Glucocerebrosidase, a Parkinson´s disease-associated protein, is imported into mitochondria and regulates complex I assembly and function

doi:10.21203/rs.3.rs-1521848/v1

Download PDF

Article

Glucocerebrosidase, a Parkinson´s disease-associated protein, is imported into mitochondria and regulates complex I assembly and function

https://doi.org/10.21203/rs.3.rs-1521848/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Mutations in the lysosomal enzyme β-glucocerebrosidase (GCase), which cause Gaucher's disease, are the most frequent genetic risk factor for Parkinson’s disease (PD). Here, we employed global proteomic and single-cell genomic approaches in stable cell lines as well as induced pluripotent stem cell (iPSC)-derived neurons and midbrain organoids to dissect the mechanisms underlying GCase-related neurodegeneration. We demonstrate that a fraction of GCase can be imported from the cytosol into the mitochondria via the translocase of the outer mitochondrial membrane (TOM) import machinery via recognition of internal mitochondrial targeting sequence-like signals. In mitochondria, GCase promotes the assembly and function of mitochondrial complex I (CI). Furthermore, GCase interacts with the mitochondrial quality control proteins HSP60 and LONP1. Disease-associated mutations impair the assembly and function of CI and enhance the interaction with the mitochondrial quality control machinery. These findings reveal a previously unknown function of GCase in mitochondria and suggest that defective CI activity and energy metabolism may drive the pathogenesis of GCase-driven neurodegeneration.

Mutations in the acid β-glucocerebrosidase (GBA1) gene cause the lysosomal storage disease Gaucher's disease (GD), which is an autosomal recessive condition that can present with both systemic and neurological symptoms. Three clinical GD subtypes have been identified: nonneuropathic (type I), acute neuropathic (type II), and chronic neuropathic (type III). GBA1 mutations have been classified as mild or severe according to their association with GD type I or II and III. After the initial reports of an increased incidence of Parkinson’s disease (PD) in patients affected by GD type I and in their family members, GBA1 mutations have been identified as the most frequent genetic risk factor for PD ^{1, 2}. To date, more than 350 pathogenic GBA1 mutations have been linked to GD, some of which are also linked to PD. Mutations associated with the most severe neuronopathic forms of GD are linked to a higher risk of developing PD ³. Among GBA1 mutations, 4 missense variants (p.E326K, p.T369M, p.N370S, and p.L444P) account for ∼87% of PD cases ⁴. Interestingly, GBA1 variants commonly observed in PD patients, such as the p.E326K variant, are not associated with GD ^{5, 6}. The p.E326K variant displays a limited impact on GCase activity and a lower penetrance compared to the severe mutation p.L444P. PD patients carrying E326K mutations have milder motor complications and a higher risk of cognitive decline ^{7, 8}. Despite the link between GBA1 and brain disease, the mechanisms underlying neurodegeneration and disease severity in mutation carriers are still elusive. GBA1 encodes glucocerebrosidase (GCase), a lysosomal enzyme that catalyzes the hydrolysis of glucosylceramide into glucose and ceramide. GBA1 effects may arise from several pathways, including lysosomal dysfunction, sphingolipid dyshomeostasis, α-synuclein (A-SYN) aggregation and defects in autophagy and protein trafficking ⁹. Furthermore, impairment of mitochondrial metabolism and dynamics has been observed in vivo as well as in a variety of cellular models of GCase deficiency and patient neurons ^{10, 11, 12, 13}.

Here, we dissected GBA1-related neurodegeneration by investigating GCase-interacting proteins via a global proteomic approach. Furthermore, we explored how distinct disease-causing mutations alter the profile of GCase interactions. Using molecular and biochemical approaches, we identified a novel function of GCase as a mitochondrial protein with a role in maintaining respiratory chain complex I (CI) integrity and cellular energy homeostasis in cell lines as well as patient induced pluripotent stem cell (iPSC)-derived dopaminergic (DA) neurons and midbrain organoids. Furthermore, we provide a mechanism for mitochondrial LONP1 protease in the folding and degradation of mitochondrial GCase. Importantly, by combining brain organoids with single-cell RNA sequencing (scRNAseq), we provide evidence for a role of GCase in the disruption of neuronal CI function and mitochondrial energy metabolism.

Analysis of the GCase interactome in stable T-Rex HEK cell lines reveals novel mitochondrial interactors

To obtain insight into the molecular mechanisms that link GCase to brain disease, we set out to identify the interactome of wild-type (WT) and mutant GCase using quantitative proteomic analysis. As most GBA1 mutations lead to a misfolded enzyme that is rapidly degraded via ER-associated degradation by the proteasome ^{14, 15}, we generated stable Flp-In™T-REx™-HEK 293 cell lines (henceforth, T-Rex HEK cells) expressing WT or mutant GCase (E326K or L444P) as a V5-FLAG-tagged protein using a tetracycline-inducible system (Figure 1A). This model provided a sufficient level of expression of both WT and mutant GCase (Figure 1B). Protein levels and GCase activity were significantly decreased in E326K- and L444P- compared to WT-GCase T-Rex HEK cells, and the degree of reduction was in accordance with the severity of the mutation (Figure 1B, Supplementary Figure 1A, B). Next, we performed FLAG immunoprecipitation followed by isotopic labeling (tandem mass tag, TMT) and MS analysis. GCase interactors were defined as proteins that were quantified by ≥2 unique peptides and enriched 1.5-fold over the control (Supplementary Dataset 1). First, we investigated the interactome of WT-GCase. The compiled list contained known GCase interactors, such as LIMP2, calnexin, calreticulin, and DNAJB11. Analysis of the 100 top hits of the WT-GCase interactome showed that proteins from different compartments, including the ER, Golgi, lysosomes, and vesicles, interact with GCase (Figure 1C). Interestingly, 19 out of the top 100 binding partners of WT-GCase were proteins associated with mitochondria (Figure 1C, Supplementary Dataset 1). The enrichment analysis of the biological processes (BP) revealed several mitochondrial pathways (Figure 1D). Interestingly, the analysis of the only interactome study of endogenous WT-GCase, which was performed using SILAC proteomics in HeLa cells ¹⁶, revealed that 8% of endogenous GCase-interacting proteins were mitochondrial proteins (Supplementary Figure 1C). Using GCase-FLAG immunoprecipitation followed by western blotting, we validated the interaction with cytosolic chaperones involved in protein folding prior to mitochondrial import (HSC70) as well as mitochondrial outer membrane (TOM70), inner membrane (TIM23 and ATP5B), and mitochondrial matrix (HSP60 and LONP1) proteins (Figure 1E). The interaction of GCase with HSP60 and LONP1 was confirmed by endogenous coimmunoprecipitation (CoIP) of GCase (Supplementary Figure 1D). Furthermore, we confirmed the specific interactions of WT-GCase with selected proteins from the cytosol, ER, and mitochondria (Supplementary Figure 1E).

Interactome and whole proteome analysis reveals the enrichment of mitochondrial proteins in L444P- and E326K-GCase mutants

Next, we compared the TMT-based quantitative interactomes of WT-, L444P-, and E326K-GCase in T-Rex HEK cells. Interestingly, among the dysregulated interactors of WT and mutant GCase, we identified proteins involved in mitochondrial protein quality control, including LONP1, as well as the mitochondrial complex I (CI) assembly factor TIMMDC1 (Figure 1F and Supplementary Dataset 1). In addition, we performed a proteome-wide analysis of TMT-labeled whole-cell lysates from WT-, E326K-, and L444P-GCase T-Rex HEK cells. The analysis of the differentially expressed proteins (DEPs) revealed an abundance of mitochondrial proteins in WT-, E326K-, and L444P-GCase T-Rex HEK cells compared to the control (15%, 25%, and 20%, respectively) (Supplementary Figure 2A, B and Supplementary Dataset 2). Gene ontology (GO) analysis of DEPs revealed an enrichment of proteins linked to mitochondrial energy metabolism and the mitochondrial membrane in the E326K- and L444P-GCase proteomes (Supplementary Figure 2C).

GCase localizes to mitochondria

The TMT-based interactome dataset supports the hypothesis that both WT and mutant GCase can be imported into mitochondria. To confirm the mitochondrial localization of GCase, we performed subcellular fractionation of WT-, E326K-, and L444P-GCase T-Rex HEK cells, and we subjected isolated mitochondria to proteinase K digestion and digitonin permeabilization followed by western blotting (Figure 2A). Possible contamination of the mitochondrial fraction with lysosomal proteins was prevented by digitonin lysis and proteinase K digestion. The lysosomal proteins hexosaminidase B and LAMP1 were successfully removed by this treatment, whereas GCase and the mitochondrial matrix protein LONP1 were still present in this fraction (Figure 2A). To confirm the mitochondrial localization of GCase, T-Rex HEK cells overexpressing WT or mutant GCase were stained for TOM20 and GCase followed by expansion microscopy (ExM), which allows superresolution imaging on a conventional fluorescence microscope ¹⁷ (Supplementary Figure 3A).

GCase contains internal MTS-like signals

As GCase does not have a conventional N-terminal mitochondrial targeting signal (MTS), we employed the TargetP predictive score consecutively for each residue to identify potential internal MTS-like signals (iMTS-ls) ¹⁸. TargetP analysis revealed the presence of 2 iMTS-ls sites with a score >0.6 (Figure 2B). To validate the potential role of these iMTS-ls in GCase mitochondrial import, we employed the split GFP system ¹⁹, where the first 10 β-strands of GFP (GFP_1-10) were targeted to mitochondria through linkage with a mitochondria-targeting sequence (MTS-GFP_1-10). MTS-GFP_1-10 was integrated into T-Rex HEK cells under a doxycycline inducible promoter (Figure 2C) and the correct localization of MTS-GFP_1-10was confirmed by assessing the colocalization between LONP1 and MTS-GFP_1-1048 h after doxycycline induction (Figure 2C). To test whether GCase traffics to mitochondria, the 9-aa linker and S11β-GFP were added at the C-terminus to the WT sequence of GCase (GBA1-β11) (Figure 2D). As a positive control, we introduced the MTS of Cox8A at the same location as the V5-Flag-tag of the N-terminus and the S11β-strand of GFP to the C-terminus in WT-GCase (MTS-GCase). Transient overexpression of WT, E326K, or L444P GBA1-β11 in cells expressing MTS-GFP_1-10 led to GFP signaling colocalizing with MitoTracker Red (Figure 2E). The deletion of the iMTS-ls with the highest score predicted by TargetP (dMTS-GCase) abolished GCase mitochondrial localization, supporting its role in the mitochondrial import of GCase (Figure 2E).

GCase is imported into mitochondria by the TOM protein import machinery

Most mitochondrial proteins are imported by the mitochondrial protein import machinery (TOM, TIM23, mitochondrial HSP70). Upon translocation through the TOM channel, proteins are transferred to the presequence translocase of the inner membrane (TIM23 complex) ²⁰. Based on the interactome analysis, GCase interacts with several components of the mitochondrial import machinery as well as mitochondrial chaperones and proteases (Figure 3A and Supplementary Dataset 1). Using a CoIP assay in T-Rex HEK cells, we confirmed that both WT and mutant GCase interact with HSC70, a cytosolic chaperone protein belonging to the HSP70 family involved in mitochondrial import ²¹ (Figure 1E, 3B). Furthermore, CoIP experiments confirmed the interaction of WT and mutant GCase with TOM70 and TIM23 (Figure 1E, Figure 3C-D). Interestingly, L444P-GCase showed a significantly higher interaction with TOM70 (Figure 3C). The GCase interactome revealed an increased interaction of L444P-GCase with the mitochondrial protease LONP1 compared to WT- and E326K-GCase (Figure 1H and Supplementary Dataset 1). CoIP experiments performed in T-Rex HEK cells confirmed the increased interaction of L444P-GCase with LONP1 compared to WT- and E326K-GCase (Figure 3E). We also detected a significantly increased interaction between L444P-GCase and the mitochondrial chaperone HSP60 (Figure 3F). Furthermore, immunostaining for the mitochondrial protease LONP1 followed by ExM confirmed the interaction between GCase and LONP1 and revealed increased colocalization between GCase and LONP1 in L444P-GCase cells compared to WT-GCase and E326K-GCase cells (Supplementary Figure 3B). To investigate the role of HSC70 and TIM23 in the folding and mitochondrial import of GCase, we employed shRNA-mediated knockdown (KD) in T-Rex HEK cells expressing the L444P mutation, which we found to have a more significant impact on the GCase interaction with mitochondrial proteins. Both, HSC70 and TIM23 KD, led to a significantly decreased interaction of L444P-GCase with LONP1 (Supplementary Figure 3C-F), supporting their role in the mitochondrial import of GCase.

GCase localizes to mitochondria and interacts with mitochondrial chaperones and proteases in human iPSC-derived neurons

To exclude potential artifacts caused by protein overexpression and to confirm these results at the endogenous level in a disease-relevant model, we generated and characterized a set of isogenic human induced pluripotent stem cells (iPSCs) carrying homozygous and heterozygous L444P GBA1 mutations, E326K heterozygous GBA1 mutations, an artificial severe recombinant variant containing the p.L444P mutation (RecNcil), and corresponding isogenic controls (Supplementary Figure 4A-C and Supplementary Table 1). First, we sought to validate the increased interaction between endogenous mutant GCase and LONP1 as well as HSP60. To this end, lysates from isogenic iPSC-derived neural precursor cells (NPCs) generated from L444P/L444P GBA1 iPSCs and isogenic controls were subjected to CoIP using an antibody against GCase. Western blot analysis confirmed the higher interaction between the severe L444P-GCase with LONP1 and HSP60 compared to WT-GCase (Supplementary Figure 4D). Furthermore, to validate the mitochondrial localization of GCase at the endogenous level, isogenic NPCs were then differentiated into midbrain dopaminergic (DA) neurons (Supplementary Figure 4E). No significant differences in the iPSC differentiation potential were observed between mutants and controls (Supplementary Figure 4F). Analysis of ExM immunofluorescence images confirmed the increased colocalization between L444P-GCase and LONP1 in iPSC-derived DA neurons (Supplementary Figure 4G, H).

GCase regulates OXPHOS complex I assembly in HEK cells and iPSC-derived dopaminergic neurons

Whole-cell lysate TMT proteomic analysis revealed dysregulation of mitochondrial CI pathways (Supplementary Figure 2A). Furthermore, the GCase interactome showed a decrease in the interaction between E326K- and L444P-GCase and TIMMDC1, which is involved in mitochondrial CI assembly ²² (Figure 1H). CoIP followed by western blot analysis confirmed the decreased interaction between E326K- and L444P-GCase and TIMMDC1 compared to WT-GCase in T-REX cells (Figure 4A, B). Similarly, the interaction between GCase and NDUFA10, a CI subunit, was significantly reduced in E326K- and L444P-GCase cells compared to WT-GCase cells (Figure 4A, B). These data suggest the involvement of GCase in CI assembly. CoIP confirmed the interaction of GCase with TIMMDC1 at the endogenous level in NPCs (Figure 4C). To determine whether GCase plays a role in maintaining the higher-order integrity of the CI complexes, we employed Blue native-PAGE to visualize the intact supercomplexes by immunoblotting to detect its subunit NDUFA9 in NPCs and DA neurons differentiated from L444P/L444P and E326K/WT iPSCs and their isogenic controls as well as GBA1 knockout (KO) iPSCs. CI-containing supercomplex was decreased in GBA1 KO and GBA1 mutant NPCs and DA neurons, suggesting a role for GCase in CI assembly (Figure 4D).

GCase modulates CI activity and respiration in HEK cells and iPSC-derived neurons

Next, we sought to further investigate the impact of GCase on CI activity. To this end, we first employed WT-, E326K-, and L444P-GCase T-Rex HEK cells. Interestingly, overexpression of GCase led to an increase in CI activity, with a significant reduction in E326K- and L444P-GCase compared to WT-GCase cells (Supplementary Figure 5A). In parallel, we measured the oxygen consumption rate (OCR) in WT-, L444P-, and E326K-GCase T-Rex HEK cells. Similar to that observed for CI activity, overexpression of GCase led to an increase in basal, ATP-linked, and maximal respiration (Supplementary Figure 5B, C). However, the OCR was significantly decreased in L444P and E326K cells compared to WT cells (Supplementary Figure 5B, C). We validated these results in DA neurons generated from GBA1 iPSCs and their corresponding isogenic controls. CI activity and the OCR were significantly decreased in L444P and E326K iPSC-derived neurons compared to the corresponding isogenic controls (Figure 4E and Supplementary Figure 5D).

GBA1 mutant midbrain organoids show alterations in CI function and perturbations of energy metabolism pathways

To investigate GBA1-related pathways in a model that better resembles the complexity of human disease, GBA1 mutant iPSC lines and corresponding isogenic controls were differentiated into midbrain organoids. To improve midbrain organoid differentiation, we developed a protocol combining a previous method that promotes midbrain floor-plate formation ²³ with a recent strategy based on the biphasic activation of WNT signaling, which promotes the reproducible 2D differentiation of midbrain DA neurons from pluripotent stem cells ²⁴ (Figure 5A). At 18 days in vitro (DIV), embryoid bodies showed uniform neuroectoderm formation, and the organoids grew up to 5-6 mm in diameter after 65 days (Figure 5B). Midbrain organoids derived from GBA1 mutant iPSCs and corresponding isogenic controls displayed similar sizes and organization (Figure 5B, 5C). Developing organoids showed tubular structures containing midbrain progenitors expressing OTX2 (Figure 5C, Supplementary Figure 6A). Cells expressing tyrosine hydroxylase (TH), the rate-limiting enzyme in DA synthesis, were observed after day 35, and the number and complexity of cellular processes increased over time (Figure 5D, Supplementary Figure 6A). We observed S100B-positive cellular processes starting at 35 days after induction (Supplementary Figure 6B). Long-term culture midbrain organoids showed DA neuron maturation and the formation of neuromelanin starting at 70 DIV, with a peak at 120 DIV (Supplementary Figure 6C, D). Midbrain organoids showed the expression of synaptic markers (Supplementary Figure 6E). GBA1 mutant organoids showed a significant reduction in GCase activity and protein levels as well as increased A-SYN levels compared to isogenic controls at 42 DIV (Figure 5E-G, Supplementary Figure 6F, G). To gain insight into cell-type-specific mechanisms, we transcriptionally profiled GBA1 mutant midbrain organoids and gene-corrected controls at the single cell level using 10X Genomics Chromium scRNA-seq. Based on the evidence of significant changes in A-SYN levels (Figure 5G), organoids were analyzed at 42 DIV. Normalization and integration were performed to reduce the batch effect and compare the genotypes. We identified 21 cell clusters present in all lines in both mutant and gene-corrected iPSC-derived organoids (Supplementary Dataset 3). After cluster annotation, subclusters were merged into five main clusters (Figure 6A), namely, progenitor cells such as radial glia-like cells (MSX1, PAX3, SOX2, and RFX4) as well as neural progenitor cells (NEUROG1 and NEUROD4), neuronal (DCX and NEFM), astrocyte (TNC and AQP1), and oligodendrocyte (SOX10 and APOD) clusters (Figure 6A-C). Overall, the cell type composition was comparable among mutant and isogenic control organoids (Figure 6B). Upon subclustering, we identified 8 neuronal subclusters, including two distinct DA neuronal clusters, two GABAergic (GAD1, GAD2, and SLC32A1) and two glutamatergic (SLC17A6 and GRIN2B) neuronal clusters and an undefined neuronal cluster (DCX, SYP, and VAMP2) (Supplementary Figure 7A, B). Synaptic transcription factor genes were highly expressed within the neuronal clusters (Figure 6D). To dissect the chronological development of midbrain organoids, we analysed organoids at 42, 65, and 100 DIV and constructed cell fate trajectories by using the lineage decision method Monocle 3 (Supplementary Figure 8A, B). The trajectory proceeded along two branches, towards cells expressing radial glia markers (SOX2, RFX4) or neural progenitor markers (NEUROG1, NEUROD4) (Supplementary Figure 8C). At 100 DIV, the trajectories led to two distinct states, oligodendrocytes and astrocytes (SOX10, APOD, AQP1, S100B) as well as pericytes (PDGFRB, DCN) (State 1), and neurons expressing high levels of DCX, EN1 and LMO3 (State 2) (Supplementary Figure 8C). Interestingly, such a late emergency of pericyte clusters has also been described in the developing human midbrain ²⁵. Confirming neuronal maturation, cells belonging to State 2 expressed high levels of CALB1, SYP, SNCA, GAD1, and TH (Supplementary Figure 8C). Differential gene expression and pathway analysis of neuronal clusters in scRNA-seq data revealed a profound dysregulation of metabolic pathways across isogenic pairs, with a stronger impact on mitochondrial ATP-related pathways in the severe mutant recombinant L444P GBA1 organoids (Figure 5E and Supplementary Figure 9). In this respect, pathway analysis revealed a selective dysregulation of mitochondrial energy metabolism in DA neuron clusters from GBA1 mutant organoids compared to isogenic controls (Supplementary Figure 9), suggesting that severe GBA1 mutations drive selective mitochondrial disturbances in DA neurons. Such selective impairment of metabolism in DA neurons was not evident in E326K and L444P mutant organoids. However, more than 50% of the neuronal mitochondrial DEGs in all isogenic couples belonged to respiratory chain complexes (Supplementary Dataset 4). In line with our findings in stable HEK cell lines and iPSC-derived neurons, CI activity was significantly reduced in GBA1 mutant organoids compared to isogenic controls (Figure 6F). To further validate the findings of the interactome analysis in midbrain organoids, we performed immunostaining for the mitochondrial protease LONP1 followed by ExM in the L444P-GCase mutant and gene-corrected controls, which confirmed the interaction between GCase and LONP1 and revealed increased colocalization between GCase and LONP1 in L444P/L444P organoids (Figure 6G).

Mitochondrial LONP1 protease contributes to the degradation of mutant GCase

Proteome, CoIP, and ExM experiments showed that the mitochondrial LONP1 protease interacts with GCase and that GBA1 mutations significantly increase this interaction. Therefore, we sought to dissect the role of LONP1 in the regulation of mitochondrial GCase. To this end, we treated T-Rex HEK cells overexpressing V5-Flag-GCase (WT, L444P, or E326K) with non-cytotoxic concentrations of 2-cyano-3,12-dioxooleana-1,9-dien-28-oicacid (CDDO-Me), which inhibits LONP1 protease ²⁶, but not the 26S proteasome ²⁷. As expected, CDDO treatment induced a significant upregulation of the mitochondrial unfolded protein response (UPR^mt) (Supplementary Figure 10A). No significant changes in the expression level of UPR^mt genes were observed when comparing WT, L444P, and E326K cells (Supplementary Figure 10A). Interestingly, CDDO treatment caused a significant reduction in CI activity in WT-GCase T-Rex cells, whereas mutant GCase cells displayed a significant increase in CI activity upon CDDO treatment (Supplementary Figure 10B). Given the link between GCase activity and A-SYN ²⁸ and the role of LONP1 in the degradation of intramitochondrial A-SYN ²⁹, we treated mutant and isogenic control DA neurons with Alexa Fluor 594-labeled A-SYN preformed fibrils (PFFs) and assessed A-SYN internalization by live-cell imaging in the presence or absence of CDDO (Supplementary Figure 10C). E326K and L444P GBA1 mutant neurons showed increased levels of intracellular A-SYN PFFs compared to isogenic controls, suggesting an increased uptake or defects in the degradation of A-SYN (Figure 7A). Upon CDDO treatment, we observed a significant increase in intracellular A-SYN PFFs in isogenic control neurons, whereas E326K and L444P GBA1 did not show significant changes (Figure 7A). To assess selective mitochondrial PFF accumulation, we treated iPSC-derived neurons with CDDO prior to incubation with A-SYN PFFs, and we performed live-cell imaging with MitoTracker. CDDO treatment led to an increase in intramitochondrial A-SYN PFFs in isogenic control and E326K mutant iPSC-derived neurons (Figure 7A). Furthermore, we observed a significant decrease in CI activity in isogenic control neurons upon CDDO treatment, whereas a nonsignificant increase was observed in mutant neurons (Figure 7B).

Here, we identified a novel function of GCase beyond the lysosome, which links this disease-relevant lysosomal protein to mitochondria. Our data provide clear evidence that GCase is associated with mitochondria. We show that the HSC70/TOM70 pathway mediates the mitochondrial import of GCase. TOM70 is an essential component of the import machinery of mitochondrial proteins with iMTS-ls ³⁰, and cytosolic ATP-dependent HSC70/HSP70 chaperones play a key role in preprotein targeting to mitochondria ²¹. Interestingly, we identified an increased interaction between mutant GCase and proteins of the mitochondrial protein quality control systems, namely, HSP60 and LONP1. Most pathogenic GBA1 mutations are missense mutations, and the mutant enzyme has sufficient residual enzymatic activity when properly folded and trafficked to the lysosome. Hence, it is possible that the increased interaction of mutant GCase with HSP60 and LONP1 reflects a mitochondria-based quality control system for cytoplasmic proteins ^{31, 32}. Indeed, it has been shown that aggregation-prone proteins are partially imported and degraded by mitochondria ³¹. Furthermore, cytosolic misfolded proteins are associated with and can be transported into mitochondria ³³. Based on this evidence, mutant GCase would be transported and degraded within mitochondria ³¹. It is also interesting to note that GBA1 mutations did not cause significant upregulation of UPR^mt genes. Besides its proteolytic activity, LONP1 also shows an intrinsic chaperone-like function and cooperates with mitochondrial HSP70 in the folding pathway in vitro ³⁴. Our results suggest that under normal conditions, LONP1 might function, to a larger degree, in GCase folding within mitochondria. In the presence of GBA1 mutations, the increased chaperone and proteolytic LONP1 activity may promote the degradation of misfolded/damaged intramitochondrial GCase. LONP1 inhibition ameliorates CI activity in GBA1 mutant HEK cells, while the opposite effect is observed in WT-GCase cell models. Similar results were obtained in iPSC neurons, even though the increase in CI activity upon LONP1 inhibition in GBA1 mutant cells was not significant. These data support the hypothesis that, in the presence of mutant GCase, LONP1 proteolytic activity is more pronounced than its chaperone function. As an alternative explanation, the increased interaction between mutant GCase and LONP1 may interfere with LONP1 folding properties leading to mitochondrial protein aggregation, which can also partially explain the decrease in CI activity and A-SYN intramitochondrial accumulation in iPSC-derived neurons (Fig. 7C). It is interesting to note that LONP1 inhibition did not cause an increase of intramitochondrial A-SYN PFF levels in iPSC-derived neurons carrying the severe L444P mutation. These results would suggest that the burden of misfolded GCase may overloads the capacity of mitochondrial proteases. The regulatory mechanisms controlling the interaction between LONP1 and folding GCase substrates versus GCase proteolytic substrates should be further explored.

As LONP1 is also involved in the folding of CI subunits ³⁴ and it cooperates with the AAA protease CLPXP in the degradation of complex I subunits in depolarized mitochondria ³⁵, its overload by mutant GCase may further contribute to the observed CI instability and mitochondrial energy defects. Under physiological conditions, GCase plays a functional role in regulating energy cellular homeostasis by interacting with TIMMDC1 and by promoting mitochondrial CI activity. Interestingly, findings from Guarani et al. already indicated a possible interaction between GCase and TIMMDC1 in human cell lines ²². GD and PD mutations negatively interfere with the interaction between GCase and CI subunits. This would explain the mechanisms underlying decreased CI activity, which has been observed in both GCase-deficient and mutant models ^{11, 13}. These data suggest a major contribution of CI and mitochondria to GBA1-driven neurodegeneration. TIMMDC1-associated CI assembly defects have been linked to a decrease in mature CI ^{22, 36}. Furthermore, CI assembly defects lead to an accumulation of CI smaller building blocks ³⁶. In line with this evidence, we have previously shown an increase in CI subunits in iPSC-derived neurons carrying GBA1 mutations compared to isogenic controls ¹³. Furthermore, the MS data reported in this study show that overexpression of mutant GCase is associated with an increase in mitochondrial proteins, including CI subunits, compared to control conditions. Interestingly, our results show that the severe L444P variant has a significantly higher impact on the degree of GCase/LONP1 interaction compared to the milder E326K variant, suggesting a stronger activation of the chaperone and proteolytic activity of this mitochondrial protease. Both variants affect CI stability and function as well as cellular respiration, with the L444P variant having a stronger, although not significant, impact compared to the E326K.

Several studies suggest mitochondrial impairment in PD, although it is still unclear whether such mitochondrial defects are primary causes of neurodegeneration or secondary effects of it. A very recent study showed that CI dysfunction may drive PD degenerative processes and clinical phenotypes in a mouse model ³⁷ and this observation is consistent with the finding of significant and specific CI deficiency in human PD substantia nigra ^{38, 39}. Our data further support the hypothesis that CI dysfunction may be a key pathogenetic driver of PD. From a therapeutic perspective, our data point toward increasing GCase levels and modulating the mitochondrial quality control system as potential therapeutic approaches in GBA1-derived neurodegeneration. Due to the decrease in GCase activity that has been observed in sporadic cases, patients affected by dementia with Lewy bodies, and aged individuals ^{40, 41, 42}, such a mechanism could be of broad therapeutic relevance. In summary, our studies link GCase toxicity directly to mitochondrial bioenergetics and propose targeting GCase mitochondrial localization as a promising therapeutic approach for neurodegeneration.

Constructs and generation of stable cell lines

Constructs were purchased from IDT (gBlocks Gene Fragments) and subcloned into pcDNA5/frt/to (Thermo Fisher Scientific, # V652020). Site-directed mutagenesis was performed according to the manufacturer’s protocol (Agilent, QuickChange II XL), and base pair exchange was confirmed by Sanger sequencing. Flp-In™ 293 T-Rex cells (Thermo Fisher Scientific, # R78007) were grown in media composed of Dulbecco’s modified Eagle’s medium (DMEM, Sigma–Aldrich), 10% fetal bovine serum (Gibco), 1% GlutaMax (Gibco) supplemented with 100 mg/ml Zeocin (InvivoGen) and 15 mg/ml blasticidin (InvivoGen). Inducible Flp-In T-Rex 293 cells were generated according to the manufacturer’s protocol (Thermo Fisher Scientific). Selection was performed with DMEM supplemented with 15 mg/ml blasticidin and 100 mg/ml hygromycin B Gold (InvivoGen) 48 h after transfection and continued until the expression of the gene of interest was induced by treating the cells with 50 ng/ml doxycycline hyclate (Sigma–Aldrich) for 48 h. List of primer sequencing is provided in Supplementary Table 2.

Coimmunoprecipitation

Cells were detached using Accutase and washed 2x with ice-cold phosphate-buffered saline (PBS, Sigma–Aldrich). Cells were lysed in 1xTBS (pH7.4) containing 0.5% NP40 and protease/phosphatase inhibitor cocktail (Pierce, #A32959). The washed Protein G Agarose beads (Merck, #16-266) were preincubated with the respective antibody for endogenous PD while rotating for 2 h. Lysates were immunoprecipitated with M2 anti-flag agarose resin (Sigma–Aldrich, #A2220), antibody-decorated beads against the respective protein (for endogenous CoIP), or control IgG for 2 h at 4°C while rotating. For Flag coimmunoprecipitation, Flag was eluted with Flag Peptide (Sigma–Aldrich, #F3290) by moving on a wheel and spun for 1 min at 1000 × g. Control immunoprecipitation using only Protein G agarose was performed. The supernatant was collected and subjected to TMT labeling or western blot analysis. For endogenous coimmunoprecipitation, Protein G agarose beads (Merck, #16-266) were washed and preincubated with the antibody against the respective protein overnight. Next, the beads were washed three times with binding buffers containing 150 mM NaCl and 50 mM Tris-HCl. Elution was performed by boiling the beads twice with 2x Laemmli buffer. Then, the beads were spun for 1 min at 10,000 × g, and the supernatant was collected and subjected to western blot analysis. The antibodies are listed in Supplementary Table 3.

TMT-based quantitative proteomics

Lysates were pretreated with Benzonase ^® Nuclease (Sigma) before further processing for LC-MS analysis. Samples were incubated with dithiothreitol at 56°C for 30 min (10 mM in 50 mM HEPES, pH 8.5) to reduce cysteine. This step was followed by sample alkylation with 2-chloroacetamide (20 mM in 50 mM HEPES, pH 8.5) while protected from light at RT for 30 min. The SP3 protocol was used as published by Hughes et al. for processing the samples ⁴³. The proteins were digested on beads using Trypsin (sequencing grade, Promega) overnight at 37°C. The ratio between trypsin and proteins was 1:50. TMT11plex Isobaric Label Reagent (Thermo Fisher) was used as described in the manufacturer’s protocol. For sample clean up, an OASIS® HLB µElution Plate (Waters) was used. Six fractions were the outcome of offline high pH reversed-phase fractionation with an Agilent 1200 Infinity high-performance liquid chromatography system equipped with a Gemini C18 column (3 μm, 110 Å, 100 x 1.0 mm, Phenomenex). A trapping cartridge (µ-Precolumn C18 PepMap 100, 5 µm, 300 µm i.d. x 5 mm, 100 Å) and an analytical column (nanoEase™ M/Z HSS T3 column 75 µm x 250 mm C18, 1.8 µm, 100 Å, Waters) were attached to the UltiMate 3000 RSLC nano LC system (Dionex). For 6 min, continuous flow of trapping solution (0.05% trifluoroacetic acid in water) onto the trapping column at 30 µL/min allowed peptide trapping. Elution of trapped peptides was achieved by running solvent A (0.1% formic acid in water, 3% DMSO) with a constant flow of 0.3 µL/min through the analytical column, with an increasing percentage of solvent B (0.1% formic acid in acetonitrile, 3% DMSO). The Nanospray Flex™ ion source in positive ion mode allowed direct attachment of the valve of the analytical column to an Orbitrap Fusion™ Lumos™ Tribrid™ Mass Spectrometer (Thermo Fisher Scientific). A Pico-Tip Emitter 360 µm OD x 20 µm ID; 10 µm tip (New Objective) was used to introduce the peptide into the Fusion Lumos using an applied spray voltage of 2.2 kV. The capillary temperature was 275°C. The resolution of the orbitrap used in profile mode was 120,000, and the mass range was set to 375-1500 m/z during the full mass scan. The maximum filling time was 50 ms. For data-dependent acquisition (DDA), the resolution of the Orbitrap was set to 30,000, and the fill time was set to 94 ms. The ion number was limited to 1x10⁵ ions. A normalized collision energy of 36 was applied. Profile mode was used to acquire MS² data. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD032155.

MS data analysis

Data were processed with IsobarQuant (https://github.com/protcode/isob/archive/1.0.0.zip) and Mascot v2.2.07 (http://www.matrixscience.com/server.html; RRID:SCR_014322). The UniProt Homo sapiens database (UP000005640, https://www.uniprot.org/proteomes/UP000005640) was used as the proteome database. The search criteria were determined as follows: carbamidomethyl (C) and TMT11 (K) (fixed modification), acetyl (N-term), oxidation (M) and TMT11 (N-term) (variable modifications). For the full scan (MS1), a mass error tolerance of 10 ppm was set, and a spectrum of 0.02 Da was used for MS/MS (MS2). Two skipped cleavages by trypsin were the maximum permitted. For protein identification, the following criteria had to be fulfilled: a) at least two distinctive peptides were recognized per protein and b) the peptide length was required to be at least seven amino acids long. The false discovery rate (FDR) at the peptide and protein levels was set to 0.01. To filter out nonspecific proteins, a limma-based differential analysis was performed comparing FLAG and IgG control samples. Hit and candidate proteins were defined based on the following criteria. Candidate proteins were defined with an FDR ≤ 0.2 and a FC ≥ 1.5, and hit proteins were based on an FDR < 0.05 and a FC > 2. Statistically enriched hit and candidate proteins were analyzed using Panther pathway analysis software version 14.0 (FDR <5%) (http://www.pantherdb.org; RRID:SCR_004869). Mitochondrial proteins were identified according to the Human Mitocarta 3.0 database (https://www.broadinstitute.org/files/shared/metabolism/mitocarta/human.mitocarta3.0.html; RRID:SCR_004869). The subcellular localization of the top 100 candidate proteins was assigned according to the GeneCards database version 5.6 (https://www.genecards.org; RRID:SCR_002773) by using a confidence level ³ 4. The visualization of the protein subcellular localization was generated with Cytoscape version 3.9.0 (https://cytoscape.org; RRID:SCR_003032). The WT-GCase interactor list was used as the source node, fold-changes as the source attribute, and the subcellular localization as the target node. The distance and distribution between the nodes were randomly assigned.

iMTS-ls profile generation

Internal N-terminal matrix-targeting-like signal (iMTS-ls) analysis was performed as described in Boos et al. ¹⁸. Multiple truncated GCase protein sequences were generated by sequentially removing N-terminal amino acids. GCase suffix sequences were then subjected to TargetP prediction (TargetP 2.0) in standard FastA format. The mTP scores obtained were plotted against the corresponding amino acid position.

Split-GFP mitochondrial localization

For mitochondrial targeting of GFP_1-10(MTS-GFP_1-10), the N-terminal matrix-targeting signal (MTS) of mitochondrial subunit VIII of cytochrome c oxidase (Cox8A) was added at the N-terminus of the GFP_1-10β-strand sequence, as described in Calì et al. ⁴⁴. The MTS-GFP_1-10construct was obtained from Integrated DNA Technologies IDT and cloned into pcDNA™5/FRT (Thermo Fisher). The stable T-Rex HEK cell line was generated as described above. For GCase split-GFP, a 27-bp linker followed by the GFP S11 sequence ⁴⁴ was added at the C-terminus of the GCase cDNA sequence. For MT-GCase split-GFP, we cloned a construct in which we inserted the mitochondrial targeting sequence of Cox8A along with a 9-amino-acid (aa)-long linker at the N-terminus and the 11^th β-strand sequence of GFP at the C-terminus of GCase. Constructs were obtained from VectorBuilder. E326K or L444P mutation was inserted into the WT construct by site-directed mutagenesis using the QuickChange XL kit (Agilent) according to the manufacturer’s protocol. For mtGFP_1-10induction, 2x10^4 mtGFP_1-10 T-Rex HEK cells were seeded on Geltrex (Thermo Fisher)-coated chambered cell culture slides (Ibidi) for live-cell imaging or immunostaining and treated with 50 to 200 ng/ml doxycycline, respectively. GCase split-GFP constructs were transfected with Viafect (Promega) according to the manufacturer’s protocol.

Live-cell imaging

For live-cell imaging, cells were washed once with OptiMEM (Gibco) and then incubated for 30 min at 37°C with 100 nM MitoTracker red CM-H₂Xros (Thermo Fisher Scientific) in OptiMEM. Cells were washed once with OptiMEM, in which the cells were kept for imaging. Images were acquired using a Leica TCS SP8 confocal microscope (Leica, Germany) equipped with a 100× /1.4 numerical aperture oil-immersion objective. Images were analyzed using Diffraction PSF 3D and DeconvolutionLab2 plugins in Fiji-ImageJ version 2.3.0/1.53q (https://fiji.sc; RRID:SCR_002285). For A-SYN PFF experiments, iPSC-derived neurons were treated with 0.25 mM Alexa Fluor 594-labelled PFFs (594-PFF) with or without cotreatment with 1 mM CDDO-Me (Cayman Chemical). After 24 h, the cells were incubated with 100 nM MitoTracker Green (Invitrogen, MA, USA) in neuronal medium for 30 min at 37 °C. A CellBrite™ Steady 488 Membrane Staining Kit (Biotium) was used to visualize cell membranes according to the manufacturer’s instructions. Images were acquired using a Leica TCS SP8 confocal microscope (Leica, Germany) with a 63 × /1.4 numerical aperture oil-immersion objective. Z-stacks were acquired for calculation of the PFF particle area and fluorescence intensity. For each condition, 5-8 images were acquired from at least four independent experiments. Data were obtained from n > 20 cells per experiment per condition. For the quantification of colocalization and image processing, images were analyzed using the “Analyze particles” and “EzColocalization” plugins in Fiji-ImageJ.

Generation of induced pluripotent stem cells and gene correction

Skin fibroblasts from all patients were obtained with informed consent approved by the Ethics Committee of the Medical Faculty and the University Hospital Tübingen. Skin fibroblasts were reprogrammed by nucleofection with pCXLE-hOct3/4 (RRID:Addgene_27076), pCXLE-hSK (RRID:Addgene_27078) and pCXLE-hUL (RRID:Addgene_27080) plasmids ⁴⁵ using the Amaxa nucleofection kit for human dermal fibroblasts (Lonza, VPD-100) and program P-022 of the Nucleofector 2b (Lonza). Nucleofected fibroblasts were plated in 6-well plates coated with Matrigel (Corning) in DMEM supplemented with 10% FBS (Gibco) and 1% GlutaMAX Supplement (Gibco). The following day, the medium was changed to DMEM +/+ (DMEM with 10% FBS and 1% GlutaMAX Supplement and 1% Pen/Strep (Millipore)) supplemented with 2 ng/ml recombinant basic human fibroblast growth factor (FGF2, Peprotech). On day 3 or 4 postnucleofection, the medium was changed to E8 medium composed of DMEM F12 with HEPES (Gibco), 128 ng/ml ascorbic acid (Sigma–Aldrich), 1x insulin-transferrin-selenium (ThermoFisher Scientific), 10 ng/mL FGF2 (Peprotech), 500 ng/ml heparin (Sigma–Aldrich), and 2 ng/ml TGFβ1 (Peprotech). E8 medium was supplemented with 100 µm sodium butyrate and 0.1% Pen/Strep. Colonies started to appear from day 14 onward. Induced pluripotent stem cells (iPSCs) were cultured on Vitronectin XF (StemCell Technologies) in E8 medium. The gene correction for the L444P mutation was performed as previously described ⁴⁶. The crRNA for the gene correction of the E326K mutation was designed with an online CRISPR design tool according to Ran et al ⁴⁷. The guides and ssODN sequences are listed in Supplementary Table 2. One hour before nucleofection, 10 µM Rock inhibitor was added to the iPSC medium. 240 nM crRNA (IDT): Atto550-labeled tracrRNA (IDT) duplex was complexed with 124 µM Cas9 to form the ribonucleoprotein complex (RNP complex). iPSCs (1.6x10^6) were nucleofected with the RNP complex and 16 µg of ssODN using 100 µl of Ingenio nucleofection solution (Mirus) with program B-016 of Nucleofector 2b. Following nucleofection, the cells were FACS sorted for Atto550-positive cells using a FACS Aria II with a 100-µm nozzle. After sorting, 1x10^4 cells were plated per 10-cm dish. Colonies were picked and sequenced by Sanger sequencing. The top 5 possible exonic off-target effects predicted by https://cctop.cos.uni-heidelberg.de:8043/ (RRID:SCR_016963) were checked by Sanger sequencing.

Differentiation of iPSCs into neuronal precursor cells and dopaminergic neurons

The generation of neuronal precursor cells (NPCs) from patient and isogenic control iPSCs as well as the differentiation from NPCs to dopaminergic neurons was performed according to Reinhardt et al. ⁴⁸. NPCs were cultured in N2/B27 medium, consisting of DMEM-Ham’s F12 medium and neurobasal medium (1:1), 0.5% N2 (Gibco), 1% B27 (Gibco), 1% P/S and 1% GlutaMAX supplemented with 3 µm CHIR 99021 (Axon Medchem), 0.5 µM puromycin (Merck Millipore), and 150 µM ascorbic acid (AA, Sigma–Aldrich). NPCs were split in a Matrigel-coated well every 5-6 days at a ratio of 1:10. To start differentiation, 1.25x10^6 cells were split into a 6-well and the following day the medium was changed to differentiation medium [N2/B27 medium supplemented with 100 ng/ml FGF8 (Peprotech), 1 µM phorbol 12-myristate 13-acetate (PMA) and 200 µM AA]. On day 8, the medium was changed to maturation medium [N2/B27 medium supplemented with 10 ng/ml brain-derived neurotrophic factor (BDNF) (Peprotech), 10 ng/ml glial cell-derived neurotrophic factor (GDNF) (Peprotech), 1 ng/ml TGF-β3 (Peprotech), 0.5 mM dibutyryl cAMP (dbcAMP) (Applichem), and 200 µM AA] supplemented with 1 µM PMA for 2 days. From day 10 onward, the cells were cultured in maturation medium without PMA. Maturation was reached after 14 days in maturation medium.

Western blot analysis

Cells were collected and washed once with PBS. The pellets were lysed, and the protein concentration was determined using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). Protein (20-30 µg) was mixed with 6x Laemmli containing 12.5% β-mercaptoethanol (Roth). Protein samples, 20-30 µl of eluate of the Flag precipitate or the eluate of the endogenous PD were loaded on self-casted 7.5-15% acrylamide gels (Applichem) or on precast NuPage 4-12% Bis-Tris Protein Gels (Thermo Fisher Scientific). Proteins were transferred to PVDF membranes. The PVDF membranes were blocked in TBS 0.1% Tween containing 5% milk (nonfat dried milk powder, PanReac AppliChem) or 5% bovine serum albumin (Serva). The primary antibody was diluted in 5% Roche Block solution (Roche) or 5% BSA in TBS-T supplemented with 0.04% sodium azide. Secondary antibodies were diluted in blocking solution. The antibodies are listed in Supplementary Table 3.

Blue native electrophoresis and in-gel activity assay

Mitochondrial pellets were isolated from NPCs and iPSC-derived neurons and processed for blue native electrophoresis (BN-PAGE) as previously described ⁴⁹ with minor modifications. In brief, samples were solubilized with digitonin at a detergent-to-protein ratio of 4:1, and 40 mg of mitochondrial protein was loaded per lane in precast NativePAGE^TM 3-12% Bis-Tris gels (InvitrogenTM). After electrophoresis, proteins were transferred to PVDF membranes at 30 V overnight and probed with the appropriate antibody.

Midbrain organoid generation

Midbrain-like organoids were generated by a protocol adapted from Jo et al. ²³ (dx.doi.org/10.17504/protocols.io.5qpvobr8xl4o/v1). At day 0, iPSC colonies were dissociated into single cells using Accutase. In each well of a 96-well U-Bottom low-attachment plate, 10^4 iPSCs were seeded in a 1:1 mix of DMEM F12 and neurobasal medium supplemented with 0,5% N-2 supplement, 1% B27 supplement without vitamin A, 1% NEAA, 0.01% β-mercaptoethanol, 1 µg/ml heparin, 10 µM SB431542, 200 ng/ml LDN, 0.7 µM CHIR99021 and 50 µM Rock inhibitor. On day 4, the medium was changed to DMEMF12/neurobasal (1:1) supplemented with 0,5% N2 supplement, 1% B27 supplement, 1% NEAA, 0.1% β-mercaptoethanol, 1 µg/ml heparin, 10 µM SB431542, 200 ng/ml LDN, 100 ng/ml SHH-C25II, 1 µM purmorphamine, 7.5 µM CHIR99021 and 100 ng/ml FGF8. On day 7, embryoid bodies (EBs) were embedded in Matrigel using a protocol adapted from Qian et al. ⁵⁰. Matrigel was diluted in a 3:2 ratio with medium and used as an embedding mixture. EBs were washed in fresh medium and embedded in Matrigel in a six-well ultra-low-attachment plate using the Matrigel-medium mixture. The Matrigel-EB mixture was incubated for 30 min at 37°C, and on day 7, medium containing the following was added: neurobasal medium supplemented with 0,5% N2 supplement, 1% diluted B27 supplement, 1% GlutaMax, 1% NEAA, 0.1% β-mercaptoethanol, 2.5 µg/ml insulin, 200 ng/ml laminin, 7.5 µM CHIR99021, 100 ng/ml SHH-C25II, 1 µM purmorphamine and 100 ng/ml FGF8. On day 8, the medium was changed to neurobasal medium supplemented with 0,5% N2, 1% B27, 1% GlutaMax, 1% NEAA, 7.5 µM CHIR99021, 0.1% β-mercaptoethanol, 10 ng/ml BDNF, 10 ng/ml GDNF, 100 µM ascorbic acid and 125 µM dbcAMP. On day 10, the medium was refreshed, and the concentration of CHIR99021 was reduced to 3 µM. On day 13, the concentration of CHIR99021 was further reduced to 0.7 µM until day 15. On day 20, the Matrigel was dissociated from the organoids.

10X genomics single-cell RNA sequencing

Midbrain organoids were dissociated into single-cell suspensions using a Worthington Papain Dissociation System kit (Worthington Biochemical), as previously described ⁵¹. Organoids were minced and incubated with 2,5 mL papain/DNAase solution was incubated on an orbital shaker for 30 min at 37°C. Organoid suspensions were triturated using 1 ml low-attachment pipette tips and then incubated on an orbital shaker for 20 min at 37°C. After enzyme inactivation, the cell suspensions were filtered with a 30-μm cell strainer and washed three times with HBSS. The final cell suspension was centrifuged at 300 × g for 7 min, and the cell pellet was resuspended in PBS-0.5% BSA. The cell concentration and viability were determined by microscopy using trypan blue staining (0.2%). scRNA-seq libraries were immediately generated using the 10X Chromium Next GEM Single Cell 3’ Reagent Kit v3.1 (Cat. 1000128) according to the manufacturer’s instructions and paired-end sequenced on an Illumina NovaSeq 6000 (SP Flow Cell) with a sequencing depth of 300 million reads per library. GEO accession number: GSE198033.

Single-cell sequencing data analysis

The 10X Genomics pipeline cellranger count was run to generate filtered gene-barcode matrices that were used as input for downstream analysis using the R package Seurat version 4.1.0 (RRID:SCR_007322) (Butler et al. 2018, Stuart et al. 2019). Low-quality cells were filtered out (detected genes > 200, counts > 500, and mitochondrial ratio < 0.2), and genes with zero counts and genes that were expressed in less than 10 cells were also removed. For clustering analysis, the dataset was SCtransformed and normalized, and variation due to the cell cycle and mitochondrial expression was regressed out. Data from different 10x runs were integrated in Seurat using individual 10X runs as grouping variables. Principle components (PCs) were determined, and using the first 40 PCs, cells were clustered using a K-nearest neighbor (KNN) graph with a clustering resolution of 0.6, resulting in 21 clusters. Cell clusters were visualized using UMAP. Conserved markers were determined across all samples for each cluster (FindConservedMarkersfunction) using default settings (min.pct=0.25, logfc.threshold=0.25) and used for assigning cell type to cell clusters. The corresponding annotation file was downloaded from https://raw.githubusercontent.com/hbctraining/scRNA-seq/master/data/annotation.csv. Pseudo-time analysis was performed using Monocle 3 (version 1.0.0). After data normalization and integration, Monocle3 object was generated using as.cell_data_set function, and cells were clustered using cluster_cells function with a resolution of 1e-3. Clusters were manually assigned using La Manno et al as a reference ²⁵. When ordering the cells along the trajectory using “order_cells” function, SOX1 expressing cells were specified as the starting state.

Immunofluorescence

Cells were washed once with PBS and fixed for 10 min with 4% paraformaldehyde (PFA, Sigma–Aldrich) at RT. After fixation, the cells were washed twice with PBS. The cells were blocked and permeabilized in PBS containing 0.1% Triton-X (PBS-T) and 10% normal goat serum (NGS) for 1 h at RT. The primary antibody was diluted in PBS-T containing 5-10% NGS and incubated overnight at 4°C. Secondary antibody conjugated with Alexa 488 or 568 (Invitrogen) was diluted in 5% NGS in PBS-T and incubated for 1 h at RT. The nuclei of the cells were stained by incubation with 1 µM 4',6-diamidino-2-phenylindole dihydrochloride (DAPI, Invitrogen), and coverslips were mounted on glass slides with mounting medium (DAKO). Images were acquired with a 63x 1.4NA plan-apochromat oil objective of a TCS SP8 confocal microscope (Leica Biosystems). Midbrain organoids were fixed in 4% (w/v) PFA for 1 h, and individual organoids were equilibrated in 30% sucrose in PBS overnight at 4°C. The next day, the organoids were embedded in blocks in optimal cutting temperature compound (OCT, Tissue-Tek). Slices with a thickness of 20 µm were cryosectioned and mounted on noncharged slides. Tissues were blocked in 10% (v/v) NGS in 0.5% Triton X-100 in PBS and incubated with primary antibodies overnight and secondary antibodies for 3 h. DAPI staining was performed with a concentration of 1 µM for 5 min, and after three PBS washes, tissues were mounted with DAKO mounting medium (DAKO) for image acquisition. The antibodies are listed in Supplementary Table 3.

Expansion microscopy

Expansion microscopy (ExM) was performed as described ⁵² with some modifications. Cells were blocked with 10% (v/v) normal goat serum (NGS) in 0.1% (v/v) Triton X-100 in PBS and incubated with primary antibodies in blocking solution overnight. After a 3-h incubation with the corresponding secondary antibody (Alexa Fluor, Invitrogen), the samples were washed and treated with 0.1 mg/ml Acryloyl-X SE solution (Thermo Scientific) in PBS for 3 h at room temperature. The freshly prepared gelling solution consisted of Stock X solution (8.6% (w/v) sodium acrylate 33% (w/v), 2.5% (w/v) acrylamide 50% (w/v), 0.15% (w/v) N,N´-methylenebisacrylamide 2% (w/v), 11.7% (w/v) NaCl 5 M, and PBS 1X), water, 10% (w/v) TEMED and 10% (w/v) APS stock solution in a 47:1:1:1 ratio. Gel digestion was performed overnight in digestion buffer (0.5% (w/v) Triton X-100, 0.2% (v/v) EDTA 0.5 M, pH 8, 5% (v/v) Tris-Cl 1 M, pH 8, 4.67% (w/v) NaCl and 8 U/ml proteinase K). The gelling solution was added to each well and covered by a 15-mm coverslip to ensure the formation of a smooth, flat and thin gel. Coverslips were then incubated for 1 h at 37°C for complete polymerization. The gel was expanded in water for 1 h and mounted in 10 µg/mL poly-L-ornithine-coated coverslips to immobilize the gel for picture acquisition. Midbrain organoids were fixed, and immunofluorescent staining was performed as described above. Sections were treated with 0.1 mg/ml acryloyl-X SE solution in PBS at room temperature overnight. Gelation was performed in a 47:1:1:1 ratio of Stock X, 10% (w/v) TEMED, 10% (w/v) APS, and 0.5% (w/v) 4-hydroxy-TEMPO stock solutions. Gel digestion and expansion were performed as described above. Images were acquired using a Leica TCS SP8 confocal microscope (Leica, Germany) equipped with a 100× /1.4 numerical aperture oil-immersion objective. For each condition, 5 images were acquired from at least three independent experiments. Images were analyzed using Diffraction PSF 3D, DeconvolutionLab2, and EzColocalization plugins in Fiji-ImageJ. GraphPad Prism version 9.0.0 (RRID:SCR_002798) was used for calculating Spearman’s rank correlation value (ρ) to identify colocalization of fluorescence signals. The antibodies are listed in Supplementary Table 3.

Mitochondrial isolation

Mitochondria were isolated from 10x10^6 cells or individual organoids using the Qproteome Mitochondrial isolation kit (Qiagen) with slight alterations to the manufacturer’s protocol. Cells or individual organoids were harvested, washed in 0.9% NaCl and incubated for 10 min at 4°C in lysis buffer. The homogenate was centrifuged at 1000 × g for 10 min at 4°C, and the supernatant was designated the cytosolic fraction. The pellet was resuspended in disruption buffer and mechanically passed through a 26-gauge needle 10 times. The enriched nuclear fraction was pelleted by centrifugation at 1000 × g for 10 min and homogenized in disruption buffer. Next, the supernatant was centrifuged at 6000 × g for 10 min at 4°C and resuspended in mitochondria storage buffer (enriched mitochondrial fraction). All buffers (except mitochondria storage buffer) were supplemented with 1:100 protease inhibitors (provided with the kit). Where indicated, cells were treated with 20 mg/ml digitonin (Sigma-Aldrich) or 0.01% Triton X-100 in PBS. For the proteinase K protection assay, freshly isolated mitochondria were resuspended in mitochondria storage buffer and treated with 2-20 mg/ml proteinase K (New England Biolabs GmbH) for 30 min on ice to digest surface-exposed proteins. The reaction was stopped by adding 2 mM PMSF Protease Inhibitor (Thermo Fisher).

Complex I activity assay

Mitochondria were isolated from HEK cells, iPSC-derived neurons, or midbrain organoids using the Qproteome Mitochondrial isolation kit as described above. Complex I (NADH oxidase/coenzyme Q reductase) was measured using the MitoCheck Complex I Activity Assay kit (Cayman Chemical, cat# 700930). The rate of NADH oxidation, which is proportional to CI activity, was determined by a decrease in absorbance at 340 nm over 15 min in the presence of ubiquinone and potassium cyanide to inhibit complex IV and prevent oxidation of ubiquinone.

A-SYN PFFs

The production and purification of recombinant human A-SYN was conducted according to Martinez et al. ⁵³. A-SYN cDNAs were cloned into pET 21D (Novagen, Merck Millipore, #69743), and the plasmids were expressed in BL21DE3 E. coli (Novagen, Merck Millipore, #69450). Cultures of 750 ml were grown to midlog phase, and isopropyl-1-thio-3-d-galactopyranoside was added to 0.4 mM. After 2 h, the cells were pelleted, washed in PBS, and resuspended in 50 ml of 20 mM HEPES, 100 mM KCl, pH 7.2. The resuspended bacteria were heated to 90°C for 5 min. Aggregated protein was removed by centrifugation (40 min at 40,000 × g at 4°C). Contaminating nucleic acids and proteins were removed by ion exchange chromatography on Q-Sepharose Hi-Trap columns equilibrated with Solution A (50 mM Tris pH 7.4; Amersham Biosciences). Purified A-SYN was eluted by loading the soluble fraction and applying an increasing gradient of Solution B (50 mM Tris, pH 7.4, 1 M KCl). SYN-containing fractions were pooled and chromatographed on a Superose 12 column (Amersham Biosciences) in 20 mM HEPES, pH 7.4, 100 mm KCl. The monomer was aliquoted and frozen at −80°C. For preparation of PFFs, A-SYN monomer was shaken at 1,000 rpm for 7 days. PFFs were validated by electron microscopy, a sedimentation assay at 100,000 × g for 60 min, the thioflavin T assay, and on primary cortical neurons by pS129-syn immunostaining. Preformed fibrils were labeled using the Alexa Fluor 594 kit (Thermo Fisher, #A10239) according to the manufacturer’s instructions. For the treatment of neurons and organoids, A-SYN PFFs, which were generated at a concentration of 5 mg/mL, were vortexed and diluted with Dulbecco’s phosphate-buffered saline to 100 μg/mL and then sonicated (10-s pulses with 30% amplitude six times every 2 min) using a HTU SONI-130 sonicator (G. Heinemann, Germany). A-SYN PFFs were then diluted in neuronal or organoid media and added to cultures.

Virus production

The verified Mission lentiviral plasmids encoding nontargeting shRNA (#SHC016) and HSPA8/HSC70 shRNA (#TRCN0000017279, #TRCN0000017281) in the pLKO.1-puro vector backbone were purchased from Sigma–Aldrich. The sequences of shRNA-TIM23 used by Goemans et al. ⁵⁴ were cloned into pLV(shRNA)-Puro-U6 (purchased from VectorBuilder); pLV[shRNA]-Puro-U6>Scramble-shRNA (purchased from VectorBuilder) was used as a nontargeting shRNA. For lentiviral production, HEK cells were transfected with the plasmids expressing the shRNAs together with the lentiviral packaging plasmid psPAX2 (RRID:Addgene_12260) and the envelope plasmid pMD2.G (RRID:Addgene_12259) using TransIT-X2 (Mirus). In short, HEK 293T cells were seeded in a 10-cm dish to reach 80% confluency the day after transfection. The next day, the medium was changed. On days 4 and 5 after transfection, the medium was collected and filtered through a 0.45-µm PVDF membrane. To concentrate the virus, the filtered supernatant was centrifuged in a Vivaspin column (Sartorius Stedim Biotech) at 3000 RCF at 4 °C until the volume reached 500-1000 µl. The virus-containing supernatant was collected, and the concentration of p24 particles was determined with Lenti-X GoStix Plus (Takara). Cells were infected with equal concentrations of p24 particles for scramble and respective shRNA.

GCase activity assay

Cells or midbrain organoids were lysed by sonication in H₂O containing 0.01% Triton, and the protein concentration was determined by BCA. For enzymatic assay measurement, 10-20 µg of protein was incubated for 30 min at room temperature with 25 µl of McIlvaine buffer 4X (0.4 M citric acid, 0.8 M Na2HPO4), pH 5.2, AMP-DNM (N-(5-adamantane-1-yl-methoxy-pentyl)-deoxynojirimycin) at a final concentration of 5 nM, and H2O to a final volume of 100 µl. 4-Methylumbelliferyl-β-D-glucopyranoside (MUB-Glc; Glycosynth, Warrington, UK) was dissolved in 200 mM sodium citrate phosphate buffer by heating to 60°C. At the end of incubation, 25 µl of 4-MU was added at a final concentration of 6 mM and incubated for 2 h at 37°C. As a control, 1 mM condurbitol B epoxide (CBE, Calbiochem) was used. The fluorescence was recorded after transferring 20 μl of the reaction mixture to a microplate and adding 180 μl of 0.25 M glycine, pH 10.2. Data were calculated as picomoles of converted substrate per milligram of cell protein per hour.

A-SYN ELISA

The levels of total A-SYN were measured in midbrain organoids homogenates using the hSYN total ELISA kit (847-0108000103, Roboscreen Diagnostic, Liepzig, Germany), according to the manufacturer’s instructions. The optical density was read at 450 nm on a microplate reader (Bio-Rad). Data were normalized on protein content.

Mitochondrial respiration

Mitochondrial function was assessed in live cells using an XFp Extracellular Flux Analyzer (Agilent Technologies). A total of 1x10^4 T-Rex HEK or 1.5x10^5 iPSC-derived neurons were seeded in XFpSeahorse microplates and allowed to adhere overnight (for HEK cells) or for seven days (iPSC-derived neurons). Measurements of the oxygen consumption rate (OCR) were performed in freshly prepared assay medium, pH 7.4 (Seahorse XF DMEM Medium), according to the manufacturer’s protocol. The OCR was measured before and after the serial addition of 20 µM or 1 μΜ oligomycin, 1 µM or 5 μΜ carbonyl cyanide p-trifluoromethoxyphenylhydrazone (CCCP), and 2 µM or 1 μΜ antimycin A and 2 µM or 1 µΜ rotenone to T-Rex HEK cells or iPSC-derived neurons, respectively (all from Sigma–Aldrich). Following each injection, three measurements for a total period of 15 min were recorded. The data were analyzed using Wave 2.6 software (RRID:SCR_014526), and OCR parameters (basal respiration, maximal respiratory capacity, respiratory reserve, and ATP-linked respiration) were calculated. At least three technical replicates per condition were used, and the experimental values were normalized to the protein content per well via a BCA assay.

Acknowledgments

The authors thank Johannes M Herrmann (University of Kaiserslautern), Peng Qui (Georgia Institute of Technology), and Frank Stein (Proteomics Core Facility, EMBL) for helpful discussions. We acknowledge the funding support of the Helmholtz Association Young Investigator Award (VH-NG-1123, to MD), Juniorprofessuren-Programm BW (Az:7365.521/(16),to MD), EU Joint Programme – Neurodegenerative Disease Research (JPND) project (GBA-PaCTS, 01ED2005B, to MD and AHVS), Instituto de Salud Carlos III-MINECO/European FEDER Funds Grant PI20-00057 (to CU), Comunidad Autónoma de Madrid/ERDF-ESF P2018/BAA-4403 (to CU), DZNE (PD research; MIGAP study, to KB and TG), and EMBO ALTF-1013-2019 (to MJP) for this project. This research was funded in part by Aligning Science Across Parkinson’s [Grant number: ASAP-000420] through the Michael J. Fox Foundation for Parkinson’s Research (MJFF) (to MD and AHVS). Images were created with BioRender.com.

Author contributions

Conceptualization, P.B., M.J.P., M.D.; formal analysis, P.B., M.J.P., H.R., F.B., G.C., C.U., M.D.; investigation,; P.B., M.J.P., S.K., H.R., M.I., C.G., F.B., M.O., A.C., C.U., M.D.; resources, G.C., KB, TG, A.H.V.S.; writing – original draft, P.B., M.D.; writing – review & editing, P.B., M.J.P., S.K., H.R., M.I., C.G., F.B., M.O., G.C., K.B., T.G., A.H.V.S., C.U., M.D; funding acquisition, C.U., A.H.V.S., M.D.; supervision, M.D.

Neudorfer O, et al. Occurrence of Parkinson's syndrome in type I Gaucher disease. QJM: monthly journal of the Association of Physicians 89, 691–694 (1996).
Sidransky E, et al. Multicenter analysis of glucocerebrosidase mutations in Parkinson's disease. N Engl J Med 361, 1651–1661 (2009).
Gan-Or Z, et al. Differential effects of severe vs mild GBA mutations on Parkinson disease. Neurology 84, 880–887 (2015).
Malek N, et al. Features of GBA-associated Parkinson's disease at presentation in the UK Tracking Parkinson's study. J Neurol Neurosurg Psychiatry 89, 702–709 (2018).
Duran R, et al. The glucocerobrosidase E326K variant predisposes to Parkinson's disease, but does not cause Gaucher's disease. Movement disorders: official journal of the Movement Disorder Society 28, 232–236 (2013).
Zhang Y, et al. Integrated Genetic Analysis of Racial Differences of Common GBA Variants in Parkinson's Disease: A Meta-Analysis. Front Mol Neurosci 11, 43 (2018).
Straniero L, et al. The SPID-GBA study: Sex distribution, Penetrance, Incidence, and Dementia in GBA-PD. Neurol Genet 6, e523 (2020).
Davis MY, et al. Association of GBA Mutations and the E326K Polymorphism With Motor and Cognitive Progression in Parkinson Disease. JAMA Neurol 73, 1217–1224 (2016).
Vieira SRL, Schapira AHV. Glucocerebrosidase mutations: A paradigm for neurodegeneration pathways. Free Radic Biol Med 175, 42–55 (2021).
Cleeter MW, et al. Glucocerebrosidase inhibition causes mitochondrial dysfunction and free radical damage. Neurochem Int 62, 1–7 (2013).
Osellame LD, et al. Mitochondria and quality control defects in a mouse model of Gaucher disease–links to Parkinson's disease. Cell metabolism 17, 941–953 (2013).
de la Mata M, et al. Pharmacological Chaperones and Coenzyme Q10 Treatment Improves Mutant beta-Glucocerebrosidase Activity and Mitochondrial Function in Neuronopathic Forms of Gaucher Disease. Sci Rep 5, 10903 (2015).
Schondorf DC, et al. The NAD + Precursor Nicotinamide Riboside Rescues Mitochondrial Defects and Neuronal Loss in iPSC and Fly Models of Parkinson's Disease. Cell Rep 23, 2976–2988 (2018).
Ron I, Horowitz M. ER retention and degradation as the molecular basis underlying Gaucher disease heterogeneity. Hum Mol Genet 14, 2387–2398 (2005).
Sawkar AR, et al. Gaucher disease-associated glucocerebrosidases show mutation-dependent chemical chaperoning profiles. Chemistry & biology 12, 1235–1244 (2005).
Tan YL, Genereux JC, Pankow S, Aerts JM, Yates JR, 3rd, Kelly JW. ERdj3 is an endoplasmic reticulum degradation factor for mutant glucocerebrosidase variants linked to Gaucher's disease. Chem Biol 21, 967–976 (2014).
Chen F, Tillberg PW, Boyden ES. Optical imaging. Expansion microscopy. Science 347, 543–548 (2015).
Boos F, Mühlhaus T, Herrmann J. Detection of Internal Matrix Targeting Signal-like Sequences (iMTS-Ls) in Mitochondrial Precursor Proteins Using the TargetP Prediction Tool.) (2018).
Cabantous S, Terwilliger TC, Waldo GS. Protein tagging and detection with engineered self-assembling fragments of green fluorescent protein. Nat Biotechnol 23, 102–107 (2005).
Harbauer AB, Zahedi RP, Sickmann A, Pfanner N, Meisinger C. The protein import machinery of mitochondria-a regulatory hub in metabolism, stress, and disease. Cell Metab 19, 357–372 (2014).
Young JC, Hoogenraad NJ, Hartl FU. Molecular Chaperones Hsp90 and Hsp70 Deliver Preproteins to the Mitochondrial Import Receptor Tom70. Cell 112, 41–50 (2003).
Guarani V, Paulo J, Zhai B, Huttlin EL, Gygi SP, Harper JW. TIMMDC1/C3orf1 functions as a membrane-embedded mitochondrial complex I assembly factor through association with the MCIA complex. Mol Cell Biol 34, 847–861 (2014).
Jo J, et al. Midbrain-like Organoids from Human Pluripotent Stem Cells Contain Functional Dopaminergic and Neuromelanin-Producing Neurons. Cell Stem Cell 19, 248–257 (2016).
Kim TW, et al. Biphasic Activation of WNT Signaling Facilitates the Derivation of Midbrain Dopamine Neurons from hESCs for Translational Use. Cell Stem Cell 28, 343–355 e345 (2021).
La Manno G, et al. Molecular Diversity of Midbrain Development in Mouse, Human, and Stem Cells. Cell 167, 566–580 e519 (2016).
Gibellini L, et al. Inhibition of Lon protease by triterpenoids alters mitochondria and is associated to cell death in human cancer cells. Oncotarget 6, 25466–25483 (2015).
Lee J, et al. Inhibition of mitochondrial LonP1 protease by allosteric blockade of ATP -binding and -hydrolysis via CDDO and its derivatives. J Biol Chem, 101719 (2022).
Mazzulli JR, et al. Gaucher disease glucocerebrosidase and alpha-synuclein form a bidirectional pathogenic loop in synucleinopathies. Cell 146, 37–52 (2011).
Lautenschlager J, et al. Intramitochondrial proteostasis is directly coupled to alpha-synuclein and amyloid beta1-42 pathologies. J Biol Chem 295, 10138–10152 (2020).
Backes S, et al. Tom70 enhances mitochondrial preprotein import efficiency by binding to internal targeting sequences. Journal of Cell Biology 217, 1369–1382 (2018).
Ruan L, et al. Cytosolic proteostasis through importing of misfolded proteins into mitochondria. Nature 543, 443–446 (2017).
Li Y, et al. A mitochondrial FUNDC1/HSC70 interaction organizes the proteostatic stress response at the risk of cell morbidity. EMBO J 38, (2019).
Cortes Sanchon A, et al. ER-misfolded proteins become sequestered with mitochondria and impair mitochondrial function. Commun Biol 4, 1350 (2021).
Shin CS, et al. LONP1 and mtHSP70 cooperate to promote mitochondrial protein folding. Nat Commun 12, 265 (2021).
Pryde KR, Taanman JW, Schapira AH. A LON-ClpP Proteolytic Axis Degrades Complex I to Extinguish ROS Production in Depolarized Mitochondria. Cell Rep 17, 2522–2531 (2016).
Wang T, et al. C9orf72 regulates energy homeostasis by stabilizing mitochondrial complex I assembly. Cell Metab 33, 531–546.e539 (2021).
Gonzalez-Rodriguez P, et al. Disruption of mitochondrial complex I induces progressive parkinsonism. Nature 599, 650–656 (2021).
Schapira AH, Cooper JM, Dexter D, Jenner P, Clark JB, Marsden CD. Mitochondrial complex I deficiency in Parkinson's disease. Lancet 1, 1269 (1989).
Schapira AH, Cooper JM, Dexter D, Clark JB, Jenner P, Marsden CD. Mitochondrial complex I deficiency in Parkinson's disease. J Neurochem 54, 823–827 (1990).
Gegg ME, et al. Glucocerebrosidase deficiency in substantia nigra of parkinson disease brains. Ann Neurol 72, 455–463 (2012).
Hallett PJ, et al. Glycosphingolipid levels and glucocerebrosidase activity are altered in normal aging of the mouse brain. Neurobiol Aging 67, 189–200 (2018).
Rocha EM, et al. Progressive decline of glucocerebrosidase in aging and Parkinson's disease. Ann Clin Transl Neurol 2, 433–438 (2015).
Hughes CS, Foehr S, Garfield DA, Furlong EE, Steinmetz LM, Krijgsveld J. Ultrasensitive proteome analysis using paramagnetic bead technology. Mol Syst Biol 10, 757 (2014).
Calì T, Ottolini D, Soriano ME, Brini M. A new split-GFP-based probe reveals DJ-1 translocation into the mitochondrial matrix to sustain ATP synthesis upon nutrient deprivation. Human Molecular Genetics 24, 1045–1060 (2015).
Okita K, et al. A more efficient method to generate integration-free human iPS cells. Nat Methods 8, 409–412 (2011).
Schöndorf DC, et al. iPSC-derived neurons from GBA1-associated Parkinson’s disease patients show autophagic defects and impaired calcium homeostasis. Nature Communications 5, (2014).
Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F. Genome engineering using the CRISPR-Cas9 system. Nat Protoc 8, 2281–2308 (2013).
Reinhardt P, et al. Derivation and Expansion Using Only Small Molecules of Human Neural Progenitors for Neurodegenerative Disease Modeling. PLoS ONE 8, e59252 (2013).
Lobo-Jarne T, et al. Multiple pathways coordinate assembly of human mitochondrial complex IV and stabilization of respiratory supercomplexes. EMBO J 39, e103912 (2020).
Qian X, Jacob F, Song MM, Nguyen HN, Song H, Ming GL. Generation of human brain region-specific organoids using a miniaturized spinning bioreactor. Nat Protoc 13, 565–580 (2018).
Perez MJ, et al. Loss of function of the mitochondrial peptidase PITRM1 induces proteotoxic stress and Alzheimer's disease-like pathology in human cerebral organoids. Mol Psychiatry, (2020).
Asano SM, Gao R, Wassie AT, Tillberg PW, Chen F, Boyden ES. Expansion Microscopy: Protocols for Imaging Proteins and RNA in Cells and Tissues. Curr Protoc Cell Biol 80, e56 (2018).
Martinez J, Moeller I, Erdjument-Bromage H, Tempst P, Lauring B. Parkinson's disease-associated alpha-synuclein is a calmodulin substrate. J Biol Chem 278, 17379–17387 (2003).
Goemans CG, Boya P, Skirrow CJ, Tolkovsky AM. Intra-mitochondrial degradation of Tim23 curtails the survival of cells rescued from apoptosis by caspase inhibitors. Cell Death & Differentiation 15, 545–554 (2008).

There is NO Competing Interest.

SupplementaryDataset1.xlsx
Supplementary Dataset 1
SupplementaryDataset2.xlsx
Supplementary Dataset 2
SupplementaryDataset3.xlsx
Supplementary Dataset 3
SupplementaryDataset4.xlsx
Supplementary Dataset 4
SupplementaryMaterial042422.pdf
SupplementaryTable3.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Glucocerebrosidase, a Parkinson´s disease-associated protein, is imported into mitochondria and regulates complex I assembly and function

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Methods

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1