Reduced ER-mitochondrial contact sites and mitochondrial Ca2+ flux in PRKN-mutant patient tyrosine hydroxylase reporter iPSC lines

Endoplasmic reticulum-mitochondrial contact sites (ERMCS) play an important role in mitochondrial dynamics, calcium signaling, and autophagy. Disruption of the ERMCS has been linked to several neurodegenerative diseases, including Alzheimer’s disease (AD), Parkinson’s disease (PD), and amyotrophic lateral sclerosis (ALS). However, the etiological role of ERMCS in these diseases remains unclear. We previously established tyrosine hydroxylase reporter (TH-GFP) iPSC lines from a PD patient with a PRKN mutation to perform correlative light-electron microscopy (CLEM) analysis and live cell imaging in GFP-expressing dopaminergic neurons. Here, we analyzed ERMCS in GFP-expressing PRKN-mutant dopaminergic neurons from patients using CLEM and a proximity ligation assay (PLA). The PLA showed that the ERMCS were significantly reduced in PRKN-mutant patient dopaminergic neurons compared to the control under normal conditions. The reduction of the ERMCS in PRKN-mutant patient dopaminergic neurons was further enhanced by treatment with a mitochondrial uncoupler. In addition, mitochondrial calcium imaging showed that mitochondrial Ca2+ flux was significantly reduced in PRKN-mutant patient dopaminergic neurons compared to the control. These results suggest a defect in calcium flux from ER to mitochondria is due to the decreased ERMCS in PRKN-mutant patient dopaminergic neurons. Our study of ERMCS using TH-GFP iPSC lines would contribute to further understanding of the mechanisms of dopaminergic neuron degeneration in patients with PRKN mutations.


Introduction
Mitochondria play an essential role in the regulation of cellular energy metabolism, redox signaling, and apoptosis.The endoplasmic reticulum-mitochondrial contact sites (ERMCS) have been shown to be associated with a variety of cellular processes, such as lipid biosynthesis, mitochondrial dynamics, calcium signaling, and autophagy (Rowland and Voeltz, 2012;Paillusson et al., 2016;Csordás et al., 2018;Gómez-Suaga et al., 2018).Mitochondrial abnormalities, including disruption of ERMCS, have been suggested to be involved in the pathogenesis of neurodegenerative diseases, such as Alzheimer's disease (AD), Parkinson's disease (PD), and amyotrophic lateral sclerosis (ALS) (Paillusson et al., 2016;Okano and Morimoto, 2022).For example, a reduction in ER-mitochondria associations in PD patients (Paillusson et al., 2017;Liu et al., 2019) and the occasional localization of PD-causing proteins at ERMCS (Ottolini et al., 2013;Guardia-Laguarta et al., 2014;Gelmetti et al., 2017) have been reported.The expression of ERMCS-associated proteins has been shown to be altered in AD brain and AD mouse models (Hedskog et al., 2013).ALS model mice have exhibited impaired ultrastructure of the ERMCS in motor neurons (Watanabe et al., 2016).However, the physiological significance of ERMCS in these diseases has not been elucidated in detail.
The Parkin RBR E3 ubiquitin protein ligase is encoded by the PRKN/PARK2 gene, which is the most common causative gene for the autosomal recessive form of PD.Parkin is a key component of mitochondrial quality control mechanisms (Abou-Sleiman et al., 2006;Narendra et al., 2008;Scarffe et al., 2014).PD patients with PRKN mutations show preferential degeneration of dopaminergic neurons in the substantia nigra pars compacta.Several studies using patient fibroblasts from patients with PRKN mutations have reported ERMCS alteration (Gautier et al., 2016;Basso et al., 2018), but just one study (McLelland et al., 2018) has reported ERMCS abnormalities using dopaminergic neurons derived from PRKN-mutant patient iPSCs.It is still unclear how ERMCS are involved in the cell death of dopaminergic neurons in PRKNrelated PD.
We have previously established tyrosine hydroxylase reporter (TH-GFP) iPSC lines from two control subjects and one PD patient with a PRKN mutation (Yokota et al., 2021).The TH-GFP iPSC lines express GFP specifically in dopaminergic neurons, making them suitable for correlative light-electron microscopy (CLEM) analysis and live cell imaging of GFP-expressing dopaminergic neurons (Yokota et al., 2021).We have shown that small and lowfunctional mitochondria unique to dopaminergic neurons and the preferential death of PRKN-mutant patient dopaminergic neurons under 30 μM a mitochondrial uncoupler, carbonyl cyanide m-chlorophenyl hydrazine (CCCP) for 24 h (Yokota et al., 2021).Meanwhile, we have not yet analyzed ERMCS in PRKN-mutant patient dopaminergic neurons derived from TH-GFP iPSC lines.
In this study, ERMCS were observed in GFP-positive dopaminergic neurons derived from TH-GFP iPSC lines using CLEM and proximity ligation assay (PLA) to clarify the alterations of ERMCS in PRKN-mutant dopaminergic neurons.The PLA suggest that ERMCS were significantly reduced in PRKN-mutant patient lines compared to that in the control lines.
In addition, mitochondrial calcium imaging provided evidence that mitochondrial Ca 2+ flux was decreased in the PRKN-mutant patient lines.These results suggest that dopaminergic neurons from PRKNmutant patients have impaired calcium flux from the ER to the mitochondria due to reduced ERMCS.Our findings would partially help us to understand the mechanisms of dopaminergic neuron degeneration in patients with PRKN mutations.

Human iPSCs
The four TH-GFP iPSC lines used for CLEM, PLA, and Ca 2+ imaging were previously established from the following original iPSC lines (Yokota et al., 2021).201B7, as a control line, was kindly provided by Dr. Shinya Yamanaka at Kyoto University (Takahashi et al., 2007).The other two control lines (WD39 and eKA4), and two PD patient lines with a PRKN mutation (PA9, exon 2-4 homozygous deletions and PB2, exon 6, 7 homozygous deletions) were established by H.O. (Imaizumi et al., 2012;Matsumoto et al., 2016).The PRKN knock-in/knock-out iPSC line (B7PA21) was established by Dr. Minoru Narita at Hoshi University School of Pharmacy and Pharmaceutical Science in a previous report (Kuzumaki et al., 2019).One control iPSC (JB6) and one PD patient iPSC with PRKN mutation (PH13, exon 3 homozygous deletion) were generated from peripheral blood mononuclear cells with CytoTune ™ -iPS 2.0 Sendai Reprogramming Kit (Nacalai) encoding the four factors (Oct3/4, Sox2, Klf4, and c-Myc) using the same method as in previous report (Ishikawa et al., 2022).The use of human iPSCs was approved by the Ethical Committees of Juntendo University School of Medicine (Approval Number 2017032).

Correlative light-electron microscopy
For CLEM analysis, dissociated neurospheres were reseeded on gridded coverslips (Matsunami) after coating with poly-L-ornithine and fibronectin.On day 7-10 of culture, cells were fixed in 2% paraformaldehyde, 0.5% glutaraldehyde, and 50 mM sucrose in 0.1 M phosphate buffer.Brightfield and fluorescence images were captured using a BZ-X710 fluorescence microscope (Keyence).Next, cells were fixed in 2% glutaraldehyde and 50 mM sucrose in 0.1 M phosphate buffer, followed by post-fixation with 1% osmium tetroxide.Fixed cells were dehydrated and embedded in Epon812 (Oken Shoji).Ultrathin sections were cut using an ultramicrotome UC6 (Leica), stained with uranyl acetate and lead citrate, and examined using a Hitachi HT7700 electron microscope (Hitachi).The electron microscope images were analyzed using ImageJ software (https://imagej.nih.gov/ij/).

Proximity ligation assay
PLA was performed using Duolink PLA (Sigma-Aldrich) to quantify ER-mitochondria interactions.On day 7-10 of culture, after reseeding dissociated neurospheres, differentiated cells were fixed with 4% paraformaldehyde in PBS for 10 min and permeabilized with 0.1% Triton X-100 in PBS for 15 min at room temperature.Cells were then blocked with Duolink Blocking Solution for 1 h at 37 °C and stained with primary antibodies for 1 h at 37 °C.The following antibodies were used: rat anti-GFP antibody (MBL; 1:500) for TH-GFP iPSC lines or sheep anti-TH antibody (PelFreez; 1:400) for other iPSC lines, rabbit anti-VDAC antibody (Proteintech; 1:500), and mouse anti-IP3R3 antibody (Santa Cruz Biotechnology; 1:100).Cells were then washed and stained with Alexa Fluor 488-conjugated donkey anti-rat secondary antibody (Jackson ImmunoResearch; 1:400) for TH-GFP iPSC lines or donkey anti-sheep secondary antibody (Thermo Fisher Scientific; 1:800) for other iPSC lines, DAPI (Thermo Fisher Scientific; 1:10,000), and Duolink PLA Probe anti-mouse PLUS and anti-rabbit MINUS for 1 h at 37 °C.PLA signals were detected using the Duolink Detection Reagents FarRed.PLA fluorescence images were captured using a confocal microscope (Zeiss LSM880 with Airyscan).Airyscan z-stack images were processed using a maximum-intensity-projection.

Immunofluorescence staining
On day 7-10 of culture, after reseeding dissociated neurospheres, differentiated cells were fixed with 4% paraformaldehyde in PBS for 10 min and permeabilized with 0.1% Triton X-100 in PBS for 5 min at room temperature.Cells were then blocked with 2% BSA in PBS for 30 min and stained with primary antibodies for 2 h at room temperature.The following antibodies were used: rat anti-GFP antibody (MBL; 1:500), mouse anti-Tom20 antibody (Santa Cruz Biotechnology; 1:100), and rabbit anti-Calnexin antibody (Abcam; 1:1,000).Cells were then washed and stained with Cy3-, Alexa Fluor 488-, or 647-conjugated donkey secondary antibodies (Jackson ImmunoResearch; 1:400 or Thermo Fisher Scientific; 1:800) and DAPI (Thermo Fisher Scientific; 1: 10,000) for 1 h at room temperature.Immunostaining images were taken using a confocal microscope (Zeiss LSM880).Airyscan images were processed using an Airyscan-processing.The fluorescence images were analyzed using a Threshold tool in ImageJ software (https://imagej.nih.gov/ij/) for selection of the staining positive area.

Mitochondrial and cytosolic Ca 2+ imaging
The pLV-EF1α-R-CEPIA3mt containing the R-CEPIA3mt ORF (Addgene plasmid #140464) (Kanemal et al., 2020) was synthesized using Vector Builder.The pLV-EF1α-R-CEPIA3mt and the packaging plasmids were co-transfected into 293T cells using TransIT-293 Reagent (Mirus Bio), and then the supernatants were concentrated using a Lenti-X Concentrator (Clontech) after 4 days of transfection.On day 1 of culture after reseeding dissociated neurospheres derived from the TH-GFP iPSCs, the cells were infected with the concentrated lentivirus at a rate of 1 in 250.On day 8 of culture, the cells were perfused with HBSS and the fluorescence intensity of R-CEPIA3mt was measured via an objective lens (PlanApo ×60, N.A. 1.20; Olympus) and a cooled CCD camera (ORCA-ER; Hamamatsu Photonics) every 1 s for 80 s using Aquacosmos 2.0 (336 × 256 resolution) (Hamamatsu Photonics) for mitochondrial Ca 2+ imaging.For cytosolic Ca 2+ imaging, the cells were incubated with 5 μM Fura 2acetoxymethyl ester (AM) (Dojindo) and 0.01% Pluronic F-127 (Biotium) in HBSS for 30 min at room temperature and perfused with HBSS, and the Fura 2-AM fluorescence ratio (F 340 /F 380 ) was measured via an objective lens (S Fluor ×40, N.A. 0.90; Nikon) and a cooled CCD camera (ORCA-ER) every 1 s for 80 s using Aquacosmos 2.0 (336 × 256 resolution) (Hamamatsu Photonics).Next, the cells were stimulated by perfusion with 10 μM histamine (Wako), the inositol triphosphate-generating agonist, and the fluorescence intensity of R-CEPIA3mt or the Fura 2-AM fluorescence ratio was measured every 1s for 300 s.The ratio of the change in fluorescence intensity was determined using the following equation.ΔF/F 0 = (fluorescence intensity-minimum fluorescence intensity before stimulation)/minimum fluorescence intensity before stimulation.The Fura 2-AM fluorescence was expressed as ratiometric value (R: F 340 /F 380 ).The localization of R-CEPIA3mt to mitochondria in the differentiated cells was confirmed by staining for 30 min with 50 nM Mitotracker Deep Red (Thermo Fisher Scientific).Live cell images were taken using a confocal microscope (Zeiss LSM880).

Statistical analysis
GraphPad Prism 8 software was used for statistical analyses.Differences between groups were evaluated using the unpaired t-test or two-way ANOVA with Bonferroni's multiple comparison test, as appropriate.Statistical significance was set at p < 0.05.

CLEM analysis for the ERMCS in GFPpositive dopaminergic neurons derived from TH-GFP iPSC lines
To examine the ER-mitochondria interface in control and PRKN-mutant patient dopaminergic neurons, we differentiated into dopaminergic neurons derived from two control TH-GFP iPSC lines (201B7 T1-3 and WD39 T1-2) and two PRKN-mutant patient TH-GFP iPSC lines (PB2 T1-1 and PB2 T1-4) (Yokota et al., 2021) using the direct neurosphere conversion method (Imaizumi et al., 2015;Matsumoto et al., 2016;Fujimori et al., 2017;Yamaguchi et al., 2020).We treated the differentiated cells with 30 μM CCCP for 24 h to investigate the change in the ERmitochondria interface in dopaminergic neurons under pathological conditions, because several studies have reported that 30 μM CCCP treatment for 24-48 h exhibited pathogenicity specific to patient dopaminergic neurons (Imaizumi et al., 2012;Suzuki et al., 2017;Yamaguchi et al., 2020;Yokota et al., 2021).We acquired brightfield and fluorescence images of the differentiated cells on gridded coverslips to determine the location of GFP-positive dopaminergic neurons, and then fixed them strongly for electron microscopy using the CLEM method described previously (Yokota et al., 2021) (Figure 1A).
We searched for GFP-positive cells on ultrathin sections using cell shapes and positions in the brightfield and fluorescence images and observed the ER tubules around mitochondria in GFP-positive cells under a transmission electron microscope (Figure 1B).
CLEM analysis of GFP-positive dopaminergic neurons revealed that the CCCP treatment significantly reduced the ER around mitochondria under CCCP treatment in control lines (Figures 1C, D; p = 0.0042).Comparing control and PRKN-mutant patient lines, PRKN-mutant patient dopaminergic neurons tended to have less ERMCS under normal conditions, but this difference was not statistically significant (Figures 1C, D; p = 0.5508).CLEM analysis using TH-GFP iPSCs indicate that dopaminergic neurons under CCCP treatment tend to have less ERMCS than dopaminergic neurons under normal conditions.

PLA for quantification of the ERMCS in dopaminergic neurons derived from TH-GFP iPSC lines
We thought that the EM analysis in ultrathin sections was not sufficient to compare ERMCS between control and patient lines, and we therefore used the in situ proximity ligation assay (PLA) and captured PLA fluorescence z-stack images using a confocal microscope to quantify ERMCS in whole dopaminergic neuronal soma.In PLA, fluorescence signals are only emitted when two target proteins are in close proximity within 40 nm of each other (Söderberg et al., 2006).In our study, PLA was performed using antibodies against VDAC and IP3R3, which are markers for the mitochondrial outer membrane and ER, respectively.PLA in dopaminergic neurons revealed that 30 μM CCCP treatment for 24 h remarkably reduced the VDAC-IP3R3 interaction in both control and PRKN-mutant patient lines (Figures 2A-C; control, p < 0.0001; PRKN, p < 0.0001), similar to the results of CLEM analysis.Comparing control and PRKN-mutant patient lines, PLA in dopaminergic neurons under normal conditions showed that the PRKN-mutant patient lines had significantly less VDAC-IP3R3 interaction than the control lines (Figures 2A, C; p < 0.0001).Furthermore, we found that the VDAC-IP3R3 interaction was significantly reduced in dopaminergic neurons from PRKNmutant patients compared to control dopaminergic neurons under CCCP treatment (Figures 2B, C; p = 0.0029).The difference in the amount of VDAC-IP3R3 interaction between PB2 T1-1 and PB2 T1-4 lines both under normal conditions and CCCP treatment (Figures 2A-C) may possibly be due to the difference in the gene expression specific to each cell lines.
To investigate whether the reduced VDAC-IP3R3 interaction in PRKN-mutant patient lines results from a reduction in the amount of mitochondria and/or ER, we stained the TH-GFP iPSCderived neurons with anti-GFP, anti-Tom20, and anti-Calnexin antibodies.Immunofluorescence staining showed that the immunoreactivity for calnexin in PRKN-mutant lines was not decreased compared to that in control lines (Supplementary Figures S1A-C).Additionally, the immunoreactivity for Tom20 in PRKN-mutant lines was rather increased compared to that in control lines under normal conditions (Supplementary Figures S1A, C; p = 0.0029).However, in PRKN-mutant patient lines the immunoreactivity for Tom20 under CCCP treatment was decreased compared to that under normal conditions (Supplementary Figures S1A-C; p < 0.0001), suggesting that the decrease in VDAC-IP3R3 interaction in PRKN-mutant patient lines under CCCP treatment is probably due to the decrease in the amount of mitochondria.
Taken together, these results suggest that PRKN-mutant patient dopaminergic neurons derived from TH-GFP iPSCs have fewer ERmitochondria interactions than control dopaminergic neurons under normal conditions, and the reduction of ERMCS in PRKN-mutant patient dopaminergic neurons is exacerbated by CCCP treatment.

PLA for the ERMCS in dopaminergic neurons derived from other iPSC lines
To verify whether other patients with PRKN mutations or PRKN-deficient cells also have a reduced ERMCS, we differentiated into dopaminergic neurons from two control iPSC lines (eKA4 and JB6) and two patient iPSC lines with PRKN mutations (PH13 and PA9), and a PRKN knock-in/knock-out iPSC line (B7/PA21) (Imaizumi et al., 2012;Matsumoto et al., 2016;Kuzumaki et al., 2019).The PLA in dopaminergic neurons under 30 μM CCCP treatment for 24 h demonstrated that the VDAC-IP3R3 interactions were markedly reduced in the PA9 and PH13 lines compared with the eKA4 and JB6 lines (Figures 3A, B; p = 0.0004), indicating that ER-mitochondria interactions were also reduced in dopaminergic neurons derived from other patients with PRKN mutations.Similarly, the PLA using the B7/PA21 line showed a reduction of the VDAC-IP3R3 signals in PRKN knock-in/knock-out dopaminergic neurons under CCCP treatment (Figures 3A, B; p = 0.0467), suggesting that the loss of Parkin resulted in the reduction of ERMCS.

Mitochondrial Ca 2+ imaging in GFP-positive dopaminergic neurons derived from TH-GFP iPSCs
The ERMCS has been reported to be associated with calcium flux from the ER into the mitochondria (Bravo et al., 2011;Rowland and Voeltz, 2012;Krols et al., 2016;Paillusson et al., 2016;Gómez-Suaga et al., 2018).To investigate whether mitochondrial Ca 2+ flux is reduced due to decreased ERMCS in dopaminergic neurons from PRKN-mutant patients, we used the red fluorescent genetically encoded mitochondrial Ca 2+ indicator R-CEPIA3mt (Kanemal et al., 2020).The CEPIA3mt can be used for concentrationdependent intramitochondrial Ca 2+ imaging (Suzuki et al., 2014;Kanemal et al., 2020).The lentivirus expressing R-CEPIA3mt was infected into the cells on day 1 after replating dissociated neurospheres derived from the TH-GFP iPSCs.On day 8 after infection, the localization of R-CEPIA3mt to mitochondria in GFP-positive dopaminergic neurons was confirmed by staining the differentiated cells with Mitotracker Deep Red (Figure 4A).The signals of R-CEPIA3mt fluorescence were detected and the intensity of that was measured in GFP-positive dopaminergic neurons derived from all the TH-GFP iPSC lines for mitochondrial Ca 2+ imaging (Figures 4B, C).
To detect the mitochondrial Ca 2+ flux, we stimulated cells with histamine, an inositol triphosphate-generating agonist for Ca 2+ release from the ER to the mitochondria.Histamine stimulation induced an increase in R-CEPIA3mt fluorescence intensity in control GFP-positive dopaminergic neurons, whereas the ratio of change in fluorescence intensity in PRKN-mutant dopaminergic neurons was small compared to that in control lines.(Figures 4C, D; p = 0.0252).To exclude the possibility that the decreased calcium flux to the mitochondria in PRKN-mutant patient lines depends on cytosolic Ca 2+ levels, we stained the TH-GFP iPSC-derived neurons with cytosolic Ca 2+ fluorescent indicator Fura 2-AM and measured the Fura 2-AM fluorescence ratio under histamine stimulation.Cytosolic Ca 2+ imaging showed no significant difference in cytosolic Ca 2+ levels between control and PRKN-mutant patient lines under histamine stimulation (Supplementary Figures S2A, B; p = 0.2037).These findings suggest that the failure of calcium flux from the ER to the mitochondria is not due to low concentrations of cytosolic Ca 2+ in PRKN-mutant dopaminergic neurons.These findings raise the possibility that the impaired ERMCS may be partly responsible for the death of dopaminergic neurons in patients with PRKN mutations.

Discussion
In this study, we showed that CCCP treatment induced a decrease in ERMCS compared to normal conditions in TH-GFP iPSC-derived dopaminergic neurons using CLEM and PLA.It is known that CCCP treatment induces Parkin to ubiquitinate Mfn2, a tethering protein at ERMCS.It was reported that Parkin-mediated ubiquitinated Mfn2 was retrotranslocated from the outer mitochondrial membrane, and that ERMCS was decreased during CCCP-induced mitophagy (McLelland et al., 2018).Further studies on non-ubiquitinated Mfn2 in PRKN-mutant dopaminergic neurons under CCCP treatment are needed to clarify the effect of Parkin deficiency on ERMCS under pathological conditions.
We also demonstrated reduced ERMCS in PRKN-mutant patient dopaminergic neurons derived from TH-GFP iPSCs under both normal conditions and CCCP treatment using PLA.In addition, we performed the quantitative analysis of ERMCS in EM images according to the methods used in previous studies (Watanabe et al., 2016;Paillusson et al., 2017;Parrado-Fernández et al., 2018;Liu et al., 2019).Our CLEM analysis did not show significant difference in ERMCS between control and PRKN-mutant patient dopaminergic neurons.This discrepancy between CLEM and PLA might be due to the difference between the twodimensional analysis using ultrathin sections and the threedimensional analysis using whole cells.Consistent with our results in PLA, a previous study has reported the reduction of ERMCS in Parkin downregulating mouse embryonic fibroblasts and PRKN-mutant patient fibroblasts (Basso et al., 2018).Furthermore, several studies on other PD-causing proteins have reported reduced ERMCS in iPSC-derived dopaminergic neurons with triplication of the α-synuclein gene (Paillusson et al., 2017), PINK1 or DJ-1 knockdown neuroblastoma cells (Parrado-Fernández et al., 2018), and DJ-1 knockout neuroblastoma cells (Liu et al., 2019).However, some studies have reported increased ERMCS in PRKN-mutant patient fibroblasts (Gautier et al., 2016) and PRKN-mutant dopaminergic neurons (McLelland et al., 2018).Whether ERMCS are increased or decreased in PD patients is still controversial.
PLA in PRKN knock-in/knock-out dopaminergic neurons indicated that the loss of Parkin resulted in the reduction of ERMCS.Our results are consistent with those of a previous study, which reported that silencing of Parkin in SH-SY5Y cells impairs ERMCS and Ca 2+ transients (Calì et al., 2013).It has been suggested that Parkin is present at ERMCS and interacts with other ERMCS-associated proteins (Gelmetti et al., 2017).Thus, Parkin deficiency may disrupt ERMCS in dopaminergic neurons.
PLA is a highly sensitive method, but it relies on the specificity of individual antibodies against endogenous proteins.Non-specific binding of the antibodies causes a high false positive rate (Wilson and Metzakopian, 2021).The use of other methods such as split fluorescence reporters of ERMCS in future studies would be valuable for the more precise validation of ERMCS.
Mitochondrial Ca 2+ imaging showed that the mitochondrial Ca 2+ flux was reduced in PRKN-mutant dopaminergic neurons after histamine-induced Ca 2+ release from the ER to mitochondria, suggesting impaired ERMCS-mediated Ca 2+ flux from the ER to mitochondria.Our results are consistent with those of several previous studies that reported decreased mitochondrial Ca 2+ uptake in PRKN-mutant patient fibroblasts (Basso et al., 2018) and DJ-1 knockout cells (Liu et al., 2019).The previous study also demonstrated decreased ATP production in DJ-1 knockout cells, suggesting mitochondrial dysfunction (Liu et al., 2019).In addition, it has been reported that dopaminergic neurons from PRKN-mutant patients show a failure of mitochondrial degradation (Imaizumi et al., 2012;Suzuki et al., 2017;Yamaguchi et al., 2020).The defects and mitochondrial Ca 2+ flux in dopaminergic neurons from PRKN-mutant patients in our study may lead to the accumulation of damaged mitochondria and subsequent cell death of PRKN-mutant dopaminergic neurons.Expression of a linker or treatment with an agonist for the ERMCS could clarify whether the reduced ERMCS affects the cell death of PRKN-mutant dopaminergic neurons under CCCP treatment.
On the other hand, cytosolic Ca 2+ levels were not increased after histamine stimulation despite decreased mitochondrial Ca 2+ flux in PRKN-mutant dopaminergic neurons.A possible cause for this is the decrease in Ca 2+ release from the ER in PRKN-mutant patient dopaminergic neurons.The Ca 2+ release from the ER could not be measured in this study, but the measurement of that in PRKNmutant patient dopaminergic neurons is necessary to be addressed in further studies.
In the present study, we first visualized the mitochondrial Ca 2+ flux specific to dopaminergic neurons using PRKN-mutant patient TH-GFP iPSC lines and R-CEPIA3mt.Our strategy of using TH-GFP iPSCs overcame the heterogeneity of iPSCderived differentiated cells composed of dopaminergic and non-dopaminergic neurons in mitochondrial Ca 2+ imaging.However, the limitation of our study is the small number of TH-GFP iPSC lines used for CLEM and mitochondrial Ca 2+ imaging.Future studies with other TH-GFP iPSC lines are needed to clarify the general differences between the control and PRKN-mutant patients.In conclusion, PLA using TH-GFP iPSC lines revealed reduced ERMCS in PRKN-mutant dopaminergic neurons under both normal conditions and CCCP treatment.Furthermore, mitochondrial Ca 2+ imaging demonstrated a decreased mitochondrial Ca 2+ flux in PRKNmutant dopaminergic neurons, suggesting impaired Ca 2+ flux from the ER to mitochondria.Our findings would provide insights into the pathogenesis of cell death in dopaminergic neurons in PRKN-mutant patients.

FIGURE 1
FIGURE 1 CLEM analysis of GFP-positive dopaminergic neurons derived from TH-GFP iPSCs.(A) The merged brightfield and GFP (left) and GFP (right) images of TH-GFP iPSC-derived cells on the gridded coverslips.Asterisks indicate GFP-positive dopaminergic neurons.Scale bar, 20 µm.(B) The merged GFP and TEM (left) and TEM (right) images of TH-GFP iPSC-derived cells in the same location as (A) in the ultrathin sections.Asterisks indicate GFP-positive dopaminergic neurons.Scale bar, 20 µm.(C) Representative TEM images of GFP-positive dopaminergic neurons under normal conditions (DMSO treatment, top) and 30 μM CCCP treatment for 24 h (bottom) derived control and PRKN-mutant patient TH-GFP iPSCs.ER and mitochondria are painted light yellow and cyan, respectively."PRKN" represents PRKN-mutant patient.Scale bar, 200 nm.(D) Quantitative analysis of ERMCS in GFP-positive dopaminergic neurons from EM images.The graph represents the ratio of mitochondrial length in contact with ER (<30 nm) to the total mitochondrial circumference in dopaminergic neurons (DMSO control; n = 16, DMSO PRKN; n = 16, CCCP control; n = 15, CCCP PRKN; n = 15)."PRKN" represents PRKN-mutant patient.Horizontal lines indicate median values.Statistical significance was evaluated using the two-way ANOVA with Bonferroni's multiple comparison test.**p < 0.01.There was no significant difference in ERMCS between DMSO and CCCP in PRKN-mutant dopaminergic neurons (p = 0.0644).There was no significant difference in ERMCS between control and PRKN-mutant patient lines under normal conditions (p = 0.5508) and CCCP treatment (p > 0.9999).

FIGURE 3
FIGURE 3 PLA in dopaminergic neurons derived from other iPSCs.(A) Representative PLA images of TH-positive dopaminergic neurons under 30 μM CCCP treatment for 24 h derived from other iPSCs.PLA signals represent the interactions between VDAC and IP3R3."PRKN" and "PRKN-KIKO" represents PRKN-mutant patient and PRKN knock-in/knock-out, respectively.Scale bar, 5 µm.(B) Quantitative analysis of the MFI of PLA signals in TH-positive dopaminergic neurons in PLA images.The MFI was measured by surrounding the cell body region of TH-positive cells (control; n = 20, PRKN; n = 22, PRKN KIKO; n = 12) in ZEN software."PRKN" and "PRKN-KIKO" represents PRKN-mutant patient and PRKN knock-in/knock-out, respectively.Horizontal lines indicate median values.Statistical significance was evaluated using the two-way ANOVA with Bonferroni's multiple comparisons test.*p < 0.05, ***p < 0.001.

FIGURE 4
FIGURE 4 Mitochondrial Ca 2+ imaging in GFP-positive dopaminergic neurons derived from TH-GFP iPSCs.(A) Live cell images of R-CEPIA3mt-expressing WD39 T1-2 dopaminergic neurons stained with Mitotracker DeepRed.Scale bar, 10 µm.The boxed areas are enlarged to show the cytoplasmic area of GFP-positive R-CEPIA3mtexpressing dopaminergic neurons.Scale bar of the enlarged images, 1 µm.(B)Representative GFP and R-CEPIA3mt images of the differentiated cells derived from control and PRKN-mutant patient TH-GFP iPSCs before mitochondrial Ca 2+ measurement.Arrows indicate GFP-positive R-CEPIA3mt-expressing dopaminergic neurons."PRKN" represents PRKN-mutant patient.The boxed areas are enlarged to show the cytoplasmic area of GFP-positive R-CEPIA3mt-expressing dopaminergic neurons.(C) Representative time course of the ratio of the change in R-CEPIA3mt fluorescence intensity after histamine treatment in GFP-positive dopaminergic neurons derived from TH-GFP iPSCs.ΔF/F 0 = (fluorescence intensity-minimum fluorescence intensity before stimulation)/ minimum fluorescence intensity before stimulation."PRKN" represents PRKN-mutant patient.(D) Quantitative analysis of maximum Ca 2+ flux in control and PRKN-mutant dopaminergic neurons derived from TH-GFP iPSCs.(F max -F 0 )/F 0 = (maximum fluorescence intensity during stimulation-minimum fluorescence intensity before stimulation)/minimum fluorescence intensity before stimulation.The quantitation was performed for six control dopaminergic neurons and six PRKN-mutant dopaminergic neurons, (Continued )