615. Choroid Plexus-Targeted Viral Gene Therapy for Lysosomal Storage Diseases

Abstract

Lysosomal storage diseases (LSDs) represent a category of approximately 60 inherited neurometabolic illnesses with considerable morbidity and mortality in children and adults. The burden of LSD is high due to the chronic progressive decline in the neurocognitive function of affected individuals that profoundly limits their societal integration. Current therapy for LSDs has focused on recombinant enzyme replacement therapy (ERT), which while promising, is not ideal for numerous reasons, including the short half-lives of ERT with weekly or monthly administration, potential immune responses against ERT may also reduce the effectiveness for the treatment and inefficient passage through the blood-brain barrier. The choroid plexuses are vascularized structures that project into the cerebrospinal fluid (CSF) and feature specialized polarized epithelia derived from neuroectoderm that are post-mitotic, i.e., do not undergo turnover, and produce CSF by transporting water and ions into the brain ventricles. We hypothesized that remodeling these epithelia to secrete missing lysosomal enzymes by one-time administration of a recombinant AAV vector with selective tropism for choroid plexus (e.g., serotype 5) would be an attractive strategy for long-term treatment of LSDs. Potentially this approach would result in steady secretion of the missing enzyme into the CSF, which normally carries molecules throughout the ventricular system into the subarachnoid space, and ultimately deliver enzyme to the entire brain. The cross-correction phenomenon in many LSDs would provide a further advantage. To evaluate this hypothesis in preclinical animal models, we chose two prototypical LSDs, α-mannosidosis and mucopolysaccharidosis type IIIB (MPS IIIB or Sanfilippo B syndrome). We cloned the respective cDNAs (human (h) LAMAN and NAGLU) into rAAV shuttle plasmids and generated high titer of rAAV5 expressing the enzymes. We administered viral particles to the lateral cerebroventricles sof homozygous mutant mice on day 2 or 3 of life at doses of 5 × 109 or 5 × 1010 vector genomes. In the LAMAN deficient mice, we documented dose-dependent transduction and hLAMAN mRNA expression confined to the choroid plexuses of rAAV5-treated animals. Brain biochemical analyses at 1, 2 and 6 months post-treatment documented sustained, highly statistically significant increases of LAMAN enzyme activity in the brain globally (olfactory bulb, cerebral cortex, cerebellum, brainstem). By 8 months of age, untreated mutant mice showed prominent lysosomal vacuoles in hippocampal neurons, in contrast to rAAV5-hLAMAN treated mutants for which brain histopathology was comparable to wild-type. Lysosomal associated membrane protein 1 (Lamp1) levels were normalized in AAV5-hLAMAN treated mutant brain. In MPS IIIB mutant mice, levels of NAGLU enzyme activity were 2-8 fold higher in brain sections six weeks after rAAV5-hNAGLU treatment compared to normal controls. β-Hexosaminidase activity, which is elevated in MPS IIIB, was reduced to heterozygote carrier levels. Tissue evaluations by immunohistochemistry showed robust hNAGLU expression in the choroid plexus epithelia. Lamp1 expression was significantly reduced in hippocampus and frontal cortex of treated mice. The approach outlined herein challenges existing treatment modalities for LSDs, by exploiting the ability of choroid plexus-targeted gene therapy to restore missing or defective lysosomal enzymes at concentrations and distributions in CSF suitable for disease correction. The potential impact on clinical practice in the field of LSD is high since, if these results extend to larger animal disease models, nonhuman primates, and human subjects, the largest current barriers to health for affected patients would be circumvented.