PMC

(A) Predicted pairing between an expI crRNA and the expI protospacer (PAM is green, spacer is blue, protospacer is red and seed sequence is bold and underlined). (B) Left; protospacer sequences in WT, ΔexpI (PCF81), ΔexpI with a single WT (RBV01), C3T (RBV04), C6T (RBV03) or G-1T (PAM) (RBV02) expI protospacer. Right; toxicity assays of a single anti-expI spacer (expI; pE1-16; white bars) or a plasmid with no spacers (none; pC1-16; black bar) when expressed in the backgrounds shown on the left. (C) Predicted folds and position of mutations for WT, G20A, C18A and C18A/G8U single CRISPR repeats. The black triangle represents the site of cleavage by Cas6f. (D) Toxicity assays of plasmids expressing a single anti-expI spacer flanked by either WT (pE1-16), G20A (pE1-16 G20A), C18A (pE1-16 C18A) and C18A/G8U repeats (pE1-16 C18A/G8U) in WT P. atrosepticum.

Plasmids containing no spacers (control; pC1-16) or one spacer targeting expI (anti-expI; pE1-16) were repressed or induced in the WT and then visualised by (A) LIVE/DEAD staining and fluorescence microscopy (white scale bar; 20 µm; all images to scale) or (B) transmission electron microscopy (TEM) (black scale bars; 2 µm) (C) Quantification of cell lengths of 60 cells from each treatment as assessed by TEM. ns, not significant; *, p-value of <0.0001 when assessed by unpaired two-tailed t-test (PRISM).

(A) P. atrosepticum CRISPR/Cas genomic organisation. The cas1 and the cas2-cas3 hybrid genes are shown in blue, csy1-3 in pale blue, cas6f in orange and the CRISPRs as grey arrows in the direction of transcription. CRISPR2 and 3 are separated by a toxin/antitoxin system (white arrows). (B) Spacer 6 of CRISPR2 (from leader) contains deviations from the repeat consensus (shown in blue) and (C) has a 100% match to a protospacer (red) within eca0560 in HAI2 in the P. atrosepticum genome. The protospacer matching spacer 6 contains a non-consensus 5′-protospacer-TG-3′ PAM (green). (D) Toxicity assays in the WT with plasmids containing consensus repeats and either no spacer (pC1-16), a single native spacer 6 (5′-protospacer-TG-3′ PAM; pTraGS6-16) or an engineered spacer (5′-protospacer-GG-3′ PAM; pTraG1-16) that targets eca0560. The protospacer locations in eca0560 of the native (pTraGS6-16) and engineered (pTraG1-16) spacers are shown below in red. (E) Sequence and pairing of the engineered anti-eca0560 spacer (in pTraG1-16) with the consensus 5′-protospacer-GG-3′ PAM (green).

(A) Strategy for chromosomal targeting (see ). Protospacers were PCR-amplified from target DNA with full or partial repeats and BbsI sites on the primers. PCR products were digested with BbsI and cloned into BbsI-digested plasmids containing a leader sequence and one repeat. This process was repeated to generate multiple spacer inserts. These arrays generate crRNAs that target the template DNA strand but not the mRNA. (B) Transformation plates of P. atrosepticum WT or a Δcas mutant (PCF80) after growth for 36 h on LBA containing Ap. Plasmids transformed contained no spacers (pC1-780), 3 scrambled spacers (pS3-780) and 3 anti-expI spacers (pE3-780). Representatives are shown from experiments performed at least in triplicate.

Plasmids containing either 3 scrambled spacers (pS3-16) or 3 spacers targeting expI (pE3-16) were induced in either (A) WT or (B) Δcas (PCF80) strains or (C) repressed in the WT. (D) Plasmids containing either no spacers (pC1-16) or 3 spacers targeting lacZ (pL3-16) were induced in either WT or Δcas strains and measured after 12 h. (E) Plasmids containing no spacers (pC1-16) or one spacer targeting expI (pE1-16) or lacZ (pL1-16) were induced in the WT. (F) Expression of the cas operon measured from a chromosomal transcriptional/translational lacZ fusion (PCF79). (G) Plasmids containing no spacers (pC1-16) or one, two, three or eight identical spacers (pTraG1-16, pTraG2-16, pTraG3-16 and pTraG8-16) display similar toxicity. (H) Viable count (cfu/ml) after 12 h with either repression or induction of plasmids encoding no spacers (none) or 1× anti-expI (expI). Limit of detection was 1×10². Data shown are the mean ± the SD of three experiments.

(A) Schematic of HAI2 inserted in the P. atrosepticum genome. The attL and attR sites are indicated by black and white boxes respectively, eca0560 and eca0573 are shown as white arrows and the protospacer in eca0560 is indicated in grey and the Km^R marker in eca0573 is depicted in black. Excision of HAI2 results in the circularised form, pHAI2, which contains the attP site and results in the generation of the attB site located within the phe-tRNA gene in the genome. In the absence of the circularised pHAI2 form, the strain is designated ΔHAI2. Primers used for strain confirmation in (B) are shown as black arrows and the cas1 gene was used as a positive control. (B) PCR results for representative class I and class II mutants and the WT. PCR was performed for cas1, eca0560, attR, attL, attP and attB and with primers shown in part (A). Schematic representations of (C) the class I mutants (designated ΔHAI2) that have precisely lost the 97,875 bp island and (D) the class II mutants 2, 5 and 14 containing a 40,227 bp deletion between TTGGCAC sequences in both eca0522 and internal to the Km^R insertion in eca0573. (E) Scale genetic map of the 7 classified HAI2 class II mutants defining the deleted regions. Black bars indicate the presence of the gene as specified. The gray vertical line represents the CRISPR-targeted eca0560 gene. The star represents three accurately sequenced junctions. The blue arrow depicts the site of Km insertion within the chromosome. Class II mutants are numbered as in and their PCR profiles are shown in .