373. Optimization of AAV-Gene Therapy for GM1-Gangliosidosis

Abstract

GM1-gangliosidosis is an autosomal recessive disease caused by mutations in the GLB1 gene encoding for the lysosomal acid beta-galactosidase (βgal) enzyme. The resulting enzymatic deficiency leads to accumulation of GM1-ganglioside in neurons and eventually massive neurodegeneration. The incidence of GM1-gangliosidosis is estimated at 1:100,000-200,000 live births, and there is currently no available treatment. Intracranial delivery of recombinant adeno-associated virus (AAV) vectors has been shown to be highly effective in animal models of this disease. The choice of injection sites and promoters driving transgene expression are important parameters to translate into clinical trials a safe and effective AAV-based therapy for GM1 gangliosidosis. We performed a study using AAVrh10 vectors to determine the safest and most effective promoter and delivery route combination for AAVrh10-βgal gene therapy in a GM1-mouse model. We used a combination of two direct intracranial injection site - bilateral thalamic and deep cerebral nuclei infusions (Th+DCN), or bilateral thalamic and CSF infusion into one cerebral lateral ventricle (Th+ICV). We tested three different promoters of varying strengths. The CBA promoter, composed of the chicken beta actin promoter with an enhancer element and an artificial intron; the CBi, identical to the previous promoter but lacking the enhancer element; the CB, the basic chicken beta-actin promoter without an enhancer or intron. All vectors were delivered at a total dose of 4E9 (Th+DCN) or 1E10 (Th+ICV) vector genomes. At four weeks post injection, we analyzed the brain histologically, for beta-galactosidase expression and decrease in lysosomal storage. Additionally, we analyzed the enzyme activity of beta-galactosidase, as well as the total GM1-gangliosidosis storage by enzymatic assays and LC-MS/MS mass spectrometry, respectively. The AAVrh10-CBA-βgal vector restored GM1 ganglioside levels to normal, and generated the highest βgal activity of all three vectors. Additionally, the distribution was more extensive for the AAVrh10-CBA vector, showing distribution to the frontal cortex, while for the other vectors the enzyme was mostly restricted to the injection site and the immediate area. However, at this vector dose we did not detect any apparent toxicity or cell death. Thus, we determined that the AAVrh10-CBA-βgal vector is safe and effective, and thus appropriate to carry out long-term therapeutic studies in GM1 mice toward a clinical trial.