Pathogen-selective killing by guanylate-binding proteins as a molecular mechanism leading to inflammasome signaling

Abstract

Inflammasomes are cytosolic signaling complexes capable of sensing microbial ligands to trigger inflammation and cell death responses. Here, we show that guanylate-binding proteins (GBPs) mediate pathogen-selective inflammasome activation. We show that mouse GBP1 and GBP3 are specifically required for inflammasome activation during infection with the cytosolic bacterium Francisella novicida. We show that the selectivity of mouse GBP1 and GBP3 derives from a region within the N-terminal domain containing charged and hydrophobic amino acids, which binds to and facilitates direct killing of F. novicida and Neisseria meningitidis, but not other bacteria or mammalian cells. This pathogen-selective recognition by this region of mouse GBP1 and GBP3 leads to pathogen membrane rupture and release of intracellular content for inflammasome sensing. Our results imply that GBPs discriminate between pathogens, confer activation of innate immunity, and provide a host-inspired roadmap for the design of synthetic antimicrobial peptides that may be of use against emerging and re-emerging pathogens.

Fig. 1. GBP1, GBP2, GBP3, and GBP5 are required for F. novicida-induced inflammasome activation.

a Immunoblot analysis of caspase-1 (Casp-1) and gasdermin D (GSDMD) in WT, Gbp1^–/–, Gbp2^–/–, Gbp3^–/–, Gbp5^–/–, Gbp7^–/–, Gbp^chr3-KO, or Aim2^–/– BMDMs left untreated (Med.) or assessed after infection with F. novicida (MOI 100) for 10 h. b The release of IL-1β and IL-18 from BMDMs after treatment as in a. c The release of LDH from BMDMs after treatment as in a. d Immunoblot analysis of Casp-1 and GSDMD in WT, Gbp1^–/–, Gbp3^–/– and Aim2^–/– BMDMs left untreated (Med.) or assessed after infection with F. novicida (MOI 100) for 10 h, infection with L. monocytogenes (MOI 100) for 20 h, infection with MCMV (MOI 10) for 10 h or transfection of poly(dA:dT) (5 µg/ml) and pcDNA (5 µg/mL) for 4 h. e The release of IL-1β, IL-18 and LDH from BMDMs after treatment as in d. f Light microscopy analysis of WT, Gbp1^–/–, Gbp3^–/– or Aim2^–/– BMDMs left untreated (Med.) or following infection with F. novicida as in d. White arrows indicate pyroptotic cells. Each symbol represents an independent experiment (b, c, and e). Scale bar, 20 µm (f). ns no statistical significance; **P < 0.01; ***P < 0.001; ****P < 0.0001 (one-way ANOVA with Dunnett’s multiple-comparisons test (b, c, and e)). Data are representative of three independent experiments (a–f; mean and s.e.m. in b, c, and e). Source data are provided as a Source data file.

Fig. 2. GBP1 and GBP3 target intracellular F. novicida and restrict its growth.

a Confocal microscopy analysis of F. novicida (green) and GBP1 (red) or GBP3 (red) in WT, Gbp1^–/–, Gbp2^–/–, Gbp3^–/– and Gbp5^–/– BMDMs left untreated (Med.) or assessed 20 h after infection with F. novicida (MOI 20). White arrows indicate bacteria colocalized with GBP. b Quantitation of GBP1-, GBP2-, GBP3-, GBP5-positive F. novicida in WT, Gbp1^–/–, Gbp2^–/–, Gbp3^–/– and Gbp5^–/– BMDMs as treated in a. c The percentages of WT and Gbp1^–/– BMDMs (left) or WT and Gbp3^–/– BMDMs (right) harboring different number of bacteria. d Confocal microscopy analysis of F. novicida (green) and DNA (blue) in WT, Gbp1^–/– and Gbp3^–/– BMDMs 0, 4, 8, and 12 h after infection with F. novicida (MOI 25). e Recovery of F. novicida (as colony-forming units (CFU)) from WT, Gbp1^–/–, Gbp3^–/– and Ifnar1^–/– BMDMs at 4, 8, or 12 h after infection with F. novicida (MOI 50). Scale bars, 7 µm (a) and 20 µm (d). ns no statistical significance; *P < 0.05, **P < 0.01; ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Dunnett’s multiple-comparisons test (b, e)). Data are from one experiment representative of three independent experiments (a, d) or pooled from three independent experiments (b, c, and e, mean and s.e.m. in b, c, and e). Source data are provided as a Source data file.

Fig. 3. GBP1 peptide binds to and kills F. novicida.

a Analysis of AMP probability (red) and charge (black) for the mouse GBP1 protein sequence and illustration of the location of putative antimicrobial stretches within the predicted mouse GBP1 structure. b Viability of F. novicida [F. nov., as percentage of CFU in relation to solvent control (Sol.Ctrl.)] assessed 6 h after incubation with GBP1^28–67, GBP1^209–238, GBP1^424–452, GBP1^558–577 or WLBU2 at 0.1, 1 or 10 μg/mL. c Viability of F. nov. following incubation with GBP1^28–67 for 6 h over a concentration range (0.01–20 µg/mL) to determine the half maximal inhibitory concentration (IC₅₀). d Flow cytometric analysis (left) and quantitation of flow cytometry plots (right) of SYTOX stained F. novicida treated with Sol.Ctrl. or 100 μg/mL of GBP1^28–67, GBP1^209–238, GBP1^424–452, GBP1^558–577 or WLBU2 for 12 h. e Confocal microscopy analysis of Hoechst-stained total bacteria (blue), FITC-GBP1^28–67 (green) and SYTOX (red) in F. novicida or E. coli treated with 10 µg/mL FITC-GBP1^28–67 or FITC-control peptide for 6 h. White arrows indicate dead bacteria covered with FITC-GBP1^28–67. f Flow cytometric quantitation of SYTOX stained F. novicida treated with 10 µg/mL of FITC-GBP1^28–67 for 6 h. g Quantitation of FITC-GBP1^28–67 bound to F. novicida and E. coli after 1 h incubation with 10 µg/mL of either FITC- GBP1^28–67 or a FITC-control peptide (in relative fluorescence units, RFU). h Quantitation of the effect of 1 M NaCl, 0.01% saponin or 0.08% sarcosyl on FITC-GBP1^28–67 binding to F. novicida. Scale bar, 5 µm (e). ns no statistical significance; *P < 0.05, **P < 0.01; ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Dunnett’s multiple-comparisons test (b, d, h), two-tailed t-test (f, g). Data are representative of three independent experiments (b–h; mean and s.e.m. in b–d, f–h). Source data are provided as a Source data file.

Fig. 4. GBP1 peptide induces membrane disruption, expulsion of cytoplasmic and membranous content from F. novicida.

a Scanning electron microscopy and (b) transmission electron microscopy analysis of the morphology of F. novicida 12 h after treatment with solvent control, 100 μg/mL of GBP1^28–67 or WLBU2. c Negative-stain transmission electron microscopy analysis of F. novicida 12 h after treatment with solvent control, 100 μg/mL of GBP1^28–67 or WLBU2. Scale bars, 200 nm (a), 500 nm (b) and 1 μm (c). Orange arrow heads indicate bacteria with disrupted cell membrane. Data are representative of three independent experiments (a–c).

Fig. 5. GBP1, GBP2, GBP3 and GBP5 provide host protection against F. novicida infection in vivo.

a Survival of 7-week-old WT mice (n = 25), Gbp1^–/– mice (n = 25) and Aim2^–/– mice (n = 32) infected subcutaneously with 1.2 × 10⁶ colony-forming units (CFUs) of F. novicida. b Body weight of 7-week-old WT mice (n = 25), Gbp1^–/– mice (n = 18) and Aim2^–/– mice (n = 25) 0–7 d after subcutaneous infection with 1.2 × 10⁶ CFUs of F. novicida, presented relative to initial body weight at day 0, set as 100%. c Survival of 7-week-old WT mice (n = 15), Gbp3^–/– mice (n = 15) and Aim2^–/– mice (n = 10) infected subcutaneously with 1.2 × 10⁶ colony-forming units (CFUs) of F. novicida. d Body weight of 7-week-old WT mice (n = 15), Gbp3^–/– mice (n = 15) and Aim2^–/– mice (n = 7) 0–7 d after subcutaneous infection with 1.2 × 10⁶ CFUs of F. novicida, presented relative to initial body weight at day 0, set as 100%. e Bacterial burden in the liver (left) and spleen (middle) and concentration of IL-18 in the serum (right) of 7-week-old WT mice (n = 17), Gbp1^–/– mice (n = 15) and Aim2^–/– mice (n = 16) on day 3 after infection with 6 × 10⁵ CFUs of F. novicida. f Bacterial burden in the liver (left) and spleen (middle) and concentration of IL-18 in the serum (right) of 7-week-old WT mice (n = 14), Gbp3^–/– mice (n = 9) and Aim2^–/– mice (n = 9) on day 3 after infection with 6 × 10⁵ CFUs of F. novicida. g Bacterial burden in the liver (left) and spleen (middle) and concentration of IL-18 in the serum (right) of 7-week-old WT mice (n = 10) and Gbp2^–/– mice (n = 9) on day 3 after infection with 6 × 10⁵ CFUs of F. novicida. h Bacterial burden in the liver (left) and spleen (middle) and concentration of IL-18 in the serum (right) of 7-week-old WT mice (n = 10) and Gbp5^–/– mice (n = 12) on day 3 after infection with 6 × 10⁵ CFUs of F. novicida. Each symbol represents an individual mouse (e–h). ns no statistical significance; *P < 0.05, **P < 0.01; ***P < 0.001, ****P < 0.0001 (log-rank test (a and c) or one-way ANOVA with Dunnett’s multiple-comparisons test (e, f, mean and s.e.m. in e, f)) or two-tailed t-test (b, d, g, h, mean and s.e.m. in b, d, g, h). Data are pooled from two or three independent experiments (a–h). Source data are provided as a Source data file.