PAM-Dependent Target DNA Recognition and Cleavage by C2c1 CRISPR-Cas Endonuclease

Abstract

C2c1 is a newly identified guide RNA-mediated type V-B CRISPR-Cas endonuclease that site-specifically targets and cleaves both strands of target DNA. We have determined crystal structures of Alicyclobacillus acidoterrestris C2c1 (AacC2c1) bound to sgRNA as a binary complex and to target DNAs as ternary complexes, thereby capturing catalytically competent conformations of AacC2c1 with both target and non-target DNA strands independently positioned within a single RuvC catalytic pocket. Moreover, C2c1-mediated cleavage results in a staggered seven-nucleotide break of target DNA. crRNA adopts a pre-ordered five-nucleotide A-form seed sequence in the binary complex, with release of an inserted tryptophan, facilitating zippering up of 20-bp guide RNA:target DNA heteroduplex on ternary complex formation. Notably, the PAM-interacting cleft adopts a "locked" conformation on ternary complex formation. Structural comparison of C2c1 ternary complexes with their Cas9 and Cpf1 counterparts highlights the diverse mechanisms adopted by these distinct CRISPR-Cas systems, thereby broadening and enhancing their applicability as genome editing tools.

Keywords: C2c1; RuvC catalytic pocket; binary complex with sgRNA; genome editing tool; sequence-specific PAM recognition; structure; ternary complex with added DNA; type V CRISPR-Cas endonuclease.

Figure 1. Overall Structure of AacC2C1-sgRNA-DNA Ternary Complex

(A) Domain organization of C2c1. (B) Two views of a ribbon representation of C2c1-sgRNA-DNA ternary complex, color-coded as defined in panels A and C. TS, target DNA strand; NTS, non-target DNA strand. (C) Schematic representation of the sgRNA-target DNA scaffold. (D) Structure of the sgRNA-target DNA in the ternary complex. See also Figure S1 and S2 and Table S1.

Figure 2. Detailed Interactions of C2c1 with sgRNA and Guide RNA-Target DNA Heteroduplex

(A) Recognition the sgRNA by the OBD, Helical-II, RuvC and Nuc domains. (B) Recognition the repeat:anti-repeat RNA duplex by the Helical-II and RuvC domains. (C) Recognition stem 1 of sgRNA by RuvC and Nuc domains. (D) Recognition of the guide:target heteroduplex by the REC and NUC lobes. (E) Recognition of the +1 phosphate group (+1P) and the proximal-PAM region of the dG1-dT6/C1-A6 base pairs of the guide:target heteroduplex by the OBD and Helical-I domains. (F) Mutational analysis of the residues involving in the binding of +1 phosphate group. See also Figure S2, S3 and S4.

Figure 3. Recognition of the PAM Duplex

(A) Recognition of the PAM duplex by the OBD and Helical-I domains. (B–D) Recognition of the dA(-3):dT(-3*) (panel B), dA(-2):dT(-2*) (panel C), and dG(-1):dC(-1*) (panel D) base pairs. (E) Recognition of the target DNA strand in the PAM duplex. (F) Recognition of the non-target DNA strand in the PAM duplex. (G) In vitro cleavage assay of the PAM-interacting residues. See also Figure S5.

Figure 4. Recognition of Excess 8-mer Substrate DNA Strand and Positioning Relative to Cleavage Site

(A) Recognition of substrate DNA by the RuvC and Nuc domains. (B) Schematic of detailed interactions of substrate DNA recognition by RuvC and Nuc domains. Hydrogen bonds and salt bridges are indicated with green lines. Hydrophobic and stacking interactions are shown by dashed orange lines. (C) Recognition of substrate DNA strand in the catalytic pocket. (D) Recognition of potential scissile phosphate of substrate DNA strand in the catalytic pocket. The modeled side chains of three Ala-mutated catalytic residues are marked by stars in panels B–D. (E) Mutational analysis of key residues lining the catalytic pocket. (F) Schematic of the engineered AacC2c1 sgRNA targeting the EMX1 DNA. The cleavage sites are indicated by red triangles. Sanger-sequencing traces are shown below. The additional non-templated adenines are denoted as N, which resulted from the polymerase used in sequencing (Clark, 1988).

Figure 5. Recognition of the Extended Target and Non-target DNA strands

(A) Stick representation of the excess 8-mer substrate DNA strand positioned in the catalytic pocket. The bound sulfate group mimics a phosphate group potentially connecting the target DNA strand and substrate DNA strand. (B) Overall structure of the extended target DNA strand recognition by the RuvC and Nuc domains. (C) Stick representation of the extended target DNA strand in the RuvC catalytic pocket. (D) Recognition of the dG21 to dG24 segment of extended target DNA strand in the catalytic pocket. The modeled side chains of three Ala-mutated catalytic residues are marked by stars. (E) Stick representation of the extended non-target strand d(T-T-T-T) segment in the RuvC catalytic pocket. See also Figure S5 and Table S1.

Figure 6. Domain Rearrangements on Ternary Complex Formation

(A) Two views of ribbon representation of C2c1-sgRNA binary complex. (B) Structural comparison between C2c1-sgRNA binary and C2c1-sgRNA-DNA ternary complexes. Vector length correlates with the domain transition scale. (C, D) Binding with target DNA widens the guide RNA-target DNA heteroduplex binding channel on proceeding from the binary complex (panel C) to the ternary complex (panel D). (E, F) Observable guide regions of crRNA in the C2c1-sgRNA binary complex (panel E) and the C2c1-sgRNA-DNA ternary complex (panel F). (G, H) Superposition of guide RNA seed segment in binary (nucleotides 1 to 5, in silver) and ternary (nucleotides 1 to 7, in magenta) complexes. The unblocking movement of Trp234 from the binary complex (in silver) to the ternary complex (in yellow) allows continuous stacking of guide RNA. Panel H is a blow up of panel G. (I, J) “Locking” of the PAM duplex on proceeding from the binary complex (panel I) to the ternary complex (panel J). The untraceable loop-α5-α6 helical I segment in the binary complex is shown by a red circle in panel I. See also Figure S6 and Table S1.

Figure 7. Structural comparison between AacC2c1 and AsCpf1 Ternary Complexes

(A–C) AacC2c1 complex: Conformational transition in Helical-I and Helical-II domains on proceeding from the binary (in silver) to ternary (in color) complex (panel A), substrate DNA positioned in the RuvC catalytic pocket containing modeled acidic residues (panel B), and overview of the ternary complex structure emphasizing the distance between last nucleotide of guide:target heteroduplex and the catalytic pocket (panel C). (D–F) AsCpf1 complex: Conformational transition in Helical-I and Helical-II domains on proceeding from the binary (in silver) to ternary (in color) complex (panel D), empty RuvC catalytic pocket (panel E), and overview of the ternary complex structure emphasizing the distance between last nucleotide of guide:target heteroduplex and the catalytic pocket (panel F). See also Figure S7.