Neurobiology of Disease 69 (2014) 76–92

Contents lists available at ScienceDirect

Neurobiology of Disease journal homepage: www.elsevier.com/locate/ynbdi

Intracellular processing of disease-associated α-synuclein in the human brain suggests prion-like cell-to-cell spread Gabor G. Kovacs a,⁎, Leonid Breydo b, Ryan Green b, Viktor Kis c, Gina Puska c, Péter Lőrincz c, Laura Perju-Dumbrava a, Regina Giera a, Walter Pirker d, Mirjam Lutz a, Ingolf Lachmann e, Herbert Budka a,1, Vladimir N. Uversky b,f,g, Kinga Molnár c,2, Lajos László c,⁎⁎,2 a

Institute of Neurology, Medical University Vienna, Austria Department of Molecular Medicine, Morsani College of Medicine, University of South Florida, Tampa, FL, USA c Department of Anatomy, Cell and Developmental Biology, Eötvös Loránd University of Sciences, Budapest, Hungary d Department of Clinical Neurology, Medical University Vienna, Austria e AJ Roboscreen GmbH, Hohmannstrasse 7, 04129 Leipzig, Germany f USF Health Byrd Alzheimer's Research Institute, Morsani College of Medicine, University of South Florida, Tampa, FL 33612, USA g Institute for Biological Instrumentation, Russian Academy of Sciences, 142290 Pushchino, Moscow Region, Russia b

a r t i c l e

i n f o

Article history: Received 19 February 2014 Revised 29 April 2014 Accepted 17 May 2014 Available online 27 May 2014 Keywords: α-synuclein endosome ependyma gap junction internalisation lysosome mitochondria nanotube prion-like spreading

a b s t r a c t Dementia with Lewy bodies (DLB), Parkinson’s disease (PD) and multiple system atrophy are characterized by the deposition of disease-associated α-synuclein. In the present study we 1) examined the molecular specificity of the novel anti-α-synuclein 5G4 antibody; 2) evaluated immunoreactivity patterns and their correlation in human brain tissue with micro- and astrogliosis in 57 cases with PD or DLB; and 3) performed a systematic immunoelectron microscopical mapping of subcellular localizations. 5G4 strongly binds to the high molecular weight fraction of β-sheet rich oligomers, while no binding to primarily disordered oligomers or monomers was observed. We show novel localizations of disease-associated α-synuclein including perivascular macrophages, ependyma and cranial nerves. α-Synuclein immunoreactive neuropil dots and thin threads associate more with glial reaction than Lewy bodies alone. Astrocytic α-synuclein is an important component of the pathology. Furthermore, we document ultrastructurally the pathway of processing of disease-associated α-synuclein within neurons and astroglial cells. Interaction of mitochondria and disease-associated α-synuclein plays a key role in the molecular–structural cytopathogenesis of disorders with Lewy bodies. We conclude that 1) the 5G4 antibody has strong selectivity for β-sheet rich α-synuclein oligomers; 2) Lewy bodies themselves are not the most relevant morphological substrate that evokes tissue lesioning; 3) both neurons and astrocytes internalize disease-associated α-synuclein in the human brain, suggesting prion-like cell-to-cell spread of α-synuclein by uptake from surrounding structures, as shown previously in experimental observations. © 2014 Elsevier Inc. All rights reserved.

Introduction α-Synuclein is a 140 amino acid protein that is physiologically found mainly in a presynaptic location (Iwai et al., 1995). It belongs to the class of intrinsically unstructured proteins and it is characterized by a remarkable conformational plasticity (Uversky, 2003). Indeed, it may stay unfolded, or adopt an amyloidogenic partially folded conformation, or fold into α-helical or β-structural species, both monomeric and ⁎ Correspondence to: G.G. Kovacs, Institute of Neurology, Medical University of Vienna, AKH 4J, Währinger Gürtel 18-20, A-1097 Vienna, Austria. Fax: +43 1 40400 5511. ⁎⁎ Correspondence to: L. László, Department of Anatomy, Cell and Developmental Biology, Eötvös Loránd University of Sciences, Budapest, Hungary. Fax: +36 1 3812184. E-mail addresses: [email protected] (G.G. Kovacs), [email protected] (L. László). Available online on ScienceDirect (www.sciencedirect.com). 1 Present address: Institute of Neuropathology, University Hospital Zurich, Switzerland. 2 These authors contributed equally.

http://dx.doi.org/10.1016/j.nbd.2014.05.020 0969-9961/© 2014 Elsevier Inc. All rights reserved.

oligomeric (Uversky, 2003). In addition to the formation of oligomers, biochemical modifications like nitration and phosphorylation occur (Kovacs et al., 2010). A major role of α-synuclein was suggested for a group of neurodegenerative diseases called α-synucleinopathies. These include disorders that show mainly neuron-related accumulation of α-synuclein, like Parkinson`s disease (PD) and Dementia with Lewy bodies (DLB), while multiple system atrophy (MSA) is a disease characterized mainly by oligodendroglial cytoplasmic inclusions (Jellinger, 2012). PD and DLB share common morphological features described as Lewy bodies and Lewy neurites. Recent advances in immunohistochemistry revealed that dot-like structures and thin neurites are also present in the brains of patients with PD and DLB, which are likely better substrates for the clinical symptoms (Alafuzoff et al., 2008b, 2009; Leverenz et al., 2008). However, the currently commercially available anti-α-synuclein antibodies detect also the monomeric physiological form in addition to the pathological inclusions of α-synucleinopathies

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

(Beach et al., 2008; Kovacs et al., 2012b), which may cause misinterpretation of the immunostaining patterns. Application of disease-associated phosphorylation-specific antibodies provided an important step to avoid this obstacle (Saito et al., 2003). Depending on the epitope, some antibodies also label astrocytes correlating with the progression of disease (Braak et al., 2007). Indeed, α-synuclein pathology progresses through stages (Braak et al., 2003), which also raised the possibility of a prion-like spreading (Visanji et al., 2013). There are experimental data that strongly support this concept by demonstrating endocytotic uptake of α-synuclein (Visanji et al., 2013). Internalization of prions is a central event in prion diseases (Arnold et al., 1995; Baron et al., 2002; Borchelt et al., 1992; Caughey et al., 1991; Laszlo et al., 1992; Magalhaes et al., 2002; Taraboulos et al., 1992, 1995), which are the model disorders of cell-to-cell transmission of disease-associated protein spreading. In spite of the accumulating evidence of a similar process in experimental models of α-synucleinopathies (Holmes et al., 2013), there is a lack of ultrastructural evidence in the human brain that demonstrates the role of the endosomal-lysosomal system (ELS) in the processing of disease-associated α-synuclein. Evidence of α-synuclein internalization would be in line with experimental observations (Guo and Lee, 2014), even when the dynamic process of spreading of pathological proteins in the human brain is difficult to visualize in the static state of an autopsy. We recently reported that the commercially available antibody 5G4 detects disease-associated forms of α-synuclein, including oligomeric aggregates with high specificity; it proved to be very useful in the neuropathological diagnostic practice (Kovacs et al., 2012b). However, amyloid oligomers are structurally extremely diverse varying from primarily disordered (Ladiwala et al., 2011) to β-sheet rich (Ma and Nussinov, 2010; Wu et al., 2010). It was not clear whether this antibody has a preference of any specific oligomeric structure. Therefore, we examined the reactivity of this antibody with several α-synuclein oligomers with known structures, and then applied it for a detailed ultrastructural mapping of its subcellular localization. Materials and methods Oligomer and fibril preparation and characterization The 4D6 antibody was from Thermo Scientific (Waltham, MA, USA). Recombinant α-synuclein was a gift from Dr. Munishkina (University of California, Santa Cruz). All other reagents were from Sigma (St. Louis, MO, USA), Thermo Scientific (Waltham, MA, USA) or VWR (Radnor, PA, USA). Sample preparation α-Synuclein fibrils (0.4 mg/ml) were prepared in 20 mM acetate buffer, pH 3.5 in the presence of 0.1 M NaCl at 37 °C. Protein was initially dissolved in 5 mM NaOH at 4 mg/ml, incubated in this solution for 1 min and diluted into the final reaction buffer. The reaction mixture was incubated at 37 °C for 2 days with stirring. Nitrated α-synuclein was prepared as described below (Yamin et al., 2003). Specifically, α-synuclein (4 mg/ml, 850 μl) was incubated in Hepes buffer (20 mM, pH 7.5). Tetranitromethane (80 μl, 1% solution in ethanol) was added, and the reaction mixture incubated for 30 min. After that another 80 μl of 1% solution of tetranitromethane in ethanol was added, and the reaction mixture incubated for another 15 min. Then the solution was dialyzed against 20 mM Hepes (pH 7.5, 2× 500 ml) for 24 h using the dialysis membrane with the MWCO 3500. The resulting nitrated α-synuclein was converted to oligomers by stirring the protein solution (1 mg/ml) in 20 mM sodium phosphate buffer (pH 7.0) in the presence of 0.1 M NaCl at 37 °C for 24 h. To separate the oligomers from the remaining monomer, 200 μl of 1 mg/ml solution of oligomers in 20 mM sodium phosphate, 0.1 M NaCl, pH 7.0 was filtered via 30 kDa MWCO centrifuge filter (Pall) by centrifugation on the tabletop

77

centrifuge at 13,000 rpm for 10 min. The retentate was diluted back to 300 μl with the same buffer. Oxidized α-synuclein was prepared as previously described by oxidizing the protein with hydrogen peroxide (Zhou et al., 2010). α-Synuclein (4 mg/ml, 400 μl) was incubated in Hepes buffer (20 mM, pH 7.5) in the presence of 0.5% hydrogen peroxide for 20 min. After that the reaction was quenched with a solution of sodium thiosulfate (130 μl, 10% solution in water), and the reaction mixture incubated for another 10 min. Then the solution was dialyzed against 20 mM Hepes (pH 7.5, 2× 500 ml) for 24 h using the dialysis membrane with the MWCO 3500. The resulting oxidized α-synuclein was converted to oligomers by incubating the protein solution (0.75 mg/ml, 400 μl) in 20 mM sodium phosphate buffer (pH 7.0) in the presence of 0.1 M NaCl at 25 °C for 48 h. To separate the oligomers from the remaining monomer, 200 μl of 0.75 mg/ml solution of oligomers in 20 mM sodium phosphate, 0.1 M NaCl, pH 7.0 was filtered via 30 kDa MWCO centrifuge filter (Pall) by centrifugation on the tabletop centrifuge at 13,000 rpm for 10 min. The retentate was diluted back to 200 μl with the same buffer. Electron microscopy 10 μl aliquots of protein solutions (0.1–0.3 mg/ml) were adsorbed onto 200 mesh formvar/carbon-coated nickel grids for 5 min. The grids were washed with water (10 μl), stained with 2% uranyl acetate for 2 min and washed with water again. The samples were analyzed with a JEM 1400 transmission electron microscope (JEOL) operated at 80 kV. Fluorescence spectroscopy Fluorescence emission spectra of thioflavin T (excitation 442 nm, emission 470–500 nm) were measured using a JASCO FP-8300 spectrofluorometer at 25 °C. A solution of thioflavin T (400 μl, 3 μM) in buffer (10 mM Hepes, pH 8.) was placed into a 0.4 cm pathlength cell, and the fluorescence emission spectra were acquired with 100 nm/min scan speed in triplicate. Protein solution (either oligomers or fibrils) was added in small aliquots and changes in fluorescence recorded. Circular dichroism (CD) spectroscopy Far-UV CD (190–260 nm) spectra of proteins were measured using a JASCO J-815 spectropolarimeter at 25 °C. A solution of protein (250 μl, 0.15 mg/ml) was placed into a 0.1 cm pathlength cell, and the CD spectra were acquired with 20 nm/min scan speed at 0.1 nm step size and 1.0 nm bandwidth under constant purging with nitrogen. Four spectra were accumulated and averaged for each sample. CD of amyloid fibrils (suspension, ~1 mg/ml) was acquired in the 0.2 mm pathlength cell. Fitting of CD spectra was performed with CDPro software. ELISA For ELISA plates were coated with 10–600 ng of different α-synuclein preparations (monomers, oligomers or fibrils) using 0.1 M sodium bicarbonate, pH 9.6 as a coating buffer for 2 h at 37 °C. The plates were washed with Tris buffered saline with 0.02% Tween (TBS-T), blocked with 3% BSA for 1 h at 37 °C, washed 3 times with TBS-T again and incubated with a 5G4 antibody (10 ng/well in TBS-T containing 3% BSA) for 1 h at 37 °C. The plate was washed 3 times with TBS-T, incubated with the HRP-labeled anti-rabbit secondary antibody (diluted 1:5,000 in TBS-T containing 3% BSA, Santa Cruz Biotechnology) for 1 h at 37 °C, washed with TBS-T again and developed with 3,3′,5,5′tetramethylbenzidine (100 μl, Thermo Scientific). When the blue color developed, the reaction was stopped by addition of 100 μl of 1 M HCl, and the absorbance read at 450 nm. SDS-PAGE and Western blots For Western blots, the samples (0.3 mg/ml, 11 μl) were mixed with 4× LDS loading buffer (4 μl) and heated at 70 °C for 5 min. Samples were run on Novex 4%–12% Bis–Tris gels (Life Technologies) and transferred to PVDF. The membrane was developed with the 5G4 antibody

78

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

(1: 3000 dilution in 5% milk) and a secondary anti-rabbit antibody (1: 10,000 dilution in 5% milk). Membranes were developed with the Supersignal WestPico ECL kit from Thermo Scientific. For SDS-PAGE with silver staining, the samples (0.3 mg/ml, 3 μl per lane) were mixed with 4 × LDS loading buffer (1 μl per lane) and heated at 70 °C for 5 min. Samples were run on Novex 4%–12% Bis–Tris gels (Life Technologies) and developed with the silver staining kit (Pierce) according to the manufacturer’s directions. Immunogold characterization of monomer and fibril α-synuclein preparations Monomer and non-monomer fibril preparations of α-synuclein were performed as described before (Kovacs et al., 2012b). Five-microliter protein sample aliquots were dropped onto formvar/carbon-coated 400-mesh nickel grids. For immunogold labeling, the grids covered by deposited proteins were floated upside down on 50 μl drops of the following antibodies: KM51 (1:100, full length α-synuclein; Novocastra, Newcastle upon Tyne, UK), and 42/αS (1:100, aa 15-123; Transduction Laboratories, Lexington, KY), phosphorylated α-synuclein (α-synuclein phosphorylated Ser129; Wako, Osaka, Japan) and the 5G4 antibody (1:100) diluted in PBS containing 1% BSA and 1.5% milk powder for 30 min at RT. After washing with PBS–BSA, the specimens were incubated on 50 μl aliquots of 10 nm gold-conjugated goat anti-mouse IgG (1:100; British BioCell, Cardiff, UK) for 30 min at RT. The samples then were negatively stained with 2% uranyl acetate prior to the investigation by a JEOL 1011 electron microscope. Density of gold particles (number/μm2) was evaluated on the monomer preparation for each antibody. Kruskal– Wallis and Mann–Whitney tests were used to compare the density values (Laszlo et al., 1990). Evaluation of human samples Case selection A total of 57 patients (35 men) with neuropathologically confirmed Lewy-related pathology were included in this study (Table 1). The cases were recruited from our tissue bank obtained from routine diagnostic practice of the Institute of Neurology. Samples were collected following local regulations and the study was performed in the frame of a study (“Molecular neuropathologic investigation of neurodegenerative diseases”) approved by the Ethical Committee of the Medical University of Vienna. Immunohistochemistry Formalin fixed, paraffin-embedded tissue blocks (2.5 × 2.0 cm) were evaluated. Post mortem delay was between 12 and 30 h and formalin fixation time was between 2 and 8 weeks. Immunostaining was performed with the mouse monoclonal 5G4 antibody using paraffinembedded tissue sections of the medulla oblongata (containing also plexus choroideus and the ventricular wall of the 4th ventricle), pons, mesencephalon, amygdala with nucleus basalis, basal ganglia, thalamus, hippocampus (containing also plexus and the ventricular wall of the inferior horn of the lateral ventricle), entorhinal, temporal, frontal and cingular cortices. Cranial nerves were available in 32 cases, in some cases from both sides (thus a maximum of 39 vagal nerves were examined). In addition, Gallyas staining was performed on selected sections. We applied 1:2000 dilutions and a pretreatment of 10 min microwaving in citrate buffer (pH 6) followed by 5 min 80% formic acid treatment. In addition, we used monoclonal anti-ubiquitin (1:50,000, Millipore, Temecula, CA, USA), anti-p62 (1:500, BD Biosciences, USA), anti-HLA-DR (clone CR3/43, 1:100, DakoCytomation, Glostrup, Denmark), and polyclonal anti-GFAP (1:1500, DakoCy tomation). The DAKO EnVision© detection kit, peroxidase/DAB, rabbit/ mouse (Dako) was used for visualization of antibody reactions. Double immunolabeling was performed using 5G4 and the polyclonal anti-GFAP. The fluorescence-labeled secondary antibodies were Alexa Fluor (AF) 555 donkey anti-mouse IgG (1:200, Molecular Probes, Inc.,

Table 1 Clinicopathological summary of cases involved in this study. Abbreviations: NP: neuritic plaques score (CERAD); NFD-BB: Braak and Braak stages of neurofibrillary degeneration (Alafuzoff et al., 2008a); LRP-B: Lewy-related pathology according to Braak staging (Braak et al., 2003); CN: Cranial nerves (bold and underlined indicates those where α-synuclein immunoreactivity was seen); AGD: argyrophilic grain disease staged according to Saito et al. (Saito et al., 2004); BGG: basal ganglia; Hipp: hippocampus; Cbll: cerebellum; n.a: not available. Nr Age Sex NP NFDBB

LRP- Other B

1 2 3 4 5 6

61 78 78 70 77 81

m w m m m w

B B B B B C

3 3 4 4 4 5

6 5 6 6 6 6

– – AGD II – AGD III + Lacunes in BGG –

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

80 83 80 62 79 80 75 81 78 80 78 84 78 85 83 82 74

m w w m m m m m w m m m w m w m m

C C C C A No No A No A A A No No No No No

5 6 6 6 2 3 2 3 1 3 3 3 3 2 1 2 2

4 5 4 5 6 5 4 4 4 6 5 5 3 4 4 4 5

– AGD II TDP-43+ – – AGD III – – – – – – – – – – –

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

83 67 75 84 78 79 84 80 80 80 80 83 90 72 77 81 84 80 83 83 83 81 79 84 73

w w m m w m w w m m m w w m m w m w m w m w m m m

No No B No B B B No No B C C C C C C B B A B B A No No A

3 2 5 3 3 6 3 2 2 2 6 6 6 5 6 5 4 4 2 4 4 3 2 2 2

6 4 5 4 3 1 2 3 4 7 6 6 6 6 6 6 6 5 5 6 5 5 3 5 5

49 50 51 52 53 54 55 56 57

82 83 80 82 83 83 79 59 67

m w w m m w w m m

A A B A B B C B B

4 2 2 2 4 4 5 3 2

4 3 5 5 4 2 4 5 6

– – – Lacunar infarcts in BGG Lacunar infarcts in BGG – – – – – TDP-43+ – Microinfarct – – – – – AGD II – Microinfarct Cbll and Hipp Lacunar infarcts in BGG – Occipital infarct Lymphoma infiltration of meninges AGD II – – Microinfarct Cbll and BGG Cortical infarcts – Wernicke encephalopathy – –

CN n.a n.a n.a n.a n.a I, II, III, V, VII, VIII, X, XII n.a n.a n.a n.a n.a X X IV n.a n.a n.a I, II, III, V, X X III, X, XII X X, XII I, II,III, VI, VII, VIII, X n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a X, XII n.a X X III X X, XII n.a V, X I, II, X X X IV X X, XII IV, X, XII X, XII X, XII X, XII V, X XII III, X X, XII –

Eugene, OR, USA), AF 488 goat anti-rabbit (1:200, Molecular Probes, Inc.). We evaluated double immunofluorescent labeling with a Zeiss LSM 510 confocal laser microscope.

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

Data analysis of histopathology Sections immunostained for α-synuclein (5G4) were scored semiquantitatively (0–1–2–3; none, some, moderate, numerous) for the different types of α-synuclein immunoreactivities; and also for the extent of microglial activation and GFAP immunoreactivity (see Supplemental Fig. 1A). For statistical analysis we used SPSS version 21.0. Spearman test was performed to correlate between the mean score of α-synuclein immunoreactivities and degree of micro- and astrogliosis. Univariate analysis (linear regression) was used to evaluate the interaction between these variables. P value below 0.05 was accepted as significant. Pre-embedding immunogold electron microscopy After incubation of the deparaffinized tissue sections (case 18, Table 1) with anti-α-synuclein antibody (5G4), we applied ultrasmall nanogold (0.8 nm) conjugated secondary antibody (Aurion, Wageningen, The Netherlands) followed by silver-enhancing method. Slides were mounted with resin and examined by light microscopy. Selected areas containing immunoreactive structures were re-embedded for ultrathin sectioning (Kovacs et al., 2012a). We used a JEOL-1011 electron microscope to analyze the ultrastructural localization of α-synuclein-specific immunolabeling in the ultrathin sections. Freeze substitution and low temperature embedding in LR White Samples from particular brain regions (substantia nigra, amygdala, putamen; case 57, Table 1) were cut into small blocks (2 × 2 × 2 mm), fixed in a freshly prepared fixative solution containing 3.2% formaldehyde, 0.2% glutaraldehyde, 1% sucrose, 3 mM CaCl2 in 0.1 M Na-cacodylate buffer for overnight at 4 C. After extensive washing in Na-cacodylate, free aldehydes were quenched in 50 mM glycin–50 mM ammonium chloride in the same buffer and cryoprotected in 30% sucrose in Na-cacodylate for 24 h. The blocks were freezed in liquid nitrogen, and then transferred to anhydrous methanol at −70 °C that in some cases contained 0.5% uranylacetate. After 6 h the temperature were raised to −20 °C and the dehydration was continued for 24 h with gentle agitation. Then specimens were infiltrated with pure LR White at −20 °C for 24 h (three changes 8 h each) then polymerized for 96 h at − 20 °C using a DL-103 12 W Ultraviolet lamp. A corresponding sample was taken from the same anatomical regions for paraffin embedding and examined under light microscopy using immunohistochemistry for antibody 5G4 to confirm that the samples indeed contain pathological protein deposits. Post-embedding immunogold labeling Double labeling with two different size gold particles conjugated antibodies was performed as described before (Arnold et al., 1995; Laszlo et al., 1992). In addition to 5G4, the following antibodies were used: anti-EEA1 antibody (early endosome marker; rabbit polyclonal IgG; ab2900, developed against a synthetic peptide derived from within residues 1350 to the C-terminus of human EEA1; Cambridge CB4 0FL, UK); anti-Rab7 (rabbit polyclonal IgG; developed using a synthetic peptide corresponding to amino acid residues 163–177 of human Rab7; Sigma, Saint Louis, Missouri, 63103 USA); and CD63 (Lamp3, lysosomal membrane glycoprotein; rabbit polyclonal antibody IgG; raised against amino acids 45–238 mapping at the C-terminus of CD63 of human origin; sc-15363, Santa Cruz Biotechnology, Inc., 69115 Heidelberg, Germany) (Laszlo et al., 1991). Results Oligomer and fibril preparation and characterization Since there are significant difficulties involved in production of homogeneous oligomer populations (Lesne, 2013), we have used chemically modified derivatives of α-synuclein. These chemical modifications inhibit conversion of α-synuclein to fibrils allowing it to form distinct populations of amyloid oligomers. While beta sheet stacking in oligomers

79

can accommodate a variety of defects (Ma and Nussinov, 2010), it's not the case for fibrils. Thus many chemical modifications, like lysine acylation (Siegel et al., 2007) or tyrosine nitration (Uversky et al., 2005) are compatible with relatively loose oligomer structure but incompatible with additional protein-protein interactions present in the fibril structure. Treatment of α-synuclein with tetranitromethane produces a protein with tyrosine residues modified to 3-nitrotyrosines (Yamin et al., 2003). Nitration of a tyrosine residue significantly decreases its pKa value allowing the phenolic hydroxyl to become ionized at physiological pH. Presence of additional negative charges can disturb formation of stable structures such as amyloid fibrils. Thus, the modified protein can be converted to oligomers. These oligomers (called nitrated oligomers here) have been previously shown to be β-sheet rich (Uversky et al., 2005) and we confirmed that. Far UV CD spectrum of these oligomers (Fig. 1A) indicates a structure containing a mixture of β-sheets and random coils. Estimation of the secondary structure from CD spectra of nitrated oligomers with CDPro software showed a mixture of β-sheets (~30%–40%), turns (about 10%–15%) and disordered regions (the rest). EM images show that the aggregates have the appearance of protofibrils (Supplemental Fig. 2A). Since both previous studies and their appearance indicated that these oligomers have relatively high molecular weight, we fractionated them by filtration via a 30 kDa MWCO filter. CD spectra showed that the retentate is composed primarily of β-sheets (Fig. 1A) but is structurally different from α-synuclein fibrils. Binding of thioflavin T to these oligomers was tested with fluorescence spectroscopy. The results (Fig. 1D) show that fluorescence of thioflavin T remained largely unchanged in the presence of nitrated α-synuclein oligomers but greatly increased in the presence of the fibrils. Treatment of α-synuclein with hydrogen peroxide leads to oxidation of methionine residues to methionine sulfoxides and upon longer exposure to methionine sulfones. Due to much higher polarity of the oxidized methionine residues, they disturb the protein structure. Even low levels of methionine oxidation interfere with fibril formation (Breydo et al., 2005) while high levels of oxidation can in some cases prevent fibril formation and lead to disaggregation of preformed fibrils (Binger et al., 2008). The protocol we used (see Materials and Methods) has been previously shown to lead to relatively high level of oxidation, and aggregation of α-synuclein modified in this manner was shown to lead to primarily disordered oligomers (called oxidized oligomers here) (Uversky et al., 2002). Far UV CD spectra confirmed that oxidized oligomers are primarily disordered (Fig. 1A). EM imaging showed that the aggregates were spherical oligomers, 5–10 nm in diameter (Supplemental Fig. 2B). After fractionation of the oligomers with 30 kDa MWCO filter, retentate remained primarily disordered (Fig. 1A). Amyloid fibrils of α-synuclein were prepared using the previously described method (Breydo et al., 2014). EM imaging (Supplemental Fig. 2C) confirmed their typical appearance as ribbons 10–20 nm in width. Their structure was confirmed by ThT binding (Fig. 1D) and FTIR spectroscopy (Breydo et al., 2014). CD spectroscopy (Fig. 1A) also showed a spectrum with a minimum at 225 nm typical for α-synuclein amyloid fibrils (Jain and Bhat, 2014).

Determination of structural preferences of the 5G4 antibody We examined binding of 5G4 antibody to various preparations of α-synuclein by ELISA and Western blot. First, we added the 5G4 antibody (15 ng/well) to a range of concentrations (10–600 ng/well) of several antigens and detected the antibody–antigen binding with direct ELISA (Fig. 1B). We found that this antibody preferentially binds to the oligomers prepared from nitrated α-synuclein while its binding to other α-synuclein aggregates or to the monomeric protein was much weaker. We then filtered the oligomers through the 30 kDa MWCO filter to separate them from the remaining monomer. When we used only the oligomeric fractions retained after filtration, the results were similar (Fig. 1C).

80

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

Fig. 1. 5G4 antibody preferentially binds to β-sheet rich α-synuclein oligomers. A: CD spectra of α-synuclein fibrils and oligomers before and after filtration through 30 kDa MWCO filter. B, C: ELISA data of reactivity of 5G4 antibody with α-synuclein monomer, fibrils and oligomers. Data represent mean ± SEM of three replicates. D: 5G4-positive α-synuclein oligomers do not bind thioflavin T. Fluorescence of 3 μM thiolavin T (expressed as a percentage of its fluorescence in the absence of proteins) in the presence of 0–1.5 μM α-synuclein fibrils or oligomers. E: Western blot data of reactivity of 5G4 antibody with α-synuclein monomer, fibrils and oligomers. Lane 1: monomers, lane 2: nitrated oligomers, lanes 3 and 4: oxidized oligomers, lane 5: fibrils. F: silver stained SDS-PAGE α-synuclein monomers, fibrils and oligomers (same samples as in Fig. 1D). Lane 1: monomers, lane 2: nitrated oligomers, lanes 3 and 4: oxidized oligomers, lane 5: fibrils.

Western blot (Fig. 1E) showed that the 5G4 antibody strongly binds to the high molecular weight fraction of nitrated oligomers. This antibody also showed weak binding to fibrils and to high molecular weight impurities present in the monomeric protein. No binding to oxidized oligomers was observed. From these results we can conclude that 5G4 antibody has strong selectivity for β-sheet rich α-synuclein oligomers. Silver staining of the same gel (Fig. 1F) confirmed the presence of high molecular weight aggregates in both fibril and nitrated oligomer samples. However, only nitrated oligomers were strongly stained with the 5G4 antibody (Fig. 1F).

Characterization of α-synuclein preparations using anti-α-synuclein antibodies Immunogold labeling of the α-synuclein preparations demonstrated that only 42αSyn and KM51 reacted with the thioflavine-negative monomer form (Fig. 2A, B, D, F, H). The density of gold particles reacting in the monomer preparation was significantly higher when 42αSyn and KM51 antibodies were used as compared to the no-first antibody control, and 5G4 and phospho-synuclein antibodies (p b 0.001). In addition 42αSyn labeled more than KM51. In contrast, the thioflavine-positive fibrils were

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

81

Fig. 2. Affinity of 42αSyn, KM51, pSyn and 5G4 α-synuclein antibodies against α-synuclein monomer and oligomer samples. A: Statistical demonstration of antibody affinity against monomer α-synuclein (antibody control: No 1st Ab). B, D, F, H: Representative parts of the sample contained only monomer α-synuclein after 18 nm immunogold labeling. C, E, G, I: Representative parts of the samples contained oligomer α-synuclein after 18 nm immunogold labeling. Scale bar is 100 nm for all images.

labeled by 42αSyn, KM51 and 5G4 (Fig. 2C, E, G, I). Antibody 5G4 showed the most specific and most intensive binding to these fibrils (Fig. 2H and I). Light microscopic morphology of immunodeposits Using antibody 5G4 we observed the following types of immunoreactivities: 1) Neuronal immunoreactivity comprise classical brainstem and cortical types of Lewy bodies and fine dot-like granular immunoreactivity

in neuronal cytoplasm, furthermore, thick neurites (Lewy-neurites) and axonal spheroids (Fig. 3A, B). 2) Thin neurites and dots in the neuropil (up to 4 μm): These always precede the appearance of cortical Lewy bodies in the neocortex and are always more abundant then Lewy bodies (Fig. 3C). In cases with rapidly progressive form of DLB (referred to frequently as Creutzfeldt–Jakob disease due to the similarity of symptoms) this type of immunoreactivity was extremely prominent (see Supplemental Fig. 1B). In contrast to cortical Lewy bodies they are observed

82

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

also in upper cortical layers (1–3) as the earliest feature of neocortical involvement. Dots are abundant also in the striatum (Fig. 3D) in cases with stages above 3 according to Braak et al. (Braak et al., 2003). 3) Glial immunoreactivities: i) Immunoreactivity of astrocytes was seen mainly in the entorhinal and temporal cortices, amygdala and striatum, and less in the hippocampus CA subregions (mostly CA1). In the cortex both the upper and lower layers showed these astrocytes but not layer 4 (Fig. 3E–G). Two morphologies were distinguished: star-like (Fig. 3H) was seen mainly the cortex and amygdala; this showed co-localization with GFAP astroglial marker (Fig. 3I) but not with ubiquitin or p62 (data not shown), moreover the majority were not visible with Gallyas silver staining. The other type of astroglial immunolabeling is characterized by the accumulation of granular immunoreactivity in the cytoplasm of astrocytes: this was seen particularly in the striatum (Fig. 3J), but also in the cortex. We observed “maturation” of these deposits (Fig. 3K–M); tiny dots first appeared around lipofuscin pigments, followed by growth and confluescence of these, and finally only a few showed the star-like appearance (Braak et al., 2003) seen in the cortex. It must be emphasized that lipofuscin-related 5G4 immunoreactivity can be seen only in regions where α-synuclein pathology is present and otherwise in all further anatomical regions lipofuscin remains completely negative. Glial α-synuclein immunoreactivity was observed only in Braak stages 4 and above (Chi2 test p = 0.002). In the neocortex this glial pattern is seen in the upper and lower layers. iii) Gallyas positive and p62 immunoreactive coiled-body like structures in the oligodendroglial cells (Wakabayashi et al., 2000) were also immunoreactive for α-synuclein (Fig. 3N). 4) Other types of immunoreactivities include deposits around vessels and in exra-CNS parts of cranial nerves, and those associated with the ependymal layer. i) Vessel-wall related immunoreactivity in 8 cases (19.5%) cases. This was mainly related to the small arteries of subcortical structures (basal ganglia and amygdala and pons). It was seen also associated to perivascular macrophages (Fig. 3O, P). There was lack of association with Braak stages of α-synuclein pathology or Braak and Braak stages of tau positive neurofibrillary degeneration (p N 0.1). Instead vessel wall fibrosis or mineralization was seen in these small vessels. ii) In the cranial nerves we observed axonal spheroids, Lewy neuritelike structures, as well as small dots around the myelin sheath. These were observed in all examined olfactory bulbs (4), in three out of four in the optic chiasm (Fig. 3Q), in two out of four cases where a total of six trigeminal nerves were examined (in two cases both sides available) (Fig. 3R), while they were not detectable in the oculomotor (6), trochlear (3), facial (2), vestibulocochlear (2), and hypoglossal (12) nerves. Immunoreactivity was most prominent in the vagus nerve: 33 out of 39 (84.6%) examined were positive. Immunoreactivity following long axons was observed in the intramedullary portion (Fig. 3S, T), while in the extramedullary portion axonal dots and small globular structures were noted in 21 (53.8%), axonal spheroids (like thick Lewy-neurites) in 12 (30.7%) (Fig. 3U–W).

iii) We observed tiny dots, thin neurites in the subependymal area, as well as tiny dots between ependymal cells as well as larger amorphous deposits (Fig. 3X1–4). These were not seen with antiubiquitin/p62 immunostaining (data not shown). These were more frequently seen in the ependymal layer of the aqueduct, followed by the 3rd ventricle, and less was seen in the lateral ventricle (including inferior horn). Indeed, aqueductal α-synuclein immunoreactivity showed a strong correlation with Braak stage of LRP (Spearman correlation R = 0.55, p = 0.001). The stage of neurofibrillary degeneration did not influence the ependymarelated α-synuclein immunopositivity in any regions examined (p N 0.3). On the light microscopic level unequivocal immunoreactivity was not seen in the choroid plexus (Fig. 3Y). Correlation of immunostaining patterns with tissue lesioning First we compared the scores of different α-synuclein immunoreactivities pooled from all examined reagions (Table 2). Spearman correlation test showed significant correlation of thin neurites, neuronal cytoplasmic, and neuropil dots with each other. Presence of astroglial α-synuclein immunoreactivity correlated only with dots and thin neurites but not with the score of thick neurites and neuronal immunoreactivity. Thin neurites, dots and neuronal cytoplasmic deposits (Lewy bodies and cytoplasmic granular immunoreactivity together), but not the thick neurite and astroglial α-synuclein immunoreactivity score correlated with Braak and Braak stages of neurofibrillary degeneration (tau pathology) and Braak stages of α-synuclein pathology (p b 0.02 for all correlations). Linear regression model showed significant interaction (p b 0.01) of Braak and Braak stages of neurofibrillary degeneration and Braak stages of α-synuclein pathology. Thus the score of thin neuritic, dot-like and neuronal cytoplasmic α-synuclein immunoreactivity was higher when the tau pathology (neurofibrillary degeneration) was in a more advanced stage. The score of the extent of microglial activation correlated with the amount of thin neurites and dots, neuronal cytoplasmic and glia-related α-synuclein deposits but not with thick neurites (see Supplemental Fig. 1. and Table 2) in the same region examined. GFAP immunoreactivity showed astroglial cells in the gray matter; although these did not exhibit hypertrophy (Halliday and Stevens, 2011); GFAP immunoreactivity correlated with reactive microgliosis (R = 0.39, p b 0.001) and the amount of α-synuclein immunopositive dots and glial α-synuclein immunoreactivity score (Table 2). In regional comparisons the GFAP immunoreactivity score correlated with the cortical but not striatal glia-related deposits (data not shown). Immunogold ultrastructural observations Both pre-embedding and post-embedding labeling in the substantia nigra confirmed that 5G4 recognizes classical Lewy neurites (online Supplementary Fig. 3A–C). Further post-embedding investigation of the amygdala revealed neuronal and glial pre-fibrillary structures corresponding to the light microscopic finding of intracytoplasmic dot-like and glial immunoreactivites (Fig. 4). In addition to the findings of fibrillized structures as seen in light microscopy, our ultrastructural observations allowed us to expand the evaluation of the spectrum of

Fig. 3. Light microscopic immunostaining patterns in representative cases of α-synuclein using antibody 5G4.A: Fine dot-like granular immunoreactivity in neuronal cytoplasm and classical brainstem Lewy body and Lewy neurites in the substantia nigra. B: Cortical Lewy body neurites and dots in the entorhinal cortex. Note that a neurofibrillary tangle indicated by an arrow remains unstained. C: Abundant thin neurites and dots in the neuropil in the temporal neocortex. D: Dots are abundant also in the striatum. E: Immunoreactivity of astrocytes was seen both the upper (enlarged in F) and lower layers (enlarged in g). H: Star-like astrocytes in the amygdala, showing co-localisation with GFAP astroglial marker (I). J: Accumulation of granular immunoreactivity in the cytoplasm of astrocytes show “maturation” (K–M). N: Coiled-body like glial structures are detectable by Gallyas silver staining (right lower inset) and immunostaining for p62 (right upper inset). O: Prominent perivascular α-synuclein immunoreactivity; dot-like profiles are seen in the perivascular macrophages as enlarged in P. Q: Lewy neurites in the optic chiasm. R: Tiny dot-like deposits in the trigeminal nerve. Long axonal staining was observed in the intramedullary portion (S) of the vagus nerve (enlarged in T) followed by immunoreactivity in the extramedullary portion (U). Globular structures and axonal spheroids in the extramedullary portion of the vagus nerve (Klüver–Barrera staining in V and α-synuclein immunostaining in W on adjacent section). X1–4: Tiny dots, thin neurites in the subependymal area, as well as tiny dots between ependymal cells as well as larger amorphous deposits. Y: Lack of unequivocal immunoreactivity in the choroid plexus. Bar in “A” represents 25 μm for A, B, D, P, Q, R, T, and u; 10 μm for I, K, L, M and N; 50 μm for c, F, G, H, J, O, V, W, and X1–4; 100 μm for S and Y; and 150 μm for E.

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

83

84

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

Table 2 Summary of the Spearmann correlation test (R: correlation coefficient) for the comparison of the scores of α-synuclein immunoreactivities, activated microglia and astroglial GFAP immunopositivity. Thick Nt αSyn Thick Nt αSyn Thin Nt αSyn Neuronal αSyn Dots αSyn Glial αSyn

R p R p R p R p R p

1.00 0.50 b0.001 0.52 b0.001 0.19 0.08 −0.16 0.14

Thin Nt αSyn

Neuronal αSyn

Dots αSyn

Glial αSyn

Microglia CR3/43

Astroglia GFAP

0.50 b0.001 1.00

0.52 b0.001 0.68 b0.001 1.00

0.19 0.08 0.66 b0.001 0.48 b0.001 1.00

−0.16 0.14 0.34 0.002 0.14 0.18 0.48 b0.001 1.00

0.10 0.35 0.41 b0.001 0.29 0.007 0.53 b0.001 0.53 b0.001

−0.15 0.18 0.17 0.13 −0.03 0.79 0.29 0.009 0.51 b0.001

0.68 b0.001 0.66 b0.001 0.34 0.002

0.48 b0.001 0.14 0.18

subcellular locations of α-synuclein revealing a complex intracellular trafficking pathway. Pre-embedding immunogold labeling shows vesicles of various size harbouring α-synuclein deposits (Fig. 5A, B). The post-embedding double immunogold labeling reveals that diseaseassociated-α-synuclein is internalized by a standard clathrin-mediated endocytotic pathway in neurons (Fig. 5C–H). This pathway is characterized by immunoreactivity for endocytotic markers. From the clathrincoated pit, the α-synuclein oligomers go to the early endocytotic subcompartiment characterized by early endosomal antibody (EEA) (Fig. 5E, F). From the early endocytotic subcompartiment there are two possibilities for the processing of α-synuclein; either recycling back to the cell surface of neurons or progression to the late endosomal structures, including CD63 immunoreactive multivesicular bodies (MVB) (Fig. 5G). From here it is transported to endolysomes (”maturation”) (Fig. 5H). Corresponding to light microscopic observations, lipofuscinrelated 5G4 immunoreactivity can be observed in neurons and in astrocytes (Figs. 6, 7). Lipofuscin-granules are parts of the endosomal–lysosomal subcompartiments in the perikaryon of these cells. Between the lipofuscin granules, small pre-fibrillary structures can be seen, furthermore, there are single gold particles in lipofuscin (Figs. 6C, D and 7E, F). Rab7 and CD63 positivity confirms the endosomal–lysosomal origin of lipofuscin (Figs. 6C, D and 7E, F and 8). The electron-lucent part of the lipofuscin granules suggests a degenerating mitochondrial origin; at high magnification the intra-mitochondrial crista membranes are clearly identifiable

0.48 b0.001

(Fig. 7E, F and G). The larger fibrillar aggregates are surrounded in a ring formation by relatively preserved mitochondria (Fig. 8A, B). Moreover, α-synuclein gold particles are seen in relation to mitochondrial membranes (Fig. 8C–E). In the subependymal regions of the ventricles we observe similar intracellular (neuronal process) distribution of the disease-associated α-synuclein as in the putamen. In addition, corresponding to the light microscopic observations (Fig. 3X), between the ependymal cells 5G4 immunoreactive α-synuclein clusters adjacent to gap junction-like structures, furthermore, they can be seen in the narrow extracellular space (Fig. 9), while in the ependymal cells themselves, we do not observe any immunoreactivity. Discussion The major implication of our study is that we document in the human brain a similar internalization process of disease-associated αsynuclein as seen for disease-associated prion protein in prion diseases, a model of transmissible neurodegenerative disorders. Indeed, this was previously shown already in animal brains of prion disease (Jeffrey, 2013; Laszlo et al., 1992). A prion-like spreading was suggested already based on the observation of the hierarchical deposition of α-synuclein in the human brain in PD (Braak et al., 2003) and of Lewy body pathology in embryonic dopamine neurons transplanted into the putamen of human PD patients (Li et al., 2008). Furthermore, there are in vitro and

Fig. 4. Intracellular deposits of pre-fibrillary structures consists of disease-associated α-synuclein. A: Intensely labelled pre-fibrillary structures consist of pathologic α-synuclein in a neuron in the amygdala (enlarged in B). C: 5G4-positive fibrillary granule in a glial cell (most likely oligodendrocyte) (enlarged in D). Scale bars: A, C: 5 μm, B, D: 500 nm.

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

85

Fig. 5. Endocytotic pathway for α-synuclein demonstrated by antibody 5G4. A: Pre-embedding ultrastructural localization of disease-associated α-synuclein by 5G4 antibody using nanogold immunolabeling followed silver-intensification in a substantia nigra neuron. Note the very intense silver particle accumulation: the black precipitation (5G4) fills almost the entire cell. 5G4 positivity can be also seen in vesicles just beneath the plasma membrane (black arrowheads). B (magnified part of A): Silver particles in endocytotic vesicle. C–D: Postembedding 5G4 labeling in substantia nigra neurons (black arrowheads). C: Significant immunopositivity in clathrin coated pit. D: Gold particles in early endocytotic vesicle. E, F: Double labeling with 5G4 (black arrowhead, 20 nm gold) and anti-early endosome (EEA1) marker (white arrowhead, 10 nm gold) antibodies. G, H: Double labeling with 5G4 (black arrowhead, 20 nm gold) and late endosomal–lysosomal marker, CD63 (white arrowhead, 10 nm gold) antibodies in multivesicular body (G) and lipofuscin (H; indicated as Lf). Scale bars: A: 2 μm, B: 150 nm; C–H: 200 nm.

in vivo (experimental animal models) evidences for a seeding effect and cell-to-cell transmission of α-synuclein (for review see (Visanji et al., 2013). While post mortem examination of human brains cannot provide details on the dynamics of this pathway, however, documentation of the association of disease-associated α-synuclein to key subcellular compartments of internalization strongly complements the dynamic experimental observations. In addition, 1) we demonstrate that interaction of mitochondria and disease-associated α-synuclein plays a key role in the molecular cytopathogenesis of disorders characterized by Lewy body pathology; 2) we show novel localizations of disease-associated α-synuclein in the brain including perivascular macrophages, ependyma and different cranial nerves; 3) we provide further support to the concept that neuropil dots and threads are more likely to evoke tissue reactions than Lewy bodies; and 4) we confirm that astrocytic α-synuclein

immunoreactivity is an important component of the protein pathology in disease with Lewy bodies. Prior to the microscopic investigations we further examined the reactivity of our 5G4 antibody. Our previous study showed that 5G4 antibody strongly binds to the high molecular weight fraction of nitrated oligomers, while no binding to oxidized oligomers was observed. From these results we conclude that 5G4 antibody has strong selectivity for β-sheet rich α-synuclein oligomers, and therefore we postulate that it detects the disease-associated forms and is suitable to map the subcellular localization of α-synuclein in the human brain. Nitration of pathological inclusions has been shown in α-synucleinopathies (Duda et al., 2000). Later antibodies recognizing native recombinant and nitrated/oxidized recombinant α-synuclein have been developed (Duda et al., 2002). These antibodies showed an abundance of previously underappreciated α-synuclein pathology particularly

86

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

Fig. 6. Double immunogold labeling demonstrates pathologic accumulation of 5G4 positive α-synuclein in the lipofuscin region of a neuron in the putamen of a Parkinson's disease brain. A: Lipofuscin region (Lf) in a neuron at low magnification after double immunogold labeling with 5G4 and CD63 antibodies (B: magnified part of A). C–D (magnified parts of b): 20 nm gold particles indicate 5G4 (black arrowheads) and 10 nm gold particles indicate CD63 (white arrowheads) immunopositivity. Scale bars: A: 2 μm, b: 1 μm; C–D: 200 nm.

in the striatum of disorders with Lewy bodies (Duda et al., 2002). One of these antibodies (Syn303; detecting oxidized α-synuclein) was used in an immunohistochemical study and it also showed diffuse staining of the neuropil, which was interpreted to be normal synaptic terminal staining (Beach et al., 2008); this synaptic staining cannot be seen when using antibody 5G4 (Kovacs et al., 2012b). One major finding of our study is that we were able to document the pathway of the processing of disease-associated α-synuclein within the neurons and astroglial cells (summarized in Fig. 10). Accumulating evidence suggests that both monomeric and oligomeric α-synuclein can be secreted into the extracellular environment in vitro and in vivo (Lee, 2008; Lee et al., 2010). Intracellular α-synuclein exocytosed into the extracellular space could be internalized and directly packaged into the endosomal vesicles (Hasegawa et al., 2011) (Fig. 10. path 1; see also Fig. 5A–F). However, it is possible that α-synuclein might be secreted through different secretory pathways depending on the size of the aggregates or physiological conditions. It might be destined for lysosomal degradation (Fig. 10. path 3) or be processed through the

MVB-exosome pathway (Alvarez-Erviti et al., 2011; Emmanouilidou et al., 2010; Hasegawa et al., 2011) (Fig. 10. path 5; see also Fig. 5G), or in case of impaired MVB-exosome biogenesis, it can be introduced into the extracellular milieu through the Rab11a-dependent recycling endosomal pathway (Hasegawa et al., 2011; Liu et al., 2009) (Fig. 10. path 2). Our observation of α-synuclein in relation to lipofuscin and mitochondria highlights an important aspect of the intracellular processing. It has been widely hypothesized that membrane interactions play a central role in the function and toxicity of α-synuclein (Auluck et al., 2010; Volles and Lansbury, 2007). Moreover, it has been demonstrated that α-synuclein selectively and preferentially bounds to the mitochondria prepared from the rat brain (Nakamura et al., 2008). Our findings corroborate this phenomenon in human brain (Fig. 10. path 6; see also Figs. 7F and 8C–E). The notion that PD and mitochondrial dysfunction are connected has been around for over two decades (Schapira and Jenner, 2011; Schapira et al., 1990). In addition, several genes encoding mitochondrial proteins are linked to inherited forms of PD (Valente et al., 2012). Two possible functions have been attributed to mitochondrial α-synuclein, i.e., regulation of mitochondrial dynamics (Xie and Chung, 2012) and maintenance of mitochondrial calcium homeostasis (Cali et al., 2012). Once incorporated into the mitochondria, α-synuclein leads to fragmented, aggregated mitochondria (Kamp et al., 2010) (see also Fig. 10. path 7). Misfolded α-synuclein accumulates within both mitochondrial membranes (Devi et al., 2008), leading to disruption of ATP synthesis and damage of the mitochondrial membrane potential (Kamp et al., 2010). Cytochrome c, a well-known electron transfer enzyme, and mediator of apoptotic cell death may be involved in the oxidative stress-induced aggregation of α-synuclein (Hashimoto et al., 1999). Cytochrome c release is downstream or simultaneous step with the fragmentation of mitochondria (Suen et al., 2008). According to this finding we identified disease-associated α-synuclein fibrillar aggregates surrounded by relatively preserved mitochondria (Fig. 8A–B). Although this fragmentation appears to be universally associated with apoptosis, excessive mitochondrial fission can occur in the absence of apoptosis (upon exposure to uncoupling agents that disrupt inner mitochondrial membrane electrochemical potential or viral infection) (Suen et al., 2008). We find no direct morphological evidence supporting mitophagy, the selective macroautophagy mechanism eliminating damaged mitochondria (Narendra et al., 2009; Tanaka, 2010; Youle and Narendra, 2011). However, Rab7 immunopositive particles in close proximity of mitochondria (Fig. 8E) suggest a cross talk or interaction between the mitochondrial and endosomal–lysosomal compartments. Lipofuscin-granules (“aging pigments“) are components of the ELS of neurons and astrocytes. Interestingly, the major source of lipofuscin is generally thought to result from incomplete degradation of damaged mitochondria (Gray and Woulfe, 2005) (Fig. 10. path 8; see also Fig 7D–G). Indeed, proteome analysis of purified lipofuscin fraction from human temporal cortex and hippocampus samples identified a common set of 49 (among N 200 total) proteins that are mainly derived from mitochondria, cytoskeleton, and cell membrane (Ottis et al., 2012). Under pathological conditions, the contents of lysosomes might leak into the cytoplasm (Laszlo et al., 1992; Nakamura et al., 1989). In case of α-synuclein, Hashimoto et al. raised the possibility that synuclein-leaking might be attributed to Lewy body formation (Hashimoto et al., 2004). Similarly to scrapie prion protein and amyloid-β, α-synuclein oligomers that escaped from the lysosomal system might initiate the filament formation in the cytosol in the neighborhood of the lipofuscin granules (see Fig. 10. path 9; see also Figs. 5H, and 6). Recent data have suggested that extracellular αsynuclein aggregates can induce intracellular synuclein pathologies supporting the hypothesis that α-synuclein pathology can spread via a "prion-like" self-templating mechanism; indeed, Sacino et al. provided an evidence that distinct PD-linked mutant α-synucleins generate structurally distinct aggregates in both neurons and astrocytes of mixed primary cultures (Sacino et al., 2013) (see also the blue dots in the neuron in Fig. 10).

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

87

Fig. 7. Double immunogold labeling demonstrating pathologic accumulation of 5G4 positive α-synuclein in the lipofuscin region of astrocytes in the putamen of a PD brain. Panel A demonstrates an astrocyte at low magnification with nucleus (NuASTR) and lipofuscin region (Lf) after 5G4 and GFAP double labeling. B (magnified parts of A): 5G4 positive fibrils decorated with 18 nm gold particles. C: GFAP fibrils labeled with 12 nm gold particles. D–F: Double immunogold labeling with 5G4 (20 nm gold) and CD63 (10 nm gold particles) antibodies. D: Astrocyte with lipofuscin region (Lf). E (magnified part of D): Lipofuscin region. F (magnified region of E): Lipofuscin region labeled by 5G4 (black arrowheads) and CD63 (white arrowhead) antibodies. Within a large lipofuscin granule the dotted line borders an injured mitochondrion (Mt) fused with dense residual body. Double-head arrows point mitochondrial crista membrane remnants. G: Lipofuscin particle in a section made by conventional electronmicroscopic preparation (Durcupan embedding). Note the swollen degenerating mitochondrion (Mt) as a component of the lipofuscin granule (matured lysosome). Scale bars: A, D: 5 μm, B, C, E, F insert: 500 nm, F: 200 nm.

In our light microscopic study, we were able to document novel localizations of disease-associated α-synuclein. Presence of α-synuclein immunoreactivity related to vessels was associated mainly to thickened vessels. This raises the possibility that in hypertensive small vessel disease associated with α-synucleinopathy, activation of perivascular macrophages can be involved in transport and clearance of the disease-associated protein, which has implications for biomarker research. Interestingly, vessel-associated distribution reminiscent of the present observation has been shown for disease-associated prion

protein in Creutzfeldt–Jakob disease and was even suggested as an alternative way for spread of disease (Koperek et al., 2002). In addition, we found prominent immunoreactivity in the ependyma mostly surrounding the aqueduct but also in the ventricle walls both by immunohistochemistry (Fig. 3X) and by immunogold electron microscopy (Fig. 9). This correlated with the Braak stages. In contrast to others, using different anti-α-synuclein antibodies, we did not find immunoreactivity in the choroid plexus (Mollenhauer et al., 2012). This calls for caution when interpreting α-synuclein immunopositivity in epithelial

88

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

Fig. 8. Fibrillar accumulation of disease-associated α-synuclein surrounded by mitochondria in the dendritic process of a neuron in the putamen region of a PD brain. A: Fibrillar accumulation surrounded by mitochondria (white arrows) after post-embedding labeling with 5G4 antibody. Insert of A: A similar neuronal process after pre-embedding 5G4 immunolabeling with nanogold intensification (white arrows show mitochondria). B (magnified part of A): Mitochondria (white arrows) on the boundary of the gold particle decorated immunpositive fibrillar accumulation. C–E: Double labeling with 5G4 (black arrowheads) and Rab7 (white arrowheads) antibodies. C: Gold particles on the mitochondria (white arrows) and between them (the black arrows indicate a pre-fibrillary profile). D, E (magnified parts of C): 20 nm (5G4, black arrowheads) and 10 nm (Rab7, white arrowheads) gold particles on mitochondria and within small vesicles in their neighborhood. Scale bars: A: 1 μm, a insert, B, C: 500 nm; D, E: 200 nm.

cells, since it might represent cross-reaction or they may detect the physiological form of α-synuclein. Our light microscopic finding supports the idea that disease-associated α-synuclein might reach the cerebrospinal fluid (CSF). Ependymal cells have well-developed adherens junctions, non-continuous tight junctions (except at some of the circumventricular organs (Del Bigio, 1995; Rodriguez and Bouchaud, 1996)) and gap-junctions (Del Bigio, 2010). This connection system does not form a barrier between the CSF and the neuropil. The ependymal layer lies on a basement membrane and underlying network of astroglial processes, and is covered by some supraependymal fibers originating from the subependymal plexus; a few dopaminergic axons were seen to run between the ependymal cells (Michaloudi and Papadopoulos, 1996). Sung et al. presented that α-synuclein can modulate (reduce) gap junctional intercellular communication in SH-SY5Y cells (Sung et al., 2007). We observed gap junctions characteristically labeled with the disease-specific 5G4 antibody in the subependymal layer (online Supplementary Fig. 4); most likely these belong to the

subependymal plexus originating from noradrenergic and dopaminergic neurons. This observation suggests that α-synuclein in human CSF is principally derived from neurons of the central nervous system (Mollenhauer et al., 2012). Moreover, detection of α-synuclein related to gap junctions is reminiscent of the theory of spreading of the protein via nanotubes (Visanji et al., 2013). Finally, the presence of α-synuclein deposition in the subependymal plexus may influence the physiological functions of the ependyma potentially leading to dysfunction of the CSF circulation (i.e. leading to ventricle enlargement). Importantly, we frequently detected the presence of disease-associated α-synuclein in the extramedullary portion of the vagus nerve (Del Tredici et al., 2010). This provides further evidence for the concept that α-synuclein pathology enters the central nervous system from the periphery in most of the cases with Lewy body pathology (Braak et al., 2003; Del Tredici et al., 2010). In addition to the well-known involvement of the olfactory bulb, we found immunoreactivity in the trigeminal nerve and also in the optic chiasm. For the trigeminal nerve

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

89

Fig. 9. Distribution of 5G4 positive α-synuclein in the ependymal layer of the lateral ventricle. A: Ependymal layer of the lateral ventricle (V) after 5G4 immunogold labeling at low magnification (NuEP: ependymal cell nucleus). Insert of a: Magnified part of the surface with cilia. B: Pathway decorated by 12 nm gold particles and cell coupling structures leads from the subependymal plexus towards the ventricle (V). C, D (magnified parts of B): Immunogold 5G4 labeling (black arrowheads) between the ependymal cells (white arrows in D show desmosomes). Scale bars: A: 2 μm, B–D: 200 nm.

the present study does not provide a clue which fibers (motor or sensory) are involved. α-Synuclein is expressed in the retina and optic nerve in humans, moreover, in transgenic mice overexpressing α-synuclein, pathological α-synuclein depositions can be seen in the optic nerve (Surguchov et al., 2001); we confirm this in diseased human tissue. This has implications both as a potential for examination of the retina or might have relevance for visual symptoms of patients with PD or DLB. Our study provided further support that Lewy bodies themselves are not the major substrates of tissue reactions in PD and DLB. Indeed, it has been shown that the amount of nigral Lewy bodies alone does not correlate with the striatal dopaminergic deficit, whereas the total burden of α-synuclein pathology, including neuritic, does (Kovacs et al., 2008). It is conceivable that the widespread deposition of α-synuclein in neuronal processes causes dysfunction leading to the clinical symptoms. In addition, we confirm the observations on astrocytic α-synuclein pathology reported by Braak and co-workers using different antibodies (Braak et al., 2007). Together with the observation of endocytosis of α-synuclein by astrocytes, these observations underpin the role of astrocytes in PD and DLB (for review see Halliday and Stevens, 2011). It must be noted that not all cases with Braak stages above IV showed this

type of pathology; most likely a threshold of local (neuron-derived) α-synuclein pathology needs to be reached before astrocytes start to accumulate α-synuclein as shown also in our study by their correlation with α-synuclein dots and thin threads in the same region examined. This is paralleled by an increase of GFAP immunoreactivity in astrocytes, even when the typical morphology of reactive astrogliosis cannot be observed in these regions (Halliday and Stevens, 2011). In contrast, we could not find evidence for phagocytic activity of microglial cells, but we observed increased amount of these cells correlating with the α-synuclein immunoreactive dots and thin threads suggesting a reactive response to neuronal damage, which also might contribute to further disease progression by releasing toxic substances (Halliday and Stevens, 2011). As reported recently in a study on Parkinson disease dementia, α-synuclein pathology strongly correlated with Braak and Braak stages of neurofibrillary degeneration (tau pathology) (Irwin et al., 2012). However, it should be noted that α-synuclein dots and thin threads appeared in cortical areas without apparent tau pathology; thus there was an anatomical discrepancy between the depositions of the two neurodegeneration-related proteins. This could suggest that a general

90

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

Fig. 10. Schematic demonstration of spreading pathways of the diseased associated α-synuclein in neurons and astrocytes as supported by the present study. Path 1: Endocytosis at the synapse and transport to the early endosome (EE); Path 2: multivesicular (MVB) body formation; Path 3: late endosome (LE) formation and fusion with the lysosome (L); Path 4: Exosome (Ex) docking and fusion with the plasma membrane; Path 5: Cytosolic disease-associated α-synuclein spreads via the mitochondrial network; Path 6: Fibrillar preaggregate- and aggregate-formation in close vicinity of mitochondria present in dendrites; Path 7: Mitochondria binding to α-synuclein oligomers fuse with late endosome and form lipofuscin granules (Lf); Path 8: α-synuclein oligomers present on the lipofuscin granules form a fibrillar network in the cytoplasm. Abbreviations: Ax: axon, Nu: nucleus, red dots: 5G4 positive disease assotiated α-synuclein oligomers, blue dots: physiologic α-synuclein monomer in neuron, red arrow: spreading of α-synuclein oligomers. Note: A small part of an astrocyte (right) in an intimate connection with the neuron. The identical numbers indicate the same cellular process in both neuron and astrocyte.

neurodegenerative effect initiating the Alzheimer`s disease (AD)-related neurofibrillary pathology enhances the process of α-synuclein pathology, which might dissociate anatomically from tau pathology and follow a specific but distinct spreading pathway. Indeed, biochemical alterations of α-synuclein may appear without immunohistochemically detectable protein deposition as exemplified by abnormal nitration of α-synuclein and has been shown in the frontal cortex of Pick`s diseased brains (Dalfo et al., 2006) or in brains of advanced AD without Lewy body pathology that had phosphorylation of α-synuclein (Ser129) in synapticenriched fractions of the frontal cortex (Muntane et al., 2008). Conclusions The 5G4 antibody has strong selectivity for β-sheet rich α-synuclein oligomers, including also in fibrils, and therefore is an excellent tool for morphological localization of disease-specific α-synuclein in the human brain. We provide evidence for the endocytotic uptake of diseaseassociated α-synuclein by neurons and astrocytes in the human brain, which supports the concept of cell-to cell propagation of this protein as found also by experiemental observations (Guo and Lee, 2014). Moreover, the importance of mitochondria and also non-Lewy body type α-synuclein pathologies is highlighted. Our observations support therapeutic concepts that aim to halt the spread of α-synuclein pathology. However, great caution is needed if an anti-α-synuclein antibody is therapeutically used, since those reacting with the physiological form of α-synuclein may cause unexpected complications. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.nbd.2014.05.020. Acknowledgment Supported by European Commission's 7th Framework Programme under GA No 278486, "DEVELAGE". This study was partly supported by the Austrian–Hungarian Action Foundation (No 82öu8; to L.L.,

K.M., G.G.K.), and by the Hungarian Scientific Research Fund (OTKANK78012; to L.L., K.M., G.P., V.K.). Ingolf Lachmann is employed by diagnostic company AJ Roboscreen. Antibody 5G4 is patented by AJ Roboscreen, inventors are Gabor G. Kovacs, Ingolf Lachmann, Awad A. Osman and Uta Wagner. References Alafuzoff, I., Arzberger, T., Al-Sarraj, S., Bodi, I., Bogdanovic, N., Braak, H., Bugiani, O., Del-Tredici, K., Ferrer, I., Gelpi, E., Giaccone, G., Graeber, M.B., Ince, P., Kamphorst, W., King, A., Korkolopoulou, P., Kovacs, G.G., Larionov, S., Meyronet, D., Monoranu, C., Parchi, P., Patsouris, E., Roggendorf, W., Seilhean, D., Tagliavini, F., Stadelmann, C., Streichenberger, N., Thal, D.R., Wharton, S.B., Kretzschmar, H., 2008a. Staging of neurofibrillary pathology in Alzheimer's disease: A study of the BrainNet Europe Consortium. Brain Pathol. 18, 484–496. Alafuzoff, I., Parkkinen, L., Al-Sarraj, S., Arzberger, T., Bell, J., Bodi, I., Bogdanovic, N., Budka, H., Ferrer, I., Gelpi, E., Gentleman, S., Giaccone, G., Kamphorst, W., King, A., Korkolopoulou, P., Kovacs, G.G., Larionov, S., Meyronet, D., Monoranu, C., Morris, J., Parchi, P., Patsouris, E., Roggendorf, W., Seilhean, D., Streichenberger, N., Thal, D.R., Kretzschmar, H., 2008b. Assessment of alpha-synuclein pathology: A study of the BrainNet Europe Consortium. J. Neuropathol. Exp. Neurol. 67, 125–143. Alafuzoff, I., Ince, P.G., Arzberger, T., Al-Sarraj, S., Bell, J., Bodi, I., Bogdanovic, N., Bugiani, O., Ferrer, I., Gelpi, E., Gentleman, S., Giaccone, G., Ironside, J.W., Kavantzas, N., King, A., Korkolopoulou, P., Kovacs, G.G., Meyronet, D., Monoranu, C., Parchi, P., Parkkinen, L., Patsouris, E., Roggendorf, W., Rozemuller, A., Stadelmann-Nessler, C., Streichenberger, N., Thal, D.R., Kretzschmar, H., 2009. Staging/typing of Lewy body related alphasynuclein pathology: A study of the BrainNet Europe Consortium. Acta Neuropathol. 117, 635–652. Alvarez-Erviti, L., Seow, Y., Schapira, A.H., Gardiner, C., Sargent, I.L., Wood, M.J., Cooper, J.M., 2011. Lysosomal dysfunction increases exosome-mediated alpha-synuclein release and transmission. Neurobiol. Dis. 42, 360–367. Arnold, J.E., Tipler, C., Laszlo, L., Hope, J., Landon, M., Mayer, R.J., 1995. The abnormal isoform of the prion protein accumulates in late-endosome-like organelles in scrapie-infected mouse brain. J. Pathol. 176, 403–411. Auluck, P.K., Caraveo, G., Lindquist, S., 2010. alpha-Synuclein: Membrane interactions and toxicity in Parkinson's disease. Annu. Rev. Cell Dev. Biol. 26, 211–233. Baron, G.S., Wehrly, K., Dorward, D.W., Chesebro, B., Caughey, B., 2002. Conversion of raft associated prion protein to the protease-resistant state requires insertion of PrP-res (PrP(Sc)) into contiguous membranes. EMBO J. 21, 1031–1040. Beach, T.G., White, C.L., Hamilton, R.L., Duda, J.E., Iwatsubo, T., Dickson, D.W., Leverenz, J.B., Roncaroli, F., Buttini, M., Hladik, C.L., Sue, L.I., Noorigian, J.V., Adler, C.H., 2008. Evaluation of alpha-synuclein immunohistochemical methods used by invited experts. Acta Neuropathol. 116, 277–288.

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92 Binger, K.J., Griffin, M.D., Howlett, G.J., 2008. Methionine oxidation inhibits assembly and promotes disassembly of apolipoprotein C-II amyloid fibrils. Biochemistry 47, 10208–10217. Borchelt, D.R., Taraboulos, A., Prusiner, S.B., 1992. Evidence for synthesis of scrapie prion proteins in the endocytic pathway. J. Biol. Chem. 267, 16188–16199. Braak, H., Del Tredici, K., Rub, U., de Vos, R.A., Jansen Steur, E.N., Braak, E., 2003. Staging of brain pathology related to sporadic Parkinson's disease. Neurobiol. Aging 24, 197–211. Braak, H., Sastre, M., Del Tredici, K., 2007. Development of alpha-synuclein immunoreactive astrocytes in the forebrain parallels stages of intraneuronal pathology in sporadic Parkinson's disease. Acta Neuropathol. 114, 231–241. Breydo, L., Bocharova, O.V., Makarava, N., Salnikov, V.V., Anderson, M., Baskakov, I.V., 2005. Methionine oxidation interferes with conversion of the prion protein into the fibrillar proteinase K-resistant conformation. Biochemistry 44, 15534–15543. Breydo, L., Reddy, K.D., Piai, A., Felli, I.C., Pierattelli, R., Uversky, V.N., 2014. The crowd you're in with: Effects of different types of crowding agents on protein aggregation. Biochim. Biophys. Acta 1844, 346–357. Cali, T., Ottolini, D., Negro, A., Brini, M., 2012. alpha-Synuclein controls mitochondrial calcium homeostasis by enhancing endoplasmic reticulum–mitochondria interactions. J. Biol. Chem. 287, 17914–17929. Caughey, B., Raymond, G.J., Ernst, D., Race, R.E., 1991. N-terminal truncation of the scrapie-associated form of PrP by lysosomal protease(s): Implications regarding the site of conversion of PrP to the protease-resistant state. J. Virol. 65, 6597–6603. Dalfo, E., Martinez, A., Muntane, G., Ferrer, I., 2006. Abnormal alpha-synuclein solubility, aggregation and nitration in the frontal cortex in Pick's disease. Neurosci. Lett. 400, 125–129. Del Bigio, M.R., 1995. The ependyma: A protective barrier between brain and cerebrospinal fluid. Glia 14, 1–13. Del Bigio, M.R., 2010. Ependymal cells: Biology and pathology. Acta Neuropathol. 119, 55–73. Del Tredici, K., Hawkes, C.H., Ghebremedhin, E., Braak, H., 2010. Lewy pathology in the submandibular gland of individuals with incidental Lewy body disease and sporadic Parkinson's disease. Acta Neuropathol. 119, 703–713. Devi, L., Raghavendran, V., Prabhu, B.M., Avadhani, N.G., Anandatheerthavarada, H.K., 2008. Mitochondrial import and accumulation of alpha-synuclein impair complex I in human dopaminergic neuronal cultures and Parkinson disease brain. J. Biol. Chem. 283, 9089–9100. Duda, J.E., Giasson, B.I., Chen, Q., Gur, T.L., Hurtig, H.I., Stern, M.B., Gollomp, S.M., Ischiropoulos, H., Lee, V.M., Trojanowski, J.Q., 2000. Widespread nitration of pathological inclusions in neurodegenerative synucleinopathies. Am. J. Pathol. 157, 1439–1445. Duda, J.E., Giasson, B.I., Mabon, M.E., Lee, V.M., Trojanowski, J.Q., 2002. Novel antibodies to synuclein show abundant striatal pathology in Lewy body diseases. Ann. Neurol. 52, 205–210. Emmanouilidou, E., Melachroinou, K., Roumeliotis, T., Garbis, S.D., Ntzouni, M., Margaritis, L.H., Stefanis, L., Vekrellis, K., 2010. Cell-produced alpha-synuclein is secreted in a calcium-dependent manner by exosomes and impacts neuronal survival. J. Neurosci. 30, 6838–6851. Gray, D.A., Woulfe, J., 2005. Lipofuscin and aging: A matter of toxic waste. Science of aging knowledge environment: SAGE KE, p. re1. Guo, J.L., Lee, V.M., 2014. Cell-to-cell transmission of pathogenic proteins in neurodegenerative diseases. Nat. Med. 20, 130–138. Halliday, G.M., Stevens, C.H., 2011. Glia: Initiators and progressors of pathology in Parkinson's disease. Mov. Disord. 26, 6–17. Hasegawa, T., Konno, M., Baba, T., Sugeno, N., Kikuchi, A., Kobayashi, M., Miura, E., Tanaka, N., Tamai, K., Furukawa, K., Arai, H., Mori, F., Wakabayashi, K., Aoki, M., Itoyama, Y., Takeda, A., 2011. The AAA-ATPase VPS4 regulates extracellular secretion and lysosomal targeting of alpha-synuclein. PLoS One 6, e29460. Hashimoto, M., Takeda, A., Hsu, L.J., Takenouchi, T., Masliah, E., 1999. Role of cytochrome c as a stimulator of alpha-synuclein aggregation in Lewy body disease. J. Biol. Chem. 274, 28849–28852. Hashimoto, M., Kawahara, K., Bar-On, P., Rockenstein, E., Crews, L., Masliah, E., 2004. The role of alpha-synuclein assembly and metabolism in the pathogenesis of Lewy body disease. J. Mol. Neurosci. 24, 343–352. Holmes, B.B., DeVos, S.L., Kfoury, N., Li, M., Jacks, R., Yanamandra, K., Ouidja, M.O., Brodsky, F.M., Marasa, J., Bagchi, D.P., Kotzbauer, P.T., Miller, T.M., Papy-Garcia, D., Diamond, M. I., 2013. Heparan sulfate proteoglycans mediate internalization and propagation of specific proteopathic seeds. Proc. Natl. Acad. Sci. U. S. A. 110, E3138–E3147. Irwin, D.J., White, M.T., Toledo, J.B., Xie, S.X., Robinson, J.L., Van Deerlin, V., Lee, V.M., Leverenz, J.B., Montine, T.J., Duda, J.E., Hurtig, H.I., Trojanowski, J.Q., 2012. Neuropathologic substrates of Parkinson disease dementia. Ann. Neurol. 72, 587–598. Iwai, A., Masliah, E., Yoshimoto, M., Ge, N., Flanagan, L., de Silva, H.A., Kittel, A., Saitoh, T., 1995. The precursor protein of non-A beta component of Alzheimer's disease amyloid is a presynaptic protein of the central nervous system. Neuron 14, 467–475. Jain, M.K., Bhat, R., 2014. Modulation of human alpha-synuclein aggregation by a combined effect of calcium and dopamine. Neurobiol. Dis. 63, 115–128. Jeffrey, M., 2013. Review: Membrane-associated misfolded protein propagation in natural transmissible spongiform encephalopathies (TSEs), synthetic prion diseases and Alzheimer's disease. Neuropathol. Appl. Neurobiol. 39, 196–216. Jellinger, K.A., 2012. Neuropathology of sporadic Parkinson's disease: Evaluation and changes of concepts. Mov. Disord. 27, 8–30. Kamp, F., Exner, N., Lutz, A.K., Wender, N., Hegermann, J., Brunner, B., Nuscher, B., Bartels, T., Giese, A., Beyer, K., Eimer, S., Winklhofer, K.F., Haass, C., 2010. Inhibition of mitochondrial fusion by alpha-synuclein is rescued by PINK1, Parkin and DJ-1. EMBO J. 29, 3571–3589. Koperek, O., Kovacs, G.G., Ritchie, D., Ironside, J.W., Budka, H., Wick, G., 2002. Diseaseassociated prion protein in vessel walls. Am. J. Pathol. 161, 1979–1984. Kovacs, G.G., Milenkovic, I.J., Preusser, M., Budka, H., 2008. Nigral burden of alphasynuclein correlates with striatal dopamine deficit. Mov. Disord. 23, 1608–1612.

91

Kovacs, G.G., Botond, G., Budka, H., 2010. Protein coding of neurodegenerative dementias: The neuropathological basis of biomarker diagnostics. Acta Neuropathol. 119, 389–408. Kovacs, G.G., Molnar, K., Keller, E., Botond, G., Budka, H., Laszlo, L., 2012a. Intraneuronal immunoreactivity for the prion protein distinguishes a subset of E200K genetic from sporadic Creutzfeldt–Jakob disease. J. Neuropathol. Exp. Neurol. 71, 223–232. Kovacs, G.G., Wagner, U., Dumont, B., Pikkarainen, M., Osman, A.A., Streichenberger, N., Leisser, I., Verchere, J., Baron, T., Alafuzoff, I., Budka, H., Perret-Liaudet, A., Lachmann, I., 2012b. An antibody with high reactivity for disease-associated alpha-synuclein reveals extensive brain pathology. Acta Neuropathol. 124, 37–50. Ladiwala, A.R., Dordick, J.S., Tessier, P.M., 2011. Aromatic small molecules remodel toxic soluble oligomers of amyloid beta through three independent pathways. J. Biol. Chem. 286, 3209–3218. Laszlo, L., Doherty, F.J., Osborn, N.U., Mayer, R.J., 1990. Ubiquitinated protein conjugates are specifically enriched in the lysosomal system of fibroblasts. FEBS Lett. 261, 365–368. Laszlo, L., Tuckwell, J., Self, T., Lowe, J., Landon, M., Smith, S., Hawthorne, J.N., Mayer, R.J., 1991. The latent membrane protein-1 in Epstein–Barr virus-transformed lymphoblastoid cells is found with ubiquitin–protein conjugates and heat-shock protein 70 in lysosomes oriented around the microtubule organizing centre. J. Pathol. 164, 203–214. Laszlo, L., Lowe, J., Self, T., Kenward, N., Landon, M., McBride, T., Farquhar, C., McConnell, I., Brown, J., Hope, J., et al., 1992. Lysosomes as key organelles in the pathogenesis of prion encephalopathies. J. Pathol. 166, 333–341. Lee, S.J., 2008. Origins and effects of extracellular alpha-synuclein: Implications in Parkinson's disease. J. Mol. Neurosci. 34, 17–22. Lee, S.J., Desplats, P., Sigurdson, C., Tsigelny, I., Masliah, E., 2010. Cell-to-cell transmission of non-prion protein aggregates. Nat. Rev. Neurol. 6, 702–706. Lesne, S.E., 2013. Breaking the code of amyloid-oligomers. Int. J. Cell Biol. 2013, 950783. Leverenz, J.B., Hamilton, R., Tsuang, D.W., Schantz, A., Vavrek, D., Larson, E.B., Kukull, W.A., Lopez, O., Galasko, D., Masliah, E., Kaye, J., Woltjer, R., Clark, C., Trojanowski, J.Q., Montine, T.J., 2008. Empiric refinement of the pathologic assessment of Lewyrelated pathology in the dementia patient. Brain Pathol. 18, 220–224. Li, J.Y., Englund, E., Holton, J.L., Soulet, D., Hagell, P., Lees, A.J., Lashley, T., Quinn, N.P., Rehncrona, S., Bjorklund, A., Widner, H., Revesz, T., Lindvall, O., Brundin, P., 2008. Lewy bodies in grafted neurons in subjects with Parkinson's disease suggest host-to-graft disease propagation. Nat. Med. 14, 501–503. Liu, J., Zhang, J.P., Shi, M., Quinn, T., Bradner, J., Beyer, R., Chen, S., Zhang, J., 2009. Rab11a and HSP90 regulate recycling of extracellular alpha-synuclein. J. Neurosci. 29, 1480–1485. Ma, B., Nussinov, R., 2010. Polymorphic C-terminal beta-sheet interactions determine the formation of fibril or amyloid beta-derived diffusible ligand-like globulomer for the Alzheimer Abeta42 dodecamer. J. Biol. Chem. 285, 37102–37110. Magalhaes, A.C., Silva, J.A., Lee, K.S., Martins, V.R., Prado, V.F., Ferguson, S.S., Gomez, M.V., Brentani, R.R., Prado, M.A., 2002. Endocytic intermediates involved with the intracellular trafficking of a fluorescent cellular prion protein. J. Biol. Chem. 277, 33311–33318. Michaloudi, H.C., Papadopoulos, G.C., 1996. Catecholaminergic and serotoninergic fibres innervate the ventricular system of the hedgehog CNS. J. Anat. 189 (Pt 2), 273–283. Mollenhauer, B., Trautmann, E., Otte, B., Ng, J., Spreer, A., Lange, P., Sixel-Doring, F., Hakimi, M., Vonsattel, J.P., Nussbaum, R., Trenkwalder, C., Schlossmacher, M.G., 2012. alpha-Synuclein in human cerebrospinal fluid is principally derived from neurons of the central nervous system. J. Neural Transm. 119, 739–746. Muntane, G., Dalfo, E., Martinez, A., Ferrer, I., 2008. Phosphorylation of tau and alpha-synuclein in synaptic-enriched fractions of the frontal cortex in Alzheimer's disease, and in Parkinson's disease and related alpha-synucleinopathies. Neuroscience 152, 913–923. Nakamura, Y., Takeda, M., Suzuki, H., Morita, H., Tada, K., Hariguchi, S., Nishimura, T., 1989. Lysosome instability in aged rat brain. Neurosci. Lett. 97, 215–220. Nakamura, K., Nemani, V.M., Wallender, E.K., Kaehlcke, K., Ott, M., Edwards, R.H., 2008. Optical reporters for the conformation of alpha-synuclein reveal a specific interaction with mitochondria. J. Neurosci. 28, 12305–12317. Narendra, D., Tanaka, A., Suen, D.F., Youle, R.J., 2009. Parkin-induced mitophagy in the pathogenesis of Parkinson disease. Autophagy 5, 706–708. Ottis, P., Koppe, K., Onisko, B., Dynin, I., Arzberger, T., Kretzschmar, H., Requena, J.R., Silva, C.J., Huston, J.P., Korth, C., 2012. Human and rat brain lipofuscin proteome. Proteomics 12, 2445–2454. Rodriguez, P., Bouchaud, C., 1996. The supra-ependymal innervation is not responsible for the repression of tight junctions in the rat cerebral ependyma. Neurobiology 4, 185–201. Sacino, A.N., Thomas, M.A., Ceballos-Diaz, C., Cruz, P.E., Rosario, A.M., Lewis, J., Giasson, B.I., Golde, T.E., 2013. Conformational templating of alpha-synuclein aggregates in neuronal-glial cultures. Mol. Neurodegener. 8, 17. Saito, Y., Kawashima, A., Ruberu, N.N., Fujiwara, H., Koyama, S., Sawabe, M., Arai, T., Nagura, H., Yamanouchi, H., Hasegawa, M., Iwatsubo, T., Murayama, S., 2003. Accumulation of phosphorylated alpha-synuclein in aging human brain. J. Neuropathol. Exp. Neurol. 62, 644–654. Saito, Y., Ruberu, N.N., Sawabe, M., Arai, T., Tanaka, N., Kakuta, Y., Yamanouchi, H., Murayama, S., 2004. Staging of argyrophilic grains: an age-associated tauopathy. J. Neuropathol. Exp. Neurol. 63, 911–918. Schapira, A.H., Jenner, P., 2011. Etiology and pathogenesis of Parkinson's disease. Mov. Disord. 26, 1049–1055. Schapira, A.H., Cooper, J.M., Dexter, D., Clark, J.B., Jenner, P., Marsden, C.D., 1990. Mitochondrial complex I deficiency in Parkinson's disease. J. Neurochem. 54, 823–827. Siegel, S.J., Bieschke, J., Powers, E.T., Kelly, J.W., 2007. The oxidative stress metabolite 4-hydroxynonenal promotes Alzheimer protofibril formation. Biochemistry 46, 1503–1510. Suen, D.F., Norris, K.L., Youle, R.J., 2008. Mitochondrial dynamics and apoptosis. Genes Dev. 22, 1577–1590.

92

G.G. Kovacs et al. / Neurobiology of Disease 69 (2014) 76–92

Sung, J.Y., Lee, H.J., Jeong, E.I., Oh, Y., Park, J., Kang, K.S., Chung, K.C., 2007. Alpha-synuclein overexpression reduces gap junctional intercellular communication in dopaminergic neuroblastoma cells. Neurosci. Lett. 416, 289–293. Surguchov, A., McMahan, B., Masliah, E., Surgucheva, I., 2001. Synucleins in ocular tissues. J. Neurosci. Res. 65, 68–77. Tanaka, A., 2010. Parkin-mediated selective mitochondrial autophagy, mitophagy: Parkin purges damaged organelles from the vital mitochondrial network. FEBS Lett. 584, 1386–1392. Taraboulos, A., Raeber, A.J., Borchelt, D.R., Serban, D., Prusiner, S.B., 1992. Synthesis and trafficking of prion proteins in cultured cells. Mol. Biol. Cell 3, 851–863. Taraboulos, A., Scott, M., Semenov, A., Avrahami, D., Laszlo, L., Prusiner, S.B., 1995. Cholesterol depletion and modification of COOH-terminal targeting sequence of the prion protein inhibit formation of the scrapie isoform. J. Cell Biol. 129, 121–132. Uversky, V.N., 2003. A protein-chameleon: Conformational plasticity of alpha-synuclein, a disordered protein involved in neurodegenerative disorders. J. Biomol. Struct. Dyn. 21, 211–234. Uversky, V.N., Yamin, G., Souillac, P.O., Goers, J., Glaser, C.B., Fink, A.L., 2002. Methionine oxidation inhibits fibrillation of human alpha-synuclein in vitro. FEBS Lett. 517, 239–244. Uversky, V.N., Yamin, G., Munishkina, L.A., Karymov, M.A., Millett, I.S., Doniach, S., Lyubchenko, Y.L., Fink, A.L., 2005. Effects of nitration on the structure and aggregation of alpha-synuclein. Brain Res. Mol. Brain Res. 134, 84–102. Valente, E.M., Arena, G., Torosantucci, L., Gelmetti, V., 2012. Molecular pathways in sporadic PD. Parkinsonism Relat. Disord. 18 (Suppl. 1), S71–S73.

Visanji, N.P., Brooks, P.L., Hazrati, L.N., Lang, A.E., 2013. The prion hypothesis in Parkinson's disease: Braak to the future. Acta Neuropathol. Commun. 1, 2. Volles, M.J., Lansbury Jr., P.T., 2007. Relationships between the sequence of alphasynuclein and its membrane affinity, fibrillization propensity, and yeast toxicity. J. Mol. Biol. 366, 1510–1522. Wakabayashi, K., Hayashi, S., Yoshimoto, M., Kudo, H., Takahashi, H., 2000. NACP/alphasynuclein-positive filamentous inclusions in astrocytes and oligodendrocytes of Parkinson's disease brains. Acta Neuropathol. 99, 14–20. Wu, J.W., Breydo, L., Isas, J.M., Lee, J., Kuznetsov, Y.G., Langen, R., Glabe, C., 2010. Fibrillar oligomers nucleate the oligomerization of monomeric amyloid beta but do not seed fibril formation. J. Biol. Chem. 285, 6071–6079. Xie, W., Chung, K.K., 2012. Alpha-synuclein impairs normal dynamics of mitochondria in cell and animal models of Parkinson's disease. J. Neurochem. 122, 404–414. Yamin, G., Uversky, V.N., Fink, A.L., 2003. Nitration inhibits fibrillation of human alphasynuclein in vitro by formation of soluble oligomers. FEBS Lett. 542, 147–152. Youle, R.J., Narendra, D.P., 2011. Mechanisms of mitophagy. Nat. Rev. Mol. Cell Biol. 12, 9–14. Zhou, W., Long, C., Reaney, S.H., Di Monte, D.A., Fink, A.L., Uversky, V.N., 2010. Methionine oxidation stabilizes non-toxic oligomers of alpha-synuclein through strengthening the auto-inhibitory intra-molecular long-range interactions. Biochim. Biophys. Acta 1802, 322–330.

Intracellular processing of disease-associated α-synuclein in the human brain suggests prion-like cell-to-cell spread.

Dementia with Lewy bodies (DLB), Parkinson's disease (PD) and multiple system atrophy are characterized by the deposition of disease-associated α-synu...
8MB Sizes 0 Downloads 3 Views