Improved genome editing by an engineered CRISPR-Cas12a

Abstract

CRISPR-Cas12a is an RNA-guided, programmable genome editing enzyme found within bacterial adaptive immune pathways. Unlike CRISPR-Cas9, Cas12a uses only a single catalytic site to both cleave target double-stranded DNA (dsDNA) (cis-activity) and indiscriminately degrade single-stranded DNA (ssDNA) (trans-activity). To investigate how the relative potency of cis- versus trans-DNase activity affects Cas12a-mediated genome editing, we first used structure-guided engineering to generate variants of Lachnospiraceae bacterium Cas12a that selectively disrupt trans-activity. The resulting engineered mutant with the biggest differential between cis- and trans-DNase activity in vitro showed minimal genome editing activity in human cells, motivating a second set of experiments using directed evolution to generate additional mutants with robust genome editing activity. Notably, these engineered and evolved mutants had enhanced ability to induce homology-directed repair (HDR) editing by 2-18-fold compared to wild-type Cas12a when using HDR donors containing mismatches with crRNA at the PAM-distal region. Finally, a site-specific reversion mutation produced improved Cas12a (iCas12a) variants with superior genome editing efficiency at genomic sites that are difficult to edit using wild-type Cas12a. This strategy establishes a pipeline for creating improved genome editing tools by combining structural insights with randomization and selection. The available structures of other CRISPR-Cas enzymes will enable this strategy to be applied to improve the efficacy of other genome-editing proteins.

Figure 1.

Importance of the bridge helix (BH) of LbCas12a protein in regulating its nuclease activities. (A) Illustration of cis- and trans-cleavage activities of Cas12a proteins. Trans-activity of Cas12a RNP can be activated by either direct binding to target ssDNA or after processing its target dsDNA. (B) Schematic presentation of LbCas12a protein. Domain assignment for LbCas12a (upper panel), protein structure (PDB ID: 5XUS) highlighting the bridge helix in LbCas12a (lower panel, left), and bridge helix sequences of different Cas12a orthologs and designed mutations from LbCas12a (lower panel, right). Point mutations are underlined, and deletions are replaced with a triple alanine sequence (AAA) which is also underlined. (C) and (D). In vitro kinetic studies of cis- and trans-cleavage activities by the wild-type (WT) protein and four designed mutants, mut1-4. Each data point is averaged from two independent assays. In the cis-cleavage assays, the non-target strand (NTS) of the dsDNA substrate was 5’-end-labeled with γ-³²P-ATP. In the trans-cleavage assays, a ssDNA substrate was 5’-end-labeled with γ-³²P-ATP. Initial reaction rates are given after each protein symbol; n.d., not detected.

Figure 2.

Mutational effect of W890A on the nuclease activities of LbCas12a. (A, B) Kinetic studies of the cis-cleavage activities on dsDNA (A) and ssDNA (B) by wild-type LbCas12a and mut2 (W890A). TS = target strand; NTS = nontarget strand; * = labeled strand. In these kinetic studies, the labeled strand was fluorescently labeled at the 5’-end with FAM. Each data point in panel (A) is averaged from three independent assays; each data point in panel (B) is averaged from two independent assays. Initial rates are presented after each symbol. (C) Mutational effect of W890A on the cleavage sites of NTS of dsDNA. (D) Genome editing workflow using tdTomato neural progenitor cells (NPCs) from Ai9 mice. The tdTomato gene will be turned on only when editing happens to remove the stop cassette. (E) Genome editing of NPCs by wild-type LbCas12a and mut2. The editing level is reflected by the percentage of tdTomato-positive NPCs (n = 3, means ± SD). In both delivery conditions, the editing efficiency of mut2 is significantly lower than the wild-type protein (***P < 0.001).

Figure 3.

Enhancing the cis-cleavage activity of mut2 by directed evolution. (A) Schematic illustration of the positive selection system used for directed evolution of mut2 LbCas12a. (B) Generation of 3 beneficial mutants from mut2 (starter) by directed evolution. Upper panel showing the region (R) divisions of LbCas12a protein used for error-prone polymerase chain reaction (PCR) mutagenesis. Lower panel showing the mutants selected followed each round of mutagenesis. Three beneficial mutants (named mut2A, mut2B, and mut2C) were generated from two rounds of selection. The mutations in each variant are listed. (C) Snapshots of LbCas12a protein structure (PDB ID: 5XUS) highlighting the locations of the beneficial mutations obtained from directed evolution.

Figure 4.

Enhanced activity of mut2A-C in vitro and in cells. (A–C) In vitro kinetic studies of the cleavage activities on NTS (A) and TS (B) in a dsDNA as well as the trans-cleavage activities (C). * indicates labeled strand. In these in vitro kinetic studies, the labeled strand was fluorescently labeled at the 5’-end with FAM. Each data point is averaged from three independent assays. All three beneficial mutants from directed evolution are more active than mut2. mut2B and mut2C display similar cis-activities on both strands as wild-type, but their trans-activities are dramatically lower relative to the wild-type. (D) Genome editing activity of mut2A-C in HEK293 cells. The upper panel shows the nontarget strand (NTS) ssDNA donors of NTS-L and NTS-S defined by the location of inserts from PAM. Specifically, the insert in NTS-L is located at 20–24nt from PAM, while the insert in NTS-S is located at 11–14nt from PAM. Red arrowheads indicate cleavage sites of LbCas12 proteins on target genomic DNA. Insert above the triangle means an exogenous restriction site is inserted as code for calculation of the rate of HDR. The length of ssDNA donors used in this study is less than 200nt, and the length of PCR amplicons is less than 250nt. The other two panels show the ratios of HDR:indel calculated from the averaged NGS data from NTS-L (middle panel) and NTS-S (bottom panel). The three beneficial mutants from the selection display much better activities in genome editing in HEK293 cells when donor NTS-L is used. Significance tests were carried out between each mutant and the wild-type protein. Values represent the replicate average ± standard error of the mean, where n = 2 for NTS-S experiments. For NTS-L experiments, n = 3 for EMX1-10856 and CCR5-457 and n = 4 for EMX1-4368 and VEGFA-1662. P-values were determined using two-sided Dunnett's test: * P < 0.05, ** P < 0.01, *** P < 0.001.

Figure 5.

Hyper-effective LbCas12a proteins. (A) Schematic presentation of the overall pathway for generating beneficial and hyper-effective (HypE) LbCas12a mutants. These two HypE-mut2B-W and HypE-mut2C-W mutants are generated by restoring W890 in the corresponding beneficial mutants, respectively. HypE-mut2C-WF is a result of the restorations of both W890 and F884 in mut2C. (B, C) In vitro Kinetic studies of the cis- and trans-activities of the HypE-LbCas12a proteins using 20 nM protein, 24 nM crRNA, and 40 nM target-strand-labeled dsDNA. For trans-cleavage assays, 120 nM of labeled ssDNA was used. Each data point is averaged from two independent assays. In these kinetic studies, the labeled strand was fluorescently labeled at the 5’-end with FAM. Initial rates shown after each protein symbol. (D) Genome editing with HypE-mut2C-W and mut2C-WF. Selected loci are difficult to edit in HEK293 cells; significance tests were carried out between each mutant and wild-type LbCas12a. Values represent the replicate average ± standard error of the mean, where n = 2 for all experiments except for IDS-1257 (n = 4). P-values were determined using two-sided Dunnett's test: * P < 0.05, ** P < 0.01, *** P < 0.001.