E3 ligase AREL1 controls perinuclear localization of lysosomes and supports Purkinje cell survival
Luyi Jiang, Jiangfen Tang, Ya-Fen Zhang, Wen-Xuan Zou, Gang Deng, Na Tian, Xiaolu Zhao, Lei Han, Kai Liu, Bao-Liang Song, Jie Luo
TL;DR
This study shows that the E3 ligase AREL1 helps position lysosomes near the nucleus in neurons, which is important for their function and for preventing cell loss in the brain.
Contribution
The study identifies a novel mechanism involving AREL1, UBAC2, and V-ATPase that regulates lysosomal positioning and function in neurons.
Findings
AREL1 interacts with V-ATPase and catalyzes ubiquitylation of V1B2, promoting perinuclear lysosome localization.
AREL1 or UBAC2 depletion leads to peripheral lysosomes with reduced acidity and degradative capacity.
Arel1 knockout mice show Purkinje cell loss, motor impairment, and lysosomal dysfunction.
Abstract
Localization of lysosomes influences their properties, e.g., perinuclear lysosomes are more acidic but less mobile compared with the peripheral ones. Furthermore, the endoplasmic reticulum (ER) can actively regulate the dynamics and functions of lysosomes via membrane contact sites. In this study, we find that ER-resident apoptosis-resistant E3 ubiquitin protein ligase 1 (AREL1) establishes membrane contacts with lysosomes by directly interacting with the Voa subunit of V-ATPase. AREL1 also catalyzes K33-linked polyubiquitylation of V-ATPase V1B2 subunit, inducing its binding to UBAC2 localized in the perinuclear ER. Depletion of AREL1 or UBAC2 increases the number of peripheral lysosomes that possess partially assembled V-ATPase, elevated luminal pH, and attenuated degradative capacity. Knockdown of ZRANB1, the deubiquitylating enzyme that antagonizes AREL1-mediated V1B2…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14
Figure 15
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 16
Figure 17- —National Natural Science Foundation of China
- —National Key R&D Program of China
- —Natural Science Foundation of Hubei Province
- —111 Project of Ministry of Education of China
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsCellular transport and secretion · Autophagy in Disease and Therapy · Endoplasmic Reticulum Stress and Disease
Introduction
Lysosomes are highly heterogeneous organelles whose spatial distributions and functions are tightly correlated. They tend to cluster in the perinuclear region under normal conditions (Jongsma et al, 2016). Compared with the peripheral ones, perinuclear lysosomes exhibit higher acidity and reduced motility, which ensure optimal activity of hydrolytic enzymes, enhanced fusion of autophagosomes with lysosomes, and eventually efficient degradation of a broad range of substrates (Johnson et al, 2016; Jongsma et al, 2016; Rayens et al, 2022). When nutrients (amino acids, lipids, and growth factors) are abundant, lysosomes are induced to transport to the cell periphery, where they serve as a signaling platform for mammalian target of rapamycin (mTOR) to sense and transduce nutritional inputs (Jia and Bonifacino, 2019; Korolchuk et al, 2011; Raiborg, 2018).
The endolysosomal transport and positioning are modulated by a complex array of factors including motors and the associated proteins as well as interorganelle contacts (Bonifacino and Neefjes, 2017; Jongsma et al, 2020; Jongsma et al, 2023), particularly ER–endolysosome membrane contact sites (MCSs) (Friedman et al, 2013; Gao et al, 2022; Palomo-Guerrero et al, 2019; Raiborg et al, 2015; Rocha et al, 2009). The ER-embedded UBE2J1/RNF26 ubiquitylation complex has been shown to act cooperatively with USP15, probably USP17 as well, to establish perinuclear positioning of the endolysosomal system (Cremer et al, 2021; Jongsma et al, 2016; Lin et al, 2022). The sorting nexin SNX19 is another ER protein responsible for constraining endolysosomes in the perinuclear region via interacting with phosphatidylinositol 3-phosphate on the endolysosomal surface (Saric et al, 2021). It remains to be determined whether there are other molecular complexes that can actively tether lysosomes to the perinuclear ER, and whether defects in lysosomal perinuclear positioning may impair animal physiology.
The vacuolar-type adenosine triphosphatase (V-ATPase) is a major determinant of luminal acidity of various organelles, including lysosomes. The active holoenzyme is composed of a cytosolic V_1_ domain for ATP hydrolysis and a membrane-embedded V_o_ domain for proton pumping (Vasanthakumar and Rubinstein, 2020). These two domains can dissociate to become autoinhibited and reassemble to confer V-ATPase activity in response to various nutrient cues (Collins and Forgac, 2020; McGuire et al, 2017). In mammalian cells, the reassembly of lysosomal V-ATPase is increased by glucose and amino acid starvation (McGuire and Forgac, 2018; Stransky and Forgac, 2015), as well as when mammalian target of rapamycin complex 1 (mTORC1) is inactive (Ratto et al, 2022) or lysosomal phosphatidylinositol 4-phosphate accumulates (Ebner et al, 2023). Intriguingly, these conditions also cause perinuclear clustering of lysosomes. The results suggest that V-ATPase assembly/disassembly and lysosomal positioning are closely correlated.
In this study, we identify apoptosis-resistant E3 ubiquitin protein ligase 1 (AREL1) as a critical regulator of lysosomal positioning and degradative function. AREL1 acts by two manners: interacting with the V_o_a subunit of the V-ATPase to establish ER–lysosome MCSs, and catalyzing K33-linked polyubiquitylation of the V_1_B2 subunit that then binds to ER-resident UBAC2 to confer perinuclear localization of lysosomes. The deubiquitylating enzyme ZRANB1 by counteracting AREL1-mediated V_1_B2 ubiquitylation reduces ER–lysosome MCSs and releases lysosomes to the cell periphery. Moreover, disruption of the AREL1−UBAC2−V-ATPase axis not only impairs lysosomal perinuclear clustering but also compromises lysosomal pH and degradative capacity. In mice, knockout of Arel1 induces age-dependent Purkinje cell loss and motor impairments, with lipofuscin granules accumulated in Purkinje cells well before the onset of ataxic phenotype. Our results highlight the importance of ER-mediated lysosomal positioning and degradative function in neuronal health and disease.
Results
AREL1 is an ER-localized, phase-separated protein and interacts with the Voa subunit of the V-ATPase
We first sought to identify proteins at ER–lysosome MCSs using the split-TurboID in proximity labeling strategy (Cho et al, 2020), in which the amino- and carboxyl-terminal fragments of TurboID—an engineered E. coli biotin ligase—were fused to LAMP1-mCherry and EGFP-Sec61β, respectively (Fig. 1A). When ER and lysosomes were closely apposed, the two inactive fragments reconstituted into a functional enzyme that conjugated nearby proteins with exogenously added biotin. As proof of the strategy, fluorescently labeled streptavidin was detected at the interfaces between the ER and lysosomes in U2OS cells, a human osteosarcoma cell line (Fig. EV1A). We next performed the split-TurboID in proximity labeling experiment coupled to mass spectrometry in HEK293T cells owing to relatively high transfection efficiency (Fig. EV1B). The subcellular fractionation assay confirmed the enrichment of biotinylated proteins in the membrane fractions (Fig. EV1C). These biotinylated membrane proteins were then subjected to streptavidin affinity purification followed by mass spectrometry analysis. Of all the proteins identified (Dataset EV1), 85 harboring transmembrane domain(s) were profoundly enriched in cells transfected with both LAMP1-mCherry-Tb(N) and Tb(C)-EGFP-Sec61β compared with cells transfected with either one. Proteins reported to localize in the membranes other than the ER or lysosomes, or with well-established functions, were further excluded. Among 18 remaining candidates, most were lysosomal or ER proteins (Fig. EV1D and Dataset EV1), including those known to mediate the formation of ER–lysosome MCSs such as VAPA (Rocha et al, 2009), MOSPD2 (Di Mattia et al, 2018), and protrudin (Raiborg et al, 2015). AREL1 is of particular interest given its uncharacterized subcellular localization. It is a HECT-type E3 ligase reported to mediate degradation of mitochondrial proapoptotic proteins and pro-interleukin-1β precursor (Kim et al, 2013; Mishra et al, 2023).Figure 1AREL1 is an ER-localized, condensed, single transmembrane protein mediating ER–lysosome membrane contact formation through binding to the V_o_a subunit of V-ATPase.(A) Schematic illustration of the split-TurboID-based in proximity labeling assay to identify proteins at ER–lysosome membrane contact sites. Tb(N) TurboID (N terminus), Tb(C) TurboID (C terminus), mCh mCherry. (B) Schematic illustration of the human AREL1 protein. TM transmembrane domain, IDR intrinsically disordered region, LCR low-complexity region, HECT homologous to E6AP C-terminus. C790 is the catalytically active site. (C) Representative confocal images showing AREL1 localization at ER–lysosome membrane contact sites. U2OS cells (a human osteosarcoma cell line) were transduced with lentiviruses expressing AREL1-EGFP and mCherry-Sec61β and then immunostained with anti-LAMP1 antibody. Cell contour and nucleus are outlined using white dashed lines. Boxed areas are enlarged sequentially (first vertically and then horizontally). Scale bars, 10 μm (top row), 2 μm (bottom row), 1 μm (right panel). (D) Representative confocal images showing the apposition of endogenous AREL1 protein to lysosomes. U2OS cells were immunostained with anti-AREL1 and anti-LAMP1 antibodies. The boxed area is enlarged and shown at the bottom. Scale bars, 10 μm (main), 2 μm (inset). (E) Representative confocal images showing the localization of indicated AREL1 variants relative to lysosomes. U2OS cells were transduced with lentiviruses expressing the indicated AREL1 variants tagged with EGFP and immunostained with anti-LAMP1 antibody. Boxed areas are enlarged and shown at the bottom. Scale bars, 10 μm (main), 2 μm (inset). WT wild-type. (F) Superplots showing Pearson’s correlation coefficient for AREL1 variants and LAMP1 per cell (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: AREL1(WT)-EGFP, 21; AREL1(C790A)-EGFP, 21; and AREL1(Δhinge region)-EGFP, 21; from three independent experiments. Unpaired two-tailed Student’s t test. Pearson’s correlation coefficient for LAMP1 and AREL1(WT)-EGFP vs Pearson’s correlation coefficient for LAMP1 and AREL1(C790A)-EGFP, P = 0.4482; Pearson’s correlation coefficient for LAMP1 and AREL1(WT)-EGFP vs Pearson’s correlation coefficient for LAMP1 and AREL1(Δhinge region)-EGFP, P < 0.0001. (G) Representative images showing droplet formation of purified recombinant proteins in the presence of the indicated concentrations of NaCl. Scale bars, 10 μm. (H) Sequence alignment of the low-complexity region (LCR) of AREL1 from the indicated species. Cationic and aromatic amino acids are in gray and red shadows, respectively. (I) Representative images showing droplet formation of indicated LCR variants in the presence of 50 mM NaCl. Scale bars, 10 μm. Δ5 K, LCR with lysine residues at positions 346, 348, 349, 351, and 352 all deleted. (J) Representative confocal images showing the localization of indicated AREL1 variants relative to the ER. U2OS cells were transduced with lentiviruses expressing the indicated AREL1 variants tagged with EGFP and mCherry-Sec61β. Scale bars, 10 μm. Δ5 K, LCR with lysine residues at positions 346, 348, 349, 351, and 352 all deleted. (K) Representative confocal images showing WT and AREL1^−/−^ U2OS cells transduced with lentiviruses expressing indicated proteins and immunostained with anti-calnexin and anti-LAMP1 antibodies, followed by proximity ligation assay. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (L) Superplots showing the number of PLA puncta per cell (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: WT cells expressing EGFP, 15; AREL1^−/−^ cells expressing EGFP, 15; AREL1^−/−^ cells expressing AREL1(WT)-EGFP, 15; AREL1^−/−^ cells expressing AREL1(ΔLCR)-EGFP, 15 and AREL1^−/−^ cells expressing AREL1(Y354A/Y356A)-EGFP, 15; from three independent experiments. Unpaired two-tailed Student’s t test. WT cells expressing EGFP vs AREL1^−/−^ cells expressing EGFP, P = 0.0375; AREL1^−/−^ cells expressing EGFP vs AREL1^−/−^ cells expressing AREL1(WT)-EGFP, P = 0.0346; AREL1^−/−^ cells expressing EGFP vs AREL1^−/−^ cells expressing AREL1(ΔLCR)-EGFP, P = 0.9859; AREL1^−/−^ cells expressing EGFP vs AREL1^−/−^ cells expressing AREL1(Y354A/Y356A)-EGFP, P = 0.9810. (M) Representative confocal images showing the localization of AREL1-EGFP relative to lysosomes in U2OS cells transduced with lentiviruses encoding negative control shRNA (shNC) and shRNA against ATP6V0A3 (shATP6V0A3). Boxed areas are enlarged and shown on the right. Scale bars, 10 μm. (N) Superplots showing Pearson’s correlation coefficient for AREL1 variants and LAMP1 per cell (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: shNC, 21 and shATP6V0A3, 21; from three independent experiments. Unpaired two-tailed Student’s t test. shNC cells vs shATP6V0A3 cells, P < 0.0001. (O) Co-immunoprecipitation (IP) analysis showing the interaction between AREL1 variants and the V_o_a3 subunit of V-ATPase. HEK293T cells were transfected as indicated and subjected to IP with anti-FLAG beads. (P) AlphaFold 3-predicted interaction of AREL1 hinge region (G_706_TGDIS) (green) and the amino-terminal domain (NTD) of the V_o_a3 subunit (pink). ipTM = 0.32, pTM = 0.79. (Q) Co-IP analysis showing the interaction between V_o_a3 variants and AREL1. HEK293T cells were transfected as indicated and subjected to IP with anti-FLAG beads. Source data are available online for this figure.
We then examined the effects of each of 18 candidates on ER–lysosome MCSs using shRNA-mediated knockdown (Fig. EV1E) followed by proximity ligation assay (PLA) in U2OS cells. PLA allows the detection of two proteins at a distance of less than 40 nm and provides a quantitative measure of ER–lysosome MCSs (Lim et al, 2019; Saric et al, 2021; Soderberg et al, 2006). Silencing of AREL1 and several other genes (ATP6AP1, ATP6AP2, VAPA, VMA21, UBE2J1, MOSPD2, ZFYVE27, and TMEM9) significantly reduced the percentage of lysosomes in contacts with the ER (Fig. EV1F,G). RNF26 is a previously identified ER-resident ubiquitin ligase that tethers the endolysosomal system in the perinuclear region (Cremer et al, 2021; Cremer et al, 2023; Jongsma et al, 2016). Depletion of RNF26 caused a 24% reduction in PLA puncta in U2OS cells (Fig. EV1F,G).
We also performed the split-Turbo ID in proximity labeling assay using AREL1 knockout (KO) U2OS cells (Fig. EV1H,I) as an unbiased confirmation of our screening. The total levels of biotinylated proteins were reduced in AREL1 KO cells (Fig. EV1J). The protein levels of VAPA in the pellets were decreased as well. These results suggest that AREL1 is responsible for establishing ER–lysosome MCSs. Knockout of AREL1 in U2OS cells and knockdown of AREL1 in HEK293T cells markedly reduced PLA puncta for LAMP1–calnexin interactions (Fig. EV1K–O).
AREL1 has a transmembrane domain at the very amino terminus, with the rest lying in the cytosol (Fig. 1B). It was mostly present in the ER-enriched membrane fractions in density gradient centrifugation (Fig. EV2A). Ectopically expressed AREL1 showed a perinuclear concentration that corresponded well with the ER marker Sec61β while encompassing LAMP1-positive lysosomes (Fig. 1C). The close apposition of endogenous AREL1 protein to lysosomes (Fig. 1D) was confirmed using an antibody whose specificity had been validated using AREL1 KO cells (Fig. EV1H,I). The antibody recognizes the carboxyl-terminal HECT domain of AREL1, and therefore failed to detect AREL1 in the purified membrane fractions exposed to trypsin regardless of the presence or absence of detergent (Fig. EV2B). However, deletion of the entire C-terminal HECT domain (AREL1(1–482)) completely dissociated AREL1 from lysosomes (Fig. EV2C–E), whereas inactivation of the ubiquitin ligase activity (AREL1(C790A)) failed to do so (Fig. 1E,F). Using a series of truncations of the HECT domain (Fig. EV2C), we found that the flexible hinge region of six amino acids (G_706_TGDIS), which connects the N-lobe and the C-lobe of the HECT domain (Singh et al, 2019), was responsible for the association of AREL1 with lysosomes (Figs. EV2C–E and 1E,F).
The puncta pattern of AREL1 prompted us to investigate whether AREL1 could undergo phase separation. Indeed, there is an intrinsically disordered region (IDR, amino acids 313–345) followed by a low-complexity region (LCR, amino acids 346–360) between the transmembrane domain and the HECT domain (Fig. 1B). The purified mEGFP-LCR protein, but not mEGFP-IDR protein, formed phase-separated droplets in vitro (Fig. 1G). The sequence of LCR is highly conserved among species (Fig. 1H), and harbors six cationic amino acids (K346, K348, K349, K351, K352, and K360) and two aromatic amino acids (Y354 and Y356). Deletion of lysine residues preceding tyrosine residues ( \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\triangle$$\end{document} 5 K) or replacement of two tyrosine residues with alanine (Y354A/Y356A) disrupted phase separation of mEGFP-LCR protein, whereas substitutions tyrosine for tryptophan (Y354W/Y356W) or phenylalanine (Y354F/Y356F) had no effects (Fig. 1I), suggesting that the cation-π interaction between lysine and aromatic amino acids drives phase separation of AREL1. Consistent with the in vitro results, deletion of LCR or lysine residues within LCR as well as substitution of tyrosine residues for alanine resulted in a punctiform-to-meshwork redistribution of AREL1 that coincided with the pattern of the ER marker (Fig. 1J). Notably, re-expression of the mutants defective in phase separation failed to revert the percentage of lysosomes in contact with the ER, as indexed by PLA puncta, as that of the wild-type (WT) protein did in AREL1 KO cells (Fig. 1K,L).
To identify the interacting partners of AREL1 on the lysosomal membrane, we employed the split-TurboID in proximity labeling strategy again, with the two enzyme halves fused with AREL1 and LAMP1, respectively (Fig. EV3A). A total of 25 endolysosomal proteins (LAMP1 as the bait was not counted) were significantly enriched following biotin treatment, and 7 among them were the subunits of V-ATPase and highly ranked (Fig. EV3B; Dataset EV2). Many ER proteins were enriched as well (Fig. EV3B; Dataset EV2).
The composition of mammalian V-ATPase is shown in Fig. EV3C. Many subunits have multiple isoforms that express in cell-, tissue-, and organelle-specific manners (Toei et al, 2010). All four isoforms of subunit V_o_a were co-immunoprecipitated with AREL1 (Fig. EV3D). These isoforms share a homologous amino-terminal domain (NTD) (Fig. EV3E). It was this NTD and the hinge region (G_706_TGDIS) that mediated the interaction between V_o_a and AREL1 (Fig. EV3F,G). The NTDs of all four V_o_a isoforms were predicted to interact with the hinge region in a similar manner by AlphaFold 3 (Fig. EV3H).
We further sought to map the exact amino acid(s) mediating the interaction between the V_o_a subunit and the hinge region of AREL1 by alanine-scanning mutagenesis. The V_o_a3 isoform (encoded by ATP6V0A3) was chosen as a representative because it was highly expressed in U2OS cells (Fig. EV3I), and knockdown of ATP6V0A3 (Fig. EV3J,K) almost completely disrupted AREL1 association with lysosomes (Fig. 1M,N). The co-immunoprecipitation (IP) of AREL1 by V_o_a3 was nearly completely eliminated by D709A mutation in the hinge region and profoundly reduced by I710A mutation, with an even greater effect observed for the double-site mutant (Fig. 1O). According to the AlphaFold 3 modeling, the D709 residue of the hinge region forms a hydrogen bond with the H297 residue of V_o_a3, and the I710 residue inserts into a hydrophobic groove (Fig. 1P). Consistently, single and double mutations of the H297 and A305 residues of V_o_a3 markedly abrogated its interaction with AREL1 (Fig. 1Q).
Together, the above results suggest that ER-resident AREL1 establishes ER–lysosome MCSs by interacting with the V_o_a subunit of V-ATPase.
AREL1 regulates the perinuclear positioning of lysosomes
The effects of AREL1 on the ER and lysosomes were next evaluated. No apparent alterations in ER morphology or lysosome numbers were detected in AREL1 KO U2OS cells (Fig. EV4A,B). Compared with WT cells whose lysosomes were mostly clustered in the perinuclear region, AREL1 KO cells showed a dispersed distribution of lysosomes throughout the cytosol (Fig. 2A), as indicated by increased distances between the nuclear envelope and lysosomal center (Fig. 2B,C). The scattering of lysosomes caused by AREL1 deficiency was completely reverted following re-expression of WT AREL1 but not the catalytically inactive C790A mutant or AREL1 lacking the hinge region, LCR region, or carrying Y354A/Y356A mutations (Fig. 2D,E). Knockdown of AREL1 binding partner ATP6V0A3 similarly caused lysosome dispersal in U2OS cells (Fig. 2F,G). The spatial distribution of EEA1-positive early endosomes remained unaltered despite the absence of AREL1 (Fig. EV4C,D). We also generated AREL1 KO HeLa cells (Fig. EV4E,F) and found a redistribution of lysosomes but not early endosomes to the cell periphery (Fig. EV4G–J). These results suggest that AREL1 is critical for the perinuclear localization of lysosomes, and that E3 ubiquitin ligase activity, lysosome contacts, and phase separation property of AREL1 are all indispensable for its regulation of lysosomal distribution.Figure 2AREL1 regulates lysosomal positioning.(A) Representative confocal images showing the distribution of lysosomes in WT and AREL1^−/−^ U2OS cells. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (B) Schematic illustration showing distance measurement between lysosomes and the nuclear envelope. (C) Superplots showing the distance from lysosomes to the nuclear envelope (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells (# of lysosomes): WT, 15 (1076) and AREL1^−/−^, 13 (1026); from three independent experiments. Mann–Whitney U test. WT cells vs AREL1^−/−^ cells, P < 0.0001. (D) Representative confocal images showing WT and AREL1^−/−^ U2OS cells transduced with lentiviruses expressing indicated proteins and immunostained with anti-LAMP1 antibody. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (E) Superplots showing the distance from lysosomes to the nuclear envelope (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells (# of lysosomes): WT cells expressing EGFP, 9 (862); AREL1^−/−^ cells expressing EGFP, 9 (950); AREL1^−/−^ cells expressing AREL1(WT)-EGFP, 9 (801); AREL1^−/−^ cells expressing AREL1(C790A)-EGFP, 9 (948); AREL1^−/−^ cells expressing AREL1(Δhinge region)-EGFP, 9 (911); AREL1^−/−^ cells expressing AREL1(ΔLCR)-EGFP, 9 (882) and AREL1^−/−^ cells expressing AREL1(Y354A/Y356A)-EGFP, 9 (838); from three independent experiments. Mann–Whitney U test. WT cells expressing EGFP vs AREL1^−/−^ cells expressing EGFP, P < 0.0001; AREL1^−/−^ cells expressing EGFP vs AREL1^−/−^ cells expressing AREL1(WT)-EGFP, P < 0.0001; AREL1^−/−^ cells expressing EGFP vs AREL1^−/−^ cells expressing AREL1(C790A)-EGFP, P = 0.7795; AREL1^−/−^ cells expressing EGFP vs AREL1^−/−^ cells expressing AREL1(Δhinge region)-EGFP, P = 0.1415; AREL1^−/−^ cells expressing EGFP vs AREL1^−/−^ cells expressing AREL1(ΔLCR)-EGFP, P = 0.4267; AREL1^−/−^ cells expressing EGFP vs AREL1^−/−^ cells expressing AREL1(Y354A/Y356A)-EGFP, P = 0.8293. (F) Representative confocal images showing the distribution of lysosomes in U2OS cells transduced with lentiviruses encoding negative control shRNA (shNC) and shRNA against ATP6V0A3 (shATP6V0A3). Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (G) Superplots showing the distance from lysosomes to the nuclear envelope (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells (# of lysosomes): shNC, 20 (1357) and shATP6V0A3, 17 (1370); from three independent experiments. Mann–Whitney U test. shNC cells vs shATP6V0A3 cells, P < 0.0001. Source data are available online for this figure.
AREL1 deficiency impairs the acidification and degradative capacity of lysosomes
Perinuclear lysosomes have been found to be more acidic than the peripheral ones (Johnson et al, 2016). By using the Oregon Green Dextran whose fluorescence emitted upon excitation near 490 nm is exquisitely pH-sensitive whereas that near 440 nm is not (DiCiccio and Steinberg, 2011), we showed that, in both WT and AREL1 KO cells, lysosomal pH was elevated as lysosomes were further away from the nuclear envelope, and that AREL1 KO cells had significantly more peripheral lysosomes with higher luminal pH (Fig. 3A). The overall fluorescence intensity of LysoSensor green, which stains the acidic organelles, was drastically reduced in AREL1 KO cells (Fig. 3B,C).Figure 3AREL1 deficiency increases the number of peripheral lysosomes with elevated luminal pH, partially assembled V-ATPase, and reduced degradative potency.(A) The correlation of lysosomal pH with its distance to the nuclear envelope in WT and AREL1^−/−^ U2OS cells. The effects of the independent variable (genotype) on two dependent variables (distance and pH) were analyzed using multivariate analysis of variance (MANOVA) in RStudio, and multivariate significance was assessed using Pillai’s trace statistic (n = 40 lysosomes from WT U2OS cells and 36 lysosomes from AREL1^−/−^ U2OS cells). WT cells vs AREL1^−/−^ cells, P < 0.0001. (B) Representative confocal images showing WT and AREL1^−/−^ U2OS cells stained with LysoSensor green. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 20 μm. (C) Superplots showing the relative intensity of LysoSensor green (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: WT, 52 and AREL1^−/−^, 59; from three independent experiments. Unpaired two-tailed Student’s t test. WT cells vs AREL1^−/−^ cells, P < 0.0001. (D) Representative confocal images showing the localization of V_1_D and LAMP1 in WT and AREL1^−/−^ U2OS cells. Boxed areas are enlarged on the right. Cell contour and nucleus are outlined using white dashed lines. PP peripheral, PN perinuclear. Scale bars, 10 μm (main), 2 μm (inset). (E) Box plots showing the distance from V_1_D-positive and V_1_D-negative lysosomes to the nuclear envelope in WT and AREL1^−/−^ U2OS cells. Data are presented as median with interquartile range. Each box-and-whisker consists of the 25th quantile (the upper border of box), median (horizontal line inside the box), 75th quantile (the lower border of box), and vertical lines extending to the minimum and maximum values. # of cells (# of lysosomes): WT (V_1_D-positive), 14 (856); WT (V_1_D-negative), 14 (112); AREL1^−/−^ (V_1_D-positive), 14 (770) and AREL1^−/−^ (V_1_D-negative), 14 (532); from three independent experiments. Mann–Whitney U test. V_1_D-positive lysosomes vs V_1_D-negative lysosomes in WT cells, P < 0.0001; V_1_D-positive lysosomes in WT cells vs V_1_D-positive lysosomes in AREL1^−/−^ cells, P = 0.0142; V_1_D-positive lysosomes vs V_1_D-negative lysosomes in AREL1^−/−^ cells, P < 0.0001; V_1_D-negative lysosomes in WT cells vs V_1_D-negative lysosomes in AREL1^−/−^ cells, P < 0.0001. (F) Percentages of lysosomes positive and negative for V_1_D in WT and AREL1^−/−^ U2OS cells. Data are presented as mean±SD. # of cells: WT, 14 and AREL1^−/−^, 14; from three independent experiments. Unpaired two-tailed Student’s t test. V_1_D-negative lysosomes in WT cells vs V_1_D-negative lysosomes in AREL1^−/−^ cells, P < 0.0001. (G) Immunoblotting analysis of cytosolic (C) and membrane (M) fractions from WT and AREL1^−/−^ U2OS cells. (H) Quantification of the relative abundance of V_1_D in membrane and cytosolic fractions. Data are presented as means ± SD (n = three independent experiments). Unpaired two-tailed Student’s t test. V_1_D in membrane fractions of WT cells vs V_1_D in membrane fractions of AREL1^−/−^ cells, P = 0.0047; V_1_D in cytosolic fractions of WT cells vs V_1_D in cytosolic fractions of AREL1^−/−^ cells, P = 0.0006. (I) Representative confocal images showing WT and AREL1^−/−^ U2OS cells incubated with DQ-OVA. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (J) Representative confocal images showing WT and AREL1^−/−^ U2OS cells stained with SiR-Lysosome. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (K) Superplots showing the relative intensity of DQ-OVA (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: WT, 82 and AREL1^−/−^, 77; from three independent experiments. Unpaired two-tailed Student’s t test. WT cells vs AREL1^−/−^ cells, P < 0.0001. (L) Superplots showing the relative intensity of SiR-Lysosome (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: WT, 23 and AREL1^−/−^, 24; from three independent experiments. Unpaired two-tailed Student’s t test. WT cells vs AREL1^−/−^ cells, P = 0.0001. Source data are available online for this figure.
The luminal pH of lysosomes is determined by the dynamic balance between proton influx through V-ATPase and proton efflux through TMEM175 (Hu et al, 2022), SLC7A11 (Zhou et al, 2025), as well as other transporters or channels. The V_o_ and V_1_ domains of V-ATPase are connected by the central stalk of D and F subunits (Fig. EV3C) (Wang et al, 2020). The association of V_1_D subunit with the lysosomal membrane protein LAMP1 provides a quantitative measure of intact, active V-ATPases. To better visualize membrane-associated V_1_ domains, we permeabilized cells with digitonin prior to fixation so that free-floating V_1_ domains could diffuse out of the cells. In both WT and AREL1 KO cells, V_1_D was perinuclearly distributed and colocalized with most, if not all, perinuclear lysosomes (Fig. 3D). Lysosomes negative for V_1_D were more distant from the nucleus than V_1_D-positive ones (Fig. 3E), suggesting that V-ATPases on the membrane of peripheral lysosomes are incomplete and therefore functionally incompetent. Depletion of AREL1 significantly increased the percentage of V_1_D-negative lysosomes as well as their distance to the nuclear envelope (Fig. 3E,F). In support of these results, V_1_D was predominantly associated with the membrane fractions isolated from WT cells but shifted to the cytosolic fractions when AREL1 was ablated (Fig. 3G,H).
The degradative potency of lysosomes was evaluated using two kinds of fluorogenic probes. DeQuenched ovalbumin (DQ-OVA) is a BODIPY-labeled substrate for lysosomal proteases that gives off fluorescence upon enzymatic digestion (Albrecht et al, 2020), whereas Silicon rhodamine (SiR)-Lysosome is a fluorophore-tagged pepstatin A that can bind specifically to active cathepsin D, the main acid hydrolase in the lysosome (Lukinavicius et al, 2016). Both probes emitted robust fluorescence in the perinuclear region of WT cells, suggesting perinuclear lysosomes have higher enzymatic activities than those in the peripheral region (Fig. 3I,J). The intensity of fluorescent signals was significantly less in AREL1 KO cells (Fig. 3K,L).
These findings are consistent with the previous reports that peripheral lysosomes have increased pH (Johnson et al, 2016), reduced proteolytic capacity (Johnson et al, 2016; Korolchuk et al, 2011), and reduced amounts of the V_1_ subunits of V-ATPase (Tang et al, 2021), underscoring an important role of AREL1 in regulating lysosomal perinuclear positioning and degradative capacity.
AREL1 ubiquitylates the V1B2 subunit of V-ATPase for binding to UBAC2 in the perinuclear ER
That the C790A mutant despite interacting with lysosomes (Fig. 1E,F) fails to anchor them perinuclearly (Fig. 2D,E) suggests a second pair of proteins, one being ubiquitylated by AREL1, is required for tethering lysosomes in the perinuclear region. We first investigated which subunit(s) of V-ATPase could be ubiquitylated by AREL1. The results showed that AREL1 selectively ubiquitylated the B2 isoform of the V_1_ domain, among the examined V-ATPase subunits, in a catalytic activity-dependent manner (Figs. 4A and EV5A). Consistent with the findings that AREL1 catalyzes K33-linked polyubiquitylation (Kristariyanto et al, 2015; Michel et al, 2015), the K33R ubiquitin mutant was the only one out of seven K-to-R mutants that failed to confer AREL1-mediated ubiquitylation of V_1_B2 (Fig. EV5B). The ubiquitin that only contains the lysine residue at position 33 (K33 only) was sufficient to support AREL1-mediated V_1_B2 ubiquitylation (Fig. EV5C).Figure 4AREL1 and ZRANB1 regulate the ubiquitylation of the V_1_B2 subunit of V-ATPase for lysosomal perinuclear positioning and degradative function.(A) HEK293T cells were transfected as indicated and subjected to immunoprecipitation (IP) with anti-FLAG beads followed by immunoblotting to analyze ubiquitylation. (B) WT and AREL1^-/-^ U2OS cells were harvested and subjected to IP with control IgG beads or anti-V_1_B2 beads, followed by immunoblotting to analyze ubiquitylation. (C) U2OS cells were harvested and subjected to subcellular fractionation. The cytosolic (C) and membrane (M) fractions were then subjected to IP with control IgG beads or anti-V_1_B2 beads, followed by immunoblotting to analyze ubiquitylation. (D) Representative confocal images showing the distribution of lysosomes in U2OS cells transduced with lentiviruses encoding negative control shRNA (shNC) and shRNA against ATP6V1B2 (shATP6V1B2). Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (E) Superplots showing the distance from lysosomes to the nuclear envelope (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells (# of lysosomes): shNC, 12 (1101) and shATP6V1B2, 12 (1045); from three independent experiments. Mann–Whitney U test. shNC cells vs shATP6V1B2 cells, P < 0.0001. (F) HEK293T cells were transfected as indicated and subjected to IP with anti-FLAG beads followed by immunoblotting to analyze ubiquitylation. (G) Representative confocal images showing U2OS cells transduced with lentiviruses expressing indicated mCherry-tagged ZRANB1 variants and immunostained with anti-LAMP1 antibody. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (H) Superplots showing the distance from lysosomes to the nuclear envelope (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells (# of lysosomes): mCherry, 12 (974), mCherry-ZRANB1 (WT), 10 (1006), and mCherry-ZRANB1 (C443S), 9 (727); from three independent experiments. Mann–Whitney U test. Cells expressing mCherry vs cells expressing mCherry-ZRANB1 (WT), P < 0.0001; Cells expressing mCherry vs cells expressing mCherry-ZRANB1 (C443S), P = 0.4803. (I) U2OS cells transduced with lentiviruses encoding negative control shRNA (shNC) and shRNA against ZRANB1 (shZRANB1) were harvested and subjected to IP with control IgG beads or anti-V_1_B2 beads, followed by immunoblotting to analyze ubiquitylation. (J) Representative confocal images showing the distribution of lysosomes in U2OS cells transduced with lentiviruses encoding negative control shRNA (shNC) and shRNA against ZRANB1 (shZRANB1). Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (K) Superplots showing the distance from lysosomes to the nuclear envelope (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells (# of lysosomes): shNC, 22 (1795) and shZRANB1, 23 (1759); from three independent experiments. Mann–Whitney U test. shNC cells vs shZRANB1 cells, P < 0.0001. (L) Representative confocal images showing shNC and shZRANB1 U2OS cells stained with LysoSensor green. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (M) Superplots showing the relative intensity of LysoSensor green (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: shNC, 80 and shZRANB1, 73; from three independent experiments. Unpaired two-tailed Student’s t test. shNC cells vs shZRANB1 cells, P < 0.0001. (N) Representative confocal images showing shNC and shZRANB1 U2OS cells incubated with DQ-OVA. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (O) Superplots showing the relative intensity of DQ-OVA (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: shNC, 60 and shZRANB1, 60; from three independent experiments. Unpaired two-tailed Student’s t test. shNC cells vs shZRANB1 cells, P < 0.0001. Source data are available online for this figure.
ATP6V1B2 was highly expressed in U2OS cells (Fig. EV6A) and endogenous V_1_B2 protein was ubiquitylated by AREL1 (Fig. 4B). We further immunoprecipitated endogenous V_1_B2 protein from membrane and cytosolic fractions, respectively, and found only membrane-associated V_1_B2 to be ubiquitylated (Fig. 4C). These results suggest that the V_1_B2 subunit in intact V-ATPase complex, rather than the unassembled cytosolic V_1_B2, is ubiquitylated by AREL1. In ATP6V1B2 knockdown cells (Fig. EV6B,C), lysosomes were no longer restrained in the perinuclear region (Fig. 4D,E), and the numbers of calnexin-LAMP1 PLA puncta were reduced (Fig. EV6D,E). Three copies of V_1_B2 and V_1_A constitute the V_1_ domain that is responsible for ATP binding and hydrolysis (Collins and Forgac, 2020; Vasanthakumar and Rubinstein, 2020). The detrimental effects of ATP6V1B2 deficiency on lysosomal acidity and degradative function were anticipated and therefore not examined.
ZRANB1 is a K29/K33-specific deubiquitylating enzyme (Licchesi et al, 2011). Overexpression of WT ZRANB1 but not catalytically inactive C443S mutant markedly reduced ubiquitin-positive signals associated with lysosomes (Fig. EV6F,G) and effectively counteracted V_1_B2 ubiquitylation by AREL1 (Fig. 4F). Lysosomes were dispersed towards the cell periphery when WT ZRANB1 was overexpressed but remained perinuclearly following C443S mutant overexpression (Fig. 4G,H). To examine whether ZRANB1 could trim ubiquitin chains from V_1_B2 protein in the endogenous context, we generated ZRANB1 knockdown U2OS cells (Fig. EV6H) and indeed found substantially increased ubiquitylation of endogenous V_1_B2 protein compared with control cells (Fig. 4I). The average distances between lysosomes and the nuclear envelope were reduced while the PLA signals of ER–lysosome MCSs were increased in ZRANB1 knockdown cells (Figs. 4J,K and EV6I,J). Silencing of ZRANB1 markedly increased the fluorescence intensities of LysoSensor green and DQ-OVA (Fig. 4L–O).
We hypothesized that the interaction between ubiquitylated V_1_B2 and membrane-anchored ubiquitin-binding protein(s) might account for perinuclear localization of lysosomes. Hundreds of the proteins harboring ubiquitin-binding domains were analyzed for the presence of transmembrane domains. Ancient ubiquitous protein 1 (AUP1), rhomboid domain-containing protein 3 (RHBDD3), and ubiquitin-associated domain-containing protein 2 (UBAC2) were the three that fit the criteria (Fig. EV7A). Whereas AUP1 and RHBDD3 appeared as the cytoplasmic foci, UBAC2 was colocalized with the ER marker Sec61β and concentrated around the nucleus (Fig. EV7B). Knockdown of UBAC2 dispersed lysosomes to the cell periphery, while that of AUP1 or RHBDD3 did not affect lysosomal positioning (Fig. EV7C–E).
UBAC2 has three transmembrane segments and a cytosolic ubiquitin-associated (UBA) domain for interacting with ubiquitylated proteins (Fig. 5A). The WT form of UBAC2 was co-immunoprecipitated with V_1_B2, whereas the UBA-deleted mutant failed to do so (Fig. 5B). Knockout of AREL1 completely abolished the interaction between UBAC2 and V_1_B2, and complementing the WT form but not the catalytically inactive C790A mutant rescued UBAC2–V_1_B2 interaction in AREL1 KO cells (Fig. 5C). Moreover, overexpression of WT ZRANB1 instead of the inactive C443S mutant abolished the interaction between UBAC2 and V_1_B2 (Fig. 5D).Figure 5UBAC2 interacts with the ubiquitylated V_1_B2 subunit and regulates lysosomal positioning and degradative function.(A) Schematic illustration of the human UBAC2 protein. UBA, ubiquitin-associated. (B) Co-immunoprecipitation (IP) analysis of HEK293T cells transfected as indicated. (C) Co-IP of WT and AREL1^-/-^ U2OS cells transfected as indicated. (D) Co-IP analysis of HEK293T cells transfected as indicated. (E) Representative confocal images showing the distribution of lysosomes in U2OS cells transduced with lentiviruses encoding negative control shRNA (shNC) and shRNA against UBAC2 (shUBAC2), transfected with the plasmids expressing indicated proteins, and immunostained with anti-LAMP1 antibody. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. (F) Superplots showing the distance from lysosomes to the nuclear envelope (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells (# of lysosomes): shNC cells expressing EGFP, 11 (930), shUBAC2 cells expressing EGFP, 12 (1112), shUBAC2 cells expressing UBAC2(WT)-EGFP, 9 (808), and shUBAC2 cells expressing UBAC2(ΔUBA)-EGFP, 9 (974); from 3 independent experiments. Mann–Whitney U test. shNC cells expressing EGFP vs shUBAC2 cells expressing EGFP, P < 0.0001; shUBAC2 cells expressing EGFP vs shUBAC2 cells expressing UBAC2(WT)-EGFP, P < 0.0001; shUBAC2 cells expressing EGFP vs shUBAC2 cells expressing UBAC2(ΔUBA)-EGFP, P = 0.1125. (G) Representative confocal images showing LysoSensor green in control and UBAC2 knockdown U2OS cells. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 20 μm. (H) Superplots showing the relative intensity of LysoSensor green (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: shNC, 26 and shUBAC2, 27; from three independent experiments. Unpaired two-tailed Student’s t test. shNC cells vs shUBAC2 cells, P < 0.0001. (I) Representative confocal images showing the localization of V_1_D and LAMP1 in U2OS cells transduced with lentiviruses encoding negative control shRNA (shNC) and shRNA against UBAC2 (shUBAC2). Cell contour and nucleus are outlined using white dashed lines. Boxed areas are enlarged on the right. Scale bars, 10 μm (main), 2 μm (inset). (J) Box plots showing the distance from V_1_D-positive and V_1_D-negative lysosomes to the nuclear envelope in shNC and shUBAC2 U2OS cells. Data are presented as median with interquartile range. Each box-and-whisker consists of the 25th quantile (the upper border of box), median (horizontal line inside the box), 75th quantile (the lower border of box), and vertical lines extending to the minimum and maximum values. # of cells (# of lysosomes): shNC (V_1_D-positive), 11 (713); shNC (V_1_D-negative), 11 (91); shUBAC2 (V_1_D-positive), 11 (581) and shUBAC2 (V_1_D-negative), 11 (454); from three independent experiments. Mann–Whitney U test. V_1_D-positive lysosomes vs V_1_D-negative lysosomes in shNC cells, P < 0.0001; V_1_D-positive lysosomes in shNC cells vs V_1_D-positive lysosomes in shUBAC2 cells, P = 0.0428; V_1_D-positive lysosomes vs V_1_D-negative lysosomes in shUBAC2 cells, P < 0.0001; V_1_D-negative lysosomes in shNC cells vs V_1_D-negative lysosomes in shUBAC2 cells, P < 0.0001. (K) Percentages of lysosomes positive and negative for V_1_D in shNC and shUBAC2 U2OS cells. Data are presented as mean±SD. # of cells: shNC, 11 and shUBAC2, 11; from three independent experiments. Unpaired two-tailed Student’s t test. V_1_D-negative lysosomes in shNC cells vs V_1_D-negative lysosomes in shUBAC2 cells, P < 0.0001. (L) Representative confocal images showing DQ-OVA in shNC and shUBAC2 U2OS cells. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. The right is superplots showing the relative intensity of DQ-OVA (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: shNC, 59 and shUBAC2, 66; from three independent experiments. Unpaired two-tailed Student’s t test. shNC cells vs shUBAC2 cells, P < 0.0001. (M) Representative confocal images showing SiR-Lysosome in shNC and shUBAC2 U2OS cells. Cell contour and nucleus are outlined using white dashed lines. Scale bars, 10 μm. Right is superplots showing the relative intensity of SiR-Lysosome (small dots) and its mean per independent experiment (large dots). Means and error bars (SD) are shown as black bars. # of cells: shNC, 24 and shUBAC2, 27; from three independent experiments. Unpaired two-tailed Student’s t test. shNC cells vs shUBAC2 cells, P < 0.0001. Source data are available online for this figure.
UBAC2 depletion induced the peripheral distribution of lysosomes, and the phenotype was completely rescued by re-expression of WT UBAC2 but not the UBA domain-deleted, V_1_B2-binding incompetent mutant (Fig. 5E,F). As seen in AREL1 KO cells (Fig. 3), LysoSensor green signal was greatly reduced in UBAC2 knockdown cells (Fig. 5G,H), together with significant increases in the percentage of V_1_D-negative lysosomes and their distance from the nucleus (Fig. 5I–K). Depletion of UBAC2 markedly attenuated the fluorescence of DQ-OVA and SiR-Lysosome (Fig. 5L,M). UBAC2 knockdown cells displayed reduced numbers of calnexin-LAMP1 PLA puncta (Fig. EV7F,G).
Deficiency of AREL1 did not affect the subcellular distribution of UBAC2 (Fig. EV7H). However, in UBAC2 knockdown cells where lysosomes were peripherally dispersed (Fig. EV7D), WT AREL1 and the C790A mutant were redistributed to the cell periphery, whereas the hinge region-deleted AREL1 still stayed perinuclearly (Fig. EV7I). Since the hinge region-deleted AREL1 failed to interact with lysosomes (Fig. 1E,F), these results suggest that the peripheral distribution of AREL1 is actually conferred by that of lysosomes in UBAC2 knockdown cells. The findings that AREL1 as an integral membrane protein can co-travel with lysosomes are not totally unexpected, since lysosomes can actively regulate ER structure and distribution (Lu et al, 2020).
To reconcile our model with rapid assembly/disassembly kinetics of V-ATPase, we performed fluorescence recovery after photobleaching (FRAP) experiments as described previously (Bodzeta et al, 2017; Sava et al, 2024). V_1_E1 tagged with mCherry exhibited a robust punctate staining pattern that was colocalized with LAMP1-EGFP (Fig. EV8A), and was chosen as a measure for the recruitment of V_1_ domain to the relatively immobile perinuclear lysosomes. The perinuclear regions of WT and ZRANB1 knockdown cells were photobleached, and the recovery of fluorescence was measured. The half-time of fluorescence recovery (τ1/2) was 24.46 s in WT cells and 44.06 s in ZRANB1 knockdown cells (Fig. EV8B,C), indicating a delayed exchange of V_1_E1 between the cytosolic pool and that bound to the lysosomal surface when ZRANB1 was depleted. These results suggest that increased ubiquitylation of V_1_B2 can help stabilize the V-ATPase holoenzyme and facilitate UBAC2-mediated perinuclear localization of lysosomes. It should also be emphasized that the recovery time in our FRAP experiments was similar to that reported by the abovementioned previous studies (Bodzeta et al, 2017; Sava et al, 2024), suggesting that V-ATPase assembly is indeed rapid but still subjected to regulation by ubiquitylation/deubiquitylation.
Together, these results suggest that AREL1 functions in concert with UBAC2 to anchor lysosomes in the perinuclear region for optimal degradative capacity.
AREL1 deficiency causes age-associated neurodegeneration in mice
To investigate the role of AREL1 in animal physiology, we generated whole-body Arel1 knockout mice (Fig. EV9A). Arel1 knockout homozygotes (Arel1^−/−^) were born at a normal Mendelian ratio and exhibited no gross abnormalities after birth. Surprisingly, Arel1^−/−^ mice at 12 months of age started to lose balance and showed circling behavior and head tilt (Fig. 6A; Movies EV1 and EV2). When lifted up by the tails, Arel1^−/−^ mice displayed a hindlimb clasping phenotype (Fig. 6B,C; Movies EV3 and EV4). To systematically assess the locomotor functions of Arel1 mice, we subjected 12-month-old males and females for footprint analysis (Fig. 6D). Compared with age-matched controls, both male and female Arel1^−/−^ mice displayed aberrant gait patterns characterized by altered stride length, sway and stance (Fig. 6E). In the rotarod test, Arel1^−/−^ males and females spent significantly less time on the rod (Fig. 6F). It is interesting that the rotarod performance was worse as compared to footprint one in males, whereas footprint parameters were a bit more altered in females. These results suggest that Arel1 deficiency causes late-onset motor impairment in mice.Figure 6Arel1^−/−^ mice show age-dependent cerebellar ataxia, Purkinje cell loss, and lipofuscin accumulation.(A) Representative images showing the walking behavior of 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice. (B) Representative images showing the postures of 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice in the tail suspension test. (C) Hindlimb clasping score of 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice. Data are presented as mean ± SD (n = 16 mice per group). Unpaired two-tailed Student’s t test. Arel1^+/+^ mice vs Arel1^−/−^ mice, P < 0.0001. (D) Representative footprints of 12-month-old Arel1^+/+^ and Arel1^−/−^ male and female mice. Scale bars, 2 cm. (E) Stride length, sway, and stance of 12-month-old Arel1^+/+^ and Arel1^−/−^ mice. Data are presented as mean ± SD (n = 12 for male mice per genotype, and n = 10 for female mice per genotype). Unpaired two-tailed Student’s t test. Stride length in Arel1^+/+^ male mice vs stride length in Arel1^−/−^ male mice, P = 0.0482; sway in Arel1^+/+^ male mice vs sway in Arel1^−/−^ male mice, P = 0.0297; stance in Arel1^+/+^ male mice vs stance in Arel1^−/−^ male mice, P = 0.0024; stride length in Arel1^+/+^ female mice vs stride length in Arel1^−/−^ female mice, P < 0.0001; sway in Arel1^+/+^ female mice vs sway in Arel1^−/−^ female mice, P = 0.9707; stance in Arel1^+/+^ female mice vs stance in Arel1^−/−^ female mice, P = 0.0055. (F) Time to fall off the rods of 12-month-old Arel1^+/+^ and Arel1^−/−^ mice. Data are presented as mean ± SD (n = 12 for male mice per genotype, and n = 10 for female mice per genotype). Unpaired two-tailed Student’s t test. Arel1^+/+^ male mice vs Arel1^−/−^ male mice, P < 0.0001; Arel1^+/+^ female mice vs Arel1^−/−^ female mice, P = 0.0203. (G) Immunoblotting analysis showing AREL1 protein expression in the indicated brain regions from 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice. (H) Representative immunohistochemical staining images showing AREL1 expression in the cerebellum of 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice. Boxed areas are enlarged and shown at the bottom. Scale bars, 100 μm (left), 50 μm (right). (I, J) Representative confocal images showing calbindin expression in the cerebellum of 12-month-old Arel1^+/+^ and Arel1^−/−^ male (I) and female (J) mice. Boxed areas are enlarged and shown on the right. Scale bars, 1 mm (main), and 50 μm (inset). Quantification of calbindin-positive Purkinje cells is presented as mean ±SD (n = 4 mice per genotype). Unpaired two-tailed Student’s t test. Arel1^+/+^ male mice vs Arel1^−/−^ male mice, P = 0.0012; Arel1^+/+^ female mice vs Arel1^−/−^ female mice, P = 0.0105. (K, L) Representative confocal images showing calbindin staining (magenta) and lipofuscin autofluorescence (green) in Purkinje cells of 12-month-old Arel1^+/+^ and Arel1^−/−^ male (K) and female (L) mice. Scale bars, 10 μm (main). Quantification of lipofuscin autofluorescence in calbindin-positive Purkinje cells is presented as mean ± SD (n = 55 and 52 cells for Arel1^+/+^ and Arel1^−/−^ males, respectively, and n = 57 and 56 cells for Arel1^+/+^ and Arel1^−/−^ females, respectively). Unpaired two-tailed Student’s t test. Arel1^+/+^ male mice vs Arel1^−/−^ male mice, P < 0.0001; Arel1^+/+^ female mice vs Arel1^−/−^ female mice, P < 0.0001. (M) Representative electron micrographs showing Purkinje cells in 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice. Lysosomes (LY) and lipofuscin (LF) granules are enlarged in A and B, respectively. Scale bars, 5 μm (main), 0.5 μm (Box A), 0.5 μm (Box B). (N) Lysosome diameters measured using electron micrographs. Data are presented as mean ± SD (n = 115 and 70 lysosomes for 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice, respectively). Unpaired two-tailed Student’s t test. Arel1^+/+^ mice vs Arel1^−/−^ mice, P = 0.0009. (O) Box plots showing the distance from lysosomes to the nuclear envelope measured using electron micrographs. Data are presented as median with interquartile range (n = 115 and 70 lysosomes for 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice, respectively). Mann–Whitney U test. Arel1^+/+^ mice vs Arel1^−/−^ mice, P = 0.0005. (P) Lipofuscin numbers per cell. Data are presented as mean ± SD (n = 15 cells for both 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice). Unpaired two-tailed Student’s t test. Arel1^+/+^ mice vs Arel1^−/−^ mice, P < 0.0001. (Q) Size distributions of lipofuscins. Data are presented as means ± SD (n = 43 and 122 for 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice, respectively). Unpaired two-tailed Student’s t test. P values from left to right: 0.2187, 0.0024, 0.2467, 0.1440, 0.0086, 0.0132, 0.0477, 0.3259, 0.0719. (R) Representative periodic acid-Schiff staining showing lipofuscin granules in the cerebellum of 12-month-old Arel1^+/+^ and Arel1^−/−^ male mice. Boxed areas are enlarged as numbered. Cell contour is outlined using black dashed lines. Red arrows indicate aggregates. Scale bars, 500 μm (main), 30 μm (upper), 15 μm (lower). (S) Working model of AREL1-mediated lysosome positioning and function. Source data are available online for this figure.
According to several tissue expression databases [BIOGPS, Expression Atlas, TissueEnrich (Jain and Tuteja, 2019), JensenLab (Palasca et al, 2018), and Human Protein Atlas], AREL1 is widely expressed across mouse tissues, with a relatively higher level in the adult brain of mice and humans. Quantitative real-time PCR analysis confirmed AREL1 expression in the mouse brain (Fig. EV9B). Using Arel1^−/−^ mice as negative controls for the antibody, we found the cerebellum was one of the brain regions expressing AREL1 protein (Fig. 6G). Further examination of cerebellar sections by immunohistochemistry showed that AREL1 was concentrated in Purkinje cells (Fig. 6H). In support of our findings, the in situ hybridization data retrieved from Allen Mouse Brain Atlas also reveal the high expression of Arel1 that corresponds to calbindin-positive Purkinje cells (Fig. EV9C). In fact, by analyzing the previously published single-nucleus transcriptomics of mouse cerebellar cortex (Kozareva et al, 2021), we found that Arel1, Zranb1, V-ATPase, and Ubac2 were highly expressed in Purkinje neurons (Fig. EV9D).
We detected no apparent abnormalities in the cerebral cortex or hippocampus of 12-month-old Arel1^−/−^ mice (Fig. EV9E). However, Arel1^−/−^ cerebellum had significantly less Purkinje cells compared with WT controls (Fig. 6I,J). Lipofuscin is composed of undigested remnants, including proteins, lipids, carbohydrates, and other cell materials due to lysosomal dysfunction over time (Heinsen, 1979; Sobaniec-Lotowska, 2001). We observed many more lipofuscin-positive puncta in the residual Purkinje cells of Arel1^−/−^ mice (Fig. 6K,L).
Under the electron microscope, lysosomes appeared as spherical organelles with uniformly dense matrices (Fig. 6M, Box A and yellow dots in Fig. EV9F), and lipofuscin granules were highly osmiophilic structures irregular in shape and size and contained one or multiple electron-lucent droplets (Fig. 6M, Box B and red patches in Fig. EV9F). In Arel1^−/−^ Purkinje cells, lysosomes were significantly larger and more distant from the nucleus (Fig. 6N,O). Lipofuscin granules were occasionally observed in WT Purkinje cells, probably as a result of aging; however, their numbers and sizes were markedly increased in Arel1^−/−^ Purkinje cells (Fig. 6P,Q). Periodic acid-Schiff staining also revealed the accumulation of lipofuscin granules in Arel1^−/−^ Purkinje cells (Fig. 6R). These results indicate lysosomal positioning and degradative function are impaired in Purkinje cells of Arel1^−/−^ mice.
To investigate whether ataxia, Purkinje cell loss, and lipofuscin accumulation seen in mid-aged Arel1^−/−^ mice occurred concomitantly or sequentially, we examined male mice at 6 months of age, when Arel1^−/−^ ones showed no deficits in footprint analysis (Fig. EV9G,H) or the rotarod test (Fig. EV9I). The numbers of Purkinje cells in Arel1^−/−^ mice were similar to those in WT mice (Fig. EV9J). However, significantly more lipofuscins were detected in Arel1^−/−^ Purkinje cells (Fig. EV9K). These results suggest that lysosomal dysfunction may cause late-onset Purkinje cell loss and ataxia. Whether Arel1 deficiency may result in other neurological impairments such as central and peripheral vestibular problems merits further investigation.
Discussion
In the current study, we demonstrate that lysosomes are anchored perinuclearly by the ER via a dual-tethering mechanism that involves two ER-embedded proteins (AREL1 and UBAC2) and two V-ATPase subunits (V_o_a and V_1_B2). AREL1 interacts with V_o_a and further catalyzes K33-linked ubiquitylation of V_1_B2. The ubiquitylated V_1_B2 then binds to perinuclearly localized UBAC2 to anchor lysosomes in the perinuclear region (Fig. 6S). The deubiquitylating enzyme ZRANB1 can hydrolyze K33-linked ubiquitin chains from the V_1_B2 subunit, thereby releasing lysosomes to the cell periphery. Via such ubiquitylation/deubiquitylation cycles, lysosomes dynamically shuttle between perinuclear and peripheral regions.
The positioning of lysosomes closely correlates with their luminal pH and functions. Perinuclear lysosomes are known to be more acidic than the peripheral ones, partly because the latter have reduced proton-pumping activity and increased rates of proton leakage (Johnson et al, 2016). The increased density of V-ATPase holoenzyme in perinuclear lysosomes can be another contributing factor as well. In fact, both V_1_A subunit, as indexed by the SidK probe, and V_1_C1 have been found to be perinuclearly concentrated (Maxson et al, 2022; Tang et al, 2021). The V_1_ subunits are, as revealed by proteomics, highly enriched on lysosomes isolated from OSW-1-treated cells and strongly decreased from amino acid-restimulated cells, where lysosomes are induced to cluster in the perinuclear and peripheral region, respectively (Ebner et al, 2023; Ratto et al, 2022). However, the mechanisms by which V_1_ subunits are perinuclearly present are poorly understood. In addition to microtubule acetylation that accounts for perinuclear V_1_C1 localization (Tang et al, 2021), we hereby show that V_1_B2 ubiquitylation by AREL1 is required for its interaction with UBAC2 in the perinuclear ER, which can help secure lysosomes for complete digestion of cargos. Without AREL1, V_1_B2 is not ubiquitylated, and lysosomes are dissociated from the perinuclear ER. It is also possible that AREL1 deficiency impairs the attachment or retention of V_1_B2 on the lysosomal surface, which also leads to peripherally localized lysosomes with less acidity and compromised degradative capacity. These findings provide new mechanistic explanations for perinuclear lysosomes being more acidic and less mobile compared with the peripheral ones (Johnson et al, 2016). Since Rabconnectin-3 is essential for the assembly and proper function of V-ATPases (Eaton et al, 2024; Einhorn et al, 2012; Jaskolka et al, 2021; Yan et al, 2009), it will be interesting to examine whether Rabconnectin-3 is involved in AREL1-regulated V-ATPase assembly.
We show that UBAC2 via regulating lysosomal positioning determines the subcellular localization of AREL1 (Fig. EV7I). However, why UBAC2 specifically resides in the perinuclear region is unclear. It has been shown that the ER-embedded proteins such as CLIMP63, p180 and KTN1, and cytoskeletons such as microtubules with different modifications and vimentin-containing intermediate filaments help define the perinuclear morphology of the ER and restrain proteins in the perinuclear ER (Cremer et al, 2023; Zheng et al, 2022). Whether these ER-shaping factors define the perinuclear localization of UBAC2 needs to be explored further.
Regarding the physiological consequences of proper lysosomal positioning, it has been shown, in cultured cells, that lysosomes can change their intracellular positioning in response to nutrient availability, with the perinuclear ones being responsible for substrate degradation (Jerabkova-Roda et al, 2024; Korolchuk et al, 2011). The RNF26-mediated perinuclear localization of lysosomes has been shown to promote lysosomal trafficking of activated EGFR and termination of EGF-induced AKT signaling (Cremer et al, 2021), as well as ER reorganization in response to proteotoxic stress (Cremer et al, 2023). In our study, we take a step further and evaluate the functional significance of AREL1-mediated lysosomal positioning at the animal level. Arel1^−/−^ mice display more peripherally localized lysosomes and, strikingly, increased numbers and sizes of lipofuscin granules—suggestive of lysosomal dysfunction and senescence—in Purkinje cells (Figs. 6K–R and EV9F). They show age-dependent loss of Purkinje cells (Fig. 6I–L), where ARE1L1 is highly expressed (Figs. 6H and EV9C,D), and progressive ataxic phenotype (Fig. 6A–F; Movies EV1–4). Consistent with these phenotypes, genes involved in the AREL1−UBAC2−V-ATPase−ZRANB1 axis were highly expressed in Purkinje cells (Fig. EV9D). These results highlight the importance of AREL1-regulated lysosomal distribution and function in highly polarized, postmitotic neurons.
Lysosomal dysfunction has been closely implicated in normal aging and neurodegenerative diseases, including Alzheimer’s disease and Parkinson’s disease, and many of the identified mutations affect the activities of lysosomal enzymes, V-ATPase, and transporters (Lie and Nixon, 2019; Tan and Finkel, 2023; Udayar et al, 2022). Our work identifies an upstream regulator of lysosomal function and suggests that disrupted ER regulation of lysosomal positioning is a previously uncharacterized driving factor for Purkinje cell loss. It will be worth investigating whether aberrant lysosomal positioning underlies neurodegeneration in various pathological conditions.
Methods
Reagents and tools tableReagent/resourceReference or sourceIdentifier or catalog number Experimental models Arel1^floxed/floxed^ *(M. musculus)*Nanjing Biomedical Research Institute of Nanjing UniversityN/ACMV-Cre *(M. musculus)*Shulaibao (Wuhan) Biotechnology Co., Ltd.YDG21032203HEK293T cells *(H. sapiens)*ATCCCRL-3216HeLa cells *(H. sapiens)*ATCCCRM-CCL-2U2OS cells *(H. sapiens)*ATCCHTB-96 Recombinant DNA pLVX-LAMP1-mCherry-TurboID(N terminus, 1-73)-IRES-PuroThis studyN/ApLVX-TurboID(C terminus, 74-320)-EGFP-Sec61β-IRES-PuroThis studyN/ApLVX-TurboID(C terminus, 74-320)-miRFP670-Sec61β-IRES-PuroThis studyN/ApLVX-EGFP-Sec61β-IRES-PuroThis studyN/ApLVX-mCherry-Sec61β-IRES-PuroThis studyN/ApLVX-AREL1(WT)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(ΔIDR, 313-348)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(ΔLCR, 346–360)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(Δ5K)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(Y354A/Y356A)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(Y354F/Y356F)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(Y354W/Y356W)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(1–482)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(C790A)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(1-712)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(1-705)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(1-668)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(1-589)-EGFP-IRES-PuroThis studyN/ApLVX-AREL1(Δ706-711)-EGFP-IRES-PuroThis studyN/ApLVX-mCherry-ZRANB1(WT)-IRES-PuroThis studyN/ApLVX-mCherry-ZRANB1 (C443S)-IRES-PuroThis studyN/ApLVX-UBAC2-mCherry-IRES-PuroThis studyN/ApLVX-UBAC2(WT)-EGFP-IRES-PuroThis studyN/ApLVX-UBAC2(ΔUBA, 304-344)-EGFP-IRES-PuroThis studyN/ApLVX-AUP1-EGFP-IRES-PuroThis studyN/ApLVX-RHBDD3-EGFP-IRES-PuroThis studyN/Ap3×FLAG-CMV-14-ATP6AP1This studyN/Ap3×FLAG-CMV-14-ATP6AP2This studyN/Ap3×FLAG-CMV-14-ATP6V0A1This studyN/Ap3×FLAG-CMV-14-ATP6V0BThis studyN/Ap3×FLAG-CMV-14-ATP6V0CThis studyN/Ap3×FLAG-CMV-14-ATP6V0D1This studyN/Ap3×FLAG-CMV-14-ATP6V1AThis studyN/Ap3×FLAG-CMV-14-ATP6V1B1This studyN/Ap3×FLAG-CMV-14-ATP6V1B2This studyN/Ap3×FLAG-CMV-14-ATP6V1C1This studyN/Ap3×FLAG-CMV-14-ATP6V1DThis studyN/Ap3×FLAG-CMV-14-ATP6V1E1This studyN/Ap3×FLAG-CMV-14-ATP6V1G1This studyN/Ap3×FLAG-CMV-14-ATP6V1HThis studyN/Ap3×FLAG-CMV-14-ATP6V0A1(NTD, 1-388)This studyN/Ap3×FLAG-CMV-14-ATP6V0A1(CTD, 389-837)This studyN/Ap3×FLAG-CMV-14-ATP6V0A2 (NTD, 1-393)This studyN/Ap3×FLAG-CMV-14-ATP6V0A2(CTD, 394-856)This studyN/Ap3×FLAG-CMV-14-ATP6V0A3(NTD, 1-385)This studyN/Ap3×FLAG-CMV-14-ATP6V0A3(CTD, 386-830)This studyN/Ap3×FLAG-CMV-14-ATP6V0A4(NTD, 1-390)This studyN/Ap3×FLAG-CMV-14-ATP6V0A4(CTD, 391-840)This studyN/Apcdna3.0-AREL1(WT)-mycThis studyN/Apcdna3.0-AREL1(Δ706-711)-mycThis studyN/Apcdna3.0-UBAC2(WT)-mycThis studyN/ApcDNA3.0-UBAC2(ΔUBA)-mycThis studyN/ApET-28a(+)-mEGFPThis studyN/ApET-28a(+)-mEGFP-IDRThis studyN/ApET-28a(+)-mEGFP-LCRThis studyN/ApET-28a(+)-mEGFP-LCR(Δ5K)This studyN/ApET-28a(+)-mEGFP-LCR(Y354A/Y356A)This studyN/ApET-28a(+)-mEGFP-LCR(Y354W/Y356W)This studyN/ApET-28a(+)-mEGFP-LCR(Y354F/Y356F)This studyN/ApEF-HA-ubiquitin(WT, K6R, K11R, K27R, K29R, K33R, K48R, K63R, K29, K33)This studyN/ApX330-U6-Chimeric_BB-CBh-hSpCas9Addgene42230pLKO.1 puroAddgene8453pMD2.GAddgene12259psPAX2Addgene12260 Antibodies Anti-alpha TubulinProteintech11224-1-APAnti-ATP6V0A3Proteintech83351-6-RRAnti-ATP6V1B2Proteintech15097-1-APAnti-ATP6V1B2Proteintech68441-1-IgAnti-ATP6V1DProteintech14920-1-APAnti-calbindinCell Signaling Technology13176Anti-calnexinProteintech10427-2-APAnti-calnexinMerckMAB3126Anti-DYKDDDDK tagProteintech20543-1-APAnti-EEA1BD Biosciences610457Anti-HA tagProteintech51064-2-APAnti-LAMP1Developmental Studies Hybridoma Bankclone H4A3Anti-LAMP1Cell Signaling Technology9091Anti-mCherryProteintech26765-1-APAnti-myc tagProteintech16286-1-APcontrol mouse IgGProteintechB900620Anti-AREL1DIA-AN (Wuhan, China)N/AAnti-EGFPHomemadeN/Aperoxidase AffiniPure goat anti-mouse IgG secondary antibodyJackson ImmunoResearch115-035-003peroxidase AffiniPure goat anti-rabbit IgG secondary antibodyJackson ImmunoResearch111-035-144HRP-conjugated goat anti-rabbit secondary antibodyProteintechPR30009Alexa Fluor 488 goat anti-mouse IgGThermo Fisher ScientificA11001Alexa Fluor 488 goat anti-rabbit IgGThermo Fisher ScientificA11008Alexa Fluor 555 donkey anti-mouse IgGThermo Fisher ScientificA31570Alexa Fluor 647 goat anti-rabbit IgGThermo Fisher ScientificA21244 Oligonucleotides and other sequence-based reagents sgRNA targeting sequences human AREL1: ACTATTTATGACTACGTGCG, CTCTCATGTCGAGCTAGCAG, CACGTGCGGCAAGCTCAAAGThis studyN/AshRNA targeting sequences negative control: CCTAAGGTTAAGTCGCCCTCGThis studyN/AshRNA targeting sequence human ANKLE2: ATGTCAGCCAGGATCGCTAAAThis studyN/AshRNA targeting sequence human AREL1: CCGGGAATGGTTTGAGCTAATThis studyN/AshRNA targeting sequence human ATP6AP1: GCATTGAGGATTTCACAGCATThis studyN/AshRNA targeting sequence human ATP6AP2: AGTCTTGACAGTGTTGCAAATThis studyN/AshRNA targeting sequence human ATP6V0A3: CAACTCCTTCAAGATGAAGATThis studyN/AshRNA targeting sequence human ATP6V1B2: GCTGAATTTCTGGCGTACCAAThis studyN/AshRNA targeting sequence human AUP1: GCAGAGCACATGAAGCGACAAThis studyN/AshRNA targeting sequence human ESYT1: GCGTCTCACCACAGTCTTAAAThis studyN/AshRNA targeting sequence human ESYT2: GCTCGCAGAGAAACAAGCTTAThis studyN/AshRNA targeting sequence human MOSPD2: CCCAGATGGTTATTGGAAATTThis studyN/AshRNA targeting sequence human MXRA7: GAGAAGGCTTCTCCTTCAAATThis studyN/AshRNA targeting sequence human RNF26: CCCTTGGAAATTGCTGAAGGAThis studyN/AshRNA targeting sequence human RHBDD3: GCCGTGTCACTGTTGGTTGGAThis studyN/AshRNA targeting sequence human STX18: GACATAAGAGAGGCCATTAAAThis studyN/AshRNA targeting sequence human TMEM9: GCATCTGTCCACCTTATAGAAThis studyN/AshRNA targeting sequence human TMEM214: GAAGGTACAGAAGTCTTTGCAThis studyN/AshRNA targeting sequence human TMEM106B: CCATTATTGGTCCACTTGATAThis studyN/AshRNA targeting sequence human TMX1: GCTGAAAGTAAAGAAGGAACAThis studyN/AshRNA targeting sequence human UBAC2: GCCATTACATTAGCATGTATTThis studyN/AshRNA targeting sequence human UBE2J1: GATGATATACCTACAACATTCThis studyN/AshRNA targeting sequence human USE1: GACGTAAGGAAGAGAACTGGAThis studyN/AshRNA targeting sequence human VAPA: GCGAAATCCATCGGATAGAAAThis studyN/AshRNA targeting sequence human VMA21: GCTCCTGTTCTTCACAGCTTTThis studyN/AshRNA targeting sequence human ZFYVE27: GTGTAACCAGACCTTGAGCAAThis studyN/AshRNA targeting sequence human ZRANB1: CACGCTGGAAAGATTGGGAATThis studyN/A Chemicals, enzymes, and other reagents DAB staining kitProteintechPR30010DAPIThermo Fisher ScientificD1306D-biotinSangon BiotechA600078DQ-OVAThermo Fisher ScientificD12053DTTSangon BiotechA620058DuoLink In Situ Orange Starter KitMerckDUO92102ECL Plus western blotting substrateThermo Fisher Scientific32132FLAG beadsMerckA2220FluorSave mounting mediumMillipore345789FuGENE HDPromegaE2311LysoSensor Green DND-189YEASEN Biotech40767ES50NeutrAvidin agarose resinThermo Fisher Scientific29204NigericinTargetmolT3092Ni-NTA AgaroseQIAGEN30210Oregon Green 488 conjugated dextran-10 kDMolecular ProbesD7170PAS staining kitSolarbioG1281PolyethyleneiminePolysciences23966Protease inhibitor cocktailMilliporeP8340-5mLProtein A/G agarose beadsSmart-LifesciencesSA032005PuromycinBiofroxx1299MG025SiR-LysosomeCytoskeletonCY-SC012TrypsinSangon BiotechA003702 Software DigitalMicrographGatanN/AFijiImageJN/ADigitalMicrographGatanN/AGraphPad PrismPrismN/APerseusPerseus Software Ltd.N/ARstudioPositN/AZEN (black edition)ZEISSN/A Other Standard chowResearch DietsD10001
Ethical statement
Mice were maintained and used in accordance with the guidelines of the Institutional Animal Care and Use Committee of Wuhan University under the protocol number WDSKY0201408.
Mice
Arel1^floxed/floxed^ mice were generated by Nanjing Biomedical Research Institute of Nanjing University using the CRISPR-cas9 system. Cas9 mRNA, single guide RNAs (sgRNAs), and donor were co-injected into zygotes. SgRNAs directed cas9 endonuclease cleavage in intron 3 and intron 8, resulting in loxP sites inserted into intron 3 and intron 8, respectively, by homologous recombination. Arel1^floxed/floxed^ mice were first crossed with CMV-Cre transgenic mice. Heterozygous offsprings were intercrossed to generate Arel1 homozygous knockout mice (Arel1^−/−^) and wild-type littermate controls.
Mice were housed in a specific pathogen-free environment under a 12-h light/12-h dark cycle, with the temperature at 21–23 °C and relative humidity at 50–60%. Mice had ad libitum access to water and standard chow (Research Diets, D10001). For behavioral experiments, mice were subjected to habituation in the behavioral test room for 1 h. Age and gender of mice used for the experiments were stated in the relevant figure legends.
Plasmids
The plasmids were generated using standard molecular cloning techniques. Those expressing various forms of ubiquitin were generated as described previously (Wang et al, 2017).
Cell culture
HEK293T (a human embryonic kidney cell line), HeLa (a human cervical cancer cell line), and U2OS (a human osteosarcoma cell line) cells were obtained from ATCC and grown as a monolayer at 37 °C in 5% CO_2_. No further authentication of the cell lines was performed before use. No test for mycoplasma contamination was performed.
Cells were maintained in DMEM containing 100 units/ml penicillin and 100 μg/ml streptomycin sulfate supplemented with 10% fetal bovine serum.
Generation of AREL1 knockout cells using CRISPR/Cas9
U2OS and HeLa cells were transfected with three pX330 vectors containing gRNAs targeting human AREL1 exons. Cells were plated into 96-well plates in single colonies. Knockout of AREL1 was verified by genomic sequencing and immunoblotting analysis.
Transfection
Transient transfection of HEK293T and U2OS cells was performed using linear polyethyleneimine (Polysciences, 23966) and FuGENE HD (Promega, E2311), respectively, following the manufacturers’ instructions.
Short hairpin RNA (shRNA)-mediated knockdown
Lentiviruses were packaged with pMD2.G and psPAX2 in HEK293T cells, and transduced into U2OS cells. After 48 h, cells were subjected to 2 μg/ml puromycin selection for 5 days to generate AREL1 knockdown stable cell lines.
Subcellular fractionation
U2OS cells were harvested and washed once with ice-cold PBS. Then, cells were lysed with a Dounce homogenizer in ice-cold hypotonic buffer (10 mM HEPES, pH 7.2, 10 mM KCl, 1.5 mM MgCl_2_, 0.1 mM EGTA, and protease inhibitors). The homogenates were centrifuged at 3000× g at 4 °C for 5 min to pellet unbroken cells and nuclei. The supernatants were subjected to centrifugation at 30,000× g at 4 °C for 10 min. The supernatants were cytosolic fractions, and the pellets were membrane fractions. Membrane fractions were washed three times with hypotonic buffer before sample preparation.
For the Na_2_CO_3_ treatment, cell homogenates were treated with 0.1 M Na_2_CO_3_ at 37 °C for 30 min and then subjected to centrifugation at 30,000×g at 4 °C for 10 min to isolate cytosolic and membrane fractions.
For the trypsin treatment, cell homogenates were treated with trypsin (Sangon Biotech, A003702) at 37 °C for 30 min and then subjected to sample preparation.
Proximity labeling using split-TurboID
HEK293T cells and AREL1 knockout U2OS cells were transfected with split-TurboID plasmids using LPEI and FuGENE HD, respectively. After 48 h, D-biotin (Sangon Biotech, A600078) was added to the culture medium to a final concentration of 100 μM in HEK293T or indicated concentrations in AREL1 knockout U2OS cells. After 4-h incubation, cells were washed once with ice-cold PBS and harvested. Then cells were subjected to membrane fractionation as described above.
To pulldown biotinylated membrane proteins, membrane fractions were homogenized in ice-cold lysis buffer (50 mM HEPES, 2.5 mM MgCl_2_, 200 mM KCl, 5% glycerol, 1% Triton X-100 plus protease inhibitors), and centrifuged at 12,000× g for 10 min to remove the non-dissolved parts. The supernatants were incubated with 100 μl high-capacity NeutrAvidin agarose resin (Thermo Scientific, 29204) at 4 °C for 4 h. After washing with the lysis buffer for three times, the biotinylated proteins were eluted by incubating beads with 2× loading buffer plus 2 mM D-biotin and 20 mM DTT for 10 min at 95 °C. The eluents were subjected to SDS–PAGE and mass spectrometry analysis.
Mass spectrometry
Enriched biotinylated proteins were separated by SDS–PAGE and digested by in-gel tryptic digestion. The gel slices were treated with 10 mM DTT and 55 mM iodoacetamide to reduce the disulfide bond and alkylate the resulting thiol group. Trypsin was added at a final concentration of 10 ng/μl for overnight digestion at 37 °C. The phosphopeptides were enriched by using a homemade TiO_2_ microcolumn. In brief, the digested peptides were loaded in 80% acetonitrile (ACN), 5% trifluoroacetic acid (TFA), 1 M glycolic acid, washed with 80% ACN, 1% TFA, and 10% ACN, 0.1% TFA, eluted with 2 M NH_3_·H_2_O, and desalted in a R3 microcolumn. LC–MS/MS was performed using EASY-nLC 1000 system interfaced to Q Exactive HF (Thermo Fisher Scientific).
To identify lysosomal proteins that interact with AREL1, the database search was performed using Maxquant software, and the quantified protein list was further analyzed using Perseus software for statistical analysis. The endogenously biotinylated proteins, namely pyruvate carboxylase (PC), propionyl-CoA carboxylase subunit a (PCCA), propionyl-CoA carboxylase subunit b (PCCB), methylcrotonoyl-CoA carboxylase subunit a (MCCC1), and methylcrotonoyl-CoA carboxylase beta chain (MCCC2), were served as the internal references for normalization. A two-sample t test was conducted to calculate P values of changed proteins using the implemented function in Perseus.
Immunostaining and confocal microscopy
Cells were grown on glass coverslips for 48 h, washed with PBS, and fixed with 4% paraformaldehyde (PFA) for 10 min at room temperature (RT). Cells were then permeabilized with liquid nitrogen and incubated with primary and secondary antibodies diluted in PBS containing 1% bovine serum albumin for 1 h at RT. After washing 3 times with PBS, coverslips were mounted on glass slides with FluorSave mounting medium (Millipore, 345789).
For endogenous AREL1 staining, cells were first fixed with PFA and then incubated with citrate buffer (pH 6.0) at 95 °C for 30 min followed by permeabilization with liquid nitrogen. Then, cells were incubated with the anti-AREL1 (20 μg/ml) overnight, followed by secondary antibody staining for 1 h at RT.
For endogenous ATP6V1D staining, cells were incubated in 25 μg/ml digitonin for 10 min on ice to remove cytosolic proteins before fixation. Then, fixed cells were permeabilized with liquid nitrogen and incubated with the anti-ATP6V1D overnight, followed by secondary antibody staining for 1 h at RT. Images were taken using a Leica SP8 LIGHTNING confocal system equipped with a ×63 objective lens.
Proximity ligation assay (PLA) and quantification analysis
PLA was conducted using a mouse primary antibody against endogenous ER protein calnexin and rabbit primary antibody against endogenous lysosomal protein LAMP1, followed by mouse and rabbit PLA probes from the DuoLink In Situ Orange Starter Kit Mouse/Rabbit (Sigma-Aldrich, DUO92102), following the manufacturer’s instructions. Images were taken using a Leica SP8 LIGHTNING confocal microscope with the 405 nm (for DAPI) and 561 nm (for PLA signal) channels.
PLA signal quantification was performed using Fiji as previously described (Saric et al, 2021). The 561 nm-channel images were imported to Fiji and converted to binary images, and a threshold was applied to eliminate background using the Image>Adjust>Threshold command. The Watershed feature was applied to separate any signal dots apparently in touch using the Process>Binary>Watershed command. Dots that were at least 5 pixels in size were counted using the Analyze>Analyze Particles command, and the total number of dots per cell (counterstained by DAPI) was measured and recorded.
Immunoblotting
Equal amounts of total proteins were resolved by SDS–PAGE and transferred to PVDF membranes. Membranes were blocked with 5% skim milk in tris-buffered saline (TBS) containing 0.075% Tween-20 (TBST), probed with indicated primary antibodies overnight at 4 °C. After washing in TBST three times, membranes were incubated with horseradish peroxidase-conjugated secondary antibodies diluted in TBST supplemented with 5% skim milk for 1 h at RT, followed by at least three washes with TBST. Chemiluminescent signals were detected using Pierce ECL Plus western blotting substrate (Thermo Scientific, 32132).
For detecting biotinylated proteins, PVDF membranes were incubated with streptavidin-HRP diluted in TBST supplemented 3% bovine serum albumin at RT for 1 h. Membranes were then washed three times with TBST, followed by chemiluminescent detection.
Protein expression and purification
pET-28a(+) plasmids encoding mEGFP fusion proteins were transformed into the Escherichia coli strain BL21(DE3), and expression was induced by 0.5 mM IPTG in LB at 16 °C overnight. Cells were harvested for sonication in PBS with protease inhibitors, followed by centrifugation at 5000× g for 5 min. The supernatants were incubated with Ni-NTA Agarose (QIAGEN, 30210) at for 4 h 4 °C, washed three times with binding buffer (20 mM sodium phosphate, 0.5 M NaCl, 40 mM imidazole, pH 7.4), and eluted with 500 mM imidazole (20 mM sodium phosphate, 0.5 M NaCl, pH 7.4). The eluents were concentrated by ultra-centrifugation and stored at −30 °C.
In vitro phase separation
Purified recombinant proteins were diluted in the buffer composed of 50 mM HEPES (pH 7.4) and 50 mM NaCl at RT. Mixed reactions were pipetted to 96-well microscopy plates and imaged by a Leica SP8 LIGHTNING confocal system.
Immunoprecipitation (IP)
Cells were collected and lysed in IP buffer (50 mM Tris-HCl, pH 7.6, 150 mM NaCl, 5 mM EDTA, 0.5% NP40, and protease inhibitors). After centrifugation at 12,000× g for 10 min, supernatants were collected and immunoprecipitated with 20 μl anti-FLAG beads (Sigma-Aldrich, A2220) at 4 °C for 4 h. Beads were centrifuged at 1000× g for 5 min and washed three times with 0.5% NP40 IP buffer. Beads were then incubated with the loading buffer to elute pulled-down proteins. Aliquots of eluents were subjected to immunoblotting analysis.
Ubiquitylation
For ubiquitylation assay in transfected HEK293T cells, cells were harvested 48 h post transfection and lysed in the ubiquitylation buffer (1% Nonidet P-40, 1% deoxycholate, 5 mM EDTA, 5 mM EGTA, protease inhibitors, and 10 mM N-ethylmaleimide in PBS). Lysates were immunoprecipitated with 20 μl anti-FLAG beads at 4 °C for 4 h. Then, the beads were washed three times with the ubiquitylation buffer, and incubated with the loading buffer to elute pulled down proteins. Aliquots were subjected to immunoblotting analysis.
For ubiquitylation assay of endogenous V_1_B2 in whole-cell lysate, U2OS cells (ten 15-cm dishes per group) were harvested and lysed in the ubiquitylation buffer (1% Nonidet P-40, 1% deoxycholate, 5 mM EDTA, 5 mM EGTA, protease inhibitors, 10 mM N-ethylmaleimide, 10 μM PR-619 in PBS). For ubiquitylation assay of endogenous V_1_B2 in cytosol and membrane fractions, U2OS cells (ten 15-cm dishes per group) were harvested and broken with Dounce homogenizer in ice-cold hypotonic buffer (10 mM HEPES, pH 7.2, 10 mM KCl, 1.5 mM MgCl_2_, 0.1 mM EGTA, protease inhibitors, 10 mM N-ethylmaleimide, and 10 μM PR-619 in PBS), followed by subcellular fractionation as described above. Isolated membrane fractions were then lysed with RIPA buffer containing protease inhibitors and deubiquitinase inhibitors.
Lysates (1 mg total protein), cytosolic fractions (750 μg total proteins), or lysed membrane fractions (250 μg total proteins) were incubated with 2 μg anti-ATP6V1B2 antibody or control mouse IgG and 100 μl Protein A/G agarose beads (Smart-Lifesciences, SA032005) at 4 °C for 12 h. Then, beads were washed 3 times with the corresponding buffer, and incubated with the loading buffer to elute pulled down proteins. Aliquots were subjected to immunoblotting analysis.
Measurement of lysosomal pH
Lysosomal pH measurement with a ratiometric pH-sensitive dye Oregon Green 488 conjugated dextran-10 kD (Molecular Probes, D7170) was performed as described previously (Hu et al, 2022). In brief, cells grown on culture plates with glass-bottom were loaded with 150 μg/mL Oregon Green 488 Dextran overnight and chased in medium without dye for 3 h before imaging. Cells were washed with Ringer’s buffer (155 mM NaCl, 5 mM KCl, 2 mM CaCl_2_, 1 mM MgCl_2_, 2 mM NaH_2_PO_4_, 10 mM HEPES, and 10 mM glucose, pH 7.4) and imaged using a Zeiss LSM 880 confocal microscope equipped with a ×60 objective lens and a temperature-controlled stage. The fluorescence emission (530 ± 20 nm) excited at 440-nm and 490-nm wavelengths were acquired with a cooled digital CCD camera.
To generate the pH calibration curve, cells were incubated with pH standard buffers (pH 3.5, 4.5, 5.5, 6.5, 7.5) supplemented with 10 μM nigericin (Targetmol, T3092). The fluorescence intensities excited at 490 nm and 440 nm in response to varied pH were quantified by Fiji. The ratios of fluorescence intensities excited at 490 nm and 440 nm in response to varied pH were fitted to a sigmoidal equation. The lysosomal pH values were then calculated with the resulting intensity ratios (490 nm/440 nm) of individual lysosomes based on the calibration curve.
LysoSensor green staining
U2OS cells were cultured in a 24-well glass-bottom dish. After 24 h, cells were incubated with 1 µM LysoSensor Green DND-189 (YEASEN Biotech, 40767ES50) diluted in culture medium at 37 °C for 30 min. Live cells were then washed twice and kept in culture medium without phenol red for imaging. Images were taken using a Leica SP8 LIGHTNING confocal system equipped with a 63× objective lens and a temperature-controlled stage. The fluorescence intensities of LysoSensor green were quantified by Fiji.
SiR-Lysosome staining
U2OS cells were cultured in a 24-well glass-bottom dish. After 24 h, live cells were incubated with 0.5 μM SiR-Lysosome (Cytoskeleton, CY-SC012) diluted in culture medium at 37 °C for 1 h. Live cells were then washed twice and kept in culture medium without phenol red for imaging. Images were taken using a Leica SP8 LIGHTNING confocal system equipped with a 63× objective lens and a temperature-controlled stage. The fluorescence intensities of SiR-Lysosome were quantified by Fiji.
Endocytosis of DQ-OVA
U2OS cells were grown on glass coverslips. After 24 h, cells were incubated with 10 µg/ml DQ-OVA (Invitrogen, D12053) at 37 °C for 3 h. Cells were washed three times with PBS, fixed with 4% PFA, and mounted for imaging. Images were taken using a Leica SP8 LIGHTNING confocal system equipped with a ×63 objective lens. The fluorescence intensities of DQ-OVA were quantified by Fiji.
FRAP
Indicated cells grown on glass-bottom dishes were transfected with the plasmid expressing V_1_E1-mCherry. After 48 h, live cells were washed twice, kept in culture medium without phenol red at a temperature-controlled stage (37 °C), and imaged using a Zeiss LSM 880 confocal microscope equipped with a ×60 objective lens.
Perinuclear lysosomes due to relatively immobility were selected for region of interest (ROI) imaging. Photobleaching of the mCherry in ROI was achieved using 20 iterations of the 561 nm laser line at 100% intensity, and post-bleached images were collected for 180 s with 5-s intervals to determine recovery of mCherry fluorescence in the bleached region. Quantitation of the time-lapse fluorescence intensity of ROI and the calculation of half-time of fluorescence recovery (τ1/2) were performed using Zen software (Zen Black 2.3 SP1).
Footprint analysis
The footprint analysis was performed as previously described (Wertman et al, 2019). In brief, mice were acclimated to the behavior rooms for 1 h prior to testing. The fore and hind paws were painted with orange and blue non-toxic water-based paints, respectively. Mice were allowed to walk in a straight line in a narrow tunnel (60 cm × 10 cm × 10 cm) on white paper, with a darkened cage used as the bait at the end of the tunnel. Mice were repeatedly tested three times. In each test, three hind paw steps from the middle portion were measured for stride length, stride width, and stance.
Hindlimb clasping
Mice were lifted clear of all surrounding objects by the tail. The positions of their hindlimbs were observed for 1 min. Hindlimb positions were scored manually as described previously (Guyenet et al, 2010; Petkovic et al, 2020): 0, the hindlimbs were consistently splayed outward, away from the abdomen; (1) one hindlimb retracted toward the abdomen for more than 50% of the time suspended; (2) both hindlimbs were partially retracted toward the abdomen for more than 50% of the time suspended; (3) both hindlimbs were entirely retracted and touching the abdomen for more than 50% of the time suspended.
Rotarod test
The rotarod test was performed as previously described (de Haas et al, 2016; Hayashi-Takagi et al, 2015; Kim et al, 2019; Shiotsuki et al, 2010). In brief, the mice were trained 3 days before test, mice were trained at 8 rpm for three times per day with a 5 min resting period between each trial. On day 4, mice were placed on an accelerating rotarod cylinder, and the latency time of the animals was measured. The speed was slowly increased from 4 to 40 rpm within 5 min. A trial ended if the animal fell off the rods.
Immunostaining of cerebellar sections
Mice were anesthetized and transcardially perfused with PBS followed by 4% PFA in PBS. Brains were post-fixed with 4% PFA for 24 h and then incubated in 30% sucrose/PBS till sinking. Brains were embedded in OCT compound and snap frozen with liquid nitrogen-cooled isopentane. OCT blocks were then sectioned in 30 μm-thick sections using a Leica Cryostat and kept at −80 °C.
For cryoimmunolabeling, cryosections were thawed at RT and then washed three times with PBS for 2 min each. Sections were permeabilized with PBS containing 0.1% Triton X-100 for 10 min, followed by three times washing with PBS for 2 min. Sections were blocked in 10% normal goat serum in PBS for 1 h at RT, and then incubated at 4 °C with the anti-calbindin antibody. After overnight incubation, sections were washed three times with PBS for 5 min and incubated with goat anti-rabbit secondary antibody at RT for 1 h. After three times washing with PBS for 5 min, sections were mounted with FluorSave mounting medium.
For lipofuscin autofluorescence detection, cryosections were first immunostained with anti-calbindin primary antibody and Alexa Fluor 555 secondary antibody.
For immunohistochemical staining, cryosections were thawed at RT and then washed 3 times with PBS for 2 min. Sections were quenched with 3% hydrogen peroxide in PBS for 15 min at RT. Then antigen retrieval was done by incubating sections with citrate buffer (pH 6.0) for 20 min at 95 °C. Then, slides were incubated with anti-AREL1 antibody overnight at 4 °C. After washing, the HRP-conjugated secondary antibody was added and incubated for 30 min at RT. After washing, the DAB staining was conducted with a DAB staining kit (Proteintech, PR30010). After washing three times with PBS, sections were dehydrated by increasing concentrations of ethanol and xylene, and then mounted with neutral balsam mounting medium.
Transmission electron microscopy
Mice were anesthetized and transcardially perfused with 4% PFA and 2% glutaraldehyde in PBS. After dissection, brains were post-fixed in the same fixative overnight. Cerebellar samples were stained with 1% osmium tetroxide, dehydrated in increasing concentrations of acetone from 30% to 100% and infiltrated and embedded in SPI-PON812 resin (SPI-CHEM). Samples were then sectioned at 2-µm thickness and stained with 1% toluidine blue for light microscopic assessment. The 70-nm ultrathin sections were cut using an ultra-microtome (Leica, UC7), followed by staining with uranyl acetate and lead citrate. Sections were mounted on copper grids and viewed with a transmission electron microscope (JEOL, JEM-1400Plus). Electron micrographs were captured by a Gatan digital camera (Gatan, Rio9) and its application software (Gatan Digital Micrograph 3.0 software).
Periodic acid Schiff (PAS) staining
PAS staining was conducted following the manufacturer’s instructions of the PAS staining kit (Solarbio, G1281). The brain sections were dehydrated by increasing concentrations of ethanol and xylene, and then mounted with neutral balsam mounting medium.
Quantification of confocal images
Cells were selected and analyzed as previously described (Saric et al, 2021; Williamson et al, 2022). Those with flat and round shapes were selected, and the narrow ones were excluded. In the cases of overexpression, cells expressing medium and similar levels of indicated proteins were chosen for analysis, and those showing high expression levels were excluded to avoid artifacts.
For vesicle distribution measurement, a total of 10 z-stacks were taken per cell from the top to the bottom and then merged. Images were exported with scale bars using Leica LAS X software. Images were then imported to Fiji, and scale bars were used to set scale using the Analyze>Set Scale command. Straight lines were manually drawn from vesicles to the nuclear envelope, and the distances of straight lines were calculated using the Analyze>Measure command and recorded.
For fluorescence intensity quantification of LysoSensor green, SiR-lysosome, and DQ-OVA, cell borders were manually outlined using the Freehand selections tool in Fiji, and total fluorescence intensities within the borders were quantified using the Analyze>Measure command and recorded. Relative fluorescence intensities were calculated by comparing experimental groups with control groups.
For colocalization quantification, images were exported in separate color channels using Leica LAS X software. Images were then imported to Fiji. Cell borders were manually outlined using the Freehand selections tool in one channel, and the borders were applied to the other channel using the Edit>Selection>Restore Selection command. The Pearson’s correlation coefficient of the two channels per cell was quantified by the Coloc 2 plugin and recorded.
For lipofuscin autofluorescence quantification, a total of 10 z-stacks were taken per cell from the top to the bottom and then merged. The calbindin signals (under the 555-nm channel) were used to draw the borders of somas of Purkinje cells, and the of lipofuscin autofluorescence (under the 488-nm channel) was quantified by Fiji.
Statistical analysis
All statistical analyses were performed using GraphPad Prism 10. Data were first analyzed for normality using Shapiro–Wilk test. The normally distributed data are presented as mean ± SD, and unpaired two-tailed Student’s t test was used for comparison between two groups. The non-normally distributed data are presented as mean ± SD (in superplots) or median with interquartile range (in box plots), with Mann–Whitney U test used for comparison between two groups.
The effects of the independent variable (genotype) on two dependent variables (distance and pH) were analyzed using multivariate analysis of variance (MANOVA). The analysis was performed in Rstudio (R version 4.4.1), and multivariate significance was assessed using Pillai’s trace statistic.
Sample sizes are stated in the relevant figure legends, and the exact P values (unless <0.0001) are shown in the figures.
Supplementary information
Peer Review File Dataset EV1 Dataset EV2 Movie EV1 Movie EV2 Movie EV3 Movie EV4 Source data Fig. 1 Source data Fig. 2 Source data Fig. 3 Source data Fig. 4 Source data Fig. 5 Source data Fig. 6 Figure EV1 Source Data Figure EV2 Source Data Figure EV3 Source Data Figure EV4 Source Data Figure EV5 Source Data Figure EV6 Source Data Figure EV7 Source Data Figure EV8 Source Data Figure EV9 Source Data Expanded View Figures
The reference list from the paper itself. Each links out to its DOI / PubMed record.