Nucleic Acid-Dependent Conformational Changes in CRISPR-Cas9 Revealed by Site-Directed Spin Labeling

Abstract

In a type II clustered regularly interspaced short palindromic repeats (CRISPR) system, RNAs that are encoded at the CRISPR locus complex with the CRISPR-associated (Cas) protein Cas9 to form an RNA-guided nuclease that cleaves double-stranded DNAs at specific sites. In recent years, the CRISPR-Cas9 system has been successfully adapted for genome engineering in a wide range of organisms. Studies have indicated that a series of conformational changes in Cas9, coordinated by the RNA and the target DNA, direct the protein into its active conformation, yet details on these conformational changes, as well as their roles in the mechanism of function of Cas9, remain to be elucidated. Here, nucleic acid-dependent conformational changes in Streptococcus pyogenes Cas9 (SpyCas9) were investigated using the method of site-directed spin labeling (SDSL). Single nitroxide spin labels were attached, one at a time, at one of the two native cysteine residues (Cys80 and Cys574) of SpyCas9, and the spin-labeled proteins were shown to maintain their function. X-band continuous-wave electron paramagnetic resonance spectra of the nitroxide attached at Cys80 revealed conformational changes of SpyCas9 that are consistent with a large-scale domain re-arrangement upon binding to its RNA partner. The results demonstrate the use of SDSL to monitor conformational changes in CRISPR-Cas9, which will provide key information for understanding the mechanism of CRISPR function.

Keywords: CRISPR; Cas9; Conformational change; EPR; Spin labeling.

Figure 1

CRISPR-Cas9 system for spin-labeling studies. (A) Domain organization of SpyCas9. The domains are represented as the following: RuvC and HNH are the endonuclease (NUC) domains that cleave non-target (n) and target (c) strands of the DNA, respectively; BH is the bridge helix that is arginine rich; HD (helical domain) I, II, III are alpha-helical rich recognition (REC) domains; and PI is the PAM Interacting domain that is essential in discriminating self from the non-self DNA. Asterisks (*) indicate the two native cysteine residues for attaching the spin label. (B) A schematics of the Cas9 ternary complex. (C) Chemical structure of the R5p label.

Figure 2

Biochemical characterizations of SpyCas9 protein function. Panel (A) shows an example of SDS PAGE of the purified proteins, panels (B) – (E) show results from DNA cleavage measurements with the ³²P label at either the n-strand [panels (B) & (C)] or the c-strand [panels (D) & (E)]. As shown, the DNA cleavage measurements were carried out at a DNA/sgRNA/SpyCas9 ratio of 1/3.5/5.0, with the DNA duplex concentration set at 100 nM. Excesses of Cas9 protein over sgRNA were used in order to ensure complete binding of the sgRNA. Neither free Cas9 nor sgRNA is capable of binding or cleaving the target DNA (6), the excess of these species did not interfere with the results reported.

Figure 3

X-band cw-EPR spectra of R5p labeled at residue 80 of SpyCas9. In each panel, shown on the left are the EPR spectra, with the spectral width being 100 G and dash lines marking the respective low- and high-field features for comparison. Asterisks (*) mark an ultra-sharp spectral feature that most likely arises from incomplete subtraction of the unreacted or detached spin labels. Shown on the right are the corresponding crystal structures, with the protein domains color-coded following the scheme used in Figure 1A, and when present, the RNA in pink and the DNA in Cyan. The Cys80 residue is indicated by the red arrow and shown in the CPK mode. (A) The apo state. The corresponding structure (PDB entry 4CMP (13)) shows that Cys80 is largely buried by the Helical-III domain. (B) The sgRNA-bound state. The corresponding structure (PDB entry 4ZT9 (17)) shows Cys80 is completely exposed to solvent. (C) The sgRNA/DNA bound state. The corresponding structure (PDB entry 4UN3 (15)) shows that Cys80 is completely exposed to solvent, and the local environment around Cys80 is identical to that of the sgRNA bound state.

Figure 4

X-band cw-EPR spectra of R5p labeled residue C574 of SpyCas9. In each panel the spectra are shown on the left. The corresponding crystal structures are shown on the right, with the same color coding scheme as that used in Figure 3. The Cys574 residue (when present in the structure) is indicated by the red arrow and shown in the CPK mode. (A) The apo state. The red circle marks the expected location of residue 574, the density of which is not observed in the corresponding crystal structure (PDB entry 4CMP (13)). (B) The sgRNA bound state. The corresponding structure is from PDB entry 4ZT9 (17). (C) The sgRNA/DNA bound state. The corresponding structure is from PDB entry 4UN3 (15).