Physiopathologie de la dégénérescence des motoneurones spinaux dans la forme adulte de la gangliosidose à GM2 à début tardif

GM2 gangliosidosis is a lysosomal storage disease caused by mutations in the HEXA and HEXB genes, characterized by a deficiency in β-hexosaminidase A (HexA) leading to an accumulation of GM2 ganglioside. GM2 gangliosidosis presents with heterogeneous neurological manifestations, which can begin at any age depending on the severity of the mutations and the degree of enzyme deficiency. In the late-onset form, motor function impairment, linked to the degeneration of the spinal motor neuron (SMN), is the earliest, most frequent, and most disabling of all clinical manifestations, not observed in children. The mechanisms underlying SMN degeneration are unknown, as there is currently no relevant animal model to study the late-onset form of GM2 gangliosidosis. I have established a novel and relevant cellular model of the late-onset form of GM2 gangliosidosis using induced pluripotent stem cell-derived SMNs from three different adult patients with the disease. I have characterized these lines and confirmed a marked reduction in the HexA α-subunit and HexA enzymatic activity. This deficiency leads in patient SMNs to early accumulation of GM2 as well as wider changes in other GM1 and GM3 ganglioside species on liquid chromatography-mass spectrometry. This is followed by the formation of GM2 clusters detected by immunostaining that partially colocalized with lysosomal (LAMP1) and late endosomal (RAB7) membrane markers. I also observed lysosomal morphological abnormalities at later stages of SMN maturation, including increased overall LAMP1 intensity and enlarged lysosomes on both immunostaining and transmission electron microscopy. These enlarged lysosomes were filled with membranous material organized in parallel layers, resembling membranous cytoplasmic bodies observed in brain tissue from GM2 gangliosidosis patients. Treatment with Venglustat, a glucosylceramide synthase inhibitor, at 1-3 μM efficiently reduced GM2 clusters when applied at a very early stage of SMN development, with a much more partial effect when the treatment was applied at a later stage. Higher concentrations (> 4 μM) led to neurodegeneration. Despite pronounced GM2 storage and lysosomal pathology, late-onset GM2 gangliosidosis SMNs displayed no signs of abnormal neuritogenesis nor early degeneration. However, neuronal activity was altered in microelectrode array recordings. Bulk RNA sequencing at two developmental stages, young SMN stage (day 14) and mature SMN stage (day 28) confirmed the successful differentiation and maturation of SMNs as well as a similar expression of motor neuron identity markers between patients and controls, supporting a similar neurodevelopmental trajectory. Gene Ontology enrichment analysis revealed a dysregulation of synapse-coding genes at day 28. Analysis of differentially expressed genes at day 28 revealed that coding for ion channels or regulator proteins (DPP6, KCNV1, RGS8), the glutamatergic synapse (GLRA1 and GRIP2) and the endolysosomal system (CTSF, PLIN3 and ARHGAP26) were among the top hits. Moreover, GM2 ganglioside reduction with Venglustat significantly reduced RGS8 and CTSF upregulation, with a nonsignificant reduction of KCNV1 and PLIN3 upregulation and increase in ARHGAP26 downregulation. In conclusion, I generated and fully characterized a disease-relevant cellular model to study SMN degeneration in late-onset GM2 gangliosidodisis. This model recapitulates hallmark pathological features of the disease and develop profound lysosomal alterations. In addition, patient-derived SMNs exhibited altered neuronal activity as well as early dysregulation of synaptic genes. This in vitro model can serve as a platform for preclinical testing of emerging therapies, and strategies aimed at correcting synaptic membrane abnormalities could represent a novel therapeutic avenue.