Fig 2: Loss of NDST3 promotes the assembly of the V-ATPase V1-V0 holoenzyme ATo examine the levels of the V-ATPase V1-V0 holoenzyme in WT and NDST3 KO RPE1 cells, cytosolic and membrane fractions were analyzed by immunoblotting using antibodies against the V-ATPase subunits V1A (ATP6V1A), V1C1 (ATP6V1C1), and V0D (ATP6V0D). LAMP1 and vinculin were used as loading controls for membrane proteins and cytosolic proteins, respectively.BQuantification of the levels of ATP6V0D relative to LAMP1 in the membrane fraction from (A) (n = 3 independent experiments, P = 0.7786).C, DRatios of ATP6V1A (C) and ATP6V1C1 (D) to ATP6V0D in the membrane fraction from (A) to represent the assembly of the V-ATPase V1-V0 holoenzyme (n = 3 independent experiments, ***P = 0.0005, ****P < 0.0001). Data information: Error bars represent ± standard deviation. Source data are available online for this figure.

Fig 3: In vitro assay for pH-dependent maturation of Cathepsin B and the colocalization of Cat B with LAMP1 (related to Fig 2) In vitro assay for Cat B maturation under various pH conditions. Recombinant human Cat B was incubated in activation buffers with different pH values (pH4.0~pH6.5) at room temperature for 10 min and then subjected to immunoblotting analysis using an antibody against Cat B. The ratio of mature Cat B to ProCat B was quantified (n = 3 independent experiments, *P = 0.0177).The colocalization of Cat B with LAMP1 is not altered by the loss of NDST3. WT and NDST3 KO cells were fixed and subjected to immunofluorescent staining using antibodies against Cat B and LAMP1. The nucleus was stained with DAPI. Representative images of the staining are shown, and Pearson’s correlation is used to evaluate the colocalization of Cat B and LAMP1 (n = 37 cells in the WT group, n = 38 cells in the NDST3 KO group, P = 0.2959). Scale bar, 10 µm. Data information: Error bars represent ± standard deviation.Source data are available online for this figure.

Fig 4: Impacts of Arg145Pro on the expression of autophagy related proteins. (A) HeLa cells were transfected with wtCLN5 or mtCLN5, and the expression levels of autophagy related proteins LC3II, LAMP-1 and p62/SQSTM1 were detected by Western blot using corresponding antibodies with GAPDH as a loading control. One representative experiment out of three is shown. (B) Relative intensity of GAPDH-loaded autophagy related proteins in wtCLN5 and mtCLN5 transfected cells from 3 independent experiments. Gray scale scanning was performed by Image J software and the relative expression level of LC3II, LAMP-1 and p62/SQSTM1 in mtCLN5 transfected cells was normalized by that in wtCLN5 transfected cells.

Fig 5: Functional analysis of p.Arg145Pro. (A,B) Impacts of Arg145Pro on the expression of CLN5. HeLa cells were transfected with wtCLN5 or mtCLN5 and total expression of CLN5 was detected by Western blot using antibody against CLN5 or anti-HA tag IgG with GAPDH as a loading control. One representative experiment out of three is shown. (C) Impacts of Arg145Pro on the cellular localization of CLN5. Cells transfected with wtCLN5 or mtCLN5 were treated with 50µM cycloheximide for 5 h and then fixed with 4% paraformaldehyde and subsequently co-stained with anti-HA tag antibody (red) and antibody against LAMP-1 specifically recognizing lysosome (green). Co-localization of CLN5 and lysosome was analyzed by confocal microscopy. Transfected cells with no treatment are control. Representative confocal images from three independent experiments are shown.