Adenosine A_2A Receptors Modulate α-Synuclein Aggregation and Toxicity

Abstract

Abnormal accumulation of aggregated α-synuclein (aSyn) is a hallmark of sporadic and familial Parkinson's disease (PD) and related synucleinopathies. Recent studies suggest a neuroprotective role of adenosine A_2A receptor (A_2AR) antagonists in PD. Nevertheless, the precise molecular mechanisms underlying this neuroprotection remain unclear. We assessed the impact of A_2AR blockade or genetic deletion (A_2AR KO) on synaptic plasticity and neuronal cell death induced by aSyn oligomers. We found that impairment of LTP associated with aSyn exposure was rescued in A_2AR KO mice or upon A_2AR blockade, through an NMDA receptor-dependent mechanism. The mechanisms underlying these effects were evaluated in SH-SY5Y cells overexpressing aSyn and rat primary neuronal cultures exposed to aSyn. Cell death in both conditions was prevented by selective A_2AR antagonists. Interestingly, blockade of these receptors did not interfere with aSyn oligomerization but, instead, reduced the percentage of cells displaying aSyn inclusions. Altogether, our data raise the possibility that the well-documented effects of A_2AR antagonists involve the control of the latter stages of aSyn aggregation, thereby preventing the associated neurotoxicity. These findings suggest that A_2AR represent an important target for the development of effective drugs for the treatment of PD and related synucleinopathies.

α-Synuclein, adenosine A_2A receptors, hippocampus, neuroprotection, Parkinson's disease

Introduction

Parkinson's disease (PD) is a progressive and chronic neurodegenerative disorder characterized by the loss of dopaminergic neurons in the “substantia nigra pars compacta”, which underlies the classical motor symptoms of the disease (Jankovic 2008). Another important neuropathological hallmark of the disease is the presence of intraneuronal inclusions known as Lewy bodies and Lewy neurites (Braak et al. 2003). These deposits occur early in the disease process and are accompanied by progressive neuronal dysfunction and, eventually, death of the afflicted neuronal populations (Braak and Del Tredici 2008). Behavioral and cognitive deficits are concomitant with these pathological changes (Turner 2002). In fact, the accumulation of Lewy bodies and Lewy neurites is not restricted to substantia nigra but is extended to several brain areas including those related to memory, such as the hippocampus and cortex (Spillantini et al. 1997; Mattila et al. 2000; Braak et al. 2004). Moreover, recent “postmortem” studies in PD patient brains suggest a correlation between cognitive deficits or dementia and the accumulation of cortical Lewy bodies (Kövari et al. 2003; Braak et al. 2005; Caviness et al. 2011). These inclusions are mainly composed of fibrillar aggregates of α-synuclein, a neuronal presynaptic protein associated with both familial and sporadic forms of PD (Satake et al. 2009; Simón-Sánchez et al. 2009; Hamza et al. 2010).

Recently, adenosine A_2A receptor (A_2AR) emerged as an attractive nondopaminergic target for the treatment of motor and nonmotor symptoms of PD. Antiparkinsonian actions are achieved through the blockade of this receptor, whose expression and function become aberrant throughout aging and in age-related pathologies, including the early stages of PD (Lopes et al. 1999; Varani et al. 2010; Villar-Menéndez et al. 2014). This strategy was also proven beneficial in other diseases associated with neuronal dysfunction, such as epilepsy, acute and chronic stress and Alzheimer's disease (Cunha 2005; Canas et al. 2009; Batalha et al. 2013; Laurent et al. 2014). In addition, epidemiological studies show an inverse correlation between the consumption of caffeine, an A_2AR antagonist, and the risk of developing PD (Ascherio et al. 2001).

Furthermore, A_2AR deregulation was suggested to play an important role in α-synuclein-mediated neurotoxicity, since α-synuclein-induced damage to striatal neurons was clearly reduced in A_2AR KO mice (Kachroo and Schwarzschild 2012). However, the extent to which A_2AR are involved in α-synuclein-associated toxicity, the underlying protective molecular mechanisms, or the impact on other brain areas is still unknown. Therefore, the purpose of this study was to gain insight into the novel concept of a crosstalk between α-synuclein and A_2AR and to explore the ability of A_2AR to modulate α-synuclein-mediated synaptic dysfunction, formation of inclusions, and neuronal death. To test these hypotheses, we first assessed the functional outcomes of the pharmacological blockade or genetic deletion of A_2AR on rodent hippocampal slices exposed to extracellular α-synuclein oligomers. Then, we set out to determine how A_2AR affect α-synuclein-mediated cell death, and more importantly, whether modulation of A_2AR function impacts on α-synuclein aggregation and oligomerization.

Here, we show for the first time that selective adenosine A_2AR antagonists rescue both exogenous and endogenous α-synuclein-associated cell death. In addition, we found that A_2AR modulation alters the formation of α-synuclein inclusions in cultured cells. Furthermore, the toxic effects of α-synuclein oligomers on synaptic function are fully prevented by A_2AR blockade or deletion, through a mechanism dependent on NMDA receptor. These findings suggest that A_2AR represent an important target for the development of effective drugs for the treatment of PD and related synucleinopathies.

Materials and Methods

Animals

Animal procedures were performed in accordance with the guidelines of the European Community guidelines (Directive 2010/63/EU), Portuguese law on animal care (1005/92), and approved by the “Instituto de Medicina Molecular” Internal Committee and the Portuguese Animal Ethics Committee (“Direcção Geral de Veterinária”). Environmental conditions were kept constant: food and water ad libitum, 21 ± 0.5°C, 60 ± 10% relative humidity, 12-h light/dark cycles. All animals were killed by decapitation after anesthesia under halothane atmosphere. Male Wistar rats (8–12 weeks old) were purchased from Harlan Interfauna Iberica. Global A_2AR KO mice with a C57Bl/6-background were generated by a standard replacement-type vector constructed to inactivate the A_2AR (Chen et al. 1999). Congenic global A_2AR KO mice were made by backcrossing KO on mixed (129-Steel × C57BL/6) genetic background to C57BL/6 mice for 13–15 generations. Heterozygous cross-breeding was used to generate WT and global KO mice. Male KO and WT mice with matched age (8–12 weeks old, male) were used for electrophysiological experiments.

Purification and Oligomerization of Recombinant α-Synuclein

α-Synuclein was prepared as previously (Diógenes et al. 2012; Vicente Miranda et al. 2013). Monomeric α-synuclein was readily used or stored at −80°C until further use. Oligomerization was induced by continuous shaking of monomeric α-synuclein (140 µM) for 6 days at 37°C in a thermomixer (Eppendorf) at 900 rpm. Samples were ultracentrifuged to remove fibrillary α-synuclein. The supernatant containing monomeric and oligomeric α-synuclein was centrifuged in Amicon filter unit with Ultracel membrane NMWL of 30 kDa (Millipore). The retained fraction containing α-synuclein oligomers (>30 kDa) was readily used or stored at −80°C until further use. The concentration of α-synuclein was determined using its molar extinction coefficient at 280 nm (i.e., ɛ₂₈₀ = 5960 L/mol/cm).

SDS–PAGE

The composition of different α-synuclein species, monomers, and oligomers was evaluated by SDS–PAGE. Five micrograms of each α-synuclein sample was separated by SDS–PAGE using a Tetra Cell (Bio-Rad) in a precast 4–15% polyacrylamide gel (Bio-Rad) using standard procedures. Proteins were transferred to a nitrocellulose membrane (Bio-Rad) using the Mini Tans-Blot system (Bio-Rad).

Prestained standard proteins were also loaded on the gel. Membrane was blocked for 1 h at room temperature (RT) with blocking solution (5% bovine serum albumin in 50 mM Tris, 150 mM NaCl, 0.1% and Tween 20, pH 7.5). The membrane was incubated overnight at 4°C with mouse anti-α-synuclein primary antibody (1:1000; BD Transduction Laboratories) diluted in blocking solution. Membrane was washed and incubated for 1 h at RT with anti-mouse-horseradish peroxide (HRP)-conjugated secondary antibody (1:10 000, Invitrogen) diluted in blocking solution. Detection procedures were performed according to ECL system (Millipore) using a chemidoc system (Bio-Rad).

Rat Primary Neuronal Cultures

Hippocampal neurons were cultured from 18 days Sprague Dawley rat (Harlan, Barcelona, Spain) embryos as previously described (Valadas et al. 2012). Briefly, embryos were collected in Hank's Balanced Salt Solution (1 mM Ca²⁺ and 1 mM Mg²⁺) and rapidly decapitated. Meninges and white mater were removed, and whole cortices (hippocampi and attached cortex) were incubated for 15 min in Hank's Balanced Salt Solution and 0.025% trypsin. Cells were centrifuged 3 times and washed with Hank's Balanced Salt Solution (10% fetal bovine serum) and finally re-suspended in Neurobasal medium. Cells were plated on poly-d-lysine-coated coverslips in 24-well plates at density of 8 × 10⁴ cells/well. Neurons were grown for 10 days at 37°C in a 5% CO₂-humidified atmosphere in Neurobasal medium with 2% B-27 supplement, 25 µm glutamate, 0.5 mm glutamine, and 2 U/mL penicillin/streptomycin, in the absence of any positive selection for neurons. Cells with 9 or 10 DIV were treated with extracellular α-synuclein species (500 nm) for 24 h or 90 min, respectively.

Electrophysiological fEPSPs Recordings

The experiments were performed in acute transverse hippocampal slices from male Wistar rats (8–12 weeks old) and in A_2AR KO and WT mice. After decapitation, the brain was rapidly removed and the hippocampi were dissected free in ice-cold artificial CSF or Krebs solution composed of (mM): NaCl 124; KCl 3; NaH₂PO₄ 1.25; NaHCO₃ 26; MgSO₄ 1; CaCl₂ 2; and D-glucose 10, previously gassed with 95% O₂ and 5% CO₂, pH 7.4. Slices (400 μm thick) were obtained with a McEwan tissue chopper and were incubated with or without extracellular α-synuclein oligomers (500 nM) for 90 min at RT in gassed artificial CSF. Incubation with SCH 58261 (50 nM) started 20 min prior to α-synuclein oligomers incubation and was kept throughout the 90 min of α-synuclein incubation (Fig. 2A). Following this incubation period, slices were superfused with artificial CSF (3 mL/min) at 30.5°C and fEPSPs were recorded as previously (Diógenes et al. 2012) in the “stratum radiatum” of the CA1 area. We first carried out input–output (I/O) curves, and then LTP was induced by a theta-burst protocol (10 trains with 4 pulses each at 100 Hz, separated by 200 ms).

SH-SY5Y Cells Inducibly Overexpressing Wild-Type α-Synuclein

Stable SH-SY5Y cell lines inducibly expressing human WT α-synuclein (kind gift from Prof. Kostas Vekrellis, Athens, Greece) were generated as previously described (Vekrellis et al. 2009). Cells were cultured in RPMI 1640^® medium (Life Technologies) supplemented with 10% fetal bovine serum, penicillin (100 U/mL), and streptomycin (100 mg/mL). SH-SY5Y cells were maintained in 250 µg/mL G418 and 50 µg/mL hygromycin B. α-synuclein expression was switched off by the addition of doxycycline (Dox) (2 µg/mL). Stock cultures were kept in the presence of Dox. Overexpressing α-synuclein cells were maintained for 12 DIV in the absence of Dox. For propidium iodide (PI) and Syto-13 uptake assay cells were plated onto 12-well plates (3.8 cm²) at a density of 6 × 10⁴ cells/well, 24 h before drug exposure. For western blot analysis, the cells were seeded into 6-well plates at a density of 15 × 10⁴ cells/well.

H4 Cells Stably Expressing VN-Syn/Syn-VC

For the α-synuclein dimerization model, human H4 neuroglioma cells stably expressing 2 α-synuclein BiFC constructs were used (Outeiro et al. 2008). This assay is based on the reconstitution of functional fluorescent proteins promoted by the interaction between, at least, 2 α-synuclein molecules, which enables the direct visualization of α-synuclein dimeric/oligomeric species formation. The 2 BiFC constructs used were generated by fusing half of the fluorescent Venus protein with α-synuclein in the N-terminal and the other half fused with α-synuclein in the C-terminal.

Cells were maintained in OPTI-MEM^® (Life Technologies) supplemented with 10% fetal bovine serum, penicillin (100 U/mL), and streptomycin (100 mg/mL) in a humidified atmosphere of 5% of CO₂ in air at 37°C. Cells were plated onto 12-well plates and 24 h before treatment. Twenty-four hours after drug treatment, cells were washed with phosphate buffer saline (PBS: NaCl 137 mM, KCl 2.7 mM, KH₂PO₄ 1.8 mM and Na₂HPO₄ 10 mM, pH 7.4) and fixed with 4% paraformaldehyde (PFA) for 10 min at RT, followed by a 10 min incubation with Hoescht 33258 dye (1 mg/mL, Life Technologies-Invitrogen) at RT. Cells were then washed and maintained in PBS and imaged on an Olympus IX81-ZDC microscope system (Olympus Germany) using the 20× objective and maintaining the same exposure time for Venus and Hoechst channels for each condition. Quantification of the number of cells and average Venus fluorescence intensity was performed using an in-house developed macro for ImageJ (http://imagej.nih.gov/ij/). Briefly, single-cell nuclei were identified using the Hoechst channel by thresholding and particle analysis, and the corresponding regions of interest (ROIs) were then used to measure the average intensity in the Venus channel. The number of cells with α-synuclein dimers was determined by counting the ROIs where the average fluorescent intensity was higher than a given threshold. Values of each condition were then averaged and statistical analysis was performed.

Human Neuroglioma H4 Cells

Human neuroglioma H4 cells were maintained in OPTI-MEM^® (Life Technologies) supplemented with 10% fetal bovine serum, penicillin (100 U/mL), and streptomycin (100 mg/mL) in a humidified atmosphere of 5% of CO₂ in air at 37 °C. Cells were plated in 12-well plates 24 h prior to transfection. Cells were transfected with equimolar amounts of the plasmids encoding the human WT α-synuclein with a C-terminal tag corresponding to a truncated fragment of EGFP (referred to as SynT) and synphilin-1 as previously (Lázaro et al. 2014). Twenty-four hours after transfection, the cells were incubated for 24 h with different A_2AR modulators and, after this period, the cells were subjected to immunocytochemistry for studying α-synuclein inclusions. Transfected cells were identified and classified in 2 groups: cells without inclusions and cells with one or more inclusions. Results were expressed as the percentage of the total number of transfected cells.

Immunocytochemistry

Forty-eight hours after transfection, H4 cells were washed with PBS and fixed with 4% PFA for 10 min at RT, followed by a permeabilization step with 0.5% Triton X-100 (Sigma–Aldrich) for 20 min at RT. After blocking in 1.5% normal goat serum (PAA)/DPBS for 1 h, the cells were incubated with mouse anti-α-synuclein primary antibody (1:1000, BD Transduction Laboratories) overnight at 4°C. After a 30-min washing with PBS, the cells were incubated with the secondary antibody Alexa Fluor 488 donkey anti-mouse IgG (Life Technologies-Invitrogen) for 2 h at RT. Finally, the cells were stained with Hoechst 33258 (1 mg/mL, Life Technologies-Invitrogen) (1:5000 in DPBS) for 10 min and maintained in PBS for epifluorescence microscopy.

Propidium Iodide and Syto-13 Uptake Assay

This protocol was used either in primary neuronal cultures or SH-SY5Y cells and performed as previously described (Valadas et al. 2012). Briefly, cells were washed with Krebs-HEPES (NaCl 117 mM, KCl 3 mM, glucose 10 mM, NaHCO₃ 26 mM, Na₂HPO₄ 1.25 mM, HEPES 10 mM, CaCl₂ 2 mM, MgCl₂ 1 mM), incubated with Syto-13 (4 µM) and PI (5 µg/mL) for 3 min at RT and directly observed using an Axiovert 200 fluorescence microscope. An average of 1400 cells were counted per condition in each experiment. Syto-13 labels with green fluorescence (emits preferentially at 509 nm when excited at 488 nm) both RNA and DNA in living cells. PI labels with red fluorescence (absorbing preferentially at 535 nm and emitting at 617 nm) cells that lost plasma membrane integrity. Cell viability was presented as the ratio between the number of living cells and the total number of cells.

Real-Time qPCR

Total RNA from H4 and SH-SY5Y cells was extracted using the RNAspin Mini RNA isolation kit (GE Healthcare). Briefly, cell cultures were washed with PBS, scraped, collected in lysis buffer and processed according to the manufacturer's instructions. RNA was quantified with the NanoDrop 2000 (Thermo scientific). Total RNA (2 μg) was reverse-transcribed using random primers and SuperScript™ First-Strand Synthesis System for RT-PCR (Invitrogen, Life technologies), and negative controls were made without reverse transcriptase. qPCR was carried out with Power SYBR Green PCR Master mix (Applied Biosystems), using 8 ng/µL of total cDNA and 0.2 µm of each primer, performed in a Rotor-Gene 6000 Real Time Rotary Analyzer (Corbett Research). The thermal cycler conditions were 10 min at 95°C, 40 cycles of a two-step PCR, 95°C for 15 s followed by 60°C for 25 s with a final thermal ramp from 72 to 95°C. The primers used in qPCR include: forward 5′-AACCTGCAGAACGTCAC-3′ and reverse 5′-GTCACCAAGCCATTGTACCG-3′ for human A_2AR (Invitrogen, HPLC purified, product size 245 bp) and forward 5′-GGACTTCGAGCAAGAGATGG-3′ and reverse 5′-AGCACTGTGTTGGCGTACAG-3′ for human β-actin (Invitrogen, HPLC purified, product size 233 bp). The qPCR products were analyzed by electrophoresis on a 2% agarose gel containing Greensafe Premium Nucleic Acid Gel Stain (Nzytech).

Co-immunoprecipitation (Co-IP)

Briefly, WT rat hippocampal slices were homogenized in IP buffer (NP40 1%, SDS 0.1%, Tris–HCl 50 mM, NaCl 150 mM, sodium deoxycholate 0,5%, EDTA 1 mM, protease inhibitors—Complete, EDTA-free Protease Inhibitor cocktail tablets; Roche). Protein extracts were incubated with protein G PLUS-Agarose (Santa Cruz Biotechnology) for 1 h at 4°C to eliminate nonspecific binding. After incubation, the precleared supernatants containing 1 mg of protein were incubated with anti-PSD-95 antibody (1:50; Cell Signaling Technology) or IgG (for negative control; Santa Cruz Biotechnology) overnight at 4°C under rotation. The day after, lysates were incubated with protein G PLUS-Agarose for 3 h with rotation at 4°C. Beads were washed 3 times with IP buffer and resuspended in 1.5× sample buffer pH (Tris 359 mM pH 6.8, glycerol 30%, sodium dodecyl sulfate 10%, dithiothreitol 600 mM, and bromophenol blue 0.012%). Bound proteins eluted from the immune complexes were denatured by heating to 95°C for 5 min and used for western blot analysis. Western blot was performed with anti-NMDA receptor subunit 2B (1:1000; Cell Signaling technology), anti-NMDA receptor subunit 1 (1:500; BD Pharmingen™), and anti-PSD-95 (1:1000; Cell Signaling Technology) (see western blot).

Western Blotting

SH-SY5Y cells were washed with cold PBS and then mechanically scrapped in radioimmunoprecipitation assay buffer pH 8.0 (RIPA buffer: NaCl 150 mM, Tris-base 50 mM, EDTA 1 mM, Nonidet P40 1%, sodium dodecyl sulfate 0.1%, proteases inhibitors—Complete, EDTA-free Protease Inhibitor cocktail tablets; Roche). Cells were centrifuged at 16 000×g during 10 min at 4°C and the pellet, including cell debris, was discarded and the supernatant used for western blot. After protein quantification using Bio-Rad DC Protein Assay kit, lysates were denatured with 5× sample buffer pH 6.8 (Tris 359 mM pH 6.8, glycerol 30%, sodium dodecyl sulfate 10%, dithiothreitol 600 mM, and bromophenol blue 0.012%) and heated at 95°C for 5 min and further processed as before (Valadas et al. 2012). Samples and the prestained molecular weight marker (BIO-RAD) were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE; 15% gel) under reducing conditions and electro-transferred to polyvinylidene difluoride membranes (0.45 µm, Immobilon) using standard procedures. Thereafter, nonspecific binding was blocked with 3% bovine serum albumin (fatty acid free) in Tris-buffered saline (pH 7.6) containing 0.1% Tween 20 (TBS-T) for 1 h at RT. Membranes were then incubated overnight at 4°C with the corresponding primary antibody, namely mouse anti-α-synuclein (1:1000; BD transduction lab), mouse anti-adenosine A_2AR (1:1000; Millipore), rabbit anti-α-tubulin (1:5000; Abcam), mouse anti-GAPDH (1:1000; Ambion), rabbit anti-PSD-95 (1:1000; Cell Signaling), rabbit anti-NMDA receptor subunit 2B (1:1000; Cell Signaling), mouse anti-NMDA receptor subunit 1 (1:500; BD Pharmingen™) diluted in blocking solution. After 3 washing periods of 10 min with TBS-T, membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-mouse or anti-rabbit secondary antibodies (1:10 000; Santa Cruz Biotechnology) (in 5% nonfat dry milk) for 1 h at RT. After 40 min of washing with TBS-T, chemiluminescent detection was performed with ECL western blotting detection reagent (GE Healthcare) using X-Ray films (Fujifilm). Densitometric quantification was determined using Image-J software and normalized to the corresponding α-tubulin band density.

Drugs

5-Amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo[4,3-e]-1,2,4-triazolo[1,5-c]pyrimidine (SCH 58261; Tocris Cookson) and 4-(2-[7-amino)]-2-(2-furyl{1,2,4}-triazolo{2,3-a{1,3,5}triazin-5-yl-aminoethyl)phenol} (ZM 241385; Tocris Cookson) were prepared as 5 mm stock solutions in dimethylsulfoxide (DMSO). 2-[p-(2-Carboxyethyl)-phenylethylamino]-5′-N-ethylcarboxamidoadenosine (CGS 21680; Tocris Cookson) was prepared as a 12 µM stock solution in DMSO. N⁶-Cyclopentyladenosine (CPA; Tocris) was prepared as a 5 mM stock solution in DMSO. 1,3-Dipropyl-8-cyclopentyladenosine (DPCPX; Tocris Cookson) was prepared as a 5 mM stock solution in 99% DMSO and 1% NaOH 1 M. DL-2-Amino-5-phosphonopentanoic acid (APV; Abcam) was prepared as a 100 mM stock solution in NaOH 100 mm. All aliquots were kept frozen at −20°C until use.

Statistics

The values presented are mean ± SEM of n independent experiments. To test the significance of the differences between 2 conditions, a Student's t test was used. In statistical tests between 3 or more conditions, a one-way ANOVA followed by a Bonferroni's multiple comparison post hoc test was used. P-values of <0.05 were considered to be statistically significant.

Results

A_2AR Blockade Rescues Neuronal Cell Death Induced by Exogenous α-Synuclein Oligomers

First, we compared the effect of different α-synuclein species on rat primary hippocampal cultures. Cells were incubated with either monomeric or oligomeric α-synuclein preparations for different time periods. The confirmation of the biochemical properties of the different species was performed by SDS–PAGE/immunoblot analysis, revealing that monomers migrated with a typical molecular weight of ∼ 15 kDa (monomeric molecular weight of α-synuclein), whereas oligomeric forms migrated with an apparent molecular weight of >30 and <180 kDa (Fig. 1D). We found no significant cytotoxicity in cells treated with either α-synuclein species (500 nM) for 90 min, as assessed by PI and Syto-13 staining. Conversely, we found that after 24 h of exposure there was a robust increase in neuronal cell death upon incubation with α-synuclein oligomers (cell viability_CTR = 68.7 ± 4.6%; cell viability_{aSyn olig, 24h} = 35.5 ± 3.1%; n = 4; *P < 0.001; Fig. 1A,B), whereas monomeric species had no effect on cell viability.

Figure 1.

Neuronal death induced by extracellular α-synuclein oligomers is prevented by A_2AR blockade. (A) Representative images of hippocampal cultures labeled with propidium iodide (PI) and Syto-13 in (i) control conditions (CTR), (ii) after 24-h incubation with α-synuclein monomers (aSyn mon, 500 nM), (iii) or with α-synuclein oligomers (aSyn olig, 500 nM) alone or in the presence of the selective A_2AR antagonists, (iv) SCH 58261 (50 nM) or (v) ZM 241385 (50 nM). Scale bar: 50 µm. (B) Cell viability upon incubation with extracellular aSyn mon or olig. Only 24-h incubation with aSyn olig leads to a decrease in cell viability compared with CTR. (C) Rescue of cell viability after exposure to aSyn olig for 24 h by the selective A_2AR antagonists, SCH 58261 (50 nM) or ZM 241385 (50 nM). (D) SDS–PAGE separation of the different aSyn species (monomers and oligomers). Monomers migrate with monomeric molecular weight (15 kDa) whereas aSyn oligomers display SDS-resistant high-molecular-weight species. *P < 0.001. Cell viability is presented as the percentage of living cells compared with the number of total cells counted. All values are mean ± SEM of 4 independent experiments.

We next investigated whether adenosine A_2AR blockade could attenuate the neurotoxic effects induced by exposure to α-synuclein oligomers. The blockade of A_2AR, using the selective antagonists SCH 58261 (50 nM) or ZM 241385 (50 nM), significantly reduced neuronal cell death induced by α-synuclein oligomers, resulting in levels similar to control (cell viability_SCH 58261+ _{aSyn olig, 24 h} = 70.6 ± 4.3%; cell viability_{ZM 241385+ aSyn olig, 24 h} = 69.7 ± 3.2%, n = 4; *P < 0.001 vs. CTR; Fig. 1A,C).

Pharmacological or Genetic Blockade of A_2AR Prevents LTP Impairment Induced by α-Synuclein Oligomers

We have previously demonstrated that α-synuclein oligomers, but not monomers or fibers, impair synaptic plasticity and increase basal synaptic transmission through NMDA receptor activation (Diógenes et al. 2012). Thus, we next set out to investigate the role of A_2AR on this impairment of synaptic function. For this, we preincubated rat hippocampal slices with α-synuclein oligomers together with the selective A_2AR antagonist SCH 58260 (50 nM, 110 min; Fig. 2A) and induced LTP in Schaffer-collaterals/CA1 pyramid glutamatergic synapses by theta-burst stimulation. The LTP amplitude was significantly reduced in slices preincubated with α-synuclein oligomers (500 nM, 90 min) when compared with control slices (LTP_CTR = 58.1 ± 7.4%; LTP_{aSyn olig} = 7.7 ± 2.3%; n = 9; *P < 0.001; Fig. 2B–D). When A_2AR were blocked by SCH 58261, the LTP magnitude was reestablished to control values (LTP_{SCH 58261+aSyn olig} = 50.6 ± 13.0%; n = 5; *P < 0.01 vs. LTP_{aSyn olig}; Fig. 2B–D). SCH 58261 alone did not affect LTP magnitude (LTP_SCH 58261 = 52.5 ± 10.6%; n = 5; *P < 0.05 vs. LTP_CTR; Fig. 2D).

Figure 2.

A_2AR blockade rescues LTP impairment induced by extracellular α-synuclein oligomers. (A) Schematic representation of hippocampal slices incubation protocol. (B) Representative traces (1) prior and (2) after LTP induction, composed of the stimulus artifact followed by the presynaptic volley and the fEPSP. (C) Changes in fEPSP slope upon LTP induced by theta-burst stimulation from hippocampal rat slices (CTR: control; aSyn olig: after preincubation with α-synuclein oligomers, 500 nM, 90 min; aSyn olig + SCH 58261: after incubation with α-synuclein oligomers in the presence of the A_2AR antagonist, SCH 58261, 50 nM, 110 min). SCH 58261 rescued LTP impairment induced by α-synuclein oligomers. *P < 0.01. (D) LTP magnitude after theta-burst stimulation (change in fEPSP slope at 50–60 min). (E) Effect of NMDA receptor antagonist APV (50 µM, 30 min) superfusion on basal fEPSP slope. SCH 58261 prevented the effect of α-synuclein oligomers on NMDA receptor contribution for basal synaptic transmission. *P < 0.05. (F) Quantification of the effects observed in (E) (change in slope between baseline and the last 10 min of APV application). (G) Input/output (I/O) curves corresponding to fEPSP slope evoked by different stimulation intensities (60–300 µA). Slices co-incubated with aSyn olig and SCH 58261 displayed higher Emax values when compared with control slices, similar to what was observed with aSyn olig alone. (H) Co-immunoprecipitation of PSD-95 in hippocampal slices. NMDA receptor subunit 2B (NMDAR2B) are enriched in aSyn olig preincubated slices whereas co-incubation with SCH 58261 reestablished NMDAR2B subunit levels. NMDA receptor subunit 1 (NMDAR1) levels were not changed in any condition. Values were normalized to PSD-95. IgG was used as a negative control (Neg CTR). *P < 0.05. All values are mean ± SEM of 3–9 independent experiments.

To assess the role of A_2AR on NMDA receptor-mediated effects, we evaluated the effect of the NMDA receptor antagonist APV (50 µM) on basal synaptic transmission. As expected, APV did not modify the fEPSP slope in control slices (Fig. 2E,F). In contrast, the acute application of APV induced a progressive reduction of the fEPSP in α-synuclein oligomer-treated slices (fEPSP_{aSyn olig} = −13.5 ± 1.8%; n = 7; *P < 0.001; Fig. 2E,F), in agreement with the previously reported impact of oligomeric α-synuclein on NMDA receptor (Diógenes et al. 2012). Interestingly, when slices were preincubated with SCH 58261 together with α-synuclein oligomers, the effect of the NMDA receptor antagonist was prevented (fEPSP_{SCH 58261+aSyn olig} = −6.4 ± 1.3%; n = 6; *P < 0.05 vs. fEPSP_{aSyn olig}; Fig. 2E,F). Accordingly, we observed an increase in NMDA receptor subunit 2B (NMDAR2B 137.4 ± 4.3%; n = 4; *P < 0.05 vs. CTR; Fig. 2H) in slices exposed to α-synuclein, but not in NMDA subunit 1 (NMDAR1). This increase was prevented by co-incubation with SCH 58261 (101.9 ± 8.3%; n = 3; *P < 0.05 vs. CTR; Fig. 2H). SCH 58261 alone did not alter NMDA receptor subunit 2B levels.

In order to evaluate whether A_2AR blockade also rescued baseline synaptic efficiency, I/O curves were recorded. Slices preincubated with α-synuclein oligomers alone showed a shift to the left in the I/O curve, as previously described (Diógenes et al. 2012). SCH 58261 co-incubation did not change this α-synuclein-induced effect (Fig. 2G).

We then assessed the synaptotoxic effects of α-synuclein oligomers on hippocampal slices from A_2AR KO mice. Remarkably, α-synuclein oligomers (500 nM, 90 min) failed to impair LTP in A_2AR KO mice (LTP_CTR = 34.3 ± 6.9%; LTP_{aSyn olig} = 31.2 ± 4.2%; n = 3; *P < 0.05; Fig. 3D,E), in contrast to the significant effect in WT mice (LTP_CTR = 64.4 ± 14.2%, n = 4; LTP_{aSyn olig} = 10.9 ± 12.9%; n = 3; *P < 0.05; Fig. 3A,B). As observed in rat hippocampal slices, α-synuclein oligomers also increased basal synaptic excitability, as observed by the shift to the left of the I/O curve, both in WT and A_2AR KO mice (Fig. 3C,F).

Figure 3.

A_2AR deletion fully prevents LTP impairment induced by extracellular α-synuclein oligomers. (A) Changes in fEPSP slope induced by theta-burst stimulation recorded from WT mice hippocampal slices in control conditions (CTR) or in the presence of α-synuclein oligomers (aSyn olig, 500 mM, 90 min). (B) Plot of the LTP magnitude (change in fEPSP slope at 50–60 min comparing to baseline) from (A). (C) I/O curves from WT hippocampal slices, corresponding to fEPSP slope evoked by various stimulation intensities (400–700 μA) in CTR and upon preincubation with aSyn olig. (D) Changes in fEPSP slope induced by theta-burst stimulation recorded from global A_2AR KO mouse hippocampal slices in control conditions (CTR) or in the presence of α-synuclein oligomers (aSyn olig). Genetic deletion of A_2AR prevented LTP impairment induced by aSyn olig in WT mice slices. (E) Plot of the LTP magnitude from experiments shown in (D). (F) I/O curves from A_2AR KO hippocampal slices obtained by the same method as in (C). aSyn olig have a comparable effect both in WT or in A_2AR KO mice. All values are mean ± SEM of 3–4 independent experiments.

A_2AR Blockade Rescues Cell Death Induced by Endogenous α-Synuclein

To evaluate the role of A_2AR on the toxicity induced by intracellular α-synuclein, we used an established Dox inducible (tet-off) neuroblastoma cell model that we verified to express A_2AR (Supplementary Fig. 1A). Maximal α-synuclein expression was reached at 12 days after Dox removal (Supplementary Fig. 1B) and, at this time point, we found a significant decrease in cell viability compared with control cells in the presence of Dox (+Dox) (cell viability_{+Dox CTR} = 91.6 ± 1.5%; cell viability_{–Dox CTR} = 68.7 ± 1.6%; n = 8–9; *P < 0.001; Fig. 4A,B); the A_2AR mRNA or protein levels were not altered (Supplementary Fig. 1C,D). This toxicity was completely prevented by a 24-h treatment with the selective A_2AR antagonists, SCH 58261 or ZM 241385 (cell viability_{−Dox+SCH 58261} = 87.1 ± 5.1%; cell viability_{−Dox+ZM 241385} = 84.8 ± 5.5%; n = 4–7; *P < 0.01 vs. +Dox CTR; Fig. 4A,B). Furthermore, activation of A_2AR with the selective A_2AR agonist CGS 21680 (30 nm) in control cells in the presence of Dox resulted in cytotoxicity, mimicking α-synuclein overexpression (−Dox CTR) (Fig. 4C). Importantly, these effects were not due to changes in α-synuclein levels induced by the different treatments (Supplementary Fig. 1E).

Figure 4.

A_2AR blockade rescues SH-SY5Y cell toxicity induced by increased levels of endogenous α-synuclein. (A) Representative images of 12 days in vitro (DIV) tet-off SH-SY5Y cells labeled with PI and Syto-13 (i) in the presence (ii) or absence of doxycycline (Dox; 2 µg/mL) and (iii–vi) under different treatments. Scale bar: 50 µm. (B) Cell viability with different cell treatments. The overexpression of α-synuclein, induced by the absence of Dox (−Dox), lead to a decrease in the number of viable cells, which was prevented by the selective A_2AR antagonists, SCH 58261 (50 nM) or ZM 241385 (50 nM). (C) Impact of the A_2AR agonist, CGS 21680 (30 nM) on cell viability. The treatment of control cells (CTR +Dox) with the A_2AR agonist lead to a similar decrease in cell death as observed in α-synuclein-overexpressing cells (CTR −Dox); this increase in cell death was prevented in both conditions upon treatment with the NMDA receptor antagonist, APV (50 µM). (D) Impact of the A₁R selective antagonist, DPCPX (50 nM) on cell viability. The treatment with the DPCPX induced a similar decrease in cell viability either in CTR +Dox or α-synuclein-overexpressing cells (CTR −Dox). Cell viability is presented as the ratio between the number of living cells and the number of total cells counted. *P < 0.01. All values are mean ± SEM.

Next, we investigated whether NMDA receptors were involved in this effect. When NMDA receptors were blocked by APV alone, the toxicity induced by overexpression of α-synuclein (−Dox) was completely rescued (cell viability_−Dox+APV = 92.5 ± 1.7%; n = 4; *P < 0.001 vs. −Dox CTR; Fig. 4A,C). This effect is the same of that observed upon A_2AR blockade. Consistently, blockade of NMDA receptor also prevented toxicity induced by direct A_2AR activation with CGS 21680 (Fig. 4A,C).

To rule out any possible contribution of the more abundant adenosine receptor, A₁ receptor (A₁R), to the α-synuclein-induced effects, we selectively blocked this receptor (DPCPX, 50 nM).

DPCPX alone (50 nM) reduced cell viability by 20%, as expected (Valadas et al. 2012). A₁R are not involved in α-synuclein-induced toxicity since this reduction was similar in either +Dox or −Dox cells (Fig. 4D). Moreover, the effect of A_2AR activation by CGS 21680 is independent of A₁R, being maintained even under A₁R blockade (+Dox + DPCPX+CGS 21680). Finally, the increase in cell viability achieved by A_2AR blockade (ZM 241385) in −Dox cells is still present even under A₁R blockade (Fig. 4D).

A_2AR Blockade Does not Affect α-Synuclein Oligomerization

We next evaluated whether A_2AR affected the initial events of α-synuclein aggregation, by monitoring the ability of α-synuclein to oligomerize, using a stable cell model of α-synuclein dimerization/oligomerization based on BiFC (Outeiro et al. 2008) (Fig. 5A). We found that α-synuclein formed oligomers in H4 control cells (CTR; Fig. 5C), and no significant differences in the oligomerization pattern were detected upon A_2AR modulation, as assessed by Venus fluorophore reconstitution (Fig. 5B,C). This lack of effect does not stem from a reduced dynamic range, as we have previously validated the reversibility of the system in response to multiple conditions (Outeiro et al. 2008; Zondler et al. 2014).

Figure 5.

A_2AR modulators do not change α-synuclein dimerization in living cells using Bimolecular Fluorescence Complementation (BiFC) assay. (A) Schematic representation of the BiFC assay. α-Synuclein BiFC constructs in anti-parallel orientation. (B) Quantification of the mean fluorescence intensity of cells showing no significant difference in cells treated with the A_2AR antagonists ZM 241385 (50 nM) or SCH 58261 (50 nM) nor with the A_2AR agonist CGS 21680 (30 nM). Results were normalized to control condition (CTR). All values are mean ± SEM of 5 independent experiments. (C) Representative images of H4 cells with different treatments. Scale bar: 50 µm.

A_2AR Blockade Decreases α-Synuclein Aggregation

Finally, we asked whether A_2AR affected latter steps in the process of α-synuclein aggregation. For this, we used an established model of α-synuclein inclusion formation in human neuroglioma H4 cells (McLean et al. 2001; Lázaro et al. 2014) that also express A_2AR (Fig. 6C,D). We found that a 24-h treatment with the A_2AR antagonist ZM 241385 (50 nM) significantly reduced the percentage of cells displaying α-synuclein inclusions (α-synuclein inclusions_CTR = 70.3 ± 2.0%; α-synuclein inclusions_{ZM 241385} = 58.4 ± 4.2%, n = 3–4; *P < 0.05; Fig. 6A,B). In contrast, activation of A_2AR with CGS 21680 (30 nM) increased the percentage of cells containing α-synuclein inclusions (α-synuclein inclusions_{CGS 21680} = 83.8 ± 1.1%, n = 3; *P < 0.05 vs. α-synuclein inclusions _CTR; Fig. 6A,B).

Figure 6.

A_2AR blockade decreases α-synuclein aggregation in a SynT-Synphilin-1 neuroglioma cell model. (A) Representative images of H4 cells in (i) control conditions (CTR), (ii) incubated for 24 h with ZM 241385 (50 nM), or (iii) CGS 21680 (30 nM). Scale bar: 50 µm. (B) ZM 241385 significantly reduced the percentage of cells containing α-synuclein inclusions whereas CGS 21680 increased the percentage of cells with inclusions, comparing with control cells. Results are expressed as the percentage of the total number of transfected cells. *P < 0.05. All values are mean ± SEM of 3 independent experiments. (C) qPCR products of A_2AR (245 bp) in H4 cells. β-Actin (233 bp) was used as housekeeping control, and RT-minus control yielded no appreciable bands in the expected band size for the primers used. (D) Western blot showing A_2AR expression in H4 cells.

Discussion

We have gathered evidence indicating that A_2AR play an important role in modulating the deleterious effects of α-synuclein. Here, we show, for the first time, that selective A_2AR antagonists are able to rescue both exogenous and intracellular α-synuclein-mediated cyto- and neurotoxicity. Furthermore, pharmacological and genetic inactivation of A_2AR fully prevents the α-synuclein-mediated toxic effects on synaptic function. This neuroprotective effect afforded by A_2AR inhibition is due to the reestablishment of glutamate NMDA receptor signaling. Finally, we found that A_2AR antagonists are also able to decrease the number of α-synuclein aggregates, which might explain their protective effects on α-synuclein-associated synaptic dysfunction and neuronal death.

α-Synuclein aggregation and progressive neuronal cell death are the neuropathological hallmarks of several neurodegenerative disorders known as synucleinopathies (Marques and Outeiro 2012). However, the precise molecular mechanisms underlying the process of α-synuclein aggregation and the exact nature of the toxic species produced during aggregation remain unclear. α-Synuclein is a soluble protein expressed ubiquitously in the central nervous system, including the cerebral cortex, hippocampus, amygdala, and olfactory bulb (Maroteaux and Scheller 1991; Abeliovich et al. 2000). Under pathological conditions, this protein exhibits a propensity to misfold and aggregate, first into small oligomeric species that are rich in β-sheet structure, and then into higher-molecular-weight insoluble fibrils (Spillantini et al. 1998; Lashuel et al. 2012). As reported for amyloid-beta plaques in Alzheimer's disease, soluble oligomeric species are thought to constitute the most neurotoxic species (Kayed et al. 2003; Glabe and Kayed 2006; Irvine et al. 2008; Outeiro et al. 2008; Emadi et al. 2009; Martin et al. 2012). Consistently, we observed that mature extracellular α-synuclein oligomers have the ability to induce synaptic impairment through an NMDA receptor-dependent mechanism (Diógenes et al. 2012). We now report that exposure to α-synuclein oligomers causes damage in primary neuronal cultures, ultimately leading to neuronal death. In contrast, α-synuclein monomers have no effect.

An imbalance in adenosine levels in the brain together with an abnormal function and increased A_2AR levels, which facilitates excitotoxicity and consequent neuronal death, has been reported in multiple conditions such as ischemia, stress, epilepsy, Alzheimer's and PD (Lopes et al. 1999; Latini and Pedata 2001; Rebola et al. 2005; Cunha et al. 2006; Varani et al. 2010; Batalha et al. 2013; Villar-Menéndez et al. 2014). Based on this idea, A_2AR antagonists started emerging as promising candidates in modulating the demise of different psychiatric and neurological disorders, including PD. In fact, it was demonstrated that consumption of caffeine, a nonspecific A_2AR antagonist, reduces the risk of developing PD (Ascherio et al. 2001). Indeed, polymorphisms in the human A_2AR gene (ADORA2A) are linked to a reduced risk of PD (Popat et al. 2011). The specific blockade of these receptors was shown to be protective in several PD models (Chen et al. 2001; Ikeda et al. 2002; Aguiar et al. 2006; Kachroo et al. 2010; Xu et al. 2010); including in α-synuclein-mediated neurotoxicity (Kachroo and Schwarzschild 2012). The crossing of A_2AKO with hm²-α-synuclein mice resulted in reduced neuronal loss, suggesting the potential involvement of A_2AR on α-synuclein-associated toxicity. However, in this report, the impact of A_2AR on α-synuclein oligomerization or aggregation was not determined (Kachroo and Schwarzschild 2012).

In order to clarify the molecular basis of A_2AR-mediated protection against α-synuclein toxicity, we now evaluated the A_2AR effect in multiple models of α-synuclein aggregation and toxicity. Our data demonstrate that the selective blockade or deletion of A_2AR prevents both synaptic plasticity impairment and neuronal death induced by extracellular α-synuclein oligomers. The ability to respond to theta-burst stimulation, when exposed to α-synuclein mature oligomers, is restored either in hippocampal slices from A_2AR KO mice, or WT slices in the presence of a specific A_2AR antagonist. As we have previously reported, α-synuclein oligomers promote an increase in basal synaptic transmission both by the activation of NMDA receptor and by the insertion of Ca²⁺-permeable AMPA receptors in the postsynaptic membrane (Diógenes et al. 2012), which leads to synapse saturation and consequent LTP impairment. Since Ca²⁺-permeable AMPA receptors are crucial for LTP maintenance (Plant et al. 2006), the complete rescue of LTP impairment by A_2AR antagonist suggests a reestablishment of this AMPA impaired trafficking. In fact, it has been described that A_2AR have the ability to modulate the membrane levels of Ca²⁺-permeable AMPA receptors (Dias et al. 2012), which can explain the observed effects. Furthermore, we show that the basal overactivation of NMDA receptor caused by α-synuclein oligomers is also prevented by A_2AR blockade, since NMDA receptor basal contribution is no longer observed. However, while glutamatergic transmission is restored, the effects of α-synuclein on basal synaptic transmission were not rescued by A_2AR blockade or in A_2AR KO mice, as reflected in the unmodified steeper I/O curve. A_2AR do not only affect glutamatergic transmission, but can also directly enhance inhibitory GABAergic transmission, leading to disinhibition of pyramidal cells (Rombo et al. 2014). A possible explanation for the lack of effect of A_2AR blockade on the I/O curve might be due to a resulting overall excitation, caused by a decrease in the inhibitory GABAergic tonus.

In addition to these effects in early synaptic dysfunction, A_2AR antagonists were also effective in preventing subsequent neuronal death in neuronal cultures exposed to α-synuclein oligomers. Our observations are consistent with previous reports showing that pharmacologic or genetic modulation of A_2AR can prevent neurotoxicity and the extent of neuronal damage in neurons affected by ischemia, hypoxia, stress, or β-amyloid exposure (Cunha 2005; Canas et al. 2009; Valadas et al. 2012). To further detail the mechanism of A_2AR action on α-synuclein-induced toxicity, we used an established SH-SY5Y cell model of endogenous α-synuclein accumulation (Vekrellis et al. 2009). Cells overexpressing α-synuclein for 12 DIV show increased cell death that is rescued by selective A_2AR antagonists. Furthermore, if we activate A_2AR in control cells (+Dox), we mimic the cytotoxic effects induced by α-synuclein accumulation (−Dox), supporting the idea of the involvement of A_2AR overactivation on α-synuclein-induced cell death. Together, these data raises the hypothesis that α-synuclein accumulation is leading to a toxic overactivation of A_2AR. This can result either from, overexpression of A_2AR in these conditions or, alternatively, to an increase of the endogenous ligand adenosine. The fact that the overexpression of α-synuclein does not alter A_2AR levels favors the latter hypothesis. Under physiological conditions, adenosine is tonically activating the abundant A₁R and producing synaptic inhibition (Dunwiddie and Diao 1994; Takahashi et al. 1995; Dittman and Regehr 1996). However, in this situation, A₁R do not seem to contribute to α-synuclein-induced toxicity, since when we blocked the receptors, we could not see any differences in cell viability upon α-synuclein accumulation. This is in accordance with different sources of adenosine activating A₁R and A_2AR (Cunha 2008).

Together our results suggest that α-synuclein-induced cell death is associated with an increase in A_2AR activation that mediates NMDA receptor overactivation, which is a prominent synaptic event leading to excitotoxicity (Besancon et al. 2008). In fact, A_2AR are known to increase NMDA receptor function in the hippocampus (Rebola et al. 2008), namely promoting Ca²⁺ entry through NMDA receptor, by (PKA)-dependent regulation (Higley and Sabatini 2010). Interestingly, the same mechanisms of Ca²⁺ entry dysfunction (Martin et al. 2012) and NMDA receptor activation (Diógenes et al. 2012) are involved in α-synuclein-associated neurotoxicity. This suggests that A_2AR blockade is probably counteracting these α-synuclein-associated effects, which then translates into the prevention of synaptic dysfunction and cell death. While non-neuronal A_2AR have been implicated in neurotoxin-based PD mechanisms (such as MPTP or 6-OHDA) (Yu et al. 2008), forebrain neuronal A_2AR are critical for the control of cortico-striatal synaptic activity. Our data reinforce the notion that counteracting neuronal A_2AR activation has benefits, specifically against aSyn-induced synaptic deficits.

Furthermore, dysfunction in Ca²⁺ has also been shown to increase α-synuclein propensity to form aggregates (Rcom-H'cheo-Gauthier et al. 2014) which are known to be associated with PD-related neurotoxicity (Conway et al. 2000; Karpinar et al. 2009). Based on these findings, we hypothesized that the observed protective effects of A_2AR blockade could be due to the modulation of α-synuclein aggregation process, as also observed to occur for the formation of mutated ataxin-3 aggregates (Gonçalves et al. 2013). To this end, we used 2 cell-based models mimicking different steps of the α-synuclein aggregation process namely dimerization/oligomerization and inclusions formation (McLean et al. 2001, 2002; Outeiro et al. 2008). We did not detect significant differences in the oligomerization pattern using the BiFC assay, upon treatment with A_2AR modulators. Since this assay does not distinguish between dimers, trimers and higher-molecular-weight α-synuclein oligomers (Outeiro et al. 2008), we cannot discard the hypothesis that A_2AR modulators might interfere with the later stages of aggregation. Indeed, A_2AR blockade decreases the number of α-synuclein inclusions in a cell model of α-synuclein inclusion formation, whereas their activation enhances inclusion formation. It has been reported that the activation of NMDA receptor can downregulate the ubiquitin proteasome system (Caldeira et al. 2013), which may consequently lead to the accumulation of proteins that are prone to aggregation, like α-synuclein. There are reports suggesting that A_2AR can directly bind and modulate the activity of ubiquitin proteasome system (Milojevic et al. 2006; Chiang et al. 2009). The observed changes are not associated with A_2AR affecting directly the levels of aSyn, in accordance with previous observations from Kachroo et al. (2012). In that study, the authors did not assess changes in the aggregation pattern. Here, we now report for the first time that the A_2AR blockade reduces the percentage of cells containing inclusions, using a very sensitive model that allows more accurate quantifications. Our results suggest that A_2AR modulation does not interfere with the initial events leading to the formation of oligomeric species but, instead, may interfere with the latter stages of the α-synuclein aggregation process. Moreover, these data strongly suggest that the recently reported ability of caffeine (a nonselective adenosine receptor antagonist) to interfere with α-synuclein aggregation might be due to its actions on A_2AR (Kardani and Roy 2015). Whether these effects are due to the attenuation of ubiquitin proteasome system dysfunction or to downstream mediators of α-synuclein toxicity remains to be clarified.

Currently, there are multiple specific A_2AR antagonists, including caffeine, progressing through Phase II and III clinical trials for the symptomatic treatment of PD (Barkhoudarian and Schwarzschild 2011). Thus, this class of agents is well positioned for clinical testing of their neuroprotective potential. The present findings strengthen the rationale for disease modification trials of A_2AR antagonism and complement epidemiological data on caffeine links to a reduced risk of PD and substantially broaden the potential use of A_2AR as therapeutic targets in synucleinopathies.

Overall, our results highlight the interplay between toxic and protective influences of A_2AR on α-synuclein aggregation and associated synaptic toxicity and neurodegeneration, raising the possibility that adenosine A_2AR antagonists produce their well-documented neuroprotective effects in PD models by preventing α-synuclein-inclusion formation and consequent associated toxicity. Furthermore, we now show that this rescue in α-synuclein-associated toxicity is being mediated via NMDA receptor, which are known to be involved in proteasome clearance system. Moreover, we also demonstrate that both intracellular overexpression of α-synuclein and extracellular addition of oligomeric α-synuclein increase cell death, suggesting that α-synuclein may induce similar toxic effects irrespective of being generated intra- or extracellulary and, more importantly, A_2AR antagonists are able to completely rescue these α-synuclein-associated toxic events.

Authors’ Contributions

D.G.F. designed and performed most of the experimental work and wrote the manuscript. V.L.B. and J.E.C. contributed to some of the electrophysiogical experiments and data analysis. R.G. performed qPCR analysis. H.V-M. was responsible for the production of α-synuclein oligomers and SDS page profiling. F.Q. and J.R. contributed to the electrophysiogical experiments in A_2AR KO mice. J.R. designed the macro for fluorescence intensity analysis. A.A.-T. and R.A.C. contributed to manuscript revision. L.V.L. and T.F.O. designed the experiments and wrote the manuscript.

Supplementary Material

Supplementary material can be found here.

Funding

T.F.O. and L.V.L. were supported by a grant from the Fritz Thyssen Stiftung (Az. 10.12.2.165), Germany. L.V.L. is an Investigator FCT, funded by Fundação para a Ciência e Tecnologia, Portugal. T.F.O. is supported by the DFG Center for Nanoscale Microscopy and Molecular Physiology of the Brain (CNMPB; Germany).

Notes

The authors acknowledge J. Baião for technical assistance and J.T.F. for figure layout design. Conflict of Interest: None declared.