACI W2 requires RasGRF1 to alter spinogenesis

Ras signaling is an important pathway for dendritic spine formation and neurodegeneration [16]. Thus, we examined whether W2 alters the levels and/or activity of Ras signaling proteins. To test this question, we examined the mRNA and protein levels of RasGRF1, a Ras activator, after treatment of primary cortical neurons with vehicle control (0.05% DMSO) or W2 (10μM). We found that W2 increased RasGRF1 mRNA and protein levels by using qPCR and western blotting, respectively (Fig. 6A–C). As an alternative approach, we used immunostaining and verified that W2 increased the expression levels of RasGRF1 in primary hippocampal neurons (Fig. 6D, E).

Next, we examined whether W2 alters spinogenesis in a RasGRF1-dependent manner. For this experiment, primary hippocampal neurons (DIV19) were cotransfected with GFP together with either RasGRF1 shRNA or pLL3.7 (control vector for RasGRF1 shRNA). After 24 h, cells were treated with vehicle control (0.05% DMSO) or W2 (10μM) for an additional 24 h, and then dendritic spine density was measured. We found that knockdown of RasGRF1 by itself decreased dendritic spine density compared to pLL3.7 control vector, whereas W2 had significantly increased dendritic spine density compared to vehicle control (Fig. 6F, G). Importantly, W2 failed to increase dendritic spine density with knockdown of RasGRF1. These results suggest that W2-mediated dendritic spine formation is dependent on RasGRF1.

HDACI W2 increases dendritic spine density through ERK activity

We then examined the activity of specific molecules in the signaling pathway downstream of Ras, including ERK and cAMP response element-binding protein (CREB). Primary hippocampal neurons were transfected with GFP and treated with vehicle control (0.05% DMSO) or W2 (10μM). After 24 h, we observed that W2 significantly increased the levels of phospho-ERK but did not alter total ERK, phospho-CREB, or total CREB (Fig. 7A–D). Similar results were obtained by western blotting of cortical neuron lysates treated with vehicle or W2 (Fig.7E–H).

To test whether ERK signaling is necessary for W2’s effect on dendritic spine formation, we transfected primary hippocampal neurons with GFP for 24 h and pre-treated with vehicle control (0.05% DMSO) or an inhibitor of MEK (an ERK activating kinase) (PD98059, 10μM). After 1 h, the cells were treated with vehicle control (0.05% DMSO) or W2 (10μM) for 24 h and dendritic spine density was then measured. Compared to vehicle, W2 significantly increased dendritic spine density, whereas PD98059-treated cells did not cause any change in dendritic spine density. However, pre-treatment with PD98059 completely inhibited W2-mediated spinogenesis (Fig. 7I, J).These results suggest that W2 increased dendritic spine density by regulating ERK pathway activity.

DISCUSSION

HDACIs, originally utilized in cancer therapy, are gaining attention as potential treatments forneurodegenerative disorders due to their regulation of Aβ and cognitive function [17 –19]. In our previous study, we demonstrated that the HDACI W2 decreased levels of Aβ as well as improved cognitive performance in a mouse model of AD [8]. Several studies have demonstrated that deficits in learning and memory are correlated with dendritic spine density [3, 20]. For instance, Ripoli et al. demonstrated that Aβ accumulation reduces dendritic spine density and impairs cognitive function [21]. In addition, HDACIs such as sodium butyrate increase long-term potentiation, increase dendritic spine density, and improve cognitive performance [22, 23]. Another study demonstrated that treatment with the HDACI phenyl butyrate restored dendritic spine density in a mouse model of AD [24]. However, the molecular mechanisms by which HDACIs regulate cognitive and synaptic function are not well studied.

In the present study, we examined the effects of HDACI W2 treatment on synaptic structure and function. We demonstrated that W2 promoted dendritic spine number in vitro. Wild-type mice treated with W2 also exhibited a higher number of cortical and hippocampal dendritic spines in vivo. Interestingly, we found that W2-injected mice showed distinct effects on dendritic spine density in apical oblique (AO) versus basal dendrites (BS) in the hippocampus and cortex. Both the hippocampus and cortex receive distinct afferent inputs to their apical oblique (AO) versus basal shaft (BS) dendrites. This difference in inputs could explain the selective impact that HDACIs may have in the cortex. For example, for electrophysiological recordings, the BS dendritic spikes seem to rely more on NMDARs in comparison to AO dendrites which depend more on voltage-gated channels [25]. Moreover, these two sub-compartments (AO versus BS) in the cortex have been shown to respond to genetic perturbations differently; in the Tg2576 AD mouse model, cortical pyramidal neurons have altered morphology of AO dendritic spines but not BS dendritic spines compared to wild-type mice [26].

Based on our findings and the literature, we hypothesize that W2 may have differential effects on AO and BS synapses in cortical layers II/III by increasing endogenous NMDA receptor subunit GluN1 (as W2 can alter GluN1 expression levels (Fig. 4)) or some unknown genes in BS (but not AO) that can lead to selective effects on dendritic spine density. Further studies are required to elucidate these effects.

Does HDACI W2 affect functional synapses? A previous study demonstrated that the HDACIs SAHA and TSA increase synaptophysin in LNCaP and LAPC4 cells [27]. In addition, the HDACI TSA was shown to increase PSD-95 clustering in primary hippocampal neurons [28]. Akhtar et al. found that TSA increases the frequency without changing the amplitude of the AMPA-mediated miniature excitatory postsynaptic current in primary hippocampal neurons [29]. Here, we found that HDACI W2-treated primary hippocampal neurons showed a significant increase in the number of puncta immunoreactive for PSD-95, a postsynaptic marker, and synaptophysin, a pre-synaptic marker, as well as by increased expression of NMDA receptor subunit GluN1 and cell surface levels of AMPA receptor subunits GluA1/2. These data suggest that HDACI W2 regulates dendritic spine number by altering excitatory synapses. However, further studies are required to examine the effect of W2 on synaptic function using electrophysiological techniques.

Next, we investigated the molecular mechanism by which the novel HDACI W2 regulates dendritic spine density. Previously, Guan et al. demonstrated that the reduced synapse number and cognitive function in HDAC2-overexpressing mice could be rescued by treatment with the HDACI SAHA, which regulates chromatin-mediated neuroplasticity [5]. However, the details of the mechanism mediating the HDACI regulation of synaptic plasticity remained unknown. We began exploring this mechanism by examining whether the HDACI W2 acted through a Ras-dependent mechanism because Ras signaling plays a critical role in both dendritic spine formation and neurodegeneration [30 –33]. For instance, RasGRF1 knockout mice exhibit impaired cognitive function [34]. Additionally, Eckel-Mahan et al. demonstrated that Ras activity and downstream ERK activation peak during the day, resulting in increased daytime learning and memory compared with that during the night [35]. Here, we observed that HDACI W2 treatment regulated the levels of RasGRF1 and p-ERK, and that W2 requires RasGRF1 and ERK signaling to alter dendritic spine formation. Our results suggest that W2 regulates dendritic spine formation via a RasGRF1/ERK-dependent signaling pathway.

Another possible mechanism is that W2 may act on the chromatin histones present at the promoter of synaptic genes. For example, Guan et al. demonstrated that HDAC2 suppressed the expression of synaptic proteins PSD-95, GluN2A, GluN2B and GluA1 using ChIP assay [5]. Although our studies did not observe a change in histone H3 acetylation in cortical neurons with W2 treatment, we cannot rule out the transcriptional effects mediated by other types of histone acetylation. Further in-depth studies will be required to understand how W2 affects RasGRF1 and whether the enhanced transcription of synaptic genes such as AMPA/NMDA receptors or PSD-95 also contributes to the increase of spine density by W2.

Taken together with our previous findings, W2 is a promising drug candidate for AD treatment. A greater understanding of HDACI characteristics will provide insight into the therapeutic potential of the novel HDACI W2, not only for AD but also for other neurodegenerative disorders.

Footnotes

ACKNOWLEDGMENTS

The monoclonal antibody PSD-95, NR1, NR2B were developed and obtained from UC Davis/NIH NeuroMabFacility. This work was supported by Georgetown University faculty tenure track start package-up and KBRI basic research program through Korea Brain Research Institute funded by the Ministry of Science, ICT & future Planning (No.2231-415) (H.S.H), Mayo Clinic (JK), and GHR Foundation (JK).

Authors’ disclosures available online ().

References

, Dai

, Huang

, Jiang

(2011) Targeting HDACs: A promising therapy for Alzheimer’s disease. Oxid Med Cell Longev 2011, 143269.

De Roo

, Klauser

, Garcia

, Poglia

, Muller

(2008) Spine dynamics and synapse remodeling during LTP and memory processes. Prog Brain Res 169, 199–207.

Vivar

, Potter

, van Praag

(2013) All about running: Synaptic plasticity, growth factors and adult hippocampal neurogenesis. Curr Top Behav Neurosci 15, 189–210.

Fischer

, Sananbenesi

, Wang

, Dobbin

, Tsai

(2007) Recovery of learning and memory is associated with chromatin remodelling. Nature 447, 178–182.

Guan

, Haggarty

, Giacometti

, Dannenberg

, Joseph

, Gao

, Nieland

, Zhou

, Wang

, Mazitschek

, Bradner

, DePinho

, Jaenisch

, Tsai

(2009) HDAC2 negatively regulates memory formation and synaptic plasticity. Nature 459, 55–60.

Mungenast

, Tsai

(2012) Cognitive function in health and disease: The role of epigenetic mechanisms. Neurodegener Dis 10, 191–194.

Konsoula

, Barile

(2012) Epigenetic histone acetylation and deacetylation mechanisms in experimental models of neurodegenerative disorders. J Pharmacol Toxicol Methods 66, 215–220.

Sung

, Lee

, Yoon

, DiBattista

, Song

, Sohn

, Moffat

, Turner

, Jung

, Kim

, Hoe

(2013) Mercaptoacetamide-based class II HDAC inhibitor lowers Abeta levels and improves learning and memory in a mouse model of Alzheimer’s disease. Exp Neurol 239, 192–201.

Pak

, Yang

, Rudolph-Correia

, Kim

, Sheng

(2001) Regulation of dendritic spine morphology by SPAR, a PSD-95-associated RapGAP. Neuron 31, 289–303.

10.

DiBattista

, Dumanis

, Song

, Bu

, Weeber

, Rebeck

, Hoe

(2015) Very low density lipoprotein receptor regulates dendritic spine formation in a RasGRF1/CaMKII dependent manner. Biochim Biophys Acta 1853, 904–917.

11.

Konsoula

, Velena

, Lee

, Dritschilo

, Jung

(2011) Histone deacetylase inhibitor: Antineoplastic agent and radiation modulator. Adv Exp Med Biol 720, 171–179.

12.

Hayashi

, Choi

, Rao

, Jung

, Lee

, Zhang

, Chattarji

, Kirkwood

, Tonegawa

(2004) Altered cortical synaptic morphology and impaired memory consolidation in forebrain- specific dominant-negative PAK transgenic mice. Neuron 42, 773–787.

13.

Morgado-Bernal

(2011) Learning and memory consolidation: Linking molecular and behavioral data. Neuroscience 176, 12–19.

14.

Lee

, Kesner

(2002) Differential contribution of NMDA receptors in hippocampal subregions to spatial working memory. Nat Neurosci 5, 162–168.

15.

Sanderson

, Good

, Seeburg

, Sprengel

, Rawlins

, Bannerman

(2008) The role of the GluR-A (GluR1) AMPA receptor subunit in learning and memory. Prog Brain Res 169, 159–178.

16.

Patterson

, Yasuda

(2011) Signalling pathways underlying structural plasticity of dendritic spines. Br J Pharmacol 163, 1626–1638.

17.

Lockett

, Wilkes

, Maleszka

(2010) Brain plasticity, memory and neurological disorders: An epigenetic perspective. Neuroreport 21, 909–913.

18.

Robakis

(2003) An Alzheimer’s disease hypothesis based on transcriptional dysregulation. Amyloid 10, 80–85.

19.

Yacoubian

, Cantuti-Castelvetri

, Bouzou

, Asteris

, McLean

, Hyman

, Standaert

(2008) Transcriptional dysregulation in a transgenic model of Parkinson disease. Neurobiol Dis 29, 515–528.

20.

Brewster

, Lugo

, Patil

, Lee

, Qian

, Vanegas

, Anderson

(2013) Rapamycin reverses status epilepticus-induced memory deficits and dendritic damage. PLoS One 8, e57808.

21.

Ripoli

, Piacentini

, Riccardi

, Leone

, Li Puma

, Bitan

, Grassi

(2013) Effects of different amyloid beta-protein analogues on synaptic function. Neurobiol Aging 34, 1032–1044.

22.

Haettig

, Stefanko

, Multani

, Figueroa

, McQuown

, Wood

(2011) HDAC inhibition modulates hippocampus-dependent long-term memory for object location in a CBP-dependent manner. Learn Mem 18, 71–79.

23.

Townsend

, Qu

, Gray

, Wu

, Seto

, Hutton

, Shearman

, Middleton

(2010) Oral treatment with a gamma-secretase inhibitor improves long-term potentiation in a mouse model of Alzheimer’s disease. J Pharmacol Exp Ther 333, 110–119.

24.

Ricobaraza

, Cuadrado-Tejedor

, Marco

, Perez-Otano

, Garcia-Osta

(2012) Phenylbutyrate rescues dendritic spine loss associated with memory deficits in a mouse model of Alzheimer disease. Hippocampus 22, 1040–1050.

25.

Antic

, Zhou

, Moore

, Short

, Ikonomu

(2010) The decade of the dendritic NMDA spike. J Neurosci Res 88, 2991–3001.

26.

Alpar

, Ueberham

, Bruckner

, Seeger

, Arendt

, Gartner

(2006) Different dendrite and dendritic spine alterations in basal and apical arbors in mutant human amyloid precursor protein transgenic mice. Brain Res 1099, 189–198.

27.

Frigo

, McDonnell

(2008) Differential effects of prostate cancer therapeutics on neuroendocrine transdifferentiation. Mol Cancer Ther 7, 659–669.

28.

Catarino

, Ribeiro

, Santos

, Carvalho

(2013) Regulation of synapse composition by protein acetylation: The role of acetylated cortactin. J Cell Sci 126, 149–162.

29.

Akhtar

, Raingo

, Nelson

, Montgomery

, Olson

, Kavalali

, Monteggia

(2009) Histone deacetylases 1 and 2 form a developmental switch that controls excitatory synapse maturation and function. J Neurosci 29, 8288–8297.

30.

Lee

, Hoe

, Pak

(2011) Plk2 Raps up Ras to subdue synapses. Small GTPases 2, 162–166.

31.

Saini

, Chisari

, Gautam

(2009) Shuttling and translocation of heterotrimeric G proteins and Ras. Trends Pharmacol Sci 30, 278–286.

32.

Stornetta

, Zhu

(2011) Ras and Rap signaling in synaptic plasticity and mental disorders. Neuroscientist 17, 54–78.

33.

, Carew

(2010) Small G protein signaling in neuronal plasticity and memory formation: The specific role of ras family proteins. Neuron 68, 340–361.

34.

Fasano

, D’Antoni

, Orban

, Valjent

, Putignano

, Vara

, Pizzorusso

, Giustetto

, Yoon

, Soloway

, Maldonado

, Caboche

, Brambilla

(2009) Ras-guanine nucleotide-releasing factor 1 (Ras-GRF1) controls activation of extracellular signal-regulated kinase (ERK) signaling in the striatum and long-term behavioral responses to cocaine. Biol Psychiatry 66, 758–768.

35.

Eckel-Mahan

, Phan

, Han

, Wang

, Chan

, Scheiner

, Storm

(2008) Circadian oscillation of hippocampal MAPK activity and cAmp: Implications for memory persistence. Nat Neurosci 11, 1074–1082.