CRISPR/Cas9 targeting events cause complex deletions and insertions at 17 sites in the mouse genome

Abstract

Although CRISPR/Cas9 genome editing has provided numerous opportunities to interrogate the functional significance of any given genomic site, there is a paucity of data on the extent of molecular scars inflicted on the mouse genome. Here we interrogate the molecular consequences of CRISPR/Cas9-mediated deletions at 17 sites in four loci of the mouse genome. We sequence targeted sites in 632 founder mice and analyse 54 established lines. While the median deletion size using single sgRNAs is 9 bp, we also obtain large deletions of up to 600 bp. Furthermore, we show unreported asymmetric deletions and large insertions of middle repetitive sequences. Simultaneous targeting of distant loci results in the removal of the intervening sequences. Reliable deletion of juxtaposed sites is only achieved through two-step targeting. Our findings also demonstrate that an extended analysis of F1 genotypes is required to obtain conclusive information on the exact molecular consequences of targeting events.

Figure 1. Targeting 17 sites in the mouse genome with CRISPR/Cas9.

STAT5 TF-binding sites (GAS motif), CTCF-binding regions and an NFIB-binding site were targeted for deletion. STAT5-binding sites in three gene loci (A, Stat5 (ref. 20); B, Socs2 (ref. 26); C, Wap19) were targeted individually (A, B and C-1). The three STAT5-binding sites in the Wap super-enhancer were deleted in four different combinations (C-1/2, C-1/3, C2/3 and C-1/2/3). Combined deletion of two or more STAT5-binding sites within the same gene locus was accomplished through a successive (two-steps) or simultaneous (one-step) targeting. A total of 11 CTCF-binding sites in two gene loci (D, Wap; E, Csn) were targeted individually and in different combinations. (F), an NFIB-binding site in the Csn locus was targeted (Supplementary Fig. 1). Target mutations were identified in 632 founder mice and 54 mutant lines were established. The positions of reference genes in the respective loci are indicated as coloured boxes. A, Stat5a; B, Socs2; C and D, Wap; E, Csn1s1; F, Csn3.

Figure 2. Targeting individual genomic sites with corresponding single sgRNAs.

Two genomic sites (B and C-1) were targeted independently with two individual sgRNAs (1 and 2). Four individual genomic sites (D-2, D-3, D-4(1) and F) were targeted with one sgRNA each. The deletion of two juxtaposed genomic sites (C-2/3) was generated in two steps. A single sgRNA-targeting site C-2 was injected into zygotes from mice carrying already a deletion in site C-3. The deletion of site C-3 was generated by TALEN. B, Socs2; C and D, Wap; E, Csn1s1; F, Csn3. The red numbers refer to the sites being targeted in the respective experiments.

Figure 3. Targeting individual genomic sites with more than one sgRNA.

Each of seven individual genomic sites (A, D-4(2), E-1, E-2, E-3, E-4 and E-5) was simultaneously targeted with two or three sgRNAs. Combined deletions of more than one genomics site (C-1/3 and D-1/2/3/4) were accomplished in two steps. Two sgRNAs targeting site C-1 were simultaneously injected into zygotes from mice carrying already a deletion in site C-3 (ref. 19). Two sgRNAs targeting site D-2 were simultaneously injected into mice carrying already a deletion in sites C-1/3/4. A, Stat5; C and D, Wap; E, Csn1s1. The red numbers refer to the sites being targeted in the respective experiments.

Figure 4. Targeting more than one genomic site with several sgRNAs.

Three juxtaposed genomic sites (D-1/3/4) were targeted simultaneously with four sgRNAs. Combined deletion of more than one genomic site (C-1/2, C-1/2/3 and D-1/2/3/4/5) was accomplished in two steps. Four sgRNAs targeting sites C-2 and C-3 were simultaneously injected into zygotes from mice carrying already a deletion in site C-1 (ref. 19). Four sgRNAs targeting sites D-2 and D-5, which are 23 kb apart from each other, were simultaneously injected into mice carrying already deletions in sites C-1/3/4. C and D, Wap. The red numbers refer to the sites being targeted in the respective experiments.

Figure 5. Asymmetric deletions.

(a) Schematic diagram of symmetric and asymmetric deletions detected in mice targeted with a single sgRNA. Red triangle, Cas9-cutting site three base pairs upstream of the PAM sequence. Symmetric deletions were defined as those with an equal or less than 1.5-fold ratio between the upstream and downstream Cas9-cutting site. In asymmetric deletions, the difference at either site was more than 1.5-fold than at the other site. (b) Percentage of symmetric and asymmetric deletions identified in CRISPR/Cas9-targeted mice (n=139). More than 80% of deletions were asymmetric and more than 70% of the deletions exceeded a two-fold difference. Only deletions obtained from a single sgRNA injection were analysed to avoid the effect of multiple variables. Deletions obtained from a single sgRNA injection: deletions targeting TF-binding site B, C-1, C-2/3, D-2, D-3, D-4(1) and F. (c) Ratio of symmetric and asymmetric deletions obtained with each sgRNA. B #1, n=8; B #2, n=18; C-1 #1, n=10; C-1 #2, n=8; C-2/3, n=12, D-2, n=40; D-3, n=12; D-4(1), n=21; F, n=10. Asterisk (*), sgRNAs with identical deletions identified in more than one half of the founders. (d) Frequency of symmetric deletions obtained with different deletion sizes. (e) Representative examples of deletions towards the 5′ end and 3′ end of sgRNA. If the deletion at the upstream Cas9-cutting site was longer (≥1.5-fold) than that at the downstream one, it was defined as a 5′ deletion and vice versa. (f) Percentage of deletions towards the 5′ end and 3′ end of sgRNA.

Figure 6. Preferential deletions at repeat sequences.

(a) Average frequency of deletions found at repeat sequences by single sgRNA injections. Repeat sequences were aligned in ∼45% at one end or both ends (total n=122; deletion at repeat sequences, n=56). Only deletions obtained from injections with single sgRNAs were analysed to avoid effects of multiple variables. Deletions obtained from single sgRNA injections: Deletions targeting TF-binding site B, C-1, C-2/3, D-2, D-3, D-4(1) and F. (b) Representative examples of repeat sequences aligned at one end (upper panel) and both ends (bottom panel). More than 60% of mutant founder mice targeting the genomic site D-4 exhibited the exact same deletion that retained only a single copy of repeat sequence. More than 30% of mutant founder mice targeting the genomic site C-2/3 showed the exact same deletion and repeat sequences remained at both ends. (c) Percentage of repeat sequences aligned at one end and both ends in founder mice carrying deletions at repeat sequences.

Figure 7. Large deletions obtained with single sgRNAs.

(a) Diagram of single and multiple sgRNAs targeting specific sites. Targeted sites are shown in purple and sgRNAs are indicated as cyan arrows. (b) Comparison of deletion sizes generated by single sgRNAs and multiple sgRNAs injected into mouse zygotes (total number of founder mice, n=243; single sgRNA injection at a single site, n=122; multiple sgRNA injections at a single site, n=121). Single sgRNA injection at a single site, deletions obtained by targeting TF-binding site B, C-1, C-2/3, D-2, D-3, D-4(1) and F; multiple sgRNA injection at a single site, deletions obtained by targeting TF-binding site A, C-1/3, D-4(2), D-1/2/3/4, E-1, E-2, E-3, E-4 and E-5. The median deletion size obtained with single sgRNAs (9 bp) was smaller than that with multiple sgRNAs (84 bp). The deletion size generated with a single sgRNA was up to 600 bp. Median, middle bar inside the box; IQR, 50% of the data; whiskers, 1.5 times the IQR.

Figure 8. Deletion sizes obtained on sequentially (two-steps) and simultaneously (one-step) targeting the mouse genome.

(a) Schematic diagram of deleting several sites in a single gene locus using sequential or simultaneous sgRNA targeting. The targeted site is shown in purple and sgRNAs are indicated as cyan arrows. (b) Comparison of deletion sizes obtained from sequential and simultaneous sgRNA-targeting approaches. Sequential sgRNA injections, deletions obtained by targeting TF-binding site C-2/3; simultaneous sgRNA injections, deletions obtained by targeting TF-binding sites C-1/2, C-1/2/3, D-1/3/4 and D-1/2/3/4/5. Deletions smaller than 400 bp and identified by typical PCR genotyping method were called ‘short deletion'. Those over 400 bp were considered ‘large deletions'. Results are shown as the mean (total number of founder mice, n=65; sequential deletion, n=20; simultaneous deletion, n=45). (c) The percentage of short and large deletions obtained from mutant mice generated by simultaneous injection with more than one sgRNA.

Figure 9. Screening for compound heterozygote deletions that are disguised as homozygote deletions.

(a) Diagrams depict misleading genotyping results caused by large deletions at one of the two alleles. Misleading results are shown in red with quotation marks. All possible genotypes generated from single site mutagenesis are shown. (b) Diagrams depict examples of potential misleading genotyping results caused by stitched large deletions and continuous large deletions. Blue and red arrows indicate PCR primers. Misleading results are shown in red with quotation marks. WT, wild-type; het, heterozygous; homo, homozygous; M1: mutated site 1; M2: mutated site 2. (c) Schematic demonstration of a mutant derived from simultaneous injection of two sgRNAs, which carries a 1 kb deletion between two targeting sites in addition to the desired short deletions at the target sites.