Tropheryma whipplei escapes LAPosome and modulates macrophage response in a xenophagy-dependent manner

Abstract

Tropheryma whipplei, the agent of Whipple's disease, is an intracellular pathogen that replicates in macrophages. The phagocytic and cellular processes leading to the formation of T. whipplei replicative vacuole remain poorly understood. Macrophage microbicidal activity is largely related to macro/autophagy which is also essential for cell homeostasis. Here, we show that T. whipplei uptake by macrophages involved LC3-associated phagocytosis (LAP). Bacteria then escaped into the cytosol from where they were recaptured by xenophagy. We also demonstrate that T. whipplei blocked the autophagic flux to build its replicative compartment. Inhibition of LAP resulted in the decrease of interleukin (IL)-10 secretion and the restoration of the autophagy flux, suggesting that modulation of autophagy during infection alters immune response and promote persistence. Our results provide new insight in the intracellular fate of the bacteria during macrophage infection and suggest the possible involvement of previously unknown virulence factors in T. whipplei infection.

Keywords: Autophagy; Immune escape; LC3-associated phagocytosis (LAP); Macrophage; Tropheryma whipplei; Whipple’s disease.

Figure 1.

T. whipplei induces LC3-associated phagocytosis in macrophages. MDMs were infected with T. whipplei (50 bacteria per cell) for 2 hours, fixed, then stained with an anti-T. whipplei antibody in red and antibodies directed against RUBCN (A), NOX2 (B), UVRAG (C), or LC3B (D) in green. Nuclei were stained with DAPI (blue). The images were visualized by confocal fluorescence microscopy. Scale bar: 20 µm. (E) Colocalization between T. whipplei and the proteins of interest was expressed as Manders coefficient. (F) MDMs were pre-treated with 10 µM of SAR405 and/or 25 µM of GSK2795039 for 4 hours before been infected for 2 hours with T. whipplei (50 bacteria per cell), washed to remove free bacteria and then lysed. Bacterial DNA copies was determined by qPCR. The experiment was performed using three different donors (N = 3), and the values represent the mean ± standard error of the mean. *p<0.05, by two-way ANOVA. (G) MDMs were infected with T. whipplei (50 bacteria per cell) for 2 hours, then were fixed and analyzed by transmission electron microscopy; arrow in (i) shows bacteria inside a simple membrane structure and arrow in (ii) shows bacteria inside a double-membrane structure. Scale bar: 500 nm. The experiments were performed on three different donors (N = 3) and representative results are shown.

Figure 2.

T. whipplei escapes before being recaptured by xenophagy. MDMs were infected 2 hours with live or paraformaldehyde-inactivated T. whipplei (50 bacteria per cell), fixed then stained with an anti-T. whipplei antibody in red, and antibodies directed against Galectin 8 (A), or NDP52 (B), ULK1 (C), Ubiquitin (D) or SQSTM1/p62 (E) in green. Nuclei were stained with DAPI (blue). The images were visualized by confocal fluorescence microscopy. Scale bar: 20 µm. Colocalization between T. whipplei and the proteins of interest was expressed as Manders coefficient (F). (G) MDMs were infected with T. whipplei (50 bacteria per cell) for 2 hours, then fixed and analyzed by transmission electron microscopy; arrow in (i) shows bacteria free in the cytosol near a simple membrane structure and arrow in (ii) shows bacteria near a double-structure membrane like autophagore elongation. Scale bar: 500 nm. The experiments were performed in triplicates (N = 3); representative results are shown.

Figure 3.

T. whipplei infection blocks autophagic flux in macrophages. MDMs were pre-treated 4 hours with 500 nM rapamycin, or starved for 16 hours, or treated with 200 nM bafilomycin A1, or 60 µM chloroquine, before infection with T. whipplei (50 bacteria per cell) for 24 hours. (A) Cells were then washed and lysed. Whole-cell lysates were then analyzed by western blot against the indicated proteins. (B) MDMs were washed, fixed and then stained with anti-LC3B in green, anti-SQSTM1-p62 in red and DAPI to identify nuclei in blue. Yellow puncta show cargoes. (C-E) MDMs were infected or not by T. whipplei (50 bacteria per cell) for 4 hours, washed and incubated for 24 hours, washed, fixed and stained with an anti-T. whipplei antibody in red, an anti-LC3B (C), Ubiquitin (D) or SQSMT1/p62 (E) in green. Nuclei were stained with DAPI (blue). (F) Colocalization between T. whipplei and proteins of interest was expressed as Manders coefficient. The images were visualized by confocal fluorescence microscopy. Scale bar: 20 µm. The experiments were performed in triplicates (N = 3); representative results are shown.

Figure 4.

T. whipplei hijacks autophagy to build up its replicative compartment. MDMs were infected with T. whipplei (50 bacteria per cell) for 4 hours, washed and incubated for 24 hours, then fixed and analyzed by transmission electron microscopy (A) or by confocal immunofluorescence after staining with anti-T. whipplei antibody in red, and antibodies directed against RUBCN (B), NDP52 (C), Galectin 8 (D), ULK1 (E), an ATG5-ATG12 (F), BECN1 (G), m-TOR (H), Raptor (I), or UVRAG (J) in green. (K) Colocalization between T. whipplei and the proteins of interest was expressed as Manders coefficient. Nuclei were stained with DAPI (blue). Scale bar: 20 µm. The experiments were performed on three different donors (N = 3); representative results are shown.

Figure 5.

LAP induction promotes autophagy blockade and alters the immune response during T. whipplei infection. (A) MDMs were infected with T. whipplei (50 bacteria per cell) for the indicated times, then washed and lysed. Whole-cell lysates were then analyzed by western blot against the indicated proteins. (B) MDMs were infected with T. whipplei (50 bacteria per cell) 4 hours, washed to remove free bacteria and incubated for 12 days (0 corresponds to 4h infection). Every 3 days, cells were washed and lysed with RIPA buffer. Whole-cell lysates were then analyzed by western blot against the indicated proteins. (C) MDMs were pre-treated with 10 µM of SAR405, 25 µM of GSK2795039 or both for 4 hours and infected with T. whipplei (50 bacteria per cell) for 2 hours, Cells were washed and lysed with RIPA buffer. Whole-cell lysates were then analyzed by western blot against the indicated proteins. (D) MDMs from different donors (N=3) were infected with T. whipplei (50 bacteria per cell) for 4 hours, washed and incubated for 12 days. Gene expression of RUBCN was monitored every 3 days (day 0 corresponds to 4 hours infection) by qRT-PCR and expressed as fold change after normalization to the ACTB endogenous control (i) or at the translational level by measuring mean fluorescence intensity of RUBCN stained cells by confocal microscopy (ii). **p<0.01, ***p<0.001, *****<0.0001 by one-way ANOVA. (iii) Representative immunofluorescences are shown with anti-T. whipplei in red, anti-RUBCN in green and DAPI in blue. (E-H) MDMs were pre-treated with both 10 µM of SAR405 and 25 µM of GSK2795039 4 hours before infection with T. whipplei (50 bacteria per cell). After 24 hours, supernatants were collected and levels of IL-1β (E), TNF (F), IL-6 (G) and IL-10 (H) were assessed by ELISA. The experiment was performed using different donors (N = 3), representative results are shown for western-blots, and the values represent the mean ± standard error. ****p<0.0001 by two-way ANOVA.

Figure 6.

T. whipplei hijacks the autophagic flux. In macrophages, T. whipplei is enveloped in a LAPosome expressing RUBCN, and NOX2, characteristic of LC3-associated phagocytosis (1). LAP induction is associated with secretion of the immunomodulatory cytokine IL-10. T. whipplei then escapes (2) in the cytosol where it is tagged by Galectin 8 and NDP52 allowing the elongation of a phagophore (3). Subsequently, and probably thanks to the secretion of yet unknown bacterial factors, the bacterium hijacks the autophagic machinery (4), notably by blocking fusion and/or cargoes fusion with lysosomes, to invest its replicative niche which expresses LC3, Lamp1 but lacks cathepsin-D. Induction of LAP and perennial over-expression of RBCN could promote inhibition of autophagy and alternate activation of macrophage, by IL-10 secretion and favoring persistence of infection.