CPEB3 selectively inhibits α‐synuclein aggregation without modulating TDP‐43 pathology

Abstract

Abnormal accumulation of misfolded proteins is a hallmark of neurodegenerative diseases. Amyloid aggregation of α‐synuclein (α‐Syn) and TAR DNA‐binding protein 43 (TDP‐43) contributes to Parkinson's disease and frontotemporal dementia, respectively. The heterotypic aggregates are increasingly recognized as highly cytotoxic. Given the frequent co‐occurrence of α‐Syn, TDP‐43, and tau pathologies, we examined whether the first prion‐like domain (PRD1) of CPEB3 modulates α‐Syn and TDP‐43 aggregation. Nuclear magnetic resonance (NMR) relaxation experiments revealed a direct interaction between PRD1 and the amyloid core of α‐Syn, suppressing its aggregation, while phase separation assays showed delayed liquid–liquid phase separation (LLPS) ‐mediated α‐Syn aggregation. In contrast, no interaction was detected with the C‐terminal domain of TDP‐43 (TDP‐43_CTD), indicating selective inhibition of α‐Syn aggregation by PRD1.

Biophysical studies show that the CPEB3 prion‐like domain 1 (PRD1) inhibits α‐synuclein fibril formation. NMR reveals that PRD1 stabilizes the N‐terminal and hydrophobic regions of α‐synuclein, thereby suppressing aggregation. Consistently, PRD1 delays α‐synuclein aggregation under LLPS conditions while having no detectable effect on fibril formation of the TDP‐43 C‐terminal domain.

Abbreviations

AD, Alzheimer's disease

ALS, Amyotrophic lateral sclerosis

CPEB3, Cytoplasmic Polyadenylation Element Binding Protein

CPMG, Car‐Purcell Meiboom‐Gill

DIC, Differential interference contrast

FTD, Frontotemporal dementia

HSQC, Heteronuclear single quantum coherence

IDP, Intrinsically disordered protein

LB, Lewy bodies

LCD, Low‐complexity domain

LLPS, Liquid–liquid phase separation

LMD, Localization module

ND, Neurodegenerative disease

NES, Nuclear export signal

NMR, Nuclear magnetic resonance

PD, Parkinson's disease

PEG, polyethylene glycol

PRD, Prion‐like domain

SDS/PAGE, sodium dodecyl sulfate‐polyacrylamide gel electrophoresis

TDP‐43, Tar‐DNA‐binding protein‐43

TEM, transmission electron microscopy

ThT, Thioflavin T

Neurodegenerative diseases have traditionally been classified according to the aggregation of a single hallmark protein that defines each disorder [1]. Increasing evidence, however, indicates that multiple aggregation‐prone proteins can misfold and accumulate within the same brain across different neurodegenerative conditions [2, 3]. This convergence of proteinopathies likely reflects the finely balanced nature of the brain's protein homeostasis network, which becomes increasingly vulnerable with age [4, 5]. The simultaneous presence of two or more distinct proteinopathies within a single brain is referred to as mixed pathology. Such pathological overlap suggests the existence of shared molecular mechanisms that collectively contribute to neuronal dysfunction and neurodegeneration [6, 7, 8]. Importantly, some neurodegenerative diseases are more prone to mixed pathology than others, with α‐synuclein and TDP‐43 pathologies commonly co‐occurring in the aging brain.

α‐Syn is an intrinsically disordered protein whose misfolding, aggregation, and subsequent amyloid fibril formation are closely associated with the pathogenesis of PD [9, 10, 11, 12]. A key pathological hallmark of PD is the presence of intraneuronal protein inclusions called Lewy bodies (LBs), which are primarily composed of α‐Syn fibrils [13]. Structurally, α‐Syn consists of three regions: an N‐terminal region, a central hydrophobic region that drives aggregation, and a highly acidic, flexible C‐terminal region. While the hydrophobic region is considered the primary driver of aggregation [14], recent evidence indicates that the N‐terminal region also plays a critical role in modulating α‐Syn misfolding and aggregation [15, 16]. The N‐terminal and central hydrophobic regions of α‐Syn promote its liquid–liquid phase separation under physiological conditions and in the presence of molecular crowding agents. LLPS is increasingly recognized as an early and potentially pathological step in the transition from soluble α‐Syn to its aggregated, disease‐associated forms [17].

A major pathological hallmark of Amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) is the presence of intracellular inclusions composed of TDP‐43, characterized by the selective loss of motor neurons, typically resulting in death within a few years of symptom onset [18]. TDP‐43 is a 414‐amino acid protein consisting of a ubiquitin‐like N‐terminal domain (NTD) containing a nuclear localization signal (NLS), two RNA recognition motifs (RRM1 and RRM2), a nuclear export signal (NES), and a highly aggregation‐prone C‐terminal domain (CTD) [19]. The CTD is enriched in glutamine/asparagine (Q/N) and glycine (Gly) residues and harbors an intrinsically disordered prion‐like domain (PrLD) that adopts transient secondary structures in aqueous environments. This domain facilitates the formation of dynamic, β‐sheet‐rich oligomers implicated in pathological aggregation [20]. Notably, pathological inclusions in TDP‐43 proteinopathies predominantly consist of C‐terminal fragments, with the low‐complexity domain (LCD) representing the core aggregation‐prone component [21, 22, 23].

The presence of both TDP‐43 and α‐synuclein pathologies has been reported in Parkinson's disease and dementia with lewy bodies. Moreover, patients with Alzheimer's disease, characterized by tau and Aβ aggregation, frequently exhibit α‐synuclein–positive inclusions and TDP‐43 C‐terminal fragments [24, 25]. Together, these observations underscore the substantial pathological overlap among major neurodegenerative disorders and highlight the tendency of multiple aggregation‐prone proteins to coexist within the same disease context [2]. A shared characteristic of these proteins is that they are capable of self‐propagation and intercellular transmission via a prion‐like mechanism [26, 27]. Clinically, the presence of mixed pathologies is associated with faster cognitive decline and more rapid disease progression compared to cases involving a single pathology [28, 29, 30, 31]. Emerging evidence suggests that diagnostic and therapeutic approaches developed for a specific neurodegenerative disorder may have broader applicability across multiple diseases characterized by overlapping proteinopathies [8]. Thus, characterization of protein–protein interactions (PPIs) is essential for understanding the molecular mechanisms that influence the rate of disease progression [32, 33].

In our previous study, we examined the interaction between tau and the RNA‐binding protein CPEB3, revealing that the prion‐like domain 1 (PRD1) of CPEB3 is capable of inhibiting tau aggregation. CPEB3 is a functional prion protein that plays a critical role in synaptic plasticity and long‐term memory formation [34, 35, 36, 37, 38]. N‐terminal region of CPEB3 is compartmentalized into three distinct subdomains: PRD1; residues 1–217, a localization module (LMD; residues 218–284), and PRD2; residues 285–449 [39]. The intrinsic aggregation propensity of CPEB3 is primarily attributed to PRD1, which is rich in disordered and aggregation‐prone sequences, characteristic of functional prion domains. Given that CPEB3 has been shown to inhibit tau aggregation, we investigate here the interactions of α‐Syn and TDP‐43 with CPEB3 in order to evaluate its potential role as a therapeutic modulator in mixed pathology conditions involving these aggregation‐prone proteins [40].

Materials and methods

Expression and purification of α‐synuclein

The construct encoding wild‐type α‐synuclein cloned into the pT7‐7 vector was obtained from Genscript and transformed into E. coli. BL21 (DE3) competent cells. Protein expression was carried out in Luria‐Bertani (LB) medium supplemented with 100 mg/mL ampicillin resistance at 37 °C. When the culture reached an optical density of 0.6 at 600 nm, expression was induced with 1 mm isopropyl‐β‐D‐thiogalactopyranoside (IPTG) and allowed to proceed for 8 h at 37 °C.

For ¹⁵N‐labeled samples used in NMR spectroscopy, the protein was expressed in M9 minimal medium containing 1 g/L ¹⁵NH₄Cl (Cambridge Isotope Laboratories, Tewksbury, MA, USA). Cells were harvested by centrifugation at 10000 ×  g for 10 min at 4 °C. The resulting pellet was resuspended in lysis buffer (50 mm Tris/HCl, pH 7.4; 20 mm NaCl; 10% glycerol; and 30 μL protease inhibitor) and subjected to heat treatment at 99 °C for 20 min with shaking at 300 rpm. The lysate was then sonicated using a Branson probe sonicator (15 s on, 30 s off, 10 cycles). Cell debris was removed by centrifugation at 10 000 rpm for 1 h at 4 °C. The supernatant was precipitated using 60% ammonium sulfate (60 g per 100 mL supernatant). The precipitated protein was dissolved in equilibration buffer (20 mM Tris/Cl, pH 6.5; 1 mm EDTA; 2 mm DTT; 1 mm PMSF), centrifuged and loaded on a pre‐equilibrated Q Sepharose anion exchange column. The protein was eluted using a linear NaCl gradient (50 mm to 1 M NaCl). Eluted fractions were analyzed by SDS/PAGE to confirm purity (Fig. S1). Protein‐containing fractions were pooled, dialyzed against 20 mm ammonium bicarbonate, lyophilized, and stored at −20 °C until further use.

PRD1 expression and purification

The PRD1 construct was obtained from GenScript as a gene insert in the pET28a vector between the restriction sites NdeI and XhoI, encoding an N‐terminal 6 × His tag. The construct was transformed into E. coli BL21 (DE3) competent cells and expressed in LB medium supplemented with 20 μg/mL kanamycin. Protein expression was induced at an OD₆₀₀ of 0.6 by the addition of 1 mm IPTG and incubating at 37 °C for 23 h.

Cell pellets acquired by centrifugation at 10000 × g for 10 min are resuspended in lysis buffer (20 mM sodium phosphate buffer, pH 8.0; 0.5 M NaCl; 1 mm PMSF; 0.05% β‐mercaptoethanol; 5% glycerol; 0.5% Triton X‐100; and 1 × protease inhibitor). The suspension was sonicated using a Branson probe sonicator (30 s on, 30 s off, total on‐time 15 min), followed by heating at 99 °C for 20 min. Cell debris was removed by centrifugation at 10 000 rpm for 45 min at 4 °C. The supernatant was subjected to ammonium sulfate precipitation (40 g per 100 mL of supernatant). The resulting protein pellet was resuspended in equilibration buffer (20 mm sodium phosphate buffer, pH 8.0; 150 mm NaCl; 0.03% NaN₃). The protein solution was incubated with pre‐equilibrated Ni‐NTA Superflow resin at 4 °C with gentle rotation (10 rpm) for 1 h. The resin was then washed, and the bound protein was eluted using a linear imidazole gradient (10–100 mm) in equilibration buffer. Eluted fractions were analyzed by SDS/PAGE to assess purity (Fig. S1). Protein‐containing fractions were pooled, concentrated using an Amicon Ultra centrifugal filter unit (10 kDa MW) cut‐off membrane and buffer‐exchanged into phosphate buffer (20 mm sodium phosphate, pH 7.4; 150 mm NaCl; 0.03% NaN₃). The purified protein was aliquoted and stored at −80 °C.

Expression and purification of TDP‐43 C‐terminal domain (TDP‐43_CTD )

The construct encoding the C‐terminal fragment of TDP‐43 (residues 267–414), containing a C‐terminal 6 × His tag and cloned into the pET‐20b(+) expression vector (ampicillin resistance), was obtained from GenScript (Piscataway, NJ, USA). The plasmid was transformed into E. coli BL21 (DE3) competent cells, and protein expression was carried out in LB medium supplemented with 100 μg/mL ampicillin. Cultures were grown at 37 °C with shaking at 200 rpm and induced with 1 mm IPTG upon reaching an optical density (OD₆₀₀) of 0.6–0.8. Following induction, culture was incubated for 16 h at 37 °C. Cells were harvested by centrifugation at 8500 rpm for 10 min at 4 °C and resuspended in lysis buffer containing 500 mm NaCl, 20 mm Tris/HCl (pH 8.0), 10 mm imidazole, and 1 mm dithiothreitol (DTT). The suspension was lysed using a Branson probe sonicator (30 s on/30 s off, 50% amplitude) for a total on‐time of 15 min. The lysate was centrifuged at 9500 rpm for 1 h at 4 °C, and the resulting pellet containing inclusion bodies was resuspended in denaturing buffer composed of 8 M urea, 500 mm NaCl, 20 mm Tris/HCl (pH 8.0), 10 mm imidazole, and 1 mm DTT. To remove insoluble debris, the suspension was centrifuged again at 8500 rpm for 1 h at 4 °C. The supernatant containing solubilized denatured protein was incubated with pre‐equilibrated Ni‐NTA Superflow resin at 4 °C for 1.5 h under gentle rotation (10 rpm) to facilitate binding. The bound protein was eluted using a linear imidazole gradient (20 mm to 1 m) in the same denaturing buffer. Eluted fractions were analyzed on 15% SDS/PAGE, and those containing the target protein were pooled. Gradual urea removal from the target protein was performed by stepwise dialysis in denaturing buffer (without imidazole) containing 6 m, 4 m, and finally 2 m urea. The partially refolded protein was subsequently desalted into 20 mm MES buffer (pH 6.1) using Zeba Spin Desalting Columns (Thermo Scientific, Pierce, USA) to obtain the protein in a native‐like state. Protein concentration was determined using a UV–visible spectrophotometer and used for further experiments.

Thioflavin T (ThT) fluorescence assay for aggregation kinetics

To assess the effect of PRD1 on the aggregation kinetics of α‐Syn and TDP‐43_CTD, ThT fluorescence assays were performed by incubating the samples at 37 °C. For α‐Syn, fluorescence measurements were conducted with varying concentrations of the protein in 20 mm phosphate‐buffered saline (PBS, pH 7.4) containing 50 μm ThT. Additionally, aggregation assays were carried out using 500 μm in the presence and absence of 500 μm PRD1. For TDP‐43_CTD, the ThT assay was conducted with 20 μm of the protein and 10 μm ThT, in the presence and absence of equimolar PRD1 (20 μm), in 20 mm MES buffer at pH 6.1.

All fluorescence measurements were performed using a TECAN F200 Pro plate reader, with readings taken every 20 min (10 min of shaking at 200 rpm followed by a 10 min rest period). Excitation and emission wavelengths were set to 440 nm and 480 nm, respectively. Each experimental condition was performed in triplicate, and the resulting fluorescence intensities were averaged. Aggregation kinetics were analyzed by fitting the averaged curves to a sigmoidal function to determine the half‐life (t_1/2) of aggregation. Blank controls containing only ThT (at the same concentration as in the experimental conditions) in the absence of protein were included to account for background fluorescence from unbound ThT. All measurements were conducted in sealed plates to prevent evaporation.

X‐34 fluorescence assay

Aggregation kinetics of α‐Syn in the presence of PRD1 were also monitored using the X‐34 fluorescence assay, following the same protocol as the ThT assay. X‐34 measurements were performed at a lower α‐Syn concentration (100 μm), which is fivefold less than that used in the corresponding ThT experiments, and 10 μm of the dye was used. Fluorescence readings were recorded with an excitation wavelength of 380 nm and an emission wavelength of 460 nm, and the measurements were performed in TECAN F200 Pro plate reader.

Transmission electron microscopy (TEM)

TEM was employed to examine the morphology of fibrils formed by α‐Syn and TDP‐43_CTD under various aggregation conditions. Monomeric α‐Syn and TDP‐43_CTD were incubated at their respective optimal concentrations in 20 mm PBS and 20 mm MES buffers respectively. Samples were maintained at 37 °C with shaking in the absence of fluorescent dyes, and aliquots for TEM analysis were collected at time points corresponding to the steady‐state phase of the ThT fluorescence kinetics (measured using a TECAN F200 Pro plate reader). To allow complete fibril maturation, the samples were further incubated for several additional days. To assess phase separation–induced aggregation, α‐Syn was incubated in PBS containing 20% polyethylene glycol (PEG‐8000) at 37 °C, while phase separation of TDP‐43_CTD was induced by the addition of 100 mm NaCl. For both proteins, aliquots of samples with and without PRD1 were collected daily for TEM analysis under their respective buffer conditions.

For TEM grid preparation, a 5 μL aliquot of the sample was diluted to 10 μL with the corresponding buffer. The diluted sample was applied to a 400‐mesh carbon‐coated copper grid and allowed to adsorb for 2 min. Grids were then negatively stained with 3% (w/v) uranyl acetate for 45 s, followed by two washes with deionized, autoclaved water (30 s each), and air‐dried for 3 min. Imaging was performed on a 120 kV FEI TECNAI G2 Spirit Bio‐Twin transmission electron microscope.

Optical microscopy

Phase separation of α‐Syn was monitored using Differential Interference Contrast (DIC) and fluorescence microscopy on an Olympus IX83 inverted microscope. For fluorescence imaging, proteins were labeled with NHS‐Rhodamine. The dye was dissolved in 5 μL of anhydrous DMSO and added to the phase‐separated protein samples at a dye‐to‐protein molar ratio of 1 : 20. A 2 μL aliquot of the labeled condensate was placed on a clean glass slide and covered with a coverslip. Images were acquired using appropriate DIC and fluorescence settings and subsequently processed using ImageJ software.

Solution NMR experiments

¹H − ¹⁵N HSQC NMR

NMR experiments were performed on Bruker Avance III spectrometers operating at a ¹H frequency of 700 MHz, equipped with triple resonance TXI probes and z‐axis gradients. All measurements were carried out at 5 °C unless otherwise specified. NMR spectra were processed using NMRPipe, and peak assignment was performed with Sparky. For α‐Syn experiments, ¹⁵N‐labeled α‐Syn (350 μm) was prepared in 20 mm PBS (pH 7.4), containing 0.05% sodium azide (NaN₃) and 5% D₂O. To investigate the effect of PRD1 on α‐Syn aggregation, samples containing 350 μm ¹⁵N‐labeled α‐Syn alone, and in combination with 350 μm unlabeled PRD1, were incubated at 37 °C with shaking at 900 rpm in a thermoshaker. ¹H–¹⁵N heteronuclear single quantum coherence (HSQC) spectra were acquired before and after incubation to assess aggregation‐induced signal changes. The relative peak intensities (I/I₀) were calculated, where I₀ corresponds to the signal before incubation and I represents the intensity after incubation. Additionally, to monitor aggregation, HSQC spectra were also recorded daily at 25 °C for the incubated samples.

To study residue‐specific interactions between PRD1 and TDP‐43_CTD, ¹H–¹⁵N HSQC spectra were acquired for 75 μm ¹⁵N‐labeled TDP‐43_CTD in 20 mm MES buffer (pH = 6.5), containing 0.05% sodium azide (NaN₃) and 10% D₂O in the presence and absence of equimolar concentration of PRD1 (75 μm). Chemical shift perturbations were calculated to identify the residues affected upon binding using the equation given below.

$$\mathrm{CSP}=\frac{\sqrt{{\mathit{\delta H}}^2+{\left(\frac{\mathit{\delta N}}{5}\right)}^2}}{2}$$ (1)

¹⁵N relaxation experiments

To examine the residue‐specific dynamics of α‐Syn, transverse ¹⁵N relaxation rate (R₂) measurements were performed with and without PRD1. Relaxation‐compensated Carr–Purcell–Meiboom–Gill (CPMG) pulse sequences were used with a constant field of 550 Hz to suppress line broadening due to chemical exchange. Spectra were recorded with 2048 complex points in the direct (¹H) dimension and 256 complex points in the indirect (¹⁵N) dimension. Variable relaxation delays used were: 10, 200, 400, 600, 800, 1000, 1200, 1400, 1600, and 1800 ms. Peak intensities were fitted using an exponential decay model to extract R₂ values. The difference in R₂ between the PRD1‐bound and unbound samples (ΔR ₂) was calculated to assess residue‐specific lifetime line broadening due to interaction.

Results and discussion

Given the clinical relevance of coexisting proteinopathies, particularly those involving α‐Syn, TDP‐43, and tau, this study is aimed to investigate whether PRD1 of CPEB3 exerts a modulatory effect on the aggregation behavior of these proteins, similar to its previously observed influence on tau‐K18. To investigate potential interactions, we employed full‐length α‐synuclein (residues 1–140) with a non‐amidated N‐terminus (Fig. 1A), the C‐terminal domain of TDP‐43 (residues 267–414), and PRD1 of CPEB3 (Fig. 1B).

Fig. 1.

Structural features of α‐Syn and CPEB3 PRD1, and the impact of PRD1 on α‐Syn aggregation. Schematic representation of (A) full‐length α‐Syn (140 aa) (B) full‐length CPEB3 and the first prion domain of CPEB3 (PRD1) as used in the study. (C) Thioflavin T fluorescence assay of α‐Syn and α‐Syn with PRD1. Error bar represents standard deviation of three replicates. Transmission electron microscopic images of (D) α‐Syn after 36 h (E) after 5 days incubation at 37°C (F) α‐Syn with PRD1 after 36 h (G) after 5 days incubation at 37 °C (Scale bar: 500 nm). Three independent experiments were performed (n = 3).

This study focuses on elucidating the direct interactions between PRD1 and α‐Syn or TDP‐43, characterizing the residue‐specific regions involved in the interaction using NMR, and identifying the consequent changes in conformational dynamics. This provides a better understanding of the broader regulatory potential of CPEB3 in modulating multiple aggregation‐prone proteins implicated in neurodegenerative disease.

PRD1 inhibits α‐Syn fibril formation

The process of fibril formation of α‐Syn was primarily monitored using ThT fluorescence assay in a time dependent manner. ThT exhibits an increase in fluorescence on binding to amyloid fibrils. This binding assay can be used to monitor the aggregation as it gives a sigmoidal curve upon fibril formation with a lag phase (nucleation phase), an exponential phase (rapid aggregation) and a saturation phase at the steady state. The half‐life (t_1/2) of aggregation, that is, the time required to achieve 50% of the final ThT fluorescence intensity serves as an indicator of the rate of protein aggregation.

ThT fluorescence assay was performed on different concentrations of α‐Syn (150 μm, 250 μm, and 500 μm). The results demonstrated that α‐Syn aggregation kinetics are concentration‐dependent, as evidenced by a progressive decrease in aggregation half‐life at higher α‐Syn concentrations (Fig. S2). To investigate the effect of PRD1 on α‐Syn aggregation, we compared the ThT fluorescence intensity of α‐Syn (500 μm) alone and α‐Syn (500 μm) co‐incubated with an equimolar concentration of PRD1 (500 μm) (Fig. 1C). α‐Syn (500 μm) alone showed an increase in fluorescence intensity within 36 h of incubation at 37 °C. But in the presence of PRD1, no significant increase in fluorescence was observed indicating that PRD1 inhibits the formation of ThT‐positive fibrils of α‐Syn. The fluorescence intensity of α‐Syn and α‐Syn in the presence of PRD1 was also assessed using X‐34, a Congo red derivative. Consistent with these findings, fluorescence measurements using X‐34, a Congo red derivative, also showed a similar inhibitory effect of PRD1 on α‐Syn aggregation (Fig. S3).

To further validate this observation, TEM was performed on both samples. After 36 h (ThT incubation time), α‐Syn alone displayed a mixture of fibrils and amorphous aggregates (Fig. 1D), Upon extended incubation for an additional 3 days at 37 °C, α‐Syn alone had fully transitioned into mature fibrils (Fig. 1E). In contrast, the α‐Syn with PRD1 sample predominantly exhibited amorphous aggregates (Fig. 1F), and no evidence of fibril formation was detected following 3 days of incubation at 37 °C, indicating that the resulting aggregates are likely mixed assemblies of α‐Syn and PRD1. This mixed aggregation appears to interfere with the typical fibril formation pathway of α‐Syn, preventing the generation of mature α‐Syn fibrils (Fig. 1G). TEM images from multiple regions were imaged to ensure consistency of the observed morphology (Fig. S4). The present results indicate that PRD1 inhibits α‐Syn fibril formation, likely by stabilizing off‐pathway intermediate species. Previous studies have shown that certain off‐pathway oligomers can be relatively stable and do not necessarily mature into amyloid fibrils [41]. However, it is also well recognized that some oligomeric species may possess inherent cytotoxicity. Thus, while PRD1‐mediated stabilization of such intermediates appears to suppress fibril formation in vitro, the net effect on disease progression remains uncertain. Whether this interaction is ultimately protective or detrimental would depend on the specific structural and functional properties of the intermediates formed. Additional studies are therefore required to characterize these species in detail and to assess their cellular toxicity in more physiologically relevant models.

NMR‐based characterization of the interaction between α‐Syn and PRD1

To understand the fundamental interactions that might exist between α‐Syn and PRD1 a series of ¹H−¹⁵N 2D HSQC NMR titration experiments were performed with ¹⁵N‐labeled α‐Syn. Initially the control ¹H−¹⁵N 2D HSQC spectrum of α‐Syn was acquired at 5 °C and pH 7.4. To this sample, equal concentration of PRD1 was added and HSQC spectra were recorded. An overlay of the spectra revealed negligible chemical shift perturbations upon the addition of PRD1 (Fig. S5), indicating that PRD1 does not induce major conformational changes in α‐Syn under these conditions. To assess whether PRD1 influences the behavior of α‐Syn under aggregation‐prone conditions, HSQC spectra were recorded at 5 °C for α‐Syn [¹⁵N] at pH 7.4 in presence and absence of PRD1 before and after incubation at 37 °C and at 900 rpm for 40 h (conditions that promote fibril formation). Before incubation, the peak intensities of the residues of α‐Syn [¹⁵N] and α‐Syn [¹⁵N] with PRD1 were comparable (Fig. 2A). Incubation resulted in a significant reduction in peak intensities of α‐Syn [¹⁵N] alone compared to α‐Syn [¹⁵N] with PRD1. The differential reduction in peak intensity was substantial in the N‐terminal region of the protein (Fig. 2B). The intensity ratio (I/I₀) was also plotted for α‐Syn [¹⁵N] and α‐Syn [¹⁵N] with PRD1 (Fig. S6). I₀ represents the signal intensity before incubation and I represents the signal intensity after incubation. In I/I₀ values we observed an overall reduction for α‐Syn compared to α‐Syn with PRD1. These results suggest that PRD1 influences the self‐association of α‐Syn [¹⁵N] and reduces its oligomerization propensity. Though the incubation was done at 37 °C, all the HSQC spectra were measured at 5 °C to maximize the number of visible peaks in the spectra for more quantitative analysis. Lower temperature also reduces any aggregation during the experiments.

Fig. 2.

NMR analysis of α‐Syn dynamics and interaction with CPEB3 PRD1. Nuclear magnetic resonance (NMR) studies of α‐Syn with first prion domain of CPEB3 (PRD1). Intensity plot of α‐Syn and α‐Syn with PRD1 (A) before incubation. (B) After 40 h incubation at 37 °C a reduction in peak intensities was observed for α‐Syn alone compared to α‐Syn with PRD1. (C) Residue‐specific change in transverse relaxation rate of α‐Syn in the presence of PRD1. Increase in transverse relaxation rate (¹⁵N‐ΔR₂) of α‐Syn (500 μm) with PRD1 (500 μm) in the aggregation prone N‐terminal and hydrophobic region. Error bars represent the standard deviation (SD). The highlighted region corresponds to the α‐Syn residues that are critical for the interaction with PRD1. Three independent experiments were performed (n = 3).

To further assess residue‐specific changes under conditions closer to ambient temperature, we additionally recorded HSQC spectra of α‐Syn [¹⁵N] and α‐Syn [¹⁵N] with PRD1 at 25 °C. In this set of experiments, the samples were incubated at 37 °C with shaking at 900 rpm, and HSQC spectra were collected daily to monitor the progression of aggregation. After 2 days of incubation, the HSQC analysis at 25 °C revealed that several peaks present in the α‐Syn [¹⁵N] with PRD1 sample (Fig. S7B) were no longer detectable in the α‐Syn [¹⁵N] sample without PRD1 (Fig. S7A). The missing peaks correspond to residues G31, G41, T54, S42, S9, T64, T72, T75, which are located within the N‐terminal and the hydrophobic regions, known for their aggregation propensity [42, 43, 44]. The loss of these signals in the α‐Syn [¹⁵N] spectrum is attributed to aggregation‐induced line broadening [45, 46]. In contrast, the preservation of signal intensity in the presence of PRD1 suggests that PRD1 inhibits α‐Syn aggregation by stabilizing these aggregation‐prone residues.

Prior work has shown that α‐Syn monomers interact with fibril surfaces primarily through their N‐terminal region [47]. Moreover, several α‐Syn aggregation inhibitors have been reported to preferentially target the N‐terminal and hydrophobic regions [48, 49, 50], in agreement with our observations. Previous studies have shown that molecular chaperones recognize a characteristic motif in α‐synuclein that includes the N‐terminal region and residues around Tyr39, allowing transient interactions that suppress protein aggregation [48]. In addition, long‐range electrostatic interactions between the positively charged N‐terminus and the negatively charged C‐terminus have been shown to confer a self‐chaperoning effect, stabilizing aggregation‐resistant conformations of α‐Syn [51]. While the precise molecular mechanism by which PRD1 inhibits α‐Syn fibril formation remains to be fully elucidated, our data are consistent with a model in which PRD1 interacts with the N‐terminal and hydrophobic regions of α‐Syn, thereby perturbing the aggregation processes mediated by the fibril core. Solid‐state NMR studies have demonstrated that the β‐sheet‐rich fibril core of α‐Syn comprises residues 38–98 [52, 53]. This region overlaps with the PRD1 interaction site, identified using NMR spectroscopy supporting the hypothesis that PRD1 interferes with the self‐assembly of α‐Syn and thereby suppresses fibril formation.

To determine the conformational dynamics of α‐Syn in the presence of PRD1, we performed relaxation NMR experiments using a relaxation‐compensated Carr−Purcell−Meiboom−Gill (CPMG) pulse sequence. The conformational dynamics of the protein will be significantly influenced by a binding process, affecting both the thermodynamic stability and kinetics of the interaction [54, 55, 56]. NMR relaxation experiments can reveal valuable information about local conformational dynamics, particularly segmental motions [57, 58]. An increase in ¹⁵N R ₂ value is indicative of a potential interaction between the two proteins, suggesting a conformational exchange on the microsecond–millisecond timescale [41].

Residue‐specific ¹⁵N‐ΔR ₂ values were obtained by subtracting the R ₂ relaxation rates of free ¹⁵N‐labeled α‐Syn from those measured in the presence of PRD1 at the same aggregation time point. Data was acquired for both the samples before and after incubation at 37 °C for 40 h. Upon incubation, an overall increase in ¹⁵N‐ΔR ₂ values was observed, with particularly higher values in the 37V–57E region (Fig. 2C). This region encompasses the segments ³⁶GVLYVGS⁴² and ⁴⁵KEGVVHGVATVAE⁵⁷, both previously identified as highly amyloidogenic [59, 60, 61]. Notably, the 47–56 segment known as the hydrophobic region has been reported to independently form amyloid‐like aggregates [50]. The higher ¹⁵N‐ΔR ₂ values within this region suggest that the interaction with PRD1 restricts the conformational flexibility of α‐Syn, thereby interfering with its self‐assembly. Taken together, these results indicate that PRD1 inhibits α‐Syn aggregation by modulating conformational dynamics and stabilizing the aggregation‐prone 37V–57E region of α‐Syn by interfering with the self‐assembly of the protein, thus limiting its participation in fibril formation. All relaxation NMR experiments were conducted at 5 °C. Few peaks were absent from the spectra due to aggregation‐induced line broadening.

α‐Syn and PRD1 coexist in a phase separated state

The structural conversion of α‐Syn into amyloid fibrils is a well‐established pathological hallmark of PD [62, 63]. While aggregation is a central feature of amyloid‐related disorders, the role of LLPS and its relationship in modulating protein aggregation has attained increasing attention. Emerging evidence suggests that factors influencing amyloid formation often exert similar effects on phase separation [64]. Having characterized the interaction between monomeric α‐Syn and PRD1 under aggregating conditions, we next sought to determine whether these proteins are also capable of coexisting within a phase‐separated state. In the presence of a molecular crowding agent such as 20% polyethylene glycol (PEG‐8000) α‐Syn at a concentration of 200 μm, formed liquid‐like droplets in vitro indicative of LLPS. Previous studies have shown that compared to physiological pH (7.4), acidic pH (5.4) requires only low critical protein concentration for LLPS formation and accelerating the aggregation [65, 66]. To explore how PRD1 influences the phase behavior of α‐Syn, we investigated their interaction under two pH conditions: 5.4 and 7.4.

At pH 7.4, α‐Syn underwent LLPS, forming liquid‐like droplets after 24 h of incubation at 37 °C. Over the next 3 days, these droplets gradually matured into fibrils. This transition was reflected in the gradual increase in ThT fluorescence intensity in the phase‐separated samples (Fig. 3A) and clearly observed in the TEM images (Fig. 3C,D). Studies show that the evolution of amyloid fibrils from liquid‐like droplets can be effectively monitored using the ThT assay [67, 68, 69]. At 24 h, TEM revealed primarily liquid‐like droplets rather than mature fibrils (Fig. 3C). At this stage, although the samples exhibited increasing ThT fluorescence, the structural conversion into detectable fibrils was not yet complete. The rise in ThT signal at intermediate time points is therefore interpreted as capturing early cross‐β structural ordering within the droplets, followed by the appearance of mature fibrils within 3 days as observed in TEM. In contrast, under the same conditions, α‐Syn incubated with PRD1 displayed markedly lower ThT fluorescence throughout the incubation period, suggesting a reduced propensity of droplets to form fibrils (Fig. 3A). Microscopy further revealed that, in the presence of PRD1, α‐Syn failed to form stable liquid‐like droplets; instead, amorphous aggregates dominated the phase‐separated samples (Fig. 3C). Even after 3 days of incubation, these aggregates did not fully transition into fibrils, confirming that PRD1 delays fibril formation by α‐Syn. (Fig. 3D). Interestingly, the aggregates that accumulated in the presence of PRD1 appeared as oligomeric or amorphous species, which may represent off‐pathway intermediates or exist in equilibrium with fibrils, with the potential to convert over longer timescales. Collectively, these findings suggest that PRD1 interferes with LLPS‐mediated aggregation of α‐Syn, most likely by stabilizing intermediate oligomeric forms. However, additional studies are needed to define the structural characteristics of these intermediates and to assess their potential cellular toxicity.

Fig. 3.

Effect of PRD1 on PEG‐induced phase separation and aggregation of α‐Syn at physiological and acidic pH. Phase separation studies of α‐Syn with 20% polyethylene glycol (PEG‐8000). (A) Thioflavin T (ThT) fluorescence studies of α‐Syn and α‐Syn with first prion domain of CPEB3 (PRD1) in presence of PEG at pH = 7.4 (B) ThT fluorescence assay of α‐Syn and α‐Syn with PRD1 in presence of PEG at pH = 5.4. Error bars indicate standard deviation of three replicates. Transmission electron microscopy (TEM) images of α‐Syn (C) within 24 h of incubation at pH = 7.4 (i) α‐Syn and (ii) α‐Syn with PRD1 (D) After 3 days incubation at 37 °C (i) α‐Syn formed fibrils and (ii) α‐Syn with PRD1 formed a mixture of fibrils and aggregates. TEM images of α‐Syn (E) within 24 h of incubation at pH = 5.4. (i) α‐Syn and (ii) α‐Syn with PRD1. (F) After 2 days of incubation at 37 °C α‐Syn droplets transformed into fibrils and in α‐Syn with PRD1 some droplets persisted. Liquid–liquid phase separation‐mediated (LLPS) amyloid aggregation of α‐Syn is delayed in the presence of PRD1 (Scale bar: 500 nm). Three independent experiments were performed (n = 3).

As expected, α‐Syn underwent droplet formation more rapidly at pH 5.4. Both α‐Syn and the α‐Syn with PRD1 mixture formed liquid‐like droplets within 24 h of incubation at 37 °C in the presence of 20% PEG‐8000 (Fig. 3E). Droplet formation was further validated by DIC microscopy and fluorescence imaging using NHS‐Rhodamine (Fig. S8). ThT fluorescence assay further confirmed the progression toward amyloid fibril formation in both samples, as evidenced by an increase in fluorescence intensity over time (Fig. 3B). While both samples eventually transitioned into fibrils, the α‐Syn with PRD1 sample retained a noticeable fraction of droplets even at later time points (Fig. 3F). This observation suggests that, similar to pH 7.4, PRD1 also delays the LLPS‐mediated aggregation of α‐Syn at pH 5.4, but by stabilizing the droplet phase and extending the time required for complete conversion into fibrils.

Interaction studies of TDP‐43_CTD with PRD1

To investigate the interaction between TDP‐43_CTD (Fig. 4A, S9) and PRD1, a combination of biophysical and imaging techniques was employed, including ThT fluorescence assay, TEM, and NMR spectroscopy. The kinetics of TDP‐43_CTD fibril formation were evaluated by monitoring ThT fluorescence intensity. As shown in Fig. 4B, ThT fluorescence profiles of 20 μm TDP‐43_CTD alone and in the presence of an equimolar concentration of PRD1 (20 μm) exhibited comparable aggregation kinetics, with no significant change in the lag phase. This suggests that PRD1 does not modulate the aggregation rate of TDP‐43_CTD.

Fig. 4.

PRD1 does not modulate the aggregation or phase separation behavior of TDP‐43_CTD. (A) Schematic representation of TDP‐43. TDP‐43 C‐terminal domain (TDP‐43_CTD) was used in the study. (B) Thioflavin T fluorescence assay of TDP‐43_CTD with first prion domain of CPEB3 (PRD1). Error bars represent standard deviation of three replicates. TEM images of (C) (i) TDP‐43_CTD and (ii) TDP‐43_CTD with PRD1 on 2 days incubation at 37 °C. (D) Phase separation studies of (i)TDP‐43_CTD and (ii) TDP‐43_CTD with PRD1 on 2 days incubation with 100 mm NaCl at 37 °C (Scale bar: 500 nm). (E) Overlaid heteronuclear single quantum coherence (HSQC) spectra of TDP‐43_CTD with PRD1. These studies show that PRD1 does not influence the aggregation of TDP‐43_CTD. Three experiments were performed independently (n = 3).

To assess whether PRD1 alters the fibril morphology of TDP‐43_CTD, TEM imaging was conducted. Samples were analyzed before and after 3 days of incubation with PRD1. Aggregation of TDP‐43_CTD alone resulted in the formation of dense, elongated fibrils [Fig. 4C(i)]. Notably, co‐incubation with PRD1 at equimolar concentration produced fibrils with similar morphology [Fig. 4C(ii)], indicating that PRD1 does not disrupt the fibril formation of TDP‐43_CTD. Previous studies have shown that under aggregating conditions, the addition of salt can induce phase separation in TDP‐43_CTD. To investigate this behavior and assess the potential influence of PRD1, we examined the phase separation of TDP‐43_CTD in the presence and absence of PRD1. A solution containing 20 μm TDP‐43_CTD in phosphate buffer (pH 6.1) with 100 mm NaCl was used to promote phase separation. Under these conditions, TDP‐43_CTD formed liquid‐like droplets [Fig. 4D(i)]. When PRD1 was added at an equimolar concentration under the same buffer conditions, TDP‐43_CTD again phase separated and formed similar droplets [Fig. 4D(ii)]. These findings indicate that PRD1 does not inhibit the salt‐induced phase separation in TDP‐43_CTD [70, 71].

To assess whether PRD1 interacts with TDP‐43_CTD at the residue level, 2D ¹H–¹⁵N HSQC NMR spectroscopy was performed using 75 μM ¹⁵N‐labeled TDP‐43_CTD in 20 mM MES buffer (pH 6.1) at 5 °C. At this concentration, TDP‐43_CTD exhibited rapid aggregation upon incubation, thus we were only able to measure the HSQC spectra before incubation. The overlaid spectra of ¹⁵N‐TDP‐43_CTD and ¹⁵N‐TDP‐43_CTD with PRD1 revealed no detectable chemical shift changes (Fig. 4E). Consistently, chemical shift perturbation (CSP) analysis (Fig. S10) showed uniformly low values across all residues, indicating the absence of measurable residue‐specific interactions. These observations suggest that under the experimental conditions employed, PRD1 does not specifically interact with TDP‐43_CTD and alter its structure.

Under physiological conditions, any interaction between these intrinsically disordered proteins may be transient or weak [72]. However, neurodegenerative states frequently alter protein localization, increase cellular stress, and promote misfolding and aggregation, thereby creating an environment that can facilitate aberrant or heterotypic interactions [73]. Although our current findings are based on in vitro studies, the biological context strongly supports the possibility that such interactions could occur in vivo, particularly under pathological conditions. Investigating this in relevant cellular or animal models represents an important direction for future work.

Conclusion

Mixed proteinopathies are clinically significant, as they are associated with accelerated cognitive decline and more rapid disease progression compared to cases involving a single pathology. In this context, we investigated whether the PRD1 of CPEB3 could modulate the aggregation behavior of α‐Syn and TDP‐43, building on its previously observed effects on tau‐K18. To our knowledge, this is the first study to explore such interactions. Our NMR analyses revealed that PRD1 interacts with and stabilizes the N‐terminal and hydrophobic regions of α‐Syn, thereby inhibiting its aggregation. Transverse relaxation data further indicated that PRD1 modulates the conformational dynamics of α‐Syn, particularly by stabilizing the 37V–57E region, which in turn reduces its contribution to fibril formation. Consistently, PRD1 was also found to delay α‐Syn aggregation under LLPS conditions.

In contrast, PRD1 did not interfere with fibril formation of the TDP‐43_CTD, suggesting a degree of specificity in its modulatory activity (Fig. 5). This differential interaction is notable, given the low sequence similarity between α‐Syn and TDP‐43_CTD. One plausible explanation for this selectivity lies in the conformational differences in their aggregation intermediates. Since intrinsically disordered proteins (IDPs) interact with their partners through weak and transient interactions, sufficient incubation time is required for productive binding before aggregation dominates. In the case of α‐Syn, PRD1 has the opportunity to interact with its hydrophobic regions and inhibit aggregation. Conversely, TDP‐43_CTD, due to its low‐complexity domain, is highly prone to rapid self‐association. In our previous work, we demonstrated that PRD1 interacts with tau through hydrophobic regions, and our current data suggest that α‐synuclein similarly engages with PRD1 via its hydrophobic segments. Thus, the absence of a comparable hydrophobic interface in the TDP‐43_CTD likely explains the lack of a detectable interaction with PRD1 and, consequently, the absence of any measurable effect on its aggregation. Taken together, these findings suggest that PRD1 acts as a selective modulator of pathological protein aggregation and emphasize the importance of the nature of interaction in dictating the outcome of IDP–partner interactions. They further highlight the need for future studies to elucidate the molecular determinants underlying this specificity.

Fig. 5.

Differential modulation of α‐Syn and TDP‐43_CTD aggregation by PRD1. Schematic illustration of the interaction of the first prion domain of CPEB3 (PRD1) with α‐Syn and TDP‐43_CTD. α‐Syn fibril formation is inhibited by PRD1, but PRD1 could not modulate the aggregation of TDP‐43_CTD.

Author contributions

ATB and VV conceived and designed the study. ATB, MFA, and HVP performed the experiments and contributed to data analysis. ATB drafted the manuscript, and VV revised and edited it. All authors reviewed and approved the final version of the manuscript.

Conflicts of interest

The authors declare no conflicts of interest.

Supporting information

Fig S1 SDS/PAGE images of α‐Syn and PRD1.

Fig. S2. Thioflavin T assay of α‐Syn at different concentrations.

Fig. S3. X‐34 fluorescence assay of α‐Syn with PRD1.

Fig. S4. TEM images of α‐Syn and α‐Syn with PRD1.

Fig. S5. Overlaid HSQC spectra of α‐Syn and α‐Syn with PRD1.

Fig. S6. Intensity ratio α‐Syn and α‐Syn with PRD1.

Fig. S7. HSQC spectra of α‐Syn and α‐Syn with PRD1 at 25°C 1H15N HSQC spectra of 350 μM (A) α‐Syn (B) α‐Syn with PRD1 at 25°C and pH = 7.4.

Fig. S8. Phase separation studies on α‐Syn with PRD1.

Fig. S9. SDS/PAGE image of TDP‐43CTD.

Fig. S10. Chemical shift perturbation plot of α‐Syn with PRD1.

Data accessibility

All the data presented in this study are available within the article and/or the supporting information.