CD40 induces macrophage anti-Toxoplasma gondii activity by triggering autophagy-dependent fusion of pathogen-containing vacuoles and lysosomes

Abstract

Many intracellular pathogens, including Toxoplasma gondii, survive within macrophages by residing in vacuoles that avoid fusion with lysosomes. It is important to determine whether cell-mediated immunity can trigger macrophage antimicrobial activity by rerouting these vacuoles to lysosomes. We report that CD40 stimulation of human and mouse macrophages infected with T. gondii resulted in fusion of parasitophorous vacuoles and late endosomes/lysosomes. Vacuole/lysosome fusion took place even when CD40 was ligated after the formation of parasitophorous vacuoles. Genetic and pharmacological approaches that impaired phosphoinositide-3-class 3 (PIK3C3), Rab7, vacuolar ATPase, and lysosomal enzymes revealed that vacuole/lysosome fusion mediated antimicrobial activity induced by CD40. Ligation of CD40 caused colocalization of parasitophorous vacuoles and LC3, a marker of autophagy, which is a process that controls lysosomal degradation. Vacuole/lysosome fusion and antimicrobial activity were shown to be dependent on autophagy. Thus, cell-mediated immunity through CD40 stimulation can reroute an intracellular pathogen to the lysosomal compartment, resulting in macrophage antimicrobial activity.

Figure 1. CD40 stimulation induces vacuole/lysosome fusion in human and mouse macrophages infected withT. gondii .

(A) Control and CD154-stimulated human macrophages were incubated with LysoTracker Red, then infected with T. gondii–YFP. Macrophages incubated with opsonized T. gondii were used as controls. Cells were examined by confocal microscopy 6 hours after infection or 2 hours after addition of opsonized parasites. Macrophages shown contain 1 tachyzoite of T. gondii. (B–D) Control and CD40-activated human macrophages were infected with T. gondii–YFP. Macrophages incubated with opsonized T. gondii were used as controls. Macrophages were incubated with anti–LAMP-1 (B), anti-CD63 (C), or anti-cathepsin D (D) Abs; this was followed by addition of secondary antibodies. Cells were examined by confocal microscopy at 6 hours after infection or 1 hour after addition of opsonized T. gondii. CD40-activated macrophages show colocalization of LAMP-1, CD63, and cathepsin D (rings) around T. gondii–containing vacuoles (arrowheads). Cath, cathepsin. Scale bars: 5 μm. (E and F) Quantification of colocalization of late endosomal/lysosomal markers around vacuoles containing T. gondii within human (E) or mouse (F) primary macrophages. Monolayers were examined at 6 and 8 hours after challenge in the case of human and mouse macrophages, respectively. In the case of macrophages incubated with opsonized T. gondii, M6PR was assessed at 15 minutes while LAMP-1, LAMP-2, CD63, and cathepsin D expression were assessed at 1 hour. Percentages indicate the mean ± SD. Results shown are representative of 3–4 independent experiments. Cath, cathepsin; ctr, control; DIC, differential interface contrast; ops, opsonized.

Figure 2. Kinetics of recruitment of endosomal and lysosomal markers.

Mouse peritoneal macrophages were incubated with control (ctr) or anti-CD40 mAbs, then infected with T. gondii. Expression of late endosomal and lysosomal markers (A) or early endosomal markers (B) was examined by immunofluorescence at different times after challenge. Opsonized tachyzoites were used as positive control. Results are shown as the mean ± SD and are representative of 3 independent experiments.

Figure 3. CD40 stimulation induces convergence of PV and late endosomes/lysosomes.

Mouse resident peritoneal macrophages were incubated with control or stimulatory anti-CD40 mAbs, then challenged with T. gondii–secRFP. Expression of LAMP-1 was assessed at 8 hours after challenge. Inset represents an extracellular tachyzoite with accumulation of fluorescence in dense granules. Macrophages were also incubated with heat-killed T. gondii–secRFP, and LAMP-1 expression was assessed at 1 hour after challenge. Arrowheads indicate colocalization of LAMP-1 around T. gondii–containing compartments. Scale bar: 5 μm. Results are representative of 4 independent experiments.

Figure 4. CD40 stimulation induces fusion of preformed PV and late endosomes/lysosomes.

Bone marrow macrophages from TNF-α^–/– mice infected with CPSII KO T. gondii–YFP were incubated overnight in the presence of uracil (0.2 mM) to allow parasite replication. After uracil removal, infected macrophages were cultured with stimulatory anti-CD40 mAb for 24 hours. TNF-α (500 pg/ml) was then added to monolayers, and cells were examined after 7 hours. Arrowheads denote colocalization of LAMP-1 (ring) around PV in CD40-activated macrophage. Scale bar: 5 μm. Results are representative of 4 independent experiments.

Figure 5. Activated CD4⁺ T cells induce vacuole/lysosome fusion through CD40/CD154 interaction.

Peritoneal macrophages from WT mice (A) or IFN-γ^–/– mice (B) were infected with T. gondii–YFP and incubated for 10 minutes; this was followed by removal of extracellular parasites. After 1 hour, either resting (R) or activated (A) CD4⁺ T cells from WT (A) or IFN-γ^–/– (B) mice were added in the presence of anti-CD54 or control mAbs. Colocalization of LAMP-1 around PV was assessed by immunofluorescence 8 hours after infection. Results are shown as the mean ± SD and are representative of 4 independent experiments.

Figure 6. Vacuole/lysosome fusion mediates macrophage antimicrobial activity induced by CD40.

Mouse peritoneal macrophages were incubated with control or anti-CD40 mAbs or with IFN-γ/LPS, then infected with T. gondii. (A and B) Pepstatin (50 μM) was added 1 hour after infection. Percentage of infected macrophages and parasite load was assessed by light microscopy at 1 and 18 hours after challenge. (C–E) Bafilomycin A (BFA; 25 nM) was added 1 hour before infection with T. gondii. (C) Expression of cathepsin D was examined by immunofluorescence 8 hours after infection. (D) Percentage of infected macrophages and parasite load were assessed at 1 and 18 hours after challenge. (E) Parasite load was determined at 18 hours. Results are shown as the mean ± SD and are representative of 4 independent experiments.

Figure 7. Blockade of vacuole/lysosome fusion ablates CD40-induced antimicrobial activity.

The effects of knockdown of PIK3C3 (A–C) and Rab7 dominant-negative mutant (D–F) on vacuole/lysosome fusion and antimicrobial activity were examined. (A) hmCD40–RAW 264.7 cells were mock transfected (M) or were transfected with sense (S) or antisense (AS) ODN against PIK3C3. Protein expression of PIK3C3 and actin were analyzed 48 hours after transfection. (B) Transfected cells were incubated with or without hCD154, then infected with T. gondii–YFP. Vacuole/lysosomal fusion was assessed by cathepsin D staining 8 hours after infection. (C) Transfected cells were treated with medium alone, hCD154, or IFN-γ/LPS; this was followed by T. gondii infection. Cells were examined by light microscopy 18 hours after infection. (D) CD40-activated and control mouse peritoneal macrophages were infected with T. gondii–YFP followed by assessment of Rab7 expression by immunofluorescence. Arrowhead denotes colocalization of Rab7 (ring) around T. gondii–containing vacuole in CD40-activated macrophage. Scale bar: 5 μm. (E) hmCD40–RAW 264.7 cells were incubated with medium with or without hCD154 followed by transfection with Rab7 WT or Rab7(T22N). Cells were infected with T. gondii–RFP. Vacuole/lysosomal fusion was assessed by cathepsin D staining 8 hours after infection. (F) Transfected hmCD40–RAW 264.7 cells treated with medium alone, CD154, or IFN-γ/LPS were challenged with T. gondii. Cells were examined by light microscopy 18 hours after challenge. Results are shown as the mean ± SD and are representative of 4 independent experiments. DN, dominant negative.

Figure 8. CD40 stimulation induces vacuole/lysosome fusion and antimicrobial activity through autophagy.

(A) Control and CD40-activated hmCD40–RAW 264.7 cells were transfected with LC3-EGFP and infected with T. gondii–RFP. Monolayers were examined by confocal microscopy at 5 hours after challenge. Arrowhead indicates accumulation of LC3 around vacuole. Scale bars: 5 μm. (B) Kinetics of colocalization of LC3 and LAMP-1 around PV in hmCD40–RAW 264.7 cells. (C) hmCD40–RAW 264.7 cells were transfected with control or Beclin 1 siRNA. Immunoblot was performed after 96 hours. Two days after transfection with control or Beclin 1 siRNA, hmCD40–RAW 264.7 cells were transfected with LC3-EGFP, then incubated in complete medium or HBSS for 1 hour. Expression of LC3 was analyzed by confocal microscopy. (D) Control or CD40-activated hmCD40–RAW 264.7 cells transfected with control or Beclin 1 siRNA were infected with T. gondii–YFP, then stained for cathepsin D 8 hours after infection. Percentages of vacuoles that colocalized with cathepsin D were assessed by immunofluorescence. (E) Control, CD40-activated, or IFN-γ/LPS–treated hmCD40–RAW 264.7 cells transfected with control or Beclin 1 siRNA were infected with T. gondii followed by assessment of parasite load by light microscopy 18 hours after infection. (F and G) Control, CD40-, or IFN-γ–activated human monocyte–derived macrophages were infected with T. gondii, then incubated with or without 3-MA (10 mM). Expression of cathepsin D was examined by immunofluorescence 6 hours after infection (F), and parasite load was assessed at 18 hours after infection (G). Results are shown as the mean ± SD and are representative of 3 to 4 independent experiments.