Characterizing large homology directed repair (HDR) insertions by CRISPR/Cas9 using MinION long-read sequencing technology

Mollie Schubert, from integrated DNA technologies, opened her talk by saying “my talk will be a bit different as we tend to use Cas9 in cells”. She went on to describe how the CRISPR/Cas9 system can be used to perform genome editing by using an RNA “guide” to cut DNA in specific places. If donor DNA is present in the cell, homology directed repair (HDR) is performed to integrate the donor DNA into the host genome at the cut site. If it is absent, non-homologous end joining occurs and can be exploited to generate knockout models.

So what are the reasons someone may want to perform homology directed repair using Cas9? Mollie showed a list of applications for this, including: correcting a mutation, tagging a protein, or inserting foreign genetic material to generate transgenic organisms. However, assaying the success of Cas9-mediated HDR is important, and Mollie said that this is where Oxford Nanopore long-read sequencing can help.

Mollie discussed how CRISPR gene editing could be used to insert large regions of DNA, and that using different sized inserts allows the quantification of the efficiency of the reaction. Mollie described the number of different outcomes that could occur during a gene editing experiment at the molecular level. Such outcomes include: no insertion, just insertions or deletions at the cut site, the correct insertion at the cut site, a partial insertion, or blunt insertion of the donor DNA where it can form duplications or inversions.

As an example, Mollie displayed sequencing data spanning a Cas9 cut site without any donor DNA, and showed that the resulting repair products were, indeed, highly variable. Explaining a number of approaches that could be used to assess the efficiency of Cas9 genome editing approaches, Mollie said that methods like digital droplet PCR and qPCR will still capture partial insertions and thus nothing is known about the overall “correctness”. Overall, high-throughput sequencing provides the best assessment as it provides high sensitivity and quantification, however, short-read sequencing cannot capture all the information required for long inserts.

Examining the same system with targeted amplicon sequencing of the insertion site, Mollie showed that Oxford Nanopore sequencing provided results that were consistent with PCR fragment analysis. Using a pipeline previously used for short-read sequencing, Mollie said that it was built with the forward compatibility for long reads. Briefly, the pipeline aligns reads and interrogates regions of interest for variants. Mollie then moved on to show how the Cas9 targeted cutting itself could be used to excise the native DNA of interest rather than using a PCR based approach. Here, Mollie showed 3,777X coverage of the target region, which related to an on-target rate of 1.74%. Again, preliminary results showed that this approach was consistent with the PCR-based methods used.

In summary, Mollie said that MinION nanopore sequencing has allowed for a more detailed analysis of HDR insertions than previous methods, but they are further investigating the PCR amplification vs. CRISPR-Cas9 approaches to identify any biases that may be associated with long sequence inserts.