Targeting lysosomal pH restores mitochondrial quality control in GBA1-mutant Parkinson’s disease Authors: Preethi Sheshadri, Maria Alicia Costa-Besada, Alessia Fisher, Szilvia Kiraly, Kritarth Singh, Ioanna Kourouzidou, Thomas S. Blacker, Jialiu Zeng, Orian S. Shirihai, Mark W. Grinstaff, Michael R. Duchen Published: 2026 ABSTRACT BackgroundHeterozygous mutations in the glucocerebrosidase gene (GBA1), which encodes the lysosomal enzyme β-glucocerebrosidase (GCase), are a genetic risk factor for Parkinson’s disease (PD). The pathophysiological consequences of GBA1 mutations on dopaminergic neuronal function, especially their impact on lysosomal function, mitophagy, and mitochondrial bioenergetics, remain unclear.MethodsFibroblasts and dopaminergic neurons generated from induced pluripotent stem cells (iPSCs) derived from patients with GBA1-PD were used in the study. Live-cell imaging was performed to measure lysosomal acidification, protease activity, mitochondrial membrane potential, and mitophagy. Mitochondrial morphology and autophagic vesicles were examined using transmission electron microscopy. Oxygen consumption rate was measured by Seahorse assay. V-ATPase assembly was quantified using fluorescence lifetime imaging with Förster resonance energy transfer (FLIM-FRET), and pharmacological interventions included rapamycin and acidic nanoparticles.ResultsGCase activity, lysosomal acidification, protease activity, mitophagy and mitochondrial bioenergetic function were all impaired in GBA1 mutant dopaminergic neurons. Mitochondria were fragmented, with reduced membrane potential and oxygen consumption. Mechanistic target of rapamycin complex 1 (MTORC1) was constitutively phosphorylated and FLIM-FRET measurements confirmed impairment of lysosomal V-ATPase assembly, which was reversed by rapamycin treatment. Rapamycin and lysosome-targeting acidic nanoparticles rescued lysosomal pH and restored mitophagy, mitochondrial membrane potential and mitochondrial oxidative phosphorylation complex level in the GBA1 mutant dopaminergic neurons.ConclusionsWe revealed a novel mechanistic link between GBA1 mutations and mitochondrial dysfunction, as the disruption of V-ATPase assembly driven by MTORC1 activation impairs lysosomal acidification. This causes impairment of mitophagy, leading to mitochondrial dysfunction, undermining dopaminergic cell function and fate. Pharmacological intervention with rapamycin or acidic nanoparticles restores lysosomal pH and rescue mitochondrial function, representing a novel therapeutic approach for GBA1-PD .Supplementary InformationThe online version contains supplementary material available at 10.1186/s40035-026-00559-z. Introduction Genome-wide association studies have revealed that mutations of the glucocerebrosidase (GBA1) gene constitute a significant risk factor in the development of Parkinson’s disease (PD) [1]. The GBA1 gene encodes the enzyme β-glucocerebrosidase (GCase), which generates glucose and ceramide from glucosylceramide within lysosomes. While homozygous GBA1 mutations cause Gaucher’s disease, which may include a significant neurodegenerative component [2], heterozygous mutations are associated with an increased risk of developing PD. Two most prevalent GBA1 mutations associated with PD are N370S and L444P mutations [3]. Intriguingly, the less common E326K mutation in GBA1 exhibits a relatively mild effect on GCase activity, does not cause Gaucher’s disease, but is correlated with PD risk [4]. Mitochondrial dysfunction appears to constitute a defining characteristic of PD [5]. The adverse impact of dysfunctional mitochondrial oxidative phosphorylation (OXPHOS) on the viability of dopaminergic (DA) neurons has been substantiated by studies using experimental models of PD that result from toxins targeting complex I [5]. Gba1 knockout mice, a model of severe neurodegeneration, exhibit severe mitochondrial dysfunction in primary neurons and astrocytes in culture, alongside neurological pathologies associated with PD, including disruption of autophagy-lysosomal pathways and accumulation of ubiquitinated proteins and α-synuclein [6]. The autophagosome-lysosome axis and the ubiquitin proteasome system are responsible for degradation of dysfunctional organelles and protein degradation; they together play a critical role in cellular quality control [7]. Lysosomal acidification generating pH values between 4.5 to 4.7 [8] is essential for normal lysosomal function and is required for the activity of lysosomal hydrolytic enzymes and effective protein degradation [9, 10]. The acidification is generated by the vacuolar-type H+ ATPase (V-ATPase), a proton pump composed of a peripheral V1 domain that hydrolyses ATP, and a membrane-integrated Vo domain responsible for the translocation of protons into the lysosomal lumen [11]. Disrupted assembly and dysfunction of V-ATPase have been reported to be associated with multiple disorders [12], including juvenile-onset PD [9], Alzheimer’s disease [13, 14], epilepsy [15, 16] and Down syndrome [17]. GBA1-linked pathologies such as Gaucher’s disease and PD are associated with dysfunctions of the autophagy-lysosome pathway, including impaired lysosomal regeneration from autolysosomes during macroautophagy [18, 19]. These findings suggest that targeting the dysfunctional lysosomal system and the autophagy-lysosome pathway may serve as a potential therapeutic strategy for PD linked to GBA1 mutations. Although numerous studies indicate that impaired lysosomal acidification plays a role in neuronal pathologies associated with various neurodegenerative disorders [20] and that GBA1-PD is characterised by lysosomal dysfunction [10], the effects of compromised lysosomal acidification in GBA1-PD have not been extensively examined. This study hypothesizes that lysosomal dysfunction caused by GBA1 mutations in PD may impair mitochondrial function, leading to cellular malfunction. The experimental design involved characterizing mitochondrial and lysosomal function in patient-derived fibroblasts and DA neurons generated from induced pluripotent stem cells (iPSCs). Our rationale was to explore the potential link between lysosomal impairment and mitochondrial dysfunction. To achieve this, fluorescent imaging and western blotting were used to assess cellular functions, alongside fluorescent lifetime imaging. The effects of rapamycin and acidic nanoparticles (NPs) on GBA1-PD patient-derived cells, were investigated to understand the mechanisms underlying cellular dysfunction. Materials and methods Cell lines and ethics committee approval The human induced pluripotent stem cells (hiPSC) and fibroblast lines used in the study, along with their sources, are detailed in Table S1. Fibroblasts were cultured in DMEM (1 ×) + GlutaMAX™ basal media (Gibco, 61965–059, Grand Island, NE), 10% (v/v) fetal bovine serum (Gibco, 10270098, London, UK) and 1% antibiotic–antimycotic (Gibco, 15240096) at 37 ºC with 5% CO2. The media were changed every 72 h, and the cells were passaged upon reaching 80% confluency using 0.25% Trypsin–EDTA (Gibco, 25200056). Passages between 9 and 15 were used for experiments. GBA1 patient-derived iPSC lines were cultured under feeder-free conditions on Geltrex-coated 6-well plates, using mTeSR Plus medium (STEMCELL Technologies, 100–0276, 100–1130, Cambridge, UK) and were passaged with 0.5 mM EDTA (Gibco, 15575020) every 4–5 days or upon reaching 80% confluency. Passage numbers between 20 and 30 were employed for differentiation purposes. Human pluripotent stem cell-derived DA neurons DA neurons were generated from iPSCs using a step-by-step differentiation protocol, as previously described [21]. Briefly, 90% − 95% confluent hiPSCs were cultured on Geltrex-coated plates and maintained in N2B27 medium, which is a 1:1 mix of Neurobasal (Gibco, 21103049) supplemented with 1 × B27 (Gibco, 17504044), Glutamax (Gibco, 35050–038), and 0.5 × Anti-Anti (Gibco, 15240096), and DMEM-F12 (Gibco, 10565018) supplemented with 1 × N2 supplement (Gibco, 17502048), 1 × MEM Non-Essential Amino Acids (Gibco, 11140–050), 0.5 × Antibiotic-Antimycotic (Gibco, 15240096), 50 μM β-mercaptoethanol (Gibco, 21985–023), and human insulin solution (Sigma, I9278, St. Louis, MO). Medium was changed daily for the first 14 days. On day 0, the medium was supplemented with 5 μM SB431542 (Tocris Bioscience, 1614/10, Bristol, UK), 2 μM dorsomorphin (Tocris Bioscience, 3093/10), and 1 μM CHIR99021 (Tocris Bioscience, 4423) until day 2. From day 2 to day 7, 1 μM purmorphamine (Merck Millipore, SML0868, Burlington, VT) was added to the mix. From day 8 to day 14, CHIR99021 and SB431542 were removed from the medium, and the cells were maintained with dorsomorphin and purmorphamine in N2B27. Cells were dissociated with 1 mg/mL Dispase (Gibco, 17105041) on days 4 and 14. After patterning for 14 days, the midbrain DA neuronal precursor cells were maintained in the N2B27 medium until day 18. On day 19, the cells were dissociated with Accutase (Gibco, A1110501) and plated onto Geltrex-coated plates at a density of 2 × 105 cells/cm2 and terminally differentiated with N2B27 medium supplemented with Compound E (Enzo Lifesciences, ALX-270–415-C250, Farmingdale, NY) and 10 µM Rho-kinase inhibitor, Y-27632 dihydrochloride (Tocris Bioscience, 1254), from day 20 until day 70–75, with medium change twice a week. DA neurons between days 70 and 75 were used for downstream experiments. NP, rapamycin and diethyl succinate treatment Control and acidic NPs, composed of poly(ethylene-succinate) and poly(ethylene tetrafluorosuccinate-co-succinate), respectively, were obtained from Prof. Mark Grinstaff (Departments of Biomedical Engineering and Chemistry, Boston University) and Prof. Orian Shirihai (Department of Medicine, David Geffen School of Medicine, University of California). The NPs were synthesized as described previously [22]. The NPs were stored at − 80 °C in aliquots and thawed fresh prior to treatment. Aliquots of 75 mg/mL NPs were thawed to room temperature, vortexed thoroughly, added to the cell cultures at a concentration of 180 µg/mL in fresh culture medium and incubated for 12–16 h before the experiment. Rapamycin (Sigma, R0395) was prepared in DMSO at 200 µM and sterile filtered before being stored as aliquots at − 20 ºC. Before use, the stock solution was diluted to 200 nM in medium, and added to cell cultures for incubation for 12–16 h. Diethyl succinate (Merck Millipore, 112402) was diluted to 3 mM in culture medium, sterile-filtered and added to the cell culture 30 min prior to mitochondrial membrane potential (ΔΨm) measurement with tetramethyl-rhodamine methyl ester (TMRM). Generation of lentivirus and transduction pLV-Ef1a-3xHA-ATP6V1B2-mNeonGreen and pLV-Ef1a-3xFlag-ATP6Voa3-mScarlet plasmids were kindly provided by Dr. Wilhelm Palm (EPR 462 and EPR 475, European Plasmid Repository). pHAGE-mt-mKeima was a generous gift from Prof. Richard Youle (Addgene, 131626, Watertown, MA). Lentiviral plasmids were packaged using psPAX2 and pMD2.G in HEK293T cells with Xtremegene-HP transfection reagent (Merck Millipore, 6366236001), according to the manufacturer’s instructions. The medium supernatant containing viruses was collected 48 and 72 h post-transfection and concentrated using Lenti-X transfection reagent (Takara, 631231, Mountain View, CA) with overnight incubation, followed by centrifugation at 2000 rpm for 1 h at 4 ºC. The pellet was then resuspended in DMEM-F12 medium and stored at − 80 ºC for further use. At 72–96 h prior to imaging, fibroblasts or DA neurons were treated with the lentiviral particles. The medium was replaced with regular culture medium at 24 h post-treatment and incubated for another 48 h before imaging. Confocal imaging DMEM-F12 medium (Gibco, 21041025) or BrainPhys medium (STEMCELL Technologies, 05796, Cambridge, UK) served as a recording buffer for all live cell imaging of fibroblasts and DA neurons, respectively. All images were acquired using a Carl Zeiss LSM 880 Confocal Laser Scanning microscope with Zen Black Software and a 63 × /1.40 oil immersion lens at 37 °C. Fixed and stained fibroblasts were imaged with the same equipment. Lysosensor Yellow/Blue DND 160 The cells were washed with Dulbecco’s Phosphate-Buffered Saline (DPBS) and incubated for 4 min in the recording buffer containing 3 μM Lysosensor Yellow/Blue DND 160 (Invitrogen, L7545, Carlsbad, CA) at 37 °C. After incubation, the dye was washed twice with PBS and imaged in fresh recording buffer using the 355 nm UV laser. The dual emission maxima of the dye (440 nm/540 nm) were detected using spectral scanning (lambda scan mode) on the Zeiss LSM 880. Images were subsequently linearly unmixed employing the ZEN Black software to separate the 440 nm and 540 nm emission images. The 440/540 nm fluorescence ratio was then calculated using the Ratio Plus plugin on ImageJ/Fiji software to determine the lysosomal acidity of fibroblasts and DA neurons. Lysosomal pH measurement The lysosomal pH was measured using Lysosensor Yellow/Blue DND-160 (Invitrogen, L7545). Fibroblasts cultured on fluorodishes were treated with pH calibration buffer containing Monensin (Sigma, M5273) and Nigericin (Sigma, N7143), with a pH ranging from 3.5 to 7.0. The cells were then stained with Lysosensor Yellow/Blue DND 160 (final concentration, 2 µM) for 5 min at room temperature and imaged as mentioned above. The 440/540 fluorescence ratio was measured for each calibration buffer. A lysosomal pH calibration curve was established by correlating the 440/540 nm ratio with the corresponding pH values using ImageJ/Fiji software. Measurement of lysosomal proteolytic activity by DQRed BSA Lysosomal proteolytic activity was assessed using the DQRed BSA trafficking assay dye (Invitrogen, D12051). DA neurons were washed with DPBS, and pre-warmed medium containing DQRed BSA was added to the fluorodishes. The dishes were incubated at 37 °C for three hours, washed with DPBS and recording buffer was added prior to imaging. DQRed BSA fluorescence was excited using a 561 nm Argon laser, and the emitted fluorescence was collected within the 564–740 nm range. The fluorescence intensity was quantified per cell using the ImageJ/Fiji software with consistent threshold settings across all samples. Measurement of ΔΨm using TMRM Fibroblasts or DA neurons were washed with DPBS and incubated in recording buffer containing 20 nM TMRM (Thermo Scientific, T668, Waltham, MA) at 37 °C for 30 min. Following incubation, fresh recording buffer containing 20 nM TMRM was added before proceeding with imaging. The cells were excited for TMRM fluorescence at 561 nm, and images were acquired as Z-stacks. The maximum intensity projection images were then used to quantify the fluorescence intensity using ImageJ/Fiji software with consistent threshold settings across all samples. The TMRM images were also used to analyze the mitochondrial morphology [23]. Measurement of Lipofuscin levels Lipofuscins are excited at 355 nm and fluoresce across a range of 480 to 700 nm [24]. Fibroblasts were washed with the recording buffer and imaged using UV illumination at 355 nm. Images were captured by spectral scanning, and subsequently linearly unmixed using ZEN Black software to separate the 460 nm and 480 nm emission images, thus distinguishing NAD(P)H from lipofuscins. The spectrally unmixed images at 480 nm were then utilized to calculate lipofuscin density per cell using ImageJ/Fiji software. Colocalization of the Rhodamine-B-labelled acidic NPs with lysosomes Fibroblasts were treated with Rhodamine B-labelled acidic NPs for 12–16 h prior to imaging. On the imaging day, cells were washed with the recording buffer and incubated with 50 nM LysoTracker Blue DND-22 dye (Invitrogen, L7525) for 2 h at 37 °C. Following the incubation, the cells were washed with the recording buffer and imaged in fresh buffer. Images were acquired using sequential excitation (405 nm for LysoTracker Blue and 561 nm for Rhodamine) and emission ranges set to 420–480 nm and 570–620 nm, respectively. Colocalization Pearson’s coefficient was quantified using ImageJ/Fiji software with consistent threshold settings across all samples. Measurement of mitophagy using mt-Keima reporter Measurement of mitophagy was performed as previously described [25]. Untreated fibroblasts and DA neurons were tranduced with lentiviral mt-Keima particles 72 h prior to imaging. At 12–16 h before imaging, the cells were treated with acidic NPs or rapamycin. The cells were imaged using two sequential excitation wavelengths (458 nm for green fluorescence and 561 nm for red fluorescence) with an emission range of 570–695 nm. The laser power was set at the minimum output to allow clear visualisation of the mt-Keima signal. The high F543:F458 ratio values were generated using the Ratio Plus plugin in ImageJ/Fiji and used as an index of mitophagy. Fluorescence lifetime imaging microscopy (FLIM) quantification of ATP6V1B2-mNeonGreen Fluorescence lifetime imaging was conducted using single-photon excitation on a multimodal time-resolved fluorescence microscope as previously described [26]. This setup encompassed an 80-MHz, near-infrared, femtosecond excitation source (Insight X3, Spectra Physics, Crewe, UK), a second harmonic generation unit (Harmonixx SHG, APE, Berlin, Germany), a laser scanning unit (DCS-120, Becker & Hickl, Berlin, Germany), an inverted microscope (Axio Observer 7, Zeiss, Cambridge, UK) featuring a high numerical aperture objective (Plan-Apochromat 63x/1.4 Oil M27, Zeiss), an ultrafast hybrid detector (HPM-100–07, Becker & Hickl), and time-correlated single photon counting electronics (SPC-180NX, Becker & Hickl). Images were acquired using 473 nm excitation to minimise the ratio of acceptor to donor excitation, along with 500–540 nm emission filtering to isolate fluorescence from the mNeonGreen donor. Photon counts were acquired for two minutes and histogrammed at 14.6 ps intervals. Curve fitting analysis was performed in SPCImage (Becker & Hickl). Lysosomal enrichment assay The lysosomal enrichment assay was conducted using the Lysosome Enrichment Kit for Tissues and Cultured Cells (Thermo Scientific, 89839) according to the manufacturer’s instructions. Briefly, the DA neurons were pelleted and lysed with lysosome enrichment reagent A, which contained 1 × protease and phosphatase inhibitors. The solution was then sonicated on ice, applying 12 bursts at 9 W of power (Thermo Scientific) and mixed with lysosome enrichment reagent B. The solution was subsequently centrifuged at 500 × g for 10 min. Then 200 µL of supernatant was aliquoted for use as cell supernatant for Western blotting. Lysosomes were isolated from the remaining solution through gradient centrifugation. The protein concentrations of the lysosomes and cell supernatant were then quantified using a BCA assay and proceeded for western blotting as discussed below. SDS-PAGE and immunoblotting Fibroblasts and DA neurons were washed with ice-cold PBS, followed by addition of 150 μL of ice-cold RIPA lysis buffer (Sigma-Aldrich, R0278) supplemented with protease inhibitors (Roche 4693116001, Basel, Switzerland), PMSF (Sigma, 93482), and phosphatase inhibitors (Roche 4906837001). Cells were scraped using a plastic scraper, and the lysates were transferred to 1.5 mL tubes. The lysates were rotated at 4 °C for 30 min and sonicated (3 cycles, 3 s each at 40% amplitude, with 5-min intervals). Samples were centrifuged at 16,000 × g for 30 min at 4 °C, and the supernatant was collected. Protein concentration was determined using a BCA assay kit (Thermo Scientific, 23227). A total of 20–30 μg of protein was diluted with RIPA buffer and mixed with NuPAGE 4 × sample buffer (Invitrogen, NP0007). The samples were heated at 95 °C for 5 min (for OXPHOS proteins, the lysates were heated at 45 ºC for 5 min). Proteins were separated on 4%–12% NuPAGE Bis–Tris polyacrylamide gels (Invitrogen, NP0335) immersed in MOPS running buffer (Invitrogen, NP0001). Proteins were transferred to PVDF membranes (Merck Millipore, IPFL00010) activated in methanol using a wet transfer system. Membranes were blocked in Superblock blocking buffer (Invitrogen, 37545) for 1 h at room temperature. The blots were cut where necessary before incubating with primary antibodies (Table S2), which were diluted in 1 × blocking buffer, and incubated with the membranes overnight at 4 °C. After three 10-min washes in TBST, the membranes were incubated with secondary antibodies (IRDye® 680RD Goat anti-Mouse IgG, 926–68070; IRDye® 800CW Goat anti-Rabbit IgG, 926–32211, Li-COR Biosciences; 1:10,000, Lincoln, NE) diluted in 1% BSA/TBST for 1 h at room temperature. The membranes were washed three additional times with TBST. Fluorescent signals were developed using a LiCOR Odyssey CLx system. Immunostaining Fibroblasts were grown on coverslips, treated, and fixed with 4% (w/v) paraformaldehyde. After fixation, the cells were permeabilised with 50 µg/mL digitonin in PBS for 10 min. The cells were then washed, blocked with 3% BSA, and incubated with primary antibodies (Table S2) for citrate synthase, LAMP1, and LC3 in 3% BSA for 1 h at room temperature, followed by incubation with Alexa Fluor 488/594/647-conjugated secondary antibodies for 1 h at room temperature. Coverslips were mounted on glass slides, and images were acquired using appropriate excitation and emission filters to capture fluorescent signals. Colocalization Pearson’s coefficient (Mito vs LC3 and Mito vs LAMP1) was quantified using ImageJ/Fiji with consistent threshold settings across all samples. Mitochondrial isolation Mitochondria were isolated according to the method described earlier [27] and modified for the neuronal cultures. Briefly, DA neuron cell pellets were resuspended in mitochondria isolation buffer (MIB1: 225 mM mannitol, 75 mM sucrose, 5 mM HEPES, 1 mM EGTA and 1 mg/mL fatty acid free BSA) and homogenised using an ice-cold Dounce tissue grinder tube (appropriate for 1–1.5 mL homogenisation volume). The homogenate was centrifuged, and the mitochondrial pellet was washed and pellet down in MIB without BSA (MIB2) according to the protocol. The final pellet was resuspended in a very small volume of MIB2, and the protein concentration was determined using BCA Protein assay kit (Thermo Scientific, 23227) according to the manufacturer’s specifications. The isolated mitochondria were then used for respiratory measurements. Measurement of oxygen consumption rate Measurements of mitochondrial respiration were conducted with the Seahorse Bioscience XFe96 bioanalyzer using the Seahorse XF Cell Mito Stress Test Kit (Agilent, 103015–100, Santa Clara, CA). Final maturation of iPSCs into DA neurons was performed in XF96 cell culture microplates (Agilent, 102416–100). On the day of the experiment, the culture medium was replaced with Seahorse XF Base medium (Agilent, 103334–100) supplemented with 1 mM pyruvate (Gibco, 11360070), 2 mM glutamine (Gibco, 25030081) and 10 mM glucose (Gibco, A2494001) and incubated for 30 min at 37 °C in a CO2-free incubator before loading into the Seahorse Analyzer. After measuring basal respiration, the drugs oligomycin (5 µM), FCCP (1.5 µM), and rotenone/antimycin A (0.5 µM/0.5 µM) were added to each well in sequential order. After the assay, cells were stained with Hoechst 33342 (5 µM; Thermo Scientific, 62249) for 30 min. ImageXpress was then used to count the number of cell nuclei (cell numbers) in each well and normalised to obtain the basal respiration rate values. The respiratory measurements of isolated mitochondria from mixed neuronal cultures were performed as described earlier [28] using tht Seahorse XFe96 cell culture plate assay. Briefly, 5 or 10 µg of mitochondrial protein was resuspended in 30 µL (1 well) of individual substrate mix (for example, pyruvate/malate substrate + mitochondria assay solution [MAS]) and plated into each well. The cell culture plate was centrifuged at 2000 × g for 20 min at 4 °C to form a uniform layer of mitochondria at the bottom. After centrifugation, 150 µL of substrate solution was carefully added to each well. Fresh injection solutions were made in MAS without BSA and loaded into the cartridge, and calibrated according to the manufacturer’s specifications. After calibration, the culture plate with mitochondria was inserted, and the assay was run essentially as described in the method. Transmission electron microscopy (TEM) TEM was performed as previously described [25]. Briefly, fibroblasts and DA neurons grown on coverslips were fixed in electron microscopy fixative containing 2% glutaraldehyde (EMS, 16365, Hatfield, PA) and 2% paraformaldehyde (EMS, 15710) in 0.1 M sodium cacodylate for 1 h. Following fixation, cells were washed with 0.1 M cacodylate buffer (EMS, 11650) and then fixed in a solution of 1% osmium tetroxide (EMS, 19150) and 1% potassium ferricyanide (EMS, 25120–20) in 0.1 M sodium cacodylate. This was followed by sequential dehydration using ethanol. Coverslips were then embedded in epoxy resin (Agar Scientific Ltd., Araldite Kit CY212, Essex, UK) according to standard protocols. The embedded samples were sectioned into 50-nm slices using an ultramicrotome equipped with a diamond knife and mounted onto copper TEM grids. The grids were stained with lead citrate for 3 min before imaging. Images were captured using a JEOL1400 Transmission Electron Microscope at magnifications ranging from 800 × to 1200 × (digital magnification). Images were imported into ImageJ/Fiji to quantify mitochondrial area, mitochondrial cristae density and mitochondrial aspect ratio. The autophagic vesicles were classified as autophagosomes, autolysosomes and lysosomes as per [29]. Mitochondria and autophagic vesicles were manually traced to measure their areas, and the structures were verified by an electron microscopy expert. Quantification was performed in a double-blind manner. Quantification and statistical analysis All statistical analyses were performed using GraphPad Prism (version10; GraphPad Software) and verified by a qualified statistician. Sample sizes (n) for each experiment represent values from at least 3 independent experiments. For comparisons between two datasets, paired t-test or two-tailed unpaired t-tests were used for normally distributed data; when normality was not met, the Mann–Whitney test was used. For multi-group comparisons, one-way ANOVA was performed, followed by the Kruskal–Wallis test, with Dunn’s multiple comparisons correction applied. For experiments involving independent comparisons within the group, ordinary two-way ANOVA with Tukey’s multiple comparisons test (single pooled variance), Šidák multiple comparisons or Uncorrected Fisher’s LSD (single pooled variance) was used. Samples normalised to a reference value of 1 were analysed using one-sample Wilcoxon rank t-tests (Table S3). Where multiple comparisons were conducted, adjusted P-values are reported following the appropriate post hoc correction. Statistical significance was defined as adjusted P < 0.05. The mitochondrial morphology parameters were heavily skewed, hence the values were log(Y) transformed before calculating the effect size, confidence intervals and unpaired t-test. Autophagic vesicle counts in neurons contained zeros; therefore, values were log(1 + Y) transformed prior to two-way ANOVA. Effect sizes are reported as partial eta-squared (partial η2) with 95% confidence intervals. Data are presented as mean ± standard deviation (SD). Effect sizes were calculated accordingly (for Mann–Whitney’s test comparison—Hedges’ g; Unpaired t-test – Cohen’s d; one-sample t-test—Glass's ∆; for one-way ANOVA—ω2; for two-way ANOVA—partial η2). The noncentral F distribution method (for ANOVA) or the Bootstrap (all others) was used to compute 95% confidence intervals to aid interpretation of the magnitude and precision of observed effects. Wherever possible, individual data points and raw data distributions are shown in figures, and all raw numerical values used to generate graphs are provided in Additional file 3. The statistical test, effect size, and confidence interval values for each graph are included in Additional file 2 to promote transparency and reproducibility. Results To characterise the impact of PD-related GBA1 mutations on lysosomal and mitochondrial function, we examined three lines of human dermal fibroblasts and six iPSC lines derived fromGBA1-PD patients carrying E326K or N370S mutation, along with two healthy controls. We also included one CRISPR-corrected isogenic control iPSC line each for GBA1-N370S and GBA1-E326K iPSCs (Table S1). We first performed the experiments in fibroblasts as a proof of principle, and subsequently confirmed the key findings in iPSC-derived DA neurons. Lysosomal function is impaired in cells with PD-related GBA1 mutations As GCase is a lysosomal enzyme, our investigation began by characterising the lysosomes in fibroblasts and DA neurons from both control and PD patient-derived iPSCs. The GBA1 protein expression levels measured by Western blot were not significantly different between the control and GBA1 mutant neurons; however, GCase enzymatic activity was significantly reduced in iPSC-DA neurons and fibroblasts carrying either GBA1 mutation (Fig. 1a, b and Fig. S1a, b). Lysosomal acidification was measured using the ratiometric pH-sensitive dye, Lysosensor Yellow/Blue DND160. The data revealed impaired acidification in both E326K and N370S DA neurons and fibroblasts (Fig. 1c and Fig. S1c, d). The lysosomal pH values were 4.74 ± 0.03 and 4.77 ± 0.02 in the control cells but between 4.89 ± 0.02 and 5.06 ± 0.03 in the mutant fibroblasts (Fig. 1d). The function of endocytic trafficking and lysosomal proteolytic activity was measured using the DQRed BSA assay. DQRed BSA intensity was significantly reduced in the GBA1 mutant neurons, indicating impaired lysosomal degradative function in GBA1-PD cells (Fig. 1e and Fig. S1e). Autofluorescence excited at 355 nm and imaged between 400–600 nm showed an unusually bright extramitochondrial component in GBA1-PD fibroblasts (Fig. 1f) that was significantly greater than control cells. Spectral scanning and linear unmixing separated the expected mitochondrial NAD(P)H signal with an emission peak at 450 nm and an extensive non-mitochondrial signal with a peak emission at 480 nm, attributed to the accumulated lipofuscin. Lipofuscins are cytoplasmic granules generated as a consequence of autophagy and phagocytosis processes, which show auto-fluorescence over a broad spectrum ranging from 480 to 700 nm when excited by ultraviolet or blue light [24]. The accumulation of lipofuscin, which serves as an indicator of ageing, is consistent with impaired lysosomal acidification and defective autophagy [30]. Accumulation of lipofuscin is also a hallmark of Batten Disease (neuronal ceroid-lipofuscinoses), a severe early-onset neurodevelopmental disorder with progressive neurodegeneration also associated with lysosomal dysfunction [31]. These data collectively indicate significant lysosomal dysfunction in the GBA1-PD cells.Fig. 1Lysosomal abnormalities in GBA1-PD DA cells. a Representative western blot images of GBA1 protein and quantification in isogenic control and GBA1-E326K and N370S DA neurons. b GCase activity in GBA1-E326K (left) and GBA1-N370S (right) DA neurons and respective isogenic controls. c Representative ratioed images and quantification of fluorescence ratio of GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls labelled with Lysosensor Yellow/Blue DND 160, indicating lysosomal pH in cells. Scale bars, 10 µm, d Histogram depicting the mean lysosomal pH values in GBA1 mutant fibroblasts. e Representative confocal images and quantification of relative intensity in GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls stained with DQRed BSA indicative of proteolytic activity. Scale bars, 10 µm, f Representative confocal images and quantification of lipofuscins in GBA1 fibroblasts. Scale bars, 10 µm. Data are presented as mean ± SD; one-sample Wilcoxon rank t-test, unpaired t-test or One-way ANOVA with Kruskal–Wallis and Dunn’s multiple comparisons test. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are provided in Additional file 2 Lysosomal abnormalities in GBA1-PD DA cells. a Representative western blot images of GBA1 protein and quantification in isogenic control and GBA1-E326K and N370S DA neurons. b GCase activity in GBA1-E326K (left) and GBA1-N370S (right) DA neurons and respective isogenic controls. c Representative ratioed images and quantification of fluorescence ratio of GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls labelled with Lysosensor Yellow/Blue DND 160, indicating lysosomal pH in cells. Scale bars, 10 µm, d Histogram depicting the mean lysosomal pH values in GBA1 mutant fibroblasts. e Representative confocal images and quantification of relative intensity in GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls stained with DQRed BSA indicative of proteolytic activity. Scale bars, 10 µm, f Representative confocal images and quantification of lipofuscins in GBA1 fibroblasts. Scale bars, 10 µm. Data are presented as mean ± SD; one-sample Wilcoxon rank t-test, unpaired t-test or One-way ANOVA with Kruskal–Wallis and Dunn’s multiple comparisons test. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are provided in Additional file 2 Mitochondria are dysfunctional in cells with GBA1-PD mutations To determine whether the impaired lysosomal activity in the presence of GBA1 mutations has a secondary impact on mitochondrial morphology and function, we measured ΔΨm using the potentiometric fluorescent reporter TMRM. ΔΨm was significantly reduced in E326K and N370S GBA1 mutant DA neurons and fibroblasts (Fig. 2a and Fig. S2a, b). The lower ΔΨm was accompanied by mitochondrial fragmentation in the N370S and E326K DA neurons and fibroblasts (Fig. 2a and Fig. S2a). Ultrastructure analysis using electron microscopy revealed significant differences in mitochondrial morphology in the GBA1 mutant neurons, including reduced cristae density and increased mitochondrial area, along with decreased mitochondrial aspect ratio, indicating mitochondrial swelling and fragmentation in the GBA1 mutant iPSC DA neurons (Fig. 2b). Measurements of oxygen consumption rates using the ‘Seahorse’ respirometry system revealed ~ 40% decrease in basal oxygen consumption rate in the GBA1 mutant DA neurons compared with the isogenic controls (Fig. 2c). To exclude changes in mitochondrial mass and to explore the substrate dependence of mitochondrial respiration, we used mitochondria isolated from the cultures. This allowed measurements of respiratory rates with different substrates, favouring complex I (pyruvate/malate) or complex II (rotenone/succinate). ADP-stimulated respiration (State 3) rate using both complex I- and complex II-dependent substrates was significantly reduced in the GBA1 mutant neurons compared to the control neurons (Fig. 2d and Fig. S2c). Western blotting of the OXPHOS complex proteins using a cocktail of antibodies to respiratory chain proteins, revealed a significant increase in the expression of complex IV in E326K DA neurons (Fig. 2e and Fig. S2d) and increased expression of complex I proteins in N370S DA neurons compared to the isogenic controls (Fig. 2e and Fig. S2d). Complex III was reduced in both E326K and N370S DA neurons compared to the isogenic controls while the expression of other respiratory chain proteins remained unaltered. None of these data suggested any defect in complex I assembly or function. These data collectively point to impaired mitochondrial bioenergetic function in fibroblasts and neurons carrying GBA1 mutations.Fig. 2Mitochondrial dysfunction in GBA1-PD DA neurons. a Representative confocal images of GBA1-E326K and GBA1-N370S DA neurons along with isogenic controls stained with TMRM and subsequent quantification of fluorescent intensity to measure ΔΨm (lower, left two) and mitochondrial fragmentation count (lower, right two). Scale bars, 10 µm. b Representative ultramicroscopic images of isogenic control and GBA1-E326K and GBA1-N370S DA neurons (upper) and Box-and-whisker plots (below). The white arrows indicate normal mitochondria; the black arrows indicate abnormal mitochondria. Scale bars, 0.5 µm. Box-and-whisker plots show the minimum, maximum, median, and interquartile range of the log(Y) values of the morphological parameters, including average number of mitochondria, average cristae density in mitochondria, average mitochondrial area and mitochondrial aspect ratio. c Histograms representing the basal respiration rate in GBA1 mutant iPSC-DA neurons from GBA1-E326K and GBA1-N370S mutant lines and respective isogenic controls. d Upper, representative seahorse plots showing oxygen consumption rates in iPSC-DA neurons from GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls, using various substrates. Lower, histograms representing State 3 (ATP Production) in mitochondria isolated from GBA1-E326K and GBA1-N370S DA neurons and isogenic controls, as measured by seahorse assay. e Representative immunoblotting images and subsequent quantification of OXPHOS complex proteins in GBA1 E326K and N370S DA neurons. Data presented as mean ± SD; Statistics: Mann–Whitney’s test, log-transformed and unpaired t-test, one-sample Wilcoxon rank test or Ordinary two-way ANOVA with Šidák multiple comparisons test with single pooled variance. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are reported in Additional file 2 Mitochondrial dysfunction in GBA1-PD DA neurons. a Representative confocal images of GBA1-E326K and GBA1-N370S DA neurons along with isogenic controls stained with TMRM and subsequent quantification of fluorescent intensity to measure ΔΨm (lower, left two) and mitochondrial fragmentation count (lower, right two). Scale bars, 10 µm. b Representative ultramicroscopic images of isogenic control and GBA1-E326K and GBA1-N370S DA neurons (upper) and Box-and-whisker plots (below). The white arrows indicate normal mitochondria; the black arrows indicate abnormal mitochondria. Scale bars, 0.5 µm. Box-and-whisker plots show the minimum, maximum, median, and interquartile range of the log(Y) values of the morphological parameters, including average number of mitochondria, average cristae density in mitochondria, average mitochondrial area and mitochondrial aspect ratio. c Histograms representing the basal respiration rate in GBA1 mutant iPSC-DA neurons from GBA1-E326K and GBA1-N370S mutant lines and respective isogenic controls. d Upper, representative seahorse plots showing oxygen consumption rates in iPSC-DA neurons from GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls, using various substrates. Lower, histograms representing State 3 (ATP Production) in mitochondria isolated from GBA1-E326K and GBA1-N370S DA neurons and isogenic controls, as measured by seahorse assay. e Representative immunoblotting images and subsequent quantification of OXPHOS complex proteins in GBA1 E326K and N370S DA neurons. Data presented as mean ± SD; Statistics: Mann–Whitney’s test, log-transformed and unpaired t-test, one-sample Wilcoxon rank test or Ordinary two-way ANOVA with Šidák multiple comparisons test with single pooled variance. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are reported in Additional file 2 Mitophagy is impaired in cells with GBA1-PD mutations We then set out to explore the underlying mechanisms that link impaired lysosomal function to impaired mitochondrial bioenergetics in the GBA1-PD fibroblasts and DA neurons. A logical mechanism linking impaired lysosomal function with mitochondrial dysfunction might operate through dysfunctional mitophagy and our earlier work also established dysfunctional mitophagy in Gba1 KO mice [6]. We quantified mitophagy using the dual excitation probe mt-Keima (Fig. 3a). mt-Keima is a pH-sensitive probe that measures the fraction of mitochondria in neutral (pH 7–7.8 in the mitochondrial matrix) versus acidic pH (pH 4.5–4.7 in lysosomes) environment [32]. The mt-Keima signal ratio was significantly reduced in the fibroblasts and DA neurons carrying the GBA1-N370S or GBA1-E326K mutation, indicating impaired mitophagy (Fig. 3a and Fig. S3a, b). Western blot analysis quantifying the status of autophagy pathways in the GBA1 mutant neurons revealed increased LAMP1 levels in E326K and N370S neurons, along with increased pMTORC1/MTORC1 ratio in the two mutant types. While expression of p62 was increased in the E326K mutant iPSC DA neurons, these were not significantly altered in the N370S DA neurons (Fig. 3b). Since the lysosomal pH was increased in GBA1-PD cells (Fig. 1d), the reduced mt-Keima signal in GBA1-PD cells could be attributed either to the altered lysosomal pH or to the impaired fusion of autophagosomes to lysosomes. To address this, we performed triple staining of fibroblasts against citrate synthase (CiS, a marker for mitochondria), LAMP1 (a marker for lysosomes), and LC3 (a marker for autophagosomes). Subsequent imaging revealed that although there was no significant difference in the density of mitophagosomes (mitochondria and autophagosome colocalization; yellow arrows) between GBA1 mutant and control fibroblasts, the density of mitolysosomes (colocalization of mitochondria and lysosomes; white arrows) was significantly increased in the GBA1 mutant fibroblasts (Fig. 3c). Characterisation of electron micrographs for specific autophagic vesicle types based on Neikirk et al., 2023 [29], showed that the numbers of autophagic vesicles, specifically autolysosomes and lysosomes, were significantly increased in both GBA1 mutant fibroblasts and neurons (Fig. 3d and Fig. S3c). Briefly, in Fig. 3d and Fig. S3c, autophagosomes typically display clear double membranes or a single membrane, darker than the surrounding tissue in TEM images, with more circular shapes inside, indicating cargo (red arrows). Lysosomal membranes show highly organised inner folds, while lysosomal enzymes are characterised by a darker, more uniform appearance (green arrows). Autolysosomes are intermediate structures larger than lysosomes, frequently containing intact autophagosome contents (yellow arrows). Distinct dark spots (black arrows) may indicate protein aggregates; further work is needed to confirm their identity. The autophagosomes also appeared to be fully enclosed, ruling out incomplete phagophore formation as the underlying mechanism of the autophagy defect. These data indicated that mitophagy is significantly impaired in cells carrying the GBA1-PD mutations and the decrease in mitophagy is likely due to the impaired pH rather than improper phagophore formation.Fig. 3Mitophagy defects in GBA1-PD DA neurons. a Representative images of GBA1-E326K and GBA1-N370S DA neurons and isogenic controls transduced with mt-Keima plasmid and imaged with excitation at 458 and 543 nm. Scale bars, 10 µm. The ratio of signals at 543/458 nm was quantified. b Representative western blots of GBA1 mutant and isogenic control lysates probed for autophagic proteins and subsequent quantifications. c Representative immunofluorescence images of control and N370S fibroblasts probed against citrate synthase (CiS), LAMP1 and LC3, labelling mitochondria, lysosomes and autophagosomes, respectively and subsequent quantification of colocalization co-efficient of mitophagosomes (yellow arrows) and mitolysosomes (white arrows). Scale bar, 10 µm. d Representative TEM images and quantification of control and E326K mutant fibroblasts capturing different autophagy phases in the cells. Black arrows indicate protein aggregates, yellow arrows indicate autolysosomes, red arrows indicate autophagosomes, and green arrows indicate lysosomes in the cells. Scale bars, 1 µm and 0.5 µm (magnifications). Data presented as mean ± SD. Statistics: Mann–Whitney’s test, one-sample Wilcoxon rank test, One-way ANOVA with Holm-Šidák multiple comparisons test or Ordinary two-way ANOVA with Šidák multiple comparisons test. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are in Additional file 2 Mitophagy defects in GBA1-PD DA neurons. a Representative images of GBA1-E326K and GBA1-N370S DA neurons and isogenic controls transduced with mt-Keima plasmid and imaged with excitation at 458 and 543 nm. Scale bars, 10 µm. The ratio of signals at 543/458 nm was quantified. b Representative western blots of GBA1 mutant and isogenic control lysates probed for autophagic proteins and subsequent quantifications. c Representative immunofluorescence images of control and N370S fibroblasts probed against citrate synthase (CiS), LAMP1 and LC3, labelling mitochondria, lysosomes and autophagosomes, respectively and subsequent quantification of colocalization co-efficient of mitophagosomes (yellow arrows) and mitolysosomes (white arrows). Scale bar, 10 µm. d Representative TEM images and quantification of control and E326K mutant fibroblasts capturing different autophagy phases in the cells. Black arrows indicate protein aggregates, yellow arrows indicate autolysosomes, red arrows indicate autophagosomes, and green arrows indicate lysosomes in the cells. Scale bars, 1 µm and 0.5 µm (magnifications). Data presented as mean ± SD. Statistics: Mann–Whitney’s test, one-sample Wilcoxon rank test, One-way ANOVA with Holm-Šidák multiple comparisons test or Ordinary two-way ANOVA with Šidák multiple comparisons test. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are in Additional file 2 V-ATPase complex formation is impaired in GBA1-PD Assembly of the pH regulatory component in lysosomes, the vacuolar-type H+ ATPase (V-ATPase), is regulated by MTORC1 [33], which was found to be constitutively phosphorylated in the GBA1-PD mutant cells in this study. Ratto et al. [33] demonstrated that under the nutrient-deprived conditions, MTORC1 is inactive and is distributed in the cytosol, enabling the peripheral ATP6V1 to bind to the membrane-bound ATP6V0 domain, forming a functional V-ATPase complex allowing proton exchange and acidification of the lysosomes. Under nutrient-rich conditions, MTORC1 is phosphorylated and remains on the lysosomal membrane, preventing the formation of a functional V-ATPase complex [33]. While the V-ATPase maintains the lysosomal pH, disruption of the complex does not hamper autolysosome formation [34]. Total protein estimation through Western blotting revealed no significant difference in the expression of ATP6V0D2, ATP6V1A or ATP6V1H between the control and mutant DA neurons (Fig. 4a). To examine their expression levels specifically in lysosomes, we performed a lysosomal enrichment assay and probed for pMTORC1, LAMP1, and the ATP6V0 and ATP6V1 components (Fig. 4b). Equal amounts of cell lysates and purified lysosomes (25 mg) were used to run the western blots. LAMP1 was used as a marker to verify lysosomal enrichment and GAPDH was used as a marker for cell lysates (Fig. 4b). For quantification (Fig. 4b), the bands in the lysosome-enriched samples were normalised to the protein levels in the lysosome-enriched samples from untreated isogenic controls. Since pMTORC1 represents the active form of MTORC1 and is found on the lysosomal membrane, we measured only pMTORC1 levels. The protein level of pMTORC1 was increased up to two folds, and that of LAMP1 was increased to ~ 1.5 folds in the GBA1-mutant DA neurons. However, ATP6V1A protein level was reduced by ~ 40% and ATP6V1H by ~ 15% within the lysosomes fractionated from GBA1 mutant DA neurons, while ATP6V0D2 levels remained unchanged between the control and the mutant iPSC-DA neurons (Fig. 4b). Furthermore, overnight treatment of GBA1 iPSC-DA neurons with the MTORC1 inhibitor rapamycin (200 nM) decreased the protein levels of pMTORC1 and LAMP1, while increasing the levels of ATP6V1A and ATP6V1H in the lysosomes. The ATP6V0D2 levels remained constant (Fig. 4b).Fig. 4Impaired V-ATPase complex formation in GBA1-PD cells. a Representative images of western blots probed against LAMP1 and V-ATPase complex proteins and quantification of protein levels in total cell lysates from isogenic control and GBA1-E326K and GBA1-N370S mutant DA neurons. b Representative immunoblotting images of cell supernatant and lysosomes separated by lysosome enrichment assay in untreated and 200 nM rapamycin-treated GBA1-PD DA neurons and isogenic controls, and subsequent quantification. c Representative FLIM images of ATP6V1B2-mNeonGreen in control and GBA1-PD fibroblasts co-transfected with ATP6Voa3-mScarlet and treated with/without 200 nM rapamycin, and subsequent quantification of mean lifetime (τm) and representative decay plots and exponential fit of ATP6V1B2-mNeongreen in the Control and N370S and E326K mutant fibroblasts. d Time-point images of ATP6V1B2-mNeongreen in control and GBA1-PD fibroblasts co-transfected with ATP6Voa3-mScarlet and treated with 1 µM rapamycin and imaged over a period of 40 min, and subsequent quantification of τm. Data presented as mean ± SD; Statistics: one-sample Wilcoxon rank test or Ordinary two-way ANOVA with Tukey’s multiple comparisons test with single pooled variance. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are in Additional file 2 Impaired V-ATPase complex formation in GBA1-PD cells. a Representative images of western blots probed against LAMP1 and V-ATPase complex proteins and quantification of protein levels in total cell lysates from isogenic control and GBA1-E326K and GBA1-N370S mutant DA neurons. b Representative immunoblotting images of cell supernatant and lysosomes separated by lysosome enrichment assay in untreated and 200 nM rapamycin-treated GBA1-PD DA neurons and isogenic controls, and subsequent quantification. c Representative FLIM images of ATP6V1B2-mNeonGreen in control and GBA1-PD fibroblasts co-transfected with ATP6Voa3-mScarlet and treated with/without 200 nM rapamycin, and subsequent quantification of mean lifetime (τm) and representative decay plots and exponential fit of ATP6V1B2-mNeongreen in the Control and N370S and E326K mutant fibroblasts. d Time-point images of ATP6V1B2-mNeongreen in control and GBA1-PD fibroblasts co-transfected with ATP6Voa3-mScarlet and treated with 1 µM rapamycin and imaged over a period of 40 min, and subsequent quantification of τm. Data presented as mean ± SD; Statistics: one-sample Wilcoxon rank test or Ordinary two-way ANOVA with Tukey’s multiple comparisons test with single pooled variance. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are in Additional file 2 To assess the V-ATPase assembly, we transfected the control and GBA1 mutant fibroblasts with ATP6V1B2-mNeongreen and ATP6Voa3-mScarlet and quantified their FRET interaction using FLIM [35]. The data were best fit to a biexponential decay function. The ATP6V0a3 construct utilizes the more rapidly maturing but photophysically heterogeneous acceptor variant mScarlet-I [36]. This exhibits at least two fluorescence lifetimes, meaning the biexponential decay of the donor is an oversimplification given the likelihood of contrasting FRET rates to each acceptor species [37, 38]. The results are therefore presented as the mean (amplitude weighted) fluorescence lifetime, τm, which will nevertheless respond to variations in FRET without having to interpret the meaning of individual decay components and amplitudes. The τm of ATP6V1B2-mNeongreen averaged 617.72 ± 25.3 ps in control fibroblasts but increased to 760.88 ± 20.36 ps in N370S and 811.03 ± 17.04 ps in E326K mutant fibroblasts, indicating an increased distance between ATP6V1B2-mNeongreen and ATP6Voa3-mScarlet signalled by a reduction in FRET. Overnight treatment with 200 nM rapamycin significantly decreased τm to 652.7 ± 19.69 ps and 726.05 ± 21.2 ps in E326K and N370S fibroblasts, respectively, while the τm of control fibroblasts was at 590.5 ± 32.7 ps (Fig. 4c). To validate the subcellular localization process within a condensed timeframe, we administered 1 µM rapamycin to the cells and performed FLIM every 10 min for 40 min. The mean fluorescence lifetime of ATP6V1B2-mNeongreen exhibited a progressive decline in the GBA1 mutant fibroblasts, decreasing from ~ 850 ps at 0 min to ~ 580 ps at 40 min, while τm in the control fibroblasts showed only a small reduction, decreasing from ~ 675 ps at 0 min to ~ 588 ps at 40 min (Fig. 4d). The FLIM-FRET data thus confirmed impaired assembly of the V-ATPase complex in GBA1 mutant cells. All these data are consistent with a model in which phosphorylation of MTORC1 at the lysosome membrane limits the formation of a functional V-ATPase complex in the GBA1-PD cells. Rapamycin as a potent MTORC1 inhibitor reduces MTORC1 activity and facilitates the formation of a functional V-ATPase complex. Acidification of lysosomes is sufficient to restore lysosomal and mitochondrial functions as well as mitophagy in GBA1-PD Our data showed that GBA1 mutations lead to significant impaiment of lysosomal acidification and mitophagy and suggest that this failure of cellular homeostasis may underlie the mitochondrial dysfunction seen in GBA1-PD cells. We therefore wondered whether restoring lysosomal pH could rescue these defects in GBA1-PD cells. Since impaired V-ATPase complex formation may be a consequence of MTORC1 hyperphosphorylation, we explored the impact of rapamycin (overnight treatment with 200 nM) on lysosomal and mitochondrial function. As an independent pH modulator, we employed novel poly(ethylene tetrafluorosuccinate-co-succinate) NPs, which acidify lysosomes [22]. We first validated the effects of rapamycin and the NPs in patient-derived fibroblasts (Fig. S4). To confirm that the NPs were localized to the lysosomes, we loaded the fibroblasts with rhodamine-tagged NPs and stained the cells with Lysotracker blue DND22 (Fig. S4a), which confirmed localization of the NPs to lysosomes. Overnight treatment with 180 µg/mL acidic NPs composed of poly(ethylene tetrafluorosuccinate-co-succinate) restored lysosomal pH in GBA1 mutant cells to levels comparable to those of control fibroblasts, as measured using the pH-sensing ratiometric probe Lysosensor Yellow/Blue DND160. Treatment with non-acidic control NPs did not significantly affect the lysosomal pH in the GBA1 mutant fibroblasts (Fig. S4b, c). TMRM staining demonstrated that both rapamycin and the acidic NPs significantly rescued the ΔΨm in the mutant fibroblasts (Fig. S4f, g). We also found that the ΔΨm was increased in Control 2 and E326K fibroblasts upon control NP treatment (Fig. S4g). We wondered whether the effect of control NPs on ΔΨm could be due to the succinate component in the poly(ethylene succinate) control NPs. However, treatment with 3 mM diethyl succinate for 30 min did not have any effect on the ΔΨm in the control and GBA1 mutant fibroblasts (Fig. S4h). We then assessed the impact of rapamycin and acidic NPs on GBA1 mutant iPSC-derived DA neurons (Fig. 5). Treatment with 200 nM rapamycin and 180 µg/mL acidic NPs restored lysosomal pH, ΔΨm, and mitophagy in the E326K and N370S iPSC DA neurons (Fig. 5a–c). Although there was no change in GBA1 protein level, GCase activity was increased in GBA1 mutant DA neurons following treatment with either rapamycin or acidic NPs (Fig. S4d, e). Additionally, western blot analysis of OXPHOS complexes indicated that the increased mitochondrial complex IV level in E326K DA neurons and the reduced complex I level in the N370S DA neurons were normalised upon treatment with either rapamycin or acidic NPs. Complex III levels were also rescued upon rapamycin or acidic NP treatment in both E326K and N370S DA neurons (Fig. S4i).Fig. 5Rapamycin and acidic NPs rescue lysosomal and mitochondrial dysfunction in GBA1 mutant DA neurons. a Left, representative ratioed images of GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls treated overnight with 200 nM rapamycin or 180 μg/mL acidic NPs, stained with Lysosensor Yellow/Blue DND 160. Right, quantification of fluorescence ratio indicating lysosomal pH. Scale bars, 10 µm. b Representative confocal images of GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls treated with 200 nM rapamycin or 180 μg/mL acidic NPs, stained with TMRM (left) to measure changes in ΔΨm (right). Scale bars, 10 µm. c Left, representative confocal images of GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls treated with 200 nM rapamycin or 180 μg/mL acidic NPs, probed with mt-Keima plasmid and imaged with excitation at 458 and 543 nm. Right, subsequent quantification of the ratio of signals at 543/458 nm. Scale bars, 10 µm. Data presented as mean ± SD; Statistics: Ordinary two-way ANOVA with Uncorrected Fisher’s LSD with single pooled variance. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are in Additional file 2 Rapamycin and acidic NPs rescue lysosomal and mitochondrial dysfunction in GBA1 mutant DA neurons. a Left, representative ratioed images of GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls treated overnight with 200 nM rapamycin or 180 μg/mL acidic NPs, stained with Lysosensor Yellow/Blue DND 160. Right, quantification of fluorescence ratio indicating lysosomal pH. Scale bars, 10 µm. b Representative confocal images of GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls treated with 200 nM rapamycin or 180 μg/mL acidic NPs, stained with TMRM (left) to measure changes in ΔΨm (right). Scale bars, 10 µm. c Left, representative confocal images of GBA1-E326K and GBA1-N370S DA neurons and respective isogenic controls treated with 200 nM rapamycin or 180 μg/mL acidic NPs, probed with mt-Keima plasmid and imaged with excitation at 458 and 543 nm. Right, subsequent quantification of the ratio of signals at 543/458 nm. Scale bars, 10 µm. Data presented as mean ± SD; Statistics: Ordinary two-way ANOVA with Uncorrected Fisher’s LSD with single pooled variance. P-values are noted on the graphs. The statistical tests, corresponding effect sizes, and confidence intervals for each graph are in Additional file 2 As rapamycin restored the assembly of the V-ATPase complex in GBA1 mutant cells, we asked whether the acidic NPs also function through the V-ATPase complex formation. FLIM-FRET experiments conducted on control and acidic NP-treated GBA1 mutant fibroblasts transduced with ATP6V1B2-mNeongreen and ATP6Voa3-mScarlet plasmids demonstrated a further increase in the τm of ATP6V1B2 in both control and mutant fibroblasts treated with acidic NPs, confirming that the acidic NPs operate independently of V-ATPase (Fig. S4j). Thus, the data presented here suggest that constitutive phosphorylation of MTORC1 in GBA1-PDcells leads to impaired lysosomal pH and compromised lysosomal proteolytic function. This results in the accumulation of mitolysosomes and impaired mitophagy. The compromised mitophagy culminates in mitochondrial dysfunction in GBA1 mutant cells. Furthermore, inhibiting MTORC1 activity or pharmacologically decreasing lysosomal pH using acidic NPs enhances lysosomal proteolytic activity, rescuing mitochondrial dysfunction in GBA1 mutant iPSC-DA neurons and fibroblasts. Discussion In this study, we report a range of abnormalities in the lysosomes and mitochondria of GBA1-PD fibroblasts and iPSC-DA neurons carrying E326K or N370S GBA1 mutations. Given that GBA1 is a lysosomal-resident protein, we hypothesized that mitochondrial dysfunction likely reflects a primary lysosomal defect. We used the ratiometric probe Lysosensor Yellow/Blue DND 160 to measure lysosomal pH. When calibrated, these measurements revealed that lysosomal pH in control cells ranged from 4.74 ± 0.03 to 4.77 ± 0.02, while in mutant cells it showed a significant shift to a range of 4.89 ± 0.02 to 5.06 ± 0.03. Reports in the literature indicate that the optimal pH for lysosomal function ranges between 4.2 and 4.8, depending on the cell type [8, 39]. Although the difference between control and mutant cells seems modest, the data we present suggest that this difference in mutant cells is functionally significant. In line with other studies, we observed a pH-dependent decrease in the lysosomal enzyme activity, an increase in lysosomal number, and disruption of mitophagy and mitochondrial metabolism [18, 20, 40–45]. There are very few reports discussing mitochondrial abnormalities associated with GBA1-PD [42, 46–49], and even fewer exploring the status of mitophagy pathways [10]. The TMRM data indicated reduced ΔΨm and also revealed mitochondrial fragmentation. Electron microscopy experiments supported these findings, showing that mitochondria were shorter, swollen, and mitochondrial aspect ratio was reduced compared to controls. These data, along with reduced density of cristae, collectively indicate structural abnormalities in the mitochondria. A decline in ΔΨm is known to destabilize inner membrane organization and promote both swelling and fragmentation [50], which aligns with the morphological shift toward more rounded, fragmented mitochondria observed in our analyses. We also found a significant reduction in basal respiratory rates and substrate-dependent mitochondrial respiration in the E326K and N370S DA neurons. The compromised bioenergetics and structural abnormalities collectively indicate dysfunctional mitochondria. Western blots revealed that the level of mitochondrial complex III was reduced in GBA1-E326K and N370S DA neurons compared to the respective isogenic controls. The mitochondrial complex III plays a key role in driving respiration and the maintenance of membrane potential [51]. Reduced expression of complex III could therefore account for the reduced mitochondrial membrane potential and reduced respiratory rate in GBA1-PD DA neurons. Reduced mitochondrial complex III expression has also been reported in platelets and lymphocytes of PD patients [52, 53]. However, in this study, increased complex IV levels were observed in only two E326K mutant DA neuron lines, while one N370S line exhibited increased complex I expression. This could be due to variation between subjects. More replications and experiments are needed to obtain conclusive data on mitochondrial complex levels in the E326K and N370S GBA1-PD lines. Mitochondrial dysfunction was also associated with impaired mitophagic flux in E326K and N370S GBA1-PD fibroblasts and iPSC-DA neurons. Numerous studies have suggested mechanistic evidence for a connection between PD and GBA1 mutations. For example, dysfunctional mitochondria in primary neurons and astrocytes have been described in the Gba1 knockout mouse model, associated with impairment of the autophagy-lysosomal pathway, resulting in the accumulation of ubiquitinated proteins and α-synuclein [6]. Various studies have explored the mechanisms behind the PD pathology in GBA1 mutant models. A recent study emphazised that decreased GCase activity in GBA1-PD midbrain dopaminergic neuronal cells leads to defective modulation of the untethering protein TBC1D15, which regulates Rab7 GTP hydrolysis for contact untethering. This impairment results in prolonged mitochondria-lysosome contacts, potentially affecting mitochondrial dynamics and function in GBA1-linked PD [48]. Baden et al. proposed an alternative function of GBA1 in the mitochondria in maintaining the integrity of complex I and support of energy metabolism [46], although in the present study we found no evidence for mitochondrial localization of GBA1 or impaired assembly of complex I. Respirometry assays using isolated mitochondria revealed a marked reduction in both maximal respiratory capacity and ATPase-linked oxygen consumption in GBA1-E326K and N370S mutant DA neurons, upon stimulation with complex I (pyruvate/malate) and complex II (rotenone/succinate) substrates. The impairment across both complexes points to a global mitochondrial dysfunction, potentially arising from structural abnormalities such as fragmented mitochondria and decreased cristae density—features observed in these mutants—or as a downstream consequence of lysosomal dysfunction. Notably, lysosome-related mitochondrial impairment has also been reported in TBCK (Tre-2/Bub2/Cdc16 (TBC1) domain containing Kinase) encephaloneuronopathy, supporting a broader link between lysosomal defects and mitochondrial pathology [54]. Recent evidence suggests that GBA1 mutation is a primary cause of the pathologies associated with lysosomes and mitochondria in PD. There appears to be a correlation between the onset of PD and lysosomal dysfunction caused by impaired GCase activity. Dysfunctional lysosomes disrupt cellular signalling networks, metabolic processes, autophagy, and regulation of calcium (Ca2⁺) signalling, all of which may contribute to PD pathology [7, 11, 19, 20, 43]. Furthermore, both lysosomal function and autophagy are intrinsically dependent on lysosomal acidification. Optimal lysosomal pH is also important for enzyme activity, substrate degradation and calcium efflux in the lysosomes [55]. Li et al. also suggest that autophagy in L444P mutant GBA1-PD neurons is impaired at two stages: the initial autophagy stage and the lysosomal degradation stage. While these reports indicate a relationship between GBA1 mutations and autophagic, lysosomal, and mitochondrial dysfunction, a mechanistic dissection of these multi-organellar dysfunctions is lacking. Dysregulation of the MTORC1 pathway has been reported in various neurodegenerative diseases, including PD and Alzheimer’s disease [56, 57]. A recent study by Chen et al. suggested that hyperphosphorylation of MTORC1 in striatal inhibitory neurons increases dopamine receptor D1 and reduces dopamine receptor D2 in medium spiny neurons in the ventral striatum, alongside dysregulated odour preference behaviour of mice [58]. Liu et al. demonstrated the role of MTORC1 in dopamine dynamics and synaptic plasticity [59]. These results elucidate the additional impacts of increased phosphorylated MTORC1 on PD. Kaempferol, an MTORC1 inhibitor, has also been shown to promote autophagy and protect DA neurons in the MPTP-induced C57BL/6 J PD mouse model [60]. All these studies suggest that hyperphosphorylated MTORC1 may play a central role in the development of PD associated with GBA1 mutations. Our study also indicates increased MTORC1 phosphorylation in GBA1-PD fibroblasts and iPSC-DA neurons. While the link between MTORC1 hyperphosphorylation and GBA1 mutations is unclear, it may reflect a cellular stress response to lipid substrate accumulation under GBA1 mutant conditions [61–63]. Notably, impaired lysosomal acidification is proposed to be a critical factor in the pathogenesis of multiple neurodegenerative diseases [13, 61, 64, 65]. Mutations in presenilin-1 (PS1) associated with familial Alzheimer’s disease have been demonstrated to inhibit the lysosomal acidification of fibroblasts, resulting in chronic alterations in autophagy and degradation, which may be restored through lysosomal reacidification [66]. Furthermore, it has been established that individuals with Down Syndrome, who are predisposed to early-onset dementia resembling Alzheimer’s disease, exhibit increased levels of the β-cleaved carboxy-terminal fragment of the amyloid precursor protein due to the presence of an additional chromosome 21, which impairs lysosomal acidification and functionality via inhibition of V-ATPase [17]. Collectively, these studies illuminate the connections between the impairment of lysosomal acidification in autophagic pathways and the pathogenesis of neurodegeneration, thereby presenting potential therapeutic avenues for targeting lysosomal function and pH to mitigate these debilitating diseases. Acidic NPs in the form of poly(lactic-co-glycolic acid) acidic NPs have been previously used to reduce lysosomal pH and attenuate disease progression in cells derived from GBA1-N370S PD patient fibroblasts and in PD mouse models [67]. Additionally, poly-succinate acidic NPs used in this study have been employed to restore lysosomal pH and rehabilitate metabolic and lysosomal functions in cases of non-alcoholic fatty liver disease [22]. As an MTORC1 inhibitor [68], rapamycin also acts as a prominent inducer of mitophagy and has been demonstrated to mitigate mitochondrial dysfunction in various disease models through inhibition of the MTORC1 pathway [69, 70]. Consistent with findings from other studies, we found that rapamycin rescued mitophagy and mitochondrial function in GBA1-PD fibroblasts and DA neurons. This could either be due to the role of MTORC1 as a regulator of mitochondrial function [71, 72] or as a downstream effect of lysosomal acidification by MTORC1 inhibition. To establish whether the impaired lysosomal acidification is the cause of these dysfunctions, we employed an independent pH modulator in the form of novel acidic poly-succinate NPs. These NPs rectify lysosomal pH in the cells while enhancing lysosomal and mitochondrial functions in GBA1-PD fibroblasts and DA neurons. The ratio of alkaline vs acidic lysosomes decreased significantly in all mutants and most controls after the acidic NP treatment, demonstrating a decrease in the mean lysosomal pH. The effect on controls could be due to the alkalinisation of the lysosomes with age, which can be reversed by the acidic NP treatment in healthy cells [73]. Importantly, treatment with acidic NPs successfully eliminated the differences in lysosomal pH between the GBA1 mutants and the controls. We conclude that the restoration of lysosomal acidification in the mutant cells improved lysosomal function (Fig. 6).Fig. 6Illustration of the proposed pathophysiological cascade in GBA1-related Parkinson’s disease and how the phenotype can be corrected by targeting lysosomal pH. GBA1-PD cells exhibit a range of abnormalities in lysosomal biology, including reduced GCase activity, increased lysosomal numbers, elevated lysosomal pH, and decreased pH-dependent proteolytic activity. Mitochondrial dysfunction is evidenced by a decreased ΔΨm, decreased oxygen consumption rate, swollen mitochondria, reduced cristae density, and the accumulation of abnormal mitochondria. Mitophagy is suppressed and autolysosomes accumulate within the GBA1-PD cells. Restored acidification of lysosomes with rapamycin or acidic poly-succinate NPs enhances lysosomal function and improves mitophagy, thereby rescuing mitochondrial dysfunction in GBA1-PD fibroblasts and lysosomes Illustration of the proposed pathophysiological cascade in GBA1-related Parkinson’s disease and how the phenotype can be corrected by targeting lysosomal pH. GBA1-PD cells exhibit a range of abnormalities in lysosomal biology, including reduced GCase activity, increased lysosomal numbers, elevated lysosomal pH, and decreased pH-dependent proteolytic activity. Mitochondrial dysfunction is evidenced by a decreased ΔΨm, decreased oxygen consumption rate, swollen mitochondria, reduced cristae density, and the accumulation of abnormal mitochondria. Mitophagy is suppressed and autolysosomes accumulate within the GBA1-PD cells. Restored acidification of lysosomes with rapamycin or acidic poly-succinate NPs enhances lysosomal function and improves mitophagy, thereby rescuing mitochondrial dysfunction in GBA1-PD fibroblasts and lysosomes Lysosomal acidification also rescued GCase activity in the cells despite GBA1 mutations. Other studies have also demonstrated the restoration of GCase activity despite GBA1 mutations under acidic pH conditions [74–76]. The N370S mutation results in a structurally rigid protein with reduced intra-lysosomal stability and reduced GCase activity [76], and the E326K mutation has little effect on the active catalytic site of the protein but affects the dimerisation and quaternary structure formation due to folding defects, likely leading to a mild reduction in GCase activity at higher pH levels [74]. GCase activity depends on protonation of amino acids at the active sites and is promoted by activators like saposin C [77]; the activity can be affected by GBA1 mutations [74, 78]. Restoring the acidic pH in the lysosomes helps restore functional GCase activity despite the mutations, possibly by stabilizing protein structure, improving folding and optimizing the active-site ionisation [74, 79]. To explore whether altered lysosomal pH is related to mitochondrial dysfunction in GBA1-PD, we evaluated ΔΨm, redox levels, and OXPHOS protein levels following treatment with NPs. Our results revealed a significant rescue of ΔΨm among GBA1-PD mutants after acidic NP treatment, indicating improved mitochondrial function. When compared to untreated healthy controls, the difference in ΔΨm was no longer statistically significant in the mutants following treatment with acidic NPs. It is noteworthy that some control samples also exhibited alterations due to acidic NP treatment, resulting in a small but significant increase in ΔΨm. We also found a significant increase in ΔΨm in control 2 and E326K fibroblasts upon control NP treatment. Since the NPs are composed of poly(ethylene-succinate), we suspected this could be a substrate-mediated (succinate) effect in control NP-treated conditions. However, 30-min treatment with 3 mM diethyl succinate did not alter the ΔΨm in the control or mutant fibroblasts. Acidic NP treatment also rescued the differential mitochondrial complex levels observed in the GBA1-E326K and N370S DA neurons. Rescuing lysosomal pH also improved mitophagy in the GBA1-PD fibroblasts and iPSC-DA neurons. Conclusions We have explored the relationship between lysosome biology, pH, mitophagy and mitochondria in cells carrying PD-associated mutations of GBA1. A detailed examination of the pathophysiology of these cells revealed a range of abnormalities in the lysosomes, including reduced GCase activity, increased lysosomal numbers, impaired lysosomal acidification, and impaired pH-dependent lysosomal proteolytic activity. Mitochondrial dysfunction is reflected by reduced ΔΨm, swollen, fragmented mitochondria, decreased cristae density, and accumulation of abnormal mitochondria, along with a reduced oxygen consumption rate. We also found a decrease in mitophagy and an accumulation of autolysosomes in the GBA1-PD cells. We show that increased pMTORC1 on the lysosomal membrane, possibly due to nutrient accumulation, hinders the formation of a functional V-ATPase complex in GBA1-PD cells, thereby compromising lysosomal acidification. We found that inhibiting MTORC1 using rapamycin rescues functional V-ATPase formation and lysosomal pH in the GBA1-PD neurons, along with mitophagy and mitochondrial function. Independent modulation of lysosomal pH using acidic poly-succinate NPs, restored lysosomal pH, rescued mitophagy and restored mitochondrial function in the GBA1-PD fibroblasts and iPSC-DA neurons (Fig. 6). Since rapamycin has broad effects, the comparison to NPs is valuable, adding a translational value for future applications. These studies identify the underlying mechanisms causing impaired lysosomal pH regulation associated with PD-related GBA1 mutations and provide evidence that restoring lysosomal pH is sufficient to rescue lysosomal function and mitophagy and therefore to rescue mitochondrial function in both GBA1-PD patient-derived fibroblasts and iPSC-DA neurons. We highlight that lysosomal pH correction is a pivotal, unifying therapeutic target in GBA1-PD. Supplementary Information Additional file 1. Table S1. List of cell lines used. Table S2. List of antibodies used in the study. Figure S1. Lysosomal abnormalities in GBA1-PD DA neurons and fibroblasts. Figure S2. Mitochondrial dysfunction in GBA1-PD DA neurons and fibroblasts. Figure S3. Mitophagy defects in GBA1-PD DA neurons and fibroblast. Figure S4. Effect of acidic NPs and rapamycin on GBA1-PD fibroblasts and DA neurons.Additional file 2. Statistical tests, corresponding effect sizes, and confidence intervals for each graph.Additional file 3. Raw data.Additional file 4. Original blots. Additional file 1. Table S1. List of cell lines used. Table S2. List of antibodies used in the study. Figure S1. Lysosomal abnormalities in GBA1-PD DA neurons and fibroblasts. Figure S2. Mitochondrial dysfunction in GBA1-PD DA neurons and fibroblasts. Figure S3. Mitophagy defects in GBA1-PD DA neurons and fibroblast. Figure S4. Effect of acidic NPs and rapamycin on GBA1-PD fibroblasts and DA neurons. Additional file 2. Statistical tests, corresponding effect sizes, and confidence intervals for each graph. Additional file 3. Raw data. Additional file 4. Original blots.