Type I CRISPR-Cas targets endogenous genes and regulates virulence to evade mammalian host immunity

Abstract

Clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated (Cas) systems in bacteria and archaea provide adaptive immunity against invading foreign nucleic acids. Previous studies suggest that certain bacteria employ their Type II CRISPR-Cas systems to target their own genes, thus evading host immunity. However, whether other CRISPR-Cas systems have similar functions during bacterial invasion of host cells remains unknown. Here we identify a novel role for Type I CRISPR-Cas systems in evading host defenses in Pseudomonas aeruginosa strain UCBPP-PA14. The Type I CRISPR-Cas system of PA14 targets the mRNA of the bacterial quorum-sensing regulator LasR to dampen the recognition by toll-like receptor 4, thus diminishing the pro-inflammatory responses of the host in cell and mouse models. Mechanistically, this nuclease-mediated RNA degradation requires a "5'-GGN-3'" recognition motif in the target mRNA, and HD and DExD/H domains in Cas3 of the Type I CRISPR-Cas system. As LasR and Type I CRISPR-Cas systems are ubiquitously present in bacteria, our findings elucidate an important common mechanism underlying bacterial virulence.

Figure 1

PA14 CRISPR-Cas is necessary for lasR mRNA repression. (A) Schematic of PA14 CRISPR-Cas locus, comprising Cas protein-coding genes csy1, cas1, csy2, csy3, cas3 and csy4, as well as two CRISPR regions (repeats indicated by vertical shadows). (B) lasR transcripts in PA14 WT, ΔTCR and complemented strains were quantified by qPCR. (C) CFU of PA14 WT and ΔTCR mutant strain transformed with lasR expressing plasmid p-lasR and empty control plasmid (EV, empty vector), and relative plasmid transformation efficiency was quantified by counting. (D) lasR transcripts in plasmid-transformed PA14 WT and ΔTCR mutant strains were quantified by qPCR. (E) Time course of lasR stability in PA14 WT and ΔTCR mutants after treatment with rifampicin. Data are representative of three experiments expressed as means ± SEM (n = 3, one-way ANOVA with Tukey's post hoc; ^*P ≤ 0.05; ^**P ≤ 0.005; ^***P ≤ 0.001).

Figure 2

Analysis of PA14 CRISPR-mediated dynamic RNA degradation. (A) lasR transcripts in PA14 WT and indicated CRISPR mutant strains were quantified by qPCR. (B) Homology comparison between lasR mRNA and crRNA1 spacer12 (yellow indicating a PAM-like sequence), and mutation strategy of lasR mRNA for 100% hybridization to crRNA1-12. (C) Effect of Cas3 concentration on lasR mRNA cleavage. (D) Effect of crRNA1-12 concentration on lasR mRNA cleavage. (E) Effect of “5′-GGA-3′” recognition site on lasR mRNA cleavage. (F) Effects of spacer-matched length on lasR mRNA cleavage. (G) Effect of gRNA-target RNA mismatch on lasR mRNA cleavage. (H) Mutation strategy of Cas3 HD, DExD/H and HelC domains. (I) lasR transcripts in indicated PA14 Cas3 mutant strains were quantified by qPCR. (J) Effect of Cas3 catalytic residues on lasR mRNA cleavage. (K) Binding of Csy1-4 and Cas3 to RNA substrate examined using an EMSA assay. BSA was used as a negative control. Data are representative of three experiments expressed as means ± SEM (n = 3, one-way ANOVA with Tukey's post hoc; ^*P ≤ 0.05; ^**P ≤ 0.005).

Figure 3

PA14 CRISPR-Cas deficiency increases PA14 virulence and host innate immune response. (A) C57BL/6J mice were challenged intranasally with 5 × 10⁶ colony forming units (CFU) of PA14 WT or indicated CRISPR-Cas and lasR mutants (n = 6, two independent experiments), and survival monitored over time and represented by Kaplan-Meier survival curves. (B) Mice were infected with 5 × 10⁶ CFU of PA14 WT, indicated CRISPR or lasR mutants per mouse for 24 h. Lung injury and inflammation were assessed by histology. Paraffin-embedded sections were analyzed by H&E staining (arrows indicating the region of insets showing tissue injury and inflammatory influx). Magnification: ×200 (insets: ×400). (C) Inflammatory cell infiltration in the lungs was quantified in B (10 random areas). (D) Bacterial burdens in the lungs after homogenization in PBS. (E, F) Superoxide production in AMs was detected by NBT assay (E) and H₂DCF assay (F). (G) Inflammatory cytokines in BALF were assessed by ELISA. (H) Expression of inflammatory cytokines of the lungs was detected by immunoblotting. Data of D-G are representative of three experiments expressed as means ± SEM (n = 3, one-way ANOVA with Tukey's post hoc; ^*P ≤ 0.05; ^**P ≤ 0.005).

Figure 4

PA14 CRISPR-Cas deficiency increases bacterial phagocytosis by host alveolar macrophages. (A, B) MH-S and MLE-12 cells were infected with PA14 WT, ΔTCR and complemented strains at an MOI of 20:1 for 2 h. Time course of survival was determined by an MTT assay. (C) Inflammatory cytokines in medium were assessed by ELISA. (D) Expression of inflammatory cytokines in MH-S cells was detected by immunoblotting. (E-G) MH-S cells were transfected with Rab5-RFP (red) for 24 h, and then were challenged with PA14 WT and ΔTCR mutant, both harboring a plasmid pMQ70-EGFP, at an MOI of 20:1 for 30 min. Colocalization (arrows) of bacteria and Rab5 observed by confocal laser scanning microscopy (CSLM). (E) Uptake of the bacteria (F) and Rab5 dots (G) were quantified (random 100 cells). Data are representative of three experiments expressed as means ± SEM (n = 3, one-way ANOVA with Tukey's post hoc; ^*P ≤ 0.05; ^**P ≤ 0.005).

Figure 5

PA14 CRISPR-Cas regulates innate immunity through TLR4/NF-κB signaling. (A) MH-S cells were infected with PA14 WT and ΔTCR mutant at an MOI of 20:1 for 2 h. qPCR primer assay of immune response-related factors in ΔTCR mutant-infected MH-S cells versus PA14 WT-infected controls. Red dots indicate increased genes and green dots indicate decreased genes (over 10-fold change and P ≤ 0.05, n = 2 biological replicates). (B) Expression of TLR2, TLR4 and indicated signaling factors in PA14 WT-, ΔTCR mutant- and complemented strain-infected MH-S cells (30 min). (C) MH-S cells were infected with PA14 WT, ΔTCR mutant and complemented strain at an MOI of 20:1 for 0, 15, 30 and 60 min. CSLM results show the translocation of p-NF-κB (p-p65) in MH-S cells using immune staining. DAPI was used for staining the nucleus (arrows showing the nuclear translocation). (D) CSLM results show the production of TLR4 in MH-S cells. (E) TLR4 expression was determined in C57BL/6J WT and TLR4 KO mice by immunoblotting. (F) Secretion of TNF-α, IL-1β and IL-6 in BALF from C57BL/6J WT and TLR4 KO mice after WT PA14, ΔTCR mutant and complemented strain infection. (G) Phagocytosis of indicated MH-S cells was measured by CFU count assay. MH-S cells were infected with an MOI of 20:1, 25:1, 30:1, 35:1 and 40:1 PA14 WT. The MOI of 20:1 ΔTCR mutant infection was a control. (H) Expression of TLR4 and production of TNF-α, IL-1β and IL-6 in PA14 WT-infected and ΔTCR mutant-infected MH-S cells were detected by immunoblotting. Data are representative of three experiments expressed as means SEM (n = 3, one-way ANOVA with Tukey's post hoc; ^*P ≤ 0.05; ^**P ≤ 0.005).

Figure 6

PA14 CRISPR-Cas modulates TLR4 signaling by temporal repression of lasR. (A) lasR transcripts in invading bacteria were quantified by qPCR. MH-S cells were infected with PA14 WT, ΔTCR mutant and the complemented strain at an MOI of 20:1. Bacteria in the medium or in the phagosome of MH-S cells were collected at 0, 1, 2, 3 and 4 h after infection. (B) Time course of transcripts of PA14 CRISPR-Cas components in MH-S cells that have phagocytized PA14 WT strain. (C) TLR4 expression in PA14 WT-, ΔTCR mutant-, ΔlasR mutant- and the complemented strain-infected MH-S phagosomes, determined by immunoblotting. (D) Transcripts of lasR in PA14 WT and mutants were quantified by qPCR. RNA was isolated when PA14 WT, ΔTCR and the complemented strains grew at an OD_{600 nm} of 0.5 (log phase) or 2.0 (stationary phase) in M9 medium. (E, F) MH-S cells were infected with PA14 WT, ΔTCR mutant and the complemented strain in log phase at an MOI of 20:1 for 2 h. Survival rates were determined by an MTT assay (E), and phagocytosis of indicated MH-S cells were measured by CFU count assay (F). Data are representative of three experiments expressed as means ± SEM (n = 3, one-way ANOVA with Tukey's post hoc; ^*P ≤ 0.05; ^**P ≤ 0.005). (G) Model of PA14 evasion from macrophages mediated by function of the CRISPR-Cas system to target lasR mRNA: (I) PA14 lacking the CRISPR-Cas system (P14 ΔTCR): when captured in phagosome, the LasR-dependent virulence factors of PA14 activate host TLR4 expression, leading to elevated inflammatory responses. (II) WT P14: once entering the phagosome, the expression of CRISPR-Cas components of PA14 is upregulated. The crRNA1-12 structure is associated with Cascade (Csy1-4 complex) and interacts with lasR mRNA through a sequence matching part of crRNA1-12 (yellow; PAM like sequence, red). lasR mRNA is then cleaved by Cas3, leading to its degradation. Reduction in lasR levels decreases TLR4 expression and TLR4-mediated recognition in macrophages, thus dampening host inflammatory responses. It is also likely that such decrease in TLR4 expression levels may in turn lead to suppressed phagocytosis, which could further compromise the immune response. The exact mechanism by which LasR of P14 regulates host phagocytosis remains undefined.