Abstract
The choroid plexuses (CP) are highly vascularized structures that project into the ventricles of the vertebrate brain. The polarized epithelia of the CP produce cerebrospinal fluid by transporting water and ions into the ventricles from the blood and normally secrete a large number of proteins. We assessed the feasibility of selective CP transduction with recombinant adeno-associated virus (rAAV) gene therapy vectors for treatment of lysosomal storage disease (LSD), a broad category of neurometabolic illness associated with significant burdens to affected patients and their families. There are no ideal or complete therapeutic options currently available, especially for the central nervous system manifestations of LSDs. Alpha-mannosidosis (AMD) is an autosomal recessive prototypical LSD caused by deficiency of lysosomal alpha-mannosidase and characterized by cerebellar ataxia, neurocognitive disability, facial and skeletal abnormalities, hearing impairment, and mild immune deficiency. In a murine model of AMD, we compared the biochemical effects of CSF-directed rAAV serotypes 1, 4, 5, 6, and 9. Recombinant AAV1 and rAAV6, two closely related serotypes whose capsid sequences differ by only six amino acids, showed the most robust transduction of CP in mouse brain, consistent with their transduction of CPE in nonhuman primates and cats, as well as in other structures. We found restoration of LAMAN enzyme activity comparable to or higher than AMD heterozygote levels in the brain globally (olfactory bulb, cortex, cerebellum, brainstem). Further IND-generating preclinical experiments will advance rAAV6-LAMAN, which appears to be the most promising choroid plexus-targeting candidate serotype for future clinical translation to treat AMD.