Edit · 2025 08 20 095455 Nature S41467 025 62755 1

Saving needs a token. Log in or register to get one. Edits are public and licensed CC-BY-SA-4.0.

# Design of Ig-like binders targeting α-synuclein fibril for mitigating its pathological activities | Nature Communications

**URL:** https://www.nature.com/articles/s41467-025-62755-1

**Saved:** 2025-08-20T16:54:58.925Z

---

Download PDF
Article
Open access
Published: 09 August 2025
Design of Ig-like binders targeting α-synuclein fibril for mitigating its pathological activities
Shuyi Zeng, Xingyu Xiong, …Dan Li Show authors

Nature Communications 
volume
 16, Article number: 7368 (2025) Cite this article

2077 Accesses

19 Altmetric

Metrics
details

Abstract

Parkinson’s disease (PD) is characterized by the accumulation and spread of pathological α-synuclein (α-syn) fibrils, which contribute to neuroinflammation and neurodegeneration. Here we show that two immunoglobulin-like (Ig-like) domains derived from α-syn receptors, the D1 domain of lymphocyte-activation gene 3 (L3D1) and the V domain of advanced glycation end-products (vRAGE), effectively block cell surface binding of α-syn fibrils, suppress fibrils-induced neuronal α-syn aggregation, and reduce inflammatory responses in microglia. Building on this, we identified two additional Ig-like binders, the D1 domain of cluster of differentiation 4 (CD4 D1) and the D1 domain of chimeric antigen receptor (CAR D1), that target the C-terminal region of α-syn fibrils and mitigate fibrils-induced pathological activities. A structure-guided mutant, CAR D1_Mut, exhibits enhanced binding affinity and functional efficacy. These findings highlight the potential of Ig-like binders as molecular tools to interfere with pathological α-syn interactions.

Similar content being viewed by others
Fibrillar α-synuclein induces neurotoxic astrocyte activation via RIP kinase signaling and NF-κB
Article Open access 31 July 2021
Phenotypic manifestation of α-synuclein strains derived from Parkinson’s disease and multiple system atrophy in human dopaminergic neurons
Article Open access 21 June 2021
Effects of innate immune receptor stimulation on extracellular α-synuclein uptake and degradation by brain resident cells
Article Open access 16 February 2021
Introduction

Parkinson’s disease (PD) is an age-related progressive neurodegenerative disorder characterized by the accumulation of intraneuronal Lewy bodies/Lewy neurites (LBs/LNs), which contain substantial amounts of α-synuclein (α-syn) fibrils1,2,3,4. These pathological α-syn fibrils possess a strong capability to propagate from cell to cell within the brain, following the Braak staging pattern5,6,7,8,9. This spread of α-syn pathology leads to neurodegenerative changes, loss of neuronal function, and exacerbation of motor and cognitive symptoms as PD progresses10,11,12,13,14,15. Additionally, α-syn fibrils trigger the release of pro-inflammatory cytokines from microglia, and induce neuroinflammation, which further damages neurons and accelerates PD progression16,17,18. Consequently, α-syn fibrils have emerged as a crucial target for developing effective therapeutic treatments for PD.

Recently, multiple receptors on neurons and microglia have been identified to directly bind to α-syn fibrils, mediating their cell-to-cell transmission and inducing neuroinflammation. For instance, receptors such as FAM171A219, amyloid precursor-like protein 1 (APLP1)20,21, lymphocyte-activation gene 3 (LAG3)21,22,23,24, and low-density lipoprotein receptor-related protein 1 (LRP1)25,26,27 on neurons facilitate the uptake and internalization of α-syn fibrils, promoting their spread throughout the brain. The binding of α-syn fibrils to advanced glycation end-products (RAGE) and Toll-like receptor 2 (TLR2) on microglia induces inflammatory responses28,29,30. Notably, mechanistic studies have revealed that these receptors, including LAG3, RAGE, and APLP1, capture α-syn fibrils by binding to a common acidic C-terminal region that is flanking on the surface of α-syn fibrils21,31. These findings indicate that the C-terminal region is a key pathological epitope of α-syn fibrils for receptor binding, making it a critical target for alleviating α-syn fibril pathology.

In this study, we explore the potential of the isolated immunoglobulin-like (Ig-like) domains from previously identified α-syn fibril receptors as molecular tools to modulate the pathological activities of α-syn fibrils. We found that, by blocking α-syn fibrils attachment to cell surfaces, the isolated D1 domain of LAG3 (L3D1) and the V domain of RAGE (vRAGE) alone can inhibit fibrils-induced neuronal α-syn aggregation and reduce neuroinflammatory responses. We further identified two additional Ig-like binders: the D1 domain of the cluster of differentiation 4 receptor (CD4 D1)32,33,34 and the D1 domain of the chimeric antigen receptor (CAR D1)35,36,37. These recombinant Ig-like binders replicate L3D1’s binding characteristics to α-syn with varying affinities and inhibit fibrils-induced neuronal α-syn aggregation and neuroinflammation. Furthermore, enhancing the positive surface charge of CAR D1 significantly increases its binding affinity for α-syn, enhancing its ability to reduce the fibrils-induced neuronal α-syn aggregation and neuroinflammation. Our findings reveal that these recombinant Ig-like binders interact with the C-terminal of α-syn through electrostatic interactions, providing molecular tools to study and potentially modulate α-syn-related pathogenesis in cellular models.

Results
L3D1 and vRAGE inhibit fibrils-induced neuronal α-syn aggregation and inflammation

Our previous studies revealed that Ig-like domains, including L3D1 and vRAGE, preferentially bind α-syn fibrils by directly interacting with their C-terminal region21,31. We hypothesized that L3D1 or vRAGE alone could serve as protein binders against α-syn fibrils attachment to cell surfaces by competing with various receptors for α-syn fibrils binding, thereby alleviating fibrils-induced neuronal α-syn aggregation and inflammation. To test this hypothesis, we overexpressed and isolated L3D1 and vRAGE and examined their effects in rat primary cortical neurons using both a well-documented fibrils neuron surface binding assay and a widely used fibrils-induced neuronal α-syn aggregation assay38.

As shown in the representative immunofluorescence images in Fig. 1a and Supplementary Fig. 1, after incubating α-syn preformed fibrils (PFF) in the culture medium for 2 h, we observed significant adherence of α-syn PFF to neurons, as indicated by the prominent fluorescence signal of α-syn (Fig. 1a and Supplementary Fig. 1). Remarkably, when either L3D1 or vRAGE was present at a molar ratio of 1:1 (α-syn PFF: L3D1/vRAGE), the amount of PFF binding to neurons was significantly reduced. Statistical analyses further confirmed that both L3D1 and vRAGE significantly decreased α-syn PFF adherence to neurons (Fig. 1b). We then characterized the impact of L3D1 and vRAGE on the pathology of α-syn PFF, through measuring the fluorescence intensity of phosphorylated Serine 129 (pS129) α-syn, a pathological marker of endogenous α-syn aggregates39,40. The results indicated that both L3D1 and vRAGE reduced α-syn PFF-induced pathological neuronal α-syn aggregation in an in vitro primary neuronal model (Fig. 1c, d).

Fig. 1: L3D1, dL3D1, and vRAGE block α-syn PFF binding to primary neurons and prevent pathological α-syn aggregation.

a Representative immunofluorescence images of PBS, 200 nM α-syn PFF, and the mixtures of 200 nM α-syn PFF-L3D1/dL3D1/vRAGE (1:1, molar ratio) binding to rat primary cortical neurons. DAPI (blue), microtubule-associated protein 2 (MAP2) (green), and α-syn (red) were stained. Scale bar, 20 μm. b Statistical analyses of 200 nM α-syn PFF-L3D1/dL3D1/vRAGE (1:1, molar ratio) mixtures binding to rat primary cortical neurons compared with α-syn PFF binding to rat primary cortical neurons. The intensity of α-syn was normalized to the intensity of DAPI for each sample. 24 images were randomly taken for each group, derived from 3 biological replicates, with 8 images per replicate. One-way ANOVA followed by Tukey’s post-hoc test. Data shown are mean ± s.d. ***p < 0.001; ****p < 0.0001. The p value for PFF vs. PFF:L3D1 = 1:1 was 0.0009; other p values were all <0.0001. c Representative immunofluorescence images of rat primary neurons treated with PBS, 200 nM α-syn PFF, and the mixtures of 200 nM α-syn PFF-L3D1/dL3D1/vRAGE (1:1, molar ratio) for 14 d, respectively. DAPI (blue), MAP2 (green), and pS129 α-syn (red) were stained. Scale bar: 20 μm. The dashed line boxes highlight areas within the solid boxes, shown at 4× magnification. d Quantitative analyses of pS129 α-syn aggregation induced by PBS, 200 nM α-syn PFF, and the mixtures of 200 nM α-syn PFF-L3D1/dL3D1/vRAGE (1:1, molar ratio). The intensity of pS129 α-syn was normalized to the intensity of DAPI for each sample. 24 images were randomly taken for each group, derived from 3 biological replicates, with 8 images per replicate. One-way ANOVA followed by Tukey’s post-hoc test. Data shown are mean ± s.d. ***p < 0.001; ****p < 0.0001. The p value for PFF vs. PFF:L3D1 = 1:1 was 0.0004; other p values were all <0.0001. Source data are provided as a Source Data file.

Full size image

Notably, L3D1 features an extra loop that interacts with the major histocompatibility complex class II (MHC-II), which is crucial for mediating LAG3’s function in immune responses and tumor escape41. To preserve its α-syn PFF binding activity while eliminating its normal function, we developed dL3D1, a truncated version of L3D1 that lacks the MHC-II binding loop but retains the ability to bind α-syn PFF. Notably, this truncated form effectively blocked α-syn PFF binding to neuronal surfaces and prevented the formation of endogenous phosphorylated α-syn aggregates (Fig. 1a–d and Supplementary Fig. 1).

We next assessed the influence of these three isolated Ig-like domains on α-syn fibrils-induced neuroinflammatory responses using BV2 cells, an immortalized microglial cell line. We first examined whether these three Ig-like binders affected α-syn PFF binding to the cell surface. In the BV2 membrane surface binding assay, L3D1, dL3D1, and vRAGE significantly blocked the binding of α-syn PFF to the BV2 cells, as evidenced by reduced fluorescence intensity of total α-syn (Fig. 2a, b, and Supplementary Fig. 2). Then to evaluate downstream inflammatory responses, BV2 cells were treated with α-syn monomer, α-syn PFF alone or in combination with Ig-like binders, or the RAGE inhibitor FPS-ZM1. Lipopolysaccharide (LPS), an endotoxin known to induce neuroinflammation, was used as a positive control. Quantitative real-time PCR (qPCR) was used to measure the mRNA levels of three pro-inflammatory cytokines: interleukin-1β (IL-1β), IL-6, and tumor necrosis factor α (TNF-α)42,43. Results showed that α-syn PFF alone significantly promoted the expression of inflammatory cytokines. However, co-administration with either L3D1, dL3D1, vRAGE, or FPS-ZM129 was suggested to markedly suppress proinflammatory cytokine release induced by α-syn PFF (Fig. 2c–e).

Fig. 2: L3D1, dL3D1, and vRAGE decrease α-syn PFF binding to BV2 cells and prevent α-syn PFF-induced inflammation.

a Representative immunofluorescence images of PBS, 200 nM α-syn PFF, and the mixtures of 200 nM α-syn PFF-L3D1/dL3D1/vRAGE (1:1, molar ratio) binding to BV2 cells. DAPI (blue) and α-syn (red) were stained. Scale bar, 20 μm. b Statistical analyses of 200 nM α-syn PFF-L3D1/dL3D1/vRAGE mixtures binding to BV2 cell surfaces compared with 200 nM α-syn PFF binding to BV2 cell surfaces. The intensity of α-syn was normalized to the intensity of DAPI for each sample. 24 images were randomly taken for each group, derived from 3 biological replicates, with 8 images per replicate. One-way ANOVA followed by Tukey’s post-hoc test. Data shown are mean ± s.d. **p < 0.01, ***p < 0.001, ****p < 0.0001. The p value for PFF vs. PFF:L3D1 = 1:1 was 0.0007; the p value for PFF vs. PFF:dL3D1 was 0.002; other p values were all <0.0001. Statistical analyses of the mRNA fold change of IL-1β (c), IL-6 (d), and TNF-α (e). BV2 cells were treated with PBS, 100 ng/mL LPS, 5 μM α-syn monomer, 5 μM α-syn PFF, and 5 μM α-syn PFF-L3D1/dL3D1/vRAGE/FPS-ZM1(1:1, molar ratio) mixtures, respectively. LPS, an endotoxin known to induce neuroinflammation, served as a positive control, while FPS-ZM1, a RAGE inhibitor, was included as a negative control. Data are shown as mean ± s.d., n  =  4 independent samples. One-way ANOVA followed by Tukey’s post-hoc test. ns, no significance; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. In (c): p values for PBS vs. Mono, PBS vs. PFF:L3D1 = 1:1, and PBS vs. PFF:FPS-ZM1 = 1:1 were 0.3069, 0.0120, and 0.0416, respectively. In (d): p values for PBS vs. Mono and PBS vs. PFF:FPS-ZM1 = 1:1 were 0.0811 and 0.0277, respectively. In (e): p values for PBS vs. Mono and PBS vs. PFF:FPS-ZM1 = 1:1 were 0.2863 and 0.1655, respectively. Other p values were all <0.0001. Source data are provided as a Source Data file.

Full size image

In conclusion, our results demonstrate that these Ig-like binders including L3D1, dL3D1, and vRAGE, which directly interact with the C-terminal of α-syn fibrils, may serve as molecular tools to inhibit fibrils-induced neuronal α-syn aggregation and neuroinflammation.

Identification of additional Ig-like binders to inhibit α-syn fibrils-induced pathology

To identify and design additional protein binders that mimic the binding and inhibitory activities of L3D1 and vRAGE for α-syn fibrils, we used dL3D1 as a template for primary sequence and 3D structure analysis (Fig. 3). This search led us to select two candidates, CD4 D1 and CAR D1, which exhibited high structural homology to dL3D1 with Cα root-mean-square deviation (RMSD) values of 1.099 Å and 1.233 Å, respectively. Additionally, CD4 D1 revealed more positive surface charges compared to CAR D1. Given the importance of positive surface charge in binding to α-syn fibrils as indicated in previous studies14,21, we hypothesize that both candidates can bind to α-syn fibrils but with different affinities, with CD4 D1 potentially exhibiting stronger binding.

Fig. 3: Ig-like binders from Ig superfamily share similar structural properties.

a Ribbon diagrams (top) and electrostatic surfaces (bottom) of dL3D1 (left, modeled by Zhang et al.21), CD4 D1 (middle, PDB: 1WIP), and CAR D1 (right, PDB: 1EAJ) are shown. b Sequence identity and structural similarity of CD4 D1 and CAR D1 compared to dL3D1.

Full size image

We overexpressed and purified the CD4 D1 and CAR D1 proteins and examined their binding affinities to α-syn using the biolayer interferometry (BLI) assay. Remarkably, CD4 D1 binds α-syn PFF with an affinity of 9.7 nM, over 100 times higher than its affinity for α-syn monomer (KD (dissociation constant) = 1.13 μM) (Fig. 4a), similar to the interaction between L3D1 and α-syn21. In contrast, CAR D1 exhibits a weaker binding affinity to α-syn PFF (KD = 0.10 μM) and minimal binding to α-syn monomer (KD ≈ 200 μM) (Fig. 4b). We then visualized the direct binding of α-syn fibrils to these proteins using an immunogold transmission electron microscopy (TEM) assay. Purified His-tagged CD4 D1 and CAR D1 were incubated with α-syn fibrils. Nanogold particles binding to the His-tag revealed demonstrated that CD4 D1 and CAR D1 attach to α-syn fibrils, as evidenced by localized nanogold labeling (Fig. 4c).

Fig. 4: CD4 D1, CAR D1, and CAR D1_Mut preferentially bind α-syn PFF and prevent α-syn PFF-induced neuronal aggregation and neuroinflammation.

The binding kinetics of CD4 D1 (a) and CAR D1 (b) with α-syn monomer (left) and α-syn PFF (right) measured by BLI assay. The association and dissociation profiles were divided by a vertical dash line. CD4 D1 (a) and CAR D1 (b) were fixed to the sensors, and the 5 concentrations of α-syn monomer (left) and PFF (right) used are indicated, respectively. c Immunogold TEM images of α-syn fibrils alone (black box) and α-syn fibrils incubated with CD4 D1 (pink box) and CAR D1 (blue box). White arrows indicate the nanogolds on α-syn fibrils. The dashed boxes in each image are zoomed in and displayed on the right side of each image. Scale bar, 100 nm. The statistical analyses of neuron surface binding assay (d), fibrils-induced neuronal α-syn aggregation assay (e), and BV2 cell surface binding assay (f) of α-syn PFF-L3D1/dL3D1/vRAGE (1:1, molar ratio) mixtures compared to 200 nM α-syn PFF-treated group. The intensity of α-syn was normalized to the intensity of DAPI for each sample. 24 images were randomly taken for each group, derived from 3 biological replicates, with 8 images per replicate. One-way ANOVA followed by Tukey’s post-hoc test. Data shown are mean ± s.d. ns, no significance; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. In (d): p values for PBS vs. PFF:CD4 D1 = 1:1, PBS vs. PFF:CAR D1_Mut = 1:1 and PFF vs. PFF:CAR D1 = 1:1 were 0.0001, 0.0013, and 0.0193, respectively. In (e): p values for PFF vs. PFF:PFF:CD4 D1 = 1:1 and PFF vs. PFF:CAR D1 = 1:1 were 0.0014 and 0.9235, respectively. In (f): p values for PFF vs. PFF:CD4 D1 = 1:1, PFF vs. PFF:CAR D1 = 1:1, and PFF vs. PFF:CAR D1_Mut = 1:1 were 0.0015, 0.0169, and 0.0043, respectively. Other p values were all <0.0001. Statistical analyses of the mRNA fold change of IL-1β (g), IL-6 (h), and TNF-α (i). BV2 cells were treated with PBS, 100 ng/mL LPS, 5 μM α-syn monomer, 5 μM α-syn PFF, and 5 μM α-syn PFF-CD4 D1/CAR D1/CAR D1_Mut/FPS-ZM1 (1:1, molar ratio) mixtures, respectively. Data are shown as mean ± s.d., n  =  4 independent samples. One-way ANOVA followed by Tukey’s post-hoc test. ns, no significance; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. In (g): p values for PBS vs. Mono, PBS vs PFF:CD4 D1 = 1:1, PFF vs. PFF: CAR D1_Mut, and PBS vs. PFF:FPS-ZM1 = 1:1 were 0.5259, 0.0089, 0.0162, and 0.1365, respectively. In (h): p values for PBS vs. Mono, PBS vs. CAR D1_Mut = 1:1, and PBS vs. PFF:FPS-ZM1 = 1:1 were 0.0692, 0.0018, and 0.0226, respectively. In (i): p values for PBS vs. Mono, PBS vs. PFF:CAR D1 = 1:1, and PBS vs. PFF:FPS-ZM1 = 1:1 were 0.3020, 0.0003, and 0.1776, respectively. Other p values were all <0.0001. Source data are provided as a Source Data file.

Full size image

To further validate the specificity of these Ig-like binders for α-syn, we introduced K19, a fibril-forming fragment of Tau44,45, as a control. BLI assays revealed no detectable interaction between K19, in either its monomeric or fibrillar form, with these Ig-like binders including L3D1, vRAGE, CD4 D1, or CAR D1, confirming their selective binding to α-syn (Supplementary Fig. 3).

Collectively, these findings indicate that both CD4 D1 and CAR D1 exhibit a much higher binding affinity to α-syn fibrils compared to α-syn monomer, akin to the interactions observed between L3D1/vRAGE and α-syn, suggesting that CD4 D1 and CAR D1 may also inhibit fibrils-induced neuronal α-syn aggregation and inflammation.

Of note, although CD4 D1 and CAR D1 can bind to α-syn monomer, monomeric α-syn alone does not induce pS129 α-syn aggregation or neuroinflammation in our cellular models (Figs. 2c–e, 4g–i), nor has it been reported to do so in the literature40. Therefore, the observed inhibitory effects are unlikely to arise from interactions with the monomeric form.

We then examined whether the binding of CD4 D1 and CAR D1 to α-syn fibrils could alleviate neuronal α-syn aggregation and inflammation induction. CD4 D1 was indicated to inhibit both the binding of α-syn PFF to primary neuron membranes and α-syn PFF-induced pathological α-syn aggregation in primary neurons (Figs. 4d, e, Supplementary Figs. 1, 4, and 5). Additionally, CD4 D1 suggested protective effects against the adhesion of α-syn PFF onto BV2 cell membranes (Fig. 4f and Supplementary Figs. 2, 6), consequently downregulating the expression of pro-inflammatory cytokines IL-1β, IL-6, and TNFα (Fig. 4g–i). While CAR D1 exhibited inhibitory effects in all these assays, its performance was weaker than CD4 D1, consistent with its lower binding affinity for α-syn PFF. Notably, cell-based competitive binding assay results suggested CD4 D1 and CAR D1 significantly reduced the attachment of α-syn PFF to the cell surface in HEK293T cells overexpressing LAG3 or RAGE (Supplementary Figs. 7, 8). However, we note that these findings reflect cellular competition assays, and direct receptor-level binding evidence remains to be established.

In conclusion, our results indicate that the two identified Ig-like binders with structures similar to L3D1 reduce α-syn fibrils-induced neuronal aggregation and inflammation in cellular models, supporting their potential as molecular tools to interfere with α-syn pathology in vitro.

Structural characterization of the binding between Ig-like binders and α-syn

To identify the binding interface between α-syn and these Ig-like binders, we first conducted solution nuclear magnetic resonance (NMR) spectroscopy to identify the key regions of α-syn monomer that bind to CD4 D1 and CAR D1 (Fig. 5a–d). Titration of CD4 D1 with 15N-labeled α-syn monomer revealed significant chemical shift differences (CSDs > 0.01 ppm) in the residues clustered in the C-terminal region of α-syn, particularly residues 115–135, which are enriched with negatively charged residues such as D119 and E131 (Fig. 5a, c). This indicates that CD4 D1 interacts with the acidic C-terminal region of α-syn, similar to L3D1 and vRAGE21,31. A similar but much weaker pattern was observed for CD4 D1 (Fig. 5b, d), consistent with its lower binding affinity measured by BLI (Fig. 4a, b). These results indicate that both CD4 D1 and CAR D1 interact with the acidic C-terminal region of α-syn monomer, in line with previously characterized Ig-like binders such as L3D1 and vRAGE21,31. To determine whether this binding mode extends to α-syn fibrils, we further analyzed interactions using BLI and immunogold TEM assays. Deletion of the C-terminal region (construct referred to as α-syn100) completely abolished the binding of both α-syn monomer and PFF to CD4 D1 and CAR D1 (Fig. 5e, f), confirming the essential role of the C-terminal in mediating these interactions. To directly assess the sufficiency of the C-terminal region in fibril binding, we engineered a chimeric fibril-forming construct, K19–α-Syn101–140, by fusing the C-terminal residues of α-syn (101–140) to the fibril-forming Tau K19 fragment. This hybrid construct exhibited strong binding to all tested Ig-like binders, with KD comparable to those of α-syn PFF (Supplementary Fig. 9). Furthermore, immunogold TEM revealed selective binding of nanogold-labeled Ig-like binders to K19–α-Syn101–140 hybrid fibrils, but not to K19 or α-syn100 fibrils (Supplementary Fig. 10). Together, these results demonstrate that the C-terminal region of α-syn, particularly residues 101–140, serves as the principal binding site for Ig-like binders in both monomeric and fibrillar contexts.

Fig. 5: CD4 D1 and CAR D1 interact with the C-terminal of α-syn.

Overlaid 2D HSQC spectra of 15N-labeled α-syn monomer alone (black) and that titrated by CD4 D1 (a, left) and CAR D1 (b, left) at a molar ratio of 1:1 (α-syn:CD4 D1/CAR D1). Representative residues with significant CSDs were zoomed in on the right. c, d Calculated CSDs of 15N-labeled α-syn monomer titrated by CD4 D1 from (a) and by CAR D1 from (b). The black dashed lines indicate the residues with CSDs > 0.01 ppm (c) and 0.005 ppm (d), respectively. The binding kinetics of CD4 D1 (e) and CAR D1 (f) with α-syn100 monomer (left) and α-syn100 PFF (right) measured by BLI assay. The association and dissociation profiles were divided by a vertical dash. Ig-like binders were fixed to the sensors, and the 5 concentrations of α-syn100 monomer (left) and PFF (right) used are indicated. N.D., not detected. Source data are provided as a Source Data file.

Full size image

Given that the NMR data revealed the critical role of the C-terminal segment, including residues 115–135, for the binding of α-syn to CD4 D1 and CAR D1, we utilized AlphaFold 346 to model the structure of CD4 D1 and CAR D1 in complex with residues 115–135 of α-syn (α-syn115–135). As shown in Fig. 6a and Supplementary Fig. 11a, the acidic α-syn115–135 is well-accommodated on the basic surface of CD4 D1. To stabilize the complex structure, three intermolecular salt bridges are formed between basic residues from CD4 D1 (K60, R83, and R84) and acidic residues from α-syn115–135 (D119, E130, and E126) (Table 1). Additionally, four hydrogen bonds are observed between residues from CD4 D1 and α-syn115–135: K71 (main chain) with N122 (side chain) and E123 (main chain), and two hydrogen bonds between L69 (main chain) and Y125 (main chain). In contrast, only one salt bridge and three hydrogen bonds between CAR D1 and α-syn115–135 are observed in the complex structure of CAR D1-α-syn115–135 (Fig. 6b, Table 1, and Supplementary Fig. 11b). Consistently, the binding interface energy47,48,49 in the CD4 D1-α-syn115–135 complex is lower than that of the CAR D1-α-syn115–135 complex (Fig. 6c), indicating a stronger binding affinity of CD4 D1 to α-syn PFF, which aligns with the BLI measurements.

Fig. 6: α-Syn uses its acidic C-terminal to bind with CD4 D1, CAR D1, and CAR D1_Mut.

Electrostatic surface of the representative complex structure of α-syn115–135-CD4 D1 (a), α-syn115–135-CAR D1 (b), and α-syn115–135-CAR D1_Mut (d) which were predicted by AlphaFold 3. The complex interfaces were zoomed in (bottom). Salt bridges are highlighted in orange lines and hydrogen bonds are in dark gray lines. c Violin plots of ΔΔG (Gibbs free energy change) and ΔΔG/ΔSASA (ΔΔG per unit change in solvent accessible surface area) for three complexes, each calculated from 200 trials. (REU, Rosetta Energy Unit). One-way ANOVA followed by Tukey’s post-hoc test. Data shown are mean ± s.d. ****p < 0.0001. All p values were < 0.0001. e Binding kinetics of CAR D1_Mut with α-syn monomer (left) and α-syn PFF (right) measured by BLI assay. The association and dissociation profiles were divided by a vertical dash. Ig-like binders were fixed to the sensors, and the 5 concentrations of α-syn monomer (left) and PFF (right) used are indicated. Source data are provided as a Source Data file.

Full size image
Table 1 Salt bridges and hydrogen bonds (H bonds) between Ig-like binders and α-syn115-135
Full size table
Structure-based design to improve the Ig-like binder’s activity

Based on the complex structure model of CAR D1 and α-syn115–135, we aimed to enhance the binding and inhibitory capability of CAR D1 towards α-syn. Given the importance of electrostatic interactions in binding to the C-terminal acidic region of α-syn, our strategy was to increase the positive charge on the interface of CAR D1. We achieved this by introducing multiple basic residues and mutating out the acidic residues within the interface of CAR D1. To this end, we designed a mutated version of CAR D1, referred to as CAR D1_Mut, incorporating seven mutations at the interface: G51R, P52R, D54G, E56I, D68R, D81R, and D82R (Supplementary Fig. 12). Using AlphaFold 3, we obtained a complex structure model of CAR D1_Mut and α-syn115–135. As designed, additional intermolecular interactions were observed, inducing five salt bridges and seven hydrogen bounds between CAR D1_Mut and α-syn115–135 (Fig. 6d, Table 1 and Supplementary Fig. 11c). Consequently, the CAR D1_Mut-α-syn115–135 complex exhibits significantly lower interface energy compared to the CAR D1-α-syn complex (Fig. 6c), indicating enhanced binding and a more stable complex structure.

Remarkably, CAR D1_Mut demonstrated increased binding affinities to both α-syn PFF and monomer. Its binding affinity to α-syn PFF is 7.91 nM (Fig. 6e), which is 20-fold higher than that of CAR D1 and similar to that of CD4 D1. TEM images also revealed extensive nanogold particles on α-syn fibrils bound to CAR D1_Mut (Supplementary Fig. 13). Additionally, the binding affinity of CAR D1_Mut to α-syn monomer is 1.51 μM, approximately a 130-fold increase compared to that of CAR D1 (Fig. 6e). Moreover, both the BLI and TEM results indicated that the enhanced binding of CAR D1_Mut is indeed mediated by the C-terminal region of α-syn (Supplementary Fig. 9, 10). Furthermore, CAR D1_Mut exhibited significantly stronger inhibitory activities, preventing both the binding of α-syn PFF to primary neuron membranes and the α-syn PFF-induced pathological α-syn aggregation in primary neurons compared to CAR D1 (Fig. 4d, e, Supplementary Fig. 4, 5). CAR D1_Mut also effectively protected against the adhesion of α-syn PFF onto BV2 cell membranes (Fig. 4f, Supplementary Fig. 6) and consequently downregulated the expression of pro-inflammatory cytokines IL-1β, IL-6, and TNFα (Figs. 4g–i). Of note, CAR D1_Mut significantly reduced α-syn PFF attachment in HEK293T cells overexpressing LAG3 or RAGE (Supplementary Fig. 7, 8). However, we note that this assay measures total cellular association rather than receptor occupancy, and alternative explanations such as extracellular fibril sequestration or receptor internalization cannot be excluded. Overall, these results demonstrate the feasibility of a structure-based design approach in developing Ig-like binders to bind α-syn fibrils and prevent its pathological activities.

Discussion

The pivotal role of α-syn fibrils in the onset and progression of PD and other synucleinopathies has led to numerous strategies aimed at preventing pathological aggregation or eliminating pre-accumulated α-syn fibrils. Approaches have included small molecules50,51,52 and proteins such as chaperones53,54,55,56,57,58 and nanobodies59,60,61,62,63,64. In this study, we explored a complementary strategy: developing Ig-like binders that target α-syn fibrils and attenuate their pathological activities in cell-based models. Inspired by previously identified native receptors of α-syn fibrils, we discovered that the fibrils-binding Ig-like domains of these receptors alone can act as potent inhibitors, preventing α-syn fibrils pathology in primary neurons and microglia. Building on the structural mechanisms of interactions between α-syn fibrils and the Ig-like binders, we identified two additional Ig-like binders from CD4 and CAR. These isolated Ig-like binders capture the C-terminal region of α-syn fibrils and reduce their membrane association and downstream aggregation in cultured neurons and microglia. Our findings suppose a model that these Ig-like binders alone can competitively bind α-syn fibrils, preventing their attachment to cell membranes and blocking subsequent pathological processes in cell models (Fig. 7). Notably, our Ig-like binders, including L3D1 and vRAGE, showed binding affinities comparable to anti-α-syn monoclonal antibodies currently undergoing clinical trials, such as Prasinezumab (from Roche, phase II) and MEDI1341 (from AstraZeneca, phase I) (Supplementary Fig. 14), highlighting their potential as protein-based tools for targeting α-syn fibrils in future studies.

Fig. 7: Hypothetical working model of how Ig-like binders prevent α-syn fibrils binding to cell surface and reduce its pathological activities.

Ig-like binders, such as L3D1, vRAGE, CD4 D1, and CAR D1, specifically bind to the acidic C-terminal of α-syn, thus inhibiting α-syn fibrils from attaching to cell membranes and further blocking the following pathological processes, including neuronal α-syn aggregation and neuroinflammation induced by α-syn fibrils in cellular models. The diagram is created in BioRender. Liu, C. (2025) https://BioRender.com/fvndt7o.

Full size image

However, it remains to be determined whether these affinities can translate into comparable in vivo efficacy or therapeutic benefit. We acknowledge that the current findings are limited to biochemical and cell-based models. While these platforms are informative for mechanistic studies and offer an initial proof-of-concept, in vivo validation will be essential to assess therapeutic potential, bioavailability, and long-term effects. In particular, future efforts should include biophysical competition assays (e.g., BLI or SPR), as well as evaluation in more physiologically relevant systems, such as transgenic PD mouse models or α-syn PFF-injection paradigms. Furthermore, issues related to in vivo delivery, stability, and immunogenicity of the Ig-like binders remain to be addressed. AAV mediated delivery of Ig-like binders could be used to assess their ability to mitigate α-syn pathology in a physiologically relevant setting. This approach would not only test their effects on pathological aggregation and propagation, but also evaluate their impact on motor behavior and disease progression. Such in vivo investigations are critical next steps toward validating the therapeutic potential of our candidates and translating these findings into future clinical applications.

Among the Ig-like binders studied, vRAGE exhibited the highest binding affinity and most effective inhibitory action against α-syn fibrils at the cellular level, making it a promising candidate for further optimization and therapeutic development65. However, it is crucial to limit its binding to other targets to avoid potential side effects. Structural insights into how vRAGE recognizes α-syn and other clients are essential for this optimization. Moreover, as our current data do not directly demonstrate receptor-level competition, future studies should assess whether these binders can displace endogenous α-syn receptors on neurons or glia. Encouragingly, in the case of L3D1, we demonstrated that deleting the key loop responsible for MHC-II binding did not affect its ability to bind α-syn fibrils and inhibit their pathology41. This suggests that it is possible to separate the normal function of an Ig-like binder from its binding to α-syn fibrils21,31. Moreover, our structure-based design could significantly improve the binding affinity and inhibitory activity of CAR D1. Nonetheless, further accurate structure-based design efforts are needed to maximize the specific binding of α-syn fibrils and minimize off-target effects of the Ig-like binders.

Our work emphasizes the C-terminal region of α-syn as a key pathological epitope for targeting protein binders and inhibitors. Beyond Ig-like domain-based designs, other approaches, such as AI-driven de novo protein design using different scaffolds or compound screening targeting this key region66,67,68, may produce promising therapeutic candidates to eliminate α-syn fibril pathology, thereby offering molecular tools for studying α-syn fibril pathology and guiding therapeutic exploration.

Methods
Ethical statement

All animal experiments were performed according to the protocols approved by the Animal Care Committee of the Interdisciplinary Research Center on Biology and Chemistry (IRCBC), Chinese Academy of Sciences (CAS) (Approval No. IACUC-20230110001).

Protein purification of recombinant full-length α-syn, α-syn100, K19, and K19-α-Syn101-140

For full-length α-syn, protein expression and purification followed the protocol described previously69. Briefly, 1 mM isopropyl-1-thio-D-galactopyranoside (IPTG) was added to induce protein expression at 16 °C for 12 h. Then cells were harvested and lysed. Following boiling, streptomycin addition, pH adjustment to 3.5, and centrifugation, the supernatant was dialyzed (25 mM Tris-HCl, pH 8.0) overnight at 4 °C. Next, an anion exchange column, Q column (GE Healthcare, 17-5156-01), and a Superdex 75 column (GE Healthcare, 28-9893-33) were used for further purification. For 15N-labeled α-syn, the purification was the same as that for unlabeled α-syn, except that cells were grown in the absence of M9 minimal medium with 15NH4Cl (1 g/L).

For α-syn100, protein expression and purification followed the previous publication70. Briefly, after cell harvest and centrifugation, the supernatant was dialyzed (25 mM Na2HPO4-NaH2PO4, pH 6.0) overnight at 4 °C. Then, a cation exchange column, a SP column (GE Healthcare, 17-1152-01), and a Superdex 75 column (GE28-9893-33) were employed for further purification. Purified proteins were stored at −80 °C.

For K19, protein expression and purification followed the protocol described previously44,71. Briefly, K19 was purified using a HighTrap HP SP column followed by a Superdex 75 gel filtration column. The purified proteins were concentrated and stored at −80 °C.

For K19-α-Syn101-140, 1 mM IPTG was added to induce protein expression at 25 °C for 12 h. Then cells were harvested and lysed. Following boiling and centrifugation, the supernatant was dialyzed (25 mM Tris-HCl, pH 8.0) overnight at 4 °C. Next, an anion exchange column, Q column and a Superdex 75 column were used for further purification.

Expression, purification, and refolding of Ig-like binders

Protein expression and purification of Ig-like binders followed the protocol previously established21. In brief, 1 mM IPTG was added to induce expression at 25 °C for 12 h. After cell lysis (50 mM Tris, pH 8.0, 500 mM NaCl) and centrifugation, the pellets were sequentially resuspended and dissolved in three types of buffer (50 mM Tris, 5% Triton X-100, pH 8.0; 50 mM Tris, 1 M NaCl, pH 8.0; 6 M guanidine hydrochloride, 50 mM Tris, pH 8.0, 50 mM NaCl). Then, a Ni-NTA column (GE Healthcare) was used, and isolated Ig-like binders were eluted with the buffer (50 mM Na2HPO4, pH 4.0, 6 M guanidine hydrochloride).

Isolated Ig-like binders were performed refolding followed the previous publication21. In short, the isolated Ig-like binders were performed a three-step dialysis at 4 °C using three different buffers: 50 mM Na2HPO4-NaH2PO4, pH 7.0, 50 mM NaCl, 2 M Guanidine hydrochloride, 5% glycerin; 50 mM Na2HPO4-NaH2PO4, pH 7.0, 50 mM NaCl, 1 M Guanidine hydrochloride, 2.5% glycerin; 50 mM Na2HPO4-NaH2PO4, pH 7.0, 50 mM NaCl. Finally, a Superdex 75 column (GE Healthcare, 28-9893-33) was used for further purification. Purified Ig-like binders were stored at −80 °C.

Preparation of α-syn fibrils and PFF

200 μM full-length α-syn, α-syn100, K19, and K19-α-Syn101-140 monomer were incubated in fibril growth buffer (50 mM Tris-HCl, pH 7.5, 150 mM KCl) in a ThermoMixer (Eppendorf), shaking at 37 °C, 900 rpm for 5 d, after filtration with 0.22 μm strainers. Then the fibril samples were first characterized by negative-staining transmission electron microscopy (NS-TEM) and then sonicated into short PFF, following the parameter setting of 20% power and 40 cycles (1 s on, 1 s off for each cycle).

NS-TEM

5 μL fibrils/PFF solution was loaded onto a 200-mesh glow discharged copper grid (Zhongjingkeyi Technology Co., Ltd., Beijing) for 45 s. Then, the grid was washed with double-distilled water followed by 2% w/v uranyl acetate for another 45 s. At last, the grid was dried in air. The TEM images were acquired using a Tecnai T12 transmission electron microscope (FEI) operated in 120 kV.

Cell culture

For primary rat cortical neuron cultures, primary rat cortical neurons were prepared from the cortex of embryonic day (E) 16 to E18 Spragu-Dawley rat (Shanghai SIPPR BK laboratory Animals Ltd.) embryos as described previously38. Briefly, isolated rat cortical neurons were plated onto coverslips with a density of 150, 000 cells each well in 24-well plates.

For BV2 and HEK293T cell cultures, cells were cultured in the 5% CO2 cell culture incubator at 37 °C with DMEM medium (Gibco, 11995065) containing 10% fetal bovine serum (FBS) (Gibco, 10099-141) and 1% penicillin-streptomycin (PS) (Gibco, 15140122). BV2 cells were plated on coverslips in 24-well plates for cell surface binding assay and in 12-well plates for the examination of proinflammatory cytokines. HEK293T cells were plated on coverslips in 24-well plates for competitive binding assay.

Cell surface binding assay

Both primary cortical neurons and BV2 cells were treated with PBS, 200 nM α-syn PFF alone, and mixtures of 200 nM PFF-Ig-like binders (1:1, molar ratio) at room temperature (RT) for 2 h, generally following an established protocol22. Mixtures were prepared by incubating 200 nM PFF with 200 nM Ig-like binders in PBS for 10 min at RT before addition to the medium. Then cells were fixed with 4% paraformaldehyde (PFA) and blocked with 3% goat serum (GS) diluted in PBS. Next, cells were incubated with primary antibodies at 4 °C overnight and incubated with secondary antibodies for 1 h at RT. Then cells were incubated with 4′,6-diamidino-2-phenylindole (DAPI) (1:10000, Yeasen, 40728ES03) diluted in PBS for nucleus staining. Finally, for membrane staining, an additional 10-min incubation with DiI dye (1:1000, MCE, HY-D0083) was performed. After rinsing, coverslips were mounted on glass slides with the mounting medium (Prolong gold antifade reagent, Invitrogen, P36930). The primary antibodies used included anti-total α-syn monoclonal rabbit antibody (1:1000, Abcam, ab138501) and anti-MAP2 monoclonal chicken antibody (1:2500, Abcam, ab5392). The secondary antibodies used included goat anti-chicken Alexa Fluor 488 (1:1000, Invitrogen, A-11039) and goat anti rabbit Alexa Fluor 568 (1:1000, Invitrogen, A-11036).

Fibrils-induced neuronal α-syn aggregation assay

At 7 d in vitro (DIV) after primary neuron culture, PBS, 200 nM α-syn PFF alone, and mixtures of PFF-Ig-like binders (PFF: Ig-like binders = 1:1 (molar ratio) following a 10 min-co-incubation were added for 14 d in a manner consistent with previous work72,73,74. Mixtures were prepared by incubating 200 nM PFF with 200 nM Ig-like binders in PBS for 10 min at RT before addition to the medium. Similarly, coverslips were first fixed with 4% PFA and then blocked with 3% GS diluted in PBS. The following was the incubation with primary antibodies diluted in blocking solution at 4 °C overnight, followed by the incubation with secondary antibodies diluted in blocking solution for 1 h at RT. Lastly, DAPI (1:10000, Yeasen) was stained. After washing, coverslips were mounted on glass slides using the mounting medium. The first antibodies used included anti-phosphorylated S129 α-syn monoclonal rabbit antibody (1:1000, Abcam, ab51253) and anti-MAP2 monoclonal chicken antibody (1:2500, Abcam, ab5392). The secondary antibodies used included goat anti-chicken Alexa Fluor 488 (1:1000, Invitrogen, A-11039) and goat anti rabbit Alexa Fluor 568 (1:1000, Invitrogen, A-11036).

Cell-based competitive binding assay

The plasmids used for transient transfections in HEK293T cells: the gene of hRAGE was inserted into a pcDNA3.1(-) vector with a 3 x FLAG tag; the gene of hLAG3 was inserted into a pfastbac vector with a GFP tag.

Constructed plasmids were delivered to HEK293T cells with PolyJetTM reagent (SignaGen Laboratories, SL 100688). After 24 h, cell surface binding assay was performed in transfected HEK293T cells following the protocol in the Cell surface binding assay mentioned above.

The primary antibodies used included anti-total α-syn monoclonal rabbit antibody (1:1000, Abcam, ab138501), anti-GFP monoclonal mouse antibody (1:500, Abmart, M20004L), and anti-FLAG monoclonal mouse antibody (1:500, Proteintech, 66008-4). The secondary antibodies used included goat anti-mouse Alexa Fluor 488 (1:1000, Invitrogen, A-11029) and goat anti rabbit Alexa Fluor 568 (1:1000, Invitrogen, A-11036).

Solution NMR spectroscopy

All NMR experiments were performed in the NMR buffer (50 mM Na2HPO4-NaH2PO4, pH 7.0, 50 mM NaCl and 10% (v/v) D2O) at 298 K using an Agilent 800 MHz spectrometer equipped with a cryogenic probe (Agilent Technologies) using the VNMRJ (rev.3.2 A) software.

The Agilent standard pulse sequence was used to collect the two-dimensional (2D) 1H-15N HSQC spectra of 50 μM 15N-labeled α-syn alone as well as 50 μM 15N-labeled α-syn titrated by 50 μM CD4 D1/CAR D1. Backbone resonance assignment of α-syn was performed based on previously published data (BMRB ID: 16300) and has been experimentally validated previously19,21,31,70,75,76. All NMR data were analyzed with NMRPipe77 and Sparky78.

Quantitative polymerase chain reaction

BV2 cells were treated with PBS, 100 ng/mL LPS, 5 μM α-syn monomer alone, 5 μM α-syn PFF alone, and mixtures of PFF-Ig-like binders/FPS-ZM1 (MCE, 945714-67-0) ((PFF: Ig-like binders/FPS-ZM1 = 1:1(molar ratio), 10 min-co-incubation) for 7 h. Then BV2 cells were collected and performed RNA extraction following manufacturer’s instructions (ZYMO, R1055). Next, the reverse transcription reaction was performed using a kit (Vazyme, R333-01) and the quantitative polymerase chain reaction was performed using SYBR Green Master Mix (Bimake, B21202) with the Step One Plus PCR Systems (ThermoFisher). The PCR conditions of each cycle included: 95 °C for 10 min, 95 °C for 15 s, and 60 °C for 1 min, 40 cycles in total. The expression level of β-actin, IL-6, IL-1β, and TNF-α were examined. The fold induction of gene expression was calculated following the ΔΔs method79. β-Actin was taken as the reference gene for value normalization. Data were analyzed and graphed using GraphPad Prism 9, with mean ± s.d.

BLI assay

BLI assay was performed using the Octet RED 96 system (Sartorius) at 37 °C80,81. 200 μL α-syn/α-syn100/K19/K19-α-syn101-140 monomers/PFF or buffer (50 mM Na2HPO4-NaH2PO4, pH 7.0, 50 mM NaCl, 0.05% Tween-20) were plated in a 96-well black flat bottom plate (Greiner Bio-One, 655209). Streptavidin (SA) biosensors (Sartorius, 2112123611) were incubated with α-syn/α-syn100/K19/K19-α-syn101-140 monomer/PFF to eliminate the non-specific binding. Then biotinylated CD4 D1/CAR D1/CAR D1_Mut (10 μg/mL) were loaded onto the SA biosensors. Lastly, different concentrations of α-syn or α-syn100 monomer/PFF were incubated with SA biosensors for association step, followed by disassociation subsequently. The results were analyzed by ForteBio Data Analysis 9 and graphed by GraphPad Prism 9.

Enzyme-linked immunosorbent assay (ELISA)

The ELISA assay followed previous publications50,59. 100 μL/well, 50 μg/mL L3D1, vRAGE, Prasinezumab and MEDI1341 was coated in 96-well plates (Corning, 3590) at 4 °C overnight. After PBST (PBS containing 0.1% Tween-20) washing, plates were incubated with 5% milk diluted in PBST for blocking for 1 h at 4 °C at first. Then, after PBST washing, plates were incubated with α-syn PFF diluted in PBS, which were serially diluted at 4x from 2 μM in triplicates, for 1 h at 4 °C, Next, after PBST washing, plates were incubated with a rabbit-derived antibody against α-syn PFF (Abcam, 138501) for 1 h at 4 °C. Next, after PBST washing, plates were incubated with an HRP-conjugated, goat anti-rabbit antibody (Thermo, 31460) for 1 h at 4 °C. Finally, a TMB substrate reagent set (BD OptEIA, 555214) was used for coloration followed by 2 M HCl as a stop solution. Absorbance at the wavelength of 450 nm was read and recorded. Values are normalized and plotted against α-syn PFF concentrations. The curve is fitted using nonlinear regression to a binding saturation equation analyzed by GraphPad Prism 9. Prasinezumab and MEDI1341 were generously provided by Mabwell Therapeutics, Inc. (Shanghai). These antibodies were prepared in-house according to the sequences disclosed in the patent documents. Specifically, Prasinezumab is listed in the IMGT/mAb-DB under ID: 769, with patent number WO2017207739A1. MEDI1341’s patent is listed as US20180002411A1.

Immunogold TEM assay

Following the previously established immunogold TEM assay21,24,31. Ig-like binders were first mixed with α-syn/α-syn100/K19/K19-α-syn101-140 fibrils at a molar ratio of 1:1 (fibrils: Ig-like binders) for 10 min at RT. Then α-syn fibrils alone and the mixtures of fibrils-Ig-like binders were incubated with 5 nm Ni-NTA-Nanogold (1:100, Nanoprobes, Yaphank, NY) for 30 min at RT. Next, 5 μL samples were loaded onto the glow discharged grids for 45 s. After rinsing with buffer containing imidazole (50 mM Na2HPO4-NaH2PO4, pH 7.0, 50 mM NaCl containing 10 mM imidazole) for 1 min at RT, grids were rinsed with double-distilled water and 3% (w/v) uranyl acetate for 45 s, respectively. The grids were finally stained with 3% (w/v) uranyl acetate for 45 s again. TEM images were acquired by a Tecnai T12 microscope (FEI Company) at 120 kV.

Confocal imaging and image analysis

For cell surface binding assay and intraneuronal propagation assay, all fluorescent images were acquired by a laser scanning confocal microscope (SP8, Leica) using a 63× water immersion objective. The parameter settings of Z stack were: 10 stacks, 1 μm/stack for primary neuron; 10 stacks, 0.1 μm/stack for BV2 cells. Image J was used for batched analyses of images. The resolution of each image was 1024 × 1024. Background subtraction was performed after all images were projected in Z direction with max intensity. The immunofluorescent intensity of the mean total α-syn or pS129 α-syn signal intensity were normalized to DAPI intensity of their own control group.

For DiI staining, images were acquired by a laser scanning confocal microscope (SP8, Leica) using a 100× oil immersion objective. Z stack was not performed. The resolution of each image was 1024 × 1024.

For cell-based competitive binding assay in HEK293T cells overexpressing LAG3 or RAGE, images were acquired by a laser scanning confocal microscope (SP8, Leica) using a 63× water immersion objective. Z stack was not performed. The resolution of each image was 1024 × 1024.

For statistical analyses of cell surface binding assay, intraneuronal propagation assay and cell-based competitive binding assay, at least three independent biological repeats were performed for each group. Statistical analysis of total α-syn and pS129 α-syn fluorescent intensity were analyzed using One-way ANOVA followed by Tukey’s post-hoc test72,82,83. Data shown are mean ± s.d. The results were graphed by GraphPad Prism 9.

Model prediction and interface analysis

The modeling of CD4 D1-α-syn115-135, CAR D1-α-syn115-135, and CAR D1_Mut-α-syn115-135 complexes were performed using AlphaFold 346 server (https://alphafoldserver.com/) (seeds: auto). The top ranked model from the predicted results of each complex is chosen as the starting point for energy minimization using Rosetta Relax48,84,85 (ver 2021.07). Each starting point underwent 200 trails of relaxation, resulting in a refined output PDB file for each relaxation. The outputs were then processed by Rosetta Interface Analyzer86, which calculates the ΔG and ΔG/ΔSASA (ΔΔG per unit change in solvent accessible surface area), representing the total energy change of all interactions on the interface, and the density of interactions on the interface, respectively87. One-way ANOVA followed by Tukey’s post-hoc test. Data shown are mean ± s.d. The results of ΔG and ΔG/dΔSASA values were graphed by GraphPad Prism 9.

Detailed Rosetta scripts for interface analysis

###scripts for Relax###

#i/o

-s input.pdb

-nstruct 200

-out:file:silent output

#relax option

-ex1

-ex2

-use_input_sc

-flip_HNQ -no_optH false

-relax:cartesian

-score:weights ref2015_cart

###scripts for interfaceAnalyzer###

#$Rosettascripts/InterfaceAnalyzer.mpi.linuxgccrelease \

-in:file:silent input.silent \

-pack_separated true \

-out:file:score_only score.sc

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All data needed to evaluate the findings of this study are available in the article and supplementary files. The source data underlying Figs. 1b, d; 2b–e; 4a, b, d–i; 5a–f; 6c, e; Supplementary Figs. S3, 7–9, 13 are provided as a Source Data file with this paper. Source data are provided with this paper.

References

Spillantini, M. G. et al. Alpha-synuclein in Lewy bodies. Nature 388, 839–840 (1997).

ADS
 
CAS
 
PubMed
 
Google Scholar

Spillantini, M. G. & Goedert, M. The alpha-synucleinopathies: Parkinson’s disease, dementia with Lewy bodies, and multiple system atrophy. Ann. N.Y. Acad. Sci. 920, 16–27 (2000).

ADS
 
CAS
 
PubMed
 
Google Scholar

Braak, H., Sandmann-Keil, D., Gai, W. & Braak, E. Extensive axonal Lewy neurites in Parkinson’s disease: a novel pathological feature revealed by alpha-synuclein immunocytochemistry. Neurosci. Lett. 265, 67–69 (1999).

CAS
 
PubMed
 
Google Scholar

Kordower, J. H., Chu, Y., Hauser, R. A., Freeman, T. B. & Olanow, C. W. Lewy body-like pathology in long-term embryonic nigral transplants in Parkinson’s disease. Nat. Med. 14, 504–506 (2008).

CAS
 
PubMed
 
Google Scholar

Uemura, N., Uemura, M. T., Luk, K. C., Lee, V. M. & Trojanowski, J. Q. Cell-to-cell transmission of tau and α-synuclein. Trends Mol. Med. 26, 936–952 (2020).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Braak, H. et al. Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol. Aging 24, 197–211 (2003).

PubMed
 
Google Scholar

Guo, J. L. & Lee, V. M. Cell-to-cell transmission of pathogenic proteins in neurodegenerative diseases. Nat. Med. 20, 130–138 (2014).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Kim, S. et al. Transneuronal propagation of pathologic α-synuclein from the gut to the brain models Parkinson’s disease. Neuron 103, 627–641.e7 (2019).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Li, J. Y. et al. Lewy bodies in grafted neurons in subjects with Parkinson’s disease suggest host-to-graft disease propagation. Nat. Med. 14, 501–503 (2008).

CAS
 
PubMed
 
Google Scholar

Desplats, P. et al. Inclusion formation and neuronal cell death through neuron-to-neuron transmission of alpha-synuclein. Proc. Natl. Acad. Sci. USA 106, 13010–13015 (2009).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Peng, C., Trojanowski, J. Q. & Lee, V. M. Protein transmission in neurodegenerative disease. Nat. Rev. Neurol. 16, 199–212 (2020).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Peng, C. et al. Cellular milieu imparts distinct pathological α-synuclein strains in α-synucleinopathies. Nature 557, 558–563 (2018).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Soto, C. & Pritzkow, S. Protein misfolding, aggregation, and conformational strains in neurodegenerative diseases. Nat. Neurosci. 21, 1332–1340 (2018).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Long, H. et al. Wild-type α-synuclein inherits the structure and exacerbated neuropathology of E46K mutant fibril strain by cross-seeding. Proc. Natl. Acad. Sci. USA 118, e2012435118 (2021).

Damier, P., Hirsch, E. C., Agid, Y. & Graybiel, A. M. The substantia nigra of the human brain. II. Patterns of loss of dopamine-containing neurons in Parkinson’s disease. Brain 122, 1437–1448 (1999).

PubMed
 
Google Scholar

Béraud, D. et al. Microglial activation and antioxidant responses induced by the Parkinson’s disease protein α-synuclein. J. Neuroimmune Pharm. 8, 94–117 (2013).

Google Scholar

Zhang, W. et al. Aggregated alpha-synuclein activates microglia: a process leading to disease progression in Parkinson’s disease. FASEB J. 19, 533–542 (2005).

CAS
 
PubMed
 
Google Scholar

Scheiblich, H. et al. Microglia jointly degrade fibrillar alpha-synuclein cargo by distribution through tunneling nanotubes. Cell 184, 5089–5106.e21 (2021).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Wu, K. M. et al. Neuronal FAM171A2 mediates α-synuclein fibril uptake and drives Parkinson’s disease. Science 387, 892–900 (2025).

CAS
 
PubMed
 
Google Scholar

Mao, X. et al. Aplp1 interacts with Lag3 to facilitate transmission of pathologic α-synuclein. Nat. Commun. 15, 4663 (2024).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Zhang, S. et al. Mechanistic basis for receptor-mediated pathological α-synuclein fibril cell-to-cell transmission in Parkinson’s disease. Proc. Natl. Acad. Sci. USA 118, e2011196118 (2021).

Mao, X. et al. Pathological α-synuclein transmission initiated by binding lymphocyte-activation gene 3. Science 353, aah3374 (2016).

Gu, H. et al. Lymphocyte activation gene 3 (Lag3) contributes to α-synucleinopathy in α-synuclein transgenic mice. Front. Cell Neurosci. 15, 656426 (2021).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Zhang, S. et al. Conformational dynamics of an α-synuclein fibril upon receptor binding revealed by insensitive nuclei enhanced by polarization transfer-based solid-state nuclear magnetic resonance and cryo-electron microscopy. J. Am. Chem. Soc. 145, 4473–4484 (2023).

CAS
 
PubMed
 
Google Scholar

Chen, K. et al. LRP1 is a neuronal receptor for α-synuclein uptake and spread. Mol. Neurodegener. 17, 57 (2022).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Pan, C. & Peng, C. LRP1: a novel receptor for the transmission of pathological α-synuclein. Mol. Neurodegener. 17, 79 (2022).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Sui, Y. T., Bullock, K. M., Erickson, M. A., Zhang, J. & Banks, W. A. Alpha synuclein is transported into and out of the brain by the blood-brain barrier. Peptides 62, 197–202 (2014).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Dzamko, N. et al. Toll-like receptor 2 is increased in neurons in Parkinson’s disease brain and may contribute to alpha-synuclein pathology. Acta Neuropathol. 133, 303–319 (2017).

CAS
 
PubMed
 
Google Scholar

Gasparotto, J. et al. Targeted inhibition of RAGE in substantia nigra of rats blocks 6-OHDA-induced dopaminergic denervation. Sci. Rep. 7, 8795 (2017).

ADS
 
PubMed
 
PubMed Central
 
Google Scholar

Jiang, X., Wang, X., Tuo, M., Ma, J. & Xie, A. RAGE and its emerging role in the pathogenesis of Parkinson’s disease. Neurosci. Lett. 672, 65–69 (2018).

CAS
 
PubMed
 
Google Scholar

Long, H. et al. Interaction of RAGE with α-synuclein fibrils mediates inflammatory response of microglia. Cell Rep. 40, 111401 (2022).

CAS
 
PubMed
 
Google Scholar

Garretti, F. et al. Interaction of an α-synuclein epitope with HLA-DRB1(∗)15:01 triggers enteric features in mice reminiscent of prodromal Parkinson’s disease. Neuron 111, 3397–3413.e5 (2023).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Gao, Y. et al. CD4(+) T-cell senescence in neurodegenerative disease: pathogenesis and potential therapeutic targets. Cells 13, 749 (2024).

Weiss, F., Labrador-Garrido, A., Dzamko, N. & Halliday, G. Immune responses in the Parkinson’s disease brain. Neurobiol. Dis. 168, 105700 (2022).

CAS
 
PubMed
 
Google Scholar

Golubovskaya, V. & Wu, L. Different subsets of T cells, memory, effector functions, and CAR-T immunotherapy. Cancers 8, 36 (2016).

Haghikia, A., Schett, G. & Mougiakakos, D. B cell-targeting chimeric antigen receptor T cells as an emerging therapy in neuroimmunological diseases. Lancet Neurol. 23, 615–624 (2024).

CAS
 
PubMed
 
Google Scholar

Kim, A.B. et al. Chimeric antigen receptor macrophages target and resorb amyloid plaques. JCI Insight 9, https://doi.org/10.1172/jci.insight.175015 (2024).

Volpicelli-Daley, L. A., Luk, K. C. & Lee, V. M. Addition of exogenous α-synuclein preformed fibrils to primary neuronal cultures to seed recruitment of endogenous α-synuclein to Lewy body and Lewy neurite-like aggregates. Nat. Protoc. 9, 2135–2146 (2014).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Fujiwara, H. et al. alpha-Synuclein is phosphorylated in synucleinopathy lesions. Nat. Cell Biol. 4, 160–164 (2002).

CAS
 
PubMed
 
Google Scholar

Luk, K. C. et al. Exogenous alpha-synuclein fibrils seed the formation of Lewy body-like intracellular inclusions in cultured cells. Proc. Natl. Acad. Sci. USA 106, 20051–20056 (2009).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Huard, B. et al. Characterization of the major histocompatibility complex class II binding site on LAG-3 protein. Proc. Natl. Acad. Sci. USA 94, 5744–5749 (1997).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Knott, C., Stern, G. & Wilkin, G. P. Inflammatory regulators in Parkinson’s disease: iNOS, lipocortin-1, and cyclooxygenases-1 and -2. Mol. Cell Neurosci. 16, 724–739 (2000).

CAS
 
PubMed
 
Google Scholar

Nagatsu, T., Mogi, M., Ichinose, H. & Togari, A. Changes in cytokines and neurotrophins in Parkinson’s disease. in Advances in Research on Neurodegeneration (eds Riederer, P. et al.) 277–290 (Springer, 2000).

Lu, J. et al. Structural basis of the interplay between α-synuclein and Tau in regulating pathological amyloid aggregation. J. Biol. Chem. 295, 7470–7480 (2020).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

von Bergen, M. et al. Assembly of tau protein into Alzheimer paired helical filaments depends on a local sequence motif ((306)VQIVYK(311)) forming beta structure. Proc. Natl. Acad. Sci. USA 97, 5129–5134 (2000).

ADS
 
Google Scholar

Abramson, J. et al. Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 630, 493–500 (2024).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Kuhlman, B. & Baker, D. Native protein sequences are close to optimal for their structures. Proc. Natl. Acad. Sci. USA 97, 10383–10388 (2000).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Rohl, C. A., Strauss, C. E., Misura, K. M. & Baker, D. Protein structure prediction using Rosetta. Methods Enzymol. 383, 66–93 (2004).

CAS
 
PubMed
 
Google Scholar

Alford, R. F. et al. The Rosetta all-atom energy function for macromolecular modeling and design. J. Chem. Theory Comput. 13, 3031–3048 (2017).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Tao, Y. et al. Structural mechanism for specific binding of chemical compounds to amyloid fibrils. Nat. Chem. Biol. 19, 1235–1245 (2023).

CAS
 
PubMed
 
Google Scholar

Horne, R. I. et al. Discovery of potent inhibitors of α-synuclein aggregation using structure-based iterative learning. Nat. Chem. Biol. 20, 634–645 (2024).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Murray, K. A. et al. Small molecules disaggregate alpha-synuclein and prevent seeding from patient brain-derived fibrils. Proc. Natl. Acad. Sci. USA 120, e2217835120 (2023).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

St Martin, J. L. et al. Dopaminergic neuron loss and up-regulation of chaperone protein mRNA induced by targeted over-expression of alpha-synuclein in mouse substantia nigra. J. Neurochem. 100, 1449–1457 (2007).

Google Scholar

Wentink, A. S. et al. Molecular dissection of amyloid disaggregation by human HSP70. Nature 587, 483–488 (2020).

ADS
 
CAS
 
PubMed
 
Google Scholar

Falsone, S. F., Kungl, A. J., Rek, A., Cappai, R. & Zangger, K. The molecular chaperone Hsp90 modulates intermediate steps of amyloid assembly of the Parkinson-related protein alpha-synuclein. J. Biol. Chem. 284, 31190–31199 (2009).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Jia, C. et al. Different heat shock proteins bind α-synuclein with distinct mechanisms and synergistically prevent its amyloid aggregation. Front. Neurosci. 13, 1124 (2019).

PubMed
 
PubMed Central
 
Google Scholar

Wang, C. et al. VAMP2 chaperones α-synuclein in synaptic vesicle co-condensates. Nat. Cell Biol. 26, 1287–1295 (2024).

Gao, X. et al. Human Hsp70 disaggregase reverses Parkinson’s-linked α-synuclein amyloid fibrils. Mol. Cell 59, 781–793 (2015).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Butler, Y. R. et al. α-Synuclein fibril-specific nanobody reduces prion-like α-synuclein spreading in mice. Nat. Commun. 13, 4060 (2022).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Jiang, Y. et al. Single-domain antibody-based noninvasive in vivo imaging of α-synuclein or tau pathology. Sci. Adv. 9, eadf3775 (2023).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Lynch, S. M., Zhou, C. & Messer, A. An scFv intrabody against the nonamyloid component of alpha-synuclein reduces intracellular aggregation and toxicity. J. Mol. Biol. 377, 136–147 (2008).

CAS
 
PubMed
 
Google Scholar

Hmila, I. et al. Novel engineered nanobodies specific for N-terminal region of alpha-synuclein recognize Lewy-body pathology and inhibit in-vitro seeded aggregation and toxicity. FEBS J. 289, 4657–4673 (2022).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Guilliams, T. et al. Nanobodies raised against monomeric α-synuclein distinguish between fibrils at different maturation stages. J. Mol. Biol. 425, 2397–2411 (2013).

CAS
 
PubMed
 
Google Scholar

Iljina, M. et al. Nanobodies raised against monomeric ɑ-synuclein inhibit fibril formation and destabilize toxic oligomeric species. BMC Biol. 15, 57 (2017).

PubMed
 
PubMed Central
 
Google Scholar

Kare, P., Bhat, J. & Sobhia, M. E. Structure-based design and analysis of MAO-B inhibitors for Parkinson’s disease: using in silico approaches. Mol. Divers 17, 111–122 (2013).

CAS
 
PubMed
 
Google Scholar

Tang, X. et al. A survey of generative AI for de novo drug design: new frontiers in molecule and protein generation. Brief Bioinform. 25, bbae338 (2024).

Jones, C. A. H. et al. Computer-aided design and biological evaluation of diazaspirocyclic D(4)R antagonists. ACS Chem. Neurosci. 15, 2396–2407 (2024).

CAS
 
PubMed
 
Google Scholar

Andrianov, A. M., Shuldau, M. A., Furs, K. V., Yushkevich, A. M. & Tuzikov, A. V. AI-driven de novo design and molecular modeling for discovery of small-molecule compounds as potential drug candidates targeting SARS-CoV-2 main protease. Int. J. Mol. Sci. 24, 8083 (2023).

Li, Y. et al. Amyloid fibril structure of α-synuclein determined by cryo-electron microscopy. Cell Res 28, 897–903 (2018).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Xu, Q. et al. α-Synuclein amyloid fibril directly binds to LC3B and suppresses SQSTM1/p62-mediated selective autophagy. Cell Res 35, 72–75 (2025).

CAS
 
PubMed
 
Google Scholar

Barghorn, S., Biernat, J. & Mandelkow, E. Purification of recombinant tau protein and preparation of Alzheimer-paired helical filaments in vitro. Methods Mol. Biol. 299, 35–51 (2005).

CAS
 
PubMed
 
Google Scholar

Sun, Y. et al. The hereditary mutation G51D unlocks a distinct fibril strain transmissible to wild-type α-synuclein. Nat. Commun. 12, 6252 (2021).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Zhao, K. et al. Parkinson’s disease-related phosphorylation at Tyr39 rearranges α-synuclein amyloid fibril structure revealed by cryo-EM. Proc. Natl. Acad. Sci. USA 117, 20305–20315 (2020).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Mahul-Mellier, A. L. et al. The process of Lewy body formation, rather than simply α-synuclein fibrillization, is one of the major drivers of neurodegeneration. Proc. Natl. Acad. Sci. USA 117, 4971–4982 (2020).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Rao, J. N., Kim, Y. E., Park, L. S. & Ulmer, T. S. Effect of pseudorepeat rearrangement on alpha-synuclein misfolding, vesicle binding, and micelle binding. J. Mol. Biol. 390, 516–529 (2009).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Huang, C. et al. The mouse nicotinamide mononucleotide adenylyltransferase chaperones diverse pathological amyloid client proteins. J. Biol. Chem. 298, 101912 (2022).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Delaglio, F. et al. NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR 6, 277–293 (1995).

CAS
 
PubMed
 
Google Scholar

Lee, W., Tonelli, M. & Markley, J. L. NMRFAM-SPARKY: enhanced software for biomolecular NMR spectroscopy. Bioinformatics 31, 1325–1327 (2015).

PubMed
 
Google Scholar

Pfaffl, M. W. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 29, e45 (2001).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Bates, T. A. et al. Biolayer interferometry for measuring the kinetics of protein-protein interactions and nanobody binding. Nat. Protoc. 20, 861–883(2024).

Frenzel, D. & Willbold, D. Kinetic titration series with biolayer interferometry. PLoS ONE 9, e106882 (2014).

ADS
 
PubMed
 
PubMed Central
 
Google Scholar

Hu, J. et al. Phosphorylation and O-GlcNAcylation at the same α-synuclein site generate distinct fibril structures. Nat. Commun. 15, 2677 (2024).

ADS
 
CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Long, H., Zeng, S. & Li, D. Cellular and animal models to investigate pathogenesis of amyloid aggregation in neurodegenerative diseases. Biophys. Rep. 8, 14–28 (2022).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Tyka, M. D. et al. Alternate states of proteins revealed by detailed energy landscape mapping. J. Mol. Biol. 405, 607–618 (2011).

CAS
 
PubMed
 
Google Scholar

Conway, P., Tyka, M. D., DiMaio, F., Konerding, D. E. & Baker, D. Relaxation of backbone bond geometry improves protein energy landscape modeling. Protein Sci. 23, 47–55 (2014).

CAS
 
PubMed
 
Google Scholar

Stranges, P. B. & Kuhlman, B. A comparison of successful and failed protein interface designs highlights the challenges of designing buried hydrogen bonds. Protein Sci. 22, 74–82 (2013).

CAS
 
PubMed
 
Google Scholar

Goudy, O. J., Nallathambi, A., Kinjo, T., Randolph, N. Z. & Kuhlman, B. In silico evolution of autoinhibitory domains for a PD-L1 antagonist using deep learning models. Proc. Natl. Acad. Sci. USA 120, e2307371120 (2023).

CAS
 
PubMed
 
PubMed Central
 
Google Scholar

Download references

Acknowledgements

We thank Dr. Zhijun Liu and Dr. Hongjuan Xue of the Nuclear Magnetic Resonance System (https://cstr.cn/31129.02.NFPS.NMRSystem) at the National Facility for Protein Science in Shanghai (https://cstr.cn/31129.02.NFPS), for providing technical support and assistance in data collection and analysis. We thank the staff members of the Large-scale Protein Preparation System at the National Facility for Protein Science in Shanghai (NFPS), Shanghai Advanced Research Institute, Chinese Academy of Sciences, China for providing technical support and assistance in BLI data collection and analysis. We thank the Cryo-Electron Microscopy center at Interdisciplinary Research Center on Biology and Chemistry, Shanghai Institute of Organic Chemistry for help with cryo-EM data collection. This work was supported by the National Natural Science Foundation (NSF) of China (32494764, 92353302, and 32170683 to D.L.; 22477130 to S.Z.; 82188101 and 22425704 to C.L.), the Science and Technology Commission of Shanghai Municipality (STCSM) (Grant No. 22JC1410400 to C.L.), Shanghai Oriental Talents Program (Excellent Academic Leader) to S.Z.; Shanghai Basic Research Pioneer Project and Shanghai Municipal Science and Technology Major Project to C.L., the Shanghai Pilot Program for Basic Research—Chinese Academy of Science, Shanghai Branch (Grant No. JCYJ-SHFY-2022-005 to C.L.), the CAS Project for Young Scientists in Basic Research (Grant No.YSBR-095 to C.L.), the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDB1060000 to C.L.), the National Key R&D Program of China (Grant No. 2020YFA0909200 to Z.W.), the National Natural Science Foundation of China (Grant No. 32101181 to Z.W.) and the Shanghai Key Laboratory of Aging Studies (Grant No. 19DZ2260400 to C.L.). Cong Liu is a SANS Exploration Scholar.

Author information
Authors and Affiliations

Bio-X Institutes, Key Laboratory for the Genetics of Developmental and Neuropsychiatric Disorders (Ministry of Education), Shanghai Jiao Tong University, Shanghai, China

Shuyi Zeng & Dan Li

Zhangjiang Institute for Advanced Study, Shanghai Jiao Tong University, Shanghai, China

Shuyi Zeng & Dan Li

School of Life Science and Technology, ShanghaiTech University, Shanghai, China

Xingyu Xiong, Bo Sun, Zhizhi Wang & Wenqing Xu

Interdisciplinary Research Center on Biology and Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, Shanghai, China

Houfang Long, Qianhui Xu, Cong Liu & Shengnan Zhang

University of the Chinese Academy of Sciences, Beijing, China

Houfang Long & Qianhui Xu

Shanghai Starriver Bilingual School, Shanghai, China

Yifan Yu

State Key Laboratory of Chemical Biology, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, Shanghai, China

Cong Liu & Shengnan Zhang

Shanghai Academy of Natural Sciences (SANS), Fudan University, Shanghai, China

Cong Liu

Shanghai Key Laboratory of Aging Studies, Shanghai, China

Cong Liu

Contributions

D.L. designed the project. S.Z. and H.L. purified proteins and prepared PFF used in this study. S.Z. and B.S. performed the neuron surface binding assay, neuron propagation assay, BV2 cell surface binding assay and qPCR assay. S.Z. and H.L. performed the BLI assay. S.Z. and S.Z. performed the NMR spectroscopy and analyzed the NMR data. S.Z. and Q.X. performed cell-based competitive binding assay. Y.Y., X.X., W.X. and Z.W. performed AlphaFold prediction and analyzed the computational calculations. All the authors are involved in analyzing the statistical data and contributed to manuscript editing. S.Z., S.Z., X.X., C.L. and D.L. wrote the manuscript.

Corresponding author

Correspondence to Dan Li.

Ethics declarations
Competing interests

The authors declare no competing interests.

Peer review
Peer review information

Nature Communications thanks George D. Mellick and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information
Supplementary Information
Reporting Summary
Transparent Peer Review file
Source data
Source Data
Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article
Cite this article

Zeng, S., Xiong, X., Long, H. et al. Design of Ig-like binders targeting α-synuclein fibril for mitigating its pathological activities. Nat Commun 16, 7368 (2025). https://doi.org/10.1038/s41467-025-62755-1

Download citation

Received
19 September 2024

Accepted
29 July 2025

Published
09 August 2025

DOI
https://doi.org/10.1038/s41467-025-62755-1

Share this article

Anyone you share the following link with will be able to read this content:

Get shareable link

Provided by the Springer Nature SharedIt content-sharing initiative

Subjects
Cellular imaging
Cellular neuroscience
Parkinson's disease
Protein aggregation

Cancel

Preview · live