Formation of seeding-competent α-synuclein aggregates in parkin-deficient iPSC-derived human neurons

Abstract

Loss-of-function mutations in PARK2 (parkin) cause early-onset familial Parkinson's disease (PD) and may also contribute to sporadic PD. While Lewy bodies, enriched in aggregated phosphorylated α-synuclein (α-Syn), are typical in PD, their presence in PARK2-mediated PD remains debated. Using human isogenic PARK2^-/- induced pluripotent stem cell-derived neurons, we investigated α-Syn pathology under parkin deficiency. PARK2^-/- neurons showed elevated intracellular aggregated and total α-Syn levels, increased α-Syn release, and higher levels of aggregation-inducing α-Syn seeds. These neurons also displayed more pSer129 α-Syn⁺ inclusions, which were further enhanced by α-Syn preformed fibril (PFF) exposure. Moreover, we identified synaptic loss in the PARK2^-/- neurons, exacerbated by PFF treatment, and dysregulated Ca²⁺ homeostasis consistent with enhanced activity of the smooth endoplasmic reticulum Ca²⁺-ATPase (SERCA). Our data provide an important contribution to the debate on the role of α-Syn in the pathology of PARK2-related PD and challenge the view of PARK2-related PD as a non-synucleinopathy.

Fig. 1. General characterization of differentiated cultures.

A Graphical overview of the applied dopaminergic (DA) differentiation protocol for differentiation of iPSC-derived PARK2^−/− and isogenic control NSCs. B Immunofluorescence staining of day 45 differentiated cultures for tyrosine hydroxylase (TH, dopaminergic neuronal marker), microtubule-associated protein 2 (MAP2, mature neuronal marker), FOXA2 (midbrain dopaminergic neuronal marker), β-tubulin-III (BTUB, general neuronal marker), glial-acidic fibrillary protein (GFAP, astrocyte marker), α-Synuclein (α-Syn), and synaptophysin (SYN, presynaptic marker). The nuclei were stained with DAPI (blue). Scale bar = 50 µm. C–E Quantification of MAP2⁺ mature neurons, TH⁺ DA neurons, and GFAP⁺ astrocytes showed no difference in the differentiation capacity of PARK2^−/− and isogenic control cells. Mean ± SEM, n = 7-15 coverslips from 3-5 independent differentiations. Statistical analysis: Student’s t-test, NS not significant.

Fig. 2. Increased α-Syn aggregation in PARK2^−/− neurons.

A–C ELISA for total α-Syn and aggregated α-Syn (aggregate specific MJF-14-6-4-2 antibody). Mean ± SEM, n = 6 cell lysate samples from 3 independent differentiations. D Release of α-Syn to the culture medium was measured using a sensitive MesoScale immunoassay for α-Syn. Mean ± SEM, n = 5-6 media samples from 3 independent differentiations. E Day 45 differentiated cultures were immunofluorescence-stained for α-Syn phosphorylated at Ser129 (pSer129, green) in combination with MAP2 (purple) to label neuronal cell bodies, and the nuclei were stained with DAPI (blue). Pictures were acquired randomly with a X20 objective on a Zeiss Observer Z1 microscope. Scale bar = 50 µm. F, G Quantification of total pSer129 α-Syn staining per cell and pSer129⁺ α-Syn inclusions/cell. Mean ± SEM, n = 10-14 coverslips from 3 independent differentiations. H The presence of templating active α-Syn species in the neuronal cell cultures was investigated using the α-Syn seed amplification assay (SAA). α-Syn aggregation was monitored by tracking the increase in thioflavin T (ThT) fluorescence over 44 h. The dotted line indicates the threshold for a positive ThT signal (defined as >5000 RFU). Lysates from PARK2^−/− (grey) and control (blue) cells, extracted in RIPA buffer from three independent differentiations, were tested alongside positive controls (α-Syn PFF) and negative controls (α-Syn without RIPA or cell lysate). Each replicate was analysed in quadruplicate, and the graph represents the mean ± SEM. The table presents the incubation time for each sample to reach a fluorescent signal of 5000 RFU (t₅₀₀₀) and the amount of RIPA buffer added to each reaction. The seeding efficiency of cell extracts correlated with the PARK2^−/− phenotype but not with the RIPA content of the samples. I The end ThT signal was measured after 44 h. Mean ± SEM, n = 3 cell lysate samples from 3 independent differentiations. Statistical analysis: Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 3. Exposure to α-Syn preformed fibrils (PFFs).

A Graphical overview of PFF treatment. PARK2^−/− and isogenic control iPSC-derived NSCs were differentiated for 38 days before treatment with human recombinant S129A-α-Syn PFF (14 µg/ml). Excess PFFs were washed out after 24 h, and cells were for 45 days before fixation. B Differentiated cultures were immunofluorescence-stained for pSer129 α-Syn (green) in combination with MAP2 (purple) to label neuronal cell bodies, and the nuclei were stained with DAPI (blue). Pictures were acquired randomly with a X60 objective on a Zeiss Observer Z1 microscope. Scale bar = 20 µm. Scalebar on zoomed images = 10 µm. C Quantification of total pSer129 staining per affected cell. D Affected cells were quantified as the percentage of cells with pSer129 α-Syn inclusion⁺. This analysis was performed on pictures taken with a X20 objective. E Number of pSer129 α-Syn⁺ inclusions per affected cell. F–H Inclusion size distribution according to small (25-100 pixels), medium (101-300 pixels), and large (>300 pixels) inclusions. Mean ± SEM, n = 10 coverslips from 3 independent differentiations. Statistical analysis: Two-way ANOVA, NS = not significant, *p < 0.05, **p < 0.01, ***p < 0.001. I I³ neurons were transfected with lentivirus encoding sgRNA targeting either PARK2, SNCA or a non-targeting guide (control). Following transfection, neurons were treated with PFFs for 14 days, and pSer129 α-Syn levels were quantified. Mean ± SEM, n = 3 independent experiments. Statistical analysis: paired t-test; dotted lines represent the experimental pairs across genotypes.

Fig. 4. Cell types affected by α-Syn aggregation.

A Immunofluorescence staining for pSer129 α-Syn (green) in combination with TH (dopaminergic neuronal marker, red), GFAP (astrocyte marker, red), or GABA (GABAergic neurons, red). The nuclei were stained with DAPI (blue). Arrows indicate pSer129 α-Syn⁺ cells. Pictures were acquired randomly with a X20 objective on a Zeiss Observer Z1 microscope. Scale bar = 50 µm. MesoScale multiplex analysis of cytokine release to the conditioned culture medium from control cultures (B) and PARK2^−/− cultures (C). Data are expressed as percentage relative to the vehicle group for each cytokine. Total cytokine release was evaluated as the mean difference per cytokine between vehicle and PFF-treated cultures and is presented for control cultures in (D) and PARK2^−/− cultures in (E). The baseline difference in cytokine release between untreated control and PARK2^−/− cultures is presented in (F). Mean ± SEM, n = 3-4 samples from 2 independent differentiations. Statistical analysis: Student’st-test, NS not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 5. Reduced synaptic density and increased intracellular Ca²⁺ in PARK2^−/− neurons.

A Immunofluorescence staining for pSer129 α-Syn (green), MAP2 (purple; neuronal cell bodies), and DAPI (blue; nuclei). Pictures were acquired randomly with a X60 objective on a Zeiss Observer Z1 microscope. Scale bar = 20 µm. Scalebar on zoomed neurites = 10 µm. B Immunofluorescence staining for synaptophysin (SYN, red), TH (dopaminergic neurons; green), and DAPI (blue; nuclei). Scale bar = 25 µm. C Quantification of the presynaptic numbers per cell in untreated PARK2^−/− and control cultures. Mean ± SEM, n = 9 coverslips from three independent differentiations. Statistical analysis: Student’s t-test, ***p < 0.001. D Representative transmission electron microscopy (TEM) images of synapses in control and PARK2^−/− neurons. PSD: postsynaptic density, SC: synaptic cleft, SVs: synaptic vesicles. Scale bar = 100 nm. E–G Live-cell Ca²⁺ imaging with Fluo-4-AM in PARK2^−/− and isogenic control cultures treated with thapsigargin (2 µM, added at frame 100 (arrow)). E ∆F/F₀ traces over time normalized to the mean control peak values, showing individual (thin lines) and average (bold) responses from two independent experiments (3 movies in each; 150–200 cells/movie). F Representative timelapse pictures of Ca²⁺ signal at selected timepoints. Scale bar = 50 µm. G Quantification of peak ∆F/F₀ values. Mean ± SEM, n = 6 movies from two independent differentiations. Statistical analysis: Student’s t-test, *p < 0.05. H–J Depolarization with KCl (25 mM) was performed at frame 100 (arrow). H ∆F/F₀ traces over time normalized to the mean control peak values, shown per experiment (thin lines) and across experiments (bold) from two independent differentiations (3 movies recorded in each; 150–200 cells/movie). I Timelapse pictures showing Ca²⁺ signal over time. Scale bar = 50 µm. J Peak ∆F/F₀ values during depolarization. Mean ± SEM, n = 6 movies from two independent differentiations. Statistical analysis: Student’s t-test, NS not significant.