Improving the efficiency of homology directed repair in gene editing

Over the past decade, CRISPR/Cas9 technology has revolutionized gene and cell therapy by enabling targeted genome and epigenome editing. Two cellular repair mechanisms can be induced when Cas9 cleaves DNA to create double-strand DNA breaks: non-homologous end joining (NHEJ) and homology directed repair (HDR). While NHEJ is active throughout the cell cycle, HDR is mostly restricted to the G2/S mitotic phases, limiting its application to non-dividing cells.

This limitation is unfortunate because HDR is more precise than NHEJ due to differences in their repair mechanisms. NHEJ is prone to errors because it causes insertions and deletions of DNA bases independently of a repair template, whereas HDR relies upon a repair template to add the correct bases. However, HDR efficiency can be low in therapeutically relevant cells, such as T cells and stem cells, and many efforts have been made to increase HDR efficiency via enrichment strategies to select for successfully edited cells.

In a similar vein, Rasmus O. Bak and colleagues at Aarhus University have introduced a new enrichment strategy in their recent article in Molecular Therapy – Methods & Clinical Development that may further improve HDR efficiency and increase the number of successfully edited cells.

Read full paper in ASGCT's Molecular Therapy – Methods & Clinical Development

Bak and colleagues’ approach relies upon CRISPR activation, or CRISPRa, to fuse catalytically deactivated Cas9 (dCas9) to a tripartite transactivator (VPR). When co-delivered with sgRNAs, the dCas9-VPR complex is directed to the region immediately upstream of the transcription start site of the target gene.

To optimize the approach, the researchers constructed an HDR template containing a GFP reporter gene with expression driven by a strong constitutive promoter, i.e., SFFV or EF1α, as well as a separate cassette and inducible miniCMV promoter to drive expression of a biologically inert cell surface marker (tNGFR). This HDR template plasmid was then co-electroporated with the dCas9-VPR complex and different combinations of sgRNAs to transfect K562 cells in vitro.

The best transfection rates were achieved using three murine CCR7 sgRNAs, resulting in tNGFR expressionin 91% of cells. Similar results were achieved when the tNGFR gene was integrated into the genome via HDR integration, followed by time in culture to diminish the expression of the episomal plasmid and then electroporation of CRISPRa targeting the reporter cassette. With CRISPRa, 96% of cells expressed tNGFR; without it, only 42% of cells expressed tNGFR. 

The HDR enrichment strategy was less effective in primary human T cells, but nevertheless achieved efficient expression of the reporter gene (tNGFR or tEGFR) in two genomic loci (HBB or AAVS1). From 31% to 58% of all CD3/CD28-activated T cells expressed the reporter gene, representing enrichment of 1.8- to 3.3-fold.

To apply their strategy to hematopoietic stem and progenitor cells (HSPCs), the researchers swapped the tNGFR and tEGFR reporter genes for tCD19 and tCD8 reporter genes because HSPCs endogenously express both NGFR and EGFR. After HDR-mediated integration and CRISPRa application, tCD19 and tCD8 were detected in 28% of HSPCs. Following immunomagnetic enrichment, the frequencies of tCD19- and tCD8-positive HSPCs were further increased to 63% and 65%, respectively. 

To test the efficacy of the HDR enrichment strategy in a therapeutically relevant context, the researchers selected CAR T cells. They replaced GFP with an antiCD19-CAR cassette expressed from an EF1α promoter (a more clinically relevant medium-strength promoter) and targeted the construct to the CCR5 locus for HDR-mediated integration. The reporter gene was tNGFR driven by the miniCMV promoter, similar to the construct used for the primary human T cells.

Upon HDR integration and CRISPRa application, 26% of T cells expressed both CAR and tNGFR, and immunomagnetic enrichment further increased the number of positive antiCD19 CAR T cells by 2.5-fold. Cytotoxicity was preserved in the enriched CAR T cells, which showed significantly greater cytotoxicity in co-culture with CD19+ NALM6 cells than did non-enriched CAR T cells.  

Altogether, these results suggest that this HDR enrichment strategy could be eventually be applied to clinical gene therapy. However, several issues must be resolved to make this approach truly viable.

The authors observed leaky expression of the tNGFR reporter gene in both T cells and HSPCs, which they attributed to the effects of the powerful SFFV promoter driving expression of GFP on the miniCMV promoter. Indeed, switching out the SFFV promoter with the EF1α promoter reduced leaky tNGFR expression to only 7% of T cells expressing tNGFR, down from 13% in T cells expressing tNGFR with the SFFV promoter. The authors also noted the tNGFR reporter gene was leakier than the other reporters used, indicating the choice of the reporter gene should be considered carefully, in addition to the promoter or other factors such as the cassette orientation.

Another limitation is the observed discrepancy between the expression of the gene of interest (GFP/CAR) and the reporter gene (tCD19) in CAR T cells. The authors found GFP cassette deletions encompassing the GFP coding sequence and/or EF1α promoter that were initiated at the site of the double strand break, indicating either prematurely terminated HDR from the repair templates or packaging and delivery of the truncated AAV vector genomes.

The final limitation is the use of electroporation to introduce the HDR template followed by another round of electroporation to introduce CRISPRa. Since repeated electroporation can damage cells, the authors examined the effects of two electroporation steps on primary human T cell viability and proliferation. Cell proliferation slowed for 4-7 days after the second electroporation, but fortunately recovered to the same rate as non-electroporated cells or cells electroporated only once. However, they did not assess the impact of electroporation on the stemness and repopulation capacity of HSPCs, which is needed to determine how this strategy will fare in stem cells.

As these issues are resolved and the approach is further refined, this HDR enrichment strategy may become a helpful addition to the gene editing toolbox.

Read full open access paper and learn more:

Mikkelsen, NS, Hernandez, SS, Jensen, TL, Schneller, JL, Bak, RO (2023). Enrichment of transgene integrations by transient CRISPR activation of a silent reporter gene. Mol Ther Methods Clin Dev. 2023; 29: P1-16. https://doi.org/10.1016/j.omtm.2023.02.010

To learn more about CRISPR/Cas9 technology, hear Jennifer Doudna, PhD, speak during the keynote at the ASGCT Annual Meeting in May.

View full Annual Meeting program

Courtney Bricker-Anthony, PhD, is ASGCT's scientific editor.

Related Articles