Single-domain antibody-based protein degrader for synucleinopathies

Abstract

Synucleinopathies are a group of neurodegenerative diseases characterized by the accumulation of α-synuclein (α-syn) in the brain, leading to motor and neuropsychiatric symptoms. Currently, there are no known cures for synucleinopathies, and treatments mainly focus on symptom management. In this study, we developed a single-domain antibody (sdAb)-based protein degrader with features designed to enhance proteasomal degradation of α-syn. This sdAb derivative targets both α-syn and Cereblon (CRBN), a substrate-receptor for the E3-ubiquitin ligase CRL4^CRBN, and thereby induces α-syn ubiquitination and proteasomal degradation. Our results indicate that this therapeutic candidate enhances proteasomal degradation of α-syn, in addition to the endogenous lysosomal degradation machinery. By promoting proteasomal degradation of α-syn, we improved clearance of α-syn in primary culture and mouse models of synucleinopathy. These findings indicate that our sdAb-based protein degrader is a promising therapeutic candidate for synucleinopathies. Considering that only a small percentage of antibodies enter the brain, more potent sdAbs with greater brain entry than whole antibodies could enhance clinical benefits of antibody-based therapies.

Fig. 1

Design and working model for a single-domain antibody (sdAb)-based protein degrader targeting synucleinopathies. The degrader molecule is composed of anti-α-syn sdAb 2D8 connected via a variable-size linker to thalidomide, an E3-ligase recruiting ligand that engages with CRL4^CRBN. This degrader molecule can enter the mouse brain after i.v. injection and target α-syn both extracellularly and intracellularly. Specifically, extracellularly, the sdAb may sequester α-syn aggregates and disrupt their assembly, and thereby collectively prevent the spread of α-syn pathology between neurons. Intracellularly, the protein degrader may bind to α-syn aggregates within the endosomal-lysosomal system and facilitate their disassembly, which would result in better access of lysosomal enzymes to degrade the aggregates. The degrader molecule also brings α-syn and E3 ligase into proximity to form a ternary complex, which is predicted to mediate α-syn ubiquitination and trigger its degradation by the proteasome. This working model provides a framework for the development of protein degraders for the treatment of synucleinopathies. Abbreviations: α-syn = α-synuclein, E3 = E3 ligase, E2 = E2 ligase, Ub = ubiquitin

Fig. 2

Affinities of unmodified sdAb 2D8 and modified sdAb 2D8-PEGn-T for different α-syn preparations. (A) Summary table of binding affinities: To conduct biolayer interferometry assay in solution phase, unmodified sdAb 2D8 and modified sdAbs 2D8-PEGn-T (n = 2, 4, 6) were loaded onto the Ni-NTA biosensor, and K_D values were determined using increasing concentration of the different α-syn preparations, including rec α-syn and soluble fraction (S1) from LBD brain. (B-I) The representative curves illustrate the wavelength shift of interference in nanometers (nm), which is interpreted as binding. The curves demonstrate the association and dissociation of sdAb and the α-syn preparations at various concentrations, with the broken line indicating the fitting curve used to calculate the K_D value ± standard deviation (SD). The K_D value was obtained from three independent experiments. Additional information on association (k_a) and dissociation (k_d) values can be found in Supplemental Table 1

Fig. 3

Neuroprotective effects of unmodified sdAb 2D8 and modified sdAb 2D8-PEGn-T. (A) M83 primary neurons were incubated with 10 µg/ml LBD S1 fraction and 5 µg/ml of either 2D8 or its molar equivalent of 2D8-PEGn-T (n = 2, 4, 6) simultaneously for 24 h. After media replacement, cells were cultured for an additional 96 h, with samples collected for analysis (n = at least 12 replicates per condition). (B) One-way ANOVA analysis revealed significant group differences in GAPDH levels (p < 0.0001). S1 alone decreased GAPDH levels by 43% compared to control at 96 h (p = 0.0003). Both modified 2D8 and modified 2D8-PEGn-T (n = 2, 4, 6) prevented S1-induced toxicity, increasing GAPDH levels by 42% (p = 0.0005, 2D8), 43% (p = 0.0003, 2D8-PEG2-T), 52% (p < 0.0001, 2D8-PEG4-T), and 39% (p = 0.0018, 2D8-PEG6-T) relative to S1 alone, respectively. (C) One-way ANOVA analysis revealed significant group differences in NeuN levels (p < 0.0001). S1 alone decreased NeuN levels by 54% compared to control at 96 h (p < 0.0001). Both unmodified 2D8 and modified 2D8-PEGn-T (n = 2, 4, 6) prevented S1-induced toxicity, resulting in increased GAPDH levels by 50% (p < 0.0001, 2D8), 45% (p = 0.0017, 2D8-PEG2-T), 52% (p < 0.0001, 2D8-PEG4-T), and 52% (p < 0.0001, 2D8-PEG6-T) relative to S1 alone, respectively. (D) One-way ANOVA analysis revealed significant group differences in total α-syn/GAPDH levels (p < 0.0001). S1 alone increased α-syn levels by 39% compared to control at 96 h (p = 0.0002). Both sdAb 2D8 and modified sdAbs 2D8-PEGn-T (n = 2, 4, 6) prevented α-syn seeding, resulting in decreased total α-syn/GAPDH levels by 35% (p = 0.0008, 2D8), 45% (p < 0.0001, 2D8-PEG2-T), 48% (p < 0.0001, 2D8-PEG4-T), and 47% (p < 0.0001, 2D8-PEG6-T) relative to S1 alone, respectively. (E) Immunoblots illustrating total α-syn, GAPDH, and NeuN levels in treated M83 cell lysate. The complete western blots and bands analyzed are shown in Supplemental Fig. 5. **, ***, ****: p < 0.01, 0.001, 0.0001 (Tukey’s post-hoc test)

Fig. 4

Enhanced intracellular clearance of α-syn by modified sdAb 2D8-PEG4-T in primary neurons. (A) M83 primary neurons were incubated with 10 µg/ml LBD S1 fraction for 24 h. Subsequently, different concentrations of sdAbs ranging from 0.625–10 µg/ml of 2D8 and molar equivalent of modified 2D8 were added, with samples collected 96 h post-antibody application. (B) Schematic diagram of PROTAC-mediated ternary complex formation and Hook effect based on PROTAC concentration. (C) One-way ANOVA analysis revealed significant differences in normalized total α-syn levels when sdAb concentration exceeded 0.625 µg/ml (p = 0.02 for 1.25 µg/ml, p = 0.0004 for 2.5 µg/ml, p = 0.0001 for 5.0 µg/ml, p = 0.0136 for 10.0 µg/ml). At 1.25–5.0 µg/ml sdAb concentration, S1 alone increased α-syn levels by 46% compared to control at 96 h (p = 0.0154 (1.25 µg/ml), p = 0.0049 (2.5 µg/ml), p = 0.0043 (5.0 µg/ml)). At 2.5–5.0 µg/ml, 2D8-PEG4-T reduced S1-induced increase in total α-syn levels (reduced by 61% at 2.5 µg/ml, p = 0.0001, and 67% at 5.0 µg/ml, p < 0.0001, relative to S1 alone). Modified sdAb 2D8-PEG4-T also lowered total α-syn levels compared to unmodified 2D8 at 2.5 and 5.0 µg/ml (reduced by 38% at 2.5 µg/ml, p = 0.0298, and 41% at 5.0 µg/ml, p = 0.0126, relative to unmodified 2D8). (D) Immunoblots depicting total α-syn and GAPDH levels in treated M83 cell lysate. Complete western blots and bands analyzed are in Supplemental Fig. 6. (E) Immunofluorescence of M83 primary mouse neurons treated with 10 µg/ml LBD S1 fraction for 24 h. Media was then exchanged, and unmodified sdAb 2D8 or modified sdAb 2D8-PEG4-T at 5.0 µg/ml or molar equivalent were added and incubated for an additional 2 h. Neurons were subsequently immuno-probed for sdAb containing His tag (HIS.H8, green) and total α-syn (211, red). Scale bar: 20 μm. Mander’s coefficient M2 quantifies the ratio of α-syn overlapping with sdAb. Calculated as the sum of overlapping intensity in α-syn divided by total intensity of α-syn. *, **, ***, ****: p < 0.05, 0.01, 0.001, 0.0001 (Tukey’s post-hoc test)

Fig. 5

α-Syn clearance by unmodified and modified sdAbs via lysosome and proteasome pathways. (A) To determine the degradation mechanism of 2D8 and 2D8-PEG4-T, the lysosome and proteasome pathways were blocked using bafilomycin A1 (Baf.A1) and MG132 inhibitors, respectively. (B) M83 primary neurons were incubated with 10 µg/ml LBD S1 fraction for 24 h to allow uptake of pathological α-syn. The media was then replaced, and the neurons were pre-treated with either the Baf.A1 or MG132 for 6 h. Subsequently, the neurons were treated with 5 µg/ml 2D8 or molar equivalent of modified 2D8 for 2 days. (C-F) One-way ANOVA analysis revealed group differences in normalized total α-syn levels (p = 0.0007 (C), p < 0.0001 (D), p = 0.0060 (E), and p = 0.0002 (F)), N = 3. For the Baf.A1 inhibitor (C-D), S1 alone increased α-syn levels by 59% compared to control (p = 0.0013 (C), p < 0.0001 (D)). Both 2D8 and 2D8-PEG4-T prevented S1-induced increase in total α-syn levels (reduced by 40% for 2D8, p = 0.0194 (C); and 83% for 2D8-PEG4-T, p < 0.0001 (D)). However, the Baf.A1 inhibitor blocked this degradation effect, leading to a 34% increase for 2D8 (p = 0.0494) with 0.5 µM of inhibitor and a 22% increase for 2D8-PEG4-T (p = 0.0054) with 1 µM of inhibitor (C-D). For the MG132 inhibitor (E-F), S1 alone increased α-syn levels by 53% compared to control (p = 0.0077 (E), p = 0.0018 for (F)). Both 2D8 and 2D8-PEG4-T prevented the S1-induced increase in total α-syn levels (reduced by 40% for 2D8, p = 0.0434 (E), and 76% for 2D8-PEG4-T, p < 0.0001 (F)). The MG132 inhibitor did not block α-syn clearance of 2D8 (E), but it blocked α-syn clearance of 2D8-PEG4-T in a concentration-dependent manner (F), resulting in a 32% increase and 48% increase in α-syn with 5 µM (p = 0.0437) and 10 µM MG132 (p = 0.0040) for 2D8-PEG4-T. (G-H) Immunoblots show total α-syn and GAPDH levels for C-F. Complete western blot analysis are in Supplemental Fig. 7. *, **, ****: p < 0.05, 0.01, 0.0001 (Tukey’s post-hoc test)

Fig. 6

Ternary complex formation and sdAb-mediated effects on pathological α-syn in M83 primary neurons. (A) Co-immunoprecipitation (co-IP) and western blot analysis demonstrate the formation of ternary complexes in M83 primary neurons upon sdAb treatment. (B) Experimental procedure: M83 primary neurons were treated with 10 µg/ml LBD brain soluble fraction (S1) for 24 h to allow uptake of pathological α-syn. The media was then exchanged, and the neurons were pre-treated with the proteasome inhibitor MG132 (10 µM) for 6 h. Next, sdAb 2D8 or sdAb 2D8-PEG4-T (5 µg/ml or its molar equivalent) were added directly into the media without media exchange. Samples were collected 2 days after antibody application. (C) Co-IP using α-syn pulldown (α-syn 211 antibody) was performed, followed by detection of α-syn (PA5-13397 antibody). Long (i) and short (ii) conjugation of the antibody with α-syn is shown. (D) Co-IP using α-syn pulldown (α-syn 211 antibody) was performed, followed by the detection of ubiquitinated proteins (P4D1 antibody). Long (i) and short (ii) ubiquitinated chain conjugation with α-syn is shown. (E) Control western blot analysis with IP input confirms the effect of modified sdAb 2D8-PEG4-T on α-syn with or without the proteasome inhibitor MG132. Complete western blot analysis can be found in Supplemental Fig. 8

Fig. 7

α-Syn brain signal following anti-α-syn sdAb administration in M83 synucleinopathy mice. (A) Schematic diagram illustrating the protocol for anti-α-syn sdAb administration in M83 synucleinopathy mice. To evaluate initial α-syn burden, all mice received an i.v. injection of fluorescently labeled anti-α-syn sdAb 2D10 (10 mg/kg), and in vivo imaging was performed using the IVIS system. Subsequently, the mice received i.v. injections of either PBS, sdAb 2D8 (3 mg/kg), or sdAb 2D8-PEG4-T (molar equivalent of 3 mg/kg 2D8) on days 4, 7, and 10. On day 13, the mice were injected with fluorescently labeled 2D10 again (10 mg/kg) to reassess α-syn burden. The mice were then perfused, and their brains extracted for tissue analyses. (B) Representative images depicting the brain signal intensity after i.v. injection of labeled sdAb-2D10 (10 mg/kg) before and after sdAb administration. The scale bar represents the maximum pixel intensity. Quantitative analysis of the brain signal was conducted by calculating the total radiant efficiency of the summed pixel intensity, derived from the area under the curve (AUC) cumulatively within 2 h post-injection (C-D). (C) Mice were assigned to groups with similar brain α-syn signal as confirmed by one-way ANOVA analysis indicating non-significant (ns) group differences in cumulative brain α-syn signal before sdAb administration (p = 0.4676). (D) One-way ANOVA revealed group differences in cumulative brain α-syn signal after sdAb administration (p = 0.0059). Compared to the PBS group, sdAb 2D8-PEG4-T decreased brain α-syn signal by 81% (**: p = 0.0049)

Fig. 8

Western blot analysis of soluble and insoluble α-syn from the in vivo study. (A) Western blot analysis of the soluble fraction probed with total α-syn antibody 211, phospho-serine129 (pS129) α-syn antibody, and loading control GAPDH antibody. The complete western blots and bands analyzed are shown in Supplemental Fig. 10. (B) One-way ANOVA revealed group differences in soluble α-syn levels as detected by total α-syn antibody 211 (p = 0.0013). sdAb 2D8-PEG4-T decreased total α-syn levels by 70% (p = 0.0072), compared to the PBS treatment (p = 0.0072), and by 72% compared to sdAb 2D8 (p = 0.0018). (C) One-way ANOVA revealed group differences in soluble α-syn levels as detected by pS129 α-syn antibody (p = 0.0002). sdAb 2D8-PEG4-T decreased pS129 α-syn levels by 90% compared to the PBS treatment (p = 0.0001), and by 85% compared to sdAb 2D8 (p = 0.0055). (D) One-way ANOVA did not reveal significant group differences in GAPDH levels (p = 0.1369), indicating lack of toxicity of the sdAb treatments. (E) Western blot analysis showing the sarkosyl pellet insoluble fraction probed with total α-syn antibody 211 and pS129 α-syn antibody. The complete western blots and bands analyzed are shown in Supplemental Fig. 10. (F-G) One-way ANOVA revealed significant group differences in insoluble α-syn levels as detected by (F) total α-syn antibody 211 (p = 0.0001) and (G) pS129 α-syn antibody (p < 0.0001), compared to the PBS treatment. sdAb 2D8 decreased total α-syn levels by 69% (p = 0.0015, F) and pS129 α-syn levels by 59% (p = 0.0016, G). sdAb 2D8-PEG4-T decreased total α-syn levels by 93% (p < 0.0001, F) and pS129 α-syn levels by 89% (p < 0.0001, G), compared to the PBS treatment. **, ***, ****: p < 0.01, 0.001, 0.0001 (Tukey’s post-hoc test)

Fig. 9

Analysis of phospho-Ser129 (pS129) immunoreactivity and astro-/microgliosis following sdAb administration in M83 mice. (A) Representative images of coronal brain sections through the midbrain region from mice treated with PBS, 2D8, and 2D8-PEG4-T, showing pS129-labeled α-syn aggregates. Distinct neuronal perikaryal inclusions (white arrows) and Lewy body (LB)-like inclusions (red arrows) were evident in the PBS treatment group. In comparison, the 2D8 treatment group exhibited reduced perikaryal inclusions (white arrows) and fewer LB-like inclusions (red arrows). Some vascular (lower half of image) and/or spheroid-like pS129 αsyn immunoreactivity (upper right of image) was also seen in this group. The 2D8-PEG4-T treatment group displayed only a few LB-like inclusions (red arrows), and some vascular immunoreactivity or perfusion artifact. Analogous α-syn pathologies were noted in the original report of this model [51]. Scale bar: 50 μm. (B) Digital quantification of brain α-syn pathology. One-way ANOVA revealed group differences in pS129 α-syn staining (p = 0.0188). 2D8-PEG4-T group had a significantly reduced α-syn pathology compared to the PBS group (70% reduction, p = 0.0146). There was no significant difference between 2D8 and 2D8-PEG4-T groups. (C) Semi-quantitative analysis of brain α-syn pathology. One-way ANOVA revealed group differences in pS129 α-syn staining (p = 0.0020). Both 2D8 (p = 0.0487) and 2D8-PEG4-T (p = 0.0079) groups had significantly reduced α-syn pathology compared to the PBS group (49% vs. 62% reduction, respectively). There was no significant difference between 2D8 and 2D8-PEG4-T. (D) Western blots of the soluble brain fraction probed with GFAP antibody, Iba-1 antibody, and loading control GAPDH antibody. (E) One-way ANOVA revealed group differences in normalized GFAP levels. (p = 0.028). sdAb 2D8-PEG4-T decreased GFAP levels by 81% compared to the PBS treatment (p = 0.022). (F) One-way ANOVA revealed group differences in normalized Iba-1 levels. (p = 0.0328). sdAb 2D8-PEG4-T decreased Iba-1 levels by 70% compared to the PBS treatment (p = 0.0429). *, **: p < 0.05, 0.01 (Tukey’s post-hoc test). The complete western blots and bands analyzed are in Supplemental Fig. 10