We have performed a series of preclinical investigations on gene therapy for Fabry disease. A recombinant retroviral gene transfer vector was designed that engineers efficient transduction of cells and expression of human alpha-Gal A activity. Enzymatic correction was observed in cultured skin fibroblasts and B cell lines derived from patients with Fabry disease. Corrected cells secreted significant quantities of alpha- Gal A into the culture medium. The secreted enzyme is taken up by uncorrected bystander cells. We have also achieved enzymatic correction in CD34+ and progenitor colony cells obtained from bone-marrow aspirates from patients with Fabry disease. A recombinant fusion protein has been engineered that combines the alpha-Gal A enzyme in frame with a carboxy- terminal eight amino acid addition called the FLAG peptide. This sequence is specific and antigenic when it is expressed. It can be detected with an antibody. This fusion protein will permit purification of extra- and intracellular forms of alpha-Gal A and has provided direct evidence of uptake into bystander cells and lysosomal localization. We are also producing retroviral vectors with selectable markers to increase expression by altering the viral or alpha-Gal A constructs or by sorting the transduced cells prior to implantation. We are also conducting experiments to optimize the transduction of CD34+ cells mobilized in peripheral blood. We have produced a high-titer clinical-grade vector under GMP conditions for gene therapy trials in Fabry disease. We are also exploring gene therapy in the alpha-Gal A knock-out mouse model of Fabry disease that we created. In addition, we have made gene transfer vector constructs based on HIV-1 that are highly effective in the transduction of quiescent cells including rat cerebellar neurons. This investigation has particular relevance to gene therapy since most stem cells in the bone marrow are non-dividing cells and those within the nervous system are in the post-mitotic state. We are initiating studies aimed at genetic correction of the Niemann-Pick C phenotype in vivo using the NPC mouse model.
