NPC1 plays a role in the trafficking of specific cargo to melanosomes

Abstract

Niemann-Pick type C1 (NPC1) protein is a multimembrane spanning protein of the lysosome limiting membrane that facilitates intracellular cholesterol and sphingolipid transport. Loss-of-function mutations in the NPC1 protein cause Niemann-Pick disease type C1, a lysosomal storage disorder characterized by the accumulation of cholesterol and sphingolipids within lysosomes. To investigate whether the NPC1 protein could also play a role in the maturation of the endolysosomal pathway, here, we have investigated its role in a lysosome-related organelle, the melanosome. Using a NPC1-KO melanoma cell model, we found that the cellular phenotype of Niemann-Pick disease type C1 is associated with a decreased pigmentation accompanied by low expression of the melanogenic enzyme tyrosinase. We propose that the defective processing and localization of tyrosinase, occurring in the absence of NPC1, is a major determinant of the pigmentation impairment in NPC1-KO cells. Along with tyrosinase, two other pigmentation genes, tyrosinase-related protein 1 and Dopachrome-tautomerase have lower protein levels in NPC1 deficient cells. In contrast with the decrease in pigmentation-related protein expression, we also found a significant intracellular accumulation of mature PMEL17, the structural protein of melanosomes. As opposed to the normal dendritic localization of melanosomes, the disruption of melanosome matrix generation in NPC1 deficient cells causes an accumulation of immature melanosomes adjacent to the plasma membrane. Together with the melanosomal localization of NPC1 in WT cells, these findings suggest that NPC1 is directly involved in tyrosinase transport from the trans-Golgi network to melanosomes and melanosome maturation, indicating a novel function for NPC1.

Keywords: NPC1; Niemann-Pick disease type C; PMEL17; endosomes; lysosome-related organelles; melanosomes biogenesis; traffic; tyrosinase.

Figure 1

Generation of the NPC1-KO cell line.A, immunoblot analysis of NPC1 protein in MNT-WT, MNT-bulk, and NPC1-KO (four clones). Calnexin (CNX) was used as loading control. B, relative lysosomal volume of NPC1-KO cells (four clones) to MNT-WT cells measured by LysoTracker staining and FACS analysis. Data analysis was based on mean fluorescence of three independent experiments and represented as mean ± SD (one-way ANOVA analysis; ∗∗∗∗p < 0.0001). C, immunofluorescence staining with anti-LAMP-2; the scale bar represents 10 μm. D, graphic representation of mean fluorescence intensity of LAMP-2 immunostaining, n = 52 cells (two-tailed student’s t test; ∗∗∗∗p < 0.0001). E and F, immunoblot analysis and quantification of LAMP-2 (E) and cathepsin D (F) in MNT-WT versus NPC1-KO (clone 2). Band densitometry plot of LAMP-2 (n = 5 biological replicates) and cathepsin D (n = 3 biological replicates) was represented as mean ± SD. Two-tailed Student’s t test; ∗∗∗∗p < 0.0001,∗p < 0.05. FACS, fluorescence-activated cell sorting; LAMP, lysosome-associated membrane protein; NPC1, Niemann–Pick type C1.

Figure 2

Characterization of the lipid storage phenotypes in the NPC1-KO cell line.A, cholesterol measurement by Amplex Red assay, data are presented as the mean ± SD, MNT-WT versus NPC1-KO, n = 3 biological replicates. B, D, and E, assay for total GSLs (B), for different GSL species (D) and for GM2, the main storage species (E) by HPLC, n = 2 biological replicates. C, representative GSLs trace from MNT-WT cells. F, the mean ± SD of GlcCer in MNT-WT and NPC1-KO, n = 3 biological replicates. NPC1, Niemann–Pick type C1; GlcCer, glucosylceramidase; GSL, glycosphingolipid.

Figure 3

NPC1-KO cells present pigmentation defects.A, visual comparison of MNT-WT and NPC1-KO cell pellets. B, C, and D, immunoblot analysis and quantification of TYR in NPC1-KO (clone 3 and 2) versus MNT-WT cells (B), in MNT-WT cells transfected with siControl (CTRL) or with siNPC1 (C) and in NPC1-KO cells transfected with empty vector—pcDNA3.1+ or with hNPC1pcDNA3.1+ (hNPC1) (D). Intensities of TYR protein bands were quantified and represented as mean ± SD (two-tailed Student’s t test; ∗∗p < 0.01; ∗p < 0.05, n = 3 biological replicates). E, co-immunoprecipitation of NPC1 protein using rabbit anti-NPC1 antibody in MNT-WT and NPC1-KO cell lysates was assessed. The resulted immunocomplexes were eluted with LB, and NPC1 and TYR were visualized by immunoblotting (n = 3 biological replicates). F, in-gel analysis of TYR activity. G, TYR relative gene expression level determined by semiquantitative qRT-PCR, n = 3 biological replicates. H–J, immunoblot analysis and quantification of MITF (H), TYRP-1 (I), and DCT (J) protein levels normalized to CNX or actin, divided by WT average and represented as mean ± SD, n = 3 biological replicates. CNX, calnexin; DCT, dopachrome-tautomerase; MITF, microphthalmia-associated transcription factor; NPC1, Niemann–Pick type C1; qRT-PCR, real time quantitative reverse transcription PCR; TYR, tyrosinase; TYRP, tyrosinase-related protein.

Figure 4

Proteome analysis of MNT-WT versus NPC1-KO cells by mass spectrometry.A, principal component analysis (PCA) plot showing clustering of replicates according to their characteristics. B, volcano plot indicating protein expression differences of MNT-WT versus NPC1-KO according to two sample t test (p < 0.05 and absolute log2FC ≥ 1), n = 3 biological replicates. C, heatmap of log2 transformed LFQ intensity values showing the statistically significant upregulated and downregulated proteins in NPC1-KO cell line. D, box plot of log2 transformed LFQ intensity values of upregulated proteins involved in lipid metabolism. E, pie chart presenting protein classes of significantly downregulated protein in NPC1-KO cell line based on the Panther database. F, dot plot of log2 transformed LFQ intensity of selected proteins belonging to melanogenesis pathway. LFQ, label-free quantitation; NPC1, Niemann–Pick type C1.

Figure 5

NPC1 impairs TYR processing and degradation.A, cells were untreated or treated with bafilomycin A (100 nM, 5 h) and analyzed by Western blotting for TYR N-glycan processing after Endo H and PNGase F glycosidase treatment of cell lysates. Relative expression levels of PNGase F-sensitive TYR and the ratio of Endo H-resistant to Endo H-sensitive TYR were plotted. Quantitative representation as mean ± SD of relative protein expressions and the ratio of Endo H resistant to Endo H sensitive bands (one-way ANOVA analysis; ∗p < 0.05, ∗∗p < 0.01, n = 3 biological replicates). B, visual comparison of cell pellets of MNT-WT and NPC1-KO cells untreated or treated with bafilomycin A1 (100 nM, 5 h). C, immunoblot analysis for TYR, LAMP-2, and CNX in cell lysates and in the enrichment of secreted extracellular vesicles from culture supernatant. D, effect of MG132 (50 μM, 5 h) or inhibitor mix (INH = 14.5 uM leupeptin and 106 uM pepstatin, 5 h) cell treatment on TYR expression, analyzed by Western blotting and densitometry, represented as fold change relative to cells treated with DMSO. Quantitative representation as mean ± SD of relative protein expression, n = 3 biological replicates (one-way ANOVA analysis; ∗p < 0.05). E, CHX treated MNT-WT and NPC1-KO cells analyzed by immunoblotting for TYR normalized to tubulin and graphic representation of TYR degradation rate as a percentage of control (CHX, 0 h) (n = 2 biological replicates). CHX, cycloheximide; CNX, calnexin; DMSO, dimethyl sulfoxide; Endo H, endoglycosidase H; LAMP, lysosome-associated membrane protein; NPC1, Niemann–Pick type C1; PNGase F, peptide-N glycosidase F; TYR, tyrosinase.

Figure 6

Altered distribution of proteins involved in melanogenesis. Co-localization of TYR with RAB38, TYRP-1, CD63, and PMEL17 (HMB-45) by confocal immunofluorescence microscopy (the scale bar represents 10 μm). The Manders’ correlation coefficient was plotted for the measurement of the co-localization of TYR with RAB38 (n = 33), TYRP-1 (n = 22), CD63 (n = 25) and HMB-45 (n = 35). Two-tailed Student’s t test was applied for statistical analysis (∗∗p < 0.01, ∗∗∗p < 0.001). HMB, human melanoma black; TYR, tyrosinase.

Figure 7

Investigation of the PMEL17 pathway in the absence of NPC1.A, immunoblot analysis of PMEL17 protein recognized by PEP-13 (immature PMEL17-P1 and M-beta) and HMB-45 (M-alpha/P2, M-alpha-C and RPT) antibodies. The insoluble fraction was extracted with 1% SDS and 50 mM DTT and boiled at 95 °C. The KO HMB-45 positive fraction was normalized to loading control, and divided by WT average and the relative fold change was represented as mean ± SD (Two-tailed Student’s t test; ∗∗∗p <0.001, n = 3 biological replicates). B, the surface expression of PMEL17 analyzed by flow cytometry represented in a bar plot as the median of fluorescence related to MNT-WT. C, immunoblot analysis of ApoE protein and graphical representation as mean ± SD of relative protein expressions(Two-tailed Student’s t test; ∗p <0.05, n = 3 biological replicates). D, subcellular fractionation of MNT-WT and NPC1-KO cell homogenates on a sucrose gradient. Twelve fractions were collected and analyzed by immunoblotting with antibodies directed against NPC1, EEA1, LAMP-2, TYR, TYRP-1, and HMB-45. Graphic representation of quantified proteins expression level as mean of two independent experiments. For TYR, three independent replicates were quantified and plotted as mean ± SD. ApoE, apolipoprotein E; HMB, human melanoma black; LAMP, lysosome-associated membrane protein; NPC1, Niemann–Pick type C1; TYR, tyrosinase; TYRP, tyrosinase-related protein.

Figure 8

Schematic representation of intracellular TYR trafficking during melanosome maturation. In this model, we propose a model of the mechanism of melanosome biogenesis by highlighting PMEL17 and the TYR pathways in NPC1-KO cells compared to MNT-WT cells. In MNT-WT, as in NPC1-KO cells the PMEL17 protein is trafficked from the TGN to the plasma membrane (PM) and is then endocytosed, while TYR is targeted to early endosomes directly or via PM. The early endosomes progressively mature into MVB, from which late endosomes, premelanosome or multiple vesicle exosomes (MVE) are developed. As illustrated, exosome secretion is not affected in NPC1 deficient cells mutant cells, while the pathways which lead to melanosome maturation are impaired in NPC1 KO cells. PMEL17 is sorted to intraluminal vesicles in premelanosome and forms a sheet-like matrix of amyloid fibrils, where melanin can be deposited. In the absence of the NPC1 protein, PMEL17 fibrils accumulates and the maturation of the immature melanosomes is perturbed (dashed arrows). In MNT-WT cells, TYR is sorted from the MVB to immature melanosomes, where it initiates the melanin biosynthesis process. In contrast, in NPC1-deficient cells, TYR delivery to immature melanosomes is reduced, due to its target misrouting for lysosomal degradation. The key to the proteins indicated in the figure are presented in the right hand panel. The modified pathways in NPC1-deficient cells are indicated by dashed red arrows. MVB, multivesicular bodies; NPC1, Niemann–Pick type C1; TGN, trans-Golgi network; TYR, tyrosinase.