Muscular dystrophies are inherited genetic disorders due to mutations in genes responsible for the structure and function of skeletal muscle fibers. A promising treatment for these diseases is gene therapy using recombinant Adeno-Associated Vectors (rAAVs). Capsid engineering to improve muscle targeting is a powerful tool to decrease costs and possible adverse effects of rAAV-based gene transfer.

Here, we performed a high throughput peptide display AAV2 library screening by intravenous injection in Dmdᵐᵈˣ mice (a model for Duchenne muscular dystrophy) and in healthy controls. Three peptides (O1, O2 and O3) were selected according to their occurrence in the target tissue (Tibialis Anterior) and based on target-to-noise ratio (Tibialis Anterior/liver). The peptides were then cloned into AAV2 and additional capsids such as AAV8, AAV9. The vectors were intravenously injected in healthy and Dmdᵐᵈˣ mice. O2 fusion in rAAV2 increased muscle-specific transduction by 1 log and transgene expression by 3 log, while liver transduction and transgene expression were decreased by 1 log. O3-rAAV2 also showed a slight increase in muscle-specific targeting. Surprisingly, O2 and O3 peptides in rAAV8 and rAAV9 did not increase the efficiency of transduction and transgene expression in muscle, but O2 decreased by 1 log the liver-specific transduction. In rAAV9rh74, O2 peptide reduced liver transduction by 3 log and transgene expression by 2 log. These results showed that virus entry does not always correlates with transgene expression, and the efficiency of capsid engineering still depends on the AAV serotype used, giving important insights for future design strategies.