Generation of SIN-Lentiviral Vectors for the Treatment of Canavan Disease Using Ex Vivo Transduction of Hematopoietic Target Cells

The autosomal recessive leukodystrophy Canavan Disease is caused by an acquired or hereditary mutation in the Aspartoacylase gene (ASPA) leading to enzymatic dysfunction of a critical component of central nervous system metabolic and osmotic homeostasis. This loss of enzyme function results in demyelination, hydrocephalus, spastic paralysis, seizures, and ultimately premature death. Metabolic treatment for Canavan Disease can slow but not stop disease progression. Gene therapy aimed at replacing the defective ASPA gene has been employed in 13 patients using AAV-2 vectors expressing the human wild-type ASPA cDNA driven by a human Neuron Specific Enolase promoter and delivered via intra-parenchymal injection. Long-term follow-up in these patients revealed minimal clinical impact as although there was measurable physiologic activity of the transgene, it was likely less than what would be sufficient to alter the disease. It has been clearly demonstrated in animal models and in human patients that vector transduced hematopoietic long-term repopulating cells can differentiate into self-renewing microglial populations that can seed the CNS following busulfan treatment. To improve gene delivery to the CNS we have chosen to target the hematopoietic system with SIN-lentiviral vectors expressing wild-type ASPA linked via picornavirus 2A sequence to a GFP reporter gene driven by the highly active MND promoter and flanked by the 650bp cHS4 insulator element. We have produced this vector at laboratory scale with unconcentrated titers in the low 10⁷ t.u./ml range. We have demonstrated transmission of full length vector genome to both cell lines and human CD34+ primary cells. Single cell jurkat transducents were expanded to clonal cell lines which all demonstrated measurable ASPA protein by Western blot and functional ASPA enzyme activity by HPLC assay. Transduction of human mobilized peripheral blood CD34+ cells at an MOI of 26 (MOI 13 x 2 exposures) and plated in methylcellulose for CFU-C assay resulted in 40.6% GFP+ cells by flow cytometry and 0.69 vector genomes/ human genome by real-time PCR. Vector and mock treated cells were transplanted into busulfan conditioned NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ (NSG) mice and allowed to engraft. After 12 weeks in vivo, mice were sacrificed and the bone marrow extracted. No difference in overall (hCD45+) or lineage specific (hCD33+, 19+, or 71+) human cell engraftment was observed between the arms indicating a lack of toxicity, with a transduction efficiency of 15.3±7.4% GFP+ hCD45+ cells by flow cytometry and 0.23±0.15 vector genomes/ human genome by real- time PCR. Based on the success of these proof-of-concept experiments, we have produced reporterless versions expressing wild-type and human codon-optimized ASPA. We plan to test these vectors for hematopoietic toxicity and transduction efficiency in human primary cells using the NSG mouse xenotransplant model and to test for functional correction in the Nur7 Canavan Disease mouse model by transducing 5-FU treated Nur7 mouse bone marrow cells and transplanting them into busulfan conditioned Nur7 recipients. If these vectors continue to perform well in in vivo preclinical assays, it is our plan to begin large-scale production with translation into clinical trials.