Mito-TEMPO improves survival rates in npc1-knockout zebrafish by reducing oxidative stress and enhancing mitophagy via Sod2

Abstract

Niemann-Pick type C (NPC) disease is a lysosomal storage disorder primarily caused by mutations in the NPC1 gene. Most patients present with early-life symptoms including hepatosplenomegaly and digestive system impairment, followed by progressive neurodegeneration. However, effective therapeutic approaches to improve survival in NPC disease remain limited. In this study, using an npc1-knockout (NPC1-KO) zebrafish model established in our laboratory, our team suggests that npc1 deficiency appears to correlate with marked down-regulation of superoxide dismutase 2 (Sod2) expression, concurrent with excessive oxidative stress (OS), mitochondrial dysfunction, and defective mitophagy. Treatment with Mito-TEMPO, a mitochondria-targeted antioxidant acting on SOD, increased survival rates and ameliorated cholesterol accumulation and liver function impairment in early-stage NPC1-KO zebrafish. The underlying mechanism may involve attenuation of OS and promotion of PINK1/Parkin-dependent mitophagic flux through SOD2 enhancement. Our findings support Mito-TEMPO as a potential therapeutic agent and SOD2 as a possible target for NPC disease.

INTRODUCTION

Niemann-Pick type C (NPC) disease is a rare lysosomal storage disorder caused by mutations in the NPC1 (95%) or NPC2 (5%) gene that disrupt intracellular cholesterol transport (1). This defect results in progressive accumulation of unesterified cholesterol and sphingolipids within lysosomal compartments (1, 2). Patients with NPC exhibit a broad range of symptoms, including neonatal jaundice, hepatosplenomegaly, ataxia, tremors, seizures, and learning difficulties. Infantile-onset cases are characterized by severe hepatic dysfunction (cholestatic jaundice and cirrhosis) and digestive system impairment, frequently culminating in premature death before 20 years of age (1, 3).

Current therapeutic strategies primarily aim to manage symptoms rather than provide curative benefits. Zavesca (Miglustat), an approved glycosphingolipid synthesis inhibitor, slows disease progression but has limited effects on systemic manifestations, and some patients experience adverse events such as upper respiratory tract infections, diarrhea, and weight loss (4). Hydroxypropyl-β-cyclodextrin (β-CD), a cholesterol-chelating agent effectively redistributing lysosomal cholesterol, is demonstrated to take good therapeutic effects in animal models but remains in early-phase clinical evaluation (5, 6). Vorinostat, a histone deacetylase inhibitor, reverses NPC1-associated liver dysfunction in treated mice yet fails to improve overall disease progression or survival (7, 8). These findings underscore the urgent need for unexplored therapeutics capable of modifying early disease progression in NPC disease.

In NPC1-deficient cells, elevated mitochondrial cholesterol induces mitochondrial dysfunction and disrupts the electron transport chain (2, 9, 10), promoting excessive reactive oxygen species (ROS) generation (11). Elevated ROS further enhances lipid synthesis and can trigger apoptosis. Increased ROS levels have been observed in multiple NPC1 models, including rats treated with the NPC1 inhibitor U18666A, Npc1-mutant or Npc1-knockout (KO) BALB/c mice, and NPC1 patient-derived fibroblasts (12–15). The superoxide dismutase (SOD) enzyme family (SOD1, SOD2, and SOD3) plays a critical role in oxidative stress (OS) defense by effectively scavenging ROS (16). Alterations in SOD activity and glutathione transferase expression have been reported in cerebrospinal fluid from patients with NPC1 (2, 17). A previous study revealed that low concentrations of allopregnanolone reduce ROS and lipid peroxidation in NPC1-deficient fibroblasts (14). Furthermore, antioxidant modulation promoting SOD2 in human adipose-derived mesenchymal stem cells alleviates pathological conditions associated with intervertebral disc degeneration (18). Therefore, enhancing SOD2 expression to suppress OS may represent a promising therapeutic approach for NPC disease.

Beyond lysosomal damage from cholesterol accumulation, NPC1-deficient cells exhibit mitochondrial structural and functional abnormalities (19), leading to impaired autophagy (including mitophagy). Autophagy is a crucial process for removing cytoplasmic macromolecules and damaged organelles such as mitochondria via lysosomal degradation (20, 21). Defective autophagy and mitophagy have been described in NPC1-deficient fibroblasts, mice, and other NPC1-deficient induced neurodegenerative disease models, where impaired fusion of autophagosomes and lysosomes causes accumulation of undegraded material and exacerbates cellular stress (22–25). A recent study showed that autophagic flux in NPC1-deficient cells can be restored by inducers such as rapamycin and Torin 1 (26). Thus, promoting autophagy (particularly mitophagy) may offer another viable therapeutic strategy for NPC disease.

Zebrafish serve as an effective model for studying rare diseases such as spinocerebellar ataxia and other genetic disorders, offering rapid reproduction and large offspring numbers (27–29). Given the high evolutionary conservation of the NPC1 protein and functional similarity between zebrafish and mammalian counterparts (29, 30), our team previously established a CRISPR/Cas9-mediated npc1-KO (NPC1-KO) zebrafish model that recapitulates the pathological characteristics of human NPC disease (31). Building on this model, we identified that Mito-TEMPO, a mitochondria-targeted antioxidant that enhances SOD2 expression, markedly improves survival and mitigates hepatic and intestinal damage in early-stage NPC1-KO zebrafish. The therapeutic mechanism likely involves attenuation of OS and activation of mitophagy.

RESULTS

Elevated OS and reduced Sod2 expression in NPC1-KO zebrafish

We previously generated an NPC1-KO zebrafish model by deleting seven base pairs from the npc1 gene, which faithfully recapitulates the cardinal pathological features of human NPC disease (31) (Fig. 1, A and B). In this study, proteomic analysis was performed between NPC1-KO and wild-type (WT) zebrafish at 1 month postfertilization (mpf). The results (Fig. 1, C and D) revealed that several mitochondrial (proteins related to Timm9, Cox5Aa, Sdhb, Bpnt1, etc.) and lysosomal (protein related to Scarb2B) proteins were markedly down-regulated in NPC1-KO zebrafish. We also found that the OS levels increased (proteins related to Gpx9, Txnrd3, Aoc2, Pxdn, etc.) and the autophagy function was blocked (proteins related to Park7, Cisd1, Tomm70a, Anxa6, etc.); of particular interest was the reduction in Sod2, a key regulator of ROS (32), indicating its potential as a therapy target for NPC disease.

Fig. 1. Establishment of NPC1-KO in zebrafish and proteomic results; Mito-TEMPO molecular docking results and zebrafish embryo toxicity test results.

(A) Identification of the effect of the npc1 genotype on NPC1-KO zebrafish establishment. bp, base pairs. (B) Fluorescence microscopy of Npc1 in zebrafish. Scale bars, 100 μm. (C) Volcano plot showing the clustering of differential proteins in WT and NPC1-KO zebrafish. FC, fold change. (D) Heatmap showing the clustering of differential proteins in WT and NPC1-KO zebrafish. n = 3 independent biological repeats. Proteins related to mitochondria, lysosomes, autophagy, lipid metabolism, OS, and apoptosis were chosen. (E) Embryonic development of WT zebrafish exposed to Mito-TEMPO and β-CD. Scale bars, 500 μm. h, hours. (F) Statistical analysis of heartbeats was performed to evaluate the toxicity of Mito-TEMPO and β-CD. n = 30 independent biological repeats. (G and H) Results of the molecular docking of Mito-TEMPO with SOD2 from different species. All data are presented as the means ± SEM. Hypergeometric test (D) and two-tailed Student’s t test (F); n.s., no significant difference compared with the control group (F).

Molecular docking between Mito-TEMPO and SOD2

Mito-TEMPO, a mitochondrial superoxide scavenger, has been reported to reduce ROS levels by modulating the expression of the SOD family (33). To further investigate the interaction between Mito-TEMPO and SOD2, molecular docking analysis was performed using AutoDock Vina 2.1.6. The results showed that Mito-TEMPO could bind to both zebrafish Sod2 (UniProt ID: Q6P980) and human SOD2 [Protein Data Bank (PDB) ID: 9BWQ] with strong binding affinities (docking scores: −7.2 and −7.0 kcal/mol, respectively). These interactions were mediated by hydrogen bonding, van der Waals forces, and π-bond interactions (Fig. 1, G and H). A binding score lower than −7.0 kcal/mol generally indicates high binding affinity (34). Our findings suggest that Mito-TEMPO may directly interact with SOD2. On the basis of these results, we selected Mito-TEMPO as a potential therapeutic agent for NPC disease.

Mito-TEMPO improves the survival rate and reduces the cholesterol and ROS levels in NPC1-KO zebrafish

In this study, we selected β-CD as a positive control drug in the in vivo experiments as it has been reported in clinical trials to alleviate lysosomal cholesterol accumulation in NPC disease (7, 35). First, an acute toxicity test was performed on zebrafish embryos using both drugs. The results indicated that 100 μM Mito-TEMPO (Mito 100) and 2.5 mM β-CD did not cause significant developmental or cardiac toxicity in zebrafish embryos (Fig. 1, E and F, and fig. S1).

Subsequently, we monitored the survival of offspring from npc1 heterozygote self-crosses following drug administration (Fig. 2A). Survival data were recorded every other day, and genotyping was performed immediately after death. The 60-day survival analysis revealed that NPC1-KO zebrafish exhibited an obvious survival decline within 30 days posthatching compared with WT littermates. Treatment with either Mito-TEMPO or β-CD significantly improved the survival rate of NPC1-KO zebrafish to a comparable extent (Fig. 2B).

Fig. 2. Mito-TEMPO therapeutic effect on NPC1-KO zebrafish.

(A) Flowchart showing Mito-TEMPO administration in zebrafish. (B) Statistical analysis of zebrafish survival rate. The life cycle was recorded from birth to 60 days, including the WT group (n = 50), NPC1-KO group (n = 43), β-CD group (n = 41), and Mito 100 group (n = 42). (C) Oil Red O staining of zebrafish showing lipid droplet accumulation mainly in the abdominal cavity, liver, and intestine of NPC1-KO zebrafish. Scale bars, 150 and 500 μm. (D) Filipin III staining of cholesterol in zebrafish, showing cholesterol accumulation primarily in the abdominal cavity and head of NPC1-KO zebrafish. Scale bars, 150 and 500 μm. (E) ROS staining in zebrafish, captured in the abdominal cavity near the liver. Scale bars, 25 μm. (F to K) Statistical results of AST, ALT, triglyceride, TC, ROS, and MDA levels measured using assay kits. n = 4 independent biological repeats. Data are presented as the mean ± 95% confidence interval (B) or the means ± SEM [(F) to (K)]. Gehan-Breslow-Wilcoxon test (B) and two-tailed Student’s t test [(F) to (K)]; n.s., not significant; *P < 0.05; **P < 0.01; ***P < 0.001 compared with the NPC1-KO group.

We further examined liver enzyme activities and lipid/cholesterol accumulation. Oil Red O and Filipin III staining revealed pronounced lipid and cholesterol accumulation in liver and intestine of NPC1-KO zebrafish compared with WT littermates (Fig. 2, C and D). Consistently, the levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), triglyceride, and cholesterol were significantly elevated at 10 days postfertilization (dpf) and 1 mpf (Fig. 2, F to I). In contrast, both Mito 100 and β-CD treatment notably reduced AST, ALT, triglyceride, and cholesterol levels, supporting favorable therapeutic efficacy for both agents on NPC1-KO zebrafish (Fig. 2, C, D, and F to I).

We also found that the overall ROS and malondialdehyde (MDA) levels were significantly elevated in NPC1-KO zebrafish compared with WT littermates but were notably reduced following Mito 100 treatment (Fig. 2, E, J, and K). Given the strong docking affinity observed between Mito-TEMPO and zebrafish Sod2, we next examined Sod2 expression. Whole-mount immunofluorescence and Western blot analyses showed a marked reduction of Sod2 in NPC1-KO zebrafish at 10 dpf and 1 mpf, whereas Mito 100 treatment significantly restored Sod2 expression at both time points (Fig. 3, A, C, and E). These results suggest that Mito 100 exerts its antioxidant effect in NPC1-KO zebrafish by regulating Sod2-mediated OS homeostasis.

Fig. 3. Mito-TEMPO effects on Sod2 protein and mitophagy in NPC1-KO zebrafish.

(A) Fluorescence microscopy of Sod2 expression in zebrafish. Scale bars, 150 and 500 μm. (B) Confocal microscopy of Park2 and Lc3b in zebrafish cells, focusing on the abdominal cavity near the liver. Scale bars, 5 μm. (C) Immunoblot analysis of Sod2, Lc3, and Park2 proteins in zebrafish. (D) Transmission electron microscopy of zebrafish cells near the intestine. Label 1: normal mitochondria; label 2: mitophagy process; label 3: lipid droplets; label 4: swollen mitochondria. The captured mitophagy process primarily represents the stage where lysosomes fuse with mitochondrial vesicles or mitochondrial vesicles enclose lysosomes. Scale bars, 1 and 0.5 μm. (E to G) Quantitative immunoblot results for Sod2, Lc3, and Park2 proteins in zebrafish. n = 3 independent biological repeats. All data are presented as the means ± SEM. Two-tailed Student’s t test; n.s., not significant; *P < 0.05; **P < 0.01 compared with the NPC1-KO group.

Mito-TEMPO promotes mitophagy in NPC1-KO zebrafish

Mitophagy, a selective form of autophagy, plays an essential role in maintaining mitochondrial homeostasis (36, 37). Previous studies reported a marked reduction of autophagy in NPC1-deficient fibroblasts (38), yet the role of mitophagy in NPC1 deficiency remains unclear. Transmission electron microscopy revealed both normal and damaged mitochondria surrounded by double-membrane vesicles in WT littermates after 24 hours of fasting. In contrast, NPC1-KO zebrafish at 10 dpf and 1 mpf exhibited swollen mitochondria and numerous lipid droplets, indicating mitochondrial injury and loss of mitophagy. Treatment with Mito 100 reversed these alterations compared with the NPC1-KO zebrafish (Fig. 3D).

We next examined the expression of representative mitophagy-related proteins, including Lc3 and Park2 (encoding Parkin). Immunofluorescence staining and Western blot analyses revealed that Lc3b expression and Lc3b puncta were markedly increased, whereas Park2 expression was significantly decreased in NPC1-KO zebrafish at 10 dpf and 1 mpf. Treatment with Mito 100 effectively reduced Lc3b expression and restored Park2 levels at both time points (Fig. 3, B, C, F, and G), suggesting that Mito-TEMPO enhances mitophagy ability in NPC1-KO zebrafish.

Mito-TEMPO effects on HepG2 cells induced by U18666A

U18666A, an NPC1 protein inhibitor, is widely used to induce endo/lysosomal cholesterol accumulation, mimicking NPC disease (39). Thus, we established an in vitro NPC1-deficient model (NPC1 model) by treating HepG2 cells with U18666A (Fig. 4A). Before functional assays, cytotoxicity assays confirmed that Mito-TEMPO exhibited no significant toxicity to HepG2 cells at concentrations no more than 100 μM throughout the experimental period (Fig. 4B).

Fig. 4. Mito-TEMPO effects on cholesterol, ROS, and mitochondrial membrane potential in model cells.

(A) Flowchart of Mito-TEMPO administration on HepG2 cells. U18666A (3 μg/ml) was added for 24 hours to modify the NPC disease model in vitro. (B) Results of the cytotoxicity test of Mito-TEMPO administration. DMSO (2.5%) was used as the positive drug for toxicity. n = 6 independent biological repeats. (C) Results of Filipin III staining for cholesterol in HepG2 cells. Scale bars, 150 μm. (D) Results of ROS staining in HepG2 cells. Scale bars, 150 μm. (E) Results of JC-1 staining for mitochondrial membrane potential in HepG2 cells. Green panel indicates JC-1 monomer (JC-1-M), and red panel indicates JC-1 aggregate (JC-1-A). n = 6 independent biological repeats. Scale bars, 150 μm. (F) Statistical results of cholesterol and cholesterol ester by the cholesterol kit in HepG2 cells. n = 4 independent biological repeats. (G) Statistical results of ROS by the ROS kit in HepG2 cells. n = 5 independent biological repeats. (H) Statistical results of JC-1 mean fluorescence intensity (MFI) by the JC-1 kit. n = 4 independent biological repeats. All data are presented as the means ± SEM. Two-tailed Student’s t test [(B), (F), (G), and (H)]; n.s., not significant; *P < 0.05; **P < 0.01; ***P < 0.001 compared with the model group.

To validate model establishment, we examined cholesterol accumulation and apoptosis following 24 hours of U18666A (3 μg/ml) treatment. Filipin III staining and cholesterol assays revealed substantial cholesterol accumulation in the model group (Fig. 4, C and F). JC-1 staining showed enhanced green fluorescence and reduced red fluorescence, indicative of mitochondrial depolarization (Fig. 4, E and H), whereas flow cytometry analysis revealed an increased proportion of late apoptotic cells (Fig. 5, A and B), indicating mitochondrial and cellular damage (40). Subsequent treatment with Mito-TEMPO significantly reduced intracellular cholesterol levels, mitochondrial impairment (Fig. 4, C, E, F, and H), and apoptosis (Fig. 5, A and B) in a dose-dependent manner (12.5, 25, and 50 μM), suggesting that Mito-TEMPO exerts similar therapeutic effects in the NPC1 in vitro model as observed in vivo.

Fig. 5. Mito-TEMPO effects on apoptosis and SOD2 protein in HepG2 cells analyzed by flow cytometry.

(A) Flow cytometry dot plots of HepG2 cell apoptosis detected by annexin V/PI staining. [Annexin V]⁺ [PI]⁺ indicates late apoptosis, and [Annexin V]⁺ [PI]⁻ indicates early apoptosis. (B) Statistical analysis of apoptosis in HepG2 cells. n = 3 independent biological repeats. (C) Flow cytometry histograms showing SOD2 expression in HepG2 cells. For each assay, a blank group was set without antibody but with FITC dye. (D) Statistical results for SOD2 expression in HepG2 cells. n = 3 independent biological repeats. All data are presented as the means ± SEM. Two-tailed Student’s t test [(B) and (D)]; n.s., not significant; *P < 0.05; **P < 0.01; ***P < 0.001 compared with the model group.

Mito-TEMPO reduces OS and increases SOD2 expression in model cells

SOD2 is specifically induced to eliminate excessive mitochondrial superoxide and mitigate protein oxidation (32). In this section, we first examined the ROS and MitoSOX levels in U18666A-induced HepG2 cells. The results revealed that, compared with the control group, ROS and MitoSOX levels were markedly elevated in the model group, whereas Mito-TEMPO treatment effectively reduced ROS and MitoSOX accumulation in a dose-dependent manner (12.5, 25, and 50 μM) (Fig. 4, D and G, and fig. S1, A and B). Consistent with these, the expression of SOD2 was significantly decreased in the model group, whereas Mito-TEMPO administration restored SOD2 expression in a dose-dependent manner (12.5, 25, and 50 μM) relative to the model group (Fig. 5, C and D). These findings suggest that Mito-TEMPO may attenuate OS production by enhancing SOD2 expression.

Mito-TEMPO takes therapeutic effects on HepG2 induced by U18666A by enhancing SOD2

To further explore whether the therapeutic effect of Mito-TEMPO in the in vitro NPC1 model was mediated by increased SOD2 expression, we constructed SOD2 knockdown plasmids [pLKO.1-SOD2 short hairpin RNA (shRNA)] and SOD2 overexpression plasmids (pcDNA3.1-mCherry-SOD2 OE) (fig. S2A). After transfection, the SOD2 shRNA group showed an obvious decrease in SOD2 expression levels, whereas the SOD2 OE group presented a significant increase (fig. S2, B and C). On the basis of this, we performed drug administration (fig. S2D) and evaluated the effects through the levels of cholesterol, ROS, and JC-1-M. Consistent with the previous findings, the model group showed obviously elevated levels of cholesterol, ROS, and JC-1-M ratio compared with the control group. In contrast, both SOD2 OE and Mito 50 treatment markedly reduced cholesterol, ROS, and JC-1-M levels compared with the model group, indicating that the SOD2 OE mimicked therapeutic effects in the in vitro NPC1 model. However, whereas compared with the Mito 50 group, the Mito 50 + SOD2 shRNA group showed apparently higher cholesterol content, ROS level, and JC-1-M ratios, suggesting that the original therapeutic effect of Mito-TEMPO might be inhibited by SOD2 shRNA (fig. S2, E to J). All these results suggest that Mito-TEMPO exerts potential therapeutic effects on U18666A-induced HepG2 cells by enhancing SOD2 expression.

Mito-TEMPO promotes the mitophagic flux in model cells

We directly visualized mitophagic flux in vitro using confocal microscopy. Initially, the overall expression of LC3B was examined. Immunofluorescence and flow cytometry analyses revealed markedly increased LC3B puncta and expression levels in the model group, suggesting that mitophagy process might be impaired. Mito-TEMPO treatment reduced LC3B levels at the tested concentrations (Figs. 6A and 7, A and G).

Fig. 6. Mito-TEMPO effects on mitophagy flux in HepG2 cells.

(A) Confocal fluorescence images of LC3B in HepG2 cells. Scale bars, 5 μm. (B) Confocal fluorescence images of HepG2 cells transfected with the pmCherry-GFP-LC3B tandem reporter. Scale bars, 5 μm. (C) Confocal fluorescence images of HepG2 cells transfected with the pmCherry-GFP-FIS1^101–152 tandem reporter. Scale bars, 5 μm. (D) Schematic illustration of the pmCherry-GFP-FIS1^101–152 tandem reporter analysis principle. (E) Quantification of vesicles positive for both GFP and mCherry (autophagosomes) and for mCherry alone (autolysosomes). Eight cells per experiment, n = 3 independent biological repeats. (F) Quantification of vesicles positive for mCherry puncta (mitolysosomes). Six cells per experiment, n = 3 independent biological repeats. All data are presented as the means ± SEM. Two-tailed Student’s t test [(E) and (F)]; n.s., not significant; *P < 0.05; **P < 0.01; ***P < 0.001 compared with the model group.

Fig. 7. Mito-TEMPO effects on mitophagy pathway proteins in HepG2 cells analyzed by flow cytometry.

(A to F) Flow cytometry histograms showing LC3B, p62, TOMM20, COX-IV, PINK1, and PARK2 expression in HepG2 cells. For each assay, a blank group was set without antibody but with FITC dye. (G to L) Statistical results for LC3B, SQSTM1/p62, TOMM20, COX-IV, PINK1, and PARK2 expression in HepG2 cells. n = 3 independent biological repeats. All data are presented as the means ± SEM. Two-tailed Student’s t test [(G) to (L)]; n.s., not significant; *P < 0.05; **P < 0.01; ***P < 0.001 compared with the model group.

Subsequently, a pmCherry-GFP-LC3B tandem fluorescent reporter was used to monitor mitophagic flux. The green fluorescence of the fusion protein is highly sensitive to the acidic environment of lysosomes and is rapidly quenched in autolysosomes, leaving only red fluorescence detectable in these structures (41). Fluorescence analysis using the Leica LAS X system showed that, compared with the control group, autolysosomes were significantly reduced in the model group, whereas autophagosomes were markedly increased, indicating inhibition of mitophagic flux. In contrast, Mito-TEMPO treatment (12.5, 25, and 50 μM) effectively reversed this effect in a dose-dependent manner (Fig. 6, B and E), indicating that Mito-TEMPO could promote mitophagic flux.

To further validate this finding, we used a pmCherry-GFP-FIS1^101–152 tandem reporter to locate mitophagy events. This reporter fuses mCherry-GFP to the mitochondrial outer membrane protein FIS1. Under steady-state conditions, mitochondria exhibit both red and green fluorescence. However, when mitophagy is activated, mitochondria are delivered to lysosomes, where the green fluorescent protein (GFP) signal is quenched whereas mCherry fluorescence remains stable, resulting in mCherry-only puncta (41) (Fig. 6D). The results showed that mCherry-only puncta were nearly absent in the model group compared with the control, whereas Mito-TEMPO significantly increased their number at the tested doses (Fig. 6, C and F). Collectively, these findings indicate that Mito-TEMPO enhances mitophagic flux in the in vitro NPC1 model.

Mito-TEMPO modulates PINK1/Parkin-mediated mitophagy via the up-regulation of SOD2

A previous study has demonstrated that the PINK1/Parkin pathway plays a crucial role in initiating mitophagy (42). Therefore, we examined the impact of Mito-TEMPO on this pathway in the NPC1 in vitro model. Flow cytometry revealed reduced PINK1 and PARK2 expression and elevated SQSTM1/p62 levels in the model group compared with the control, whereas Mito-TEMPO treatment (12.5, 25, and 50 μM) increased PINK1 and PARK2 and decreased SQSTM1/p62 expression (Fig. 7, B, E, F, H, K, and L). We also evaluated mitochondrial membrane proteins TOMM20 (outer membrane) and COX-IV (inner membrane), which are closely related to mitophagy. Compared with the control, both TOMM20 and COX-IV were significantly up-regulated in the model group but were down-regulated by Mito-TEMPO in a dose-dependent response (Fig. 7, C, D, I, and J). Collectively, these results suggest that Mito-TEMPO enhances mitophagy in association with the PINK1/Parkin-mediated pathway in the in vitro NPC1 model.

On the basis of the SOD2 OE and SOD2 shRNA plasmid construction, we also detected the expression of LC3B, PINK1, and PARK2 to further explore the hypothesis that increased PINK1/Parkin-mediated mitophagy of Mito-TEMPO might be associated with enhanced SOD2. Consistent with the previous findings, the model group showed significantly elevated LC3B expression and reduced PINK1 and PARK2 expression compared with the control group. In contrast, both SOD2 OE and Mito 50 treatment evidently reduced LC3B expression and increased PINK1 and PARK2 expression. However, compared with the Mito 50 group, the Mito 50 + SOD2 shRNA group showed obviously higher LC3B expression and lower PINK1 and PARK2 expression, indicating that Mito-TEMPO–induced enhancement of mitophagy was abolished (fig. S3, A to F). Collectively, all these results suggest that Mito-TEMPO is associated with enhanced PINK1/Parkin-mediated mitophagy via up-regulating SOD2 expression, which may partly underlie its therapeutic effect on U18666A-induced HepG2 cells.

DISCUSSION

Zebrafish has a single npc1 gene (NCBI Gene ID: 553330) located on chromosome 2 (43). The amino acid sequence of zebrafish npc1 shares 60% identity and 66% similarity with the human NPC1 ortholog (44), making zebrafish an ideal model organism for NPC disease research. In our previous work, CRISPR/Cas9 gene editing targeting exon 2 of NPC1 was used to generate an NPC1-KO zebrafish model (31). The mutation site was located in the cysteine-rich loop of the lumen A region (amino acid residue position C27), corresponding to the conserved NPC1-specific cysteine-rich domain (45, 46). The NPC1-KO zebrafish displayed phenotypes resembling human NPC disease and exhibited typical characteristics of liver disease or other digestive tract impairments in early stage, including jaundice, hepatosplenomegaly, emaciation, and abdominal lipid accumulation. Most NPC1-KO zebrafish died within 1 month, paralleling clinical observations in patients with the p.C31Wfs*26 mutation, who typically succumb during early infancy (47, 48). Thus, improving early-stage survival is a critical therapeutic objective. Because of the rarity of NPC disease and the limitations of mammalian models, effective early-life therapies such as β-CD remain under Phase 1 and Phase 2 clinical trials (35). Building on our established NPC1-KO zebrafish model, this study aimed to explore alternative treatments capable of enhancing early survival in NPC disease.

Loss of NPC1 function leads to lipid accumulation (including cholesterol and sphingolipids) in late endosomal/lysosomal compartments (49). Abnormal cholesterol storage disrupts lysosomal function and affects other organelles such as the endoplasmic reticulum (ER), Golgi apparatus, and mitochondria, resulting in OS and impaired autophagic flux (50–52). OS contributes to excessive lipid and cholesterol synthesis (lipid peroxidation) and autophagy dysregulation (53–55). Moreover, abnormal autophagy induces ER stress, triggers the unfolded protein response, and hinders degradation of damaged organelles through processes such as mitophagy, ER-phagy, and pexophagy, leading to amyloid accumulation and cellular injury in NPC1 disease (15, 51, 52). Consistent with this, our NPC1-KO zebrafish exhibited elevated OS, blocked autophagic flux, and shortened lifespan compared with WT controls. Accumulating evidence has demonstrated that hepatocytes degrade lipid droplets via macrolipophagy or microlipophagy, thereby alleviating hepatic steatosis (56–58). Thus, autophagy induction could represent a rational therapeutic approach for NPC disease. However, several studies suggest that autophagy activation alone cannot rescue cholesterol accumulation in NPC1-deficient cells (59), mitophagy impairment and cholesterol accumulation seem to be two parallel pathological mechanisms in NPC disease. Rather, combining autophagy activators with cholesterol-lowering agents may synergistically restore cellular homeostasis (23, 26, 59). In addition, β-CD–threaded polyrotaxanes have been shown to both reduce cholesterol accumulation and enhance autophagy in NPC1 fibroblasts (60). Considering these findings and our zebrafish results, effective NPC therapies may require multitarget agents that simultaneously reduce OS and restore autophagy.

The SOD family plays a vital role in OS regulation (61). In this study, proteomic analysis of NPC1-KO zebrafish identified significantly decreased SOD2 expression, whereas SOD1 and SOD3 remained unchanged. Reduced SOD2 expression can lead to excessive ROS accumulation (62). A similar reduction has been reported in NPC1^I1061T cells, where N-acetylcysteine or histone deacetylase inhibitors alleviated OS by up-regulating SOD2 and preventing apoptosis (5, 17). Zinc ions have also been shown to facilitate metabolic and bioenergetic recovery following spinal cord injury by activating microglial mitophagy via the FOXO3a-SOD2 pathway (63). Nevertheless, no prior in vivo studies have investigated whether SOD2 enhancement can directly ameliorate NPC disease. We found that Mito-TEMPO, a mitochondrial antioxidant acting through SOD promotion (64), might serve as a potential therapeutic candidate. Mito-TEMPO has exhibited significant therapeutic efficacy in neurodegenerative disease and metabolic disorder models: 50 to 100 μM Mito-TEMPO improves neuronal survival in Alzheimer’s disease models by increasing SOD activity (65, 66). Similarly, 70 μM Mito-TEMPO has a therapeutic effect on ferroptosis in pancreatic β cells through the mitochondrial ROS autophagy-lysosomal pathway (67). However, the therapeutic effects of Mito-TEMPO on NPC disease have not been reported in either in vitro or in vivo models. On the basis of the previously reported effective doses of Mito-TEMPO and our zebrafish developmental toxicity and cytotoxicity assays, we selected 100 μM as the treatment concentration for zebrafish in vitro models and 12.5, 25, and 50 μM as the concentration gradients for in vitro cell culture models in this study. Mito-TEMPO treatment significantly improved early-stage survival in NPC1-KO zebrafish, reduced hepatic cholesterol and lipid accumulation, and mitigated liver function impairment, with effects comparable to β-CD. Furthermore, Mito-TEMPO reduced OS and restored dysregulated autophagy levels in NPC1 model.

Proteomic and ultrastructural analyses also revealed mitochondrial structural damage and down-regulation of PINK1 and Parkin(key mitophagy-related proteins) in NPC1-KO zebrafish, implying suppression of mitophagy. Mitophagy can proceed through Parkin-dependent or Parkin-independent mechanisms, with the PINK1/Parkin pathway being the most characterized (68, 69). In this pathway, PINK1 accumulation recruits and activates Parkin, initiating mitophagy to preserve cellular homeostasis (70). A previous study has shown that cadmium induces PINK1/Parkin-mediated mitophagy to protect cortical neurons from apoptosis (71), whereas taurine activates the same pathway to reduce OS and mitigate intervertebral disc degeneration (72). However, few studies have linked mitophagy modulation to NPC1 pathology. Our results indicate that Mito-TEMPO restored mitochondrial integrity, enhanced mitophagic flux, and up-regulated PINK1/Parkin signaling, consistent with the idea that its therapeutic efficacy in NPC disease involves the enhancement of Parkin-dependent mitophagy and attenuation of apoptosis.

Generally, Mito-TEMPO is a mitochondrion-targeted SOD mimetic especially for SOD2 (65, 73), which may underpin its ability to reverse the SOD2 down-regulation and degradation induced by NPC disease. However, our preliminary molecular docking analysis revealed strong binding affinities between Mito-TEMPO and SOD2 in both zebrafish and human models, despite minor structural variations in SOD2 orthologs between these two species. These findings are consistent with the possibility that Mito-TEMPO might interact with SOD2 to stabilize the protein, potentially contributing to the up-regulation of SOD2 expression and activity. Notably, SOD2 knockdown in our NPC1 models partially attenuated the therapeutic efficacy (including reduced cholesterol and ROS levels and ameliorated mitochondrial membrane potential) and reduced PINK1/Parkin signaling triggered by Mito-TEMPO, indicating that its therapeutic effects were partially suppressed in association with the lack of sufficient SOD2 binding partners. Consistently, our data suggest that Mito-TEMPO reduced cholesterol accumulation in the NPC1 model by alleviating OS and restoring autophagic flux by enhancing SOD2. We observed that SOD2 OE also improved cholesterol homeostasis in the in vitro NPC1 model, which supports the potential feasibility of SOD2-targeted therapy for NPC disease in vivo.

There still exist several limitations in this study. First, the random nature of point mutations within NPC1-KO sites, combined with current genetic engineering limitations, precludes the reliable establishment of stable NPC1-KO cell lines that accurately replicate the specific loss-of-function mutations found in NPC1-KO zebrafish. Therefore, an in vitro NPC1 dysfunction model was established using U18666A, a potent inhibitor of late endosomal trafficking, to investigate the disease’s underlying mechanisms. Besides, NPC disease is also characterized by neurodegenerative symptoms such as ataxia and muscular tremors. Nevertheless, this study primarily focused on therapeutic interventions aimed at improving survival and cholesterol homeostasis rather than addressing neurological pathology as neurodegenerative symptoms typically occur later than lipid accumulation. Accordingly, hepatic and intestinal tissues were selected as the primary target organs, given their central roles in lipid metabolism. Last, although Mito-TEMPO exhibited activity through docking with SOD2 and showed potential therapeutic effects in the NPC1 models, Mito-TEMPO may also exert its therapeutic effects via alternative molecular pathways or targets. Similarly, enhancing SOD2 expression alleviated cholesterol accumulation in the in vitro NPC1 model, and the study could not conclusively determine whether direct targeting of SOD2 alone would yield comparable outcomes in vivo. Future studies will focus on elucidating the precise molecular mechanisms (Parkin or PINK1 KO models) and evaluating nervous system development (spinal cord and brain) of the NPC disease model to further clarify the therapeutic potential of Mito-TEMPO. In addition, we will assess the merits of combination therapy involving Mito-TEMPO and other promising compounds (e.g., β-CD), as well as the therapeutic value of SOD2-targeted interventions in vivo for NPC disease.

Collectively, using the NPC1-KO zebrafish model, our findings support that Mito-TEMPO effectively improves survival, reduces cholesterol accumulation, and restores liver function in NPC1-KO zebrafish. Its mechanism is likely consistent with the suppression of OS and the promotion of Parkin-dependent mitophagy via Sod2 activity enhancement (Fig. 8). These findings provide valuable insights into potential therapeutic approaches for NPC disease, particularly for highly aggressive genotypes manifesting in early developmental stages.

Fig. 8. Therapeutic effects of Mito-TEMPO on early-stage NPC1-KO zebrafish.

The reduced Sod2 levels and abnormal mitophagy were observed in NPC1-KO zebrafish. Mito-TEMPO effectively promotes Sod2 expression, alleviates OS, and enhances mitophagy, thereby mitigating NPC1-related symptoms and improving early-stage survival.

MATERIALS AND METHODS

Molecular docking

The molecular structure of Mito-TEMPO was obtained from the PubChem Compound Database (https://pubchem.ncbi.nlm.nih.gov/), and SOD2 protein structure data were retrieved from the PDB (https://rcsb.org/) or UniProt (https://uniprot.org/). Water molecules were first removed from both protein and ligand structures, polar hydrogen atoms were added, and the protein was enclosed within a defined grid box. Docking simulations were performed using AutoDock Vina 2.1.6 with 50 iterations. The conformation exhibiting the lowest binding energy and most frequently recurring binding mode was selected as the final output. The docking results were visualized using PyMOL 3.1 (Schrödinger, USA) and Discovery Studio 2019 (BIOVIA, USA).

Animals

The Tübingen strain of zebrafish used was obtained from the China Zebrafish Resource Center (Wuhan, China). All experiments were approved by the Animal Research and Ethics Committee of the Institute of Hydrobiology, Chinese Academy of Sciences, and conducted in accordance with institutional guidelines. Adult zebrafish were maintained in a recirculating system at 28°C under a 14-hour light/10-hour dark cycle. Zebrafish larvae were raised in 90-mm dishes until 5 dpf and subsequently transferred to the circulation system. Larvae were fed vitelline and paramecia twice daily from 5 to 14 dpf, and juvenile fish (≥14 dpf) were fed brine shrimp twice daily. All the animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of the Institute of Hydrobiology, Chinese Academy of Sciences (IHB/LL/2024077).

Generation of NPC1-KO zebrafish

The NPC1-KO zebrafish were generated by intercrossing heterozygous npc1 carriers. Genotyping was performed using polymerase chain reaction (PCR) according to previously established methods (31). Both male and female zebrafish were used as no sex-specific differences in NPC1 disease onset or progression have been reported (74).

Developmental toxicity assays

Developmental toxicity testing was performed in 6-well plates with 10 WT embryos per well, incubated at 28°C following OECD Test Guideline 236 (OECD, 2013) (75). Embryos were exposed to 2.5 mM β-CD or Mito-TEMPO (12.5, 25, 50, and 100 μM) from 12 to 96 hours postfertilization (hpf). Each treatment group included three biological replicates. Heart rate was assessed under an inverted microscope (Z16 APO, Leica, Germany) by counting beats over a 30-s interval at 48 and 96 hpf.

Zebrafish drug treatment and sample collection

For drug treatment, offspring from npc1 heterozygote crosses were divided into three groups: WT zebrafish, untreated NPC1-KO zebrafish, and drug-treated NPC1-KO zebrafish. Drug treatments were administered by immersion for 24 hours every other day [quaque altera die (QOD)] starting at hatching and continued for at least 60 days. Treatment concentrations were 100 μM Mito-TEMPO (MedChemExpress, HY-112879) or 1 mM β-CD (Sigma-Aldrich, 332607). Genotyping and sampling were conducted at 10 dpf and 1 mpf. If fish died at intermediate time points, samples were collected immediately for genotyping (Fig. 2A).

For sample collection, fish were fasted overnight before processing. Groups of 10 larvae (10 dpf) or 2 to 3 larvae (1 mpf) fish were anesthetized with 0.05% eugenol (MedChemExpress, HY-N0337) and prepared for biochemical and Western blot analyses. Whole fish were fixed in 4% paraformaldehyde (Beyotime Biotechnology, P0099) or electron microscopy fixative (Servicebio, G1102), embedded in optimal cutting temperature (OCT) compound (Sakura Finetek USA, 4583), frozen, and sectioned at 8 μm in thickness using a cryostat (Cryotome E, Thermo Fisher Scientific, USA). Remaining tissues were snap frozen in liquid nitrogen and stored at −80°C until further use.

Cell culture, transfection, and drug treatment

HepG2 cells (American Type Culture Collection, HB-8065) were cultured in Dulbecco’s modified Eagle’s medium (DMEM; BBI LIFE SCIENCES CORPORATION, E600003-0500) supplemented with 10% fetal bovine serum (FBS; Gibco, 10099-141) and 1% penicillin–streptomycin-amphotericin B (Beyotime Biotechnology, C0224) at 37°C in a humidified incubator with 10% CO₂. Cells were digested using 0.05% trypsin-EDTA (Beyotime Biotechnology, C0205) and seeded at 1 × 10⁵ cells/ml in 12-well plates or 1 × 10⁴ cells/ml in 96-well plates.

For transfection, cells were transfected with plasmids using EndoFectin HepG2 reagent (Procell, 164416) following the manufacturer’s instructions. Plasmids pmCherry-GFP-LC3B (MiaoLing Plasmid Sharing Platform, P0202) and pmCherry-GFP-FIS1^101–152 were kindly provided by R. Zhou’s laboratory (41).

For cell viability testing, cells were treated with phosphate-buffered saline (PBS) or Mito-TEMPO (0.78 to 500 μM) for 48 hours. A 2% dimethyl sulfoxide (DMSO) (Sigma-Aldrich, D5879) was used as the positive control.

For staining and flow cytometry assays, cells were pretreated with 3 μM U18666A (Sigma-Aldrich, HY-107433) for 24 hours to induce the NPC1 model, followed by PBS or Mito-TEMPO (12.5, 25, 50, and 100 μM). A 2-hour starvation period was included to induce mitophagy (Fig. 4A).

Survival assays

Zebrafish survival was monitored every other day for 60 days. Mortality was recorded as 1, and accidental loss or survival as 0, for survival curve analysis.

Liver enzyme assays

Liver function was evaluated by measuring AST (Beyotime Biotechnology, P2715S) and ALT (Beyotime Biotechnology, P2711S) levels using assay kits according to the manufacturer’s protocols. Fluorescence intensity was measured with a microplate reader (SpectraMax i3x, Molecular Devices, China) at 560/590 nm.

Transmission electron microscopy assays

Whole zebrafish tissues were fixed in electron microscopy fixation solution under light-protected conditions at 4°C, cut into 1-mm³ sections, rinsed in 0.1 M PBS, and postfixed in 1% osmium tetroxide (Ted Pella, 18456) for 2 hours. Samples were dehydrated through a graded acetone series, infiltrated with epoxy resin for 2 to 3 hours, embedded, and polymerized at 70°C for 30 min using Epon 812 epoxy resin (SPI-CHEM, 90529-77-4) in predried molds. Resin blocks were then cured at 40°C for 12 hours and 60°C for 48 hours to form hard plastic. After trimming, ultrathin sections (60 to 80 nm) were prepared, mounted on copper grids, and stained with lead citrate and uranyl acetate (SPI-CHEM, 02624-AB). Sections were examined and imaged using a transmission electron microscope (HT7800, Hitachi, Japan).

OS assays

For the cellular ROS assay, cells were incubated with 10 μM dichlorodihydrofluorescein diacetate (DCFH-DA; Beyotime Biotechnology, S0033S) for 20 min at 37°C in the dark following drug treatment. Cells were then washed with cold PBS. Fluorescence images were captured using a fluorescence microscope (Ti-E, Nikon, Japan), and quantitative data were obtained with a microplate reader (SpectraMax i3x, Molecular Devices, China) at 488/525 nm.

For zebrafish ROS assays, frozen tissue sections were incubated with 10 μM dihydroethidium (DHE; Servicebio, G1746) for 30 min at 37°C in the dark, followed by fluorescence microscopy examination.

For zebrafish MDA detection (Beyotime Biotechnology, S0131S), whole-tissue homogenates were prepared on ice, centrifuged at 1000g for 10 min, and the supernatant was collected for measurement. MDA levels were determined commercial assay kits according to the manufacturer’s instructions.

Lipid and cholesterol assays

For Filipin III staining (Absin, Abs42018484), cells were fixed with 4% paraformaldehyde for 10 min, washed with PBS, stained with Filipin III dye for 20 min at room temperature (RT) in the dark, and imaged using a fluorescence microscope (DM4B, Leica, German). Frozen zebrafish sections were processed similarly.

For Oil Red O staining, frozen zebrafish sections were stained with Oil Red O solution (Beyotime Biotechnology, C0157S) for 20 min and imaged with a slide scanner (Aperio VERSA B, Leica, Germany).

For zebrafish triglyceride (Beyotime Biotechnology, S0219S) and cholesterol (Beyotime Biotechnology, S0211S) quantification, whole-tissue homogenates were analyzed using commercial assay kits following the manufacturer’s instructions, and fluorescence intensity was measured at 560/590 nm using a microplate reader.

For the cellular cholesterol efflux assay, HepG2 cells were seeded in 96-well plates at 1 × 10⁴ cells/ml. Cholesterol levels were determined using the Amplex Red cholesterol assay (Beyotime Biotechnology, S0211S) according to the manufacturer’s instructions. Reactions were conducted with and without cholesterol esterase to quantify total cholesterol (TC) and free cholesterol (FC). Cholesteryl esters (CEs) were calculated as TC−FC, and the CE/FC ratio was determined. Fluorescence intensity was measured at 560/590 nm using a microplate reader.

Western blotting assays

Whole zebrafish tissues were lysed in radioimmunoprecipitation assay (RIPA; Beyotime Biotechnology, P0038) buffer on ice for 30 min. Supernatants were collected after centrifugation at 12,000g for 15 min. Protein concentration was determined using a BCA assay kit (Beyotime Biotechnology, P0010). Equal amounts of protein (20 to 40 μg) were resolved by SDS–polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride (PVDF) membranes. Membranes were blocked in 5% bovine serum albumin (BSA; BioFroxx, 4240GR025) for 1 hour at RT and then incubated overnight at 4°C with primary antibodies: SOD2 (1:200; GeneTex, GTX124294), LC3 (1:200; ABclonal, A15591), PARK2 (1:200; ABclonal, A0968), and β-actin (1:1000; Proteintech, 81115-1-RR), followed by horseradish peroxidase (HRP)–conjugated secondary antibodies (Kirkegaard & Perry Laboratories, 074-1506) for 1 hour at RT. Protein bands were visualized using enhanced chemiluminescence (ECL; Beyotime Biotechnology, P0018M) and imaged with a ChemiDoc system (5200, Tanon, China). Densitometric analysis was performed using the ImageJ software.

Cell viability assays

Cell viability was measured using the Cell Counting Kit-8 (CCK-8; Beyotime Biotechnology, C0038). After drug treatment, CCK-8 reagent was added to each well, and cells were incubated for 2 hours at 37°C. Optical density (OD) was measured at 450 nm using a microplate reader.

Mitochondrial transmembrane potential assays

Cells were seeded in 96-well plates and subjected to drug treatment, followed by staining with JC-1 dye according to the JC-1 Mitochondrial Membrane Potential Assay Kit (Beyotime Biotechnology, C2006) protocol. Fluorescence was recorded using a microplate reader at 490/530 nm or 525/590 nm,

Immunofluorescence microscopy

For zebrafish immunofluorescence staining, tissue sections underwent antigen retrieval using citric acid buffer at 95°C, followed by blocking with H₂O₂ for 25 min at RT, permeabilization with 0.5% Triton X-100 for 10 min, and blocking with 5% BSA in PBS for 1 hour at RT. Sections were incubated overnight at 4°C with primary antibodies (1:25) diluted in 0.1% Triton X-100 and 2% BSA in PBS. After washing, sections were incubated with Cy3-conjugated secondary antibodies (1:100; Beyotime Biotechnology, A0516) for 1 hour at RT. Antigen retrieval and immune elution were performed for 30 min before incubation with a second antibody, followed by fluorescein isothiocyanate (FITC)–conjugated secondary antibody (1:100; Beyotime Biotechnology, A0562) for 1 hour at RT. After 4′,6-diamidino-2-phenylindole (DAPI) staining (Beyotime Biotechnology, R0306S-6) for 5 min, images were acquired using a fluorescence or confocal microscope (TCS SP8, Leica, Germany).

For cell immunofluorescence staining, treated cells were washed with PBS, fixed with 4% paraformaldehyde for 10 min at RT, permeabilized with 0.5% Triton X-100 for 10 min, and blocked with 5% BSA in PBS for 1 hour. Cells were incubated with primary antibodies (1:100) overnight at 4°C, followed by appropriate fluorescent secondary antibodies for 1 hour and DAPI counterstaining for 5 min at RT. Images were captured using a confocal microscope.

For mitophagy flux analysis, cells transfected with tandem fluorescence reporters were fixed with 4% paraformaldehyde for 15 min at RT after treatment, and fluorescence images were captured using a confocal microscope.

Flow cytometry

For apoptosis analysis, cell apoptosis was evaluated using the Annexin V Apoptosis Detection Kit (Beyotime Biotechnology, C1062M) following the manufacturer’s instructions. A total of 5 × 10⁵ cells were washed with PBS and resuspended in 100 μl of binding buffer containing 5 μl of annexin V–FITC and 10 μl of propidium iodide (PI). After incubation for 15 min at RT in the dark, 400 μl of binding buffer was added, and samples were immediately analyzed on a flow cytometer (CytoFLEX S, Beckman, USA). The data were analyzed by FlowJo V (FlowJo LLC, USA).

For intracellular staining of LC3B, SQSTM1/p62, PINK1, PARK2, TOMM20, COX-IV, and SOD2, cells were fixed and permeabilized (Multisciences, GAS005) and then blocked for 1 hour at RT with PBS containing 5% BSA. Cells were incubated with rabbit anti-mouse primary antibodies against LC3B (1:100, ABclonal, A11923), SQSTM1/p62 (1:100, ABclonal, A11483), PINK1 (1:100, ABclonal, A7131), PARK2 (1:100, ABclonal, A0968), TOMM20 (1:100, ABclonal, A6774), COX-IV (1:100, ABclonal, HB, A6564), or SOD2 (1:100, GeneTex, GTX124294) for 4 hours, followed by FITC-conjugated goat anti-rabbit IgG (1:100, Thermo Fisher Scientific, F-2765) for 1 hour at RT. After washing, fluorescence intensity was detected using a flow cytometer and the data were analyzed by FlowJo V.

Proteomic analysis

WT and NPC1-KO zebrafish at 1 mpf were collected for proteomic profiling. Three biological replicates were analyzed per group. Samples were rinsed with ultrapure water, and protein extraction and trypsin digestion were performed. Frozen samples were ground into powder using liquid nitrogen, transferred to 1.5-ml centrifuge tubes, and lysed with buffer containing 8 M urea, 1 mM phenylmethylsulfonyl fluoride (Sinopharm, XW020003), and 2 mM EDTA (Sinopharm, 10009717). Ultrasonic lysis was conducted for 5 min on ice, followed by centrifugation at 15,000g and 4°C for 10 min to collect the supernatant.

A total of 100 μg of protein per sample was diluted to 200 μl with 8 M urea, reduced with 5 mM dithiothreitol (DTT; Solarbio, D8220) at 37°C for 45 min, and alkylated with 11 mM iodoacetamide (Aladdin, I131590) for 15 min in the dark at RT. The reaction was diluted with 800 μl of 25 mM ammonium bicarbonate, and digestion was initiated with 2 μg of trypsin overnight at 37°C. The pH of the digested peptides was adjusted to 2 to 3 using 20% trifluoroacetic acid (TFA), and samples were desalted with C18 columns.

Peptide separation was performed using a Vanquish Neo UHPLC system. The mobile phase A consisted of 0.1% formic acid in water, and mobile phase B was 0.1% formic acid in acetonitrile. A trap-and-elute dual-column setup was used, with a PepMap Neo Trap Cartridge (300 μm by 5 mm, 5 μm) as the trapping column and an Easy-Spray PepMap Neo UHPLC column (150 μm by 15 cm, 2 μm) as the analytical column. The column temperature was maintained at 55°C, injection volume at 200 ng, flow rate at 2.5 μl/min, gradient time at 6.9 min, and total run time at 8 min.

Chromatographically separated peptides were analyzed using the Orbitrap Astral high-resolution mass spectrometer (Orbitrap Astral, Thermo Fisher Scientific, USA) operating in positive ion mode under data-independent acquisition (DIA). The precursor ion scan range was 380 to 980 mass/charge ratio (m/z), with an MS1 resolution of 240,000 (at 200 m/z), normalized automatic gain control (AGC) target of 500%, and maximum injection time (IT) of 5 ms. MS2 analysis was performed using DIA mode with 299 scan windows, an isolation window of 2 Th, higher-energy collisional dissociation (HCD) collision energy of 25%, normalized AGC target of 500%, and maximum IT of 3 ms. Differentially expressed proteins (DEPs) were identified and annotated using Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and subcellular localization analyses.

Quantification and statistical analysis

Raw mass spectrometry (MS) data were processed using DIA-NN (v1.8.1) in library-free mode. The UniProtKB database (uniprotkb_proteome_UP000000437_banmayu_2024_08_06. fasta; 46 559 sequences) was used to generate a spectral library through neural network–based deep learning algorithms. The “match between runs” (MBR) option was applied to create a DIA-based spectral library, followed by reanalysis with this library. The false discovery rate (FDR) for both protein and precursor ion identification was controlled below 1%.

All quantitative data were analyzed using GraphPad Prism 10 (GraphPad Software, USA). Statistical comparisons between two groups were conducted using Student’s t test or the Gehan-Breslow-Wilcoxon test where appropriate. Unless otherwise indicated, results are expressed as means ± SEM. Significance levels were denoted as *P < 0.05; **P < 0.01; ***P < 0.001; nonsignificant (n.s.), P ≥ 0.05.

Supplementary Materials

This PDF file includes:

Supplementary Methods

Table S1

Figs. S1 to S3