Introduction

Bi-allelic mutations in the GBA1 gene, which encodes the lysosomal enzyme glucocerebrosidase (GCase), are the primary cause of the lysosomal storage disorder Gaucher disease (GD). Moreover, heterozygous GBA1 mutations represent a significant genetic risk factor for Parkinson’s disease (PD) (Migdalska-Richards & Schapira, 2016). GCase plays a crucial role in lysosomal homeostasis by facilitating the breakdown of glucosylceramide (GluCer) into ceramide and glucose (Boer et al., 2020). Loss of normal GCase function in GD leads to the accumulation of lipid-laden Gaucher cells in multiple organs, primarily in the liver, spleen, and bone marrow. These cells may also accumulate in other tissues, including the lymphatic system, lungs, skin, eyes, kidneys, heart, and, in rare cases, the nervous system (Cox, 2010). While the neurological and haematological manifestations of GD are well-documented, the impact of GBA1 mutations on renal function remains an underexplored but emerging area of interest. Furthermore, the potential involvement of the renal system in PD is also becoming increasingly recognised (Meléndez-Flores & Estrada-Bellman, 2021).

Gaucher macrophages have been identified within endothelial and interstitial regions of the kidney, where they may contribute to severe proteinuria. However, renal dysfunction remains a rare symptom in GD patients (Smith et al., 1978; Chander et al., 1979; Siegel et al., 1981; Halevi et al., 1993). GluCer accumulation has also been observed in the glomeruli of patients with diabetic nephropathy and polycystic kidney disease (PKD), as well as in rodent models (Shayman et al., 1991; Zador et al., 1993; Deshmukh et al., 1994; Natoli et al., 2010). Furthermore, treatment with GluCer synthase inhibitors has been shown to ameliorate glomerular degeneration in PKD, and clinical trials are underway, highlighting the critical role of GluCer regulation in maintaining renal health (Natoli et al., 2010; Shayman et al., 2016).

Loss of function mutations in LIMP2 (also known as SCARB2), a sorting receptor that specifically directs GCase to the lysosome, have been identified in multiple families as the cause of progressive action myoclonus and renal failure (AMRF) (Berkovic et al., 2008; Balreira et al., 2008; Chaves et al., 2011; Hopfner et al., 2011). In addition to myoclonus, AMRF is marked by proteinuria, weight gain, and fluid retention leading to oedema. The mechanism underlying LIMP2-associated AMRF is likely distinct from GD, as GluCer-filled macrophages are not observed in tissue samples from AMRF patients. Although not directly implicated in renal pathology, AMRF patients and Limp2 knockout mice show a several fold increase in circulating levels of glucosylsphingosine, a water soluble deacylated form of GluCer that can be excreted in urine. This increase is also seen in GD (Dekker et al., 2011; Gaspar et al., 2014). Furthermore, renal failure is a major complication in Fabry disease, another lysosomal storage disorder characterised by the accumulation of the complex glycosphingolipid globotriaosylceramide (GB3) (Nakao et al., 2003).

An emerging link between PD and kidney dysfunction is becoming increasingly evident. Large cohort studies involving patients with nephrotic syndrome, end-stage renal disease (ESRD), and chronic kidney disease (CKD) have consistently reported a significantly higher incidence of PD or parkinsonism (Huang et al., 2018; Wang et al., 2014; Melendez-Flores, 2021). However, it remains unclear whether renal and neurological diseases directly influence one another or whether they share a common upstream pathological mechanism. Evidence from large datasets, including South Korean populations and UK Biobank samples, suggests that a reduced glomerular filtration rate (GFR) and elevated proteinuria are associated with an increased risk of developing PD (Nam et al., 2019; Peng et al., 2024). In individuals with PD, an inverse correlation has been observed between GFR and serum levels of neurofilament light chain (NfL), a biomarker of neuronal damage, as well as between GFR and cognitive decline (Qu et al., 2023). These findings imply that kidney dysfunction is closely linked to the severity of PD. Notably, both PD and various forms of kidney disease are characterised by elevated oxidative stress, suggesting that shared pathophysiological mechanisms may underlie this association (Pizzino et al., 2017).

Fruit flies are increasingly used as models for renal disorders. First, they serve as a kidney stone model, as xanthic/uric acid crystals can form in the Drosophila Malpighian tubules (MTs) (Dow & Romero, 2010). Second, they are used to study podocyte biology, as the slit diaphragm structure of mammalian podocytes is conserved in fly nephrocytes, a large class of cells adjacent to the heart tube that filter the haemolymph through fluid-phase endocytosis (Weavers et al., 2009). Water retention can result from the loss of filtration efficiency in the MTs, caused by the breakdown of occluding junctions between epithelial cells or impaired renewal of differentiated renal epithelial cells (Martinez-Corrales et al., 2016; Denholm et al., 2010; Dornan et al., 2023).

Using our previously characterized GD Drosophila model lacking Gba1b, the primary fly orthologue of GBA1, we demonstrate progressive degeneration of the renal system, loss of organismal water homeostasis and ionic hypersensitivity. Specifically, Gba1b^-/- flies exhibit reduced glomerular filtration capacity in pericardial nephrocytes, along with widespread cellular degeneration in the MTs. Furthermore, renal system dysfunction negatively impacts both organismal and neuronal health in Gba1b^-/-flies, including autophagic-lysosomal status in the brain. Additionally, we show that the loss of Parkin, a familial PD-linked E3 ubiquitin ligase, similarly leads to renal defects. Notably, renal dysfunction in GD flies can be alleviated through rapamycin administration, highlighting a potential therapeutic avenue for lysosomal and neurodegenerative diseases linked to GBA1 dysfunction.

Materials and methods

Fly stocks and rearing

Flies were reared at 25°C and 50% relative humidity in a 12:12LD cycle on SYA Food (1.5% agar, 5% sucrose, 10% dried yeast [Leiber Bierhefe], 0.3% nipagin, 0.3% propionic acid). Experimental flies were raised at a defined density, mated for 1 day and females were subsequently maintained at a density of 15 flies per vials and changed to fresh food every 2-3 days. The following fly stocks were acquired from the Bloomington Drosophila Stock Center: Gba1b^CRIMIC (BL#78943), park¹ (BL#34747), C724-Gal4 (BL#99567), c42-Gal4 (BL#30825), tub-Gal4 (BL#5138), UAS-CD8RFP (BL#27392), UAS-mcherry.nls (BL#38424), UAS-mitoGFP (BL#8442), UAS-yfpPTS (BL#64247), klf15 (BL#18979), UAS-Sod1 (BL#24754), UAS-Sod2 (BL#24494), UAS-CatA (BL#24621), UAS-mitoroGFP2.grx1 (BL#67664), UAS-cytoroGFP2.grx1 (BL#67662), UAS-Park (BL#51651). Gba1b^-/- flies were previously constructed by homologous recombination (Kinghorn et al., 2016). UAS-Nrf2 and UAS-Keap1 were gifts from Dirk Bohmann (Sykiotis & Bohmann, 2008) and UAS-mitoQC was a gift from Alex Whitworth (University of Cambridge; Lee et al., 2016).

Water weight assays

Flies were frozen in liquid nitrogen and immediately weighed on an AT-201 Mettler Toledo Analytical Balance. Flies were then dried out for 24 hrs in an 80°C oven and re-weighed. The total water volume was calculated as the weight lost from drying.

Hemolymph extraction

Groups of 15 flies were beheaded, placed upside-down in a truncated pipette tip in a 1.5ml Eppendorf and centrifuged for 10 minutes at 1500 rpm at 4°C. The haemolymph was then flash-frozen in liquid nitrogen.

Drinking assays

Assay 1: Groups of 15 female flies were placed in vials containing dried fly food and a 1.5 mL Eppendorf tube filled with water. The Eppendorf tubes were weighed before and after a 24-hr period. The difference in weight, adjusted for evaporation using a control tube lacking a fly, was calculated and normalised per fly. Assay 2: Cohorts of flies were starved for 5 hours before being provided with a water source for 1 hour. To estimate water consumption, the mean weight of water in the starved cohort was subtracted from the weight of flies given access to water.

Cold-stress assays

>75 flies of each genotype were placed in empty glass tubes and uniformly submerged in an ice bucket for 10 minutes. To assess chill-coma recovery, flies were filmed and considered recovered once they stood upright and began walking. Survival was scored periodically over a 3-day period.

Lifespan

Lifespan assays were conducted as previously described (Atilano et al., 2023). Eggs were collected over a 24-hour period on grape juice agar plates and then pipetted onto SYA food at a uniform density. Female flies were split into groups of 15 in vials, transferred to fresh food every 2–3 days, and monitored for survival.

Ionic stress and natriuresis assays

Experimental flies were raised on SYA food as described above for three weeks before being transferred to food containing 4% NaCl or KCl, with or without a water source provided by 1% agar in a pipette tip (Van Dam et al., 2020). For dehydration assays, flies were raised without a water source on active dry yeast (Saf-Levure).

FRUMS (Fluorescent Renal Uptake and Metabolite Secretion) assay

A volume of 23 nL of 2.5% (w/v) Brilliant Blue FCF-E133 (Florio Colori) in ddH₂O was injected into the haemolymph using a Drummond Nanoject II Microinjector. Blue flies were then scored manually every hour for qualitative clearance of dye from the hemolymph and subsequently from the MT and gut.

Tissue preparation and immunostaining

MTs and brains were dissected in ice-cold PBS and immediately fixed in 4% PFA for 20 minutes. The tissues were then washed three times for 20 minutes each in PBS-T (0.5% Triton-X), followed by a 30-minute incubation in blocking solution (5% Normal Goat Serum in PBS-T). Primary antibody incubation was performed overnight at 4°C with gentle nutation. Tissues were then washed three times for 20 minutes in PBS-T and incubated with secondary antibodies in blocking solution for 2 hours at room temperature. This was followed by three additional 20-minute washes in PBS-T before mounting in Vectashield Mounting Medium with DAPI (Vector Laboratories). Primary antibodies used were mouse anti-cora C615.15 (DSHB,1:200), rabbit anti-Ref(2)P (ab178440, 1:200), mouse anti-FK2 (Sigma 04263, 1: 800). The secondary antibodies used were goat anti-mouse Alexa488 (A11001) and goat-rabbit Alexa568 (A11036), both at a dilution of 1:250. For BODIPY experiments, 20 µM BODIPY^TM 505/515 was incubated 1:100 in addition to the primary antibody.

Tissues with genetically encoded fluorescent markers (YFP-PTS, mito-Grx1-roGFP2 and cyto-Grx1-roGFP2) were dissected in 20 mM NEM in PBS, fixed in 4% PFA for 20 minutes, washed for 20 minutes in PBS-T, and mounted in PBS.

Dextran uptake assay

Fly abdomens were dissected from 28-day-old flies and incubated in 0.33 mg/mL Texas Red 10,000 MW dextran (Thermo Fisher) in PBS for 5 minutes. The abdomens were then washed twice in PBS, fixed in 4% PFA for 20 minutes, and washed for 20 minutes in PBS-T. Finally, the samples were mounted in VECTASHIELD mounting medium with DAPI and imaged on a Zeiss 700 confocal microscope.

Live imaging

Tissues were dissected at room temperature and transferred to 50 nM Lysotracker^TM Red DND-99 (Thermo Fisher), Mitotracker^TM Red CMH₂XROS, or DHE (Thermo Fisher) in PBS, followed by immediate imaging at 20x magnification using a Zeiss 700 or 880 confocal microscope. For mitoQC experiments, MTs were dissected in cold PBS and immediately imaged. For lipid peroxidation analysis, MTs were dissected in PBS and incubated in 20 µM BODIPY^TM 581/591 in Schneider’s medium for 30 minutes. After two PBS washes, tissues were mounted in PBS and immediately imaged

Reduced glutathione (GSH) assay

GSH levels were measured using a Glutathione Assay Kit (Cayman Chemicals) according to the manufacturer’s instructions. Briefly, groups of 10 flies were homogenised with glass beads (Sigma) in MES buffer using a Precellys Ribolyser. The homogenate was then centrifuged at 10000 g for 10 minutes and the resulting supernatant was transferred to a new tube. Samples were deproteinated in 50 mg/ml metaphosphoric acid and basified with 1/20 4M triethanolamine. A 50 µl aliquot of each sample was loaded into Greiner transparent flat-bottom 96-well plates (Cellstar) containing 150 µl enzyme buffer. Absorbance was measured at 405 nm at 5-minute intervals for 30 minutes using a Tecan Infinite M200 plate reader. GSH values were calculated from the slope of a standard curve generated using known Glutathione Persulfide (GSSG) concentrations.

NAD/NADH assay

NAD/NADH quantification was performed using a NAD+/NADH Quantification Colorimetric Kit (Biovision) according to the manufacturer’s instructions. Groups of 10 flies were flash-frozen in liquid nitrogen and homogenized in 300 µL of extraction buffer. Samples were deproteinated by centrifugation at 10000 rpm for 10 minutes at 4°C using Amicron Ultra 0.5 mL 10,000 MW filters. The samples were then divided into NAD and NADH groups, where NAD was decomposed by incubation at 60°C for 1 hour. Following this, samples were incubated in NAD cycling buffer and developer, and absorbance readings were taken at 450 nm using a Tecan Infinite M200 plate reader.

Statistical analyses

Data was analysed in Graphpad Prism 8.0.2 and the statistical tests used are described in the figure legends. p<0.05 was considered statistically significant in all cases. Log-rank test on lifespan data were performed in Microsoft Excel (available at http://piperlab.org/resources/).

Results

Flies lacking Gba1b display age-dependent MT degeneration

Since both sphingolipid accumulation and PD are linked to renal disease, we aimed to assess the renal system in a previously characterised GD Drosophila model, lacking the main fly orthologue of GBA1, known as Gba1b (Kinghorn et al., 2016). Fly MTs consist of a tubulated epithelial sheet composed of principal cells (PCs) and stellate cells (SCs) surrounding a lumen. To investigate structural integrity, we stained the MTs with coracle (Cora), a core component of the septate junction (SJ), a specialised cell-cell junction which is essential for barrier function and transcellular water flow (Dornan et al., 2023). In control fly MTs, Cora marked clearly defined cell vertices throughout the tubule. However, in Gba1b^-/-fly MTs, the number of well-defined SJ vertices was significantly reduced, and Cora was frequently mislocalised throughout the cytoplasm (Figure 1A). Notably, these structural abnormalities were absent in day 1 Gba1b^-/- fly MTs, indicating that the loss of tissue architecture is progressive.

Figure 1.

Since lipid aggregation is a hallmark of metabolic disorders, we stained MTs with the lipid marker BODIPY, which revealed a significant increase in luminal lipids in Gba1b^-/- flies compared to controls (Figure 1B). Moreover, counter-staining with the nuclear marker DAPI identified a significant increase in small nuclei proliferating from the ureter, presumptively corresponding to renal stem cells (RSCs), extending into the absorptive zone of the MT in Gba1b^-/- flies compared to age-matched control MTs. RSCs are known to proliferate in response to stress and inflammation (Singh et al., 2007), resulting in decreased water flux (Martinez-Corrales et al., 2019).

Transcellular water flux principally occurs through SCs, and given the observed loss of normal tissue architecture, we quantified SCs in Gba1b^-/- fly MTs through expression of the membrane-localised CD8:RFP using the SC-specific driver C724-Gal4. We observed a significant reduction in SC numbers in Gba1b^-/- fly MTs compared to those seen in control flies, alongside a loss of their characteristic star or bar morphology (Figure 1C).

The gut and brain of Gba1b^-/- flies, similar to macrophages in GD patients, are characterised by enlarged lysosomes (Kinghorn et al., 2016, Atilano et al 2023). Consistent with this, staining of MTs with LysoTracker^TM, an acidic vesicle dye, revealed a significant increase in fluorescence intensity in Gba1b^-/- fly MTs, consistent with pronounced lysosomal expansion in a subset of renal cells (Figure 1D).

As MTs play a critical role in redox balance, with mitochondrial activity in PCs being pivotal to physiological function, we examined mitochondrial morphology using mito-GFP expression. In control PCs, mitochondria appeared fused and tubulated, whereas in Gba1b^-/- fly MTs, they were predominantly punctate, indicative of defective mitochondrial fusion-fission homeostasis (Figure 1E). We next stained with MitoTracker^TM CM-H2XRos, a mitochondrial-specific fluorescent probe used to measure reactive oxygen species (ROS). This demonstrated reduced staining intensity in the MTs of Gba1b^-/- flies. This is likely consistent with a decrease in mitochondrial ROS production, due to impaired mitochondrial membrane potential and dysfunction of the electron transport chain (ETC) (Figure 1F). In keeping with this, NADH, but not NAD+, was decreased in whole Gba1b^-/- flies, suggesting impaired mitochondrial energy metabolism (Figure S1A).

Finally, to assess MT physiological function, we employed the FRUMs assay, in which blue food dye is injected into the hemolymph, and its clearance into the tubules and subsequent excretion serve as a proxy for water homeostasis. Gba1b^-/- flies exhibited increased retention of blue dye in both the hemolymph and subsequently the tubules, indicating impaired MT-mediated hemolymph clearance (Figure 1G).

Ion and water homeostasis is dysregulated in Gba1b^-/- flies

To investigate the impact of these defects in MT morphology and function on water homeostasis, we measured water weight in aged Gba1b^-/-flies. Our analysis revealed an increase in both total water weight and percentage body water weight (Figure 2A), partially attributable to an elevated hemolymph volume in Gba1b^-/- flies compared to controls (Figure 2B). Notably, this increase was not due to heightened water intake, as Gba1b^-/- flies exhibited either reduced or unchanged drinking behaviour across multiple assays (Figure S2A-B).

Figure 2.

To determine the physiological consequences of this water retention, we subjected Gba1b^-/- flies to a desiccative stress assay and observed that they exhibited a small but significant dehydration resistance compared to controls (Figure 2C). Gba1b^-/- flies also retained a higher level of water throughout the course of the dehydration assay (Figure S2C). As MTs play a crucial role in regulating ion homeostasis, we subjected Gba1b^-/-flies to ionic stress assays using food supplemented with 4% NaCl or KCl. We observed that Gba1b mutants were highly sensitive to both ionic stress conditions (Figure 2D, E). To determine whether the lethality observed in these assays was due to desiccative stress from water loss accompanying excess salt efflux through the MTs, or direct cellular ionic stress, we hypothesised that providing a water source should enable flies with functional MTs to maintain homeostasis under ionic stress conditions. To test this, we measured both water levels and lifespan under ionic stress while supplementing a water source via 1% agar-filled pipette tip (Figure 2F, S2D). Water supplementation led to a substantial lifespan extension in control flies exposed to 4% NaCl, with a smaller yet significant extension in Gba1b^-/-flies (Figure 2F). NaCl exposure led to a reduction in water weight, which was rescued by water supplementation in control but not Gba1b^-/- flies (Figure S2D). These observations suggest that the early mortality in control flies under ionic stress primarily results from dehydration rather than direct ionic toxicity. On the other hand, although dehydration contributes to Gba1b^-/- mortality, a significant component of their susceptibility to ionic stress is likely due to defective natriuresis, leading to impaired ion homeostasis.

In Drosophila, cold tolerance is closely linked to altered ion homeostasis. To investigate this, we subjected aged control and Gba1b^-/-flies to a chill-coma stress assay and found that flies lacking Gba1b exhibited a remarkable resistance to chill-induced damage, whereas control flies experienced rapid mortality within 24 hours (Figure 2G).

Hemolymph clearance and nephrocyte function is diminished in Gba1b^-/-flies

Podocyte-like nephrocytes filter small macromolecules through a size-excluding slit-diaphragm system, after which they are then taken up non-selectively through fluid-phase endocytosis and degraded in the lysosome (Weavers et al., 2019). To assess endocytic function, we incubated dissected pericardial nephrocytes with Texas-red dextran to visualise early endosomes. At 4 weeks of age, Gba1b^-/- flies exhibited a population of endosome-negative nephrocytes, indicating a loss of endosomal activity, and consequently, hemolymph filtration (Figure 3A). Consistent with reduced filtration, protein concentration analysis using a BCA assay on extracted hemolymph revealed a significant increase in circulating soluble protein in Gba1b^-/- flies relative to controls (Figure 3B). These findings suggest that defects in both MT function and glomerular filtration contribute to increased circulating waste, potentially leading to secondary toxic consequences.

Figure 3.

To determine the contribution of nephrocyte dysfunction to disease pathology in Gba1b loss of function, we impaired nephrocyte maturation using a Klf15 LOF mutant, which is exclusively expressed in nephrocytes and required for their differentiation (Ivy et al., 2018). Loss of Klf15 significantly shortened the lifespan of both Gba1b^-/-and control flies, demonstrating that nephrocyte dysfunction is deleterious to both healthy flies and those lacking GCase activity (Figure 3C). To determine the consequence of abnormal nephrocyte function on brain health, we measured lysosomal area in the brain with LysoTracker^TM, or stained for polyubiquitin. Klf15 LOF had no significant effect on lysosomal or polyubiquitin area relative to healthy controls, nor an additive effect on the increased lysosomal volume (Figure 3D) and polyubiquitin accumulation (Figure 3E) seen in Gba1b^-/- fly brains. This suggests that loss of nephrocyte viability itself does not exacerbate brain pathology.

Gba1b^-/- MTs display loss of redox homeostasis capacity

The renal system is key to organismal redox homeostasis, a process thought to play a key role in the pathogenic mechanisms of GD and PD progression (Kishi et al., 2024). Glutathione (GSH), the most abundant antioxidant in Drosophila, plays a crucial role in redox homeostasis by reducing hydrogen peroxide (H₂O₂) and lipid hydroperoxides (ROOH) via GSH peroxidases. To evaluate redox balance, we measured total GSH levels in whole flies from both control and Gba1b^-/- groups. Our results revealed a significant decrease in GSH in Gba1b^-/-flies, indicating a compromised ability to maintain redox equilibrium (Figure 4A). To further evaluate the glutathione redox potential (E_GSH) in MTs, we utilised the redox-sensitive green, fluorescent biosensor Grx1-roGFP2, targeted to both the mitochondria and cytosol (Albrecht et al., 2011). Our findings revealed a shift toward a more reduced glutathione state in the mitochondria of Gba1b^-/-fly tubules compared to controls (Figure 4B). Interestingly, while cytosolic glutathione typically maintains a less oxidised GSH/GSSG ratio than mitochondria, it was paradoxically more oxidised in Gba1b^-/- tubules relative to controls (Figure 4C). This indicates that despite a potential reduction in mitochondrial ROS production, the tubules fail to effectively regulate basal ROS levels. These abnormalities may contribute to the impaired physiological function of the MTs.

Figure 4.

We then incubated MTs with the nuclear-localised superoxide indicator dihydroethidium (DHE), revealing a significant decrease in superoxide levels in Gba1b^-/- tubules (Figure 4D). Since glutathione primarily detoxifies H₂O₂ and ROOH, rather than superoxide, this reduction could indicate lower mitochondrial ROS production or altered superoxide metabolism (Winterbourn, 2016).

To investigate the role of lipids in redox dyshomeostasis, we measured lipid peroxidation in MTs using the polyunsaturated lipid peroxidation marker BODIPY^TM 581/591 C11. Gba1b^-/- fly tubules exhibited higher lipid peroxidation ratios than controls, suggesting increased oxidative damage and potential deficiencies in antioxidant defences or lipid turnover mechanisms (Figure 4E). We conclude that Gba1b^-/- flies, particularly their MTs, have a diminished capacity to regulate redox balance.

To determine the endogenous antioxidant defence in the tubules, we profiled peroxisomal morphology using the marker YFP-PTS. We found a significant decrease in peroxisome density and area in Gba1b^-/- fly tubules, potentially indicating defects in peroxisome maturation and function (Figure S3A). We further investigated peroxisomal and cytosolic antioxidant defences by immunostaining for superoxide dismutase 1 (Sod1) and catalase (CatA). Sod1 showed a significant decrease in staining intensity and a transition from a compartmentalised to diffuse localisation in Gba1b^-/- fly tubules (Figure S3B). Taken together, these findings indicate that the MTs of Gba1b^-/- flies exhibit redox imbalance characterised by mitochondrial, peroxisomal, and lipid oxidative stress. Moreover, the inability to effectively regulate antioxidant responses may have broader implications for physiological functions across multiple tissues in Gba1b deficiency.

Induction of antioxidant effector Nrf2 worsens renal and neurological phenotypes in Gba1b^-/- flies

To investigate the contribution of redox dyshomeostasis to Gba1b^-/-fly disease phenotypes, we overexpressed the antioxidant regulator Nrf2/CnC in PCs, a manipulation previously shown to prematurely impair renal function (Burbridge et al., 2021). Consistent with this, whilst Nrf2 overexpression resulted in a modest but statistically significant lifespan reduction in control flies, its overexpression in Gba1b^-/- flies led to a dramatic lifespan shortening (Figure 5A). This suggests that Gba1b^-/- flies have an increased susceptibility to antioxidant-mediated tubular dysfunction. In addition, we observed that Nrf2 overexpression in Gba1b^-/- flies resulted in an increase in renal stem cell proliferation at the ureter (Figure 5B). However, despite these effects, there was a significant increase in water weight in controls overexpressing Nrf2, but not in Gba1b^-/- flies (Figure 5C). Collectively, these findings indicate that Gba1b^-/- flies exhibit disproportionate negative renal and systemic responses to activation of the antioxidant response. We next tested the consequence of Nrf2 overexpression in PCs on neuronal health in Gba1b loss of function. This intervention led to a significant increase in lysosomal puncta number, as assessed by LysoTracker^TM staining (Figure 5D), and exacerbated protein dyshomeostasis, as indicated by polyubiquitin accumulation and increased levels of the ubiquitin-autophagosome trafficker Ref(2)p/p62 in Gba1b^-/- fly brains (Figure 5E). Interestingly, Nrf2 overexpression had no significant effect on lysosomal area or ubiquitin puncta in control brains, demonstrating that the antioxidant response specifically in Gba1b^-/- flies negatively impacts disease states in the brain and renal system.

Figure 5.

Gba1b^-/-lifespan is shortened by shifting antioxidant state at an organismal level and in renal tubules

To assess the ability of Gba1b^-/- flies to respond to exogenous antioxidants, we subjected them to a diet supplemented with ascorbic acid, a vitamin C analogue previously shown to extend lifespan in healthy flies (Bahadorani et al., 2008). While ascorbic acid led to a mild, non-significant increase in lifespan in control flies, it resulted in a significant reduction in lifespan in Gba1b^-/- flies, indicative of an increased sensitivity to antioxidant supplementation (Figure 6A).

Figure 6.

Given that Nrf2 overexpression shortens lifespan in Gba1b^-/- flies, we investigated the effects of overexpressing its downstream antioxidant targets, Sod1, Sod2, and CatA, both ubiquitously using the tub-Gal4 driver and with c42-Gal4, which expresses in PCs. While antioxidant gene overexpression provided variable benefits in healthy control flies, ubiquitous or tubule overexpression was detrimental in Gba1b^-/- flies, shortening the lifespan (Figure 6B-F), suggesting that MT and organismal responses to oxidative stress may advance disease phenotypes and contribute to systemic redox imbalance.

Furthermore, we examined the effects of PGC1α overexpression, a key regulator of mitochondrial biogenesis and an upstream activator of Nrf2. Ubiquitous overexpression of PGC1α extended lifespan in healthy controls but did not significantly impact the lifespan of Gba1b^-/- flies (Figure S3C). Expression of either PGC1α or NDI, a yeast NADH dehydrogenase that bypasses mitochondrial complex I (Sanz et al., 2010), using c42-Gal4 led to a reduction in lifespan in Gba1b^-/- flies, suggesting that mitochondrial or ETC defects in tubules may not be the key driver of redox-stress in Gba1b^-/-(Figure S3C, D).

Given that antioxidant activation is detrimental in Gba1b^-/-flies, we investigated whether repressing the antioxidant response could have a protective effect. To this end, we expressed the Nrf2 repressor Keap1 ubiquitously and specifically in the MTs of Gba1b^-/- flies. However, Keap1 overexpression, whether ubiquitous or using the PC driver c42-Gal4, significantly reduced lifespan in both Gba1b^-/- flies and controls (Figure 6G, H). Taken together, our results demonstrate that flies lacking Gba1b experience redox dyshomeostasis and are highly susceptible to both oxidative and reductive stress.

Parkin mutants phenocopy the renal dysfunction observed in Gba1b^-/- flies

As the kidney-PD axis remains underexplored in animal models of disease, we investigated whether renal pathology is also evident in other PD models. While PD is primarily idiopathic, rare familial forms have been identified. PARK2, an E3 ubiquitin ligase, plays a crucial role in mitophagy initiation, and autosomal recessive mutations in the fly Parkin gene, park, lead to early-onset PD through mitochondrial dysfunction and subsequent redox stress (Whitworth & Pallanck, 2017). We found that loss-of-function (LOF) park mutants (park^1/1) exhibited increased water weight and body water percentage (Figure 7A), which correlated with degenerative MT pathology. These mutants displayed elevated RSC proliferation (Figure 7B), increased tubule diameter, reduced mitochondrial membrane potential (Figure 7C), and a significant shift toward a more reduced mitochondrial redox state (Figure 7D), phenotypes that closely resemble those observed in Gba1b^−/− fly mutants.

Figure 7.

Mitophagy occurs in a spatially segregated manner in the mammalian kidney, where intense mitolysosome staining is visible in the proximal convoluted tubule, the site of high active transport demands and, consequently, elevated mitochondrial activity and turnover (Mcwilliams et al., 2016). To determine if mitophagy defects were present in park^1/1 mutant fly MTs, we expressed the mitophagy reporter mitoQC in PCs. This revealed a significant reduction in acidified mitolysosomes, indicating a mitophagy defect in park^1/1 tubules (Figure 7E).

Consistent with reduced MT function, park1/1 nephrocytes showed a more severe phenotype than Gba1b^-/-flies, with a complete loss of endocytically active nephrocytes by 28 days, as assessed by a dextran uptake assay (Figure 7F). In addition, park^1/1mutants exhibited increased resistance to cold stress (Figure 7G), along with a significant reduction in renal clearance of haemolymph, as assessed by the FRUMS assay (Figure 7H).

To determine whether the effects of park loss were tissue-autonomous, we overexpressed park in park^1/1 mutants using the PC-driver c42-Gal4. This was sufficient to rescue water content, indicating that mitochondrial clearance and function are critical for maintaining tubule integrity (Figure 7I). In contrast, brain-restricted Parkin expression had no impact on tubule health (Figure S4A). These findings support a model in which mitophagy defects due to loss of park drive mitochondrial dysfunction, leading to physiological impairment and eventual tubular degeneration.

Rapamycin alleviates disease phenotypes in the MTs of Gba1b^-/- flies

To identify therapeutics with potential relevance to renal defects in our disease models, we first treated Gba1b^-/- flies with rapamycin, which we previously showed extends lifespan in association with autophagy induction (Kinghorn et al 2016; Atilano et al., 2023). Rapamycin treatment significantly reduced water retention (Figure 8A), suppressed RSC over proliferation (Figure 8B) and increased FRUMS dye clearance from the body (Figure S5C, D), indicating that mTOR inhibition confers renal benefits in Gba1b^-/- flies. Additionally, rapamycin normalised lipid peroxidation in Gba1b^-/-MTs to control levels (Figure 8C).

Figure 8.

Using tubule-specific expression of mitoQC, we found that Gba1b^-/-fly MTs exhibited a reduced ratio of non-acidified to acidified mitochondria compared to control flies, suggesting that unlike park^1/1 tubules, mitophagy remains functional and potentially overactive. Moreover, the mitoQC ratio in Gba1b^-/- fly MTs was not affected by rapamycin treatment (Figure 8D). The decreased mitochondrial membrane potential of Gba1b^-/- MTs was not rescued by rapamycin treatment (Figure 8E). This indicates that some aspects of mitochondrial dysfunction in Gba1b^-/- flies are independent of mTOR-regulated pathways.

In contrast to the MTs of Gba1b^-/- flies, park^1/1fly tubules displayed decreased lysosomal content (Figure 8F), consistent with a block in mitophagy. Rapamycin failed to rescue the elevated water weight (Figure 8G), or mitochondrial redox state (Figure 8H) in park^1/1 flies, indicating that renal dysfunction in this model may stem from mitophagy defects that are unresponsive to mTOR inhibition. However, rapamycin did rescue both ureter-associated RSC over proliferation and lipid peroxidation in park^1/1 fly tubules (Figures 8I, J). Together, these findings suggest that although both Gba1b^-/- and park^1/1 mutant flies exhibit pronounced renal defects, the underlying mechanisms differ. park^1/1 fly mutants display a mitophagy block and renal dysfunction that cannot be reversed by rapamycin, whereas the renal pathology in Gba1b^-/- flies is influenced by rapamycin, potentially through altered systemic lipid metabolism.

Discussion

Despite growing epidemiological evidence linking renal disease to neurodegenerative disorders (Meléndez-Flores & Estrada-Bellman, 2021), the mechanistic and directional relationships between these conditions remain unresolved. Here, we identify progressive renal dysfunction as a previously underappreciated feature in Drosophila models of GD and PD. Flies lacking either the main GBA1 orthologue, Gba1b, or the mitophagy regulator Parkin, park, exhibit age-dependent degeneration of the renal system. This is characterised by disorganised MTs, impaired nephrocyte function, redox imbalance, and lipid accumulation. These renal impairments contribute to organismal decline, including fluid imbalance, ionic hypersensitivity, and worsening neurodegeneration.

A key finding is that Gba1b-deficient flies exhibit a collapse of redox homeostasis that does not align with classical oxidative stress. Redox-sensitive probes revealed lipid peroxidation and cytosolic oxidation alongside modest mitochondrial reduction in Gba1b⁻^/⁻ MTs. This divergence was further supported by opposite responses to redox-targeted interventions. In Gba1b^⁻/⁻ flies, overexpression of Nrf2 or antioxidant enzymes (SOD1, SOD2, CatA), and supplementation with ascorbic acid, all led to worsened survival. These results suggest a critical loss of redox buffering capacity, where both oxidative and reductive shifts are deleterious. This redox fragility sets Gba1b-related pathology apart from conventional oxidative stress models (Dias et al., 2014) and echoes recent findings that both oxidative and reductive stress can impair renal function in Drosophila (Burbridge et al., 2021). These observations underscore the potential risks of unselective antioxidant therapies in GBA1-related diseases and suggest that redox tone may serve as a biomarker of tissue vulnerability and therapeutic responsiveness.

In contrast to Gba1b^-/- flies, those lacking park benefit from antioxidant pathway activation (Whitworth et al., 2005), consistent with oxidative stress as a key driver of pathology. This model-specific disparity underscores the need for tailored therapeutic approaches. Clinically, mitophagy has been shown to protect against renal injury and nephrotoxicity (Tang et al., 2018; Wang et al., 2018), reinforcing parkin’s role in renal health.

Notably, the source of renal damage differs between models. In park^1/1 flies, renal defects are tissue-autonomous: tubule-specific re-expression of Parkin restores water regulation. Conversely, Gba1b is not expressed in the renal system, pointing to a non-cell-autonomous mechanism of dysfunction. This suggests that systemic factors, such as circulating glycosphingolipids or a loss of extracellular vesicle-mediated delivery of GCase, may mediate renal toxicity (Thomas et al., 2018). Supporting this hypothesis, sphingolipid accumulation is a hallmark of GD, and lipid dysregulation has been directly implicated in renal pathology. For example, mutations in the ceramide synthase gene CERS2 cause renal dysfunction in mice (Imgrund et al., 2009). Moreover, ceramide buildup induces podocyte apoptosis (Yoo et al., 2015) and Drosophila sply mutants, deficient in sphingosine-1-phosphate lyase, also display nephrotic-like phenotypes with slit diaphragm defects (Lovric et al., 2017). It should be noted, however, that these are developmental and static in nature, unlike the progressive degeneration in Gba1b⁻^/⁻ flies.

Mitochondrial analysis further revealed that Gba1b deficiency impairs mitochondrial morphology, reduces membrane potential, and limits NADH availability, all hallmarks of dysfunction seen in Gba1-deficient mammalian neurons (Osellame et al., 2013). However, mitophagy appears preserved in Gba1b⁻^/⁻ MTs, in contrast to park^1/1 flies. A reduction in intact mitochondria in both models suggests that excessive degradation may contribute to renal stress, albeit through divergent upstream mechanisms.

Strikingly, rapamycin, a known mTOR inhibitor and autophagy inducer (Bjedov et al., 2010), rescued renal structure and function in Gba1b^⁻/⁻ flies, including water retention, RSC proliferation, and lipid peroxidation. This aligns with findings that mTOR overactivation contributes to podocyte damage in glomerular disease (Godel et al., 2011). These effects occurr despite unaltered mitophagy, pointing toward alternative mechanisms such as improved lipid handling or redox buffering. In contrast, rapamycin had limited effects on renal function in park¹/¹ flies, highlighting mechanistic divergence between lysosomal and mitochondrial models. In GD, where lysosomal clearance is impaired, rapamycin may enhance autophagic flux to bypass the enzymatic block (Atilano et al., 2023).

Together, these findings establish renal degeneration as a driver of systemic decline in Drosophila models of GD and PD. Recent evidence for a brain-kidney axis in PD, where α-synuclein accumulates in renal tissue and may retrogradely propagate to the brain under renal failure (Yuan et al., 2025), supports the relevance of the renal system to PD. We propose that renal oxidative stress could influence such inter-organ signalling, with potential feedback on neurodegenerative cascades. Future validation of these mechanisms in mammalian models will be essential to define translational relevance.

If confirmed, our findings could inform new biomarker strategies and therapeutic targets for GBA1 mutation carriers and other at-risk groups. Maintaining renal health may represent a modifiable axis of intervention in neurodegenerative disease. Our results also reinforce the need for precision in redox modulation, particularly where antioxidant therapies may cause harm. Finally, the opposing effects of rapamycin in Gba1b and park models demonstrate that effective treatments must account for the underlying molecular defect. Renal imaging or biomarker assays in GBA1 mutation carriers may provide a means to monitor disease progression or therapeutic response. Identifying redox fragility and mTOR sensitivity as actionable parameters opens new avenues for screening in both invertebrate and vertebrate systems.

In conclusion, we uncover a critical and previously overlooked role for the renal system in GD and PD pathogenesis. We demonstrate that redox dyshomeostasis, renal degeneration, and mTOR-responsive pathways are central features of disease in Gba1b-deficient flies, and mechanistically distinct from those in Park mutant models. These findings support a broader, integrated view of neurodegenerative disease and pave the way for organ-specific interventions in disorders long considered purely neurocentric. Renal health, redox homeostasis and autophagy status emerge as key modulators of disease trajectory, offering a new conceptual and therapeutic framework beyond the brain.

Acknowledgements

We thank all members of the Kinghorn lab and the UCL Institute of Healthy Ageing, Michael Duchen and Antonella Spinazzola for critical comments and Shu Yi Koay for early contributions to the project. This work was supported by the Wellcome Trust (Wellcome Career Development Awards to K.J.K., 214589/Z/18/Z and 309250/Z/24/Z), funding from the Rosetrees Trust (K.J.K. and A.H. M701 and M701A), the MRC (K.J.K and A.H. MR/X011070/1), and the Vivensa Foundation (K.J.K. PF2302/8).

Additional files

Figure S1. S1A) Quantification of NAD+/NADH ratios in whole ctrl and Gba1b^-/-flies (unpaired t-test NAD+ ctrl vs Gba1b^-/-, p=0.7770; NADH ctrl vs Gba1b^-/-, p=0.0208*). S1B) Example image showing the localisation of Gba1b expression within Drosophila MTs. Gba1b^CRIMIC>mCherry.nls, DAPI (Blue) and mCherry (Red). Scale-bar = 100µm.

Figure S2. S2A) Drinking assay using differential fly weights from a starved mean, demonstrating a significant decrease in water ingestion by Gba1b^-/- flies compared to ctrl (unpaired t-test, *p=0.0057). S2B) Drinking assay by weight differential of a water source, demonstrating no significant difference in water ingestion in Gba1b^-/- flies compared to ctrl (unpaired t-test, p=0.50). S2C) Ctrl and Gba1b^-/- fly water weight across a desiccation time course (two-way-ANOVA, ctrl vs Gba1b^-/-***p<0.0001, 0hr vs 5hr vs 10hr ***p<0.0001). S2D) Ctrl and Gba1b^-/- fly water weight across a natriuresis assay (one-way ANOVA with Tukey’s multiple comparison’s test. Ctrl SYA vs ctrl NaCl, ***p<0.0001; ctrl SYA vs ctrl NaCl + H₂O, p=0.9748; ctrl NaCl vs ctrl NaCl + H₂O, ***p<0.0001; Gba1b^-/- SYA vs Gba1b^-/- NaCl, ***p<0.0001; Gba1b^-/- SYA vs Gba1b^-/- NaCl + H₂O, p=0.1299; Gba1b^-/- NaCl vs Gba1b^-/- NaCl + H₂O, **p=0.0041) S2E) Ctrl and Gba1b^-/- fly water weight before and after chill coma stress assay (two-way ANOVA ctrl vs Gba1b^-/-, ***p<0.0001, 0 hr vs 10 hr, p=0.8536)

Figure S3. S3A) Example images (20x) and quantification of peroxisomes of ctrl and Gba1b^-/- MTs. expressing tub>YFP:PTS. Quantification of mean peroxisome area (µm²) (unpaired t-test, p=0.0322), percentage of MT that is peroxisome +ve (unpaired t-test, **p=0.0019) and number of peroxisomal per µm² (unpaired t-test, p***=0.0003), All quantification was conducted on segmented max projection z-stacks. S3B) Example images (20x) of immunostaining for Sod1 (red; unpaired t-test, **p=0.0014) and CatA (red) (unpaired t-test, p=0.0801) in control and Gba1b^-/- MTs. Scale-bar = 50µm. S3C) Survival curve for tub>PGC1α; Gba1b^-/- (log-rank test, tub> vs tub> PGC1α, ***p<0.0001; > PGC1α; vs tub> PGC1α, ***p<0.0001, tub>; Gba1b^-/- vs tub> PGC1α; Gba1b^-/- ***p<0.0001; > PGC1α; Gba1b^-/- vs tub> PGC1α; Gba1b^-/-, ***p<0.0001). S3D) Survival curve for c42> PGC1α; Gba1b^-/- (Log-rank test, c42>; Gba1b^-/- vs c42> PGC1α; Gba1b^-/-, ***p<0.0001; > PGC1α; Gba1b^-/- vs c42> PGC1α; Gba1b^-/-, ***p<0.0001). S3E) Survival curve for c42>Ndi; Gba1b^-/- (log-rank test, c42>; Gba1b^-/- vs c42>Ndi; Gba1b^-/-, ***p<0.0001; >Ndi; Gba1b^-/- vs c42>Ndi; Gba1b^-/-, ***p<0.0001).

Figure S4. S4A) Example images and quantification of DAPI-stained nuclei proliferation from the ureter fork for elav>park; park^1/1and ctrl. (Elav>+; park^1/1 vs elav>park; park^1/1, p=0.0743, one-way ANOVA with Tukey’s multiple comparisons test). Scale-bar = 100µm. S4B) Climbing assay time course for klf15^-/-; park^1/1demonstrating no additive effect of nephrocyte ablation on climbing phenotypes (age-effect, **p=0.0002; genotype effect, ***p<0.0001; ctrl vs klf15^-/-p=0.1735; ctrl vs park^1/1, ***p<0.0001, park^1/1 vs klf15^-/- park^1/1, ***p=0.0003, two-way ANOVA with Tukey’s multiple comparisons test).

Figure S5. S5A) Example images corresponding to figure 8B), ctrl and Gba1b^-/- 3-week-old tubules treated with rapamycin and stained with DAPI. Scale-bar = 100µm. Arrow denotes the ureter fork. S5B) Example images corresponding to figure 8F), ctrl and park^1/1 3-week-old tubules stained with LysoTracker^TM, Scale-bar = 100µm. S5C) Quantification of FRUMS assay clearance from the hemolymph in ctrl and Gba1b^-/- (ctrl vs Gba1b^-/-, ***p<0.0001, log-rank test; ctrl vs ctrl rapa, p=0.666, log-rank test; Gba1b^-/-vs Gba1b^-/- rapa, p=0.0005, log-rank test). S5D) Quantification of FRUMS assay clearance from the whole body in ctrl and Gba1b^-/- flies (ctrl vs Gba1b^-/- cleared from haemolymph, ***p<0.0001, log-rank test; ctrl vs ctrl rapa p=0.561, log-rank test; Gba1b^-/-vs Gba1b^-/- rapa, **p=0.0038, log-rank test).

Additional information

Funding

Wellcome Trust

https://doi.org/10.35802/214589

Wellcome Trust (309250/Z/24/Z)

Medical Research Council (MR/X011070/1)

Dunhill Medical Trust (PF2302/8)

Rosetrees Trust (M701)

Rosetrees Trust (M701A)