Crystal structure of Cas9 in complex with guide RNA and target DNA

Abstract

The CRISPR-associated endonuclease Cas9 can be targeted to specific genomic loci by single guide RNAs (sgRNAs). Here, we report the crystal structure of Streptococcus pyogenes Cas9 in complex with sgRNA and its target DNA at 2.5 Å resolution. The structure revealed a bilobed architecture composed of target recognition and nuclease lobes, accommodating the sgRNA:DNA heteroduplex in a positively charged groove at their interface. Whereas the recognition lobe is essential for binding sgRNA and DNA, the nuclease lobe contains the HNH and RuvC nuclease domains, which are properly positioned for cleavage of the complementary and noncomplementary strands of the target DNA, respectively. The nuclease lobe also contains a carboxyl-terminal domain responsible for the interaction with the protospacer adjacent motif (PAM). This high-resolution structure and accompanying functional analyses have revealed the molecular mechanism of RNA-guided DNA targeting by Cas9, thus paving the way for the rational design of new, versatile genome-editing technologies.

Figure 1. Overall structure of the Cas9–sgRNA–DNA ternary complex

(A) Domain organization of S. pyogenes Cas9. BH, Bridge helix. (B) Schematic representation of the sgRNA:target DNA complex. (C) Ribbon representation of the Cas9–sgRNA–DNA complex. Disordered linkers are shown as red dotted lines. (D) Surface representation of the Cas9–sgRNA–DNA complex. The active sites of the RuvC (D10A) and HNH (H840A) domains are indicated by dashed yellow circles. (E) Electrostatic surface potential of Cas9. The HNH domain is omitted for clarity. Molecular graphic images were prepared using CueMol (http://www.cuemol.org). See also Figures S1, S2 and Table S1.

Figure 2. REC lobe and PI domain

(A) Structure of the REC lobe. The REC2 domain and the Bridge helix are colored dark gray and green, respectively. The REC1 domain is colored gray, with the repeat-interacting and anti-repeat-interacting regions colored pale blue and pink, respectively. The bound sgRNA:DNA is shown as a semi-transparent ribbon representation. (B) Mutational analysis of the REC lobe. Schematics show the truncation mutants. The bar graph shows indel mutations generated by the truncation mutants, measured by the SURVEYOR assay (n = 3, error bars show mean ± S.E.M., N.D., not detectable). (C) Western blot showing the expression of the truncation mutants in HEK 293FT cells. (D) Structure of the PI domain. The bound sgRNA is shown as a semi-transparent ribbon representation. (E) Mutational analysis of the PI domain. Schematics show wild-type SpCas9 and St3Cas9, chimeric Sp-St3Cas9 and St3-SpCas9, and the SpCas9 PI domain truncation mutant. Cas9s were assayed for indel generation at target sites upstream of either NGG (left bar graph) or NGGNG (right bar graph) PAMs (n = 3, error bars show mean ± S.E.M., N.D., not detectable). See also Figures S3–S5.

Figure 3. NUC lobe

(A) Structure of the RuvC domain. The core structure of the RNase H fold is highlighted in cyan. The active-site residues are shown as stick models. (B) Structure of the T. thermophilus RuvC dimer in complex with a Holliday junction (PDB ID 4LD0). The two protomers are colored cyan and gray, respectively. (C) Mutational analysis of the RuvC and HNH domains. The sequences (top) illustrate Cas9 nicking targets on opposite strands of DNA. Targets 1 and 2 are offset by a distance of 4-bp in between. The cleavage sites by the HNH and RuvC domains are indicated by pink and cyan triangles, respectively. The heatmap (bottom) shows the ability of each catalytic mutant to induce double- (with either sgRNA 1 or 2) or single-stranded breaks (only with both sgRNAs together). Gray boxes, not assayed. (D) Indel formation by Cas9 nickases depends on the off-set distance between sgRNA pairs. The off-set distance is defined as the number of base pairs between the PAM-distal (5′) ends of the guide sequence of a given sgRNA pair (n = 3, error bars show mean ± S.E.M., N.D., not detectable). (E) Structure of the HNH domain. The core structure of the ββα-metal fold is highlighted in magenta. The active-site residues are shown as stick models. (F) Structure of the T4 Endo VII dimer in complex with a Holliday junction (PDB ID 2QNC). The two protomers are colored pink and gray, respectively, with the ββα-metal fold core highlighted in magenta. The bound Mg²⁺ ion is shown as an orange sphere. (G) Superimposition of the Cas9 HNH domain and T4 Endo VII (PDB ID 2QNC). See also Table S2.

Figure 4. sgRNA and its target DNA

(A) Schematic representation of the sgRNA:target DNA complex. The guide and repeat regions of the crRNA sequence are colored sky blue and blue, respectively. The tracrRNA sequence is colored red, with the linker region colored violet. The target DNA and the tetraloop are colored yellow and gray, respectively. The numbering of the 3′ tails of the tracrRNA is shown on a red background. Watson-Crick and non-Watson-Crick base pairs are indicated by black and gray lines, respectively. Disordered nucleotides are boxed by dashed lines. (B) Structure of the sgRNA:target DNA complex. (C) Close-up view of the repeat:anti-repeat duplex and the three-way junction. Key interactions are shown with gray dashed lines. (D) Effects of sgRNA mutations on the ability to induce indels. Base changes from the sgRNA(+85) scaffold are shown at the respective positions, with dashes indicating unaltered bases (n = 3, error bars show mean ± S.E.M., p values based on unpaired Student’s t-test, N.D., not detectable). See also Figure S6.

Figure 5. Schematic representation of sgRNA:target DNA recognition by Cas9

Residues that interact with the sgRNA:DNA via their main chain are shown in parentheses. Note that water-mediated hydrogen-bonding interactions are not shown, for clarity.

Figure 6. sgRNA:target DNA recognition by Cas9

(A and C–J) Recognition of the guide (A), the guide:target heteroduplex (C), the repeat (D), the anti-repeat (E), the three-way junction (F), stem loop 1 (G), the linker (H), stem loop 2 (I) and stem loop 3 (J). Hydrogen bonds and salt bridges are shown as dashed lines. In (A), the target DNA is omitted, for clarity. (B) Effects of Cas9 (top) and sgRNA (bottom) mutations on the ability to induce indels (n = 3, error bars show mean ± S.E.M., p values based on unpaired Student’s t-test. N.D., not detectable).

Figure 7. Structural flexibility of the complex and a model for RNA-guided DNA cleavage by Cas9

(A) Structural comparison of Mol A and Mol B. In Mol A (left), the disordered linker between the RuvC and HNH domains is indicated by a dotted line. In Mol B (right), the disordered HNH domain is shown as a dashed circle. The flexible connecting segment (α39 and α40) in the RuvC domain is colored orange. (B) Superimposition of the Cas9 proteins in Mol A and Mol B. The two complexes are superimposed based on the core β-sheet of the two RuvC domains. The HNH domain and the bound sgRNA:target DNA complex were omitted, for clarity. (C) Superimposition of the sgRNA:target DNA complex in Mol A and Mol B. After superimposition of the two complexes as in (B), the Cas9 proteins were omitted to show the sgRNA:target DNA complex. (D) Molecular surface of Cas9. The HNH domain and the sgRNA:target DNA complex were omitted, for clarity. (E) Model of RNA-guided DNA cleavage by Cas9.