In vitro modelling and characterization of cells involved in Gaucher Disease-related bone defects

Gaucher disease (GD) is an autosomal recessive lysosomal storage disorder caused by mutations in the acid β-glucosidase encoding gene (GBA1), resulting in deficient activity of acid β-glucosidase (GCase) and the subsequent accumulation of glucosylceramide (GlcCer) within lysosomes. Up to 90% of GD patients report bone symptoms that severely reduce their quality of life. The molecular basis of bone involvement in GD is unclear. However, several lines of evidence suggest that both osteoblasts and osteoclasts play a role in the process, whereas almost no information is available about cartilage cells. For these reasons, the general objective of this project was to explore the role played by different cells in GD bone pathology. Specifically, we (1) developed GD models of osteoclasts, osteoblasts, and chondrocytes (GBA KO cells) by using CRISPR/Cas technology to knock out (KO) the GBA1 gene in continuous cell lines relevant to study bone pathophysiology, (2) characterized the different features of wild-type (wt) and GBA KO cells, and (3) examine cell-to-cell interactions. Firstly, we modelled GD monocytes using the THP-1 cell line. We observed increased osteoclastogenesis of THP-1 GBA KO compared with THP-1 wt. As the number of differentiated osteoclasts was reduced by providing THP-1 GBA KO with the recombinant GCase, an inhibitor of GlcCer synthesis, or an anti-inflammatory compound, we demonstrated that lipid accumulation and inflammation play a central role in this process. Secondly, we modelled GD osteoblasts using the osteosarcoma cell line SaOS and characterized these cells in terms of type I collagen, which is produced by osteoblasts and is the main constituent of the organic component of bone extracellular matrix (ECM), and alkaline phosphatase (ALP), an enzyme required for the formation of hydroxyapatite crystals, which represent the inorganic component of bone ECM. Compared with SaOS wt, SaOS GBA KO showed a slight decrease in ALP mRNA expression and calcium deposition, and a significant decrease in type I collagen mRNA expression and deposition. Moreover, we performed the miRNA profile of these cells in cell lysates and exosomal preparations using NGS technology. Among the significantly up- and down-regulated candidates, we focused on miR-488-3p, which was strongly overexpressed by GBA KO cells. We further validated the overexpression of miR-488-3p in primary osteoblasts treated with the GCase inhibitor CBE to mimic GD. Furthermore, to characterize the functional impact of miR-488-3p overexpression, we transfected it into SaOS wt and we showed that it downregulates the expression of ALP- and collagen type I-encoding genes, probably by downregulating their transcription factor RUNX2. Taken together, these data suggest an impairment of matrix deposition by GD osteoblasts that can be explained, at least in part, by the action of the overexpressed miR-488-3p. Thirdly, we model GD chondrocytes using the C28I2 cell line. As we found decreased expression of collagen type II- and increased expression of collagen type X- encoding genes in GBA KO cells, we hypothesize changes in matrix composition and a different fate of GD chondrocytes compared with normal cells. In addition, these cells release more interleukin-8 and attract more neutrophils than wt: these observations may explain the pain experienced by GD patients. Finally, we examined the role of GD osteoblasts and chondrocytes in osteoclastogenesis. We found increased osteoclastogenesis when the conditioned medium (CM) of SaOS GBA KO cells was used to differentiate THP-1 wt, likely due to increased release of the receptor activator of NF-kB ligand (RANKL). Furthermore, in a preliminary experiment, we found a similar trend in osteoclast formation when C28I2 GBA KO-CM was used, suggesting a possible role of GD chondrocytes in enhancing osteoclastogenesis. In conclusion, we provide evidence for the role of osteoclasts, osteoblasts, and chondrocytes in GD.