Genetic modification of cell and animal models enables discovering the mechanistic underpinnings of disease states. This knowledge is instrumental for developing diagnostics strategies, identifying prognostics biomarkers, and zeroing on therapeutic targets.
Several programmable endonuclease systems including first the zinc finger nucleases (ZFNs), followed by transcription activator-like effector nucleases (TALENs), and lastly the CRISPR-Cas nucleases, have progressively facilitated gene editing workflows. All these systems share a common mechanism by which induced DNA double-strand breaks (DSB) activate endogenous cellular repair processes, such as non-homologous end joining (NHEJ) and homology-directed repair (HDR). Activation of these DNA repair mechanisms ultimately leads to targeted sequence changes. (1, 2)
Among these two repair processes, NHEJ occurs with more frequency and results in the formation of insertions and deletions (indels). Therefore, NHEJ provides a means to knock-out gene expression. On the other hand, HDR which results in the recombination of sequences with homologous ends, may be exploited to knock-in specific sequences (e.g., exogenously provided DNA template or donor DNA).
A major challenge for the efficiency of knock-in approaches lies in the low frequency of HDR DNA repair, which can be as low as 1% in some cell types. (1) Investigators have identified over the years several strategies to improve the efficiency of CRISPR/CAS knock-in editing.
Designing robust guide RNAs. Available software tools which predict on-target and off-target effects support careful design of guide RNAs.
Multiple overlapping guide RNAs- knock-in efficiency of single-stranded oligonucleotide (ssODN) donors may be improved by the use of overlapping guide RNAs which share ~5 bases of the target sequence. (3)
Choosing the right donor DNA format. The use of double-stranded DNA (dsDNA) donors is associated with poor knock-in efficiency and increased off-target insertions, which is driven by genomic integration of dsDNA through the NHEJ repair process. (4, 5)
Modified ssODN donors- Use of chemically modified ssODN donors (e.g., phosporothioate or LNA) improves knock-in efficiency by stabilizing the donor template. (7)
Long ssDNA donors- In the Easi-CRISPR (Efficient additions with ssDNA inserts-CRISPR) method, long single-stranded DNA templates (~1.5kb) with ~50-100 bases homology arms are used in combination with RNP complexes, which are microinjected into zygotes for knock-in mice generation. (5) Compared to dsDNA, the use of ssDNA donor templates, including long ssDNA, is associated with improved insertion efficiency and viability of embryos during animal model generation. (5, 6)
Enhancing localization of donor DNA to target sites. Various strategies aimed at ensuring the contact between Cas and the donor DNA once inside the cell have been developed.
HUH endonuclease RNP fusion complex- Cas9 fused to an HUH endonuclease (e.g., Porcine Circovirus 2 (PCV) Rep protein), which allows for covalent binding of ssODN donors carrying a 13 bp recognition sequence. (8)
Streptavidin RNP fusion complex- Cas9 is fused to streptavidin, which allows interaction with biotinylated DNA template. (9)
Shifting the balance between the frequency of NHEJ and HDR. Several methods may be leveraged to increase the frequency of HDR, however their success may be cell type specific.
Promoting cell synchronization- strategies may be implemented to induce cell arrest as HDR is restricted to the cell cycle’s late S and G2 phases. For example, by using inhibitors of microtubule polymerization to arrest cells in the G2/M phase (e.g., nocodazole). (10, 11)
Inhibiting NHEJ- targeting enzymes involved in NHEJ DNA repair such as DNA ligase IV for inhibition or degradation (e.g., SCR7). Inhibiting the expression of Ku proteins by siRNA or shRNA. (12)
Direct interaction between Cas9 and donor template- Addition of Cas9 target sequences (tCTSs) to the ends of the DNA template for HDR favors nuclear targeting. (13)
Overall, various strategies have been developed to improve CRISPR/Cas mediated knock-in efficiency. In some instances, the combination of more than one of these methods has proved to further improve editing. One potential draw back for the quick and straightforward application of these methods is that they have been validated under different experimental conditions (e.g., different-cell lines, -donor DNA), therefore requiring careful troubleshooting.