Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Neurosci. Author manuscript; available in PMC 2012 June 7.
Published in final edited form as:
PMCID: PMC3278324

This article has been retractedRetraction in: J Neurosci. 2013 December 11; 33(50): 19735    See also: PMC Retraction Policy

Epigenetic enhancement of BDNF signaling rescues synaptic plasticity in aging


Aging-related cognitive declines are well documented in humans and animal models. Yet the synaptic and molecular mechanisms responsible for cognitive aging are not well understood. Here we demonstrated age-dependent deficits in long-term synaptic plasticity and loss of dendritic spines in the hippocampus of aged Fisher 344 rats, which were closely associated with reduced histone acetylation, upregulation of histone deacetylase 2 (HDAC2), and decreased expression of a histone acetyltransferase. Further analysis showed that one of the key genes affected by such changes was the brain-derived neurotrophic factor (Bdnf) gene. Age-dependent reductions in H3 and H4 acetylation were detected within multiple promoter regions of the Bdnf gene, leading to a significant decrease in BDNF expression and impairment of downstream signaling in the aged hippocampus. These synaptic and signaling deficits could be rescued by enhancing BDNF and trkB expression via HDAC inhibition or by directly activating trkB receptors with 7, 8-dihydroxyflavone, a newly identified, selective agonist for trkB. Taken together, our findings suggest that age-dependent declines in chromatin histone acetylation and the resulting changes in BDNF expression and signaling are key mechanisms underlying the deterioration of synaptic function and structure in the aging brain. Furthermore, epigenetic or pharmacological enhancement of BDNF-trkB signaling could be a promising strategy for reversing cognitive aging.


Brain aging is marked by a gradual decline in cognitive function, often linked with age-dependent deterioration of synaptic function in brain regions crucial for memory formation and consolidation such as the hippocampus and prefrontal cortex (Lynch et al., 2006; Tapia-Arancibia et al., 2008). Aging is also known to significantly increase the vulnerability of neural cuicuits to Alzheimer’s disease in the same brain regions (Kern and Behl, 2009). Yet the molecular mechanisms underlying aging-related neural and synaptic vulnerability are mostly unknown. Recent evidence indicates that chromatin remodeling via histone acetylation plays a crucial role in regulating synaptic and cognitive function (Levenson and Sweatt, 2005; Haggarty and Tsai, 2011). Increasing histone acetylation by inhibition of histone deacetylase (HDAC) enhances gene transcription and improves hippocampal long-term potentiation (LTP) or memory function in several experimental models of neurological diseases (Fischer et al., 2007; Vecsey et al., 2007; Francis et al., 2009), indicating involvement of chromatin acetylation dysregulation in certain forms of cognitive impairments. However, given the broad effect of the HDAC inhibitor on gene transcription, further studies are in need to determine the precise sites of histone acetylation alterations, key genes affected, and associated signaling changes involved in this regulatory mechanism. Moreover, the previous studies were mostly conducted in young rodents; few have addressed this issue in the context of cognitive aging.

The present study demonstrates that the aging process results in reduced acetylation of H3 and H4 in the promoter regions of the brain derived neurotrophic factor (Bdnf) gene. BDNF is a trkB-acting neurotrophin that provides essential support for long-term synaptic plasticity in adult hippocampus (Korte et al., 1995; Patterson et al., 1996; Pang et al., 2004). Our findings show that age-dependent declines in histone acetylation lead to reduced BDNF expression and impaired activity of key signaling pathways for synaptic plasticity. Such changes appear to contribute substantially to aging-related deficits in hippocampal synaptic function and structure, which can be rescued by HDAC inhibition or by selective activation of trkB receptors.

Materials and Methods

Animals, brain slice preparation, and drug treatment

Male Fischer 344 (F344) rats were divided into three age groups: young (1–2 months old), middle-aged (12–14 months old), and aged (22–23 months old). Hippocampal slices 500 µm thick were prepared as previously described (Xie et al., 2000). After recovering for 1–2 hr in a holding chamber containing oxygenated artificial cerebrospinal fluid (ACSF), slices were treated with an HDAC inhibitor or vehicle for 3 hr and subsequently used for various assays. The HDAC inhibitor trichostatin (TSA) was purchased from A.G. Scientific (San Diego, CA) and sodium butyrate (SB) from Sigma-Aldrich (St. Louis, MO). The trkB agonist 7, 8-dihydroxyflavone (7, 8-DHF) was purchased from TCI America (Portland, OR). TSA was dissolved in ethanol as stock solution and diluted with ACSF before application with a final ethanol concentration of 0.06% or less. 7, 8-DHF was dissolved in dimethyl sulfoxide (DMSO) and diluted before application with a final DMSO concentration of 0.02%. ACSF containing the same concentration of ethanol or DMSO served as vehicle controls when appropriate. To determine age-dependent and drug-induced changes, rats of different age groups were included in the same set of experiments, and multiple slices from the same animal were used for different drug treatments.

Electrophysiological recording

Hippocampal slices were placed in a submerged recording chamber and perfused with 29–30°C oxygenated ACSF at 2–3 ml/min. Field excitatory postsynaptic potentials (fEPSPs) were recorded from the Schaffer collateral-CA1 pathway using standard extracellular recording techniques as previously described (Xie et al., 2000). Basal synaptic transmission was assessed using the synaptic input-output curve constructed by plotting fEPSP slopes vs. presynaptic fiber volley amplitudes evoked by an incremental sequence of stimulus intensities (40 – 600 µA). LTP was induced using a single high frequency stimulation (HFS) at 100 Hz for 1 sec and was expressed as the percent change of the fEPSP slope from its baseline value. The average amount of LTP recorded during the period of 45–60 min after the HFS was used for inter-group comparisons.

Western blot analysis

The protein levels in hippocampal lysates were measured using a previously described protocol (Zeng et al., 2010) with modifications. Hippocampal samples were collected and homogenized in 50 mM Tris/HCl buffer containing Protease Inhibitors Complete (Roche Molecular Biochemicals, Indianapolis, IN) and Protein Phosphatase Inhibitor Cocktail I and II (Sigma-Aldrich). Sample aliquots containing 20 µg protein each were electrophoresed on 12% SDS/PAGE gels. The proteins were transferred onto low-fluorescent PVDF membranes, probed overnight at 4°C with the primary antibody, and incubated for another 2 h with a HRP-conjugated secondary antibody. The immunoreactive bands were visualized with ECL-plus fluorescence and quantified with Imagequant 5.2 software (GE Healthcare). All blots were reprobed with a β-actin antibody and the signal value was normalized to that of β-actin. The percent change in the protein of interest was calculated relative to that of vehicle-treated young controls run in the same experiments. The primary antibodies used to detect acetyl-H3K9, acetyl-H4K12, H3, H4, HDAC1, HDAC2, Pro-BDNF, trkB, and phospho-Ser831-GluR1 were from Abcam (Cambridge, MA). The antibodies for BDNF were from Abcam and Santa Cruz Biotechnology (Santa Cruz, CA). The antibodies for CREB binding protein (CBP), extracellular signal-regulated kinase 1/2 (Erk1/2), phospho-Erk1/2, Ca2+/calmodulin-dependent protein kinase II (CaMKII), and GluR1 were from Millipore (Billerica, MA), and the antibody for phospho-Thr286-CaMKII was from Thermo Scientific (Rockford, IL).

Enzyme-linked immunosorbent assay (Elisa)

The BDNF level in rat hippocampal tissues was quantified using a ChemiKine BDNF sandwich ELISA kit from Millipore following the manufacturer’s instructions. Hippocampi were dissected, flash-frozen in liquid nitrogen, and stored at −80°C before use. The frozen tissues were homogenized and centrifuged at 14,000×g for 30 minutes. The resulting supernatants were applied to microplates pre-coated with a rabbit anti-BDNF polyclonal antibody and incubated at 4°C overnight. After washes, a biotinylated, mouse anti-BDNF monoclonal antibody was added for incubation for 2 hr at room temperature. The streptavidin-HRP conjugate solution, TMB/E substrate and stop solution were then added sequentially. The amount of BDNF was determined by the optical density reading at 450 nm and expressed as the ratio of BDNF to the total soluble protein content (pg/mg).

Chromatin immunoprecipitation (ChIP)

ChIP was performed with antibodies against acetyl-H3K9/27 (Cosmo Bio) or acetyl-H4K5/8/12/16 (Upstate) as described previously (Kimura et al., 2008). Hippocampal slices were cut into 2 mm pieces, cross-linked with 1% formaldehyde, and stored in −80°C before use. Nuclei were extracted, lysed, and sonicated using a microtip until the DNA fragments were reduced to 200–1000 bp in length. Dynal protein-G beads (40 µl, Invitrogen) were preincubated with 3 µg antibody for at least 6 hr. Immunoprecipitation was then performed overnight at 4°C with antibody-conjugated protein-G beads. DNA/protein complexes were eluted from the beads and reverse cross-linked at 65 °C overnight. The DNA was digested with RNase A and proteinase K, followed by phenol/chloroform extraction and ethanol precipitation. Real-time quantitative PCR (q-PCR) was performed in an iCycler using iQ SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA). The amount of immnoprecipitated DNA was calculated in comparison to the total input DNA. Data were derived from four independent amplifications. The values of the control samples obtained from vehicle-treated young rats were set to 1.0, and others were expressed as fold changes relative to the young controls. The primer sequences used were as follows:


Golgi staining and dendritic spine analysis

Golgi staining was performed using a FD Rapid Golgi Stain Kit (FD NeuroTechnologies, Ellicott City, MD) following the manufacturer’s protocol. Hippocampal slices were immersed in an impregnation solution for 1 week at room temperature, transferred into solution C for 24 hr at 4°C, and cut into 150 µm sections using a Leica Vibratome. Bright-field images of CA1 pyramidal neurons and their apical dendrites in stratum radiatum were taken at 100X magnification using a Zeiss Axioplan 2 microscpoe. Dendritic spines, which were identified as small protrusions extending ≤3.0 µm from their parent dendrites, were counted using Neurolucida software 9.0 (MicroBrightField, Williston, VT). Segment-dendrite analysis was used to quantify the number of spines on apical proximal dendrites arising from the soma and within 50–100 µm from the center of the soma, and data were expressed as the number of spines per 10 µm segment. To determine relative spine density, spines on multiple dendritic branches from a single neuron were counted to obtain the average spine number per 10 µm, and these values were expressed as percentages relative to vehicle-treated young controls. Analysis was conducted blind to sample identity on batches of slices that were sectioned and processed in the same experiments. Twelve neurons in each rat and 5 rats per group were analyzed.

Immunohistochemical staining

Hippocampal slices were fixed using 4% paraformaldehyde, cryoprotected with 30% sucrose, and cut into 40 µm sections. Cells were permeabilized with 0.1% Triton X-100 and incubated with a rabbit anti-synaptophysin polyclonal antibody (Epitomics, Burlingame, CA) followed by a Cy3-conjugated secondary antibody. Fluorescent images of synaptophysin labeling were acquired using a Zeiss LSM510 confocal microscope or an ultra-sensitive Andor iXon EMCCD camera. The images were processed using the ImageJ software (NIH). The optical density was calculated as the mean pixel intensity from a 400 µm2 sampling zone in CA1 stratum radiatum and used as an estimate of synapse density.

Statistical analysis

Data are presented as group means ± s.e.m. Statistical analysis was performed using two-sample Student’s t-test or one-way ANOVA followed by Tukey post hoc test when appropriate. Two-way ANOVA was used to compare the drug effect at different ages. Statistical significance was defined as P<0.05.


Age-dependent LTP decline is reversed by HDAC inhibitors in a BDNF-dependent manner

Hippocampal LTP is widely regarded as a form of activity-dependent synaptic plasticity underlying learning and memory. To determine age-dependent changes in synaptic function, we compared basal synaptic transmission and LTP induction at Schaffer collateral-CA1 synapses in brain slices derived from young, middle-aged, and aged F344 rats. The strength of basal synaptic transmission appeared unaffected by age (Fig. 1A). However, HFS-induced LTP was reduced in an age-dependent manner (Fig.1B, H). The LTP decline in aged slices was prevented by pretreatment with the HDAC inhibitor TSA (2 µm) or SB (2.5 mM) for 3 hr before LTP induction (Fig. 1D, E, H). TSA treatment also enhanced LTP in young slices but to a lesser extent (Fig. 1C, H). Two-way ANOVA revealed the significant main effect of age (F(2, 31) = 222.2, p<0.01) and drug treatment (F(1, 31) = 359.2, p<0.01) on average LTP, as well as a significant interaction between these two factors (F(2, 31) = 22.1, p<0.01). Importantly, TSA-induced LTP enhancement was blocked by the transcription inhibitor actinomycin D (Act-D) or by a trkB-IgG fusion protein that sequesters BDNF and blocks its activity (Fig. 1F, G). These results suggest that age-dependent LTP decline can be reversed by HDAC inhibition through transcription- and BDNF-dependent mechanisms.

Figure 1
HDAC inhibition reverses age-dependent decline in hippocampal LTP. A. The basal synaptic transmission at Schaffer collateral-CA1 synapses, measured by the synaptic input-output curve, was not affected by age. B. HFS-induced LTP was reduced in brain slices ...

Age-dependent changes in histone acetylation, HDAC2 and CBP

Next we investigated whether alterations in chromatin histone acetylation contributed to the LTP decline with age. Western blot analysis showed that acetylation of histone H3 at lysine 9 (K9) and H4 at lysine 12 (K12) decreased significantly with age in the hippocampus (Acetyl-H3K9: F (2, 90) = 9.9, p<0.01; Acetyl-H4K12: F (2, 90) = 12.7, p<0.01). TSA pretreatment that prevented the LTP decline also effectively restored acetyl-H3K9 and acetyl-H4K12 levels in the aging hippocampus (Fig. 2A, B, D). The total H3 and H4 levels, on the other hand, were affected neither by age nor TSA treatment (p>0.05 for both). As histone acetylation is catalyzed by histone acetyltransferases (HAT) and reversed by HDAC, any significant changes in the HAT/HDAC balance could alter the cellular level of acetylated histones. We therefore examined the effect of aging on HDAC and HAT expression in the hippocampus (Fig. 2C, E). Compared to the young controls, aged hippocampal tissues expressed a significantly higher level of HDAC2 and lower level of CREB-binding protein (CBP), which is a transcriptional coactivator with intrinsic HAT activity (Ogryzko et al., 1996). Expression of HDAC1, another member of HDAC family, was not altered by aging. These results suggest that a combination of increased HDAC2 and reduced HAT levels may account for acetyl-histone deficits in the aged brain. While not affecting expression of HDACs, TSA treatment unexpectedly raised CBP levels in the aged but not young slices (Fig. 2C, E). Thus the consequence of elevated HDAC2 expression in the aged brain appears to be two-fold, both promoting histone deacetylation and limiting HAT-mediated acetylation through suppression of CBP expression.

Figure 2
Age-dependent histone acetylation deficits. A and B. The hippocampal levels of acetyl-H3K9 (Ac-H3K9) and acetyl-H4K12 (Ac-H4K12) were reduced with age but restored by TSA treatment. C. HDAC2 and CREB binding protein (CBP) levels were altered by aging. ...

Histone acetylation deficits lead to reduced BDNF and trkB expression

Chromatin remodeling by histone acetylation plays an important role in the regulation of Bdnf gene transcription (Martinowich et al., 2003; Aid et al., 2007). We hypothesize that aging-related deficits in H3 and H4 acetylation can lead to repression of BDNF expression and hence impairment of synaptic plasticity. Previous studies on BDNF expression in aged rats have yielded mixed results, depending upon the strain, brain region, and assay used (Tapia-Arancibia et al., 2008). Using both Western blot analysis and ELISA, we detected age-dependent reductions in hippocampal BDNF levels in male F344 rats (Fig. 3A, B), which were accompanied by decreases in the BDNF precursor protein, pro-BDNF, and trkB receptors (Fig. 3C, D). TSA or SB treatment in the aged slices restored the expression of these proteins to the levels comparable to those of young controls (Fig. 3A–F).

Figure 3
BDNF expression is reduced with age but enhanced by HDAC inhibition. A–E. Age-dependent reductions in pro-BDNF, BDNF and trkB levels in hippocampal lysates, reversible by a 3-hr treatment with either TSA (0.66 or 2 µM) or SB (2.5 mM). ...

Transcription of the rat Bdnf gene is controlled by a set of 9 distinct promoters regulated by neuronal activity, epigenetic modifications, and several transcription factors (Aid et al., 2007; Tapia-Arancibia et al., 2008). To establish the specific pattern of changes in histone acetylation-regulated BDNF transcription during aging, we examined acetylation of H3 and H4 within individual Bdnf gene promoter regions in hippocampal tissues (Table 1). ChIP followed by q-PCR detected marked reductions in H3 and H4 acetylation in multiple Bdnf promoter regions in aged hippocampal tissues (Fig. 3G), which was consistent with age-dependent reductions in BDNF expression. TSA treatment (2 µM, 3 hr) in the aged slices significantly increased H3 acetylation in 3 out of 4 promoter regions affected by aging (P2, P4, and P8) and also increased H4 acetylation in the P4 region. The effect of TSA was less consistent in tissues from young rats, showing a trend to increase H4 but reduce H3 acetylation (Fig. 3G). Together these findings support a causative link between histone acetylation deficits and BDNF downregulation during aging. It is also indicated that BDNF expression can be enhanced by HDAC inhibition in a promoter-selective and age-dependent manner.

Table 1
Histone acetylation levels in the Bdnf promoter regions

HDAC inhibition activates BDNF-regulated signaling pathways

To better understand how the reduced BDNF expression could contribute to aging-related synaptic plasticity decline, we examined age-dependent changes in two key components of BDNF-regulated signaling cascades, CaMKII and Erk. Both kinases can be activated by BDNF and play crucial roles in LTP and learning (Minichiello, 2009; Zeng et al., 2010). Although total CaMKII and Erk2 levels were not affected by aging or TSA treatment, the levels of phosphorylated, highly active forms of the kinases (pCaMKII and pErk2) were significantly reduced with aging and could be restored by TSA treatment (Fig. 4A, C). Consistent with the change in CaMKII activity, phosphorylation of AMPA receptors at a CaMKII-dependent site (Ser831-GluR1) was reduced with age but enhanced by TSA (Fig. 4B). Furthermore, TSA-induced increases in p-CaMKII and pErk2 levels were prevented by co-treatment with Act-D or trkB-IgG (Fig. 4D–F), suggesting that gene transcription and BDNF function were required to restore the activity of these two kinases in the aged rats. These results confirm that aging-related changes in the BDNF-trkB system leads to reduced activity of downstream signaling cascades that are crucial for synaptic plasticity. Interestingly, application of Act-D or trkB-IgG also blocked TSA-induced increases in histone acetylation and expression of BDNF and trkB (Fig. 4D–F), suggesting the presence of a positive feedback loop where enhancement of BDNF expression by HDAC inhibition can further facilitate histone acetylation and BDNF-trkB signaling (Fig. 7).

Figure 4
HDAC inhibition enhances the activity of BDNF-regulated signaling pathways. A–C, TSA treatment (0.6 or 2 µM, 3 h) increased phosphorylation of CaMKII, Erk2, and GluR1, but not their total protein levels in the aged hippocampus. D, E, Representative ...
Figure 7
A proposed model for the interactions between aging, histone acetylation and BDNF signaling in regulating hippocampal synaptic plasticity. Age-dependent changes in the HAT/HDAC balance results in histone acetylation deficits, which lead to down-regulation ...

A selective trkB agonist mimics the effect of HDAC inhibitors

If indeed BDNF downregulation is a major cause of aging-related synaptic dysfunction, one would expect improvement of synaptic plasticity by exogenously applied BDNF or its functional mimetics in aged animals. We therefore tested the effect of 7, 8-DHF, a selective, small-molecule trkB agonist (Jang et al., 2010). As shown in Figure 5A, acute perfusion of aged hippocampal slices with 10 µM 7, 8-DHF effectively rescued HFS-induced LTP without affecting baseline synaptic transmission, and this effect could be completely blocked by the trkB inhibitor K252a (Tapley et al., 1992). Similar to previously described actions of BDNF (Pang et al., 2004), 7, 8-DHF perfusion did not significantly enhance normal LTP induced in the young slices (data not shown). Closely resembling the action of TSA, 7, 8-DHF treatment increased phosphorylation of CaMKII, GluR1 and Erk, but not their total protein levels (Fig. 5B–D, F). Application of 7, 8-DHF also increased histone acetylation and BDNF levels in the aged slices (Fig. 5E, F), further confirming the bidirectional positive interactions between histone acetylation and BDNF-trkB signaling.

Figure 5
The selective trkB agonist 7, 8-DHF mimics the effect of HDAC inhibitors. A. 7, 8-DHF perfusion (10 µM), as indicated by the horizontal bar, enhanced CA1 LTP without affecting basal synaptic transmission in the aged slices. This effect was blocked ...

HDAC inhibition or trkB activation increases dendritic spine number and synapse density

Last, we examined whether age-dependent changes in histone acetylation-regulated BDNF expression could affect structural plasticity of hippocampal neurons. Dendritic spines serve as the postsynaptic structure in 90% of excitatory synapses in the brain. Actin cytoskeleton that supports dendritic spines can undergo dynamic reorganization during neuronal activity, changing the number and shape of the spines (Johnson and Ouimet, 2006). Such plastic changes modulate synapse formation and contribute to LTP consolidation (Rex et al., 2007). We compared dendritic spine number and synapse density in the hippocampal CA1 region of young and aged rats. Dendritic spines along individual apical dendrites of CA1 pyramidal neurons were visualized using Golgi staining (Fig. 6A). Quantitative analysis (Fig. 6B, C) revealed age-dependent reductions in spine number and density. Incubation of aged slices with TSA increased the spine number in a BDNF-dependent manner, which was reversed by co-treatment with trkB-IgG. Incubation with BDNF or 7, 8-DHF also significantly increased the spine number in aged slices. Their effects were blocked by the trkB inhibitor K252a. Immunohistochemical staining of synaptophysin, a presynaptic marker of functional synapses, confirmed that synapse density was diminished in the CA1 apical dendritic layer but restored by TSA treatment in the aged hippocampus (Fig. 6D, E). These results suggest that aging-related decline in functional plasticity is accompanied by dendritic spine loss and hence reduced synapse formation. Similar to the functional impairment, the structural deficit in the aged hippocampus can be rescued by HDAC inhibition or trkB activation.

Figure 6
HDAC inhibition or trkB activation restores dendritic spine number and synapse density in the aged hippocampus. A. Images of Golgi staining of dendritic spines in the CA1 region of young (Y) and aged (A) rats. Scale bar, 10 µm. K: K252a, 0.2 µM. ...


Chromatin remodeling via histone acetylation is a key mechanism to control gene transcription. Acetylation of histones at lysine residues by HATs relaxes the chromatin structure, allowing recruitment and initiation of transcriptional machinery, whereas deacetylation of histones by HDACs is generally associated with transcriptional repression (Roth and Sweatt, 2009). Previous studies, mostly conducted in young rodents, have implicated these processes in the regulation of synaptic plasticity and memory under both physiological and pathological conditions (Levenson et al., 2004; Fischer et al., 2007; Vecsey et al., 2007; Francis et al., 2009). The present study revealed age-dependent declines in histone acetylation, coupled with LTP deficits and loss of dendritic spines in the hippocampus of aged F344 rats. Consistent with previous findings that synaptic plasticity and memory begin to decline well in advance of old age (Granger et al., 1996; Park et al., 2002; Lynch et al., 2006), deficits in both LTP and histone acetylation were evident in the middle-aged and further advanced in the aged rats. Moreover, we observed elevated HDAC2 levels and reduced expression of CBP, a known HAT, in the aged brain. Application of TSA or SB, both blocking Class I HDACs including HDAC2, restored histone acetylation and concurrently enhanced hippocampal LTP and spine density in the aged rats. These findings clearly indicate that a shift in the HAT/HDAC balance in favor of HDAC2 activity may account for age-dependent histone acetylation deficits and subsequent changes in synaptic function and structure. HDAC2 is known to negatively regulate memory formation and is likely the most relevant isoform for HDAC inhibitor-induced memory enhancement (Salisbury and Cravatt, 2007; Guan et al., 2009). Hence aging-related HDAC2 upregulation and histone acetylation deficits may represent a compromised epigenetic environment which can ultimately reduce the “plastic” capacity of the aging brain for learning and memory. Indeed a recent report has shown that behavioral learning-induced H4K12 acetylation and gene transcription are impaired in the hippocampus of 16-month-old C57BL/6 mice (Peleg et al., 2010). Interestingly, the basal levels of HDAC activity and histone acetylation remain relatively normal in these mice. It is most likely that aging-related changes in histone acetylation can occur at the basal and/or regulatory level depending upon the advancing in age, species or strain differences.

In addition to histones, many other cellular proteins are subject to posttranslational modifications via acetylation (Polevoda and Sherman, 2002). Notably, acetylation of nuclear factor-κB, a transcription factor, improves formation of long-term fear memory (Yeh et al., 2004). Acetylation of α-tubulin promotes axonal transport, dendrite extension and arborization (Janke and Kneussel, 2010). Selective inhibition of HDAC6, a major tubulin deacetylase (Hubbert et al., 2002), fails to enhance memory formation nevertheless (Guan et al., 2009), suggesting that tubulin acetylation may not play a significant role in cognitive regulation. Further studies are needed to sort out the potential contribution of non-histone substrates in aging- or HDAC-induced synaptic and cognitive alterations.

Transcriptional activity promoted by histone acetylation regulates a wide range of cellular activity. While HDAC inhibition has emerged as a promising strategy for enhancing cognitive function, clinical application of currently available, broad-acting HDAC inhibitors may be limited by undesirable side-effects (Prince et al., 2009). An alternative approach is to identify the key genes involved in cognitive enhancement by HDAC inhibitors and develop specific strategies targeting such genes or their associated signaling pathways. To this end, we have identified the Bdnf gene as a major target for aging-related transcriptional changes. The rodent Bdnf gene consists of eight 5’ noncoding exons linked to separate promoters and one 3’ exon encoding the BDNF protein. The multiple Bdnf promoters are subject to differential regulation, providing precise regulation of BDNF production in an activity-dependent and tissue-specific manner (Metsis et al., 1993; Timmusk et al., 1993). Promoter I, II and IV (P1, P2, and P4), the predominant promoters used in the rat brain, can be activated by neuronal activity and are sensitive to modulation through histone deacetylation (Tao et al., 2002; Martinowich et al., 2003; Aid et al., 2007). These regulatory mechanisms may be compromised during aging, as we detected significant reductions in H3 and H4 acetylation in multiple Bdnf promoter regions as well as decreased pro-BDNF and BDNF levels in aged hippocampal tissues. Given the crucial role of activity-dependent BDNF production in supporting hippocampal synaptic plasticity (Thoenen, 1995; Lu, 2003), disruption of this mechanism could be a major contributor to aging-related plasticity declines. This notion has been substantiated by our finding that HDAC inhibition rescued LTP in aged rats in a transcription-and BDNF-dependent manner.

BDNF binds to trkB receptors to activate several intracellular signaling cascades involving Erk, CaMKII, phosphatidylinositol 3-kinase (PI3K), and phospholipase C (Minichiello, 2009; Zeng et al., 2010). Our data show that the impact of BDNF downregulation could be exacerbated by a concurrent reduction in trkB expression in aged rats. The combined effect impaired the activation of the downstream signaling pathways, as reflected by reduced pCaMKII and pErk levels in hippocampal neurons. The single-HFS protocol used in this study normally induces the early-phase LTP, in which CaMKII plays an indispensible role (Silva et al., 1992). Activation of CaMKII enhances AMPA receptor function (Derkach et al., 1999) and trafficking (Hayashi et al., 2000), promotes activity-dependent growth of dendritic spines and formation of new synapses (Jourdain et al., 2003), and increases neurotransmitter release (Benfenati et al., 1992). Impairment of these CaMKII-dependent mechanisms could lead to the decline in early LTP seen in aged rats. Conversely, activation of CaMKII via increased BDNF production may be partially responsible for enhancement of LTP and synapse formation by HDAC inhibitors. Furthermore, Erk has well established roles in the late-phase LTP, a more persistent, transcription- and translation-dependent form of synaptic plasticity (Impey et al., 1998; Kelleher et al., 2004). Reminiscent of late LTP, TSA treatment in the aged slices enabled induction of a LTP component sensitive to the transcription inhibitor Act-D. Erk is known to phosphorylate CBP and enhance its HAT activity (Ait-Si-Ali et al., 1999). Activation of Erk during learning tasks increases H3 acetylation in the hippocampus (Levenson et al., 2004) and insular cortex (Swank and Sweatt, 2001). We thus speculate that Erk serves as a vital link in a positive feedback loop for BDNF expression, where an increase in BDNF production by HDAC inhibition leads to Erk activation which in turn further increases histone acetylation and BDNF transcription, thereby effectively rescuing synaptic plasticity (Fig. 7).

Because of its neurotrophic actions, BDNF has drawn intense interests for its therapeutic potentials in brain injury and neurodegenerative diseases. However, clinical trials using recombinant BDNF have so far yielded disappointing results (Ochs et al., 2000), possibly caused by poor delivery, short half-life, and other pharmacokinetic limitations of the BDNF protein. In the present study the selective trkB agonist 7, 8-DHF mimicked the effect of HDAC inhibitors and BDNF, rescuing synaptic plasticity and dendritic spine density in the aged rats. 7, 8-DHF is a flavonoid compound with high affinity for trkB receptors and proven to activate downstream signaling both in vitro and in vivo (Jang et al., 2010). The effectiveness of 7, 8-DHF in the aged animals despite their reduced trkB levels suggests that strong activation of the remaining trkB by this agonist could compensate for the reduced receptor level in our preparations. 7, 8-DHF treatment not only activated downstream signaling but also restored histone acetylation in the aged brain, further supporting the notion that engaging the bidirectional positive interactions between BDNF-trkB signaling and histone acetylation can be an effective approach to improve synaptic plasticity during aging. Different from BDNF, 7, 8-DHF and its derivatives are small molecules that can readily penetrate the blood-brain barrier and are orally bioavailable (Liu et al., 2010). Such advantages may confer these compounds promising therapeutic potentials for clinical applications. Evidence in vivo has shown neuroprotective actions of 7, 8-DHF in experimental models of excitotoxicity, stroke and Parkinson disease (Jang et al., 2010). More recently, we have observed significant improvement of spatial memory following systemic application of 7, 8-DHF in a transgenic mouse model of Alzheimer’s disease (Y. Zeng, et al., unpublished observations). In light of the present findings, 7, 8-DHF and other functional mimetics of BDNF could emerge as a new class of drugs for cognitive aging and aging-related neurodegenerative diseases.


This study was supported by National Institute of Health grants R01AG017542 and P30AG028748 to C.W.X. and R21MH089518 to Y.E.S. J.K. is a Research Fellow supported by the Japan Society for the Promotion of Science. We thank Xin Liu and Mochtar Pribadi for help with ELISA and Golgi staining.


Conflict of interests: None


  • Aid T, Kazantseva A, Piirsoo M, Palm K, Timmusk T. Mouse and rat BDNF gene structure and expression revisited. J Neurosci Res. 2007;85:525–535. [PMC free article] [PubMed]
  • Ait-Si-Ali S, Carlisi D, Ramirez S, Upegui-Gonzalez LC, Duquet A, Robin P, Rudkin B, Harel-Bellan A, Trouche D. Phosphorylation by p44 MAP Kinase/ERK1 stimulates CBP histone acetyl transferase activity in vitro. Biochem Biophys Res Commun. 1999;262:157–162. [PubMed]
  • Benfenati F, Valtorta F, Rubenstein JL, Gorelick FS, Greengard P, Czernik AJ. Synaptic vesicle-associated Ca2+/calmodulin-dependent protein kinase II is a binding protein for synapsin I. Nature. 1992;359:417–420. [PubMed]
  • Derkach V, Barria A, Soderling TR. Ca2+/calmodulin-kinase II enhances channel conductance of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate type glutamate receptors. Proc Natl Acad Sci U S A. 1999;96:3269–3274. [PubMed]
  • Fischer A, Sananbenesi F, Wang X, Dobbin M, Tsai LH. Recovery of learning and memory is associated with chromatin remodelling. Nature. 2007;447:178–182. [PubMed]
  • Francis YI, Fa M, Ashraf H, Zhang H, Staniszewski A, Latchman DS, Arancio O. Dysregulation of histone acetylation in the APP/PS1 mouse model of Alzheimer's disease. J Alzheimers Dis. 2009;18:131–139. [PubMed]
  • Granger R, Deadwyler S, Davis M, Moskovitz B, Kessler M, Rogers G, Lynch G. Facilitation of glutamate receptors reverses an age-associated memory impairment in rats. Synapse. 1996;22:332–337. [PubMed]
  • Guan JS, Haggarty SJ, Giacometti E, Dannenberg JH, Joseph N, Gao J, Nieland TJ, Zhou Y, Wang X, Mazitschek R, Bradner JE, DePinho RA, Jaenisch R, Tsai LH. HDAC2 negatively regulates memory formation and synaptic plasticity. Nature. 2009;459:55–60. [PMC free article] [PubMed]
  • Haggarty SJ, Tsai LH. Probing the role of HDACs and mechanisms of chromatin-mediated neuroplasticity. Neurobiol Learn Mem. 2011;96:41–52. [PMC free article] [PubMed]
  • Hayashi Y, Shi SH, Esteban JA, Piccini A, Poncer JC, Malinow R. Driving AMPA receptors into synapses by LTP and CaMKII: requirement for GluR1 and PDZ domain interaction. Science. 2000;287:2262–2267. [PubMed]
  • Hubbert C, Guardiola A, Shao R, Kawaguchi Y, Ito A, Nixon A, Yoshida M, Wang XF, Yao TP. HDAC6 is a microtubule-associated deacetylase. Nature. 2002;417:455–458. [PubMed]
  • Impey S, Obrietan K, Wong ST, Poser S, Yano S, Wayman G, Deloulme JC, Chan G, Storm DR. Cross talk between ERK and PKA is required for Ca2+ stimulation of CREB-dependent transcription and ERK nuclear translocation. Neuron. 1998;21:869–883. [PubMed]
  • Jang SW, Liu X, Yepes M, Shepherd KR, Miller GW, Liu Y, Wilson WD, Xiao G, Blanchi B, Sun YE, Ye K. A selective TrkB agonist with potent neurotrophic activities by 7,8-dihydroxyflavone. Proc Natl Acad Sci U S A. 2010;107:2687–2692. [PubMed]
  • Janke C, Kneussel M. Tubulin post-translational modifications: encoding functions on the neuronal microtubule cytoskeleton. Trends Neurosci. 2010;33:362–372. [PubMed]
  • Johnson OL, Ouimet CC. A regulatory role for actin in dendritic spine proliferation. Brain Res. 2006;1113:1–9. [PubMed]
  • Jourdain P, Fukunaga K, Muller D. Calcium/calmodulin-dependent protein kinase II contributes to activity-dependent filopodia growth and spine formation. J Neurosci. 2003;23:10645–10649. [PubMed]
  • Kelleher RJ, 3rd, Govindarajan A, Jung HY, Kang H, Tonegawa S. Translational control by MAPK signaling in long-term synaptic plasticity and memory. Cell. 2004;116:467–479. [PubMed]
  • Kern A, Behl C. The unsolved relationship of brain aging and late-onset Alzheimer disease. Biochim Biophys Acta. 2009;1790:1124–1132. [PubMed]
  • Kimura H, Hayashi-Takanaka Y, Goto Y, Takizawa N, Nozaki N. The organization of histone H3 modifications as revealed by a panel of specific monoclonal antibodies. Cell Struct Funct. 2008;33:61–73. [PubMed]
  • Korte M, Carroll P, Wolf E, Brem G, Thoenen H, Bonhoeffer T. Hippocampal long-term potentiation is impaired in mice lacking brain-derived neurotrophic factor. Proc Natl Acad Sci U S A. 1995;92:8856–8860. [PubMed]
  • Levenson JM, Sweatt JD. Epigenetic mechanisms in memory formation. Nat Rev Neurosci. 2005;6:108–118. [PubMed]
  • Levenson JM, O'Riordan KJ, Brown KD, Trinh MA, Molfese DL, Sweatt JD. Regulation of histone acetylation during memory formation in the hippocampus. J Biol Chem. 2004;279:40545–40559. [PubMed]
  • Liu X, Chan CB, Jang SW, Pradoldej S, Huang J, He K, Phun LH, France S, Xiao G, Jia Y, Luo HR, Ye K. A Synthetic 7,8-Dihydroxyflavone Derivative Promotes Neurogenesis and Exhibits Potent Antidepressant Effect. J Med Chem. 2010;53:8274–8286. [PMC free article] [PubMed]
  • Lu B. BDNF and activity-dependent synaptic modulation. Learn Mem. 2003;10:86–98. [PubMed]
  • Lynch G, Rex CS, Gall CM. Synaptic plasticity in early aging. Ageing Res Rev. 2006;5:255–280. [PubMed]
  • Martinowich K, Hattori D, Wu H, Fouse S, He F, Hu Y, Fan G, Sun YE. DNA methylation-related chromatin remodeling in activity-dependent BDNF gene regulation. Science. 2003;302:890–893. [PubMed]
  • Metsis M, Timmusk T, Arenas E, Persson H. Differential usage of multiple brain-derived neurotrophic factor promoters in the rat brain following neuronal activation. Proc Natl Acad Sci U S A. 1993;90:8802–8806. [PubMed]
  • Minichiello L. TrkB signalling pathways in LTP and learning. Nat Rev Neurosci. 2009;10:850–860. [PubMed]
  • Ochs G, Penn RD, York M, Giess R, Beck M, Tonn J, Haigh J, Malta E, Traub M, Sendtner M, Toyka KV. A phase I/II trial of recombinant methionyl human brain derived neurotrophic factor administered by intrathecal infusion to patients with amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord. 2000;1:201–206. [PubMed]
  • Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y. The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell. 1996;87:953–959. [PubMed]
  • Pang PT, Teng HK, Zaitsev E, Woo NT, Sakata K, Zhen S, Teng KK, Yung WH, Hempstead BL, Lu B. Cleavage of proBDNF by tPA/plasmin is essential for long-term hippocampal plasticity. Science. 2004;306:487–491. [PubMed]
  • Park DC, Lautenschlager G, Hedden T, Davidson NS, Smith AD, Smith PK. Models of visuospatial and verbal memory across the adult life span. Psychol Aging. 2002;17:299–320. [PubMed]
  • Patterson SL, Abel T, Deuel TA, Martin KC, Rose JC, Kandel ER. Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice. Neuron. 1996;16:1137–1145. [PubMed]
  • Peleg S, Sananbenesi F, Zovoilis A, Burkhardt S, Bahari-Javan S, Agis-Balboa RC, Cota P, Wittnam JL, Gogol-Doering A, Opitz L, Salinas-Riester G, Dettenhofer M, Kang H, Farinelli L, Chen W, Fischer A. Altered histone acetylation is associated with age-dependent memory impairment in mice. Science. 2010;328:753–756. [PubMed]
  • Polevoda B, Sherman F. The diversity of acetylated proteins. Genome Biol. 2002;3 reviews0006. [PMC free article] [PubMed]
  • Prince HM, Bishton MJ, Harrison SJ. Clinical studies of histone deacetylase inhibitors. Clin Cancer Res. 2009;15:3958–3969. [PubMed]
  • Rex CS, Lin CY, Kramar EA, Chen LY, Gall CM, Lynch G. Brain-derived neurotrophic factor promotes long-term potentiation-related cytoskeletal changes in adult hippocampus. J Neurosci. 2007;27:3017–3029. [PubMed]
  • Roth TL, Sweatt JD. Regulation of chromatin structure in memory formation. Curr Opin Neurobiol. 2009;19:336–342. [PMC free article] [PubMed]
  • Salisbury CM, Cravatt BF. Activity-based probes for proteomic profiling of histone deacetylase complexes. Proc Natl Acad Sci U S A. 2007;104:1171–1176. [PubMed]
  • Silva AJ, Stevens CF, Tonegawa S, Wang Y. Deficient hippocampal long-term potentiation in alpha-calcium-callmodulin kinase II mutant mice. Science. 1992;257:201–206. [PubMed]
  • Swank MW, Sweatt JD. Increased histone acetyltransferase and lysine acetyltransferase activity and biphasic activation of the ERK/RSK cascade in insular cortex during novel taste learning. J Neurosci. 2001;21:3383–3391. [PubMed]
  • Tao X, West AE, Chen WG, Corfas G, Greenberg ME. A calcium-responsive transcription factor, CaRF, that regulates neuronal activity-dependent expression of BDNF. Neuron. 2002;33:383–395. [PubMed]
  • Tapia-Arancibia L, Aliaga E, Silhol M, Arancibia S. New insights into brain BDNF function in normal aging and Alzheimer disease. Brain Res Rev. 2008;59:201–220. [PubMed]
  • Tapley P, Lamballe F, Barbacid M. K252a is a selective inhibitor of the tyrosine protein kinase activity of the trk family of oncogenes and neurotrophin receptors. Oncogene. 1992;7:371–381. [PubMed]
  • Thoenen H. Neurotrophins and neuronal plasticity. Science. 1995;270:593–598. [PubMed]
  • Timmusk T, Palm K, Metsis M, Reintam T, Paalme V, Saarma M, Persson H. Multiple promoters direct tissue-specific expression of the rat BDNF gene. Neuron. 1993;10:475–489. [PubMed]
  • Vecsey CG, Hawk JD, Lattal KM, Stein JM, Fabian SA, Attner MA, Cabrera SM, McDonough CB, Brindle PK, Abel T, Wood MA. Histone deacetylase inhibitors enhance memory and synaptic plasticity via CREB:CBP-dependent transcriptional activation. J Neurosci. 2007;27:6128–6140. [PMC free article] [PubMed]
  • Xie CW, Sayah D, Chen QS, Wei WZ, Smith D, Liu X. Deficient long-term memory and long-lasting long-term potentiation in mice with a targeted deletion of neurotrophin-4 gene. Proc Natl Acad Sci U S A. 2000;97:8116–8121. [PubMed]
  • Yeh SH, Lin CH, Gean PW. Acetylation of nuclear factor-kappaB in rat amygdala improves long-term but not short-term retention of fear memory. Mol Pharmacol. 2004;65:1286–1292. [PubMed]
  • Zeng Y, Zhao D, Xie CW. Neurotrophins enhance CaMKII activity and rescue amyloid-beta-induced deficits in hippocampal synaptic plasticity. J Alzheimers Dis. 2010;21:823–831. [PMC free article] [PubMed]