Introduction

Neurons form functional networks by extending axons and dendrites (collectively termed neurites) that can connect via synapses to other neurons and cells. Molecules that promote effective neurite growth may have therapeutic potential in the treatment of a wide variety of disorders of the human nervous system. The statins are important cholesterol lowering agents for patients at risk of cardiovascular disease. However, many studies show that statins may have additional neuroprotective properties independent of their effect on cholesterol synthesis. Statins can reduce the region of focal cerebral ischemia1 and protect cortical neurons from excitotoxity2. Recent studies reported that statins can preventively and therapeutically attenuate traumatic brain injury and spinal cord injury. These beneficial effects of statins include upregulation of brain-derived neurotrophin factor (BDNF) and vascular endothelial growth factor (VEGF) and activation of the Akt-mediated signaling pathways3,4,5. Epidemiological studies have established a link between the use of statins and a lowered risk of developing Alzheimer's disease (AD)6. While it is unclear exactly how statin treatment leads to a reduction in risk, there is significant evidence to suggest that statins affect amyloid precursor protein (APP) processing and consequently reduce plaque formation7,8,9,10. In addition, the treatment of cells in culture with statins induces a number of striking morphological changes. For example, statin treatment has been reported to promote neurite outgrowth11,12,13,14. Furthermore, enhancement of neurite outgrowth by statins treatment is probably mediated by inhibition of Rho isoprenylation, a cholesterol-independent mechanism12. In addition to the Rho family of proteins, the serine/threonine kinase Akt, also known as protein kinase B, has also been revealed as a key mediator of several aspects of neurite outgrowth. Akt is the major effector of the phosphatidylinositol 3-kinase (PI3K) signaling pathway. Downstream of Akt, several substrates have been identified that are likely to play key roles in Akt-mediated neurite outgrowth, such as mammalian target of rapamycin (mTOR) and glycogen synthase kinase 3β (GSK-3β)15. mTOR is a serine/threonine protein kinase that regulates multiple cellular functions including neurite outgrowth. PI3K/Akt/mTOR signaling has been shown to promote growth and branching in hippocampal neurons16. The serine/threonine protein kinase GSK-3β is a key substrate. Activated Akt phosphorylates at GSK-3β at Ser-9 to inactivate its kinase activity17. It has been reported that statins can induce the phosphorylation of Akt3,18,19. These observations suggest the possibility that statins can affect neurite outgrowth by activation of the Akt signaling pathway. In the present study, we examine the effects of the widely employed hydrophilic statin atorvastatin on neurite growth in primary cultured cortical neurons and investigate in more detail the major signaling pathways that mediate these effects. Our data demonstrate that atorvastatin promotes neurite outgrowth. The mechanism mediating this effect involves the PI3K/Akt/mTOR and PI3K/Akt/GSK-3β signaling pathways.

Materials and methods

Atorvastatin was obtained from LKT Laboratories, Inc (St Paul, MN, USA). LY294002 and wortmannin (PI3K inhibition), LY303511 (inactive analogue of LY294002), PD98059 and U0126 (MEK inhibition), and rapamycin (mTOR inhibition) were from Calbiochem (Darmstadt, Germany). SB415286 (GSK3β inhibition) and triciribine (TCBN; Akt inhibition) were from Santa Cruz Biotechnology, Inc (Santa Cruz, CA, USA).

Primary cortical neuron cultures

Primary cultures were obtained from the cerebral cortex of 0- to 24-h-old Sprague-Dawley rats. After removal of the meninges and white matter, the cerebral cortex was collected in Hanks' solution without Ca2+ and Mg2+ (D-Hanks). The cortex was 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 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 then plated onto 35-mm cell culture dishes (∼150 cells/mm2) or 6-well culture plates pre-coated with poly-L-lysine (0.1 g/L, Sigma, St Louis, MO, USA) and kept at 37°C in a humidified atmosphere of 5% CO2/95% O2. Forty-eight hours after plating, the media were removed and replaced with DMEM containing glutamine (3 mg/mL), B-27 (2%; Life Technologies, Gaithersburg, MD, USA) and cytosine arabinofuranoside (final concentration 5 μmol/L, Sigma) to inhibit the proliferation of non-neuronal cells. After 4 d in culture, the cells were incubated for various lengths of time in media containing atorvastatin (0.05–10 μmol/L). For pharmacological experiments, inhibitors were added 30 min prior to addition of atorvastatin. Control cultures received a similar amount of DMSO, which was the organic solvent for all inhibitors. Microtubule-associated protein immunostaining indicated that about 90% of the cells were neurons after arabinofuranoside treatment for 3 d.

Immunofluorescence for MAP-2

To examine the effect of treatment on neurite outgrowth, neurons were incubated with media containing the appropriate test compound for an appropriate period of time. Following the treatment period, neurons were fixed with 4% paraformaldehyde/PBS at room temperature for 30 min and permeabilized in 0.2% Triton X-100/PBS for 5 min. Following washes with PBS, the neurons were incubated for 1 h in blocking solution (TBS containing 5% BSA at room temperature). Mouse monoclonal anti MAP-2 antibody (1:500, Abcam Inc, Cambridge, MA, USA) diluted in the blocking buffer was applied and incubated overnight at 4°C. After extensive washes, Cy3-conjugated anti-mouse secondary antibody (Life Technologies, Gaithersburg, MD, USA) was applied and incubated at 37°C for 1 h .

Neuronal imaging and analysis

Neuronal morphology was analyzed using an Olympus CKX41SF microscope, a Universal Imaging SPOT CCD camera and the SPOT Advanced Imaging software. Briefly, images of neurons were analyzed for total neurite branch length (TNBL), neurite number, terminal branch number, and soma area. TNBL and soma area were calculated using image measuring software, whereas neurite number and terminal branch number were counted manually. Five to six fields per well were chosen at random and only neurons distinguishable from neighboring neurons were evaluated to ensure the precision of the measurements; approximately 50–60 neurons were quantified per treatment group per experiment. The experimental group was kept blind until all analysis was done, and then the data from the same group of experiments were pooled together. Each experiment was performed in triplicate, unless otherwise noted.

Western blot analyses

Cells treated with atorvastatin or inhibitor described above were lysed in RIPA Lysis Buffer [Beyotime Institute of Biotechnology; 50 mmol/L Tris (pH 7.4), 150 mmol/L NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 1 mmol/L sodium orthovanadate, 10 mmol/L sodium fluoride, 1 mmol/L phenylmethylsulphonyl fluoride, and 1×Halt protease inhibitor cocktail], after which detergent-insoluble materials were removed by centrifugation at 12 000×g for 10 min. Protein concentration in the soluble fraction was measured using an Enhanced BCA protein assay kit (Beyotime Institute of Biotechnology, Haimen, China). Equal amounts of protein were then separated by SDS-PAGE, transferred onto nitrocellulose membranes, and probed with primary antibodies against the following proteins: rabbit anti-phospho-PDK1 (Ser241), rabbit anti-PDK1, rabbit anti-phospho-Akt (Ser473), rabbit anti-Akt, rabbit anti-phospho-PTEN (Ser380) (phosphatase and tensin homolog deleted on chromosome 10), rabbit anti-PTEN, rabbit anti-phospho-mTOR (Ser2448), rabbit anti-mTOR, rabbit anti-phospho-p70S6K (Thr389), rabbit anti-p70S6K, rabbit anti-phospho-4EBP1 (Thr37/46), anti-4EBP1, rabbit anti-phospho-GSK-3β (Ser9), and anti-GSK-3β (all from Cell Signaling Technology, Beverly, MA, USA and diluted 1:1000). Bound antibodies were detected with horseradish peroxidase-conjugated anti-rabbit immunoglobulin G (IgG) (Cell Signaling Technology) each diluted 1:2000 and Supersignal West Pico chemiluminescense substrate (Pierce, Rockford, IL, USA). Staining intensity was quantified from four blots derived from four independent experimental trials. The density of each band was quantified with Image J software and normalized to total kinase or β-actin expression. The protein levels reported in the figures were obtained as a ratio between the band intensity for the protein of interest and the band intensity of total kinases or β-actin (Sigma), used as loading control.

Statistical analysis

Statistical analyses were conducted using multifactor ANOVA including appropriate variables or t-test when suitable. Values represent mean±SEM. P values <0.05 were considered statistically significant.

Results

Atorvastatin increases neurite outgrowth and soma size in cortical neurons

To test the effects of atorvastatin on neurite outgrowth, we used dissociated postnatal cortical neuronal cultures as our model system. The initial set of experiments was designed to investigate whether atorvastatin affects neurite outgrowth in cultured cortical neurons. Atorvastatin (0.05–10 μmol/L) was added to cultures of cortical neurons at 4 DIV, at a stage in which neurites mature by elongating and branching. TNBL, neurite number, terminal branch number, and soma area were measured after an additional 48 h. As shown in Figure 1, incubation of cortical neurons with atorvastatin (0.05–10 μmol/L) for 48 h resulted in a dose-dependent increase in the soma size. Both the neurite number and terminal branch number were significantly increased, resulting in a net increase of TNBL. The maximum dose of atorvastatin was 10 μmol/L for neurite outgrowth. Therefore, we chose this treatment protocol to identify the underlying mechanisms of this event in all subsequent experiments. In our culture of cortical neurons, cells were classified as pyramidal or nonpyramidal on the basis of morphological features. However, there was no difference in the effect of atorvastatin on pyramidal and nonpyramidal neurons. In a time course analysis, increased TNBL and terminal branch number were detected as early as 12 h after atorvastatin treatment (Figure 2A and 2B). A significant increase in neurite number was detected at 24 h after atorvastatin treatment (Figure 2C). A significant increase in soma area was detected at 48 h after atorvastatin treatment (Figure 2D).

Figure 1
figure 1

Atorvastatin (Ator) promotes neurite growth in rat cortical neurons in a dose-dependent manner. (A) Examples of neurons taken using phase-contrast microscopy depicting cortical neurons either in the absence (left panel) or presence of 10 μmol/L atorvastatin (right panel). (B) Dendritic structures are confirmed by immunostaining for the dendritic marker MAP-2. Cultured cortical cells were treated at 4 DIV either with vehicle solution (Control, 0.1% DMSO) or with 10 μmol/L atorvastatin for 48 h. Following the treatment period, phase-contrast digital images of the cells were taken using a phase-contrast microscope. The graphs show mean±SEM for TNBL (C), neurite number (D), terminal branch number (E), and soma area (F). Data are from at least three independent experiments (n=50–60 cells). bP<0.05, cP<0.01 demonstrates statistical significance by one-way ANOVA, followed by least significant difference (LSD)'s post-hoc test.

Full size image
Figure 2
figure 2

Atorvastatin (10 μmol/L) significantly enhanced neurite outgrowth after 12, 24, and 48 h of treatment. The graphs show mean±SEM for TNBL (A), terminal branch number (B), neurite number (C), and soma area (D). Data are from at least three independent experiments (n=50–60 cells). bP<0.05, cP<0.01 demonstrates statistical significance by one-way ANOVA, followed by least significant difference (LSD)'s post-hoc test.

Full size image

Atorvastatin promotes neurite outgrowth via the PI3K-Akt pathway in cortical neurons

Having established that atorvastatin treatment stimulated neurite outgrowth, we next sought to determine the specific signaling pathways that contribute to these effects. At first, we focused on the PI3K-Akt and MEK-MAPK signaling pathways which have recently been shown to be important for the control of neurite outgrowth. In addition, statins are reported to activate the PI3K pathway, resulting in the phosphorylation of Akt20 in endothelial cells. Therefore, the requirement for PI3K activation was first examined. Pretreatment of the cortical neurons with the specific PI3K inhibitor LY294002 (30 μmol/L) significantly blocked atorvastatin-induced neurite outgrowth as reflected by neurite number, terminal branch number and soma area (Figure 3). Similar results were also observed with the use of another structurally unrelated PI3K inhibitor, wortmannin (5 μmol/L) (data not shown). In contrast the inactive analogue of LY294002, LY303511 (20 μmol/L) did not affect the effect of atrovastatin (data not shown). These results show that the activation of PI3K is required for atorvastatin-induced increases in neurite outgrowth in cultured cortical neurons. Pretreatment of the cortical neurons with the specific MEK inhibitor U0126 (10 μmol/L) also partly blocked the atorvastatin-induced increase in TNBL, terminal branch number, and neurite number. PD98059 (30 μmol/L; MEK inhibitor) showed similar results. Remarkably, the atorvastatin-induced increase in soma area was blocked by LY294002 but not by U0126 (Figure 3E). Concurrent application of LY294002 and U0126 to block both the PI3K and MAPK signaling pathways completely blocked atorvastatin-induced increases in neurite outgrowth. Interestingly, neither LY294002 nor U0126 changed any of the neurite parameters in control neurons (Figure 3). Together, the data reported above indicate that atorvastatin promotes neurite outgrowth and branching through both the PI3K/Akt and MAPK pathways, whereas atorvastatin increases soma area through only the PI3K/Akt pathway.

Figure 3
figure 3

Atorvastatin (Ator) increases neurite outgrowth and soma size through PI3K/Akt and MAPK pathways. (A) Cultured neurons at 4 DIV were pretreated with vehicle, LY294002 (LY, 30 μmol/L), U0126 (10 μmol/L), or tricribine (TCBN, 1 μmol/L) for 30 min and then exposed to atorvastatin (10 μmol/L) for 48 h in the presence of vehicle or inhibitor. Following the treatment period, phase-contrast digital images of the cells were taken using a phase-contrast microscope. The graphs show mean±SEM for TNBL (B), terminal branch number (C), neurite number (D), and soma area (E). Quantifications were performed from at least three independent experiments (n=50–60 neurons). bP<0.05, cP<0.01 compared with control group; eP<0.05, fP<0.01 compared with 10 μmol/L atorvastatin group.

Full size image

Pretreatment of the cortical neurons with tricribine, which blocks all Akt isoforms (Akt1, Akt2, and Akt3) without inhibiting known upstream activators PDK1 and PI3K, completely eliminated the effects of atorvastatin on TNBL, terminal branch number, neurite number, and soma size. Interestingly, tricribine alone also reduced soma size in control neurons (Figure 3).

To further ascertain the requirement for the activation of the PI3K/Akt signaling pathway in the atorvastatin-induced increases in neurite outgrowth, we used phospho-specific antibodies to measure the relative levels of phosphorylated, active forms of the protein kinases phosphoinositide-dependent kinase-1 (PDK1) and Akt after cultured neurons were treated with atorvastatin. For PDK1, we used an antibody specific for phosphorylated serine 241, which is on the activation loop of PDK1 and is essential for kinase activity. For Akt, we used antibody specific for phosphorylated serine 473, which is necessary for maximal activation of Akt21. As expected, incubation of cultured cortical neurons at 4 DIV with atorvastatin (0.1, 1, and 10 μmol/L) for 48 h caused marked increases in phospho-PDK1 and phospho-Akt (Ser473) in a concentration-dependent manner without a concurrent increase in the total levels of PDK1 and Akt as measured on Western blots of lysates of neurons (Figure 4A). Next, to further confirm that the observed effect of atorvastatin on neurite outgrowth is mediated by elevated level of PIP3, we tested whether atorvastatin treatment can downregulate the PIP3 phosphatase PTEN, which is a negative regulator of the PI3K signaling pathway. The results showed that atorvastatin treatment did not affect the levels of phospho-PTEN (Ser380) expression (Figure 4A). In a time course analysis, increased phospho-PDK1 levels were detected as early as 6 h and leveled off between 48 to 96 h after atorvastatin treatment (Figure 4B). It is well documented that PDK1 is downstream of PI3K, therefore, we determined whether LY294002 (30 μmol/L) or wortmannin (5 μmol/L) could block the atorvastatin-induced increases in the phosphorylation of PDK1 and Akt. We found that LY294002 (30 μmol/L) significantly blocked the increases in PDK1 and Akt phosphorylation induced by atorvastatin. Similar results were also observed with the use of wortmannin (5 μmol/L) (data not shown). LY303511 (20 μmol/L) did not affect this action of atrovastatin. None of LY294002, wortmannin, or LY303511 alone altered the basal levels of phosphorylated PDK1 and Akt (Figure 4C and 4D). Neither PD98059 (10 μmol/L) nor U0126 pretreatment blocked the action of atorvastatin on phosphorylated PDK1 and Akt (data not shown). Pretreatment of cortical neurons with tricribine (1 μmol/L) for 30 min completely blocked atorvastatin-induced increases in phospho-Akt levels. Tricribine (1 μmol/L) alone inhibited the basal levels of phosphorylated Akt (Figure 4D). Together, these results indicate that PI3K mediates the action of atorvastatin on PDK1 and Akt levels.

Figure 4
figure 4

(A) Representative Western blot showing that atorvastatin induced increases in phosphorylation of PDK1 and Akt in a concentration-dependent manner, whereas atorvastatin did not have significant effects on phospho-PTEN. Group data showing the normalization of phospho-PDK1, phospho-Akt, and phospho-PTEN proteins to total PDK1, Akt, and proteins was determined in each group of four experiments. (B) Representative Western blot showing the level of phospho-PDK1 taken at different times after adding atorvastatin (10 μmol/L). Group data showing the normalization of phospho-PDK1 proteins to total PDK1 protein was determined in each group of four experiments. (C) Representative Western blot showing that LY294002 inhibited atorvastatin-induced increases in phosphorylation of PDK1, whereas LY303511 did not have an effect. Group data showing the normalization of phospho-PDK1 to total PDK1 protein was determined in each group of four experiments. (D) Representative Western blot showing that both LY294002 and tricribine inhibited atorvastatin-induced increases in phosphorylation of Akt (Ser473). Group data showing normalization of phosphorylated Akt to total kinase was determined in each group of four experiments. Mean±SEM. bP<0.05, cP<0.01 compared with control group.

Full size image

Atorvastatin-induced neurite outgrowth is mTOR-dependent

mTOR is an important downstream node in the PI3K-Akt pathway. We next asked whether the increase in neurite outgrowth by atorvastatin occurs through an mTOR-coupled mechanism. We used rapamycin, a highly specific mTOR inhibitor, to further elucidate the downstream signaling mechanisms. This treatment decreased the phosphorylation of p70S6K on Thr389, which depends on mTOR activity22. As shown in Figure 5A–5E, pretreatment of cortical neurons with rapamycin (100 nmol/L) almost completely blocked atorvastatin-increased neurite outgrowth and soma size, indicating that a rapamycin-sensitive pathway is involved. We conclude therefore that atorvastatin stimulates neurite outgrowth in cultured cortical neurons in a rapamycin-sensitive manner.

Figure 5
figure 5

Atorvastatin enhances neurite outgrowth as well as soma size in an mTOR-dependent manner. (A) Cultured neurons at 4 DIV were pretreated with vehicle, rapamycin (100 nmol/L) for 30 min and then exposed to atorvastatin (10 μmol/L) for 48 h in the presence of vehicle or inhibitor. Following the treatment period, phase-contrast digital images of the cells were taken using a phase-contrast microscope. The graphs show mean±SEM for TNBL (B), terminal branch number (C), neurite number (D), and soma area (E). Quantifications were performed from at least three independent experiments (n=50–60 neurons in each experiment). bP<0.05, cP<0.01 compared with control group; eP<0.05, fP<0.01 compared with 10 μmol/L atorvastatin group. (F) Representative Western blot showing both LY294002 and tricribine inhibited atorvastatin-induced increases in phosphorylation of mTOR (Ser2448). Group data showing normalization of phosphorylated mTOR to total kinase was determined in each group of four experiments. Mean±SEM. bP<0.05, cP<0.01 compared with control group.

Full size image

To further ascertain the requirement for mTOR pathway activation in atorvastatin-induced neurite outgrowth, we used phospho-specific mTOR antibody (Ser2448), which has been shown to be important in the control of mTOR activity23,24,25, to measure the relative levels of phosphorylated, active forms of mTOR. As expected, atorvastatin (10 μmol/L) induced a significant increase in phosphorylated mTOR without a concurrent increase in the total level of this kinase (Figure 5F). Next, we tested the dependence of atorvastatin-induced mTOR phosphorylation on PI3K/Akt pathway. The results showed that LY294002 (30 μmol/L) and tricribine (1 μmol/L) completely blocked the atorvastatin-induced increase in phosphorylated mTOR (Figure 5F).

Atorvastatin stimulates the phosphorylation of p70S6K and 4E-BP1

mTOR stimulates translation by activating p70S6K, a kinase that increases synthesis of several components of the translation machinery24. The suppression of p70S6K impairs dendrite branching similarly to inhibition of mTOR16,26. We then asked if the atorvastatin-induced increases in neurite outgrowth are p70S6K-dependent by examining the phosphorylation state of p70S6K after atorvastatin treatment. We used an antibody specific for phosphorylated threonines 389, a site whose phosphorylation is vital for p70S6K activation27. When cultured cortical neurons were treated with atorvastatin (10 μmol/L) for 48 h, the phosphorylation of p70S6K was significantly increased. Furthermore, the atorvastatin-induced increases in the phosphorylation of p70S6K were completely blocked by LY294002 (30 μmol/L) and rapamycin (100 nmol/L) pretreatment, respectively (Figure 6A).

Figure 6
figure 6

Atorvastatin-induced increases in the phosphorylation of p70S6K and 4E-BP1 are PI3K- and mTOR-dependent. Cultured cortical neurons at 4 DIV were pretreated with vehicle, LY294002 (30 μmol/L) or rapamycin (100 nmol/L) for 30 min and then exposed to atorvastatin (10 μmol/L) for 48 h in the presence of vehicle or inhibitor. (A) Representative Western blot showing both LY294002 and rapamycin inhibited atorvastatin-induced increases in phosphorylation of p70S6K (Thr389). Group data showing normalization of phosphorylated p70S6K to total kinase was determined in each group of four experiments. (B) Representative Western blot showing both LY294002 and rapamycin inhibited atorvastatin-induced increases in phosphorylation of 4E-BP1 (Thr70). Group data showing normalization of phosphorylated 4E-BP1 to total kinase was determined in each group of four experiments. Mean±SEM. cP<0.01 compared with control group.

Full size image

In addition, because activation of mTOR can also contribute to translational initiation by phosphorylation of 4E-binding protein 1 (4E-BP1), which binds to and represses the function of the cap-binding translation factor elF4E, we asked if this regulatory protein may have a role in atorvastatin-induced neurite outgrowth. We used an antibody specific for phosphorylated threonines 37/46, a site those phosphorylation is required for the inactivation of 4E-BP1 and its dissociation from the elF4E complex28. Incubation of the cultured cortical neurons with atorvastatin (10 μmol/L) for 48 h increased the phosphorylation of 4E-BP1. Furthermore, atorvastatin-induced increases in the phosphorylation of 4E-BP1 was completely blocked by LY294002 (30 μmol/L) and rapamycin (100 nmol/L) pretreatment, respectively (Figure 6B). These results support the suggestion that activation of both p70S6K and 4E-BP1 are required for atorvastatin-induced neurite outgrowth.

Atorvastatin increases neurite outgrowth through the Akt/GSK3β pathway

Signaling via the Akt/GSK-3β pathway can regulate neurite morphology by regulating phosphorylation of MAP217. Akt phosphorylates GSK-3β to inactivate it. Inactivation of GSK-3β leads to decreased phosphorylation of MAP2 and its increased binding and stabilization of microtubules. To test whether GSK-3β mediates the effects of atorvastatin on neurite outgrowth, we treated cultured cortical neurons with SB415286, a specific GSK-3β inhibitor. GSK-3β inhibition by SB415286 (5 μmol/L) caused an increase in TNBL, terminal branch number, neurite number, and soma area. Treatment of neurons with a combination of atorvastatin (10 μmol/L) and SB415286 (5 μmol/L) caused an additive effect on TNBL, neurite number, terminal branch number, and soma area (Figure 7A–7E). To further test whether atorvastatin can promote neurite growth by regulating the Akt/GSK-3β pathway in cultured cortical neurons, we tested for phosphorylation of GSK-3β in cells treated with atorvastatin. Phosphorylation of GSK-3β increased in response to treatment with atorvastatin (10 μmol/L) (Figure 7F). To confirm that GSK-3β mediates the effect of atorvastatin on neurite outgrowth, we treated cultured cortical neurons with either LY294002 or tricribine. Pre-treatment with LY294002 (30 μmol/L) or tricribine (1 μmol/L) before adding atorvastatin completely blocked atorvastatin-induced phosphorylation of GSK-3β (Figure 7F). These treatments did not affect GSK-3β protein levels. Together, these results suggest that atorvastatin enhances neurite outgrowth through the Akt/GSK-3β pathway.

Figure 7
figure 7

Atorvastatin enhances neurite outgrowth through the PI3K/Akt/GSK-3β pathway. (A) Cultured neurons at 4 DIV were pretreated with vehicle or SB415286 (5 μmol/L) for 30 min and then exposed to atorvastatin (10 μmol/L) for 48 h in the presence of vehicle or inhibitor. Following the treatment period, phase-contrast digital images of the cells were taken using a phase-contrast microscope. The graphs show mean±SEM for TNBL (B), terminal branch number (C), neurite number (D), and soma area (E). Quantifications were performed from at least three independent experiments (n=50–60 neurons). bP<0.05, cP<0.01 compared with control group; eP<0.05 compared with 10 μmol/L atorvastatin group. hP<0.05, iP<0.01 compared with atorvastatin+SB415286 group. (F) Representative Western blot showing both LY294002 and tricribine inhibited atorvastatin-induced increases in phosphorylation of GSK-3β (Ser9). Group data showing normalization of phosphorylated GSK-3β to total kinase was determined in each group of four experiments. Mean±SEM. bP<0.05, cP<0.01 compared with control group.

Full size image

Discussion

This study demonstrates that atorvastatin promotes neurite outgrowth by activating the PI3K/Akt/mTOR and Akt/GSK-3β signal pathways in cultured cortical neurons. Our findings are consistent with previous reports that statins significantly induce neurite outgrowth in cultured hippocampal neurons12, explants of rat embryonic cortex and post-natal spinal cord13 and PC12 cells11. In contrast, statins have also been reported to inhibit neurite outgrowth in cultured cortical neurons, either by reducing cholesterol levels29 or by inhibition of isoprenylation30. However, these studies started statin treatment within 24 h of plating primary neuronal cells. In the present study, we grew the neurons in culture for 4 d before starting treatment to better understand the effect that statin treatment might have more developmentally mature neurons. A possible explanation for this discrepancy is increased sensitivity of immature neurons to toxicity induced by inhibition of mevalonate synthesis. Comparison of these results also suggests that statins may either stimulate or inhibit neuronal growth, depending on the stage of neuronal development. How can enhancement of neurite outgrowth by atorvastatin be explained? Our results are summarized in Figure 8, which will serve as a guide to the discussion. A previous report has demonstrated that pravastatin's stimulation of neurite outgrowth is mediated by reduction of RhoA signaling via inhibition of geranylgeranylation12. In addition to the Rho family proteins31, Akt has been revealed as a key mediator of several aspects of neurite outgrowth, including elongation, branching and caliber15. Several in vitro and in vivo studies provided evidence that statins activate the PKB/Akt pathway3,32,33. Though the mechanism of Akt activation by statins is not known, a study on mammalian endothelial cells demonstrated that simvastatin activates Akt in these cells, while treatment with wortmannin or LY294002 blocked this activation of Akt, suggesting that PI3-kinase signaling is involved33. Our study is the first demonstration to our knowledge that the PI3K/Akt pathway is involved in enhancement of neurite outgrowth by atorvastatin because this process was blocked by wortmannin or LY294002. Furthermore, Western blotting reveals that atorvastatin increased PDK1 and Akt phosphorylation. Therefore, activation of Akt by atorvastatin as demonstrated here is one explanation for the effect of the drug on neurite outgrowth in cortical neurons. The question remains of how atorvastatin activates specific protein kinase such as Akt. The involvement of Ras in the activation of PI3K has been reported34. It has been reported that lovastatin stimulates rapid activation of Ras, which associates with and activates PI3K in the plasma membrane, which in turn regulates Akt35. Based on these previous reports, we propose the hypothesis that atorvastatin activates Akt through activation of Ras and further facilitates neurite outgrowth in cortical neurons. We need to further clarify the specific downstream signal of Akt that promotes neurite outgrowth in the presence of atorvastatin.

Figure 8
figure 8

Schematic view of the effects of atorvastatin on PI3K pathways. Findings based on the present study are in bold and highlighted by shaded boxes. The inhibitors used in the study are boxed in red.

Full size image

Akt phosphorylates or interacts with a number of proteins that may positively influence the development of neuronal morphology. The PI3K/Akt pathway is known to regulate the mTOR pathway, and mTOR has been shown to be a direct substrate of Akt. PI3K/Akt/mTOR signaling has been shown to promote growth and branching in hippocampal neurons16. In recent reports, mTOR promoted axonal regeneration in the adult central nervous system and increased the axonal growth of injured peripheral nerves36,37. In our experiment, atorvastatin increased the activation of mTOR and p70S6K while rapamycin, an inhibitor of mTOR, decreased the effect of atorvastatin on neurite outgrowth. These results implicated mTOR as a signal of neurite outgrowth in neurons stimulated by atorvastatin. Akt is the main upstream regulator of mTOR in many cellular functions38, and in our experiments the inhibition of PI3K-dependent Akt activation by LY294002 decreased the activation of mTOR. In addition, both LY294002 and Akt inhibitor blocked neurite outgrowth facilitated by atorvastatin. Based on these results, we believe that atorvastatin increases the phosphorylation of mTOR and neurite outgrowth through PI3K and Akt. The PI3K pathway to mTOR is mediated by Akt and results in inhibition of tuberin, a GAP (GTPase-activating protein) for Rheb (Ras homolog enriched in brain); elevated GTP-bound Rheb then stimulates mTOR39,40,41. In addition, atorvastatin may activate the mTORC2 complex, thereby stimulating phosphorylation of Akt at Ser47342,43. mTOR activation stimulated translational initiation involving both 4E-BP1 and p70S6K22, which stimulate translation of 5′ capped and 5′TOP (5′-terminal oligo pyrimidine tract) mRNAs, respectively24,44. mTOR and its substrates 4E-BP1 and p70S6Ks are present in dendrites45. p70S6Ks requires hierarchical phosphorylation by mTOR and MAPK for optimal activation46. Similarly, 4E-BP1 is also phosphorylated by mTOR and MAPK at multiple sites47. These might be partially accountable for the involvement of both PI3K and MAPK in neurite formation. The phosphorylated 4E-BP1 is replaced by eIF4G, a scaffolding translation factor that assembles several critical components of the cap-dependent translation initiation complex on the mRNA via its interaction with cap-bound eIF4E48. The activated p70S6K then phosphorylates ribosomal protein S6, which can result in enhanced translation of 5′-oligopyrimidine tract-containing mRNAs that encode numerous components of translation machinery48. In our study, atorvastatin treatment leads to increased phosphorylation of 4E-BP1 and p70S6K, both of which are inhibited by rapamycin. Thus, activation of mTOR by atorvastatin may enhance both cap-dependent translation (eIF4E pathway) and initiation of 5′-oligopyrimidine tract-containing mRNAs (p70S6K-ribosomal protein S6 pathway). It has been shown that suppression of p70S6K or overexpression of 4E-BP1 impairs dendrite branching similarly to inhibition of mTOR16. Additional investigation into the potential role of p70S6K and 4E-BP1 in atorvastatin-induced neurite outgrowth would be of great interest.

Activated Akt phosphorylates GSK-3β at Ser9 to inactivate its kinase activity17; thus, GSK-3β activity is inversely correlated with PI3K/Akt signaling activity49. Some studies have observed GSK-3β phosphorylation following Akt activation and subsequently neurite outgrowth50,51. Our study demonstrates that treatment of cultured cortical neurons with atorvastatin increased phosphorylation of GSK-3β (Ser9). Inhibiting PI3K with LY294002 or directly inhibiting Akt with tricribine blocked the atorvastatin-induced increase in phosphorylated GSK-3β (Ser9). Another study also demonstrates that simvastatin increases phosphorylation of GSK-3β (Ser9) in the hippocampus in rats after traumatic brain injury3. GSK-3β, which is inactivated upon phosphorylation by Akt, is implicated in the regulation of microtubule dynamics by phosphorylating MAPs, including adenomatosis polyposis coli (APC), MAP1, and MAP252. Phosphorylation of MAPs is a key step in the regulatation of neurite initiation, growth, and branching in hippocampal neurons17. Decreased activity of GSK-3β reduces phosphorylation of MAP2, a step that can enhance microtubule polymerization and dendritic growth53.

Taken together, we have shown for the first time that atorvastatin can regulate neurite outgrowth via the PI3K/Akt/mTOR and PI3K/Akt/GSK-3β pathways in cultured cortical neurons. It will be important to investigate the effects of atorvastatin on neurite growth in vivo. We believe that statins have great potential for treating several nervous disorders. Our results indicate a novel cholesterol-independent effect of statins and call for further studies to reveal the responsible mechanisms.

Author contribution

Ying JIN designed the research; Hai-juan SUI, Yan DONG, Qi DING, Wen-hui QU, Sheng-xue YU, and Ying-xin JIN performed the research; Hai-juan SUI and Wen-hui QU analyzed the data; and Ying JIN wrote the paper.