Harnessing Type I and Type III CRISPR-Cas systems for genome editing

Abstract

CRISPR-Cas (clustered regularly interspaced short palindromic repeats-CRISPR-associated) systems are widespread in archaea and bacteria, and research on their molecular mechanisms has led to the development of genome-editing techniques based on a few Type II systems. However, there has not been any report on harnessing a Type I or Type III system for genome editing. Here, a method was developed to repurpose both CRISPR-Cas systems for genetic manipulation in Sulfolobus islandicus, a thermophilic archaeon. A novel type of genome-editing plasmid (pGE) was constructed, carrying an artificial mini-CRISPR array and a donor DNA containing a non-target sequence. Transformation of a pGE plasmid would yield two alternative fates to transformed cells: wild-type cells are to be targeted for chromosomal DNA degradation, leading to cell death, whereas those carrying the mutant gene would survive the cell killing and selectively retained as transformants. Using this strategy, different types of mutation were generated, including deletion, insertion and point mutations. We envision this method is readily applicable to different bacteria and archaea that carry an active CRISPR-Cas system of DNA interference provided the protospacer adjacent motif (PAM) of an uncharacterized PAM-dependent CRISPR-Cas system can be predicted by bioinformatic analysis.

Figure 1.

Schematic of the CRISPR-based genome editing in S. islandicus. (A) An example of a crRNA (top strand) and its corresponding DNA target designed for lacS gene editing with data presented in Figure 2. The target sequence is positioned from +933 to +972 relative to the ATG start codon (+1)) in the lacS gene, including the protospacer (underlined) and a CCT-PAM (protospacer adjacent motif, shown in red). The presence of the 5′-PAM ensures the Type I-A CRISPR-Cas mediated DNA interference while the mismatch between the 5′-repeat handle and the target sequence (labeled as ‘Mismatch’) induces DNA interference by the Type III-B Cmr-α system. (B) Donor DNAs contain a DNA segment homologous to that flanking the chromosomal target site but it is altered either by Deletion, or Insertion, or Point mutation, such that it is not to be targeted by the endogenous CRISPR systems. (C) Two alternative fates for S. islandicus cells transformed with a pGE plasmid. pGE carrying an artificial mini-CRISPR locus with a single spacer and a donor DNA fragment. The target site is composed of a protospacer and its adjacent sequence, which is to be recognized by a Type I or Type III-B DNA interference system. If recombination did not occur during transformation, the CRISPR DNA interference selectively targets the wild type gene for degradation, leading to cell death; if recombination yielded the mutant gene on the chromosome, the mutant cell is devoid of the CRISPR immunity, forming colonies on plates.

Figure 2.

Generating deletion mutants of lacS gene by CRISPR-Cas systems in S. islandicus. (A) Schematic of the mutant identification by PCR. DNA target is detailed in Figure 1A, including a protospacer (underlined) and a CCT-PAM (protospacer adjacent motif, shown in red), and it serves as a target both for the Type I-A and III-B CRISPR-Cas systems. PCR with primers F1/R1 amplifies the donor DNA from pGE-lacS1 whereas PCR with primers F2/R2 yields PCR products both for the wild-type lacS gene (wt lacS) and for mutant lacS lacking the target site (lacS ⁻). (B) Screening of the enzymatic activity of lacS in S. islandicus strains using X-gal. X-gal was added into S. islandicus cultures and incubated at 78°C for 1 h. E233 – S. islandicus pyrEF mutant as the genetic host for genome editing. E233lacS⁻– Mutants obtained from pGE-lacS1 transformation. (C) PCR screen of lacS deletion in S. islandicus strains. PCR with primers F1/R1 specific for pGE vector amplifies the non-target lacS allele (1283 bp) from the plasmid whereas PCR with primers F2/R2 yields 485-bp and 442-bp DNA fragments that are derived from the wild-type lacS and the deletion mutant allele, respectively. The double bands indicate an escape mutant (indicated with an arrow). (D) Representatives of the chromatographs of the sequencing results of the wild-type lacS and its mutant allele lacS⁻. The protospacer and the CCT-PAM motif (on the opposite strand) that are highlighted in blue and red, respectively, are present in the wild-type lacS but absent from lacS⁻.

Figure 3.

In situ gene tagging of cmr-2α. (A) Schematic of in situ gene tagging strategy. The protospacer is underlined in which the TGA stop codon of cmr-2α is indicated with an asterisk. Mismatches between the 3-flanking DNA stretch of the protospacer and the 5′ repeat handle of the crRNA facilitate DNA interference to the wild-type cells by Type III-B Cmr-α as experimentally demonstrated previously (36). Insertion of 6x His codons in the middle of the protospacer yields a non-target cmr-2α gene. Therefore, cells that carry His-tagged cmr-2α gene are devoid of the DNA interference and selectively retained. (B) PCR screening of cmr-2α-His recombinants generated with S. islandicus Δcas3. Primers F3 is specific for genome DNA while R3 contains the 18 nt encoding the 6xHis tag. All 12 tested strains carry the in situ tagged gene. (C) Representative chromatographs of the sequencing results of the wild-type cmr-2α and cmr-2α-His recombinants. Precise insertion of His-tag coding sequence (highlighted in blue) in front of the stop codon was observed in all analyzed strains. (D) Co-purification of the Cmr-α complex by nickel affinity chromatography purification of His-tagged Cmr-2α. Proteins were stained with Silver staining. (E) Western analysis of His-tagged Cmr-2α using antibody specifically recognizes the His-tag peptide. A peptide of the predicted size is hybridized. The DNA replication protein PCNA3 was used as a reference.

Figure 4.

Mutagenesis of cmr-2α HD domain. (A) A sequence alignment of the N-termini of S. islandicus Cmr-2α (SiRe-0894), Cmr-2β (SiRe-0598) and its Pyrococcus furiosus homolog PF1129. Four conserved amino acids were chosen for constructing substitution mutations (indicated with red asterisks). (B) Schematic of the mutagenic strategy. The I-A target site is underlined and bases highlighted in red are to be mutated whereas bases highlighted in blue represent mutated bases present in DNA donor. The donor DNA containing mutations on the PAM and/or the seed sequence that inactivate the Type I-A DNA interference activity is provided for recombination. (C) Chromatographs of the sequencing results for the three types of cmr-2α mutant. Sequencing of seven transformants yielded three different types of mutant, carrying either HD mutation, or HD+K mutation or all four designed mutations.