Numerous AAV capsid optimization approaches, such as directed evolution and rational design, have been developed to enhance tissue-specific tropism and expression efficiency. However, these methods face challenges like low screening efficiency due to limited library diversity or insufficient scientific foundation. Only a few valuable capsids specific to tissues like muscle or CNS have been identified through these systems. To overcome these limitations, we utilize AAV to display human genomic sequences, named as Human-Originated Peptide Evolution (HOPE) system. First, we cloned random sequences from human genomic DNA or cDNA via primer libraries. Then, these sequences were inserted into the parental AAV9 capsid at positions 588/589, which was packaged into AAV library (>4M, NGS). The AAV library was enriched in vitro and in vivo over several rounds. Lead candidates were identified according to enrichment fold and ratio in specific tissues. These candidates were further optimized through machine learning and rational design to improve targeting properties and CMC profile.
After two rounds of screening in iPSC-derived myotubes and one round in mice, a lead candidate, 8S09, demonstrated significantly higher transduction efficiency in muscle tissue compared to AAV9 (50–200 times). After introducing liver de-targeting mutations, LD-8S09 retained strong delivery (29–164 times) and protein expression (>1000 times) compared to AAV9 in the muscles of mice. The muscle targeting of LD-8S09 outperformed other myotropic AAV, including MyoAAV4E, SLB101 and LICA1 in mice.The biodistribution of LD-8S09 is being investigated in NHP and data will be available in Jan 2026. Also, DMD base editor drug delivered by LD-8S09 is being tested in DMD mice.
We successfully identified a super myotropic AAV capsid only through 3 rounds enrichment, indicating the high screening efficiency of HOPE platform. Screening of the other tissue-specific AAV variants in NHP is underway. Collectively, this innovative approach represents a significant advancement in engineering AAV for more tissue-targeted applications.