Materials and Methods

The Rsm (Csr) post-transcriptional regulatory pathway coordinately controls multiple CRISPR–Cas immune systems

MATERIALS AND METHODSStrains and growth conditionsThe bacterial strains used in this study are listed in Supplementary Table S1. All plasmids are listed in Supplementary Table S2, including details of their construction, and oligonucleotides used are listed in Supplementary Table S3. All strains and plasmids were confirmed by Sanger sequencing. Serratia sp. ATCC 39006 strains were grown at 30°C, Pectobacterium atrosepticum SCRI1043 at 25°C and E. coli strains at 37°C in lysogeny broth (LB) medium and on plates containing 1.5% (w/v) agar (LBA). Media was supplemented with antibiotics at the following concentrations: Kanamycin (Km) at 50 μg/ml; Ampicillin (Ap) at 100 μg/ml; Chloramphenicol (Cm) at 25 μg/ml, Spectinomycin (Sp) at 50 μg/ml, Gentamicin (Gm) at 50 μg/ml and Tetracycline (Tc) at 10 μg/ml. Five-aminolevulinic acid (ALA; 50 μg/ml) was added for growth of ST18. Bacterial growth was measured as optical density using a Jenway 6300 Spectrophotometer at 600 nm (OD600). When grown in 96-well plates, plates were incubated in an IncuMix shaker and optical density was measured using a Varioskan LUX Multimode Reader (Thermo Fisher Scientific) at 600 nm.Reporter plasmids and gene expression assaysSerratia strains, each containing a plasmid with the fluorescent reporter zsGreen fused to a promoter of a different cas gene, were used in plate reader assays to measure gene expression (Supplementary Figure S1). The major transcriptional start sites were determined from an in-house unpublished transcriptional start site RNA seq dataset. Plasmids used to measure expression were: cas3 (type I-E, pPF1973), cas1 (type I-F, pPF1891), cas10 (type III-A, pPF1890) and rsmB (pPF1976) and were compared with the empty vector (pPF1854). Pigmentless (prodigiosin-deficient) versions of all the strains were used for these assays to remove any overlap with fluorescence measurements (PCF396 ΔpigA-O background). Strains were generated by generalised transduction of marked mutations into the PCF396 ΔpigA-O background as described previously (31). Mutant derivatives of PCF396 were used in these assays as follows: PCF398 (rsmA), PCF406 (rsmB), PCF629 (pigQ), PCF630 (pigW), PCF631 (pigX), PCF694 (rsmS), PCF703 (rsmA, rsmS), PCF675 (rsmA, rsmB), PCF704 (rsmA, pigX), PCF705 (rsmS, rsmB), PCF706 (rsmS, pigX) and PCF717 (rsmB, pigX). Strains carrying appropriate plasmids were grown in a 96 deep-well plate overnight in 1 ml LB containing Gm. For all reporter assays, the OD600 of the overnight cultures was measured and adjusted to 0.02 to inoculate a 96-well black sided, clear bottom plate (with the appropriate antibiotics) to monitor OD600 as well as fluorescence with excitation at 490 nm and emission at 510 nm for 23 h. For all endpoint reporter data presented, fluorescence was normalised by the OD600.Complementation assays were performed similarly using the strains listed above, but containing either an empty vector control (pPF781) or the expression vector for each gene: rsmA (pPF1958), rsmB (pPF1959), pigQ (pPF1960), pigW (pPF1961), pigX (pPF1962) and rsmS (pPF1964). Complementation strains were grown in LB containing Gm + Cm.Sample and library preparation for RNA sequencingOvernight cultures of Serratia LacA (WT) and the rsmA mutant (NMW7) were subcultured into 25 ml LB medium in 250 ml flasks in biological triplicates to a starting OD600 of 0.05. The cultures were grown in LB broth for 12 h at 30°C with shaking at 200 rpm, and 2 ml samples were collected for RNA extraction. The twelve-hour time point (early stationary phase) was previously established as a point of elevated CRISPR–Cas activity due to the rising density of bacterial populations (18). Bacterial cells were harvested by centrifugation and the resulting cell pellets were resuspended in RNAlater (Thermo Fisher Scientific) and stored at -20°C until further processing. The Qiagen RNeasy kit was used to extract total cellular RNA. TurboDNase (Thermo Fisher Scientific) was added to remove genomic DNA (gDNA). Samples were confirmed to be gDNA-free by means of PCR analysis with primers PF796 and PF797 that are designed to amplify the flhDC operon of Serratia. Quality control checks of the resulting RNA samples were performed using the nanodrop (Thermo Fisher Nanodrop one) and 2100 Bioanalyzer (Agilent Genomics).Ribosomal RNA (rRNA) was depleted using a RiboCop rRNA depletion kit (Lexogen). Synthesis of antisense cDNA was initiated through ligation of a TruSeq adaptor sequence (Illumina) to the 3′ OH end of the fragmented RNA. Next, the antisense cDNA was purified, followed by a ligation of a 5′ sequencing adaptor to the 3′ end of the antisense cDNA. The cDNA was then amplified using PCR and the resulting products were gel fractionated to satisfy the size requirements for Illumina sequencing. Lastly, cDNA libraries were sequenced with the Illumina HiSeq 2500 V2 Rapid sequencing system to an average depth of around 10 million reads per library, generating an output in the form of 100 bp demultiplexed reads in FASTQ format.RNA sequencing data analysisGenerated reads in FASTQ format were initially processed by removing adaptors and low-quality reads using Trimmomatic (32). Additionally, quality assessment of the reads was carried out using FastQC (Babraham Bioinformatics, Bowtie2 (33) was used with default parameters for mapping reads to the reference genome of Serratia sp. ATCC 39006 (accession number: {"type":"entrez-nucleotide","attrs":{"text":"CP025085","term_id":"1315974439","term_text":"CP025085"}}CP025085), followed by a conversion to BAM format for analysis using SAMtools (34). Statistical analysis was performed using the default settings of the DESeq2 package of R/Bioconductor to identify differentially expressed transcripts with a false discovery rate (FDR) of less than 5% and a fold change >1.5 (i.e. log2 >0.58) (35). Full RNA-seq outputs from DESeq2 are provided in Supplementary Table S4.Type III-A CRISPR–Cas interference assaysCRISPR–Cas interference assays are based on conjugation efficiency and were carried out as described previously (18). Briefly, E. coli ST18 was used as the donor for conjugation of an untargeted control (pPF781) and type III-A targeting plasmid (pPF1043). Plasmid pPF1043 has a protospacer targeted by a spacer in CRISPR3. Strains were grown overnight in LB with appropriate antibiotics, following which, 1 ml of culture was centrifuged at 17 000 × g and the supernatant removed. Cell pellets were washed with 500 μl LB + ALA to remove antibiotics and resuspended in 1 ml LB + ALA. Washed and resuspended cells were adjusted to an OD600 of 1. E. coli donors and Serratia recipients (WT or NMW7 (rsmA)) were mixed in a 1:1 ratio and 20 μl spotted on LBA + ALA + glucose (0.2% w/v) and incubated at 30°C for 20 h. Conjugation spots were scraped from the plate and added to 500 μl phosphate buffered saline (PBS) to resuspend cells. Resuspended cells were serially diluted (undiluted to 10−8) and plated onto LBA + arabinose (0.02% w/v) for recipient counts or LBA + arabinose + Cm for selection of transconjugants. Arabinose was included to induce the protospacer transcription necessary for type III-A targeting. CRISPR–Cas interference was assessed using conjugation efficiency, which was measured as a ratio of transconjugants divided by recipients. In addition, to assess the effects of rsmA overexpression on interference, conjugation efficiency assays were performed similarly with the following recipients (NMW7 (rsmA) + pQE-80L-oriT (control) and NMW7 (rsmA) + pPF513 (RsmA)). Bacteria were plated onto LBA + Ap + arabinose (0.02%) for recipient counts or LBA + Ap + Cm + arabinose for selection of transconjugants. Conjugation efficiency data was log-transformed and normality was assessed using Shapiro-Wilk and Kolmogorov-Smirnov tests. A Student's t-test (two-sided) was used on log-transformed data to determine statistical significance.Type I CRISPR–Cas interference assaysSerratia type I conjugation efficiency assays were performed the same as type III, but with the following variations. An untargeted control (pPF719) and type I-E (pPF724) or type I-F (pPF722) targeting plasmids were used. Plasmids pPF724 and pPF722 each contain a protospacer targeted by a spacer from either CRISPR1 (type I-E) or CRISPR2 (type I-F). Recipient Serratia strains used were WT (LacA) and an rsmA mutant (NMW7). Following conjugation, bacteria were plated onto LBA for recipient counts or LBA + Tc for selection of transconjugants. In addition, to assess the effects of rsmA overexpression on interference, conjugation efficiency assays were performed with the following recipients (NMW7 (rsmA) + pPF781 (control) and NMW7 (rsmA) + pPF1958 (RsmA)), which were grown with arabinose (0.1% w/v). Bacteria were plated onto LBA + Cm + arabinose (0.1% w/v) for recipient counts or LBA + Cm + Tc + arabinose for selection of transconjugants. Pectobacterium type I-F conjugation efficiency assays were carried out the same as for Serratia type I systems, with the following variations. An untargeted control (pPF571) and type I-F targeting plasmid (pPF572) were used. Plasmid pPF572 contains a protospacer targeted by a spacer from CRISPR1. Strains used were WT (P. atrosepticum SCRI1043) and an rsmA mutant (AE9). Conjugation efficiency data was log-transformed and normality was assessed using Shapiro-Wilk and Kolmogorov-Smirnov tests. A Student's t-test (two-sided) was used on log-transformed data to determine statistical significance.Primed CRISPR adaptation assayPrimed CRISPR adaptation was analysed using priming plasmids that lead to acquisition of new spacers. Plasmids (pPF953, naïve; pPF1233, I-E primed; pPF1236, I-F primed) were conjugated into Serratia WT (LacA) and rsmA (NMW7) from E. coli ST18 donors and the transconjugants were grown in LB with Tc selection and 25 μM IPTG for 24 h. The I-E and I-F primed plasmids contain a protospacer with a non-canonical protospacer adjacent motif, that promotes acquisition of new spacers through priming. From those cultures, new overnight cultures without antibiotic selection, but containing IPTG, were inoculated and grown for 24 h. From those cultures, 10 μl were used to inoculate new overnight cultures containing 5 ml of LB without selection and grown for 24 h at 30°C. New cultures were inoculated every 24 h for up to two days to observe priming. Aliquots of culture from each day were mixed 1:1 with 50% glycerol and frozen at -80°C for future use. CRISPR array expansion (indicative of adaptation) was assessed via PCR (20 cycles) with primers PF633/PF2177 for type I-E (CRISPR1) and PF1888/PF1990 for type I-F (CRISPR2). PCR samples were run on a 2% (w/v) agarose gel with ethidium bromide in sodium borate buffer. To assess the effect of rsmA expression on CRISPR adaptation, plasmids (pPF953, naïve; pPF1233, I-E primed; pPF1236, I-F primed) were conjugated into strains NMW7 (rsmA) + pPF781 (control) and NMW7 (rsmA) + pPF1958 (RsmA) in the presence of Cm for expression plasmid maintenance and arabinose (0.1% w/v) for rsmA expression. Serial passaging and PCR was performed as described above, with the addition of Cm and arabinose (0.1% w/v) to culture media.Construction of a FLAG-tagged RsmA strainA C-terminal rsmA-3xFLAG chromosomal fusion was constructed using allelic exchange mutagenesis with sucrose selection. A plasmid (pPF1811) was constructed using a gBlock (PF3562) containing 500 bp upstream and downstream the rsmA gene and the 3xFLAG, and the sacB-containing suicide plasmid pPF1117 as a backbone. The plasmid pPF1811 was then conjugated into Serratia and plated onto LBA containing Cm to select for recombination of the plasmid with the chromosome. An overnight culture without selection was plated on LBA with 10% (w/v) sucrose to select for recombination leading to plasmid (sacB) loss. The rsmA region was screened by PCR (PF3563 and PF3565) and the mutant confirmed by Sanger sequencing.Co-IP and RIP-seq sample preparationCo-immunoprecipitation of RsmA-3xFLAG with an anti-FLAG antibody and Protein A-Sepharose beads was performed from Serratia lysates of WT (LacA control) and isogenic RsmA-3xFLAG strain (PCF624). The cells were grown overnight and 100 ml LB cultures were inoculated to an OD600 of 0.05 in 500 ml flasks and grown to late exponential phase (OD600 of 0.6) at 30°C as described earlier. Cells were harvested by centrifugation at 6000 × g for 20 min at 4°C. The cell pellets corresponding to a total OD600 of 60 were then resuspended in 1 ml of Buffer A (20 mM Tris–HCl pH 8.0, 150 mM KCl, 1 mM MgCl2 and 1 mM dithiothreitol (DTT)) and subsequently harvested by centrifugation (10 min, 11 000 × g, 4°C). The pellets were snap-frozen in liquid nitrogen and stored at -80°C.When required, the pellets were thawed on ice and resuspended in 1 ml of lysis buffer (Buffer A containing 1 mM PMSF, 0.2% (v/v) Triton X-100, DNase I 0.02 U/μl and RNase inhibitor 0.4 U/μl) and transferred to FastPrep tubes (MP biomedicals) containing silica spheres. Cells were lysed using FastPrep24 running two lysis steps at the speed of 4 m/s for 15 s, with 5 min cooling on ice between each step. Cell debris was then separated by centrifugation at 17 000 × g for 10 min at 4°C and the supernatant (lysate fraction) was collected into a new tube. Incubation of the samples with 35 μl of anti-FLAG antibody (Monoclonal anti-FLAG M2, Sigma, #F1804) was carried out for 30 min at 4°C on an orbital shaker. Then 75 μl of ProteinA-Sepharose (Sigma P6649 6MB aqueous ethanol suspension) were washed with Buffer A and added to the lysate containing the antibody. Solutions were then incubated for a further 30 min at 4°C on an orbital shaker. The beads with the lysate were then subjected to centrifugation at 15 000 × g for 1 min at 4°C. After centrifugation, the supernatant was removed. The beads were washed 5 times with 500 μl of Buffer A. A phenol–chloroform–isoamyl (PCI) extraction was performed, adding 500 μl of Buffer A and 500 μl of PCI to the beads, followed by inversion of the tube and 5 min incubation at room temperature (RT). A 30 min centrifugation step at 17 000 × g at 4°C was performed to pellet the RNA.An overnight precipitation at –20°C was performed by adding the supernatant to a tube containing 1 ml of 30:1 mix (absolute ethanol:sodium acetate) plus 1 μl of GlycoBlue. The RNA precipitate was subjected to centrifugation at 17 000 × g for 30 min at 4°C and washed with 500 μl of ice-cold 70% ethanol followed by a further 10 min centrifugation in the same conditions. Then the pellet was left to dry by opening the tube at room temperature. A DNase I digestion was carried out to eliminate all possible DNA, followed by a PCI extraction, overnight precipitation in 30:1 mix (absolute ethanol:sodium acetate) and final wash in 70% ice-cold ethanol as previously explained. The absence of DNA contamination was confirmed by PCR of the pigQ gene (PF3143 and PF3144).SDS-PAGE and western blotting of RsmA-3xFLAGProtein samples (equivalent to an OD600 of 1) were collected during Co-IP experiments to confirm enrichment of RsmA-3xFLAG by western blotting. These cell pellets were resuspended in the appropriate volume of protein loading buffer (62.5 mM Tris–HCl pH 6.8, 100 mM DTT, 10% (v/v) glycerol, 2% (w/v) SDS and 0.01% (w/v) bromophenol blue) to make final concentration of 0.1 OD600/μl per sample and boiled at 95°C for 8 min. A total volume of 20 μl for culture protein (C), lysate (L), supernatant 1 (SN1), supernatant 2 (SN2) and wash (W) fractions (corresponding to 0.2 OD600) and 20 μl from the elution (E) fraction (equivalent to an OD600 of 10) were resolved using 15% SDS-polyacrylamide gels and transferred onto nitrocellulose membrane using semi-dry blotting. After transfer, membranes were blocked for 1 h in a 10% milk powder solution in 1× phosphate-buffered saline containing 0.01% sodium azide (PBS-A) buffer followed by washing. An overnight incubation at 4°C of the primary antibody (monoclonal anti-FLAG 1:1000 or monoclonal antibody specific for GroEL 1:10 000 were diluted in PBS-A) was performed, followed by washing with 1× Tris-buffered saline with 0.001% Tween 20 (TBST20) and incubation with the horseradish peroxidase (HRP)-coupled secondary antibody (anti-mouse IgG 1:10 000 in TBST20 for anti-FLAG or anti-rabbit IgG 1:10 000 in TBST20 for anti-GroEL) for 1 h at 4°C. After washing, the membrane was developed using the HRP catalyzed oxidation of its substrate luminol, in presence of hydrogen peroxidase and p-hydroxicoumaric acid as an enhancer. The chemiluminescence output was detected using a CCD camera-equipped Image Quant LAS 4000 machine.RIP-seq and data analysisRNA was converted to cDNA using the adapter ligation method by Vertis Biotechnologie AG, Germany ( Briefly, the eluted RNA after DNase I treatment was subjected to oligonucleotide adapter ligation on the 3′ end, first-strand cDNA synthesis using M-MLV reverse transcriptase (Agilent), and Illumina TruSeq sequencing adapter ligation on the 3′ end. The resulting cDNA was PCR-amplified using Herculase II Fusion DNA Polymerase (Agilent) with 13 amplification cycles following the manufacturer's instructions, purified using Agencourt AMPure XP kit (Beckman Coulter Genomics) following the manufacturer's instructions, and analyzed by capillary electrophoresis. After cDNA purification and quantification with capillary electrophoresis, two cDNA pools were constructed, according to ratios calculated from the quantification, for Illumina NextSeq sequencing (data collected from a protocol provided by Vertis Biotechnologie AG). The resulting samples were then run on an Illumina NextSeq 500 instrument with 76 cycles in single-read mode (Supplementary Table S5).RIP-seq data analysis was performed as follows. First, demultiplexing was performed using bcl2fastq v2.20.0.422. Illumina reads were quality controlled and adapter trimmed with Cutadapt (36) version 2.5 using a cutoff Phred score of 20 in NextSeq mode and reads without any remaining bases were discarded (command line parameters: --nextseq-trim=20 -m 1 -a AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC). Afterwards, we applied the pipeline READemption (37) version 0.4.5 to align all reads longer than 11 nt (-l 12) to the Serratia reference genome (RefSeq assembly accession: GCF_002847015.1) using segemehl version 0.2.0 (38) with an accuracy cut-off of 95% (-a 95). We used READemption gene_quanti to quantify aligned reads overlapping genomic features by at least 10 nts (-o 10) on the sense strand (-a). For this, we applied RefSeq annotations (antisense_RNA, CDS, ncRNA, riboswitch, RNase_P_RNA, rRNA, SRP_RNA, tmRNA, tRNA, direct_repeat) for assembly GCF_002847015.1 (annotation date 16 March 2017) in GFF format supplemented with annotations for strand-specific intergenic regions, and sRNAs based upon homology to Salmonella sRNAs. For annotation-based identification of enriched genomic regions based on biological duplicates, we used DESeq2 (35) version 1.24.0. Size factors for normalization were calculated manually by dividing the total number of aligned reads for each library by the minimum over all libraries. Fold-change shrinkage was applied by setting ‘betaPrior = TRUE’ and testing only for enrichment in the RsmA-3xFLAG compared to the WT CoIP libraries was conducted by setting altHypothesis = ‘greater’. Features were considered significantly enriched with a adjusted P-value <0.005 (Supplementary Table S6). In addition, we applied READemption to generate coverage plots representing the numbers of mapped reads per nucleotide. Here, we used sequencing depth-normalized plots from output folder coverage-tnoar_min_normalized for visualization in the Integrated Genome Browser (IGB) (39).Peak calling and RsmA motif analysisPeak calling for RIP-seq libraries was performed using the sliding window approach implemented in the tool PEAKachu (, version 0.1.0, commit c869dc5583c0ccd9981c0576d38ce388f2df958c). The tool was run using as input READemption BAM files for the two replicates of RsmA-3xFLAG and WT, respectively. By calling peakachu window, we calculated library-specific read count values of genomic regions using window size (-w) 25 and step size (-l) 5. Window count normalization and detection of significantly enriched windows subsequently merged into peaks was conducted via methods implemented in DESeq2 (35) using the following PEAKachu parameters: normalization method (-n) manual, statistical test (-d) deseq, mad-multiplier (-m) 6.0, fold change (-f) 10.0 and Benjamini-Hochberg adjusted P-value (-Q) 0.05. Size factors for normalization (-s) were calculated by dividing the total number of aligned reads for each library by the minimum over all libraries and annotations in GFF format (see above) were used to map overlapping features to called peaks. Enriched Co-IP sequences from PEAKachu that reached significance were analyzed by MEME version 5.2.0 to generate consensus motifs ( (40). The analysis was run using default settings, except for restricting the motif width to 6-10 characters.Phage infection assaysTo test for the effects of RsmA on phage sensitivity independently of CRISPR immunity, the WT (LacA) and an rsmA mutant (NMW7) were grown overnight in LB with no supplements. Stationary phase cultures were normalised to an OD600 of 0.05 in 180 μl LB in a 96-well plate. Dilutions of ΦJS26 lysate were prepared in phage buffer (10 mM Tris–HCl pH 7.4, 10 mM MgSO4 and 0.01% (w/v) gelatine) and 20 μl was added to each well to produce the desired MOI. For the uninfected controls, 20 μl of phage buffer was added to wells. Plates (96-well) were incubated in a plate reader (Varioskan Flash, Thermo Scientific) at 30°C with shaking at 300 rpm and OD600 measurements taken every 12 min for 20 h. Each condition was repeated in triplicate with data plotted as the mean +/− the standard deviation.To assess CRISPR–Cas protection against phage infection, plasmids harbouring mini-CRISPR arrays expressing anti-ΦJS26 spacers (pPF1473 - type III; pPF1485 - type I-E; pPF1489 - type I-F) or a control plasmid containing no mini-CRISPR array (pPF260 - control) were introduced into strains WT (LacA) and an rsmA mutant (NW64). Strains were grown overnight in LB with kanamycin (50 μg/ml) for plasmid maintenance and IPTG (0.1 mM) for CRISPR-array expression. Stationary phase cultures were normalised to an OD600 of 0.05 in 180 μl LB + kanamycin + IPTG in a 96-well plate and the growth conditions performed as described in the previous paragraph.Identification and analyses of RsmA in phage genomesWe first searched Caudovirale, Kalamavirale, Levivirale, Mindivirale, Tubulavirale and Vinavirale genomes in GenBank for RsmA homologs via a hidden Markov model (HMM) search using the Pfam CsrA/RsmA HMM (PF02599) with HMMER3 ( We used a full sequence E-value cut-off of 10−4 and a 50% coverage threshold for the HMM and 30% coverage of the target sequence. The phage CsrA/RsmA sequences were filtered to remove duplicates using CD-HIT (41), aligned using MUSCLE (42), manually curated, then used to build a phage-CsrA/RsmA HMM with HMMER3. The phage-CsrA/RsmA HMM was used to search the GenBank Caudovirale, Kalamavirale, Levivirale, Mindivirale, Tubulavirale and Vinavirale genomes and non-eukaryotic viral contigs in the IMG/VR database v3 (September 2020). We applied a full sequence E-value cut-off of 10−4 and a 50% coverage threshold for the HMM and 30% coverage of the target sequence. Classification of whether host bacteria encoded CsrA/RsmA was based on the prevalence of RsmA homologs within sequenced genomes (RefSeq) of the corresponding host taxonomic genus. First, we used an HMM search (as above) to determine whether each bacterial genome in RefSeq (September 2020) encoded CsrA/RsmA. For each bacterial genus, based on the NCBI taxonomy, we then determined the proportion of CsrA/RsmA-encoding genomes, which revealed that CsrA/RsmA is generally either very common or almost entirely absent from any given genus (Supplementary Figure S9). Genera where <20% of genomes encoded CsrA/RsmA were classed as typically lacking CsrA/RsmA, whereas genera with >80% of genomes encoding CsrA/RsmA were classed as hosts with CsrA/RsmA. Alignment of CsrA/RsmA homologs was performed using MUSCLE (42). Protein accessions (NCBI) for Figure ​Figure6E6E are: {"type":"entrez-protein","attrs":{"text":"NP_417176.1","term_id":"16130603","term_text":"NP_417176.1"}}NP_417176.1 (E. coli K12), {"type":"entrez-protein","attrs":{"text":"WP_005972168.1","term_id":"492811525","term_text":"WP_005972168.1"}}WP_005972168.1 (Serratia ATCC 39006), {"type":"entrez-protein","attrs":{"text":"QBQ72211.1","term_id":"1603721009","term_text":"QBQ72211.1"}}QBQ72211.1 (Serratia phage Parlo), {"type":"entrez-protein","attrs":{"text":"AXF51437.1","term_id":"1433277135","term_text":"AXF51437.1"}}AXF51437.1 (Erwinia phage Pavtok), {"type":"entrez-protein","attrs":{"text":"AEZ50864.1","term_id":"374427970","term_text":"AEZ50864.1"}}AEZ50864.1 (Burkholderia phage DC1), {"type":"entrez-protein","attrs":{"text":"AXG67699.1","term_id":"1435207262","term_text":"AXG67699.1"}}AXG67699.1 (Ralstonia phage GP4) and {"type":"entrez-protein","attrs":{"text":"QIW86647.1","term_id":"1829638232","term_text":"QIW86647.1"}}QIW86647.1 (Klebsiella phage LASTA).Open in a separate windowFigure 6.Mutation of rsmA results in sensitivity to phage infection, but is overcome by CRISPR–Cas immunity. (A) Mapping of the Co-IP RNA-seq reads demonstrates direct binding of RsmA to flagella loci, including the mRNA of the flhDC master regulator. The WT control (grey) and RsmA-3xFLAG Co-IP (green) are shown. Replicate 1 (R1) samples are shown for the top strand (no significant enrichment was observed on the bottom strand). (B) The rsmA mutant has increased sensitivity to phage JS26 compared with the WT strain (n = 3 biologically independent samples). (C) CRISPR immunity conferred by the type I-E, type I-F and type III-A CRISPR–Cas systems provides phage resistance to both the WT and rsmA mutant strains. In graphs, lines represent the mean ± the standard deviation. MOI = multiplicity of infection. (D) Bioinformatic identification of phages encoding RsmA. *For the IMG/VR dataset, only viral assemblies with an estimated completeness of >80% were searched. Inference of the hosts encoding CsrA/RsmA is based on whether the corresponding host genus typically encodes CsrA/RsmA. *Only host genera with five or more sequenced phages or viral contigs in the dataset are included. (E) Sequence comparison of select CsrA/RsmA homologs from bacteria and phages. See also Supplementary Figure S9.

Article TitleThe Rsm (Csr) post-transcriptional regulatory pathway coordinately controls multiple CRISPR–Cas immune systems


The RNA-seq and RIP-seq data in this publication have been deposited in NCBI’s Gene Expression Omnibus (91) and are accessible through GEO Series accession number{"type":"entrez-geo","attrs":{"text":"GSE161713","term_id":"161713"}}GSE161713.

Login or Signup to leave a comment
Find your community. Ask questions. Science is better when we troubleshoot together.
Find your community. Ask questions. Science is better when we troubleshoot together.

Have a question?

Contact or check out our support page.