Phagosomal rupture by Mycobacterium tuberculosis results in toxicity and host cell death

Abstract

Survival within macrophages is a central feature of Mycobacterium tuberculosis pathogenesis. Despite significant advances in identifying new immunological parameters associated with mycobacterial disease, some basic questions on the intracellular fate of the causative agent of human tuberculosis in antigen-presenting cells are still under debate. To get novel insights into this matter, we used a single-cell fluorescence resonance energy transfer (FRET)-based method to investigate the potential cytosolic access of M. tuberculosis and the resulting cellular consequences in an unbiased, quantitative way. Analysis of thousands of THP-1 macrophages infected with selected wild-type or mutant strains of the M. tuberculosis complex unambiguously showed that M. tuberculosis induced a change in the FRET signal after 3 to 4 days of infection, indicating phagolysosomal rupture and cytosolic access. These effects were not seen for the strains M. tuberculosisΔRD1 or BCG, both lacking the ESX-1 secreted protein ESAT-6, which reportedly shows membrane-lysing properties. Complementation of these strains with the ESX-1 secretion system of M. tuberculosis restored the ability to cause phagolysosomal rupture. In addition, control experiments with the fish pathogen Mycobacterium marinum showed phagolysosomal translocation only for ESX-1 intact strains, further validating our experimental approach. Most importantly, for M. tuberculosis as well as for M. marinum we observed that phagolysosomal rupture was followed by necrotic cell death of the infected macrophages, whereas ESX-1 deletion- or truncation-mutants that remained enclosed within phagolysosomal compartments did not induce such cytotoxicity. Hence, we provide a novel mechanism how ESX-1 competent, virulent M. tuberculosis and M. marinum strains induce host cell death and thereby escape innate host defenses and favor their spread to new cells. In this respect, our results also open new research directions in relation with the extracellular localization of M. tuberculosis inside necrotic lesions that can now be tackled from a completely new perspective.

Figure 1. M. marinum is able to rupture the phagosome of THP-1 cells unlike an ESAT-6/CFP-10 secretion-deficient strain.

THP-1 cells were infected with M. marinum expressing DsRed, M. smegmatis or M. marinum M^VU at a MOI of 1 for the indicated time and then loaded with CCF-4 molecule for 2 h. After PFA fixation, cells were imaged by fluorescence widefield microscope Nikon Ti with 20X objective (A). Picture acquisition was done randomly and automatically for each condition on 36 fields to follow the FRET signals and the fluorescent bacteria. Data represented in panels B and C were obtained via specialized algorithms using the Metamorph software. The plots are representative of 3 independent experiments.

Figure 2. Unlike the attenuated BCG vaccine, virulent M. tuberculosis is able to induce phagosomal rupture in THP-1 macrophages.

THP-1 cells were infected with BCG DsRed (A,C) or M. tuberculosis DsRed (B,D) at a MOI of 1 for the indicated time and then loaded with CCF-4 molecule for 2 h. After PFA fixation, cells were imaged using a fluorescence widefield microscope (Nikon Ti) with an 40X objective (A, B). Picture acquisition was achieved randomly and automatically for each condition on 49 fields in duplicates and further 450/535 nm intensity ratio measurements shown in panels C and D were obtained through analysis by a specialized algorithm using the Metamorph software. The plots are representative of 3 independent experiments.

Figure 3. Analysis of ESX-1 deletion-, truncation- and complementation-mutants highlights the link between phagosomal rupture and functional ESX-1 secretion.

THP-1 cells were infected with M. tuberculosisΔRD1 (A,D), BCG::RD1 (B,E) or BCG::RD1-ESAT-6Δ84–95 (C,F) at a MOI of 1 for the indicated time and then loaded with CCF-4 molecule for 2 h. After PFA fixation, cells were imaged using a fluorescence widefield microscope (Nikon Ti) equipped with a 40X objective (A,B,C). Picture acquisition was achieved randomly and automatically for each condition on 49 fields in duplicates and further 450/535 nm intensity ratio measurement (D,E,F) was obtained through analysis by a specialized algorithm using the Metamorph software. The plots are representative of 3 independent experiments.

Figure 4. Evolution of cell numbers during the time course of infection with different M. tuberculosis complex members.

THP-1 cells were infected with M. tuberculosis PFA killed, BCG, BCG::RD1, M. tuberculosis or M. tuberculosisΔRD1 at a MOI of 1 for the indicated time. Arrows indicate the decrease of the number of cells with virulent strains. After PFA fixation, cells were stained by incubation with WGA (wheat germ agglutinin) and imaged using a Nikon Ti widefield microscope. Pictures represent a mosaic of 49 individual images from 96 well plates acquired with a 20X objective (A). Cytofluorometry analysis of THP-1 cells infected with various wild-type and recombinant M. tuberculosis and BCG strains at day 7 post-infection. THP-1 cells were stained with FLICA Poly-Caspases SR-VAD-FMK (apoptosis) and Live/Dead Pacific Blue (necrosis) (B).

Figure 5. Necrosis induction does not lead to the release of mycobacteria from the phagolysosome to the cytosol.

THP-1 cells were infected with BCG at a MOI of 1 for 2 days before necrosis induction using 100 µg/ml concanavalin A for 24 h or 48 h. After CCF-4 loading for 2 h, cells were incubated 5 minutes in the presence of 2 µg/ml propidium iodide (PI) and then subjected to PFA fixation. Cells were imaged using fluorescence widefield Nikon Ti microscope with 40X objective (A). Picture acquisition was achieved randomly and automatically for each condition on 49 fields and further 450/535 nm intensity ratio measurements (B,C) were obtained through analysis by a specialized algorithm using the Metamorph software. The plots were representative of 2 independent experiments.