Report

Oxidative Damage Linked to Neurodegeneration by Selective α-Synuclein Nitration in Synucleinopathy Lesions

See allHide authors and affiliations

Science  03 Nov 2000:
Vol. 290, Issue 5493, pp. 985-989
DOI: 10.1126/science.290.5493.985

Abstract

Aggregated α-synuclein proteins form brain lesions that are hallmarks of neurodegenerative synucleinopathies, and oxidative stress has been implicated in the pathogenesis of some of these disorders. Using antibodies to specific nitrated tyrosine residues in α-synuclein, we demonstrate extensive and widespread accumulations of nitrated α-synuclein in the signature inclusions of Parkinson's disease, dementia with Lewy bodies, the Lewy body variant of Alzheimer's disease, and multiple system atrophy brains. We also show that nitrated α-synuclein is present in the major filamentous building blocks of these inclusions, as well as in the insoluble fractions of affected brain regions of synucleinopathies. The selective and specific nitration of α-synuclein in these disorders provides evidence to directly link oxidative and nitrative damage to the onset and progression of neurodegenerative synucleinopathies.

Oxidative injury has been implicated in the pathogenesis of neurodegenerative disorders including Alzheimer's disease (AD) (1), Parkinson's disease (PD) (2, 3), dementia with Lewy bodies (DLB) (4), amyotrophic lateral sclerosis (5), and Huntington's disease (6). Oxidative injury occurs when the compensatory antioxidant capacity of cells is overwhelmed by excess production of reactive species that damage lipids, nucleic acids, proteins, and other cellular components. Both reactive oxygen and nitrogen species are produced in vivo and may act synergistically to form nitrating agents that modify proteins as well as other biomolecules such as thiols, aldehydes, and lipids (7, 8). For example, superoxide reacts with nitric oxide to generate peroxynitrite that forms biologically active nitrating agents in the presence of CO2 or other catalysts (redox active metals, or metalloproteins) (9,10), which can convert native tyrosine residues in proteins into 3-nitrotyrosine (3-NT).

Initially described over 10 years ago (11), α-synuclein (α-syn) is a 140–amino acid, highly conserved protein that is abundant in neurons, especially presynaptic terminals (12, 13). Two mutations in the α-syn gene have been shown to be pathogenic for familial PD in rare kindreds (14, 15), and it has been demonstrated that α-syn is the major component of Lewy bodies (LBs) and Lewy neurites (LNs) in PD, DLB, and the LB variant of AD (LBVAD) (16–18). Moreover, α-syn also appears to be a major component of glial and neuronal cytoplasmic inclusions (GCIs and NCIs) in multiple system atrophy (MSA) brains (19,20), as well as of the LB-like inclusions, neuraxonal spheroids, and LNs in neurodegeneration with brain iron accumulation type 1 (NBIA1), a rare disorder previously termed Hallervorden-Spatz disease (19, 21).

Immunoreactive 3-NT was detected in LBs of the PD brain and appears to be a common feature of α-syn lesions in many synucleinopathies (22), but no protein building blocks of disease-specific lesions that are modified with 3-NT have been identified in any neurodegenerative disorder. Thus, to determine if α-syn in the hallmark lesions of synucleinopathies are specifically nitrated, we raised monoclonal antibodies (mAbs) to nitrated α-syn (23) and screened them by enzyme-linked immunosorbent assay (ELISA) to identify anti–nitrated α-syn mAbs using nitrated and unmodified α-syn as well as other nitrated proteins, i.e., recombinant human β-syn, ribonuclease A (RNase A), cytochrome c, bovine superoxide dismutase–1 (SOD-1), and phospholipase A2 (PLA2). Several mAbs (e.g., nSyn 8, 14, and 24) specifically recognized nitrated α- and β-syn but not nonnitrated α-syn or any other nitrated proteins tested (Fig. 1A). mAb nSyn 12 reacted with nitrated α-syn, as well as with nitrated β-syn, RNase A, and PLA2, but not with nonnitrated α-syn. Western blots were used to further characterize mAbs that recognized nitrated synucleins by ELISA, and nSyn 24 bound specifically to nitrated α- and β-syn, but not to nonnitrated α-syn or other nitrated proteins (Fig. 1B). Indeed, nSyn 24 did not cross-react with these other proteins even after longer Western blot exposure times. However, nSyn 24 also recognized dimerized nitrated α-syn with an apparent molecular mass of about 35 kD (Fig. 1B). In contrast, anti–α-syn mAb LB509 (18) recognized nitrated and nonnitrated α-syn, whereas polyclonal anti–3-NT (24) recognized all the nitrated proteins tested. However, because the cytochrome c and PLA2 immunoreactive bands were weak, these proteins may be nitrated to a lesser extent, or the antibody may have less affinity for 3-NT in these proteins.

Figure 1

Characterization of mAbs to nitrated α-syn. (A) ELISA assessment of the specificities of representative mAbs with unmodified human recombinant α-syn and nitrated (n) human recombinant α-syn, human recombinant β-syn, RNase A, cytochrome c (cyto c), bovine superoxide dismutase–1 (SOD-1), and phospholipase A2 (PLA2). All nitrated proteins were prepared as described (31); 25 ng of each protein was immobilized onto ELISA plates and the binding of mAbs to antigens was detected with 3,3′,5,5′-tetramethylbenzidine. (B) Specificity of mAb nSyn 24 as determined by Western blot analysis. Fifty nanograms of each protein were loaded in each lane of 12% polyacrylamide gels, separated by electrophoresis, and transferred to nitrocellulose membranes that were probed with either mAb nSyn 24, mAb LB509, or rabbit anti–3-NT antibody and developed by enhanced chemiluminescence. (C) Epitope mapping of anti–nitrated α/β-syn mAbs. Fifty nanograms of full-length wild-type or truncated recombinant human α-syn or mutagenized species thereof were resolved on 15% polyacrylamide gels and probed with each antibody for Western blot analysis. Recombinant α-syn proteins were expressed in Escherichia coli and purified as described (36).

The epitopes recognized by the anti–nitrated α/β-syn mAbs were characterized by Western blot performed on nitrated full-length and truncated recombinant human α-syn proteins, as well as on nitrated recombinant α-syn proteins that were mutagenized to substitute Phe for one or more of the four Tyr (Y) residues in human α-syn (i.e., Y39, Y125, Y133, and/or Y136). These studies showed that mAbs nSyn 8, nSyn 12, and nSyn 24 recognized the COOH-terminal region of α-syn because none recognized nitrated α-syn amino acids 1 to 110 or 1 to 120; however, the substitution of any single Tyr residue with Phe was not sufficient to abolish the binding of these antibodies to nitrated α-syn (Fig. 1C) (25). The binding of nSyn 12 was dependent on nitration of either Y125 or Y136, but the recognition of nitrated Y136 dominated that of nitrated Y125 (Fig. 1C). The epitope recognized by nSyn 24 also was dependent on nitration of two Tyr residues, and Y133 appeared to be the dominant protein, with the reactivity for nitrated Y133 being more robust than nitrated Y125. The epitope recognition of nSyn 14 is dependent on the nitration of Y39 in α-syn or β-syn (Fig. 1C). The conservation of all four Tyr residues and the amino acid sequences proximal to them in human α- and β-syn may explain the recognition of both nitrated synuclein proteins by all the mAbs reported here, as verified by Western blot analysis and ELISA.

In immunohistochemical studies, numerous LBs, LNs, GCIs, NCIs, LB-like inclusions, and neuraxonal spheroids in brains with different synucleinopathies were robustly labeled by the mAbs nSyn 8, 12, and 24, which recognize the nitrated COOH-terminal region of α-syn (Fig. 2). For example, in DLB and LBVAD brains, staining with nSyn 8 detected abundant cortical and nigral LBs (Fig. 2, A to C), but LNs were not as extensively labeled, whereas many GCIs and occasional NCIs were detected by this mAb in MSA brains (Fig. 2D). Further, many LB-like inclusions and neuraxonal spheroids, as well as occasional GCIs, were detected by nSyn 8 in the NBIA1 brain (Fig. 2E). Similarly, the nSyn 12 mAb labeled large numbers of LBs (Fig. 2F) and GCIs (Fig. 2G) in DLB and MSA brains, respectively. The nSyn 24 mAb robustly stained cortical and nigral LBs, LNs, and neuraxonal spheroids in PD and DLB brains (Fig. 2, K to O), GCIs in MSA brains (Fig. 2, P and Q), and LB-like inclusions and neuraxonal spheroids in the NBIA1 brain (Fig. 2R). In addition, staining with the nSyn 14 mAb detected nigral LBs (Fig. 2H), and cortical LBs (Fig. 2I) in LBVAD and GCIs in MSA (Fig. 2J), although not as robustly as the antibodies directed against COOH-terminal epitopes. Because the nSyn 8, nSyn 14, and nSyn 24 mAbs detect nitrated α- or β-syn and previous studies demonstrated α-syn but not β-syn in the lesions mentioned above (16–20, 26), we conclude that these mAbs detect only nitrated α-syn in these lesions. Staining with the mAbs labeled the peripheral region more intensely than the central core of classical LBs. In sharp contrast, none of these mAbs stained AD amyloid plaques, neurofibrillary tangles (NFTs), or any lesions in progressive supranuclear palsy, corticobasal degeneration, or Pick's disease brains (Fig. 2S) (27). As an additional control for antibody specificity, nSyn 24 was preincubated with 5 μg of either an unnitrated or nitrated synthetic polypeptide corresponding to amino acids 115 to 140 of human α-syn. Preincubation with the nitrated peptide almost completely eliminated staining (Fig. 2T), compared with antibody absorbed with the unnitrated peptide (Fig. 2U). Collectively, these studies show that nitration of α-syn in synucleinopathies occurs at both the Tyr residues within the COOH-terminal region and at Y39 located within the KTKEGV repeat region.

Figure 2

Immunostaining of diverse synucleinopathy lesions with anti–nitrated α/β-syn mAbs as described (37). Immunostaining by the nSyn 8 mAb of nigral LBs in DLB (A) and LBVAD (B), cortical LBs in LBVAD (C), GCIs and NCIs in MSA (D), and LBs and neuraxonal spheroids of NBIA1 (E). Staining with the nSyn 12 mAb labels cortical LBs in DLB (F, with high power view in inset), as well as GCIs in MSA (G). Staining with the nSyn 14 mAb labels nigral LBs (H), and cortical LBs (I), in LBVAD and GCIs in MSA (J). Immunostaining with the nSyn 24 mAb labels LBs and LNs (K) and a neuraxonal spheroid (L) in PD, and nigral LBs (M), hippocampal CA2/3 LNs (N), and cortical LBs (O) in DLB. nSyn 24 also labels GCIs in MSA (P and Q) and LB-like inclusions and neuraxonal spheroids in NBIA1 (R). Staining of AD hippocampus with nSyn 24 reveals no immunoreactivity (S), and preabsorption of nSyn 24 with nitrated α-syn COOH-terminal peptide (amino acids 115 to 140) substantially reduced immunostaining (T) compared to absorption with nonnitrated peptide (U). Bar, 10 μm (A to D, F inset, G, H, J to N, Q, and R) and 30 μm (E, F, I, O, P, and S to U).

Using double-label immunofluorescence, we analyzed the abundance of α-syn lesions detected with anti–nitrated α/β-syn mAbs—for example, white-matter GCIs in MSA cerebellum were double-labeled with nSyn 24 and anti–α-syn rabbit antibody SNL 4 (28), which was subsequently visualized separately (Fig. 3, A and B) or simultaneously (Fig. 3C). Cortical LBs also were double-labeled with these antibodies (Fig. 3D). Because more extensive analysis revealed that most LBs, GCIs, and spheroids were labeled by both antibodies, nitrated α-syn is a widespread and abundant component of these lesions.

Figure 3

Double-label immunofluorescence of GCIs with the SNL 4 and nSyn 24 antibodies. Detection of GCIs from MSA with the mouse anti–nitrated α/β-syn mAb nSyn 24 (A), affinity-purified rabbit anti–α-syn antibody SNL 4 (B), or both (C) with a goat anti-mouse immunoglobulin G (IgG) Alexa Fluro 488–conjugated antibody and a goat anti-rabbit IgG Alexa Fluro 594–conjugated antibody (Molecular Probes, Eugene, Oregon). The sections were covered with Vectashield-DAPI (4′,6′-diamidino-2-phenylindole) mounting medium (Vector Laboratories, Burlingame, California). Most GCIs are double-labeled. (D) Double-label immunofluorescence demonstrating colocalization of SNL 4 and nSyn 24 immunoreactivities in a cortical LB. Bar, 30 μm (A to C) and 10 μm (D).

Immunoelectron microscopy revealed that the nSyn 24 mAb labeled filamentous structures in the major α-syn lesions (Fig. 4). For example, nSyn 24 intensely labeled most filaments in the GCIs of MSA (Fig. 4A). Moreover, although nSyn 24 labeled filaments more intensely in the periphery than in the core of LB-like inclusions in NBIA1 (Fig. 4B), cortical LBs (25), spheroidlike inclusions (Fig. 4C), and LN-like inclusions in NBIA1 (Fig. 4D) were labeled throughout. The immunolabeling of these lesions was restricted to tubular, filamentous profiles and rare straight filamentous structures (Fig. 4D, inset). Thus, nitrated α-syn is an integral component of the α-syn filaments that form the defining lesions of diverse synucleinopathies.

Figure 4

Silver enhanced immunoelectron microscopy of synucleinopathy lesions with anti–nitrated α/β-syn mAb, nSyn 24. (A) Immunopositive GCI filaments in MSA. The arrow indicates the nucleus of the oligodendrocyte. Classical LB-like (B), spheroidlike (C), and LN-like (D) inclusions in NBIA 1. (Insets) Immunolabeled tubular, filamentous structures at higher magnification. The inset in (D) depicts a rare straight filamentous structure in spheroid and LN-like inclusions. Bar, 2 μm for low-magnification images and 200 nm for insets. Immunopositive structures visualized with DAB were silver-enhanced as described (38).

Because previous studies showed that normal α-syn isolated from control brains is soluble in high-salt (HS) and Triton X-100 buffers, but abnormal forms of α-syn are recovered from HS/Triton X-100–insoluble fractions of synucleinopathy brains (18,19, 29), we examined whether nitrated α-syn was present in soluble and insoluble extracts of LBVAD and control brains. In Western blot studies with the nSyn 24 mAb to detect nitrated α-syn, an immunoreactive band with the same electrophoretic mobility as α-syn was present in the HS/Triton-insoluble fraction of the LBVAD brains, but not in control brain or in any of the HS-soluble fractions (Fig. 5). We confirmed and extended these findings by using an anti–α-syn mAb Syn 208 (28) to demonstrate α-syn in the HS-soluble fractions of control and LBVAD brains as well as monomeric and aggregated α-syn in the HS/Triton-insoluble fractions of the LBVAD brain (Fig. 5). Further, an antibody specific for β-syn (Syn 207) (19, 28) detected β-syn in the HS fraction of LBVAD and normal control brains, but not in HS/Triton-insoluble fractions (30). The higher molecular mass α-syn immunoreactive bands detected with the Syn 208 antibody were not apparent in blots probed with nSyn 24, and we speculate that these bands are dityrosine–cross-linked α-syn polymers (31).

Figure 5

Nitrated α-syn is in the HS/Triton-insoluble fraction of LBVAD brains. Gray-matter extracts of cingulate gyrus (A) or amygdala (B) from neuropathologically normal brain (control) or LBVAD brains were biochemically fractionated into HS/Triton-soluble and -insoluble fractions (39) followed by ECL-based Western blot analysis with anti–α-syn–specific antibody Syn 208 or anti–nitrated α/β-syn antibody nSyn 24 as probes. The arrow indicates the presence of nitrated α-syn in the HS/Triton-insoluble fraction of LBVAD brains. Higher molecular mass species detected by nSyn 24 mAb are nonspecific bands because they are present in both control and LBVAD samples.

The experiments described here provide evidence that α-syn is a specific target of nitrating agents in several diverse neurodegenerative synucleinopathies. We demonstrated this by showing that mAbs specific for nitrated α/β-syn (i) stain numerous LBs, LNs, GCIs, NCIs, LB-like inclusions, and neuraxonal spheroids in synucleinopathies but not in NFTs or senile plaques; (ii) label the filamentous α-syn structures that form these lesions; and (iii) detect nitrated α-syn only in HS/Triton-insoluble fractions of LBVAD brain tissue. Although these mAbs recognized both nitrated α-syn and β-syn, previous observations have shown that β-syn does not accumulate in these lesions (16–20,26). Thus, our study provides evidence that nitrated α-syn is present in these lesions. Because previous studies only used antibodies specific for 3-NT to detect 3-NT immunoreactivity in NFTs in AD and LBs in PD brains (22, 32, 33), the exact molecular target of tyrosine nitration in these pathological lesions had not been identified. However, by using the mAbs specific for nitrated α/β-syn described here, we demonstrated that α-syn is indeed a target of nitration in the hallmark lesions of diverse synucleinopathies.

Our observation that nitrated α-syn is present only in the HS/Triton-insoluble fractions of LBVAD cortex suggests that the solubility of α-syn may be reduced by tyrosine nitration and that this modification may promote the fibrillogenesis of α-syn and/or its aggregation into hallmark lesions of synucleinopathies. Alternatively, nitration of α-syn may occur only after it has assembled into filamentous structures. Nitration may render α-syn more resistant to proteolysis or alter other properties of this synaptic protein, thereby playing a mechanistic role in the formation and/or stability of α-syn lesions, as well as in the onset and progression of synucleinopathies.

Although we detected nitrated α-syn in the major signature lesions of diverse synucleinopathies, the extent of this nitration remains to be determined. Further, because nitrating agents also can oxidize tyrosine residues to form o-o′-dityrosine, and this can result in the covalent cross-linking of α-syn and the formation of stable α-syn polymers (31), more than one nitrating process may contribute to the pathogenesis of α-syn lesions. Because the formation of 3-NT and o-o′-dityrosine can occur simultaneously under oxidative and nitrative conditions, species of α-syn that have been modified byo-o′-dityrosine cross-linking as well as by 3-NT modification may contribute to the pathogenesis of synucleinopathies.

Our findings suggest that impairment of cellular antioxidative mechanisms or overproduction of reactive species may be a primary event leading to the onset and progression of neurodegenerative synucleinopathies. Thus, elucidation of the role of oxidative and nitrative injury in mechanisms underlying these and other neurodegenerative disorders may lead to the identification of therapeutic targets to prevent or reverse these diseases.

  • * These authors contributed equally to this work.

  • To whom correspondence should be addressed. E-mail: vmylee{at}mail.med.upenn.edu

REFERENCES AND NOTES

View Abstract

Navigate This Article