Mitochondria damaged by Oxygen Glucose Deprivation can be Restored through Activation of the PI3K/Akt Pathway and Inhibition of Calcium Influx by Amlodipine Camsylate

Amlodipine, a L-type calcium channel blocker, has been reported to have a neuroprotective effect in brain ischemia. Mitochondrial calcium overload leads to apoptosis of cells in neurologic diseases. We evaluated the neuroprotective effects of amlodipine camsylate (AC) on neural stem cells (NSCs) injured by oxygen glucose deprivation (OGD) with a focus on mitochondrial structure and function. NSCs were isolated from rodent embryonic brains. Effects of AC on cell viability, proliferation, level of free radicals, and expression of intracellular signaling proteins were assessed in OGD-injured NSCs. We also investigated the effect of AC on mitochondrial structure in NSCs under OGD by transmission electron microscopy. AC increased the viability and proliferation of NSCs. This beneficial effect of AC was achieved by strong protection of mitochondria. AC markedly enhanced the expression of mitochondrial biogenesis-related proteins and mitochondrial anti-apoptosis proteins. Together, our results indicate that AC protects OGD-injured NSCs by protecting mitochondrial structure and function. The results of the present study provide insight into the mechanisms underlying the protective effects of AC on NSCs.

effect of Ac on nSc intracellular signaling. To investigate the effects of AC on signaling proteins associated with the survival of NSCs, we analyzed PI3K, phospho-Akt (Ser473), phospho-GSK3β (Ser9), and phospho-p44/42 MARK (Erk1/2) expression by western blotting (Fig. 1F). Immunoreactivities of PI3K, phospho-Akt (Ser473), phospho-GSK3β (Ser9), and phospho-p44/42 MARK (Erk1/2) were higher in NSCs co-treated with AC than in NSCs treated with OGD alone. To determine whether NSCs were affected by OGD, we also measured levels of HIF-1α in the NSCs by western blotting. HIF-1α expression was significantly increased in NSCs treated with OGD alone, but it was markedly decreased in NSCs co-treated with AC and OGD. AC (up to 10 μM) did not affect viability and was not cytotoxic to NSCs. (C) To measure the effects of AC on OGD-induced neurotoxicity, NSCs were simultaneously treated with several concentrations of AC (0, 0.1, 1, or 10 μM) and OGD for 8 h. AC up to 1 μM protected NSCs from OGD, but 10 μM AC was not protective. DAPI and TUNEL staining (D) and Annexin V/PI FACS analysis (E) showed that AC effectively blocked apoptotic cell death of NSCs induced by OGD. (F) AC restored the levels of each of these signaling proteins in NSCs under OGD. Data are means (% of control) ± SD from five independent experiments. Treatment groups were compared with the control group using Tukey's test after one-way ANOVA. *p < 0.05 and **p < 0.01 (vs. control group), # p < 0.05 and ## p < 0.05 (vs. the group treated only with OGD). Full-length blots/gels are presented in Supplementary Fig. S1.
Protective effect of AC on mitochondrial cristae ultrastructure and function in NSCs injured by oGD. Transmission electron microscopy (TEM) was used to evaluate ultrastructural changes of the cristae in the mitochondria of NSCs subjected to OGD and/or AC treatment (Fig. 3A). Structural enlargement of the mitochondrial matrix and deformation of the mitochondrial cristae were observed in NSCs treated with OGD for 8 hours compared with the control. However, co-treatment of NSCs with AC plus OGD for 8 hours resulted in significant preservation of mitochondrial cristae ultrastructure in a concentration-dependent manner. Low magnification TEM images are presented in Supplementary Fig. S3. In addition, mitochondrial function was evaluated in NSCs treated with OGD and/or AC by western blotting using mitochondrial biogenesis-related marker antibodies (Fig. 3B). Significantly higher expression of mitochondrial biogenesis-related proteins was observed in NSCs co-treated with AC and OGD than in cells treated with OGD alone. effect of Ac on oGD-induced free radical production and mitochondrial damage in nScs.
Oxidative damage was evaluated by detecting production of reactive oxygen species (ROS) in NSCs. NSCs were treated with re-oxygenation for 4 hours following OGD for 8 hours and ROS levels were assessed using DCF The OGD-induced decrease in proliferative activity of NSCs was significantly recovered by treatment with AC. BrdU assay (B) and Ki67 staining (C) also showed that AC effectively increased proliferation of NSCs subjected to OGD. (D) A cell migration assay showed that the migratory activity of NSCs was reduced by OGD but strongly increased by AC treatment. Data are means (% of control) ± SD from five independent experiments. Treatment groups were compared with the control group using Tukey's test after one-way ANOVA. *p < 0.05 and **p < 0.01 (vs. control group), # p < 0.05 (vs. the group treated only with OGD). Full-length blots/gels are presented in Supplementary Fig. S2. fluorescence (Fig. 4A). Treatment with OGD for 8 hours followed by re-oxygenation for 4 hours resulted in significantly increased free radical production in NSCs compared with the control group. However, co-treatment with AC (0.1 and 1 μM) resulted in a significant decrease in ROS production in a concentration-dependent manner. Additional analyses were conducted to evaluate mitochondrial changes in NSCs treated with OGD and/or AC. Intracellular calcium level and mitochondrial membrane potential were measured, and an ATP assay was performed. Intracellular calcium levels were significantly lower in NSCs co-treated with AC than in cells treated with OGD alone (Fig. 4B). In addition, mitochondrial membrane potential and ATP concentrations were significantly higher in NSCs co-treated with AC and OGD than cells treated with OGD alone (Fig. 4C,D). Using FACS, lower JC-1 monomer expression was also confirmed in NSCs co-treated with AC compared to NSCs treated with OGD alone (Supplementary Fig. S5).

Effect of AC on mitochondrial apoptosis-related proteins in NSCs.
The effect of AC on mitochondrial apoptosis-related proteins was evaluated in NSCs that were injured by OGD using western blotting (Fig. 4E). Expression of the survival-related protein Bcl-2 and mitochondrial cytochrome c was significantly increased in NSCs co-treated with AC than NSCs treated with OGD alone. However, the expression of the apoptosis-related protein Bax, cytosolic cytochrome c, and cleaved PARP decreased in response to co-treatment with AC compared to cells treated with OGD alone.
The PI3K/Akt pathways play a critical role in mediating AC-induced protective effects and restoration of mitochondrial damage in NSCs injured by OGD. The roles of the PI3K/Akt pathways in mediating the protective effect of AC on the NSCs were investigated. NSCs were pre-treated with LY294002, a PI3K inhibitor, before OGD and/or AC treatment. LY294002 blocked the protective effect of AC on NSCs damaged by OGD, and cell viability was decreased by approximately 18%, compared to NSCs that were not pre-treated with the PI3K inhibitor (Fig. 5A). The levels of the PI3k/Akt pathway related signaling proteins, and mitochondrial structure and function related proteins, were also measured. The expression of PI3K p110, phospho-Akt (Ser473), Drp1, Mfn1 and Mfn2 was significantly decreased in NSCs pre-treated with LY294002 ( Fig. 5B,C). showing that expression levels of proteins associated with mitochondria biogenesis of NSCs, such as phospho-p38, PGC1α, NRF-1, Drp1, Mfn 1, and Mfn 2, were decreased by OGD treatment but increased with AC treatment. Data are means (% of control) ± SD from five independent experiments. Treatment groups were compared with the control group using Tukey's test after oneway ANOVA. * p < 0.05 (vs. control group), # p < 0.05 (vs. the group treated only with OGD). Full-length blots/ gels are presented in Supplementary Fig. S4.

Discussion
Previously, we reported that amlodipine had possible neuroprotective effects on oxidative stress-injured cortical neurons and NSCs 5,6 . In the present study, we focused on the effects of AC on mitochondrial structure and function in OGD-injured NSCs. AC increased cell viability, PI3K survival signaling, and NSC proliferation (Figs 1 and 2). AC also effectively preserved the mitochondrial structure of NSCs based on TEM findings (Fig. 3A). In addition, AC enhanced the expression of mitochondrial biogenesis-related proteins (Fig. 3B) and inhibited the expression of mitochondrial anti-apoptosis proteins (Fig. 4). Based on these findings, we hypothesize that AC improves cell survival by protecting mitochondria and inhibiting apoptosis in OGD-injured NSCs. Although previous studies have reported possible neuroprotective effects of amlodipine, this study is the first to identify the protective effects of AC on mitochondria in OGD-injured NSCs 7,8,10,11 . In addition, we identified that the PI3K/ Akt pathway plays an important role in mediating the AC-induced protective effects, and restoration of mitochondrial damage in NSCs injured by OGD (Fig. 5) 15 .
Neurotoxicity is mediated by elevation of cytosolic Ca 2+ , and acute neural cell death correlates with the absolute amount of intracellular calcium 16,17 . Mitochondria are a major calcium-regulating organelle and mitochondrial calcium uptake and release mechanisms are key regulators of cell life or death 18,19 . Calcium overload in mitochondria and subsequent dysfunction leads to cell death after ischemic and traumatic brain injury 20,21 . The MPT pore is located in the inner mitochondrial membrane and allows solutes with molecular masses of up to 1500 Da to enter or exit the mitochondrial matrix 14 . Activation of the MPT pore is the most frequently observed consequence of extensive Ca 2+ accumulation in the mitochondria 20 . Opening of the MPT pore results in osmotic swelling of the mitochondrial matrix that is associated with a decrease in membrane potential and release of cytochrome c with other apoptotic factors 14,20 . Elevated Ca 2+ in mitochondria is thought to be a critical determinant of whether cells undergo oxidative stress-induced apoptosis 22 . Via its calcium inhibitory properties, amlodipine, a calcium channel blocker used as an antihypertensive drug 23 , can block calcium influx and reduce calcium storage in cells. Similar to our findings, a previous study reported that administration of azelnidipine, a L-type calcium antagonist, reduced not only cellular Ca 2+ accumulation but also mitochondrial Ca 2+ accumulation in renal cells 24 . In addition to stabilizing cellular and mitochondrial Ca 2+ homeostasis, azelnidipine protects renal cells from apoptosis induced by hypoxic injury through inhibition of the MPT, cytochrome c release, and downstream signaling cascades. Amlodipine is also an L-type calcium antagonist 25 . Therefore, we hypothesized that AC would inhibit intracellular calcium influx and reduce Ca 2+ accumulation in both the cellular cytoplasm with OGD for 30 m and free radical generation was measured using the fluorescent probe DCFH-DA. OGD significantly increased levels of intracellular reactive oxygen species (ROS). However, co-treatment with AC dramatically decreased ROS levels in NSCs. (B) Measurement of intracellular Ca 2+ levels suggested that intracellular Ca 2+ influx into NSCs was increased by OGD but is decreased by AC treatment. (C) Mitochondrial membrane potential was measured in NSCs under diverse conditions and it decreased significantly upon OGD treatment, whereas AC effectively restored membrane potential to almost normal. (D) ATP level decreased with OGD while AC increased ATP level. (E) Western blots show that expression of proteins associated with cell survival, such as Bcl-2 and mitochondrial cytochrome c, decreased in response to OGD treatment but increased with AC treatment. Proteins related to cell death, including Bax, cytosolic cytochrome c, and cleaved PARP increased with OGD treatment but decreased with AC treatment. Data are means (% of control) ± SD from five independent experiments. Treatment groups were compared with the control group using Tukey's test after oneway ANOVA. *p < 0.05 and **p < 0.01 (vs. control group), # p < 0.05 (vs. the group treated only with OGD). Full-length blots/gels are presented in Supplementary Fig. S6.  (2019) 9:15717 | https://doi.org/10.1038/s41598-019-52083-y www.nature.com/scientificreports www.nature.com/scientificreports/ and mitochondria of NSCs. This reduction in mitochondrial Ca 2+ accumulation could potentially inhibit MPT pore opening, thereby preventing cytochrome c release, the downstream cascade, and, apoptosis of NSCs, as we demonstrated in this study.
We found that AC effectively enhanced the expression of PI3K pathway proteins, ERK, and anti-apoptotic protein Bcl-2 in OGD-injured NSCs. Similar to our findings, a previous study showed that amlodipine markedly activated the PI3K/Akt pathway in neonatal rat cardiac muscle 4 . It is well documented that the PI3K/Akt signaling pathway plays a survival role and is involved in neuroprotection by increasing levels of downstream survival-related proteins including phosphorylated Akt, phosphorylated GSK-3β, and Bcl-2, and by decreasing levels of apoptosis-related proteins including cytosolic cytochrome c and cleaved caspase 3 [26][27][28][29][30] . A previous study also reported that pretreatment of human umbilical venous endothelial cells with amlodipine resulted in an increase in the Bcl-2/Bax ratio 31 . Bcl-2-family proteins can modulate intracellular Ca 2+ transport systems at the endoplasmic reticulum, mitochondria, and plasma membrane 32 . In addition, Bcl-2-family proteins finely regulate mitochondrial calcium uptake and can maximize the positive effects of Ca 2+ on mitochondrial bioenergetics while protecting the mitochondria from Ca 2+ overload leading to cell death 32 . A previous study reported that Bcl-2 proteins reduce Ca 2+ at the endoplasmic reticulum, leading to reduction of mitochondrial calcium and consequently, inhibition of cell apoptosis 19 .
We also observed in the present study that AC upregulated the expression of biogenesis-related proteins. Both PGC-1 and NRF-1 participate in mitochondrial biogenesis and respiration 33,34 . Phosphorylated dynamin-related protein (Drp) 1 reduces mitochondrial fission and promotes cell survival 35 . Mfn1 and 2 play important roles in mitochondrial fusion, and disruption of mitochondrial fusion results in mitochondrial heterogeneity and dysfunction 36 .
Based on our results, we propose the following mechanism for the protective effects of AC in NSCs under OGD (Fig. 6). The mechanisms that were not established in this study, were referenced from above studies [33][34][35][36] . Under OGD, extracellular calcium influx occurs in NSCs with an increase in ROS and hypoxia/ischemia related protein HIF-1α. This process causes mitochondrial structural changes and activates the apoptosis-related protein Bax in the mitochondrial membrane. Mitochondrial cytochrome c is then released from the mitochondria to the cytosol and this process increases caspase-3 activity. Ultimately, caspase-3 induces cleavage of PARP and cell apoptosis. In contrast, when NSCs under OGD are treated with AC, calcium influx is blocked and the PI3K cell survival signaling pathway is upregulated. In addition, the survival-related protein Bcl-2, which is activated by AC, may contribute to the maintenance of the mitochondrial membrane potential and inhibit release of mitochondrial cytochrome c. Furthermore, AC increases the expression of mitochondrial biogenesis-related proteins such as mitofusin (Mfn) which may serve to maintain mitochondrial morphology and function in NSCs under OGD. Together, these findings indicate that AC protects NSCs from OGD and further oxidative damage and promotes cell survival via mitochondrial protection and inhibition of apoptosis. NSCs were pre-treated with LY294002, a PI3K inhibitor, (10 µM) and the graph shows significant inhibition of the protective effect of AC on NSCs injured using OGD. (B) Western blots show that expression of the PI3K/ Akt pathway associated proteins increased in response to AC treatment in NSCs injured using OGD, and treatment with LY294002 decreased expression. (C) LY294002 also inhibited the effect of amlodipine on the expression of the mitochondrial structure and function-related proteins such as Drp1, Mfn1, and Mfn2 in NSCs injured using OGD. Data are means (% of control) ± SD from five independent experiments. Treatment groups were compared with the control group using Tukey's test after one-way ANOVA. *p < 0.05 and **p < 0.01 (vs. control group), # p < 0.05 and ## p < 0.01 (vs. the group treated only with OGD). Full-length blots/gels are presented in Supplementary Figs S7  www.nature.com/scientificreports www.nature.com/scientificreports/ The present study had several limitations. First, because this study was an in vitro study, the neuroprotective effects of AC need be re-confirmed in experimental animal models. Second, this study was conducted with AC only, which limits the generalizability of our findings. Therefore, neuroprotective effects of other amlodipine salt formulations should be investigated under the same conditions. However, considering our previous report that amlodipine besylate and AC have similar neuroprotective effects 5 , it is likely that other amlodipine salts might have similar protective effects.
In conclusion, our results provide evidence that AC protects OGD-injured NSCs by inhibiting cellular and mitochondrial calcium influx, activating the PI3K pathway, and enhancing the expression of mitochondrial biogenesis-related proteins. In addition, AC has a preservative effect on mitochondrial structure and the functions of NSCs injured by OGD. The results of the present study provide insight into the mechanisms underlying the protective effects of amlodipine on NSCs; namely, AC protect NSCs from OGD by protecting mitochondria.

Methods
Materials. AC was a kind gift from Hanmi Pharmaceutical Corporation (Seoul, South Korea). AC was dissolved in dimethyl sulfoxide with a final concentration in culture medium of 1% (vol/vol).

culture of neural stem cells and production of a hypoxic environment. All procedures involving
animals were performed in accordance with the Hanyang University guidelines for the care and use of laboratory animals and were approved by the Institutional Animal Care and Use Committee (IACUC) of Hanyang University. Every effort was made to minimize the number of animals used and to limit animal suffering. Each animal was used only once.
To evaluate the effect of AC on NSCs, cells were treated with several concentrations of AC alone for 1 and 8 h. Finally, to measure the protective effect of AC against OGD, we treated NSCs with several concentrations of AC under OGD for 1 and 8 h, as described above.
To investigate whether the PI3K/Akt pathway plays a critical role in mediating the neuroprotection afforded by AC against OGD in NSCs, we also treated NSCs with 10 μM LY294002, a PI3K inhibitor, 30 min before treatment with 1 μM AC in the background of OGD. Cell viability and expression of proteins were assessed using TBS and western blot, respectively. NSCs were separated into 4 groups as follows: control (group 1), 1 or 8 h OGD alone (group 2), 1 or 8 h OGD + 1 μM AC (group 3), 10 μM LY294002 + 1 or 8 h OGD + 1 μM AC (group 4). www.nature.com/scientificreports www.nature.com/scientificreports/ Trypan blue staining and the lactate dehydrogenase (LDH) release assay to measure cell viability.
For trypan blue staining, 10 μl aliquots of cells were incubated with 10 μl of trypan blue solution for 2 m. Unstained live cells were counted with a hemocytometer. A colorimetric assay kit (Roche Boehringer-Mannheim, IN, USA) was used to quantify LDH released from cultured NSCs according to the manufacturer's instructions. Cell viability was assessed using an ELISA plate reader (Synergy H1 Hybrid reader, BioTek Instruments, Winooski, VT, USA) at 490 nm with a reference wavelength of 690 nm. All results were normalized to the OD of an identical well without cells 41 . DApi and tUneL staining to evaluate apoptosis. NSCs were seeded on collagen-coated 13 mm diameter glass cover slips and treated with AC without exposure to OGD, without AC but with exposure to OGD for 8 h, or with exposure to OGD in the presence of AC (0.1 and 1 μM) for 8 h. Cells were then rinsed twice with PBS, air-dried, and fixed with 4% paraformaldehyde in PBS for 1 h at room temperature.
Annexin V and propidium iodide (PI) staining and detection. Annexin V/PI staining was measured using a FITC Annexin V Apoptosis Detection Kit I (Beckton Dickinson, Franklin Lakes, NJ, USA) according to the manufacturer's instructions. Briefly, cortical neurons were exposed to OGD with AC (0.1 and 1 µM) and incubated for 8 h at 37 °C. Cells were then rinsed twice with cold PBS and resuspended in 1× Binding Buffer. They were transferred (100 μl of the solution; 1 × 10 5 cells) to a 1.5 ml culture tube followed by application of 5 μl of FITC Annexin V and 10 μl PI. Samples were gently mixed and incubated for 15 m at room temperature in the dark. Finally, 400 μl of 1× Binding Buffer was added to each tube and samples were analyzed by flow cytometry (Accuri C6 Flow cytometer, BD Biosciences). Data were acquired and analyzed with BD Accuri C6 software. colony-forming unit assay. Proliferation of NSCs was measured via a colony-forming unit (CFU) assay.
Approximately 0.5 × 10 4 cells were seeded in a 60-mm grid plate and treated with OGD and AC (0.1 and 1 µM) for 8 h. Cells were then washed with DPBS, and the culture medium was changed. After 14 days, cells were washed again with DPBS and stained with 0.5% crystal violet (Sigma) in methanol for 30 m at room temperature. After staining, plates were washed with DPBS and allowed to dry. Colonies were counted under a dissecting microscope and colonies less than 2 mm in diameter or faintly stained were excluded.
Migration assays. Cell migration assays were performed using a QCM 24-well colorimetric cell migration assay kit (Chemicon, Temecula, CA, USA) according to the manufacturer's instructions. Briefly, OGD and AC (0.1 and 1 µM) were added into the upper chamber of the plate followed by incubation for 24 h at 37 °C. Cells that had migrated through the membrane were stained, and the number of migrated cells was determined by measuring the absorbance at 560 nm using a spectrophotometric plate reader (Synergy H1, BioTek Instruments) 42 .
BrdU cell proliferation assay. NSCs were incubated in BrdU-labeling medium (10 μM BrdU) for 5 h, and cell proliferation was measured using a BrdU Labeling and Detection Kit (Roche Boehringer-Mannheim, IN, USA) according to the manufacturer's instructions. Cell proliferation was assessed using an ELISA plate reader (Synergy H1, BioTek Instruments) at 370 nm with a reference wavelength of 492 nm. All results were normalized to the OD of an identical well without cells. electron microscopy. NSCs (1 × 10 6 ) were seeded on 150 mm dishes and exposed to OGD and AC (0.1 and 1 µM) for 8 h. Cells were then washed twice with PBS and fixed with EM fixing buffer for 1 h at room temperature. Twenty-four hours after MPP + (1-methyl-4-phenylpyridinium) treatment, cells were washed with DMEM and fixed in 0.1 M cacodylate buffer (pH 7.0, TED PELLA INC., Redding, CA, USA) containing 2% paraformaldehyde (Merck) and 0.5% glutaraldehyde (EMS, Hatfield, PA, USA) at 37 °C for 15 m, after which fixation was allowed to proceed at room temperature. Cells were washed with the same buffer three times, post-fixed with 1% osmium tetroxide (Heraeus, Hanau, Germany) for 30 m, and EM bloc-stained with 0.2% uranyl acetate (EMS) solution for 1 h. Subsequently, cells were dehydrated through an ascending ethanol series and embedded in an Epon mixture. Thin sections of 80 nm were cut using a Reichert-Jung Ultracut E ultramicrotome (Leica, Wetzlar, Germany) and mounted on a 200-mesh grid. Electron microscopy observation was performed with a Hitachi H-7500 transmission electron microscope (Hitachi, Ltd., Tokyo, Japan) with 80 kV acceleration voltage.
Determination of free radical production. To measure free radical production, NSCs were exposed to OGD for 30 m and incubated with the fluorescent probe 2′,7′-dichlorodihydrofluorescein diacetate (H 2 DCF-DA) (Molecular Probes Inc., Eugene, OR, USA) for 15 m. Then, NSCs treated with different concentrations of AC were exposed to OGD for 8 h followed by sample reoxygenation for 4 h. After incubation with 10 µM H 2 DCF-DA at 37 °C for 15 m, cells were washed with PBS three times. H 2 DCF-DA freely crosses cell membranes and is hydrolyzed by cellular esterases to 2′,7′-dichlorodihydrofluorescein (H 2 DCF), which is oxidized to 2′7′-dichlorofluorescein (DCF), which fluoresces in the presence of peroxides. Therefore, DCF fluorescence indicates the level of intracellular hydrogen peroxide, not superoxide. Accumulation of DCF in cells was assessed using an ELISA plate reader (Synergy H1 reader, BioTek Instruments) by measuring fluorescence (excitation 488 nm; emission 530 nm). All results were normalized to the OD of an identical well without cells.