Adenosine A1 receptor ligands bind to α-synuclein: implications for α-synuclein misfolding and α-synucleinopathy in Parkinson’s disease

Elisabet Jakova; Mohamed Taha Moutaoufik; Jeremy S. Lee; Mohan Babu; Francisco S. Cayabyab

doi:10.1186/s40035-022-00284-3

Adenosine A1 receptor ligands bind to α-synuclein: implications for α-synuclein misfolding and α-synucleinopathy in Parkinson’s disease

Jakova, Elisabet; Moutaoufik, Mohamed Taha; Lee, Jeremy S.; Babu, Mohan; Cayabyab, Francisco S. 2022-02-10 00:00:00 Background: Accumulating α-synuclein (α-syn) aggregates in neurons and glial cells are the staples of many synu- cleinopathy disorders, such as Parkinson’s disease (PD). Since brain adenosine becomes greatly elevated in ageing brains and chronic adenosine A1 receptor (A1R) stimulation leads to neurodegeneration, we determined whether adenosine or A1R receptor ligands mimic the action of known compounds that promote α-syn aggregation (e.g., the amphetamine analogue 2-aminoindan) or inhibit α-syn aggregation (e.g., Rasagiline metabolite 1-aminoindan). In the present study, we determined whether adenosine, A1R receptor agonist N -Cyclopentyladenosine (CPA) and antago- nist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX) could directly interact with α-syn to modulate α-syn aggregation and neurodegeneration of dopaminergic neurons in the substantia nigra (SN). Methods: Nanopore analysis and molecular docking were used to test the binding properties of CPA and DPCPX with α-syn in vitro. Sprague–Dawley rats were administered with 7-day intraperitoneal injections of the A1R ligands and 1- and 2-aminoindan, and levels of α-syn aggregation and neurodegeneration were examined in the SN pars compacta and hippocampal regions using confocal imaging and Western blotting. Results: Using nanopore analysis, we showed that the A1R agonists (CPA and adenosine) interacted with the N-ter- minus of α-syn, similar to 2-aminoindan, which is expected to promote a “knot” conformation and α-syn misfolding. In contrast, the A1R antagonist DPCPX interacted with the N- and C-termini of α-syn, similar to 1-aminoindan, which is expected to promote a “loop” conformation that prevents α-syn misfolding. Molecular docking studies revealed that adenosine, CPA and 2-aminoindan interacted with the hydrophobic core of α-syn N-terminus, whereas DPCPX and 1-aminoindan showed direct binding to the N- and C-terminal hydrophobic pockets. Confocal imaging and Western blot analyses revealed that chronic treatments with CPA alone or in combination with 2-aminoindan increased α-syn expression/aggregation and neurodegeneration in both SN pars compacta and hippocampus. In contrast, DPCPX and 1-aminoindan attenuated the CPA-induced α-syn expression/aggregation and neurodegeneration in SN and hippocampus. *Correspondence: frank.cayabyab@usask.ca Department of Surgery, College of Medicine, University of Saskatchewan, Saskatoon, SK, Canada Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 2 of 26 Conclusions: The results indicate that A1R agonists and drugs promoting a “knot” conformation of α-syn can cause α-synucleinopathy and increase neuronal degeneration, whereas A1R antagonists and drugs promoting a “loop” con- formation of α-syn can be harnessed for possible neuroprotective therapies to decrease α-synucleinopathy in PD. Keywords: Alpha-synucleinopathy, Adenosine A1 receptor, N -cyclopentyladenosine, 8-cyclopentyl-1,3- dipropylxanthine, 1-aminoindan, 2-aminoindan, Neuroprotection, Neurodegeneration, Protein misfolding Background subthalamonigral pathway may be clinically relevant in Adenosine is a nucleoside that is involved in many physi- improving tardive dyskinesia in PD patients by reducing ological activities including cell proliferation, migra- the glutamatergic outputs of the SN pars reticulata dopa- tion of dendritic cells, and the release of small proteins minergic neurons. Moreover, most compounds that tar- called cytokines which are vital for cell signalling from get A1R activation were believed to be neuroprotective in periphery to secondary lymphoid organs, vascular reac- both the SN and hippocampus. For example, activation tivity, apoptosis and most importantly, the passage of of A1R is involved in paeoniflorin (a chemical compound neuronal stem cells [1–5]. Adenosine is also implicated derived from Paenoia lactiflora)-induced neuroprotec - in central nervous system (CNS) disorders such as tion in cerebral ischemia in Sprague–Dawley rats [27]. ischemia, trauma, epilepsy, neuropsychiatric disorders However, we recently reported that chronic stimulation and cancer [6–11]. Moreover, various roles of adenosine of A1Rs by intraperitoneal (i.p.) injection of the A1R have garnered intense investigations in many ageing- agonist CPA in rats for 3 days is sufficient to induce neu - related neurodegenerative diseases such as ischemic rodegeneration in the hippocampus [24]. Additionally, stroke, Alzheimer’s disease (AD) and Parkinson’s disease longer-term chronic A1R stimulation with CPA increases (PD) [12–15]. PD is the second most prevalent ageing- sortilin expression that promotes α-syn upregulation in related neurodegenerative disease after AD [16]. The dopaminergic MN9D cells and SN dopaminergic neu- pathophysiology of PD directly involves the imbalance rons of Sprague–Dawley rats [25]. Since highly upregu- of dopaminergic signalling pathways and accumulation lated α-syn can be found in the hippocampus and SN of of protein aggregates of α-synuclein (α-syn) in inclu- rodent synucleinopathy models [25, 28–30], we therefore sions (Lewy bodies) causing the characteristic motor tested the possibility that the commonly used A1R-selec- and cognitive deficits commonly observed in PD patients tive agonist ligand CPA can bind to α-syn and enhance [17–19]. Recently, some neuroprotective drugs have neurotoxicity, whereas the A1R-specific antagonist ligand been found to bind to α-syn and prevent further aggre- DPCPX can bind to α-syn and promote neuroprotec- gation, including caffeine, nicotine, 1-aminoindan and tion. By 7-day chronic injections in Sprague–Dawley metformin [20]. Additionally, there are other drugs such male rats, we determined if chronic stimulation with as methamphetamine, cocaine, 2-aminoindan and the CPA causes dopaminergic neuron loss and increases herbicides, paraquat and rotenone, which appear to be expression of α-syn in the SN. We then co-administered neurotoxic because they increase α-syn misfolding and DPCPX as a method to control neurodegeneration and can be correlated with a higher incidence of PD [20–23]. decrease aggregation of α-syn caused initially by CPA. Chronic adenosine A1 receptor (A1R) stimulation has Fluoro-Jade C (FJC) and Thioflavin S (Thio-S) staining recently been reported to cause hippocampal and sub- of hippocampal and SN brain regions were performed to stantia nigra (SN) neuronal death, as well as increasing assess neurodegeneration and α-syn aggregation, respec- α-syn accumulation in dopaminergic SN neurons [24, tively [31, 32]. 25]. Since a primary therapeutic goal of management Nanopore analysis and molecular docking are use- of PD is to minimize α-syn misfolding and aggregation, ful analytical tools for studying intrinsically disordered we investigated whether adenosine and the A1R agonist proteins like cellular prion proteins, β-amyloid as well N -cyclopentyladenosine (CPA), as well as antagonist as α-syn and α-syn/drug complexes [20, 21, 33–39]. 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), can bind Nanopores are single-molecule counters consisting of to and modulate α-syn misfolding. a nanometre aperture that allows the fluxes of ions and A1Rs are expressed at high levels in the limbic sys- small charged polypeptides through an insulating mem- tem especially in the hippocampus as well as in the SN brane. Applying a voltage across this membrane results region. A1R stimulation with CPA reduces glutamate in an electrochemical gradient that drives ions through and gamma-aminobutyric acid release from nerve termi- the α-hemolysin (α-HML) toxin derived from Staphylo- nals of the SN pars reticulata region of the rat brain [26]. coccus aureus [33]. A single α-syn protein interacts with This presynaptic inhibition of glutamate release from the the α-HML pore, causing a blockade current (I) for an Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 3 of 26 Fig. 1 Nanopore analysis setup and α-synuclein (α-syn) interaction with the α-Hemolysin pore. a The patch-clamp setup at 100 mV direct current (DC) allows the ions to flow in the pore and create an ionic current b The interruption of the current when α-syn interacts with the pore forming three distinguishable blockade current events: b1 Translocation events, where α-syn goes through the pore causing a large current blockade (as seen in Fig. 1 c); b2 Intercalation events, where α-syn is trapped in the pore entrance, but will diffuse back after a period of time causing an intermediate current blockade; b3 Bumping events, where α-syn approaches the pore, but diffuses away without entering causing a small current blockade. c Disruption of the blockade current and time caused by α-syn when the protein translocates the pore. d Full sequence of α-syn. e The domains of α-syn used in the nanopore setup consisting of: N-terminus (blue); ΔNAC, the entire sequence of α-synuclein without the non-amyloid β-component region (blue and red); and C-terminus (red) amount of time (T) (Fig. 1) [39]. When α-syn translocates of single molecules. Moreover, this technique has been through the pore, a large current blockade is observed for widely used to determine if a drug binds to α-syn. When a long translocation time. Conversely, if the α-syn pro- a protein-drug complex is formed, there is an increase of tein approaches the pore but then diffuses away without bumping events and a decrease in translocation events, entering, a small current blockade is observed for a short which indicates that the drug causes folding of the time. This type of event is called bumping [40, 41]. The protein. most important advantage of nanopore analysis is that Additionally, molecular docking is a computer simula- molecules can be detected without labelling and at very tion technique that allows prediction of the binding con- low concentrations. Uniquely, this technique requires formation of a desired protein or peptide to a chemical less than an hour to non-destructively analyze thousands compound or other small molecules, making molecular Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 4 of 26 docking analysis one of the best techniques for structure- 22 ± 1 °C with an applied potential of 100 mV at a band- based drug design [42, 43]. Therefore, using these com - width of 10 kHz using an Axopatch 200B amplifier (Axon plimentary biophysical and computational techniques in Instruments, San Jose, CA) under voltage-clamp condi- combination with our in vivo studies, we aimed to elu- tions using a Clampfit software (Molecular Devices, San cidate the effects of adenosine and other A1R ligands on Jose, CA). As discussed elsewhere, temperature changes α-syn conformations and dopaminergic neuron loss in due to Joule heating are expected to be negligible [46]. PD. Data analysis Methods The blockade amplitudes and duration times obtained Animal protocol with Clampfit were transferred to Origin 7 graph - Animals were housed and treated humanely in accord- ing software (OriginLab Corporation, Northampton, ance with the guidelines from the following governing MA) and were used to construct blockade current and bodies: National Research Council (US) Committee for time histograms. The blockade amplitudes were plot - the Update of the Guide for the Care and Use of Labora- ted as statistical histograms and each event population tory Animals (Washington DC, 2011); Canadian Coun- (e.g., translocation, intercalation and bumping) was fitted cil on Animal Care (CCAC); and the University of with a Gaussian function to obtain the peak/population Saskatchewan Animal Research Ethics Board (AREB) blockade current value (I). The duration time data for that approved our Animal Use Protocol (#20070090). each population were plotted separately and the data fit - Male Sprague–Dawley rats (20–30 days old varying from ted with a single exponential decay function to obtain the 250 to 300 g) were used for immunofluorescence confo - characteristic time (T). Each experiment was repeated cal imaging and biochemical studies as described below. at least three times and the event profiles were added The animals were housed in cages of two, with free access together. The error in the peak current was estimated to to food pellets and water. be < ± 1 pA and the proportion (%) of the events in each peak is reported in Tables as means ± SEM. Reagents Structural modeling and docking α-Syn (rPeptide, Bogart, GA) was dissolved in nucle- Distinct structural subpopulations of α-syn monomers ase-free water at a final concentration of 1 μM. Adeno - (C1–C8, see Additional file 1: Appendix Fig. 1) have sine was purchased from Millipore-Sigma (Oakville, recently been identified [47] and were subsequently used Canada), CPA was purchased from Abcam (Toronto, in the present study (with the exception of C6, which Canada), DPCPX from Tocris (Burlington, Canada), 1- is believed to be membrane-bound) in our molecular and 2-aminoindan were purchased from Sigma-Aldrich docking simulations to predict the drug-protein com- (Oakville, Canada). For the nanopore analysis, all drugs plexes. The respective conformation of each of the α-syn were dissolved in methanol (MeOH) and used at a final structures was taken from PDB-DEV (Entry: PDB- concentration of 10 μM. For the 7-day chronic i.p. injec- DEV_00000082) [48]. The chemical structures of adeno - tion, CPA, DPCPX, 1- and 2-aminoindan were dissolved sine, CPA, DPCPX, 1-aminoindan and 2-aminoindan in 0.1% dimethyl sulfoxide (DMSO) (Sigma-Aldrich, were obtained from PubChem (CIDs: 60961, 53477947, Oakville, Canada) in 0.9% sodium chloride at a final con - 1329, 123445, 76310, respectively). The molecular dock - centration of 3 mg/ml. ing study was carried out using Autodock Vina module implemented in PyRx tool (La Jolla, CA) [43]. Protein and Nanopore analysis ligand interactions were analyzed and visualized through Instrument setup Pymol (New York, NY) and LigPlot + (Cambridge, UK). The standard direct current (DC) setup has been described in detail previously by our lab [39, 44, 45]. In In vivo drug treatments to study α‑syn aggregation brief, a lipid bilayer was painted onto a 150-μm aperture and neurodegeneration in a Teflon perfusion cup. The two buffer compartments In support of the in vitro data, a full in vivo study con- on either side of the lipid bilayer each contained a 1-ml sisting of 7-day chronic i.p. injections of eight reagents total volume. Five microliters of 1 μg/ml α-HML (Milli- or their combinations in 28-day old male Sprague– pore-Sigma, Oakville, Canada) were added to the cis side Dawley rats was performed. The eight treatments con - of the membrane and the current was monitored until sisted of (1) Control (0.1% DMSO in 0.9% saline), (2) stable pore insertion was achieved. Consistent results CPA, (3) DPCPX, (4) 1-aminoindan, (5) 2-aminoin- were achieved with one to four pores. The peptides were dan, (6) CPA + DPCPX, (7) 1-aminoindan + CPA , and added to the cis side of the pore with a positive electrode (8) 2-aminoindan + CPA. Although 1-aminoindan and on the trans-side. The experiments were carried out at Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 5 of 26 Thioflavin‑S 2-aminoindan have very similar structures, they have Thio-S (Sigma-Aldrich, Oakville, Canada) is a fluores - been shown to possess very different properties in vitro cent marker that detects α-syn aggregates and amyloid [21, 36], therefore we suggest they will exhibit differ - plaques. Coronal slices of 40 μm were firstly treated with ent physiological properties in vivo as well. All drugs 0.3% KMnO for 4 min, followed by a 30-min incuba- were dissolved at 3 mg/ml in DMSO, and each drug tion with 1 M phosphate buffered saline at 4 °C. These was administered to the animals by daily i.p. injections slices were then stained with 0.05% Thio-S in 50% etha - (3 mg/kg body weight) for 7 consecutive days. After the nol in the dark for 8 min, rinsed with 80% ethanol twice, first injection with DPCPX, 1-aminoindan or 2-aminoin - followed by three rinses with ultra-pure water for 30 s. dan, the animals were returned to their cages for 30 min Finally, the slices were incubated again with 1 M phos- before a subsequent CPA injection was administered. phate buffered saline for 30 min at 4 °C before starting Then on the eighth day following the final injections, the the DAPI stain. The FITC filter (488 nm laser line) was animals were sacrificed and processed for brain immuno - used to image Thio-S using a Zeiss LSM700 confocal histochemistry/confocal imaging or Western blotting as microscope (Carl Zeiss Group, Canada) and images were described below. analyzed with ImageJ (Public Domain). Immunohistochemistry FluoroJ ‑ ade C Anesthetized rats were transcardially perfused with 0.9% FJC is a fluorescent marker for neurodegeneration (Mil - saline, and then fixed with 4% paraformaldehyde. The lipore-Sigma, Oakville, Canada). Coronal slices of 40 μm extracted brains were put in 30% cryoprotected sucrose were mounted on 5% gelatin-coated super-frost plus solution for 48 h prior to slicing. The brains were initially microscope slides (Thermo Fisher Scientific, Waltham, frozen at − 40 °C (BFS-30 mp controllers) and sliced with MA) and dried overnight at 4 °C. Initially, the micro- the help of a microtome (Leica SM2010 R Sliding con- scope slides were immersed in 1% NaOH/80% ethanol for troller). Coronal slices of 40 μm were then washed three 5 min followed by 2-min immersion in 70% ethanol. The times in 0.1 M phosphate buffered saline followed by slides were then rinsed for 2 min with ultra-pure water. 1-h blocking at room temperature with blocking buffer. The microscope slides were further immersed in 0.06% The buffer solution components have been previously KMnO for 10 min, followed by additional rinse for 2 min described [49]. The slices were then incubated over - with ultra-pure water. The slides were then stained with night at 4 °C with the following primary antibodies: 1:200 0.004% FJC in 0.1% acetic acid for 20 min with gentle mouse monoclonal to α-syn (Abcam Inc, Toronto, Can- shaking on an orbital shaker. Lastly, the slides were rinsed ada) and 1:200 rabbit polyclonal to tyrosine hydroxylase three times in ultra-pure water for 1 min each, mak- (TH) (Millipore-Sigma, Oakville, Canada). Subsequently, ing sure to remove all the excess water after each rinse. slices were then incubated for 1 h in the dark at room The slides were then rinsed in xylene and allowed to dry temperature with the following secondary antibodies: overnight at 4 °C. Then they were treated with Prolong AlexaFluor-555-conjugated anti-mouse and AlexaFluor- Gold Antifade Reagent from Invitrogen (Thermo Fisher 647-conjugated anti-rabbit (1:1000) purchased from Inv- Scientific, Waltham, MA), and respective images were itrogen (Thermo Fisher Scientific, Waltham, MA). Slices taken using a Zeiss LSM700 confocal microscope (Carl were then treated with Thio-S (see further details below). Zeiss Group, Canada) and analyzed with ImageJ (Public Lastly, the slices were incubated for 5 min at room tem- Domain). FJC fluorescence was obtained by exciting the perature with DAPI (2 mg/ml) from Invitrogen (Thermo dye with 488 nm laser. Fisher Scientific, Waltham, MA) and images were taken using a Zeiss LSM700 confocal microscope (Carl Zeiss Nigral slice preparation for Western blotting Group, Canada) and analyzed with ImageJ (Public Nigral slices (400 μm) from male Sprague–Dawley rats Domain). Images of the hippocampal CA1 pyramidal were prepared with the help of the vibratome tissue slicer layer and the SN pars compacta were obtained using (Leica VT1200 S). The rat was initially anaesthetized the Zeiss Plan-Apochromat 63X/1.4 oil objective lens with halothane and rapidly decapitated. Once the brain (Carl Zeiss). Images were acquired as Z-stack images of was extracted it was placed immediately in an ice-cold hippocampal or SN regions with 12–13 Z-stack images sucrose dissection medium and oxygenated with 95% taken at 1-µm intervals near the middle of brain slices. oxygen with 5% carbon dioxide. The slices were then Two Z-stack images were taken along the hippocam- equilibrated in the oxygenated artificial cerebrospinal pal CA1 or SN pars compacta region for each slice, and fluid for 1 h. Nigral slices were transferred into homog - immunofluorescence signals were averaged using densi - enization lysis buffer containing 1% NP-40 detergent tometry analysis. and supplemented with protease inhibitors. After tissue Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 6 of 26 homogenization, the protein concentration was meas- MeOH were similar to those found in α-syn alone (i.e., ured with Bradford Assay using the DC protein assay blockade current peaks around − 85 pA). Moreover, the dye (Bio-Rad, Canada). Protein lysates (50 μg/lane) from blockade populations of the translocation and bumping the different treatment groups were separated in 12% peaks in the presence of 10% MeOH were not signifi - SDS-PAGE gels, and transferred to polyvinylidene dif- cantly different from the α-syn alone control recordings luoride (PVDF) membranes (Millipore, ThermoFisher (Additional file 1: Table S1); therefore, all subsequent Scientific, USA) using 30 V overnight at 4 °C. The PVDF recordings with different drugs described below were membranes were then treated with mouse monoclonal performed with 10% MeOH in the recording solution. [4D6] anti-α-syn (Abcam Inc, Canada) primary anti- When 1 µM α-syn and 10 μM of the drug (in 10% body overnight at 4 °C after 1-h blocking with 5% non- MeOH) were inserted on the cis side, we observed fat milk in Tris-buffered saline with Tween-20. The changes in the blockade current events (See Additional next day, the membranes were incubated for 1 h with file 1: Fig. S1 for details of current of α-HML pore at the appropriate secondary antibody at room tempera- + 100 mV and the changes in blockade current events ture. The membranes were then finally re-probed with once α-syn and/or drugs such as CPA and DPCPX were chicken polyclonal antibody against Tubulin-III. Proteins added in the cis side of the perfusion cup). First, we were visualized using enhanced chemiluminescence and tested the potential binding of adenosine with α-syn in ChemiDoc (Bio-Rad, Canada). Densitometry analysis our nanopore setup. Adenosine appeared to have weak was performed using ImageJ (Public Domain). All the binding affinity to α-syn. The majority of the events of the above-mentioned solutions and procedures have been −86 pA blockade current was related to α-syn transloca- previously described [49]. tion; however, there were fewer events observed in the translocation peak in the α-syn and adenosine histogram Statistical analyses (52%) compared to α-syn alone (66%) (Fig. 2d). Interest- For nanopore, histological and Western blot analyses, ingly, CPA and DPCPX appeared to bind to α-syn as well statistical analyses were conducted with GraphPad Prism (Fig. 2e, f). For the first time, we observed a decrease 8 software (San Diego, CA) with one-way ANOVA fol- in the blockade current of the translocation peak from lowed by Student-Neuman-Keuls multiple comparison −85 pA for α-syn alone to −89 pA for the α-syn and post-hoc test. The significances are indicated as: ns, non- CPA complex, as the percentage of events decreased significant; *P < 0.05; **P < 0.01; and ***P < 0.001. from 66% to 40% for CPA (Fig. 2e). Taken together, the observed effects of CPA and, to a lesser extent, adeno - Results sine on the α-syn translocation events are clear signs of CPA and DPCPX bind to α‑syn in vitro binding. Conversely, DPCPX caused a small increase of Initially, both CPA and DPCPX were tested by nanopore the blockade current to -84 pA, which was accompanied analysis. Once a stable pore was created, a final concen - by a decrease in the number of events in the transloca- tration of 1 μM α-syn was inserted on the cis side of the tion peak (Fig. 2f ). The blockade times of translocation perfusion cup. At first, a few bumping events at 30 pA and and bumping peaks of α-syn with and without adeno- a higher number of translocation events around 85 pA sine, CPA or DPCPX were also calculated. Representa- were observed, whereas intercalation events were rarely tive exponential time graphs are shown in Fig. 3. The encountered (Fig. 1b2). Thus, these observations confirm times of translocation and bumping events for α-syn similar recordings of stable blockade and bumping cur- alone were well established [33]. The times of transloca - rents of α-syn, as previously reported [33]. Previous work tion events (α-syn alone, 0.52 ms) decreased when α-syn using confocal single-molecule fluorescence techniques was combined with adenosine (0.46 ms), CPA (0.47 ms) indicated the formation of oligomers of α-syn in the pres- or DPCPX (0.42 ms), which further indicates a potential ence of DMSO [50]. These oligomers have a high binding binding of these drugs to the protein. On the other hand, affinity to lipid membranes. Therefore, to avoid potential we observed increased bumping times when α-syn was aggregation of α-syn as well as binding of the oligom- incubated with adenosine, CPA or DPCPX (Fig. 3e–h). ers to the membrane and disruption of the lipid bilayer, A full summary of the blockade populations and times is which could lead to further issues with α-HML assembly shown in Table 1. and conductivity, we decided to use MeOH to dissolve all Caffeine, a nonselective inhibitor of all the adenosine the drugs in our nanopore studies. As shown in Fig. 2b receptors (A1, A2A, A2B and A3) [51], has previously and c, 1% and 10% final concentrations of MeOH did been shown by nanopore analysis to bind to the N- and not significantly change the blockade current histograms C-termini of α-syn, thereby promoting a neuroprotective compared to control (α-syn alone, Fig. 2a). The transloca - loop conformation [37]. It is known that caffeine compet - tion and bumping peaks in the presence of 1% and 10% itively antagonizes adenosine’s effects [52, 53]. Although Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 7 of 26 Fig. 2 Representative blockade current histograms of 1 μM α-synuclein alone (a) and with 1% methanol (b), 10% methanol (c), 10 μM adenosine (d), 10 μM CPA (e) and 10 μM DPCPX (f) at 100 mV DC, indicating binding to the protein. Each experiment was run in triplicates and the standard error of the mean estimated for the percentage of events was < ± 10% (see Table 1 and Additional file 1: Table S1) Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 8 of 26 Fig. 3 Representative blockade time profiles of translocation (a–d) and bumping events (e–h) for 1 μM α-synuclein alone and in the presence of 10 μM adenosine, 10 μM CPA or 10 μM DPCPX. Each experiment was run in triplicates. For mean and SEM values of populations of translocation and bumping and blockade times in the absence or presence of adenosine, CPA or DPCPX, please see Table 1 Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 9 of 26 Table 1 Populations and blockade times of translocations and bumping events for α-syn alone and α-syn complexes with adenosine, CPA, and DPCPX Protein‑ drug complex α‑syn α‑syn + Adenosine α‑syn + CPA α‑syn + DPCPX Population of translocation 66% 52% [*] 40% [**] 37% [**] SEM 1% 2% 1% 1% Population of bumping 24% 25% [ns] 37% [**] 46% [**] SEM 1% 3% 2% 2% Time of translocation 0.52 ms 0.46 ms [ns] 0.47 ms [ns] 0.42 ms [ns] SEM 0.05 ms 0.05 ms 0.05 ms 0.04 ms Time of bumping 0.05 ms 0.12 ms [***] 0.09 ms [**] 0.07 ms [*] SEM < 0.01 ms 0.01 ms < 0.01 ms < 0.01 ms ns, non-significant *P < 0.05, **P < 0.01, and ***P < 0.001 vs α-syn alone (one-way ANOVA, followed by Student–Newman–Keuls multiple comparison test) the blockade populations of adenosine and CPA showed or DPCPX (Fig. 4c, f, i). With the addition of CPA, the some similarities to the histogram of caffeine binding N-terminus proportion of bumping events decreased sig- to α-syn [37], both adenosine and CPA decreased the nificantly from 70% to 46% (Fig. 4a vs b). Conversely, the blockade current of the translocation events of α-syn widespread block of events between − 50 and − 100 pA to − 86 and − 89 pA, respectively (Fig. 2d, e). Previous in the N-terminus control developed into a well-defined results using 5 μM caffeine showed that the translocation broad translocation peak at − 26 pA with a population population decreased to 44% from 81% of α-syn alone, of 35%. The changes observed for ΔNAC after addition whereas the bumping population significantly increased of CPA were remarkable and demonstrated clear signs to 38% compared to 9% of α-syn alone [37]. Similarly, of binding (Fig. 4d vs e). The broad translocation peak here DPCPX + α-syn decreased the translocation popula- at − 86 pA was reduced into a small cluster of events, tion (37%) and increased the bumping population (46%) whereas the small bumping peak significantly increased (Fig. 2f ), suggesting that like caffeine, DPCPX may poten - in population from 19% to 66% and shifted to − 36 pA tially bind to the N- and C-termini of α-syn, forming a from − 27 pA. Interestingly, the C-terminal domain his- loop conformation. togram profiles in the absence and the presence of CPA did not show significant differences (Fig. 4g vs h), which Alpha‑synuclein domain investigations of CPA and DPCPX indicates that CPA does not interact with the C-terminus. To further probe the exact binding of both CPA and In contrast, DPCPX produced different profile histo - DPCPX, separate domains of α-syn, namely the N- and grams of each of the α-syn domains when compared to C-termini, and the ΔNAC construct, i.e., α-syn with dele- both the control and CPA. The N-terminus histogram tion of the non-amyloid β-component region (Fig. 1d, e), profile showed that DPCPX caused a decrease in the pro - were tested against CPA or DPCPX. The behaviour of portion of bumping events from 70% to 41%, and DPCPX each domain was different in a standard nanopore analy - also revealed two additional peaks, namely the translo- sis at a direct current voltage of 100 mV (Fig. 4a, d, g). The cation peak at − 72 pA and an intercalation peak at − 51 blockade current histogram of the N-terminus had a sin- pA (Fig. 4a vs c). The intercalation peak at − 51 pA had a gle Gaussian peak at − 30 pA due to bumping events. The low population of events (21%) whereas the translocation N-terminus is positively charged (+ 4). Consequently, it peak was broader and had a similar proportion to the will be difficult for this N-terminal fragment to translo - translocation peak of CPA. The ΔNAC translocation peak cate through the pore under the applied positive trans- disappeared with the addition of DPCPX; instead, two membrane voltage. Conversely, the C-terminus contains peaks with similar proportion of events were observed a total of 12 negative charges, which permits transloca- at − 24 and − 39 pA, representing the bumping and tion through the pore. The blockade current histogram intercalation peaks, respectively (Fig. 4d vs f ). Lastly, the had a large and wide translocation peak at − 69 pA and C-terminus histogram of DPCPX indicated a decrease of a fairly small bumping peak at − 30 pA. The ΔNAC had the bumping events from 20% to 7%, an emergence of an two peaks, a large peak at − 86 pA due to translocation intercalation peak at − 31 pA, and a significant decrease and a smaller one at − 27 pA due to bumping. of the translocation peak from 77% to 38% (Fig. 4g vs i). Figure 4 shows the blockade current histograms of For convenience, all the blockade intensities and popula- each α-syn domain in the presence of CPA (Fig. 4b, e, h) tions events are shown in Table 2. Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 10 of 26 Fig. 4 Representative blockade current histograms of 10 μM CPA and 10 μM DPCPX with N-terminus (a–c), ΔNAC (d–f) and C-terminus (g–i) of α-synuclein at 100 mV DC. Each experiment was run in triplicates and the error estimated for the percentage of events was < ± 10% (see Table 2) The changes in the shape of histograms (Figs. 2 and not infer what the structure was (Fig. 2d–f). In other 4) upon addition of a drug demonstrated drug binding. cases, there were many intermediate events that suggest Further, the increase in bumping events and decrease in the presence of many different structures (Fig. 4a, d–f ). translocation events demonstrate that the drug caused However, there may be other drug/α-syn interactions protein folding. Nanopore analysis showed that the drug that are transient or dependent on initial drug concentra- binding resulted in either a “knot” or a “loop” α-syn tions, which could alter the proportions of translocation, conformation (Fig. 5). Interestingly, we observed inter- bumping and intermediate events as previously reported mediate gaussian peaks suggestive of the presence of a for the drug Rasagiline [36]. Since α-syn is an intrinsi- particular partially folded structure though we could cally disordered protein, it is assumed to have an infinite Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 11 of 26 Table 2 Summary of the intensity and the population of current blockades for all the domains in the absence or presence of CPA or DPCPX Domain‑ drug complex I P I P I P Trans Trans Inter Inter Bump Bump N-term – – – – − 30 ± 1 pA 70% ± 2% N-term + CPA −66 ± 6 pA 35% ± 3% – – − 26 ± 1 pA 46% ± 2% [***] N-term + DPCPX −72 ± 7 pA 36% ± 10% − 51 ± 2 pA 21% ± 12% − 30 ± 2 pA 41% ± 1% [***] C-term −69 ± 2 pA 77% ± 4% – – − 30 ± 2 pA 20% ± 3% C-term + CPA −66 ± 3 pA 70% ± 8% – – − 38 ± 4 pA 23% ± 3% [ns] [ns] C-term + DPCPX −64 ± 1 pA 38% ± 2% − 31 ± 1 pA 33% ± 4% − 19 ± 1 pA 7% ± 1% [***] [ns] ΔNAC −86 ± 2 pA 58% ± 2% – – − 27 ± 1 pA 19% ± 4% ΔNAC + CPA – – − 36 ± 3 pA 66% ± 3% – – ΔNAC + DPCPX – – − 39 ± 1 pA 35% ± 5% − 24 ± 2 pA 34% ± 3% Mean ± SEM. I and P represent intensity and population of the current blockade, respectively ns, non-significant P > 0.05, *P < 0.05, **P < 0.01, and ***P < 0.001 vs N-terminus or C-terminus alone (one-way ANOVA, followed by Student–Newman–Keuls multiple comparison test) number of conformations. Therefore, it is possible that that could be toxic to neurons; C4 has low propensity for some of these α-syn structures (bound or unbound by α-helical structure like cluster C3 and, hence, is unlikely ligands) could minimally be present in our nanopore to be membrane-bound; some C5 structures interact recordings, and consequently had non-negligible contri- with membranes and might be important for synap- butions to the current histograms that were not covered tic functions, while other C5 structures form tetramers by the gaussian fitting in Fig. 2a, b, d–f and Fig. 4a, b, d–f. in vivo, which are believed to promote protection against Taken together, the altered translocation and bumping neurodegenerative disorders; C6 N-terminal residues (both blockade current peaks and populations of events) adopt an α-helical structure that targets and anchors and the appearance of intercalation events in the N-ter- α-syn to membrane of synaptic vesicles; C7 is similar to minus, ΔNAC, and C-terminus domains of α-syn indi- C1, C3 and C4, having low propensity for α-helical for- cate that DPCPX binds to both the N- and C-termini of mation (i.e., higher β-strand propensity), hence, α-syn α-syn. monomers are likely in aqueous solution; and C8 struc- ture plays a role in fibril formation and has high α-helix Molecular docking simulations reveal interactions propensity like C2, C5 and C6 [47]. of DPCPX and 1‑aminoindan with N‑ and C‑termini of α‑syn Four of the eight structures of α-syn were selected to To confirm the biophysical results from nanopore analy - study the α-syn–DPCPX drug complex: C2, C5, C7 and sis, we further characterized the α-syn–drug complexes C8 [47]. For the C2 structure of α-syn, DPCPX showed by performing molecular docking studies of the three hydrophobic interaction with the N-terminus of α-syn, A1R ligands as well as 1- and 2-aminoindan with α-syn. specifically the negatively charged glutamic acid 20 Conformational ensemble of α-syn in solution as deter- (E20) and positively charged lysine 21 (K21), and with mined by discrete molecular dynamics simulations and the very end of the C-terminus (amino acids E139 and further confirmed by far-UV circular dichroism and alanine 140 (A140)) (Fig. 6a). However, for the C5 struc- cross-linking mass spectrometry, has recently revealed ture (Fig. 6b), DPCPX was shown to form hydrophobic stable monomeric α-syn clusters of structure [47] (C1- bonds with the N-terminus (amino acids phenylalanine 4 C8, see Additional file 1: Appendix 1). We used these (F4) and E20) and hydrogen bonds with K60, and hydro- structures in the molecular docking simulations to deter- phobic bonds with the NAC region (amino acids E61 mine if any of these structures correlates with the drug/ and F94). DPCPX also formed hydrogen bonds with the α-syn complex as predicted from the nanopore analysis. NAC region (amino acids glutamine 62 (Q62) and valine The 8 clusters of structural subpopulations have the fol - 63 (V63)). Similar as the interactions with C2, DPCPX lowing features: C1 structure forms a dimer and plays binding to the N-terminus (hydrogen bond with E28, a role in fibril formation; C2 and C3 are precursors for with additional hydrophobic interactions with V15 and oligomer formation; C3 has antiparallel β-sheets in the V40) and the C-terminus (hydrogen bond with tyrosine aggregate-prone NAC segment and forms oligomers 136 (Y136) with additional hydrophobic interactions with Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 12 of 26 Fig. 5 Eec ff ts of adenosine A1 receptor (A1R) ligands on α-synuclein (α-syn) expression and folding patterns in in vivo and in vitro studies. a A1R agonist CPA (and adenosine) increases α-syn expression and aggregation in the rat substantia nigra. Nanopore analysis and molecular docking simulations predicted binding of A1R agonist CPA (and adenosine) to the N-terminus of α-syn, leaving the NAC domain intact and able to promote aggregation. b (b1) Adenosine, CPA and 2-aminoindan bind to and stabilize α-syn to adopt a “knot” conformation which has been shown to induce aggregation and neurodegeneration. In contrast (b2), DPCPX and 1-aminoindan bind to both the N- and C-termini of α-syn, which does not promote aggregation and neurodegeneration. Created using BioRender.com Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 13 of 26 Fig. 6 Molecular docking simulation of α-synuclein (α-syn) structures C2 (a), C5 (b), C7 (c), and C8 (d) bound to DPCPX. Below full 3D representations show magnified binding pocket of α-syn and the locations of amino acid residues responsible for each drug binding. Bold black dashed lines and amino acid residues indicate hydrogen bonding, while the grey dashed lines and amino acid residues indicate hydrophobic interactions. Hydrogen bonding of DPCPX with both the N- and C-terminal amino acid residues is observed in C7 α-syn structure (c). DPCPX also forms hydrogen bond with either the N-terminal (C5 α-syn structure in b) or C-terminal amino acids (C8 α-syn structure in d) and also hydrophobic bonds with portions of the NAC region. N-and C-terminal binding of DPCPX is also observed without hydrogen bonding (C2 α-syn structure in a). The molecular docking study was carried out using Autodock Vina module implemented in PyRx tool. Protein and ligand interactions were analyzed and visualized through Pymol and LigPlot + aspartic acid 135 (D135) and E137) of the C7 structure that other α-syn structures can bind DPCPX via hydro- was revealed (Fig. 6c). Furthermore, DPCPX was shown gen bonding with the N-terminus (H50 and K12 of C3 to bind to the C8 structure in the N-terminus (amino structure) or C-terminus (E137 of C1 structure) and acids K32, glycine 36 (G36), V37, threonine 44 (T44) and hydrophobic interactions with the NAC region (Addi- V48) and the C-terminus with a hydrogen bond at the tional file 1: Fig. S2). E139 and additional hydrophobic interactions at G106, Like DPCPX, four α-syn structures, C1, C4, C5 and A107 and proline 108 (P108). For all the structures C2, C8, were selected to analyze the binding mode of C5, C7 and C8, DPCPX was revealed to reside within a 1-aminoindan to α-syn. For the C1 structure, 1-ami- closed globular conformation and interacts with both the noindan was shown to form hydrogen bond with E104 N- and C-termini of α-syn. However, it is also possible of the C-terminus as well as hydrophobic interactions Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 14 of 26 Fig. 7 Molecular docking simulation of α-synuclein (α-syn) structures C1 (a), C4 (b), C5 (c) and C8 (d) bound to 1-aminoindan. Below full 3D representations show the magnified binding domains of α-syn and the amino acid residues in both the N- and C- termini of α-syn that facilitate drug binding. Bold black dashed lines and amino acid residues indicate hydrogen bonding, whereas the grey dashed lines indicate hydrophobic interactions. Hydrogen bonding of 1-aminoindan with C-terminal amino acid residues is observed in C1, C4, C5 and C8 α-syn structures. In addition, hydrophobic interactions occur between 1-aminoindan and N-terminal amino acid residues (C1 and C4 α-syn structures) and also between 1-aminoindan and portions of the NAC region (C5 and C8 α-syn structures). The molecular docking study was carried out using Autodock Vina module implemented in PyRx tool. Protein and ligand interactions were analyzed and visualized through Pymol and LigPlot + with the aromatic positively charged cleft of the However, the C5 and C8 structures were shown to N-terminus (amino acids G31, K32, V37, and Y39) form different drug-protein complexes that may reg- (Fig. 7a). For the C4 structure (Fig. 7b), 1-aminoin- ulate tetramer formation (C5) or fibrillation (C8). dan was shown to form hydrogen bonds with the polar 1-Aminoindan appeared to bind only to the C-termi- serine 129 (S129) and hydrophobic interactions with nal polar cleft between T92, Q99 and Q134 of the C5 E131 and Y136 of the C-terminus, and interact with structure (Fig. 7c). Also, 1-aminoindan was shown to the N-terminus of C4 structure, forming hydrophobic form further hydrophobic bonds with G101 and P128 bonds with A18. These indicate the formation of a loop and a hydrogen bond with L100. Lastly, 1-aminoindan conformation between the N- and C-termini of the had hydrophobic interactions with the NAC region protein when 1-aminoindan binds to the C4 structure. of the C8 structure (amino acids V71, T72, A76, and Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 15 of 26 Fig. 8 Molecular docking simulation of α-synuclein (α-syn) structures C4 (a) and C5 (b) bound to adenosine. Below full 3D representations show the magnified binding pocket of α-syn and the amino acid residue locations responsible for each drug binding. Bold black dashed lines and amino acid residues indicate hydrogen bonding, whereas the grey dashed lines and amino acid residues indicate hydrophobic interactions. Adenosine only formed hydrogen bonds and hydrophobic interactions with N-terminal amino acid residues in C5 α-syn structure. In addition, adenosine also formed hydrogen bonds with amino acid residues within the N-terminus and NAC region in C4 α-syn structure. The molecular docking study was carried out using Autodock Vina module implemented in PyRx tool. Protein and ligand interactions were analyzed and visualized through Pymol and LigPlot + V77), and in the C-terminus (amino acids methionine mainly bound to the N-terminus of α-syn (blue alpha- 116 (M116), P117 and V118) (Fig. 7d). 1-Aminoindan helix region) (Fig. 8). For the C4 structure, adenosine also formed a hydrogen bond with D119 in the C-ter- formed hydrogen bonds with V15 and K21 in the posi- minus of the structure. tively charged cleft in the C4 N-terminus and with G68, These results indicated that the A1R antagonist A78 and Q79 in the C4 NAC region. Other hydrophobic DPCPX and Rasagiline metabolite 1-aminoindan pos- interactions were revealed at G14 and A17 in the N-ter- sess similar binding interactions with α-syn. As the minus and at T72, G73 and V74 in the NAC region. For full crystal structure of α-syn is not yet available, we the C5 structure, adenosine bound only to the N-termi- suggest that using the four conformations of α-syn in nus, forming hydrogen bonds with the polar negatively our molecular docking studies could provide more charged cleft of the N-terminus (amino acids E20, Q24, complete information on the binding interactions of and A53) and hydrophobic interactions with G25 and DPCPX and 1-aminoindan with α-syn. K60. CPA is a chemical derivative of adenosine that shows Molecular docking simulations reveal that CPA, adenosine greater selectivity as an A1R agonist, and thus is expected and 2‑aminoindan bind to the N‑terminus of α‑syn to interact with α-syn similarly as adenosine. Similar to Based on the results of nanopore analysis, molecular adenosine, CPA formed hydrogen bonds with G47 of the docking simulation of α-syn binding to adenosine, A1R N-terminus and G68 of the NAC region of the C2 α-syn agonist CPA, and methamphetamine analog 2-ami- structure (Fig. 9a). CPA also had various hydrophobic noindan was conducted. Results showed that adenosine interactions in the N-terminus (amino acids V26, G31, (See figure on next page.) Fig. 9 Molecular docking simulation of α-synuclein (α-syn) structures C2 (a), C5 (b), and C8 (c) bound to CPA; C2 (d), C5 (e), and C8 (f) bound to 2-aminoindan. Below full 3D representations show the magnified binding pocket of α-syn and the amino acid residue locations responsible for each drug binding. Bold black dashed lines and amino acid residues indicate hydrogen bonding, whereas the grey dashed lines and amino acid residues indicate hydrophobic interactions. Both CPA and 2-aminoindan formed hydrogen bonds and hydrophobic interactions with the N-terminal amino acids only (C5 and C8 α-syn structures) (b-c and e–f, respectively). CPA also forms hydrogen bond and hydrophobic interactions with amino acids within the N-terminal and the NAC region (C2 α-syn structure) (a). In contrast, 2-aminoindan only forms hydrophobic interactions with the N-terminus and NAC domain in C2 α-syn structure (d). The molecular docking study was carried out using Autodock Vina module implemented in PyRx tool. Protein and ligand interactions were analyzed and visualized through Pymol and LigPlot + Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 16 of 26 Fig. 9 (See legend on previous page.) Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 17 of 26 E35, K43, V48, and K58) and in the NAC region (amino K34 (inosine, guanosine, thymidine, and uridine) and acids Q62, V63, G67, and V71). For the C5 structure, CPA with the negatively charged E20 (guanosine, cytidine, formed a hydrogen bond with the negatively charged E20, and uridine) (Additional file 1: Fig. S3). Inosine had simi- as well as the positively charged K60, Q24 and A53 in the lar hydrophobic interactions with Q24, G25 and A53 as N-terminus (Fig. 9b). It also formed hydrophobic interac- adenosine, and formed additional hydrophobic bonds tions with G25, K34 and V49. For the C8 structure, CPA with C5 α-syn conformation at H50 and E57. Inosine interacted with similar negatively charged N-terminal formed hydrogen bonds with V49 and T54 in the N-ter- cleft containing E28 and A29 (Fig. 9c), and also formed minus as well. Guanosine formed hydrogen bonds with various hydrophobic bonds in the N-terminus (amino the polar cleft of Q24 and A53 of the N-terminus same acids L8, K10, E35, L38, Y39, and V40). as adenosine (Additional file 1: Fig. S3, Fig. 8b). Con- Similar to CPA, 2-aminoindan was also shown to have versely, cytidine (Additional file 1: Fig. S3c) interacted various hydrophobic interactions with the N-terminus with the same amino acids, but through hydrophobic (amino acids V26, E35, K43, G47, and V48) and the NAC interactions (amino acids A19, Q24, G25, A53 and E57). region (amino acids V63, and V71) of the C2 structure Moreover, thymidine and uridine shared the same hydro- of α-syn (Fig. 9d). Moreover, 2-aminoindan also formed phobic interactions with the C5 α-syn structure (Addi- a hydrogen bond with E57 and hydrophobic interac- tional file 1: Fig. S3d, c), and both bound to G25, A53 and tions with other residues within the N-terminus of the E57. Uridine also formed other hydrophobic bonds with C5 structure (amino acids G25, V26, A30, G31, K34, V49, A19 and Q24, like cytidine (Additional file 1: Fig. S3c, e). and A53) (Fig. 9e). Similar to CPA, 2-aminoindan formed hydrogen bonds with A29 and E35 and hydrophobic A1R agonist CPA and drugs that bind to α‑syn N‑terminus interactions with other residues inside the N-terminus of increase α‑syn expression and aggregation in SN the C8 structure (amino acids L8, Serine 9 (S9), K10, and and hippocampal neurons T22) (Fig. 9f ). To investigate whether drug binding to the N- and/or Taken together, the molecular docking studies con- C-terminus of α-syn can affect the levels of α-syn expres - firmed the results of nanopore analyses, that adeno - sion and aggregation in vivo, we administered the drugs sine, CPA and 2-aminoindan mainly interacted with the individually or in combination with the A1R agonist CPA. N-terminus of α-syn (C5 cluster); however, other sub- Representative images of the SN pars compacta region populations of α-syn clusters showed hydrogen bonding labelled with DAPI, TH, and α-syn showed that α-syn of CPA and 2-aminoindan with other N-terminal resi- was localized in the soma (cytosol, nuclei) and presum- dues only (C8) or hydrogen bonding of adenosine with ably the dendrites of dopaminergic neurons (Fig. 10b). the N-terminal and proximal NAC amino acid residues Interestingly, α-syn expression was increased by CPA (C4). The interactions of the drugs with α-syn N-termi - in the absence or presence of DPCPX, 1-aminoindan, nus are expected to promote α-syn aggregation. In con- or 2-aminoindan (Fig. 10b, and 11a, b) compared to the trast, DPCPX and 1-aminoindan showed binding to both control (0.1% DMSO in 0.9% saline). The 2-aminoin - N- and C-terminal regions of α-syn, which may promote dan + CPA treatment induced the highest level of α-syn an α-syn conformation that prevents α-syn aggregation. protein in the SN pars compacta (Fig. 11a, b). In contrast, DPCPX, 1-aminoindan or 2-aminoindan alone did not Comparison of α‑syn binding of adenosine with that of significantly increase the α-syn protein level (Fig. 11a, b). other standard nucleosides using molecular docking To determine whether these changes in α-syn protein simulation level correlated with the level of α-syn aggregation, the Adenosine is not only an endogenous agonist of all the SN pars compacta was co-labelled with α-syn marker purinergic G protein-coupled receptors, but also a purine and Thio-S. As shown in Fig. 11a-c, treatments with ribonucleoside [55]. To determine whether adenosine CPA, DPCPX + CPA , 1-aminoindan + CPA, and 2-ami- binds specifically to α-syn, we performed additional noindan + CPA all increased the Thio-S level. Treat - molecular docking simulations to compare α-syn binding ment with 2-aminoindan alone also enhanced Thio-S of adenosine with that of the four other standard nucleo- labelling, and co-administration of 2-aminoindan with sides (guanosine, cytidine, thymidine, and uridine) and CPA caused a significant further elevation of Thio-S the adenosine metabolite inosine, using the C5 α-syn compared to 2-aminoindan alone (Fig. 11b). In contrast, structure (Additional file 1: Fig. S3). treatments with DPCPX or 1-aminoindan alone did not All the purine nucleosides were shown to bind to a significantly increase Thio-S labelling or attenuate the similar hydrophobic pocket in the N-terminus of the C5 CPA-induced increase in Thio-S level. The colocalization α-syn conformation. Similar to adenosine, almost all of of Thio-S signal with α-syn was increased in all the treat - them formed hydrogen bonds with the positively charged ments compared to control (about a two-fold increase Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 18 of 26 in Pearson correlation coefficients, Fig. 11c). In conclu- 2-aminoindan or 1-aminoindan treatment alone group, sion, our results showed that 7-day systemic administra- compared to CPA, CPA + DPCPX, or control treatment tion of CPA, alone or in combination with 2-aminoindan, (P < 0.001). increased expression and aggregation of α-syn in the SN Moreover, when the 15 kDa, 30 kDa and 75 kDa α-syn pars compacta (Figs. 10 and 11). densitometry values were added, we still detected signifi - A1R is widely distributed in other regions of the brain cantly higher levels of total α-syn in the CPA (P < 0.01), including the hippocampus. Therefore, the CA1 region of DPCPX + CPA (P < 0.033), and 2-aminoindan + CPA the hippocampus was also analyzed for α-syn expression (P < 0.001) groups (Fig. 10c). It is noteworthy that signifi - and aggregation. Similar to the nigral tissue, treatments cant accumulation of α-syn in the SN lysate was associ- with CPA and 2-aminoindan + CPA increased the levels ated with treatments with compounds that were found to of both α-syn and Thio-S (Additional file 1: Fig. S4a). In bind only to the N-terminus of α-syn (i.e., the A1R ago- particular, CPA and 2-aminoindan + CPA induced a four- nist CPA and 2-aminoindan); moreover, this elevation fold increase of α-syn compared to the control (Addi- could be attenuated by co-treatments with compounds tional file 1: Fig. S4b). This increase was also observed for that were found to bind to both the N- and C-termini colocalization of Thio-S with α-syn (Additional file 1: Fig. of α-syn (e.g., DPCPX, 1-aminoindan). In addition, the S4c). However, co-administration of CPA with DPCPX observed higher molecular weight band at 75 kDa likely or 1-aminoindan caused significant attenuation of α-syn indicates the presence of C5 α-syn structures that differ - accumulation compared to CPA treatment (Additional entially bind to DPCPX (Fig. 6b), 1-aminoindan (Fig. 7c), file 1: Fig. S4b, c). and 2-aminoindan (Fig. 9e), since the C5 structure of α-syn is known to be a precursor for the formation of Western blotting analysis of α‑syn in SN tetramers [47]. Treatment with the A1R agonist CPA and 2-aminoin- dan + CPA induced a significant 1.5-fold increase in the CPA and 2‑aminoindan increase neurodegeneration of SN level of monomeric α-syn (15 kDa band) in SN; however, pars compacta dopaminergic neurons and hippocampal the 1-aminoindan treatment group showed an approxi- pyramidal neurons mately three-fold decrease in α-syn compared to the Having shown that CPA and 2-aminoindan alone or in DMSO/saline control group (Fig. 10c). The CPA-induced combination can increase α-syn aggregation, we then increase in α-syn was partially attenuated by co-adminis- determined whether these treatments could lead to neu- tration with DPCPX and was fully restored to control lev- ronal damage. We used FJC as a common fluorescent els by 1-aminoindan co-treatment. Prominent signals for marker for neurodegeneration in the CNS [31]. FJC stain- 30 kDa α-syn were shown in the SN immunoblots, but ing was performed in nigral slices −5.30 to −5.60 mm not detectable in the hippocampal lysate immunoblots from the bregma as well as hippocampal slices −3.80 to (data not shown). In contrast to the monomeric 15 kDa −4.16 mm from the bregma. Representative high-mag- α-syn band, the 30 kDa α-syn signal was not significantly nification images of FJC staining in the pars compacta altered by CPA, DPCPX + CPA , 2-aminoindan + CPA , or region of SN indicate that CPA alone, 2-aminoindan 2-aminoindan treatment, compared to control (DMSO/ alone, and 2-aminoindan + CPA co-administration all saline) or naive group; however, the 30 kDa α-syn sig- increased the level of FJC fluorescence (Fig. 12a). In nal was significantly decreased by DPCPX (P < 0.01), contrast, DPCPX or 1-aminoindan alone did not signifi - 1-aminoindan + CPA (P < 0.01), and 1-aminoindan cantly increase FJC staining, but both were effective in alone (P < 0.001) treatments. Interestingly, the level of attenuating the CPA-induced increase in neurodegenera- 75 kDa α-syn was significantly higher in the DPCPX, tion (Fig. 12a). Similar results were observed in the CA1 region of the hippocampus, except that 2-aminoindan (See figure on next page.) Fig. 10 Summary of the surface area analysis of the pars compacta region of the substantia nigra for DAPI, tyrosine hydroxylase ( TH), and α-synuclein (α-syn). (a) Image of a 40-μm nigral brain slice in the DMSO/Saline control group, with 3,3’-diaminobenzidine (DAB) and TH staining at 4 × magnification with a light microscope. (b) Representative images of DAPI (Blue), TH (Green, Alexa Fluor 555), and α-syn (Red, Alexa Fluor 647) staining in the substantia nigra pars compacta of rats with 7-day chronic intraperitoneal injections of the following agents: Control (DMSO/ Saline), CPA, DPCPX + CPA, 1-aminoindan + CPA, and 2-aminoindan + CPA. CPA with or without 2-aminoindan increased α-syn immunofluorescence compared to control. The CPA-induced increase in α-syn was attenuated by DPCPX or 1-aminoindan. Scale bar, 20 μm. (c) Western blots from total lysates of the substantia nigra and quantification of α-syn level in the substantia nigra. CPA increased the level of 15 kDa α-syn monomers, which was attenuated by DPCPX and 1-aminoindan but not by 2-aminoindan. DPCPX and 1-aminoindan alone significantly reduced the level of 30 kDa α-syn dimers. In contrast, DPCPX, 2-aminoindan, and 1-aminoindan alone significantly increased the 75 kDa α-syn, which likely represent the α-syn tetramers. All values were normalized to β-tubulin III. n = 4 animals in each treatment group. Mean ± SEM. ns, non-significant; *P < 0.05; **P < 0.01; and ***P < 0.001 (one-way ANOVA followed by Student–Newman–Keuls post-hoc multiple comparison test) Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 19 of 26 Fig. 10 (See legend on previous page.) Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 20 of 26 Fig. 11 Summary of surface area analysis of α-synuclein (α-syn) and Thioflavin S in the substantia nigra (SN) pars compacta region. (a) Confocal microscopic images of DAPI, α-syn and Thioflavin S staining in 40-μm nigral brain slices of rats with the following treatments: Control (DMSO/ Saline), CPA, DPCPX, 1-aminoindan, 2-aminoindan, DPCPX + CPA, 1-aminoindan + CPA, and 2-aminoindan + CPA. Scale bar, 50 μm. (b) The mean area intensities of α-syn and Thioflavin S in the SN pars compacta. The fluorescence intensity was quantified in a 100 × 100 μm region and normalized by subtracting the fluorescence intensity in a 50 × 50 μm background non-cell body bottom area. CPA increased the levels of α-syn and aggregated α-syn, and these levels were further enhanced by co-treatments with 2-aminoindan. (c) Pearson correlation coefficient of α-syn and Thioflavin S in the SN pars compacta with CPA. Average intensity values and correlation coefficients in bars represent mean ± SEM from n = 4 independent experiments. ns, non-significant; *P < 0.05; **P < 0.01; and ***P < 0.001 (one-way ANOVA followed by Student–Newman–Keuls post-hoc multiple comparison test) Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 21 of 26 Fig. 12 Fluoro-Jade C (FJC) staining in the SN pars compacta (a) and CA1 of hippocampus (b) of rats with 7-day chronic intraperitoneal injection of Control (DMSO/saline), CPA, DPCPX, 1-aminoindan, 2-aminoindan, DPCPX + CPA, 1-aminoindan + CPA, and 2-aminoindan + CPA. Scale bar 50 μm. Summary bar graphs show significant increases in the relative fluorescence intensity of FJC staining in pars compacta after CPA, 2-aminoindan, and CPA + 2-aminonindan treatments (a). In contrast, only CPA and CPA + 2-aminoindan treatments significantly increased FJC fluorescence in the CA1 hippocampal neurons (b). FJC fluorescence intensity in a 100 × 100 μm region was normalized to the control group (100%). Values are shown as mean ± SEM. The average FJC fluorescence values were obtained from n = 4 independent experiments. ns, non-significant; *P < 0.05; **P < 0.01; and ***P < 0.001 (one-way ANOVA followed by Student–Newman–Keuls post-hoc multiple comparison test) alone did not cause significant neurodegeneration 1-aminoindan with CPA prevented neurodegeneration in (Fig. 12b). CPA and 2-aminoindan + CPA treatments hippocampal slices. demonstrated much higher levels of degenerating pyram- Taken together with the above results from nanopore idal neurons compared to the other treatments. Similar analysis, molecular docking and Thio-S labelling, these to the nigral slices, the co-administration of DPCPX or results suggest that compounds that bind to both N- and C-termini of α-syn (e.g., DPCPX and 1-aminoindan) Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 22 of 26 may be effective in attenuating the neurotoxic effects and thus could contribute to α-syn protein misfolding of compounds that bind to α-syn and promote α-syn and the development of α-synucleinopathy in PD. We accumulation and misfolding (e.g., CPA, adenosine and observed that CPA interacted not only with the α-syn 2-aminoindan). N-terminus but also with α-syn lacking the NAC region (ΔNAC). However, the A1R antagonist DPCPX sig- nificantly altered the population histograms from the Discussion nanopore analysis when the N-terminus, C-terminus One of the most commonly used treatments for PD or the ΔNAC regions were studied, indicating DPCPX patients is Levodopa (a dopamine precursor). However, binding to all these α-syn polypeptide domains. Based this treatment is associated with numerous adverse on the nanopore analysis, we propose that the two effects such as early loss of voluntary movement, severe A1R ligands CPA and DPCPX have different binding dyskinesia episodes and most predominantly end-of-dose interactions with α-syn. Therefore, as we previously worsening [56, 57]. Previous studies on dopamine recep- suggested for 2-aminoindan [20, 21], CPA could bind tor agonists, catechol-O-methyltransferase antagonists, to the N-terminus of α-syn similar to 2-aminoindan, monoamine oxidase inhibitors, and antagonists of dopa- thereby leaving the NAC domain free to misfold and mine transporters have shown promising results in slow- cause a higher chance of protein aggregation. In con- ing the progression and alleviating the symptoms of the trast, DPCPX, similar to 1-aminoindan, metformin and disease [58–61]. However, these therapies are unable to caffeine, appeared to bind to both the N- and C-termini prevent PD progression without causing other significant of α-syn, creating a “loop” conformation of the protein side effects including an increased risk of cardiac-valve (Fig. 5b2) [20, 21]. This loop conformation has been regurgitation, hypertension, confusion and hallucinations suggested to prevent the NAC domain from misfolding [62]. and hence promote neuroprotection by stopping aggre- Adenosine binds to its inhibitory A1Rs (coupled to gation of misfolded α-syn protein fibrils and formation G ) and its excitatory A2A receptors (A2AR, coupled αi of Lewy bodies. Therefore, the two possible conforma - to G ), and A2ARs are believed to contribute to the αs tions adopted by α-syn in the presence of CPA (knot pathogenesis of PD, which has prompted the develop- conformation) or DPCPX (loop conformation) (Fig. 5) ment of small molecular agents for potential PD ther- are expected to contribute to increased or decreased apy, including apedenoson, preladenant, regadenoson α-syn aggregation, respectively. Additionally, the block- and SYN-115 [24, 63]. Recently, several studies have ade currents by CPA and DPCPX resembled those by reported that A2AR antagonism produces far better 2-aminoindan and 1-aminoindan, respectively. The results in slowing down the pathology and progres- metabolite of Rasagiline, 1-aminoindan, is an irrevers- sion of PD and improving symptom management [24, ible inhibitor of the monoamine oxidase type B enzyme 64–68]. Unfortunately, the majority of these drugs which is administered in mono- and/or poly-thera- have failed in clinical trials except for istradefylline, peutic route to treat early symptoms of PD as well as which has been pursued as a potential PD drug in cognitive impairments and fatigue [75, 76]. Conversely, Phase III clinical trials in Japan [69, 70] but recently 2-aminoindan, an amphetamine analogue, is shown to been approved by US Food and Drug Administration cause PD-like symptoms in long-term users and addicts as the only non-dopaminergic add-on therapy for the [77]. Recent case reports indicated that amphetamine treatment of so-called “off phenomenon” and motor and methamphetamine users have a three-fold risk of fluctuations of Levodopa therapy in PD [71]. We have developing PD compared with non-users. Interestingly, suggested that a possible cross-talk between A1Rs and this risk is particularly high in women and it manifests A2ARs could contribute to PD and other neurodegen- even at 30 years of age [78]. Although 1-aminoindan erative diseases due to the elevated levels of adenosine and 2-aminoindan have very similar structures, they in the ageing brain, which may increase A2AR activa- have been shown to hold very different blockade cur - tion [24, 72–74]. More recently, we have reported that rent properties [20, 21]. chronic stimulation of A1Rs with the A1R agonist CPA The blockade histograms for 1-aminoindan demon - leads to increased expression of α-syn both in the SN strate that most events are related to translocation, as region of Sprague–Dawley rat brain and in the dopa- observed previously [20, 21]. In the presence of 10 μM minergic MN9D cells [25]. Here, our in vitro findings 1-aminoindan, a broad translocation peak at − 73 pA and suggest that drugs that bind to A1Rs may play a major a small bumping peak at − 34 pA were observed. Nano- role in synucleinopathy, independent of their canoni- pore analysis with the use of different α-syn domains cal function as A1R agonist or antagonist. We show for suggests that 1-aminoindan binds to both the N- and the first time that these adenosine-related compounds C-termini of α-syn, and by doing so, the drug-protein could bind differentially to different regions of α-syn, Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 23 of 26 complex is expected to adopt a “loop” conformation that fide A1R ligands, also bind to α-syn to modulate its mis - promotes neuroprotection by preventing α-syn mis- folding and aggregation patterns. folding. Our results from both nanopore analysis and Previous studies using nanopore analysis and a yeast molecular docking suggest that DPCPX resembles 1-ami- model of PD reported that several compounds that bind noindan in binding pattern and drug-protein conforma- to α-syn, including 1-aminoindan and the dimer com- tion. Conversely, our previous report has also established pounds containing caffeine linked to 1-aminoindan, nico - that the 2-aminoindan histograms have two peaks with tine or metformin, can indeed prevent α-syn aggregation similar proportions of events, namely, the proportion of and promote survival in yeast [21, 36, 54]. Moreover, we translocation events at 48% and bumping events at 40% previously reported that prolonged A1R activation in [20, 21]. However, 2-aminoindan appears to bind with the brain produced by i.p. injections of CPA (3 mg/kg) higher affinity to α-syn due to the high binding constant once daily for three days led to significant neuronal loss 4 −1 5 × 10 M derived from isothermal titration calorim- in rat hippocampus in vivo [24]. More recently, we also etry (ITC), and most importantly, because it binds only to reported that longer-term A1R stimulation with 5 mg/ the N-terminus of the protein. This N-terminal binding kg CPA (i.p. injections daily for 5 weeks) led to increased of 2-aminoindan has been suggested to promote a neu- α-syn expression and accumulation in the SN neurons, rotoxic “knot” conformation that will lead to increased which was associated with motor and cognitive defi - α-syn aggregation. Here, the nanopore analysis using cits in Sprague–Dawley rats [25]. In the present study, full-length and different domains of α-syn combined with we also showed that 7-day co-administration of either molecular docking attempts using aqueous, membrane- DPCPX or 1-aminoindan with CPA prevented the CPA- unbound α-syn structures [47] showed that both adeno- induced α-syn accumulation and aggregation in both SN sine and the selective A1R agonist CPA closely resembled pars compacta and hippocampal CA1 region, and this 2-aminoindan in the binding interactions and drug-pro- coincided with significantly reduced neurodegenera - tein conformations, indicating that these A1R agonists tion of dopaminergic neurons in the SN pars compacta bind prominently to the N-terminus of α-syn. Although and pyramidal neurons in the hippocampus. However, our results indicate a broad agreement between in vitro future studies are required to further differentiate the and in silico techniques, both methods have their own precise roles of A1R ligand stimulation and α-syn–drug inherent limitations. Our nanopore analysis gives an binding to promote or attenuate α-syn aggregation and excellent prediction of the compound binding site, but it subsequent neurodegeneration using rational mutagen- does not provide precise amino acid residue(s) and posi- esis design or using knockdown studies involving A1R or tions of the drug binding. At best, our nanopore analysis α-syn genes. indicated formation of a “knot” or “loop” conformation According to our previous report [25], the neuro- (Fig. 5) when the ligand-interacting residues within the toxic effect of CPA observed in the present study was N-terminus, NAC region, and/or C-terminus of α-syn are likely mediated in part by A1R-induced downstream bound to promote protein folding. On the other hand, for activation of JNK/c-Jun and sortilin-dependent bind- molecular docking, the residues that bind to each ligand ing and accumulation of α-syn in dopaminergic neu- may be dependent on the starting model used; that is, rons. Since we found in the present study that CPA and compared to the aqueous α-syn cluster structures used DPCPX could also bind directly to α-syn, in the future, in the present study, we found distinct ligand-interacting it would be important to test in A1R knockout mice or in amino acid residues that were bound by CPA and DPCPX MN9D dopaminergic neurons with CRISPR/Cas9 gene when the micelle-bound 1XQ8 model of α-syn was used knockdown of A1Rs to determine whether chronic CPA (data not shown) [79]. Moreover, it appears that confor- administration could still induce upregulation and accu- mational selection may result in only one or several of mulation of misfolded α-syn, and also whether DPCPX the conformations used in the present study to be pop- or 1-aminoindan can prevent this CPA-induced neuro- ulated. Admittedly, the lack of site-directed mutational degeneration of dopaminergic neurons in the absence evidence that could help validate the docking results and of functional A1Rs. Since adenosine elevation is widely directly confirm the ligand interactions, is a major weak - known to occur in the ageing brain and adenosine and ness of the present study. Based on the molecular dock- CPA were shown to bind to α-syn N-terminus and cause ing results, future studies using site-directed mutagenesis aggregation, there is a possibility that the increased brain of these residues and retesting with nanopore analysis adenosine may be a risk factor for increased α-syn mis- are needed to understand the conformational changes folding observed in α-synucleinopathy in PD patients. of α-syn upon binding to CPA, DPCPX, and other com- The results also suggest that targeting α-syn with pounds. Taken together, our results show that CPA and adenosine-related compounds (e.g., DPCPX and caf- DPCPX, in addition to their canonical function as bona feine) or compounds with similar binding profiles (e.g., Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 24 of 26 1-aminoindan, metformin, nicotine) may be an attrac- alone and with 10 μM CPA and DPCPX (dissolved in methanol). Fig. S2: tive therapeutic approach to reducing neurodegeneration Molecular docking simulation of α-synuclein structures C1 and C3 bound to DPCPX. Fig. S3: Molecular docking simulation of α-synuclein C5 struc- associated with increased accumulation of adenosine in ture bound to the five ribonucleosides. Fig. S4: Summary of the surface aging-related neurodegenerative diseases. area analysis of the CA1 region of the hippocampus of DAPI, α-synuclein and Thio-S. Table S1: Populations and the blockade times of each of translocations and bumping events for α-synuclein alone and α-synuclein complexes with 1% and 10% methanol. Conclusion Nanopore analysis and molecular docking techniques are Acknowledgements excellent complimentary tools to probe potential protein- This work was supported by funding from Heart and Stroke Foundation of drug complexes. Importantly, the combination of nano- Canada, NSERC Discovery Grant, Saskatchewan Health Research Foundation pore analysis and molecular docking with our in vivo (SHRF), and College of Medicine (University of Saskatchewan) to FSC. JSL was supported by NSERC Discovery Grant. MB received support from NSERC rodent model of α-synucleinopathy in the present study Discovery Grant, Canadian Institutes of Health Research (CIHR), and SHRF. EJ has provided novel insights into the structure and mis- received Doctoral Award from College of Medicine, University of Saskatch- folding pattern of the inherently disordered α-syn pro- ewan. MTM received SHRF Postdoctoral Fellowship and Parkinson Canada Postdoctoral Fellowship. All the biosafety and approvals for the ethical and tein in the presence of adenosine A1R ligands. Here, we humane use of animals were obtained prior to the start of the study. All proce- demonstrated that adenosine and the A1R agonist CPA dures performed in this study were in accordance with the ethical standards bind to the N-terminus of α-syn, similar to 2-aminoin- of the University of Saskatchewan. The study was approved by the Animal Review and Ethics Board (AREB) of the University of Saskatchewan (Animal Use dan, thereby promoting a more compact, neurotoxic Protocol # 20070090). knot conformation that leads to increased α-syn aggrega- tion and neurodegeneration in the SN and hippocampus Authors’ contributions E.J. performed and analyzed the nanopore assay, animal chronic injections (Fig. 5). In contrast, the A1R antagonist DPCPX binds and postmortem histological and biochemical experiments. M.T.M. carried out to both the N- and C-termini of α-syn, similar to 1-ami- the structural modeling analysis. E.J. and F.S.C. wrote the manuscript; M.T.M., noindan, causing the protein to adopt a neuroprotective J.S.L. and M.B. helped revise the manuscript. All authors read and approved the final manuscript. loop conformation and thereby reducing neurodegenera- tion under persistent A1R stimulation. Still, the underly- Funding ing mechanisms of CPA-mediated α-syn accumulation This research was funded by the Natural Sciences and Engineering Research Council of Canada Grant (FSC), the Saskatchewan Health Foundation Collabo- and aggregation and subsequent neurodegeneration rative Innovation and Development Grant (FSC and JSL), the Heart and Stroke require further studies. Our recent study demonstrated Foundation of Canada (FSC), and Canada Foundation for Innovation Leaders that chronic A1R stimulation with CPA leads to A1R- Opportunity Fund (FSC). MTM was funded by Postdoctoral Fellowships from Saskatchewan Health Research Foundation and Parkinson Canada. dependent accumulation of α-syn [25], but other plau- sible explanation for its intracellular accumulation may Availability of data and materials also involve downstream A1R signaling, reduced vesicu- All the data generated or analyzed during this study are included in this manuscript. Original raw data are available from the University of Saskatch- lar trafficking of α-syn to the surface membranes, or ewan (Department of Surgery) and can be readily furnished upon request. increased protein stability and reduced degradation of α-syn upon direct binding with A1R ligands. Therefore, Declarations stable inhibition of chronic adenosine A1R stimula- tion occurring in aged brains of PD patients by clinically Ethical approval and consent to participate Not applicable. approved drugs that also promote a “loop” conformation of α-syn may be beneficial neuroprotective therapies to Consent for publication decrease α-synucleinopathy in PD. Not applicable. Competing interests The authors declare that they have no competing interests. Abbreviations DPCPX: 8-Cyclopentyl-1,3-dipropylxanthine; A1R: Adenosine A1 receptor; 6 Author details α-syn: Alpha-synuclein; CPA: N -cyclopentyladenosine; FJC: Fluoro-Jade C; TH: Department of Surgery, College of Medicine, University of Saskatchewan, Tyrosine hydroxylase. Saskatoon, SK, Canada. Department of Chemistry and Biochemistry, Faculty of Science, University of Regina, Regina, SK, Canada. Department of Bio- Supplementary Information chemistry, Microbiology and Immunology, College of Medicine, University of Saskatchewan, Saskatoon, SK, Canada. The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40035- 022- 00284-3. Received: 7 December 2020 Accepted: 21 January 2022 Additional file 1: Appendix 1: Summary of the eight α-synuclein structures as determined by discrete molecular dynamics simulations and further confirmed by far-UV circular dichroism and cross-linking mass spectrometry [47]. Fig. S1: Nanopore analysis recordings of α-synuclein Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 25 of 26 References 25. Lv YC, Gao AB, Yang J, Zhong LY, Jia B, Ouyang SH, et al. Long-term 1. Gillis P, Malter JS. The adenosine-uridine binding factor recognizes the adenosine A1 receptor activation-induced sortilin expression promotes AU-rich elements of cytokine, lymphokine, and oncogene mRNAs. J Biol α-synuclein upregulation in dopaminergic neurons. Neural Regen Res. Chem. 1991;266(5):3172–7. 2020;15(4):712. 2. Bours MJ, Swennen EL, Di Virgilio F, Cronstein BN, Dagnelie PC. Adenosine 26. Shen KZ, Johnson SW. Presynaptic GABAB and adenosine A1 receptors 5’-triphosphate and adenosine as endogenous signaling molecules in regulate synaptic transmission to rat substantia nigra reticulata neurones. immunity and inflammation. Pharmacol Ther. 2006;112(2):358–404. J Physiol. 1997;505(1):153–63. 3. Touyz RM, Briones AM. Reactive oxygen species and vascular biology: 27. Liu DZ, Xie KQ, Ji XQ, Ye Y, Jiang CL, Zhu XZ. Neuroprotective effect implications in human hypertension. Hypertens Res. 2011;34(1):5–14. of paeoniflorin on cerebral ischemic rat by activating adenosine A1 4. Burnstock G, Verkhratsky A. Long-term (trophic) purinergic signalling: receptor in a manner different from its classical agonists. Br J Pharmacol. purinoceptors control cell proliferation, differentiation and death. Cell 2005;146(4):604–11. Death Dis. 2010;1(1):e9–e9. 28. Wang Y T, Lin HC, Zhao WZ, Huang HJ, Lo YL, Wang HT, et al. Acrolein acts 5. Cao X, Li LP, Qin XH, Li SJ, Zhang M, Wang Q, et al. Astrocytic Adenosine as a neurotoxin in the nigrostriatal dopaminergic system of rat: involve- 5′-Triphosphate release regulates the proliferation of neural stem cells in ment of α-synuclein aggregation and programmed cell death. Sci Rep. the adult hippocampus. Stem Cells. 2013;31(8):1633–43. 2017;7(1):1–9. 6. Abbracchio MP, Burnstock G, Verkhratsky A, Zimmermann H. Puriner- 29. Tapias V, Hu X, Luk KC, Sanders LH, Lee VM, Greenamyre JT. Synthetic gic signalling in the nervous system: an overview. Trends Neurosci. alpha-synuclein fibrils cause mitochondrial impairment and selective 2009;32(1):19–29. dopamine neurodegeneration in part via iNOS-mediated nitric oxide 7. Eltzschig HK, Eckle T. Ischemia and reperfusion—from mechanism to production. Cell Mol Life Sci. 2017;74(15):2851–74. translation. Nat Med. 2011;17(11):1391–401. 30. Hu Q, Ren X, Liu Y, Li Z, Zhang L, Chen X, et al. Aberrant adenosine A2A 8. Pagonopoulou O, Efthimiadou E, Asimakopoulos B, Nikolettos NK. receptor signaling contributes to neurodegeneration and cogni- Modulatory role of adenosine and its receptors in epilepsy: possible tive impairments in a mouse model of synucleinopathy. Exp Neurol. therapeutic approaches. Neurosci Res. 2006;56(1):14–20. 2016;283:213–23. 9. Dale N, Frenguelli BG. Release of adenosine and ATP during ischemia and 31. Schmued LC, Stowers CC, Scallet AC, Xu L. Fluoro-Jade C results in ultra epilepsy. Curr Neuropharmacol. 2009;7(3):160–79. high resolution and contrast labeling of degenerating neurons. Brain Res. 10. Ferré S, Ciruela F, Canals M, Marcellino D, Burgueno J, Casadó V, et al. 2005;1035(1):24–31. Adenosine A2A-dopamine D2 receptor–receptor heteromers. Targets for 32. Lee HJ, Shin SY, Choi C, Lee YH, Lee SJ. Formation and removal of neuro-psychiatric disorders. Parkinsonism Re Disord. 2004;10(5):265–271. α-synuclein aggregates in cells exposed to mitochondrial inhibitors. J Biol 11. Saitoh M, Nagai K, Nakagawa K, Yamamura T, Yamamoto S, Nishizaki T. Chem. 2002;277(7):5411–7. Adenosine induces apoptosis in the human gastric cancer cells via an 33. Madampage CA, Tavassoly O, Christensen C, Kumari M, Lee JS. Nanopore intrinsic pathway relevant to activation of AMP-activated protein kinase. analysis. Prion. 2012;6(2):116–23. Biochem Pharmacol. 2004;67(10):2005–11. 34. Baran C, Smith GTS, Bamm VV, Harauz G, Lee JS. Divalent cations induce 12. Manwani B, McCullough LD. Function of the master energy regulator a compaction of intrinsically disordered myelin basic protein. Biochem adenosine monophosphate-activated protein kinase in stroke. J Neurosci Biophys Res Commun. 2010;391(1):224–9. Res. 2013;91(8):1018–29. 35. Lee JS, Madampage CA, Stefureac RI, Napper S, Andrievskaia O. Nanopore 13. Angulo E, Casadó V, Mallol J, Canela EI, Viñals F, Ferrer I, et al. A1 Adeno- analysis of the interaction of metal ions and antibodies with prion pro- sine receptors accumulate in neurodegenerative structures in Alzheimer’s teins. Biochem Cell Biol. 2010;88(2):400–2. disease and mediate both amyloid precursor protein processing and tau 36. Kakish J, Tavassoly O, Lee JS. Rasagiline, a suicide inhibitor of mono- phosphorylation and translocation. Brain Pathol. 2006;13(4):440–51. amine oxidases, binds reversibly to α-synuclein. ACS Chem Neurosci. 14. Jansen KLR, Faull RLM, Dragunow M, Synek BL. Alzheimer’s disease: 2014;6(2):347–55. Changes in hippocampal N-methyl-d-aspartate, quisqualate, neuroten- 37. Tavassoly O, Kakish J, Nokhrin S, Dmitriev O, Lee JS. The use of nanopore sin, adenosine, benzodiazepine, serotonin and opioid receptors—an analysis for discovering drugs which bind to α-synuclein for treatment of autoradiographic study. Neuroscience. 1990;39(3):613–27. Parkinson’s disease. Eur J Med Chem. 2014;88:42–54. 15. Morelli M, Carta AR, Jenner P. Adenosine A2A receptors and Parkinson’s 38. Tavassoly O, Lee JS. Methamphetamine binds to α-synuclein and causes disease. Handb Exp Pharmacol. 2009;193:589–615. a conformational change which can be detected by nanopore analysis. 16. de Lau LML, Breteler MMB. Epidemiology of Parkinson’s disease. Lancet FEBS Lett. 2012;586(19):3222–8. Neurol. 2006;5(6):525–35. 39. Jakova E, Lee JS. Superposition of an AC field improves the discrimination 17. Ingelsson M. Alpha-synuclein oligomers—neurotoxic molecules in between peptides in nanopore analysis. Analyst. 2015;140(14):4813–9. Parkinson’s disease and other Lewy body disorders. Front Neurosci. 40. Christensen C, Baran C, Krasniqi B, Stefureac RI, Nokhrin S, Lee JS. Eec ff t of 2016;10:408. charge, topology and orientation of the electric field on the interaction of 18. Daue W, Przedborski S. Parkinson’s disease: mechanisms and models. peptides with the α-hemolysin pore. J Pept Sci. 2011;17(11):726–34. Neuron. 2003;39(6):889–909. 41. Meng H, Detillieux D, Baran C, Krasniqi B, Christensen C, Mad- 19. Spillantini MG, Crowther RA, Jakes R, Hasegawa M, Goedert M. ampage C, et al. Nanopore analysis of tethered peptides. J Pept Sci. α-Synuclein in filamentous inclusions of Lewy bodies from Parkin- 2010;16(12):701–8. son’s disease and dementia with Lewy bodies. Proc Natl Acad Sci U S 42. Malty RH, Jessulat M, Jin K, Musso G, Vlasblom J, Phanse S, et al. Mito- A. 1998;95(11):6469–73. chondrial targets for pharmacological intervention in human disease. J 20. Kakish J, Lee D, Lee JS. Drugs that bind to α-synuclein: neuroprotective or Proteome Res. 2015;14(1):5–21. neurotoxic? ACS Chem Neurosci. 2015;6(12):1930–40. 43. Moutaoufik MT, Malty RH, Amin S, Zhang Q, Phanse S, Gagarinova A, 21. Jakova E, Lee JS. Behavior of α-synuclein–drug complexes during et al. Rewiring of the human mitochondrial interactome during neuronal nanopore analysis with a superimposed AC field. Electrophoresis. reprogramming reveals regulators of the respirasome and neurogenesis. 2017;38(2):350–60. iScience. 2019;19:1114–1132. 22. Tripathi R, Saber H, Chauhan V, Tripathi K, Factor S. Parkinson disease from 44. Rotem D, Jayasinghe L, Salichou M, Bayley H. Protein detection by nano- long term drug abuse: Meta-analysis of amphetamine/methampheta- pores equipped with aptamers. J Am Chem Soc. 2012;134(5):2781–7. mine and Parkinson disease. Neurology. 2018;90(15):p6.079. 45. Venkatesan BM, Bashir R. Nanopore sensors for nucleic acid analysis. Nat 23. Pregeljc D, Teodorescu-Perijoc D, Vianello R, Umek N, Mavri J. How impor- Nanotechnol. 2011;6:615. tant is the use of cocaine and amphetamines in the development of Par- 46. Cooper JC, Hall EAH. The nature of biosensor technology. J Biomed Eng. kinson disease? A computational study. Neurotox Res. 2020;37(3):724–31. 1988;10(3):210–9. 24. Stockwell J, Jakova E, Cayabyab FS. Adenosine A1 and A2A receptors in 47. Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural the brain: current research and their role in neurodegeneration. Mol- heterogeneity of α-synuclein is governed by several distinct subpopula- ecules. 2017;22(4):676. tions with interconversion times slower than milliseconds. Structure. 2021;29(9):1048–64. Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 26 of 26 48. Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural 69. Kondo T, Mizuno Y, Japanese Istradefylline Study Group. A long-term ensemble of alpha-synuclein monomer. Zenodo. 2021; https:// zenodo. study of istradefylline safety and efficacy in patients with Parkinson org/ record/ 47216 17#. YeM3a tHMLOg disease. Clin Neuropharmacol. 2015;38(2):41–46. 49. Chen Z, Xiong C, Pancyr C, Stockwell J, Walz W, Cayabyab FS. Prolonged 70. Vorovenci RJ, Antonini A. The efficacy of oral adenosine A2A antagonist adenosine A1 receptor activation in hypoxia and pial vessel disruption istradefylline for the treatment of moderate to severe Parkinson’s disease. focal cortical ischemia facilitates clathrin-mediated AMPA receptor endo- Expert Rev Neurother. 2015;15:1383–90. cytosis and long-lasting synaptic inhibition in rat hippocampal CA3-CA1 71. Chen, JF, Cunha RA. The belated US FDA approval of the adenosine A 2A synapses: differential regulation of GluA2 and GluA1 subunits by p38 receptor antagonist istradefylline for treatment of Parkinson’s disease. MAPK and JNK. J Neurosci. 2014;34(29):9621–43. Purinergic Signal. 2020:1–8. 50. Högen T, Levin J, Schmidt F, Caruana M, Vassallo N, Kretzschmar H, et al. 72. Stockwell J, Chen Z, Niazi M, Nosib S, Cayabyab FS. Protein phosphatase Two different binding modes of α-synuclein to lipid vesicles depending role in adenosine A1 receptor-induced AMPA receptor trafficking and rat on its aggregation state. Biophys J. 2012;102(7):1646–55. hippocampal neuronal damage in hypoxia/reperfusion injury. Neurop- 51. Fredholm BB, Bättig K, Holmén J, Nehlig A, Zvartau EE. Actions of caffeine harmacology. 2016;102:254–65. in the brain with special reference to factors that contribute to its wide- 73. Chen Z, Stockwell J, Cayabyab FS. Adenosine A1 receptor-mediated spread use. Pharmacol Rev. 1999;51(1):83–133. endocytosis of AMPA receptors contributes to impairments in long-term 52. Daly JW, Butts-Lamb P, Padgett W. Subclasses of adenosine receptors in potentiation (LTP) in the middle-aged rat hippocampus. Neurochem Res. the central nervous system: interaction with caffeine and related methyl- 2016;41(5):1085–97. xanthines. Cell Mol Neurobiol. 1983;3(1):69–80. 74. Qin X, Zaki MG, Chen Z, Jakova E, Ming Z, Cayabyab FS. Adenosine signal- 53. Biaggioni I, Paul S, Puckett A, Arzubiaga C. Caffeine and theophylline ing and clathrin-mediated endocytosis of glutamate AMPA receptors in as adenosine receptor antagonists in humans. J Pharmacol Exp Ther. delayed hypoxic injury in rat hippocampus: Role of casein kinase 2. Mol 1991;258(2):588–93. Neurobiol. 2021;58(5):1932–51. 54. Kakish J, Allen KJH, Harkness TA, Krol ES, Lee JS. Novel dimer com- 75. Olanow CW, Rascol O, Hauser R, Feigin PD, Jankovic J, Lang A, et al. 2009. pounds that bind α-synuclein can rescue cell growth in a yeast model A double-blind, delayed-start trial of rasagiline in Parkinson’s disease. N overexpressing α-synuclein. A possible prevention strategy for Parkinson’s Engl J Med. 2009;361(13):1268–1278. disease. ACS Chem Neurosci. 2016;7(12):1671–1680. 76. Group, Parkinson Study. A controlled, randomized, delayed-start study of 55. Bennett LLEE, Schnebli HP, Vail MH, Allan PW, Montomery JA. Purine ribo- rasagiline in early Parkinson disease. Arch Neurol. 2004;61(4):561–6. nucleoside kinase activity and resistance to some analogs of adenosine. 77. Callaghan RC, Cunningham JK, Sykes J, Kish SJ. Increased risk of Parkin- Mol Pharmacol. 1966;2(5):432–43. son’s disease in individuals hospitalized with conditions related to the use 56. Tomlinson CL, Stowe R, Patel S, Rick C, Gray R, Clarke CE. Systematic of methamphetamine or other amphetamine-type drugs. Drug Alcohol review of levodopa dose equivalency reporting in Parkinson’s disease. Depend. 2012;120(1):35–40. Mov Disord. 2010;25(15):2649–53. 78. Curtin K, Fleckenstein AE, Robison RJ, Crookston MJ, Smith KR, Hanson 57. Marsden CD, Parkes JD. Success and problems of long-term levodopa GR. Methamphetamine/amphetamine abuse and risk of Parkinson’s therapy in Parkinson’s disease. Lancet. 1977;309(8007):345–9. disease in Utah: a population-based assessment. Drug Alcohol Depend. 58. Olanow CW, Obeso JA, Stocchi F. Continuous dopamine-receptor treat- 2015;146:30–8. ment of Parkinson’s disease: scientific rationale and clinical implications. 79. Ulmer TS, Bax A, Cole NB, Nussbaum RL. Structure and dynamics of Lancet Neurol. 2005;5(8):677–87. micelle-bound human α-synuclein. J Biol Chem. 2005;280(10):9595–603. 59. Bonifácio MJ, Palma PN, Almeida L, Soares-da-Silva P. Catechol-O- methyltransferase and its inhibitors in Parkinson’s disease. CNS Drug Rev. 2007;13(3):352–79. 60. Group, Parkinson Study. A randomized placebo-controlled trial of rasagiline in levodopa-treated patients with parkinson disease and motor fluctuations: the presto study. Arch Neurol. 2005;62(2):241–8. 61. Horstink M, Tolosa E, Bonuccelli U, Deuschl G, Friedman A, Kanovsky P, et al. Review of the therapeutic management of Parkinson’s disease. Report of a joint task force of the European Federation of Neurologi- cal Societies and the Movement Disorder Society–European Sec- tion. Part I: early (uncomplicated) Parkinson’s disease. Eur J Neurol. 2006;13(11):1170–1185. 62. Schade R, Andersohn F, Suissa S, Haverkamp W, Garbe E. Dopamine agonists and the risk of cardiac-valve regurgitation. N Engl J Med. 2007;356(1):29–38. 63. Ferreira DG, Batalha VL, Vincente Miranda H, Coelho JE, Gomes R, Gon- çalves FQ, et al. Adenosine A2A receptors modulate α-synuclein aggrega- tion and toxicity. Cereb Cortex. 2017;27(1):718–30. 64. Schwarzschild MA, Agnati L, Fuxe K, Chen JF, Morelli M. Targeting adenosine A2A receptors in Parkinson’s disease. Trends Neurosci. 2006;29(11):647–54. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : 65. Factor SA, Wolski K, Togasaki DM, Huyck S, Cantillon M, Ho TW, et al. 2013. Long-term safety and efficacy of preladenant in subjects with fluctuating fast, convenient online submission Parkinson’s disease. Mov Disord. 2013;28(6):817–820. thorough peer review by experienced researchers in your ﬁeld 66. Mizuno Y, Hasegawa K, Kondo T, Kuno S, Yamamoto M. Clinical efficacy of istradefylline (KW-6002) in Parkinson’s disease: a randomized, controlled rapid publication on acceptance study. Mov Disord. 2010;25(10):1437–43. support for research data, including large and complex data types 67. Prediger RDS. Eec ff ts of caffeine in Parkinson’s disease: from neuro - • gold Open Access which fosters wider collaboration and increased citations protection to the management of motor and non-motor symptoms. J Alzheimers Dis. 2010;20(1):205–20. maximum visibility for your research: over 100M website views per year 68. Postuma RB, Lang AE, Munhoz RP, Charland K, Pelletier A, Moscov- ich M, et al. Caffeine for treatment of Parkinson disease. Neurology. At BMC, research is always in progress. 2012;79(7):651–8. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Translational Neurodegeneration Springer Journals http://www.deepdyve.com/lp/springer-journals/adenosine-a1-receptor-ligands-bind-to-synuclein-implications-for-4ciPbIB5Ch

Loading next page...

References (88)

(2015)
Methamphetamine/amphetamine abuse and risk of Parkinson’s disease in Utah: a population-based assessment
Drug Alcohol Depend, 146
(2007)
Catechol-O-methyltransferase and its inhibitors in Parkinson’s disease
CNS Drug Rev, 13
(2010)
Clinical efficacy of istradefylline (KW-6002) in Parkinson’s disease: a randomized, controlled study
Mov Disord, 25
P Gillis (1991)
3172
J Biol Chem, 266
(2014)
Rasagiline, a suicide inhibitor of monoamine oxidases, binds reversibly to α-synuclein
ACS Chem Neurosci, 6
(1988)
The nature of biosensor technology
J Biomed Eng, 10
(1983)
Subclasses of adenosine receptors in the central nervous system: interaction with caffeine and related methylxanthines
Cell Mol Neurobiol, 3
(2010)
Nanopore analysis of the interaction of metal ions and antibodies with prion proteins
Biochem Cell Biol, 88
(Tripathi R, Saber H, Chauhan V, Tripathi K, Factor S. Parkinson disease from long term drug abuse: Meta-analysis of amphetamine/methamphetamine and Parkinson disease. Neurology. 2018;90(15):p6.079.)
Tripathi R, Saber H, Chauhan V, Tripathi K, Factor S. Parkinson disease from long term drug abuse: Meta-analysis of amphetamine/methamphetamine and Parkinson disease. Neurology. 2018;90(15):p6.079.
Tripathi R, Saber H, Chauhan V, Tripathi K, Factor S. Parkinson disease from long term drug abuse: Meta-analysis of amphetamine/methamphetamine and Parkinson disease. Neurology. 2018;90(15):p6.079., Tripathi R, Saber H, Chauhan V, Tripathi K, Factor S. Parkinson disease from long term drug abuse: Meta-analysis of amphetamine/methamphetamine and Parkinson disease. Neurology. 2018;90(15):p6.079.
(Moutaoufik MT, Malty RH, Amin S, Zhang Q, Phanse S, Gagarinova A, et al. Rewiring of the human mitochondrial interactome during neuronal reprogramming reveals regulators of the respirasome and neurogenesis. iScience. 2019;19:1114–1132.)
Moutaoufik MT, Malty RH, Amin S, Zhang Q, Phanse S, Gagarinova A, et al. Rewiring of the human mitochondrial interactome during neuronal reprogramming reveals regulators of the respirasome and neurogenesis. iScience. 2019;19:1114–1132.
Moutaoufik MT, Malty RH, Amin S, Zhang Q, Phanse S, Gagarinova A, et al. Rewiring of the human mitochondrial interactome during neuronal reprogramming reveals regulators of the respirasome and neurogenesis. iScience. 2019;19:1114–1132., Moutaoufik MT, Malty RH, Amin S, Zhang Q, Phanse S, Gagarinova A, et al. Rewiring of the human mitochondrial interactome during neuronal reprogramming reveals regulators of the respirasome and neurogenesis. iScience. 2019;19:1114–1132.
(2007)
Dopamine agonists and the risk of cardiac-valve regurgitation
N Engl J Med, 356
(2004)
A controlled, randomized, delayed-start study of rasagiline in early Parkinson disease
Arch Neurol, 61
(2006)
Targeting adenosine A2A receptors in Parkinson’s disease
Trends Neurosci, 29
MJ Bours (2006)
358
Pharmacol Ther, 112
(Factor SA, Wolski K, Togasaki DM, Huyck S, Cantillon M, Ho TW, et al. 2013. Long-term safety and efficacy of preladenant in subjects with fluctuating Parkinson’s disease. Mov Disord. 2013;28(6):817–820.)
Factor SA, Wolski K, Togasaki DM, Huyck S, Cantillon M, Ho TW, et al. 2013. Long-term safety and efficacy of preladenant in subjects with fluctuating Parkinson’s disease. Mov Disord. 2013;28(6):817–820.
Factor SA, Wolski K, Togasaki DM, Huyck S, Cantillon M, Ho TW, et al. 2013. Long-term safety and efficacy of preladenant in subjects with fluctuating Parkinson’s disease. Mov Disord. 2013;28(6):817–820., Factor SA, Wolski K, Togasaki DM, Huyck S, Cantillon M, Ho TW, et al. 2013. Long-term safety and efficacy of preladenant in subjects with fluctuating Parkinson’s disease. Mov Disord. 2013;28(6):817–820.
(1997)
Presynaptic GABAB and adenosine A1 receptors regulate synaptic transmission to rat substantia nigra reticulata neurones
J Physiol, 505
(2006)
A1 Adenosine receptors accumulate in neurodegenerative structures in Alzheimer’s disease and mediate both amyloid precursor protein processing and tau phosphorylation and translocation
Brain Pathol, 13
(2010)
Nanopore analysis of tethered peptides
J Pept Sci, 16
(2015)
Mitochondrial targets for pharmacological intervention in human disease
J Proteome Res, 14
X Cao (2013)
1633
Stem Cells, 31
(2015)
The efficacy of oral adenosine A2A antagonist istradefylline for the treatment of moderate to severe Parkinson’s disease
Expert Rev Neurother., 15
(2016)
Alpha-synuclein oligomers—neurotoxic molecules in Parkinson’s disease and other Lewy body disorders
Front Neurosci, 10
(2020)
How important is the use of cocaine and amphetamines in the development of Parkinson disease? A computational study
Neurotox Res, 37
(2011)
Effect of charge, topology and orientation of the electric field on the interaction of peptides with the α-hemolysin pore
J Pept Sci, 17
(2006)
Adenosine 5'-triphosphate and adenosine as endogenous signaling molecules in immunity and inflammation
Pharmacol Ther, 112
(2009)
Release of adenosine and ATP during ischemia and epilepsy
Curr Neuropharmacol, 7
(2017)
Synthetic alpha-synuclein fibrils cause mitochondrial impairment and selective dopamine neurodegeneration in part via iNOS-mediated nitric oxide production
Cell Mol Life Sci, 74
(2013)
Astrocytic Adenosine 5′-Triphosphate release regulates the proliferation of neural stem cells in the adult hippocampus
Stem Cells, 31
(2015)
Drugs that bind to α-synuclein: neuroprotective or neurotoxic?
ACS Chem Neurosci, 6
(2006)
Modulatory role of adenosine and its receptors in epilepsy: possible therapeutic approaches
Neurosci Res, 56
(2005)
A randomized placebo-controlled trial of rasagiline in levodopa-treated patients with parkinson disease and motor fluctuations: the presto study
Arch Neurol, 62
(2021)
Adenosine signaling and clathrin-mediated endocytosis of glutamate AMPA receptors in delayed hypoxic injury in rat hippocampus: Role of casein kinase 2
Mol Neurobiol, 58
G Burnstock (2010)
e9
Cell Death Dis, 1
O Pagonopoulou (2006)
14
Neurosci Res, 56
(2012)
Two different binding modes of α-synuclein to lipid vesicles depending on its aggregation state
Biophys J, 102
(2012)
Nanopore analysis
Prion, 6
(2012)
Methamphetamine binds to α-synuclein and causes a conformational change which can be detected by nanopore analysis
FEBS Lett, 586
(2010)
Effects of caffeine in Parkinson’s disease: from neuroprotection to the management of motor and non-motor symptoms
J Alzheimers Dis, 20
(2011)
Ischemia and reperfusion—from mechanism to translation
Nat Med, 17
(2017)
Adenosine A2A receptors modulate α-synuclein aggregation and toxicity
Cereb Cortex, 27
(2016)
Aberrant adenosine A2A receptor signaling contributes to neurodegeneration and cognitive impairments in a mouse model of synucleinopathy
Exp Neurol, 283
(2012)
Increased risk of Parkinson’s disease in individuals hospitalized with conditions related to the use of methamphetamine or other amphetamine-type drugs
Drug Alcohol Depend, 120
(2005)
Structure and dynamics of micelle-bound human α-synuclein
J Biol Chem, 280
(Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural ensemble of alpha-synuclein monomer. Zenodo. 2021; https://zenodo.org/record/4721617#.YeM3atHMLOg)
Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural ensemble of alpha-synuclein monomer. Zenodo. 2021; https://zenodo.org/record/4721617#.YeM3atHMLOg
Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural ensemble of alpha-synuclein monomer. Zenodo. 2021; https://zenodo.org/record/4721617#.YeM3atHMLOg, Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural ensemble of alpha-synuclein monomer. Zenodo. 2021; https://zenodo.org/record/4721617#.YeM3atHMLOg
(1977)
Success and problems of long-term levodopa therapy in Parkinson’s disease
Lancet, 309
HK Eltzschig (2011)
1391
Nat Med, 17
(Ferré S, Ciruela F, Canals M, Marcellino D, Burgueno J, Casadó V, et al. Adenosine A2A-dopamine D2 receptor–receptor heteromers. Targets for neuro-psychiatric disorders. Parkinsonism Re Disord. 2004;10(5):265–271.)
Ferré S, Ciruela F, Canals M, Marcellino D, Burgueno J, Casadó V, et al. Adenosine A2A-dopamine D2 receptor–receptor heteromers. Targets for neuro-psychiatric disorders. Parkinsonism Re Disord. 2004;10(5):265–271.
Ferré S, Ciruela F, Canals M, Marcellino D, Burgueno J, Casadó V, et al. Adenosine A2A-dopamine D2 receptor–receptor heteromers. Targets for neuro-psychiatric disorders. Parkinsonism Re Disord. 2004;10(5):265–271., Ferré S, Ciruela F, Canals M, Marcellino D, Burgueno J, Casadó V, et al. Adenosine A2A-dopamine D2 receptor–receptor heteromers. Targets for neuro-psychiatric disorders. Parkinsonism Re Disord. 2004;10(5):265–271.
(2005)
Continuous dopamine-receptor treatment of Parkinson’s disease: scientific rationale and clinical implications
Lancet Neurol, 5
(2005)
Fluoro-Jade C results in ultra high resolution and contrast labeling of degenerating neurons
Brain Res, 1035
(2017)
Acrolein acts as a neurotoxin in the nigrostriatal dopaminergic system of rat: involvement of α-synuclein aggregation and programmed cell death
Sci Rep, 7
(1966)
Purine ribonucleoside kinase activity and resistance to some analogs of adenosine
Mol Pharmacol, 2
(2010)
Divalent cations induce a compaction of intrinsically disordered myelin basic protein
Biochem Biophys Res Commun, 391
(2012)
Caffeine for treatment of Parkinson disease
Neurology, 79
(2014)
Prolonged adenosine A1 receptor activation in hypoxia and pial vessel disruption focal cortical ischemia facilitates clathrin-mediated AMPA receptor endocytosis and long-lasting synaptic inhibition in rat hippocampal CA3-CA1 synapses: differential regulation of GluA2 and GluA1 subunits by p38 MAPK and JNK
J Neurosci, 34
RM Touyz (2011)
5
Hypertens Res, 34
(2010)
Systematic review of levodopa dose equivalency reporting in Parkinson’s disease
Mov Disord, 25
(2012)
Protein detection by nanopores equipped with aptamers
J Am Chem Soc, 134
(Kondo T, Mizuno Y, Japanese Istradefylline Study Group. A long-term study of istradefylline safety and efficacy in patients with Parkinson disease. Clin Neuropharmacol. 2015;38(2):41–46.)
Kondo T, Mizuno Y, Japanese Istradefylline Study Group. A long-term study of istradefylline safety and efficacy in patients with Parkinson disease. Clin Neuropharmacol. 2015;38(2):41–46.
Kondo T, Mizuno Y, Japanese Istradefylline Study Group. A long-term study of istradefylline safety and efficacy in patients with Parkinson disease. Clin Neuropharmacol. 2015;38(2):41–46., Kondo T, Mizuno Y, Japanese Istradefylline Study Group. A long-term study of istradefylline safety and efficacy in patients with Parkinson disease. Clin Neuropharmacol. 2015;38(2):41–46.
(2004)
Adenosine induces apoptosis in the human gastric cancer cells via an intrinsic pathway relevant to activation of AMP-activated protein kinase
Biochem Pharmacol, 67
(2010)
Long-term (trophic) purinergic signalling: purinoceptors control cell proliferation, differentiation and death
Cell Death Dis, 1
(2017)
Adenosine A1 and A2A receptors in the brain: current research and their role in neurodegeneration
Molecules, 22
(2009)
Adenosine A2A receptors and Parkinson’s disease
Handb Exp Pharmacol., 193
(Kakish J, Allen KJH, Harkness TA, Krol ES, Lee JS. Novel dimer compounds that bind α-synuclein can rescue cell growth in a yeast model overexpressing α-synuclein. A possible prevention strategy for Parkinson’s disease. ACS Chem Neurosci. 2016;7(12):1671–1680.)
Kakish J, Allen KJH, Harkness TA, Krol ES, Lee JS. Novel dimer compounds that bind α-synuclein can rescue cell growth in a yeast model overexpressing α-synuclein. A possible prevention strategy for Parkinson’s disease. ACS Chem Neurosci. 2016;7(12):1671–1680.
Kakish J, Allen KJH, Harkness TA, Krol ES, Lee JS. Novel dimer compounds that bind α-synuclein can rescue cell growth in a yeast model overexpressing α-synuclein. A possible prevention strategy for Parkinson’s disease. ACS Chem Neurosci. 2016;7(12):1671–1680., Kakish J, Allen KJH, Harkness TA, Krol ES, Lee JS. Novel dimer compounds that bind α-synuclein can rescue cell growth in a yeast model overexpressing α-synuclein. A possible prevention strategy for Parkinson’s disease. ACS Chem Neurosci. 2016;7(12):1671–1680.
(1991)
Caffeine and theophylline as adenosine receptor antagonists in humans
J Pharmacol Exp Ther, 258
(Horstink M, Tolosa E, Bonuccelli U, Deuschl G, Friedman A, Kanovsky P, et al. Review of the therapeutic management of Parkinson’s disease. Report of a joint task force of the European Federation of Neurological Societies and the Movement Disorder Society–European Section. Part I: early (uncomplicated) Parkinson’s disease. Eur J Neurol. 2006;13(11):1170–1185.)
Horstink M, Tolosa E, Bonuccelli U, Deuschl G, Friedman A, Kanovsky P, et al. Review of the therapeutic management of Parkinson’s disease. Report of a joint task force of the European Federation of Neurological Societies and the Movement Disorder Society–European Section. Part I: early (uncomplicated) Parkinson’s disease. Eur J Neurol. 2006;13(11):1170–1185.
Horstink M, Tolosa E, Bonuccelli U, Deuschl G, Friedman A, Kanovsky P, et al. Review of the therapeutic management of Parkinson’s disease. Report of a joint task force of the European Federation of Neurological Societies and the Movement Disorder Society–European Section. Part I: early (uncomplicated) Parkinson’s disease. Eur J Neurol. 2006;13(11):1170–1185., Horstink M, Tolosa E, Bonuccelli U, Deuschl G, Friedman A, Kanovsky P, et al. Review of the therapeutic management of Parkinson’s disease. Report of a joint task force of the European Federation of Neurological Societies and the Movement Disorder Society–European Section. Part I: early (uncomplicated) Parkinson’s disease. Eur J Neurol. 2006;13(11):1170–1185.
(2011)
Reactive oxygen species and vascular biology: implications in human hypertension
Hypertens Res, 34
(2017)
Behavior of α-synuclein–drug complexes during nanopore analysis with a superimposed AC field
Electrophoresis, 38
(1998)
α-Synuclein in filamentous inclusions of Lewy bodies from Parkinson’s disease and dementia with Lewy bodies
Proc Natl Acad Sci U S A, 95
(2015)
Superposition of an AC field improves the discrimination between peptides in nanopore analysis
Analyst, 140
(2013)
Function of the master energy regulator adenosine monophosphate-activated protein kinase in stroke
J Neurosci Res, 91
(Olanow CW, Rascol O, Hauser R, Feigin PD, Jankovic J, Lang A, et al. 2009. A double-blind, delayed-start trial of rasagiline in Parkinson’s disease. N Engl J Med. 2009;361(13):1268–1278.)
Olanow CW, Rascol O, Hauser R, Feigin PD, Jankovic J, Lang A, et al. 2009. A double-blind, delayed-start trial of rasagiline in Parkinson’s disease. N Engl J Med. 2009;361(13):1268–1278.
Olanow CW, Rascol O, Hauser R, Feigin PD, Jankovic J, Lang A, et al. 2009. A double-blind, delayed-start trial of rasagiline in Parkinson’s disease. N Engl J Med. 2009;361(13):1268–1278., Olanow CW, Rascol O, Hauser R, Feigin PD, Jankovic J, Lang A, et al. 2009. A double-blind, delayed-start trial of rasagiline in Parkinson’s disease. N Engl J Med. 2009;361(13):1268–1278.
(2002)
Formation and removal of α-synuclein aggregates in cells exposed to mitochondrial inhibitors
J Biol Chem, 277
(Chen, JF, Cunha RA. The belated US FDA approval of the adenosine A 2A receptor antagonist istradefylline for treatment of Parkinson’s disease. Purinergic Signal. 2020:1–8.)
Chen, JF, Cunha RA. The belated US FDA approval of the adenosine A 2A receptor antagonist istradefylline for treatment of Parkinson’s disease. Purinergic Signal. 2020:1–8.
Chen, JF, Cunha RA. The belated US FDA approval of the adenosine A 2A receptor antagonist istradefylline for treatment of Parkinson’s disease. Purinergic Signal. 2020:1–8., Chen, JF, Cunha RA. The belated US FDA approval of the adenosine A 2A receptor antagonist istradefylline for treatment of Parkinson’s disease. Purinergic Signal. 2020:1–8.
(2021)
The structural heterogeneity of α-synuclein is governed by several distinct subpopulations with interconversion times slower than milliseconds
Structure, 29
(2003)
Parkinson’s disease: mechanisms and models
Neuron, 39
(1990)
Alzheimer’s disease: Changes in hippocampal N-methyl-d-aspartate, quisqualate, neurotensin, adenosine, benzodiazepine, serotonin and opioid receptors—an autoradiographic study
Neuroscience, 39
(2006)
Epidemiology of Parkinson’s disease
Lancet Neurol, 5
(2011)
Nanopore sensors for nucleic acid analysis
Nat Nanotechnol, 6
(1999)
Actions of caffeine in the brain with special reference to factors that contribute to its widespread use
Pharmacol Rev, 51
(2016)
Adenosine A1 receptor-mediated endocytosis of AMPA receptors contributes to impairments in long-term potentiation (LTP) in the middle-aged rat hippocampus
Neurochem Res, 41
(2020)
Long-term adenosine A1 receptor activation-induced sortilin expression promotes α-synuclein upregulation in dopaminergic neurons
Neural Regen Res, 15
(2005)
Neuroprotective effect of paeoniflorin on cerebral ischemic rat by activating adenosine A1 receptor in a manner different from its classical agonists
Br J Pharmacol, 146
(2014)
The use of nanopore analysis for discovering drugs which bind to α-synuclein for treatment of Parkinson’s disease
Eur J Med Chem, 88
MP Abbracchio (2009)
19
Trends Neurosci, 32
(2016)
Protein phosphatase role in adenosine A1 receptor-induced AMPA receptor trafficking and rat hippocampal neuronal damage in hypoxia/reperfusion injury
Neuropharmacology, 102
N Dale (2009)
160
Curr Neuropharmacol, 7
(1991)
The adenosine-uridine binding factor recognizes the AU-rich elements of cytokine, lymphokine, and oncogene mRNAs
J Biol Chem, 266
(2009)
Purinergic signalling in the nervous system: an overview
Trends Neurosci, 32

Publisher: Springer Journals
Copyright: Copyright © The Author(s) 2022
eISSN: 2047-9158
DOI: 10.1186/s40035-022-00284-3
Publisher site: See Article on Publisher Site

Abstract

Background: Accumulating α-synuclein (α-syn) aggregates in neurons and glial cells are the staples of many synu- cleinopathy disorders, such as Parkinson’s disease (PD). Since brain adenosine becomes greatly elevated in ageing brains and chronic adenosine A1 receptor (A1R) stimulation leads to neurodegeneration, we determined whether adenosine or A1R receptor ligands mimic the action of known compounds that promote α-syn aggregation (e.g., the amphetamine analogue 2-aminoindan) or inhibit α-syn aggregation (e.g., Rasagiline metabolite 1-aminoindan). In the present study, we determined whether adenosine, A1R receptor agonist N -Cyclopentyladenosine (CPA) and antago- nist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX) could directly interact with α-syn to modulate α-syn aggregation and neurodegeneration of dopaminergic neurons in the substantia nigra (SN). Methods: Nanopore analysis and molecular docking were used to test the binding properties of CPA and DPCPX with α-syn in vitro. Sprague–Dawley rats were administered with 7-day intraperitoneal injections of the A1R ligands and 1- and 2-aminoindan, and levels of α-syn aggregation and neurodegeneration were examined in the SN pars compacta and hippocampal regions using confocal imaging and Western blotting. Results: Using nanopore analysis, we showed that the A1R agonists (CPA and adenosine) interacted with the N-ter- minus of α-syn, similar to 2-aminoindan, which is expected to promote a “knot” conformation and α-syn misfolding. In contrast, the A1R antagonist DPCPX interacted with the N- and C-termini of α-syn, similar to 1-aminoindan, which is expected to promote a “loop” conformation that prevents α-syn misfolding. Molecular docking studies revealed that adenosine, CPA and 2-aminoindan interacted with the hydrophobic core of α-syn N-terminus, whereas DPCPX and 1-aminoindan showed direct binding to the N- and C-terminal hydrophobic pockets. Confocal imaging and Western blot analyses revealed that chronic treatments with CPA alone or in combination with 2-aminoindan increased α-syn expression/aggregation and neurodegeneration in both SN pars compacta and hippocampus. In contrast, DPCPX and 1-aminoindan attenuated the CPA-induced α-syn expression/aggregation and neurodegeneration in SN and hippocampus. *Correspondence: frank.cayabyab@usask.ca Department of Surgery, College of Medicine, University of Saskatchewan, Saskatoon, SK, Canada Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 2 of 26 Conclusions: The results indicate that A1R agonists and drugs promoting a “knot” conformation of α-syn can cause α-synucleinopathy and increase neuronal degeneration, whereas A1R antagonists and drugs promoting a “loop” con- formation of α-syn can be harnessed for possible neuroprotective therapies to decrease α-synucleinopathy in PD. Keywords: Alpha-synucleinopathy, Adenosine A1 receptor, N -cyclopentyladenosine, 8-cyclopentyl-1,3- dipropylxanthine, 1-aminoindan, 2-aminoindan, Neuroprotection, Neurodegeneration, Protein misfolding Background subthalamonigral pathway may be clinically relevant in Adenosine is a nucleoside that is involved in many physi- improving tardive dyskinesia in PD patients by reducing ological activities including cell proliferation, migra- the glutamatergic outputs of the SN pars reticulata dopa- tion of dendritic cells, and the release of small proteins minergic neurons. Moreover, most compounds that tar- called cytokines which are vital for cell signalling from get A1R activation were believed to be neuroprotective in periphery to secondary lymphoid organs, vascular reac- both the SN and hippocampus. For example, activation tivity, apoptosis and most importantly, the passage of of A1R is involved in paeoniflorin (a chemical compound neuronal stem cells [1–5]. Adenosine is also implicated derived from Paenoia lactiflora)-induced neuroprotec - in central nervous system (CNS) disorders such as tion in cerebral ischemia in Sprague–Dawley rats [27]. ischemia, trauma, epilepsy, neuropsychiatric disorders However, we recently reported that chronic stimulation and cancer [6–11]. Moreover, various roles of adenosine of A1Rs by intraperitoneal (i.p.) injection of the A1R have garnered intense investigations in many ageing- agonist CPA in rats for 3 days is sufficient to induce neu - related neurodegenerative diseases such as ischemic rodegeneration in the hippocampus [24]. Additionally, stroke, Alzheimer’s disease (AD) and Parkinson’s disease longer-term chronic A1R stimulation with CPA increases (PD) [12–15]. PD is the second most prevalent ageing- sortilin expression that promotes α-syn upregulation in related neurodegenerative disease after AD [16]. The dopaminergic MN9D cells and SN dopaminergic neu- pathophysiology of PD directly involves the imbalance rons of Sprague–Dawley rats [25]. Since highly upregu- of dopaminergic signalling pathways and accumulation lated α-syn can be found in the hippocampus and SN of of protein aggregates of α-synuclein (α-syn) in inclu- rodent synucleinopathy models [25, 28–30], we therefore sions (Lewy bodies) causing the characteristic motor tested the possibility that the commonly used A1R-selec- and cognitive deficits commonly observed in PD patients tive agonist ligand CPA can bind to α-syn and enhance [17–19]. Recently, some neuroprotective drugs have neurotoxicity, whereas the A1R-specific antagonist ligand been found to bind to α-syn and prevent further aggre- DPCPX can bind to α-syn and promote neuroprotec- gation, including caffeine, nicotine, 1-aminoindan and tion. By 7-day chronic injections in Sprague–Dawley metformin [20]. Additionally, there are other drugs such male rats, we determined if chronic stimulation with as methamphetamine, cocaine, 2-aminoindan and the CPA causes dopaminergic neuron loss and increases herbicides, paraquat and rotenone, which appear to be expression of α-syn in the SN. We then co-administered neurotoxic because they increase α-syn misfolding and DPCPX as a method to control neurodegeneration and can be correlated with a higher incidence of PD [20–23]. decrease aggregation of α-syn caused initially by CPA. Chronic adenosine A1 receptor (A1R) stimulation has Fluoro-Jade C (FJC) and Thioflavin S (Thio-S) staining recently been reported to cause hippocampal and sub- of hippocampal and SN brain regions were performed to stantia nigra (SN) neuronal death, as well as increasing assess neurodegeneration and α-syn aggregation, respec- α-syn accumulation in dopaminergic SN neurons [24, tively [31, 32]. 25]. Since a primary therapeutic goal of management Nanopore analysis and molecular docking are use- of PD is to minimize α-syn misfolding and aggregation, ful analytical tools for studying intrinsically disordered we investigated whether adenosine and the A1R agonist proteins like cellular prion proteins, β-amyloid as well N -cyclopentyladenosine (CPA), as well as antagonist as α-syn and α-syn/drug complexes [20, 21, 33–39]. 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), can bind Nanopores are single-molecule counters consisting of to and modulate α-syn misfolding. a nanometre aperture that allows the fluxes of ions and A1Rs are expressed at high levels in the limbic sys- small charged polypeptides through an insulating mem- tem especially in the hippocampus as well as in the SN brane. Applying a voltage across this membrane results region. A1R stimulation with CPA reduces glutamate in an electrochemical gradient that drives ions through and gamma-aminobutyric acid release from nerve termi- the α-hemolysin (α-HML) toxin derived from Staphylo- nals of the SN pars reticulata region of the rat brain [26]. coccus aureus [33]. A single α-syn protein interacts with This presynaptic inhibition of glutamate release from the the α-HML pore, causing a blockade current (I) for an Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 3 of 26 Fig. 1 Nanopore analysis setup and α-synuclein (α-syn) interaction with the α-Hemolysin pore. a The patch-clamp setup at 100 mV direct current (DC) allows the ions to flow in the pore and create an ionic current b The interruption of the current when α-syn interacts with the pore forming three distinguishable blockade current events: b1 Translocation events, where α-syn goes through the pore causing a large current blockade (as seen in Fig. 1 c); b2 Intercalation events, where α-syn is trapped in the pore entrance, but will diffuse back after a period of time causing an intermediate current blockade; b3 Bumping events, where α-syn approaches the pore, but diffuses away without entering causing a small current blockade. c Disruption of the blockade current and time caused by α-syn when the protein translocates the pore. d Full sequence of α-syn. e The domains of α-syn used in the nanopore setup consisting of: N-terminus (blue); ΔNAC, the entire sequence of α-synuclein without the non-amyloid β-component region (blue and red); and C-terminus (red) amount of time (T) (Fig. 1) [39]. When α-syn translocates of single molecules. Moreover, this technique has been through the pore, a large current blockade is observed for widely used to determine if a drug binds to α-syn. When a long translocation time. Conversely, if the α-syn pro- a protein-drug complex is formed, there is an increase of tein approaches the pore but then diffuses away without bumping events and a decrease in translocation events, entering, a small current blockade is observed for a short which indicates that the drug causes folding of the time. This type of event is called bumping [40, 41]. The protein. most important advantage of nanopore analysis is that Additionally, molecular docking is a computer simula- molecules can be detected without labelling and at very tion technique that allows prediction of the binding con- low concentrations. Uniquely, this technique requires formation of a desired protein or peptide to a chemical less than an hour to non-destructively analyze thousands compound or other small molecules, making molecular Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 4 of 26 docking analysis one of the best techniques for structure- 22 ± 1 °C with an applied potential of 100 mV at a band- based drug design [42, 43]. Therefore, using these com - width of 10 kHz using an Axopatch 200B amplifier (Axon plimentary biophysical and computational techniques in Instruments, San Jose, CA) under voltage-clamp condi- combination with our in vivo studies, we aimed to elu- tions using a Clampfit software (Molecular Devices, San cidate the effects of adenosine and other A1R ligands on Jose, CA). As discussed elsewhere, temperature changes α-syn conformations and dopaminergic neuron loss in due to Joule heating are expected to be negligible [46]. PD. Data analysis Methods The blockade amplitudes and duration times obtained Animal protocol with Clampfit were transferred to Origin 7 graph - Animals were housed and treated humanely in accord- ing software (OriginLab Corporation, Northampton, ance with the guidelines from the following governing MA) and were used to construct blockade current and bodies: National Research Council (US) Committee for time histograms. The blockade amplitudes were plot - the Update of the Guide for the Care and Use of Labora- ted as statistical histograms and each event population tory Animals (Washington DC, 2011); Canadian Coun- (e.g., translocation, intercalation and bumping) was fitted cil on Animal Care (CCAC); and the University of with a Gaussian function to obtain the peak/population Saskatchewan Animal Research Ethics Board (AREB) blockade current value (I). The duration time data for that approved our Animal Use Protocol (#20070090). each population were plotted separately and the data fit - Male Sprague–Dawley rats (20–30 days old varying from ted with a single exponential decay function to obtain the 250 to 300 g) were used for immunofluorescence confo - characteristic time (T). Each experiment was repeated cal imaging and biochemical studies as described below. at least three times and the event profiles were added The animals were housed in cages of two, with free access together. The error in the peak current was estimated to to food pellets and water. be < ± 1 pA and the proportion (%) of the events in each peak is reported in Tables as means ± SEM. Reagents Structural modeling and docking α-Syn (rPeptide, Bogart, GA) was dissolved in nucle- Distinct structural subpopulations of α-syn monomers ase-free water at a final concentration of 1 μM. Adeno - (C1–C8, see Additional file 1: Appendix Fig. 1) have sine was purchased from Millipore-Sigma (Oakville, recently been identified [47] and were subsequently used Canada), CPA was purchased from Abcam (Toronto, in the present study (with the exception of C6, which Canada), DPCPX from Tocris (Burlington, Canada), 1- is believed to be membrane-bound) in our molecular and 2-aminoindan were purchased from Sigma-Aldrich docking simulations to predict the drug-protein com- (Oakville, Canada). For the nanopore analysis, all drugs plexes. The respective conformation of each of the α-syn were dissolved in methanol (MeOH) and used at a final structures was taken from PDB-DEV (Entry: PDB- concentration of 10 μM. For the 7-day chronic i.p. injec- DEV_00000082) [48]. The chemical structures of adeno - tion, CPA, DPCPX, 1- and 2-aminoindan were dissolved sine, CPA, DPCPX, 1-aminoindan and 2-aminoindan in 0.1% dimethyl sulfoxide (DMSO) (Sigma-Aldrich, were obtained from PubChem (CIDs: 60961, 53477947, Oakville, Canada) in 0.9% sodium chloride at a final con - 1329, 123445, 76310, respectively). The molecular dock - centration of 3 mg/ml. ing study was carried out using Autodock Vina module implemented in PyRx tool (La Jolla, CA) [43]. Protein and Nanopore analysis ligand interactions were analyzed and visualized through Instrument setup Pymol (New York, NY) and LigPlot + (Cambridge, UK). The standard direct current (DC) setup has been described in detail previously by our lab [39, 44, 45]. In In vivo drug treatments to study α‑syn aggregation brief, a lipid bilayer was painted onto a 150-μm aperture and neurodegeneration in a Teflon perfusion cup. The two buffer compartments In support of the in vitro data, a full in vivo study con- on either side of the lipid bilayer each contained a 1-ml sisting of 7-day chronic i.p. injections of eight reagents total volume. Five microliters of 1 μg/ml α-HML (Milli- or their combinations in 28-day old male Sprague– pore-Sigma, Oakville, Canada) were added to the cis side Dawley rats was performed. The eight treatments con - of the membrane and the current was monitored until sisted of (1) Control (0.1% DMSO in 0.9% saline), (2) stable pore insertion was achieved. Consistent results CPA, (3) DPCPX, (4) 1-aminoindan, (5) 2-aminoin- were achieved with one to four pores. The peptides were dan, (6) CPA + DPCPX, (7) 1-aminoindan + CPA , and added to the cis side of the pore with a positive electrode (8) 2-aminoindan + CPA. Although 1-aminoindan and on the trans-side. The experiments were carried out at Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 5 of 26 Thioflavin‑S 2-aminoindan have very similar structures, they have Thio-S (Sigma-Aldrich, Oakville, Canada) is a fluores - been shown to possess very different properties in vitro cent marker that detects α-syn aggregates and amyloid [21, 36], therefore we suggest they will exhibit differ - plaques. Coronal slices of 40 μm were firstly treated with ent physiological properties in vivo as well. All drugs 0.3% KMnO for 4 min, followed by a 30-min incuba- were dissolved at 3 mg/ml in DMSO, and each drug tion with 1 M phosphate buffered saline at 4 °C. These was administered to the animals by daily i.p. injections slices were then stained with 0.05% Thio-S in 50% etha - (3 mg/kg body weight) for 7 consecutive days. After the nol in the dark for 8 min, rinsed with 80% ethanol twice, first injection with DPCPX, 1-aminoindan or 2-aminoin - followed by three rinses with ultra-pure water for 30 s. dan, the animals were returned to their cages for 30 min Finally, the slices were incubated again with 1 M phos- before a subsequent CPA injection was administered. phate buffered saline for 30 min at 4 °C before starting Then on the eighth day following the final injections, the the DAPI stain. The FITC filter (488 nm laser line) was animals were sacrificed and processed for brain immuno - used to image Thio-S using a Zeiss LSM700 confocal histochemistry/confocal imaging or Western blotting as microscope (Carl Zeiss Group, Canada) and images were described below. analyzed with ImageJ (Public Domain). Immunohistochemistry FluoroJ ‑ ade C Anesthetized rats were transcardially perfused with 0.9% FJC is a fluorescent marker for neurodegeneration (Mil - saline, and then fixed with 4% paraformaldehyde. The lipore-Sigma, Oakville, Canada). Coronal slices of 40 μm extracted brains were put in 30% cryoprotected sucrose were mounted on 5% gelatin-coated super-frost plus solution for 48 h prior to slicing. The brains were initially microscope slides (Thermo Fisher Scientific, Waltham, frozen at − 40 °C (BFS-30 mp controllers) and sliced with MA) and dried overnight at 4 °C. Initially, the micro- the help of a microtome (Leica SM2010 R Sliding con- scope slides were immersed in 1% NaOH/80% ethanol for troller). Coronal slices of 40 μm were then washed three 5 min followed by 2-min immersion in 70% ethanol. The times in 0.1 M phosphate buffered saline followed by slides were then rinsed for 2 min with ultra-pure water. 1-h blocking at room temperature with blocking buffer. The microscope slides were further immersed in 0.06% The buffer solution components have been previously KMnO for 10 min, followed by additional rinse for 2 min described [49]. The slices were then incubated over - with ultra-pure water. The slides were then stained with night at 4 °C with the following primary antibodies: 1:200 0.004% FJC in 0.1% acetic acid for 20 min with gentle mouse monoclonal to α-syn (Abcam Inc, Toronto, Can- shaking on an orbital shaker. Lastly, the slides were rinsed ada) and 1:200 rabbit polyclonal to tyrosine hydroxylase three times in ultra-pure water for 1 min each, mak- (TH) (Millipore-Sigma, Oakville, Canada). Subsequently, ing sure to remove all the excess water after each rinse. slices were then incubated for 1 h in the dark at room The slides were then rinsed in xylene and allowed to dry temperature with the following secondary antibodies: overnight at 4 °C. Then they were treated with Prolong AlexaFluor-555-conjugated anti-mouse and AlexaFluor- Gold Antifade Reagent from Invitrogen (Thermo Fisher 647-conjugated anti-rabbit (1:1000) purchased from Inv- Scientific, Waltham, MA), and respective images were itrogen (Thermo Fisher Scientific, Waltham, MA). Slices taken using a Zeiss LSM700 confocal microscope (Carl were then treated with Thio-S (see further details below). Zeiss Group, Canada) and analyzed with ImageJ (Public Lastly, the slices were incubated for 5 min at room tem- Domain). FJC fluorescence was obtained by exciting the perature with DAPI (2 mg/ml) from Invitrogen (Thermo dye with 488 nm laser. Fisher Scientific, Waltham, MA) and images were taken using a Zeiss LSM700 confocal microscope (Carl Zeiss Nigral slice preparation for Western blotting Group, Canada) and analyzed with ImageJ (Public Nigral slices (400 μm) from male Sprague–Dawley rats Domain). Images of the hippocampal CA1 pyramidal were prepared with the help of the vibratome tissue slicer layer and the SN pars compacta were obtained using (Leica VT1200 S). The rat was initially anaesthetized the Zeiss Plan-Apochromat 63X/1.4 oil objective lens with halothane and rapidly decapitated. Once the brain (Carl Zeiss). Images were acquired as Z-stack images of was extracted it was placed immediately in an ice-cold hippocampal or SN regions with 12–13 Z-stack images sucrose dissection medium and oxygenated with 95% taken at 1-µm intervals near the middle of brain slices. oxygen with 5% carbon dioxide. The slices were then Two Z-stack images were taken along the hippocam- equilibrated in the oxygenated artificial cerebrospinal pal CA1 or SN pars compacta region for each slice, and fluid for 1 h. Nigral slices were transferred into homog - immunofluorescence signals were averaged using densi - enization lysis buffer containing 1% NP-40 detergent tometry analysis. and supplemented with protease inhibitors. After tissue Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 6 of 26 homogenization, the protein concentration was meas- MeOH were similar to those found in α-syn alone (i.e., ured with Bradford Assay using the DC protein assay blockade current peaks around − 85 pA). Moreover, the dye (Bio-Rad, Canada). Protein lysates (50 μg/lane) from blockade populations of the translocation and bumping the different treatment groups were separated in 12% peaks in the presence of 10% MeOH were not signifi - SDS-PAGE gels, and transferred to polyvinylidene dif- cantly different from the α-syn alone control recordings luoride (PVDF) membranes (Millipore, ThermoFisher (Additional file 1: Table S1); therefore, all subsequent Scientific, USA) using 30 V overnight at 4 °C. The PVDF recordings with different drugs described below were membranes were then treated with mouse monoclonal performed with 10% MeOH in the recording solution. [4D6] anti-α-syn (Abcam Inc, Canada) primary anti- When 1 µM α-syn and 10 μM of the drug (in 10% body overnight at 4 °C after 1-h blocking with 5% non- MeOH) were inserted on the cis side, we observed fat milk in Tris-buffered saline with Tween-20. The changes in the blockade current events (See Additional next day, the membranes were incubated for 1 h with file 1: Fig. S1 for details of current of α-HML pore at the appropriate secondary antibody at room tempera- + 100 mV and the changes in blockade current events ture. The membranes were then finally re-probed with once α-syn and/or drugs such as CPA and DPCPX were chicken polyclonal antibody against Tubulin-III. Proteins added in the cis side of the perfusion cup). First, we were visualized using enhanced chemiluminescence and tested the potential binding of adenosine with α-syn in ChemiDoc (Bio-Rad, Canada). Densitometry analysis our nanopore setup. Adenosine appeared to have weak was performed using ImageJ (Public Domain). All the binding affinity to α-syn. The majority of the events of the above-mentioned solutions and procedures have been −86 pA blockade current was related to α-syn transloca- previously described [49]. tion; however, there were fewer events observed in the translocation peak in the α-syn and adenosine histogram Statistical analyses (52%) compared to α-syn alone (66%) (Fig. 2d). Interest- For nanopore, histological and Western blot analyses, ingly, CPA and DPCPX appeared to bind to α-syn as well statistical analyses were conducted with GraphPad Prism (Fig. 2e, f). For the first time, we observed a decrease 8 software (San Diego, CA) with one-way ANOVA fol- in the blockade current of the translocation peak from lowed by Student-Neuman-Keuls multiple comparison −85 pA for α-syn alone to −89 pA for the α-syn and post-hoc test. The significances are indicated as: ns, non- CPA complex, as the percentage of events decreased significant; *P < 0.05; **P < 0.01; and ***P < 0.001. from 66% to 40% for CPA (Fig. 2e). Taken together, the observed effects of CPA and, to a lesser extent, adeno - Results sine on the α-syn translocation events are clear signs of CPA and DPCPX bind to α‑syn in vitro binding. Conversely, DPCPX caused a small increase of Initially, both CPA and DPCPX were tested by nanopore the blockade current to -84 pA, which was accompanied analysis. Once a stable pore was created, a final concen - by a decrease in the number of events in the transloca- tration of 1 μM α-syn was inserted on the cis side of the tion peak (Fig. 2f ). The blockade times of translocation perfusion cup. At first, a few bumping events at 30 pA and and bumping peaks of α-syn with and without adeno- a higher number of translocation events around 85 pA sine, CPA or DPCPX were also calculated. Representa- were observed, whereas intercalation events were rarely tive exponential time graphs are shown in Fig. 3. The encountered (Fig. 1b2). Thus, these observations confirm times of translocation and bumping events for α-syn similar recordings of stable blockade and bumping cur- alone were well established [33]. The times of transloca - rents of α-syn, as previously reported [33]. Previous work tion events (α-syn alone, 0.52 ms) decreased when α-syn using confocal single-molecule fluorescence techniques was combined with adenosine (0.46 ms), CPA (0.47 ms) indicated the formation of oligomers of α-syn in the pres- or DPCPX (0.42 ms), which further indicates a potential ence of DMSO [50]. These oligomers have a high binding binding of these drugs to the protein. On the other hand, affinity to lipid membranes. Therefore, to avoid potential we observed increased bumping times when α-syn was aggregation of α-syn as well as binding of the oligom- incubated with adenosine, CPA or DPCPX (Fig. 3e–h). ers to the membrane and disruption of the lipid bilayer, A full summary of the blockade populations and times is which could lead to further issues with α-HML assembly shown in Table 1. and conductivity, we decided to use MeOH to dissolve all Caffeine, a nonselective inhibitor of all the adenosine the drugs in our nanopore studies. As shown in Fig. 2b receptors (A1, A2A, A2B and A3) [51], has previously and c, 1% and 10% final concentrations of MeOH did been shown by nanopore analysis to bind to the N- and not significantly change the blockade current histograms C-termini of α-syn, thereby promoting a neuroprotective compared to control (α-syn alone, Fig. 2a). The transloca - loop conformation [37]. It is known that caffeine compet - tion and bumping peaks in the presence of 1% and 10% itively antagonizes adenosine’s effects [52, 53]. Although Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 7 of 26 Fig. 2 Representative blockade current histograms of 1 μM α-synuclein alone (a) and with 1% methanol (b), 10% methanol (c), 10 μM adenosine (d), 10 μM CPA (e) and 10 μM DPCPX (f) at 100 mV DC, indicating binding to the protein. Each experiment was run in triplicates and the standard error of the mean estimated for the percentage of events was < ± 10% (see Table 1 and Additional file 1: Table S1) Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 8 of 26 Fig. 3 Representative blockade time profiles of translocation (a–d) and bumping events (e–h) for 1 μM α-synuclein alone and in the presence of 10 μM adenosine, 10 μM CPA or 10 μM DPCPX. Each experiment was run in triplicates. For mean and SEM values of populations of translocation and bumping and blockade times in the absence or presence of adenosine, CPA or DPCPX, please see Table 1 Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 9 of 26 Table 1 Populations and blockade times of translocations and bumping events for α-syn alone and α-syn complexes with adenosine, CPA, and DPCPX Protein‑ drug complex α‑syn α‑syn + Adenosine α‑syn + CPA α‑syn + DPCPX Population of translocation 66% 52% [*] 40% [**] 37% [**] SEM 1% 2% 1% 1% Population of bumping 24% 25% [ns] 37% [**] 46% [**] SEM 1% 3% 2% 2% Time of translocation 0.52 ms 0.46 ms [ns] 0.47 ms [ns] 0.42 ms [ns] SEM 0.05 ms 0.05 ms 0.05 ms 0.04 ms Time of bumping 0.05 ms 0.12 ms [***] 0.09 ms [**] 0.07 ms [*] SEM < 0.01 ms 0.01 ms < 0.01 ms < 0.01 ms ns, non-significant *P < 0.05, **P < 0.01, and ***P < 0.001 vs α-syn alone (one-way ANOVA, followed by Student–Newman–Keuls multiple comparison test) the blockade populations of adenosine and CPA showed or DPCPX (Fig. 4c, f, i). With the addition of CPA, the some similarities to the histogram of caffeine binding N-terminus proportion of bumping events decreased sig- to α-syn [37], both adenosine and CPA decreased the nificantly from 70% to 46% (Fig. 4a vs b). Conversely, the blockade current of the translocation events of α-syn widespread block of events between − 50 and − 100 pA to − 86 and − 89 pA, respectively (Fig. 2d, e). Previous in the N-terminus control developed into a well-defined results using 5 μM caffeine showed that the translocation broad translocation peak at − 26 pA with a population population decreased to 44% from 81% of α-syn alone, of 35%. The changes observed for ΔNAC after addition whereas the bumping population significantly increased of CPA were remarkable and demonstrated clear signs to 38% compared to 9% of α-syn alone [37]. Similarly, of binding (Fig. 4d vs e). The broad translocation peak here DPCPX + α-syn decreased the translocation popula- at − 86 pA was reduced into a small cluster of events, tion (37%) and increased the bumping population (46%) whereas the small bumping peak significantly increased (Fig. 2f ), suggesting that like caffeine, DPCPX may poten - in population from 19% to 66% and shifted to − 36 pA tially bind to the N- and C-termini of α-syn, forming a from − 27 pA. Interestingly, the C-terminal domain his- loop conformation. togram profiles in the absence and the presence of CPA did not show significant differences (Fig. 4g vs h), which Alpha‑synuclein domain investigations of CPA and DPCPX indicates that CPA does not interact with the C-terminus. To further probe the exact binding of both CPA and In contrast, DPCPX produced different profile histo - DPCPX, separate domains of α-syn, namely the N- and grams of each of the α-syn domains when compared to C-termini, and the ΔNAC construct, i.e., α-syn with dele- both the control and CPA. The N-terminus histogram tion of the non-amyloid β-component region (Fig. 1d, e), profile showed that DPCPX caused a decrease in the pro - were tested against CPA or DPCPX. The behaviour of portion of bumping events from 70% to 41%, and DPCPX each domain was different in a standard nanopore analy - also revealed two additional peaks, namely the translo- sis at a direct current voltage of 100 mV (Fig. 4a, d, g). The cation peak at − 72 pA and an intercalation peak at − 51 blockade current histogram of the N-terminus had a sin- pA (Fig. 4a vs c). The intercalation peak at − 51 pA had a gle Gaussian peak at − 30 pA due to bumping events. The low population of events (21%) whereas the translocation N-terminus is positively charged (+ 4). Consequently, it peak was broader and had a similar proportion to the will be difficult for this N-terminal fragment to translo - translocation peak of CPA. The ΔNAC translocation peak cate through the pore under the applied positive trans- disappeared with the addition of DPCPX; instead, two membrane voltage. Conversely, the C-terminus contains peaks with similar proportion of events were observed a total of 12 negative charges, which permits transloca- at − 24 and − 39 pA, representing the bumping and tion through the pore. The blockade current histogram intercalation peaks, respectively (Fig. 4d vs f ). Lastly, the had a large and wide translocation peak at − 69 pA and C-terminus histogram of DPCPX indicated a decrease of a fairly small bumping peak at − 30 pA. The ΔNAC had the bumping events from 20% to 7%, an emergence of an two peaks, a large peak at − 86 pA due to translocation intercalation peak at − 31 pA, and a significant decrease and a smaller one at − 27 pA due to bumping. of the translocation peak from 77% to 38% (Fig. 4g vs i). Figure 4 shows the blockade current histograms of For convenience, all the blockade intensities and popula- each α-syn domain in the presence of CPA (Fig. 4b, e, h) tions events are shown in Table 2. Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 10 of 26 Fig. 4 Representative blockade current histograms of 10 μM CPA and 10 μM DPCPX with N-terminus (a–c), ΔNAC (d–f) and C-terminus (g–i) of α-synuclein at 100 mV DC. Each experiment was run in triplicates and the error estimated for the percentage of events was < ± 10% (see Table 2) The changes in the shape of histograms (Figs. 2 and not infer what the structure was (Fig. 2d–f). In other 4) upon addition of a drug demonstrated drug binding. cases, there were many intermediate events that suggest Further, the increase in bumping events and decrease in the presence of many different structures (Fig. 4a, d–f ). translocation events demonstrate that the drug caused However, there may be other drug/α-syn interactions protein folding. Nanopore analysis showed that the drug that are transient or dependent on initial drug concentra- binding resulted in either a “knot” or a “loop” α-syn tions, which could alter the proportions of translocation, conformation (Fig. 5). Interestingly, we observed inter- bumping and intermediate events as previously reported mediate gaussian peaks suggestive of the presence of a for the drug Rasagiline [36]. Since α-syn is an intrinsi- particular partially folded structure though we could cally disordered protein, it is assumed to have an infinite Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 11 of 26 Table 2 Summary of the intensity and the population of current blockades for all the domains in the absence or presence of CPA or DPCPX Domain‑ drug complex I P I P I P Trans Trans Inter Inter Bump Bump N-term – – – – − 30 ± 1 pA 70% ± 2% N-term + CPA −66 ± 6 pA 35% ± 3% – – − 26 ± 1 pA 46% ± 2% [***] N-term + DPCPX −72 ± 7 pA 36% ± 10% − 51 ± 2 pA 21% ± 12% − 30 ± 2 pA 41% ± 1% [***] C-term −69 ± 2 pA 77% ± 4% – – − 30 ± 2 pA 20% ± 3% C-term + CPA −66 ± 3 pA 70% ± 8% – – − 38 ± 4 pA 23% ± 3% [ns] [ns] C-term + DPCPX −64 ± 1 pA 38% ± 2% − 31 ± 1 pA 33% ± 4% − 19 ± 1 pA 7% ± 1% [***] [ns] ΔNAC −86 ± 2 pA 58% ± 2% – – − 27 ± 1 pA 19% ± 4% ΔNAC + CPA – – − 36 ± 3 pA 66% ± 3% – – ΔNAC + DPCPX – – − 39 ± 1 pA 35% ± 5% − 24 ± 2 pA 34% ± 3% Mean ± SEM. I and P represent intensity and population of the current blockade, respectively ns, non-significant P > 0.05, *P < 0.05, **P < 0.01, and ***P < 0.001 vs N-terminus or C-terminus alone (one-way ANOVA, followed by Student–Newman–Keuls multiple comparison test) number of conformations. Therefore, it is possible that that could be toxic to neurons; C4 has low propensity for some of these α-syn structures (bound or unbound by α-helical structure like cluster C3 and, hence, is unlikely ligands) could minimally be present in our nanopore to be membrane-bound; some C5 structures interact recordings, and consequently had non-negligible contri- with membranes and might be important for synap- butions to the current histograms that were not covered tic functions, while other C5 structures form tetramers by the gaussian fitting in Fig. 2a, b, d–f and Fig. 4a, b, d–f. in vivo, which are believed to promote protection against Taken together, the altered translocation and bumping neurodegenerative disorders; C6 N-terminal residues (both blockade current peaks and populations of events) adopt an α-helical structure that targets and anchors and the appearance of intercalation events in the N-ter- α-syn to membrane of synaptic vesicles; C7 is similar to minus, ΔNAC, and C-terminus domains of α-syn indi- C1, C3 and C4, having low propensity for α-helical for- cate that DPCPX binds to both the N- and C-termini of mation (i.e., higher β-strand propensity), hence, α-syn α-syn. monomers are likely in aqueous solution; and C8 struc- ture plays a role in fibril formation and has high α-helix Molecular docking simulations reveal interactions propensity like C2, C5 and C6 [47]. of DPCPX and 1‑aminoindan with N‑ and C‑termini of α‑syn Four of the eight structures of α-syn were selected to To confirm the biophysical results from nanopore analy - study the α-syn–DPCPX drug complex: C2, C5, C7 and sis, we further characterized the α-syn–drug complexes C8 [47]. For the C2 structure of α-syn, DPCPX showed by performing molecular docking studies of the three hydrophobic interaction with the N-terminus of α-syn, A1R ligands as well as 1- and 2-aminoindan with α-syn. specifically the negatively charged glutamic acid 20 Conformational ensemble of α-syn in solution as deter- (E20) and positively charged lysine 21 (K21), and with mined by discrete molecular dynamics simulations and the very end of the C-terminus (amino acids E139 and further confirmed by far-UV circular dichroism and alanine 140 (A140)) (Fig. 6a). However, for the C5 struc- cross-linking mass spectrometry, has recently revealed ture (Fig. 6b), DPCPX was shown to form hydrophobic stable monomeric α-syn clusters of structure [47] (C1- bonds with the N-terminus (amino acids phenylalanine 4 C8, see Additional file 1: Appendix 1). We used these (F4) and E20) and hydrogen bonds with K60, and hydro- structures in the molecular docking simulations to deter- phobic bonds with the NAC region (amino acids E61 mine if any of these structures correlates with the drug/ and F94). DPCPX also formed hydrogen bonds with the α-syn complex as predicted from the nanopore analysis. NAC region (amino acids glutamine 62 (Q62) and valine The 8 clusters of structural subpopulations have the fol - 63 (V63)). Similar as the interactions with C2, DPCPX lowing features: C1 structure forms a dimer and plays binding to the N-terminus (hydrogen bond with E28, a role in fibril formation; C2 and C3 are precursors for with additional hydrophobic interactions with V15 and oligomer formation; C3 has antiparallel β-sheets in the V40) and the C-terminus (hydrogen bond with tyrosine aggregate-prone NAC segment and forms oligomers 136 (Y136) with additional hydrophobic interactions with Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 12 of 26 Fig. 5 Eec ff ts of adenosine A1 receptor (A1R) ligands on α-synuclein (α-syn) expression and folding patterns in in vivo and in vitro studies. a A1R agonist CPA (and adenosine) increases α-syn expression and aggregation in the rat substantia nigra. Nanopore analysis and molecular docking simulations predicted binding of A1R agonist CPA (and adenosine) to the N-terminus of α-syn, leaving the NAC domain intact and able to promote aggregation. b (b1) Adenosine, CPA and 2-aminoindan bind to and stabilize α-syn to adopt a “knot” conformation which has been shown to induce aggregation and neurodegeneration. In contrast (b2), DPCPX and 1-aminoindan bind to both the N- and C-termini of α-syn, which does not promote aggregation and neurodegeneration. Created using BioRender.com Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 13 of 26 Fig. 6 Molecular docking simulation of α-synuclein (α-syn) structures C2 (a), C5 (b), C7 (c), and C8 (d) bound to DPCPX. Below full 3D representations show magnified binding pocket of α-syn and the locations of amino acid residues responsible for each drug binding. Bold black dashed lines and amino acid residues indicate hydrogen bonding, while the grey dashed lines and amino acid residues indicate hydrophobic interactions. Hydrogen bonding of DPCPX with both the N- and C-terminal amino acid residues is observed in C7 α-syn structure (c). DPCPX also forms hydrogen bond with either the N-terminal (C5 α-syn structure in b) or C-terminal amino acids (C8 α-syn structure in d) and also hydrophobic bonds with portions of the NAC region. N-and C-terminal binding of DPCPX is also observed without hydrogen bonding (C2 α-syn structure in a). The molecular docking study was carried out using Autodock Vina module implemented in PyRx tool. Protein and ligand interactions were analyzed and visualized through Pymol and LigPlot + aspartic acid 135 (D135) and E137) of the C7 structure that other α-syn structures can bind DPCPX via hydro- was revealed (Fig. 6c). Furthermore, DPCPX was shown gen bonding with the N-terminus (H50 and K12 of C3 to bind to the C8 structure in the N-terminus (amino structure) or C-terminus (E137 of C1 structure) and acids K32, glycine 36 (G36), V37, threonine 44 (T44) and hydrophobic interactions with the NAC region (Addi- V48) and the C-terminus with a hydrogen bond at the tional file 1: Fig. S2). E139 and additional hydrophobic interactions at G106, Like DPCPX, four α-syn structures, C1, C4, C5 and A107 and proline 108 (P108). For all the structures C2, C8, were selected to analyze the binding mode of C5, C7 and C8, DPCPX was revealed to reside within a 1-aminoindan to α-syn. For the C1 structure, 1-ami- closed globular conformation and interacts with both the noindan was shown to form hydrogen bond with E104 N- and C-termini of α-syn. However, it is also possible of the C-terminus as well as hydrophobic interactions Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 14 of 26 Fig. 7 Molecular docking simulation of α-synuclein (α-syn) structures C1 (a), C4 (b), C5 (c) and C8 (d) bound to 1-aminoindan. Below full 3D representations show the magnified binding domains of α-syn and the amino acid residues in both the N- and C- termini of α-syn that facilitate drug binding. Bold black dashed lines and amino acid residues indicate hydrogen bonding, whereas the grey dashed lines indicate hydrophobic interactions. Hydrogen bonding of 1-aminoindan with C-terminal amino acid residues is observed in C1, C4, C5 and C8 α-syn structures. In addition, hydrophobic interactions occur between 1-aminoindan and N-terminal amino acid residues (C1 and C4 α-syn structures) and also between 1-aminoindan and portions of the NAC region (C5 and C8 α-syn structures). The molecular docking study was carried out using Autodock Vina module implemented in PyRx tool. Protein and ligand interactions were analyzed and visualized through Pymol and LigPlot + with the aromatic positively charged cleft of the However, the C5 and C8 structures were shown to N-terminus (amino acids G31, K32, V37, and Y39) form different drug-protein complexes that may reg- (Fig. 7a). For the C4 structure (Fig. 7b), 1-aminoin- ulate tetramer formation (C5) or fibrillation (C8). dan was shown to form hydrogen bonds with the polar 1-Aminoindan appeared to bind only to the C-termi- serine 129 (S129) and hydrophobic interactions with nal polar cleft between T92, Q99 and Q134 of the C5 E131 and Y136 of the C-terminus, and interact with structure (Fig. 7c). Also, 1-aminoindan was shown to the N-terminus of C4 structure, forming hydrophobic form further hydrophobic bonds with G101 and P128 bonds with A18. These indicate the formation of a loop and a hydrogen bond with L100. Lastly, 1-aminoindan conformation between the N- and C-termini of the had hydrophobic interactions with the NAC region protein when 1-aminoindan binds to the C4 structure. of the C8 structure (amino acids V71, T72, A76, and Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 15 of 26 Fig. 8 Molecular docking simulation of α-synuclein (α-syn) structures C4 (a) and C5 (b) bound to adenosine. Below full 3D representations show the magnified binding pocket of α-syn and the amino acid residue locations responsible for each drug binding. Bold black dashed lines and amino acid residues indicate hydrogen bonding, whereas the grey dashed lines and amino acid residues indicate hydrophobic interactions. Adenosine only formed hydrogen bonds and hydrophobic interactions with N-terminal amino acid residues in C5 α-syn structure. In addition, adenosine also formed hydrogen bonds with amino acid residues within the N-terminus and NAC region in C4 α-syn structure. The molecular docking study was carried out using Autodock Vina module implemented in PyRx tool. Protein and ligand interactions were analyzed and visualized through Pymol and LigPlot + V77), and in the C-terminus (amino acids methionine mainly bound to the N-terminus of α-syn (blue alpha- 116 (M116), P117 and V118) (Fig. 7d). 1-Aminoindan helix region) (Fig. 8). For the C4 structure, adenosine also formed a hydrogen bond with D119 in the C-ter- formed hydrogen bonds with V15 and K21 in the posi- minus of the structure. tively charged cleft in the C4 N-terminus and with G68, These results indicated that the A1R antagonist A78 and Q79 in the C4 NAC region. Other hydrophobic DPCPX and Rasagiline metabolite 1-aminoindan pos- interactions were revealed at G14 and A17 in the N-ter- sess similar binding interactions with α-syn. As the minus and at T72, G73 and V74 in the NAC region. For full crystal structure of α-syn is not yet available, we the C5 structure, adenosine bound only to the N-termi- suggest that using the four conformations of α-syn in nus, forming hydrogen bonds with the polar negatively our molecular docking studies could provide more charged cleft of the N-terminus (amino acids E20, Q24, complete information on the binding interactions of and A53) and hydrophobic interactions with G25 and DPCPX and 1-aminoindan with α-syn. K60. CPA is a chemical derivative of adenosine that shows Molecular docking simulations reveal that CPA, adenosine greater selectivity as an A1R agonist, and thus is expected and 2‑aminoindan bind to the N‑terminus of α‑syn to interact with α-syn similarly as adenosine. Similar to Based on the results of nanopore analysis, molecular adenosine, CPA formed hydrogen bonds with G47 of the docking simulation of α-syn binding to adenosine, A1R N-terminus and G68 of the NAC region of the C2 α-syn agonist CPA, and methamphetamine analog 2-ami- structure (Fig. 9a). CPA also had various hydrophobic noindan was conducted. Results showed that adenosine interactions in the N-terminus (amino acids V26, G31, (See figure on next page.) Fig. 9 Molecular docking simulation of α-synuclein (α-syn) structures C2 (a), C5 (b), and C8 (c) bound to CPA; C2 (d), C5 (e), and C8 (f) bound to 2-aminoindan. Below full 3D representations show the magnified binding pocket of α-syn and the amino acid residue locations responsible for each drug binding. Bold black dashed lines and amino acid residues indicate hydrogen bonding, whereas the grey dashed lines and amino acid residues indicate hydrophobic interactions. Both CPA and 2-aminoindan formed hydrogen bonds and hydrophobic interactions with the N-terminal amino acids only (C5 and C8 α-syn structures) (b-c and e–f, respectively). CPA also forms hydrogen bond and hydrophobic interactions with amino acids within the N-terminal and the NAC region (C2 α-syn structure) (a). In contrast, 2-aminoindan only forms hydrophobic interactions with the N-terminus and NAC domain in C2 α-syn structure (d). The molecular docking study was carried out using Autodock Vina module implemented in PyRx tool. Protein and ligand interactions were analyzed and visualized through Pymol and LigPlot + Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 16 of 26 Fig. 9 (See legend on previous page.) Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 17 of 26 E35, K43, V48, and K58) and in the NAC region (amino K34 (inosine, guanosine, thymidine, and uridine) and acids Q62, V63, G67, and V71). For the C5 structure, CPA with the negatively charged E20 (guanosine, cytidine, formed a hydrogen bond with the negatively charged E20, and uridine) (Additional file 1: Fig. S3). Inosine had simi- as well as the positively charged K60, Q24 and A53 in the lar hydrophobic interactions with Q24, G25 and A53 as N-terminus (Fig. 9b). It also formed hydrophobic interac- adenosine, and formed additional hydrophobic bonds tions with G25, K34 and V49. For the C8 structure, CPA with C5 α-syn conformation at H50 and E57. Inosine interacted with similar negatively charged N-terminal formed hydrogen bonds with V49 and T54 in the N-ter- cleft containing E28 and A29 (Fig. 9c), and also formed minus as well. Guanosine formed hydrogen bonds with various hydrophobic bonds in the N-terminus (amino the polar cleft of Q24 and A53 of the N-terminus same acids L8, K10, E35, L38, Y39, and V40). as adenosine (Additional file 1: Fig. S3, Fig. 8b). Con- Similar to CPA, 2-aminoindan was also shown to have versely, cytidine (Additional file 1: Fig. S3c) interacted various hydrophobic interactions with the N-terminus with the same amino acids, but through hydrophobic (amino acids V26, E35, K43, G47, and V48) and the NAC interactions (amino acids A19, Q24, G25, A53 and E57). region (amino acids V63, and V71) of the C2 structure Moreover, thymidine and uridine shared the same hydro- of α-syn (Fig. 9d). Moreover, 2-aminoindan also formed phobic interactions with the C5 α-syn structure (Addi- a hydrogen bond with E57 and hydrophobic interac- tional file 1: Fig. S3d, c), and both bound to G25, A53 and tions with other residues within the N-terminus of the E57. Uridine also formed other hydrophobic bonds with C5 structure (amino acids G25, V26, A30, G31, K34, V49, A19 and Q24, like cytidine (Additional file 1: Fig. S3c, e). and A53) (Fig. 9e). Similar to CPA, 2-aminoindan formed hydrogen bonds with A29 and E35 and hydrophobic A1R agonist CPA and drugs that bind to α‑syn N‑terminus interactions with other residues inside the N-terminus of increase α‑syn expression and aggregation in SN the C8 structure (amino acids L8, Serine 9 (S9), K10, and and hippocampal neurons T22) (Fig. 9f ). To investigate whether drug binding to the N- and/or Taken together, the molecular docking studies con- C-terminus of α-syn can affect the levels of α-syn expres - firmed the results of nanopore analyses, that adeno - sion and aggregation in vivo, we administered the drugs sine, CPA and 2-aminoindan mainly interacted with the individually or in combination with the A1R agonist CPA. N-terminus of α-syn (C5 cluster); however, other sub- Representative images of the SN pars compacta region populations of α-syn clusters showed hydrogen bonding labelled with DAPI, TH, and α-syn showed that α-syn of CPA and 2-aminoindan with other N-terminal resi- was localized in the soma (cytosol, nuclei) and presum- dues only (C8) or hydrogen bonding of adenosine with ably the dendrites of dopaminergic neurons (Fig. 10b). the N-terminal and proximal NAC amino acid residues Interestingly, α-syn expression was increased by CPA (C4). The interactions of the drugs with α-syn N-termi - in the absence or presence of DPCPX, 1-aminoindan, nus are expected to promote α-syn aggregation. In con- or 2-aminoindan (Fig. 10b, and 11a, b) compared to the trast, DPCPX and 1-aminoindan showed binding to both control (0.1% DMSO in 0.9% saline). The 2-aminoin - N- and C-terminal regions of α-syn, which may promote dan + CPA treatment induced the highest level of α-syn an α-syn conformation that prevents α-syn aggregation. protein in the SN pars compacta (Fig. 11a, b). In contrast, DPCPX, 1-aminoindan or 2-aminoindan alone did not Comparison of α‑syn binding of adenosine with that of significantly increase the α-syn protein level (Fig. 11a, b). other standard nucleosides using molecular docking To determine whether these changes in α-syn protein simulation level correlated with the level of α-syn aggregation, the Adenosine is not only an endogenous agonist of all the SN pars compacta was co-labelled with α-syn marker purinergic G protein-coupled receptors, but also a purine and Thio-S. As shown in Fig. 11a-c, treatments with ribonucleoside [55]. To determine whether adenosine CPA, DPCPX + CPA , 1-aminoindan + CPA, and 2-ami- binds specifically to α-syn, we performed additional noindan + CPA all increased the Thio-S level. Treat - molecular docking simulations to compare α-syn binding ment with 2-aminoindan alone also enhanced Thio-S of adenosine with that of the four other standard nucleo- labelling, and co-administration of 2-aminoindan with sides (guanosine, cytidine, thymidine, and uridine) and CPA caused a significant further elevation of Thio-S the adenosine metabolite inosine, using the C5 α-syn compared to 2-aminoindan alone (Fig. 11b). In contrast, structure (Additional file 1: Fig. S3). treatments with DPCPX or 1-aminoindan alone did not All the purine nucleosides were shown to bind to a significantly increase Thio-S labelling or attenuate the similar hydrophobic pocket in the N-terminus of the C5 CPA-induced increase in Thio-S level. The colocalization α-syn conformation. Similar to adenosine, almost all of of Thio-S signal with α-syn was increased in all the treat - them formed hydrogen bonds with the positively charged ments compared to control (about a two-fold increase Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 18 of 26 in Pearson correlation coefficients, Fig. 11c). In conclu- 2-aminoindan or 1-aminoindan treatment alone group, sion, our results showed that 7-day systemic administra- compared to CPA, CPA + DPCPX, or control treatment tion of CPA, alone or in combination with 2-aminoindan, (P < 0.001). increased expression and aggregation of α-syn in the SN Moreover, when the 15 kDa, 30 kDa and 75 kDa α-syn pars compacta (Figs. 10 and 11). densitometry values were added, we still detected signifi - A1R is widely distributed in other regions of the brain cantly higher levels of total α-syn in the CPA (P < 0.01), including the hippocampus. Therefore, the CA1 region of DPCPX + CPA (P < 0.033), and 2-aminoindan + CPA the hippocampus was also analyzed for α-syn expression (P < 0.001) groups (Fig. 10c). It is noteworthy that signifi - and aggregation. Similar to the nigral tissue, treatments cant accumulation of α-syn in the SN lysate was associ- with CPA and 2-aminoindan + CPA increased the levels ated with treatments with compounds that were found to of both α-syn and Thio-S (Additional file 1: Fig. S4a). In bind only to the N-terminus of α-syn (i.e., the A1R ago- particular, CPA and 2-aminoindan + CPA induced a four- nist CPA and 2-aminoindan); moreover, this elevation fold increase of α-syn compared to the control (Addi- could be attenuated by co-treatments with compounds tional file 1: Fig. S4b). This increase was also observed for that were found to bind to both the N- and C-termini colocalization of Thio-S with α-syn (Additional file 1: Fig. of α-syn (e.g., DPCPX, 1-aminoindan). In addition, the S4c). However, co-administration of CPA with DPCPX observed higher molecular weight band at 75 kDa likely or 1-aminoindan caused significant attenuation of α-syn indicates the presence of C5 α-syn structures that differ - accumulation compared to CPA treatment (Additional entially bind to DPCPX (Fig. 6b), 1-aminoindan (Fig. 7c), file 1: Fig. S4b, c). and 2-aminoindan (Fig. 9e), since the C5 structure of α-syn is known to be a precursor for the formation of Western blotting analysis of α‑syn in SN tetramers [47]. Treatment with the A1R agonist CPA and 2-aminoin- dan + CPA induced a significant 1.5-fold increase in the CPA and 2‑aminoindan increase neurodegeneration of SN level of monomeric α-syn (15 kDa band) in SN; however, pars compacta dopaminergic neurons and hippocampal the 1-aminoindan treatment group showed an approxi- pyramidal neurons mately three-fold decrease in α-syn compared to the Having shown that CPA and 2-aminoindan alone or in DMSO/saline control group (Fig. 10c). The CPA-induced combination can increase α-syn aggregation, we then increase in α-syn was partially attenuated by co-adminis- determined whether these treatments could lead to neu- tration with DPCPX and was fully restored to control lev- ronal damage. We used FJC as a common fluorescent els by 1-aminoindan co-treatment. Prominent signals for marker for neurodegeneration in the CNS [31]. FJC stain- 30 kDa α-syn were shown in the SN immunoblots, but ing was performed in nigral slices −5.30 to −5.60 mm not detectable in the hippocampal lysate immunoblots from the bregma as well as hippocampal slices −3.80 to (data not shown). In contrast to the monomeric 15 kDa −4.16 mm from the bregma. Representative high-mag- α-syn band, the 30 kDa α-syn signal was not significantly nification images of FJC staining in the pars compacta altered by CPA, DPCPX + CPA , 2-aminoindan + CPA , or region of SN indicate that CPA alone, 2-aminoindan 2-aminoindan treatment, compared to control (DMSO/ alone, and 2-aminoindan + CPA co-administration all saline) or naive group; however, the 30 kDa α-syn sig- increased the level of FJC fluorescence (Fig. 12a). In nal was significantly decreased by DPCPX (P < 0.01), contrast, DPCPX or 1-aminoindan alone did not signifi - 1-aminoindan + CPA (P < 0.01), and 1-aminoindan cantly increase FJC staining, but both were effective in alone (P < 0.001) treatments. Interestingly, the level of attenuating the CPA-induced increase in neurodegenera- 75 kDa α-syn was significantly higher in the DPCPX, tion (Fig. 12a). Similar results were observed in the CA1 region of the hippocampus, except that 2-aminoindan (See figure on next page.) Fig. 10 Summary of the surface area analysis of the pars compacta region of the substantia nigra for DAPI, tyrosine hydroxylase ( TH), and α-synuclein (α-syn). (a) Image of a 40-μm nigral brain slice in the DMSO/Saline control group, with 3,3’-diaminobenzidine (DAB) and TH staining at 4 × magnification with a light microscope. (b) Representative images of DAPI (Blue), TH (Green, Alexa Fluor 555), and α-syn (Red, Alexa Fluor 647) staining in the substantia nigra pars compacta of rats with 7-day chronic intraperitoneal injections of the following agents: Control (DMSO/ Saline), CPA, DPCPX + CPA, 1-aminoindan + CPA, and 2-aminoindan + CPA. CPA with or without 2-aminoindan increased α-syn immunofluorescence compared to control. The CPA-induced increase in α-syn was attenuated by DPCPX or 1-aminoindan. Scale bar, 20 μm. (c) Western blots from total lysates of the substantia nigra and quantification of α-syn level in the substantia nigra. CPA increased the level of 15 kDa α-syn monomers, which was attenuated by DPCPX and 1-aminoindan but not by 2-aminoindan. DPCPX and 1-aminoindan alone significantly reduced the level of 30 kDa α-syn dimers. In contrast, DPCPX, 2-aminoindan, and 1-aminoindan alone significantly increased the 75 kDa α-syn, which likely represent the α-syn tetramers. All values were normalized to β-tubulin III. n = 4 animals in each treatment group. Mean ± SEM. ns, non-significant; *P < 0.05; **P < 0.01; and ***P < 0.001 (one-way ANOVA followed by Student–Newman–Keuls post-hoc multiple comparison test) Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 19 of 26 Fig. 10 (See legend on previous page.) Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 20 of 26 Fig. 11 Summary of surface area analysis of α-synuclein (α-syn) and Thioflavin S in the substantia nigra (SN) pars compacta region. (a) Confocal microscopic images of DAPI, α-syn and Thioflavin S staining in 40-μm nigral brain slices of rats with the following treatments: Control (DMSO/ Saline), CPA, DPCPX, 1-aminoindan, 2-aminoindan, DPCPX + CPA, 1-aminoindan + CPA, and 2-aminoindan + CPA. Scale bar, 50 μm. (b) The mean area intensities of α-syn and Thioflavin S in the SN pars compacta. The fluorescence intensity was quantified in a 100 × 100 μm region and normalized by subtracting the fluorescence intensity in a 50 × 50 μm background non-cell body bottom area. CPA increased the levels of α-syn and aggregated α-syn, and these levels were further enhanced by co-treatments with 2-aminoindan. (c) Pearson correlation coefficient of α-syn and Thioflavin S in the SN pars compacta with CPA. Average intensity values and correlation coefficients in bars represent mean ± SEM from n = 4 independent experiments. ns, non-significant; *P < 0.05; **P < 0.01; and ***P < 0.001 (one-way ANOVA followed by Student–Newman–Keuls post-hoc multiple comparison test) Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 21 of 26 Fig. 12 Fluoro-Jade C (FJC) staining in the SN pars compacta (a) and CA1 of hippocampus (b) of rats with 7-day chronic intraperitoneal injection of Control (DMSO/saline), CPA, DPCPX, 1-aminoindan, 2-aminoindan, DPCPX + CPA, 1-aminoindan + CPA, and 2-aminoindan + CPA. Scale bar 50 μm. Summary bar graphs show significant increases in the relative fluorescence intensity of FJC staining in pars compacta after CPA, 2-aminoindan, and CPA + 2-aminonindan treatments (a). In contrast, only CPA and CPA + 2-aminoindan treatments significantly increased FJC fluorescence in the CA1 hippocampal neurons (b). FJC fluorescence intensity in a 100 × 100 μm region was normalized to the control group (100%). Values are shown as mean ± SEM. The average FJC fluorescence values were obtained from n = 4 independent experiments. ns, non-significant; *P < 0.05; **P < 0.01; and ***P < 0.001 (one-way ANOVA followed by Student–Newman–Keuls post-hoc multiple comparison test) alone did not cause significant neurodegeneration 1-aminoindan with CPA prevented neurodegeneration in (Fig. 12b). CPA and 2-aminoindan + CPA treatments hippocampal slices. demonstrated much higher levels of degenerating pyram- Taken together with the above results from nanopore idal neurons compared to the other treatments. Similar analysis, molecular docking and Thio-S labelling, these to the nigral slices, the co-administration of DPCPX or results suggest that compounds that bind to both N- and C-termini of α-syn (e.g., DPCPX and 1-aminoindan) Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 22 of 26 may be effective in attenuating the neurotoxic effects and thus could contribute to α-syn protein misfolding of compounds that bind to α-syn and promote α-syn and the development of α-synucleinopathy in PD. We accumulation and misfolding (e.g., CPA, adenosine and observed that CPA interacted not only with the α-syn 2-aminoindan). N-terminus but also with α-syn lacking the NAC region (ΔNAC). However, the A1R antagonist DPCPX sig- nificantly altered the population histograms from the Discussion nanopore analysis when the N-terminus, C-terminus One of the most commonly used treatments for PD or the ΔNAC regions were studied, indicating DPCPX patients is Levodopa (a dopamine precursor). However, binding to all these α-syn polypeptide domains. Based this treatment is associated with numerous adverse on the nanopore analysis, we propose that the two effects such as early loss of voluntary movement, severe A1R ligands CPA and DPCPX have different binding dyskinesia episodes and most predominantly end-of-dose interactions with α-syn. Therefore, as we previously worsening [56, 57]. Previous studies on dopamine recep- suggested for 2-aminoindan [20, 21], CPA could bind tor agonists, catechol-O-methyltransferase antagonists, to the N-terminus of α-syn similar to 2-aminoindan, monoamine oxidase inhibitors, and antagonists of dopa- thereby leaving the NAC domain free to misfold and mine transporters have shown promising results in slow- cause a higher chance of protein aggregation. In con- ing the progression and alleviating the symptoms of the trast, DPCPX, similar to 1-aminoindan, metformin and disease [58–61]. However, these therapies are unable to caffeine, appeared to bind to both the N- and C-termini prevent PD progression without causing other significant of α-syn, creating a “loop” conformation of the protein side effects including an increased risk of cardiac-valve (Fig. 5b2) [20, 21]. This loop conformation has been regurgitation, hypertension, confusion and hallucinations suggested to prevent the NAC domain from misfolding [62]. and hence promote neuroprotection by stopping aggre- Adenosine binds to its inhibitory A1Rs (coupled to gation of misfolded α-syn protein fibrils and formation G ) and its excitatory A2A receptors (A2AR, coupled αi of Lewy bodies. Therefore, the two possible conforma - to G ), and A2ARs are believed to contribute to the αs tions adopted by α-syn in the presence of CPA (knot pathogenesis of PD, which has prompted the develop- conformation) or DPCPX (loop conformation) (Fig. 5) ment of small molecular agents for potential PD ther- are expected to contribute to increased or decreased apy, including apedenoson, preladenant, regadenoson α-syn aggregation, respectively. Additionally, the block- and SYN-115 [24, 63]. Recently, several studies have ade currents by CPA and DPCPX resembled those by reported that A2AR antagonism produces far better 2-aminoindan and 1-aminoindan, respectively. The results in slowing down the pathology and progres- metabolite of Rasagiline, 1-aminoindan, is an irrevers- sion of PD and improving symptom management [24, ible inhibitor of the monoamine oxidase type B enzyme 64–68]. Unfortunately, the majority of these drugs which is administered in mono- and/or poly-thera- have failed in clinical trials except for istradefylline, peutic route to treat early symptoms of PD as well as which has been pursued as a potential PD drug in cognitive impairments and fatigue [75, 76]. Conversely, Phase III clinical trials in Japan [69, 70] but recently 2-aminoindan, an amphetamine analogue, is shown to been approved by US Food and Drug Administration cause PD-like symptoms in long-term users and addicts as the only non-dopaminergic add-on therapy for the [77]. Recent case reports indicated that amphetamine treatment of so-called “off phenomenon” and motor and methamphetamine users have a three-fold risk of fluctuations of Levodopa therapy in PD [71]. We have developing PD compared with non-users. Interestingly, suggested that a possible cross-talk between A1Rs and this risk is particularly high in women and it manifests A2ARs could contribute to PD and other neurodegen- even at 30 years of age [78]. Although 1-aminoindan erative diseases due to the elevated levels of adenosine and 2-aminoindan have very similar structures, they in the ageing brain, which may increase A2AR activa- have been shown to hold very different blockade cur - tion [24, 72–74]. More recently, we have reported that rent properties [20, 21]. chronic stimulation of A1Rs with the A1R agonist CPA The blockade histograms for 1-aminoindan demon - leads to increased expression of α-syn both in the SN strate that most events are related to translocation, as region of Sprague–Dawley rat brain and in the dopa- observed previously [20, 21]. In the presence of 10 μM minergic MN9D cells [25]. Here, our in vitro findings 1-aminoindan, a broad translocation peak at − 73 pA and suggest that drugs that bind to A1Rs may play a major a small bumping peak at − 34 pA were observed. Nano- role in synucleinopathy, independent of their canoni- pore analysis with the use of different α-syn domains cal function as A1R agonist or antagonist. We show for suggests that 1-aminoindan binds to both the N- and the first time that these adenosine-related compounds C-termini of α-syn, and by doing so, the drug-protein could bind differentially to different regions of α-syn, Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 23 of 26 complex is expected to adopt a “loop” conformation that fide A1R ligands, also bind to α-syn to modulate its mis - promotes neuroprotection by preventing α-syn mis- folding and aggregation patterns. folding. Our results from both nanopore analysis and Previous studies using nanopore analysis and a yeast molecular docking suggest that DPCPX resembles 1-ami- model of PD reported that several compounds that bind noindan in binding pattern and drug-protein conforma- to α-syn, including 1-aminoindan and the dimer com- tion. Conversely, our previous report has also established pounds containing caffeine linked to 1-aminoindan, nico - that the 2-aminoindan histograms have two peaks with tine or metformin, can indeed prevent α-syn aggregation similar proportions of events, namely, the proportion of and promote survival in yeast [21, 36, 54]. Moreover, we translocation events at 48% and bumping events at 40% previously reported that prolonged A1R activation in [20, 21]. However, 2-aminoindan appears to bind with the brain produced by i.p. injections of CPA (3 mg/kg) higher affinity to α-syn due to the high binding constant once daily for three days led to significant neuronal loss 4 −1 5 × 10 M derived from isothermal titration calorim- in rat hippocampus in vivo [24]. More recently, we also etry (ITC), and most importantly, because it binds only to reported that longer-term A1R stimulation with 5 mg/ the N-terminus of the protein. This N-terminal binding kg CPA (i.p. injections daily for 5 weeks) led to increased of 2-aminoindan has been suggested to promote a neu- α-syn expression and accumulation in the SN neurons, rotoxic “knot” conformation that will lead to increased which was associated with motor and cognitive defi - α-syn aggregation. Here, the nanopore analysis using cits in Sprague–Dawley rats [25]. In the present study, full-length and different domains of α-syn combined with we also showed that 7-day co-administration of either molecular docking attempts using aqueous, membrane- DPCPX or 1-aminoindan with CPA prevented the CPA- unbound α-syn structures [47] showed that both adeno- induced α-syn accumulation and aggregation in both SN sine and the selective A1R agonist CPA closely resembled pars compacta and hippocampal CA1 region, and this 2-aminoindan in the binding interactions and drug-pro- coincided with significantly reduced neurodegenera - tein conformations, indicating that these A1R agonists tion of dopaminergic neurons in the SN pars compacta bind prominently to the N-terminus of α-syn. Although and pyramidal neurons in the hippocampus. However, our results indicate a broad agreement between in vitro future studies are required to further differentiate the and in silico techniques, both methods have their own precise roles of A1R ligand stimulation and α-syn–drug inherent limitations. Our nanopore analysis gives an binding to promote or attenuate α-syn aggregation and excellent prediction of the compound binding site, but it subsequent neurodegeneration using rational mutagen- does not provide precise amino acid residue(s) and posi- esis design or using knockdown studies involving A1R or tions of the drug binding. At best, our nanopore analysis α-syn genes. indicated formation of a “knot” or “loop” conformation According to our previous report [25], the neuro- (Fig. 5) when the ligand-interacting residues within the toxic effect of CPA observed in the present study was N-terminus, NAC region, and/or C-terminus of α-syn are likely mediated in part by A1R-induced downstream bound to promote protein folding. On the other hand, for activation of JNK/c-Jun and sortilin-dependent bind- molecular docking, the residues that bind to each ligand ing and accumulation of α-syn in dopaminergic neu- may be dependent on the starting model used; that is, rons. Since we found in the present study that CPA and compared to the aqueous α-syn cluster structures used DPCPX could also bind directly to α-syn, in the future, in the present study, we found distinct ligand-interacting it would be important to test in A1R knockout mice or in amino acid residues that were bound by CPA and DPCPX MN9D dopaminergic neurons with CRISPR/Cas9 gene when the micelle-bound 1XQ8 model of α-syn was used knockdown of A1Rs to determine whether chronic CPA (data not shown) [79]. Moreover, it appears that confor- administration could still induce upregulation and accu- mational selection may result in only one or several of mulation of misfolded α-syn, and also whether DPCPX the conformations used in the present study to be pop- or 1-aminoindan can prevent this CPA-induced neuro- ulated. Admittedly, the lack of site-directed mutational degeneration of dopaminergic neurons in the absence evidence that could help validate the docking results and of functional A1Rs. Since adenosine elevation is widely directly confirm the ligand interactions, is a major weak - known to occur in the ageing brain and adenosine and ness of the present study. Based on the molecular dock- CPA were shown to bind to α-syn N-terminus and cause ing results, future studies using site-directed mutagenesis aggregation, there is a possibility that the increased brain of these residues and retesting with nanopore analysis adenosine may be a risk factor for increased α-syn mis- are needed to understand the conformational changes folding observed in α-synucleinopathy in PD patients. of α-syn upon binding to CPA, DPCPX, and other com- The results also suggest that targeting α-syn with pounds. Taken together, our results show that CPA and adenosine-related compounds (e.g., DPCPX and caf- DPCPX, in addition to their canonical function as bona feine) or compounds with similar binding profiles (e.g., Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 24 of 26 1-aminoindan, metformin, nicotine) may be an attrac- alone and with 10 μM CPA and DPCPX (dissolved in methanol). Fig. S2: tive therapeutic approach to reducing neurodegeneration Molecular docking simulation of α-synuclein structures C1 and C3 bound to DPCPX. Fig. S3: Molecular docking simulation of α-synuclein C5 struc- associated with increased accumulation of adenosine in ture bound to the five ribonucleosides. Fig. S4: Summary of the surface aging-related neurodegenerative diseases. area analysis of the CA1 region of the hippocampus of DAPI, α-synuclein and Thio-S. Table S1: Populations and the blockade times of each of translocations and bumping events for α-synuclein alone and α-synuclein complexes with 1% and 10% methanol. Conclusion Nanopore analysis and molecular docking techniques are Acknowledgements excellent complimentary tools to probe potential protein- This work was supported by funding from Heart and Stroke Foundation of drug complexes. Importantly, the combination of nano- Canada, NSERC Discovery Grant, Saskatchewan Health Research Foundation pore analysis and molecular docking with our in vivo (SHRF), and College of Medicine (University of Saskatchewan) to FSC. JSL was supported by NSERC Discovery Grant. MB received support from NSERC rodent model of α-synucleinopathy in the present study Discovery Grant, Canadian Institutes of Health Research (CIHR), and SHRF. EJ has provided novel insights into the structure and mis- received Doctoral Award from College of Medicine, University of Saskatch- folding pattern of the inherently disordered α-syn pro- ewan. MTM received SHRF Postdoctoral Fellowship and Parkinson Canada Postdoctoral Fellowship. All the biosafety and approvals for the ethical and tein in the presence of adenosine A1R ligands. Here, we humane use of animals were obtained prior to the start of the study. All proce- demonstrated that adenosine and the A1R agonist CPA dures performed in this study were in accordance with the ethical standards bind to the N-terminus of α-syn, similar to 2-aminoin- of the University of Saskatchewan. The study was approved by the Animal Review and Ethics Board (AREB) of the University of Saskatchewan (Animal Use dan, thereby promoting a more compact, neurotoxic Protocol # 20070090). knot conformation that leads to increased α-syn aggrega- tion and neurodegeneration in the SN and hippocampus Authors’ contributions E.J. performed and analyzed the nanopore assay, animal chronic injections (Fig. 5). In contrast, the A1R antagonist DPCPX binds and postmortem histological and biochemical experiments. M.T.M. carried out to both the N- and C-termini of α-syn, similar to 1-ami- the structural modeling analysis. E.J. and F.S.C. wrote the manuscript; M.T.M., noindan, causing the protein to adopt a neuroprotective J.S.L. and M.B. helped revise the manuscript. All authors read and approved the final manuscript. loop conformation and thereby reducing neurodegenera- tion under persistent A1R stimulation. Still, the underly- Funding ing mechanisms of CPA-mediated α-syn accumulation This research was funded by the Natural Sciences and Engineering Research Council of Canada Grant (FSC), the Saskatchewan Health Foundation Collabo- and aggregation and subsequent neurodegeneration rative Innovation and Development Grant (FSC and JSL), the Heart and Stroke require further studies. Our recent study demonstrated Foundation of Canada (FSC), and Canada Foundation for Innovation Leaders that chronic A1R stimulation with CPA leads to A1R- Opportunity Fund (FSC). MTM was funded by Postdoctoral Fellowships from Saskatchewan Health Research Foundation and Parkinson Canada. dependent accumulation of α-syn [25], but other plau- sible explanation for its intracellular accumulation may Availability of data and materials also involve downstream A1R signaling, reduced vesicu- All the data generated or analyzed during this study are included in this manuscript. Original raw data are available from the University of Saskatch- lar trafficking of α-syn to the surface membranes, or ewan (Department of Surgery) and can be readily furnished upon request. increased protein stability and reduced degradation of α-syn upon direct binding with A1R ligands. Therefore, Declarations stable inhibition of chronic adenosine A1R stimula- tion occurring in aged brains of PD patients by clinically Ethical approval and consent to participate Not applicable. approved drugs that also promote a “loop” conformation of α-syn may be beneficial neuroprotective therapies to Consent for publication decrease α-synucleinopathy in PD. Not applicable. Competing interests The authors declare that they have no competing interests. Abbreviations DPCPX: 8-Cyclopentyl-1,3-dipropylxanthine; A1R: Adenosine A1 receptor; 6 Author details α-syn: Alpha-synuclein; CPA: N -cyclopentyladenosine; FJC: Fluoro-Jade C; TH: Department of Surgery, College of Medicine, University of Saskatchewan, Tyrosine hydroxylase. Saskatoon, SK, Canada. Department of Chemistry and Biochemistry, Faculty of Science, University of Regina, Regina, SK, Canada. Department of Bio- Supplementary Information chemistry, Microbiology and Immunology, College of Medicine, University of Saskatchewan, Saskatoon, SK, Canada. The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40035- 022- 00284-3. Received: 7 December 2020 Accepted: 21 January 2022 Additional file 1: Appendix 1: Summary of the eight α-synuclein structures as determined by discrete molecular dynamics simulations and further confirmed by far-UV circular dichroism and cross-linking mass spectrometry [47]. Fig. S1: Nanopore analysis recordings of α-synuclein Jako va et al. Translational Neurodegeneration (2022) 11:9 Page 25 of 26 References 25. Lv YC, Gao AB, Yang J, Zhong LY, Jia B, Ouyang SH, et al. Long-term 1. Gillis P, Malter JS. The adenosine-uridine binding factor recognizes the adenosine A1 receptor activation-induced sortilin expression promotes AU-rich elements of cytokine, lymphokine, and oncogene mRNAs. J Biol α-synuclein upregulation in dopaminergic neurons. Neural Regen Res. Chem. 1991;266(5):3172–7. 2020;15(4):712. 2. Bours MJ, Swennen EL, Di Virgilio F, Cronstein BN, Dagnelie PC. Adenosine 26. Shen KZ, Johnson SW. Presynaptic GABAB and adenosine A1 receptors 5’-triphosphate and adenosine as endogenous signaling molecules in regulate synaptic transmission to rat substantia nigra reticulata neurones. immunity and inflammation. Pharmacol Ther. 2006;112(2):358–404. J Physiol. 1997;505(1):153–63. 3. Touyz RM, Briones AM. Reactive oxygen species and vascular biology: 27. Liu DZ, Xie KQ, Ji XQ, Ye Y, Jiang CL, Zhu XZ. Neuroprotective effect implications in human hypertension. Hypertens Res. 2011;34(1):5–14. of paeoniflorin on cerebral ischemic rat by activating adenosine A1 4. Burnstock G, Verkhratsky A. Long-term (trophic) purinergic signalling: receptor in a manner different from its classical agonists. Br J Pharmacol. purinoceptors control cell proliferation, differentiation and death. Cell 2005;146(4):604–11. Death Dis. 2010;1(1):e9–e9. 28. Wang Y T, Lin HC, Zhao WZ, Huang HJ, Lo YL, Wang HT, et al. Acrolein acts 5. Cao X, Li LP, Qin XH, Li SJ, Zhang M, Wang Q, et al. Astrocytic Adenosine as a neurotoxin in the nigrostriatal dopaminergic system of rat: involve- 5′-Triphosphate release regulates the proliferation of neural stem cells in ment of α-synuclein aggregation and programmed cell death. Sci Rep. the adult hippocampus. Stem Cells. 2013;31(8):1633–43. 2017;7(1):1–9. 6. Abbracchio MP, Burnstock G, Verkhratsky A, Zimmermann H. Puriner- 29. Tapias V, Hu X, Luk KC, Sanders LH, Lee VM, Greenamyre JT. Synthetic gic signalling in the nervous system: an overview. Trends Neurosci. alpha-synuclein fibrils cause mitochondrial impairment and selective 2009;32(1):19–29. dopamine neurodegeneration in part via iNOS-mediated nitric oxide 7. Eltzschig HK, Eckle T. Ischemia and reperfusion—from mechanism to production. Cell Mol Life Sci. 2017;74(15):2851–74. translation. Nat Med. 2011;17(11):1391–401. 30. Hu Q, Ren X, Liu Y, Li Z, Zhang L, Chen X, et al. Aberrant adenosine A2A 8. Pagonopoulou O, Efthimiadou E, Asimakopoulos B, Nikolettos NK. receptor signaling contributes to neurodegeneration and cogni- Modulatory role of adenosine and its receptors in epilepsy: possible tive impairments in a mouse model of synucleinopathy. Exp Neurol. therapeutic approaches. Neurosci Res. 2006;56(1):14–20. 2016;283:213–23. 9. Dale N, Frenguelli BG. Release of adenosine and ATP during ischemia and 31. Schmued LC, Stowers CC, Scallet AC, Xu L. Fluoro-Jade C results in ultra epilepsy. Curr Neuropharmacol. 2009;7(3):160–79. high resolution and contrast labeling of degenerating neurons. Brain Res. 10. Ferré S, Ciruela F, Canals M, Marcellino D, Burgueno J, Casadó V, et al. 2005;1035(1):24–31. Adenosine A2A-dopamine D2 receptor–receptor heteromers. Targets for 32. Lee HJ, Shin SY, Choi C, Lee YH, Lee SJ. Formation and removal of neuro-psychiatric disorders. Parkinsonism Re Disord. 2004;10(5):265–271. α-synuclein aggregates in cells exposed to mitochondrial inhibitors. J Biol 11. Saitoh M, Nagai K, Nakagawa K, Yamamura T, Yamamoto S, Nishizaki T. Chem. 2002;277(7):5411–7. Adenosine induces apoptosis in the human gastric cancer cells via an 33. Madampage CA, Tavassoly O, Christensen C, Kumari M, Lee JS. Nanopore intrinsic pathway relevant to activation of AMP-activated protein kinase. analysis. Prion. 2012;6(2):116–23. Biochem Pharmacol. 2004;67(10):2005–11. 34. Baran C, Smith GTS, Bamm VV, Harauz G, Lee JS. Divalent cations induce 12. Manwani B, McCullough LD. Function of the master energy regulator a compaction of intrinsically disordered myelin basic protein. Biochem adenosine monophosphate-activated protein kinase in stroke. J Neurosci Biophys Res Commun. 2010;391(1):224–9. Res. 2013;91(8):1018–29. 35. Lee JS, Madampage CA, Stefureac RI, Napper S, Andrievskaia O. Nanopore 13. Angulo E, Casadó V, Mallol J, Canela EI, Viñals F, Ferrer I, et al. A1 Adeno- analysis of the interaction of metal ions and antibodies with prion pro- sine receptors accumulate in neurodegenerative structures in Alzheimer’s teins. Biochem Cell Biol. 2010;88(2):400–2. disease and mediate both amyloid precursor protein processing and tau 36. Kakish J, Tavassoly O, Lee JS. Rasagiline, a suicide inhibitor of mono- phosphorylation and translocation. Brain Pathol. 2006;13(4):440–51. amine oxidases, binds reversibly to α-synuclein. ACS Chem Neurosci. 14. Jansen KLR, Faull RLM, Dragunow M, Synek BL. Alzheimer’s disease: 2014;6(2):347–55. Changes in hippocampal N-methyl-d-aspartate, quisqualate, neuroten- 37. Tavassoly O, Kakish J, Nokhrin S, Dmitriev O, Lee JS. The use of nanopore sin, adenosine, benzodiazepine, serotonin and opioid receptors—an analysis for discovering drugs which bind to α-synuclein for treatment of autoradiographic study. Neuroscience. 1990;39(3):613–27. Parkinson’s disease. Eur J Med Chem. 2014;88:42–54. 15. Morelli M, Carta AR, Jenner P. Adenosine A2A receptors and Parkinson’s 38. Tavassoly O, Lee JS. Methamphetamine binds to α-synuclein and causes disease. Handb Exp Pharmacol. 2009;193:589–615. a conformational change which can be detected by nanopore analysis. 16. de Lau LML, Breteler MMB. Epidemiology of Parkinson’s disease. Lancet FEBS Lett. 2012;586(19):3222–8. Neurol. 2006;5(6):525–35. 39. Jakova E, Lee JS. Superposition of an AC field improves the discrimination 17. Ingelsson M. Alpha-synuclein oligomers—neurotoxic molecules in between peptides in nanopore analysis. Analyst. 2015;140(14):4813–9. Parkinson’s disease and other Lewy body disorders. Front Neurosci. 40. Christensen C, Baran C, Krasniqi B, Stefureac RI, Nokhrin S, Lee JS. Eec ff t of 2016;10:408. charge, topology and orientation of the electric field on the interaction of 18. Daue W, Przedborski S. Parkinson’s disease: mechanisms and models. peptides with the α-hemolysin pore. J Pept Sci. 2011;17(11):726–34. Neuron. 2003;39(6):889–909. 41. Meng H, Detillieux D, Baran C, Krasniqi B, Christensen C, Mad- 19. Spillantini MG, Crowther RA, Jakes R, Hasegawa M, Goedert M. ampage C, et al. Nanopore analysis of tethered peptides. J Pept Sci. α-Synuclein in filamentous inclusions of Lewy bodies from Parkin- 2010;16(12):701–8. son’s disease and dementia with Lewy bodies. Proc Natl Acad Sci U S 42. Malty RH, Jessulat M, Jin K, Musso G, Vlasblom J, Phanse S, et al. Mito- A. 1998;95(11):6469–73. chondrial targets for pharmacological intervention in human disease. J 20. Kakish J, Lee D, Lee JS. Drugs that bind to α-synuclein: neuroprotective or Proteome Res. 2015;14(1):5–21. neurotoxic? ACS Chem Neurosci. 2015;6(12):1930–40. 43. Moutaoufik MT, Malty RH, Amin S, Zhang Q, Phanse S, Gagarinova A, 21. Jakova E, Lee JS. Behavior of α-synuclein–drug complexes during et al. Rewiring of the human mitochondrial interactome during neuronal nanopore analysis with a superimposed AC field. Electrophoresis. reprogramming reveals regulators of the respirasome and neurogenesis. 2017;38(2):350–60. iScience. 2019;19:1114–1132. 22. Tripathi R, Saber H, Chauhan V, Tripathi K, Factor S. Parkinson disease from 44. Rotem D, Jayasinghe L, Salichou M, Bayley H. Protein detection by nano- long term drug abuse: Meta-analysis of amphetamine/methampheta- pores equipped with aptamers. J Am Chem Soc. 2012;134(5):2781–7. mine and Parkinson disease. Neurology. 2018;90(15):p6.079. 45. Venkatesan BM, Bashir R. Nanopore sensors for nucleic acid analysis. Nat 23. Pregeljc D, Teodorescu-Perijoc D, Vianello R, Umek N, Mavri J. How impor- Nanotechnol. 2011;6:615. tant is the use of cocaine and amphetamines in the development of Par- 46. Cooper JC, Hall EAH. The nature of biosensor technology. J Biomed Eng. kinson disease? A computational study. Neurotox Res. 2020;37(3):724–31. 1988;10(3):210–9. 24. Stockwell J, Jakova E, Cayabyab FS. Adenosine A1 and A2A receptors in 47. Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural the brain: current research and their role in neurodegeneration. Mol- heterogeneity of α-synuclein is governed by several distinct subpopula- ecules. 2017;22(4):676. tions with interconversion times slower than milliseconds. Structure. 2021;29(9):1048–64. Jakova et al. Translational Neurodegeneration (2022) 11:9 Page 26 of 26 48. Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural 69. Kondo T, Mizuno Y, Japanese Istradefylline Study Group. A long-term ensemble of alpha-synuclein monomer. Zenodo. 2021; https:// zenodo. study of istradefylline safety and efficacy in patients with Parkinson org/ record/ 47216 17#. YeM3a tHMLOg disease. Clin Neuropharmacol. 2015;38(2):41–46. 49. Chen Z, Xiong C, Pancyr C, Stockwell J, Walz W, Cayabyab FS. Prolonged 70. Vorovenci RJ, Antonini A. The efficacy of oral adenosine A2A antagonist adenosine A1 receptor activation in hypoxia and pial vessel disruption istradefylline for the treatment of moderate to severe Parkinson’s disease. focal cortical ischemia facilitates clathrin-mediated AMPA receptor endo- Expert Rev Neurother. 2015;15:1383–90. cytosis and long-lasting synaptic inhibition in rat hippocampal CA3-CA1 71. Chen, JF, Cunha RA. The belated US FDA approval of the adenosine A 2A synapses: differential regulation of GluA2 and GluA1 subunits by p38 receptor antagonist istradefylline for treatment of Parkinson’s disease. MAPK and JNK. J Neurosci. 2014;34(29):9621–43. Purinergic Signal. 2020:1–8. 50. Högen T, Levin J, Schmidt F, Caruana M, Vassallo N, Kretzschmar H, et al. 72. Stockwell J, Chen Z, Niazi M, Nosib S, Cayabyab FS. Protein phosphatase Two different binding modes of α-synuclein to lipid vesicles depending role in adenosine A1 receptor-induced AMPA receptor trafficking and rat on its aggregation state. Biophys J. 2012;102(7):1646–55. hippocampal neuronal damage in hypoxia/reperfusion injury. Neurop- 51. Fredholm BB, Bättig K, Holmén J, Nehlig A, Zvartau EE. Actions of caffeine harmacology. 2016;102:254–65. in the brain with special reference to factors that contribute to its wide- 73. Chen Z, Stockwell J, Cayabyab FS. Adenosine A1 receptor-mediated spread use. Pharmacol Rev. 1999;51(1):83–133. endocytosis of AMPA receptors contributes to impairments in long-term 52. Daly JW, Butts-Lamb P, Padgett W. Subclasses of adenosine receptors in potentiation (LTP) in the middle-aged rat hippocampus. Neurochem Res. the central nervous system: interaction with caffeine and related methyl- 2016;41(5):1085–97. xanthines. Cell Mol Neurobiol. 1983;3(1):69–80. 74. Qin X, Zaki MG, Chen Z, Jakova E, Ming Z, Cayabyab FS. Adenosine signal- 53. Biaggioni I, Paul S, Puckett A, Arzubiaga C. Caffeine and theophylline ing and clathrin-mediated endocytosis of glutamate AMPA receptors in as adenosine receptor antagonists in humans. J Pharmacol Exp Ther. delayed hypoxic injury in rat hippocampus: Role of casein kinase 2. Mol 1991;258(2):588–93. Neurobiol. 2021;58(5):1932–51. 54. Kakish J, Allen KJH, Harkness TA, Krol ES, Lee JS. Novel dimer com- 75. Olanow CW, Rascol O, Hauser R, Feigin PD, Jankovic J, Lang A, et al. 2009. pounds that bind α-synuclein can rescue cell growth in a yeast model A double-blind, delayed-start trial of rasagiline in Parkinson’s disease. N overexpressing α-synuclein. A possible prevention strategy for Parkinson’s Engl J Med. 2009;361(13):1268–1278. disease. ACS Chem Neurosci. 2016;7(12):1671–1680. 76. Group, Parkinson Study. A controlled, randomized, delayed-start study of 55. Bennett LLEE, Schnebli HP, Vail MH, Allan PW, Montomery JA. Purine ribo- rasagiline in early Parkinson disease. Arch Neurol. 2004;61(4):561–6. nucleoside kinase activity and resistance to some analogs of adenosine. 77. Callaghan RC, Cunningham JK, Sykes J, Kish SJ. Increased risk of Parkin- Mol Pharmacol. 1966;2(5):432–43. son’s disease in individuals hospitalized with conditions related to the use 56. Tomlinson CL, Stowe R, Patel S, Rick C, Gray R, Clarke CE. Systematic of methamphetamine or other amphetamine-type drugs. Drug Alcohol review of levodopa dose equivalency reporting in Parkinson’s disease. Depend. 2012;120(1):35–40. Mov Disord. 2010;25(15):2649–53. 78. Curtin K, Fleckenstein AE, Robison RJ, Crookston MJ, Smith KR, Hanson 57. Marsden CD, Parkes JD. Success and problems of long-term levodopa GR. Methamphetamine/amphetamine abuse and risk of Parkinson’s therapy in Parkinson’s disease. Lancet. 1977;309(8007):345–9. disease in Utah: a population-based assessment. Drug Alcohol Depend. 58. Olanow CW, Obeso JA, Stocchi F. Continuous dopamine-receptor treat- 2015;146:30–8. ment of Parkinson’s disease: scientific rationale and clinical implications. 79. Ulmer TS, Bax A, Cole NB, Nussbaum RL. Structure and dynamics of Lancet Neurol. 2005;5(8):677–87. micelle-bound human α-synuclein. J Biol Chem. 2005;280(10):9595–603. 59. Bonifácio MJ, Palma PN, Almeida L, Soares-da-Silva P. Catechol-O- methyltransferase and its inhibitors in Parkinson’s disease. CNS Drug Rev. 2007;13(3):352–79. 60. Group, Parkinson Study. A randomized placebo-controlled trial of rasagiline in levodopa-treated patients with parkinson disease and motor fluctuations: the presto study. Arch Neurol. 2005;62(2):241–8. 61. Horstink M, Tolosa E, Bonuccelli U, Deuschl G, Friedman A, Kanovsky P, et al. Review of the therapeutic management of Parkinson’s disease. Report of a joint task force of the European Federation of Neurologi- cal Societies and the Movement Disorder Society–European Sec- tion. Part I: early (uncomplicated) Parkinson’s disease. Eur J Neurol. 2006;13(11):1170–1185. 62. Schade R, Andersohn F, Suissa S, Haverkamp W, Garbe E. Dopamine agonists and the risk of cardiac-valve regurgitation. N Engl J Med. 2007;356(1):29–38. 63. Ferreira DG, Batalha VL, Vincente Miranda H, Coelho JE, Gomes R, Gon- çalves FQ, et al. Adenosine A2A receptors modulate α-synuclein aggrega- tion and toxicity. Cereb Cortex. 2017;27(1):718–30. 64. Schwarzschild MA, Agnati L, Fuxe K, Chen JF, Morelli M. Targeting adenosine A2A receptors in Parkinson’s disease. Trends Neurosci. 2006;29(11):647–54. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : 65. Factor SA, Wolski K, Togasaki DM, Huyck S, Cantillon M, Ho TW, et al. 2013. Long-term safety and efficacy of preladenant in subjects with fluctuating fast, convenient online submission Parkinson’s disease. Mov Disord. 2013;28(6):817–820. thorough peer review by experienced researchers in your ﬁeld 66. Mizuno Y, Hasegawa K, Kondo T, Kuno S, Yamamoto M. Clinical efficacy of istradefylline (KW-6002) in Parkinson’s disease: a randomized, controlled rapid publication on acceptance study. Mov Disord. 2010;25(10):1437–43. support for research data, including large and complex data types 67. Prediger RDS. Eec ff ts of caffeine in Parkinson’s disease: from neuro - • gold Open Access which fosters wider collaboration and increased citations protection to the management of motor and non-motor symptoms. J Alzheimers Dis. 2010;20(1):205–20. maximum visibility for your research: over 100M website views per year 68. Postuma RB, Lang AE, Munhoz RP, Charland K, Pelletier A, Moscov- ich M, et al. Caffeine for treatment of Parkinson disease. Neurology. At BMC, research is always in progress. 2012;79(7):651–8. Learn more biomedcentral.com/submissions

Journal

Translational Neurodegeneration – Springer Journals

Published: Feb 10, 2022

Keywords: Alpha-synucleinopathy; Adenosine A1 receptor; N6-cyclopentyladenosine; 8-cyclopentyl-1,3-dipropylxanthine; 1-aminoindan; 2-aminoindan; Neuroprotection; Neurodegeneration; Protein misfolding

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Adenosine A1 receptor ligands bind to α-synuclein: implications for α-synuclein misfolding and α-synucleinopathy in Parkinson’s disease

Adenosine A1 receptor ligands bind to α-synuclein: implications for α-synuclein misfolding and α-synucleinopathy in Parkinson’s disease

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Adenosine A1 receptor ligands bind to α-synuclein: implications for α-synuclein misfolding and α-synucleinopathy in Parkinson’s disease

Adenosine A1 receptor ligands bind to α-synuclein: implications for α-synuclein misfolding and α-synucleinopathy in Parkinson’s disease

References (88)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies