ORIGINAL RESEARCH published: 17 December 2019 doi: 10.3389/fnmol.2019.00306 ZPD-2, a Small Compound That Inhibits α-Synuclein Amyloid Aggregation and Its Seeded Polymerization Samuel Peña-Díaz 1,2† , Jordi Pujols 1,2† , María Conde-Giménez 3 , Anita Čarija 1,2 , Esther Dalfo 4,5‡ , Jesús García 6 , Susanna Navarro 1,2 , Francisca Pinheiro 1,2 , Jaime Santos 1,2 , Xavier Salvatella 6,7 , Javier Sancho 3 and Salvador Ventura 1,2,7* 1 Institut de Biotecnologia i Biomedicina, Universitat Autonoma de Barcelona, Barcelona, Spain, 2 Departament de Bioquimica i Biologia Molecular, Universitat Autonoma de Barcelona, Barcelona, Spain, 3 Department of Biochemistry and Molecular and Cell Biology, Institute for Biocomputation and Physics of Complex Systems (BIFI), University of Zaragoza, Zaragoza, Spain, 4 Faculty of Medicine, M2, Universitat Autonoma de Barcelona, Barcelona, Spain, 5 Faculty of Medicine, University of Vic – Central University of Catalonia, Vic, Spain, 6 Institute for Research in Biomedicine, The Barcelona Institute of Science and Technology, Barcelona, Spain, 7 Catalan Institute for Research and Advance Studies, Barcelona, Spain α-Synuclein (α-Syn) forms toxic intracellular protein inclusions and transmissible amyloid Edited by: structures in Parkinson’s disease (PD). Preventing α-Syn self-assembly has become one Sandra Macedo-Ribeiro, University of Porto, Portugal of the most promising approaches in the search for disease-modifying treatments for Reviewed by: this neurodegenerative disorder. Here, we describe the capacity of a small molecule Yoshitaka Nagai, (ZPD-2), identified after a high-throughput screening, to inhibit α-Syn aggregation. ZPD- Osaka University, Japan Daniel E. Otzen, 2 inhibits the aggregation of wild-type α-Syn and the A30P and H50Q familial variants Aarhus University, Denmark in vitro at substoichiometric compound:protein ratios. In addition, the molecule prevents *Correspondence: the spreading of α-Syn seeds in protein misfolding cyclic amplification assays. ZPD-2 is Salvador Ventura active against different α-Syn strains and blocks their seeded polymerization. Treating salvador.ventura@uab.es † These with ZPD-2 two different PD Caenorhabditis elegans models that express α-Syn either authors have contributed equally to this work in muscle or in dopaminergic (DA) neurons substantially reduces the number of α-Syn ‡ ORCID: inclusions and decreases synuclein-induced DA neurons degeneration. Overall, ZPD- Esther Dalfo 2 is a hit compound worth to be explored in order to develop lead molecules for orcid.org/0000-0003-4677-8515 therapeutic intervention in PD. Received: 10 September 2019 Keywords: Parkinson’s disease, α-synuclein, amyloid, protein aggregation, aggregation inhibitor, Caenorhabditis Accepted: 28 November 2019 elegans, neurodegeneration Published: 17 December 2019 Citation: Peña-Díaz S, Pujols J, INTRODUCTION Conde-Giménez M, Čarija A, Dalfo E, García J, Navarro S, Pinheiro F, Parkinson’s disease (PD) is a neurodegenerative disorder that affects about 0.3% of the population Santos J, Salvatella X, Sancho J and and >1% of people over 60 years of age (4% over 80 years) (Nussbaum and Ellis, 2003; Dexter Ventura S (2019) ZPD-2, a Small and Jenner, 2013). It is characterized by the loss of dopaminergic (DA) neurons in substantia Compound That Inhibits α-Synuclein Amyloid Aggregation and Its Seeded nigra pars compacta, which compromises the motor capacity of PD-suffering patients, producing Polymerization. tremor, rigidity, and bradykinesia (Marti et al., 2003). Additionally, since the disease spreads to the Front. Mol. Neurosci. 12:306. cerebral cortex (Braak et al., 2003), symptoms could include emotional and cognitive impairment doi: 10.3389/fnmol.2019.00306 (Marti et al., 2003). Nowadays, treatments are focused on alleviating the above mentioned motor Frontiers in Molecular Neuroscience | www.frontiersin.org 1 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation symptoms, mostly using dopamine replacement by administration of dopamine precursor (L-DOPA), combined with carbidopa, a L-DOPA decarboxylase inhibitor, and/or catechol-O-methyl transferase inhibitors and monoamine oxidase-B inhibitors (Dexter and Jenner, 2013). However, these treatments do not prevent the progression of PD and they lose efficacy as the disease advances. Parkinson’s disease is pathologically characterized by the accumulation of protein aggregates in the neuronal body, Lewy’s bodies (LB), and/or fibrils deposited in neuronal processes, Lewy’s neurites (LN), of affected neurons (Spillantini et al., 1997). These inclusions are mainly composed of α-synuclein (α-Syn), FIGURE 1 | Chemical structure of the compound ZPD-2. ZPD-2 corresponds a protein predominantly expressed in the synaptic termination to 4-cyclohexyl-2-((2-nitro-4-(trifluoromethyl)phenyl)thio)-6-oxo-1,6- of DA neurons (Bendor et al., 2013). This evidence, together dihydropyrimidine-5-carbonitrile. with the identification of mutations in the gene that encodes for this protein (SNCA) as the cause behind familial cases of PD (Polymeropoulos et al., 1997) and the observation that integrates thioflavin-T (Th-T) fluorescence and light scattering duplications and triplications of the SNCA gene lead to highly measurements, transmission electron microscopy (TEM), and penetrant forms of the disease (Singleton et al., 2003; Ibanez et al., protein misfolding cyclic amplification assays (PMCA). This 2004) directly connect PD and α-Syn. In fact, the presence of approach allowed us to identify ZPD-2 (Figure 1) as a novel aggregated α-Syn in the brain is a common feature of a group small molecule able to inhibit the aggregation of wild-type (WT) of diseases named synucleinopathies, which, in addition to PD, α-Syn, as well as that of the A30P (Kruger et al., 1998) and H50Q include Dementia with Lewy’s bodies (DLB) and multiple system (Appel-Cresswell et al., 2013) familial mutants, being active atrophy (MSA), among others (Marti et al., 2003). against the seeded polymerization of different α-Syn strains. In solution, α-Syn is a 140 amino acid intrinsically disordered The compound displayed low toxicity for neuronal human cells protein whose function seems to be related with vesicle trafficking and demonstrated significant inhibitory capacity in two well- (Bendor et al., 2013). In vitro it forms thermodynamically established Caenorhabditis elegans models of PD (van Ham et al., stable amyloid aggregates (Serpell et al., 2000) that can display 2008; Harrington et al., 2012). different conformational features (Li et al., 2018). The formation of amyloids by α-Syn follows the typical sigmoidal kinetics, reflecting a nucleation-polymerization process (Sabate et al., MATERIALS AND METHODS 2003); although secondary nucleation reactions might also occur (Xue et al., 2010). In vivo, α-Syn assemblies exert a toxic effect Protein Purification (Winner et al., 2011) and could be transmitted from cell to cell Protein expression and purification of WT α-Syn and its variants in a prion-like manner by seeding native α-Syn aggregation in (H50Q and A30P) were carried out as previously described previously unaffected neurons (Hansen et al., 2011). (Pujols et al., 2017) and the resulting purified protein was Preventing α-Syn aggregation seems to hold the potential to lyophilized and kept at −80◦ C until its use. achieve significant therapeutic impact. Several strategies have been developed toward this objective: SNCA gene-silencing In vitro Aggregation of α-Syn approaches to decrease the protein levels (McCormack et al., α-Syn was resuspended in sterile PBS and filtered through 2010), methods to increase the clearance of aggregated α-Syn by 0.22 µm membranes to remove small aggregates. Aggregation autophagic and proteasomal machineries (Gao et al., 2019), and was performed in a sealed 96-well plate, containing 70 µM molecules intended to avoid the formation and/or propagation α-Syn (WT, A30P or H50Q), 40 µM Th-T in PBS 1×, a of aggregated α-Syn (Dehay et al., 2015; Hauser, 2015). One 1/800 diameter Teflon polyball (Polysciences Europe GmbH, of the main limitations of this last strategy is the absence of a Eppelheim, Germany) and 100 µM ZPD-2 or DMSO (in control well-defined structure of monomeric α-Syn in solution, due to samples) in a total volume of 150 µL per well. The plate was its intrinsically disordered nature, which hampers the rational incubated at 100 rpm and 37◦ C after having been fixed in an design of inhibitors. High-throughput screening protocols have orbital culture shaker Max-Q 4000 (ThermoScientific, Waltham, been developed to circumvent this problem (Silva et al., 2011; MA, United States). Measurements of Th-T fluorescence were Levin et al., 2014). A number of promising small molecules done every 2 h in a Victor3.0 Multilabel Reader (PerkinElmer, have been discovered with this approach, including anle138b Waltham, MA, United States), exciting through a 430–450 nm (Levin et al., 2014), BIOD303 (Moree et al., 2015), fasudil filter and collecting the emission signal with a 480–510 filter. Each (Tatenhorst et al., 2016), squalamine (Perni et al., 2017), or assay was done in triplicate. The values of the aggregation kinetics SynuClean-D (SCD) (Pujols et al., 2018). In this context, we have were fitted to the following Eq. 1 (Crespo et al., 2016): developed a robust screening and validation protocol to analyze large chemical libraries in the search for effective inhibitors of 1 ∝=1− (1) α-Syn aggregation (Pujols et al., 2017). The in vitro pipeline kb (eka t − 1) + 1 Frontiers in Molecular Neuroscience | www.frontiersin.org 2 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation where kb and ka constitute the homogeneous nucleation rate Sixty microliters of this α-Syn solution was added into 200-µL constant and the secondary rate constant (fibril elongation and PCR tubes containing 1.0 mm silica beads (Biospec Products, secondary nucleation), respectively (Crespo et al., 2016). Bartlesville, OK, United States). Samples were exposed to 24-h Titration assays were done by applying different ZPD-2 cycles of 30 s sonication and 30 min incubation at 37◦ C, using concentrations (200, 150, 100, 75, 50, 25, and 10 µM). Time- a Misonix 4000 sonicator, at 70% power. After every 24 h-cycle, dependent assays were developed by adding 100 µM of ZPD-2 1 µL of the incubated sample was added to a new PCR-tube at different time points after the beginning of the reaction (4, 8, containing fresh α-Syn. This process was repeated for 5 days. 12, 16, 20, and 24 h). In all cases a fixed concentration of α-Syn at In the case of treated samples, ZPD-2 was added in each cycle 70 µM was maintained. to the fresh non-sonicated sample to a final concentration of Strains were generated as previously described (Bousset et al., 128 µM, which corresponds to the 0.7:1 α-Syn:ZPD-2 ratio 2013; Peelaerts et al., 2015; Carija et al., 2019). Briefly, lyophilized of the previous set of aggregation kinetics assays. Untreated α-Syn was resuspended in PBS 1× and dialyzed for 24 h in a samples were prepared adding the same concentration of DMSO 1:1000 (v/v) ratio with buffer B (50 mM Tris–HCl pH 7.0), or (0.26%) present in the treated mixtures. All the reactions were buffer C (50 mM Tris–HCl pH 7.0 supplemented with 150 mM made in triplicate. NaCl). Then, the protein was filtered through 0.22 µm membrane At the end of each cycle, 10 µL of the incubated and incubated at 70 µM in presence or absence of 100 µM ZPD- samples were diluted 1:10 with 90 µL of PBS 1×, 40 µM 2 in a 96-well plate as described above. For the seeding assays, Th-T. Th-T fluorescence was measured in a Cary Eclipse α-Syn pre-formed fibrils were sonicated for 5 min and then Fluorescence Spectrophotometer (Agilent, Santa Clara, CA, added to the aggregation reaction at ratios of 1% (v/v) for each United States), exciting at 445 nm and collecting the emission condition. The plate was then incubated, and Th-T fluorescence signal between 460 and 600 nm. measured as previously indicated. The soluble fraction was obtained for subsequent Proteinase K Digestion quantification by centrifuging 300 µL of aggregated sample For protein digestion, 6 µL of Proteinase K (5 µg/mL final at 16,900 × g for 90 min. The supernatant was then recovered concentration) was added to 18 µL of PMCA aggregated samples and loaded into a Tricine–SDS-PAGE gel. Gels were stained and incubated for 30 min at 37◦ C. After the incubation, 8 µL of with Blue safe. Finally, the density of the α-Syn bands was loading buffer containing 1% β-mercaptoethanol was added and calculated using Quantity One software (Bio-Rad, Hercules, CA, the enzyme was thermally inactivated at 95◦ C for 10 min. Finally, United States). Experiments were done at least in triplicate. 7 µL of the incubated and stained samples was loaded into a Tricine–SDS-PAGE gel together with unstained Protein Standard Transmission Electron Microscopy markers (ThermoFisher Scientific, Waltham, MA, United States). End-point α-Syn aggregates incubated for 32 h were collected, Gels were stained with Blue safe. diluted 1:10 with PBS 1× and sonicated for 5 min. Five microliters of these sonicated samples was placed rapidly on a Nuclear Magnetic Resonance carbon-coated copper grid and incubated for 5 min. The grids Expression of 15 N-labeled human WT α-Syn was carried out in were dried with a filter paper to withdraw the excess of sample Escherchia coli BL21 DE3 strain. First, cells were grown in LB and immediately washed twice with miliQ water. Finally, 5 µL medium until an OD600 of 0.6. The culture was then centrifuged of 2% (w/v) uranyl acetate was added to the top of the grid and at 3000 rpm for 15 min and the pellets collected and resuspended incubated for 2 min. The excess of uranyl acetate was removed in 1 L minimal medium, composed of: 768 mL of miliQ water with a filter paper and grids were left to air-dry for 10 min. with 1 mL of ampicillin 100 mg/mL, 100 µL CaCl2 1 M, Images were obtained using a TEM Jeol 1400 (Peabody, MA, 2 mL MgSO4 2 M, 20 mL glucose 20%, 10 mL vitamins 100× United States) operating at an accelerating voltage of 120 kV. (Sigma–Aldrich, Darmstadt, Germany), 200 mL salts M9, and A minimum of 30 fields were screened per sample, in order to 1 g 15 NH4 (Cambridge Isotope Laboratories, Inc., Tewksbury, collect representative images. MA, United States). Cells were incubated for 1 h at 37◦ C and 250 rpm. After that, protein expression was induced for 4 h Light Scattering with 1 mM IPTG. Protein was purified as previously described End-point α-Syn aggregates were collected, placed into a (Pujols et al., 2017). quartz cuvette, and analyzed in a Cary Eclipse Fluorescence 1 H-15 N HSQC spectra were obtained at 20◦ C on a Bruker Spectrophotometer (Agilent, Santa Clara, CA, United States). The 600 MHz NMR spectrometer equipped with a cryoprobe in a sample was excited at 300 nm and the subsequent scattering at mixture containing 70 µM 15 N-labeled α-Syn, PBS buffer (pH 90◦ monitored between 280 and 320 nm. 7.4), 2.5% d6-DMSO, and 10% D2 O in the absence or in the presence of 100 µM ZPD-2. Protein Misfolding Cyclic Amplification The PMCA assay was carried out as previously described Toxicity Assays (Herva et al., 2014). Briefly, α-Syn was resuspended to a final Neuroblastoma cells were incubated 24 h in DMEM concentration of 90 µM in Conversion Buffer (PBS 1×, 1% Triton medium in a 96-well plate before the addition of different X-100, 150 mM NaCl), supplemented with Complete Protease concentrations of ZPD-2 (from 1 µM to 1 mM). Cells Inhibitor Mixture (Roche Applied Science, Penzberg, Germany). were incubated for 48 h at 37◦ C and PrestoBlue reagent R Frontiers in Molecular Neuroscience | www.frontiersin.org 3 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation (ThermoFisher Scientific, Waltham, MA, United States) was distinguishable from surrounding fluorescence. ImageJ software added to analyze cell death. Treated and untreated cells were was used for measuring the number of cellular aggregates incubated with PrestoBlue for 10 min at 37◦ C. Finally, R considering the area dimensions. For the quantification of fluorescence emission was measured by exciting at 560 nm and α-syn aggregates in C. elegans one single image was taken collecting at 590 nm. from each animal. Every image contained among 30–45 stacks (1 µm) that allowed to detect aggregates at different Caenorhabditis elegans Assays animal positions. At least 30 animals were imaged for each Maintenance assayed condition. Animals synchronization was carried out by bleaching and overnight hatching in M9 (3 g/L KH2 PO4 , 6 g/L Na2 HPO4 , 5 g/L C. elegans Lifespan Analysis NaCl, 1 M MgSO4 ) buffer. Thus, nematodes were cultured at L4-stage synchronized C. elegans were exposed to 10 µM of ZPD- 20◦ C on growth media plates (NGM) containing 1 mM CaCl2 , 2 or DMSO (controls) during lifespan analysis. The worms were 1 mM MgSO4 , 5 µg/mL cholesterol, 250 M KH2 PO4 pH 6.0, classified as alive, dead, or censored every 2 days by determining 17 g/L Agar, and 3 g/L NaCl. Plates were previously seeded with their movement and response to nose and tail tap. The numbers E. coli OP50 strain. Nematodes were maintained using standard of alive and dead worms were recorded until all worms perished. protocols (Brenner, 1974). The data were plotted as a Kaplan–Meier survival curve and groups compared using a Wilcoxon-test. Strains Strain NL5901, unc-119(ed3) III; pkIs2386 [Punc-54:α-SYN:YFP; C. elegans Neurodegeneration Assays unc-119(+)] was obtained from the C. elegans Genetic Center Worms were analyzed for α-Syn-induced DA neurodegeneration (CGC). For the α-Syn-induced DA degeneration analysis, as described previously (Harrington et al., 2012). Briefly, 20–30 strain UA196 (Harrington et al., 2012), gifted generously L4-staged animals were transferred to ZPD-2 – NGM plates and by the laboratory of Dr. Guy Caldwell (Department of make them grow up to 7 days (L4 + 7 days of development) Biological Science, The University of Alabama, Tuscaloosa, after which the DA cell death induced by the over-expression AL, United States), was used; [sid-1(pk3321); baIn33 (Pdat- of α-Syn was analyzed by fluorescence. Plates containing only 1:sid-1, Pmyo-2:mCherry); baIn11 (Pdat-1:α-SYN; Pdat-1:GFP)]. 0.2% DMSO, without ZPD-2, were used as control. Worms were In the main text, this strain was named Pdat-1:GFP; Pdat- transferred daily to avoid cross contamination. 1:α-SYN. The six anterior DA neurons (four CEP and two ADE DA neurons) were scored for neurodegeneration according to ZPD-2 Administration previously described criteria (Sulston et al., 1975; Harrington After cooled, the autoclaved NGM agar medium (1 mM CaCl2 , et al., 2012). Worms were considered normal when all six anterior 1 mM MgSO4 , 5 µg/mL cholesterol, 250 M KH2 PO4 pH 6.0, DA neurons (four CEP, cephalic, and two ADE, anterior deirid) 17 g/L Agar, and 3 g/L NaCl) was enriched with 100 µM of a were present without any visible signs of degeneration. If a worm stock of ZPD-2 in 0.2% DMSO to a final concentration of 10 µM. displayed degeneration in at least one of the six neurons, it After 2 days, plates were seeded with 250 µL of E. coli OP50 with was scored as exhibiting degeneration. For each independent 10 µM of ZPD-2. Nematodes were placed on the plates at larval experiment, 30 worms of each treatment were examined under stages L4 and exposed either to ZPD-2 or DMSO (controls) for a Nikon Eclipse E800 epifluorescence microscope equipped with 7 days. Daily transfer was done to avoid cross progeny. an Endow GFP HYQ filter cube (Chroma Technology Corp., Bellows Falls, VT, United States). Aggregate Quantification The number of cellular inclusions was quantified as previously Microscopy and Imaging described (van Ham et al., 2008; Munoz-Lobato et al., Animals were placed in a 1 mM solution of sodium azide 2014). Briefly, NL5901 (Punc-54:α-SYN:YFP) worms were age- and mounted with a coverslip on a 4% agarose pad. Animals synchronized and left overnight to hatch. Nematodes in were visualized with a Nikon Eclipse E800 epifluorescence phase L1 were cultured and grown into individual NGM microscope. The system acquires a series of frames at specific plates seeded with E. coli OP50. When animals reached L4 Z-axis position (focal plane) using a Z-axis motor device. developmental stage, they were transferred onto either ZPD- Animals were examined at 100× magnification to examine 2-treated plates or DMSO-treated plates (negative control). α-Syn-induced DA cell death and at 40× to examine α-Syn Every day, animals were transferred into a new plate to apparent aggregate. avoid cross contamination. At stage L4 + 7, the aggregates in the anterior part of every single animal were counted. Statistical Analysis For each experiment, 30 7-day-old nematodes per treatment All graphs were generated with GraphPad Prism 6.0 software were analyzed using a Nikon Eclipse E800 epifluorescence (GraphPad Software Inc., La Jolla, CA, United States). Data microscope equipped with an Endow GFP HYQ filter cube were analyzed by two-way ANOVA Tukey’s HSD test using (Chroma Technology Corp., Bellows Falls, VT, United States) SPSS software version 20.0 (IBM Analytics, Armonk, NY, and each experiment was carried out in triplicate. Inclusions United States) and t-test using GraphPad software version could be described as discrete bright structures, with edges 6.0 (GraphPad Software Inc., La Jolla, CA, United States). Frontiers in Molecular Neuroscience | www.frontiersin.org 4 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation All data are shown as means and standard error of mean To address the time window in which ZPD-2 is active, (SEM). p < 0.05 was considered statistically significant. In we set up aggregation reactions with a constant amount the graphs ∗ , ∗∗ , and ∗∗∗ indicate p < 0.05, p < 0.01, and of ZPD-2 added at different time points after the reaction p < 0.001, respectively. begins. A time-dependent response was observed (Figure 3B), with a very significant inhibition when ZPD-2 was added at early (4–8 h) and intermediate (12–16 h) times, and a less pronounced effect when it was added at the plateau RESULTS phase (20–24 h). This indicates that ZPD-2 is mostly active against the species formed early in the aggregation reaction, ZPD-2 Reduces and Delays the consistent with its highest impact on the nucleation rate Aggregation of Human α-Synuclein constant kb . Importantly, NMR studies using isotopically labeled in vitro monomeric and soluble α-Syn indicated that ZPD-2 does We designed and optimized a screening protocol that allows not interact with its native form, since we could not detect to follow the aggregation kinetics of α-Syn by monitoring Th- any perturbations in chemical shifts or peak intensities in T fluorescence emission for 32 h. This approach permitted α-Syn in the presence of a molar excess of the molecule us to study the inhibitory potential of more than 14,000 (Supplementary Figure S2). compounds (Pujols et al., 2017, 2018). The activity of molecules Several α-Syn single point mutations are connected with able to reduce significantly the final amount of Th-positive the onset of familial cases of PD (Kruger et al., 1998; Appel- material and/or impact the nucleation or elongation rates of Cresswell et al., 2013). We studied the ability of ZPD-2 to the reaction was further confirmed using light scattering and prevent the aggregation of two of the most frequent and TEM measurements at the end of the reaction. This allowed us aggressive variants, H50Q and A30P. The molecule was also to identify 30 active compounds, most of which seem not to active against these α-Syn forms in kinetic assays (Figure 4A). be connected in terms of structure, precluding QSAR studies. According to the relative Th-T signal at the end of the We have previously described the properties of SCD a molecule reaction in ZPD-2-treated and non-treated samples, the molecule that acts preferentially on top of α-Syn proto-fibrillar or fibrillar inhibited the aggregation of A30P and H50Q by 96 and 94%, assemblies (Pujols et al., 2018). Here, we describe the properties respectively (Figure 4B). of ZPD-2 (Figure 1), a compound that differs in its mechanism of action. SCD and ZPD-2 share a benzotrifluoride group, which suggested that it could constitute the minimal inhibitory ZPD-2 Prevents α-Syn Seeded unit; however, this group is devoid of any anti-aggregation Aggregation in Protein Misfolding Cyclic activity by itself (unpublished), indicating that, most likely, it Amplification Assays only acts as a framework for the different active groups in Protein misfolding cyclic amplification assays, initially developed the two molecules. to study the polymerization and propagation process of the The incubation of 70 µM of α-Syn in the presence and prion protein (Barria et al., 2012; Morales et al., 2012), absence of 100 µM of ZPD-2 revealed that the compound has been recently adapted for α-Syn amyloid aggregation modulated the protein aggregation, reducing the formation of (Herva et al., 2014). Essentially, cycles of incubation at Th-T positive structures at the end of the reaction by an 37◦ C are followed by vigorous sonication in order to allow 80%, while extending t 50 by 8 h (Figure 2A). The analysis fibril growth and subsequent fibrillar rupture, thus producing of the kinetics revealed a reduction in the nucleation rate α-Syn seeds. These preformed seeds are used to trigger constant in presence of ZPD-2 (kb = 0.008833) by threefold, the aggregation of fresh protein in the following cycle, when compared to the control reaction (kb = 0.02754). The amplifying the fibrillar content. At 90 µM of α-Syn, PMCA autocatalytic rate constant was also lower in the treated sample produced amyloid structures resistant to protease K (PK) (ka = 0.2432 h−1 ) than in the control (ka = 0.3230 h−1 ). Light digestion, as observed by SDS-PAGE, with the maximum scattering measurements at 300 nm confirmed that the observed protection arising after four rounds (Figure 5A, middle). Th- reduction in Th-T fluorescence corresponds to an effective T fluorescence measurements of the same samples indicated decrease in the levels of α-Syn aggregates, with a 67% decrease that this protection correlates with an increasing presence in the dispersion of light in the presence of ZPD-2 (Figure 2B). of amyloid-like assemblies (Figure 5B). In sharp contrast, in TEM images corroborated that the samples incubated with ZPD- the presence of ZPD-2, the amount of PK-resistant protein 2 (Figure 2D) contained less fibrils per field than the non- after four rounds is negligible (Figure 5A, right), Th-T treated ones (Figure 2C). In good agreement with these data, fluorescence signal being also significantly low relative to quantification of soluble α-Syn at the end of the aggregation control samples at this stage (Figure 5B). These results reaction indicated that its level was threefold higher in ZPD-2- suggested that ZPD-2 was strongly interfering with the PMCA- treated samples (Supplementary Figure S1A). promoted seeding of α-Syn amyloids. The fact that Th-T Further analysis of the inhibition capacity of ZPD-2 indicated decrease becomes significant only at pass 4, likely indicates that it exhibited a dose-dependent effect, displaying a statistically that the aggregated non PK-resistant species generated at significant effect even at 10 µM (1:7 compound:protein ratio) early steps still retain certain Th-T binding ability, since (Figure 3A), where the final Th-T signal was reduced by 49%. SDS-PAGE analysis indicates that the levels of PK-resistant Frontiers in Molecular Neuroscience | www.frontiersin.org 5 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation FIGURE 2 | ZPD-2 inhibits the aggregation of wild-type α-synuclein in vitro. (A) Aggregation kinetics of α-Syn in absence (black) and presence (green) of ZPD-2. Intensity of Th-T fluorescence is plotted as a function of time. (B) Light scattering of end-point aggregates is measured at 300 nm for untreated (white) and ZPD-2-treated samples (green). (C,D) Representative TEM images of untreated (C) and ZPD-2-treated (D) samples. Th-T fluorescence is expressed as normalized means. Final points were obtained at 48 h after the aggregation reaction begin. Error bars are shown as standard errors of mean values, ∗∗∗ p < 0.001. FIGURE 3 | Analysis of the inhibitory capacity of ZPD-2. (A) Titration of the effect of different concentrations of ZPD-2 on 70 µM α-Syn aggregation. (B) Th-T fluorescence of α-Syn end-point aggregates after the addition of ZPD-2 at different time points during the aggregation kinetics. Th-T fluorescence is plotted as normalized means. End-points were obtained at 48 h of α-Syn incubation. Error bars are shown as standard errors of mean values, ∗∗∗ p < 0.001. protein is already decreased in treated samples at passes 1–3 depending on the environmental conditions (Li et al., 2018); a (Supplementary Figure S3). property that has been linked with its spreading in the brain and the manifestation of different synucleinopathies (Peelaerts et al., 2015). We analyzed the capacity of ZPD-2 to prevent the ZPD-2 Prevents the Aggregation of aggregation of α-Syn into different previously described amyloid Different α-Synuclein Amyloid conformations (Bousset et al., 2013; Carija et al., 2019). We refer Conformations them as strain B (buffer B, 50 mM Tris–HCl pH 7.0) and strain C The aggregation of α-Syn has been described to lead to (buffer C, 50 mM Tris–HCl pH 7.0 supplemented with 150 mM the formation of different amyloid conformations, or strains, NaCl), to keep the original strain nomenclature. ZPD-2 was Frontiers in Molecular Neuroscience | www.frontiersin.org 6 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation FIGURE 4 | ZPD-2 inhibits the aggregation of α-synuclein familial variants. (A) Aggregation kinetics of WT (black), H50Q (blue), and A30P (green) variants of α-Syn in presence (dotted) and absence (continuous) of ZPD-2, using Th-T as reporter. (B) End-point measurements of the aggregation of WT, H50Q, and A30P variants of α-Syn in presence (dotted) or absence (continuous) of ZPD-2 Th-T fluorescence are expressed as normalized means. Error bars are shown as standard errors of mean values. FIGURE 5 | PMCA of α-synuclein in presence of ZPD-2. (A) Tricine–SDS-PAGE gels of untreated (middle) and ZPD-2-treated (right) PMCA samples before (–) and after (+) being digested with proteinase K. (B) Th-T fluorescence of different PMCA cycles of treated (green) and untreated (black) samples. Soluble α-Syn and PMCA steps 4 and 5 are shown. Th-T fluorescence is plotted as normalized means. Error bars are shown as standard errors of mean values. active in both cases (Figures 6A,E), inhibiting by up to 90% the addition of 1% (v/v) of seeds effectively accelerated the formation formation of the amyloid strains B and C, as monitored by Th- of both B and C strains (Figures 7A,B). However, the presence T fluorescence. Light scattering measurements (Figures 6B,F) of ZPD-2 abrogates this effect, reducing the final amount of and TEM imaging (Figures 6C,D,G,H) and soluble protein amyloid-like structures in seeded reactions by an 87% for strain quantification at the end of the reaction (Supplementary Figure B (Figure 7A) and a 90% for strain C (Figure 7B), according to S1B) of the different samples confirmed the inhibitory activity Th-T fluorescence. Again, light dispersion measured at 300 nm of ZPD-2 against the two strains. Non-fibrillar aggregates might revealed a significant decrease of aggregates by 57 and 70% in the be necessary for fibril formation (obligate), able to convert into case of strains B and C, respectively (Figures 7C,D). fibrils, but not indispensable for fibril formation (on-pathway), or unable of converting directly to fibrils (off-pathway). The difference between the large reduction in Th-T fluorescence ZPD-2 Reduces the Formation of promoted by ZPD-2 in strain C aggregation kinetics and the α-Synuclein Inclusions in a C. elegans moderate impact the molecule has in light scattering and soluble Model of PD protein levels might indicate the formation of Th-T negative off- We assessed the toxicity of ZPD-2 for human neuroblastoma pathway aggregates in these conditions, since they do not evolve cells. No significant toxicity was observed when the molecule into fibrils. However, their size should be rather small, since we was added to the cell culture up to 80 µM (Supplementary did not observe any large amorphous aggregate in ZPD-2-treated Figure S4). We skipped efficacy studies on neuroblastoma cells, samples (Figure 6H). because, with more than 20 different compounds analyzed, we We addressed whether the strong inhibitory capability of could not find a straightforward connection between the potency ZPD-2 at neutral pH can be overridden by the presence of of the molecules in cell cultures and that in our C. elegans models preformed fibrils able to seed the aggregation reaction. The of PD. We first analyzed the effect of ZPD-2 in the C. elegans Frontiers in Molecular Neuroscience | www.frontiersin.org 7 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation FIGURE 6 | ZPD-2 blocks the aggregation of two different α-synuclein strains. (A,E) Aggregation kinetics of α-Syn strains B (A) and C (E) in absence (black) and presence (green) of ZPD-2. (B,F) Light scattering final point measurements at 300 nm of untreated (white) and ZPD-2-treated samples (green) of strains B (B) and C (F). (C,D,G,H) Representative TEM images of untreated α-Syn aggregates (C,G) and treated (D,H) samples for strains B and C, respectively. Th-T fluorescence is expressed as normalized means. Final points were obtained at 48 h after the aggregation reaction begins. Error bars are shown as standard errors of mean values, where p < 0.05 and p < 0.001 were indicated by ∗ and ∗∗∗ , respectively. strain NL5901. This strain over-expresses human α-Syn fused (Sulston et al., 1975). The existence of six anterior DA neurons to the yellow fluorescent protein (YFP), under the control of has been recently used to analyze PD-related processes in a model the muscular unc-54 promoter, transgene phIs2386 (Punc-54:α- (strain UA196) that expresses both human α-Syn and GFP under SYN:YFP). The expression of human α-Syn in the muscle of the control of the dopamine transporter promoter (Pdat-1:GFP; this nematode has been successfully used to identify modifier Pdat-1:α-SYN) (Kim et al., 2018). Human α-Syn expression in genes (Hamamichi et al., 2008; van Ham et al., 2008). Animals these DA neurons induces a progressive degeneration process at the fourth larval stage (L4) were incubated in the presence or (Cao et al., 2005). At 9 days post-hatching, the number of absence of 10 µM ZPD-2 and analyzed at 9 days post-hatching remaining functional neurons of untreated (Figure 9A) and (L4 + 7). These aged worms, which mimic aged PD patients, were ZPD-2-treated (Figure 9B) worms was analyzed. As an average, then analyzed by epifluorescent microscopy and the number of in control worms 48.1% of DA neurons are non-functional, visible α-Syn inclusions was quantified (Figures 8A,B). In these whereas in treated animals this value decreases to 40.4% (p- assays, ZPD-2 moderately, but significantly, reduced the number value = 0.038, Wilcoxon unpaired test). Despite the difference of apparent aggregates (25.7 ± 1.3) when compared to untreated between both means is rather low, the distribution of the data worms (31.8 ± 1.7) (Figure 8C). In addition, worms treated indicated a displacement in the DA neurons survival profile with ZPD-2 showed an increase in their mean lifespan of 14.2%, (Figure 9C and Supplementary Figure S6) in the presence of relative to untreated animals (p-value = 0.015, Wilcoxon unpaired ZPD-2 when compared to the control worms. As a result, there test) (Supplementary Figure S5). is a significant increase in the number of worms containing more than three functional neurons in the anterior region in the presence of ZPD-2 (51.0 ± 4.8%) when compared to the controls Neuroprotective Role of ZPD-2 in a (29.1 ± 3.1%) (Figure 9D). C. elegans Model of PD The loss of DA neurons is one of the most important characteristics of PD and an important target in the search for DISCUSSION a future treatment for this disorder. C. elegans presents a total of four pairs of DA neurons, three of them in the anterior part Protein aggregation is tightly connected with neurodegenerative (CEPD, CEPV, and ADE) and one pair in the posterior part (PDE) disorders such as Alzheimer’s and PDs. Immediately after the Frontiers in Molecular Neuroscience | www.frontiersin.org 8 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation FIGURE 7 | Seeding assays with three different strains. (A,B) Aggregation kinetics of α-Syn, buffer B (50 mM Tris–HCl pH 7.0) (A), or buffer C (50 mM Tris–HCl pH 7.0 supplemented with 150 mM NaCl) (B), reported by Th-T fluorescence, in absence of compounds and seeds (black), in presence of 1% (v/v) of preformed seeds at the specific condition (gray dotted line) and in presence of seeds and 100 µM of ZPD-2 (green). Light dispersion of treated (green) and untreated (white) seeded samples at final point of strain B (C) and strain C (D). Error bars are shown as standard errors of mean values, ∗∗∗ p < 0.001. FIGURE 8 | In vivo anti-aggregational assays in Caenorhabditis elegans. Representative images of apparent α-Syn aggregates in C. elegans body wall muscle cells obtained by epifluorescence microscopy of NL5901 worms treated without (A) and with ZPD-2 (B). (C) Quantification of α-Syn muscle inclusions in the absence (white) and presence of ZPD-2 (green). ∗∗ p < 0.01. identification of α-Syn as the main fibrillar component in LBs and that stabilize α-Syn and thus prevent or delay its aggregation, as it LNs (Spillantini et al., 1997, 1998) it became evident that targeting has been successfully done for globular proteins like transthyretin the aggregation of this protein might hold therapeutic potential (Bulawa et al., 2012; Sant’Anna et al., 2016), difficult. In this (Tatenhorst et al., 2016). scenario, evaluation of large chemical libraries appears as one of Nevertheless, the absence of a defined three-dimensional the few strategies we have to discover an effective inhibitor of structure for the functional state of α-Syn due to its intrinsically α-Syn deposition and, indeed, this approach has already rendered disordered nature makes the rational design of effective inhibitors promising molecules (Levin et al., 2014; Moree et al., 2015; Frontiers in Molecular Neuroscience | www.frontiersin.org 9 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation FIGURE 9 | Neuroprotective activity of ZPD-2 in a Caenorhabditis elegans model of PD. Representative images of GFP and α-Syn expressing anterior DA neurons in worms treated without (A) and with ZPD-2 (B) for 7 days after L4. Healthy neurons are labeled with white arrows. (C) Distribution of DA surviving neurons in the anterior region of worms. (D) Percentage of worms containing three neurons functional or less and more than three functional neurons after 7 days post-hatching. White bars indicate control samples while the green ones correspond to treated samples. ∗ p < 0.05. Tatenhorst et al., 2016; Perni et al., 2017; Pujols et al., 2018). In and H50Q (94% inhibition) familial variants, provided that the present work, we describe the discovery of ZPD-2, a small both mutations facilitate oligomerization, H50Q favoring also molecule able to prevent up to 90% the in vitro aggregation of fibrillation (Marvian et al., 2019). WT α-Syn and of familial mutants of the protein when used ZPD-2 is able to inhibit the aggregation of α-Syn under in a 0.7:1 (protein:ZPD-2) ratio, delaying also significantly the different solution conditions. This ability opens a possibility completion of the reaction. Its inhibitory capacity was confirmed for its use in different synucleinopathies, where different α-Syn by orthogonal techniques such as light scattering and TEM. strains might occur (Bousset et al., 2013; Peelaerts et al., 2015). Further analysis demonstrated that ZPD-2 was able to prevent Importantly, ZPD-2 is one of a few small molecules shown the aggregation in a concentration-dependent manner, with to inhibit efficiently α-Syn seeded aggregation, where the lag ∼50% inhibition at a 7:1 protein:compound ratio. This, together phase of the reaction is shortened or abrogated because the with solution NMR measurements indicate that ZPD-2 does soluble protein can be directly incorporated on top of the not interact significantly with soluble monomeric α-Syn, which preformed fibrillar fragments. This seeding-blocking activity suggests that it will not interfere with the functional state of explains why ZPD-2 is so effective preventing the formation the protein. In addition, the inhibitory potential of ZPD is of PK-resistant/Th-T-positive species in PMCA assays, which time-dependent, being more significant at early (0–8 h) stages, promote both templated seeding and aggregates amplification. in fair contrast with SC-D, a compound we identified in the This effect might respond to the ability of the compound to same screening campaign, whose activity was time-independent, either destabilize small aggregates or to prevent their elongation, being able to target late species (Pujols et al., 2018). The largest a property that can be very relevant to prevent the cell-to-cell affinity of ZPD-2 for early aggregating species is also inferred spreading of misfolded α-Syn. from the fact that it mainly impacts the nucleation constant, ZPD-2 had not detectable toxic effect for neuronal cells at reducing it by threefold. This might also explain why, at a 0.7:1 10 µM, a concentration at which it reduces the presence of α-Syn ratio, the molecule works well for the A30P (96% inhibition) inclusions in a C. elegans model of PD expressing human α-Syn Frontiers in Molecular Neuroscience | www.frontiersin.org 10 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation in body wall muscle cells and extends lifespan. Not surprisingly, MINECO (BIO2015-70092-R) and the European Research this anti-aggregational activity translates in reduced DA neurons Council (Contract 648201). degeneration in a C. elegans model that over-expresses human α-Syn exclusively in these cells, increasing significantly the proportion of animals that keep > 50% of their anterior part ACKNOWLEDGMENTS DA neurons intact. We thank the Infraestructura Científica y Técnica Singular NMR facility at Centres Científics i Tecnològics de la Universitat de CONCLUSION Barcelona for help with NMR, the Servei de Microscòpia at Universitat Autònoma de Barcelona for their help with TEM, and In conclusion, ZPD-2 properties make this molecule a promising Anna Villar-Pique for help with plasmid construction. The worm hit for the sake of developing leads able to tackle α-Syn strain UA196 used for neurodegeneration assays was a generous aggregation and seeds propagation in PD and, potentially, other gift of Dr. Guy A. Caldwell. synucleinopathies. SUPPLEMENTARY MATERIAL DATA AVAILABILITY STATEMENT The Supplementary Material for this article can be found All datasets generated for this study are included in the online at: https://www.frontiersin.org/articles/10.3389/fnmol. article/Supplementary Material. 2019.00306/full#supplementary-material FIGURE S1 | α-Synuclein soluble fraction at the end of the aggregation. (A) Soluble fraction of α-Syn when incubated in absence (black) or presence (green) AUTHOR CONTRIBUTIONS of ZPD-2 in PBS solution. (B) Soluble fraction of strains B and C at final point of the aggregation when incubated in presence (green) or absence (black) of ZPD-2. SP-D, JP, XS, JavS, ED, and SV conceived and designed the FIGURE S2 | Lack of interaction between monomeric α-synuclein and ZPD-2 experiments, and analyzed the results. SP-D, JP, FP, JaiS, and assessed by NMR. Superposition of the 1H-15N HSQC NMR spectra of MC-G performed the aggregation assays. AČ expressed and 15N-labeled α-Syn (70 µM) in absence (black) and presence (green) of purified the H50Q and A30P variants. SP-D and SN performed 100 µM of ZPD-2. the PMCA assays. SN performed the toxicity assays. JG and FIGURE S3 | PMCA assay at early stages. Tricine–SDS-PAGE gels of untreated XS performed the NMR assays. SP-D and ED performed the (middle) and ZPD-2-treated (right) PMCA samples before (−) and after (+) being C. elegans tests. SP-D, JP, and SV wrote the manuscript. digested with proteinase K. Soluble α-Syn and PMCA steps 1–3 are shown. FIGURE S4 | Toxicity assays. Analysis of neuronal cells culture survival in presence of different concentration of ZPD-2. Survival is potted as normalized FUNDING means. Error bars are shown as standard error of means values, where p < 0.001 was indicated by ∗∗∗ . SV was supported by the Ministerio de Economía y FIGURE S5 | C. elegans lifespan analysis. Effect of ZPD-2 treatment (green) on Competitividad (MINECO) (BIO2016-78310-R), the ICREA the survival of PD model animals, in comparison with untreated PD worms (black). (ICREA-Academia 2015) and the Fundación La Marató de The data represent the survival ratio (approximately 60–80 animals per group). TV3 (Ref. 20144330). JavS was supported by the MINECO FIGURE S6 | Distribution of functional neurons in the C. elegans dopaminergic (BFU2016-78232-P) and the Gobierno de Aragón (E45_17R). model. Normal distribution of the remaining functional dopaminergic (DA) neurons ED was supported by the Instituto de Salud Carlos III in transgenic animals when treated with ZPD-2 (green) or vehicle (gray). The (PH613883/ERDF/ESF). JG and XS were supported by the dashed line delimits animals having four or more functional DA neurons. REFERENCES Brenner, S. (1974). The genetics of Caenorhabditis elegans. Genetics 77, 71–94. Bulawa, C. E., Connelly, S., Devit, M., Wang, L., Weigel, C., Fleming, J. A., et al. Appel-Cresswell, S., Vilarino-Guell, C., Encarnacion, M., Sherman, H., Yu, I., Shah, (2012). Tafamidis, a potent and selective transthyretin kinetic stabilizer that B., et al. (2013). Alpha-synuclein p.H50Q, a novel pathogenic mutation for inhibits the amyloid cascade. Proc. Natl. Acad. Sci. U.S.A. 109, 9629–9634. Parkinson’s disease. Mov. Disord. 28, 811–813. doi: 10.1002/mds.25421 doi: 10.1073/pnas.1121005109 Barria, M. A., Gonzalez-Romero, D., and Soto, C. (2012). Cyclic amplification of Cao, S., Gelwix, C. C., Caldwell, K. A., and Caldwell, G. A. (2005). Torsin-mediated prion protein misfolding. Methods Mol. Biol. 849, 199–212. doi: 10.1007/978-1- protection from cellular stress in the dopaminergic neurons of Caenorhabditis 61779-551-0_14 elegans. J. Neurosci. 25, 3801–3812. doi: 10.1523/jneurosci.5157-04.2005 Bendor, J. T., Logan, T. P., and Edwards, R. H. (2013). The function of alpha- Carija, A., Pinheiro, F., Pujols, J., Bras, I. C., Lazaro, D. F., Santambrogio, C., et al. synuclein. Neuron 79, 1044–1066. doi: 10.1016/j.neuron.2013.09.004 (2019). Biasing the native alpha-synuclein conformational ensemble towards Bousset, L., Pieri, L., Ruiz-Arlandis, G., Gath, J., Jensen, P. H., Habenstein, B., compact states abolishes aggregation and neurotoxicity. Redox Biol. 22:101135. et al. (2013). Structural and functional characterization of two alpha-synuclein doi: 10.1016/j.redox.2019.101135 strains. Nat. Commun. 4:2575. doi: 10.1038/ncomms3575 Crespo, R., Villar-Alvarez, E., Taboada, P., Rocha, F. A., Damas, A. M., and Martins, Braak, H., Del Tredici, K., Rub, U., de Vos, R. A., Jansen Steur, E. N., and Braak, P. M. (2016). What can the kinetics of amyloid fibril formation tell about off- E. (2003). Staging of brain pathology related to sporadic Parkinson’s disease. pathway aggregation? J. Biol. Chem. 291, 2018–2032. doi: 10.1074/jbc.M115. Neurobiol. Aging 24, 197–211. doi: 10.1016/s0197-4580(02)00065-9 699348 Frontiers in Molecular Neuroscience | www.frontiersin.org 11 December 2019 | Volume 12 | Article 306 Peña-Díaz et al. ZPD-2 Inhibits α-Synuclein Aggregation Dehay, B., Bourdenx, M., Gorry, P., Przedborski, S., Vila, M., Hunot, S., Peelaerts, W., Bousset, L., Van der Perren, A., Moskalyuk, A., Pulizzi, R., Giugliano, et al. (2015). Targeting alpha-synuclein for treatment of Parkinson’s disease: M., et al. (2015). alpha-Synuclein strains cause distinct synucleinopathies mechanistic and therapeutic considerations. Lancet. Neurol. 14, 855–866. doi: after local and systemic administration. Nature 522, 340–344. doi: 10.1038/ 10.1016/S1474-4422(15)00006-X nature14547 Dexter, D. T., and Jenner, P. (2013). Parkinson disease: from pathology to Perni, M., Galvagnion, C., Maltsev, A., Meisl, G., Muller, M. B., Challa, P. K., et al. molecular disease mechanisms. Free Radic. Biol. Med. 62, 132–144. doi: 10. (2017). A natural product inhibits the initiation of alpha-synuclein aggregation 1016/j.freeradbiomed.2013.01.018 and suppresses its toxicity. Proc. Natl. Acad. Sci. U.S.A. 114, E1009–E1017. Gao, J., Perera, G., Bhadbhade, M., Halliday, G. M., and Dzamko, N. (2019). Polymeropoulos, M. H., Lavedan, C., Leroy, E., Ide, S. E., Dehejia, A., Dutra, A., Autophagy activation promotes clearance of alpha-synuclein inclusions in et al. (1997). Mutation in the alpha-synuclein gene identified in families with fibril-seeded human neural cells. J. Biol. Chem. 294, 14241–14256. doi: 10.1074/ Parkinson’s disease. Science 276, 2045–2047. doi: 10.1126/science.276.5321. jbc.RA119.008733 2045 Hamamichi, S., Rivas, R. N., Knight, A. L., Cao, S., Caldwell, K. A., and Caldwell, Pujols, J., Pena-Diaz, S., Conde-Gimenez, M., Pinheiro, F., Navarro, S., Sancho, G. A. (2008). Hypothesis-based RNAi screening identifies neuroprotective J., et al. (2017). High-throughput screening methodology to identify alpha- genes in a Parkinson’s disease model. Proc. Natl. Acad. Sci. U.S.A. 105, 728–733. synuclein aggregation inhibitors. Int. J. Mol. Sci. 18:E478. doi: 10.3390/ doi: 10.1073/pnas.0711018105 ijms18030478 Hansen, C., Angot, E., Bergstrom, A. L., Steiner, J. A., Pieri, L., Paul, G., et al. (2011). Pujols, J., Pena-Diaz, S., Lazaro, D. F., Peccati, F., Pinheiro, F., Gonzalez, D., et al. alpha-Synuclein propagates from mouse brain to grafted dopaminergic neurons (2018). Small molecule inhibits alpha-synuclein aggregation, disrupts amyloid and seeds aggregation in cultured human cells. J. Clin. Invest. 121, 715–725. fibrils, and prevents degeneration of dopaminergic neurons. Proc. Natl. Acad. doi: 10.1172/JCI43366 Sci. U.S.A. 115, 10481–10486. doi: 10.1073/pnas.1804198115 Harrington, A. J., Yacoubian, T. A., Slone, S. R., Caldwell, K. A., and Sabate, R., Gallardo, M., and Estelrich, J. (2003). An autocatalytic reaction as a Caldwell, G. A. (2012). Functional analysis of VPS41-mediated neuroprotection model for the kinetics of the aggregation of beta-amyloid. Biopolymers 71, in Caenorhabditis elegans and mammalian models of Parkinson’s disease. 190–195. doi: 10.1002/bip.10441 J. Neurosci. 32, 2142–2153. doi: 10.1523/jneurosci.2606-11.2012 Sant’Anna, R., Gallego, P., Robinson, L. Z., Pereira-Henriques, A., Ferreira, N., Hauser, R. A. (2015). alpha-Synuclein in Parkinson’s disease: getting to the core of Pinheiro, F., et al. (2016). Repositioning tolcapone as a potent inhibitor of the matter. Lancet. Neurol. 14, 785–786. doi: 10.1016/s1474-4422(15)00136-2 transthyretin amyloidogenesis and associated cellular toxicity. Nat. Commun. Herva, M. E., Zibaee, S., Fraser, G., Barker, R. A., Goedert, M., and Spillantini, 7:10787. doi: 10.1038/ncomms10787 M. G. (2014). Anti-amyloid compounds inhibit alpha-synuclein aggregation Serpell, L. C., Berriman, J., Jakes, R., Goedert, M., and Crowther, R. A. (2000). Fiber induced by protein misfolding cyclic amplification (PMCA). J. Biol. Chem. 289, diffraction of synthetic alpha-synuclein filaments shows amyloid-like cross-beta 11897–11905. doi: 10.1074/jbc.M113.542340 conformation. Proc. Natl. Acad. Sci. U.S.A. 97, 4897–4902. doi: 10.1073/pnas. Ibanez, P., Bonnet, A. M., Debarges, B., Lohmann, E., Tison, F., Pollak, P., et al. 97.9.4897 (2004). Causal relation between alpha-synuclein gene duplication and familial Silva, B., Einarsdóttir, O., Fink, and Uversky, V. (2011). Modulating α-synuclein Parkinson’s disease. Lancet 364, 1169–1171. doi: 10.1016/s0140-6736(04) misfolding and fibrillation in vitro by agrochemicals. Res. Rep. Biol. 2, 43–56. 17104-3 Singleton, A. B., Farrer, M., Johnson, J., Singleton, A., Hague, S., Kachergus, J., et al. Kim, H., Calatayud, C., Guha, S., Fernandez-Carasa, I., Berkowitz, L., Carballo- (2003). alpha-Synuclein locus triplication causes Parkinson’s disease. Science Carbajal, I., et al. (2018). The small GTPase RAC1/CED-10 is essential 302:841. doi: 10.1126/science.1090278 in maintaining dopaminergic neuron function and survival against alpha- Spillantini, M. G., Crowther, R. A., Jakes, R., Cairns, N. J., Lantos, P. L., synuclein-induced toxicity. Mol. Neurobiol. 55, 7533–7552. doi: 10.1007/ and Goedert, M. (1998). Filamentous alpha-synuclein inclusions link s12035-018-0881-7 multiple system atrophy with Parkinson’s disease and dementia with Kruger, R., Kuhn, W., Muller, T., Woitalla, D., Graeber, M., Kosel, S., et al. (1998). Lewy bodies. Neurosci. Lett. 251, 205–208. doi: 10.1016/s0304-3940(98)00 Ala30Pro mutation in the gene encoding alpha-synuclein in Parkinson’s disease. 504-7 Nat. Genet. 18, 106–108. Spillantini, M. G., Schmidt, M. L., Lee, V. M., Trojanowski, J. Q., Jakes, R., and Levin, J., Schmidt, F., Boehm, C., Prix, C., Botzel, K., Ryazanov, S., et al. (2014). Goedert, M. (1997). Alpha-synuclein in lewy bodies. Nature 388, 839–840. The oligomer modulator anle138b inhibits disease progression in a Parkinson Sulston, J., Dew, M., and Brenner, S. (1975). Dopaminergic neurons in the mouse model even with treatment started after disease onset. Acta Neuropathol. nematode Caenorhabditis elegans. J. Comp. Neurol. 163, 215–226. 127, 779–780. doi: 10.1007/s00401-014-1265-3 Tatenhorst, L., Eckermann, K., Dambeck, V., Fonseca-Ornelas, L., Walle, H., Lopes Li, B., Ge, P., Murray, K. A., Sheth, P., Zhang, M., Nair, G., et al. (2018). Cryo-EM of da Fonseca, T., et al. (2016). Fasudil attenuates aggregation of alpha-synuclein in full-length alpha-synuclein reveals fibril polymorphs with a common structural models of Parkinson’s disease. Acta Neuropathol. Commun. 4:39. doi: 10.1186/ kernel. Nat. Commun. 9:3609. doi: 10.1038/s41467-018-05971-2 s40478-016-0310-y Marti, M. J., Tolosa, E., and Campdelacreu, J. (2003). Clinical overview of the van Ham, T. J., Thijssen, K. L., Breitling, R., Hofstra, R. M., Plasterk, R. H., synucleinopathies. Mov. Disord. 18(Suppl. 6), S21–S27. and Nollen, C. (2008). elegans model identifies genetic modifiers of alpha- Marvian, A. T., Koss, D. J., Aliakbari, F., Morshedi, D., and Outeiro, T. F. (2019). synuclein inclusion formation during aging. PLoS Genet. 4:e1000027. doi: 10. In vitro models of synucleinopathies: informing on molecular mechanisms and 1371/journal.pgen.1000027 protective strategies. J. Neurochem. 150, 535–565. doi: 10.1111/jnc.14707 Winner, B., Jappelli, R., Maji, S. K., Desplats, P. A., Boyer, L., Aigner, S., et al. McCormack, A. L., Mak, S. K., Henderson, J. M., Bumcrot, D., Farrer, M. J., and Di (2011). In vivo demonstration that alpha-synuclein oligomers are toxic. Proc. Monte, D. A. (2010). Alpha-synuclein suppression by targeted small interfering Natl. Acad. Sci. U.S.A. 108, 4194–4199. doi: 10.1073/pnas.1100976108 RNA in the primate substantia nigra. PLoS One 5:e12122. doi: 10.1371/journal. Xue, W. F., Hellewell, A. L., Hewitt, E. W., and Radford, S. E. (2010). Fibril pone.0012122 fragmentation in amyloid assembly and cytotoxicity: when size matters. Prion Morales, R., Duran-Aniotz, C., Diaz-Espinoza, R., Camacho, M. V., and Soto, C. 4, 20–25. doi: 10.4161/pri.4.1.11378 (2012). Protein misfolding cyclic amplification of infectious prions. Nat. Protoc. 7, 1397–1409. doi: 10.1038/nprot.2012.067 Conflict of Interest: The authors declare that the research was conducted in the Moree, B., Yin, G., Lazaro, D. F., Munari, F., Strohaker, T., Giller, K., et al. absence of any commercial or financial relationships that could be construed as a (2015). Small molecules detected by second-harmonic generation modulate the potential conflict of interest. conformation of monomeric alpha-synuclein and reduce its aggregation in cells. J. Biol. Chem. 290, 27582–27593. doi: 10.1074/jbc.M114.636027 Copyright © 2019 Peña-Díaz, Pujols, Conde-Giménez, Čarija, Dalfo, García, Munoz-Lobato, F., Rodriguez-Palero, M. J., Naranjo-Galindo, F. J., Shephard, F., Navarro, Pinheiro, Santos, Salvatella, Sancho and Ventura. This is an open-access Gaffney, C. J., Szewczyk, N. J., et al. (2014). Protective role of DNJ-27/ERdj5 in article distributed under the terms of the Creative Commons Attribution License Caenorhabditis elegans models of human neurodegenerative diseases. Antioxid. (CC BY). The use, distribution or reproduction in other forums is permitted, provided Redox Signal. 20, 217–235. doi: 10.1089/ars.2012.5051 the original author(s) and the copyright owner(s) are credited and that the original Nussbaum, R. L., and Ellis, C. E. (2003). Alzheimer’s disease and Parkinson’s publication in this journal is cited, in accordance with accepted academic practice. No disease. N. Engl. J. Med. 348, 1356–1364. use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Molecular Neuroscience | www.frontiersin.org 12 December 2019 | Volume 12 | Article 306