Rational selection of CRISPR/Cas9 guide RNAs for homology directed genome editing

Abstract Homology-directed repair (HDR) of a DNA break allows copying of genetic material from an exogenous DNA template and is frequently exploited in CRISPR/Cas9 genome editing. However, HDR is in competition with other DNA repair pathways, including non-homologous end joining (NHEJ) and microhomology-mediated end joining (MMEJ), and the efficiency of HDR outcomes is not predictable. Consequently, to optimize HDR editing, panels of CRISPR/Cas9 guide RNAs (gRNA) and matched homology templates must be evaluated. We report here that CRISPR/Cas9 indel signatures can instead be used to identify guide RNAs (gRNAs) that maximize HDR outcomes. Specifically, we show that the frequency of deletions resulting from MMEJ repair, characterized as deletions greater than or equal to 3bp, better predicts HDR frequency than consideration of total indel frequency. We further demonstrate that tools that predict gRNA indel signatures can be repurposed to identify gRNAs to promote HDR. Finally, by comparing indels generated by S. aureus and S. pyogenes Cas9 targeted to the same site, we add to the growing body of data that the targeted DNA sequence is a major factor governing genome editing outcomes. Tatiossian and colleagues examined indel signatures from a series of CRISPR/Cas9 reagents and asked whether they could predict successful homology-directed repair (HDR) outcomes when homologous donor repair templates were present. They show that indel signatures are characteristic of either MMEJ or NHEJ repair, and that the frequency of MMEJ indels predicts the likelihood of HDR outcomes.