The neuropathology of Parkinson’s disease (PD) is well characterized by the presence of pathological inclusions of the protein α-synuclein (α-syn), called Lewy bodies (LBs) or Lewy neurites (LNs), as well as the degeneration of dopaminergic neurons located in the substantia nigra pars compacta [1]. The presence of α-syn inclusions is not limited to PD but is also observed in other neurodegenerative diseases, including multiple system atrophy and dementia with Lewy body, which are collectively referred to as α-synucleinopathies. Despite ongoing research, the exact cause of PD is not fully understood, but it is widely believed that the alteration of α-syn into aggregates plays a crucial role in the pathogenesis of PD [2].

A growing body of evidence supports the notion that the progression of PD may be due to the prion-like spread of α-syn from one neuron to another. This theory is based on the observation that Lewy pathology progressively involves more regions of the nervous system, starting in the olfactory bulb and enteric nervous system and eventually spreading to cortical areas [3]. The prion-like propagation of α-syn has been widely studied both in vitro and in vivo [4-6], and this research has advanced our understanding of the mechanisms underlying the pathogenesis of PD and has led to the development of novel therapeutic targets aimed at limiting the progression of PD.

To understand the role of α-syn in the pathogenesis of PD and to develop effective therapeutic strategies, several cellular and animal model systems have been established to monitor α-syn aggregation and propagation. These models include the use of preformed recombinant α-syn fibrils (PFF) to induce intracellular aggregation in cells [7, 8], and the use of cell-derived α-syn to monitor the spread of aggregates between cells [9, 10]. Injection of PFF into various regions of the brain has been reported to result in the formation of LB-like inclusions, providing evidence for the prion-like propagation of α-syn in vivo [11-13]. To further study the intracellular aggregation of α-syn induced by PFF in a rapid and efficient manner, we developed an in vitro model system using A53T α-syn-EGFP overexpressing SH-SY5Y cells and validated its usefulness for high-throughput screening of potential therapeutic targets. This in vitro model provides a simple and convenient platform for the study of α-syn propagation and the development of strategies to limit its progression in PD and related neurodegenerative disorders.

Materials

Monomeric α-syn were prepared as described previously [14, 15]. To prepare α-syn fibrils, five mg/ml monomeric α-syn was incubated at 37℃ with continuous agitation at 2×g for 1 week, and stored at -80℃ until use as α-syn fibrils. The status of α-syn fibrils was determined by thioflavin T binding assay and transmission electron microscopy. α-syn fibrils were sonicated 1 min on ice for on-off switch 1 sec at 100W using an ultrasonic processor VC 505 (Sonics & Materials, Inc., Newtown, CT) before use. Antibody for pSer129 α-syn (#015-25191) were purchased from Wako (Richmond, VA). Antibodies for GFP (SC-9996) and GAPDH (SC-32233) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Saracatinib (11497) was purchased from Cayman Chemical Company (Ann Arbor, MI). Rapamycin (R0395) and chloroquine (C6628) and thioflavin S (T1892-25G) were purchased from Sigma-Aldrich (St. Louis, MO). Hoechst 33342 (H1399) was purchased from Thermo Fisher Scientific (Cleveland, OH, USA).

Cell culture

SH-SY5Y cells were grown in Dulbecco’s modified Eagle’s Medium (DMEM) with 10% fetal bovine serum and maintained at 37℃ in a humidified atmosphere of 5% CO₂ and 95% air. A53T α-syn-EGFP and EGFP only overexpressing SH-SY5Y cells were prepared as described previously [10]. c-Src KD SH-SY5Y cells were prepared as described previously [14]. A53T α-syn-EGFP overexpressing mouse embryonic fibroblast (MEF) cells was generated using lentiviral transfection of A53T α-syn-EGFP in MEF cells and selected using a FACS Aria III (BD Biosciences, Piscataway, NJ). Cytotoxicity assay was performed according to the manufacturer’s instructions (EZ-Cytox, DoGenBio, Seoul, Korea).

Transmission electron microscopy

Two μl of PFF diluted with 18 μl PBS and then pipet to mix. Then, PFF were treated onto a carbon-coated copper grid (#CF200-CU, Electron Microscopy Sciences, Ft. Washington, PA, USA) followed by negative staining with 2% uranyl acetate for 2 min. The grid was gently washed three times 1 min with distilled water. After drying the grid, it was observed by SIGMA500 electron microscope (Carl Zeiss, Oberkochen, Germany).

Western blot

Cells were lysed in ice-cold RIPA buffer (25 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% sodium deoxycholate, 1% Triton X-100, and 0.1% SDS) and a protease inhibitor mixture (Calbiochem, San Diego, CA) for 20 min on ice after sonication for 3 sec. Lysates were cleared by centrifugation at 15,928×g for 30 min at 4℃. The supernatants were collected, mixed with sample buffer, resolved by SDS-PAGE, transferred to a nitrocellulose membrane, and immunoblotted with the indicated antibodies. They were then detected using an enhanced chemiluminescence system (West Save Gold, AbFrontier, Seoul, Korea).

Immunocytochemistry and confocal microscopy

Cells cultured on the coverslips were washed three times with PBS and fixed with 4% paraformaldehyde for 10 min at room temperature; the fixed cells were then washed with PBS and incubated with PBS containing 0.1% Triton X-100 for 10 min at room temperature. After washing with PBS, the cells were blocked with PBS containing 1% bovine serum albumin (GenDEPOT, Katy, TX) for 1 h at room temperature and then incubated overnight with primary antibody at 4℃. Preparations were then stained with fluorescence-conjugated secondary antibody for 1 h. The samples were incubated with 20 μg/μl Hoechst 33342 for 10 min. They were mounted and observed under a confocal microscope (Zeiss, Jena, Germany). For thioflavin S staining, the samples were incubated with 0.025% thioflavin S in 50% ethanol for 8 min. Then, the samples were washed with 80% ethanol 3 times for 5 min. They were mounted and observed under a confocal microscope.

Imaging analysis

To analyze A53T α-syn-EGFP aggregate puncta, five random fields were selected and more than 250 cells per group were analyzed. Imaging was analyzed with Metamorph (versions 7.7.8.0 or 7.10.4.407) as manufacturer’s recommendation. Briefly, the captured confocal images were imported into Metamorph. The confocal images were divided into RGB pictures using a module called ‘color separate’. The images divided by RGB were combined with z-stack using a module called ‘Stack arithmetic’. The combined images of the stack were analyzed using a module called ‘Granularity’. In the granularity module, the minimum, maximum diameter, and intensity of dots were measured. Nuclear images based on Hoechst were precluded. A dot was defined as 1.66 μm to 7.43 μm in diameter, 0.68341 μm² or more in area, and 1,200 or more in intensity.

Statistical analysis

All values are expressed as means ± SEM. Statistical significance was evaluated using the unpaired t-test or one-way ANOVA (Graphpad Prism software, San Diego, CA, USA)

The understanding of prion-like α-syn propagation has opened up new possibilities for therapeutic targets in the treatment of PD and related neurodegenerative disorders. Several in vitro model systems have been established to monitor α-syn propagation [7-10, 16]. However, some of these model systems have limitations for high-throughput screening. For example, some in vitro models have relied on further staining for phosphorylated Ser129 α-syn to analyze the formation of α-syn inclusions [8, 16, 17]. Additionally, some in vitro models have used PFF with liposomes to enhance internalization into cells [8, 18-21]. However, these models do not allow for the identification of the structures on the plasma membrane involved in PFF internalization or the PFF-induced signaling pathways because PFF are also internalized via receptor-mediated endocytosis [22]. In some in vitro models, the number of cells with α-syn aggregation has been used as an index without further analysis [18-21]. These limitations highlight the need for more advanced in vitro model systems for high-throughput screening of α-syn propagation.

In the present study, two SH-SY5Y stable cell lines were established, one overexpressing EGFP only and the other overexpressing A53T α-syn-EGFP. The use of A53T-α-syn, a variant of α-syn that has been linked to familial PD, allowed for acceleration of aggregation [23], while the EGFP tag facilitated efficient analysis without additional manipulation. Before use, we sonicated PFF to facilitate the internalization (Fig. 1A). Both cell lines showed no significant intensity differences in EGFP expression (Fig. 1B). Next, A53T α-syn-EGFP SH-SY5Y cells were treated with PFF without liposomes at different doses for three days. Exogenously added PFF induced the formation of A53T α-syn-EGFP aggregates in a dose-dependent manner (Fig. 1C). There was no difference in cell viability after 5 days of treatment with 1 μM PFF (Fig. 1E). Various parameters of aggregation were analyzed, including the number of dots per cell, size of dots, intensity of dots, and percentage of cells containing aggregation puncta. The results showed that PFF increased all of these indexes in a dose-dependent manner, with different responses for each index. The number of dots per cell and percentage of cells with aggregates increased at lower concentrations of PFF, while the size of dots and intensity of dots increased at higher concentrations (Fig. 1C). Further analysis showed that when treated with 250 nM PFF, the number of dots per cell and percentage of cells with aggregates increased from as early as 1 day and reached a plateau, while the size of dots and intensity of dots increased from 4 days after treatment. (Fig. 1D), suggesting that PFF increased all indexes of A53T α-syn-EGFP aggregation puncta in a dose- and time-dependent manner.

To verify the specificity of A53T α-syn-EGFP aggregation puncta, A53T α-syn-EGFP SH-SY5Y cells were treated with α-syn monomer at various times and doses. No aggregation puncta were observed (Fig. 2A), suggesting that PFF specifically induces aggregation puncta, as supported by previous studies [8, 24]. Additionally, treating SH-SY5Y cells overexpressing EGFP only with PFF did not result in aggregation puncta (Fig. 2B), indicating that α-syn is necessary for the formation of such puncta, also supported by a previous study [25]. To examine the cell type specificity, MEF cell line, well-known for its ease of manipulation and usefulness in genetic manipulation studies, overexpressing A53T α-syn-EGFP was generated, and A53T α-syn-EGFP SH-SY5Y cells and A53T α-syn-EGFP MEF cells showed no significant intensity differences in EGFP expression (Fig. 2C). These cells showed similar effects to SH-SY5Y cells when treated with PFF, but more PFF was required to induce the same degree of aggregation in MEF cells (Fig. 2D). This may be due to differences in the types of receptors expressed on each cell, which can affect PFF-induced aggregation [22], as well as the expression level of endogenous α-syn [26].

Interestingly, when we treated A53T α-syn-EGFP cells with liposome-free PFF, we observed an increase in the number of smaller aggregates in the cytosol of both cells, compared with previous studies [8, 18-21]. Although the exact reason for this discrepancy is still not fully understood, it is possible that the difference lies in the method of delivering PFF. Previous studies utilized liposome-aided PFF, which directly introduces PFF into the cytosol, whereas we used liposome-free PFF. α-Syn species can enter cells through various mechanisms, including receptor-mediated endocytosis, and once inside, they must escape to the cytosol in order to form aggregates with endogenous α-syn [27], which may contribute to the discrepancy in results. Further study will be needed to better understand the underlying mechanisms and confirm these observations.

LBs and LNs are known to contain α-syn that is structured as β-pleated sheets and stained by thioflavin S [28]. Additionally, LBs and LNs are also marked by phosphorylated α-syn at Ser129 [29]. When we stained PFF-treated A53T α-syn-EGFP cells with thioflavin S, we found that most aggregation puncta were colocalized with thioflavin S (Fig. 3A). However, when we stained these cells with an antibody against pSer129 α-syn, not all aggregation puncta were stained with this antibody. We observed that a few aggregation puncta were colocalized with pSer129 α-syn after one day and that the colocalization between aggregation puncta of A53T α-syn-EGFP and pSer129 α-syn increased in a time-dependent manner (Fig. 3B). Our observations showed that aggregation puncta with higher intensity and larger size were more likely to be colocalized with pSer129 α-syn (Fig. 3C). Given that the intensity and size of these aggregation puncta increased over time, it is possible that the aggregation puncta of A53T α-syn-EGFP which are not colocalized with pSer129 α-syn, represent those during the maturation step to LB-like inclusions, considering that LB formation involves several stages [30-33]. Additionally, it has been suggested that phosphorylation at Ser129 is not necessary for the nucleation-dependent polymerization of α-syn within cells, as the frequency of inclusion bodies observed in seed-transduced cells expressing S129A α-syn was similar to that in seed-transfected cells expressing wild-type α-syn cells [34]. It has also been reported that pS129 α-syn occurs after initial protein aggregation [35], which is consistent with our results.

To further validate the relevance of our model for high-throughput screening of potential therapeutic targets, we conducted several experimental tests using both pharmacological interventions and genetic modifications. It has been previously established that aggregated α-syn can be degraded by the autophagy and lysosomal pathways [36]. To evaluate the effect of autophagy on the formation of PFF induced A53T α-syn-EGFP aggregates, we treated A53T α-syn-EGFP SH-SY5Y cells with rapamycin and chloroquine, which are known to enhance and inhibit autophagy, respectively, in the presence of PFF for 1 day. Our results showed that treatment with rapamycin effectively reduced the number of aggregation puncta per cell and the percentage of cells containing such puncta but did not affect the size or intensity of the puncta (Fig. 4A), which is consistent with previous studies [14, 37]. Treatment with chloroquine increased the number of aggregation puncta per cell, the size and intensity of the puncta and the percentage of cells containing such puncta, although the number of aggregation puncta per cell was not statistically significant (Fig. 4B). We have reported that c-src regulates α-syn propagation and inhibition of c-src activity attenuates α-syn propagation in both in vitro and in vivo systems [38]. When we treated A53T α-syn-EGFP SH-SY5Y cells with saracatinib, a c-src/c-abl dual inhibitor, and PFF, we observed that reduced the number of aggregation puncta per cell and the percentage of cells containing such puncta but did not affect the size or intensity of the puncta (Fig. 4C). Additionally, we treated A53T α-syn-EGFP/c-src KD SH-SY5Y cells with PFF, the similar results with the treatment with saracatinib were also observed (Fig. 4D), suggesting that this model system will be useful as a screening method for developing strategies against α-syn propagation.

However, this model system has some limitations. Due to its resolution limitations, multiple dots grouped together may be recognized as a single dot. Diffuse large aggregates may not be recognized as dots, which can result in incorrect of the number, size and intensity of dots. Despite these limitations, it can be useful due to its rapid and convenient assay.

Recently, it has been shown that different strains of aggregation-prone proteins can lead to distinct phenotypes [39-41]. In a previous study, a model system using tau RD-YFP cells was used to categorize diverse strains of tau [42]. Similarly, A53T α-syn-EGFP cell lines can be used to study the characteristics of different PFF samples.

In conclusion, we have developed an in vitro model system using A53T α-syn-EGFP overexpressing SH-SY5Y cells. Treatment with PFF induced the formation of aggregation puncta in these cells, which were analyzed using four indices. Four indices are the reliable indicators of the effectiveness of interventions against α-syn aggregation in the one-day treatment model to minimize the screening time. This simple and efficient in vitro model system can be utilized for high-throughput screening to discover new targets for inhibiting α-syn propagation.

This research was supported by the National Research Foundation of Korea (NRF) grants funded by the Korean government (Ministry of Science and ICT) (Grant No. NRF-2019R1A5A2026045).