Tongue and Hypoglossal Motoneuron Gene Therapy in a Rat Model of Pompe Disease

Pompe disease is a progressive lysosomal storage disorder resulting from deficiency in the lysosomal enzyme acid α‐glucosidase (GAA). The only approved treatment is human recombinant GAA (enzyme replacement therapy) which does not cross the blood brain barrier and thus does not mitigate central nervous system pathology. Lingual dysfunction is prominent, leading to common impairments such as dysphagia, dysarthria, and/or disordered breathing. Towards the goal of new treatments that address the underlying neural pathology of lingual impairment, we tested the hypothesis that a single intramuscular delivery of adeno‐associated virus (AAV) encoding GAA delivered to the tongue can drive GAA expression throughout lingual myofibers and associated hypoglossal motoneurons in a rat model of Pompe disease. In addition, we evaluated immune response to the transgene product and sought to determine if an FDA‐approved immunosuppressive drug which inhibits T‐cell activation (Abatacept®) can enhance the efficacy of gene transfer. Experiments were completed using a Gaa ‐/‐ rat model of Pompe disease that recapitulates many aspects of the clinical condition. An immunosuppression group (n=3) received three loading doses of Abatacept (10mg/kg) via tail vein 30 days, 14 days, and 1 day prior to intralingual injection of rAAV9‐des‐hGAA (100µl, 1.55x10 13 vg/ml) followed by a maintenance dose 30 days later. Blood and serum samples were taken at 3, 7, 14, 30, and 60 days after AAV9 injection and tongue and brainstem tissues were harvested after 60 days. Immunohistochemistry and blood assay results were compared to Pompe rats receiving gene therapy without immunosuppression (n=3), sham injection (n=3) and “wild type” control rats (n=3). Immunohistochemistry with anti‐GAA antibodies unequivocally confirmed that lingual rAAV9‐des‐hGAA delivery drove GAA expression in tongue and hypoglossal motoneurons. In addition, Periodic Acid Shiff (PAS) staining revealed decreased glycogen accumulation in tissues in which GAA immunostaining was prominent. Initial qualitative assessment suggests that Abatacept treatment increases gene therapy efficacy. Anti‐GAA IgG assay results indicate that Abatacept significantly blunted the immune response to the GAA transgene product at 14 days (f(4)= 6.021, p=0.002) but not 30 days post injection (f(4)=6.021, p=0.0736). Lingual injection with rAAV9‐des‐hGAA results in GAA protein expression and glycogen clearance in the tongue as well as the hypoglossal nucleus in animals with Pompe disease. The immunosuppression regimen tested here blunted the initial immune response to the transgene product, and this may have therapeutic benefit. However, modification of this immunosuppression regimen by providing additional maintenance doses may be required for maximal benefit. These data represent a significant step toward a new clinically viable treatment for lingual dysfunction. The results show conclusively that lingual gene therapy can drive expression of a therapeutic protein to correct brainstem and tongue muscle histopathology in a rat model of Pompe disease, and indicate possible additional benefit of immunosuppression.