Enhancing the Astrocytic Clearance of Extracellular Α-Synuclein

Cellular and Molecular Neurobiology (2019) 39:1017–1028 https://doi.org/10.1007/s10571-019-00696-2

ORIGINAL RESEARCH

Enhancing the Astrocytic Clearance of Extracellular α‑Synuclein Aggregates by Ginkgolides Attenuates Neural Cell Injury

Jun Hua1,2 · Nuo Yin1 · Shi Xu1 · Qiang Chen1,3 · Tingting Tao1,3 · Ji Zhang4 · Jianhua Ding1 · Yi Fan1,3 · Gang Hu1,5

Received: 9 November 2018 / Accepted: 1 June 2019 / Published online: 5 June 2019 © Springer Science+Business Media, LLC, part of Springer Nature 2019

Abstract The accumulation of aggregated forms of the α-Synuclein (α-Syn) is associated with the pathogenesis of Parkinson’s disease (PD), and the efcient clearance of aggregated α-Syn represents a potential approach in PD therapy. Astrocytes are the most numerous glia cells in the brain and play an essential role in supporting brain functions in PD state. In the present study, we demonstrated that cultured primary astrocytes engulfed and degraded extracellular aggregated recombinant human α-Syn. Meanwhile, we observed that the clearance of α-Syn by astrocytes was abolished by proteasome inhibitor MG132 and autophagy inhibitor 3-methyladenine (3MA). We further showed that intracellular α-Syn was reduced after ginkgolide B (GB) and bilobalide (BB) treatment, and the decrease was reversed by MG132 and 3MA. More importantly, GB and BB reduced indirect neurotoxicity to neurons induced by α-Syn-stimulated astrocytic conditioned medium. Together, we frstly fnd that astrocytes can engulf and degrade α-Syn aggregates via the proteasome and autophagy pathways, and further show that GB and BB enhance astrocytic clearance of α-Syn, which gives us an insight into the novel therapy for PD in future.

Keywords α-Synuclein · Astrocytes · Autophagy · Ginkgolide B · Bilobalide

Abbreviations BCA Bicinchoninic acid α-Syn α-Synuclein CM Conditioned medium ALP Autophagy–lysosome pathways DAPI 4,6-Linked amidine-2-phenylindole BB Bilobalide DLB Dementia with LBs DMSO Dimethyl sulfoxide DTT DL-Dithiothreitol Jun Hua and Nuo Yin have contributed equally to this work. GB Ginkgolide B * Yi Fan GBE Ginkgo biloba extract [email protected] GFAP Glial fbrillary acidic protein * Gang Hu LBs Lewy bodies [email protected] LC3 Microtubule-associated protein 1A/1B-light chain 3 1 Jiangsu Key Laboratory of Neurodegeneration, Department LDH Lactate dehydrogenase of Pharmacology, Nanjing Medical University, 101 Longmian Road, Nanjing 211166, Jiangsu, China LNs Lewy neuritis MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra- 2 Department of Neurology & Psychology, The Fourth Clinical Medical College of Guangzhou University of Chinese zolium Medicine, 1 Fuhua Road, Shenzhen 518033, Guangdong, PBS Phosphate bufered saline China PD Parkinson’s disease 3 Neuroprotective Drug Discovery Key Laboratory of Nanjing SNpc Substantia Nigra pars compacta Medical University, 101 Longmian Road, Nanjing 211166, ThT Thiofavin T Jiangsu, China 3MA 3-Methyladenine 4 Division of Clinical Pharmacy, Department of Pharmacy, UPS Ubiquitin proteasome system The First Afliated Hospital of Nanjing Medical University, Nanjing 210029, Jiangsu, China 5 Nanjing University of Chinese Medicine, 138 Xianlin Avenue, Nanjing 210023, Jiangsu, China

Vol.:(0123456789)1 3 1018 Cellular and Molecular Neurobiology (2019) 39:1017–1028

Introduction overexpression of PLK2 reduced intra-neuronal human α-Syn accumulation, and suppressed dopaminergic neu- Alpha-Synuclein (α-Syn), a 14 kDa protein (140 amino rodegeneration induced by α-Syn overexpression in a rat acids) physiologically found in presynaptic terminals of genetic model of PD (Oueslati et al. 2013). Moreover, neurons, is the major fbrillary protein in Lewy bodies clearance of α-Syn monomers and aggregates occurs via (LBs) and Lewy neuritis (LNs) (Vekrellis et al. 2011). direct proteolysis by matrix metalloprotease 9 (MMP-9) Alpha-Syn has been suggested to play important roles in (Iwata et al. 2003) or binding to molecular chaperones brain lipid metabolism, modulation of synaptic transmis- such as the proteasome (McNaught et al. 2002) and sion, and synaptic vesicle biogenesis (Kim and Lee 2008). autophagy (Oueslati et al. 2013). However, mounting evidence has indicated that α-Syn We have previously shown that aggregated α-Syn induced malfunction, especially the accumulation of misfolded cell apoptosis, and Ginkgo biloba extract (GBE), including α-Syn aggregates and genetic mutations, can also contrib- ginkgolide B (GB) and bilobalide (BB), attenuated aggre- ute to synucleinopathies such as dementia with LBs (DLB) gated α-Syn-induced cell apoptosis (Hua et al. 2017). In the and sporadic and inherited Parkinson’s disease (PD) (Kalia present study, we showed that astrocytes internalized and and Lang 2015; Spillantini et al. 1998). eliminated α-Syn aggregates in vitro, and GB and BB pro- PD is primarily characterized by selective loss of dopa- moted the degradation of intercellular α-Syn aggregates via minergic neurons in the substantia nigra pars compacta the autophagy and proteasome pathways. Moreover, GB and (SNpc) and abrogation of dopaminergic tone along the BB reduced the indirect neurotoxicity of α-Syn to neurons nigrostriatal pathway (Dauer and Przedborski 2003; Obeso in vitro. Herein, we elucidate that the promotion of astro- et al. 2017). However, previous studies have indicated cytic clearance of α-Syn may be a potential approach in PD that the progressive neuronal accumulation of aggregated therapy. α-Syn and the formation of LBs might trigger neuroinfammation, neurotoxicity, and synaptic dysfunction in PD (Gao et al. 2011; Ma et al. 2014; Volpicelli-Daley et al. Materials and Methods 2011). Notably, the release of α-Syn aggregates from neuronal cells was signifcantly increased under PD condition Experimental Animals and Drug Administration (Jang et al. 2010), and extracellular α-Syn aggregates may cause paracrine efects on neighboring cells. Transmission All study protocols were approved by the Institutional Ani- of α-Syn aggregates from neuron-to-neuron (Tyson et al. mal Care and Use Committee of Nanjing Medical University 2017) and neuron-to-glia has been suggested as the under- (IACUC). C57BL/6 mice were provided by Model Animal lying mechanism of neuroinfammation and neurotoxicity Resource Center (MARC, Nanjing, China). All mice were in PD (Choi et al. 2018). Hence, an efcient clearance of maintained on a 12/12 h light/dark cycle with free access to this protein might represent a potential approach in PD food and water. therapy. GB and BB (purity > 99%, supplied by the National Insti- Astrocytes are the most numerous and multifunctional tutes for Food and Drug Control, China) were dissolved in type of glia cells in the central nervous system (CNS). dimethyl sulfoxide (DMSO) as stock solutions for in vitro Astrocytes are crucial to the neuronal microenvironment studies, respectively. due to its abilities to maintain tissue structure, provide nutrition and energy to neurons, and regulate synaptic Cell Culture and Treatment development and neurotransmitter uptake. Previously studies have shown that extracellular α-Syn acted on neu- Primary astrocytes were prepared from brains of 1- to ronal Toll-like receptor 2 (TLR2) might result in neuro- 2-day-old C57BL/6 mice, as previously described (Zeng degeneration by inhibition of autophagy via AKT/mTOR et al. 2007). Briefy, neonates (P1–2) were killed by rapid signaling (Kim et al. 2013), while extracellular α-Syn decapitation. The midbrain was removed, and then the tis- activated microglia via NF-κB and p38 MAPK pathway, sue was dissociated with 0.25% tryptase (VWR Life Sci- thereby resulted in neuroinfammatory factors produc- ence, USA) at 37 °C and terminated by Dulbecco’s Modifed tion (Kim et al. 2015). While extracellular α-Syn aggre- Eagle’s Medium (DMEM, Gibco Lab., USA) supplemented gates can be taken up by neurons (Lee et al. 2008a, b) and with 10% Fetal Bovine Serum (FBS, Gibco Lab., USA), glial cells (Lee et al. 2008a, b) by endocytosis, it is not and 100 mg/mL streptomycin (Sigma-Aldrich Biotechnol- clear whether astrocytes are responsible for α-Syn clear- ogy, USA) and 100 U/mL penicillin (Gibco Lab., USA). ance in PD. A recent study reported that Polo-like kinase After centrifugation at 1500 rpm for 5 min, the cell pellets 2 (PLK2) enhanced α-Syn autophagic degradation, and were re-suspended and plated on a poly-lysine-treated fask (Sigma, St Louis, MO). The cultures were maintained at

1 3 Cellular and Molecular Neurobiology (2019) 39:1017–1028 1019

37 °C in a humidifed 5% CO2–95% air atmosphere. Culture and 5 μL α-Syn sample were added to the 96-well plate to media were changed 24 h later to complete medium and sub- a fnal volume of 200 μL. Phosphate bufered saline (PBS) sequently twice a week. The purity of astrocyte was > 95% was as a blank control. The reaction was carried out at room as determined with GFAP immunocytochemistry. Before temperature for 1 min, and the absorbance value was read experimental treatments, astrocytic cultures were passaged by a full-wavelength microplate reader. The excitation light once. is 446 nm and the absorption light is 482 nm. ThT binding Human neuroblastoma SH-SY5Y cells were purchased (unit a.u.) = Absorbance of Sample − Absorbance of PBS from American Type Culture Collection (ATCC, USA). blank control. Cells were cultured in DMEM (Gibco Lab., USA) containing 10% FBS (Gibco Lab., USA), 100 mg/mL streptomycin (Sigma-Aldrich Biotechnology, USA), and 100 U/mL peni- FITC‑labeled α‑Syn Aggregates cillin (Gibco Lab., USA). The cultures were maintained at 37 °C in a humidifed atmosphere of 5% CO 2 and 95% air. The α-Syn aggregates were placed in 4-mL ultrafltration Before the experiments, cells were inoculated into 6-well tubes (30 kDa, Millipore, UFC803008) and centrifuged at plate or 96-well plate, and cultured in DMEM containing 7500 g for 20 min, following replaced PBS with a freshly 10% FBS for 24 h. Then the medium was replaced with prepared cross-linked reaction solution (7.56 g NaHCO3, serum-free medium with α-Syn (1 μM) or GB (10 μM)/ 1.06 g Na 2CO3, 7.36 g NaCl, pH 9.0, plus ddH2O fxed BB (10 μM) or MG132 (20 μM)/3MA (5 mM) in diferent volume to 1 L). Freshly prepared 1 mg/mL fuorescein iso- experimental parts. thiocyanate (FITC, Sigma, # 3326-32-7) was dissolved in DMSO against the light. 50 μL FITC was added to 1 mL Preparation of Recombinant α‑Syn Aggregates protein solution, gently shaking to mix well with the protein, then, keeping in dark at 4 °C for 8 h. Afterwards, 5 M The wild-type human α-Syn was purifed as previously NH 4Cl was added to protein solution at a fnal concentra- described (Hua et al. 2017). Briefly, α-Syn gene was tion of 50 mM and the reaction was stopped at 4 °C for obtained by digesting the plasmid pcDNA3.1(−)-α-Syn 2 h. The production was transferred to ultrafltration tubes using restriction enzymes NcoI and XhoI. PCR fragments (4 mL), centrifuged at 7500 g for 20 min, and the bufer was of pcDNA3.1(−)-α-Syn were obtained by gel electrophore- replaced with PBS. Labeled FITC-α-Syn aggregates were sis, respectively, and the accuracy of pET-28b(+)-α-Syn was protected at 4 °C, and detected with anti-α-Syn antibody by verifed in NCBI. The purifed α-Syn was inserted into NcoI/ immunofuorescence. XhoI sites of pET-28b(+) to form a recombinant expression plasmid pET-28b(+)-α-Syn, and then the construct was transformed into E. coli JM109 competent cells. The In Vitro α‑Syn Internalization Analysis fusion protein pET-28b(+)-α-Syn was expressed by addition of IPTG and purifed by ammonium sulfate, ammonium Primary astrocytes were seeded in a 24-well dish (2 × 104 acetate, and anhydrous ethanol according to its acid stability. cells/well) and cultured for 24 h. Astrocytes were treated After analyzed by MALDI analysis, the purifed α-Syn pro- with α-Syn monomers (1 μM) or aggregates (1 μM) for dif- tein was lyophilized and stored at − 80 °C until further use. ferent time-point (0.5, 2, 6, 12, and 24 h), the content of The forms of α-Syn samples (1 mg/mL) were prepared α-Syn in cell supernatant and cytoplasm was detected by by dissolving the α-Syn stock in Tris bufer (50 mM Tris, western blotting assay. 150 mM NaCl, 10 mM EDTA, pH 8.0). Aggregated α-Syn was generated by incubating the samples at 37 °C for 7 days (with shaking). Electron microscope and Thiofavin T (ThT) In Vitro α‑Syn Degradation Analysis were used to identify α-Syn morphologies before subsequent experiments. Primary astrocytes were seeded in a 24-well dish (2 × 104 cells/well) and cultured for 24 h. Astrocytes were pretreated The Thiofavin T (ThT) Assay with α-Syn aggregates (1 μM) for 1 h, then the α-Syn- containing culture medium was removed, and cells were ThT binding assay is widely used to observe the formation of washed with PBS for three times, followed by incubation aggregates. During the preparation of α-Syn aggregates, ali- with normal culture medium, medium with MG132 and/or quots of 10 μL α-Syn were removed from the incubated solu- 3MA, medium with GB or BB, or medium with MG132 and/ tion for ThT binding experiments. 1 mg ThT was dissolved or 3MA in the presence of GB or BB. The content of α-Syn in 1 mL methanol, and stored at in the dark 4 °C. Then, 5 μM in cell cytoplasm was detected by western blotting assay and freshly prepared ThT (dissolved in 50 mM glycine, pH 8.5) immunofuorescence.

1 3 1020 Cellular and Molecular Neurobiology (2019) 39:1017–1028

Immunofuorescence commercially available assay (Jiancheng, Nanjing, China) according to the manufacturer’s instructions. The ratio of Primary astrocytes were seeded in a 24-well dish (2 × 104 LDH activity in the supernatant to the total LDH activity cells/well) and cultured for 24 h. The cells were then incu- was taken as the percentage of cell death. For each experi- bated with α-Syn and GB/BB according to study design. ment, four wells per concentration were averaged and each The control group was treated only with the same amount of experiment was performed in triplicate. solvent for 24 h. Cells were washed three times with 0.01 M PBS (5 min/wash), blocked with 5% BSA (Bovine Serum Albumin) at room temperature for 30 min, and incubated Purifcation of Cell Culture Supernatant Protein with primary antibody against α-Syn (1:200, #610787, BD- Bioscience) and GFAP (Glial fbrillary acidic protein, 1:500, The cell culture supernatant was collected, centrifuged #MAB360, Millipore) overnight at 4 °C. In the following to remove dead cells, and the supernatant medium was day, cells were incubated with the secondary antibody A488 transferred into new tubes. Then, 500 μL methanol and goat anti-mice IgG (1:1000, #A11001, Invitrogen) or A555 125 μL chloroform were added to precipitate medium, vor- (1:1000, #A21422, Invitrogen) for 2 h at room temperature, texed, and centrifuged for 5 min at 16,000 g. The upper washed three times with 0.01 M PBS (15 min/wash), and phase was discarded without touching the proteins disk, incubated with 100 ng/mL DAPI or Hoechst for 3 min. and 500 μL methanol was added and centrifuged for 5 min Images were captured using a fluorescence microscope at 16,000 g for cleaning. Then, the medium was removed, (Nikon, Japan) equipped with 20× objectives. Images were and the pellet was dried at 37 °C for 5 min. Finally, 50 μL collected using sequential scanning with the 405-, 488-, and 2.5× loading bufer was added with DTT and vortexed. 594-nm laser lines to produce three color overlays.

Conditioned Medium Transfer Experiments Western Blotting Analysis

Cultured primary astrocytes were incubated with GB Detergent solubility was performed by adding Triton (10 μM) or BB (10 μM) for 24 h before α-Syn aggregate X-100 to total cell lysates (fnal concentration 1%) and treatment for another 24 h. Then, the culture mediums were incubating for 30 min on ice followed by centrifugation collected and centrifuge at 4000×g for 5 min at 20 °C to (15,000 g, 60 min, 4 °C). The supernatant was designated remove debris. Checking a small aliquot of CM at the micro- Triton X-100 soluble fraction, and the pellet was re-dis- scope for the presence of debris, and the medium was used solved in 2% SDS-containing lysis bufer and sonicated for as conditioned medium to stimulate SH-SY5Y cells. 10 s (Triton X-100 insoluble fraction). Protein concentration was determined using a Bicinchoninic acid (BCA) MTT Analysis protein assay. α-Syn monomer existed in Triton X-100 soluble and insoluble fraction, while aggregates only existed Cell viability was detected by 3-(4,5-dimethylthiazol-2-yl)- in Triton X-100 insoluble fraction. After stimulation, other 2,5-diphenyltetrazolium (MTT) assay. Cells were treated protein samples were extracted by the protein extraction with conditioned medium from treated astrocytes and kit (Nanjing Keygen, China). incubated at 37 °C and 5% CO2 for 24 h, and then the cell Proteins were separated by SDS-PAGE electrophore- supernatant was collected for LDH assay. After discarding sis and transferred to PVDF membranes with the elec- serum-free DMEM medium in 96-well plate and adding trophoretic transfer system (Bio-Rad, USA). Non-specifc 150 μL 0.5 mg/mL MTT (Sigma-Aldrich Biotechnology, binding sites were blocked with 10% non-fat dry milk in USA) dissolved in PBS solution, the cells were maintained Tris-Hcl Bufer Saline containing 0.1% Tween 20 (TBS- at 37 °C in a humidifed atmosphere of 5% CO2 and 95% T) for 1 h at room temperature. The membranes were then air for 4 h. Subsequently, cell supernatant was removed fol- incubated with primary antibody against α-Syn (1:500, lowed by incubation with 200 μL DMSO (Aladdin Indus- #610787, BD-Bioscience), LC3 (Microtubule-associated trial Corporation, Shanghai, China) at room temperature for protein 1A/1B-light chain 3, 1:1000, #2775, CST), p62 20 min. The absorbance of each hole was detected at 450 nm (1:1000, #5114, CST), Beclin-1 (1:1000, #3495, CST), wave length. and β-actin (1:1000, Sigma-Aldrich,) overnight at 4 °C. Then, the membranes were incubated with corresponding LDH Analysis secondary antibody at room temperature for 1 h. Finally, immunoblots were scanned and determined by Omega16IC Lactate dehydrogenase (LDH) activity-based cytotoxicity (Ultra-Lum, Claremont, CA, USA). assay LDH activity measurements were performed using a

1 3 Cellular and Molecular Neurobiology (2019) 39:1017–1028 1021

Statistical Analysis 24 h. The viability of SH-SY5Y cells was decreased after incubation with α-Syn monomers at dose of 10 μM and All statistical analysis were done using GraphPad Prism aggregates at dose of 0.1, 1, and 10 μM (Fig. 1b). Mean- version 5.0 (GraphPad software, Inc., CA, USA) and while, the viability of primary astrocytes was decreased Adobe Illustrator CS5 software. Values were presented as after incubation with monomers and aggregates only at mean ± standard error of the mean (S.E.M.) and analyzed dose of 10 μM (Fig. 1c). These results suggested the suc- using one-way analysis of variance (ANOVA) followed cess of α-Syn aggregate preparation, and the dose of 1 μM by Tukey post hoc test. The signifcance level was set at was chosen for the following experiments. p < 0.05. To determine whether astrocyte-mediated phagocytosis was triggered in response to recombinant α-Syn, cultured primary astrocytes were incubated with 1 μM α-Syn Results monomers and aggregates for diferent times (Fig. 1d). As shown in Fig. 1e and f, we observed a decreased content of Astrocytes Engulfed Recombinant α‑Syn Aggregates both α-Syn monomers and aggregates in the cell superna- In Vitro tant after treatment for 6 h. However, compared to rapidly clearance of monomers in cytoplasm, increased content We first constructed and purified recombinant human of α-Syn aggregates in cytoplasm was observed during α-Syn protein in vitro as previously described (Hua et al. 24 h after incubation. We further investigated whether 2017). The ThT assay for α-Syn protein showed increased α-Syn aggregates were engulfed by astrocytes by immu- binding force during 3–5 days, reaching a plateau level nofuorescence staining. A totally co-localization of FITC- after 6 day (Fig. 1a). To observe the biological function of labeled α-Syn and anti-α-Syn protein is shown in Fig. 1g. recombinant α-Syn aggregates, both SH-SY5Y cells and In addition, FITC-labeled α-Syn was majorly expressed primary astrocytes were exposed to various concentrations in astrocyte, shown as co-localization of α-Syn and GFAP (0.1, 1, and 10 μM) of α-Syn monomers and aggregates for (Fig. 1h). These results indicated that astrocyte may engulf extracellular recombinant α-Syn protein.

Fig. 1 Internalization of α-Syn aggregates in astrocytes. a Biophysi- analysis of α-Syn monomers (1 μM) and aggregates (1 μM) in cell cal characterization of the aggregation of α-Syn proteins monitored supernatant and cytoplasm. g Immunocytofuorescence imaging of by ThT fuorescence. b, c MTT assay was used to identify cell viabil- anti-α-Syn antibody (red) and FITC-α-Syn (green) in primary astro- ity of SH-SY5Y cell and primary astrocytes after treated with MPP+, cytes. h Immunocytofuorescence imaging of GFAP (red) and FITC α-Syn monomers, and aggregates. Data are representative of three (green) in astrocytes treated with FITC (upper line). Co-location of independent experiments (mean ± S.E.M). *p < 0.05, **p < 0.01, and GFAP and FITC-α-Syn in α-Syn-treated astrocytes (lower line). The ***p < 0.01 compared with the Vehicle (Veh) group. d Primary astro- cells covered with 4,6-linked amidine-2-phenylindole (DAPI, blue) cytes were prepared and treated with 1 μM of FITC-α-Syn protein for nucleus staining. Scale bar: 20 μm for several time-points (0.5, 2, 6, 12, and 24 h). e, f Western blotting

1 3 1022 Cellular and Molecular Neurobiology (2019) 39:1017–1028

Fig. 2 Clearance of α-Syn aggregates in astrocytes. a Primary astro- independent experiments (mean ± S.E.M). *p < 0.05, **p < 0.01, cytes were treated with 1 μM of α-Syn aggregates for 1 h and washed ***p < 0.001 compared with Con (0 h) group. c Immunocytofuores- twice with fresh medium, followed by incubation in fresh medium for cence imaging of GFAP (red) and FITC-α-Syn aggregates (green) in the indicated times. b Triton X-100-insoluble protein extracts were astrocytes. DAPI (blue) for nucleus staining. Scale bar: 20 μm analyzed using western blotting. Data are representative of three

Astrocytes Eliminated α‑Syn Aggregates In Vitro To further investigate whether recombinant α-Syn protein can be degraded by UPS or ALP, cultured primary Since phagocytosis and degradation was a dynamic process, astrocytes were pretreated with 1 μM α-Syn aggregates for we verifed the cellular presence of α-Syn for diferent times 1 h. Afterwards, α-Syn-containing medium was removed (0, 0.5, 2, 6, 12, 24 h) after 1-h treatment (Fig. 2a). We found and cells were incubated with MG132 (20 μM) and 3MA that the intracellular α-Syn aggregates were decreased in a (5 mM)-containing medium for 24 h (Fig. 3b). Western time-dependent manner after treatment (Fig. 2b). To further blotting (Fig. 3c) and immunocytochemistry assay (Fig. 3d) confrm the elimination of α-Syn aggregates in astrocyte, showed that α-Syn was signifcantly increased after MG132 immunocytochemistry was used to identify FITC-labeled and 3MA treatment. These results may prove that astrocytes α-Syn aggregates in astrocytes. Notably, FITC-labeled engulf and degrade α-Syn aggregates via both the protea- α-Syn was co-expressed with GFAP positive astrocyte, and some and autophagy pathways. content of α-Syn was decreased over time (Fig. 2c). These data demonstrated that astrocyte eliminated intracellular Ginkgolides Enhanced Astrocytic Uptake recombinant α-Syn protein. and Degradation of α‑Syn Aggregates

We previously have found that Ginkgolides, including GB Both the Ubiquitin–Proteasome and Autophagy– and BB, could attenuate aggregated α-Syn-induced cell Lysosome Pathways were Involved in the Astrocytic apoptosis in SH-SY5Y cells (Hua et al. 2017). To further Clearance of α‑Syn Aggregates evaluate the efect of astrocytic clearance on aggregated α-Syn-induced cell apoptosis, we frstly investigated the It has been reported that α-Syn can be degraded by the ubiq- efects of GB and BB on the phagocytosis α-Syn protein uitin proteasome system (UPS) and autophagy–lysosome in astrocytes. The cultured primary astrocytes were treated pathways (ALP) (Webb et al. 2003). In order to identify with GB or BB for 2 h prior to α-Syn aggregate (1 μM) whether astrocytes clear α-Syn through the lysosomal path- incubation for 24 h (Fig. 4a). As a result, the contents of way, astrocytes were incubated with LysoTracker and FITC- the α-Syn in cytoplasm were decreased after GB or BB α-Syn aggregates for 1 h. Immunofuorescence microscopy treatment by Western blotting analysis (Fig. 4b), and total showed the presence of α-Syn and lysosomes in astrocytes content of α-Syn protein was detected decreased by immu- was co-located (Fig. 3a), suggesting that astrocytes may nocytochemistry (Fig. 4c). These results showed that GB clear the α-Syn aggregates through the lysosomal pathway. and BB reduced α-Syn aggregates in cytoplasm, indicating

1 3 Cellular and Molecular Neurobiology (2019) 39:1017–1028 1023

Fig. 3 Astrocytes degraded α-Syn aggregates through the protea- 24 h. c Western blotting analysis of Triton X-100-insoluble protein some system and autophagy–lysosome pathway. a Co-location of extracts in astrocytes treated with α-Syn aggregates and MG132 or/ LysoTracker and FITC-α-syn aggregates in astrocytes. Fluorescence and 3MA. **p < 0.01, ***p < 0.001 compared with Con (0 h) group, imaging of astrocytes after incubating with LysoTracker (red) and #p < 0.05, ##p < 0.01 compared with Vehicle group. Data are repre- FITC- α-syn aggregates (green) for 1 h. DAPI (blue) for nucleus sentative of three independent experiments (mean ± S.E.M). d Immu- staining. Scale bar: 20 μm. b Astrocytes were treated with 1 μM of nocytofuorescence imaging of α-Syn (green) of astrocytes treated α-Syn aggregates for 1 h and washed twice with fresh medium, fol- with aggregated α-Syn. Hoechst (blue) for nucleus staining. Scale lowed by the treatment of MG132 (20 μM) or/and 3MA (5 mM) for bar: 100 μm an enhancement of astrocytic clearance of α-Syn aggregates and 3MA (5 mM) were added for 24 h (Fig. 5a). As showed by GB and BB. in Fig. 5b, and Fig. 5c, the content of the α-Syn was signif- To further confrm the efect of GB and BB on the deg- cant increased with MG132 (20 μM) and/or 3MA (5 mM) radation of α-Syn aggregates internalized in astrocytes, cul- decreased after GB (10 μM) or BB (10 μM) treatment. tured primary astrocytes were pretreated with 1 μM α-Syn Interestingly, 3MA inhibited the decrease of α-Syn more aggregates to let α-Syn transfer into cells. After 1-h incu- signifcant than MG132, whereas no distinguishable change bation, α-Syn-containing medium was removed, and cells of intracellular α-Syn was observed between 3MA-treated were incubated with GB or BB-containing medium for 24 h group and MG132 + 3MA combination. (Fig. 4d). As shown in Fig. 4e, the content of the α-Syn in Several lines of evidence have suggested that autophagy cell cytoplasm was reduced after GB or BB treatment, and enhancers can promote the clearance of aggregate-prone intracellular α-Syn was detected decreased by immunocyto- proteins (Casarejos et al. 2011). During autophagy initiation chemistry (Fig. 4f). Collectedly, our results showed that GB and autophagosome formation, Beclin-1 binds microtubule- and BB can enhance astrocytic phagocytosis and degrada- associated protein-1 light chain 3 (LC3-I) that is converted tion of α-Syn aggregates. to LC3-II (the membrane bound form) and interacts with the ubiquitin-binding protein p62 (Park et al. 2013). As Ginkgolides Promoted the Astrocytic Elimination showed in Fig. 5d, western blot analysis showed that the of α‑Syn Mainly via Regulating Autophagy treatment of GB and BB signifcantly reduced the expression of p62 after 24-h incubation. Meanwhile, protein expres- To further determine whether GB and BB can accelerate sion of autophagy-related genes Beclin-1 and the ratio of elimination via promoting the UPS or ALP pathway, pri- LC3-II/LC3-I increased after GB or BB treatment compared mary astrocytes were pretreated with 1 μM α-Syn aggregates with Vehicle group. As showed in Fig. 5d, western blot for 1 h. Then, α-Syn-containing medium was removed and analysis showed that the treatment of GB and BB signif- GB/BB-containing medium mixed with MG132 (20 μM) cantly reduced the expression of p62 after 24-h incubation.

1 3 1024 Cellular and Molecular Neurobiology (2019) 39:1017–1028

Fig. 4 Ginkgolide B and bilobalide accelerated the clearance of α-Syn in astrocytes. a After the pretreatment with GB or BB for 2 h, astrocytes were incubated with aggregated α-Syn for 24 h. b Triton X-100-insoluble protein extracts of astrocytes were analyzed by western blotting. Data are representative of three independent experiments (mean ± S.E.M). ***p < 0.001 compared with Vehicle group, ##p < 0.01, ###p < 0.001 compared with Vehicle group. c Immunocyto- fuorescence imaging of α-Syn of astrocytes treated with α-Syn aggregates and GB or BB. Scale bar: 100 μm. d Astrocytes were treated with 1 μM of α-Syn aggregates for 1 h and washed twice with fresh medium, followed by incubation with GB or BB for 24 h. e, f α-Syn content was analyzed using western blotting and immunocytofuorescence. Scale bar: 100 μm. Data are representative of three independent experiments (mean ± S.E.M). **p < 0.01 compared with Con (0 h) group, #p < 0.05, ##p < 0.01 compared with Vehicle group

Meanwhile, the level of Beclin-1 and the ratio of LC3-II/ whether it may protect neurons from the indirect neuro- LC3-I increased after GB or BB treatment compared with toxicity induced by α-Syn-stimulated astrocytic condi- Vehicle group. Taken together, GB and BB reduced astro- tioned medium (CM). Primary astrocytes were pretreated cytic expression of α-Syn via proteasome and autophagy, with GB (10 μM) or BB (10 μM) 24 h. Then, GB/BB- and autophagy may be more efciently than proteasome in containing medium was removed before α-Syn aggregates astrocytes. (1 μM) stimulation. Cellular culture supernatants (CM) were harvested, and protein levels of α-Syn were detected Ginkgolides Indirectly Protected the SH‑SY5Y Cells by western blot 24 h later. As shown in Fig. 6a, GB and from the Neurotoxicity Induced by α‑Syn‑Stimulated BB treatment can both reduce the α-Syn content in cell Astrocytic Conditioned Medium culture supernatant. When human SH-SY5Y cells were treated with the CM, a signifcantly cell injury was found Since we found that GB and BB can accelerate the elimination of α-Syn aggregates in astrocyte, we next investigated

1 3 Cellular and Molecular Neurobiology (2019) 39:1017–1028 1025

Fig. 5 GB and BB accelerated the clearance of α-Syn aggregates in in GB and BB groups. d Expression of p62, Beclin-1, and LC3II/I astrocytes through an enhanced autophagy pathway. a Astrocytes was analyzed using western blotting in astrocytes treated with GB were treated with 1 μM of α-Syn aggregates for 1 h and washed twice and BB for 24 h. Data are representative of three independent experi- with fresh medium, followed by incubation with MG132 (20 μM) or/ ments (mean ± S.E.M). *p < 0.05, **p < 0.01 compared with Con and 3MA (5 mM) in the presence of GB or BB for 24 h. b, c Triton group, #p < 0.05, ##p < 0.01 compared with Vehicle group, $p < 0.05 X-100-insoluble α-Syn content was analyzed using western blotting compared with MG132 group in α-Syn aggregate-treated group detected by MTT and Discussion LDH assay, and GB and BB can increase cell vitality and decrease cell LDH release (Fig. 6b and c). In the present study, we constructed and purifed recombinant human α-Syn protein, and focused on astrocytic properties to clear extracellular recombinant α-Syn in vitro. Our results showed that astrocytes internalized and eliminated

1 3 1026 Cellular and Molecular Neurobiology (2019) 39:1017–1028

Fig. 6 GB and BB amelio- rated astrocytic conditioned medium-induced SH-SY5Y injury. a Western blotting analysis of α-Syn aggregates in α-Syn aggregate-stimulated astrocytic conditioned medium treated with GB and BB. b, c MTT assay and LDH assay of SH-SY5Y treated with astrocytic conditioned medium. Data are representative of three independent experiments (mean ± S.E.M). *p < 0.05 compared with Vehicle group. #p < 0.05 compared with Vehicle+α-Syn group

recombinant human α-Syn aggregates in vitro, and GB and plasma and cerebrospinal fuid, and the levels of oligomeric BB promoted the degradation of intercellular α-Syn aggre- α-Syn in these body fuids are elevated in PD patients (El- gates via the pathway of autophagy and proteasome. Moreo- Agnaf et al. 2006; Foulds et al. 2011). Therefore, the clear- ver, GB and BB reduced the indirect neurotoxicity of α-Syn ance of released α-Syn may be benefcial to PD patients, to neurons in vitro. and astrocytic intake and degradation may provide efcient Synucleinopathies such as PD and DLB destroy life of probabilities in PD therapy. aging population over 65 years, and approaches to interrupt It has been suggested that the clearance of α-Syn was α-Syn misfolding and aggregation of represent promising determined by assembly state of the protein. α-Syn is paradigms for therapeutic interventions (Lee et al. 2014). degraded by both the UPS and ALP in healthy neurons, Currently, mounting evidence indicated that transmission keeping axons and dendrites functional (Ebrahimi-Fakhari of toxic α-Syn between neurons and glia cells is the major et al. 2012). Monomeric and dimeric α-Syn species are way causing neurodegeneration. Released α-Syn can pro- degraded by UPS system and chaperone-mediated autophagy mote formation of inclusion bodies and induce cell death (Mak et al. 2010), while autophagy/lysosomal pathway is in neurons and pro-infammatory responses from astrocytes involved in the clearance of oligomeric and fbrilar species (Desplats et al. 2009). Recent studies have suggested LAG3 of α-Syn (Sacino et al. 2017). When autophagic clearance (lymphocyte-activation gene 3) might be a receptor that con- is impaired in neurons, α-Syn accumulates in axons, starts tributes to α-Syn transmission (Mao et al. 2016), but the forming oligomers, and further aggregates and becomes details are not clear. In our study, we treated primary astro- deposited in Lewy bodies (Plotegher and Civiero 2012). A cytes with α-Syn aggregates, and intracellular α-Syn can nearly study indicated that the overexpression of α-Syn pro- be identifed by western blotting and immunocytochemistry teins promoted the decrease of LC3-II and the increase of assay, indicating that extracellular α-Syn can be engulfed by p62 protein levels, suggesting the inhibition of autophagy astrocytes. Moreover, Changyoun Kim et al. have showed (Erustes et al. 2018). Autophagy is a process that allows that α-Syn released from differentiated SH-SY5Y cells bulk degradation of cytoplasmic contents. It involves the might interacted with TLR2 and induced activation of formation of autophagosome, a double membrane structure, microglia (Kim et al. 2013), but the relationship between which fuses with lysosome to form autolysosome where the released α-Syn and astrocytes is still largely unknown. contents are degraded (Xie and Klionsky 2007; Yorimitsu Afterwards, we detected the content of α-Syn in astro- and Klionsky 2005). In the present study, we found FITC- cytes by western blotting and immunofuorescence assay α-Syn protein and LysoTracker had a partly co-localization after α-Syn incubation for 1 h. As expected, we found that in immunofuorescence assay, indicating a localization of the content of intracellular α-Syn aggregates was decreased α-Syn in lysosome of astrocyte. We further treated pri- in a time-dependent manner after treatment, indicating that mary astrocytes with proteasome inhibitor MG132 or/and astrocyte can also eliminate intracellular recombinant α-Syn autophagy inhibitor 3MA after α-Syn aggregate incuba- protein after internalization. Through gene expression analy- tion, and Western blotting and immunocytochemistry assay sis, astrocytes were found to express a lot of genes that have showed that α-Syn protein was signifcantly increased after been implicated in engulfment and phagocytosis (Cahoy MG132 or/and 3MA treatment. These results proved that et al. 2008). In addition, α-Syn can be detected in blood the inhibition of autophagy and the proteasome in astrocyte

1 3 Cellular and Molecular Neurobiology (2019) 39:1017–1028 1027 signifcantly decreased the elimination of α-Syn protein, in vitro α-synuclein propagation assay. JH and YF wrote the manu- and methods to enhance astrocytic autophagy may be ideal script. All authors reviewed and approved the fnal manuscript. approaches in α-Syn clearance. Compliance with Ethical Standards A large number of studies have been conducted on the pharmacological activities of GBE and commercial available Conflict of interest The authors declare that they have no confict of extract (EGb-761). GBE has been popular in many parts of interests. the world for its multi-potential properties (Yoshitake et al. 2010). Studies have demonstrated the antioxidant (Guo Ethical Approval All study protocols were approved by the Institu- et al. 2015), anticancer (Hoenerhof et al. 2013), antidia- tional Animal Care and Use Committee of Nanjing Medical University (IACUC). betic (Hirata et al. 2015), and cardio-protective (Mozet et al. 2009) activities of GB and BB. GB is one of the main active ingredients of GBE, and it is a strong and specifc platelet activating factor antagonist and free radical scavenger. GB can inhibit the apoptosis of neurons induced by multiple References factors and play a neuroprotective role (Lin et al. 2008; Shi et al. 2010; Xiao et al. 2010). Meanwhile, as the only ses- Cahoy JD, Emery B, Kaushal A, Foo LC, Zamanian JL, Christopherson KS et al (2008) A transcriptome database for astrocytes, neurons, quiterpene in GBE, BB is shown to play benefcial efect on and oligodendrocytes: a new resource for understanding brain PD (Di Matteo and Esposito 2003), which is closely related development and function. J Neurosci 28:264–278 to anti-excitotoxicity and anti-apoptosis. In the present study, Casarejos MJ, Solano RM, Gomez A, Perucho J, de Yebenes JG, Mena we further found that GB and BB accelerated elimination of MA (2011) The accumulation of neurotoxic proteins, induced by proteasome inhibition, is reverted by trehalose, an enhancer α-Syn via enhancing autophagy in astrocyte. Previous stud- of autophagy, in human neuroblastoma cells. Neurochem Int ies reported that induction of autophagy reduces the levels 58:512–520 of mutant huntingtin and protects against its toxicity in cells Choi MG, Kim MJ, Kim DG, Yu R, Jang YN, Oh WJ (2018) Seques- and in transgenic Drosophila and mouse models of Hunting- tration of synaptic proteins by alpha-synuclein aggregates leading to neurotoxicity is inhibited by small peptide. PLoS ONE ton’s Disease (Ravikumar et al. 2004; Sarkar et al. 2005). 13:e195339 Therefore, the enhancement of astrocytic autophagy to clear Dauer W, Przedborski S (2003) Parkinson’s disease: mechanisms and α-Syn protein may protect neurons from extracellular α-Syn- models. Neuron 39:889–909 mediated cytotoxicity under PD condition. Desplats P, Lee HJ, Bae EJ, Patrick C, Rockenstein E, Crews L et al (2009) Inclusion formation and neuronal cell death through neu- However, the role of autophagy in neurodegenerative dis- ron-to-neuron transmission of alpha-synuclein. Proc Natl Acad ease is controversial, as excessive autophagy can also induce Sci USA 106:13010–13015 cell apoptosis (Mizushima and Komatsu 2011; Shintani Di Matteo V, Esposito E (2003) Biochemical and therapeutic efects of and Klionsky 2004). We next investigated how did neurons antioxidants in the treatment of Alzheimer’s disease, Parkinson’s disease, and amyotrophic lateral sclerosis. Curr Drug Targets CNS response to α-Syn-stimulated astrocytic CM with or without Neurol Disord 2:95–107 GB/BB treatment. Our results demonstrated that GB or BB Ebrahimi-Fakhari D, Wahlster L, McLean PJ (2012) Protein degra- treatment reduce the neurotoxicity caused by α-Syn in astro- dation pathways in Parkinson’s disease: curse or blessing. Acta cytic conditioned medium. Collectively, GB and BB may Neuropathol 124:153–172 El-Agnaf OM, Salem SA, Paleologou KE, Curran MD, Gibson MJ, promote astrocytic engulfment and degradation by enhanc- Court JA et al (2006) Detection of oligomeric forms of alpha- ing autophagy. These data also indicated that the timely and synuclein protein in human plasma as a potential biomarker for efective clearance of cerebral α-Syn is protective to PD Parkinson’s disease. FASEB J 20:419–425 patients. Further experiments are needed to examine how Erustes AG, Stefani FY, Terashima JY, Stilhano RS, Monteforte PT, Da SPG et al (2018) Overexpression of alpha-synuclein in an astrocytes engulf α-Syn and the selectivity between mono- astrocyte cell line promotes autophagy inhibition and apoptosis. mer and aggregate, and the underlying mechanisms of GB J Neurosci Res 96:160–171 and BB regulate autophagy and the exact pathway. Foulds PG, Mitchell JD, Parker A, Turner R, Green G, Diggle P et al (2011) Phosphorylated alpha-synuclein can be detected in blood Acknowledgements We are thankful to Prof. Zhuohua Zhang for pro- plasma and is potentially a useful biomarker for Parkinson’s dis- viding the plasmid pcDNA3.1(−)-α-Syn. This research was supported ease. FASEB J 25:4127–4137 by the grants from the National Natural Science Foundation of China Gao HM, Zhang F, Zhou H, Kam W, Wilson B, Hong JS (2011) (Grant Nos. 81673408 and 81703485). Neuroinfammation and alpha-synuclein dysfunction potentiate each other, driving chronic progression of neurodegeneration in Author contributions a mouse model of Parkinson’s disease. Environ Health Perspect YF and GH conceived the study and analyzed 119:807–814 the data. JH and NY constructed lentiviral vectors and involved in Guo RZ, Liu XG, Gao W, Dong X, Fanali S, Li P et al (2015) A manuscript preparation. JH and SX designed and performed animal strategy for screening antioxidants in Ginkgo biloba extract by experiment. NY and QC performed biochemical analysis. TT and JZ comprehensive two-dimensional ultra high performance liquid performed immunohistochemical and immunolabeling analysis. JH and chromatography. J Chromatogr A 1422:147–154 JD performed in vitro cell culture experiments. NY and JD performed

1 3 1028 Cellular and Molecular Neurobiology (2019) 39:1017–1028

Hirata BK, Banin RM, Dornellas AP, de Andrade IS, Zemdegs JC, and suppresses its toxicity in vivo. Proc Natl Acad Sci USA Caperuto LC et al (2015) Ginkgo biloba extract improves insulin 110:E3945–E3954 signaling and attenuates infammation in retroperitoneal adipose Park JM, Huang S, Wu TT, Foster NR, Sinicrope FA (2013) Prognostic tissue depot of obese rats. Mediat Infamm 2015:419106 impact of Beclin 1, p62/sequestosome 1 and LC3 protein expres- Hoenerhof MJ, Pandiri AR, Snyder SA, Hong HH, Ton TV, Peddada sion in colon carcinomas from patients receiving 5-fuorouracil as S et al (2013) Hepatocellular carcinomas in B6C3F1 mice treated adjuvant chemotherapy. Cancer Biol Ther 14:100–107 with Ginkgo biloba extract for two years difer from spontaneous Plotegher N, Civiero L (2012) Neuronal autophagy, alpha-synuclein liver tumors in cancer gene mutations and genomic pathways. clearance, and LRRK2 regulation: a lost equilibrium in parkinso- Toxicol Pathol 41:826–841 nian brain. J Neurosci 32:14851–14853 Hua J, Yin N, Yang B, Zhang J, Ding J, Fan Y et al (2017) Ginkgolide Ravikumar B, Vacher C, Berger Z, Davies JE, Luo S, Oroz LG et al B and bilobalide ameliorate neural cell apoptosis in alpha-synu- (2004) Inhibition of mTOR induces autophagy and reduces tox- clein aggregates. Biomed Pharmacother 96:792–797 icity of polyglutamine expansions in fy and mouse models of Iwata A, Maruyama M, Akagi T, Hashikawa T, Kanazawa I, Tsuji Huntington disease. Nat Genet 36:585–595 S et al (2003) Alpha-synuclein degradation by serine protease Sacino AN, Brooks MM, Chakrabarty P, Saha K, Khoshbouei H, Golde neurosin: implication for pathogenesis of synucleinopathies. Hum TE et al (2017) Proteolysis of alpha-synuclein fbrils in the lysoso- Mol Genet 12:2625–2635 mal pathway limits induction of inclusion pathology. J Neurochem Jang A, Lee HJ, Suk JE, Jung JW, Kim KP, Lee SJ (2010) Non-classical 140:662–678 exocytosis of alpha-synuclein is sensitive to folding states and Sarkar S, Floto RA, Berger Z, Imarisio S, Cordenier A, Pasco M promoted under stress conditions. J Neurochem 113:1263–1274 et al (2005) Lithium induces autophagy by inhibiting inositol Kalia LV, Lang AE (2015) Parkinson’s disease. Lancet 386:896–912 monophosphatase. J Cell Biol 170:1101–1111 Kim C, Lee SJ (2008) Controlling the mass action of alpha-synuclein Shi C, Wu F, Yew DT, Xu J, Zhu Y (2010) Bilobalide prevents apop- in Parkinson’s disease. J Neurochem 107:303–316 tosis through activation of the PI3K/Akt pathway in SH-SY5Y Kim C, Ho DH, Suk JE, You S, Michael S, Kang J et al (2013) Neuron- cells. Apoptosis 15:715–727 released oligomeric alpha-synuclein is an endogenous agonist of Shintani T, Klionsky DJ (2004) Autophagy in health and disease: a TLR2 for paracrine activation of microglia. Nat Commun 4:1562 double-edged sword. Science 306:990–995 Kim C, Rockenstein E, Spencer B, Kim HK, Adame A, Trejo M et al Spillantini MG, Crowther RA, Jakes R, Hasegawa M, Goedert M (2015) Antagonizing neuronal toll-like receptor 2 prevents synu- (1998) alpha-Synuclein in flamentous inclusions of Lewy bod- cleinopathy by activating autophagy. Cell Rep 13:771–782 ies from Parkinson’s disease and dementia with lewy bodies. Proc Lee HJ, Suk JE, Bae EJ, Lee JH, Paik SR, Lee SJ (2008a) Assembly- Natl Acad Sci USA 95:6469–6473 dependent endocytosis and clearance of extracellular alpha-synu- Tyson T, Senchuk M, Cooper JF, George S, Van Raamsdonk JM, Brun- clein. Int J Biochem Cell Biol 40:1835–1849 din P (2017) Novel animal model defnes genetic contributions Lee HJ, Suk JE, Bae EJ, Lee SJ (2008b) Clearance and deposition of for neuron-to-neuron transfer of alpha-synuclein. Sci Rep 7:7506 extracellular alpha-synuclein aggregates in microglia. Biochem Vekrellis K, Xilouri M, Emmanouilidou E, Rideout HJ, Stefanis L Biophys Res Commun 372:423–428 (2011) Pathological roles of alpha-synuclein in neurological dis- Lee HJ, Bae EJ, Lee SJ (2014) Extracellular alpha-synuclein-a novel orders. Lancet Neurol 10:1015–1025 and crucial factor in Lewy body diseases. Nat Rev Neurol Volpicelli-Daley LA, Luk KC, Patel TP, Tanik SA, Riddle DM, Stieber 10:92–98 A et al (2011) Exogenous alpha-synuclein fbrils induce Lewy Lin Y, Wang R, Wang X, He RR, Wu YM (2008) Efects of ginkgolide body pathology leading to synaptic dysfunction and neuron death. B on neuronal discharges in paraventricular nucleus of rat hypo- Neuron 72:57–71 thalamic slices. Neurosci Bull 24:345–350 Webb JL, Ravikumar B, Atkins J, Skepper JN, Rubinsztein DC (2003) Ma KL, Song LK, Yuan YH, Zhang Y, Han N, Gao K et al (2014) The Alpha-Synuclein is degraded by both autophagy and the protea- nuclear accumulation of alpha-synuclein is mediated by importin some. J Biol Chem 278:25009–25013 alpha and promotes neurotoxicity by accelerating the cell cycle. Xiao Q, Wang C, Li J, Hou Q, Li J, Ma J et al (2010) Ginkgolide B Neuropharmacology 82:132–142 protects hippocampal neurons from apoptosis induced by beta- Mak SK, McCormack AL, Manning-Bog AB, Cuervo AM, Di Monte amyloid 25–35 partly via up-regulation of brain-derived neuro- DA (2010) Lysosomal degradation of alpha-synuclein in vivo. J trophic factor. Eur J Pharmacol 647:48–54 Biol Chem 285:13621–13629 Xie Z, Klionsky DJ (2007) Autophagosome formation: core machinery Mao X, Ou MT, Karuppagounder SS, Kam TI, Yin X, Xiong Y et al and adaptations. Nat Cell Biol 9:1102–1109 (2016) Pathological alpha-synuclein transmission initiated by Yorimitsu T, Klionsky DJ (2005) Autophagy: molecular machinery for binding lymphocyte-activation gene 3. Science 353:aah3374 self-eating. Cell Death Difer 12(Suppl 2):1542–1552 McNaught KS, Mytilineou C, Jnobaptiste R, Yabut J, Shashidharan Yoshitake T, Yoshitake S, Kehr J (2010) The Ginkgo biloba extract P, Jennert P et al (2002) Impairment of the ubiquitin-proteasome EGb 761(R) and its main constituent favonoids and ginkgolides system causes dopaminergic cell death and inclusion body forma- increase extracellular dopamine levels in the rat prefrontal cortex. tion in ventral mesencephalic cultures. J Neurochem 81:301–306 Br J Pharmacol 159:659–668 Mizushima N, Komatsu M (2011) Autophagy: renovation of cells and Zeng XN, Sun XL, Gao L, Fan Y, Ding JH, Hu G (2007) Aquaporin-4 tissues. Cell 147:728–741 defciency down-regulates glutamate uptake and GLT-1 expres- Mozet C, Martin R, Welt K, Fitzl G (2009) Cardioprotective efect of sion in astrocytes. Mol Cell Neurosci 34:34–39 EGb 761 on myocardial ultrastructure of young and old rat heart and antioxidant status during acute hypoxia. Aging Clin Exp Res Publisher’s Note Springer Nature remains neutral with regard to 21:14–21 jurisdictional claims in published maps and institutional afliations. Obeso JA, Stamelou M, Goetz CG, Poewe W, Lang AE, Weintraub D et al (2017) Past, present, and future of Parkinson’s disease: a special essay on the 200th anniversary of the shaking palsy. Mov Disord 32:1264–1310 Oueslati A, Schneider BL, Aebischer P, Lashuel HA (2013) Polo-like kinase 2 regulates selective autophagic alpha-synuclein clearance

1 3