Subtilase cytotoxin from Shiga-toxigenic Escherichia coli impairs the inflammasome and exacerbates enteropathogenic bacterial infection

Abstract

Subtilase cytotoxin (SubAB) is an AB₅ toxin mainly produced by the locus of enterocyte effacement-negative Shiga-toxigenic Escherichia coli (STEC) strain such as O113:H21, yet the contribution of SubAB to STEC infectious disease is unclear. We found that SubAB reduced activation of the STEC O113:H21 infection-induced non-canonical NLRP3 inflammasome and interleukin (IL)-1β and IL-18 production in murine macrophages. Downstream of lipopolysaccharide signaling, SubAB suppressed caspase-11 expression by inhibiting interferon-β/STAT1 signaling, followed by disrupting formation of the NLRP3/caspase-1 assembly. These inhibitions were regulated by PERK/IRE1α-dependent endoplasmic reticulum (ER) stress signaling initiated by cleavage of the host ER chaperone BiP by SubAB. Our murine model of SubAB-producing Citrobacter rodentium demonstrated that SubAB promoted C. rodentium proliferation and worsened symptoms such as intestinal hyperplasia and diarrhea. These findings highlight the inhibitory effect of SubAB on the NLRP3 inflammasome via ER stress, which may be associated with STEC survival and infectious disease pathogenicity in hosts.

Keywords: Bacteriology; Biochemistry; Microbiology; Protein.

Graphical abstract

Figure 1

SubAB inhibits production of IL-1β and IL-18 and activation of inflammasome-related caspases in J774.1 cells (A) J774.1 cells were infected with STEC O113 WT or O113 ΔsubAB (MOI [multiplicity of infection] = 20) in the presence or absence of SubABwt or SubABmt (0.5 μg/mL) for 16 h, and culture supernatants were analyzed for IL-1β by using ELISA. (B and C) Culture supernatants of J774.1 cells infected for 16 h under the indicated condition were concentrated by methanol/chloroform precipitation and were analyzed by Western blotting (WB) using antibody against caspase-1 (B) or caspase-11 (C). CBB, Coomassie brilliant blue. (D) Total cell lysate samples of J774.1 cells infected for 16 h under the indicated conditions were analyzed by using WB with antibodies against caspase-11, GAPDH, or BiP. The bar graph shows a densitometric analysis of the WB of caspase-11 p43. Data are means ± SD (n = 3). ∗p< 0.05; ∗∗p< 0.01; NS, not significant; ND, not detected. See also Figure S1.

Figure 2

SubAB inhibits pro-IL-1β expression via PERK- and IRE1α-dependent pathways in J774.1 cells (A) Cells were infected with STEC O113 WT or O113 ΔsubAB (MOI = 20) for 16 h in the presence or absence of SubABwt or SubABmt. Total cell lysate samples were analyzed by WB with anti-IL-1β and anti-actin antibodies. (B) Cells were treated with LPS (100 ng/mL) for 4 h with or without SubABwt or SubABmt. Total RNA extracted from the cells was subjected to qRT-PCR with primers for IL-1β and GAPDH. (C) Different siRNAs—negative control (NC), IRE1α, PERK, or ATF6—were transfected into J774.1 cells. Cell lysates were subjected to WB with anti-IRE1α,-PERK,-ATF6, or-actin antibodies. (D) siRNA-transfected cells were treated with LPS for 4 h with or without SubABwt. (E) Cells were pre-treated with 0.1 or 1 μM GSK2656157 (GSK) for 1 h and were then treated with LPS for 4 h with or without SubABwt or SubABmt. (F) Cells were pre-treated with 50 or 100 μM STF-083010 (STF) for 1 h and were then treated with LPS for 4 h with or without SubABwt or SubABmt. (G) Cells were pre-treated with 1 μM GSK or 100 μM STF and were then treated with LPS for 4 h with or without SubABwt or SubABmt. Total cell lysate samples were analyzed by using WB with anti-IL-1β or-actin antibodies. Band intensity in each case was analyzed by densitometry, and results appear under the WB images (D–G). (H) Negative control (NC) or IRE1α knockdown cells were treated with 1 μM GSK for 1 h, followed by LPS transfection (LPS TF) with or without SubABwt or SubABmt as indicated in method details and Figure S3A. Culture supernatants were subjected to ELISA for IL-1β. Data are means ± SD (n = 3). ∗p< 0.05; ∗∗p< 0.01; NS, not significant. See also Figures S2 andS3.

Figure 3

SubAB inhibits formation of the NLRP3 inflammasome complex by suppressing caspase-11 expression in J774.1 cells (A) Gene knockdown of caspase-11 (Casp11) by siRNA. Cells were transfected for 72 h with NC or caspase-11 siRNA. After LPS priming, cells were transfected with LPS as indicated in method details and Figure S3A. (B) IL-1β production by LPS transfection (LPS TF) in culture supernatants of caspase-11 knockdown cells. Culture supernatants were subjected to ELISA to detect IL-1β. (C–E) Suppression by SubAB of LPS TF-induced IL-1β production (C), caspase-11 activation (D), and pro-caspase-11 expression (E). (F) Cells were infected with STEC O113 WT or O113 ΔsubAB (MOI = 20) for 16 h in the presence or absence of SubABwt or SubABmt. Total cell lysate samples were analyzed by means of immunoprecipitation (IP) with anti-caspase-1 antibody. NLRP3 binding was analyzed via WB with anti-NLRP3 antibody. (G) Cells were stimulated with LPS TF in the presence or absence of SubABwt or SubABmt. Total cell lysate samples were analyzed by IP with anti-caspase-1 antibody. Hc in (F) and (G) indicates the heavy chain of anti-caspase-1 IgG. Data are means ± SD (n = 3). ∗p< 0.05; ∗∗p< 0.01; ND, not detected. See also Figure S3.

Figure 4

SubAB inhibits STAT1 phosphorylation through IRE1α-mediated attenuation of IFN-β production in J774.1 cells (A) J774.1 cells were infected with STEC O113 WT and STEC O113 ΔsubAB for 6 h in the presence or absence of SubABwt or SubABmt. Phosphorylation of STAT1 (P-STAT1) α and β was analyzed by using WB with anti-P-STAT1 (pY701) and total STAT1 antibodies. (B) Cells were treated with LPS as indicated and analyzed by using WB with anti-P-STAT1 and total STAT1 antibodies. Relative P-STAT1 α and β amounts were quantified by densitometry (right panel). (C) Cells were treated with LPS for 6 h as indicated. IFN-β production in culture supernatants was analyzed by means of ELISA. (D) Cells were treated with LPS for 4 h with or without SubABwt or SubABmt. Total RNA extracted from the cells was subjected to RT-PCR with IFN-β and GAPDH primers. (E) Cells were pre-treated with 100 μM STF or 1 μM GSK for 1 h and were then treated with LPS for 6 h in the presence or absence of SubABwt or SubABmt. IFN-β production in culture supernatants was analyzed by using ELISA. (F–H) Cells were transfected with negative control (NC), IRE1α (F), PERK (G), or ATF6 (H) siRNA. After 72 h, cells were stimulated with LPS as indicated, and total cell lysate samples were analyzed by using WB with anti-P-STAT1 (pY701) and total STAT1 antibodies. Relative P-STAT1 α and β amounts were quantified by densitometry (lower panels). Data are means ± SD (n = 3). ∗p< 0.05; ∗∗p< 0.01; NS, not significant. See also Figure S3.

Figure 5

SubAB inhibits caspase-11 expression through PERK-mediated attenuation of IFN-β-induced STAT1 phosphorylation in J774.1 cells (A) Cells were pre-treated with SubABwt or SubABmt for 3 h or were not treated and were then stimulated with IFN-β (100 pg/mL) for the indicated times or were not so stimulated. (B) Cells were pre-treated with or without SubABwt. After 2 h, cells were treated with GSK for 1 h at one of the indicated concentrations, followed by treatment with IFN-β for 1 h. (C) Cells were treated with 1 μM GSK or 100 μM STF for 1 h, followed by stimulation with LPS (100 ng/mL) for 4 h with or without SubABwt or SubABmt. Total cell lysate samples were analyzed by using WB with anti-P-STAT1 (pY701) and total STAT1 antibodies (A–C). Relative P-STAT1 α and β levels were quantified by using densitometry (A–C, lower panels). (D) WB with anti-caspase-11 antibody for the cells in (C). The lower panel shows densitometric analysis for p43 band intensity. Actin served as the loading control. (E) Cells were treated with 1 μM GSK or 100 μM STF for 1 h, followed by infection with STEC O113 WT or STEC O113 ΔsubAB as indicated. After 16 h, culture supernatants were subjected to ELISA for IL-1β. Data are means ± SD (n = 3). ∗p< 0.05; ∗∗p< 0.01; NS, not significant; ND, not detected. See also Figure S4.

Figure 6

SubAB inhibits caspase-1 activation and production of IL-1β and IL-18 in vivo and promotes intestinal survival of C. rodentium (A) Preparation of SubAB-expressing C. rodentium (Cr-SubAB). WB image for expression of SubA (upper panel) and SubB (lower panel) in C. rodentium strains. GAPDH served as the loading control. (B) C57BL/6 mice received oral administration of Cr-SubABwt or Cr-SubABmt as described in the STAR Methods. Body weights of groups of five mice that received PBS (Mock), Cr-SubABwt, or Cr-SubABmt, expressed as relative changes from 0 days post infection (dpi). Data are averages (±SD) of 5 mice per group. (C) Tissue homogenates of intestines from infected mice at 11 dpi were subjected to ELISA for IL-18. Data are means ± SD (n = 3 or 4). (D) Tissue homogenates of intestines from infected mice were analyzed at 11 dpi by using WB with anti-IL-1β antibody. LPS-treated J774.1 cell lysate was used as a positive control for pro-IL-1β. (E) Tissue homogenates of intestines from infected mice were analyzed at 11 dpi using WB with anti-caspase-1 and anti-BiP antibodies. Actin (D and E) served as the loading control. (F) Mouse feces were collected at 11 dpi from intestines from three mice (three feces per mouse). The number of viable bacteria was determined by CFU analysis of plates; results are given as log₁₀ CFU per gram of feces. Data are means ± SD (n = 9 per group). (G) Representative gross colon images from mice that received PBS (Mock), Cr-SubABwt, or Cr-SubABmt at 15–16 dpi. The right panel shows quantitative results of colon length for the image on the left. Data are means ± SD (n = 3 per group). (H) Immunostained micrographs of neutrophils (Gr-1, upper) and macrophages (Iba1, lower) in liver sections from mice in each group at 15–16 dpi. (I) The quantitative results of the infiltration in images from (H). Data are means ± SD (n = 3). ∗p< 0.05; ∗∗p< 0.01. See also Figures S5 and S6.