The Off-Targets of Clustered Regularly Interspaced Short Palindromic Repeats Gene Editing

Abstract

The repurposing of the CRISPR/Cas bacterial defense system against bacteriophages as simple and flexible molecular tools has revolutionized the field of gene editing. These tools are now widely used in basic research and clinical trials involving human somatic cells. However, a global moratorium on all clinical uses of human germline editing has been proposed because the technology still lacks the required efficacy and safety. Here we focus on the approaches developed since 2013 to decrease the frequency of unwanted mutations (the off-targets) during CRISPR-based gene editing.

Keywords: CRISPR/Cas9; Cas9 variants; DNA repair; gene editing; off-targets.

FIGURE 1

The Cas9:sgRNA ribonucleoprotein complex targets the genomic DNA when present in the cell nucleus. It will target PAM-adjacent sequences and may encounter several potential recognition sites. If there is a perfect base pairing between the target DNA and the sgRNA, a DSB is induced by Cas9. If there is a mismatch between the sgRNA and the DNA sequences present within the first 7–12 bp (proximal to the PAM) segment, the base pairing will not be sufficient to induce a DSB. If one or more mismatches between the DNA and sgRNA sequences are found in the PAM-distal (5′ end of the sgRNA) nucleotides, a DSB can be induced by Cas9, leading to off-target activity. The scenario of sgRNA sequence-independent off-targets is not depicted.

FIGURE 2

Strategies to decrease CRISPR/Cas9 off-target activity. The procedures to mitigate undesired CRISPR/Cas9 gene editing start (I) in silico, namely in the design of the sgRNAs. Several bioinformatic tools are available to determine the predicted off-targets of a given sgRNA, and one should choose a sequence that has high theoretical on-target activity and a low probability of generating off-target DSBs. The second step is (II) the delivery type of the CRISPR/Cas9 system players, which determines how long the cell becomes exposed to the RNP complex. Furthermore, (III) the engineering of the sgRNAs is an efficient strategy to decrease off-target activity. One additional major focus has been to (IV) engineer the highly specific SpCas9 forms with decreased off-target activity. Moreover, (V) the SpCas9 orthologs have a repertoire of PAM specificities, which increases the options for suitable sgRNA and may also decrease the off-target activity if the complexity of the PAM is higher than that of SpCas9. (VI) Cas9-mediated DSBs may be repaired by the HDR or NHEJ pathways, which generate distinct repair products. As the HDR repairs the DNA damage restoring the original sequence, it has been argued and shown that the repair of off-targets by this pathway does not generate undesired gene edits. HDR repair is promoted either by pharmacological inhibition or cell-cycle-regulation. More recently, the use of Base Editors (VII), where a catalytically dead Cas9 or a Cas9 nickase is fused to a cytidine deaminase that drives a single nucleotide modification, has the potential to reduce off-targets, due to sequence requirements and the lack of DSBs. The blockage of off-targets by GUARD RNAs (VIII) has also been shown to reduce off-targets of Cas9. Finally, (IX) the chromatin organization manipulation may skew the off-target activity.

FIGURE 3

sgRNA choice comparison between available online tools. Using a total set of 36 exons from different genes with an average size of 386 ± 530 bp (list below), the top 10 sgRNA of the CCTop, CasOFFinder, CRISPRScan, GuideScan, CHOPCHOP, and DeepCRISPR were determined. Afterward, the common top sgRNAs between any two tools were identified for each, and the mean number of common guides is shown in the matrix based on a heatmap. Each square represents the comparison of the corresponding horizontal and vertical position. The different choice of sgRNAs between the available tools for a given genomic region target is associated to the choice of different collections of potential off-targets. The list of genes used: SPHK1 (Hs), SPHK2 (Hs), IL7RA (Hs), IDE (Hs), HLA (Hs), BLM (Hs), GUSB (Hs), AZIN1 (Hs), ADRA2C (Hs), ADRB2 (Hs), MAPT (Hs), SMAD3 (Hs), EIF2AK4 (Hs), MVK (Hs), Anxa1 (Mm), Adra2c (Mm), KLHL (Hs), Mlh1 (Mm), Msh6 (Mm), KRAS (Hs), DHX37 (Hs), RYR2 (Hs), NR3C1 (Hs), ODF2 (Hs), Hnf4a (Mm), Braf (Mm), POLMU (Hs), LYPLA2 (Hs), PPT1 (Hs), ESRRA (Hs), TCRB (Hs), BTBD9 (Hs), GALNT1 (Hs), MAN2A (Hs), and ST6GAL1 (Hs).