1 INTRODUCTION Parkinson’s disease (PD) is the second most prevalent neurodegenerative disease and the most common movement disorder, affecting millions of people worldwide [1]. Besides hallmark motor signs, which include tremor, rigidity, bradykinesia and postural instability, PD patients develop a number of important and debilitating non-motor symptoms. Olfactory deficits, anxiety, depression, deficits on working, spatial and motor memories, psychosis, and sleep disturbances are amongst the most common non-motor symptoms, which precede motor dysfunction by up to four years [2-4]. No current treatment can halt or effectively slow down neurodegeneration in PD, and the development of suitable animal models that recapitulate key features of the disease is a critical step towards identification of novel therapeutic strategies. The neuropathology of PD involves extensive loss of dopaminergic neurons in the substantia nigra (SN) and accumulation of intracellular aggregated protein deposits known as Lewy Bodies (LB) [5]. The most abundant component of LB is -synuclein (α-syn), a 140 amino acid-long presynaptic protein [6-8]. When bound to the pre-synaptic membrane, α-syn has a predominantly αhelical structure. However, it can alternatively fold into a β- sheet-rich structure that easily polymerizes into amyloid fibrils and aggregates [9, 10]. Even though the causes of biophysical alterations in α-syn structure remain incompletely understood, extensive evidence suggests that this is a key event leading to dopaminergic neurodegeneration in PD patients and animal models [11-13]. Animal models of PD have been useful in providing information concerning the etiology, pathology and molecular mechanisms of this devastating disease. Experimental models of PD have been useful for testing symptom-relieving therapies or studying neurodegeneration as seen in late stages of PD [14-17]. The most commonly used rodent models of PD involve transgenic animals or the administration of neurotoxins, such as MPTP and 6-OH dopamine, that cause selective degeneration of dopaminergic neurons. Although some neurotoxin-based models are able to reproduce both non-motor and motor symptoms seen in PD patients [18, 19], they are unable to mimic the temporal sequence of these symptoms [14, 16] and usually fail to induce LB-like α-syn pathology [14-17]. On the other hand, genetically modified mice overexpressing either wild-type or mutant human α-syn successfully mimic rare familial forms of early-onset PD and are associated with extensive brain deposits of fibrillary α-syn, but none exhibit the typical degeneration of neurons in the SN [20-23]. Recent studies demonstrate that injection of α-syn fibrils into the striatum of non-transgenic rodents seeds the pathological structural conversion of endogenous α-syn, resulting in progressive loss of dopaminergic neurons in the SN and motor deficits [24-26]. Despite the importance of those studies in establishing a mechanistic link between transmission of pathological α-syn and motor symptoms, the relation between non-motor deficits and brain accumulation of α-syn has not been evaluated. Moreover, recent findings suggest that intermediate-sized aggregates are more neurotoxic than the α-syn fibrils or LB deposits, leading to the hypothesis that formation of such oligomeric assemblies is an early and central event in PD pathogenesis [27-29]. Here, we show that wild-type α-syn oligomers (-SYOs) impair dopamine release from midbrain dopaminergic neurons. In non-transgenic adult mice, a single infusion of -SYOs into the lateral brain ventricle evoked non-motor and motor symptoms in a temporal sequence reminiscent of the clinical development of PD. We further show that an early consequence of intracerebroventricular (i.c.v.) infusion of -SYOs is a marked reduction in dopamine content in the olfactory bulb, which is accompanied by early and persistent olfactory dysfunction. Motor control and striatal dopaminergic content are only affected at later time points in -SYO-injected mice. This model holds potential as a tool to unravel molecular mechanisms of pathogenesis, and for drug screening and development of novel therapeutics in PD.

2 METHODS 2

Expression, purification and oligomerization of recombinant human α- syn. E. coli strain BL21DE3pLysS (in 500 mL of LB culture medium) was transformed with the pRK172 plasmid coding for recombinant human α-syn (a kind gift from Prof. John Trojanowski, University of Pennsylvania), and was grown for 16 hours at 37 °C and 250 rpm without induction. Purification followed an established protocol [30] including two cycles of freeze-thawing, sonication, boiling and two steps of ammonium sulfate precipitation, followed by ion-exchange chromatography (QSepharose; GE Healthcare, Little Chalfont, UK). The most concentrated fractions were bufferexchanged into 10 mM Tris-Cl, pH 7.4, quantified considering an absorption coefficient of 5,600 M-1cm-1 at 274 nm and a molecular weight of 14,462 Da, lyophilized overnight and stored at -80 °C. For oligomerization, lyophilized α-syn fractions were resuspended at 6 g/L in PBS (150 mM NaCl, 6.5 mM sodium phosphate, pH 7.2) and incubated at 65 °C, without agitation and in the dark, for 3-5 days [30]. Alternatively, syn was oligomerized following previously described [31]. In brief, syn precipitated from boiled E. coli lysate using streptomycin and ammonium sulfate was resuspended in 100 mM ammonium acetate, and dialyzed against deionized water. Lyophilized aliquots were resuspended at 2 mg/ml in PBS and incubated at 37 °C for 48h. Characterization of α-syn oligomers (α-SYOs). Presence of oligomers in the preparations obtained by either of the two protocols described above was determined by size exclusion high-performance liquid chromatography (SEC-

HPLC) on a GPC-100 column (Eprogen, Downers Grove, IL), detecting both absorption at 214 nm and tyrosine fluorescence (excitation at 275 nm, emission at 305 nm). Samples were centrifuged at 14,100×g for 1 min before chromatography. For isolation of oligomer-enriched fractions, 100 μL of α-syn preparation was chromatographed on the same column, and 100 μL fractions were collected after passage by the fluorescence detector. Ten microliters (10 μL) from each fraction were analyzed on the same column to determine the proportion of oligomers and monomers, and 25 μL were used for protein determination (BCA assay; Thermo-Fisher Scientific, Rockford, IL). The presence of oligomers was also confirmed by dot blots. Briefly, 1 µg of protein was spotted onto a nitrocellulose membrane and allowed to dry at room temperature. Blots were incubated with 5% nonfat dry milk in Tween- TBS at room temperature for 1 h and were then incubated for 2 h with oligomer-specific A11 primary antibody [32] (Millipore EMD, Billerica, MA) diluted 1:200 in blocking solution. Blots were washed and incubated with horseradish peroxidase-conjugated secondary antibody (1:50,000) at room temperature for 1 h, revealed using SuperSignal West Pico chemiluminescent detection (Thermo Scientific), imaged and analyzed using NIH Image J (Windows version). For Western blots, 30 µg protein were prepared in sample buffer containing SDS, resolved (without previous boiling) in 4–20% polyacrylamide Tris-glycine gels (Invitrogen) and electrotransferred to nitrocellulose membranes at 300 mA for 1 h. Blots were incubated with 5% nonfat dry milk in Tween-TBS at room temperature for 2 h and were incubated at 4 °C overnight with anti-human α-syn primary antibody (Syn211; Invitrogen, Carlsbad, CA) diluted 1:200 in blocking buffer. Membranes were incubated

with horseradish peroxidase-conjugated secondary antibody (1:50,000) at room temperature for 2 h, probed with SuperSignal West Pico chemiluminescent detection (Thermo Fisher Scientific), imaged and quantified using NIH Image J. 3

Animals. Male Swiss mice used for i.c.v. injections were obtained from our animal facility and were 2.5-3 month-old at the beginning of experiments. Animals were housed in groups of five in each cage with free access to food and water, under a 12 h light/dark cycle, with controlled room temperature and humidity. Wistar rats were mated overnight and the formation of a vaginal plug was used as an indication of successful mating, after which a control of embryonic date was performed. All procedures used in the present study followed the ``Principles of Laboratory Animal Care'' (US National Institutes of Health) and were approved by the Institutional Animal Care and Use Committee of the Federal University of Rio de Janeiro (protocol # FARMACIA 10-05/16). Four independent groups of animals were used: -Group 1 was submitted to olfactory discrimination 4 dpi, object recognition 20 dpi, Vanilin essence discrimination 40 dpi, pole test 45 dpi and were killed at 45 dpi for immunohistochemistry assays. -Group 2 was submitted to object recognition 3 dpi, rotarod 4 dpi and killed at 8 dpi for dopamine quantification. -Group 3 was submitted to pole test 4 dpi, open field and elevated plus maze test 20 dpi, olfactory discrimination 45 dpi and were killed 45 dpi for dopamine quantification. -Group 4 was submitted to the forced swim test only, as this test can interfere with behavior in subsequent tests. Experiments performed in Groups 1, 2 and 3 were further confirmed in new groups of animals submitted to identical treatment and behavioral evaluation. Ventral midbrain neuronal cultures. Primary rat ventral midbrain cultures were obtained from embryonic day 14 (E14) rat embryos of both sexes. Embryos were decapitated and the ventral midbrain was dissected out using a standard method [33]. With this method, we could maintain cultures of ventral midbrain neurons for 21 days containing ≥10% tyrosine hydroxylase positive (TH+) neurons (Supplementary Figure 1A). Assays described in this study were performed using cultures after 12-15 days in vitro. All procedures were approved by the Institutional Animal Care and Use Committee of the Federal University of Rio de Janeiro. For dopamine release assays, cultures were treated with 50 pmols (added to 2 mL culture medium/well) of -SYOs or -syn monomers (-SYMs) for 24 h. After this period, 1 mM tyrosine or 0.5 mM L-dopa were added to the cultures for 2 h. Cultures were then washed five times with calcium-free artificial cerebrospinal fluid (aCSF) (126 mM NaCl, 12 mM MgCl2, 2.5 mM KCl, 21.4 mM NaHCO3, 11.1 mM glucose, 1.2 mM NaH2PO4, 400 µM ascorbic acid, pH 7.4). Fresh DMEM/F12 was added to culture wells, and the entire medium was collected/replaced every ten minutes. Baseline levels of dopamine in the medium were compared to levels obtained when cultures were stimulated by 200 µM glutamate [33]. Perchloric acid (0.1 M final concentration) was added to collected media and to cultured cells. Determination of dopamine levels was 4

performed by HPLC-ED, as described above. Total dopamine levels were defined as the sum of released (basal and stimulated) plus remaining cell- associated dopamine content. Values were normalized to total protein content in the corresponding sample, measured using the BCA kit. Intracerebroventricular infusion of -syn oligomers (α-SYOs) and monomers (α-SYMs). For i.c.v. infusion of α-syn, animals were anesthetized for 7 min with 2.5% isoflurane (Cristália, São Paulo, Brazil) using a vaporizer system (Norwell, MA) and were gently restrained only during the injection procedure itself, as recently described [34]. A 2.5 mm-long needle was inserted 1 mm to the right of the midline point equidistant from each eye and parallel to a line drawn through the anterior base of the eye to reach lateral ventricle of mice [34, 35]. Fifty (50) pmol of α- syn monomers or oligomers (concentration expressed as monomers) in PBS were injected in a total volume of 3 µL. Dose was chosen based on preliminary experiments from our group, which suggested that lower doses of α-SYOs (15 pmol) evoked early olfactory deficits but had no effect on motor capacity of mice (data not shown). Control mice received the same volume of vehicle (PBS). After injection, the needle was kept in place for at least 30 seconds to avoid backflow. At the end of experiments, injection of methylene blue was employed to verify the accuracy of injection into the lateral ventricle. Mice showing any signs of misplaced injections or brain hemorrhage (~ 5% of animals throughout our study) were excluded from further analysis. In all experimental group had no mortality. Behavioral tests started from the second day of experiment. Each behavioral experiment was replicated at least twice, and graphs show data from one representative experiment. Mice submitted to olfactory evaluation at 4 dpi were also submitted to open field and novel object recognition tests 20 dpi, and to evaluation of motor function at later timepoints (Rotarod at 30 dpi and pole test at 45 dpi). Mice submitted to pole test 4 dpi were also submitted to novel object recognition 3 dpi, elevated plus maze test (20 dpi) and late olfactory function assessment (45 dpi). Independent groups of mice were used for evaluation of depressive-like behavior in the forced-swim task (4 dpi) and for biochemical experiments. Familiar Bedding olfactory discrimination task. All sessions were performed between 8 am and 2 pm. The olfactory discrimination protocol was performed as described [36]. Briefly, the task consisted of placing each mouse for 5 min in a box divided into two identical compartments (30 × 30 × 20 cm each) separated by an open door. The box was placed in a room with dim light and no background noise. A handful of unchanged sawdust was taken from the animals’ housing cage and placed in one of the compartments (familiar compartment) and the same amount of fresh clean sawdust was placed in the other compartment (non-familiar compartment). Animals were allowed to choose between exploring these compartments, and time spent in each one of them was recorded by an experienced researcher. Healthy mature male mice are able to discriminate between the familiar and the non- familiar compartments based upon the odor of the sawdust, spending significantly more time in the familiar compartment, as they prefer their own odor compared to a neutral odor. Vanillin discrimination test. All sessions were performed between 8 and 12 am. This test was adapted from Petit and colleagues [37]. Mice were placed in an arena measuring 30 x 30 x 45 cm, which contained two cartridges placed on opposite sides. One cartridge was filed with water and the other with artificial vanillin odor (1:10,000 in water). Total time spent sniffing each cartridge was quantified in a 5-min session, and it is known that normal mice instinctively spend more time sniffing the new odor. The arena and objects were thoroughly cleaned between trials with 20% ethanol to eliminate olfactory cues and both cartridges were changed every three sessions for new ones with fresh essence. Results were expressed as discrimination index between vanillin essence and water [37]. Open field task. All sessions were performed between 8 and 12 am. For the open field test, each animal was placed in the center of an arena made of wood (30 x 30 x 45 cm) and divided into quadrants by lines drawn on the floor. Total locomotor activity and time spent in the center and 5

periphery of the box were recorded for 5 minutes. The arena was thoroughly cleaned with 20% ethanol in between trials to eliminate olfactory cues and illuminated with an indirect source of light (~100 lux). The longer the amount of time mice spend in the central area of the box, the less anxious they can be considered [38]. Novel object recognition test. All sessions were performed between 8 am and 3pm. The test was carried out in an arena measuring 30 x 30 x 45 cm. Test objects were made of glass or plastic and had different shapes, colors, sizes and textures. During behavioral sessions, objects were fixed to the box using tape to prevent displacement caused by exploratory activity of the animals. Preliminary tests showed that none of the objects used in our experiments evoked innate preference. Before training, each animal was submitted to a 5 minute-long habituation session, in which it was allowed to freely explore the empty arena. Training consisted in a 5 minute-long session during which animals were placed at the center of the arena in the presence of two identical objects. The amount of time spent exploring each object was recorded by a trained researcher. Sniffing and touching the object were considered as exploratory behavior. The arena and objects were cleaned thoroughly with 20% ethanol in between trials to eliminate olfactory cues. Two hours after training, animals were again placed in the arena for the test session, in which one of the objects used in the training session was replaced by a new one. Again, the amount of time spent exploring familiar and novel objects was measured. Results were expressed as percentage of time exploring each object during the training or test sessions [34]. Pole test. All sessions were performed between 8 and 12 am. Animals were placed head-up at the top of a vertical wooden pole (height: 50 cm; diameter: 1.5 cm) covered by rough clothing. The base of the pole was placed in a mouse homecage containing clean sawdust. When placed at the top of the pole, animals orient themselves downwards and descend the length of the pole back into the cage. Animals underwent 2 days of training with five trials each. On the third day, animals were again placed at the top of the pole and the amount of time to orient downwards (t-turn) and the total time to descend the pole (t-descend) were measured; the three best performances of each animal were selected and used for statistical analysis, as described [39, 40]. Rotarod test. All sessions were performed between 8 and 12 am. Mice were initially subjected to a 300 second training session in order to learn how to keep their balance on the slowly rotating rod (16 rpm). After one hour, mice were again placed in the rod, and the amount of time (latency) until a mouse fell off the rod rotating under continuous acceleration (increasing every 30 sec, from 37 to 46 rpm in ~5 min) was measured [41]. Forced-swim test. All sessions were performed between 8 and 12 am. One day prior to the experiment, mice were individually placed in a cylindrical tank (diameter 16 cm, height 25 cm) filled with water at room temperature for 15 min. The volume of water in the tank was such that animals’ tails would not touch the bottom and so that animals would not be able to climb out from the tank. Twenty-four hours later, mice were placed in the same tank and immobility time was recorded by an experienced researcher for 6 min [42]. Water in the tanks was replaced in between trials. Elevated Plus maze test. The elevated plus maze apparatus consisted of a platform comprising two open and two closed arms (each 4.5 cm wide, 30 cm in length), connected by a central square (6 × 6 cm), 60 cm high from the ground. All sessions were performed between 8 and 12 am, and mice were individually placed at the center of the apparatus facing one of the closed arms. During the 5min long sessions, an experienced researcher recorded the number of entries and time each animal spent in the open and closed arms of the maze [43]. Measurement of dopamine content. Brain monoamines were measured by HPLC coupled with electrochemical detection (HPLC-ED). Briefly, animals were decapitated, brains were removed, and olfactory bulb and striatum were dissected out in ice-cold PBS. Perchloric acid was added to each sample to a final concentration of 0.1 M. Samples were sonicated (2 pulses of 5 s, 50 Hz) and centrifuged (10,000×g) for 10 min to remove precipitated proteins. Supernatants were used for 6

HPLC analysis. For normalization, pellets were resuspended and protein contents were quantified with a BCA kit (Thermo- Fisher Scientific, Rockford, IL). Fast isocratic separation was obtained using a reverse phase LC-18 column (4.6 × 250 mm; Sigma) with the following mobile phase: 20 mM disodium hydrogen phosphate, 20 mM citric acid, 10% methanol, 0.12 mM Na2EDTA, and 566 mg/L 1-heptanesulfonic acid, pH 2.64 [44]. Immunohistochemistry. Mice were perfused with saline followed by 4% PFA. Brains were collected, cryoprotected in a sucrose gradient and frozen at -20 oC. Frozen tissue specimens were cut at 10 μm on a cryostat (Leica) and permeabilized with 0.5% Triton X-100 for 30 min at room temperature. Nonspecific sites were blocked with 3% bovine serum albumin (BSA; Sigma, St. Louis, MO), 5% normal goat serum (NGS; Sigma, St. Louis, MO) in TBS for 1 h before immunoreaction with the following antibodies: rabbit anti- tyrosine hydroxylase (1:500; Millipore, Darmstadt, Germany) and mouse anti- dopamine transporter (1:500; Abcam, Inc., Cambridge, UK). After overnight incubation with primary antibody, tissue was washed with TBS and incubated with secondary antibodies (Alexa Fluor 546-conjugated goat antirabbit IgG, 1:1,000; Alexa Fluor 488-conjugated goat anti-mouse IgG, 1:300; biotinylated goat anti-rabbit IgG, 1:300, Molecular Probes, Paisley, UK) for 2 h at room temperature. For DAT and TH analysis in caudate putament, nuclei were counterstained with DAPI (Sigma, St. Louis, MO). Sections were mounted with Mounting Media (DAKO Cytomation, Glostrup, Denmark) and imaged on a confocal microscope (Leica TCS SPE). Densitometry of immunohistochemistry images was performed using integrated density values generated in Image J software (National Institutes of Health, USA). Analyses were performed in at least 2-3 slices from each animal and 3 images per section at 40X magnification were obtained in the caudate-putamen. For immunohistochemistry of TH in substantia nigra (SN), slices were incubated with biotinylated secondary antibodies for 1 hour at room temperature, washed twice with PBS and incubated with streptavidin-biotin-peroxidase (Vector Laboratories, CA) for 45 min. Slides were then covered with 3,3’-diaminobenzidine solution (0.06% DAB in PBS containing 2% DMSO and 0.018% H2O2) for 1 to 5 min or until a brown precipitate could be observed. Identical conditions and reaction times were used for slides from different animals (run in parallel), and reaction was stopped by immersion of slides in distilled water. Slides were imaged using a Sight DS-5M-L1 digital camera (Nikon, Melville, NY) connected to an Eclipse 50i light microscope (Nikon) at 200x magnification. For quantification of TH-positive cells, we evaluated 4 sagittal slices (with interval of 100 m between slices) of each animal, counting 2-4 fields of SN of each slice, using 100x magnification. The number of Iba-1 positive cells was determined using NIH ImageJ 1.36b imaging software (NIH, Bethesda, MD). Data are expressed as number of TH-positive cells per mm2. Data analysis. No previous statistical calculation was employed to determine sample size. Instead, sample sizes in our experiments were chosen based on usual procedures and best practices in the field. Each behavioral experiment was replicated at least twice, and graphs show data from one representative experiment. All analyses were performed with GraphPad Prism6® (GraphPad Software, La Jolla, CA). Values are expressed as means ± SEM. Gaussian distribution of data was assessed using the D’Agostino-Pearson normality test. Experiments of dopamine release in mesencephalic midbrain neuronal cultures were analyzed by two-way ANOVA followed by Bonferroni. Data from object recognition, olfactory discrimination and vanillin recognition test experiments are expressed as Discrimination Index as described in [45], and difference between groups was assessed by performing one-way ANOVA followed by Tukey. DAT and TH immunofluorescence intensity was analyzed by Student's t test. In all other experiments, data were analyzed by One-way ANOVA followed by Tukey. In the Supplementary Table 1 we show a Table with F and P values for behavioral data.

3 Results 7

1.1 α-synuclein

oligomers production

(α-SYOs)

impair

dopamine

and release from midbrain neurons in

culture Human α-SYOs were routinely characterized and isolated by size- exclusion chromatography (SEC-HPLC). The main peak in our preparations eluted at approximately 2.0 mL, and was identified as monomeric -syn (α- SYMs) by comparison with the elution profile of purified α-syn not submitted to the oligomerization protocol (Fig. 1A). In addition to the monomer fraction, chromatograms revealed a smaller peak eluting at ~1.8 mL and corresponding to 5-10% of the total soluble protein (Fig. 1A, red arrow). This fraction was collected and reinjected into the SEC-HPLC system, and consisted essentially of -SYOs (Fig. 1B). Characterization of the -SYOs fraction was further performed by dot immunoblot using antioligomer antibody A11 (Fig. 1C) and by Western blot probed with an anti-human α-syn antibody (Fig. 1D). The oligomer-enriched fraction comprised mostly dimers (~30 KDa) and larger oligomers likely corresponding to trimers/tetramers (49-64 kDa), while the monomer fraction was devoid of larger aggregates and consisted almost entirely of a ~15 kDa protein band, corresponding to the expected molecular mass for α-syn monomers (Fig. 1D). The biological actions of α-SYOs and α-SYMs were initially investigated in primary cultures of rat midbrain neurons. Cultures were exposed to 25 μM -SYMs or -SYOs for 24 h, followed by 2 h incubation with L-tyrosine or L-dopa. Dopamine release was then evoked by stimulation with glutamate. Using L-tyrosine as substrate for dopamine production, we found that -SYOs significantly impaired dopamine release to the medium (Fig. 2A; F(2,75)=22.89; **p=0.006 and ***p<0.0001). Total dopamine content in the cells was also diminished upon exposure to -SYOs (Fig 2B; F(2,15)=1.064; ***p<0.0001). -SYMs had no effect on TH-dependent dopamine release (Fig. 2A, grey line), but induced a slight decrease in total cell-associated dopamine content (Fig. 2B; *p=0.0019). Interestingly, using L-dopa as substrate, SYM- or -SYO-exposed neurons exhibited normal dopamine release (Fig. 2C; F(4,111)=0.3121). These results indicate that -SYOs specifically impair the activity of tyrosine hydroxylase. Although -SYMs or -SYOs had no effect on cellular viability in midbrain cultures (Supplementary Fig. 1B, C; F(2,19)=1.053), total dopamine content was reduced in -SYO-treated cultures when L-dopa was used as dopamine precursor (Fig. 2D; F(2,27)=9.924, p=0.0004). 1.2 α-Synuclein oligomers (α -SYOs) induce early and persistent olfactory dysfunction in mice We next investigated the effect of α-SYOs in mice by injecting an oligomer-enriched fraction (50 pmol -SYOs, isolated by SEC-HPLC as shown in Fig. 1A) into the lateral brain ventricle of mice as described [34]. Because impaired olfactory function is a frequent and early symptom in PD patients, animals were subjected to an olfactory discrimination task four days after the injection (4 dpi). This task is based on the innate preference of mice for the smell of their own homecage when compared to a neutral odor. α-SYO-injected mice failed to recognize the familiar odor, spending comparable amounts of time in the compartments corresponding to familiar and non- familiar odors (Fig. 3A; F(2,25)=0.3219; *p=0.0111). Similar to control (vehicle-injected) animals, mice injected with α-SYMs explored the familiar compartment for a significantly longer amount of time, showing that monomeric α-syn does not impair olfactory information processing. In an attempt to elucidate the biochemical mechanisms underlying this result, and considering that the olfactory bulb is crucial for odor recognition and discrimination, we measured dopamine and 8

tyrosine hydroxylase (TH) levels in the olfactory bulb of vehicle- or -SYO-infused mice. We found that the oligomer-enriched -syn fraction (isolated by SEC-HPLC) induced a marked decrease in dopamine levels in the olfactory bulb (Fig. 3B; F(2,22)=9.372; p=0.0042), whereas αsyn monomers had no effect. Consistent with a reduction in dopamine levels, infusion of -SYOs induced a decrease in TH immunoreactivity in the glomerular layer (GL) of the olfactory bulb compared with vehicle-injected mice (Fig. 3C). We further investigated the olfactory function of mice in a paradigms less influenced by anxiety and other unspecific behaviors in mice, the vanillin essence identification task. In this task, performed 40 days after i.c.v. injections, -SYO-injected mice also showed an impaired sense of smell (Fig. 3D; F(2,20)=4.058; #p=0.0764). However, SYMs also induced an impaired performance in the vanillin task (*p=0.0401). To determine whether -SYO-induced olfactory dysfunction was persistent, mice were subjected to the OD task 45 days after i.c.v. infusion of either monomer- or oligomer-enriched fractions. Similar to the findings at 4 dpi, oligomer-infused mice showed no preference for the familiar compartment at 45 dpi, suggesting that their olfactory function was persistently compromised (Fig. 3E; F(2,27)=8.979; *p=0.0091; **p=0.0017). Surprisingly, mice injected with α-SYMs failed to discriminate the familiar odor in OD task when tested 45 dpi. We next measured dopamine content in the olfactory bulb and found that, at this later time point, dopamine levels were similar in vehicle- and α-SYO-injected mice (Fig. 3F; F(2,17)=16.14). Altogether, results suggest that α-SYOs infused into the lateral ventricle of mice induce early and persistent olfactory dysfunction in mice. This behavioral effect was accompanied by decreased dopamine levels and TH immunoreactivity in the olfactory bulb at early stages, but persisted even after dopamine levels were normalized at later time points. Intriguingly, α- SYMs also induced olfactory dysfunction at later time points, although the underlying mechanisms for this impact remain to be established. 1.3 α-Synuclein oligomers (α-SYOs) trigger additional non-motor symptoms in mice Non-motor symptoms, notably mood alterations, accompany almost all PD cases, causing significant distress and debilitation to patients [2, 3, 34]. We thus asked whether depressive-like and anxious behaviors could be evoked by α-SYOs. Initially, mice were subjected to the open field task for evaluation of locomotor and exploratory activities, as well as anxious behavior. Numbers of crossings (Fig. 4A; F(2,27)=0.0871) and rearings (Fig. 4B; F(2,27)=0.1029) were similar between control (vehicle-injected) mice and mice infused i.c.v. with α-SYMs or α-SYOs, indicating that α-syn does not affect locomotor or exploratory behaviors. We also measured the percentage of time spent in the center of the open field arena, which can be correlated to anxiety levels. Interestingly, mice injected with α-SYOs, but not α-SYMs or vehicle, spent significantly less time in the center of the arena (Fig. 4C; F(2,23)=3.220; *p=0.048), suggesting that oligomers caused anxious behavior in mice. To further investigate this effect, mice were tested in the elevated plus maze, a paradigm widely used to evaluate anxiolytic and anxiogenic effects of drugs and transgenes in rodents [43]. Mice infused with α-SYOs spent less time (Fig. 4D; F(2,23)=3.650; p=0.0400) and did fewer entries (Fig. 4E; F(2,20)=2.476, p=0.0979) in the open arms of the maze compared to vehicle-injected mice. Animals subjected to i.c.v. injection of α-SYMs showed a decrease (albeit not statistically significant) in the amount of time spent in the open arms of the apparatus. Total numbers of arm entries (Fig. 4F; F(2,19)=0.8730) were similar amongst the three groups, ruling out the possibility that α-SYO-injected mice were less mobile in the maze during the trial. Results suggest that α-SYOs induce anxious behavior in mice, thus mimicking an important component of non-motor symptoms in PD patients. 9

Early-stage PD patients frequently present visuospatial disturbances, depression and decreased cognitive capacity, all of which precede the development of motor symptoms [2, 3, 34]. To determine whether memory impairment was induced by α-syn in our model, α-SYO-injected mice were tested in a spatial cued version of the novel object recognition test, at both 3 (Fig. 4G; F(2,27)=0.5474) and 20 dpi (Fig. 4H; F(2,21)=0.2672). In both assessments, animals demonstrated normal object recognition memory, as shown by increased exploratory behavior towards the novel object used in the test session. Mice were further tested in the forced swim test (FST) to evaluate depressive-like behavior. Mice infused with α-SYOs or α-SYMs did not show depressive-like behavior when evaluated in FST task (Fig. 4I; F(2,28)=1.130). Intriguingly, α- SYO-infused mice exhibited shorter immobility time in the forced swim test compared to vehicle- or α-SYM-injected animals (Fig. 4I). It appears likely that increased anxiety induced by α-SYOs (see above) interfered with performance in the forced swim task. 1.4 α-SYOs, but not α-SYMs, induce motor dysfunction and striatal dopaminergic degeneration in mice Motor dysfunction is the hallmark of PD, and several mouse models of the disease have recapitulated this outcome of striatal degeneration by using behavioral paradigms such as the rotarod and pole test tasks. We thus asked whether i.c.v. infusion of α-SYOs induced motor dysfunction in mice. Neither α-SYOs nor α-SYMs had any significant effect on performance in the pole test (Fig. 5A; F(2,26)=1.524; and B; F(2,31)=3.231) or rotarod (Fig. 5C; F(2,27)=1.464) tasks at 4 dpi. Consistent with these findings, striatal levels of dopamine were similar between all groups at 8 dpi (Fig. 5D; F(2,22)=1.723), suggesting that motor dysfunction is not an early event in this model. Interestingly, α-SYO-infused mice showed significantly increased times to turn (Fig. 5E; F(2,23)=5.818, p=0.0171) and descend (Fig. 5F; F(2,42)=11.55, p=0.0001) in the pole test when animals were evaluated 45 days after injection. α-Syn oligomers further impaired performance in the rotarod paradigm (Fig. 5G; F(2,32)=3.125; p=0.1114). Evaluation of striatal dopamine content (Fig. 5H; F(2,15)=5.012, p=0.0166) showed that α-SYOs affected dopamine levels at later time points (45 dpi). Immunoreactivty for the dopamine transporter DAT (Fig. 5I; t(1,6)=3.874; p=0.0082) was also significantly reduced in the caudate putamen of mice that received a single infusion of α-SYOs into the lateral ventricle. Moreover, a reduced number of TH positive cells was also seen in the caudate putamen (Fig. 5J) and substantia nigra 45 dpi (Fig. 5K; t(1,5)=2.201, p=0.0395) further supporting that our findings resembles PD-like symptoms. Results thus suggest that dopaminergic neuron loss and motor deficits are induced late following i.c.v. infusion of αSYOs, and are preceded by non-motor symptoms frequently seen in PD patients, including olfactory deficits and increased anxiety.

4 Discussion Aggregation of α-synuclein (α-syn) is thought to play a central role in the pathophysiology of PD. However, the germane link between α-syn misfolding and dopaminergic neuronal loss as well as which is the most toxic assembly remain uncertain, despite the recent advances in the field [14-17, 21]. Several studies have shown that the injection of protofibrilar and fibrilar α-syn or its overexpression drived by lentiviral systems in the SN or striatum of mice cause loss of dopaminergic neurons in the nigrostriatal pathway [24-26, 29]. α-Syn fibrils injected into the striatum resulted in motor dysfunction in mice [25]. Whereas the ability of α-syn to form neurotoxic oligomeric species in vitro has already been described, reports on the behavioral effect of these species in vivo were still lacking [29]. In this study, we described that the infusion of wild-type α-syn oligomers (α-SYOs) into the lateral ventricle of mice results in early olfactory dysfunction (4 dpi), an initial PD symptom, and in loss of dopaminergic neurons in the olfactory bulb. This non-motor symptom was followed by late 10

striatal dopaminergic loss culminating in motor deficits, 30 days after icv injection. Our study is the first to establish a PD experimental model based on a single injection of α-SYOs into lateral ventricle of mice, which recapitulates some important motor and non-motor PD-symptoms. Our preparation of α-SYOs was also shown to induce a decrease in dopamine content in midbrain neurons in culture in a TH-dependent manner, since dopamine release was only impaired by αSYOs in tyrosine pre- incubated neurons, and not in neurons pre-incubated with L-dopa. Importantly, our results also show that cell viability was not affected by α-SYOs. Several studies have suggested that large α-syn aggregates can reduce TH activity [46], expression [47], and phosphorylation [48], which may lead to reduced dopamine synthesis. This, however, is the first report to show that oligomeric species of α-syn (α-SYO), but not monomeric species (αSYMs), directly affect dopamine production, by altering TH activity. Surprisingly, even though αSYMs had no effect on dopamine release, total dopamine levels were also altered, to some extent, by α-SYMs in cultures pre-incubated with tyrosine. One possible explanation for this finding, which still remains to be confirmed, is that monomeric -syn originated oligomeric forms at the end of the incubation period. Then, we speculate that this effect may be, at least in part, due to partial oligomerization of α-SYMs after 24 h. An impaired sense of smell is a common and early pre-motor symptom in PD [49, 50]. The olfactory bulb is one of the initial brain structures to undergo α-syn deposition [51, 52], and the olfactory impairment shows good correlation with Lewy body pathology in this brain region [53]. Defects in the sense of smell have already been shown in dopamine-selective neurotoxin-based models [18, 36, 54, 55] and in α-syn transgenic aged mice [39, 56-60]. In the current study, we report for the first time that i.c.v. administration of wild type α-syn oligomers in non-transgenic mice induces olfactory disruption. Results show that α-SYOs induced early olfactory dysfunction (4 dpi) and loss of dopaminergic neuron in olfactory bulb of animals. Remarkably, olfactory dysfunction was initially detected on the 4th day after oligomer infusion, and persisted for at least 40 days after a single icv injection of α-SYOs. Along with behavioral alterations, we also found that the olfactory bulb from α-SYO- injected mice showed decreased levels of dopamine and smaller TH immunoreactivity shortly after oligomer administration. In agreement to our results, previous studies also showed disrupted dopamine content in the olfactory bulb of animals with α-syn deposition [60]. However, our findings suggest that α-SYO-induced dopamine depletion in the olfactory bulb may not be the unique cause of olfactory dysfunction, since the levels of this neurotransmitter returned to basal levels 45 days after α-SYOs infusion, even though olfactory dysfunction still remained. Remarkably, animals infused with monomer preparations (α-SYMs) had no olfactory damage soon after injection (4 dpi), but showed olfactory dysfunction when evaluated at later time points (40 dpi). Several evidence support that high concentrations of α-syn monomers can induce PD-like symptoms [39, 56, 57, 60]. It is possible that late olfactory dysfunction induced by α-SYMs results from in vivo oligomerization of injected α-syn monomeric assemblies. Moreover, small contaminations with oligomers in the monomeric preparation could act as ``seeds'' [24], inducing in vivo oligomerization of monomeric α-syn assemblies. Further studies are required to evaluate these hypotheses. Anxiety is the most frequent mood disorder in PD, and can be present in several stages of the disease [61, 62]. The link between α-syn and anxiety- like behavior is controversial. Peña-Oliver and colleagues (2010) suggested a lack of involvement of -syn in unconditioned anxiety [63]. Moreover, spontaneously hypertensive (SHR) inbred rats, considered a genetic model of anxiety, show increased levels of α-syn mRNA and protein when compared to control animals [64]. Herein, we found that i.c.v. infusion of α-SYOs, but not α- SYMs, induced anxiety-like behavior in mice evaluated in the elevated-plus maze test at 20 dpi. Our findings show for the first time that infusion of α-SYOs into the brains of rodents reproduces the anxiety-like behavior seen in PD patients, suggesting that α-syn may modulate this mood dysfunction in this neurodegenerative disease. 11

Very few studies using PD-animal models have evaluated depressive- like behavior, and the link between α-syn and this mood dysfunction remains unclear [65]. An important hurdle in the experimental studies of depressive-like behaviors in rodents is the severe motor impairment induced by the most PD models, which is a confounding factor in many behavioral tests [66]. Increased immobility time in the forced swimming test, tail suspension test and decreased sucrose consumption, all indicating a depressive-like behavior, have been reported in pre-motor PD toxinbased rodent model with 6-OHDA administration, i.e., without motor alterations [19, 67]. Similar results have been found in MPTP/MPP+ [68-70] rodent models, although it is not a consensus [66, 71]. In genetic animal models, the demonstration of age-related anxiety and depressive-like behavior was present mainly in the VMAT-2 deficient mouse model [66, 72]. Recently, Caudal and colleagues (2015) described that a recombinant adeno-associated virus vector expressing human αsyn induces depressive-like behavior, in mice evaluated in the forced swim paradigm [73]. Herein, we found that icv infusion of SYOs induce non-motor symptoms as early olfactory deficits and anxiety-like behavior, but not depressive-like behavior in mice. We evaluated the performance of α-SYO- or α-SYM- infused mice in the forced swim test at an early time-point (4 dpi), when the animals had no motor impairment. Surprisingly, α-SYO-injected mice showed a decreased immobility time when compared to controls groups, as opposed to what is expected from animals with depressive-like behavior. Thus, the possible connection between depressive-like behavior and α-syn accumulation in the brain remains to be further elucidated by employing tasks that can overcome the difficulty of interpreting behavioral data in PD animal models which possibly show motor impairment and/or increased anxiety levels. Relatively subtle cognitive disturbances may be present from the initial stages of PD, and in many patients this may progress to more severe memory impairment and dementia [74]. Here, we show that icv infusion of two different assemblies of α-syn (α-SYOs and α-SYMs) in mice do not affect novel object recognition memory processing when evaluated 3 or 20 dpi. Previous studies reported that aged (12-month-old) transgenic mice overexpressing α-syn show cognitive impairment, with widespread Lewy body-like pathology and concomitant motor impairment [75, 76]. Cognitive dysfunction in PD patients involves mainly deficits in working, spatial and motor memories, while memory tasks dependent on the temporal remain intact [77]. These findings have been recapitulated in several PD animal models, including MPTP-injected mice, parkin-deficient mice and Y39C human α-synuclein transgenic mice [78-81]. Goldberg and colleagues [82] used the novel object recognition and novel place recognition tasks to show that α-syn overexpression interferes with memory performance. However, since most studies show that recognition memory is only affected in aged animals and at later stages of α-syn deposition [3, 56, 83-86], additional studies appear warranted to assess α-SYO-infused mice in other behavioral paradigms for detailed memory evaluation. Motor symptoms are the hallmark of PD, and occur as a consequence of progressive loss of dopaminergic neurons and deregulation of dopamine- modulated pathways in the basal ganglia [1, 15]. Several α-syn transgenic models show age-dependent motor dysfunctions [17, 21, 57-59, 87]. In addition, a single inoculation of synthetic α-syn fibrils into the stratum of mice was shown to induce dopaminergic loss in the substantia nigra, culminating in motor deficits [26]. Of note, our findings show that a single i.c.v. injection of wild type α-SYOs induced motor deficits only at later time points after oligomer administration. -SYOs-infused mice showed no motor impairment in the rotarod or pole test tasks until 10 dpi, but showed a significant motor dysfunction when evaluated 40 days after the injection. Monomeric α-syn, on the other hand, had no effect on motor function of animals at any evaluated time points. Although reduced locomotor/exploratory activity and freezing behavior have been described in other rodent models of PD [88], we found no changes in locomotor behavior in an open field following icv infusion of -SYMs or -SYOs in mice. The progressive nature of motor deficits appears as an interesting feature of the current model, and, to our knowledge, it has never been reproduced by any other PD model. In addition, current results 12

show a reduction in striatal dopamine content, a principal cause of motor deficit, temporally matching motor dysfunction following α-SYOs injection. Together, our findings indicate that administration of α-SYOs into the lateral ventricle of mice recapitulates neurodegenerative aspects as well as some of the most relevant behavioral features of PD, suggesting a mechanistic link between α-SYOs toxicity and the cardinal features of PD. This model holds potential as a tool to unravel molecular mechanisms of the disease and open new avenues for drug screening purposes in PD. Acknowledgments

This work was supported by grants from Brazilian funding agencies Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (F.S.N., C.F.B., L.R., F.C.A.G., F.G.D.F, S.T.F, J.R.C., C.P.F., M.G.), Conselho Nacional de Desenvolvimento Científico e Tecnológico (R.L.F., F.B.A., C.F.B., A.F.D.B., L.P.D., F.C.A.G., F.G.D.F., S.T.F., J.R.C., C.P.F., M.G.), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (J.D.O, D.F.E.), and National Institute for Translational Neuroscience (INNT) (STF). We thank Maíra Oliveira, Katia Laia and Mariangela M. Viana for technical support, and Ana Claudia Rangel for competent lab and project management. References [1] A.J. Lees, J. Hardy, T. Revesz,;1; Parkinson's disease, Lancet 373(9680) (2009) 2055-66. [2] K.R. Chaudhuri, D.G. Healy, A.H. Schapira, E.;1; National Institute for Clinical, Non-motor symptoms of Parkinson's disease: diagnosis and management, Lancet Neurol 5(3) (2006) 235-45. [3] K.R. Chaudhuri, A.H. Schapira,;1; Non-motor symptoms of Parkinson's disease: dopaminergic pathophysiology and treatment, Lancet Neurol 8(5) (2009) 464-74. [4] P. Martinez-Martin, C. Rodriguez-Blazquez, M.M. Kurtis, K.R. Chaudhuri, N.V. Group,;1; The impact of non-motor symptoms on health-related quality of life of patients with Parkinson's disease, Mov Disord 26(3) (2011) 399-406. [5] J.M. Savitt, V.L. Dawson, T.M. Dawson,;1; Diagnosis and treatment of Parkinson disease: molecules to medicine, J Clin Invest 116(7) (2006) 1744-54. [6] M. Goedert,;1; Alpha-synuclein and neurodegenerative diseases, Nat Rev Neurosci 2(7) (2001) 492-501. [7] M. Goedert, NEURODEGENERATION. Alzheimer's and;1; Parkinson's diseases: The prion concept in relation to assembled Abeta, tau, and alpha-synuclein, Science 349(6248) (2015) 1255555. [8] M.G. Spillantini, M.L. Schmidt, V.M. Lee, J.Q. Trojanowski, R. Jakes, M. Goedert,;1; Alphasynuclein in Lewy bodies, Nature 388(6645) (1997) 839-40. [9] P.K. Auluck, G. Caraveo, S. Lindquist,;1; alpha-Synuclein: membrane interactions and toxicity in Parkinson's disease, Annu Rev Cell Dev Biol 26 (2010) 211-33. [10] D.P. Karpinar, M.B. Balija, S. Kugler, F. Opazo, N. Rezaei-Ghaleh, N. Wender, H.Y. Kim, G. Taschenberger, B.H. Falkenburger, H. Heise, A. Kumar, D. Riedel, L. Fichtner, A. Voigt, G.H. Braus, K. Giller, S. Becker, A. Herzig, M. Baldus, H. Jackle, S. Eimer, J.B. Schulz, C. Griesinger, M. Zweckstetter,;1; Pre-fibrillar alpha-synuclein variants with impaired beta-structure increase neurotoxicity in Parkinson's disease models, EMBO J 28(20) (2009) 3256-68. [11] K.A. Conway, J.D. Harper, P.T. Lansbury,;1; Accelerated in vitro fibril formation by a mutant alpha-synuclein linked to early-onset Parkinson disease, Nat Med 4(11) (1998) 1318-20.

13

[12] K.M. Danzer, D. Haasen, A.R. Karow, S. Moussaud, M. Habeck, A. Giese, H. Kretzschmar, B. Hengerer, M. Kostka,;1; Different species of alpha-synuclein oligomers induce calcium influx and seeding, J Neurosci 27(34) (2007) 9220-32. [13] G.K. Tofaris, P. Garcia Reitbock, T. Humby, S.L. Lambourne, M. O'Connell, B. Ghetti, H. Gossage, P.C. Emson, L.S. Wilkinson, M. Goedert, M.G. Spillantini,;1; Pathological changes in dopaminergic nerve cells of the substantia nigra and olfactory bulb in mice transgenic for truncated human alpha-synuclein(1-120): implications for Lewy body disorders, J Neurosci 26(15) (2006) 3942-50. [14] J. Blesa, S. Phani, V. Jackson-Lewis, S. Przedborski, Classic and;1; new animal models of Parkinson's disease, J Biomed Biotechnol 2012 (2012) 845618. [15] W. Dauer, S. Przedborski,;1; Parkinson's disease: mechanisms and models, Neuron 39(6) (2003) 889-909. [16] S. Hisahara, S. Shimohama, Toxin-induced and;1; genetic animal models of Parkinson's disease, Parkinsons Dis 2011 (2010) 951709. [17] V. Jackson-Lewis, J. Blesa, S. Przedborski,;1; Animal models of Parkinson's disease, Parkinsonism Relat Disord 18 Suppl 1 (2012) S183-5. [18] R.D. Prediger, A.S. Aguiar, Jr., F.C. Matheus, R. Walz, L. Antoury, R. Raisman-Vozari, R.L. Doty,;1; Intranasal administration of neurotoxicants in animals: support for the olfactory vector hypothesis of Parkinson's disease, Neurotox Res 21(1) (2012) 90-116. [19] M.T. Tadaiesky, P.A. Dombrowski, C.P. Figueiredo, E. Cargnin-Ferreira, C. Da Cunha, R.N. Takahashi,;1; Emotional, cognitive and neurochemical alterations in a premotor stage model of Parkinson's disease, Neuroscience 156(4) (2008) 830-40. [20] P.M. Antony, N.J. Diederich, R. Balling,;1; Parkinson's disease mouse models in translational research, Mamm Genome 22(7-8) (2011) 401-19. [21] T.M. Dawson, H.S. Ko, V.L. Dawson,;1; Genetic animal models of Parkinson's disease, Neuron 66(5) (2010) 646-61. [22] M.H. Polymeropoulos, J.J. Higgins, L.I. Golbe, W.G. Johnson, S.E. Ide, G. Di Iorio, G. Sanges, E.S. Stenroos, L.T. Pho, A.A. Schaffer, A.M. Lazzarini, R.L. Nussbaum, R.C. Duvoisin,;1; Mapping of a gene for Parkinson's disease to chromosome 4q21-q23, Science 274(5290) (1996) 1197-9. [23] A.B. Singleton, M. Farrer, J. Johnson, A. Singleton, S. Hague, J. Kachergus, M. Hulihan, T. Peuralinna, A. Dutra, R. Nussbaum, S. Lincoln, A. Crawley, M. Hanson, D. Maraganore, C. Adler, M.R. Cookson, M. Muenter, M. Baptista, D. Miller, J. Blancato, J. Hardy, K. Gwinn-Hardy,;1; alpha-Synuclein locus triplication causes Parkinson's disease, Science 302(5646) (2003) 841. [24] K.C. Luk, V. Kehm, J. Carroll, B. Zhang, P. O'Brien, J.Q. Trojanowski, V.M. Lee,;1; Pathological alpha-synuclein transmission initiates Parkinson-like neurodegeneration in nontransgenic mice, Science 338(6109) (2012) 949-53. [25] K.C. Luk, V.M. Kehm, B. Zhang, P. O'Brien, J.Q. Trojanowski, V.M. Lee,;1; Intracerebral inoculation of pathological alpha-synuclein initiates a rapidly progressive neurodegenerative alphasynucleinopathy in mice, J Exp Med 209(5) (2012) 975-86. [26] K.L. Paumier, K.C. Luk, F.P. Manfredsson, N.M. Kanaan, J.W. Lipton, T.J. Collier, K. SteeceCollier, C.J. Kemp, S. Celano, E. Schulz, I.M. Sandoval, S. Fleming, E. Dirr, N.K. Polinski, J.Q. Trojanowski, V.M. Lee, C.E. Sortwell,;1; Intrastriatal injection of pre-formed mouse alphasynuclein fibrils into rats triggers alpha-synuclein pathology and bilateral nigrostriatal degeneration, Neurobiol Dis 82 (2015) 185-99. [27] R.A. Crowther, S.E. Daniel, M. Goedert,;1; Characterisation of isolated alpha-synuclein filaments from substantia nigra of Parkinson's disease brain, Neurosci Lett 292(2) (2000) 128-30. [28] I.F. Tsigelny, L. Crews, P. Desplats, G.M. Shaked, Y. Sharikov, H. Mizuno, B. Spencer, E. Rockenstein, M. Trejo, O. Platoshyn, J.X. Yuan, E. Masliah,;1; Mechanisms of hybrid oligomer 14

formation in the pathogenesis of combined Alzheimer's and Parkinson's diseases, PLoS One 3(9) (2008) e3135. [29] B. Winner, R. Jappelli, S.K. Maji, P.A. Desplats, L. Boyer, S. Aigner, C. Hetzer, T. Loher, M. Vilar, S. Campioni, C. Tzitzilonis, A. Soragni, S. Jessberger, H. Mira, A. Consiglio, E. Pham, E. Masliah, F.H. Gage, R. Riek,;1; In vivo demonstration that alpha-synuclein oligomers are toxic, Proc Natl Acad Sci U S A 108(10) (2011) 4194-9. [30] M.S. Celej, R. Sarroukh, E. Goormaghtigh, G.D. Fidelio, J.M. Ruysschaert, V. Raussens,;1; Toxic prefibrillar alpha-synuclein amyloid oligomers adopt a distinctive antiparallel beta-sheet structure, Biochem J 443(3) (2012) 719-26. [31] C. Follmer, L. Romao, C.M. Einsiedler, T.C. Porto, F.A. Lara, M. Moncores, G. Weissmuller, H.A. Lashuel, P. Lansbury, V.M. Neto, J.L. Silva, D. Foguel,;1; Dopamine affects the stability, hydration, and packing of protofibrils and fibrils of the wild type and variants of alpha-synuclein, Biochemistry 46(2) (2007) 472-82. [32] R. Kayed, E. Head, J.L. Thompson, T.M. McIntire, S.C. Milton, C.W. Cotman, C.G. Glabe,;1; Common structure of soluble amyloid oligomers implies common mechanism of pathogenesis, Science 300(5618) (2003) 486-9. [33] J. Pruszak, L. Just, O. Isacson, G. Nikkhah,;1; Isolation and culture of ventral mesencephalic precursor cells and dopaminergic neurons from rodent brains, Curr Protoc Stem Cell Biol Chapter 2 (2009) Unit 2D 5. [34] C.P. Figueiredo, J.R. Clarke, J.H. Ledo, F.C. Ribeiro, C.V. Costa, H.M. Melo, A.P. MotaSales, L.M. Saraiva, W.L. Klein, A. Sebollela, F.G. De Felice, S.T. Ferreira,;1; Memantine rescues transient cognitive impairment caused by high-molecular-weight abeta oligomers but not the persistent impairment induced by low-molecular-weight oligomers, J Neurosci 33(23) (2013) 962634. [35] S.E. Laursen, J.K. Belknap,;1; Intracerebroventricular injections in mice. Some methodological refinements, J Pharmacol Methods 16(4) (1986) 355-7. [36] R.D. Prediger, L.C. Batista, R. Medeiros, P. Pandolfo, J.C. Florio, R.N. Takahashi,;1; The risk is in the air: Intranasal administration of MPTP to rats reproducing clinical features of Parkinson's disease, Exp Neurol 202(2) (2006) 391-403. [37] G.H. Petit, E. Berkovich, M. Hickery, P. Kallunki, K. Fog, C. Fitzer-Attas, P. Brundin,;1; Rasagiline ameliorates olfactory deficits in an alpha-synuclein mouse model of Parkinson's disease, PLoS One 8(4) (2013) e60691. [38] L. Prut, C. Belzung,;1; The open field as a paradigm to measure the effects of drugs on anxiety-like behaviors: a review, Eur J Pharmacol 463(1-3) (2003) 3-33. [39] S.M. Fleming, N.A. Tetreault, C.K. Mulligan, C.B. Hutson, E. Masliah, M.F. Chesselet,;1; Olfactory deficits in mice overexpressing human wildtype alpha-synuclein, Eur J Neurosci 28(2) (2008) 247-56. [40] N. Ogawa, Y. Hirose, S. Ohara, T. Ono, Y. Watanabe,;1; A simple quantitative bradykinesia test in MPTP-treated mice, Res Commun Chem Pathol Pharmacol 50(3) (1985) 435-41. [41] C. Monville, E.M. Torres, S.B. Dunnett,;1; Comparison of incremental and accelerating protocols of the rotarod test for the assessment of motor deficits in the 6-OHDA model, J Neurosci Methods 158(2) (2006) 219-23. [42] R.D. Porsolt, M. Le Pichon, M. Jalfre,;1; Depression: a new animal model sensitive to antidepressant treatments, Nature 266(5604) (1977) 730-2. [43] R.G. Lister,;1; The use of a plus-maze to measure anxiety in the mouse, Psychopharmacology (Berl) 92(2) (1987) 180-5.

15

[44] D. Beckman, L.E. Santos, T.A. Americo, J.H. Ledo, F.G. de Mello, R. Linden,;1; Prion Protein Modulates Monoaminergic Systems and Depressive-like Behavior in Mice, J Biol Chem 290(33) (2015) 20488-98. [45] T. Ozawa, K. Yamada, Y. Ichitani,;1; Long-term object location memory in rats: effects of sample phase and delay length in spontaneous place recognition test, Neurosci Lett 497(1) (2011) 37-41. [46] R.G. Perez, J.C. Waymire, E. Lin, J.J. Liu, F. Guo, M.J. Zigmond,;1; A role for alphasynuclein in the regulation of dopamine biosynthesis, J Neurosci 22(8) (2002) 3090-9. [47] S. Yu, X. Zuo, Y. Li, C. Zhang, M. Zhou, Y.A. Zhang, K. Ueda, P. Chan,;1; Inhibition of tyrosine hydroxylase expression in alpha-synuclein-transfected dopaminergic neuronal cells, Neurosci Lett 367(1) (2004) 34-9. [48] T.N. Alerte, A.A. Akinfolarin, E.E. Friedrich, S.A. Mader, C.S. Hong, R.G. Perez,;1; Alphasynuclein aggregation alters tyrosine hydroxylase phosphorylation and immunoreactivity: lessons from viral transduction of knockout mice, Neurosci Lett 435(1) (2008) 24-9. [49] N.I. Bohnen, S. Gedela, H. Kuwabara, G.M. Constantine, C.A. Mathis, S.A. Studenski, R.Y. Moore,;1; Selective hyposmia and nigrostriatal dopaminergic denervation in Parkinson's disease, J Neurol 254(1) (2007) 84-90. [50] A. Haehner, T. Hummel, H. Reichmann,;1; Olfactory loss in Parkinson's disease, Parkinsons Dis 2011 (2011) 450939. [51] H. Braak, K. Del Tredici, H. Bratzke, J. Hamm-Clement, D. Sandmann-Keil, U. Rub,;1; Staging of the intracerebral inclusion body pathology associated with idiopathic Parkinson's disease (preclinical and clinical stages), J Neurol 249 Suppl 3 (2002) III/1-5. [52] H. Braak, K. Del Tredici, U. Rub, R.A. de Vos, E.N. Jansen Steur, E. Braak,;1; Staging of brain pathology related to sporadic Parkinson's disease, Neurobiol Aging 24(2) (2003) 197-211. [53] T.G. Beach, C.H. Adler, L. Lue, L.I. Sue, J. Bachalakuri, J. Henry-Watson, J. Sasse, S. Boyer, S. Shirohi, R. Brooks, J. Eschbacher, C.L. White,;1; 3rd, H. Akiyama, J. Caviness, H.A. Shill, D.J. Connor, M.N. Sabbagh, D.G. Walker, C. Arizona Parkinson's Disease, Unified staging system for Lewy body disorders: correlation with nigrostriatal degeneration, cognitive impairment and motor dysfunction, Acta Neuropathol 117(6) (2009) 613-34. [54] A. Bonito-Oliva, D. Masini, G. Fisone,;1; A mouse model of non-motor symptoms in Parkinson's disease: focus on pharmacological interventions targeting affective dysfunctions, Front Behav Neurosci 8 (2014) 290. [55] R.D. Prediger, A.S. Aguiar, Jr., A.E. Rojas-Mayorquin, C.P. Figueiredo, F.C. Matheus, L. Ginestet, C. Chevarin, E.D. Bel, R. Mongeau, M. Hamon, L. Lanfumey, R. Raisman-Vozari,;1; Single intranasal administration of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine in C57BL/6 mice models early preclinical phase of Parkinson's disease, Neurotox Res 17(2) (2010) 114-29. [56] K.F. Farrell, S. Krishnamachari, E. Villanueva, H. Lou, T.N. Alerte, E. Peet, R.E. Drolet, R.G. Perez,;1; Non-motor parkinsonian pathology in aging A53T alpha-synuclein mice is associated 16

with progressive synucleinopathy and altered enzymatic function, J Neurochem 128(4) (2014) 53646. [57] C. Hansen, T. Bjorklund, G.H. Petit, M. Lundblad, R.P. Murmu, P. Brundin, J.Y. Li,;1; A novel alpha-synuclein-GFP mouse model displays progressive motor impairment, olfactory dysfunction and accumulation of alpha-synuclein-GFP, Neurobiol Dis 56 (2013) 145-55. [58] J. Neuner, S. Filser, S. Michalakis, M. Biel, J. Herms,;1; A30P alpha-Synuclein interferes with the stable integration of adult-born neurons into the olfactory network, Sci Rep 4 (2014) 3931. [59] J. Neuner, S.V. Ovsepian, M. Dorostkar, S. Filser, A. Gupta, S. Michalakis, M. Biel, J. Herms,;1; Pathological alpha-synuclein impairs adult-born granule cell development and functional integration in the olfactory bulb, Nat Commun 5 (2014) 3915. [60] S. Zhang, Q. Xiao, W. Le,;1; Olfactory dysfunction and neurotransmitter disturbance in olfactory bulb of transgenic mice expressing human A53T mutant alpha-synuclein, PLoS One 10(3) (2015) e0119928. [61] N.N. Dissanayaka, A. Sellbach, S. Matheson, J.D. O'Sullivan, P.A. Silburn, G.J. Byrne, R. Marsh, G.D. Mellick,;1; Anxiety disorders in Parkinson's disease: prevalence and risk factors, Mov Disord 25(7) (2010) 838-45. [62] K. Walsh, G. Bennett,;1; Parkinson's disease and anxiety, Postgrad Med J 77(904) (2001) 8993. [63] Y. Pena-Oliver, V.L. Buchman, D.N. Stephens,;1; Lack of involvement of alpha-synuclein in unconditioned anxiety in mice, Behav Brain Res 209(2) (2010) 234-40. [64] S. Chiavegatto, G.S. Izidio, A. Mendes-Lana, I. Aneas, T.A. Freitas, A.S. Torrao, I.M. Conceicao, L.R. Britto, A. Ramos,;1; Expression of alpha-synuclein is increased in the hippocampus of rats with high levels of innate anxiety, Mol Psychiatry 14(9) (2009) 894-905. [65] N. Schintu, X. Zhang, P. Svenningsson,;1; Studies of depression-related states in animal models of Parkinsonism, J Parkinsons Dis 2(2) (2012) 87-106. [66] H.S. Lindgren, S.B. Dunnett,;1; Cognitive dysfunction and depression in Parkinson's disease: what can be learned from rodent models?, Eur J Neurosci 35(12) (2012) 1894-907. [67] I. Branchi, I. D'Andrea, M. Armida, T. Cassano, A. Pezzola, R.L. Potenza, M.G. Morgese, P. Popoli, E. Alleva,;1; Nonmotor symptoms in Parkinson's disease: investigating early-phase onset of behavioral dysfunction in the 6-hydroxydopamine-lesioned rat model, J Neurosci Res 86(9) (2008) 2050-61. [68] M. Moretti, V.B. Neis, F.C. Matheus, M.P. Cunha, P.B. Rosa, C.M. Ribeiro, A.L. Rodrigues, R.D. Prediger,;1; Effects of Agmatine on Depressive-Like Behavior Induced by Intracerebroventricular Administration of 1-Methyl-4-phenylpyridinium (MPP(+)), Neurotox Res 28(3) (2015) 222-31.

17

[69] A. Mori, S. Ohashi, M. Nakai, T. Moriizumi, Y. Mitsumoto,;1; Neural mechanisms underlying motor dysfunction as detected by the tail suspension test in MPTP-treated C57BL/6 mice, Neurosci Res 51(3) (2005) 265-74. [70] R.M. Santiago, J. Barbieiro, M.M. Lima, P.A. Dombrowski, R. Andreatini, M.A. Vital,;1; Depressive-like behaviors alterations induced by intranigral MPTP, 6-OHDA, LPS and rotenone models of Parkinson's disease are predominantly associated with serotonin and dopamine, Prog Neuropsychopharmacol Biol Psychiatry 34(6) (2010) 1104-14. [71] M.G. Vuckovic, R.I. Wood, D.P. Holschneider, A. Abernathy, D.M. Togasaki, A. Smith, G.M. Petzinger, M.W. Jakowec,;1; Memory, mood, dopamine, and serotonin in the 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine-lesioned mouse model of basal ganglia injury, Neurobiol Dis 32(2) (2008) 319-27. [72] T.N. Taylor, W.M. Caudle, K.R. Shepherd, A. Noorian, C.R. Jackson, P.M. Iuvone, D. Weinshenker, J.G. Greene, G.W. Miller,;1; Nonmotor symptoms of Parkinson's disease revealed in an animal model with reduced monoamine storage capacity, J Neurosci 29(25) (2009) 8103-13. [73] D. Caudal, A. Alvarsson, A. Bjorklund, P. Svenningsson,;1; Depressive-like phenotype induced by AAV-mediated overexpression of human alpha-synuclein in midbrain dopaminergic neurons, Exp Neurol 273 (2015) 243-52. [74] D. Aarsland,;1; Cognitive impairment in Parkinson's disease and dementia with Lewy bodies, Parkinsonism Relat Disord 22 Suppl 1 (2016) S144-8. [75] D. Amschl, J. Neddens, D. Havas, S. Flunkert, R. Rabl, H. Romer, E. Rockenstein, E. Masliah, M. Windisch, B. Hutter-Paier,;1; Time course and progression of wild type alpha-synuclein accumulation in a transgenic mouse model, BMC Neurosci 14 (2013) 6. [76] E. Masliah, E. Rockenstein, I. Veinbergs, M. Mallory, M. Hashimoto, A. Takeda, Y. Sagara, A. Sisk, L. Mucke,;1; Dopaminergic loss and inclusion body formation in alpha-synuclein mice: implications for neurodegenerative disorders, Science 287(5456) (2000) 1265-9. [77] G.S. Watson, J.B. Leverenz,;1; Profile of cognitive impairment in Parkinson's disease, Brain Pathol 20(3) (2010) 640-5. [78] D. Rial, A.A. Castro, N. Machado, P. Garcao, F.Q. Goncalves, H.B. Silva, A.R. Tome, A. Kofalvi, O. Corti, R. Raisman-Vozari, R.A. Cunha, R.D. Prediger,;1; Behavioral phenotyping of Parkin-deficient mice: looking for early preclinical features of Parkinson's disease, PLoS One 9(12) (2014) e114216. [79] W. Zhou, J.B. Milder, C.R. Freed,;1; Transgenic mice overexpressing tyrosine-to-cysteine mutant human alpha-synuclein: a progressive neurodegenerative model of diffuse Lewy body disease, J Biol Chem 283(15) (2008) 9863-70. [80] B.W. Kim, S. Koppula, H. Kumar, J.Y. Park, I.W. Kim, S.V. More, I.S. Kim, S.D. Han, S.K. Kim, S.H. Yoon, D.K. Choi,;1; alpha-Asarone attenuates microglia-mediated neuroinflammation by inhibiting NF kappa B activation and mitigates MPTP-induced behavioral deficits in a mouse model of Parkinson's disease, Neuropharmacology 97 (2015) 46-57. 18

[81] B.W. Kim, S. Koppula, S.Y. Park, Y.S. Kim, P.J. Park, J.H. Lim, I.S. Kim, D.K. Choi,;1; Attenuation of neuroinflammatory responses and behavioral deficits by Ligusticum officinale (Makino) Kitag in stimulated microglia and MPTP-induced mouse model of Parkinson's disease, J Ethnopharmacol 164 (2015) 388-97. [82] N.R. Goldberg, J. Caesar, A. Park, S. Sedgh, G. Finogenov, E. Masliah, J. Davis, M. BlurtonJones,;1; Neural Stem Cells Rescue Cognitive and Motor Dysfunction in a Transgenic Model of Dementia with Lewy Bodies through a BDNF-Dependent Mechanism, Stem Cell Reports 5(5) (2015) 791-804. [83] B. Dubois, B. Pillon,;1; Cognitive deficits in Parkinson's disease, J Neurol 244(1) (1997) 2-8. [84] S.J. Lewis, A. Dove, T.W. Robbins, R.A. Barker, A.M. Owen,;1; Cognitive impairments in early Parkinson's disease are accompanied by reductions in activity in frontostriatal neural circuitry, J Neurosci 23(15) (2003) 6351-6. [85] J. Pagonabarraga, J. Kulisevsky,;1; Cognitive impairment and dementia in Parkinson's disease, Neurobiol Dis 46(3) (2012) 590-6. [86] G.T. Stebbins, J.D. Gabrieli, F. Masciari, L. Monti, C.G. Goetz,;1; Delayed recognition memory in Parkinson's disease: a role for working memory?, Neuropsychologia 37(4) (1999) 50310. [87] D.R. Graham, A. Sidhu,;1; Mice expressing the A53T mutant form of human alpha-synuclein exhibit hyperactivity and reduced anxiety-like behavior, J Neurosci Res 88(8) (2010) 1777-83. [88] C. Qi, S. Varga, S.J. Oh, C.J. Lee, D. Lee,;1; Optogenetic Rescue of Locomotor Dysfunction and Dopaminergic Degeneration Caused by Alpha-Synuclein and EKO Genes, Exp Neurobiol 26(2) (2017) 97-103.

Figure 1. Characterization and fractioning of α-syn oligomers (α-SYOs) by SEC-HPLC. Human wild-type α-syn was expressed, purified and incubated to allow oligomerization (See Methods). (A) Initial preparation, composed of a mixture of α-syn monomers (α-SYMs) and oligomers (α-SYOs), was routinely characterized by SEC-HPLC. The main peak (eluting at ~ 2.0 mL) corresponds to α-SYMs, since its elution profile matches the profile of purified non-oligomerized α-syn (dashed line). A smaller peak eluting at 1.8 mL (red arrow) was re-injected (B) into the SEC-HPLC. (C) Dot immunoblot analysis shows that both α-SYMs and α-SYOs show immunoreactivity to anti-α-syn antibody, but only α-SYOs shows significant reactivity to anti-oligomer A11 antibody (top). (D) Western immunoblot using an anti-α-syn antibody shows that α-SYOs fraction comprised oligomers

19

ranging from 30 to 64 kDa, while the α-SYMs fraction contained essentially ~15 kDa monomeric α-syn.

Figure 2. α-syn oligomers (α-SYOs) impair dopamine release from midbrain neurons in culture. Rat midbrain neurons were exposed for 24 h to 25 µM of α-syn monomers (α-SYMs) or oligomers (α-SYOs). After this period, 1mM tyrosine (A, B) or 0.5 mM L-dopa (C, D) were added. In response to stimulation by 0.2 mM glutamate, tyrosineincubated midbrain dopaminergic neurons exposed to α-SYOs, but not to α-SYMs, released less dopamine than controls (A). Total dopamine levels were reduced by α-SYMs and αSYOs in the presence of tyrosine (B). When L-dopa was used as substrate for dopamine production, there was no difference in glutamate-evoked dopamine release in cultures exposed to α-SYMs or α-SYOs (C), although total levels of dopamine were reduced in αSYO-treated cultures (D). Data are expressed as means ± S.E.M, and were obtained from three independent cultures for each experiment. In A: **p=0.0006 and ***p < 0.0001, twoway ANOVA followed by Bonferroni. In B: *p=0.0019 and ***p<0.0001; and in D: ***p=0.0004, one-way ANOVA followed by Tukey post-hoc test.

Figure 3. Intracerebroventricular (i.c.v.) injection of α-syn oligomers (αSYOs) induces olfactory dysfunction and dopaminergic neurotoxicity in the olfactory bulb (OB) of mice. Mice received a single i.c.v. injection of vehicle (VEH; n=9), α-syn monomers (α-SYMs; n=10) or α-SYOs (50 pmol; n=10), and were tested in olfactory discrimination tasks. (A) Discrimination indexes shows that only α-SYO-treated mice have olfactory dysfunction when evaluated 4 dpi in the familiar bedding paradigm, which is accompanied by a decrease in dopamine content in the OB (B, 8 dpi; n=10 VEH, 10 αSYMs, 5 α-SYOs). (C) Qualitative immunohistochemistry analysis shows that OB from αSYO-injected mice (4 dpi) shows less immunoreactivity to tyrosine hydroxylase (TH) compared to VEH-treated animals (Scale bar = 75 µm). Forty (D) and forty-five days (E) after i.c.v. injection, both α-SYM- and α-SYO-injected mice showed olfactory deficit (n= 11 VEH, 6 α-SYMs, 8 α-SYOs) in the vanillin and familiar bedding tasks, respectively, while dopamine levels in α-SYOs mice returned to basal levels at 45 dpi (F; n= 7 VEH, 6 αSYOs). Data are expressed as means ± S.E.M. In A: *p=0.0111; in B: *p=0.0042; in D: *p=0.0401 and #p=0.0764; in E: *p=0.0091 and **p=0.0017, one-way ANOVA followed by Tukey.

20

Figure 4. Intracerebroventricular injection of α-syn oligomers (α-SYOs) induces anxiety-like behavior, but not cognitive impairment or depressive-like behavior in mice. Mice received a single i.c.v. injection of vehicle (VEH), α-syn monomers (α-SYMs) or α-SYOs (50 pmol) and were tested in the open field task 20 dpi. Number of crossings (A), rearings (B) and percentage of time spent in the center of the arena (C) were measured during 5 min (n= 11 VEH, 7 α-SYMs, 12 α-SYOs). Mice were also assessed in the elevated plus maze test 20 dpi, and the percentage of time in the open arms (D), number of entries in the open arms (E) and total number of arm entries (F) were recorded in a 5min-long session (n= 11 VEH, 7 α-SYMs, 12 α-SYOs). Cognitive performance of VEH-, αSYM- or α-SYO-injected mice was evaluated in the novel object recognition paradigm 3 (G; n= 8 VEH, 7 α-SYMs, 7 α-SYOs) and 20 dpi (H; n= 11 VEH, 7 α-SYMs; 7 α-SYOs). A separate group of mice was tested in the forced swim protocol 4 dpi, in order to assess whether injection of α-SYOs induced depressive-like behavior (I; n= 11 VEH, 10 α-SYMs, 10 α-SYOs). Data are expressed as means ± S.E.M. In C: *p=0.0480; in D: p=0.0400; in E: 0.0979,one-way ANOVA followed by Tukey.

Figure 5. Intracerebroventricular injection of α-syn oligomers (α-SYOs) induces late motor deficits and striatal dopamine loss in mice. Mice received a single i.c.v. injection of vehicle (VEH), α-syn monomers (α-SYMs) or α-SYOs (50 pmol) and were assessed in the pole test (A,B,E,F) and rotarod (C,G) motor paradigms. Time to orient downwards (t-turn) and total time to descend the pole (t-descend) were measured at 4 dpi (A,B; n =11/group) or 45 dpi (E,F, n= 11 VEH, 7 α-SYMs, 8 α-SYOs). No differences between groups were seen in rotarod test performed early after injection (4 dpi) (C; n= 10/group). Forty-five days after i.c.v. injection, α-SYO-injected mice showed longer latencies to turn (E) and to descend (F) the pole, indicative of motor dysfunction. Rotarod test also showed that α-SYO-injected mice had late motor dysfunction (30 dpi) (G; n= 10/group). Striatal dopamine content was measured by HPLC 8 (D; n = 10 VEH, 10 αSYMs, 10 α-SYOs) and 45 (H; n= 6 VEH, 6 α-SYMs, 10 α-SYOs) days after i.c.v. injection of vehicle, α-SYMs or α-SYOs. (I) Representative images of DAT immunolabeling in the caudate putamen of VEH- and α-SYO-injected mice (Scale bar = 75 µm), along with quantification of DAT immunoreactivity (n = 4 VEH, 4 α-SYOs). (J) Representative images of TH immunolabeling (n = 4 VEH, 4 α-SYOs) in the caudate putamen of VEH- and αSYO-injected mice 45 dpi (scale bar = 75 µm), along with quantification of TH immunoreactivity. (K) Representative images of TH-positive cells in the substantia nigra of 21

VEH- and α-SYO-injected mice 45 dpi (scale bar = 100 μm), along with quantification of TH-positive cells per um2 (n = 4 VEH, 3 α-SYOs). Data are expressed as means ± S.E.M. In E: *p=0.0171, in F: *p=0.0001, in G: *p=0.1114, in H: p=0.0166, one-way ANOVA followed by Tukey. In I: **p = 0.0082, in K: *p= 0.0395, Student’s t-test

TDENDOFDOCTD

22