613. A Comparison of CNS Transduction After Systemic versus Cranial Delivery of an AAV2/9 CamKII Promoter-eGFP Vector in Mice

Abstract

Niemann-Pick type C (NPC) disease is a rare but fatal lysosomal storage disorder characterized by accumulation of unesterified cholesterol and other lipids within the lysosome. Clinical manifestations include ataxia, dementia and hepatosplenomegaly. 95% of cases are caused by an NPC1 gene mutation resulting in a lack of functional NPC1 protein, a putative cholesterol transport protein found in the lysosomal limiting membrane. The neuronal pathology seen in NPC1−/− mice begins with cerebellar axonal swelling, progresses to gliosis and visible lipid accumulation within lysosomes, then is followed by a marked loss of Purkinje cells. Weight loss and ataxia accompany disease progression. We have previously demonstrated that systemic delivery of an AAV9 vector designed to express the human NPC1 gene under the control of the neuronal-specific promoter, mouse calcium/calmodulin-dependent protein kinase II (CamKII), to Npc1−/− mice resulted in a modest but significant increase in survival. Transgene expression, assayed by immunohistochemistry (IHC), revealed widespread NPC1 expression within the brain, broadly correlating with the endogenous CamKII expression pattern. Delayed loss of Purkinje cells in the AAV treated Npc1−/− mice was also observed. Despite the increase in lifespan, expression of NPC1 in the Purkinje cells of the AAV treated Npc1−/− mice was limited, and the mice eventually succumbed to NPC1 disease. This result is notable due to the association between Purkinje cell loss and NPC1 disease progression. In an effort to improve neuronal transduction, particularly in Purkinje cells, we have explored various central nervous system (CNS) delivery routes with an AAV9 reporter, configured to express eGFP under the CamKII promoter. Three groups of mice (n=3 mice per group) received stereotactic-guided AAV9 CamKII-eGFP injections into either the lateral ventricles (dose: 3.8 × 1011 GC), cisterna magna (dose: 3.1 × 1011 GC), or a combination of both (dose: 3.8 × 1011 GC). After two weeks, eGFP expression was assessed using IHC and microscopy. Lateral ventricle and combination injections yielded widespread neuronal transduction throughout the brain, but expression in the cerebellum was limited and sporadic. Injections into the cisterna magna - expected to target the cerebellum - did not improve cerebellar expression compared to the other routes. Overall, aside from a heightened intensity of transgene expression, cranial injections did not appear to increase neuronal targeting compared to systemic delivery by retro-orbital injections (dose: 1×1012 GC). Based on preliminary evidence from a limited number of mice, gross differences in transduction between Npc1−/− and control mice were not observed, suggesting that the disease state at the time of assessment did not influence vector tropism. Our results will help define the optimal delivery route for future NPC1 vector optimization studies, and may be informative for others who seek to correct neurodegenerative mouse models using AAV9-mediated gene therapy.