Genetic characterization of antiplasmid immunity through a type III-A CRISPR-Cas system

Abstract

Many prokaryotes possess an adaptive immune system encoded by clustered regularly interspaced short palindromic repeats (CRISPRs). CRISPR loci produce small guide RNAs (crRNAs) that, in conjunction with flanking CRISPR-associated (cas) genes, combat viruses and block plasmid transfer by an antisense targeting mechanism. CRISPR-Cas systems have been classified into three types (I to III) that employ distinct mechanisms of crRNA biogenesis and targeting. The type III-A system in Staphylococcus epidermidis RP62a blocks the transfer of staphylococcal conjugative plasmids and harbors nine cas-csm genes. Previous biochemical analysis indicated that Cas10, Csm2, Csm3, Csm4, and Csm5 form a crRNA-containing ribonucleoprotein complex; however, the roles of these genes toward antiplasmid targeting remain unknown. Here, we determined the cas-csm genes that are required for antiplasmid immunity and used genetic and biochemical analyses to investigate the functions of predicted motifs and domains within these genes. We found that many mutations affected immunity by impacting the formation of the Cas10-Csm complex or crRNA biogenesis. Surprisingly, mutations in the predicted nuclease domains of the members of the Cas10-Csm complex had no detectable effect on antiplasmid immunity or crRNA biogenesis. In contrast, the deletion of csm6 and mutations in the cas10 Palm polymerase domain prevented CRISPR immunity without affecting either complex formation or crRNA production, suggesting their involvement in target destruction. By delineating the genetic requirements of this system, our findings further contribute to the mechanistic understanding of type III CRISPR-Cas systems.

FIG 1

Assembly of the Cas10-Csm ribonucleoprotein complex is required for antiplasmid CRISPR immunity. (A) Organization of the type III-A CRISPR system in S. epidermidis RP62a. This system contains nine CRISPR-associated (cas and csm) genes, four direct repeats (DR), and three spacers, the first of which (spc1) targets the nickase gene in pG0400, a staphylococcal conjugative plasmid. (B) crRNA biogenesis in S. epidermidis. Transcription of the repeat-spacer array generates a precursor crRNA that is subjected to two cleavage events: primary processing, which entails endoribonucleolytic cleavages within repeats to yield ∼71-nucleotide crRNA intermediates, and maturation, during which further trimming of the 3′ end generates mature crRNA species that are 31, 37, 43, 49, and 55 nucleotides in length. Base pairing between spc1 crRNA and the target region of pG0400 results in the prevention of plasmid transfer, presumably due to the destruction of the target by Cas nucleases. (C) Conjugation of pG0400 into S. epidermidis LM1680 harboring WT and mutant pcrispr plasmids. Conjugation was carried out by filter mating; the mean values for two independent experiments ± standard deviations (SD) obtained for recipients and transconjugants are shown. (D) Purification of Cas10-Csm from cells carrying wild-type and mutant pcrispr plasmids. His₆ tags were placed on the N terminus of Csm2 or the C terminus of Csm4 in the case of the Δcsm2 mutant for complex purification from S. epidermidis LM1680 extracts using Ni²⁺ affinity chromatography. Purified proteins were resolved by SDS-PAGE. Protein identities were assigned as described in reference . (E) Nucleic acids were extracted from the indicated complexes, radiolabeled at the 5′ end, and resolved using denaturing PAGE.

FIG 2

Cas6 is sufficient for primary processing within repeat sequences. (A) SDS-PAGE of purified His₁₀-SUMO-Cas6. Cas6 was cloned into a pET28b vector and expressed and purified from E. coli BL21. (B) Diagram of the substrate used for the nuclease assay. It contains 86 nt of leader sequence followed by a full repeat (blue square; 36 nt), spc1 (yellow; 35 nt), and a partial second repeat that lacks the sequences required for primary processing (28 nt). There is a single Cas6 cleavage site in the substrate within the first repeat (arrowhead) that produces two RNAs of 114 and 71 nt. The substrate was generated and internally labeled by in vitro transcription. (C) Cleavage of the radiolabeled substrate by Cas6. Different concentrations of Cas6 (0, 5, 50, or 500 nM) were incubated with the substrate for 20 min at 37°C, and RNA was extracted and visualized by denaturing PAGE. (D) spc1-directed CRISPR interference against pG0400 in the presence of the indicated Cas6 mutations. S. epidermidis LM1680 isolates expressing WT and cas6 mutant pcrispr plasmids were used as recipients. Conjugation was carried out by filter mating; the mean values for two independent experiments ± SD obtained for recipients and transconjugants are shown.

FIG 3

The Csm3 and Csm5 G-rich loops are required for protein stability. (A) Diagram of Cas proteins for which catalytic activity has been predicted, including the RAMPs and the Cas10 “CRISPR polymerase.” Putative catalytic histidines and aspartates are indicated by red and black asterisks, respectively. Glycine-rich regions (G-rich loops) in the RAMPs are indicated by red rectangles. Also highlighted are the HD nuclease domain (green) and Palm polymerase domain (orange) of Cas10 and the RAMP-like RRM domain (yellow) of the RAMPs. (B) Conjugation of pG0400 S. epidermidis LM1680 cells harboring wild-type and mutant pcrispr plasmids carrying mutations in the G-rich loops of Casm3, Csm4, and Csm5. Conjugation was carried out by filter mating; the mean values for two independent experiments ± SD obtained for recipients and transconjugants are shown. (C) Purification of Cas10-Csm wild type and G-rich loop mutant complexes. His₆ tags were placed on the N terminus of Csm2, and constructs were expressed in S. epidermidis LM1680. Whole-cell lysates were subjected to Ni²⁺ affinity chromatography, and protein extracts were resolved by SDS-PAGE. (D) CrRNAs were extracted from the indicated His-tagged Cas10-Csm complexes and visualized using denaturing PAGE. The length of each crRNA species in nucleotides is noted on the left of the gel..

FIG 4

The Cas10 Palm domain is required for antiplasmid CRISPR immunity. (A) Conjugative transfer of pG0400 S. epidermidis LM1680 recipient cells harboring wild-type and mutant pcrispr plasmids. Conjugation was carried out by filter mating; the mean values for two independent experiments ± SD obtained for recipients and transconjugants are shown. (B) Purification of wild-type and Palm domain mutant Cas10-Csm complexes. His₆ tags were placed on the N terminus of Csm3 (for the wild-type construct) or the N terminus of Csm2 (For the Cas10 Palm mutant construct). Constructs were expressed in S. epidermidis LM1680, whole-cell lysates were subjected to Ni²⁺ affinity chromatography, and the purified proteins were resolved by SDS-PAGE. (C) CrRNAs were extracted and visualized from each of the His-tagged Cas10-Csm complexes containing the indicated mutations. The length of each crRNA species in nucleotides is noted on the left of the gel.