Gaucher disease iPSC-derived osteoblasts have developmental and lysosomal defects that impair bone matrix deposition

Abstract

Gaucher disease (GD) is caused by bi-allelic mutations in GBA1, the gene that encodes acid β-glucocerebrosidase (GCase). Individuals affected by GD have hematologic, visceral and bone abnormalities, and in severe cases there is also neurodegeneration. To shed light on the mechanisms by which mutant GBA1 causes bone disease, we examined the ability of human induced pluripotent stem cells (iPSC) derived from patients with Types 1, 2 and 3 GD, to differentiate to osteoblasts and carry out bone deposition. Differentiation of GD iPSC to osteoblasts revealed that these cells had developmental defects and lysosomal abnormalities that interfered with bone matrix deposition. Compared with controls, GD iPSC-derived osteoblasts exhibited reduced expression of osteoblast differentiation markers, and bone matrix protein and mineral deposition were defective. Concomitantly, canonical Wnt/β catenin signaling in the mutant osteoblasts was downregulated, whereas pharmacological Wnt activation with the GSK3β inhibitor CHIR99021 rescued GD osteoblast differentiation and bone matrix deposition. Importantly, incubation with recombinant GCase (rGCase) rescued the differentiation and bone-forming ability of GD osteoblasts, demonstrating that the abnormal GD phenotype was caused by GCase deficiency. GD osteoblasts were also defective in their ability to carry out Ca2+-dependent exocytosis, a lysosomal function that is necessary for bone matrix deposition. We conclude that normal GCase enzymatic activity is required for the differentiation and bone-forming activity of osteoblasts. Furthermore, the rescue of bone matrix deposition by pharmacological activation of Wnt/β catenin in GD osteoblasts uncovers a new therapeutic target for the treatment of bone abnormalities in GD.

Figure 1.

Characterization of mesenchymal stem cells and osteoblasts derived from control and GD iPSC. (A) Flow cytometry analysis of iPSC-derived control MSC. Scatter plots show staining for the specific markers of MSC, CD29, CD44 and HLA-ABC, and staining with anti-CD45 as a negative control. Isotype controls are shown at the left. (B) qRT-PCR analysis showing the expression of osteoblast markers in iPSC-derived control and GD MSC and osteoblasts as indicated. Results are expressed as fold-change of each osteoblast line compared with its corresponding MSC line (mean ± SEM). P values for control, GD2a and GD3a for each marker are as follows: ALP (0.003, 0.016 and 0.002), Col1 (0.002, 0.372 and 0.049), RUNX2 (0.020, 0.034 and 0.049). (C) Alkaline phosphatase stain in control and GD2 osteoblasts. (D) Alizarin red stain showing the mineral deposits in control and GD2 osteoblast cultures. Scale bar, 50 µm.

Figure 2.

GCase expression in control and GD osteoblasts and phenotypic rescue by rGCase. (A) Left panel, representative immunoblot of GCase expression in control and GD2 iPSC-derived osteoblasts using specific antibodies to GCase. Right panel, quantitation of immunoblot analysis of GCase expression in control and GD2 osteoblasts. Results are expressed as fold-change compared with control osteoblasts (n = 3, ***P < 0.001). (B) GD2 iPSC-derived MSC were differentiated to osteoblasts in the absence or presence of 0.24 U/ml rGCase for 3 weeks. After differentiation, the cultures were stained for alkaline phosphatase (ALP) (upper panel) or Alizarin red (lower panel). Scale bar, 50 µm.

Figure 3.

GD osteoblasts have lower levels of total and active β-catenin. (A) Control and GD2 iPSC-derived osteoblasts were incubated in the absence or presence of 0.24 U/ml rGCase for 5 days. The cultures were then stained using anti-β catenin antibody (red). Nuclei were stained with DAPI (blue). Magnification, ×40. Scale bar, 25 µm. (B) Control and GD2 osteoblasts were incubated in the absence or presence of 0.24 U/ml rGCase for 5 days. The cultures were then stained with antibodies to active β catenin (non-phosphorylated) (red). Nuclei were stained with DAPI (blue). Magnification, ×40. Scale bar, 25 µm. (C) Left panel, representative immunoblot showing the expression of active β catenin in control and GD2 osteoblasts that were incubated in the absence or presence of 0.24 U/ml rGCase. Right panel, quantitation of immunoblot analysis of active β catenin expression in control and GD2 osteoblasts; Results are expressed as fold-change with respect to non-treated control osteoblasts (mean ± SEM, n = 3). ***P < 0.001 (one-way ANOVA). ***P < 0.001 between control versus GD2, **P < 0.01 between GD2 versus GD2 + rGCase.

Figure 4.

Lysosomal depletion in GD osteoblasts and rescue by rGCase. (A) Control and GD2 osteoblasts were incubated in the absence (NT) or presence of 0.24 U/ml rGCase for 5 days. The cultures were then stained with anti-LAMP1 antibody (green). Nuclei were stained with DAPI (blue). Magnification, ×40. Scale bar, 25 µm. (B) Quantitation of mean fluorescence intensity (MFI) of LAMP1 expression in control and GD2 osteoblasts. Results are expressed as fold-change with respect to non-treated control osteoblasts (mean ± SEM, n = 3). ***P < 0.001 (one-way ANOVA). ***P < 0.001 between control versus GD2, **P < 0.01 between GD2 versus GD2 + rGCase. (C) Representative immunoblot showing the expression of LAMP1 in control and GD2 osteoblasts. (D) Quantitation of immunoblot analysis of LAMP1 expression in control and GD2 osteoblasts; results are expressed as fold-change compared with control osteoblasts (mean ± SEM, n = 3, *P < 0.05).

Figure 5.

Reduced lysosomal enzyme activity and exocytosis in GD osteoblasts. (A) Cell lysates from control (green) and GD2 osteoblasts (blue) were assayed for activity of the lysosomal enzymes Cathepsins B, L, D and acid sphingomyelinase (ASM) as described in the section ‘Materials and Methods’. Plots represent total enzymatic activities in control and GD2 osteoblasts. (B) Lysosomal exocytosis of Cathepsins B, L and ASM in SLO-permeabilized control and GD2 osteoblasts. Control and GD2 osteoblasts were treated with SLO, and the activities of Cathepsins B, L and ASM (mean ± SEM, n = 3, **P < 0.01) in the supernatant were assayed as described in the section ‘Materials and Methods’. Green, control osteoblasts; blue, GD2 osteoblasts. RFU, relative fluorescence units. (C) Exocytosis efficiency for Cathepsin L in control and GD2 osteoblasts was measured and expressed as the percentage of total Cathepsin L activity present in whole cell lysates of each cell type, as described in the section ‘Materials and Methods’. Green, control osteoblasts; blue, GD2 osteoblasts. The results show that after SLO permeabilization, GD2 osteoblasts have defective Ca²⁺-dependent lysosomal exocytosis.

Figure 6.

GD osteoblasts exhibit defective plasma membrane repair. Untreated and SLO-treated control and GD2 osteoblasts were incubated in the presence or absence of Ca²⁺ as described in the section ‘Materials and Methods’, and the samples were analyzed by FACS. (A) Cells without SLO and without Ca²⁺. (B) Cells with SLO and without Ca²⁺. (C) Membrane repair in the presence of both SLO and Ca²⁺. Red, control osteoblasts; blue GD2 osteoblasts. In the assay shown, which is representative of several independent experiments, the extent of membrane repair was 48.9% in GD osteoblasts, and 96.2% in control osteoblasts.