Current adeno-associated virus (AAV) gene therapy using nature-derived AAVs is limited by non-optimal tissue targeting. In the treatment of muscular diseases (MD), high doses are therefore often required, but can lead to severe adverse effects. To lower treatment doses, we rationally designed an AAV that specifically targets skeletal muscle. We employed a novel computational design that integrated binding motifs of integrin alpha V beta 6 (αVβ6) into a liver-detargeting AAV capsid backbone to target the human αVβ6 complex – a selected AAV receptor for skeletal muscle. After sampling the low-energy capsid mutants, all in silico designed AAVs showed higher productivity compared to their parent. We confirmed in vitro that the enhanced transduction is due to the binding to the αVβ6 complex. Thanks to inclusion of αVβ6-binding motifs, the designed AAVs exhibited enhanced transduction efficacy in human differentiated myotubes as well as in murine skeletal muscles in vivo. One notable variant, LICA1, showed similar muscle transduction to other published myotropic AAVs, while being significantly more strongly liver-detargeted. We further examined the efficacy of LICA1, in comparison to AAV9, in delivering therapeutic transgenes in two mouse MD models at a low dose of 5E12 vg/kg. At this dose, AAV9 was suboptimal, while LICA1 transduced effectively and significantly better than AAV9 in all tested muscles. Consequently, LICA1 corrected the myopathology, restored global transcriptomic dysregulation, and improved muscle functionality. These results underline the potential of our design method for AAV engineering and demonstrate the relevance of the novel AAV variant for gene therapy treatment of MD.