Juvenile CLN3 disease is a lysosomal cholesterol storage disorder: similarities with Niemann-Pick type C disease

Abstract

Background: The most common form of neuronal ceroid lipofuscinosis (NCL) is juvenile CLN3 disease (JNCL), a currently incurable neurodegenerative disorder caused by mutations in the CLN3 gene. Based on our previous work and on the premise that CLN3 affects the trafficking of the cation-independent mannose-6 phosphate receptor and its ligand NPC2, we hypothesised that dysfunction of CLN3 leads to the aberrant accumulation of cholesterol in the late endosomes/lysosomes (LE/Lys) of JNCL patients' brains.

Methods: An immunopurification strategy was used to isolate intact LE/Lys from frozen autopsy brain samples. LE/Lys isolated from samples of JNCL patients were compared with age-matched unaffected controls and Niemann-Pick Type C (NPC) disease patients. Indeed, mutations in NPC1 or NPC2 result in the accumulation of cholesterol in LE/Lys of NPC disease samples, thus providing a positive control. The lipid and protein content of LE/Lys was then analysed using lipidomics and proteomics, respectively.

Findings: Lipid and protein profiles of LE/Lys isolated from JNCL patients were profoundly altered compared to controls. Importantly, cholesterol accumulated in LE/Lys of JNCL samples to a comparable extent than in NPC samples. Lipid profiles of LE/Lys were similar in JNCL and NPC patients, except for levels of bis(monoacylglycero)phosphate (BMP). Protein profiles detected in LE/Lys of JNCL and NPC patients appeared identical, except for levels of NPC1.

Interpretation: Our results support that JNCL is a lysosomal cholesterol storage disorder. Our findings also support that JNCL and NPC disease share pathogenic pathways leading to aberrant lysosomal accumulation of lipids and proteins, and thus suggest that the treatments available for NPC disease may be beneficial to JNCL patients. This work opens new avenues for further mechanistic studies in model systems of JNCL and possible therapeutic interventions for this disorder.

Funding: San Francisco Foundation.

Keywords: Cholesterol; Human brain tissue; Juvenile CLN3 disease; Lysosomes; Niemann–Pick type C disease; Proteomics.

Fig. 1

Lipidomic profiling of LE/Lys isolated from brain tissue of JNCL, NPC and control individuals. Quantification of 34 lipid classes by LC-MS/MS. Grey bars: controls (n = 6 individuals); dark blue bars: JNCL (n = 5 individuals); light blue bars: NPC (n = 3 individuals). ∗, ∗∗ and ∗∗∗ stand for p < 0.05, p < 0.01 and p < 0.001 in multiple unpaired t-tests with Welch's correction, respectively. Red asterisks indicate comparisons with controls, while blue asterisks indicate comparisons between JNCL and NPC. All values are given as mean ± s.e.m.

Fig. 2

Venn diagram of unique and shared proteins detected in LE/Lys isolated from brain tissue of JNCL, NPC and control individuals. Blue: controls (n = 6 individuals); red: JNCL (n = 5 individuals); green: NPC (n = 4 individuals). Representation was generated with https://bioinformatics.psb.ugent.be/webtools/Venn/.

Fig. 3

Gene ontology (GO) analysis of differentially expressed proteins in JNCL LE/Lys samples compared to controls. a, upregulated proteins. b, downregulated proteins. Significantly changed protein abundance was determined by multiple t-tests with a threshold for significance of p < 0.05 (with a permutation-based FDR correction). Only proteins that were upregulated or downregulated by more than 1.5 fold change (FC) were considered for analysis. GO enrichment was significant with FDR<1%.

Fig. 4

Gene ontology (GO) analysis of differentially expressed proteins in NPC LE/Lys samples compared to controls. a, upregulated proteins. b, downregulated proteins. Significantly changed protein abundance was determined by multiple t-tests with a threshold for significance of p < 0.05 (with a permutation-based FDR correction). Only proteins that were upregulated or downregulated by more than 1.5 fold change (FC) were considered for analysis. GO enrichment was significant with FDR<1%.

Fig. 5

Overrepresented functional pathways in differentially expressed proteins. a, JNCL vs. control samples. Left panel, upregulated proteins. Right panel, downregulated proteins. b, NPC vs. control samples. Left panel, upregulated proteins. Right panel, downregulated proteins. Representation was generated with reactome.org.