Exosome release and neuropathology induced by α-synuclein: new insights into protective mechanisms of Drp1 inhibition
Targeting alpha-synuclein (α-syn) as a therapeutic strategy for Parkinson’s disease (PD) has been intensively pursued largely due to its well-recognized pathogenic role. Since its discovery as the first familial link to PD over two decades ago, this protein has been associated with multiple neurotoxic mechanisms, such as mitochondrial dysfunction and impaired autophagic flux. We report here that blocking dynamin-related protein 1 (Drp1) improved both mitochondrial function and autophagic flux in experimental models of α-syn. Using rat dopaminergic neuronal cells with inducible wild-type human α-syn, we observed excessive mitochondrial fragmentation and increased Drp1 levels 48 h after gene induction. Functionally, these cells exhibited lower mitochondrial membrane potential, reduced ATP production rate and mitochondrial spare respiratory capacity, as well as increased levels of mitochondrial reactive oxygen species. To evaluate the protective role of Drp1 inhibition, we used three complementary approaches: gene silencing mediated by siRNA, overexpression of Drp1-dominant negative and the small molecule mitochondrial division inhibitor-1 (mdivi-1). Both morphological and functional defects induced by α-syn were attenuated by these strategies. Importantly, Drp1 inhibition reduced proteinase K-resistant α-syn aggregates. Based on that observation, we investigated the involvement of autophagy. Through a combination of stable autophagy reporter cells and immunoreactivity for LC3 and p62 in neuronal cells with either α-syn overexpression or treatment of human α-syn preformed fibrils (PFF), we observed that Drp1 inhibition abolished autophagic impairment induced by α-syn. Consistent with its role in improving autophagy function, Drp1 inhibition reduced exosome release and spread of α-syn pathology from neurons to neurons and from microglia to neurons. In summary, this study highlights new insights that Drp1 inhibition confers neuroprotection through both mitochondrial and autophagy-lysosomal pathways, further strengthening the therapeutic potential of targeting Drp1.
KeywordsParkinson’s disease Mitochondrial dynamics Neurodegeneration Alpha-synuclein Autophagy Protein aggregation
Parkinson’s disease (PD) is a complex and multifactorial disorder involving both genetic mutations and environmental factors . Since the identification of the first mutation in SNCA , the gene encoding α-synuclein (α-syn), the list of additional mutations linked to PD has expanded rapidly and become rather complex [28, 29, 53]. To date, the most investigated PD-linked gene is SNCA. Missense mutations as well as gene duplications and triplications of SNCA have been identified in familial PD [3, 34, 38, 50, 61, 73]. The discovery of increasing the gene dosage of SNCA by two to three fold can also cause PD  signifies that elevated wild-type (WT) α-syn alone is sufficient to cause the disease. α-syn is prominently present in Lewy bodies, which are intra-neuronal proteins aggregates commonly observed in PD . Although mutation in this gene is rare, the SNCA locus has been demonstrated to have genome-wide significant association with PD development . Genome-wide association studies (GWAS) have identified SNCA as a major gene associated with sporadic PD [26, 46, 59]. The fact that α-syn is involved in both familial and sporadic PD makes it a significant and attractive protein to investigate pathogenic mechanisms and therapeutic target for this neurological disorder. Neurotoxic mechanisms associated with α-syn have therefore been at the forefront of the PD research and have greatly contributed to the current understanding of the disease pathology.
α-syn has been demonstrated to induce neurotoxicity through multiple but non-mutually exclusive mechanisms [7, 17, 22, 28], including impairment in mitochondrial and autophagy-lysosomal function resulting in protein aggregation, mitochondrial impairment, oxidative stress and exosome release – all of which are the topics of interest in the present study. Relevant to this study we recently published data demonstrating that by using the small molecule Mitochondrial Division Inhibitor-1 (mdivi-1), a putative inhibitor of the mitochondrial fission Dynamic-Related Protein-1 (Drp1), we were able to reduce neuropathology induced by α-syn-A53T in rats . However, some critical questions remained from that study. First, mdivi-1 was used to block Drp1 function . Although this inhibitor has been widely reported to produce effects consistent with blocking mitochondrial fission and GTPase function of Drp1 [42, 63], questions have been raised whether this inhibitor blocks Drp1 function . Second, α-syn-A53T mutation was used to model PD. Given that this missense mutation is rare and responsible for a very small fraction of PD cases, the significance of that study in relation to sporadic PD needs to be validated in models with wild-type (WT) human α-syn. Third, to date, Drp1 is commonly referred to as a “mitochondrial fission” protein. However, most of Drp1 resides, not on mitochondria, but elsewhere in the cell. Indeed, a previous study estimated that only about 3% of Drp1 is localized to mitochondria under normal physiological condition . Although under pathological condition, post-translational modifications such as phosphorylation of Drp1 at S616, would induce its translocation to mitochondria, a significant portion most likely still remains in the cytosol. It is critical to investigate additional protective mechanisms of this protein. The present study addresses these three issues and we report here that blocking Drp1 genetically improved neuropathological hallmarks associated with mitochondrial dysfunction and autophagy flux impairment induced by human α-syn-WT. Consistent with these mechanisms, we observed improved mitochondrial function as well as reduced α-syn aggregation and exosome release when Drp1 was inhibited.
Materials and methods
Generation of stable and inducible human wild-type α-synuclein in dopaminergic neuronal cells
Stable cells with inducible α-syn-WT expression were generated by stably transfecting the rat dopaminergic neuronal cells N27 (1RB3A) [51, 52] (provided by Dr. Anumantha Kanthasamy, Iowa State University) with an ecdysone inducible system, the Complete Control Inducible Mammalian Expression System (Stratagene), as we previously described in details for the PINK1 models . The cDNA of full length human wild-type SNCA was subcloned into the multiple cloning site. Stably transformed cells were selected and maintained in RPMI containing 10% FBS, G418 (500 μg/ml) and hygromycin (200 μg/ml). Of note, due to the potential concern that N27 cells in some laboratories have a mixed population because of extensive passaging over time, we performed immunocytochemistry and immunoblotting and confirmed that approximately 91% of our N27 cells have tyrosine hydroxylase (a dopaminergic marker) immunoreactivity (Additional file 1: Figure S1). Quantification was performed by counting DAPI with or without TH immunoreactivity.
Our initial attempt was to select inducible cells with high expression of GFP using flow cytometry. However, due to undetectable GFP intensity (which could occur for a protein expressed downstream of IRES) , we directly selected cells with high expression of α-syn using SMARTFLARE human SNCA-Cy3 mRNA probe (Millipore, SF-1254). This non-toxic probe was taken up by living cells through endocytosis. Once inside the cell, it recognizes and binds to human α-syn mRNA, resulting in fluorescent signal. For sorting, cells were treated with 20 μM PonA (or ethanol vehicle control) for 24 h and SMARTFLARE probe (200 nM) was added ~ 16 h prior to sorting using FACSAria II. Cells were maintained in RPMI 1640 containing10% FBS, G418 (500 μg/ml) and hygromycin B (200 μg/ml) (complete medium). The expression of α-syn upon PonA induction was confirmed using immunostaining and western blot.
Stable autophagy reporter HeLa cells
HeLa cells with stable overexpression of mRFP-GFP-LC3 were designed to monitor autophagy flux as previously described . We have successfully used these cells to measure autophagy flux [9, 10]. These cells were maintained in DMEM + 10%FBS + G418 (100 μg/ml) + 1% penicillin/streptomycin. Of note, because we needed to performed immunostaining to identify cell with α-syn overexpression, we chose to use fixed cells instead of live cell imaging in this study. Due to the potential concern that fixation may bring back the quenched GFP signal, we directly compared the signal intensity of cells with or without PFA fixation. As seen below Additional file 1: Figure S2, GFP signal was not affected by fixation; otherwise, every single cell would have had a green signal – indicating autophagy flux is functional and the GFP signal is quenched when autophagosomes fuse with the acidic environment of the lysosomes.
Primary cultured mouse microglia were prepared from post-natal day 0 (P0) newborn C57BL/6 pups as described . Briefly, meninges-free cortices were isolated and trypsinized. Cells were cultured in complete DMEM-F12 with 10% FBS and penicillin/streptomycin. Murine monocyte colony stimulating factor (M-CSF, 10 ng/ml, cat. no.315–02, Peprotech, CA) was added to the medium 6 days after plating. After 15 days, the cultures were shaken (4 h; 260 rpm on a rotary shaker) to remove microglia.
SH-SY5Y cells and BV2 cells
Human neuroblastoma SH-SY5Y and immortalized mouse microglia BV-2 cells were cultured in DMEM containing 10% FBS, penicillin G and streptomycin sulfate.
siRNA-mediated Drp1 knockdown
Pre-designed siRNA against rat dnm1l and human DNM1L (gene encoding Drp1) were purchased from Dharmacon Research, Inc. SMARTpool: siGENOME Rat Dnm1l siRNA was used for the N27 cells and SMARTpool: siGENOME Human Dnm1l siRNA was for HeLa cells. Each of this product is a mixture of four individual siRNA duplexes that target four separate sequences of the gene to maximize efficiency of gene silencing. To enhance transfection efficiency, an “in-tube” transfection procedure  was used with the following modifications: Cell suspension (80,000–100,000cells/ml) were mixed with jetPRIME™ DNA and siRNA Transfection Reagent (Polyplus-transfection®SA). For every 500 μL of cell suspension (RPMI + 10% FBS), 50 μL of JetPRIME buffer and 2 μL JetPRIME reagent were used. Cells were plated and left in transfection medium overnight, then media was changed the following day. Gene silencing efficiency was confirmed using western blot, with 10 nM of siRNA achieving 75–90% knockdown compared to scrambled control (siGENOME Non-Targeting siRNA Control Pools, Cat# D-001206, Dharmacon Inc) after 48 h.
Preparation of human α-syn pre-formed fibrils (PFF)
α-syn monomers were obtained from the Michael J. Fox Foundation and the generation of PFF was performed according to the accompanying protocol (https://www.michaeljfox.org/files/accelerate/models/PFF%20Protocol%202017b.pdf). Briefly, the frozen aliquot was thawed on ice, centrifuged at 15,000×g for 10 min at 4 °C with a SORVALL legend micro 21R centrifuge (Thermo Scientific). Protein concentration was determined using BCA assay (Thermo Scientific Pierce) and the samples were diluted in sterile PBS to 5 mg/ml in a 1.5 ml Eppendorf protein low-bind tube. The sample was quickly centrifuged and placed in an Eppendorf ThermoMixer C orbital shaker (with thermo top on), shaken at 1000 RPM for seven consecutive days at 37 °C. The stock samples were aliquoted and stored at − 80 °C. To assess morphology of PFF, transmission electron microscopy (TEM, Phillip CM120) was used. To this end, the recombinant protein was diluted into 1 mg/ml in dPBS and sonicated using either QSonica XL-2000 at power level 2 for a total of 30 pulses (1 s each) or Fisher Scientific 120 Sonic Dismembrator equipped with CL-18 microtip (20% power) and then transferred to carbon-coated 200-mesh copper electron microscopy grids separately. Subsequently, PFF was negatively stained with 1% uranium acetate and its morphology was identified by TEM.
PFF treatment in cell cultures
SH-SY5Y and BV-2 cells were transfected with siRNA-Drp1(50 nM) or scramble control for 12 h, followed by PFF (2 μg/ml) treatment for 24 h, and cultured for additional 36 h after PFF withdrawal. For mdivi-1 treatment, PFF was added concurrently with mdivi-1 (20 μM). Conditional media (CM) were then collected for exosome isolation.
Mouse primary microglia were treated with PFF for 24 h in the presence or absence of mdivi-1, and further cultured for 36 h after PFF withdrawal. To activate microglia, cells were treated with LPS (1 μg/ml) for 3 h followed by 15 min of ATP (5 mM) before harvesting. Conditioned media (CM) were collected for exosome isolation.
Exosomes. Cultured cells were maintained in media containing exosome-depleted FBS (cat# EXO-FBS-50A-1, SBI System Biosciences). Mouse primary microglia were treated with PFF for 24 h in the presence or absence of mdivi-1, and further cultured for 36 h after PFF withdrawal. To activate microglia, cells were treated with LPS (1 μg/ml) for 3 h followed by 15 min of ATP (5 mM) before harvesting. Conditioned media (CM) were collected for exosome isolation. Cell culture media (20 ml) from two 100 mm plates of primary microglia was collected 24 h after PFF removal, centrifuges at 3000×g at room temperature for 15 min to remove cell debris. Exosomes in the resultant supernatant was precipitated using ExoQuick-TC PLUS Exosome Purification Kit (Cat# EQPL10TC-1, SBI System Biosciences) according to the manufacturer’s instructions. Exosome pellets were re-suspended in culture medium for cell treatment. To visualize exosomes, pellets of exosomes were re-suspended in sterile water and transferred to carbon-coated 200-mesh copper electron microscopy grids and incubated for 10 min at room temperature. Exosomes were then incubated with 2% phosphotungstic acid for 3 min. Micrographs were observed under a transmission electron microscope (Phillip CM120).
Cells were transfected with plasmids (0.3 μg/well for 24-well plate and 1.5 μg/well for 6-well plate) using either Lipofectamine™ 3000 (Thermo Fisher Scientific) or jetPRIME™ DNA and siRNA Transfection Reagent (Polyplus-transfection®SA) following the manufacture’s protocol.
Cells were grown on borosilicate cover slips pre-coated with poly-D-lysine in 24-well plates. Prior to immunostaining, cells were fixed with 4% formaldehyde (Thermo Scientific™ Pierce™, #28906), in warm cell culture media at 37 °C for 20 min. Please refer to the table in Additional file 1 for a list of primary antibodies and dilutions used. Corresponding Alexa Fluor®(350, 488, 586, and 633) conjugate secondary antibodies (Molecular Probes) were used at 1:500–1:1000 dilution. Slides were mounted using Prolong™ gold anti-fade mount with or without DAPI (Molecular Probes). Images were captured using Olympus Fluoview 1200 confocal microscope (with the exception of live cell imaging, which was Olympus Fluoview10i automated confocal laser-scanning microscope).
For cytosolic samples, cells were lysed with RIPA buffer (150 mM NaCl,1%(v/v) IGEPAL, 0.5% Sodium Deoxycholate, 0.1% SDS, 50 mM Tris, pH 8.0) containing 1X Halt Protease and Phosphatase Inhibitor (Thermo Scientific), dounced for 20–30 times in a 1 mL glass homogenizer and centrifuged at 16,000×g for 15 min at 4 °C. Supernatants were collected and 30-45 μg of protein per well (depending on respective antibodies) were separated in SDS-PAGE.
For experiments related to exosome release, cells were sonicated in RIPA buffer. To measure the amount of exosomes release, exosomes were extracted from same amount of culture medium and immunoblotted for exosomal markers (Alix and Tsg101). To quantify the levels of α-syn in exosomes, the same amount of protein from exosomes (40 μg) were loaded. Proteins were separated by SDS-PAGE and then transferred onto nitrocellulose membranes. The blots were then incubated overnight at 4 °C with following primary antibodies: anti-Alix, anti-Tsg101, and anti-alpha synuclein. Secondary antibodies conjugated with horseradish peroxidase (HRP) were used, and immunoreactivity was visualized with chemiluminescence (SuperSignal Ultra, Pierce, Rockford, IL,USA). Protein bands were analyzed and quantified using Scion Image system (Scion Corporation).
N27 stable cells were grown on poly-D-lysine-coated glass coverslips. Mitochondria were visualized with transfected DsRed-Mito as we previously described  and images were captured using Olympus Fluoview 1200 confocal microscope. Mitochondrial morphology was quantified blinded using Image J . More than 500 clearly identifiable mitochondria from randomly selected 30–50 cells per experiment were measured in four independent experiments. Roundness: 4 × ([Area])/(π × [Major axis]2). Aspect ratio is a measurement of major / minor axes , using ImageJ’, (https://imagej.nih.gov/ij/). Both of these values approach 1 as the particle becomes circular.
Mitochondrial membrane potential (ΔΨm)
ΔΨm in N27 was quantified using 50 nM tetramethylrhodamine methyl ester (TMRM) as previously described . As a positive control, cells were treated with 20 μM carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP) to collapse ΔΨm. Fluorescent signal was analyzed by BD Accuri C6 flow cytometer using FL-2 channel.
Reactive oxygen species (ROS) quantification
Cellular ROS was measured using the superoxide indicator dihydroethidium (DHE, Invitrogen) as described . Briefly, N27 stable cells grown in 24-well plates were treated with 10 μM DHE in cell culture medium and incubated for 20 min at 37 °C. The dye was then washed, cells were trypsinized, and analyzed using flow cytometry (BD Accuri C6),. As a positive control, cells were treated with 30 μM H2O2.
Mitochondrial ROS was assessed using MitoSOX red (M36008, Molecular Probes). N27 stable cells were grown in 96-well plate and incubated with 2.5 μM of MitoSOX working solution (in HBSS) at 37 °C for 20 min. Cells were then washed three times with PBS and fluorescent intensity was measured using a plate reader (Biotek Synergy H1 Hybrid Multi-Mode Reader) at Ex/Em 510/595. After which cells were washed twice with PBS, incubated with DAPI (5 μg/ml) in the dark at room temperature for 10 min and washes three times before the DAPI signal was quantified using the Biotek plate reader (Ex/Em 358/461). Treatment of 50 nM rotenone, a mitochondrial complex I inhibitor, for 48 h was used as a positive control to generate mitochondrial ROS.
Mitochondrial function in live cells was assessed using the Seahorse XFe96 Extracellular Flux Analyzer (Seahorse Biosciences Inc). Cells were grown in the Seahorse 96-well plates overnight for attachment. On the following day, cell culture medium was washed and replaced with 175 μl of serum free assay medium (Dulbecco’s Modified Eagle’s Medium with - 5.5 mM Glucose, 1.0 mM Sodium Pyruvate, 4 mM Glutamine, 2 mM HEPES, pH 7.4), incubated in a 37 °C non-CO2 incubator for at least 30 min before being loaded into the analyzer. Mitochondrial respiration was measured using the Mito-Stress Test (Seahorse Biosciences Inc) as instructed by the manufacturer. Oligomycin (1 μg/ml), FCCP (0.5 μM), rotenone (1 μM) and antimycin A (1 μM) were sequentially added to cells to determine mitochondrial respiration. Each oxygen consumption rate data point refers to the mean rates during each measurement cycle, which consists of a mixing time of 30s followed by a data acquisition period of 3 min. Three data points were acquired after each injection, and four data points were recorded for basal respiration . For normalization, cells were fixed with 4% PFA, followed by incubation of DAPI (80 μl of 5 μg/ml in PBS) for 10 min. The plate was then washed 3 time with PBS, and fluorescent signal DAPI for was quantified at Ex/Em 358/461 nm using the Biotak Synergy H1 Hybrid Reader.
Quantification of autophagic vesicles
The analysis of autophagy flux in stable HeLa cells expressing mRFP-GFP-LC3 was performed as described . Green vesicles represent autophagosomes because when the autophagosome fuses with the lysosomes, the pH sensitive GFP signal is quenched by the acidic environment in the lysosomes. Red vesicles are made up of both autophagosomes and autolysosomes. The number of autolysosomes was obtained by subtracting the number of green vesicles from that of the red vesicles. For analysis, cells were imaged with an Olympus Fluoview with 60x time objective, autophagosomes and autolysosomes from at least 50 cells per treatment group were counted using ImageJ. For N27 cells, autophagy blockage was assessed through quantification of LC3-mcherry puncta together with the immunostained p62 puncta.
Proteinase K digestion
To determine the formation of aggregation in cells, Proteinase K digestion was performed. Freshly fixed cultured cells were washed 3X5min with PBS, then treated with Proteinase K (~ 0.34 U/ml, Sigma P4850). The plates were then incubated in the dark at room temperature for 10 min with gentle shaking, followed by 3X5min washes with PBS and subsequently immunostained for α-syn.
N27 stable cells were transfected with Drp1 siRNA (10 nM) or scramble control (10 nM) overnight and then induced with PonA for 48 h. Overnight rapamycin treatment (1 μM) was used as a control to inhibit mTOR. Cells were then harvested for western blotting. The phosphorylated form of mTOR substrate protein 4E-BP1 was probed using Phospho-4E-BP1 (Thr37/46) (236B4) Rabbit monoclonal antibody (Cell Signaling Technology, catalogue #2855), 1:500 dilution, and subsequently probed with goat anti-rabbit IgG HRP conjugate (Bio-rad) with 1:5000 dilution.
Data represent mean ± SEM. For normally distributed data, differences between means were analyzed using one-way ANOVA, followed by Newman-Keuls post hoc testing for pairwise comparison. The null hypothesis was rejected when the p value < 0.05.
Drp1 inhibition reduces mitochondrial fragmentation induced by α-syn in rat dopaminergic neuronal cells
After the successful generation of these stable α-syn dopaminergic neuronal cells, we first assessed the impact of α-syn on mitochondrial morphology. Using DsRed-Mito transfection to visualize mitochondria, we observed that these organelles were fragmented in cells with α-syn overexpression (Fig. 1a). However, stable cells with empty vector control displayed tubular mitochondria. This observation appeared to be mediated by Drp1 because increased levels of this fission protein were observed at the mitochondrial and total levels, as demonstrated using immunocytochemistry (Fig. 1a) and immunoblotting (Fig. 1b), respectively. To investigate the effects of Drp1 inhibition on mitochondrial fragmentation induced by α-syn, we used complementary genetic and pharmacological approaches to reduce Drp1 function as we previously described [4, 15, 56]. First, we transfected cells with siRNA-Drp1, which achieved approximately 70–80% of Drp1 knockdown efficiency (Additional file 1: Figure S4). Second, we transfected cells using the dominant negative mutant Drp1-K38A. Third, the small molecule mitochondrial division inhibitor-1 (mdivi-1). As demonstrated morphologically (Fig. 1c-e) and quantitatively (Fig. 1f), all three strategies blocked mitochondrial fragmentation induced by α-syn. Of note, siRNA-Drp1 did not appear to reduce the levels of α-syn as compared to the group that received scrambled siRNA. To quantify more objectively the levels of α-syn in these two groups of cells, we performed immunoblotting and confirmed that the levels of α-syn between cells transfected with scramble-siRNA and Drp1-siRNA were not statistically different (6.72 ± 0.31 vs 6.16 ± 1.76, data represent mean ± SEM from 3 independent experiments using actin as a loading control).
Drp1 inhibition improves mitochondrial function and reduces oxidative stress induced by α-syn
An impairment in mitochondrial function invariably would lead to free radical production such as reactive oxygen species (ROS). To determine whether such ROS production would originate from mitochondria resulting in a higher total cellular ROS, we quantified MitoSOX Red and Dihydroethidium (DHE) signals to detect mitochondrial superoxide and cellular ROS levels, respectively (Fig. 2h-k). Rotenone, a complex I inhibitor, was used as a positive control to generate ROS from the ETC blockade. Consistent with its inhibitory effect on mitochondrial function, α-syn increased ROS levels, an adverse effect that was blunted by Drp1 inhibition (Fig. 2h-k). Taken together, these mitochondrial studies support the negative effects of α-syn on the mitochondrial ETC and that blocking Drp1 is protective.
Drp1 inhibition attenuates autophagic blockade and protein aggregation in dopaminergic N27 neuronal cells with inducible α-syn
Drp1 inhibition prevents autophagy flux impairment induced by α-syn in autophagy reporter HeLa cells
Human α-syn preformed fibrils (PFF) has been used in recent years to induce the release of exosomes and the spread of α-syn from one cell to another in vitro [24, 40, 65] and in vivo . Small seeds of PFF generated from recombinant α-syn can be endocytosed by neurons where it recruits endogenous α-syn to form phosphorylated and insoluble aggregates . We obtained α-syn monomers from the Michael J. FOX Foundation and generated PFF according to the accompanying protocol. With transmission electron microscopy (TEM), we confirmed the morphology and size of PFF and its sonicated form (Additional file 1: Figure S6). Using the autophagy reporter HeLa cells, we confirmed that PFF blocked autophagy flux and siRNA-Drp1 attenuated this impairment (Fig. 4i & j). Furthermore, this treatment also reduced α-syn protein aggregation (Fig. 4i & j). Together, in a cell model designed to monitor autophagy flux, our genetic and pharmacological data provide strong evidence that blocking Drp1 alleviates the negative impact of α-syn on autophagy flux and therefore supporting a novel protective mechanism of Drp1 inhibition.
Drp1 inhibition attenuates lysosomal impairment and inhibits mTOR activity
Blocking Drp1 reduces exosome release from SH-SY5Y cells treated with PFF
Drp1 inhibition reduces protein aggregation induced by exosomes from PFF-treated microglia
Mitochondrial dysfunction and impaired autophagy flux represent two major pathogenic mechanisms in PD. After the discovery of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) as the chemical causing parkinsonism , mitochondrial dysfunction mediated by blockade of the electron transport chain has been quite well-investigated in PD. However, a more recent approach is to understand the impact of mitochondrial fusion and fission, not only for PD but also other neurodegenerative diseases [2, 31, 66]. It is now recognized that a balance in mitochondrial fusion and fission is critical to neuronal function and viability. Mitochondrial fusion requires the coordination of both the inner (IMM) and the outer (OMM) mitochondrial membranes. The OMM proteins Mitofusin 1 & 2 (Mfn1/2) coordinate with the IMM Optic Atrophy-1 (Opa1) to join the membrane. Mitochondrial fission is governed by a separate set of proteins: Mitochondrial Fission Factor (Mff), Fission-1 (Fis1), as well as Mitochondrial Dynamics Proteins of 49 and 51 kDa (MiD49 and MiD51, respectively) are anchored to the OMM where they recruit cytosolic Dynamin-related protein-1 (Drp1), which then oligomerizes and forms a ring-like structure around the mitochondria to constrict and split them [36, 49]. Because Drp1 can bind to multiple downstream fission proteins to sever mitochondria, it has earned the reputation as a master regulator of mitochondrial fission. However, as demonstrated in the present study, in addition to this well-established function, Drp1 has a novel role in autophagy.
Autophagy plays a critical role in removing misfolded proteins, including α-syn [14, 37, 67]. Therefore, impairment in autophagy leads to accumulation of α-syn, which further exacerbating the blockade of autophagy as demonstrated cell culture and animal models [4, 11, 19, 68, 72], thus creating a bidirectional positive feedback loop of neurotoxicity . Strikingly, recent data from other laboratories and ours show that blocking Drp1 is capable of reducing protein aggregation. Reddy and colleagues, for example, reported than crossing Drp1+/− mice with either the AβPP (Tg2576) or the Tau P301L transgenic mouse models of (AD) reduced accumulation of toxic proteins in these animals [30, 41]. We recently demonstrated that pharmacological blocking of Drp1 dramatically reduced α-syn aggregates in nigral DA neurons of rats overexpressing α-synclein-A53T . Together, these results suggest that one possible mechanism by which blocking Drp1 reduces protein aggregation is through an improvement in autophagic function.
Given the well-established function of Drp1 in mitochondrial fission and its potential role in autophagy, the present study evaluated the protective effects of Drp1 inhibition mediated through these two mechanisms. Using primarily genetic approaches to inhibit Drp1, we report here that reduced Drp1 function conferred protection against α-syn induced impairment in both mitochondria and autophagy. In stable dopaminergic neuronal cells with inducible expression of α-syn, reducing Drp1 function genetically (Drp1 knockdown and overexpression of Drp1-dominant negative) and pharmacologically (mdivi-1 treatment) attenuated the negative impacts of α-syn on mitochondrial morphology and function (membrane potential, reactive oxygen species, respiration and spare respiratory capacity (SRC). A reduction in SRC leads to energy crisis when energy demand exceeds the supply ability of mitochondria. Indeed, SRC has been considered as a major factor that defines the survival of the neuron . Importantly, Drp1 inhibition drastically reduced proteinase K-resistant α-syn aggregates as demonstrated using complementary approaches of reducing Drp1 function in multiple cell models with overexpressing human wild-type α-syn or the treatment of recombinant α-syn (PFF). Consistent with these observations of protein clearance, we demonstrated that Drp1 inhibition abolished autophagic impairment induced by α-syn in multiple cell models. Furthermore, we assessed mTOR activity by quantifying the levels of its substrate, phosphor-4E-BP1, using immunoblotting of N27 cells overexpressing α-syn with siRNA-Drp1 or scramble control. These results demonstrate that reducing Drp1 function blocks mTOR activity and thereby increasing autophagy flux. In combination, our data indicate that that Drp1 inhibition confers protection against α-syn by both improving mitochondrial function and autophagy flux.
Blockade of the mitochondrial electron transport chain and autophagy flux has been shown to induce the spread of α-syn inter-cellularly. For example, exposure of enteric neurons to rotenone, a mitochondrial complex I inhibitor, promotes the release of α-syn, which is subsequently taken up by and form aggregates in the recipient neurons . Emerging evidence indicates that α-syn can spread inter-cellularly through exosome release, primarily because of its ability to impair mitochondria and autophagy [1, 25]. Exosomes are small extracellular vesicles with a typical size of 40–100 nm. Because these vesicles carry cargos such as mRNA and proteins, they could play a role in the spread of misfolded proteins such as α-syn . It has been demonstrated that impaired autophagy induces exosome-mediated α-syn spread to other neurons , forming aggregates and inducing death in the receiving cell [18, 27]. As demonstrated in this study, in addition to neurons, microglia are capable of releasing exosomes. Our data also indicate that when activated by LPS, microglia release drastically more exosomes. The role of activated microglia in causing neuroinflammation by releasing molecules such as TNF-α, IL-1β and IL-6 has been well-documented and proposed to be involved in PD pathogenesis. Relevant to this study, LPS has been reported to induce mitochondrial fission in microglia leading to neuroinflammation. Using primary microglia  and the BV2 murine microglial cells , two independent studies show that LPS induces mitochondrial fragmentation and neuroinflammation via a Drp1-dependent mechanism. Blocking Drp1 using shDrp1 and mdivi-1 reduces LPS-induced release of pro-inflammatory molecules. In addition to microglia, astrocytes are capable of causing neuroinflammation  - although to a lesser extent than microglia. In primary mouse astrocytes and the human U373 astrocytes, manganese (Mn) reduces oxidative phosphorylation, increases mitochondrial fragmentation and neuroinflammation - especially in the presence of supplemented aggregated α-syn . The mitochondrial targeted antioxidant mito-apocynin was demonstrated to be highly protective against Mn-induced such alterations in astrocytes , suggesting impaired mitochondria as the cause of neuroinflammation. Together, these studies indicate that enhanced mitochondrial fission in microglia and astrocytes are linked to their production of proinflammatory molecules. It is most likely that the reduction of exosome release observed in this study was mediated through mechanisms related to improved mitochondrial function, reduced neuroinflammation and improved autophagy when Drp1 was inhibited.
Mitochondrial dysfunction, impaired autophagy flux, oxidative stress and α-syn pathology (aggregation and spread) have all been proposed to play a dominant pathogenic role in PD. Blocking Drp1 function as a potential therapeutic strategy has gained interest in recent years for neurodegenerative diseases. We have reported that blocking Drp1 is neuroprotective in cell  and animal models of PD [4, 56]. A complementary approach using a peptide known as P110 to block the binding of Drp1 to Fis1 has also been shown to be protective in the MPTP models [20, 54]. However, all these previous studies focused on the mitochondrial fission pathway. Using models of α-syn, which is relevant to familial and sporadic PD, the present study provides the following novel observations: First, in addition to improving mitochondrial morphology and function, blocking Drp1 improved autophagy flux. Second, Drp1 inhibition reduced protein aggregation and spread from one cell to another via exosomes and non-cell autonomous mechanism. Together, these discoveries highlight new insights that Drp1 inhibition confers neuroprotection through both mitochondrial and autophagy-lysosomal pathways, further strengthening the therapeutic potential of targeting Drp1. The discovery that blocking Drp1 in microglia confers protection in neurons suggest that Drp1 should also be considered to be reduced in glial cells, not just in the affected neuronal population.
We thank Dr. Tan Li (Department of Biostatistics, Florida International University) for his advices on statistical analysis. Dr. Jeremy Chambers for proofreading this manuscript and constructive comments.
Conceived and designed the experiments: KT, MC, RF, MG, SL. Performed the experiments: RF, MG. Analyzed the data: KT, MC, RF, MG. Contributed reagents/materials: SL provided the HeLa autophagy reporter cells. All authors read and approved the final manuscript.
This work was supported in part by the National Institute of Environmental Health Services (R01-ES022274 and R35-ES030523, KT), Medical Research Council UK (MR/L022079/1, KT), Robert Stempel College of Public Health & Social Work, Florida International University and Peninsula Schools of Medicine and Dentistry (KT,RF, MG), Plymouth University, Plymouth, UK (RF).
Ethics approval and consent to participate
Consent for publication
All authors have read the manuscript and accepted responsibility for the manuscript’s content. The authors confirmed there are no conflicts of interest.
The authors declare that they have no competing interests.
- 5.Bjorkoy G, Lamark T, Brech A, Outzen H, Perander M, Overvatn A, Stenmark H, Johansen T (2005) p62/SQSTM1 forms protein aggregates degraded by autophagy and has a protective effect on huntingtin-induced cell death. J Cell Biol 171:603–614. https://doi.org/10.1083/jcb.200507002 CrossRefPubMedPubMedCentralGoogle Scholar
- 7.Bourdenx M, Dovero S, Engeln M, Bido S, Bastide MF, Dutheil N, Vollenweider I, Baud L, Piron C, Grouthier V et al (2015) Lack of additive role of ageing in nigrostriatal neurodegeneration triggered by alpha-synuclein overexpression. Acta Neuropathol Commun 3:46. https://doi.org/10.1186/s40478-015-0222-2 46-0222CrossRefPubMedPubMedCentralGoogle Scholar
- 19.Ebrahimi-Fakhari D, Cantuti-Castelvetri I, Fan Z, Rockenstein E, Masliah E, Hyman BT, McLean PJ, Unni VK (2011) Distinct roles in vivo for the ubiquitin-proteasome system and the autophagy-lysosomal pathway in the degradation of α-synuclein. J Neurosci 31:14508–14520. https://doi.org/10.1523/jneurosci.1560-11.2011 CrossRefPubMedPubMedCentralGoogle Scholar
- 20.Filichia E, Hoffer B, Qi X, Luo Y (2016) Inhibition of Drp1 mitochondrial translocation provides neural protection in dopaminergic system in a Parkinson's disease model induced by MPTP. Sci Rep, City, p 32656Google Scholar
- 23.Freeman D, Cedillos R, Choyke S, Lukic Z, McGuire K, Marvin S, Burrage AM, Sudholt S, Rana A, O’Connor C et al (2013) Alpha-synuclein induces lysosomal rupture and cathepsin dependent reactive oxygen species following endocytosis. PLoS One 8:e62143. https://doi.org/10.1371/journal.pone.0062143 CrossRefPubMedPubMedCentralGoogle Scholar
- 30.Kandimalla R, Manczak M, Fry D, Suneetha Y, Sesaki H, Reddy PH (2016) Reduced Dynamin-related protein 1 protects against phosphorylated tau-induced mitochondrial dysfunction and synaptic damage in Alzheimer's disease. Hum Mol Genet, 2016;25(22):4881–4897Google Scholar
- 39.Lill CM, Roehr JT, McQueen MB, Kavvoura FK, Bagade S, Schjeide BM, Schjeide LM, Meissner E, Zauft U, Allen NC et al (2012) Comprehensive research synopsis and systematic meta-analyses in Parkinson's disease genetics: the PDGene database. PLoS Genet 8:e1002548. 1002510.1001371/journal.pgen.1002548. https://doi.org/10.1371/journal.pgen.1002548 CrossRefPubMedPubMedCentralGoogle Scholar
- 41.Manczak M, Kandimalla R, Fry D, Sesaki H, Reddy PH (2016) Protective effects of reduced dynamin-related protein 1 against amyloid beta-induced mitochondrial dysfunction and synaptic damage in Alzheimer's disease. Hum Mol Genet, 2016;25(23):5148–5166Google Scholar
- 45.Nair S, Sobotka KS, Joshi P, Gressens P, Fleiss B, Thornton C, Mallard C, Hagberg H (2019) Lipopolysaccharide-induced alteration of mitochondrial morphology induces a metabolic shift in microglia modulating the inflammatory response in vitro and in vivo. Glia 67:1047–1061. https://doi.org/10.1002/glia.23587 CrossRefPubMedGoogle Scholar
- 46.Nalls MA, Pankratz N, Lill CM, Do CB, Hernandez DG, Saad M, DeStefano AL, Kara E, Bras J, Sharma M et al (2014) Large-scale meta-analysis of genome-wide association data identifies six new risk loci for Parkinson's disease. Nat Genet 2014(07/27):989–993. https://doi.org/10.1038/ng.3043 CrossRefGoogle Scholar
- 47.Pan-Montojo F, Schwarz M, Winkler C, Arnhold M, O'Sullivan GA, Pal A, Said J, Marsico G, Verbavatz JM, Rodrigo-Angulo M et al (2012) Environmental toxins trigger PD-like progression via increased alpha-synuclein release from enteric neurons in mice. Sci Rep 2:898. https://doi.org/10.1038/srep00898 CrossRefPubMedPubMedCentralGoogle Scholar
- 49.Pernas L, Scorrano L (2016) Mito-Morphosis: mitochondrial fusion, fission, and cristae remodeling as key mediators of cellular function. Annu Rev Physiol 78:505–531. https://doi.org/10.1146/annurev-physiol-021115-105011 CrossRefPubMedGoogle Scholar
- 57.Ravikumar B, Vacher C, Berger Z, Davies JE, Luo S, Oroz LG, Scaravilli F, Easton DF, Duden R, O'Kane CJ et al (2004) Inhibition of mTOR induces autophagy and reduces toxicity of polyglutamine expansions in fly and mouse models of Huntington disease. Nat Genet 36:585–595. https://doi.org/10.1038/ng1362 CrossRefPubMedGoogle Scholar
- 58.Sarkar S, Malovic E, Harischandra DS, Ngwa HA, Ghosh A, Hogan C, Rokad D, Zenitsky G, Jin H, Anantharam V et al (2018) Manganese exposure induces neuroinflammation by impairing mitochondrial dynamics in astrocytes. Neurotoxicology 64:204–218. https://doi.org/10.1016/j.neuro.2017.05.009 CrossRefPubMedGoogle Scholar
- 66.Wang W, Karamanlidis G, Tian R (2016) Novel targets for mitochondrial medicine. Sci Transl Med 8:326rv323Google Scholar
- 68.Winslow AR, Chen CW, Corrochano S, Acevedo-Arozena A, Gordon DE, Peden AA, Lichtenberg M, Menzies FM, Ravikumar B, Imarisio S et al (2010) Alpha-Synuclein impairs macroautophagy: implications for Parkinson's disease. J Cell Biol 190:1023–1037. https://doi.org/10.1083/jcb.201003122 CrossRefPubMedPubMedCentralGoogle Scholar
- 69.Wu M, Neilson A, Swift AL, Moran R, Tamagnine J, Parslow D, Armistead S, Lemire K, Orrell J, Teich J et al (2007) Multiparameter metabolic analysis reveals a close link between attenuated mitochondrial bioenergetic function and enhanced glycolysis dependency in human tumor cells. Am J Physiol Cell Physiol 292:C125–C136. https://doi.org/10.1152/ajpcell.00247.2006 CrossRefPubMedGoogle Scholar
- 72.Yu WH, Dorado B, Figueroa HY, Wang L, Planel E, Cookson MR, Clark LN, Duff KE (2009) Metabolic activity determines efficacy of macroautophagic clearance of pathological oligomeric alpha-synuclein. Am J Pathol 175:736–747. https://doi.org/10.2353/ajpath.2009.080928 CrossRefPubMedPubMedCentralGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.