Inhibition of the mitochondrial calcium uniporter inhibits Ab-induced apoptosis by reducing reactive oxygen species-mediated endoplasmic reticulum stress in cultured microglia
Abstract
Amyloid-beta (Ab) has been shown to induce microglial apoptosis, which is itself sensitive to disturbed mitochondrial calcium (Ca2+) homeostasis. The mitochondrial calcium uniporter (MCU) plays an impor- tant regulatory role in mitochondrial Ca2+ homeostasis, but its role in Ab-induced microglia apoptosis is unknown. In this study, we found increased mitochondrial Ca2+ concentration in Ab-treated primary microglia and BV-2 cells; also, the MCU inhibitor Ru360 significantly attenuated Ab-induced microglial apoptosis, whereas the MCU activator spermine augmented it. In addition, Ru360 significantly attenuated Ab-induced mitochondrial reactive oxygen species (ROS) production, as well as endoplasmic reticulum (ER) stress characterized by glucose-regulated protein 78 (GRP78) and C/-EBP homologous protein (CHOP) expression. Spermine, however, exerted the opposite effects on mitochondrial ROS production and ER stress. We also found that mitochondria-targeted antioxidant (Mito-TEMPO) treatment decreased GRP78 and CHOP expression in Ab-treated microglia. Moreover, blocking endogenous CHOP expression using a CHOP small interfering RNA (siRNA) attenuated Ab-induced cell death. Altogether, our data sug- gested that 1) inhibition of MCU exerts a neuroprotective effect on Ab-induced microglia apoptosis, and
2) that the underlying mechanism may be related to reducing mitochondrial ROS-mediated ER stress.
1. Introduction
Neuroinflammation is a hallmark of Alzheimer’s disease (AD), the progressive neurodegenerative disorder affecting elderly peo- ple worldwide that is characterized by amyloid-beta (Ab) deposi- tion in the brain. AD-related neuroinflammation is largely mediated by microglia, the brain’s main immune cells (Wes et al., 2016). Normally active microglia clear up aggregated pro- teins such as Ab, but sustained activation (‘‘overactivation”) causes release of cytotoxins that lead to neurotoxicity. Alternatively, persistent activation of microglia ultimately culminates in apoptosis, contributing to a subsequent uncontrolled inflammatory response (Hao et al., 2013). However, the underlying mechanisms regulating Ab-induced microglial apoptosis remain obscure.
Mounting evidence suggests that functions of activated micro- glia like phagocytosis and the release of proinflammatory factors such as tumor necrosis factor-a (TNF-a) and nitric oxide (NO) are Ca2+ dependent(Farber and Kettenmann, 2006; Ikeda et al., 2013). The endoplasmic reticulum (ER) is a major intracellular calcium storage pool, and is sensitive to changes in intracellular homeosta- sis. ER-mitochondria Ca2+ transfer is thought to be involved in Ab- induced apoptotic neuronal cell death (Ferreira et al., 2015). The mitochondrial calcium uniporter (MCU) is a selective Ca2+ ion channel localized in the inner mitochondrial membrane that is required for Ca2+ buffering (Baughman et al., 2011). Excessive uptake of Ca2+ into mitochondria through MCU is detrimental to mitochondrial function and leads to mitochondrial reactive oxygen species (ROS) production in AD (Toglia et al., 2016). Oxidative stress can disrupt ER function, and induces ER stress, in which the ER activates an unfolded protein response (UPR) to promote correct protein folding and degrade abnormally folded proteins. However, if the stress is prolonged, it may lead to apoptosis. Recent studies suggest that Ab leads to ER stress, and that mitochondrial dysfunction can enhance the Ab-induced ER stress response (Costa et al., 2013). However, the role of MCU in Ab-induced ER stress in microglia has never been examined.
In the current study, we investigated the role of MCU in Ab- induced apoptosis in primary microglia and BV-2 cell cultures; we analyzed its possible mechanism using the MCU activator sper- mine (Zhang et al., 2006) and inhibitor Ru360 (Zazueta et al., 1999). In addition, we used Mito-TEMPO treatment to determine whether mitochondrial ROS production is linked to ER stress, and C/-EBP homologous protein (CHOP) small interfering RNA (siRNA) inhibition to confirm the role of ER stress in Ab-induced microglial apoptosis.
2. Results
2.1. Effect of Ru360 and spermine on Ab-induced microglia apoptosis
To determine the effect of Ru360 and spermine on Ab-treated microglia, BV-2 and primary microglia were pre-incubated with Ru360 and spermine, respectively, at different concentrations for 1 h, prior to 24 h of exposure to Ab (20 lM). Cell viability was detected using a 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazo lium bromide (MTT) assay. As shown in Fig. 1A, Ab significantly reduced cell viability, and incubation with Ru360 (1–10 lM) concentration-dependently reversed Ab-induced cell death in pri- mary microglia. However, incubation with spermine (2–20 lM) concentration-dependently enhanced Ab-induced cell death in pri- mary microglia. In addition, Ru360 and spermine themselves had no detectable effects on cell viability (Fig. 1C, D). Similar results were observed in BV-2 cells treated with Ab (data not shown).
To determine whether apoptosis played a role in Ab-induced cell death, we pretreated primary microglia and BV-2 cells for 1 h with 5 lM Ru360 or 10 lM spermine, and then treated them with 20 lM Ab for 24 h. Apoptotic cells were analyzed either via a ter- minal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay (Fig. 2A) and flow cytometric analysis (Fig. 2B). We showed that Ru360 significantly decreased microglia apopto- sis, while spermine increased microglial apoptosis compared to cells treated with Ab alone.
2.2. Effect of Ru360 and spermine on mitochondrial Ca2+ concentration change
As shown in Fig. 3, mitochondrial Ca2+ concentration increased with Ab treatment compared with the control group in primary microglia, and pretreatment with Ru360 significantly decreased mitochondrial Ca2+ concentration compared with Ab treatment alone. However, pretreatment with spermine augmented the Ab- induced mitochondrial Ca2+ concentration increase. Similar results were observed in BV-2 cells treated with Ab (data not shown).
2.3. Effect of Ru360 and spermine on mitochondrial ROS production
Our results showed that the level of mitochondrial ROS forma- tion increased with Ab treatment relative to that of the control group in primary microglia, and pretreatment with Ru360 signifi- cantly decreased the levels of mitochondrial ROS compared with Ab treatment alone (Fig. 4). However, pretreatment with spermine increased Ab-induced mitochondrial ROS production versus Ab treatment alone (Fig. 4). Similar results were observed in BV-2 cells treated with Ab (data not shown).
2.4. Effect of Ru360 and spermine on glucose-regulated protein (GRP78) and CHOP expression
Our results showed that the levels of GRP78 and CHOP increased after Ab treatment compared with the control group in primary microglia, and pretreatment with Ru360 significantly decreased the expression of GRP78 and CHOP. However, pretreat- ment with spermine increased the expression of GRP78 and CHOP compared to Ab treatment alone (Fig. 5). Similar results were observed in BV-2 cells treated with Ab (data not shown).
2.5. Mitochondrial ROS production modulated activation of ER stress in Ab-treated microglia
Mito-TEMPO has been recently reported as a mitochondria- targeted antioxidant with low toxicity, making it a potential candi- date for mitochondrial ROS experiments (Trnka et al., 2009). Our results showed that the elevated level of Ab-induced mitochondrial ROS was effectively reduced by treating primary microglia with Mito-TEMPO at a concentration of 200 lM (Fig. 6A). In addition, Mito-TEMPO treatment led to a decrease in the expression of GRP78 and CHOP (Fig. 6B). Similar results were observed in BV-2 cells treated with Ab (data not shown). Altogether, these results indicated that mitochondrial ROS production modulated the acti- vation of ER stress in Ab-treated microglia.
2.6. ER stress played an important role in Ab-induced toxicity in microglia
Having determined that ER stress was upregulated following Ab exposure, we attempted to link the ER stress with Ab-induced tox- icity in microglia. Since CHOP is critical in ER stress-induced apop- tosis, we next sought to determine its role in Ab-induced cell death using siRNA to block endogenous CHOP expression in BV-2 cells. We found that there was a significant reduction in CHOP protein in BV-2 cells transfected with CHOP siRNA compared to the cells treated with scrambled (non-specific) siRNA (Fig.7A). Cell viability was examined in BV-2 cells transfected with CHOP or scrambled siRNA, and exposed to Ab. As shown in Fig. 7B, the viability of cells transfected with CHOP siRNA was significantly higher, suggesting that ER stress was a key mechanism in regulating Ab-induced microglial cell death.
3. Discussion
Previous studies have shown that ER-mitochondria crosstalk is involved in Ab-induced neuronal apoptosis, and that mitochondrial dysfunction enhances the ER stress response induced by Ab (Costa et al., 2012). ER stress can activate the UPR to restore homeostasis. The UPR is initiated by activation of three ER-resident transmem- brane proteins: PERK, IRE1 and ATF6. Under physiological condi- tions, the luminal domains of the three proteins are occupied by the ER chaperone GRP78, which inactivates them. When the UPR is initiated, GRP78 dissociates from the three transmembrane pro- teins, leading to their activation. Our results demonstrated that the expression of GRP78 was increased in Ab-treated microglia. Ru360 significantly decreased the mitochondrial Ca2+ concentration and prevented increased expression of GRP78, while spermine increased the mitochondrial Ca2+ concentration and augmented the expression of GRP78. In addition, Ru360 and spermine them- selves had no effects on the levels of mitochondrial Ca2+ concentra- tion under normal conditions. We suspected that Ru360 and spermine exerted effects based on MCU activation stimulated by Ab treatment (de Jesus Garcia-Rivas et al., 2005; Yu et al., 2016). These results suggest that ER stress occurred upon microglial exposure to Ab, and inhibition of MCU could attenuate Ab-induced ER stress in microglia. This is consistent with the hypothesis that mitochondrial dysfunction enhances the ER stress response induced by Ab in neurons.
ER stress usually occurs for a short time to allow the recovery of ER homeostasis needed for cell survival, but prolonged ER stress leads to apoptosis (Zhang et al., 2012). ER stress induces apoptotic cell death through activation of caspase-12, CHOP and the JNK pathway (Shimodaira et al., 2014). CHOP is expressed at low levels under physiological conditions, but it is highly induced in response to ER stress at the level of transcription. CHOP can induce ER stress-mediated apoptosis through down-regulation of Bcl-2 expression and exaggerated production of ROS (McCullough et al., 2001). Our results showed that the expression of CHOP increased in Ab-treated microglia. Ru360 pretreatment prevented increased expression of CHOP, whereas spermine augmented its expression. Moreover, we also found that blocking CHOP expres- sion using siRNA attenuated Ab-induced cell death, which further confirmed the importance of ER stress in modulating Ab-induced microglial toxicity. These results suggest that inhibition of MCU could attenuated Ab-induced microglial apoptosis via an ER stress-mediated mechanism.
Oxidative stress contributes to the pathogenesis of degenerative neurological diseases such as AD (Huang et al., 2016). Mitochon- drial calcium overload plays an important role in this oxidative stress-induced neuronal cell death (Peng and Jou, 2010). In this study, we found that Ru360 prevented the overproduction of mito- chondrial ROS and reversed Ab-induced oxidative stress, while spermine potentiated it; these findings are consistent with previ- ous studies demonstrating that spermine could increase ischemia-reperfusion-induced oxidative stress (Dong et al., 2014; Zhang et al., 2014), and that Ru360 could reduce ROS production caused by iron overload in brain and cardiac cells (Sripetchwandee et al., 2013; Sripetchwandee et al., 2014). How- ever, a recent study has also shown that spermine or MCU overex- pression attenuated lead-induced oxidative stress, while Ru360 or MCU knockdown augmented it (Yang et al., 2014). Although the different effects of MCU activity in oxidative stress responses need to be further explored, together these studies imply that MCU plays an important role in oxidative stress mediation.
ER stress are closely linked to oxidative stress in many patho- logical conditions(Ashraf and Sheikh, 2015; Mota et al., 2015). Pre- vious research has demonstrated that Mito-TEMPO attenuates stress-induced apoptosis via regulation of mitochondrial ROS pro- duction (Wang et al., 2013). To confirm the importance of mito- chondrial ROS in the expression of ER stress in Ab-treated cells, we examined the changes of the UPR in response to Mito-TEMPO treatment. Our results demonstrated that the up-regulated expres- sion of GRP78 and CHOP were markedly suppressed by Mito- TEMPO pre-treatment in BV-2 cells, a finding consistent with pre- vious studies showing that redox signaling pathways govern the ER stress process (Wang et al., 2016; Zhang et al., 2016; Zhang et al., 2015). Consequently, our results indicate that mitochondrial ROS renders cells more susceptible to Ab-induced ER stress.
In conclusion, our findings suggest that inhibition of MCU atten- uates Ab-induced mcroglial apoptosis, and that it does so through modulation of ROS-mediated ER stress. Understanding the appar- ent links between ER stress, oxidative stress and MCU could lead to the development of therapeutic strategies targeting Abtaining. The experimental procedures were approved by the Com- mission of Zhengzhou University for ethics of experiments on animals in accordance with international standards (No. YFY2016015).
BV-2 cell culture. The cells were purchased from Cell Center of the Peking Union Medical College (Beijing, China) and cultured in DMEM containing 5% FBS and 1% penicillin/streptomycin. Cultures were incubated at 37 °C and 5% CO2 in a fully humidified incubator. Primary microglia and BV-2 cells were pre-treated with Mito- TEMPO for 1 h, followed by stimulation with Ab for 24 h.
4. Experimental procedure
4.1. Cell culture
Microglial cultures. Primary microglial cells were isolated from mixed glial cultures as previously described (Xie et al., 2014). Briefly, primary mixed glial cultures were prepared from postnatal day 1–2 BALB/c mice. Primary microglial cells were co-cultured with astrocytes in DMEM containing 10% FBS and 1% penicillin/ streptomycin. After 10–14 days, microglia were harvested by gently shaking the culture and collecting the floating cells. The cells were seeded into tissue culture flasks and incubated at 37 °C for 1 h, and the nonadherent cells were removed by replacing the medium. The cells were grown in DMEM containing 10% FBS and incubated at 37 °C and 5% CO2. The purity of microglia was﹥ 95% verified by Recinus communis agglutinin-1 (RCA-1) Ab25-35 oligomers were prepared as described previously (Lobos et al., 2016; Tanokashira et al., 2017). Briefly, Ab25-35 peptide (Sigma, USA) were prepared as a dried hexafluoro-2-propanol (HFIP) film and stored at —80 °C. Prior to use, this peptide film was dissolved in sterile DMSO to 5 mM (stock solution). To prepare Ab25-35 oligomers, the solution was subsequently diluted to 100 lM using cold PBS, aged overnight at 4 °C and centrifuged at 14,000g for 10 min at 4 °C to remove any insoluble aggregates (protofibrils and fibrils). The supernatant containing soluble Ab25- 35 oligomers was transferred to clean tubes and stored at 4 °C. Only fresh Ab25-35 oligomers preparations (2 days-old maximum) were used in all experiments. During the experiment, the peptide stock solution was added directly to the solution bathing the cells to achieve a final concentration of 20 lM.
4.3. Cell viability assay
Cell viability was measured using the MTT assay. Briefly, cells were grown in 96-well plates. After incubating for 48 h, 20 ll of MTT solution was added to each well and the plates were further incubated at 37 °C for 4 h. The absorbance of each well was mea- sured at 570 nm on an Elisa plate reader using 200 ll MTT solubi- lization solution.
4.4. Determination of mitochondrial calcium concentration
Primary microglia and BV-2 cells were first washed with HEPES- buffered salt solution (HBSS) containing 140 mM NaCl, 5.4 mM KCl,
1.8 mM CaCl2, 0.8 mM MgCl2, 10 mM glucose and 10 mM HEPES, and then incubated with HBSS containing 1 lM rhod-2/AM (Life Technologies, Carlsbad, CA) at room temperature for 15 min according to the manufacturer’s instructions, as previously described (Peng et al., 1998; Sharma et al., 2000). The fluorescence was measured at an excitation wavelength of 575 nm and an emis- sion wavelength of 605 nm using a spectrophotometer.
4.5. TUNEL assay
The apoptotic cells were detected using the TUNEL assay and the In situ Cell Death Detection kit (Roche Diagnostics, Indi- anapolis, IN) per the manufacturer’s protocol as described previ- ously (Xie et al., 2010). The percentage of apoptotic cells was evaluated by counting approximately 500 cells.
4.6. Flow cytometry analysis
The apoptotic cells were quantified by flow cytometry as described previously (Toh et al., 2010). Briefly, cells were harvested and washed twice with ice-cold PBS and then resuspended in 200 ll binding buffer. Once resuspended, 5 ll of annexin V-FITC and 5 ll of propidium iodide were added to each sample, which was then incubated at room temperature in the dark for 10 min. Analysis was performed on a FACScan flow cytometer. All data are comprised of three independent experiments.
4.7. siRNA transfection
siRNA targeted against CHOP mRNA was obtained from Thermo Scientific Dharmacon RNAi Technologies (ON-TARGETplus SMART- pool). siRNA transfection of cells was performed according to the manufacturer’s instructions. Briefly, DharmaFECT-1 transfection reagent (Dharmacon) was combined with serum-free DMEM med- ium (Invitrogen Life Technologies) at room temperature for 5 min. CHOP or a scrambled siRNA control were added to the mixture described above and incubated at room temperature for 20 min, and then the mixture was added to the cells to achieve a concen- tration of 100 nM of the respective siRNAs. The cell culture plate was shaken for 5 s and then incubated at 37 °C for 24 h. Knock- down efficiencies were verified by western blotting.
4.8. Mitochondrial ROS production
Mitochondrial ROS production was assessed using the MitoSOX fluorophore (Invitrogen). Trypsinized primary microglia and BV-2 cells were stained with 5 lM MitoSOX at 37 °C for 15 min, washed in PBS, and then analyzed by flow cytometry. The ROS levels were expressed as arbitrary units of fluorescence intensity and normal- ized to the control group.
4.9. Western blot analysis
Western blotting was performed as described previously (Xie et al., 2014). Briefly, 30 lg protein was separated by SDS- polyacrylamide gel electrophoresis and then transferred to nitro- cellulose membranes. The membranes were incubated at room temperature in a blocking solution composed of 5% skim milk pow- der for 1 h. The membranes were then probed with antibodies against GRP78 (1:1000; CST, USA) or CHOP (1:1000; CST, USA).The same membranes were stripped and reprobed with anti-b- actin (1:4000; Santa Cruz, USA) antibodies. The membranes were then incubated with secondary antibody for 1 h at 37 °C. Immunoreactivity was visualized by chemiluminescence and exposure to a film. Band intensities were quantified by densitomet- ric analysis using a densitometer.
4.10. Statistical analysis
All data were expressed as mean ± S.D. The statistical signifi- cance was analyzed using one-way ANOVA and the Newman- Keuls test. All statistical analyses were performed using SPSS ver- sion 17.0 and P<0.05 was considered statistically significant.