Endoplasmic reticulum and lysosomal Ca²⁺ stores are remodelled in GBA1-linked Parkinson disease patient fibroblasts

Abstract

Mutations in β-glucocerebrosidase (encoded by GBA1) cause Gaucher disease (GD), a lysosomal storage disorder, and increase the risk of developing Parkinson disease (PD). The pathogenetic relationship between the two disorders is unclear. Here, we characterised Ca(2+) release in fibroblasts from type I GD and PD patients together with age-matched, asymptomatic carriers, all with the common N370S mutation in β-glucocerebrosidase. We show that endoplasmic reticulum (ER) Ca(2+) release was potentiated in GD and PD patient fibroblasts but not in cells from asymptomatic carriers. ER Ca(2+) signalling was also potentiated in fibroblasts from aged healthy subjects relative to younger individuals but not further increased in aged PD patient cells. Chemical or molecular inhibition of β-glucocerebrosidase in fibroblasts and a neuronal cell line did not affect ER Ca(2+) signalling suggesting defects are independent of enzymatic activity loss. Conversely, lysosomal Ca(2+) store content was reduced in PD fibroblasts and associated with age-dependent alterations in lysosomal morphology. Accelerated remodelling of Ca(2+) stores by pathogenic GBA1 mutations may therefore feature in PD.

Keywords: Ca(2+); Endoplasmic reticulum; Gaucher disease; Lysosomes; Parkinson disease.

Graphical abstract

Fig. 1

Pathogenic GBA1 disrupts ER Ca²⁺ release. (A–D) ER Ca²⁺ release in GBA1^wt/wt₅₅, GBA1^mut/mut₅₅^GD and GBA1^wt/mut₅₅^PD cells (young cohort). (A) Cytosolic Ca²⁺ recordings from individual fibroblasts challenged with thapsigargin (1 μM) from the indicated representative populations. Experiments were performed in the absence of extracellular Ca²⁺. (B) Summary data (mean ± SEM) quantifying the magnitude of thapsigargin-evoked Ca²⁺ signals in the indicated number of fields of view. Results are from 5 to 9 independent passages analysing 154–367 cells. (C) Cytosolic Ca²⁺ recordings from individual fibroblasts stimulated with cADPR-AM (25 μM). Experiments were performed in the presence of extracellular Ca²⁺. (D) Summary data quantifying the percentage of cells responsive to cADPR. Results are from 2 to 3 independent passages analysing 39–75 cells. (E) Similar to A except thapsigargin-evoked Ca²⁺ release was assessed in GBA1^wt/wt₅₅, GBA1^wt/mut₅₈^ASX and GBA1^wt/mut₅₅^PD cells. (F) Summary data from 4 independent passages analysing 46–127 cells. (G) Similar to C except cADPR-evoked Ca²⁺ release was assessed in GBA1^wt/wt₅₅, GBA1^wt/mut₅₈^ASX and GBA1^wt/mut₅₅^PD cells. (H) Summary data from 3 to 6 independent passages analysing 73–257 cells. *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.

Fig. 2

ER Ca²⁺ defects are age-dependent. (A) Cytosolic Ca²⁺ recordings from individual fibroblasts challenged with thapsigargin (1 μM) from representative populations of GBA1^wt/wt₇₈ and GBA1^wt/mut₇₅^PD cells (aged cohort). (B) Summary data 3 independent passages analysing 112–117 cells. (C) Similar to A except, ER Ca²⁺ release was assessed in GBA1^wt/wt₈₂ and GBA1^wt/mut₈₀^ASX cells. (D) Summary data from 3 independent passages analysing 131–134 cells. (E) ER Ca²⁺ release from GBA1^wt/wt fibroblasts with increasing age. (F) Summary data from 1 to 14 independent passages analysing 30–483 cells. (G) Magnitude of ER Ca²⁺ release versus age for both the young and aged cohort. All experiments were performed in the absence of extracellular Ca²⁺.

Fig. 3

Inhibition of β-glucocerebrosidase does not affect ER Ca²⁺ release. (A) Cytosolic Ca²⁺ recordings from individual control GBA1^wt/wt fibroblasts challenged with thapsigargin (1 μM) from a representative population treated with 10 μM CBE for 8 days. (B) Summary data from 2 independent treatments analysing 87–90 cells. (C–F) Cytosolic Ca²⁺ recordings from individual SH-SY5Y cells challenged with thapsigargin (1 μM) from a representative population treated with 10 μM CBE for 10–11 days (C) or stably expressing either scrambled shRNA (GBA1^+/+) or shRNA targeting GBA1 (GBA1^−/−) (E). Summary data from 3 independent treatments analysing 117–204 cells (D) and 3 independent passages analysing 150–143 cells (F). All experiments were performed in the absence of extracellular Ca²⁺. ns, not significant. Inset (F) is a Western blot using antibodies to β-glucocerebrosidase (top) or actin (bottom) and homogenates (14 μg) from SH-SY5Y cells treated with the indicated shRNA.

Fig. 4

Pathogenic GBA1 disrupts lysosomal morphology. (A–H) Representative confocal fluorescence images of LAMP1 staining (white) in the indicated fibroblasts from the young (A–D) and aged (E–H) cohort. Nuclei were stained with DAPI (blue). Zoomed images are displayed in the right panels. Scale bars, 10 μm. (I) Summary data quantifying LAMP1 intensity as a percentage of the indicated age-matched control (82–654 cells). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Pathogenic GBA1 disrupts lysosomal Ca²⁺ content. (A–D) Cytosolic Ca²⁺ measurements from individual fibroblasts stimulated with GPN (200 μM). Experiments were performed in the presence of extracellular Ca²⁺ and following a 12.5 min pre-treatment with 100 μM 2APB prior to recording. (A) Recordings from a representative population of GBA1^wt/wt₅₅, GBA1^mut/mut₅₅^GD and GBA1^wt/mut₅₅^PD cells. (B) Summary data from 2 independent passages analysing 72–88 cells. (C) Recordings from a representative population of control GBA1^wt/wt fibroblasts treated with 10 μM CBE for 7–9 days. (D) Summary data from 2 independent treatments analysing 43–72 cells. ***p < 0.001. ns, not significant.