Gaucher disease (GD) is caused by biallelic mutations in GBA1 gene. These mutations impair the function of the β-glucocerebrosidase (GCase) enzyme, leading to the accumulation of its substrate within lysosomes. Enzyme therapy fails to control the neuronopathic form of GD (nGD) due to its inability to cross the blood-brain barrier and reach the central nervous system (CNS). Previous studies have shown that lysosomal enzymes can be secreted and taken up by neighboring cells. Furthermore, under specific conditions, hematopoietic stem cells (HSC) can colonize the CNS and differentiate into microglia. Therefore, introducing a gene that drives high-level GCase expression into HSCs could represent an effective strategy to provide GCase to neural cells and treat patients with nGD. Mouse models do not properly reproduce the phenotype of nGD. As such, they are not suitable for evaluating the efficacy of therapeutic approaches.

Objective: To develop a multicellular in vitro model that recapitulates nGD phenotype.

Methods: We generated cerebral organoids (COs) derived from induced pluripotent stem cells (iPSCs) of a healthy individual, a patient with nGD (GBA1 ^L444P/L444P), and corrected iPSCs re-expressing wild-type (WT) GCase. We also developed microglia-containing organoids (MCO) by incorporating IPSC-derived erythromyeloid progenitors into 1-month-old COs.

Results: All MCOs, both mutant, normal and corrected contained neurons, astrocytes, and microglia in similar proportions. However, while differences in transcriptome and structural abnormalities of mitochondria (MT) and rough endoplasmic reticulum (RER) as well as astrogliosis were observed in mutant organoids compared to WT, they were reduced or absent in organoids derived from corrected cells. These results indicate that the abnormal phenotypes caused by the frequent L444P mutation involved in nGD can be rescued by providing the WT protein.

Conclusion: We established a multicellular 3D model that reproduces key features of nGD, including GCase deficiency, substrate accumulation, mitochondrial and RER abnormalities, as well as transcriptomic differences related to the presence or absence of WT GCase. Furthermore, we successfully generated MCOs. This model represents a promising tool to evaluate gene therapy approaches using lentiviral vector-transduced microglia.