Calpains are involved in calcium-induced neuronal cell toxicity, which is associated with the pathophysiology of Alzheimer's disease (AD). The activity of calpains is regulated by the inhibitor calpastatin, and increased activity of calpains and decreased calpastastin are often found in AD. Histone deacetylase (HDAC) inhibitors are implicated in AD treatment through the improvement of learning and memory but the underlying mechanism is yet to be understood. Here, using SH-SY5Y neuroblastoma cells and a calcium ionophore ionomycin, we examined whether and how HDAC inhibitor trichostatin A (TSA) inhibits calcium-induced neuronal cell death. TSA increased both the mRNA and protein levels of calpastatin, with no alterations in those of calpain 1 and calpain 2. Furthermore, TSA-stimulated increase of calpastatin was accompanied by a significant attenuation of ionomycin-induced autolysis of calpain 1, but not of calpain 2, and calpain-dependent 150 kDa αII spectrin cleavage. Under these conditions, however, caspase activity was unaltered. Moreover, ectopic expression of small interfering RNA of calpastatin reversed the inhibitory effect of TSA on ionomycin-induced calpain 1 autolysis and αII spectrin cleavage. Chromatin immunoprecipitation assay revealed the increased levels of acetylation at lysine 5 of histone H4 (H4K5-Ac), H3K9-Ac and H3K14-Ac within the calpastatin promoter region in TSA-treated cells relative to control cells. Finally, TSA significantly decreased ionomycin-induced cell toxicity. This study demonstrates that TSA attenuates calcium-induced neuronal cell death by the inhibition of calpain activity which is mediated in part by increased calpastatin expression via histone hyperacetylation within the calpastatin promoter region. Our study provides a novel mechanism for the neuroprotective effect of HDAC inhibitors on AD.
acetylation at lysine 9 of histone H3 or histone H3 acetylated at lysine 9
acetylation at lysine 14 of histone H3 or histone H3 acetylated at lysine 14
acetylation at lysine 5 of histone H4 or histone H4 acetylated at lysine 5
spectrin breakdown products
small interfering RNA
Dysregulation of calcium homeostasis in neuronal cells is implicated in a number of neurodegenerative diseases including Alzheimer's disease (AD). Excessive intracellular calcium load activates proteases such as calpains, thereby cleaving a number of substrates responsible for neuronal cell function and consequently triggering cell death. Two isoforms of calpain, μ-calpain (calpain 1) and m-calpain (calpain 2), are expressed in neuronal cells, and are activated by the level of micromolar and millimolar calcium concentration, respectively . Although physiological functions of calpains include cell motility and attachment, membrane fusion, cell cycle regulation and long-term potentiation , the details of their molecular mechanism are yet to be fully understood. In contrast, the pathological role of calpains in neuronal cells is well documented. For example, the immunoreactivity of calpain 1 and calpain 2 is highly detected in senile plaques of AD brain [3-5]. Furthermore, the inhibition of calpains exerts neuroprotection in various models of brain injuries, such as ischemic- or excitotoxicity-induced neuronal death [6, 7]. Concurrently, treatment with a synthetic calpain inhibitor improves memory and synaptic transmission in the mouse model of AD .
In addition to calpains, neuronal cells also express calpastatin, a protein that inhibits calpain . As expected, the brains of calpastatin-deficient mice show high sensitivity to neurotoxic kainite , while calpastatin-transgenic (TG) mouse brains display reduced cell loss when neurotoxic insult is applied . A recent study also showed that cleavage of the cyclin-dependent kinase 5 activator p35 to p25 was increased in brain extracts from calpastatin knockout mice, while p25 generation was not detected in those from calpastatin-TG mice , suggesting a role for calpastatin in regulating neuronal calpains. Furthermore, calpastatin overexpression causes a remarkable decrease of tau phosphorylation and amyloid plaque formation, a hallmark of AD, in a TG mouse model of AD . In aggregate, the data suggest that the ratio of calpains to calpastatin is one of the key factors in the development of AD; however, studies investigating the regulation of activity and/or expression of these two proteins are very limited .
Histone deacetylase (HDAC) inhibitors are now recognized as one of the promising therapeutic agents for AD. For example, HDAC inhibitors such as sodium butyrate or trichostatin A (TSA) are reported to restore the learning process and to enhance the access to long-term memories in the AD mouse model . Valproic acid, another HDAC inhibitor, also reduces amyloid β peptide (Αβ) production in Swedish amyloid precursor protein (APP) transfected cells and the AD mouse model [14, 15]. Recently, HDAC activity has been reported to be involved in AD pathogenesis via epigenetic regulation of AD-associated genes including neprilysin . Therefore, we tested whether calpains and/or calpastatin are also epigenetically regulated by HDAC inhibitor TSA and found that increased calpastatin by histone modification significantly inhibits calcium-mediated neuronal cell toxicity, suggesting a role for HDAC inhibitors in neuroprotection in AD development.
TSA increases the protein and mRNA levels of calpastatin
It was reported that calpains and calpastatin are significantly involved in calcium-dependent neuronal cell death, which is implicated in AD . A potential use of HDAC inhibitors in AD treatment prompted us to examine whether TSA, an HDAC inhibitor, alters the expression of calpains and calpastatin in SH-SY5Y cells. As shown in Fig. 1(A,B), TSA did not alter the protein levels of calpain 1 and calpain 2. In contrast, TSA significantly increased the expression of calpastatin protein in a dose-dependent manner. Maximal increase in calpastatin protein levels was observed after 300 nm TSA treatment for 12–24 h (~ 2-fold of control), and this fold increase did not alter further at 600 nm TSA for 24 h or 300 nm TSA for 48 h. Therefore, all subsequent experiments were performed using 300 nm TSA for 24 h. Like protein expression, real-time PCR analysis also showed that TSA (300 nm for 24 h) significantly increased calpastatin mRNA expression by ~ 3.5-fold (Fig. 1E), although neither calpain 1 mRNA nor calpain 2 mRNA was altered (Fig. 1C,D). These results suggest that TSA may modulate calpastatin expression at transcriptional level.
TSA inhibits ionomycin-induced calpain 1 autolysis and spectrin cleavage
To test whether increased calpastatin expression induced by TSA attenuates calpain activity, which may provide the mechanism by which HDAC inhibitors rescue AD-associated neuronal cell death, cells were treated with a calcium ionophore ionomycin. As expected, ionomycin treatment significantly increased intracellular calpain activity, as evidenced by increased calpain 1 autolysis and calpain-mediated 150 kDa αII spectrin cleavage (Fig. 2A). No alteration in either calpain 2 autolysis or caspase activity was found. Furthermore, our data clearly showed that TSA pretreatment significantly lowered both calpain 1 autolysis and 150 kDa αII spectrin breakdown products (SBDP), all of which were induced by ionomycin. As expected, under our conditions TSA induced calpastatin expression and hyperacetylation of histone H3 (H3-Ac). To further examine whether calpastatin directly mediates TSA-stimulated decrease in calpain 1 activity, cells were ectopically transfected with small interfering RNA (siRNA) against calpastatin. As shown in Fig. 2(B), we found that ectopic expression of calpastatin siRNA significantly reversed the inhibitory effect of TSA on ionomycin-induced calpain 1 autolysis and αII spectrin cleavage.
Histone acetylations are associated with TSA-induced increase in calpastatin expression
Since calpain 1 was not likely to be transcriptionally regulated (Fig. 1C), we next explored whether TSA increases calpastatin through hyperacetylation of the calpastatin promoter region. As shown in Fig. 3, chromatin immunoprecipitation (ChIP) analysis clearly revealed that TSA increased levels of acetylation at lysine 5 of histone H4 (H4K5-Ac), H3K9-Ac and H3K14-Ac within a promoter region of calpastatin gene, which are all known to be active sites. These results indicate that the TSA-induced increase in calpastatin expression may be mediated at least in part by the regulation of lysine hyperacetylation on histone H4 and H3.
TSA inhibits ionomycin-induced cytotoxicity
To test whether both the increased calpastatin expression and subsequent decreased calpain 1 activity induced by TSA (Fig. 2) indeed suppressed calcium-activated neuronal cell death, the differentiated SH-SY5Y cells were treated with 5 μm ionomycin for the indicated times in the absence or presence of 300 nm TSA pretreatment for 24 h. As shown in Fig. 4, lactate dehydrogenase (LDH) release assay revealed that TSA treatment alone for 36 h (i.e. 24 h TSA pretreatment plus a further 12 h incubation without ionomycin) induced cytotoxicity by ~ 15% of control, although a little toxicity was found for 24 h treatment (i.e. 24 h TSA pretreatment only). Control cells [dimethylsulfoxide (DMSO) treated, open bars] did not show any significant cytotoxicity throughout all the experimental periods. However, ionomycin treatment (grey bars) clearly induced cytotoxicity by ~ 33% at 1 h, and this toxicity greatly increased in a time-dependent manner (~ 49%, 58% and 66% of control at 3, 6 and 12 h treatment, respectively). Lastly, cells co-treated with both TSA and ionomycin (solid bars) showed a significant reduction of cell toxicity relative to cells treated with ionomycin alone. Thus we re-analyzed cytotoxicity after adjusting for the cytotoxicity by TSA alone at each time point (Fig. 4, inset). From this re-analysis, we found that TSA clearly inhibited ionomycin-stimulated cell toxicity in a time-dependent manner.
5-Aza-2′-deoxycytidine (5-Aza) does not alter the expression of calpain 1, calpain 2 or calpastatin
Since cross-talk between histone modification and DNA methylation is frequently observed in the epigenetic regulation of gene expression [17, 18], we finally tested whether 5-Aza, a DNA demethylating agent, also alters calpain 1, calpain 2 or calpastatin expression. Our data showed no alterations in all three proteins after 2 μm 5-Aza treatment for 72 h (Fig. 5A). In agreement with alterations in the protein levels, neither the mRNA level of calpain 1 (Fig. 5B), calpain 2 (Fig. 5C) or calpastatin (Fig. 5D) was altered, suggesting that DNA methylation is unlikely to be involved in the expression of calpain 1, calpain 2 and calpastatin under our experimental conditions.
In this study we found that TSA significantly increased calpastatin expression. Furthermore, this increase was due to epigenetic regulation of histone acetylation but not of DNA methylation. Since calpastatin was reported to inhibit both calpain 1 and calpain 2, which digest numerous cellular proteins associated with neuronal cell toxicity in a calcium-dependent manner, our findings may provide novel molecular mechanisms by which TSA exerts beneficial effects on neuronal cell survival in AD patients and animals.
Our data show that TSA significantly alleviates calcium-mediated neuronal cell death through the inhibition of calpain 1 activity. Previously, calpain inhibition using inhibitors E64 and BDA-410 is reported to restore normal synaptic function in hippocampus and improve memory in the AD animal model , which is significantly associated with neuronal cell death . Under our experimental conditions, calpastatin induced by TSA is likely to be a key molecule which contributes to calpain inhibition, resulting in decreasing cell toxicity. These results are largely consistent with the previous study showing that calpastatin overexpression in differentiated PC12 cells inhibits Aβ-derived or calcium-mediated calpain 1 autolysis, 150 kDa SBDP production and increased membrane permeability .
Although TSA increased calpastatin protein in the basal state (Fig. 1), it is interesting to find that TSA alone significantly decreased cell viability compared with control (DMSO only; Fig. 4). Based on the inhibitory effects of calpastatin on the proteolytic property of calpains, it is readily expected that the decreased ratio of calpain/calpastatin by TSA causes an increase in cell viability. Although unexpected results obtained from our study are yet to be fully understood, we assume that TSA-induced cytotoxicity in the basal (not calcium-stimulated) state may be attributable to mechanism(s) other than the calpain/calpastatin signaling pathway. In this regard, it should not be excluded that TSA regulates a number of other loci that may cause a negative effect on cell viability. Furthermore, a previous study also showed that TSA-stimulated apoptosis is calpain-independent and is associated with dissipated mitochondrial membrane potential and enhanced the release of Omi/HtrA2 and AIF protein from the mitochondria to the cytosol . Whether the mitochondrial function is also involved in TSA-induced cell death in SH-SY5Y cells needs further investigation. Nonetheless, it is clear that the calpain/calpastatin signaling pathway plays an important role in cytotoxicity in calcium ionophore-stimulated cells (Fig. 4).
Under our experimental conditions there seems to be no clear evidence that the calpains, important enzymes responsible for calcium-mediated neuronal cell toxicity, are epigenetically regulated, because their mRNA expressions are not altered. However, calpastatin, an endogenous inhibitor for calpains, is probably significantly regulated by histone modifications. This finding suggests that calpastatin is likely to be a more important player in calpain-mediated adverse effects in AD progression than we previously thought. In this regard, it was reported previously that in control mice calpastatin was higher in the cerebellum than in the hippocampus and frontal and temporal cortex, while calpain levels are relatively similar in all these brain regions . In a TG model of AD, however, calpain was activated through decreasing calpastatin levels only in brain regions affected in AD, particularly in the hippocampus, but not in the cerebellum, a region not exhibiting AD-like pathology. This previous study suggested that calpain activation and perhaps development of subsequent AD pathology in various brain regions are likely to be predominantly attributable to alterations in calpastatin levels. Thus, our finding of the epigenetic mechanism underlying calpastatin regulation may be useful in future AD research, given that calpastatin is an important factor in the regulation of the calpain-induced protein degradation responsible for AD progression.
Although protein kinase C and extracellular-signal-regulated kinase are reported to increase calpain activity [23, 24], the underlying mechanism(s) have not been determined. Furthermore, except for the MEF-2/E4-box, no detailed analysis of calpain promoter regions has been performed. However, promoter analysis for calpastatin identified several putative regions including a cAMP-responsive element , which may play a role in differential expression patterns. Together with these previous data, our finding of epigenetic regulation of calpastatin, via histone acetylation, suggests again that calpastatin is more likely to be regulated at its expression level in cells.
It is well known that acetylations of H4K5, H3K9 and H3K14 generate a loose chromatin configuration via decreased ionic interaction between positively charged lysines of histone and negatively charged DNA. Thus, it is now likely that the acetylation of all these histone lysine groups induced by TSA makes the euchromatin configuration accessible to transcriptional factors, including a cAMP-responsive element binding protein which is responsible for increased calpastatin expression, but further studies are needed to clarify the molecular mechanism of this linkage.
It was recently reported that the levels of histone H4 acetylation decreased by ~ 50% in the hippocampus in the mouse model of AD compared with wild-type littermates . In this previous study, treatment with TSA, an HDAC inhibitor, rescued the hypoacetylated H4 levels with concomitant increased synaptic function and memory function. Furthermore, in a mouse model of aging, dysregulation of histone H4K12 acetylation was seen in aged mice and cognitive ability was recovered if H4K12 acetylation was restored . A separate study also showed that HDAC2, but not HDAC1, was involved in memory formation and synaptic plasticity in the mouse model . All these studies clearly show that altered histone acetylation is associated with the memory impairment associated with AD and aging. Furthermore, these and our studies encourage the development of HDAC-selective inhibitors to counteract those problems. To date, however, only a few genes responsible for the memory impairment associated with AD have been revealed, which are directly regulated by histone acetylation. Our data demonstrate clearly the potential of HDAC inhibitors for AD treatment via the inhibition of calcium-induced calpain activation.
Materials and methods
Cell culture and drug treatments
The human neuroblastoma cell line, SH-SY5Y, was purchased from the Korean Cell Line Bank (Seoul, Korea) and maintained in Dulbecco's modified Eagle's medium (Gibco-BRL, Grand Island, NY, USA) containing 10% fetal bovine serum as described previously . Except for the transfection study, cells were differentiated into neuronal phenotypes using 10 μm all-trans retinoic acid (Sigma, St Louis, MO, USA) for 5–6 days as described previously . Differentiated cells were incubated with various doses of TSA (Sigma) for the indicated times. In some experiments, after treatment with 300 nm TSA for 24 h, cells were further co-treated with 5 μm ionomycin (Calbiochem, Darmstadt, Germany) for the indicated times. In a separate experiment, cells were also treated with 2 μm of 5-Aza (Sigma) for 72 h.
Western blot analysis
For western blot analysis, cells were treated with indicated doses of either TSA or 5-Aza, washed with ice-cold Dulbecco's phosphate buffered saline and then lysed in lysis buffer [20 mm Tris/HCl pH 7.5, 150 mm NaCl, 1% Triton X-100, 1 mm EDTA, 1 mm EGTA, 1 mm phenylmethylsulfonyl fluoride, 10 mm β-glycerophosphate, 1 mm NaF, 1 mm Na3VO4 and Protease Inhibitor Cocktail™ (Roche Molecular Biochemicals, Indianapolis, IN, USA)]. The protein concentrations were then determined using the BCA protein assay kit (Sigma). Equal quantities of the protein (20 μg) were separated on SDS/polyacrylamide gel under reducing conditions, after which they were electrophoretically transferred onto nitrocellulose membranes. The blots were then probed with antibodies directed against calpain 1 large subunit, calpain 2 large subunit (Cell Signaling Technology, Beverly, MA, USA), calpastatin, αII spectrin (Chemicon, Temecula, CA, USA), caspase-3 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), acetylated histone H3 (H3-Ac; Upstate, Lake Placid, NY, USA; each 1 : 1000 dilution) and β-actin (Sigma; 1 : 5000 dilution) followed by the corresponding secondary antibody, and finally developed using enhanced chemiluminescence reagents (GE Healthcare, Piscataway, NJ, USA).
Reverse transcription and real-time PCR
Total RNA was extracted from SH-SY5Y cells using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) as described previously . Briefly, cells were homogenized in 1 mL of TRIzol reagent and then RNA was used for the reverse transcription reaction to obtain cDNA. The reaction was performed using 200 units of SuperScript™ III reverse transcriptase (Invitrogen), 10 pmol of oligo-dT and 2 mm dNTPs in a 20 μL reaction mix containing 1 μg RNA for 1.5 h at 42 °C. The mRNA level of each gene was quantified by real-time PCR using a SensiMixPlus SYBR kit (Quantace, London, UK) with a Rotor-gene 6000 analyzer (Corbett Research, Sydney, Australia), according to the manufacturer's instructions. The reaction conditions were as follows: 95 °C for 10 min, 40 cycles at 95 °C for 10 s, 60 °C for 15 s and 72 °C for 20 s. The following PCR primer pairs were designed to detect each gene: calpain 1-F, 5-ACA TGG AGG CCA TCA CTT TC-3; calpain 1-R, 5-GGT CCA CGT TGT TCC ACT CT-3; calpain 2-F, 5-AGG ATG AGG ACG AGG AGG AT-3; calpain 2-R, 5-TGG TCT GCC CAC TTA ACT CC-3; calpastatin-F, 5-CAA AAA GCC TAC CCA AGC AG-3; calpastatin-R, 5-CAG CAA CAC TCT CTC CAC CA-3; GAPDH-F, 5-GAA GGT GAA GGT CGG AGT C-3; and GAPDH-R, 5-GAA GAT GGT GAT GGG ATT TC-3. At the end of each cycle, the fluorescence was measured and used for quantification.
Transfection with siRNA
SH-SY5Y cells grown in six-well plates were transiently transfected using FuGENE HD reagent (Roche) with 100 nm calpastatin siRNA (sc-29889; Santa Cruz Biotechnology) or control siRNA (sc-37007; Santa Cruz Biotechnology). Transfected cells were incubated with 300 nm TSA for 24 h and then further co-treated with 10 μm ionomycin for 30 min.
A ChIP assay was performed using a ChIP assay kit (Upstate), according to the manufacturer's instructions and as described previously . Briefly, cells treated with TSA or DMSO were crosslinked in 1% formaldehyde and 0.1 m glycine and suspended in SDS lysis buffer (1% SDS, 10 mm EDTA and 50 mm Tris/HCl, pH 8.1). Next, the chromatin solution was sonicated, pre-cleared and immunoprecipitated with 2 μg of the desired antibodies and protein A-agarose/Salmon Sperm DNA beads. Antibodies against histone H4 acetylated at lysine 5 (H4K5-Ac), H3K9-Ac and H3K14-Ac were purchased from Upstate. Mock samples were prepared by the immunoprecipitation procedure without antibody. Input (total chromatin extract), mock and ChIP samples were recovered and then used for PCR analysis. PCR amplification of a cDNA encoding each targeted gene was conducted in a total volume of 20 μL containing 0.5 units of TaKaRa Ex Taq HS polymerase, 10 pmol of each primer, 250 μm of dNTPs and 1 μL of ChIP sample. The PCR primer pairs for promoter regions of each gene for the ChIP assay were as follows: calpastatin-F, 5-AAA CTC TCG CAG CTA AAG CG-3; calpastatin-R, 5-TGG TCA GAA AGA AGG GGT TG-3; β-actin-F, 5-CCA ACG CCA AAA CTC TCC C-3; and β-actin-R, 5-AGC CAT AAA AGG CAA CTT TCG-3. The amplified products were separated using 1.5% agarose gel in TAE buffer (40 mm Tris/acetate, pH 8.0, 1 mm EDTA). The bands of target genes were visualized using 1 mg·mL−1 ethidium bromide under UV.
LDH release assay
The LDH release assay was used as an index of cell toxicity and was done as described previously  with minor modifications. In this assay, LDH activity in the cultured medium represented LDH release from SH-SY5Y cells treated with ionomycin or vehicle (DMSO only) for the indicated times in the absence or presence of TSA pretreatment. In brief, cells cultured in 12-well plates were pretreated with 300 nm TSA for 24 h and thereafter co-treated further with 5 μm ionomycin for the indicated times. For low control, cells were cultured without treatment. A high control was prepared by treating cells with 1% Triton X-100. Test samples were prepared by treating cells with DMSO, TSA, ionomycin or TSA plus ionomycin. An aliquot (100 μL) of the cultured medium was collected into a 96-well plate and incubated with 100 μL of reaction mixture, and the LDH release assay was performed using an LDH cytotoxicity assay kit (BioVision, Mountain View, CA, USA), according to the manufacturer's instructions. Cytotoxicity was calculated using the equation cytotoxicity (%) = [(test sample − low control)/(high control − low control)] × 100.
All results are expressed as means ± standard deviation (SD) with n indicating the number of experiments. Statistical significance of difference was determined using Student's t test for paired data. A value of P <0.05 was considered significant.
Conflict of interest
The authors declare that there is no conflict of interest.
This work was supported in part by grants NRF-2008-0061393, 2012R1A2A2A01004914 (awarded to I. Jo) and NRF-2011-0016127 (to S.A. Jo) from the National Research Foundation, Korea. S.A. Jo was also supported by a grant from the Institute of Bio-Science and Technology at Dankook University in 2011.