Introduction

Alzheimer's disease (AD) is a neurodegenerative disorder characterized by abnormal accumulations of extracellular amyloid plaques and intracellular neurofibrillary tangles (NFTs) throughout cortical and limbic brain regions. Cognitive deficits in AD are believed to result from progressive synaptic dysfunction and neurodegeneration initiated by soluble amyloid β oligomers (AβOs) and further involving aggregates of hyperphosphorylated Tau1,2. AβOs, which are increasingly considered to act as proximal neurotoxins in AD, interact with glutamate receptors at the dendritic membrane, induce abnormal calcium influx and oxidative stress, block long-term potentiation (LTP), and facilitate long-term depression, ultimately leading to synapse failure3,4. It has recently been suggested that the deposition of Aβ might trigger a series of cellular events that lead to posttranslational changes in Tau followed by neurite degeneration2,5. Furthermore, experiments in both cultured rodent hippocampal neurons and transgenic mice demonstrate that Tau is intrinsically involved in Aβ-mediated neuronal dysfunction and cell death6,7. Taken together, the data suggest that there is an intrinsic relationship between Aβ and Tau dysfunction. It has been shown that the deposition of Aβ induced the activation of calpain-1 and caspase-3 proteases2,5,8. These proteases cleave Tau proteins at specific sites, generating toxic Tau fragments or enhancing the aggregation properties of Tau protein2,5,9. The activation of these proteases, Tau cleavage, and the extent of neurodegeneration are both time- and dose-dependent2,5,9. Inhibition of these proteases completely prevented the Tau proteolysis that leads to the generation of a 17-kDa fragment, significantly reducing Aβ-induced neuronal death2,5. Furthermore, hippocampal neurons transfected with 17-kDa Tau showed signs of degeneration, such as tortuous processes, varicosities along the neurites, and retraction of neuritic processes5. Collectively, these data suggest that a reduction in factors that prevent the activation of calpain and caspase-3, and hence Tau cleavage, could have some bearing in the pathophysiology of AD.

Several prospective studies have indicated that statins could prevent the onset of AD10,11,12. A recent meta-analysis by Wong et al also showed that statins have preventive effects on AD13. Evidence from cell culture experiments and animal studies has suggested that statins have many pleiotropic effects, such as reducing Aβ production, suppressing inflammatory responses, protecting neurons from Aβ-induced neurotoxicity, apoptosis and oxidative stress, and promoting synaptogenesis14,15,16,17. Recently, a transgenic mouse model of tauopathy showed a reduction in NFTs in response to statin treatment in both early and late stages of disease progression18. It has been reported that statins reduce the number of phosphorylated Tau-positive neurites in aged amyloid precursor protein (APP) transgenic mice19. Atorvastatin is a member of the statin family. Clarke et al demonstrated that rats treated with atorvastatin for 3 weeks were protected against a deficiency in LTP caused by the acute injection of Aβ1–4220. Our previous results revealed that atorvastatin prevented AβO-induced synaptotoxicity, which leads to memory dysfunction through a p38MAPK-dependent pathway17. However, the mechanisms underlying the neuroprotective effects of statins have not been fully elucidated. In the present study, we analyzed whether atorvastatin exerts its neuroprotective effect against AβO-induced neurotoxicity by preventing Tau cleavage. Our results showed that atorvastatin blocked the activation of calpain and caspase-3 and hence decreased the generation of 17-kDa Tau fragments. Treatment with atorvastatin also decreased neurite degeneration in cultured hippocampal neurons treated with AβOs.

Materials and methods

Reagents

Atorvastatin was obtained from LKT Laboratories (St Paul, MN, USA). The calpain inhibitor Z-L-Abu-CONH-ethyl was from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and the caspase inhibitor benzyloxycarbonyl-VAD-fluoromethyl ketone was from Sigma-Aldrich (St Louis, MO, USA).

Preparation of AβOs

Human Aβ1–42 (product No A9810) was purchased from Sigma-Aldrich (St Louis, MO, USA). AβOs were prepared according to our previously described method17.

Primary hippocampal neuron cultures

Hippocampal cultures were prepared from embryonic Sprague-Dawley rats as previously described21. The hippocampi of E18/19 rat fetuses were collected in Hanks' solution without Ca2+ and Mg2+ (D-Hanks). The hippocampi were then mechanically fragmented, transferred to D-Hanks' solution containing 0.125% trypsin, and incubated for 15 min at 37 °C. Following trypsinization, cells were washed twice with Dulbecco's modified Eagle's medium (DMEM) and re-suspended in DMEM/F12 medium containing 10% heat-inactivated fetal bovine serum, 10% horse serum, glutamine (3 mg/mL), insulin (0.25 mg/mL), penicillin (50 U/mL), and streptomycin (50 mg/mL). The cells were plated on poly-L-lysine-coated 16-mm-diameter coverslips (∼150 cells/mm2) for immunocytochemistry assays, on 6-well culture plates (1×106 cells/well) for Western blot analysis, or on 96-well dishes (1×104 cells/well) for cell viability assays. Neurons were grown at 37 °C in a humidified atmosphere of 5% CO2/95% O2. After 16 h, the medium was changed to neurobasal medium supplemented with glutamine (3 mg/mL) and B-27 (2%; Life Technologies, Gaithersburg, MD, USA). Subsequent half-medium changes were performed every 3–4 d for 14 d, at which time AβOs treatments were initiated.

Treatment of the cultures

Immediately after preparation of soluble AβOs, the solution was diluted to between 0.16 and 2.5 μmol/L in neuronal culture medium. The hippocampal neurons, which had been cultured for 14 d, were incubated with AβOs for various time periods, ranging from 10 min to 48 h, with or without atorvastatin. In co-incubation experiments, atorvastatin or inhibitor was added to the neurons 1 h prior to incubation with AβOs. Inhibitors were added to the cell cultures 1 h prior to incubation with atorvastatin. The calpain inhibitor Z-L-Abu-CONH-ethyl was used at 1 μmol/L. The caspase inhibitor benzyloxycarbonyl-VAD-fluoromethyl ketone was used at 50 μmol/L.

Cell viability assays

To examine the effect of atorvastatin treatment on cell viability, hippocampal neurons were plated directly in 96-well dishes (1×104 cells/well) and incubated with AβOs for 48 h with or without atorvastatin (0.1, 0.5, 1, and 2.5 μmol/L). In co-incubation experiments, atorvastatin was added to the neurons 1 h prior to incubation with AβOs. At the end of incubation, cell viability was determined by adding 100 μL of CellTiter-Glo® luminescent reagent (Promega, Madison, WI, USA), which produces light in direct proportion to the amount of ATP and the number of viable cells present. The luminescence was measured using a BioTek Synergy 2 microplate reader (BioTek Instruments, Winooski, VT, USA). The luminescent values were normalized between untreated control cells (100% viable) and a cell death control treatment of 200 μmol/L staurosporine (Santa Cruz Biotechnology, Santa Cruz, CA, USA) for 24 h (0% viable), which consistently gave luminescent readings only slightly above background (wells with culture medium but without cells). The CytoTox-ONETM (Promega, Madison, WI, USA) assay measures lactate dehydrogenase (LDH) released into the culture medium through compromised cell membranes22. One hundred microliters of medium was removed from the treated neurons and placed into 96-well dishes. LDH activity in the medium was measured according to the manufacturer's instructions, and fluorescent values were normalized between untreated controls (100% viable) and Triton X lysis controls to represent the maximum amount of LDH available for release (0% viable). Each experiment was performed in triplicate.

Immunofluorescence microscopy

After fixation with 4% paraformaldehyde for 30 min, neurons cultured on coverslips were permeabilized in PBS with 0.2% Triton X-100 for 5 min and incubated with 3% BSA in PBS for 30 min for the immunocytochemical analysis of microtubule-associated protein-2 (MAP-2, a dendritic marker), tubulin, Tau-5, PSD-95, and Bassoon. The cells were incubated with a mouse monoclonal anti-MAP-2 antibody (1:100; Novus Biologicals, CO, USA), mouse monoclonal anti-Tau (Tau-5, 1:100; Abcam Inc), rabbit polyclonal anti-Bassoon (1:100; Abcam Inc), or rabbit polyclonal anti-PSD-95 (1:50; Santa Cruz Biotechnology, Santa Cruz, CA, USA) antibody overnight at 4 °C. After extensive washes with PBS, the cells were incubated with an anti-mouse or anti-rabbit secondary antibody conjugated with fluorescein (FITC, 1:1000; The Jackson Labs, West Grove, PA, USA) or rhodamine (TRITC, 1:1000; The Jackson Labs, West Grove, PA, USA).

Protease activity assays

Neurons cultured in 96-well dishes were directly measured for calpain and caspase-3/7 protease activities using substrates that release luminescent signals when cleaved. Calpain activity was measured using the Calpain-Glo assay (Promega, Madison, WI, USA). Luminescent measurements from AβO-treated wells (triplicate) were normalized to control, mock-treated wells (set to 1). Caspase-3/7 activity was measured using the Caspase-Glo 3/7 assay (Promega, Madison, WI, USA) and analyzed in the same fashion as the calpain assay.

Western blot analysis

Western blot analysis was performed as previously described17. Whole-cell lysates were prepared from hippocampal neurons cultured in 6-well dishes using radioimmune precipitation assay lysis buffer. After 30 min of lysis buffer treatment at 4 °C, insoluble material was removed by centrifugation for 30 min at 12 000×g, and the protein concentration of the lysate was determined using a BCA assay. Equal amounts of protein were then separated by SDS-PAGE, transferred onto nitrocellulose membranes, and probed with primary antibodies against the following proteins: mouse monoclonal anti-spectrin αII (Santa Cruz Biotechnology, Santa Cruz, CA, USA), rabbit monoclonal anti-vimentin (1:1000, Abcam, New Territories, HK, USA), mouse monoclonal anti-Tau (Tau-5, 1:500, Abcam, New Territories, HK, USA), rabbit polyclonal anti-p35 (C-19, 1:500, Santa Cruz Biotechnology, Santa Cruz, CA, USA), rabbit monoclonal anti-phospho-GSK3β (Ser9, 1:1000, Cell Signaling Technology, Beverly, MA, USA), rabbit monoclonal anti-phospho-Erk1/2 (Thr202/Tyr204, 1:1000, Cell Signaling Technology, Beverly, MA, USA), and rabbit monoclonal anti-phospho-Akt (Ser473, 1:1000, Cell Signaling Technology, Beverly, MA, USA). After being washed with PBS, membranes were incubated with horseradish peroxidase-conjugated anti-rabbit IgG secondary antibody (1:2000, Cell Signaling Technology, Beverly, MA, USA). Then, the membranes were washed and visualized using the ECL kit (Pierce, Rockford, IL, USA). The membranes were then reprobed for β-actin immunoreactivity using a mouse anti-β-actin antibody (1:2000, Cell Signaling Technology). To determine the phosphorylation ratio of Akt, GSK3β, and Erk1/2, the membranes were reprobed with rabbit monoclonal anti-Akt (1:1000, Cell Signaling Technology, Beverly, MA, USA), rabbit monoclonal anti-GSK3β (1:1000, Cell Signaling Technology, Beverly, MA, USA), or rabbit monoclonal anti-Erk1/2 (1:1000, Cell Signaling Technology). Staining intensity was quantified from 4 blots derived from 4 independent experimental trials. The density of each band was quantified using Image J software and normalized to total kinase or β-actin expression. The protein levels were expressed as a ratio of the band intensity for the protein of interest to that for total kinases or β-actin, which were used as loading controls.

Statistical analysis

The data are expressed as the mean±SEM and were compared by one-way ANOVA followed by an LSD post hoc multiple-comparison test. P<0.05 was considered statistically significant.

Results

Atorvastatin protected hippocampal neurons against soluble AβO-induced neurotoxicity

First, we determined the ability of different doses of soluble AβOs to induce neuronal death in primary rat hippocampal neurons using two independent assays. Our initial cell viability assay, measuring ATP content in neurons, revealed a dose-dependent effect, with soluble AβO treatment (0.625-2.5 μmol/L) generating a significant decrease in cell viability over a 48 h exposure (Figure 1A). We used an independent cell viability assay that measures the release of lactate dehydrogenase into medium, reflecting a compromised membrane structure. The pattern of LDH release in this assay was similar to the loss of ATP in the cell viability measurements described above (Figure 1A). A concentration of 1.25 μmol/L AβOs was chosen for all subsequent analyses because this concentration is comparable with many previous cell culture studies that assessed Aβ-induced cell death. When neuronal cell viability was measured over 48 h of exposure to 1.25 μmol/L AβOs, pretreatment with atorvastatin produced significant protection (Figure 1B). We also examined the effect of atorvastatin on AβO-mediated neuronal cell death using the LDH assay, holding the AβO concentration at 1.25 μmol/L. As shown in Figure 1B, atorvastatin provided significant, dose-dependent protection against Aβ-mediated cell damage. We next used protease inhibitors to determine whether the AβO-induced increase in protease activities contributed to Aβ-mediated neuronal cell damage. As shown in Figure 1B, both inhibitors provided considerable protection against neuronal death with the calpain inhibitor providing more protection than the caspase-3/7 inhibitor.

Figure 1
figure 1

Atorvastatin protected neuronal viability upon AβO treatment in cultured hippocampal neurons. Hippocampal neurons cultured for 8 d were incubated with the indicated concentrations of AβOs (A) for 48 h with or without the indicated concentrations of atorvastatin (B). In co-incubation experiments, atorvastatin was added to the neurons 1 h prior to incubation with AβOs. Cell viability was measured by ATP content and by LDH released into the medium. Data are expressed as the mean±SEM of 3 independent experiments. cP<0.01 compared with the control group; fP<0.01 compared with the AβO-treated group.

PowerPoint slide

Full size image

The experiments described above suggested that atorvastatin could block AβO-induced neurotoxicity. We next examined neuronal cell morphology 48 h after AβO treatment using an immunofluorescence microscope to image MAP-2, tubulin, and Tau. As shown in Figure 2, apparent signs of degeneration were detected 48 h after hippocampal neurons were treated with soluble AβOs for 48 h. The anti-tubulin images demonstrated that neurons displayed neuritis with varicosities as well as neuritis undergoing retraction. In contrast, no significant degenerative changes were detected in neurons treated with atorvastatin before treatment with AβOs (Figure 2). The anti-MAP2 images demonstrated that the dendrites of AβO-treated neurons were thinner and shorter than those of untreated neurons with a frequently fragmented or beaded appearance23. However, neurons exposed to AβOs plus atorvastatin, a calpain inhibitor or a caspase-3/7 inhibitor for 48 h retained relatively normal dendritic morphology (Figure 2). The anti-Tau (Tau-5) images demonstrated that Tau was mainly located in neurites in control-treated neurons. However, treatment of neurons with AβOs (1.25 μmol/L, 48 h) induced a significant decrease in Tau signal in the distal regions of the neurites and a clear redistribution of Tau into the somata. In contrast, atorvastatin treatment blocked the AβO-induced decrease in Tau signal and Tau missorting (Figure 2). Similarly, both inhibitors prevented the AβO-induced decrease in Tau signal and Tau missorting (Figure 2).

Figure 2
figure 2

Atorvastatin prevented AβO-induced neurotoxicity in cultured hippocampal neurons. Hippocampal neurons cultured for 14 d were treated with atorvastatin (1 μmol/L) 1 h before incubation with soluble AβOs (1.25 μmol/L) for 48 h. No signs of neurite degeneration were detected in untreated controls. In contrast, severe neurite degeneration was observed in cultures incubated with soluble AβOs. Atorvastatin significantly reduced the appearance of dystrophic neurites induced by AβOs. Insets are close-ups of the degeneration shown in the boxed regions. The redistribution of Tau into the somatodendritic compartment is indicated by arrowheads.

PowerPoint slide

Full size image

Because synaptic damage likely plays a more important role than neuronal loss in the early stages of AD, we investigated whether atorvastatin could block AβO-induced synaptic damage. Two different synaptic markers were used to identify synapses: Bassoon, a large cytosolic scaffolding protein that is specifically concentrated at the presynaptic active zones of synapses24, and PSD-95, a postsynaptic scaffolding protein that plays a critical role in synaptic plasticity25. Immunohistochemical detection of the Bassoon protein revealed that Bassoon exhibited a small punctate pattern on hippocampal neurons (Figure 3). Treatment of hippocampal neurons in culture with AβOs for 48 h induced a marked reduction in the density and signal intensity of Bassoon puncta. Pretreatment with atorvastatin (1 μmol/L) blocked the AβO-induced decrease in Bassoon immunoreactivity (Figure 3). To determine whether calpain and caspase are involved in the AβO-induced decrease in Bassoon expression, we observed the effect of both protease inhibitors on Bassoon immunoreactivity. After pretreatment for 1 h with a calpain or caspase inhibitor, we exposed neurons to AβOs for 48 h. Immunohistochemical detection demonstrated that both inhibitors protected against the AβO-induced decrease in Bassoon expression (Figure 3).

Figure 3
figure 3

Atorvastatin attenuated AβO-induced synaptic PSD-95 and Bassoon downregulation in cultured hippocampal neurons. Primary rat hippocampal neurons cultured for 14 d were pretreated (1 h) with atorvastatin (1 μmol/L) or vehicle before exposure to AβOs (1.25 μmol/L). After fixation, neurons were immunostained for PSD-95 or Bassoon. Higher magnification micrographs of the boxed regions are shown in the panel to the right.

PowerPoint slide

Full size image

PSD-95 is predominantly localized at synapses. To assess whether atorvastatin prevented AβO-induced decreases in synaptic PSD-95, we treated hippocampal neurons (14 d) with soluble AβOs with or without atorvastatin for 48 h and immunostained them for PSD-95. Immunofluorescence images of PSD-95 demonstrated that treatment with AβOs decreased both the density and intensity of synaptic PSD-95-labeled puncta (Figure 3). Pretreatment with atorvastatin markedly attenuated the effect of AβOs on synaptic puncta density and intensity (Figure 3). In addition, both protease inhibitors prevented AβO-induced effects on the density and intensity of synaptic PSD-95-labeled puncta. These results suggest that calpain and caspase-3/7 contribute to AβO-induced neurotoxicity.

Atorvastatin prevented soluble AβO-induced activation of calpain and caspase-3/7 proteases

To further clarity the mechanism by which atorvastatin protects hippocampal neurons against AβO-induced neurotoxicity, we performed calpain and caspase-3/7 activity assays on hippocampal neurons as a function of time in the presence of AβOs using a Calpain-GloTM protease assay and a Caspase-Glo® 3/7 assay (Figure 4A and 4B). We observed rapid induction of calpain activity, reaching a maximum at 10 min (a 3-fold increase) and remaining elevated throughout the rest of the period analyzed (Figure 4A). Caspase-3/7 activity was induced at 10 min but was smaller in magnitude (∼1.4 fold) and returned to control levels after 1 h of AβO treatment (Figure 4B). Pretreatment of neurons with atorvastatin (1 μmol/L) diminished AβO-mediated calpain and caspase-3/7 activation (Figure 4A and 4B).

Figure 4
figure 4

Atorvastatin prevented AβO-induced activation of calpain and caspase-3/7 proteases. (A and B) Direct activity measurements were performed as described under “Materials and methods” as a function of time exposed to AβOs with or without atorvastatin pretreatment. Data were normalized to 1 for mock-treated controls. (C and D) An immunoblot analysis verifies protease activity with substrate cleavage. The calpain substrate spectrin and the caspase substrate vimentin demonstrate cleavage into smaller molecular weight fragments upon AβO treatment; cleavage of both is prevented by atorvastatin pretreatment. The bar chart shows the semiquantitative analysis of spectrin and vimentin. Data are expressed as the mean±SEM of 3 independent experiments. cP<0.01 compared with the control group. eP<0.05, fP<0.01 compared with the AβO-treated group.

PowerPoint slide

Full size image

The inhibitory effects of atorvastatin on AβO-mediated activation of calpain and caspase-3/7 were verified using immunoblot analysis of known substrates for calpain and caspase-3/7 in the lysates. Cleavage of spectrin from its full-length form (240 kDa) to a 150-kDa fragment is specific for calpain activity and is commonly used a surrogate marker of this protease's activity5. AβO treatment led to a significant decrease in full-length spectrin (240-kDa) and a concomitant increase in the 150-kDa degradation fragment in cultured hippocampal neurons (Figure 4C). In contrast, preincubation with atorvastatin significantly attenuated AβO-induced calpain activation, as shown by decreased 150-kDa spectrin immunoreactivity and normal full-length spectrin levels (Figure 4C). Additionally, AβO treatment led to the fragmentation of vimentin (a known caspase-3/7 substrate), and this fragmentation was effectively attenuated by pretreatment with atorvastatin (Figure 4D).

Atorvastatin hindered AβO-induced Tau cleavage

Because we observed the inhibitory effect of atorvastatin on calpain and caspase protease activities in AβO-treated neurons, we next used immunoblotting analysis to assess whether atorvastatin prevented Tau cleavage in AβO-treated neurons. For these experiments, we harvested whole cell lysates from neurons that had been treated in the absence or presence of AβOs, atorvastatin, or both. Blots of these samples were labeled with a phosphorylation-independent Tau antibody (Tau-5), which detects both full-length Tau (∼48 and ∼66 kDa) and the 17-kDa Tau fragment26. As shown in Figure 5, AβO treatment for 24 h caused a reduction in full-length Tau immunoreactivity and a concomitant increase in 17-kDa Tau band density compared with untreated controls. In contrast, no reduction of full-length Tau was detected and the production of 17-kDa Tau was almost completely abrogated when atorvastatin (1 μmol/L and 2.5 μmol/L) was added to the culture medium of hippocampal neurons 1 h before the addition of AβO (Figure 5). To obtain insights into the roles of calpain and caspase-3 in AβO-induced Tau cleavage leading to the generation of the 17-kDa fragment, we blocked the activation of these proteases using specific inhibitors. The caspase-3/7 inhibitor significantly reduced the generation of the 17-kDa Tau fragment in the AβO-treated neurons (Figure 5). In contrast, the calpain inhibitor completely prevented the generation of this Tau fragment in AβO-treated hippocampal neurons (Figure 5).

Figure 5
figure 5

Atorvastatin protected against Tau degradation in hippocampal neurons. Hippocampal neurons cultured for 14 d were treated with atorvastatin (0.5, 1, or 2.5 μmol/L) or inhibitor 1 h before incubation with soluble AβOs (1.25 μmol/L) for 24 h. The blot was labeled with Tau-5 antibody. The bar chart shows the semiquantitative analysis of the 17-kDa Tau fragment. Data are expressed as the mean±SEM of 3 independent experiments. cP<0.01 compared with the control group; fP<0.01 compared with the AβO-treated group.

PowerPoint slide

Full size image

Atorvastatin inhibited AβO-induced increases in Tau-targeting kinase GSK3β and Cdk5 activity

GSK3β activity could play a central role in controlling apoptosis and in the development of Alzheimer's disease. Because GSK3β activity is inhibited by phosphorylation on Ser-9, we assayed the effect of atorvastatin on AβO-induced changes in the level of Ser-9 phospho-GSK3β. As shown in Figure 6A, AβO-treated hippocampal neurons demonstrated a remarkable reduction in the level of phospho-GSK3β. After pretreatment for 1 h with atorvastatin, we exposed hippocampal neurons to AβOs for 24 h. Atorvastatin treatment prevented the AβO-induced decrease in the level of Ser-9 phospho-GSK3β in a dose-dependent manner.

Figure 6
figure 6

Atorvastatin inhibited the AβO-induced increases in GSK3β and Cdk5 activities. (A) Representative immunoblot comparing phospho-GSK3β to total GSK3β in hippocampal neurons. Group data showing the normalization of phospho-GSK3β to total GSK3β, determined in each group from 3 experiments. (B) Representative immunoblot comparing p35 and the production of its proteolytic fragment, p25 (an activator of Cdk5). Group data showing the normalization of p35 or p25 to β-actin, determined in each group from 3 experiments. Cultured hippocampal neurons at d 14 were pre-incubated with a vehicle solution (control, 0.1% DMSO) or atorvastatin (0.5, 1, or 2.5 μmol/L) for 1 h and then exposed to AβOs (1.25 μmol/L) for 24 h in the presence of vehicle or atorvastatin. Data are expressed as the mean±SEM of 3 independent experiments. cP<0.01 compared with the control group. fP<0.01 compared with the AβO-treated group.

PowerPoint slide

Full size image

We next investigated whether atorvastatin inhibited the AβO-induced increase in Cdk5 activity, which is widely believed to regulate normal and pathological Tau activity27. Because p25 is generated by proteolytic processing of p3528, we indirectly assessed Cdk5 activity by immunoblotting with an antibody that recognizes both p25 and p35 and determined the relative abundance of each. Again, we observed significant production of p25 after AβO treatment. Pretreatment of neurons with atorvastatin effectively diminished the AβO-mediated Cdk5 activity (Figure 6B).

Atorvastatin suppressed AβO-induced downregulation of phospho-Akt and phospho-Erk1/2 protein expression

GSK3β activity has been shown to be regulated by Akt29. More specifically, Akt can phosphorylate GSK3β at serine 9, thereby suppressing GSK3β kinase activity30. Akt is activated by phosphorylation at serine 473, and treatment of neurons with AβOs significantly reduces the levels of phospho-Akt (Figure 7A). Therefore, we tested the effect of atorvastatin on the AβO-induced reduction in phospho-Akt. Pretreatment of neurons with atorvastatin effectively prevented the AβO-induced decrease in phospho-Akt levels (Figure 7A). In addition, AβO-treated hippocampal neurons demonstrated a significant reduction in the levels of phosphorylated Erk1/2 (Figure 7B). Pretreatment of neurons with atorvastatin also inhibited the AβO-induced decrease in phospho-Erk1/2 (Figure 7B).

Figure 7
figure 7

Atorvastatin prevented AβO-induced downregulation of phospho-Akt and phospho-Erk1/2 expression. (A) Representative immunoblot comparing phospho-Akt to total Akt in hippocampal neurons after the treatments indicated. Group data showing the normalization of phospho-Akt to total Akt as determined in each group from 3 experiments. (B) Representative immunoblot comparing phospho-Erk1/2 in hippocampal neurons after the treatments indicated. Group data showing the normalization of phospho-Erk1/2 to total Erk1/2 as determined in each group from 3 experiments. Data are expressed as the mean±SEM of 3 independent experiments. cP<0.01 compared with the control group; eP<0.05, fP<0.01 compared with the AβO-treated group.

PowerPoint slide

Full size image

Discussion

Aβ and Tau are the two major aggregating proteins in AD. The toxic effects of Aβ are upstream of changes in Tau; however, Tau is considered necessary for the progression of neurodegeneration6,7. The aim of this study was to determine whether atorvastatin could prevent AβO-induced Tau cleavage and neurotoxicity in cultured hippocampal neurons. We chose rat hippocampal neurons as an experimental system because the hippocampus is affected early in AD. AβOs were prepared from Aβ1–42. This preparation contains a mixture of AβOs plus some monomers17. We assessed the effects of atorvastatin on hippocampal neurons that were acutely exposed to AβOs. These conditions are likely to be more relevant to early-stage AD, when hyperphosphorylated and fragmented Tau have not reached detrimental concentrations31. The results described above indicate that atorvastatin blocked AβO-induced calpain and caspase-3 activation in cultured hippocampal neurons. Furthermore, atorvastatin prevented Tau cleavage, leading to the generation of toxic fragments. Through this mechanism, atorvastatin could ameliorate the synaptic impairment and neurite degeneration associated with the pathological cleavage of proteins. Our findings identify an alternative molecular mechanism by which atorvastatin could, at least to some extent, protect hippocampal neurons from AβO-induced neurotoxicity.

Several clinical studies have previously shown that treatment with statins reduced the risk of AD32,33 and improved cognitive function and neuropathological changes in AD patients34,35. Recently, it has been reported that statins could reduce the number of senile plaques and Tau-positive neurites in APP transgenic mice19. Of particular interest, some studies have shown that statin treatment markedly reduced Tau levels and NFTs in cellular and mouse models of tauopathy18,36. A clinical observation showed that simvastatin decreased the amount of phosphorylated Tau in the cerebrospinal fluid of AD patients37. Statins were also reported to reduce NFTs more significantly than senile plaques in cognitively normal subjects35. These observations suggest that statins have a direct effect on Tau pathology regardless of plasma cholesterol levels. In addition, atorvastatin suppressed Tau hyperphosphorylation induced by excess cholesterol in rat brains38. In agreement with these findings, our present results showed that atorvastatin treatment could inhibit the AβO-induced activation of calpain and caspase-3 that leads to the generation of Tau-cleaved products with toxic functions in hippocampal neurons.

Direct evidence that Tau is involved in Aβ-induced neurotoxicity in central neurons has recently been obtained6. This study showed that neurons expressing either mouse or human Tau proteins degenerated in the presence of Aβ, whereas Tau-depleted neurons showed no signs of degeneration in the presence of Aβ6. More recently, it has been suggested that proteolytic cleavage of Tau is also involved in the pathogenesis of AD. Caspase-3- and calpain-mediated Tau cleavage appeared to occur early in Aβ-induced neurotoxicity and to precede Tau hyperphosphorylation2,5,9. Caspase-3-truncated Tau has been detected before the formation of NFTs and cell death39. In addition, Aβ-induced calpain-mediated Tau cleavage that leads to the generation of the 17-kDa fragment has also been detected before enhanced Tau phosphorylation or neurite degeneration in cultured hippocampal neurons5. Time-course experiments showed that the most rapid events upon treatment of hippocampal neurons with Aβ were induction of calpain and caspase-3/7 activity. Furthermore, the early protease activity affected Tau function, degraded full-length Tau and created low molecular weight Tau fragments2. Indeed, calpain inhibition completely eliminated Aβ-mediated production of 24- and 17-kDa Tau fragments, and caspase inhibition reduced the production of the 17-kDa fragment. How cleaved Tau causes neuronal degeneration and/or neuronal death is not clear. One potential mechanism involves truncated Tau, which lacks the C-terminal 20 aa, assembles more rapidly into filaments than full-length Tau9. The formation of Tau filaments could then interfere with different cellular processes. Another potential mechanism is that Aβ-mediated production of Tau fragments might promote Tau aggregation, which could be inherently toxic, as has been suggested in previous cell culture studies40. In addition, cell death could be caused by the loss of normal Tau activity leading to misregulation of microtubule dynamics and microtubule function41,42. Another possibility is that Tau fragment-mediated neuronal death occurs through the activation of N-methyl-D-aspartate (NMDA) receptors43. These results emphasize the potential of calpain and caspase-3 as therapeutic targets in neurodegenerative diseases. In our study, we found that the inhibitory effect of atorvastatin on the AβO-induced activation of calpain and caspase-3 was concentration-dependent. Our finding that atorvastatin was effective in preventing the AβO-induced increase in calpain- and caspase-3-mediated Tau cleavage provides insight into the potential mechanism discussed above.

We also found that treatment of hippocampal neurons with AβOs induced clear redistribution of Tau into the cell bodies. These findings are similar to the results of a study by Zempel et al44. In addition to early loss of synapses, the missorting of endogenous Tau from mainly axonal to somatodendritic compartments is among the earliest visible changes in AD. In dendritic regions from which Tau has been lost there is a depletion of spines, a local increase in Ca2+, and a breakdown of microtubules. Tau in these regions shows elevated phosphorylation at certain sites that are diagnostic of AD-Tau44. Atorvastatin pretreatment prevented AβO-induced Tau missorting. This might be an alternative mechanism by which atorvastatin prevents AβO-induced synaptotoxicity. The mechanisms responsible for this effect require further examination.

In the present study, we also found that AβO treatment significantly reduced levels of phospho-Akt and phospho-GSK-3β in hippocampal neurons, suggesting decreased activity of phospho-Akt and increased activity of GSK-3β. These findings are similar to the results of a study by Reifert et al2. It has been reported that within minutes of AβO treatment, a progression of calcium/calmodulin-dependent CaN activation is observed, first in dendritic spines and minutes to hours later in the cell body45. Activated CaN dephosphorylates inhibitor-1, which leads to activation of protein phosphatase 1 (PP1)46. PP1 activates GSK3β by dephosphorylation of Ser-947. It has been postulated that GSK3β activity might have a central role in the development of AD. GSK3β activity was implicated in Tau phosphorylation, APP processing, Aβ production, and neurodegeneration48. In addition, AβO-induced GSK3β over-activity could lead to glutamate receptor internalization, spine retraction, LTP blockage, and LTD facilitation4,49. The blockade of either GSK3β expression or activity decreased Aβ production and plaque accumulation50, improved performance in memory tests, preserved the dendritic structure, and reduced the Tau-dependent pathology in AD transgenic models51,52. In our study, we found that atorvastatin pretreatment prevented AβO-induced decreases in Ser-9 phosphorylation in hippocampal neurons, suggesting decreased activity of GSK3β. Additionally, our previous study demonstrated that atorvastatin enhanced neuronal neurite outgrowth through up-regulation of the Akt/GSK3β signaling pathway21.

Consistent with previous reports2,53, we observed that phosphorylation of Erk1/2 was inhibited by AβOs in hippocampal neurons. However, our observation appears to contradict earlier reports54,55 that suggested that synthetic Aβ causes activation of Erk1/2 in hippocampal slices or slice cultures. There are three key differences among these studies. First, we examined the direct effect of soluble AβOs on cultured hippocampal neurons, whereas Dineley et al documented the effect of Aβ1–42 on hippocampal slices. Second, our experiments used a much higher concentration of soluble AβOs than the nanomolar quantities of Aβ used by Dineley et al. Third, Dineley et al reported rapid activation of Erk within 5 min that was followed by inactivation by 10 min.

In summary, our results indicated that atorvastatin prevented AβO-induced calpain- and caspase-3-mediated Tau cleavage. The decrease in the levels of Tau proteolytic forms was accompanied by increased cell survival in AβO-treated neurons. More studies are needed to fully elucidate the mechanisms by which atorvastatin prevents AβO-induced Tau cleavage and subsequent neuronal degeneration. Nevertheless, these observations provide further evidence of the beneficial effects of statins in the prevention of AβO-induced toxicity in central neurons.

Author contribution

Ying JIN designed the research; Ling-ling ZHANG, Hai-juan SUI, and Zhou LIU performed the research; Hai-juan SUI analyzed the data; and Ying JIN wrote the paper.