BE AWARE: GENOME EDITING WITH DNA-PKCS INHIBITOR AZD7648 INDUCES SIGNIFICANT GENOMIC ALTERATIONS – PUBLISHED IN NATURE BIOTECHNOLOGY

Genome editing creates double-strand breaks (DSBs) that can be repaired through either non-homologous end joining (NHEJ), microhomology-mediated end joining or homology-directed repair (HDR). While NHEJ is quick but error-prone, HDR uses a DNA template for precise edits, allowing targeted changes from single-nucleotide fixes to large gene insertions, greatly benefiting biomedical research and therapies. However, HDR is a relatively inefficient process, and ongoing efforts aim to improve its efficiency, including the use of small molecule inhibitors targeting DNA repair. One such highly potent inhibitor, AZD7648, selectively targets DNA-PKcs, redirecting DNA repair from the error-prone NHEJ pathway to the more precise HDR pathway in both transformed cell lines and primary human cells. However, the potential unintended consequences of its use in genome editing remain largely unexplored.

Are you curious to find out more?

Check out our recent advancement, led by postdoc Grégoire Cullot in a collaboration with the Gehart group (IMHS), Cathomen group (University of Freiburg, Germany) and the Gene Therapy research group of CSL Behring.

This work demonstrated that genome editing with a single Cas9-induced DSB in combination with AZD7648 leads to an increase in HDR, but this was accompanied by Cas9-induced genomic instability at on-target sites, where small-scale NHEJ outcomes were transformed into larger genetic alterations that cannot be detected by short-read sequencing.

Through the use of long-read sequencing, droplet digital PCR (ddPCR) for copy number analysis, single-cell RNA sequencing, and unbiased translocation detection, we discovered that AZD7648 significantly amplifies the frequency of kilobase-scale deletions, chromosome arm loss, and translocations across various cell types.

More broadly, genome editing-induced large-scale genomic alterations might still be largely underestimated. Indeed, these large-scale genomic alterations evade classical genome editing detection assays, typically short-read next-generation sequencing (NGS) and necessitate specific techniques that are not currently commonly used in the genome editing field. This means that clinical genome editing groups might be unaware of potential induced genomic instability and safety risks. Of note, AZD7648 is being tested by many clinical genome editing groups, however our results urge caution when deploying it during genome editing and reinforce the need to investigate genetic outcomes beyond those accessible to short-read target amplicon next-generation sequencing.

For more info check out our new paper in Nature Biotechnology!