Introduction

Lysosomes are essential organelles for the degradation and recycle of damaged complex substrates and organelles (Xu and Ren 2015). In recent years, lysosomes have emerging roles in plasma membrane repair, external environmental sensing, autophagic cargo sensing, and proinflammatory response, and thereby regulating fundamental processes such as cellular clearance and autophagy (Tsunemi, Perez-Rosello et al. 2019). Mutations in genes encoding lysosomal proteins could result in more than 70 different lysosomal storage disorders (Fraldi, Klein et al. 2016). Niemann-Pick disease type C (NPC) is a rare lysosomal storage disorder caused by mutations in either NPC1 or NPC2 gene, an endolysosomal cholesterol transporter. Deficiency in NPC1 or NPC2 protein results in late endosomal/lysosomal accumulation of unesterified cholesterol (Sarkar, Carroll et al. 2013, Spampanato, Feeney et al. 2013). Clinical symptoms of NPC include hepatosplenomegaly, progressive neurodegeneration and central neurol system dysfunction ie. seizures, motor impairment, and decline of intellectual function (Carstea, Morris et al. 1997). So far there is no FDA-approved specific therapy for NPC, although miglustat, approved to treat type I Gaucher disease, has been used for NPC treatment in countries including China, Canada and European Union (Pineda, Wraith et al. 2009, Pineda, Perez-Poyato et al. 2010, Wraith, Vecchio et al. 2010, Chien, Peng et al. 2013)

Transcriptional factor EB (TFEB), a member of the microphthalmia/TFE transcriptional factor (MiTF) family, is identified as a master regulator of lysosome and autophagy function by controlling the “coordinated lysosomal expression and regulation” (CLEAR) network, covering genes associated to lysosomal exocytosis and biogenesis, and autophagy (Sardiello and Ballabio 2009, Martini-Stoica, Xu et al. 2016, Napolitano and Ballabio 2016). In normal condition, TFEB is phosphorylated by mTOR kinase and kept in cytosol inactively (Martina and Rosa 2018, Napolitano, Esposito et al. 2018). Under stress conditions ie. starvation or oxidative stress, TFEB is dephosphorylated and actively translocates into nucleus, promoting the expression of genes associated with lysosome and autophagy (Medina, Diego et al., Zhang, Cheng et al. 2016, Puertollano, Ferguson et al. 2018). TFEB overexpression or activation results in increased number of lysosomes, autophagy flux and exocytosis (Medina, Fraldi et al. 2011, Settembre, Di Malta et al. 2011, Giatromanolaki, Kalamida et al. 2015, Xu, Du et al. 2020), which may trigger cellular clearance. In fact, it was reported that TFEB overexpression promotes cellular clearance ameliorating the phenotypes in a variety of neurodegenerative diseases such as Alzheimer’s, Parkinson’s, and Huntington’s. (Sardiello, Palmieri et al. 2009, Tsunemi and La Spada 2012, Decressac and Björklund 2013, Polito, Li et al. 2014, Ballabio 2016, Napolitano and Ballabio 2016). Moreover, up-regulation of TFEB has been reported to benefit lysosomal storage diseases (LSDs), such as Pompe disease (Argüello, Balboa et al. 2021). However, it is not known whether upregulation of TFEB by genetic or pharmacological methods, is sufficient to increase lysosome functions and alleviate NPC phenotypes in vitro and in vivo. If so, TFEB may be a putative target for NPC treatment and manipulating lysosome function via TFEB with small molecules may have broad therapeutic potential for NPC.

Results

Transgenic overexpression or pharmacological activation of TFEB reduces cholesterol accumulation in various human NPC1 cell models

TFEB and TFE3 are identified as key transcriptional factors that regulating lysosome and autophagy biogenesis (Sardiello, Palmieri et al. 2009). To investigate the role of TFEB/TFE3 in cellular cholesterol levels in NPC1 cells, HeLa cells were treated with U18666A (2.5 μM, an inhibitor of the endosomal/lysosomal cholesterol transporter NPC1), a drug that has been widely used to induce the NPC phenotype in cell models (Poh, Shui et al. 2012) (Thereafter HeLa NPC1 cells represent U18666A-treated HeLa cells). Cellular cholesterol levels were measured using the well-known cholesterol-marker filipin staining (Lu, Liang et al. 2015). As shown in Fig. 1A, B, in HeLa NPC1 cells overexpressing with TFEB-mCherry or TFEB^S211A- mCherry (S211 non-phosphorylatable mutant, a constitutively active TFEB) (Wang, Gao et al. 2015), the cellular cholesterol levels (filipin) were significantly diminished compared to non-expressed or mCherry-only transfected NPC1 cells. In contrast, TFE3-GFP overexpression displayed no obvious reduction of cholesterol levels in HeLa NPC1 cells (Fig. S1A, B). Hence, TFEB but not TFE3 contributes to cholesterol reduction in NPC1 cells.

Figure 1.

Pharmacological activation of TFEB is an emerging strategy for LSD treatment (Levine and Kroemer 2008). We have previously identified a natural TFEB agonist-sulforaphane (SFN), which is also an activator of cellular antioxidant NFE2/Nrf2 pathway, dramatically mitigating oxidative stress commonly associated with metabolic and age-related diseases including NPC diseases (Corssac, Campos-Carraro et al. 2018, Li, Shao et al. 2021, Liu, Fu et al. 2021). Based on previous data that SFN potently activates TFEB, leading to upregulate lysosomal function, we hypothesize that SFN maybe contribute to cholesterol clearance in NPC disease. To evaluate this hypothesis, we first examined the effect of SFN on cellular cholesterol levels in HeLa NPC1 cell model using Filipin staining. As shown in Fig. 1C, D, HeLa cells treated with U18666A showed significantly increased filipin signals in bright perinuclear granules, which were well co-localized with lysosomal marker LAMP1, but not with ER marker calnexin (Fig. S2A, B) (indicative of cholesterol accumulation in the late endosome/lysosome compartment, but not in ER). Moreover, when HeLa NPC1 cells were further challenged with SFN (15 μM, 12-24 h), a dramatically reduction of cholesterol accumulation (Filipin intensity) by more than 30% was observed in lysosome (Fig. 1C-E, Fig. S2C, D). These results also confirmed that U18666A interferes with the egress of free cholesterol from endosomes/lysosomes as previously reported by others (Lange and Steck 1994, Davis, Shin et al. 2021).

Next, we validated whether SFN can activate TFEB in HeLa NPC1 cell model by performing immunofluorescence experiments. In HeLa NPC1 cells, SFN (15 μM, 24 h) induced robust TFEB translocation from the cytosol to the nucleus (Fig. 1F, G). Consistent with this result, SFN treatment (15 μM, 24 h) resulted an increase on the mRNA levels of TFEB target genes, including genes required for autophagy (ULK1, SQSTM1 and ATG5) and lysosome biogenesis-LAMP1, using quantitative real-time PCR (Q-PCR) (Fig. 1H). A key mechanism of TFEB activation is Ca²⁺-dependent dephosphorylation of TFEB by protein phosphatases (Medina, Diego et al., Martina and Puertollano 2018). We next investigated the specific phosphorylated site on TFEB by SFN in NPC1 cells. Following 24 h exposure to SFN, S211 and S142-TFEB phosphorylation were significantly decreased (Fig. 1I, J), while S122 phosphorylation was not affected. These results indicate that TFEB is dephosphorylated at S211 and S142 residues by SFN in HeLa NPC1 model. TFEB nuclear shuffling is regulated by the activity of MTOR, which is regulated by its phosphorylation status (Martina, Chen et al. 2012). SFN reportedly activates TFEB in a mTOR-independent manner (Li, Shao et al. 2021). Consistent with previous report, no significant inhibition of phosphorylated MTOR (p-MTOR) or RPS6KB1 (p-RPS6KB1) can be observed in HeLa NPC1 cells treated with (15 μM, 24 h) (Fig. S3A, B). Therefore, in HeLa NPC1 cells SFN-induced TFEB activation is unlikely to be mediated by MTOR inhibition. Give the well-known role of SFN as a NFE2/Nrf2 inducer, we also validated whether SFN can induce Nrf2 activation in NPC cells. As shown in Fig. S3C, D, SFN (15 μM, 24 h) induced robust Nrf2 nuclear translocation from the cytosol to the nucleus in HeLa NPC1 and NPC primary mouse embryonic fibroblasts (MEF) cells.

We further verified the effects of SFN using another NPC1 cell model by knockdown (KD) NPC1 with specific siRNA (Liao, Wang et al. 2015, Hoglinger, Burgoyne et al. 2019). In NPC1 KD HeLa cells with more than 80 % knockdown efficiency (Fig. 2A), SFN can induce TFEB nuclear translocation (Fig. 2B, C) and upregulate its downstream gene expression (Fig. 2D). Notably, SFN (15 μM, 24 h) treatment significantly reduced cholesterol levels in NPC1 KD cells (Fig. 2E, F). Furthermore, we observed similar results in human NPC1-patient fibroblast cells, SFN treatment robustly promoted TFEB nuclear translocation (Fig. 2G, H) and cholesterol clearance (Fig. 2I, J). Taken together, these experiments demonstrate that pharmacological activation of TFEB activation by SFN could promote cellular cholesterol clearance in various NPC in vitro models.

Figure 2.

TFEB is required in SFN -promoted cholesterol clearance in NPC1 cells

We next investigated whether TFEB is required for SFN-promoted cholesterol clearance using two strategies. On one hand, HeLa cells were treated with the siRNAs specifically against TFEB or transiently overexpressed with mCherry-TFEB. The efficiency of siRNA and overexpression was evaluated by western blot (Fig. 3A). In the siTFEB-transfected HeLa NPC1 cells, SFN-promoted (15 μM, 24 h) cholesterol clearance was almost abolished (Fig. 3B, C). In contrast, in scrambled siRNA-transfected NPC1 cells, the cholesterol levels were significantly reduced by more than 30% by SFN (Fig. 3B, C). Notably, TFEB overexpression-induced reduction of cholesterol can be further boosted with SFN treatment by 20 % (Fig. 3B, C). On the other hand, HeLa TFEB KO cells was used and obtained by the CRISPR/Cas9 tool (Fig. 3D). Consistently, SFN failed to diminish cholesterol in HeLa TFEB KO cells, whereas re-expression of a recombinant TFEB restored SFN-induced cholesterol clearance (Fig. 3E, F). Collectively, these data indicate that TFEB is required for lysosomal cholesterol reduction upon SFN treatment.

Figure 3.

SFN promotes lysosomal exocytosis and biogenesis in human NPC1 models in a TFEB- dependent manner

We next studied the mechanisms by which SFN led to diminish cellular cholesterol in NPC1 cells. Considering the established role of TFEB activation in lysosomal exocytosis, we next verified the hypothesis that SFN-induced cholesterol clearance through upregulation of lysosomal exocytosis by using surface LAMP1 immunostaining (Fig. 4A). When lysosomal exocytosis processes, luminal lysosomal membrane proteins can be detected on the extracellular side of the plasma membrane (PM) by measuring surface expression of LAMP1 (lysosomal associated membrane protein 1), a resident marker protein of late endosomes and lysosomes (referred to as “lysosomes” for simplicity hereafter) with a monoclonal antibody against a luminal epitope of LAMP1 (Reddy, Caler et al. 2001). After incubation with SFN (15 μM) for 24 h, HeLa NPC1 cells exhibited a dramatic increase in the signal of LAMP1 staining in the PM compared to untreated control cells (Fig. 4A, B), suggesting an increase in lysosomes that are attached to this membrane and in the lysosomal exocytic process. Likewise, exposure to SFN induced lysosomal exocytosis in primary macrophage cells (Fig. S4).

Figure 4.

A direct consequence of lysosomal exocytosis is the release of lysosomal contents into the cell culture medium (Rodriguez, Webster et al. 1997). We then quantified the release of free cholesterol/cholesteryl ester into the medium upon SFN treatment in HeLa NPC1 cells using the Cholesterol/Cholesteryl Ester assay Kit. This assay allows to detect total cholesterol or free cholesterol in the presence or absence of cholesterol esterase; the levels of cholesteryl ester are determined by subtracting the levels of free cholesterol from total cholesterol. As shown in Fig. 4C, the levels of free cholesterol released into the medium treated were significantly increased with SFN (15 μM, 24 h) treatment compared to untreated control, suggesting that SFN can stimulate the release of free cholesterol into medium. Notably, the difference of released cholesteryl ester with or without SFN treatment has not been observed and the percentage of cholesteryl ester in total cholesterol released to medium were less than 10% under all conditions tested. Meanwhile, we further examined the effect of SFN on the release of lysosomal enzymes -β-hexosaminidase (NAGase) and acid phosphatase (ACP). In HeLa NPC1 cells treated with SFN (15 μM, 24 h), significantly higher levels of lysosomal enzymes (NAGases and ACP) were detected in the medium compared with untreated controls, but not in TFEB KO cells (Fig. 4D). Taken together, these results indicate that SFN induces an active movement of lysosomes toward the PM and increases lysosomal exocytosis in a TFEB-dependent manner.

Previously we reported that SFN could increase lysosome biogenesis and regulate lysosomal function, which contribute to ROS reduction in NPC models (Li, Shao et al. 2020). We then studied whether SFN activate lysosomal machinery in NPC1 cells. In HeLa NPC1 cells, SFN (15 μM, 12 h) treatment significantly increased the immunofluorescence intensity of LAMP1 (Fig. 4E, F). Likewise, similar results were observed in NPC1 KD cells (Fig. S5A, B). Lysosomal enzymes operate better under acidic conditions, and the degradation-active lysosomes can be tracked using LysoTracker, a fluorescent acidotropic probe (Li, Gu et al. 2019). We observed significant increases of LysoTracker staining in HeLa NPC1 cells following 12 h treatment with SFN (15 μM) (Fig. 4G, H) as well as in NPC1 KD cells (Fig. S5C, D), yet LAMP1 staining was also increased (Fig. 4E, F; Fig. S5A, B). Thus lysosomal pH was more accurately determined using a ratiometric dye-pHrodo Green Dextran. When the fluorescence ratios (pHrodo Green Dextran /CF555) were calibrated to pH values, we found that SFN treatment induced significant lysosomal hyperacidity (Fig. 4I, J). Collectively, these results suggest that SFN promotes lysosome function and biogenesis in human NPC1 model cells, consistent with our previous report.

Notably, we then examined the TFEB expression in two NPC cell lines (HeLa NPC1 cells and NPC1-patient fibroblast cells). As shown in Fig. S6A, B, TFEB expression levels in both NPC cell models were significantly decreased compared to WT, indicating that TFEB expression may be inhibited in NPC cells. Interestingly, the basal levels of lysosome biogenesis in NPC cells were comparable with WT cells (Fig. 4D-G, Fig. S5A-D), suggestive of compensatory changes caused by TFEB downregulation. Taken together, excessive activation of TFEB in NPC cells can be targeted for cholesterol clearance via upregulation of lysosomal function and biogenesis.

SFN alleviates cholesterol accumulation in primary NPC1 ^-/- MEF cells

To address the possible relevance of the SFN/TFEB axis in NPC pathology of mice experiments, we then investigated whether SFN is sufficient to reduce cholesterol and regulates lysosomal function via TFEB activation in primary murine cells. Primary mouse embryonic fibroblasts (MEF) were freshly prepared from NPC1^-/- mice (from Jackson’s laboratory). SFN (15 μM) treatment for 24 h dramatically increased TFEB nuclei signal in the NPC1^-/- MEF cells (Fig. 5A, B). Next, we analyzed the effect of SFN on cholesterol clearance in NPC1^-/- MEF cells. SFN (15 μM) treatment for 72 h exhibited substantial cholesterol reduction, whereas SFN treatment for 24 h showed relative weaker cholesterol clearance in MEF cells (Fig. 5C, D) compared with human NPC1 cells (Fig. 1C), suggesting that human NPC cells are more sensitive to SFN treatment compared to mouse NPC cells. We further tested whether SFN promotes lysosomal biogenesis and function in NPC1^-/- MEF cells. Following 24 h treatment with SFN (15 μM), a significant increase of and LAMP1 staining (Fig. 5E, F) and LysoTracker intensity (Fig. 5G, H) were observed in NPC1^-/- MEF cells. Collectively, these results suggest that SFN regulates TFEB-mediated lysosomal function axis and promotes cellular cholesterol clearance in NPC MEF cells.

Figure 5.

SFN alleviates the loss of Purkinje cells and body weight in Npc1^-/- mice

Considering that SFN promotes lysosomal cholesterol clearance in both human and murine NPC1 cell models (Fig. 1, 2, 5) and reportedly penetrates blood–brain barrier (BBB) (Kim, Kim et al. 2012, Mao, Yang et al. 2018, Tavakkoli, Iranshahi et al. 2019), SFN could be a good therapeutic candidate for neuropathology in NPC disease. To verify whether SFN targets/activates TFEB in brain, 4-week-old BALB/cJ mice were intraperitoneally injected with SFN (50 mg/kg) or vehicle for 12 h and brain tissues including cerebellum and hippocampus were collected and pS211-TFEB/TFEB levels were measured by Western blotting. As shown in Fig. 6A, B, we observed a significant decrease of pS211-TFEB protein in brain tissues with SFN treatment compared to vehicle, suggesting that TFEB in the brain was directly targeted by SFN treatment. This is the first time that SFN was shown to directly active TFEB in the brain. We then evaluated the in vivo therapeutic efficacy of SFN, NPC1^-/- mice (4-week-old) were treated with SFN (30 or 50 mg/kg) by daily intraperitoneal injection for 4 weeks. Purkinje cells located in the cerebellum are the most susceptible to NPC1 loss and exhibit significant selective loss in the anterior part of the cerebellum (Sarna, Larouche et al. 2003). Purkinje cells in cerebellum sections were measured by Calbindin staining and quantified by recording the number of surviving cells in lobules/mm of Purkinje cell layer. As shown in Fig. 6C, D, little survival of Purkinje cells in vehicle-treated NPC1^-/- cerebellum, in contrast, daily injection of SFN (50 mg/kg) in NPC1 mice prevented a degree of Purkinje cell loss particularly in the lobule IV/V of cerebellum. Body weight is another important indicator of therapeutic efficacy in NPC mice. Typically, NPC1 mice weight plateaus at 6-7 weeks of age, and then progressively decline. Meanwhile, we observed that i.p. injection of SFN (50 mg/kg) daily into NPC mice exhibited a significantly improvement in weight loss (Fig. 6E). However, SFN treatment has no effect on liver and spleen enlargement of NPC1 mice (data not shown). Collectively, our results demonstrated that pharmacological activation of TFEB by small-molecule agonists can mitigate NPC neuropathological symptoms in vivo.

Figure 6.

Figure 7.

Discussion

We have demonstrated in the current study that genetic overexpression of TFEB, but not TFE3 can dramatically mitigate cholesterol accumulation in NPC cells. Pharmacological activation of TFEB by SFN, a previously identified TFEB agonist (Li, Shao et al. 2021), significantly promoted cholesterol clearance in human and mouse NPC cells, while genetic inhibition (KO) of TFEB blocked SFN-induced cholesterol clearance. This clearance effect exerted by SFN was associated with upregulated lysosomal exocytosis and biogenesis. Notably, SFN is reportedly BBB-permeable, assuring a good candidate for efficient delivery to the brain, which is essential for targeting neurodegenerative phenotypes in neurological diseases including NPC. In the NPC mouse models, SFN exhibits in vivo efficacy of suppressing the loss of Purkinje cells and maintaining body weight. Hence, genetically or pharmacologically targeting TFEB may represent a promising approach to treat NPC and manipulating lysosome function with small-molecule TFEB agonists may have broad therapeutic potentials.

The MiT/TFE family contains four factors: MITF, TFEB, TFE3, and TFEC, which share an identical basic region for DNA binding, and highly similar HLH and Zip regions for dimerization (Haq and Fisher 2011). Many of the mechanistic insights into MiT regulation have been focused on TFEB and TFE3, which shares some overlapping functions, Surprisingly, in this study we found that overexpression of TFEB, but not TFE3 alleviated lysosomal cholesterol accumulation in NPC cells (Fig. 1A, B, Fig. S1A, B). Moreover, studies reported that only TFEB overexpression, but not other MiT members upregulates lysosomal gene expression (Sardiello, Palmieri et al. 2009). Thus, these results suggest that the functions exerted by TFEB and TFE3 in NPC may appear to be specialized.

As a proof of the role of TFEB activation in NPC, pharmacological activation of TFEB by SFN, a natural small-molecule TFEB agonist, promotes a dramatic lysosomal cholesterol-lowering effect in several genetic and pharmacological NPC cell models (Fig. 1C, 5C). SFN-induced lysosomal exocytosis and the increased population of lysosomes ready to fuse with the PM contribute to the cholesterol clearance by SFN (Fig. 4A-G). Previously we have shown that SFN can mitigate oxidative stress via TFEB-mediated lysosomal function and autophagy flux (Li, Shao et al. 2021). SFN, an electrophilic compound enriched in cruciferous vegetables such as broccoli, is a known potent inducer of NFE2L2/NRF2 (nuclear factor, erythroid 2 like 2) in various cell types including NPC cells (Fig. S3C,D), a transcriptional factor that controls the expression of multiple detoxifying enzymes through antioxidant response elements (AREs) (Yamamoto, Kensler et al. 2018). Notably, the promoter region of the Nfe2l2 gene harbors a CLEAR site, and TFEB activation can upregulate the expression of Nfe2l2 (Mansueto, Armani et al. 2017). Hence, TFEB, together with NFE2L2, function as key regulators of cellular redox. Oxidative stress is a major feature of NPC and has been attributed to neuronal damage, leading to the pathogenesis and progression of NPC (Vazquez, Balboa et al. 2012). Elevated oxidative stress has been observed in the brain of NPC patient and mice (Smith, Wallom et al. 2009, Zampieri, Mellon et al. 2009). NPC patients also show decreased antioxidant capacity (expressed as Trolox equivalents) and diminished activity of different antioxidant enzymes, which indicates a decrease in antioxidant defenses (Fu, Yanjanin et al. 2010). Hence, the protective effect of SFN against NPC could attribute to a combination with antioxidant activity and cholesterol clearance via the lysosome-dependent, TFEB-mediated regulation. Therefore, pharmacological activation of TFEB such as SFN may serve as a potential therapeutic strategy for NPC.

Materials and methods

Mammalian cell culture

HeLa cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM; Thermo Fisher Scientific, 11195-065) supplemented with 10 % fetal bovine serum (FBS; Thermo Fisher Scientific, 10091148). NPC1 patient-derived fibroblast cells (clone GM03123) and a healthy control (clone GM03440) were obtained from the Coriell Institute for Medical Research (NJ, USA). Human fibroblast cells were maintained in modified Eagle’s medium (Thermo Fisher Scientific, 1964643) supplemented with 15 % FBS, 2 mM glutamine (Thermo Fisher Scientific, 25030081) and 1 % penicillin-streptomycin. NPC1^-/- MEF cells were cultured in DMEM with 10 % FBS, 1 % penicillin-streptomycin and 1 % antibiotic-antimycotic (Thermo Fisher Scientific, 15240062). Macrophage cells were cultured in RPMI 1640 medium (Gibco, B122656) supplemented with 20 % FBS. Unless otherwise indicated, all cell cultures were maintained at 37 °C in a humidified 5 % CO₂ incubator.

Stable and CRISPR/Cas9 KO cell lines

GFP-TFEB stable HeLa cell line was kindly provided by Dr. Shawn Ferguson (Yale School of Medicine) (Zhang, Cheng et al. 2016). TFEB CRISPR-Cas9 KO HeLa cells were generated and characterized as reported previously (Li, Shao et al. 2020).

Plasmids/siRNA transfection

Plasmids including mCherry-TFEB, mCherry-TFEB^S211A and TFE3-GFP were maintained in our laboratory as previously described (Nezich, Wang et al. 2015, Li, Shao et al. 2021). The siRNA sequences targeting human TFEB (5ʹ-GAA AGG AGA CGA AGG UUC AAC AUC A-3ʹ) were purchased from Invitrogen. The siRNA sequences targeting human NPC1 (5ʹ- CAA UUG UGA UAG CAA UAU UTT-3ʹ) were chemically synthesized by GenePharma (Shanghai, China). HeLa cells were transfected with plasmids or siRNA using Lipofectamine 3000 reagent (Thermo Fisher Scientific, 2163785) or RNAi-Max reagent (Thermo Fisher Scientific, 13778150) in Opti-MEM (Thermo Fisher Scientific, 11058021), respectively. The efficiency of transfection was examined by Western blotting or Q-PCR.

Filipin staining

Cellular unesterified cholesterol were detected using a cell-based cholesterol assay kit (Abcam, ab133116). It provides a simple fluorometric method to detect the interaction of cholesterol and filipin III, which alters the filipin absorption and fluorescence spectra allowing visualization with a fluorescence microscope. Cells were cultured at 5×10² cells/well in black, clear-bottom 96-well plates and treated with compounds for the indicated conditions. After rinsing with PBS twice, cells were fixed with fixation solution for 10min followed by a PBS rinse twice. The cells were then stained with filipin III solution for 1 h at room temperature in the dark. The images were then captured by an Olympus IX73/Zeiss microscope. Image analysis was conducted using the image J software.

Immunofluorescence and confocal imaging

For immunofluorescence detection of TFEB and Nrf2, cells were grown on coverslips, fixed with 4 % PFA, permeabilized with 0.3 % Triton X-100 (Solarbio, T8200), followed by blocking in 1 % bovine serum albumin (BSA, Merck, B2064) in PBS for 1 h. Cells were then incubated with anti-TFEB (Cell Signaling Technology, 4240) or anti-Nrf2 antibody (Abcam, ab623521) at 4 °C overnight. After four washes with PBS, coverslips were incubated with secondary antibodies for 1 h and counterstained with DAPI for 10 min. For LAMP1 immunostaining, cells were fixed with 100 % methanol (−20 °C) for 10 min and then blocked with 1 % BSA in PBS for 1 h. The primary anti-LAMP1 (Abcam, ab24170) was used in this study. Finally, coverslips were mounted with Fluoromount-G (Southern Biotech, 0100-01) and ready for imaging.

Western Blotting

Cells were lysed with ice-cold RIPA buffer (Solarbio, R0010) in the presence of 1×protease inhibitor cocktail (Merck, P8340) and 1×phosphatase inhibitor cocktail (Abcam, GR304037-28) on ice for 20 min. Cells were then centrifuged and the supernatant was collected. The protein concentration of the supernatant was determined using BCA Protein Assay (Thermo Scientific, UA269551). Protein samples (20-40 μg) were then loaded and separated on SDS-polyacrylamide gradient gels (GenScript, M00654) followed by the transfer to polyvinylidene difluoride membranes (Merck, R7DA8778E). Western blot analysis was performed using primary antibodies against TFEB (1:500, Cell Signaling Technology, 4240 for cells, 1:1000, Bethyl Laboratories, A303-673A for mice tissue), pS122- TFEB (1:500, Cell Signaling Technology, 86843), pS142-TFEB (1:500, Millipore, 3321796), pS211-TFEB (1:500, Cell Signaling Technology, 37681), NPC1 (1:1000, Abcam, ab134113), LAMP1 (1:500, Abcam, ab24170), MTOR (Cell Signaling Technology, 2972), p-MTOR (Sigma-Aldrich, SAB4504476), p-RPS6KB1/S6K1 (Cell Signaling Technology, 9234), RPS6KB1/S6K1 (Cell Signaling Technology, 2708) and GAPDH (1:10000, Sigma-Aldrich, G9545). Bound proteins were then detected with secondary antibodies against horseradish peroxidase-conjugated and enhanced chemiluminescence reagents (Thermo Fisher Scientific, 203-17071). The membranes were visualized by using an Li-COR Biosciences Odyssey Fc system and the band intensity was quantified using Image J software.

LAMP1 surface labeling

Cells were pretreated with SFN as indicated time. Nonpermeabilized cells were then labeled with anti-human LAMP1 antibody (1:500, DSHB, H4A3), which recognizes a luminal epitope, at 4 °C for 1 h. Cells were then fixed in 2 % paraformaldehyde for 30 mins and incubated with Alexa Flour 488-conjugated secondary antibody at room temperature for 1 h. After PBS wash for three times, cells were counterstained with DAPI for 10 min and images were captured by Zeiss confocal microscope.

Lysosomal pH imaging

To measure lysosomal luminal pH, live cells were treated with chemicals as indicated condition, followed by incubation with 50 nM LysoTracker Red DND- 99 (Thermo Scientific, L7528) for 15-30 min. Cells were then washed with PBS for three times. Images were captured by an Olympus/Zeiss microscope. Fluorescence intensities were quantified using the Image J software.

RNA extraction and RT-QPCR

Total RNA was extracted using TRIzol according the manufacturer’s protocol (Thermo Scientific, 191002). cDNA was generated with 100-500 ng of total RNA using GoScript Reverse Transcription System (Promega, 0000316057). Q-PCR was performed using SYBR Green (TOYOBO, 563700) in CFX Connect Optics (BIORAD). The changes in the mRNA expression of target genes were normalized to that of the housekeeping gene HPRT. Primer sequences used in this study are listed as following:

HPRT: For 5ʹ- TGGCGTCGTGATTAGTGATG -3ʹ, Rev 5ʹ-CTGTTCTCGTCCAGCAGACACT -3ʹ

LAMP1: For 5ʹ-CGTGTCACGAAGGCGTTTTCAG -3ʹ, Rev 5ʹ-CTGTTCTCGTCCAGCAGACACT -3ʹ

ULK1: For 5’-TCATCTTCAGCCACGCTGT-3’, Rev 5’-CACGGTGCTGGAACATCTC-3’

SQSTM1: For 5ʹ- CTGGGACTGAGAAGGCTCAC-3ʹ; Rev 5ʹ- GCAGCTGATGGTTTGGAAAT-3ʹ

ATG5: For 5ʹ- TGCGGTTGAGGCTCACTTTATGTC-3’; Rev 5ʹ- GTCCCATCCAGAGCTGCTTGTG-3’

mGAPDH: For 5’-TGA ATA CGG CTA CAG CA -3’; Rev 5’-AGG CCC CTC CTG TTATTA TG-3’

mSQSTM1: For 5ʹ-AGGAGGAGACGATGACTGGACAC-3’; Rev 5ʹ-TTGGTCTGTAGGAGCCTGGTGAG-3’

mLC3: For 5ʹ- CAAGCCTTCTTCCTCCTGGTGAATG-3’; Rev 5ʹ-CCATTGCTGTCCCGAATGTCTCC-3’

mCTSF: 5ʹ-ACGCCTATGCAGCCATAAAG -3’; Rev 5ʹ-CTTTTGCCATCTGTGCTGAG-3’

Measurement of NAGase and ACP activity

Activities of NAGase and ACP enzymes were measured using microplate assay kits (NAGase, absin, abs580171; ACP, Solarbio, BC2135) respectively. Following the manufacturer’s instructions, cells were seeded in 6-well plates and treated with chemicals in FBS-free medium as indicated condition. Aliquot of supernatant medium was collected and put on ice for extracellular (medium) enzyme activity detection. Cells were collected and 100 μl of assay buffer was added. The cell suspension was then sonicated and centrifuged for 8000 x g for 10 mins, supernatant was collected for cellular enzyme activity detection. NAGase activity was measured in a 96-well microtiter plate containing 25 μl of sample (medium or cell lysates) and 25 μl of substrate in each well, which was mixed and incubated at 37 °C for 20 mins, and then 50 μl of stop solution was added to stop the reaction. ACP activity was measured in a microtiter plate containing 20 μl of sample, 40 μl of Reagent I and 40 μl of Reagent II. The plate was incubated at 37 °C for 15 min and then 120 μl of Reagent III was added. Finally, the absorbance of NAGase /ACP was recorded at 405 nm/ 510 nm using a FlexStation 3 Multi-Mode microplate reader (Molecular Devices) respectively. NAGase /ACP activity released to medium (%) = Enzyme activity in medium / total enzyme activity (medium + cell lysates).

Detection of Cholesterol content

The levels of total and free cholesterol released in medium were measured using a colorimetric cholesterol/cholesteryl ester detection kit (Abcam, ab102515) according to the manufacturer’s instructions. Briefly, aliquot of supernatant medium was collected and air-dried at 50 °C. The dried mixture was dissolved in cholesterol assay buffer. Aliquot of samples were mixed with cholesterol assay buffer, substrate and cholesterol enzyme with or without cholesterol esterase and incubated at 37 °C for 30 min. The absorbance was measured at 450 nm using a FlexStation 3 Multi-Mode microplate reader (Molecular Devices). The amount of cholesterol was calculated by standard curve and normalized with cellular protein content.

Animals

NPC1^-/- mice (BALB/cNctr-Npc1 m1N/J, also known as NPC1^NIH) were purchased from The Jackson Laboratory (USA). All the experimental procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals in the Zhejiang University of Technology (Hangzhou, China) and conformed to the National Institutes of Health Guide for Care and Use of Laboratory Animals. Unless stated otherwise, mice were fed with free access to water and standard diet under specific pathogen-free conditions. Genotypes were identified using a PCR-based screening (Amigo, Mendoza et al. 2002).

Histological analysis

Mice perfusion was performed with PBS. Then, mice cerebellums were post-fixed overnight at 4 °C and after this, placed in serial dilutions of sucrose (10-30 %) in PBS at 4°C overnight, respectively. Then cerebellums were cut in 5 μM thick sagittal sections by cryostat at (Leika) at −20 °C. Permeabilized slices with 0.1 % Triton X-100 were blocked in 1 % BSA in PBS for 1 h. Slices were incubated with anti-calbindin (Abcam, 108404) overnight at 4 °C and followed by incubation with the secondary antibody conjugated with Alexa Fluor 488 for 2 h. The slices were then washed with PBS three times and incubated with DAPI in dark for 10 min. All the images were captured by an Olympus or Zeiss confocal microscope.

Isolation of Primary MEF and macrophage cells

MEF cells were prepared from neonatal mice, which were euthanized by CO₂. Skin of neonatal mice were cut with scissors and gently clipped and added to 0.25 % trypsin-EDTA in the 37 °C incubator for 20 min. The cell suspension was then transferred to a 50 ml tube and 10 ml of DMEM media was added to inactivate the trypsin reaction for 5 min. The supernatant was transferred to a 60 mm dish and keep in incubator at 37 °C for 3 h, then the medium was replaced by DMEM with 10 % FBS, 1 % penicillin/streptomycin until confluency (2 to 4 days). Primary macrophages were established from hindlimb femurs and tibias of newborn mice. The marrow cells were flushed from the bones with PBS and centrifuged. Cells were then resuspended in RPMI 1640 medium (Gibco, B122656) supplemented with 20 % FBS. Cells were then seeded in culture dishes coated with 2 % gelatin and allowed to adhere for 2 h at 37 ℃.

Reagents

Chemicals used in this study including SFN (Sigma-Aldrich, S4441), DMSO (Sigma-Aldrich, D2660), Filipin III (Cayman Chemical, 70440), U18666A (MedChemExpress, HY-107433), Triton X-100 (MCE, HY-Y1883A).

Data analysis

Data are presented as mean ± s.e.m. from at least 3 independent experiments. Statistical comparisons were performed with analyzes of variance (ANOVA) or Student’s t- test with paired or unpaired wherever appropriate. A p value < 0.05 was considered as statistically significant.

Acknowledgements

This work was supported by a NSFC grant (31600823 to D. L). Additional support was provided by Calygene Biotechnology Inc. (XT [2016]008@). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. We are grateful to Dr. Shawn M. Ferguson for the GFP-TFEB stable cells and Dr. Haoxing Xu for the TFEB KO cells.

Disclosure statement

No potential conflict of interest was reported by the authors.

Abbreviations

• ACP: acid phosphatase

• AREs: antioxidant response elements

• ANOVA: analyses of variance

• BBB: blood–brain barrier

• BSA: bovine serum albumin

• CLEAR: coordinated lysosomal expression and regulation

• FBS: fetal bovine serum

• KD: knockdown

• KO: knockout

• LAMP1: lysosomal associated membrane protein 1

• LSDs: lysosomal storage diseases

• MEF: mouse embryonic fibroblast

• MiTF: microphthalmia/TFE transcriptional factor

• MTOR: mechanistic target of rapamycin kinase complex 1

• NAGase: β- hexosaminidase

• NPC: Niemann-Pick type C disease

• OE: overexpression

• PM: plasma membrane

• RPS6 KB1/ p70S6K: ribosomal protein S6 kinase B1

• SFN: sulforaphane

• TFEB: transcription factor EB.