key: cord-0275996-8410w4ra
authors: Landeck, Natalie; Strathearn, Katherine E.; Ysselstein, Daniel; Buck, Kerstin; Dutta, Sayan; Banerjee, Siddhartha; Zhengjian, Lv; Hulleman, John D.; Hindupur, Jagadish; Lin, Li-Kai; Padalkar, Sonal; McCabe, George P.; Stanciu, Lia A.; Lyubchenko, Yuri L.; Kirik, Deniz; Rochet, Jean-Christophe
title: Two C-terminal sequence variations determine differential neurotoxicity between human and mouse α-synuclein
date: 2019-07-12
journal: bioRxiv
DOI: 10.1101/700377
sha: e0d2b7fcc76dc76bd9c60b16b5329fe2a8c31026
doc_id: 275996
cord_uid: 8410w4ra

α-Synuclein (aSyn) aggregation is thought to play a central role in neurodegenerative disorders termed synucleinopathies, including Parkinson’s disease (PD). Mouse aSyn contains a threonine residue at position 53 that mimics the human familial PD substitution A53T, yet in contrast to A53T patients, mice show no evidence of aSyn neuropathology even after aging. Here we studied the neurotoxicity of human A53T, mouse aSyn, and various human-mouse chimeras in cellular and in vivo models as well as their biochemical properties relevant to aSyn pathobiology. We report that mouse aSyn is less neurotoxic than the human A53T variant as a result of inhibitory effects of two C-terminal amino acid substitutions on membrane-induced aSyn aggregation and aSyn-mediated vesicle permeabilization. Our findings highlight the importance of membrane-induced self-assembly in aSyn neurotoxicity and suggest that inhibiting this process by targeting the C-terminal domain could slow neurodegeneration in PD and other synucleinopathy disorders.

Parkinson's disease (PD) is a common progressive neurodegenerative disorder characterized 2 clinically by motor symptoms attributed to a loss of dopaminergic neurons in the substantia nigra 3 (SN). At post-mortem examination neurons in various brain regions of PD patients present with 4 cytosolic inclusions named Lewy bodies that contain amyloid-like fibrils of the presynaptic 5 protein α-synuclein (aSyn) (1) . A number of patients with familial forms of PD have been found 6 to harbor mutations in the SNCA gene, including point mutations encoding the substitutions 7 A30P, E46K, H50Q, G51D, A53E, and A53T and gene multiplications (2). Genetic and 8 neuropathological findings in humans and data from animal model studies suggest that aSyn 9

self-assembly plays a central role in the pathogenesis of PD and other neurodegenerative 10 disorders involving an accumulation of aSyn aggregates in the brain, collectively referred to as 11 synucleinopathies (3). A detailed understanding of molecular mechanisms by which aSyn forms 12 neurotoxic aggregates is critical for developing therapies aimed at slowing neurodegeneration in 13 the brains of patients with PD and other synucleinopathy disorders. 14 aSyn has been reported to adopt a natively unfolded, monomeric structure in solution (4) 15 and to exist as a compact disordered monomer in mammalian cells (5), although additional 16 evidence suggests that it can also exist as an oligomer in the cytosol (6). The protein is typically 17 expressed as a 14.4 kDa polypeptide and consists of 3 domains. The N-terminal domain 18 spanning residues 1-67 contains five conserved, lysine-rich repeats. The central region 19

spanning residues 61-95 contains a sixth lysine-rich repeat and is highly hydrophobic. A key 20 feature of this region is the presence of a segment spanning residues 71-82 that is required for 21 aSyn aggregation (7). The C-terminal region spanning residues 96-140 is enriched with proline 22

and acidic residues and is thought to regulate aSyn aggregation through auto-inhibitory long-23 range interactions (8, 9) , with electrostatic interactions mediated by the acidic residues playing a 24 major role in increasing the fibrillization lag time (10). aSyn binds to anionic phospholipid 25 vesicles by forming an amphipathic α-helix with varying lengths, including a short N-terminal 26 helix spanning residues ~1-25 and a longer helix spanning residues ~1-97 encompassing both 27 the N-terminal and central hydrophobic domains (11, 12) . Although membrane binding 28 apparently plays an important role in the normal function of aSyn related to regulation of 29 synaptic vesicle trafficking (12, 13), the protein has also been shown to undergo accelerated 30 aggregation when incubated in the presence of phospholipid vesicles at high protein:lipid ratios 31 (14-16). aSyn aggregation at the membrane surface is likely stimulated by the exposure of 32 hydrophobic residues as the membrane-bound protein shifts from the long-helix form to the 33 short-helix form (11, 16) , and by the fact that molecular interactions needed for aSyn self-34 assembly likely occur with a higher probability on the two dimensional surface of the lipid bilayer 35 than in solution (17, 18). Evidence from our laboratory suggests that membrane-induced aSyn 36 aggregation plays a key role in neurotoxicity (16), potentially via a mechanism involving 37 membrane permeabilization (19) (20) (21) (22) . 38 toxic oligomers (25, 34, 35). As a corollary, we reasoned that the human-to-mouse substitutions 180 could stimulate aSyn fibrillization, which in turn could account for the reduced neurotoxicity of m-181 aSyn and h-aSyn Chimera compared to h-aSyn A53T. To address this hypothesis, we 182 monitored the fibrillization of human, mouse, and chimeric aSyn variants using a thioflavin T 183 fluorescence assay. Consistent with previous results, we found that m-aSyn and h-aSyn A53T 184 formed fibrils with a markedly reduced lag time compared to h-aSyn WT, although the decrease 185 in lag time was more pronounced for the mouse protein ( Figure 5A ). Analysis of the chimeric 186 aSyn variants revealed that h-aSyn Chimera formed fibrils with a longer lag time compared to h-187 aSyn A53T, whereas the rate of fibrillization was only slightly decreased for m-aSyn Chimera 188 compared to m-aSyn ( Figure 5B ). 

Next, we examined the impact of human/mouse aSyn mismatches on fibril morphology 204 and dimensions based on evidence that different types of aSyn fibrils (e.g. cylindrical fibrils 205 versus flatter ribbons) can trigger different levels of neurotoxicity in cellular and animal models 206 (36). Examination of incubated samples of human, mouse, and chimeric aSyn by atomic force 207 microscopy (AFM) revealed that all of the variants formed straight, rigid, unbranched fibrils, 208 organized either as individual fibrils or in fibril bundles ( Figure 6A-D) . Two types of fibrils 209 ('twisted' and 'non-twisted') were detected in samples of m-aSyn and h-aSyn-Chimera, whereas 210 only non-twisted fibrils were observed in samples of h-aSyn-A53T and m-aSyn-Chimera under 211 our experimental conditions. The heights of the non-twisted fibrils ranged from ~8 to ~11 nm, 212

whereas the heights of the 'peak' and 'valley' regions of the twisted fibrils ranged from ~11 to 213 ~12 nm and ~7 to ~9 nm, respectively. The average pitch of twisted fibrils formed by m-aSyn 214 and h-aSyn Chimera was 123 ± 5 nm and 91 ± 8 nm, respectively ( Figure 6E ,F). 215

Further analysis of the fibrils by TEM also revealed the presence of straight, rigid, 216

unbranched fibrils for all of the variants ( Figure 6G -L). The fibrils had similar properties as those 217 described for previously reported aSyn amyloid-like fibrils (e.g. diameter of ~10-15 nm; apparent 218 winding of two fibrils around each other via a helical twist) (25). Similar features were also 219 observed for fibrils formed by aSyn variants with the S87N or N87S substitution. There were no 220 striking differences in fibril morphology among the different aSyn variants, except that individual 221 fibrils formed by h-aSyn-Chimera had a twisted appearance similar to that observed for this 222 variant by AFM (in contrast, twisted fibrils of m-aSyn were not observed by TEM). It is unclear 223 why different results were obtained for m-aSyn and h-aSyn Chimera via AFM versus TEM, 224 although this discrepancy can potentially be explained by the fact that (i) the variants formed 225 two different types of twisted fibrils, as implied by their different pitch values ( Figure 6E ,F), and 226 (ii) different types of fibrils adsorb with different efficiencies to substrates with different 227 properties such as the hydrophilic mica and hydrophobic carbon-coated grids used here for 228 AFM and TEM, respectively (37). 229

Overall, these data suggest that differences in the morphologies of aSyn fibrils formed 230 under the self-assembly conditions used in this study could potentially contribute to differences 231 in neurotoxicity among the human, mouse, and chimeric aSyn variants (Figures 2-4) . 232 233 Effects of human/mouse aSyn mismatches on membrane-induced aSyn aggregation. 234

Based on our earlier finding that lipid-induced self-assembly plays a key role in aSyn 235 neurotoxicity (16), we hypothesized that human-to-mouse substitutions should lead to a 236 decrease in aSyn aggregation propensity at membrane surfaces. Our rationale was that 237 although aSyn aggregation is primarily driven by the central hydrophobic region (7, 11, 16) , 238

modifications to the C-terminal domain have been shown to modulate aSyn self-assembly (38) 239 and aSyn-membrane interactions (39) through effects on long-range interactions. 240

As a first step, we characterized the human, mouse, and chimeric variants in terms of 241 membrane affinities, based on the premise that differences in affinity could lead to differences in 242 the variants' propensity to undergo membrane-induced aggregation (16). Recombinant h-aSyn 243 WT, h-aSyn A53T, h-aSyn Chimera, m-aSyn, and m-aSyn Chimera in a monomeric state were 244 titrated with small unilamellar vesicles (SUVs) composed of egg phosphatidylglycerol (PG) and 245 egg phosphatidylcholine (PC) (1:1 mol/mol). Analysis of the samples via far-UV CD, a method 246 used to monitor the increase in aSyn α-helical structure that results from binding of the protein 247 to phospholipids, yielded similar lipid titration curves, Kd values, and maximum α-helical 248 contents for all of the aSyn variants ( Figure 7A ; Table S1 ). These results suggested that the C-249 terminal human/mouse mismatches have little effect on the affinity of aSyn for phospholipid 250 membranes, or on the degree of α-helical structure adopted by the protein in the presence of 251 saturating lipid. 252

Next, we compared the human, mouse, and chimeric aSyn variants in terms of their 253 ability to undergo membrane-induced aggregation using a lipid-flotation assay. The proteins 254

were incubated with PG:PC SUVs under conditions that promote aSyn self-assembly, and the membrane fraction was isolated by gradient centrifugation and analyzed via Western blotting. 256

The lane loaded with h-aSyn WT contained immunoreactive bands at ~45 and 60 kDa, whereas 257 the lane containing h-aSyn A53T or m-aSyn Chimera displayed more extensive laddering, 258 including bands at 75 and 100 kDa, and an intense smear of aSyn immunoreactivity in the 259 region of the gel above 245 kDa ( Figure 7B ). In contrast, high molecular-weight species were 260 markedly less abundant in the lanes loaded with h-aSyn Chimera or m-aSyn. Densitometry 261 analysis revealed a ~2-fold greater ratio of total oligomer band intensity to monomer band 262 intensity in the h-aSyn A53T and m-aSyn Chimera samples compared to the h-aSyn Chimera 263 and m-aSyn samples, respectively ( Figure 7B ). Although the results revealed a trend towards 264 an increase in the oligomer/monomer ratios for h-aSyn A53T and m-aSyn Chimera compared to 265 the human WT protein, these effects did not reach statistical significance. Collectively, these 266 data suggested that the human-to-mouse substitutions at positions 121 and 122 interfere with 267 membrane-induced aSyn aggregation. cultured neurons and in mouse brain. Here we report that m-aSyn has a reduced propensity to 305 elicit dopaminergic cell death compared to h-aSyn A53T in primary midbrain cultures and 306 exhibits a similar effect when expressed from an AAV vector in rat SN. Furthermore, we show 307 for the first time that this differential toxicity is associated with differences in aSyn-mediated 308 vesicle disruption and aggregation at the membrane surface, rather than differences in aSyn 309 fibrillization previously suggested to be important for modulating its toxicity (42). Most 310 importantly, we provide evidence that these characteristics of the overtly toxic h-aSyn A53T suggests that different types of fibrils (e.g. cylindrical fibrils versus ribbons distinguishable by 399 EM) can act as prion-like 'strains' with different abilities to undergo cell-to-cell transmission 400 throughout the brain (36). Our AFM and TEM data imply that m-aSyn and h-aSyn Chimera 401 could potentially form different fibrillar strains (characterized by increased amounts of twisted 402 fibrils) compared to the other aSyn variants, and, therefore, the differences in in vivo 403 neurotoxicity reported here could be due to differences in the variants' abilities to seed aSyn 404 neuropathology in rat brain. However, this interpretation is challenged by evidence that (i) h- Importantly, our data reveal a strong correlation between aggregation propensity at 457 membrane surfaces and neurotoxicity among the human, mouse, and chimeric aSyn variants, 458 and thus they further support the hypothesis that lipid-induced self-assembly plays a key role in 459 aSyn-mediated neurodegeneration (11, 16, 18, 22) . The importance of membrane-induced aSyn Unexpectedly, we found that h-aSyn WT triggered vesicle disruption to a similar extent as h-485 aSyn A53T and m-aSyn Chimera. This observation was inconsistent with the fact that h-aSyn 486

WT exhibited a non-significant trend towards less extensive membrane-induced self-assembly, 487 or with evidence from our earlier studies that h-aSyn WT is considerably less toxic than h-aSyn 488 A53T in primary midbrain cultures (16) and causes neurodegeneration less rapidly than h-aSyn 489 A53T in rat midbrain (63). As one possibility, the current findings can be interpreted to mean 490 that h-aSyn WT engages in different molecular interactions during aggregation at the membrane 491 surface compared to h-aSyn A53T or m-aSyn Chimera, resulting in an increase in the extent of 492 membrane permeabilization relative to the total amount of aggregate formed. Consistent with 493 this idea, evidence from multiple studies suggests that h-aSyn WT and familial aSyn mutants 494

can form different types of amyloid-like fibrils (64-66). Overall, our data reveal a strong 495 correlation between vesicle disruption and neurotoxicity among the human, mouse, and 496 chimeric aSyn variants, and thus they support the notion that membrane permeabilization 497 associated with lipid-induced self-assembly plays a key role in aSyn-mediated 498 neurodegeneration. These findings are consistent with evidence that h-aSyn E57K, a variant 499 with a high propensity to form aggregates on a supported lipid bilayer and elicit membrane 500 damage, causes extensive dopaminergic cell death when expressed in rat SN (19, 34) . Conclusions We report for the first time that m-aSyn has a reduced ability to elicit 515 dopaminergic cell death compared to h-aSyn A53T in primary mesencephalic cultures and in rat 516 midbrain. Our results highlight the central importance of membrane-induced aSyn aggregation 517 and vesicle disruption in aSyn neurotoxicity, and they reveal a C-terminal sequence motif 518 encompassing residues 121 and 122 with an important role in modulating these activities. 519

Collectively, these findings provide a strong rationale for developing therapies aimed at 520

inhibiting aSyn aggregation at membrane surfaces, including interventions that target the C- 

and h-aSyn A53T were described previously (16, 71). cDNAs encoding h-aSyn Chimera, 593 A53T/D121G, A53T/N122S, A53T/S87N, and h-aSyn Chimera S87N (obtained using the pT7-7 594 constructs outlined above as PCR templates) were subcloned as KpnI-XhoI fragments into the 595 entry vector pENTR1A. cDNAs encoding m-aSyn, m-aSyn Chimera, and m-aSyn Chimera 596 N87S were subcloned as SalI-XhoI fragments into pENTR1A. Inserts from the pENTR1A 597 sequence. HEK293 cells were co-transfected at a confluency of 70-80% using the calcium-613 phosphate precipitation method. Plasmids used here encode essential adenoviral packaging 614 and AAV5 capsid genes as previously described (72). Three days after transfection, cells were 615 harvested in PBS and lysed by performing three freeze-thaw cycles in a dry ice/ethanol bath. 616

The lysate was then treated with benzonase and purified using a discontinuous iodixanol 617 gradient followed by Sepharose Q column chromatography (73). Vectors were concentrated 618 using a 100 kDa molecular weight cut-off column, and titers of the stock solution were 619 determined by qPCR using primers and probes targeting the ITR sequence. Before being used 620 in an experiment, vectors were diluted in PBS, pH 7.4 and re-titered, yielding the values 621 reported in the Results and in Figure 3 . it was slowly removed from the brain. After closing the wound with clips, Antisedan and 663

Temgesic were administered s.c. as an analgesic treatment and to reverse the anesthesia. 664

Histology. An overdose of sodium pentobarbital was used to kill rats 8 weeks after vector 666 delivery. Animals were perfused via the ascending aorta first with 50 mL of 0.9% (w/v) NaCl 667 followed by 250 mL of ice-cold 4% (w/v) PFA in 0.1 M phosphate buffer, pH 7.4, for 5 min. Densitometry. The optical density of TH + and VMAT2 + fibers was measured on digital images 720 of coronal striatal sections using the Zeiss microscope (Axio Zoom.V16, Zeiss, Germany). The 721 striatum of every 24 th section in the rostro-caudal axis -in total 6 sections per brain -was 722 outlined using ImageJ, and optical density readings were corrected for non-specific background 723 collected from the 5% iodixanol fraction at the top of the gradient, concentrated using a 10 kDa 803 spin filter, and analyzed via Western blotting using a primary antibody specific for aSyn (Syn-1) 804 For membrane disruption experiments, monomeric aSyn variants were isolated as 817 fibrils (10 µL, prepared as described above under 'aSyn fibrillization' with an incubation time of 836 ~100 h and diluted 1/20 in deionized water) was deposited onto the APS-mica surface, and the 837 sample was incubated for 2 min, rinsed with deionized water, and dried under an argon stream. 838

The sample was imaged with an AFM Nanoscope VIII system (Bruker, Santa Barbara, CA) 839 using MSNL probes (Cantilever F with spring constant 0.6 N/m), operating in air in peak force 840 mode. Images were acquired over a few randomly selected locations. Images were analyzed 841 using Gwyddion and FemtoScan online software (Advanced Technologies Center, Moscow, 842 Russia) (84, 85). 843 TEM analysis of aSyn amyloid-like fibrils. The morphology of aSyn amyloid-like fibrils 845 (prepared as described above under 'aSyn fibrillization' with an incubation time of ~100 h) was 846 analyzed by negative stain biological TEM (86). For biological sample preparation by negative 847 staining, which is a sample preparation technique that imparts the necessary contrast for 848 viewing during biological TEM imaging, 3 µL of aSyn sample solution (35 µM) was pipetted on a 849 discharged carbon-coated copper TEM grid substrate. Subsequently, the sample was washed 850 with deionized water carefully without letting it dry and stained with a 1% (w/v) phosphotungstic 851 acid solution (3 µL), which was left in contact with the protein on the grid for 1 min. The excess 852 solution was then removed by blotting with filter paper, and the sample was imaged using an 853 CA). Primary neuron viability data, in vivo stereology and densitometry data, densitometry data 866 from Western blots, and calcein dye release data were analyzed via ANOVA followed by 867

Tukey's multiple comparisons post hoc test for normally distributed measurements. In analyzing 868 percentage dye release data and percentage cell viability data by ANOVA, square root 869 transformations were carried out to conform to ANOVA assumptions. Normalized FFN-102 870 fluorescence data were subjected to a log transformation to account for skewness in the data. 871

The log-transformed data were analyzed using an approach that accounts for comparison Table S2 ). The data in (B) and (C) are presented as the mean ± SEM, n = 4 (B) or 1208 n = 3 (C). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, one-way ANOVA followed by Tukey's 1209 multiple comparisons post hoc test (a square root transformation was carried out on the data in 1210 panel C). 1211 

Synuclein 888 in Lewy bodies

Phenotypic spectrum of alpha-synuclein mutations: 890 New insights from patients and cellular models

The remarkable conformational 893 plasticity of alpha-synuclein: blessing or curse? Trends in molecular medicine

NACP, a protein implicated 896 in Alzheimer's disease and learning, is natively unfolded

Structural disorder 899 of monomeric alpha-synuclein persists in mammalian cells

alpha-Synuclein occurs physiologically as a helically folded 901 tetramer that resists aggregation

A hydrophobic stretch of 12 amino acid 903 residues in the middle of a-synuclein is essential for filament assembly

Mapping 906 long-range interactions in alpha-synuclein using spin-label NMR and ensemble molecular 907 dynamics simulations

Release 909 of long-range tertiary interactions potentiates aggregation of natively unstructured alpha-910 synuclein

Conserved C-912 terminal charge exerts a profound influence on the aggregation rate of alpha-synuclein

Multiple tight phospholipid-binding modes of alpha-915 synuclein revealed by solution NMR spectroscopy

Stabilization of alpha-synuclein secondary 917 structure upon binding to synthetic membranes

Wade-Martins R. alpha-Synuclein and dopamine at the 919 crossroads of Parkinson's disease

Membrane-bound a-synuclein has a high aggregation propensity 921 and the ability to seed the aggregation of the cytosolic form

Lipid 923 vesicles trigger alpha-synuclein aggregation by stimulating primary nucleation

Effects of 926 impaired membrane interactions on alpha-synuclein aggregation and neurotoxicity

A role for helical intermediates in amyloid formation by natively 929 unfolded polypeptides?

A novel pathway for 931 amyloids self-assembly in aggregates at nanomolar concentration mediated by the 932 interaction with surfaces

Mechanism of 934 membrane interaction and disruption by alpha-synuclein

Structural intermediates 937 during alpha-synuclein fibrillogenesis on phospholipid vesicles

Radiating amyloid fibril formation 940 on the surface of lipid membranes through unit-assembly of oligomeric species of alpha-941 synuclein

Endosulfine-alpha inhibits membrane-induced alpha-synuclein aggregation and protects 944 against alpha-synuclein neurotoxicity

The A53T mutation is key in defining the 949 differences in the aggregation kinetics of human and mouse alpha-synuclein

Inhibition of fibrillization and accumulation of 952 prefibrillar oligomers in mixtures of human and mouse a-synuclein

Characterization of conformational and dynamic 955 properties of natively unfolded human and mouse alpha-synuclein ensembles by NMR: 956 implication for aggregation

alpha -Synucleinopathy and 958 selective dopaminergic neuron loss in a rat lentiviral-based model of Parkinson's disease

Toxic effects of human and rodent variants of alpha-synuclein 961 in vivo. The European journal of neuroscience

Role of 963 alpha-synuclein carboxy-terminus on fibril formation in vitro

Aggregated alpha-966 synuclein mediates dopaminergic neurotoxicity in vivo

Selective loss of nigral 968 dopamine neurons induced by overexpression of truncated human alpha-synuclein in mice

Tyrosine and serine 971 phosphorylation of alpha-synuclein have opposing effects on neurotoxicity and soluble 972 oligomer formation

Multiple phosphorylation 974 of alpha-synuclein by protein tyrosine kinase Syk prevents eosin-induced aggregation. 975 FASEB journal : official publication of the Federation of American Societies for 976 Experimental Biology

In vivo demonstration 978 that alpha-synuclein oligomers are toxic

Pre-fibrillar 980 alpha-synuclein variants with impaired beta-structure increase neurotoxicity in Parkinson's 981 disease models

Structural and 983 functional characterization of two alpha-synuclein strains

Alternate aggregation pathways of the 986 Alzheimer beta-amyloid peptide: Abeta association kinetics at endosomal pH

989 Phosphorylation at Ser-129 but not the phosphomimics S129E/D inhibits the fibrillation of 990 alpha-synuclein

Allostery in a disordered protein: oxidative 992 modifications to alpha-synuclein act distally to regulate membrane binding

Neuropathology in 995 mice expressing mouse alpha-synuclein

Induction of 997 de novo alpha-synuclein fibrillization in a neuronal model for Parkinson's disease

Definition of a Molecular Pathway Mediating alpha-1000

Mutant LRRK2(R1441G) BAC 1002 transgenic mice recapitulate cardinal features of Parkinson's disease

Nigrostriatal 1005 dopaminergic deficits and hypokinesia caused by inactivation of the familial Parkinsonism-1006 linked gene DJ-1

Parkinson-like 1008 syndrome induced by continuous MPTP infusion: Convergent roles of the ubiquitin-1009 proteasome system and {alpha}-synuclein

Histochemical observations on rodent brain melanin. Brain research 1011 bulletin

Neuromelanin in human 1013 dopamine neurons: comparison with peripheral melanins and relevance to Parkinson's 1014 disease

Intrastriatal injection of pre-formed mouse alpha-synuclein fibrils into rats triggers alpha-1017 synuclein pathology and bilateral nigrostriatal degeneration

Structural comparison 1019 of mouse and human alpha-synuclein amyloid fibrils by solid-state NMR

Exogenous 1022 alpha-synuclein fibrils induce Lewy body pathology leading to synaptic dysfunction and 1023 neuron death

Pathological alpha-1025 synuclein transmission initiates Parkinson-like neurodegeneration in nontransgenic mice

Fibrils formed in vitro from a-synuclein and two 1028 mutant forms linked to Parkinson's disease are typical amyloid

Molecular and Biological 1031 Compatibility with Host Alpha-Synuclein Influences Fibril Pathogenicity

Elucidating the role of 1034 C-terminal post-translational modifications using protein semisynthesis strategies: alpha-1035 synuclein phosphorylation at tyrosine 125

Elucidating the Role of Site-Specific Nitration 1037 of alpha-Synuclein in the Pathogenesis of Parkinson's Disease via Protein Semisynthesis 1038 and Mutagenesis

alpha-Synuclein assembles into higher-order multimers 1040 upon membrane binding to promote SNARE complex formation

Conformation-specific binding of alpha-synuclein to novel protein partners detected by 1044 phage display and NMR spectroscopy

Disentangling 1046 the relationship between lewy bodies and nigral neuronal loss in Parkinson's disease. 1047 Journal of Parkinson's disease

Functionally 1049 different alpha-synuclein inclusions yield insight into Parkinson's disease pathology

Contribution of somal Lewy bodies to neuronal death

Distinct 1054 region-specific alpha-synuclein oligomers in A53T transgenic mice: implications for 1055 neurodegeneration

Selective neuronal vulnerability in Parkinson 1057 disease

Viral vector mediated overexpression of human 1059 alpha-synuclein in the nigrostriatal dopaminergic neurons: a new model for Parkinson's 1060 disease

Solid-state NMR reveals 1062 structural differences between fibrils of wild-type and disease-related A53T mutant alpha-1063 synuclein

Site-1065 specific perturbations of alpha-synuclein fibril structure by the Parkinson's disease 1066 associated mutations A53T and E46K

Conversion of 1068 wild-type alpha-synuclein into mutant-type fibrils and its propagation in the presence of 1069 A30P mutant

Vesicle permeabilization by protofibrillar a-synuclein is 1071 sensitive to Parkinson's disease-linked mutations and occurs by a pore-like mechanism

Vesicle 1074 permeabilization by protofibrillar a-synuclein: implications for the pathogenesis and 1075 treatment of Parkinson's disease

Visualization of Syk-antigen 1077 receptor interactions using green fluorescent protein: differential roles for Syk and Lyn in 1078 the regulation of receptor capping and internalization

Global analysis of human nonreceptor tyrosine kinase specificity using high-density peptide 1082 microarrays

Methionine sulfoxide 1084 reductase A protects dopaminergic cells from Parkinson's disease-related insults

Novel tools for production and purification of 1087 recombinant adenoassociated virus vectors

Recombinant 1089 adeno-associated virus purification using novel methods improves infectious titer and yield

Design-based stereology in neuroscience

Stereological methods for estimating the total number of neurons and synapses: 1094 issues of precision and bias

The efficiency of systematic sampling in stereology and its 1096 prediction

Specificity and kinetics of 1098 alpha-synuclein binding to model membranes determined with fluorescent excited state 1099 intramolecular proton transfer (ESIPT) probe

Parameters of helix-coil transition 1101 theory for alanine-based peptides of varying chain lengths in water

The mode of alpha-synuclein binding 1104 to membranes depends on lipid composition and lipid to protein ratio

Serum-induced leakage of liposome contents. Biochimica et 1107 biophysica acta

Nanoprobing of the effect of Cu(2+) 1111 cations on misfolding, interaction and aggregation of amyloid beta peptide

AFM for analysis of structure and dynamics of DNA and 1115 protein-DNA complexes

Dynamics of nucleosomes assessed with time-lapse 1117 high-speed atomic force microscopy

Remodeling of RecG Helicase at the DNA 1119 Replication Fork by SSB Protein

A negative staining method for high resolution electron microscopy 1121 of viruses

(i) on the left show typical AFM topographic images of fibrils formed by each 1174 aSyn variant. Dashed boxes indicate areas shown at a higher magnification in a spectral color 1175 scheme in panels A(ii) -D(ii). Panels A(iii)-D(iii) on the right show cross-section profiles

Regions of the fibrils from which the profiles are taken are shown with white solid lines on the 1177 AFM images in panels A(ii)-D(ii). Twisted fibrils were observed for m-aSyn (B) and h-aSyn 1178

Twisted and non-twisted fibrils account for 65% and 35% (respectively) of total 1179 fibrils formed by m-aSyn, and 76% and 24% (respectively) of total fibrils

Cross-section profiles in panels A(iii) and D(iii) reveal heights of non-twisted fibrils

Cross-section profiles 1 and 2 reveal heights of twisted fibrils at peak regions

nm) and valley regions (~7-9 nm), respectively) in panels B(iii) and C(iii). (E, F) Histograms 1183 showing distances between peaks of twisted fibrils formed by m-aSyn (E) and h-aSyn Chimera 1184

The distances were determined from cross-section profiles obtained along the long axis of 1185 the fibrils

Scale bar, 100 nm. (L) Higher-1187 magnification images generated from boxed regions of panels G, I, and J show a pair of fibrils 1188 wound around each other via a helical twist (I), two fibrils aligned in parallel (II), or individual 1189 fibrils with a twisted morphology

Figure 7. Human/mouse mismatches at positions 121 and 122 affect membrane-induced 1192

The data were fit to equation 3 (see 1195 'Experimental Procedures'), and values for KD, minimum [ ] ,222 , and maximum α-helical 1196 content were determined from the values of the fit parameters