High-throughput screening to identify human microRNAs enhancing CRISPR/Cas9-based homologous recombination for cardiac gene correction

The definitive treatment of genetic diseases through precise gene editing has been a long sought goal of gene therapy, yet hitherto unachieved at the clinical level. The CRISPR/Cas9-based technology ensures simple and efficient modification of the cellular genome and thus promises to change the perspective of treatment of hereditary disorders, including genetic cardiomyopathies. However, in mammals, gene correction through the homologous recombination (HR) machinery is largely less efficient than gene inactivation through the error prone, non-homologous end joining (NHEJ) route. This is particularly true in adult post-mitotic tissues, such as the heart, since HR commonly relies on S to G2/M-phase transition. The main goal of this project is to identify genetic treatments able to enhance frequency of CRISPR/Cas9-induced HR in vitro and in the heart. By robotic high-throughput, high-content microscopy we systematically screened a library of 2,024 human microRNAs to search for regulators of HR-mediated gene correction. We identified 20 miRNAs that significantly increase CRISPR/Cas9-induced HR events compared to controls (P<0.001). Interestingly, 10 among the top identified miRNAs belong to only two associated miRNAs families sharing the same seed sequence. A common and distinctive feature of these miRNAs is to induce accumulation of key proteins of the HR pathway, including MRE11, NBN, RAD50 and RAD51. Another highly effective miRNA, not belonging to either of these two families, regulates expression of p38 beta, suggesting involvement of this MAPK in the HR process. To study the effect of these miRNAs in the cardiac context, we first designed a CRISPR-AAV-based HR detection tool allowing precise, in frame insertion of a promoterless GFP gene into the last exon of the Myosin regulating light chain 2 (Myl2) gene in Cas9-expressing cardiomyocytes. The selected miRNAs also markedly enhanced Cas9-induced HR in this model. The same approach was then tested in vivo by producing AAV9 particles coding for the HR reporter system and the miRNAs and systemically injecting them in newborn Cas9+ mice. Also in this setting, the selected miRNAs increased the HR frequency up to 4-folds over control. Together, these results are encouraging in indicating that transient cardiac treatment with miRNAs enhancing HR together with the use of pro-recombinogenic AAVs might be sufficient to increase gene correction to a therapeutically sufficient level.