Abstract
Unbiased screening of patients seen in nephrology clinics indicate that ~ 10% of patients have a monogenetic cause of kidney disease. For these patients and possibly for the broader diaspora of patients living with chronic kidney disease, gene therapy may be efficacious and possibly curative. A major limitation of gene therapy to the kidney parenchyma thus far, has been seemingly poor delivery of cargo to cell types of interest, utilizing existing adeno-associated viral capsids (e.g. AAV9). Novel AAV capsid serotypes may enhance transduction efficiency, but compared to FDA-approved capsids for extrarenal diseases, safety and tolerability data in humans is unknown. We systematically varied promoter, cargo, genome configuration, enzymatic priming, capsid serotype, dose, and route of administration, and show that optimizing these parameters yields some of the highest kidney transduction efficiencies reported in mice-40 to 60% of kidney tubules transduced with systemic AAV9-Cbh-mCherry-WPRE-by multiple unbiased quantification methods, including a novel machine-learning-based image analysis tool. We also demonstrate AAV transduction in live human kidney using ex vivo normothermic perfusion. These data provide a roadmap for ongoing preclinical studies to enable translation of AAV-mediated kidney gene therapy into the clinic.