Mechanisms and engineering of a miniature type V-N CRISPR-Cas12 effector enzyme

Abstract

Type V CRISPR-Cas12 systems are highly diverse in their functionality and molecular compositions, including miniature Cas12f1 and Cas12n genome editors that provide advantages for efficient in vivo therapeutic delivery due to their small size. In contrast to Cas12f1 nucleases that utilize a homodimer structure for DNA targeting and cleavage with a preference for T- or C-rich PAMs, Cas12n nucleases are likely monomeric proteins and uniquely recognize rare A-rich PAMs. However, the molecular mechanisms behind RNA-guided genome targeting and cleavage by Cas12n remain unclear. Here, we present the cryo-electron microscopy (cryo-EM) structure of Rothia dentocariosa Cas12n (RdCas12n) bound to a single guide RNA (sgRNA) and target DNA, illuminating the intricate molecular architecture of Cas12n and its sgRNA, as well as PAM recognition and nucleic-acid binding mechanisms. Through structural comparisons with other Cas12 nucleases and the ancestral precursor TnpB, we provide insights into the evolutionary significance of Cas12n in the progression from TnpB to various Cas12 nucleases. Additionally, we extensively modify the sgRNA and convert RdCas12n into an effective genome editor in human cells. Our findings enhance the understanding of the evolutionary mechanisms of type V CRISPR-Cas12 systems and offer a molecular foundation for engineering Cas12n genome editors.

Fig. 1. Cryo-EM structure of the RdCas12n–sgRNA–target DNA ternary complex.

a Unrooted phylogenetic tree from representative Cas12n nucleases. AcCas12n-like and RdCas12n-like clades are shown in blue and pink, respectively. Arrows indicate Cas12n genes experimentally characterized in this study (RdCas12n, AcCas12n, CgCas12n and MlCas12n). b The tested Cas12n nucleases for purification. RdCas12n was used for final cryo-EM analysis. c Diagram of the sgRNA and target DNA used for cryo-EM analysis. d The domain structure of RdCas12n. CTD, C-terminal domain. Residues 1–10 and 268–539 (the NUC lobe) were not included in the final model. e Cryo-EM density map of the RdCas12n–sgRNA–target DNA ternary complex structure 1 (PDB ID: 9J09). f The overall structure of the RdCas12n–sgRNA–target DNA ternary complex structure 1 (PDB ID: 9J09). Disordered regions are indicated as dotted lines. The NUC region of the protein exhibits a disordered structure, and the delineation of the NUC lobe is tentative, marking a predicted domain where the NUC lobe is likely to be situated.

Fig. 2. Target DNA recognition.

a Recognition of the target DNA. The PAM duplex is bound to the cleft formed by the WED and REC domains. PAM upstream DNA exhibits additional interactions with the WED domain. b Detailed interactions of RdCas12n with the PAM duplex and PAM upstream DNA. c–f Major interactions between the PAM duplex and RdCas12n. Hydrogen bonding and electrostatic interactions are shown as blue dashed lines, and van der Waals interactions are shown as green dashed lines. The residues that interact with the nucleic acids through their main chains are shown in parentheses. g In vitro DNA cleavage activities of WT-RdCas12n and different PAM-recognition mutants. A linearized target DNA segment of 2.2-kb, comprising a 20-nucleotide target sequence and at a concentration of 10 nM, was subjected to incubation with the RNP complex at a concentration of 250 nM, at 37 °C across various time durations, as depicted by the gradient gray labels. Data are represented as mean ± SD (n = 3 biologically independent samples). Source data are provided as a Source Data file. h–i Major interactions between the PAM upstream DNA and RdCas12n. j Location of residues involved in PAM upstream DNA interactions. k Schematic of the molecular design for extra-WED region deletion mutants. l In vitro DNA cleavage activities of WT-RdCas12n, the PAM upstream DNA recognition mutants, and the extra-WED domain deletion mutants. Data are represented as mean ± SD (n = 3 biologically independent samples). Source data are provided as a Source Data file. m, n Guide:TS DNA recognition. Hydrogen bonding and electrostatic interactions are shown as blue dashed lines. The residues that interact with the nucleic acids through their main chains are shown in parentheses.

Fig. 3. sgRNA architecture.

a Schematic of the sgRNA and target DNA. The upper stem regions of Stem 2 (−142G to −107A), Stem 3 (−91C to −76G), Stem 5 (−39U to −12A) and 5′ region (−194C to −181A) are disordered in the determined structure, suggesting the flexibility of these regions. These disordered regions are shown in dashed line frames. b Structure of the sgRNA scaffold. The disordered regions are denoted by dotted lines. c–j Detailed interactions and triple pairing inside the RNA scaffold. Hydrogen bonding and electrostatic interactions are shown as blue dashed lines.

Fig. 4. sgRNA recognition.

a Recognition of the sgRNA scaffold by RdCas12n. RdCas12n is shown as a surface model. The sgRNA scaffold is recognized mainly through the WED and RuvC domains. b Recognition between the stem 1 and WED. c Recognition between the stem 3 and REC. d Recognition between the stem 3 and WED. e Recognition between the stem 4 and WED. Hydrogen bonding and electrostatic interactions are shown as blue dashed lines and van der Waals interactions are shown as green dashed lines. f Electrophoretic mobility shift assay (EMSA) analysis of binding interactions among RdCas12n or its truncation mutants, sgRNA, and dsDNA. A schematic representation of the RdCas12n variants analyzed in this assay is depicted above the corresponding data panels. Compared to full-length RdCas12n, the NUC lobe-truncated variant displayed reduced binding affinity for dsDNA targets. Experiments were conducted in three independent biological replicates, with consistent results observed across trials. g Scheme of the design for the transcriptional activation assay in E. coli Plasmids carrying various CRISPRa effectors and target-based reporters were co-transformed into E. coli S2060 cells. Successful on-target DNA binding triggers transcription of the downstream luciferase reporter. h Transcriptional activation activities of full length RdCas12n and N-RdCas12n at 9 different target sites. Data are represented as mean ± SD (n = 3 biologically independent samples). Source data are provided as a Source Data file.

Fig. 5. sgRNA engineering improves the genome editing activity of RdCas12n.

a Schematic of different sgRNA truncations for bacterial genome targeting. b Bacterial genome targeting assay using different truncated sgRNAs. +, with spacer; −, without spacer. Effective truncations are highlighted in a red box. Uncropped images are provided in the Source Data file. c Comparison of the genome editing activities of the modified sgRNAs at three different genomic loci of HEK293T cells. Data are represented as mean ± SD (n = 4 biologically independent samples). NT stands for non-targeting spacer. d Schematic of the best sgRNA version (T19) performed in HEK293T cells. e Indel efficiency comparison of the engineered version (T19) with the original version across the 19 different genomic sites (Locus_1 to 19, as specified in Supplementary Data 2) in the HEK293T cells with the 5′-VAAC-3′ PAM (where V = A, C, or G). Data are represented as mean ± SD (n = 4 biologically independent samples). NT stands for non-targeting spacer. f Impact of sgRNA-target DNA mismatch on indel activity in the HEK293T cells. Data are represented as mean ± SD (n = 4 biologically independent samples). Source data are provided as a Source Data file.

Fig. 6. The evolutionary trajectory of type V CRISPR-Cas12 family effectors.

Structural comparison of RdCas12n (PDB ID: 9J09) with TnpB (PDB ID: 8H1J), Un1Cas12f1 (PDB ID: 7C7L), AsCas12f1 (PDB ID: 8J12), Cas12m (PDB ID: 8HHL), Cas12e (PDB ID: 6NY2), and Cas12a (PDB ID: 6I1K). The second Cas12f1 molecule (Mol. 2), as well as the specific REC expansions in Cas12n and Cas12m and REC2 insertions in Cas12e and Cas12a, are highlighted in blue. The region corresponding to the REC expansion in Cas12m that is already present in Cas12n, as well as the REC expansion in Cas12a that is already present in Cas12m, are highlighted in gray. REC1 expansion in Cas12f1, Cas12e and Cas12a are highlighted in green. The WED domain expansions from AsCas12f1 to Cas12a are highlighted in orange, the PI domain in Cas12a is highlighted in red. Other major differences are indicated by dashed line frames.