ATP-Mediated Killing of Intracellular Mycobacteria by Macrophages Is a P2X₇-Dependent Process Inducing Bacterial Death by Phagosome-Lysosome Fusion

Cells and bacteria

The murine macrophage cell line J774a.1 was obtained from the American Type Culture Collection (Manassas, VA). The cells were cultured in DMEM (Life Technologies, Paisley, U.K.) supplemented with 10% FCS and 2 mM glutamine. The murine macrophage cell lines RAW 246.7, wild-type (WT) RAW 7.5R, and RAW 10S were a kind gift from J. M. Blackwell (Cambridge University, Cambridge, U.K.). The 7.5R and 10S lines are stable transfectants, the generation of which has been described previously (22). These were cultured in RPMI (Life Technologies) supplemented with 10% FCS and 2 mM glutamine. P2X₇^−/−, p47^phox−/−, and inducible NO synthase (iNOS)^−/− mice were kind gifts from I. P. Chessell (Glaxo Wellcome, Cambridge, U.K.), A. W. Segal (University College of London, London, U.K.), and F. Y. Liew (University of Glasgow, Glasgow, U.K.), respectively, and have been previously described (23, 24, 25). The genetic backgrounds of these knockout mice were, respectively, C57/129, C57, and 129. WT C57BL/6 and 129 were used as controls, as appropriate. Bone marrow-derived macrophages (BMDMs) were obtained as follows. Femurs and tibias were flushed with DMEM, and the bone marrow was washed and plated out overnight in DMEM supplemented with 10% FCS, 10% GM-CSF-containing medium, and 2 mM glutamine. Nonadherent cells were then removed, washed, and plated at 10⁶/ml in the above medium. The cells were cultured for a further 9 days, with changes of the medium on days 3 and 6. On day 9, the cells were adherent and morphologically looked like mature macrophages.

M. bovis BCG Evans strain was obtained from Evans Medical (Speke, U.K.) and maintained in log phase growth in 7H9 broth supplemented with 10% albumin dextrose catalase (ADC) enrichment medium and 0.2% Tween 80. Green fluorescent protein expressing BCG (GFP-BCG) was a kind gift from D. Young (Imperial College of London, London, U.K.). These contained the gene encoding a FACS-optimized GFP protein (26) constitutively expressed under the control of the mycobacterial heat shock protein 60 promoter in a pSMT3 shuttle vector (27) construct. Stock aliquots, stored in glycerol at −70°C, were grown to log phase in 7H9 broth supplemented with 10% ADC enrichment medium, 0.2% Tween 80, and 50 μg/ml hygromycin (Sigma, St. Louis, MO).

BCG viability assay

Macrophages were grown to subconfluence in 200 μl/well cultures in 96-well round-bottom microtiter plates. This approximates to 5 × 10⁴ cells in 200 μl/well. A total of 5 × 10⁵ BCG Evans per well was added. To aid infection, plates were centrifuged at 700 × g for 10 min and incubated at 37°C for 60 min. Excess BCG were removed by washing. As assessed by CFU, ∼10% of BCG was taken up, giving a ratio of one viable BCG per macrophage. Where appropriate, the following metabolic inhibitors were then added for 60 min. N^G-Monomethyl-L-arginine monoacetate (L-NMMA) (400 μM), superoxide dismutase (15 U/ml), catalase (0.1 mg/ml), and butan-1-ol (0.05–1%) were all purchased from Sigma. A 30 mM ATP (Boehringer Mannheim, Indianapolis, IN) solution was prepared in culture medium, adjusted to pH 7 with sodium bicarbonate, and passed through a 0.22-μM filter (Millipore, Bedford, MA) to sterilize. It was added to the macrophage cultures to give a final concentration 3 mM, and incubated at 37°C, 5% CO₂. At appropriate time points, the plates were centrifuged at 700 × g for 10 min. A total of 150 μl supernatant was removed and replaced with 130 μl 7H9 medium, supplemented as above, plus 20 μl 2% saponin solution. Cell lysis was promoted by vigorous pipetting and then incubation at 37°C for 120 min. Fifty-microliter aliquots from each well were serially diluted in 7H9 supplemented with 10% ADC enrichment and 0.2% glycerol and plated in 96-well flat-bottom plates. The plates were then incubated at 37°C for 12 days. Microcolonies were counted using an inverted microscope (×100 magnification). All experiments were performed in quadruplets.

Lucifer yellow uptake

Pore formation was assessed using a method adapted from that described by Steinberg et al. (28). Briefly, macrophages adherent to coverslips were incubated with 500 μg/ml lucifer yellow dye, 250 μM sulfinpyrazone (Sigma), and 3 mM ATP in PBS for 5 min. They were washed in 250 μM sulfinpyrazone in PBS and viewed immediately on a Zeiss confocal laser-scanning microscope (Zeiss, Oberkochen, Germany). Cells with pores larger than 463 Da exhibit intracellular staining.

Chromium-release assay

Mouse BMDMs were cultured in 96-well round-bottom microtiter plates, as previously described. Aliquots of ⁵¹Cr (ICN Pharmaceuticals, Costa Mesa, CA) were added at a concentration of 2 μCi/well, and the plates were incubated for 16 h. Cells were washed and incubated with 3 mM ATP for a further 16 h. Supernatants were removed, and their activity was counted on a Wallac 1260 Multi Gamma II gamma counter (PerkinElmer Wallac, Gaithersburg, MD). The cells themselves were dissolved in 1% Triton X-100 solution and also counted. Results are expressed as the percentage of the total activity in the supernatant.

Confocal fluorescent microscopy

Macrophages were grown to subconfluence on 13-mm glass coverslips in 24-well plates. This approximated to 5 × 10⁵ cells/coverslip. GFP-BCG were resuspended in DMEM, and a single cell culture was prepared by water bath sonication for 5 min. Any remaining clumps were removed by slow speed centrifugation (100 × g for 5 min). The bacterial concentration and confirmation of a single cell suspension were determined using a thoma chamber. The BCG were then added to the macrophages to give a final concentration of 2.5 × 10⁷/coverslip. The plate was centrifuged at 700 × g for 10 min. The cells were incubated for 1 h at 37°C, 5% CO₂, washed three times in DMEM to remove any extracellular bacteria, and incubated for an additional hour at 37°C, 5% CO₂.

LysoTracker (Molecular Probes, Eugene, OR) was added to the cells (final concentration of 50 nM) for 1 h before stimulation. ATP was then added (final concentration of 3 mM), and the cells were incubated for the indicated times.

Before fixation, coverslips were washed three times in HBSS (Sigma). They were then fixed in 3% paraformaldehyde containing 10 mM lysine, 140 mM saline, and 40 mM HEPES (pH 7.3) for 15 min at room temperature. After washing in PBS plus 5% dextrose, the coverslip cultures were mounted on glass slides with Perma Fluor (Lipshaw Manufacturing, Detroit, MI), and the edges were sealed with nail varnish.

Slides were viewed with a Zeiss confocal laser-scanning microscope. Twenty fields were examined per slide, and representative pictures were taken.

Electron microscopy

The 3-(2,4-dinitroanilino)-3′-amino-N-methyl dipropylamine (DAMP; Molecular Probes) staining was adapted from Crowle et al. (29). BMDMs were cultured as described above on 13-mm plastic coverslips and infected with BCG Evans at an infection ratio of 5 BCG to 1 macrophage. ATP (3 mM final) was then added, and 30 min before the end of the experiment DAMP (50 μM final) was also added. The coverslips were washed three times in PBS and fixed in 2% gluteraldehyde in 0.1 M calcodylate and 2.5% sucrose (pH 7.35) for 60 min at 4°C. They were washed in 50 mM ammonia chloride, and then progressively dehydrated in ethanol, infiltrated with London Resin white (Agar Scientific, Stanstead, U.K.), and polymerized with UV light. Ultrathin sections were collected on nickel grids, which were incubated for 10 min with 0.1 M HCl. These were then blocked with 0.2% BSA and 0.2% goat serum in PBS for 45 min at room temperature. The primary Ab was a mouse antidinitrophenol monoclonal (NovoCastra, Newcastle, U.K.). A mouse IgG1 control Ab (BD PharMingen, San Diego, CA) was also used to exclude nonspecific staining. These were added at a concentration of 0.5 ng/ml for 16 h at 4°C. The secondary was a goat antimouse polyclonal Ab conjugated to 5 nM gold particles (British Biocell International, Cardiff U.K.). This was used at a dilution of 1/50 and incubated with the cells for 90 min at room temperature. Grids were counterstained with urinyl acetate and viewed on a Joel Temscan (Wellwyn Garden City, U.K.) electron microscope. Pictures were taken on Agfa Scientia (Agar Scientific) EM film. The number of gold particles per vacuole was counted. Counts were performed for 20 vacuoles, involving at least 10 macrophages for each experiment.

Statistics

The Student’s t test was used for all statistical analysis.

Macrophage BCG killing kinetics post-ATP stimulation

ATP stimulation of BCG-infected J774 cells resulted in ∼50% loss of mycobacterial CFU within 60 min posttreatment (Fig. 1⇓). Loss of mycobacterial viability was detectable within 15 min and maximal at 2 h posttreatment. Subsequently, all viability studies were performed at 2 h post-ATP stimulation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

BCG viability post-ATP stimulation of J774 macrophages. Macrophages of the murine cell line J774 were infected with BCG for 60 min. Extracellular BCG were washed off, and the macrophages were then stimulated with 3 mM ATP. After the indicated time, they were lysed and assayed for BCG CFU. The time course shows rapid cell death within 15 min, with a maximum response within 60 min. SDs for quadruplets are shown, and the results are representative of three experiments. Similar results were found with human monocyte-derived macrophages, RAW macrophages, and BMDMs from BALB/c, C57BL/6, and 129 mice.

Similar BCG killing kinetics was observed in another murine macrophage cell line (RAW), and with the human monocyte cell line (THP-1), and in murine BMDMs from various mouse strains (C57BL/6, BALB/c, and 129) (results not shown).

The ATP/P2X₇ effector mechanism was confirmed as a cell-mediated function, as ATP had no effect on BCG grown in macrophage-free cultures (results not shown).

ATP-mediated pore formation and macrophage death are P2X₇ dependent

BMDMs from P2X₇^−/− mice were examined for their responses following ATP stimulation. Cells from WT mice were shown to rapidly take up the cell-impermeable dye lucifer yellow following treatment with ATP for 5 min, indicating P2X₇ pore formation had occurred (Fig. 2⇓, A and B), whereas cells from P2X₇^−/− mice failed to do so (Fig. 2⇓, C and D). WT BMDMs were also shown to die in response to ATP stimulation over a 16-h period, as indicated by ⁵¹Cr release, whereas macrophages from P2X₇^−/− mice remained viable (Fig. 3⇓a).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Macrophage membrane permeabilization following ATP stimulation. Macrophages were cultured from the bone marrow of P2X₇^−/− mice and WT littermates. They were stimulated with 3 mM ATP for 5 min in the presence of the cell-impermeable dye lucifer yellow. The cells were then washed and immediately viewed with confocal fluorescent (A and C) and transmission (B and D) microscopy. A and B demonstrate stimulated WT macrophages that are permeable to the dye, whereas C and D demonstrate stimulated macrophages from P2X₇^−/− mice whose membranes have remained impermeable. The pictures are representative of numerous fields from three experiments.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

a, Chromium release post-ATP stimulation of P2X₇^−/− BMDMs. b, Intracellular BCG killing by ATP-stimulated P2X₇^−/− BMDMs. BMDMs were cultured from P2X₇^−/− mice and WT littermates. a, Macrophages were labeled overnight with ⁵¹Cr. Excess chromium was removed. Where indicated, they were then stimulated for 16 h with 3 mM ATP. γ Activity was counted in both the supernatant and cells. Results are expressed as percentage released over total. SDs for quadruplets are shown, and the results are representative of three experiments. After stimulation, the amount of chromium released by the WT macrophages increases, indicating that cell death had occurred. There was no response in the P2X₇^−/− macrophages. b, The macrophages were infected with BCG for 60 min, and extracellular BCG were washed off. Where appropriate, macrophages were stimulated with 3 mM ATP for 120 min. They were then lysed and assayed for BCG CFU. The percentage of intracellular BCG killed is calculated by comparing the CFU from stimulated with the CFU from unstimulated macrophages. SDs for four experiments, each done in quadruplet, are shown. WT, but not P2X₇^−/−, macrophages are effective at killing BCG.

ATP-mediated killing of intracellular mycobacteria is P2X₇ dependent

Having demonstrated that macrophages from P2X₇^−/− mice had no P2X₇-associated cytotoxic activity, they were then assessed for their ability to kill intracellular mycobacteria following ATP stimulation. BMDMs were infected with BCG and exposed to ATP (3 mM) for 2 h. Compared with WT cells, macrophages from P2X₇^−/− mice were shown to be ineffective at killing BCG (Fig. 3⇑b). The percentage of BCG killed was reduced from 50% in ATP-treated WT cells to <10% in P2X₇^−/− cells. Although some residual mycobactericidal activity was still occurring, the results show that P2X₇ is the primary transducer of the intracellular mycobactericidal effector pathway activated following ATP stimulation of infected macrophages.

PLD activity is essential for ATP/P2X₇-mediated BCG killing, but not macrophage death

PLD is involved in a number of signaling cascades via the production of phosphatidic acid from phosphorylcholine. If cells are pretreated with butan-1-ol, PLD preferentially uses the primary alcohol as a substrate producing the nonsignaling phosphatidyl butanol (30). Treatment with butan-1-ol after BCG infection, but before ATP stimulation, showed a concentration-dependent inhibition of BCG killing post-ATP (Fig. 4⇓a), but had no effect on macrophage death, as measured by chromium release (Fig. 4⇓b).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

a, Effect of PLD inhibition on BCG killing by J774 macrophages post-ATP stimulation. J774 macrophages were infected with BCG and then treated for 60 min with the indicated concentrations of butan-1-ol to inhibit PLD activity. They were then stimulated with 3 mM ATP for 2 h. BCG viability was assessed and expressed as percentage of BCG killed. Inhibition of PLD activity completely inhibits BCG killing. SDs for quadruplets are shown, and the results are representative of three experiments. b, Effect of PLD inhibition on macrophage death post-ATP stimulation. ⁵¹Cr-labeled macrophages were pretreated with the indicated concentrations of butan-1-ol for 60 min before 16-h stimulation with 3 mM ATP. Inhibition of PLD activity has no effect on macrophage death. SDs for quadruplets are shown, and the results are representative of three experiments.

Radical oxygen intermediates and radical nitrogen intermediates are not required for BCG killing post-ATP/P2X₇ stimulation

Experiments were performed to investigate whether the ATP-induced mycobactericidal effects observed in murine macrophages were attributable to the generation of reactive oxygen or reactive nitrogen intermediates.

The high levels of NO required for killing mycobacteria within murine macrophages are generated by the enzyme iNOS after appropriate priming and stimulation of the cells. Macrophages from iNOS^−/− mice are therefore unable to up-regulate nitrogen radical production (25). p47^phox is a key enzyme in the NADPH reductase pathway, and mice with a targeted disruption of this gene are unable to produce reactive oxygen radicals (24).

BMDMs from iNOS^−/− and p47^phox−/− mice and WT littermates were infected with BCG and treated with ATP (3 mM) for 2 h. The results (Fig. 5⇓) reveal that ATP was equally effective in inducing bacterial death within iNOS^−/− and p47^phox−/− macrophages as in WT cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Intracellular BCG killing by BMDMs does not require oxygen or nitrogen radicals. BMDMs were cultured from p47^phox−/−, iNOS^−/−, and WT littermates. They were infected with BCG before being stimulated with 3 mM ATP for 2 h. BCG viability was assessed by CFU. Results are expressed as percentage of BCG killed. Where indicated, 400 μM L-NMMA, 15 U/ml superoxide dismutase, or 0.1 mg/ml catalase was added 60 min before stimulation. SDs for quadruplets are shown, and the results are representative of two to three experiments. In all scenarios, the macrophages remained effective at killing BCG, indicating that neither nitrogen nor oxygen radicals are required.

To confirm that production of neither oxygen or nitrogen radicals was required for the ATP effect, nitrogen radical production was blocked in the p47^phox−/− macrophages using the NO inhibitor L-NMMA, and oxygen radical production was blocked in the iNOS^−/− macrophages using superoxide dismutase and catalase (Fig. 5⇑). ATP killing of BCG was shown to be equally as effective in all cell treatments. The data reveal that oxygen and nitrogen radicals do not have a role in the ATP effector mechanism.

Killing post-P2X₇ stimulation is not affected by Nramp1 genotype

The susceptibility of inbred mice strains to BCG is associated with a single point mutation of Nramp1 (31). Both the RAW and J774 macrophage cell lines are derived from BALB/c mice, which carry the susceptible mutation of Nramp1 (31).

To examine the possible influence of Nramp1 on the ATP-bacterial killing effect, RAW cells, stably transfected with either the resistant Nramp1 (RAW 7.5R) or the susceptible Nramp1 (RAW 10S), were used. ATP treatment of BCG-infected 7.5R and 10S cells was shown to be equally as effective in inducing killing of BCG, as was observed in the WT RAW cells (Fig. 6⇓). This indicates that expression of Nramp1 does not affect the ability of RAW cells to kill BCG in response to ATP stimulation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

ATP-stimulated killing of BCG by RAW macrophages transfected with resistant or susceptible Nramp1. The 7.5R and 10S RAW macrophages, which have been respectively transfected with the resistant and susceptible Nramp1 genes, were infected with BCG and stimulated with 3 mM ATP for 2 h. BCG viability was assessed by CFU and expressed as the percentage of BCG killed. Ability to kill post-ATP is independent of which Nramp1 gene has been transfected. SDs for quadruplets are shown, and the results are representative of three experiments.

Further support for this finding is provided by the observation that BMDMs from BALB/c, C57BL/6, and 129 mice were all equally effective at killing intracellular mycobacteria following ATP stimulation. BALB/c and C57BL/6 mice are known to be susceptible to BCG infection and carry the susceptible Nramp1 mutation, whereas 129 mice are a resistant strain and carry the resistant gene (31). In total, the data suggest that Nramp1 does not have a role in ATP-mediated killing of intracellular mycobacteria.

BCG-containing phagosomes fuse with lysosomes post-P2X₇ stimulation: confocal fluorescent microscopy studies

Mycobacteria survive within macrophages by inhibiting the maturation of their phagosomes, thus avoiding fusion with lysosomes (10). Lysosomes provide a potentially toxic environment for mycobacteria containing various hydrolases and cationic proteins in an acidic environment. The promotion of phagosome-lysosome fusion by ATP was therefore investigated as a potential mycobacterial-killing mechanism.

Confocal fluorescent microscopy was used to show that P2X₇ stimulation of infected BMDMs resulted in the rapid fusion of the BCG-containing phagosomes with lysosomes (Fig. 7⇓). The acidophilic dye LysoTracker (Molecular Probes) was used to prelabel the cell lysosomes red, and GFP-BCG were used to infect the BMDMs. When analyzed by confocal fluorescent microscopy, the BCG in non-ATP-treated cells (Fig. 7⇓, A and B) appeared green, indicating that they were residing in nonacidic vesicles separate from the red-staining lysosomes. However, in ATP-treated cells (Fig. 7⇓, C and D), an increasing percentage of the BCG appeared yellow (due to colocalization with the red LysoTracker), indicating that they were residing within acidic organelles that had fused with lysosomes. To quantify the colocalization of the BCG, serial observations were expressed as the percentage of colocalized BCG over total intracellular BCG (Fig. 8⇓). The results illustrate the progressive acidification of the BCG-containing phagosomes. These experiments show that BCG-containing phagosomes rapidly fuse with lysosomes following ATP stimulation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Confocal fluorescent microscopy images showing BCG and lysosome localization. Confocal fluorescent (A, C, and E) and transmission (B, D, and F) microscopy of macrophages infected with GFP-BCG and incubated with LysoTracker to stain lysosomes red. A and B, An unstimulated macrophage containing a green BCG separate from the red lysosomes. C and D, A macrophage that has been stimulated for 240 min with 3 mM ATP. The intracellular BCG now appear yellow, indicating that colocalization and phagosome-lysosome fusion have occurred. The extracellular BCG remains green. E and F, A P2X₇^−/− macrophage that has been stimulated with ATP. No colocalization has occurred. Representative pictures from at least three experiments are shown.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 8.

Colocalization between intracellular, BCG, and lysosomes following ATP stimulation. The results from experiments similar to those that gave the images shown in Fig. 7⇑ were quantified. BCG that appeared yellow were said to be colocalized and expressed as a percentage of total intracellular BCG. Over 200 BCG from at least 30 macrophages and three different experiments were assessed. SDs are shown. Increasing colocalization after ATP stimulation is demonstrated.

Experiments using UTP, GTP, and AMP failed to show phagosome-lysosome fusion (results not shown), suggesting that ATP was acting via the P2X₇ receptor. In addition, phagosome-lysosome fusion was absent in ATP-stimulated P2X₇^−/− macrophages (Fig. 7⇑, E and F). The data demonstrate that the phagosome-lysosome fusion that occurs following ATP stimulation is dependent on and is a function of P2X₇ activation.

BCG-containing phagosomes fuse to form acidified multibacillary vacuoles post-P2X₇ stimulation: EM studies

High resolution transmission EM was used to investigate the morphological changes accompanying the fusion of BCG-containing phagosomes with lysosomes following ATP stimulation. ATP-induced acidification of BCG-containing phagosomes was confirmed by the prestaining of cells with DAMP, an acidophilic dye that has a preferential affinity for acidic organelles such as lysosomes. DAMP staining was then detected using a gold-labeled anti-dinitrophenol Ab. Fig. 9⇓, A and B, illustrates a non-ATP-stimulated macrophage in which the BCG-containing phagosome has little to no gold labeling, and is therefore shown not to be acidic. This supports previous findings (10, 11, 32) that BCG normally reside within individual nonacidic phagosomes that do not fuse with lysosomes. Within 30 min of ATP stimulation, marked changes could be observed within infected macrophages (Fig. 9⇓, C and D). At low power (×5000), the BCG were observed to coalesce into multibacillary vacuoles (Fig. 9⇓C). These are heavily labeled with DAMP (Fig. 9⇓D), indicative of vacuole acidification, in contrast to the non-DAMP-labeled individual BCG-containing phagosomes seen in unstimulated cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 9.

EM studies of BMDMs infected with BCG and stimulated with ATP. BMDMs were cultured from P2X₇^−/− mice and WT littermates. They were infected with BCG and, where indicated, stimulated with ATP. Before fixation, cells were incubated with DAMP that localizes to acidic organelles. The DAMP was then gold labeled. A, C, and E, Magnification, ×5,000; B and F, ×50,000; and D, ×40,000. A and B, An unstimulated macrophage. BCG reside in individual phagosomes that are not gold labeled, indicating they are not acidic. C and D, A macrophage that has been stimulated with 3 mM ATP for 30 min. The BCG-containing phagosomes have coalesced, and the resulting vacuole shown in D is heavily gold labeled, indicating it is acidic. E and F, A macrophage from a P2X₇^−/− mouse after ATP stimulation for 30 min. The BCG-containing phagosomes have neither coalesced nor acidified. Representative pictures from at least three experiments are shown.

By counting the number of gold particles per vacuole, the amount of DAMP staining in the various phagosomes/vacuoles was quantified (Fig. 10⇓). Overall, the results confirm and illustrate the rapid acidification of phagosomes that occurs post-ATP stimulation. In addition, the EM studies illustrate that this involves the fusion of individual BCG-containing phagosomes both with each other and with lysosomes.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 10.

Acidity of BCG-containing vacuoles in ATP-stimulated J774 macrophages. Electron micrographs were prepared as in Fig. 9⇑. BCG-infected J774 macrophages were stimulated with 3 mM ATP for the indicated times. The number of gold particles per BCG-containing phagosome or vacuole is proportional to its acidity.

To confirm that the formation of acidified multibacillary vacuoles was specifically due to ATP stimulation of P2X₇ receptors, the EM study was repeated using cells derived from P2X₇^−/− mice. Fig. 9⇑, E and F, illustrates that at 2 h post-ATP stimulation of P2X₇^−/− BMDMs, the BCG-containing phagosomes have neither acidified nor coalesced.