Fasudil attenuates aggregation of α-synuclein in models of Parkinson’s disease

Parkinson’s disease (PD) is the most common neurodegenerative movement disorder, yet disease-modifying treatments do not currently exist. Rho-associated protein kinase (ROCK) was recently described as a novel neuroprotective target in PD. Since alpha-synuclein (α-Syn) aggregation is a major hallmark in the pathogenesis of PD, we aimed to evaluate the anti-aggregative potential of pharmacological ROCK inhibition using the isoquinoline derivative Fasudil, a small molecule inhibitor already approved for clinical use in humans. Fasudil treatment significantly reduced α-Syn aggregation in vitro in a H4 cell culture model as well as in a cell-free assay. Nuclear magnetic resonance spectroscopy analysis revealed a direct binding of Fasudil to tyrosine residues Y133 and Y136 in the C-terminal region of α-Syn. Importantly, this binding was shown to be biologically relevant using site-directed mutagenesis of these residues in the cell culture model. Furthermore, we evaluated the impact of long-term Fasudil treatment on α-Syn pathology in vivo in a transgenic mouse model overexpressing human α-Syn bearing the A53T mutation (α-SynA53T mice). Fasudil treatment improved motor and cognitive functions in α-SynA53T mice as determined by CatwalkTM gait analysis and novel object recognition (NOR), without apparent side effects. Finally, immunohistochemical analysis revealed a significant reduction of α-Syn pathology in the midbrain of α-SynA53T mice after Fasudil treatment. Our results demonstrate that Fasudil, next to its effects mediated by ROCK-inhibition, directly interacts with α-Syn and attenuates α-Syn pathology. This underscores the translational potential of Fasudil as a disease-modifying drug for the treatment of PD and other synucleinopathies. Electronic supplementary material The online version of this article (doi:10.1186/s40478-016-0310-y) contains supplementary material, which is available to authorized users.


Introduction
In the coming decades the number of patients affected by neurodegenerative disorders, including Parkinson's disease (PD), will inevitably increase [12]. Neurodegeneration in PD is accompanied by the appearance of Lewy bodies, intra-neuronal protein aggregates with α-synuclein (α-Syn) as major component [4]. Initially established as histopathological marker for PD, aggregated α-Syn is now regarded as key pathogenic agent [4]. Although α-Syn is a relatively small molecule of 140 amino acids and can access different cellular compartments [66], it is predominantly located in presynaptic terminals, associated to membranous structures like synaptic vesicles [4]. The discovery that point mutations in α-Syn like A53T or A30P as well as α-Syn gene multiplications facilitate the development of PD, underlines the importance of α-Syn in this disease [28,47,52]. Furthermore, α-Syn pathology is not restricted to dopaminergic cells, but was also identified in neuronal cell populations of the enteric nervous system and olfactory bulb, explaining the variety of non-motor symptoms in PD [48]. Several animal models using α-Syn mutants were generated, either virus-mediated [38], or using transgenic approaches. Transgenic mouse lines expressing human α-Syn-A30P or -A53T develop motor and cognitive impairments, being in part similar to human PD symptoms, along with nigrostriatal degeneration and aggregate pathology [16-19, 31, 40].
In previous studies in cell culture and animal models of PD, Rho-associated protein kinase (ROCK) was identified as novel molecular neuroprotective target [27,50,58,59,61]. Pharmacological ROCK inhibition protected against MPTP toxicity, attenuating dopaminergic cell death and increasing regenerative sprouting in vivo [59]. Since α-Syn aggregation is a major step in the pathogenesis of PD, we now evaluated the anti-aggregative potential of pharmacological ROCK inhibition using the isoquinoline derivative Fasudil, a small molecule inhibitor already approved for clinical use in humans in Japan [44].
In a cell culture model in H4 human neuroglioma cells [42], we studied the effects of Fasudil treatment on α-Syn aggregation. In a cell-free aggregation assay the direct interaction of Fasudil and α-Syn was investigated and nuclear magnetic resonance (NMR) spectroscopy was then used to identify the binding sites of Fasudil on α-Syn. Finally, a long-term treatment study in the α-Syn A53T mouse model [16] was performed to investigate the impact of Fasudil application on α-Syn pathology in vivo. Our results identify Fasudil as a drug with high translational potential as disease-modifying treatment for PD and other synucleinopathies.

Materials and methods
Cell free α-Syn aggregation assay Recombinant α-Syn was expressed and purified as described previously [22]. Samples containing α-Syn were diluted to a concentration of 100 μM in a 50 mM HEPES (Sigma, Taufkirchen, Germany) buffer with 100 mM NaCl (Sigma) and 0.03 % NaN 3 (Sigma) at pH 7.4 in a 500 μl total volume and subsequently subjected to aggregation-prone conditions (37°C,~300 rpm) in low binding protein 1.5 ml Eppendorf tubes during at least 14 days in the absence and presence of increasing concentrations of Fasudil (LC Laboratories, Woburn, MA) and Y-27632 (Tocris Bioscience, Bristol, UK) corresponding to compound to protein ratios from 1:1 to 40:1. At different incubation times, a 10 μl aliquot of the sample was mixed with 1 ml of a 100 mM ThT (Sigma) solution. Its fluorescence emission was measured between 460 and 600 nm on a Varian Cary Eclipse fluorescence spectrophotometer (Agilent Technologies, Santa Clara, CA) with excitation wavelength of 442 nm at 20°C and with a 5 nm excitation and emission slit. The averaged intensity values of the maximum peak (~480 nm) were fitted to a sigmoidal equation and presented as a function of time.
Electron microscopy of α-Syn fibrils A solution containing protein was applied to glowdischarged carbon coated grids and stained with 1 % uranyl acetate (Sigma). Images were taken in a Philips CM120 electron microscope (Philips, Amsterdam, The Netherlands) at a defocus of 2.3 μm using a TemCam 224A slow scan CCD camera (TVIPS, Gauting, Germany).

SEC-HPLC and dotblot analysis
H4 neuroglioma cells were collected 24 h after transfection in a phosphate buffer (1X PBS with 0.5 % TritonX-100) freshly supplemented with protease inhibitor cocktail (Roche, Mannheim, Germany) and centrifuged for 10 min at 10,000 g. 2-3 mg of total protein in a maximum volume of 500 μl was filtered using a 0.45 μm Spin-X centrifuge filter before loading onto a Superose 6 (Superose 6 10/ 300GL. GE Healthcare Life Science, Sweden) column and subsequent high-performance liquid chromatography (HPLC) (Äkta Purifier 10, GE Healthcare). The run was performed with a flow rate of 0.5 ml/min and fractions of 500 μl were collected. For the dot blot assay, these fractions were boiled at 95°C for 10 min and centrifuged at 10,000 g for 5 min. Afterwards they were loaded to a nitrocellulose membrane with the following procedure being identical to the previously described Western blot method.

Animal experiments
Animals were treated according to the regulations of the local animal research council and legislation of the State of Lower Saxony, Germany (33.19-42502-04-12/0884). Heterozygous breeding couples of the transgenic mouse strain B6;C3-Tg(Prnp-SNCA*A53T)83Vle/J (short: α-Syn A53T ) were bought from Jax Labs (J004479; Bar Harbor, ME) and maintained at the Central Animal Care Unit of the University Medicine Göttingen, Germany. Offspring were genotyped using a qPCR protocol according to Jax Labs' instructions to discriminate homozygous, heterozygous and wt animals. We selected homozygous transgenic mice and wt littermates for our experiments. Mice were housed in groups of five in individually ventilated cages (IVC, Tecniplast, Hohenpeißenberg, Germany) under a 12 h light/12 h dark cycle with free access to food and water. To monitor for changes in weight, all mice were weighed once a week. From day 50 on mice were treated with Fasudil at a dosage of 10 mg or 30 mg/kg bodyweight (bw) via the drinking water. A daily drinking amount of 6 ml per animal was taken as a basis and closely monitored to ensure an adequate dosage as described before [58][59][60]. Treatment groups were as follows: wildtype control (wt ctrl), α-Syn A53T untreated control (A53T ctrl), α-Syn A53T mice treated with 10 mg/kg bw (A53T Fas10) and α-Syn A53T mice treated with 30 mg/kg bw (A53T Fas30). To assess possible acute effects of Fasudil treatment, a separate cohort of α-Syn A53T mice was treated with 30 mg/kg bw Fasudil in the drinking water for 24 h, or with 20 mg/kg bw Fasudil via oral gavage. Before each treatment, α-Syn A53T mice were tested on rotarod and Catwalk to obtain a basic score. After 24 h Fasudil treatment via drinking water, or 30 min after oral gavage, α-Syn A53T mice were tested again on rotarod and Catwalk, and the results obtained were compared to baseline scores.
Blood of randomly selected mice was analyzed to monitor for alterations of standard clinical chemistry and hematology parameters due to Fasudil treatment. Approximately 200 μl blood were collected for hematological analysis from the retrobulbar venous plexus and transferred to ethylene diamine tetra-acetic acid (EDTA) tubes, the age of mice ranged from 200 to 700 days. For biochemical analysis, about 1 ml blood was collected under deep anesthesia with ketamine 200 mg/kg bw (Medistar, Ascheberg, Germany) and xylazine 10 mg/kg bw (Ecuphar, Greifswald, Germany) by heart puncture and transferred to heparin tubes, mice were sacrificed immediately afterwards. Here, the age of mice ranged from 100 to 700 days. Blood samples were analyzed in the core facility of the Clinical Chemistry Department of the University Medicine Göttingen, Germany. Since α-Syn A53T mice develop a severe movement disorder starting with a paresis of the hind limbs, earliest onset after about 8 months of age [16], mice were closely monitored from day 220 on regarding weight and motor behavior on the rotarod. After clinical symptoms manifested, mice underwent for the last time additional behavioral tests as described below. To prevent animals from suffering further paresis, mice were sacrificed after behavioral tests were performed.

Rotarod
To measure motor balance and coordination, mice were tested on an accelerated rotarod (rotarod for mice 47600, Ugo Basile, Comerio, Italy) ranging from 5 to 40 rpm within 5 min as described before [58,59]. All mice were pre-trained on the rotarod in order to reach a stable performance. Rotarod test was performed after 100 d, 200 d, and from day 220 on twice a week, each time in three sessions with an intertrial interval of 30 min to reduce stress and fatigue. Rotarod performance in all three runs was recorded and the average time on the rotarod was evaluated. The continuous rotarod testing twice a week from day 220 on was used to detect the onset of clinical pathology described for this mouse model [16] as early as possible for each animal. When rotarod performance of a single mouse dropped below 50 % of its long term average, the respective animal was termed as 'disease onset' and ultimately tested in Catwalk gait analysis and novel object recognition test.

Catwalk gait analysis
Catwalk XT gait analysis system (Noldus, Wageningen, The Netherlands) was used to monitor gait performance of α-Syn A53T mice and wt controls as described before [50]. To address possible changes in gait performance during disease progression, mice were tested before possible clinical symptoms at 200 days, as well as at 'disease onset'. During the test animals were placed in a walkway of 4 cm width and videotaped from below. Footprints were automatically detected by the Catwalk XT 10.0 software. Detection settings were as follows: camera gain 20; intensity threshold 0.10; max. allowed speed variation 60 %. Three compliant runs per animal were recorded and means out of these runs were analyzed over treatment groups with Catwalk XT 10.0 gait analysis software.

Novel object recognition test
To monitor for cognitive deficits in α-Syn A53T mice, the novel object recognition (NOR) test was performed as described before [6]. Briefly, single mice were placed in an arena of 48 × 35 cm for 3 min to habituate, and then two identical objects were put into the arena to be explored by the mouse for 5 min to familiarize. After a 10 min recovery in the home cage, mice were put back into the arena for 5 min, where one of the familiar objects had been changed to a novel one, which the mouse had never seen before. During the test mice were videotaped, and the time spent with familiar and novel object, as well as general movement parameters (mobility, velocity, distance moved) were detected by Ethovision XT 8.5 software (Noldus ; Fig 6a, b). The discrimination ratio (time spent with novel object/time spent with both objects) was calculated as described before [6] and compared between the treatment groups.

Immunohistochemistry
Mice were sacrificed by exposure to carbon dioxide and directly afterwards perfused transcardially with PBS, followed by 4 % paraformaldehyde (PFA; Applichem) in PBS (pH 7.4). Brains were dissected, post-fixed in PFA overnight at 4°C, and finally embedded into paraffin. Brains were coronally sectioned into 5 μm slices, mounted on slides and deparaffinized by xylene (Sigma) treatment for 2x 10 min, followed by a descending ethanol (EtOH) row (2x 3 min 100 % EtOH (Applichem), 96 %, 90 %, 70 %, 50 % EtOH for 3 min each). For immunohistochemistry, sections were incubated for 30 min in citrate buffer (10 mM citric acid (Sigma), 0.05 % Tween 20 (Applichem), pH 6.0 at 80°C. After cooling down, sections were either treated with 10 μg/ ml proteinase K (PK, Applichem) in PBS at 55°C for 12 min to quantify α-Syn pathology, or just washed with PBS, followed by incubation in 0.1 g sudan black (Applichem) per 100 ml 70 % EtOH. After rinsing with water and PBS washing, sections were incubated for 20 min in 25 mM glycine (Applichem) in PBS. Afterwards sections were blocked for 1.5 h in a solution of 10 % normal horse serum (NHS, PAA), 5 % bovine serum albumin (BSA, Applichem), 0.3 % Triton-X 100 (Applichem) and 25 mM glycine in PBS. First antibodies against α-Syn (1:500, Covance), TH (1:1000, Zytomed), TPH2 (1:100, Millipore) or NeuN (1:100, Millipore) diluted in blocking solution were applied to the sections and incubated in a wet chamber over night at 4°C. After 3x washing in PBS respective secondary antibodies were used (goat antimouse Cy3, 1:250; goat anti-rabbit Cy5, 1:250) and incubated for 1.5 h at room temperature. After 3x washing in PBS, cell nuclei were counterstained with DAPI (Sigma) for 2 min, after final washing in PBS sections were dried and embedded with Mowiol (Sigma). From each brain region, namely red nucleus (RN), substantia nigra (SN) and dorsal raphe nucleus (DR), three to five sections were stained and evaluated by a blinded investigator. Cells with proteinase-K resistant α-Syn aggregates as a marker for α-Syn pathology, as well as neuronal cells were counted in the respective brain regions. The means of at least three independent stainings were calculated and normalized to the respective region's area as determined by immunohistochemical staining against NeuN for the RN, TH for the SN and TPH2 for the DR.

Statistical analysis
Statistical analyses were conducted using Kyplot software (Version 2.0, KyensLab Incorporated, Tokyo, Japan). Comparisons of two groups were done by unpaired Student's T-Test, multiple group comparisons by one-way ANOVA with Dunnett post-hoc test. The statistical test and number of in vitro experiments or animals used for analysis is indicated in each figure legend. Data are presented as mean ± SD or mean ± SEM, as indicated. Differences were considered significant when p < 0.05 (*p < 0.05; **p < 0.01; ***p < 0.001; n.s. = not significant).

Fasudil alters synT aggregation in the H4 aggregation model
Co-transfection of synT and SYPH1 leads to formation of intracellular inclusions within 24 h after transfection [30,42]. Treatment with Fasudil, beginning 24 h before transfection, reduced the number of transfected H4 cells with inclusions in a dose-dependent manner within 24 h (Fig. 1a, b).
As compared to untreated control (ctrl: 21.8 ± 1.95 %), treatment with 20 μM Fasudil significantly reduced the percentage of cells with inclusions by almost 50 % (Fas20: 12.8 ± 0.82 %, p = 0.007, ANOVA with Dunnett post-hoc test) (Fig. 1c). Western blot analysis of protein lysates collected 24 h after transfection showed that Fasudil treatment did not significantly change synT protein levels (Fig. 1d). In comparison, Y-27632 treatment did not have any significant effect on the number of transfected H4 cells with inclusions (Additional file 1: Figure S1). Therefore, a concentration of 20 μM Fasudil was used in the ongoing cell culture experiments.
SEC-HPLC was used to separate protein fractions according to size with regard to gel filtration standard components. Protein fractions were collected and analyzed by dot-blot against synT. Fasudil treatment significantly decreased the number of high molecular weight aggregates, and simultaneously increased the protein amount in the fraction <17 kDa (fraction 1 > 670 kDa: 44 (Fig. 1e).

Fasudil attenuates α-Syn aggregation by direct C-terminal interaction
α-Syn is a natively unfolded protein that upon exposure to aggregation-prone conditions forms amyloid fibrils that are reactive with Thioflavin T (ThT) and are visible by electron microscopy (EM). To probe the influence of Fasudil on the aggregation kinetics of α-Syn, we monitored the change in ThT fluorescence of samples both in the presence and absence of Fasudil as well as the structurally unrelated ROCK-inhibitor Y-27632, a 4-aminopyridine derivative (Fig. 2a). The data show a clear delay in α-Syn aggregation in presence of Fasudil, increasing the lag phase from~2.5 to~8 days. Comparison of Fasudil (Fas) to control (ctrl) after 9 days revealed a significant difference (ctrl: 1.01 ± 0.16, n = 3; Fas: 0.12 ± 0.06; n = 3, p = 0.0002, ANOVA with Dunnett post-hoc test), while the ROCK inhibitor Y-27632 had no effect on α-Syn aggregation in this assay (Y-27632: 0.95 ± 0.13, n = 3, p = 0.79, ANOVA with Dunnett post-hoc test; Fig. 2a). It is worth noting that the slope of the curve associated with elongation of the fibril was steeper in the control and Y-27632 samples as compared to Fasudil, suggesting that Fasudil not only affects formation of aggregation seeds but also the elongation of amyloid fibrils of α-Syn. The lack of defined amyloid fibrils after 10 days of aggregation in the presence of Fasudil was further corroborated by EM (Fig. 2b), as opposed to the classical mature fibrillar structures formed in the absence of Fasudil.
In order to obtain insight into the molecular nature of the interaction between α-Syn and Fasudil, we took advantage of solution nuclear magnetic resonance (NMR) analysis. NMR resonances are a reflection of the microenvironment in which individual amino acids are found and, therefore, are highly sensitive probes of protein-protein and/or protein-ligand interactions, allowing a detailed description of interaction interfaces. Here, we made use of two-dimensional 15 N-1 H HSQC experiments in the presence of increasing amounts of Fasudil (Fig. 2c). The perturbation in the relative position of signals in the NMR spectra allowed us to determine that Fasudil preferentially interacts with the C-terminal region of α-Syn, especially with the aromatic residues Y133 and Y136, as determined by the large chemical shift perturbation of their NMR resonances (Fig. 2d). A structural representation of α-Syn and Fasudil's interaction through the Cterminal region, specifically in residues Y133 and Y136, is given in Fig. 2e. The α-Syn model was taken from a molecular dynamics simulation [46] with Fasudil placed manually in the proximity of residues Y133 and Y136 with the molecular visualization software VMD [24].

Effect of Fasudil on SynT aggregation is attenuated after Y133/136A substitution
To address the influence of Y133 and Y136 on α-Syn aggregation in the H4 cell culture model in vitro, both residues were substituted with alanine in human synT by site-directed mutagenesis, either as single (synT-Y133A and synT-Y136A) or double mutant (synT-Y133,136A). Number, size and distribution of inclusions in single H4 cells were not affected after transfection with the respective mutant synT as compared to wildtype (wt) synT (Fig. 3a).

Effects of long-term treatment of α-Syn A53T mice with Fasudil
To investigate the influence of long-term Fasudil treatment on α-Syn pathology in vivo we used a mouse model expressing human α-Syn A53T under the prion protein (Prnp) promoter [16]. Homozygous α-Syn A53T mice develop a neuronal α-synucleinopathy with intraneuronal inclusions of α-Syn, triggering a severe movement disorder beginning earliest at 8 months of age with weight loss, followed by a paresis of the hind limbs, finally leading to paralysis and death. In this study α-Syn A53T mice were treated with 10 or 30 mg/kg bw Fasudil from day 50 on until death (Fig. 4a), and compared to untreated α-Syn A53T mice and wt controls were weighed weekly to monitor for weight changes over the experimental time course. Regarding their maximum weight, α-Syn A53T mice were insignificantly lighter (~33 g) than wt controls (~35 g). After onset of clinical symptoms, α-Syn A53T mice rapidly lost weight as previously described [16], being statistically significant at time of sacrifice as compared to wt controls (wt ctrl: 34.0 ± 1.68 g; A53T ctrl: 26.7 ± 0.79 g, p = 0.003, T-Test; A53T Fas10: 27.7 ± 1.23 g; A53T Fas30: 26.1 ± 1.29 g). Fasudil treatment did not have any significant effect on the weight of α-Syn A53T mice (Fig. 4b, c).
Blood analysis did not show any significant differences between Fasudil-treated and untreated animals. Leukocyte (LEU) counts were detected between 3.1 and 9.0 × 10 3 /μl, erythrocytes (ERY) between 7.35 and 12.6 × 10 6 /μl, and thrombocytes (THR) ranged between 388 and 2102 × 10 3 / μl. Hemoglobin (HGB) ranged from 9.1 to 17.0 g/dl and hematocrit (HCT) from 26.4 to 50.6 %. Biochemical analysis of renal and hepatic function parameters revealed maximum values between 0.15 and 0.21 mg/dl of creatinine (CREA), aspartate aminotransferase (AST) was measured between 53 and 312 U/l, alanine aminotransferase (ALT) maximum values ranged between 18 and 113 U/l, and gamma-glutamyl transferase (GGT) was below detection level of <4 U/l (Additional file 2: Figure S2). Fasudil treatment does not affect weight, rotarod performance or overall survival of α-Syn A53T mice. a 50 days after birth (DOB = day of birth) of α-Syn A53T mice, Fasudil treatment with 10 mg/kg bw and 30 mg/kg bw was started. After 100 and 200 days, mice were tested on a rotarod (Rot). From day 220 on, rotarod was performed twice a week to monitor for onset of clinical symptoms of α-Syn A53T mice. b, c Weight of α-Syn A53T mice was monitored weekly. The last 10 weight measurements are displayed (b), as well as last weight at day of sacrifice (c). d, e Rotarod analysis of α-Syn A53T mice treated with 10 mg/kg bw and 30 mg/kg bw Fasudil as compared to untreated α-Syn A53T controls and wildtype controls. The last 10 rotarod performances before sacrifice are displayed for all animals in the respective treatment groups (d), as well as the last rotarod performance before sacrifice, showing no significant differences between α-Syn A53T mice with regard to Fasudil treatment (e). f Survival of α-Syn A53T mice treated with 10 mg/kg bw and 30 mg/kg bw Fasudil as compared to untreated α-Syn A53T controls, showing no significant differences between the groups. All data are given as means ± SEM; n(wt ctrl) = 10, n(A53T ctrl) = 10, n(A53T Fas10) = 10, n(A53T Fas30) = 9; **p < 0.01, ***p < 0.001, n.s. = not significant; T-Test for comparisons between ctrl groups; ANOVA with Dunnett post-hoc test for multiple group comparisons between α-Syn A53T groups Rotarod performance of α-Syn A53T mice is impaired with disease onset α-Syn A53T mice and wt controls were pre-trained on the rotarod to obtain a stable performance. Rotarod testing after acute Fasudil treatment did not show any significant effects (Additional file 3: Figure S3). Rotarod testing at 100 and 200 days of age showed a stable performance without any signs of impairment in α-Syn A53T mice (data not shown). Beginning from day 220 on, mice were tested twice a week on the rotarod, to monitor motor impairment as indication of onset of the clinical symptoms described for this model [16] (Fig. 4a). All mice maintained a stable rotarod performance over time (Fig. 4d), until with the onset of clinical symptoms α-Syn A53T animals displayed a massive rotarod failure as a clear marker for declining motor function. Fasudil treatment did not improve rotarod performance as compared to untreated α-Syn A53T mice. Figure 4e shows means ± SEM of last rotarod performance before sacrifice (wt ctrl: 277 ± 8 s; A53T ctrl: 57 ± 13 s; A53T Fas10: 47 ± 10 s; A53T Fas30: 55 ± 14 s).

Survival of α-Syn A53T mice is not altered by Fasudil treatment
After the onset of clinical symptoms and when mice undercut 50 % of the initial rotarod performance, mice were assessed for abnormalities in gait (Catwalk gait analysis) and cognition (novel object recognition test). Due to the fast progression of the disease in this particular model, α-Syn A53T mice were sacrificed after the last behavioral test when hind limb paresis was present, to prevent animals from suffering. With regard to Fasudil treatment, no significant differences in survival time between the groups were detected (A53T ctrl: 509 ± 37 d; A53T Fas10: 428 ± 31 d; A53T Fas30: 523 ± 41 d) (Fig. 4f ).

Catwalk gait analysis reveals motor improvement after Fasudil treatment
To detect more subtle differences in gait of α-Syn A53T mice after treatment with Fasudil, mice were tested with the Catwalk gait analysis system, since this method is more sensitive than the rotarod [62]. α-Syn A53T mice were tested after 'disease onset' as detected by rotarod failure (Fig. 5a). Footprints of the mice were automatically detected, and various gait parameters were analyzed (Fig. 5 b, c). Before disease onset at day 200, no significant differences between the groups were apparent (data not shown). Acute Fasudil treatment also did not have any significant effects in the multiparametric gait analysis (Additional file 4: Figure S4). After mice displayed clinical symptoms, long-term Fasudil treatment significantly improved gait performance of α-Syn A53T mice as compared to untreated controls. In detail, run average speed was improved dose-dependently as compared to α-Syn A53T control, with high Fasudil dosage reaching significance (wt ctrl: 28.6 ± 2.11 cm/s; A53T ctrl: 13.6 ± 1.20 cm/s, p = 7.05x10 -6 , T-test; A53T Fas10: 19.7 ± 1.85 cm/s; A53T Fas30: 21.7 ± 3.24 cm/s, p = 0.03, ANOVA with Dunnett post-hoc test; Fig. 5d).

Syn pathology
To investigate the effect of Fasudil treatment on histopathology in the brain of α-Syn A53T mice, an immunohistochemical analysis was performed after 200 days as well as after disease onset (Fig. 7a). Cells showing α-Syn aggregates after proteinase-K digestion were quantified in three different brain regions, namely red nucleus (RN), substantia nigra pars compacta (SN) and dorsal raphe nucleus (DR), and compared between the different treatment groups and wt littermate controls. After 200 days no proteinase-K resistant α-Syn aggregates could be detected in α-Syn A53T mice (data not shown). As expected, no α-Syn pathology was detected in the wt controls, whereas at diseaseonset α-Syn A53T mice showed α-Syn aggregates in all investigated brain regions. Interestingly, Fasudil treatment significantly decreased the number of cells showing α-Syn aggregates.
Discussion α-Syn aggregation is a major pathophysiological step in the development of PD. Since its discovery [41] and consequent identification as most abundant protein component in Lewy bodies [53], α-Syn was shown to be involved in numerous cellular processes [4,14,29,66]. Therapeutic approaches addressing the aggregation of α-Syn thus hold the promise to result in true diseasemodification and attenuate progression in PD.
Recently, ROCK was characterized as promising new molecular therapeutic target. Its pharmacological inhibition resulted in prevention of neurodegeneration, fostered axonal regeneration and beneficially regulated microglial dysfunction in models of PD and other neurodegenerative disorders [3,8,26,27,50,[58][59][60][61]63]. Here, we evaluated the anti-aggregative potential of pharmacological ROCK inhibition using Fasudil, which is already approved for clinical use in humans [44].
In contrast to the structurally unrelated 4-aminopyridine derivative Y-27632, Fasudil treatment significantly reduced the number of H4 cells with SYPH1/synT inclusions in vitro, which was confirmed by SEC-HPLC followed by dotblot analysis. To understand the molecular basis of this reduced aggregate formation, we performed a cell-free thioflavin-T aggregation assay. Interestingly, again in contrast to Y-27632, Fasudil significantly delayed α-Syn aggregation, which was affirmed by electron microscopy and indicated a direct inhibiting function of Fasudil on α-Syn aggregation. Since this effect was observed in a cell-free assay, it was not mediated by the ROCK-inhibiting properties of Fasudil. NMR spectroscopy revealed a direct binding of Fasudil to two tyrosine residues in the C-terminal region of α-Syn, namely Y133 and Y136, and sitedirected mutagenesis of these residues confirmed the biological relevance of this interaction in the H4 aggregation model in vitro. The C-terminal region was recently shown to be involved in α-Syn fibrillization [25], with C-terminal truncation accelerating the ability of α-Syn to form fibrils in vitro [32] and in vivo [35,37], whereas negative charge in the Cterminus of full-length α-Syn delayed fibril formation [23,25]. Interestingly, Izawa and colleagues suggested Y136 as critical element in promoting fibril formation [25]. Our data now show that Fasudil binds to precisely this particular C-terminal tyrosine. This data prompted us to assess the effects of Fasudil treatment in an in vivo-model of α-Syn aggregation.
The influence of long-term Fasudil treatment on α-Syn pathology in vivo was investigated in the α-Syn A53T mouse model [16], corresponding to a mutation found in a familial form of PD [47]. Animals were treated with Fasudil from day 50 on until death, and no acute Fasudil effects with regard to motor behavior were detected. The average survival time of α-Syn A53T mice with high Fasudil treatment was more than 500 days, and did not differ significantly from untreated α-Syn A53T mice. Fasudil has already been used in animal models of neurodegeneration and was well tolerated in treatment paradigms for up to four months [58][59][60]. Here we present the first longterm animal study supporting the favorable safety profile of Fasudil over more than 15 months average treatment time, without any obvious unwanted side effects. Importantly, blood analysis revealed no significant changes after Fasudil treatment, with all parameters being in the normal range for laboratory mice [7,9,11,49]. Disease onset could be clearly determined by rotarod failure, which is in line with previous findings [16]. No rotarod failure of α-Syn A53T mice was detected before disease onset, although recent studies report contradictory findings in the same model [19,45,68].
To further investigate motor performance, the Catwalk gait analysis system was used. In contrast to the rotarod, here the animals walk freely and are not forced due to an automated treadmill, which allows a highly sensitive detection of subtle gait changes [62]. α-Syn A53T mice showed a massive decline in stride length and velocity at disease onset, which could be translated to gait deficits observed in PD patients [33,54] and is in line with previous findings in this model [17,45]. Interestingly, we show that numerous gait parameters were significantly improved by long-term Fasudil treatment in a dosedependent manner.
Since Fasudil attenuated aggregation of α-Syn in vitro, we analyzed whether its application in the α-Syn A53T mouse model also results in a histological improvement in vivo. After 200 days no α-Syn pathology could be detected in α-Syn A53T mice, which is in line with previous findings in this model [16]. Remarkably, the amount of neurons in the RN and SN showing α-Syn pathology after proteinase-K digestion at disease-onset was significantly decreased after Fasudil treatment, whereas the number of NeuN-positive cells in the RN was not affected. The number of THpositive cells in the SN was also not affected, which is in line with previous findings showing that TH neurons in the SN of α-Syn A53T mice do not display the same selective vulnerability as in humans [16]. Since RN and SN are highly involved in motor function and coordination in rodents, increased α-Syn pathology in these regions was suggested to compromise their function and thus contribute to the motor deficits [16]. We hypothesize that specific binding of Fasudil to α-Syn, as demonstrated by NMR analysis, contributes to the reduction of α-Syn pathology in RN and SN, thereby improving fine motor function.
In addition to motor deficits, PD patients show numerous non-motor symptoms, including cognitive decline [5,10]. We therefore analyzed the presence of cognitive impairments and the putative therapeutic effect of Fasudil in α-Syn A53T mice. The NOR-test has been extensively used to detect cognitive deficits in numerous disease models [6,20,39]. In contrast to a previous study in a synucleinopathy mouse model [40], we did not detect any significant differences in cognitive behavior after 200 days in our α-Syn A53T mice. However, at onset of motor symptoms there was a significant alteration in the novelty seeking behavior, which was reversed by Fasudil treatment. This is in line with previous findings showing significant deficits in short term memory in this mouse line, postulated to be mediated by a dysfunctional dopamine system [36,45,51,67]. In addition to dopaminergic projections, other neurotransmitter systems are known to contribute to motor and particularly non-motor symptoms in PD. For example, the serotonergic system is involved in memory function in PD [21]. The reduction of α-Syn pathology in the DR together with the recovery of recognition memory by Fasudil suggests that drugs acting in a disease-modifying manner by modification of α-Syn aggregation may also successfully target non-motor symptoms induced by non-dopaminergic pathology.
In regard to transfer of these findings into clinical trials, Fasudil has a highly promising translational potential, as it is already approved for clinical use in humans in Japan, showing a very favorable safety profile [44]. Fasudil is active after oral administration and is capable of crossing the blood-brain-barrier [56,60]. Although other pharmacological ROCK inhibitors, e.g. Y-27632, are more selective towards ROCK, the latter compound did not affect α-Syn aggregation. Interestingly, Y-27632 was shown to improve degradation of mutant huntingtin as well as motor performance in the R6/2 mouse model of Huntington's disease [34]. Besides, the effect of NSAIDs on lowering the amount of Aβ42 in models of Alzheimer's disease is mediated by ROCK inhibition and can be mimicked by Y-27632 [69]. Thus, in addition to the specific anti-aggregative effect of Fasudil through direct molecular interaction with α-Syn, ROCK inhibition may also contribute to a depletion of aggregationprone proteins. Increased levels of ROCK were also recently discovered in brains of patients with progressive supranuclear palsy and corticobasal degeneration, and ROCK inhibition attenuated tau pathology in Drosophila [15]. Thus, different lines of evidence suggest that ROCK is involved in the pathogenesis of aggregation disorders and that ROCK inhibition represents a promising therapeutic strategy [50,[58][59][60][61].
Here we describe for the first time to our knowledge the anti-aggregative potential of the ROCK-inhibitor Fasudil via direct binding to α-Syn, which suggests a new treatment approach employing a double function of Fasudil. Although in recent years a number of α-Syn aggregation inhibitors have been described [1,13,55,57], most of them are "pan-assay interference compounds" with a questionable clinical perspective [2]. High-throughput screening revealed more specific compounds like the α-Syn-oligomer-binding Anle-138b [65] or recently discovered monomeric α-Syn-binding BIOD303 [43], appearing more auspicious. One of the major obstacles of these new compounds, however, is the so far missing approval for human treatment with largely unclear safety and tolerability issues. Since Fasudil is already approved for clinical human treatment, the repurposing of this drug for the treatment of PD and other synucleinopathies appears therefore highly promising.

Conclusion
We describe here the anti-aggregative potential of the ROCK-inhibitor Fasudil via direct binding to α-Syn, which opens a new treatment avenue employing a ROCK-independent function of Fasudil in PD and other synucleinopathies. This is of particular interest, as Fasudil is already licensed in Japan for the treatment of subarachnoid haemorrhage-induced vasospasms and has been widely used with a beneficial safety profile. Since disease-modifying therapies for neurodegenerative diseases are urgently needed, the translation of our findings into a clinical pilot study appears realistic in the near future.
Ethical approval "All applicable international, national, and/or institutional guidelines for the care and use of animals were followed."

Additional files
Additional file 1: Figure S1. Y-27632 treatment has no effect on synT aggregation in H4 cells in vitro. a H4 neuroglioma cells were seeded in the presence or absence of Y-27632 and 24 h later transfected with plasmids encoding for synT and SYPH1. 24 h after transfection, cultures were investigated by immunocytochemistry (ICC). b ICC of synT and SYPH1 in H4 cells 24 h after transfection, treated with 5 or 20 μM Y-27632. Scale bar: 50 μm. c Quantification of transfected H4 cells with inclusions 24 h after transfection and treatment with different Y-27632 concentrations. n.s. = not significant, ANOVA, n = 3. Data are given as mean ± SEM, n.s. = not significant, ANOVA. (TIF 723 kb) Additional file 2: Figure S2. Blood cell counts and biochemical analysis of renal and hepatic function parameters in α-Syn A53T mice. a Leucocytes (LEU). b Erythrocytes (ERY). c Thrombocytes (THR). d