Adaptive immunity of type VI CRISPR-Cas systems associated with reverse transcriptase–Cas1 fusion proteins

Figure 1.. Acquisition of new spacers by CRISPR-Cas-type VI-A/RT systems. (A) The genomic arrangement of representative variants of the type VI-A/RT systems studied here is illustrated, with the phylogeny of RT–Cas1 proteins depicted on the left. The schematic phylogenetic tree was similar to that previously reported (17). Entries corresponding to unlabeled branches are listed in Supplementary Table S4. Diagrams are drawn approximately to scale. Gray boxes indicate the predicted leader regions of arrays. Rhodovulum k. DSM 19783, NZ_QAYC01000027.1; E. rectale AF25‐15, NZ_QRUJ01000006.1; Eubacteriaceae bacterium CHKCI004, NZ_FCNR01000048.1. (B) Exogenous RT activity was measured by assessing the incorporation of [32P]-dTTP into a poly(rA) substrate primed with oligo(dT)18. Polymerization is presented as a percentage of the RT–Cas1 activity of Rk_WT. The activity of the RT–Cas1 protein of a CRISPR-Cas Type III-D system from Vibrio vulnificus (16) is included as a control. The data were derived from three technical replicates and at least one biological replicate and are presented as the mean ± standard error of the mean (SEM). Rk (R. kholense, Er (E. rectale), CH (Eubacteriaceae bacterium), Vv…

Figure 2.. Characterization of spacers acquired by the Eubacteriaceae bacterium CHKCI004 adaptation system. (A) Distribution of newly detected spacers by length identified either in the E. coli chromosome or the plasmid. (B) Distribution of GC (Guanine + Cytosine) content in the genome and plasmid-aligned spacers. The dotted lines represent the mean GC content of the plasmid and the genome. (C) Protospacer composition analysis for GC content, frequency and statistical probability. Due to the low numbers of spacers acquired from the genome, plasmid and genome-derived spacers were combined and analyzed together. The spacer sequence is shaded and the spacer boundaries are marked as dotted lines. Owing to the different lengths of the detected spacers, the protospacer sequence was split into a 5′ protospacer (upper panels) and a 3′ protospacer (lower panels). Dotted lines on the GC content plot correspond to the mean % GC in the E. coli genome. On the statistical probability graph [-log10 (P)], the size of the nucleotides represented indicates their relative probability (average across all positions) in the protospacer sequences analyzed according to k-mer length = 1, 2, 3, 4 in a binomial test.…

Figure 3.. CRISPR-Cas type VI-B systems associated with RTs. (A) Schematic diagram of the phylogenetic tree constructed from the alignment of 228 Cas13b protein sequence variants, distinguishing between subtypes VI-B1 (loci with Csx27) and VI-B2 (loci with Csx28). A distinct cluster (VI-B1/RT) within subtype VI-B1 (see Supplementary Table S4 and Supplementary Figure S3A), characterized by the absence of Csx27 loci, appeared to be the origin of the variant associated with RTs. Black dots indicate the cas13b genes within the loci studied here. The corresponding tree newick file is provided as Supplementary File S2. (B) Schematic phylogenetic tree derived from the alignment of 124 sequences, including the magnesium transport protein CorA, and related proteins: the zinc transport protein ZntB, the mitochondrial inner membrane magnesium transporter Mrs2, and the magnesium transporters Alr1 and Alr2 from the IPR002523 protein family. Branches corresponding to CorA-like proteins associated with CRISPR-Cas III-B or subtype VI-B systems are differenciated. Black dots indicate the corA-like genes within the loci studied. The corresponding tree newick file is provided as Supplementary File S3. The…

Figure 4.. Spacer acquisition in type VI-B CRISPR systems (A) Spacer acquisition by the adaptation modules of the three type VI-B systems (described in Figure 3C) in the heterologous E. coli system. Black bars represent the frequency of unique aligned spacers per million reads for each RT-CRISPR-Cas system: (Hm) metagenomic assembly, (Pm) P. muris and (Ds) Dysgonomonas sp. A negative control is included for each system in which the cas2 gene is deleted (ΔCas2). The error bars indicate the range for at least three biological replicates. (B) Percentage of aligned spacers relative to the P. muris WT system normalized per sequencing read of the different mutants is represented: ΔCas2 lacks the Cas2 protein sequence in the adaptive operon; RT-, RT–Cas1 protein with a mutation affecting the catalytic core for RT activity; YADD → YAAA; and Cas1-, the RT–Cas1 coding gene is modified to replace a glutamate residue with alanine (E665A), a substitution predicted to abolish Cas1 function. Using the pUC constructs 97% of the acquired spacers were from the genome and 3% from the plasmid (C) Exogenous RT activity measured as in Figure 1. Polymerization is presented as a percentage of the RT–Cas1 activity…

Figure 5.. Characteristics of the spacers acquired by the P. muris adaptation system. (A) Distribution of newly detected spacers by length identified either in the E. coli chromosome or in the plasmid encoding the adaptation operon. (B) GC content distribution in the genome and plasmid-aligned spacers. The dotted lines represent the mean GC content. (C) Strand bias in pools of newly acquired spacers relative to the source transcript. Bar plot representing the proportion of newly acquired spacers with the WT RT–Cas1 in the sense or antisense strand of coding genes (86.75%) or in IGRs (13.25%) of the E. coli genome (n = 9). (D) Genome-derived protospacer composition analysis for GC content, frequency and statistical probability. The spacer sequence is shaded and the spacer boundaries are marked as dotted lines. Due to the different lengths of the detected spacers, the protospacer sequence is split into a 5′ protospacer (upper panel) and a 3′ protospacer (lower panels). The dotted line on the GC content plot corresponds to the mean % GC in the E. coli genome. On the statistical probability graph [−log10 (P)], nucleotides are represented such that their size indicates their relative probability…

Figure 6.. PmCas13b CRISPR array processing and in vivo interference activity. (A) Protospacer design for the MS2 phage drop plaque assay to test RNA interference. (B) Schematic diagram of the pBAD derivatives containing Cas13b gene sequences. (C) Cas13b encoded by the P. muris VI-B locus processes the CRISPR array. Northern blot analysis of exponential growth phase E. coli cell cultures in the presence of 1% arabinose, with the constructs depicted in (B). Antisense strand-specific oligonucleotides against the DR (DRAS probe) were used to probe total RNA. The expected sizes of the bands are indicated. At the bottom, the 5S RNA was probed as an RNA loading control. (D) MS2_Drop plaque assay for pBAD (pEmpty) and derivatives containing cas13b as depicted in (B). (E) MS2 phage growth assays measuring phage production over time for E. coli strains carrying the indicated plasmids and infected with the MS2 phage at an MOI of 0.1. Data are shown as mean ± SEM for n = 4 biological replicates per treatment. Statistical significance was calculated using repeated-measures ANOVA (ANalysis Of VAriance) by using the Greenhouse–Geisser and Huynh–Feldt corrections, followed by Dunnett’s multiple comparison…

Figure 7.. The presence of Cas13b does not change the spacer acquisition pattern in P. muris. (A) Schematic diagram of the pBAD derivatives containing Cas13b sequences. (Pm) A plasmid containing the RTCas1–Cas2 sequences and a minimal unit of the CRISPR array expressed under the control of an arabinose-inducible promoter. (Pm1) Pm plasmid with the gene encoding the corresponding Cas13b, Cas13b mutated at the two HEPN sites (Pm1-D1D2) and Cas13b mutated at the two K residues of the Lid region (Pm1-KK). (B) Strand bias in pools of newly acquired spacers relative to the source transcripts. Proportion of newly acquired spacers with the four different constructs in the sense (CDSs) or antisense (CDSas) strand of coding genes or in IGRs of the E. coli genome (n = 3). The Pm and derivative systems using pBAD constructs in strain BW25113 acquired 80% of the spacers from the genome and 20% from the plasmid. (C) Protospacer composition analysis for GC content at each position along the acquired spacers in the sense (above) or antisense (below) strand of coding genes obtained from cultures of E. coli containing the various constructs (figure legend) is represented. The spacer sequence is shaded. Due…

Figure 8.. Spacer acquisition from RNA by the P. muris adaptation system. (Top panel) Schematic diagram of td intron-containing constructs. We determined whether the spacers originated from RNA by using a self-splicing transcript that produces an RNA sequence junction not encoded by DNA. High-level coverage of spacers annealed with the tdI gene is shown (sense and antisense spacers are identified in different colors). (Bottom panel) Alignment of newly acquired exon junction-spanning spacers (considered to have been acquired from an RNA target), indicated as arrows. Note that both spacers are in the antisense orientation relative to the direction of transcription of the tdI intron. Flanking exons are shaded in different colors.