Coordinated host responses during pyroptosis: caspase-1-dependent lysosome exocytosis and inflammatory cytokine maturation

Abstract

Activation of caspase-1 leads to pyroptosis, a program of cell death characterized by cell lysis and inflammatory cytokine release. Caspase-1 activation triggered by multiple nucleotide-binding oligomerization domain-like receptors (NLRs; NLRC4, NLRP1b, or NLRP3) leads to loss of lysosomes via their fusion with the cell surface, or lysosome exocytosis. Active caspase-1 increased cellular membrane permeability and intracellular calcium levels, which facilitated lysosome exocytosis and release of host antimicrobial factors and microbial products. Lysosome exocytosis has been proposed to mediate secretion of IL-1β and IL-18; however, blocking lysosome exocytosis did not alter cytokine processing or release. These studies indicate two conserved secretion pathways are initiated by caspase-1, lysosome exocytosis, and a parallel pathway resulting in cytokine release, and both enhance the antimicrobial nature of pyroptosis.

Figure 1. Caspase-1 activation and lysosome exocytosis during pyroptosis

(A) Bone marrow derived macrophages were incubated with TMR-dextran to label lysosomes (red) and infected for 20 minutes with wild type (w.t.) or T3SS-null (SipB-) Salmonella (ST), or left uninfected (UI), in the presence of FAM-YVAD-FMK to label active caspase-1 (green). DNA was stained using TO-PRO3 (blue). (B) Macrophages were infected with GFP-expressing Salmonella (green) for 20 minutes, or left uninfected, and surface exposure of LAMP-1 (red) was determined by immunofluorescence staining of intact cells. DNA was stained using TO-PRO3 (blue). Scale bars are 10μm. (C) The percentage of cells with surface LAMP-1; data are means and standard deviations calculated from multiple fields from 3 experiments. *P<0.0001.

Figure 2. Active caspase-1 mediates pore formation and calcium influx

(A) Wild-type and Casp1^−/− macrophages were infected with GFP-expressing Salmonella (green) in the presence of glycine for the indicated period of time and stained with ethidium bromide (red) to identify macrophages with membrane pores and SYTO62 (blue) to label all macrophages. (B–D) Wild-type and Casp1^−/− macrophages were loaded with the calcium indicator fluo-4, infected with Salmonella, and images were taken every 15 seconds for 20 minutes. Representative images are shown (B), with arrowheads indicating macrophages with increased intracellular calcium. Scale bars are 20μm. (C) The change in mean fluorescence intensity of individual wild-type and Casp1^−/− macrophages during Salmonella infection. (D) The percentage of cells with increased intracellular calcium calculated from multiple fields containing greater than 100 cells total. Representative of 3 experiments.

Figure 3. Lysosome exocytosis requires caspase-1 activity and extracellular calcium

(A–C) Macrophages were infected with Salmonella (ST) for 20 minutes in the presence of 200μM Ac-YVAD-CMK (+YVAD) or the absence of extracellular calcium (-Ca2+). (A) Surface exposure of LAMP-1 (red) determined by immunofluorescence staining of intact cells. DNA was stained using TO-PRO3 (blue). Scale bars are 20 μm. (B) Quantification of cells with surface LAMP-1; data are means and standard deviations calculated from multiple fields from 3 experiments. (C) Secretion of lysosomal cathepsin D into the supernatant was analyzed by western blot. Representative of 3 experiments. *P<0.0001.

Figure 4. Activation of caspase-1 by the NLRs Nalp1b and Nalp3 also leads to the conserved process of lysosome exocytosis

(A) Wild-type and Casp1^−/− C57BL/6 macrophages were treated with LPS for 3 hours followed by addition of nigericin for 45 minutes to activate caspase-1 via NLRP3. (B) Caspase-1 was activated via NLRP1b by treating Balb/c macrophages with B. anthracis lethal toxin for 60 minutes. Macrophages were treated with 200μM Ac-YVAD-CMK (+YVAD) where indicated. (A–B) Surface exposure of LAMP1 (red) was determined by immunofluorescence staining of intact cells. DNA was stained using TO-PRO3 (blue). (C) LPS pretreated wild-type and Casp1^−/− macrophages were loaded with the calcium indicator fluo-4, treated with ATP, and images were taken every 15s for 20 minutes. (t=0 and 30s shown) (D) Wild-type and Casp1^−/− C57BL/6 macrophages were treated with LPS for 3 hours followed by addition of ATP for 45 minutes. Wild-type macrophages were treated in the absence of extracellular calcium (-Ca2+) where indicated. Surface exposure of LAMP-1 (red) was determined by immunofluorescence staining of intact cells. DNA was stained using TO-PRO3 (blue). All scale bars are 20μm. All images are representative of 3 experiments.

Figure 5. Secretion of the caspase-1 substrates IL-18 and IL-1β occurs independently of lysosome exocytosis

Western blot detection of cleaved IL-18 and IL-1β released into the supernatant by macrophages (A) 20 minutes after Salmonella infection in the presence of 200μM Ac-YVAD-CMK (+YVAD) or the absence of extracellular calcium (-Ca2+) or (B) 45 minutes of ATP treatment in the absence of caspase-1 (Casp1^−/−) or extracellular calcium (-Ca2+). Representative of 2 experiments.

Figure 6. During pyroptosis, lysosome exocytosis mediates release of microbial products and antimicrobial factors

Wild-type (A,B) and Casp1^−/− (C) macrophages were loaded with Alexa-488 dextran to label lysosomes (green) and Alexa-594 yeast particles (red) and infected with Salmonella. Images were taken every 15–20s for 20 minutes and representative images show yeast particles within dextran-containing lysosomes (green+red) and subsequent lysosome exocytosis and release of yeast particles (red). (A) Prior to taking the final image, the membrane-impermeant dye trypan blue was added to quench extracellular fluorescence. Lower panels show the Alexa-594 channel alone. Dextran-positive compartments containing (a) and lacking (b) zymosan particles were released over the 20 min period of pyroptosis induction (kinetics shown in Fig. S1 E). Arrows designate (a) particles with reduced fluorescence upon quenching, thus demonstrating that they have been released from the macrophage during lysosome exocytosis. Not all yeast particles were released during this 20 min period (c). Several Alexa-594 yeast particles, labeled (d), did not colocalize with Alexa-488 dextran at t=0 and their fluorescence was quenched by trypan blue, suggesting that they remained extracellular during induction of pyroptosis. B–C) Representative images showing the kinetics of zymosan release from lysosomes of wild-type macrophages (B); release is caspase-1-dependent and therefore absent in Casp1^−/− macrophages (C). Scale bars are 10μm. (D) The percentage of macrophages releasing one or more particles during infection; data are means and standard deviations from duplicate samples from multiple experiments. *P<0.05. (E) Supernatants from Salmonella-infected wild-type macrophages undergoing lysosome exocytosis (in the presence of calcium, S+LE) or in the absence of lysosome exocytosis (in the absence of calcium, S−LE) were concentrated and incubated with Salmonella for 1 hour and plated for CFU. All are representative of 3 experiments. *P=0.0004