Inhibition of Histone Acetylation by ANP32A Induces Memory Deficits

Abstract

There is accumulating evidence that decreased histone acetylation is involved in normal aging and neurodegenerative diseases. Recently, we found that ANP32A, a key component of INHAT (inhibitor of acetyltransferases) that suppresses histone acetylation, increased in aged and cognitively impaired C57 mice and expressing wild-type human full length tau (htau) transgenic mice. Downregulating ANP32A restored cognitive function and synaptic plasticity through upregulation of the expressions of synaptic-related proteins via increasing histone acetylation. However, there is no direct evidence that ANP32A can induce neurodegeneration and memory deficits. In the present study, we overexpressed ANP32A in the hippocampal CA3 region of C57 mice and found that ANP32A overexpression induced cognitive abilities and synaptic plasticity deficits, with decreased synaptic-related protein expression and histone acetylation. Combined with our recent studies, our findings reveal that upregulated ANP32A induced-suppressing histone acetylation may underlie the cognitive decline in neurodegenerative disease, and suppression of ANP32A may represent a promising therapeutic approach for neurodegenerative diseases including Alzheimer’s disease.

Keywords

Alzheimer’s disease ANP32A cognition histone acetylation synaptic-related protein

INTRODUCTION

Alzheimer’s disease (AD) is a common neurodegenerative disorder that is characterized by two neuropathological hallmarks: extracellular senile plaque deposits composed of amyloid-β (Aβ) and intracellular neurofibrillary tangles composed of hyperphosphorylated tau [1, 2]. One of the initial clinical manifestations of AD is a reduced ability to recall past experiences and the severity of memory loss increasing with disease progression. Various deficits occur in addition to memory loss, including language, spatial orientation, attention and executive function deficits [3]. Although five medications are currently used to treat the cognitive problems of AD: tacrine, rivastigmine, galantamine, donepezil (all acetylcholinesterase inhibitors), and memantine (an NMDA receptor agonist), the benefit from their use is small [4, 5].

Epigenetic modifications, particularly histone acetylation in the nervous system, play a critical role in regulation of gene expression for long-term memory [6, 7]. Acetylation diminishes the electrostatic affinity between neighboring histones and the DNA and, consequently, can promote a more open chromatin structure that allows for memory-related gene transcription [8]. In recent decade, several studies including ours have reported sporadic cases of reduced histone acetylation in patients with AD and models, which are characterized by cognitive decline [9, 10]. Accordingly, pharmacological treatments aimed at increasing histone acetylation have shown promising results in rescuing cognitive functions in some of these models by the use of nonselective histone deacetylase (HDAC) inhibitors [11, 12]. However, the causative agent that disturbs histone acetylation and induces cognitive impairment remain unknown. One likely candidate is acidic leucine-rich nuclear phosphoprotein-32A (ANP32A) [13], a key component of the inhibitor of histone acetyltransferase complex (INHAT) [14].

INHAT suppresses histone acetylation through a mechanism termed histone-masking, in which it binds to histones and masks their access to acetyltransferases [9]. We previously found that ANP32A protein level increased in aged and cognitive impaired C57 mice and ∼12 m expressing wild-type human full-length tau (htau) transgenic mice with decreasing acetylated level of histone. Meanwhile, downregulation of ANP32A in htau mice or aged and cognitive impaired C57 mice restored cognitive function and synaptic plasticity through upregulation of the expression of synaptic-related proteins via increasing histone acetylation [9, 10]. Notably, levels of ANP32A messenger RNA (mRNA) and protein were significantly increased in brains affected by AD [15]. These results indicate that ANP32A plays a key role in neurodegenerative diseases including AD. Thus, we hypothesized that cognitive capacities in the neurodegenerating brain are constrained by an epigenetic blockade of gene transcription mediated by ANP32A.

To verify this hypothesis, we overexpressed ANP32A in the hippocampal CA3 region of C57 mice by stereotaxic injection of adeno-associated virus expressing ANP32A, and found that ANP32A overexpression impaired learning and memory, reduced field excitatory postsynaptic potentials slope (fEPSPs) and decreased the spine density including mushroom and thin-shaped spines. We also found that overexpressing ANP32A decreased synaptic-related protein expression and histone acetylation.

MATERIALS AND METHODS

Antibodies

Polyclonal antibody (pAb) anti-H4K8 (acetylated histone H4 at lysine 8), pAb anti-NR2A, pAb anti-synapsin-1, pAb anti-synaptophysin, pAb anti-GluR2, and monoclonal antibody (mAb) anti-GluR1 were purchased from Millipore (Billerica, MA). PAb anti-H4K12 (acetylated histone H4 at lysine 12), pAb anti-NR2B, pAb anti-ANP32A, pAb anti-histone H4, pAb anti-histone H3, and pAb anti-H3K9 (acetylated histone H3 at lysine 9) were from Abcam (Cambridge, MA). PAb anti-H3K14 (acetylated histone H3 at lysine 14) was from Cell Signaling Technology (Cambridge Danvers, MA). MAb anti-DM1A (total α-tubulin) was purchased from Sigma-Aldrich (St. Louis, MO).

Production and brain infusion of adeno-associated virus

Adeno-associated virus-green fluorescent protein (GFP) expressing ANP32A (AAV-ANP) or the control AAV (AAV-C), whose titer was 1×10¹² TU/mL, was purchased from Heyuan Biochemical Inc (Shanghai, China). In order to avoid the possible influence of GFP in the subcellular localization of ANP32A, GFP sequence is not fused with ANP32A sequence.

For brain stereotactic injections, ∼3-month-old male C57 mice, purchased from the animal center of Tongji Medical College, Huazhong University of Science and Technology, were positioned respectively in a stereotaxic instrument, then AAV-ANP32A (AAV-ANP) and its control AAV virus (2μL) were bilaterally injected into the hippocampal CA3 region (AP±2.0, ML – 1.5, DV – 2.0) at a rate of 0.50μL/min. The needle was left in place for ∼3 min, and then withdrawn. All mice were kept at 24±2°C on daily 12-h light-dark cycles with ad libitum access to food and water. Fluorescence staining and western blot analysis were used to measure the in vivo expression efficiency after virus injection into the hippocampal CA3 region of mice brains. All animal experiments were performed according to the ‘Policies on the Use of Animals and Humans in Neuroscience Research’ revised and approved by the Society for Neuroscience in 1995, and the animal study was approved by the Institutional Animal Care and Use Committee at Tongji Medical College, Huazhong University of Science and Technology.

Behavioral tests

The effects of overexpression of ANP32A in the cognitive functions of C57 mice were detected by Morris water maze (MWM) in 4 weeks after brain infusion of the viral vectors. For the MWM test [16], briefly, mice were trained to find a hidden platform in the water maze for 6 consecutive days, and underwent four training trials (once per quadrant) per day with a 30-s interval from 14:00 to 20:00 pm. In each training trial, mice were placed into the water from a semi random set of start locations in each quadrant and faced the pool wall and ended when the animal climbed on the platform. If the mice did not locate the platform within 60 s, they were guided to the platform to stay for another 30 s. Spatial memory was tested 1 day after the last training. The longer a mouse stayed in the previous platform-located quadrant, the better it scored the spatial memory. A video camera, fixed to the ceiling, 1.5 m from the water surface, was used to record the swimming path and the time, which was used to find the platform (latency) or pass through the previous platform-located quadrant each day. The camera was connected to a digital-tracking device attached to an IBM (Armonk, NY) computer.

1 week after the MWM test, the Barnes maze test was performed [17]. Briefly, animals were trained to locate a dark escape chamber hidden underneath a hole positioned around the perimeter of a disk. To provide an aversive stimulus, four overhead halogen lamps were used to brightly illuminate the disk. The maze was manufactured from acrylic plastic to form a disk 1.5 cm thick and 115 cm in diameter, with 20 evenly spaced 7-cm holes at its edges. A video camera, which was installed directly overhead at the center of the maze, was used to record all trials simultaneously. Animals were trained with two trials per day for 5 consecutive days. After the animal was placed in the center of the maze, which was covered under a cylindrical start chamber, a trial was started; the start chamber was raised after a 10-s delay. If the animal had entered the escape chamber or when a predetermined time (5 min) had elapsed, whichever came first, a training session ended. In order to eliminate possible olfactory cues from preceding animals, all surfaces were routinely cleaned before and after each trial.

Real-time polymerase chain reaction

Total RNA (3μg in 25μL) was isolated using TrizolTM (Invitrogen, Carlsbad, CA), and then was reverse transcribed. The produced complementary DNA (1μL) was used for real-time polymerase chain reaction with primer sets [18], synaptophysin: 5^′-CAAACAATACCGAAGGGCACAG-3^′ and 5^′-AAGAGGGCTAGATAATCAGAAGACAGA-3^′, synapsin I: 5^′- CTTTGCTTGTTTATTTTGCTTC-3^′ and 5^′-CCAATGTGTTTATCTGTGACTG-3^′, NR2B: 5’-TATGCTCTTTGGGTCAGTCTCGTT-3^′ and 5^′- GTCCCTTTATCCTCCGTCTTTCTT-3^′, NR2A: 5^′- TCAAGGAAAGCAGAAGGGGAAA-3^′ and 5^′-TGTGGAATGGAATGATAGGCGA-3^′, GluR1:5^′-CACGTTTTCTCGGTAGGCATT-3^′, and 5’-CACATGTAGCCGGAGTGATG-3^′, GluR2:5^′-CACTCAAGAGGATGGGGAAA-3^′, and 5^′-ATTTCGGGTAGGGATGGTTC-3^′, and β-actin: 5^′-AGCCTTCCTTCTTGGGTAT-3^′ and 5’-GCTCAGTAACAGTCCGCCTA-3^′.

Golgi impregnation and dendritic morphology analysis

For Golgi stain, after the anesthetized mice were perfused transcardially with 4% paraformaldehyde, brain tissue was processed as described [19]. Individual sections were incubated in water solution of 3.5% K₂Cr₂O₇ and 0.4% OsO₄ overnight, and then, sandwiched in two glass slides and incubated in 1% AgNO_3 (aq) for 5 h in the dark at room temperature. The sections were mounted on gel-coated slides (0.5% porcine gelatin) after the slide assemblies were dismantled in water. A series of graded ethanol was used to dehydrate the sections, and then, the sections were cleared with Histoclear (Sigma-Aldrich) and cover slipped with Cytoseal (ThermoFisher Scientific, Waltham, MA). Olympus BX60 (Tokyo) was used to take the images.

The segments of dendrites at a distance of 190–210μm (distal) from the soma were used to determine the spine density. To acquire images for spine analysis, the dendritic segments were imaged under bright-field illumination on a Zeiss Axioimager microscope with a 63×oil immersion objective, and spine morphology was analyzed according to a previously reported method [20], which does not assess spine density in a three-dimensional manner but focuses on spines parallel to the plane of the section. Although the method may underestimate the total number of spines, it facilitates a direct comparison of treatment groups when they are analyzed in an identical manner. Linear spine density, presented as the number of spines per 10μm of dendrite length, was calculated by Image J software [21]. Spines were classified into thin (spines with a long neck and a visible small head) and mushroom (big spines with a well-defined neck and a very voluminous head) on the basis of morphology. Data from 5–7 neurons were averaged per animal and used in further statistical analysis.

Electrophysiology

A vibration microtome was used to cut mice brains into horizontal sections of 400-μm thickness in cold artificial cerebrospinal fluid (aCSF) (NaCl 126 mM, KCl 2.5 mM, NaHCO₃ 26 mM, NaH₂PO₄ 1.25 mM, CaCl₂ 2 mM, MgCl₂ 2 mM, glucose 10 mM, equilibrated with 95% O₂ and 5% CO₂). Then, the hippocampal slices were transferred into oxygen-enriched aCSF to recover for 30 min. Individual slices were laid down over an 8×8 array of planar microelectrodes, each 50×50 mm in size, with an interpolar distance of 450 mm (MED-P5455; Alpha MED Sciences, Kadoma, Japan) and kept submerged in aCSF (4 mL/min; 30°C) with a nylon mesh glued to a platinum ring. The MED64 System (Alpha MED Sciences) was used to acquire voltage signals. fEPSPs were recorded from CA3 in hippocampus by stimulating mossy fibers. Stimulation intensity was adjusted to evoke fEPSPs amplitudes that were 30% of maximal size. One train of high-frequency stimulation (100 Hz, 1-s duration at test strength) was applied to induce long-term potentiation.

Western blotting

Western blotting was performed as the method established in our laboratory [22]. After the spatial memory retention test, the animals were decapitated. The hippocampal CA3 region (where the virus infected) was rapidly removed and homogenized at 4°C using a Teflon glass homogenizer with the tissue homogenate (Tris-HCl 50 mM (pH 7.4), NaCl 150 mM, NaF 10 mM, Na₃VO₄ 1 mM, EDTA 5 mM, benzamidine 2 mM, and phenylmethylsulfonyl fluoride 1 mM). The extract was mixed with sample buffer (3:1, v/v) containing Tris-HCl 200 mM (pH 7.6), 8% sodium dodecyl sulfate, 40% glycerol, 40 mM dithiothreitol, and then boiled for 10 min. After separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, the proteins were transferred to nitrocellulose membranes. 5% nonfat milk, which dissolved in Tris buffered saline/Tween-20 (Tris-HCl 50 mM (pH 7.6), NaCl 150 mM, 0.2% Tween-20), was used to block the membranes for 1 h, and then the membranes were probed with primary antibody at 4°C overnight. Finally, the blots were incubated with anti-rabbit or anti-mouse immunoglobulin G conjugated to IRDyeTM (800CW) for 1 h at 15°C to 25°C and visualized using the Odyssey Infrared Imaging System (Licor Biosciences, Lincoln, NE).

Chromatin immunoprecipitation (ChIP) assay

The hippocampal CA3 region (where the virus infected) was rapidly removed and homogenized, then cross-linked with 1% formaldehyde and incubated with 0.125 M glycine. After centrifugation, the pellet was re-suspended in PBS and centrifugated, and then, was re-suspended in Lysis Buffer with Protease inhibitors, incubated on ice. After centrifugation, the pellet was re-suspended in breaking buffer (Tris-HCl 50 mM (pH 8.0), EDTA 1 mM, NaCl 150 mM, 1% SDS, 2% Triton X-100, protease inhibitors) and sonicated 5∼10 s, and Triton buffer was added (Tris-HCl 50 mM (pH 8.0), EDTA 1 mM, NaCl 150 mM, 0.1% Triton X-100). An aliquot was reserved as the input, and the remainder was divided to immunoprecipitate with control mouse IgG (Milipore), H4K12 or H3K14 antibody followed by incubation with protein G beads. Samples were washed three times in Triton buffer, SDS buffer was added (Tris-HCl 62.5 mM (pH 6.8), NaCl 200 mM, 2% SDS, DTT 10 mM, 2μl of proteinase K (40 mg/ml)), and then samples were vortexed and incubated at 65 °C overnight to reverse cross-linking. DNA was isolated using phenol/chloroform extraction and re-suspended in distilled H₂O. Primers used for ChIP PCR were as follows: NR2B: forward and reverse primer, 5’- CCTTAGGAAGGGGACGCTTT-3’, 5’-GGCAATTAAGGGTTGGGTTC-3’ [18]. PCR products were analyzed by 2% agarose gel electrophoresis.

Statistical analysis

The data were expressed as mean±standard deviation or mean±standard error and statistical comparisons were performed using SPSS 12.0 statistical software (SPSS Inc., Chicago, IL). Student’s unpaired t-test was used to determine the statistical significance of means.

RESULTS

Upregulation of ANP32A induced memory deficits in young wild-type mice

To verify whether ANP32A indeed plays a critical role in cognitive performance, we overexpressed ANP32A in the hippocampal CA3 region of 3-month-old C57 mice and analyzed learning and memory at 4 weeks by MWM and Barnes maze tests after the AAV injection. First, the expression of ANP32A in hippocampal CA3 subset was confirmed by direct fluorescence imaging and western blotting (Fig. 1a-c). In the MWM test, the AAV-ANP32A mice exhibited significantly longer escape latency than AAV-C mice during the 6-day training (Fig. 1d), indicating learning deficits. In memory testing, the average latency was dramatically longer in AAV-ANP32A mice (47±11.3 s) than the AAV-C mice (23±8.4 s) (Fig. 1e), and the AAV-ANP32A mice spent less time in the platform quadrant (Fig. 1f) and less frequently crossed the platform than the AAV-C mice (Fig. 1 g) with no significant differences in speed (Fig. 1 h). 1 week after the MWM test, we also analyzed the cognitive performance by Barnes maze test. The results showed that the mice with overexpression of ANP32A exhibited longer latency and less right times (the times to enter the escape box) than AAV-C mice (Fig. 1i, j). These data demonstrate the sufficiency of ANP32A upregulation in inducing learning and memory impairments.

Fig.1

ANP32A overexpression induces cognitive deficits. Adeno-associated virus coexpressing GFP and human ANP32A (AAV-ANP) or expressing GFP alone (AAV-C) (2μL, 1×10¹² TU/mL) were injected into the hippocampal CA3 region of 3-month-old C57 mice under a stereotaxic apparatus. Four weeks after injection, the mice were used for the following measurements. a) Representative images of the expression of the injected AAV virus in the hippocampal CA3 region. b, c) Western blotting analysis and quantification of ANP32A levels. Note that the injection of AAV-ANP significantly elevates ANP32A level. d) Escape latencies of AAV-C (n = 11), and AAV-ANP (n = 11) mice in the Morris water maze task. AAV-ANP32A mice exhibit longer escape latencies than the AAV-C mice, indicating the learning deficits. *p < 0.05; **p < 0.01 versus AAV-C. e-g) Latency to find the platform (e), time spent in the target quadrant (f), and times crossing the place (g) where the platform was placed during the training (n = 11 for each group). h) Swim speed of AAV-C and AAV-ANP animals in the water maze task. Note that no significant difference was observed between the AAV-C and AAV-ANP mice. i, j) Latency (i) and the right times (j) to the target hole during testing with the Barnes maze test (n = 11 mice per group). *p < 0.05, **p < 0.01 versus AAV-C (mean±SD).

Overexpressing ANP32A impaired synaptic morphology and the function in young wild-type mice

To explore the mechanisms that may underlie the effects of ANP32A, we first measured the synaptic transmission by electrophysiological recording in the acute brain slices. We observed that overexpression of ANP32A by AAV-ANP32A in 3-month-old wild-type mice significantly reduced fEPSPs slope, which was recorded from CA3 in hippocampus by stimulating mossy fibers, when compared with the AAV-C mice (Fig. 2a, b).

Fig.2

Overexpressing ANP32A impairs the synaptic plasticity. a, b) Field excitatory postsynaptic potentials (fEPSPs) slopes (a) and average slopes of fEPSPs (b) in hippocampal CA3 area of AAV-C and AAV-ANP32A animals (n = 7 to 8 slices from 4 mice each). c-e). The representative photomicrographs of spines (c) and quantitative analysis of spine density (d) or density of mushroom- or thin-shaped spines (e) in the hippocampal CA3 neurons of AAV-C or AAV-ANP animals. Scale bar = 5μm. *p < 0.05, **p < 0.01 versus AAV-C (mean±standard deviation).

The spine generation and mushroom spine formation play a crucial role in learning and memory; therefore, we measured the spine density and the morphologies in the neurons of the hippocampal CA3 region by Golgi stain. We observed that overexpression of ANP32A led to a significant decrease in spine density (Fig. 2c-e).

Overexpressing ANP32A decreased synapse-associated proteins levels in young wild-type mice

To verify the molecular mechanisms underlying the altered synapse morphology and electrophysiology, we measured the levels of synapse-associated proteins. We observed that overexpression of ANP32A in normal mice drastically reduced the mRNA and protein levels of presynaptic proteins including synaptophysin (Syp) and synapsin I (Syn1), and postsynaptic proteins including N-methyl-D-aspartate receptor type 2A (NR2A) and N-methyl-D-aspartate receptor type 2B (NR2B), and AMPA receptor subunits GluR1 and GluR2 in the hippocampal CA3 regions where AAV virus was injected (Fig. 3a-c).

Overexpressing ANP32A decreased histone acetylation

Fig.3

The synaptic-related protein expression decreased by ANP32A overexpression. a, b) The levels of synaptic proteins in the hippocampal CA3 region of AAV-C and AAV-ANP animals were detected by western blotting and quantitative analysis (n = 4 mice per group). c) The mRNA levels of synaptic proteins in the hippocampal CA3 region of AAV-C and AAV-ANP animals were detected by quantitative reverse transcription polymerase chain reaction (n = 4 mice per group). *p < 0.05, **p < 0.01 versus AAV-C (mean±standard deviation).

Histone acetylation regulates expression of synaptic proteins [18]. To investigate the mechanisms of the decreased synapse-associated proteins expression which was induced by overexpressing ANP32A, we detected the acetylated levels of histone H3 and H4. We observed that overexpression of ANP32A remarkably inhibits acetylation of H3 at K9 and K14 (H3K9, H3K14) and H4 at K5 and K12 (H4K5, H4K12) without changing the total level of H3 and H4 subunits (Fig. 4a, b). By chromatin immunoprecipitation (CHIP) assay, we also found that overexpression of ANP32A remarkably decreased binding of acetylation of H3 at K14 (H3K14) and H4 at K12 (H4K12) to the promoter of NR2B gene (Fig. 5a, b).

Fig.4

Overexpressing ANP32A decreased the level of acetylated histone. a, b) The acetylated histone levels in the hippocampal CA3 region of AAV-C and AAV-ANP mice were detected by western blotting and quantitative analysis (n = 4 mice per group). *p < 0.05, **p < 0.01 versus AAV-C (mean±standard deviation).

DISCUSSION

ANP32A, also known as phosphoprotein pp32, leucine-rich acidic nuclear protein, Inhibitor 1 of PP2A (I₁^PP2A) [23, 24], and putative histocompatibility leukocyte antigen class II-associated protein I, is a member of the acidic nuclear phosphoprotein 32 family, which consists of 32 evolutionarily conserved proteins [25]. ANP32A is highly abundant in nucleus and is also present on endoplasmic reticulum as well as cytoplasm [26, 27]. ANP32A is a multifaceted protein involved in a variety of cellular functions including tumor suppression [28 –30], cell-mediated cytotoxicity [31, 32], signal transduction [33], and regulation of microtubule function [34, 35]. We have shown that ANP32A increased in aged and cognitively impaired C57 mice and htau transgenic mice associated with decreasing histone acetylation. Downregulation of ANP32A restored cognitive function and synaptic plasticity through upregulation of the expressions of synaptic-related proteins via increasing histone acetylation [9, 10]. In the current study, we found that ANP32A overexpression induced cognitive abilities and synaptic plasticity deficits, with decreased synaptic-related proteins expression and histone acetylation. Our results indicate that the memory deficits in neurodegeneration and AD brain are likely to be mediated, at least in part, by elevated ANP32A via disturbed histone acetylation.

Fig.5

Acetylated histones binding to the coding region of NR2B are decreased by overexpressing ANP32A. a, b) Representative chromatin immunoprecipitation experiments and quantitative PCR results of (a) H3K14- and (b) H4K12-immunoprecipitated chromatin in the hippocampal CA3 region of AAV-C and AAV-ANP mice (n = 3 mice per group). *p < 0.05 versus AAV-C (mean±standard deviation).

There is accumulating evidence that epigenetic mechanisms such as altered histone acetylation are involved in cognitive aging, neurodegeneration, and AD. It is recognized that regulation of the acetylation of histone, and then altering the expression of some important genes involved in the study and memory, may be a potential therapeutic strategy for neurodegenerative diseases including AD. Acetylation and deacetylation of histone is catalyzed by histone acetyltransferases (HATs) and histone deacetylases (HDACs), respectively. Recent studies have found that HDAC6 and HDAC2 are significantly increased in cortex and hippocampus of the brains of patients with AD compared with normal controls [18, 36]. Moreover, pharmacological treatments aimed at increasing histone acetylation have shown promising results in restoring cognitive functions in neurodegenerative disease models by the use of nonselective HDAC inhibitors [18 , 38].

In addition, histone acetylation is also suppressed by a cellular complex termed INHAT [33, 34]. INHAT, a complex composed of three essential subunits, TAF-Iα, SET/TAF-Iβ, and ANP32A, binds to histones and prevents their access to HATs to be acetylated [14]. INHAT had no interaction with either p300 or p300/CBP associated factor, and histone binding is a prerequisite for INHAT and its subunits to inhibit HAT activity [14, 39]. Five major families of histones exist: H1/H5, H2A, H2B, H3, and H4. Histones H2A, H2B, H3, and H4 are known as the core histones, whereas histones H1/H5 are known as the linker histones. INHAT and its subunits have different histone binding specificities.

Histone H3 and H4 acetylation decreased in the brain affected by AD, and abnormal acetylation of histone is causally linked to the cognitive decline in AD [9 , 41]. Although ANP32A subunit has a high affinity to bind to and inhibit acetylation of histone H2B and H3, this H2B preference is lost when ANP32A is incorporated into the INHAT complex, which predominantly binds to and inhibits acetylation of histones H3 and H4 [14, 39]. In the current study, we focus on the acetylated levels of histones H3 and H4. We found that overexpression of ANP32A decreased the acetylated levels of H3K9, H3K14, H4K5, and H4K12. Acetylated histones including H3K9, H4K8, and H4K12 are found to bind to the promoter region of the synaptic genes, including glutamate receptor subunits GluR1, GluR2, NR2A, and NR2B and presynaptic proteins Syp and Syn1, and regulate their transcriptions [18, 40]. Genes such as GluR1, GluR2, NR2A and NR2B, and Syp and Syn1, which relate to synaptic plasticity, are found to downregulate in human brains affected by AD [42 –46]. ANP32A mRNA and protein level increased in the temporal and entorhinal cortices of patients with AD [15], whereas the acetylated levels of histone H3 and H4 decreased [9 , 18]. All the studies suggested that the increased ANP32A may be or is partly responsible for the decreased synaptic proteins levels, and thus attenuation of synaptic transmission and synapse density.

Our study found that overexpressing ANP32A induced memory deficits by decreasing histone acetylation and thus downregulating expression of synapse-associated proteins with impairment of synapse plasticity. Combined with our previous studies, downregulation of ANP32A may represent a promising therapeutic approach for preserving cognitive capacity.

Footnotes

ACKNOWLEDGMENTS

This work was supported by grants from the National Natural Science Foundation of China (81501211 and 81601121), Natural Science Foundation of Jiangsu Province (BK20150163) and Sanming Project of Medicine in Shenzhen.

Authors’ disclosures available online ().

References

Grundke-Iqbal

, Johnson

, Wisniewski

, Terry

, Iqbal

(1979) Evidence that Alzheimer neurofibrillary tangles originate from neurotubules. Lancet 1, 578–580.

Ballard

, Gauthier

, Corbett

, Brayne

, Aarsland

, Jones

(2011) Alzheimer’s disease. Lancet 377, 1019–1031.

Perry

, Hodges JR (1999) Attention and executive deficits in Alzheimer’s disease. A critical review. Brain 122 (Pt 3), 383–404.

(2012) Drugs for Alzheimer’s disease: Best avoided. No therapeutic advantage. Prescrire Int 21, 150.

Birks

, Grimley Evans

(2015) Rivastigmine for Alzheimer’s disease. Cochrane Database Syst Rev, CD001191.

Kosik

, Rapp

, Raz

, Small

, Sweatt

, Tsai

(2012) Mechanisms of age-related cognitive change and targets for intervention: Epigenetics. J Gerontol A Biol Sci Med Sci 67, 741–746.

Vecsey

, Hawk

, Lattal

, Stein

, Fabian

, Attner

, Cabrera

, McDonough

, Brindle

, Abel

, Wood

(2007) Histone deacetylase inhibitors enhance memory and synaptic plasticity via CREB:CBP-dependent transcriptional activation. J Neurosci 27, 6128–6140.

Jenuwein

, Allis

(2001) Translating the histone code. Science 293, 1074–1080.

Feng

, Chai

, Wang

, Hu

, Sun

, Li

, Ma

, Li

, Ke

, Wang

, Liu

(2017) Knockdown of pp32 increases histone acetylation and ameliorates cognitive deficits. Front Aging Neurosci 9, 104.

10.

Chai

, Feng

, Wang

, Hu

, Sun

, Li

, Ke

, Li

, Liu

, Wang

(2017) Downregulating ANP32A rescues synapse and memory loss via chromatin remodeling in Alzheimer model. Mol Neurodegener 12, 34.

11.

Brownell

, Allis

(1996) Special HATs for special occasions: Linking histone acetylation to chromatin assembly and gene activation. Curr Opin Genet Dev 6, 176–184.

12.

Haggarty

, Tsai

(2011) Probing the role of HDACs and mechanisms of chromatin-mediated neuroplasticity. Neurobiol Learn Mem 96, 41–52.

13.

Matilla

, Radrizzani

(2005) The Anp32 family of proteins containing leucine-rich repeats. Cerebellum 4, 7–18.

14.

Seo

, McNamara

, Heo

, Turner

, Lane

, Chakravarti

(2001) Regulation of histone acetylation and transcription by INHAT, a human cellular complex containing the set oncoprotein. Cell 104, 119–130.

15.

Tanimukai

, Grundke-Iqbal

, Iqbal

(2005) Up-regulation of inhibitors of protein phosphatase-2A in Alzheimer’s disease. Am J Pathol 166, 1761–1771.

16.

Morris

, Garrud

, Rawlins

, O’Keefe

(1982) Place navigation impaired in rats with hippocampal lesions. Nature 297, 681–683.

17.

Nagahara

, Merrill

, Coppola

, Tsukada

, Schroeder

, Shaked

, Wang

, Blesch

, Kim

, Conner

, Rockenstein

, Chao

, Koo

, Geschwind

, Masliah

, Chiba

, Tuszynski

(2009) Neuroprotective effects of brain-derived neurotrophic factor in rodent and primate models of Alzheimer’s disease. Nat Med 15, 331–337.

18.

Graff

, Rei

, Guan

, Wang

, Seo

, Hennig

, Nieland

, Fass

, Kao

, Kahn

, Su

, Samiei

, Joseph

, Haggarty

, Delalle

, Tsai

(2012) An epigenetic blockade of cognitive functions in the neurodegenerating brain. Nature 483, 222–226.

19.

Woolley

, McEwen

(1992) Estradiol mediates fluctuation in hippocampal synapse density during the estrous cycle in the adult rat. J Neurosci 12, 2549–2554.

20.

Magarinos

, Li

, Gal Toth

, Bath

, Jing

, Lee

, McEwen

(2011) Effect of brain-derived neurotrophic factor haploinsufficiency on stress-induced remodeling of hippocampal neurons. Hippocampus 21, 253–264.

21.

Spires-Jones

, Kay

, Matsouka

, Rozkalne

, Betensky

, Hyman

(2011) Calcineurin inhibition with systemic FK506 treatment increases dendritic branching and dendritic spine density in healthy adult mouse brain. Neurosci Lett 487, 260–263.

22.

Liu

, Wei

, Zhou

, Shi

, Liu

, Chai

, Yao

, Zhang

, Peng

, Hu

, Li

, Wang

(2013) Silencing PP2A inhibitor by lenti-shRNA interference ameliorates neuropathologies and memory deficits in tg2576 mice. Mol Ther 21, 2247–2257.

23.

, Makkinje

, Damuni

(1996) Molecular identification of I1PP2A, a novel potent heat-stable inhibitor protein of protein phosphatase 2A. Biochemistry 35, 6998–7002.

24.

Chen

, Li

, Grundke-Iqbal

, Iqbal

(2008) I1PP2A affects tau phosphorylation via association with the catalytic subunit of protein phosphatase 2A. J Biol Chem 283, 10513–10521.

25.

Opal

, Garcia

, McCall

, Xu

, Weeber

, Sweatt

, Orr

, Zoghbi

(2004) Generation and characterization of LANP/pp32 null mice. Mol Cell Biol 24, 3140–3149.

26.

Fan

, Beresford

, Zhang

, Xu

, Novina

, Yoshida

, Pommier

, Lieberman

(2003) Cleaving the oxidative repair protein Ape1 enhances cell death mediated by granzyme A. Nat Immunol 4, 145–153.

27.

Gallouzi

, Steitz

(2001) Delineation of mRNA export pathways by the use of cell-permeable peptides. Science 294, 1895–1901.

28.

Bai

, Brody

, Kadkol

, Pasternack

(2001) Tumor suppression and potentiation by manipulation of pp32 expression. Oncogene 20, 2153–2160.

29.

Adegbola

, Pasternack

(2005) Phosphorylated retinoblastoma protein complexes with pp32 and inhibits pp32-mediated apoptosis. J Biol Chem 280, 15497–15502.

30.

Schafer

, Parrish

, Wright

, Margolis

, Marks

, Deshmukh

, Kornbluth

(2006) Enhanced sensitivity to cytochrome c-induced apoptosis mediated by PHAPI in breast cancer cells. Cancer Res 66, 2210–2218.

31.

Beresford

, Kam

, Powers

, Lieberman

(1997) Recombinant human granzyme A binds to two putative HLA-associated proteins and cleaves one of them. Proc Natl Acad Sci U S A 94, 9285–9290.

32.

Wang

, Blanchard

, Tung

, Grundke-Iqbal

, Iqbal

(2015) Inhibition of protein phosphatase-2A (PP2A) by I1PP2A leads to hyperphosphorylation of tau, neurodegeneration, and cognitive impairment in rats. J Alzheimers Dis 45, 423–435.

33.

Vaesen

, Barnikol-Watanabe

, Gotz

, Awni

, Cole

, Zimmermann

, Kratzin

, Hilschmann

(1994) Purification and characterization of two putative HLA class II associated proteins: PHAPI and PHAPII. Biol Chem Hoppe Seyler 375, 113–126.

34.

Ulitzur

, Humbert

, Pfeffer SR (1997)Proc Natl Acad Sci U S A 94, 5084–5089 Mapmodul, Aossible modulator of the interaction of microtubule-associated proteins with microtubules.

35.

Ulitzur

, Rancano

, Pfeffer SR (1997) Biochemical characterization of mapmodulin, a protein that binds microtubule-associated proteins. J Biol Chem 272, 30577–30582.

36.

Ding

, Dolan

, Johnson

(2008) Histone deacetylase 6 interacts with the microtubule-associated protein tau. J Neurochem 106, 2119–2130.

37.

Govindarajan

, Rao

, Burkhardt

, Sananbenesi

, Schluter

, Bradke

, Lu

, Fischer

(2013) Reducing HDAC6 ameliorates cognitive deficits in a mouse model for Alzheimer’s disease. EMBO Mol Med 5, 52–63.

38.

, Dai

, Huang

, Jiang

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

39.

Seo

, Macfarlan

, McNamara

, Hong

, Mukai

, Heo

, Chakravarti

(2002) Regulation of histone acetylation and transcription by nuclear protein pp32, a subunit of the INHAT complex. J Biol Chem 277, 14005–14010.

40.

Graff

, Kim

, Dobbin

, Tsai

(2011) Epigenetic regulation of gene expression in physiological and pathological brain processes. Physiol Rev 91, 603–649.

41.

Peleg

, Sananbenesi

, Zovoilis

, Burkhardt

, Bahari-Javan

, Agis-Balboa

, Cota

, Wittnam

, Gogol-Doering

, Opitz

, Salinas-Riester

, Dettenhofer

, Kang

, Farinelli

, Chen

, Fischer

(2010) Altered histone acetylation is associated with age-dependent memory impairment in mice. Science 328, 753–756.

42.

Wakabayashi

, Narisawa-Saito

, Iwakura

, Arai

, Ikeda

, Takahashi

, Nawa

(1999) Phenotypic down-regulation of glutamate receptor subunit GluR1 in Alzheimer’s disease. Neurobiol Aging 20, 287–295.

43.

Counts

, He

, Che

, Ginsberg

, Mufson

(2009) Galanin fiber hyperinnervation preserves neuroprotective gene expression in cholinergic basal forebrain neurons in Alzheimer’s disease. J Alzheimers Dis 18, 885–896.

44.

, Sze

(2002) N-methyl-D-aspartate receptor subunit NR2A and NR2B messenger RNA levels are altered in the hippocampus and entorhinal cortex in Alzheimer’s disease. J Neurol Sci 200, 11–18.

45.

Callahan

, Vaules

, Coleman

(1999) Quantitative decrease in synaptophysin message expression and increase in cathepsin D message expression in Alzheimer disease neurons containing neurofibrillary tangles. J Neuropathol Exp Neurol 58, 275–287.

46.

Mufson

, Counts

, Ginsberg

(2002) Gene expression profiles of cholinergic nucleus basalis neurons in Alzheimer’s disease. Neurochem Res 27, 1035–1048.