Structural basis for target DNA cleavage and guide RNA processing by CRISPR-Casλ2

Abstract

RNA-guided CRISPR-Cas nucleases are widely used as versatile genome-engineering tools. Among the diverse CRISPR-Cas effectors, CRISPR-Casλ-also referred to as Cas12n-is a recently identified miniature type V nuclease encoded in phage genomes. Given its demonstrated nuclease activity in both mammalian and plant cells, Casλ has emerged as a promising candidate for genome-editing applications. However, the precise molecular mechanisms of Casλ family enzymes remain poorly understood. In this study, we report the identification and detailed biochemical and structural characterizations of CRISPR-Casλ2. The cryo-electron microscopy structures of Casλ2 in five different functional states unveiled the dynamic domain rearrangements during its activation. Our biochemical analyses indicated that Casλ2 processes its precursor crRNA to a mature crRNA using the RuvC active site through a unique ruler mechanism, in which Casλ2 defines the spacer length of the mature crRNA. Furthermore, structural comparisons of Casλ2 with Casλ1 and CasΦ highlighted the diversity and conservation of phage-encoded type V CRISPR-Cas enzymes. Collectively, our findings augment the mechanistic understanding of diverse CRISPR-Cas nucleases and establish a framework for rational engineering of the CRISPR-Casλ-based genome-editing platform.

Fig. 1. Discovery and biochemical characterization of Casλ2.

a CRISPR-Cas12 effectors exhibit significant diversity, with Casλ1 and Casλ2 clustering together alongside other representative type V Cas systems. Protein sequences were classified through global alignment, using free end gaps and similarity scoring based on the Blosum80 matrix.b Schematic of the E. coli negative selection screen. c Depletion activity of effector using a pooled depletion screen. Points represent UMIs corresponding to a specific spacer sequence. The depletion score for spacer sequences was measured as the minimum score among replicates of the ratio between normalized input reads to normalized output reads. (left) Sequence logos of spacer sequences with depletion scores greater than 3. d In vitro DNA cleavage activities of Casλ2 with 20-nt guide crRNAs. The linearized plasmid targets bearing the TTN PAMs were incubated with the Casλ2–crRNA complex at 37 °C for 30 min. The cleavage products were then analyzed by a MultiNA microchip electrophoresis system. e In vitro DNA cleavage activities of Casλ2 with 20-nt guide crRNA-temperature optimization. The linearized plasmid target bearing the TTA PAM was incubated with the Casλ2–crRNA complex at 25, 37, 45, 55, and 65 °C for 5 and 30 min, in reaction buffer containing 50 mM NaCl. The cleavage products were then analyzed by a MultiNA microchip electrophoresis system. Data are mean ± s.d. (n = 3). f In vitro DNA cleavage activities of Casλ2 with 20-nt guide crRNA-NaCl concentration optimization. The linearized plasmid target bearing the TTA PAM was incubated with the Casλ2–crRNA complex at 37 °C for 5 and 30 min, in reaction buffers with NaCl concentrations varying between 25, 50, 100, and 200 mM. The cleavage products were then analyzed by a MultiNA microchip electrophoresis system. Data are mean ± s.d. (n = 3). g In vitro DNA cleavage activities of Casλ2 with 12–22-nt guide crRNAs. The linearized plasmid target bearing the TTA PAM was incubated with the Casλ2–crRNA complex at 37 °C for 5 and 30 min, in reaction buffer containing 50 mM NaCl. The cleavage products were then analyzed by a MultiNA microchip electrophoresis system. Data are mean ± s.d. (n = 3). h Cleavage sites in the target DNA.

Fig. 2. Cryo-EM structure of the Casλ2–crRNA–target DNA complex.

a Domain structure of Casλ2. Cryo-EM density maps (b) and ribbon models (c) of the Casλ2–crRNA–target DNA complex. Magnesium ions are shown as green spheres. The disordered NTS is indicated by dashed lines. TS target DNA strand, NTS non-target DNA strand. d Schematic of the crRNA and target DNA. The disordered regions are enclosed by dashed boxes. e Structure of the crRNA.

Fig. 3. Recognition of the crRNA and target DNA.

a Recognition of the crRNA and target DNA by Casλ2. The Casλ2 proteins are shown as surface models. Recognition of the stem (b), stem loop (c), and U-rich loop (d). Hydrogen bonds are shown as dashed lines. Recognition of the first base pair of the guide–target heteroduplex (e), the end of the guide–target heteroduplex (f), the PAM duplex (g), and the substrate NTS within the RuvC active site (h). Hydrogen bonds and Mg²⁺ coordination are shown as dashed lines.

Fig. 4. Cryo-EM structures of Casλ2 in distinct functional states.

Cryo-EM density maps (top) and structural models (bottom) of the Casλ2–crRNA–target DNA complexes in the catalytically incompetent (a), intermediate (c, and NTS-cleaving (e) states. The disordered region is indicated by a dashed circle. Close-up views around the RuvC active site in the catalytically incompetent (b), intermediate (d), and NTS-cleaving (f) states. The disordered region is indicated by a dashed circle. g Cryo-EM density map (left) and structural model (right) of the Casλ2–crRNA–target DNA complexes in the TS-cleaving state. h Recognition of the extended TS within the RuvC active site.

Fig. 5. Pre-crRNA processing by Casλ2.

a Schematic of the pre-crRNAs used for the pre-crRNA processing experiments. Top, spacer-repeat-spacer (srs-type) pre-crRNA; Bottom, repeat-spacer-repeat (rsr-type) pre-crRNA. b In vitro pre-crRNA processing activities of WT Casλ2 and dCasλ2 (D324A). While the srs-type pre-crRNA was not cleaved by Casλ2 proteins, the rsr-type pre-crRNA was efficiently cleaved by WT Casλ2, indicating that Casλ2 processes its pre-crRNA on the 3′ side by using the RuvC active site. The cleavage products were analyzed by 10% denaturing urea-PAGE. c Schematic of the pre-crRNAs used for determining the pre-crRNA processing pattern. d In vitro pre-crRNA processing activities of WT Casλ2 for pre-crRNAs with different spacer lengths (17 and 24 nt) and those with a 20-nt polyA sequence at the 3′ end. The cleavage products were analyzed by 10% denaturing urea-PAGE. e Comparison of the lengths of pre-crRNA cleavage products (resulting in repeat-spacer) with those of synthesized crRNAs with different spacer lengths. f Cryo-EM density map (left) and structural model (right) of the dCasλ2–pre-crRNA binary complex. g Cryo-EM density map (left) and recognition of the pre-ordered spacer by Casλ2 (right). In the cryo-EM density map, crRNA is colored red. h Proposed model of Casλ2 pre-crRNA processing by the ruler mechanism.

Fig. 6. Stepwise mechanisms of action of Casλ2.

Schematic showing conformational changes in Casλ2 during DNA binding and cleavage. In the binary complex, Casλ2 adopts a closed conformation, accommodating a relatively long 12-nt spacer within the central groove to protect it from being cleaved. The remaining downstream spacer is bent and spontaneously directed toward the RuvC active site, resulting in the cleavage of pre-crRNA at a position 16–18 nt downstream from the crRNA scaffold. Upon target DNA binding, Casλ2 adopts an open conformation to accommodate the guide–target heteroduplex, with the REC2 domain dissociating from the TNB domain, and part of the RuvC domain and the entire TNB domain becoming disordered (incompetent state). Subsequently, the RuvC and TNB domains become ordered, accompanying the formation of the RuvC active site, although the substrate DNA is not yet bound (intermediate state). A subsequent closing movement of the REC2 domain toward the TNB domain, along with a local conformational rearrangement of the TNB domain, allows the NTS to enter the RuvC active site (NTS-cleaving state). Cleavage of the NTS is followed by the TS being directed along the positively charged surface of the REC3 domain to the RuvC active site (TS-cleaving state). The catalytically competent RuvC active site is indicated with an orange star.

Fig. 7. Structural comparison of Casλ2 with other Cas12 effectors.

Structural comparisons of Casλ2 with Casλ1 (PDB: 8DC2), CasΦ (PDB: 7LYT), and Cas12a (PDB: 6I1K). Structural models are fully colored to represent their domain configurations (left) or selectively colored to highlight regions involved in recognizing the PAM-distal region of the guide–target heteroduplex (center).