Novel bicistronic lentiviral vectors correct beta-Hexosaminidase deficiency in neural and hematopoietic stem cells and progeny: implications for in vivo and ex vivo gene therapy of GM2 gangliosidosis

The favorable outcome of in vivo and ex vivo gene therapy approaches in several Lysosomal Storage Diseases suggests that these treatment strategies might equally benefit GM2 gangliosidosis. Tay-Sachs and Sandhoff disease (the main forms of GM2 gangliosidosis) result from mutations in either the HEXA or HERB genes encoding, respectively, the alpha- or beta-subunits of the lysosomal beta-Hexosaminidase enzyme. In physiological conditions, alpha- and beta-subunits combine to generate beta-Hexosaminidase A (HexA, alpha beta) and beta-Hexosaminidase B (HexB, 1313). A major impairment to establishing in vivo or ex vivo gene therapy for GM2 gangliosidosis is the need to synthesize the alpha- and beta-subunits at high levels and with the correct stoichiometric ratio, and to safely deliver the therapeutic products to all affected tissues/organs. Here, we report the generation and in vitro validation of novel bicistronic lentiviral vectors (LVs) encoding for both the murine and human codon optimized Hexa and Hex!) genes. We show that these LVs drive the safe and coordinate expression of the alpha- and beta-subunits, leading to supranormal levels of beta-Hexosaminidase activity with prevalent formation of a functional HexA in SD murine neurons and glia, murine bone marrow-derived hematopoietic stem/progenitor cells (HSPCs), and human SD fibroblasts. The restoration/overexpression of beta-Hexosaminidase leads to the reduction of intracellular GM2 ganglioside storage in transduced and in cross-corrected SD murine neural progeny, indicating that the transgenic enzyme is secreted and functional. Importantly, bicistronic LVs safely and efficiently transduce human neurons/glia and CD34 + HSPCs, which are target and effector cells, respectively, in prospective in vivo and ex vivo GT approaches. We anticipate that these bicistronic LVs may overcome the current requirement of two vectors co-delivering the alpha- or beta-subunits genes. Careful assessment of the safety and therapeutic potential of these bicistronic LVs in the SD murine model will pave the way to the clinical development of LV-based gene therapy for GM2 gangliosidosis.