CRISPR/Cas9 deletions induce adverse on-target genomic effects

Clustered regularly interspaced short palindromic repeats (CRISPR/Cas9) have transformed genome engineering techniques. Numerous toolsets have been created to enable easy and efficient loss-of-function perturbations of functional genomic sites.

The CRISPR/Cas9 system's elements are exogenously injected into human cells to change the genome. The expressed Cas9 endonuclease generates a double-strand break (DSB) at a specific genomic region in the presence of a guide RNA (gRNA) that is complementary to the target site and next to a protospacer adjacent motif (PAM). If an external homologous DNA template is provided, the homology-directed repair is used to mend the DSB and induce specified mutations or sequence insertions.

RNA polymerase III (Pol III) recognizes internal promoter binding motifs and transcribes nuclear-encoded tRNA genes. To assess tRNA gene utilization, Pol III occupancy of tRNA genes is commonly utilized. Furthermore, Pol III-bound tRNA genes are found in euchromatic genomic areas with active histones, such as histone 3 lysine 4 trimethylation (H3K4me3). During translation, the anticodon region of a tRNA molecule detects the complementary codon of a messenger RNA and attaches an amino acid to the developing polypeptide chain, making these approximately 73 nucleotides (nt) long RNA molecules physical adapter molecules.

In a recent study posted to the preprint server bioRxiv*, a team of researchers utilized a CRISPR/Cas9 system to evaluate the usage of tRNA by deleting two tRNA genes from the genomes of hyper hepatocellular carcinoma (HepG2) and human near-haploid chronic myeloid leukemia (HAP1) cells. The authors discovered many unexpected genomic modifications at the target region using an improved droplet-based target enrichment approach (Xdrop) followed by Oxford Nanopore Technology (ONT) long-read sequencing (LRS).

The target region remained detectable and functional in Cas9 deletion clones

The authors chose a pair of tRNA genes that are close together on human chromosome 17 to assess the efficiency of CRISPR/Cas9 for eliminating these genes. They designed gRNAs to map to the unique 5' and 3' flanking regions of tRNA genes because they belong to one of the biggest multi-copy gene groups, in which individual gene family members are identical in sequence composition. To improve transfection efficiency, two Cas9 plasmids containing one of the two gRNAs and a small size plasmid were transfected into HAP1 and HepG2 cells.

The authors utilized antibiotic selection to detect positively transfected cells and generated single cell-derived clones because the CRISPR/Cas9 vector contained the puromycin resistance gene. A PCR using flanking region-specific primers was used to confirm the clonal deletion of the target region. When evaluated by agarose gel electrophoresis, the size of the PCR result indicated a successful deletion, and its sequence content was confirmed by Sanger sequencing. A total of 94 HAP1 and 90 HepG2 single-cell clones were produced. A deletion was found in 5 HAP1 clones and 17 HepG2 clones.

The use of Xdrop in deletion clones validated the target region's genomic remodeling

CRISPR/Cas9 genome alterations have recently been validated using the Xdrop technology. The Xdrop technique was used to enrich sequences containing the CRISPR/Cas9-targeted genomic area in the HAP1 Δt72 and HepG2 Δt15 deletion clones to evaluate the on-target editing results in the Cas9 deletion clones. Following that, the authors used ONT LRS to determine the sequence composition of the Xdrop-enriched molecules. In the HAP1 Δt72 and HepG2 Δt15 deletion clones, the authors got an average of 217,000 and 179,000 reads with a median size of 4,600 and 5,200 base pairs (bp). Enrichment was determined by read coverage at the target locus in each cell clone. Sharp drops in coverage at the two DSB sites were found when both adjusted and raw readings were aligned to the human reference genome, and no read covered the two DSB sites in these two deletion clones.

On-target genomic alterations occurred frequently

The authors assessed whether on-target insertion events occurred in the other HAP1 and HepG2 Δt clones anticipated to have the deletion, to determine the approximate frequency of Cas9-induced genomic changes. Due to the unusually lengthy amplicon length, they were unable to estimate the proportion of on-target occurrences by PCR using primers spanning the 5' and 3' flanking regions, as the Xdrop-LRS contig revealed the integration of genomic sequences greater than 7,900 bp. As a result, they created a pair of primers that recombine within the target region. Because no target region-specific PCR product was discovered, the authors examined five HAP1 clones, three of which carried the expected homozygous deletion. Two bands were seen in the HAP1 Δt72 deletion clone, indicating a target area duplication. A PCR product with a length of roughly 340 bp was identified in the HAP1 Δt19 deletion clone in addition to the HAP1 Δt72 deletion clone, indicating a potential on-target genomic change.

Implications

The method utilized in this study demonstrates that CRISPR/Cas9 can result in the integration of endogenous and exogenous DNA fragments and also produce inversions, duplications, and local insertions of target-derived functional fragments. Although previous research has reported that when using the dual gRNA mechanism, a genomic region of interest can be duplicated or inverted, this research presents evidence that a combination of duplication and inversion, as well as the integration of exogenous DNA fragments and grouped interchromosomal rearrangements, can happen simultaneously.

Additionally, it was shown for the first time that the target-derived fragments were nonetheless functional despite these modifications, which can complicate mechanistic explanations. These findings reveal a new example of unintentional CRISPR/Cas9 editing occurrences that can go unnoticed and have a significant impact on the conclusions gained from experimental readouts.

*Important notice

bioRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

Written by

Colin Lightfoot

Colin graduated from the University of Chester with a B.Sc. in Biomedical Science in 2020. Since completing his undergraduate degree, he worked for NHS England as an Associate Practitioner, responsible for testing inpatients for COVID-19 on admission.