Targeted genomic manipulation by Cas9 can efficiently generate knockout cells and organisms via error-prone nonhomologous end joining (NHEJ), but the efficiency of precise sequence replacement by homology-directed repair (HDR) is substantially lower1, 2. Here we investigate the interaction of Cas9 with target DNA and use our findings to improve HDR efficiency. We show that dissociation of Cas9 from double-stranded DNA (dsDNA) substrates is slow (lifetime ~6 h) but that, before complete dissociation, Cas9 asymmetrically releases the 3′ end of the cleaved DNA strand that is not complementary to the sgRNA (nontarget strand). By rationally designing single-stranded DNA (ssDNA) donors of the optimal length complementary to the strand that is released first, we increase the rate of HDR in human cells when using Cas9 or nickase variants to up to 60%. We also demonstrate HDR rates of up to 0.7% using a catalytically inactive Cas9 mutant (dCas9), which binds DNA without cleaving it.
At a glance
- Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science 346, 1258096 (2014). &
- CRISPR-Cas: new tools for genetic manipulations from bacterial immunity systems. Annu. Rev. Microbiol. 69, 209–228 (2015). &
- Structural basis of PAM-dependent target DNA recognition by the Cas9 endonuclease. Nature 513, 569–573 (2014). , , &
- Crystal structure of Cas9 in complex with guide RNA and target DNA. Cell 156, 935–949 (2014). et al.
- A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816–821 (2012). et al.
- Genome engineering with zinc-finger nucleases. Genetics 188, 773–782 (2011).
- ZFN, TALEN, and CRISPR/Cas-based methods for genome engineering. Trends Biotechnol. 31, 397–405 (2013). , &
- DNA interrogation by the CRISPR RNA-guided endonuclease Cas9. Nature 507, 62–67 (2014). , , , &
- Determining kinetics and affinities of protein interactions using a parallel real-time label-free biosensor, the Octet. Anal. Biochem. 377, 209–217 (2008). , , &
- Kinetics of DNA double-strand break repair throughout the cell cycle as assayed by pulsed field gel electrophoresis in CHO cells. Int. J. Radiat. Biol. 59, 1325–1339 (1991). &
- Highly efficient RNA-guided genome editing in human cells via delivery of purified Cas9 ribonucleoproteins. Genome Res. 24, 1012–1019 (2014). , , , &
- Dynamics of CRISPR-Cas9 genome interrogation in living cells. Science 350, 823–826 (2015). et al.
- Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell 154, 1380–1389 (2013). et al.
- Increasing the efficiency of homology-directed repair for CRISPR-Cas9-induced precise gene editing in mammalian cells. Nat. Biotechnol. 33, 543–548 (2015). et al.
- Enhanced homology-directed human genome engineering by controlled timing of CRISPR/Cas9 delivery. eLife 3, e04766 (2014). , , &
- Increasing the efficiency of precise genome editing with CRISPR-Cas9 by inhibition of nonhomologous end joining. Nat. Biotechnol. 33, 538–542 (2015). et al.
- Optimization of scarless human stem cell genome editing. Nucleic Acids Res. 41, 9049–9061 (2013). et al.
- High-frequency genome editing using ssDNA oligonucleotides with zinc-finger nucleases. Nat. Methods 8, 753–755 (2011). et al.
- CAS9 transcriptional activators for target specificity screening and paired nickases for cooperative genome engineering. Nat. Biotechnol. 31, 833–838 (2013). et al.
- Genome editing using Cas9 nickases. Methods Enzymol. 546, 161–174 (2014). &
- Targeted genome editing in human repopulating haematopoietic stem cells. Nature 510, 235–240 (2014). et al.
- Genome editing with Cas9 in adult mice corrects a disease mutation and phenotype. Nat. Biotechnol. 32, 551–553 (2014). et al.
- Cas9 specifies functional viral targets during CRISPR-Cas adaptation. Nature 519, 199–202 (2015). et al.
- FokI dimerization is required for DNA cleavage. Proc. Natl. Acad. Sci. USA 95, 10570–10575 (1998). , , &
- In vitro assessment of zinc finger nuclease activity. Methods Mol. Biol. 649, 227–235 (2010). &
- Dimeric CRISPR RNA-guided FokI nucleases for highly specific genome editing. Nat. Biotechnol. 32, 569–576 (2014). et al.
- Homology-directed repair of DNA nicks via pathways distinct from canonical double-strand break repair. Proc. Natl. Acad. Sci. USA 111, E924–E932 (2014). &
- Conservative repair of a chromosomal double-strand break by single-strand DNA through two steps of annealing. Mol. Cell. Biol. 26, 7645–7657 (2006). , , , &
- MMEJ repair of double-strand breaks (director's cut): deleted sequences and alternative endings. Trends Genet. 24, 529–538 (2008). &
- Microhomology-mediated end joining: a back-up survival mechanism or dedicated pathway? Trends Biochem. Sci. 40, 701–714 (2015). &
- In vitro enzymology of Cas9. Methods Enzymol. 546, 1–20 (2014). &
- Cas9 RNP nucleofection for cell lines using Lonza 4D Nucleofector. protocols.io doi:10.17504/protocols.io.dm649d (13 August 2015). &
- Chromatin immunoprecipitation for determining the association of proteins with specific genomic sequences in vivo. Curr. Protoc. Mol. Biol. 69, 21.3.1–21.3.33 (2005). et al.
- Supplementary Figure 1: Supporting data for Figure 1 (224 KB)
(A) Schematic of BLI assay used to measure dissociation. 5′ monobiotinylated substrate DNA (identical to λ1, Figure 2) is associated with streptavidin-coated sensor tips (black oval) and baseline signal is established (left panel). Association phase (right panel) loads Cas9 onto substrate dsDNA and measures response. Dissociation phase (not shown) transfers the tip into buffer and monitors dissociation of Cas9. (B) Cas9, Cas9D10A, and Cas9H840A cleave DNA while dCas9 does not. Cas9 nucleases were incubated with or without sgRNA for 30 minutes and associated with λ1 substrate DNA (Figure 2A) for ten minutes. Untagged (pCR1002 and pCR1003, Document S3) and NLS-tagged (pCR1053-pCR1056, Document S3) Cas9 variants were tested and found to have equivalent activity. Reaction products were resolved on a 10% TBE-Urea gel. Open arrow, uncut substrate DNA; *, excess Cy5 labeled ssDNA; ‡, excess Cy3 labeled ssDNA. Data presented is representative of n=2 experiments. (C) Fit between BLI data (thick trace) and calculated kinetic values (maroon trace) for Cas9 (black) and dCas9 (brown). Replicate data is shown. (D) Cas9 interacts specifically with substrate dsDNA. BLI traces show no interaction of apoCas9 (no sgRNA) with substrate dsDNA (maroon trace) or Cas9 with substrate dsDNA lacking a PAM (blue trace). (n=2). (E) Gel densitometry of Figure 1B. Mean ± SD normalized intensity of + strand (blue) and − strand (red) shifted products were plotted as a function of time. The indicated regression lines were used to calculate koff. Equations and standard errors of the regression coefficient (se) are presented for each trace.
- Supplementary Figure 2: Schematic of reagents and experimental design for EMSA experiments. (249 KB)
Potential supershift products are presented where appropriate.
- Supplementary Figure 3: Supporting data for Figure 2 (171 KB)
(A) The non-target strand is released on the PAM-distal side of the cut. One Cas9 molecule was loaded onto substrate DNA fluorescently labeled at the 5′ or the 3′ terminus of each strand (Supplementary Figure 3). Only the 5′ non-target strand can be removed from the complex by a challenge DNA. Open arrow, RNP-DNA complex. (B) Removal of the non-target strand depends upon the concentration of the challenge DNA but is independent of the labeling fluorophore. Single RNP EMSA was conducted as described in Figure 2A, except challenge concentration was varied from 0-1500nM (0, 30, 75, 150, 300, 600,1500nM). Catalytically inactive dCas9 was used in lane 8 (dashed box) to demonstrate that nuclease activity is required for strand extrusion activity. Substrate DNA fluorescently labeled at the 5′ termini with Cy5 or Cy3 as indicated. Open arrow, RNP-DNA complex. (C) Strand annealing occurs in single-RNP substrates when the non-target strand is left intact. Cas9 or dCas9 variants were loaded onto substrate DNA as indicated and as described in Supplementary Figure 3. Challenge concentration was varied from 0-5uM (0, 500, 1500, 2500, 5000nM). Open arrow, RNP-DNA complex; solid black arrow, supershifted products.
- Supplementary Figure 4: Model for challenge-mediated non-target strand removal activity. (83 KB)
1) After duplex cleavage, Cas9 holds onto three ends of the target DNA (white crossed circles), but the PAM-distal non-target strand is released from the Cas9-DNA complex. 2) Complementary DNA anneals to released strand. 3) Branch migration results in extrusion from the Cas9-DNA complex.
- Supplementary Figure 5: The non-target strand is available for enzymatic modification in cells. (58 KB)
Cas9 was targeted to either strand of the AAVS1 locus (AAVS1-F or AAVS1-R) and terminal transferase was introduced to 3′ end-label cut DNA with biotin. After streptavidin immunoprecipitation, end-labeling on either side of the break was determined by the ability to qPCR amplify sequences using the indicated primer pairs (Left and Right). Results are presented as the mean +/− SD fold enrichment (n=3) of labeled DNA over uncut control DNA (ACT1).
- Supplementary Figure 6: Strand-bias for optimized donor DNA is independent of genomic locus and gene transcription. (61 KB)
Cas9 targets the template strand of the EMX1 locus as diagrammed at left. Target strand (blue) or non-target strand (orange) donor ssDNAs were generated with the indicated overlaps on either side of the Cas9 cut site at EMX1. The sequences of the unedited and edited loci are presented inset (PAM sequence, underlined; cut site, magenta arrow; PciI site, bold font). HDR frequencies for editing with each donor are presented at right as mean +/− SD for n≥2 two independent experiments
- Supplementary Figure 7: Editing the EMX1 locus with and without donor DNA. (175 KB)
(A) The EMX1 locus is not cut as efficiently as the BFP locus. PCR amplification and T7E1 digestion were performed on cells edited using the indicated donor DNA (N/A – no donor, N/C – no Cas9). %Cut was quantified by gel densitometry. Compare to 95% total editing seen at the BFP locus (Supplementary Figure 9A). (B) HDR incorporation of a PciI site into the EMX1 locus shows donor strand-bias. PCR amplification (−) or PCR amplification and PciI digestion (+) was performed on cells edited using the indicated donor DNA (N/A – no donor). %Cut was quantified by gel densitometry and used to generate bar graphs in Supplementary Figure 6. Each nucleofection was performed in biological duplicate
- Supplementary Figure 8: Asymmetric donors stimulate HDR at the CXCR4 and CCR5 loci in HEK293 and K562 cells. (271 KB)
Cas9 was targeted to the CXCR4 or CCR5 loci in HEK293 and K562 cells. Target strand donors with the diagrammed overlaps were generated for each locus. The sequences of the unedited and edited loci are presented inset (PAM sequence, underlined; cut site, magenta arrow). T7 editing (T) and HDR frequencies (P; underlined) are presented at the bottom of each gel. Each nucleofection was performed in biological duplicate.
- Supplementary Figure 9: Sample flow cytometry plots. (382 KB)
- Supplementary Figure 10: Structural data is consistent with asymmetric release of substrate by Cas9. (332 KB)
A surface electrostatic view of Cas9, sgRNA (orange), and non-target (purple) or target (grey) DNA strands7. PAM-Cas9 interaction, white arrow; putative path of non-target strand, purple dots; presumed direction of non-target strand extrusion, black arrow.
- Supplementary Text and Figures (8.1 MB)
Supplementary Figures 1–10 and Supplementary Notes 1–4