Fig 1.
Mtb and Mbv differentially replicate and localise within human or bovine macrophages.
(A) Confocal images of monocyte-derived GM-CSF differentiated-human or bovine Mϕ (hMϕ or bMϕ) infected with RFP-expressing M. tuberculosis H37Rv (Mtb-RFP) or M. bovis (Mbv-RFP) for 2, 24 and 72 h. Brightfield was used to visualize the cells. Cell nuclei were stained with DAPI (blue) and bacteria-RFP are visualized in red. Scale bar: 20 μm. (B and C) Quantification of intracellular growth expressed in bacteria area (μm2) per infected cell of Mtb-RFP and Mbv-RFP within hMϕ (B) and bMϕ (C). (D and E) Electron microscopy images of hMϕ (D) and bMϕ (E) infected with Mtb or Mbv for 24 h. Asterisks mark the intracellular bacteria. Images were selected to illustrate free cytosolic bacteria (left-hand panels), phagosomal bacteria (middle panels) and bacteria surrounded by multiple membranes (right-hand panels). Scale bar, 200 nm. (F) Quantification by stereology of the proportion of bacteria contained in each compartment. Black represents the proportion of cytosolic bacteria; green, the proportion of single membrane bound bacteria, and grey, the bacteria surrounded by multiple membranes. “n” represents the number of cells analysed. The total number of bacteria analysed corresponds to the number in brackets. (G) Quantification by RT-qPCR of the relative fold change mRNA expression of interferon-β (IFNB1) in hMϕ (left graph) and bMϕ (right graph) infected with Mtb or Mbv for 24 h. Data are normalized to Mtb. (Housekeeping gene used: GAPDH). Non-infected cells (NI) were use as control. For all the figures, p-value is considered significant when < 0.05 and indicate as follow: *p<0.05; ** p<0.01; *** p<0.001; ns: not significant.
Fig 2.
Mbv specifically induces the formation of MNGCs in bMϕ.
(A) Fluorescence confocal images of bMϕ infected with Mtb-RFP or Mbv-RFP for 24 h. Non-infected cells (NI) were used as a control. The bacteria are visualized in red, the cell actin cytoskeleton is in white (phalloidin-488) and cell nuclei (DAPI) in cyan. Arrows highlight the cells containing 2 or more nuclei, corresponding to multinucleated giant cells (MNGCs). Scale bar, 50 μm. (B) Quantification of the percentage of MNGCs in bMϕ infected with Mtb-RFP or Mbv-RFP for 24 hours. Non-infected cells (NI) were used as a control. Each dot represents one donor tested. (C) Quantification of the percentage of MNGCs in hMϕ infected with Mtb-RFP or Mbv-RFP for 24 h. Non-infected cells (NI) were used as a control. (D) Electron microscopy image of an uninfected bystander-MNGC containing four distinct nuclei (marked by red arrows). The images displayed on the right side correspond to the magnification of each region (a-c) delimited by black squares in the main image. Scale bar are D: 40 μm; Da: 2 μm; Db: 2 μm; Dc: 1 μm. (E) Maximum projection (14 Z-stacks with an interval of 0.8 μm) from confocal images of bovine MNGCs without intracellular Mbv-RFP (Byst-MNGC) or containing intracellular Mbv-RFP (Mbv-MNGC). In the latter image, the presence of intracellular bacteria is highlighted in the region delimited by a white square, magnified in the right corner. Bacteria are visualized in red, the cells with brightfield image, and nuclei in cyan. A dashed line delimits the edge of MNGCs. (F) Quantification of the percentage of bystander and Mbv-MNGCs in the bMϕ population for two independent donors. Each bar chart represents the total MNGC population (100%); in black is percentage of Mbv-infected MNGCs, and in grey the percentage of uninfected bystander-MNGCs.
Fig 3.
Secreted bacterial MPB70 contributes to MNGC formation in bMϕ.
(A) Culture filtrate from Mbv wild-type (WT), Mbv ΔMPB70, Mbv ΔMPB70/MPB70 (Mbv-Compl.) were assessed for immuno-labelling of MPB70 (23 kDa). Immuno-labelling of Ag85 (38 kDa) was used as gel loading control (B) Fluorescence confocal images of bMϕ infected with Mbv WT, Mbv ΔMPB70, Mbv-Compl. for 24 h. Uninfected cells had been used as a negative control. Cell nuclei are stained with DAPI (cyan) and actin is visualized in grey. Yellow arrows point the multi-nucleated cells. Scale bar, 50 μm (C) Quantification of MNGCs from each experiment displayed in B. Graph represents the quantification of MNGCs for each condition tested where each dot represents one bovine donor. Data shown are representative of four biological repeats. (D) Fluorescence confocal images of bMϕ infected with Mbv WT, Mtb, Mtb-MPB70+ for 24 h. Uninfected cells had been used as a negative control. Cell nuclei were stained with DAPI (cyan); bacteria-RFP and actin are visualized in red and grey, respectively. Yellow arrows point the multi-nucleated cells. Scale bar, 50 μm (E) Quantification of MNGCs from each experiment displayed in D. Graph represents the quantification of MNGCs for each condition tested where each dot represents one bovine donor. Data shown are representative of two independent experiments.
Fig 4.
Extracellular vesicles produced by Mbv-infected bMϕ induce cell multinucleation.
(A) A total of 192 proteins were determined to be differentially regulated using an ANOVA and Tukey’s post-hoc test in the secretome of infected bMϕ and sorted according to Gene Ontology Cellular Component (GOCC). (B) List of 27 proteins differentially expressed in the secretome of Mtb-infected or Mbv-infected bMϕ clustered into 3 groups: extracellular vesicles, membrane trafficking and cell adhesion or others. The log2 fold change Mtb or Mbv, both normalised to uninfected (NI), samples are displayed in the left panel and the log2 fold change Mbv/Mtb is displayed in the right panel. Proteomics data were obtained from 3 independent experiments carried out with 3 different bovine donors. (C) EV purification procedure summarized in 3 steps. (D) Electron microscopy micrographs of EV-enriched fraction. The white arrows show the extracellular vesicles. Scale bar: 0.1 μm. (E) Fluorescence confocal images of naïve bMϕ stimulated for 24 h with EV-enriched fraction from uninfected (NI), Mtb-infected or Mbv-infected bMϕ. As EVs were purified in PBS, a similar volume of PBS was added to the bMϕ as a control. Cell nuclei are stained with DAPI (cyan) and actin is visualized in grey. Yellow arrows point the MNGCs. Scale bar: 50 μm (F) Quantification of MNGCs after bMϕ stimulation with EV-enriched fractions. The graph represents the quantification of MNGCs for each condition tested where each dot represents one bovine donor. Data are representative of 4 independent experiments.
Fig 5.
Granulomas from Mbv-infected cattle contain a higher number of MNGCs than Mtb-infected cattle.
(A, left image) Histological Haematoxylin and Eosin (H&E) staining of granulomatous lesions in thoracic lymph nodes of cattle challenged for 10 weeks with Mtb H37Rv strains. Scale bar, 200 μm. (A, right image) higher magnification of multinucleated giant cells (white square). Scale bar, 50 μm. (B, left image) H&E staining of granulomatous lesions in thoracic lymph nodes of cattle challenged for 6 weeks with Mbv AF2122/97. Scale bar, 200 μm. (B, right image) higher magnification of multinucleated giant cells (white square). Scale bar, 50 μm. (A and B) Granulomas are classified from early stage of maturation (type I and II) to late granulomas (type III and IV) (C) Quantification of the number of granulomas counted for each condition. (D) Quantification of bacterial load in each granuloma based on acid fast staining of the tissue section and expressed as the mean Acid Fast Bacilli (AFB) ± SEM. (E) Quantification of the number of MNGCs per granuloma ± SEM.
Fig 6.
MNGCs provide a restrictive environment for Mbv.
(A) (upper panel) Tile scan of 3 successive confocal images showing the diversity of bMφ infected with Mbv-GFP for 24 h. Cells were fixed and immunolabelled for LAMP-1 (red), Mbv-GFP (green) and cell nuclei in blue (DAPI). The lower panel represents a crop of the region marked by a white square. The different cell population are marked as follow: *Mono: Mononucleated cells; **MNGC (early): Multinucleated cells with 2 or 3 nuclei; ***MNGC (mature): Multinucleated cells with over than 3 nuclei. Right panel: the quantification of the intracellular bacterial load in mononucleated and early or mature multinucleated cells is expressed in bacteria area per cell area. (Mono, n = 60 cells; MNGC (early), n = 22 cells; MNGC (mature), n = 28 cells). (B, C, D) bMφ infected with Mbv-GFP (displayed in green) for 24 h were stained with (B) LysoTracker Red (LTR, in red), (C) the pan cysteine-cathepsin activity-based probe BMV109 (in grey) and (D) immuno-labelled with a LC3B antibody (in red). DAPI was used to label nuclei (in blue). (B) LTR (number of bacterial object analysed (n)); Mono, n = 633; MNGC (early), n = 80; MNGC (mature), n = 88). (C) BMV109; Mono, n = 368; MNGC (early), n = 52; MNGC (mature), n = 76). (D) LC3B; Mono, n = 1374; MNGC (early), n = 243; MNGC (mature), n = 133). Each micrograph is a representative image of mononucleated cells (Mono), multinucleated giant cells (MNGC) with 2 or 3 nuclei (early) or over 3 nuclei (mature). The lower panels correspond to a zoom of the region marked by a white square. The quantification of each marker’s association with bacteria is shown on the right panels. The dashed lines show the fluorescence cut-off applied. The data are representative from two independent experiments from 2–3 donors.