Alpha‐synuclein dynamics in induced pluripotent stem cell‐derived dopaminergic neurons from a Parkinson’s disease patient (PARK4) with SNCA triplication

Alpha‐synuclein is a major component of Lewy bodies, which are a histological hallmark of Parkinson's disease (PD). In this study, we investigated the dynamics of α‐synuclein during the early stage of immature dopaminergic neurons differentiated from human‐induced pluripotent stem cells derived from either a PD patient with SNCA triplication or a healthy donor.

Parkinson's disease (PD) is a neurodegenerative disorder caused by the selective loss of dopaminergic (DA) neurons in the substantia nigra pars compacta (SNc). Lewy bodies (LBs), another histological hallmark of PD, are observed in patients with familial or sporadic PD. The therapeutic potential of reducing the accumulation of α-synuclein, a major LB component, has been investigated, but it remains unknown whether the formation of LBs results in the loss of DA neurons. PARK4 patients exhibit multiplication of the α-synuclein gene (SNCA) without any pathological mutations, but their symptoms develop relatively early. Therefore, study of PARK4 might help elucidate the mechanism of α-synuclein aggregation. In this study, we investigated the dynamics of α-synuclein during the early stage of immature DA neurons, which were differentiated from human-induced pluripotent stem cells (hiPSCs) derived from either a PARK4 patient with SNCA triplication or a healthy donor. We observed increased α-synuclein accumulation in PARK4 hiPSC-derived DA neurons relative to those derived from healthy donor hiPSCs. Interestingly, α-synuclein accumulation disappeared over time in the PARK4 patient-derived DA neurons. Moreover, an SNCA-specific antisense oligonucleotide could reduce α-synuclein levels during the accumulation stage. These observations may help reveal the mechanisms that regulate α-synuclein levels, which may consequently be useful in the development of new therapies for patients with sporadic or familial PD.
PARK4 patients have a multiplication of the α-synuclein gene (SNCA) without any pathological mutation, although their symptoms develop relatively early [11,12]. Therefore, PARK4 might be a useful target to investigate the mechanism through which DA neurons regulate α-synuclein aggregation before the clinical onset of PD and could lead to the development of new therapies for patients with sporadic or familial PD.
In this study, to examine the effects of α-synuclein aggregation on DA neurons, we investigated the dynamics of α-synuclein in DA neurons differentiated from human-induced pluripotent stem cells (hiPSCs) derived from a PARK4 patient with SNCA triplication and a healthy donor. Furthermore, we examined the feasibility of a therapeutic strategy to reduce α-synuclein levels using ASOs.

Ethics statement
This study was conducted in accordance with the principles of the Declaration of Helsinki. The use of hiPSCs was approved by the ethics committee of Osaka National Hospital (approval no. 110 and 120) and Keio University School of Medicine (approval no: 20080016).
SNCA gene copy number analysis

Karyotype analysis
Karyotype analysis was performed by conventional Giemsa staining and G-banding, and diagnosis was performed based on the 2013 international system for human cytogenetic nomenclature (ISCN 2013) [18].

Quantitative RT-PCR (qPCR)
For total RNA extraction, the cells were lysed in QIAzol Lysis Reagent (cat. 79306; Qiagen), and RNA was extracted using a chloroform and isopropanol precipitation method. Total RNA was then treated with Turbo DNase (cat. AM2239; Thermo Fisher Scientific) according to the manufacturer's protocol to reduce genomic DNA contamination. cDNA was synthesized from total RNA using a PrimeScript RT Reagent kit (cat. RR037A; Takara Bio, Shiga, Japan) according to the manufacturer's specifications. The cDNA was subjected to quantitative PCR analysis using gene-specific primers/probes (Tables 1 and 2 Determination of the OCT4/TRA-1-60-positive proportion of hiPSCs hiPSCs were dissociated into single cells using TrypLE, and then, they were fixed in a solution of 4% paraformaldehyde in phosphate buffer (cat. 163-20145; FUJIFILM Wako Pure Chemical) for 15 min at room temperature (RT). Subsequently, the cells were washed three times with PBS. The fixed cells were suspended in a 0.1% solution of Triton X-100 in PBS and seeded into a 96-well plate (cat. 161093; Thermo Fisher Scientific) at a density of 2-3 × 10 4 cells/ well and allowed to settle for 10 min at RT. The 96-well plate was then centrifuged at 300 g for 3 min at RT. The supernatant was removed carefully from each well, and the plate was dried for 10-15 min at RT. Finally, 100 µL of PBS was added to each well, and immunocytochemical staining was performed as described in the next section. The OCT4/TRA-1-60-positive population was evaluated using a high-content screening system (ArrayScan XTI HCA Reader; Thermo Fisher Scientific).

Immunocytochemical staining (ICC)
The fixed cells were permeabilized and treated with blocking buffer containing PBS, 0.1% Triton X-100, and 10% normal goat serum for 1 h at RT. Next, the cells were stained overnight with primary antibodies (Table 3) in blocking buffer at 4°C. Then, the cells were washed three times with 0.01% Triton X-100 containing PBS (TPBS) and were incubated with a solution containing secondary antibodies (Table 3) and DAPI (nuclear stain; cat. D212; Dojindo, Kumamoto, Japan) for 1 h at RT. Finally, the cells were washed three times with TPBS and were observed with a high-content screening system or a confocal microscope (LSM700; Carl Zeiss, Oberkochen, Germany).

Analysis of α-synuclein accumulation
A confocal microscope (LSM700) was used to obtain zstacked fluorescent images of α-synuclein. The z-stacked images were then binarized to extract fluorescent signals of α-synuclein and passed to the mean filter using IMAGEJ software (Fiji package; National Institutes of Health, Bethesda, MD, USA) [24,25]. Accumulated α-synuclein was defined by the outlying data points based on a boxplot of the clump size derived from healthy hiPSC-derived neurons (201B7 and 409B2). Total accumulation was calculated as the sum of accumulated α-synuclein in each image, and it was normalized by the thickness of each z-stack. The values of total α-synuclein accumulation are presented as arbitrary units (a.u.). The resulting values from healthy hiPSCderived neurons and PARK4 hiPSC-derived neurons were compared statistically.

Evaluation of cleaved-caspase-3 in α-synuclein accumulated cells
Cells exhibiting perinuclear α-synuclein accumulation were manually identified in fluorescence images obtained by a confocal microscope (LSM700). We then evaluated whether these cells were positive or negative for cleaved-caspase-3 (n = 81).

Comparison of the average fluorescence signal intensity of cleaved-caspase-3 between 409B2 and PARK4-4 cells
The average fluorescence signal intensities of cleavedcaspase-3 and DAPI were determined from scanned images obtained by confocal microscopy (LSM700). For normalization, the average fluorescence signal intensity of cleaved-caspase-3 was divided by the average fluorescence signal intensity of DAPI in each image. The normalized average fluorescence signal intensity was compared between 409B2 and PARK4-4 cells on day 28 (n = 10).

Antisense oligonucleotide (ASO) treatment
The ASOs used in this study are listed in Table 4. In particular, hSNCA-121-AmNA was specific for α-synuclein Table 1. Primers for SYBR Green-based quantitative PCR.

Gene
Forward primer

Statistical analysis
The statistical significance of two group comparisons was performed using Welch's t-test. The statistical significance of multigroup comparisons was performed using Dunnett's test or two-way analysis of variance (ANOVA) followed by a post hoc multiple comparison test with Benjamini--Hochberg (BH) correction [26]. P values of < 0.05, < 0.01, and < 0.001 were considered to indicate statistical significance. Additional details of the methods are presented in the figure legends.

Results
Characterization of PARK4 patient-derived hiPSCs and healthy donor-derived hiPSCs for investigation of the dynamics of α-synuclein in DA neurons To investigate the dynamics of α-synuclein in DA neurons, which are selectively lost in patients with PD, we established an hiPSC line derived from a PARK4 patient with SNCA triplication (PARK4-4), which had a typical hiPSC-like morphology resembling the two healthy donor-derived hiPSC lines (201B7 and 409B2) (Fig. 1A). These three hiPSC lines were composed of cells with a high population of pluripotent stem cell markers (OCT4 and TRA-1-60) just before starting induction of the midbrain floor plate (Fig. 1B,C). We also confirmed that PARK4-4 cells had an increased copy number of SNCA compared to 201B7 and 409B2 cells (Fig. 1D) but had a normal karyotype (Fig. 1E).
Midbrain floor plate was successfully induced from PARK4 patient-derived hiPSCs and healthy donor-derived hiPSCs We performed experiments to efficiently generate midbrain DA neurons from hiPSC lines using previously reported methods with slight modifications [19,20] ( Fig. 2A). PARK4 patient-derived hiPSCs (PARK4-4) and healthy donor-derived hiPSCs (201B7 and 409B2) were first induced into neural aggregates with properties of the midbrain floor plate for 11 days (Fig. 2B). Thereafter, they continued differentiating to DA neurons for additional 3 days before culturing on Matrigel (Fig. 2B). Quantitative RT-PCR results showed that pluripotency marker genes-Oct4 and NANOG-were downregulated, whereas the midbrain floor plate marker genes-FOXA2, LMX1A, and CORIN-were upregulated on day 11 of midbrain floor plate induction (Fig. 2C). Although the expression of another pluripotency marker gene-LIN28A-was not low, that of the DA neuron marker gene-TH (tyrosine hydroxylase)-was slightly high just before terminal differentiation on day 11 (Fig. 2C). The population of TH-positive DA neuron within pan-neuronal markerpositive neurons was almost the same between 409B2 (25.0%, n = 108) and PARK4-4 (21.5%, n = 149) (Fig. 2D) on day 49 of differentiation. We also observed some remaining SOX2-positive undifferentiated cells and GFAP-positive glial cells in the differentiation culture on day 49 (Fig. 2D). The maturation marker of DA neuron, GIRK2, was observed in both 409B2 and PARK4-4 cells (Fig. 2E).

Higher α-synuclein accumulation in PARK4-4 cells than in 201B7 and 409B2 cells disappeared with time during terminal differentiation
Next, we confirmed the expression levels of α-synuclein on mRNA. We found that the SNCA mRNA level was higher in PARK4-4 than 409B2 cells at the hiPSC stage and on day 14 of differentiation but nearly the same after day 28 (Fig. 2F). We also investigated whether α-synuclein accumulates only in PARK4 patient-derived DA neurons (PARK4-4) but not in healthy donor-derived DA neurons (201B7 and 409B2) during terminal differentiation. In addition to neurites, perinuclear α-synuclein accumulation was readily detected in PARK4-4, but the accumulation was limited in both 201B7 and 409B2 until day 35 (Fig. 3A). These α-synuclein accumulations were identified not only in DA neurons with strong TH positivity but also in weakly TH-positive/TH-negative cells (Fig. 3A). The total amount of accumulated α-synuclein was significantly high in PARK4-4 compared with that in 201B7 on day 21 until day 42 (Fig. 3B). After day 42, however, the level of α-synuclein accumulation in PARK4-4 cells decreased to that in 201B7 and 409B2, but accumulation was still observed on some neurites (Fig. 3A,B). To confirm whether perinuclear α-synuclein accumulation disappeared because of cell death, we immunostained the cells for an apoptosis marker, cleaved-caspase-3. A total of 96.3% of cells with perinuclear α-synuclein accumulations were negative for cleaved-caspase-3 for PARK4-4. We also confirmed that the difference in comparative cleaved-caspase-3/ DAPI intensity between 409B2 and PARK4-4 cells was not significant (Fig. 3C,D).
Treatment with an α-synuclein-specific ASO decreased the levels of α-synuclein mRNA and protein Finally, we investigated whether the use of a recently developed ASO specific for α-synuclein mRNA (SNCA) [8] can decrease the levels of α-synuclein in PARK4-4 cells. Cultures of midbrain DA neurons were fed every other day with terminal differentiation medium without ASOs (untreated control; ASO (−)) or with media supplemented with ASOs (an SNCA-specific or scrambled sequences as negative controls; Fig. 4A) from days 14 to 28. SNCA mRNA expression increased gradually during terminal differentiation in the cells from days 14 to 21 and then significantly decreased in cells treated with an SNCA-specific ASO as well as negative control ASOs from days 21 to 28 (Fig. 4B, left). On both days 21 and 28 (after 7-and 14-day ASO treatment), SNCA mRNA expression levels decreased further in cells treated with an SNCA-specific ASO than in those treated with negative control ASOs or without SNCA silencing [ASO (−)] (Fig. 4B, right). The results of immunofluorescence analysis showed that α-synuclein accumulation obviously appeared in ASO (−) cells and in cells treated with negative control ASOs, but accumulation was extremely limited in cells treated with SNCA-specific ASO (Fig. 4C). Although the total amount of accumulated α-synuclein tended to be lower in cells treated with negative control ASOs than with ASO (−), similar to the results of gene expression analysis (Fig. 4B), the levels were significantly low in cells treated with SNCAspecific ASO (Fig. 4D).

Discussion
Parkinson's disease is a major neurodegenerative disease wherein the pathological symptoms are manifested later in life [1,2]. Although causative genes have been identified for familial forms of PD [3,4], they remain unknown in~90% cases of sporadic PD [5]. Among the pathological genomic loci of PD, PARK4 -which involves SNCA duplication or triplication without pathological mutations-has been identified [11]. As α-synuclein is a major component of LBs, a histological hallmark of PD in addition to loss of DA neurons in the SNc, larger amounts of α-synuclein in the brain are not favorable. In fact, symptom onset in PARK4 patients occurs at earlier ages, for example, 34 years on average [11,27], even without any pathological mutation in the α-synuclein gene. Moreover, numerous genetic variations associated with the sporadic form of PD have been identified based on genome-wide association studies [28]. A recent report showed that the transcriptional deregulation of SNCA is caused by a common PD-associated risk variant located in the noncoding distal enhancer element of SNCA [29]. Since PARK4 patients, as well as those with the sporadic form of PD, do not express neurological symptoms before PD onset, we hypothesized that a similar mechanism of α-synuclein accumulation might be underway in both PD types. In other words, the mechanisms accurately regulating α-synuclein levels before symptom onset might be similar in both types of PD. Thus, newer therapies for sporadic PD could be developed by investigating PARK4 patients because it is reasonable to consider that these α-synuclein regulation mechanisms may be functional at higher levels in DA neurons derived from PARK4 patients with SNCA triplication or duplication [30].
In this study, we investigated α-synuclein dynamics during development in DA neurons differentiated from a PARK4-4 hiPSC line (derived from a PARK4 patient with SNCA triplication) and 201B7 and 409B2 hiPSC lines (derived from a healthy donor) (Fig. 1A--E) [13,15,16]. To induce differentiation to midbrain DA neurons, which are selectively lost in patients with PD, we applied previously developed methods comprising two stages: midbrain floor plate induction and terminal differentiation [19,20] (Fig. 2A). Midbrain floor plates were successfully generated (Fig. 2B,C), and DA neurons were efficiently produced during the terminal differentiation stage not only from 201B7 and 409B2 cells but also from PARK4-4 cells (Figs 2D,E and 3A). A maturation marker of DA neurons, GIRK2, was detected, along with some GFAP-positive glial cells (Fig. 2D,E), under our culture conditions.
The SNCA mRNA level was higher in PARK4-4 cells than in 409B2 cells at the hiPSC stage and on day 14 of differentiation but almost the same on day 49 of differentiation (Fig. 2F). Therefore, α-synuclein was increased in both healthy donor-derived cells and PARK4 patient-derived cells during neuronal differentiation. Interestingly, however, the levels of α-synuclein accumulation at the perinucleus and neurites in PARK4-4 cells were significantly higher than those in 201B7 and 409B2 cells until day 42 of terminal differentiation (Fig. 3A,B). This accumulation of α-synuclein is consistent with that described previously [30][31][32]. Although we considered that we reproduced the pathology of PD, α-synuclein accumulation decreased to levels observed in 201B7 and 409B2 cells by day 49 of differentiation (Fig. 3A,B). Therefore, we hypothesized that this obliteration of α-synuclein accumulation was the consequence of perinuclear α-synuclein accumulation-induced cellular apoptosis. However, only a few cells with α-synuclein accumulation exhibited immunopositivity for cleaved-caspase-3, an apoptotic marker, clearly indicating that α-synuclein accumulation did not significantly induce cell death during terminal differentiation (Fig. 3C,D). Because we did not detect signals for phosphorylated Ser 129 of α-synuclein (data not shown), we considered that this accumulated α-synuclein itself may not be a    pathological species. As hiPSC-derived neurons are considered immature [33], α-synuclein accumulation may be appropriately regulated in the early developmental stage even in neurons with SNCA triplication. Otherwise, we might need a longer culture system or some unrevealed factors to induce pathological aggregation of accumulated α-synuclein. Together, these findings indicated the difficulty of reproducing the full phenotype of late-onset neuronal degenerative disease like PD within a limited short in vitro culture time, and these issues should be addressed in future studies. Although the mechanisms underlying α-synuclein accumulation in LBs remain unknown, therapies aimed at decreasing α-synuclein levels show potential as therapeutic strategies, and several systems have already been developed for this purpose such as SNCA-specific ASOs [8,9] and α-synuclein-specific antibodies [10]. We investigated whether a previously developed SNCA-specific ASO [8] effectively prevents α-synuclein accumulation in PARK4-4-derived DA neurons (Fig. 4A). Because α-synuclein accumulation was obliterated with time during differentiation, the effect of ASOs was analyzed from days 14 to 28 of terminal differentiation (Fig. 4A). The SNCA-specific ASO could downregulate SNCA mRNA (Fig. 4B) and α-synuclein protein expression (Fig. 4C). Therefore, SNCA-specific ASOs may be a feasible chemoprophylactic approach in patients with familial or sporadic PD.
One of the study limitations is that we only obtained snapshots of α-synuclein accumulation in cells during terminal differentiation, and some cells with α-synuclein accumulation were weakly positive or negative in terms of staining for TH. We could not determine whether these cells express TH at later time points or eventually died. Therefore, further studies are needed to elucidate how cells with α-synuclein accumulation remain alive using techniques such as time-lapse imaging.

Conclusions
Midbrain DA neurons, differentiated from PARK4 patient-derived hiPSCs with SNCA triplication, showed a significant amount of α-synuclein accumulation at the perinucleus and neurites compared with that of DA neurons differentiated from healthy donor-derived hiPSCs. However, these α-synuclein accumulations disappeared with time during their terminal differentiation. This phenomenon might further elucidate the mechanism underlying the regulation of α-synuclein levels in PARK4 patients, who are healthy even before the earlier onset of PD, and also in sporadic PD patients who account for~90% of all PD cases. tools, prepared the figures and/or tables, and reviewed drafts of the paper. KB conceived of and designed the experiments, contributed reagents/materials/analysis tools, and reviewed drafts of the paper. TS conceived of and designed the experiments, contributed reagents/ materials/analysis tools, and reviewed drafts of the paper. DI conceived of and designed the experiments, contributed reagents/materials/analysis tools, and reviewed drafts of the paper. HO conceived of and designed the experiments, contributed reagents/materials/analysis tools, and reviewed drafts of the paper. HM conceived of and designed the experiments, wrote the paper, and reviewed drafts of the paper. YK conceived of and designed the experiments, wrote the paper, and reviewed drafts of the paper.