Characterization of brain transduction capability of a BBB-penetrant AAV vector in mice, rats and macaques reveals differences in expression profiles

Abstract

Different screening methods are being developed to generate adeno-associated viral vectors (AAV) with the ability to bypass the blood-brain barrier (BBB) upon intravenous administration. Recently, the AAV9P31 stood out as the most efficient version among a library of peptide-displaying capsids selected in C57BL/6 mice using RNA-driven biopanning. In this work we have characterized in detail its biodistribution in different mouse strains (C57BL/6 and Balb/c), as well as in Sprague Dawley rats and non-human primates (Macaca fascicularis). Using GFP and NanoLuc reporter genes, we confirmed homogeneous infection and transgene expression across the CNS of mice injected intravenously with AAV9P31. A more restricted pattern was observed upon either intracerebroventricular or intraparenchymal injection. Following intravenous delivery, region- and cell-specific differential patterns of transduction were observed in the mouse brain, including a preferential transduction of astrocytes and neurons in the cerebral cortex and striatum, whereas neurons were the only transduced cell type in subcortical locations across the hippocampus, thalamus, hypothalamus, mesencephalon, brainstem and cerebellum. Furthermore, transduced microglial cells were never found in any CNS location. Peripheral organs transduced upon intravenous administration included lung, liver, peritoneum, heart and skeletal muscle. However, a comparable performance of AAV9P31 to bypass the BBB in rats and macaques was not observed, although a more limited neuronal transduction was found in the brainstem of rats upon intravenous delivery. Finally, intracerebroventricular delivery in macaques resulted in neuronal transduction in cortical, subcortical structures and cerebellum following a patchy pattern. In conclusion, the widespread CNS transduction obtained in mice upon intravenous delivery of AAV9P31 represents a powerful tool for modeling a wide variety of neurological disorders as well as an appealing choice for the evaluation of gene therapy-based therapeutics.