ATP Stimulates Human Macrophages to Kill Intracellular Virulent Mycobacterium tuberculosis Via Calcium-Dependent Phagosome-Lysosome Fusion

Materials and bacteria

Unless noted, all materials were from previously published sources (6, 9). The H37Rv and Erdman strains of M. tuberculosis were obtained from the American Tissue Type Culture Collection (Manassas, VA). The CDC1551 M. tuberculosis strain was generously provided by Dr. T. Shinnick (Centers for Disease Control and Prevention, Atlanta, GA). Virulent strains of M. tuberculosis (H37Rv, Erdman, and CDC1551) were cultured and prepared for use in experiments as noted previously (17). Briefly, aliquots of frozen M. tuberculosis stocks in 7H9 broth were thawed, cultured for 9 days on 7H11 agar at 37°C in 5% CO₂-95% air, scraped from agar plates, and suspended in RPMI 1640 by vortexing briefly. After settling, the supernatant was transferred to a new tube and allowed to settle once again. An aliquot of this final M. tuberculosis suspension was counted in a Petroff-Hauser chamber, and the concentration of bacteria was adjusted for use in experiments. M. tuberculosis preparations contained ≥90% single bacteria, with ≥80% viability by determination of CFUs (17). The effects of various experimental manipulations on the viability of M. tuberculosis were determined by analysis of CFUs.

Preparation of Mφ

PBMC were isolated from healthy, purified protein derivative-negative adult volunteers and were cultured in Teflon wells for 5 days in RPMI 1640 (pH 7.4) with 20% fresh autologous serum as previously described (17). Mφ (∼2 × 10⁶/sample) were purified by adherence to six-well plastic tissue-culture plates or chromic acid-cleaned glass coverslips (25 mm in diameter) for 2 h at 37°C in 5% CO₂. Monolayers were washed repeatedly and then incubated in RPMI 1640 with 2.5% autologous serum, without antibiotics, for use in experiments. Effects of experimental manipulations on Mφ viability were assessed by exclusion of trypan blue, and monolayer density was determined by nuclei counting with napthol blue-black stain (17, 18).

Infection of Mφ with M. tuberculosis and analysis of intracellular survival

Monocyte-derived Mφ in RPMI 1640, 20 mM HEPES, and 2.5% human serum were infected with M. tuberculosis at a bacteria:Mφ ratio (multiplicity of infection, MOI) of 1:1 and then incubated for 1 h at 37°C. Following infection, monolayers were washed three times with RPMI 1640 at 37°C and incubated with RPMI 1640 and 10% FBS for 24 h, before addition of ATP or buffer control. The MOI of 1:1 was used because it permits long-term cultivation of infected human Mφ and the determination of intracellular viability of M. tuberculosis at ≥24 h following infection. We have previously used higher levels of MOI, including 10:1, 30:1, and 100:1, but the resultant rapid lysis of the Mφ monolayer precludes accurate determination of the bacterium’s intracellular survival (9). Mφ cultures were incubated at 37°C for an additional 24 h after the addition of ATP or buffer, followed by quantitation of the growth of M. tuberculosis by determination of CFUs or by the Bactec method, as previously described (6). Because primary human Mφ exhibit a wide variance in the number of M. tuberculosis that grow intracellularly, the results of the CFU assay are generally presented as cumulative data (mean ± range) expressed as a percentage of the paired control samples. The absolute number of M. tuberculosis CFUs in the control samples has been designated in the figure legends. Determinations of mycobacterial viability by the Bactec method were in excellent agreement with the results of the CFU assay (6, 19).

Determination of Mφ [Ca²⁺]_c and modulation of Ca²⁺ levels

Calcium measurements were performed as previously described (9). Briefly, Mφ were adhered to collagen-coated glass coverslips and incubated in 10 μM fura 2-acetylmethyl ester (fura 2) in HBSS for 30 min at 37°C. Levels of [Ca²⁺]_c in single Mφ, or the mean [Ca²⁺]_c of groups of 10–20 cells, was determined using a Photoscan II spectrofluorometer (Photon Technology International, Lawrenceville, NJ). [Ca²⁺]_c was determined from the ratio of fluorescence emission intensities at 510 nm following excitation at 340 and 380 nm, respectively, as previously described (20). To analyze the contribution of [Ca²⁺]_c-mediated signaling pathways to ATP-dependent mycobactericidal activity, both extracellular and intracellular [Ca²⁺]_c were experimentally modified. Extracellular Ca²⁺ was depleted by washing Mφ twice in Ca²⁺-free buffer containing 5 mM EGTA, followed by resuspension in the same buffer and incubation for 5 min at 37°C before addition of ATP. Inhibition of changes in [Ca²⁺]_c were also produced by preincubation of Mφ with the intracellular Ca²⁺ chelator, bis-(2-amino-S-methylphenoxy)ethane-N,N,N′,N′-tetraacetic acid tetraacetoxymethyl ester (MAPTAM) (25 μM), in EGTA-containing buffer for 20 min at 37°C. The effects of EGTA- and MAPTAM-induced alterations in [Ca²⁺]_c were verified in parallel experiments by direct determination of [Ca²⁺]_c in control and treated Mφ via fluorescence of fura 2 (20). Levels of [Ca²⁺]_c were <20 nM in MAPTAM-treated cells following addition of ATP or control Ca²⁺-mobilizing agonists, platelet-activating factor (100 nM) or complement-opsonized zymosan, demonstrating the efficacy of MAPTAM-induced [Ca²⁺]_c buffering (9, 21). The effects of these modulators of extracellular and intracellular Ca²⁺ on Mφ viability and monolayer density were determined in parallel experiments, and no significant differences from control cells were observed. EGTA and MAPTAM did not alter the viability of M. tuberculosis, either when incubated directly with mycobacteria in 7H9 medium for 24 h at 37°C or when added to infected Mφ in the absence of ATP.

Confocal microscopy

The degree of maturation of phagosomes containing live or killed M. tuberculosis was assessed by colocalization of the bacilli with the acidophilic dye LysoTracker Red (Molecular Probes, Eugene, OR) and the lysosomal protein markers cathepsin D, CD63, and lysosome-associated membrane protein (LAMP)-1, as described previously (9, 22). LysoTracker Red was incubated at a 1/10,000 dilution with Mφ monolayers in RPMI 1640, 20 mM HEPES, and 2.5% autologous serum for 2 h at 37°C. Unincorporated dye was removed by washing, followed by infection with M. tuberculosis. After removal of nonadherent bacilli, LysoTracker Red was added to each well at the same concentration used for initial labeling. At 30 min postinfection, Mφ were fixed in 3.75% paraformaldehyde for 15 min and permeabilized with ice-cold methanol-acetone (1:1). Detection of the lysosomal protein markers LAMP-1, cathepsin D, or CD63 was accomplished by incubating coverslips with blocking buffer (PBS, 5% BSA, and 10% goat serum) for 1 h, followed by the appropriate 1° Abs (diluted in blocking buffer) for 1 h, repeated washings, and incubation with Texas Red-conjugated secondary anti-IgG Ab for 1 h, all at 25°C. The localization of M. tuberculosis was determined by incubating monolayers with auramine for 20 min at 25°C, followed by a 3-min incubation in 0.5% acid alcohol. Following repeated washings, coverslips were mounted with buffered glycerol solution and nail polish.

Confocal microscopy was performed on a Zeiss Laser Scan inverted 510 microscope (Zeiss, Oberkochen, Germany). An argon-krypton laser (excitation, 488 nm; emission band pass, 505–530 nm) was used for detection of auramine fluorescence, and a helium-neon laser (excitation, 543 nm; emission limit of pass, 585 nm) was used for detection of Texas Red and LysoTracker Red. The percentage of M. tuberculosis phagosomes colocalizing with the marker of interest was determined by counting 25 phagosomes from each sample. The effects of modulation of Mφ [Ca⁺²]_c on the response to ATP_e was performed by preincubation with EGTA or MAPTAM, respectively. Neither EGTA nor MAPTAM directly affected the fluorescence of auramine, LysoTracker Red, or Texas Red (9). At a MOI of 1:1, <10% of Mφ contained two or more bacteria. Therefore, confocal images of Mφ containing single bacilli are shown as most representative.

For quantitative analysis of phagosomal maturation, the range of fluorescence intensity of the Texas Red 2° Ab or LysoTracker Red was recorded along the major axis of the bacillus (defined by auramine staining). Samples in which the mean fluorescent intensity of the given lysosomal marker was greater than the mean of control, uninfected Mφ were scored as positive. For a given phagosome to be scored as a mature phagolysosome, each of the four lysosomal markers (LAMP-1, cathepsin D, CD63, and LysoTracker Red) must have all been positive. Because the confocal measurements were acquired in a blinded fashion, inherent limitations of the technique (e.g., the lack of a single, fully specific marker of Mφ lysosomes) should be distributed equally among the experimental groups. The use of this technique by our laboratory (9, 22) and others (23, 24, 25) has resulted in strong interobserver correlations, as well as excellent agreement with the results of immunoelectron microscopy (25, 26, 27).

Data analysis

Data from each experimental group were subjected to an analysis of normality and variance. Differences between experimental groups composed of normally distributed data were analyzed for statistical significance using Student’s t test. Nonparametric evaluation of other data sets was performed with the Mann-Whitney rank sum test. Analysis of correlation was performed with the Spearman rank order test (28).

ATP-induced killing of virulent M. tuberculosis within human Mφ requires increases in [Ca²⁺]_c

To characterize the biochemical determinants of ATP_e-stimulated killing of intracellular M. tuberculosis, we used an in vitro model system in which blood monocytes are differentiated to a Mφ phenotype by cultivation for 5 days, followed by purification via adherence to tissue culture plates (6, 17, 22). Mφ were infected with the virulent H37Rv strain of M. tuberculosis, at a MOI of 1:1 for 24 h, followed by incubation with 3 mM ATP. Treatment with ATP_e reduced the viability of intracellular M. tuberculosis by 83 ± 7% at 24 h and by 3.62 ± 0.41 logs at day 7 compared with control samples treated with buffer alone (p < 0.001 for each time point, n = 9) (6).

To begin to evaluate the Ca²⁺-dependence of this ATP_e-induced bactericidal activity, we determined the effect of ATP on levels of [Ca²⁺]_c in M. tuberculosis-infected Mφ labeled with the fluorescent Ca²⁺ indicator fura 2. ATP_e stimulated a rapid and sustained increase in [Ca²⁺]_c in infected Mφ that was indistinguishable from the [Ca²⁺]_c response of uninfected cells (Fig. 1⇓A) (29, 30, 31). This ATP-induced Ca²⁺ response was due to stimulation of two classes of Mφ cell surface purinergic receptors, P₂Y₂, G protein-coupled receptors and P₂X₇, pore-forming receptors, because the P₂Y₂-selective agonist, UTP, produced the initial rapid increase, but not the prolonged elevation, in [Ca²⁺]_c (Fig. 1⇓B), whereas the P₂X₇-selective agonist 3′-O-(4-benzoyl)benzoyl-ATP, produced a delayed but sustained increase in [Ca²⁺]_c in the absence of the initial rapid response (Fig. 1⇓C) (30). To determine the source of the increase in [Ca²⁺]_c following stimulation by ATP_e, we used the extracellular Ca²⁺ chelator EGTA as well as the intracellular Ca²⁺ buffer MAPTAM (9). Incubation of Mφ in Ca²⁺-free medium containing 3 mM EGTA significantly decreased the magnitude and duration of the ATP_e-induced increase in [Ca²⁺]_c (Fig. 1⇓D). Addition of the intracellular Ca²⁺ chelator MAPTAM (25 μM) to Mφ in EGTA-containing medium completely eliminated any change in [Ca²⁺]_c in response to ATP_e (Fig. 1⇓D).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

ATP-induced increases in [Ca²⁺]_c in M. tuberculosis-infected Mφ are due to stimulation of P₂Y₂ and P₂X₇ receptors and are derived from both intra- and extracellular sources of Ca²⁺. A, Human Mφ infected with H37Rv M. tuberculosis for 24 h (iMP) and uninfected control Mφ (uMP) were labeled with the Ca²⁺-sensitive fluorescent indicator fura 2. Following removal of unincorporated dye, Mφ were incubated with buffer control (Ctr) or 3 mM ATP (↓), and levels of [Ca²⁺]_c were monitored via fura 2 fluorescence. Infected Mφ were incubated with 1 mM UTP (B) or 300 μM 3′-O-(-benzoyl)benzoyl-ATP (C), and [Ca²⁺]_c was determined as in A. D, Infected Mφ were incubated in buffer control, 5 mM EGTA, or 25 μM MAPTAM and 3 mM EGTA for 20 min before addition of 3 mM ATP and determination of [Ca²⁺]_c. Data are representative of results from seven identical experiments.

In parallel experiments, the effects of EGTA with or without MAPTAM on ATP_e-induced killing of intracellular H37Rv M. tuberculosis were analyzed by determination of CFUs in detergent extracts of infected Mφ 24 h after addition of ATP. EGTA treatment reduced the level of mycobacterial killing by 83 ± 6% (Fig. 2⇓A, p < 0.001 and n = 7). Mφ incubated with MAPTAM and EGTA (which exhibit no elevation of [Ca²⁺]_c in response to ATP_e) demonstrated complete inhibition of ATP-induced bactericidal activity toward H37Rv M. tuberculosis (Fig. 2⇓A, p < 0.001 and n = 7). To verify the results of the CFU assay for mycobacterial viability, we used a complimentary Bactec method based on mycobacterial-specific oxidation of [¹⁴C]palmitate to [¹⁴C]CO₂. These studies confirmed the Ca²⁺-dependence of ATP_e-induced reductions in the intracellular viability of H37Rv M. tuberculosis (Fig. 2⇓B).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Inhibition of ATP-induced increases in Mφ [Ca²⁺]_c blocks killing of intracellular M. tuberculosis. A, Mφ were infected with H37Rv M. tuberculosis for 24 h, as described in the Fig. 1⇑ legend. Samples were incubated with buffer (control, Ctr), 3 mM ATP (ATP), 3 mM EGTA followed 20 min later by ATP (E + ATP), or 25 μM MAPTAM and 3 mM EGTA for 20 min followed by ATP (M + ATP). The viability of intracellular M. tuberculosis was determined 24 h later by plating lysates of infected Mφ on 7H11 agar and quantitation of CFUs. Data are expressed as the percentage of CFUs in each sample, relative to that of control samples (mean ± range). Control Mφ yielded 1.34 ± 0.12 × 10⁶ CFUs (mean ± SEM). B, Mφ were infected with the indicated virulent strains of M. tuberculosis, H37Rv, Erdman, or CDC1551 for 24 h. Samples were incubated with buffer control (▪), 3 mM ATP (□), or 25 μM MAPTAM plus 3 mM EGTA followed 20 min later by 3 mM ATP (▨). Viability of each strain was determined by the Bactec method and expressed as the growth index (mean ± SEM). Data in A and B represent results from five identical experiments, each performed in triplicate.

The inhibitory effect of Ca²⁺ chelation on ATP-induced killing of M. tuberculosis was reversed by increasing the concentration of extracellular Ca²⁺ to exceed that buffered by EGTA and MAPTAM (data not shown). Reversal of microbicidal activity by excess CaCl₂ supports the specificity of the Ca²⁺ chelators and strengthens the conclusion that Ca²⁺ is required for ATP-induced mycobacterial killing. Finally, the killing of two other well-characterized, virulent strains of M. tuberculosis, Erdman (26, 27) and CDC1551 (32), by ATP_e exhibited a similar Ca²⁺ dependence (Fig. 2⇑B, p < 0.001 for each strain and n = 5). Taken together, these data demonstrate that ATP_e-induced killing of virulent M. tuberculosis within human Mφ requires an increase in [Ca²⁺]_c. Because specific stimulation of Mφ P₂Y₂ receptors via UTP does not result in killing of intracellular M. tuberculosis (6), these data are consistent with a requirement for P₂X₇-mediated influx of extracellular Ca²⁺ for tuberculocidal activity.

ATP induces Ca²⁺-dependent maturation of M. tuberculosis phagosomes

The specific antimicrobial mechanism(s) by which ATP_e induces intracellular mycobacterial killing are unknown (6, 8). Because Ca²⁺ regulates several phagocyte antimicrobial responses including generation of reactive oxygen species (33) and reactive nitrogen intermediates (34, 35), granule secretion (36), synthesis of cytokines (37), and, in certain cases, phagosome-lysosome (P-L) fusion (9), the Ca²⁺-dependence of ATP_e-induced killing of M. tuberculosis may be due to multiple effects. We tested the hypothesis that increases in [Ca²⁺]_c promote ATP_e-induced killing of M. tuberculosis via enhancement of P-L fusion based on the following rationale: 1) During initial infection of naive human Mφ by M. tuberculosis, mycobacterial inhibition of Mφ Ca²⁺ signaling is tightly coupled to inhibition of P-L fusion and promotion of intracellular bacterial viability (9); 2) pharmacologic reversal of M. tuberculosis-induced inhibition of Mφ Ca²⁺ signaling results in increased levels of both P-L fusion and mycobacterial killing (9); and 3) ATP_e-stimulated killing of BCG does not involve reactive oxygen or reactive nitrogen intermediates species and has not been linked to synthesis or secretion of inflammatory mediators by infected Mφ (8). Concerning the first two points above, we recognize that biochemical mechanisms that regulate intracellular viability of M. tuberculosis during primary infection of naive Mφ may be distinct from those mechanisms involved in ATP_e-mediated reductions in mycobacterial viability following established infection of Mφ.

To test the hypothesis that the Ca²⁺ requirement in ATP_e-induced mycobacterial killing is due its promotion of P-L fusion, we determined levels of phagosomal maturation in control Mφ as well as in those treated with ATP with or without MAPTAM/EGTA. The extent of maturation of M. tuberculosis-containing phagosomes was quantitated by laser scanning confocal microscopy using three protein markers of lysosomes/late endosomes (CD63, cathepsin D, and LAMP-1) and the acidophilic dye LysoTracker Red (9). Although there is no single, unambiguous marker for lysosomes, the relative accumulation of several distinct proteins (both soluble and membrane associated) and pH-sensitive dyes provides accurate determination of the identity of vesicular compartments (22, 25, 26, 27, 29, 38). As previously demonstrated, phagosomes containing live H37Rv M. tuberculosis exhibited low levels of the lysosomal markers, consistent with an immature maturational state, i.e., inhibition of P-L fusion (Fig. 3⇓). Incubation with ATP for 30 min resulted in marked increases in each of the lysosomal markers, consistent with promotion of P-L fusion. Inhibition of ATP_e-induced increases in [Ca²⁺]_c with MAPTAM/EGTA resulted in a complete reversal of ATP-stimulated phagosomal maturation (Fig. 3⇓). In fact, the level of lysosomal markers on M. tuberculosis phagosomes in Mφ incubated with MAPTAM/EGTA followed by ATP was lower than that present in control Mφ. These data support the novel hypotheses that ATP_e-induced antituberculous activity involves the promotion of P-L fusion, and the obligatory role of Ca²⁺ in ATP_e-stimulated mycobacterial killing is due, at least in part, to the Ca²⁺ dependence of P-L fusion.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

ATP induces Ca²⁺-dependent fusion of M. tuberculosis P-L. Mφ monolayers were infected with H37Rv M. tuberculosis at an MOI of 1:1 for 24 h. Nonadherent bacilli were removed by washing, and samples were incubated with buffer control, 3 mM ATP, or 25 μM MAPTAM plus 3 mM EGTA for 20 min followed by ATP. Thirty minutes after ATP addition, cells were fixed, permeabilized, and stained for lysosomal markers and M. tuberculosis, as described in Materials and Methods. A, Data from a single experiment using Ab to the lysosomal protein CD63. B, Cumulative data for all four lysosomal markers from 25 phagosomes/sample from six identical experiments. Results are expressed as the percentage (mean ± range) of M. tuberculosis-containing phagosomes that stained positively for LysoTracker Red (Lyso), CD63, cathepsin D (Cath D), or LAMP-1.M + ATP refers to samples incubated in 25 μM MAPTAM and 3 mM EGTA for 20 min followed by ATP.

A quantitative assessment of the strength of the associations between three variables, 1) ATP_e-induced elevation of [Ca²⁺]_c, 2) P-L fusion, and 3) reductions in intracellular viability of M. tuberculosis, was made via an analysis of correlation using the Spearman rank order correlation test. The strength of the proposed association was determined via an analysis of correlation rather than regression, because the former makes no assumptions about the directionality between variables and is therefore more rigorous in complex, multivariate situations in which unambiguous designations of “independent” and “dependent” variables cannot be assigned (28). The level of [Ca²⁺]_c in ATP-stimulated Mφ was highly correlated with P-L fusion (defined as presence of all four lysosomal markers; r = 0.916, p < 0.001, and n = 9). Similarly, the degree of P-L fusion correlated highly with reductions in M. tuberculosis viability (r = 0.926 and p < 0.001). Finally, ATP_e-induced levels of [Ca²⁺]_c correlated highly with reductions in mycobacterial viability (r = 0.942 and p < 0.001). Although these high degrees of correlation between each of the variables do not prove a causal connection, they strongly support our hypothesis.

ATP-induced killing of intracellular M. tuberculosis exhibits distinct requirements for both Ca²⁺ and PLD activity

We have recently demonstrated that ATP_e-stimulated killing of M. tuberculosis within human Mφ is tightly coupled to activation of the PLD signal transduction pathway (6). Therefore, we sought to determine whether the requirements for increases in [Ca²⁺]_c and PLD activity were independent functions of distinct signaling pathways or whether they exhibited some degree of interdependence. The importance of this distinction arises from the fact that both Ca²⁺-dependent PLD activities as well as PLD-dependent increases in [Ca²⁺]_c have been previously demonstrated (10, 39).

To determine whether the ATP_e-induced increases in Mφ [Ca²⁺]_c require stimulation of PLD activity, we used two well-characterized, mechanistically distinct inhibitors of PLD activity, 2,3-diphosphoglycerate (2,3-DPG) and ethanol. Although no specific pharmacologic inhibitor of PLD has been identified, the combined use of these complementary inhibitors has yielded substantial insights into the role of PLD in diverse physiologic functions (6, 17). 2,3-DPG is a competitive inhibitor of PLD that exhibits low toxicity to intact cells (17, 40). Preincubation of Mφ monolayers with 2,3-DPG (0.1–3 mM) for 15 min at 37°C produced dose-dependent inhibition of ATP-stimulated PLD activity (6) with a maximal reduction of 91% (range 87–93% reduction, p < 0.001 and n = 5) at 5 mM 2,3-DPG. Despite this marked inhibition of ATP_e-stimulated PLD activity, 2,3-DPG had no detectable effect on the accompanying increase in [Ca²⁺]_c (Fig. 4⇓A).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Relationship between ATP-stimulated increases in [Ca²⁺]_c and PLD activity. A, Mφ infected with H37Rv M. tuberculosis were loaded with fura 2 and then incubated in buffer control, 2,3-DPG (5 mM), or 1.0% ethanol for 15 min before addition of 3 mM ATP. Levels of Mφ [Ca²⁺]_c were determined by the fluorescence of fura 2. The tracings are representative of results from five identical experiments for each condition. B, [³H]Oleic acid-labeled infected Mφ were incubated in 3 mM EGTA and sufficient CaCl₂ to yield the indicated concentrations of extracellular Ca²⁺. PLD activity was determined 15 min after addition of 3 mM ATP by quantitation of [³H]PEt in the presence of 0.5% ethanol. PLD activity is expressed as cpm in [³H]PEt per 10⁵ total cpm in phospholipid. Data represent the mean ± SEM of five identical experiments performed in triplicate.

Ethanol (and other short-chain primary alcohols) inhibit PLD-mediated generation of phosphatidic acid by substituting for water as the attacking nucleophile, yielding the metabolically inert alternate product, phosphatidylethanol (PEt) (17, 41). Thus, ethanol inhibits PLD-dependent responses without blocking enzyme turnover, providing a complementary method of PLD inhibition that is mechanistically distinct from that of 2,3-DPG. Incubation of Mφ with ethanol (0.03–1.0%, v/v) for 2 min at 37°C before addition of ATP resulted in concentration-dependent inhibition of phosphatidic acid generation (6) with a 93% reduction produced by addition of 1.0% ethanol (between 89 and 95% reduction, p < 0.001 and n = 5). Similar to the results with 2,3-DPG, ethanol had no significant effect on ATP_e-stimulated increases in Mφ [Ca²⁺]_c (Fig. 4⇑A). At the concentrations used, 2,3-DPG and ethanol had no effect on Mφ viability or monolayer density (data not shown). Taken together, the results with 2,3-DPG and ethanol demonstrate that ATP_e-induced increases in Mφ [Ca²⁺]_c are independent of the accompanying stimulation of PLD activity.

To evaluate whether ATP-stimulated increases in PLD activity were dependent on [Ca²⁺]_c, Mφ were radiolabeled with [³H]oleate for 18 h at 37°C. Following removal of unincorporated label, Mφ were incubated in various concentrations of extracellular Ca²⁺ (buffered with EGTA) before addition of 3 mM ATP. PLD activity was determined by accumulation of the PLD-specific product PEt in the presence of 0.5% ethanol. In Mφ incubated in 25 μM MAPTAM/3 mM EGTA, ATP_e stimulated significant levels of PLD activity (Fig. 4⇑B). Increases in extracellular [Ca²⁺]_c resulted in progressive increases in ATP_e-stimulated PLD activity. Thus, Ca²⁺ was not required for activation of PLD but, rather, augmented the level of PLD activity stimulated by ATP_e. These data on the relationship between Ca²⁺ and PLD activity in M. tuberculosis-infected human Mφ are in close agreement with previous work by el-Moatassim and Dubyak in uninfected murine Mφ (30, 42).

Our previous report (6) and the current data strongly suggest that ATP_e-induced killing of intracellular M. tuberculosis requires both elevation of [Ca²⁺]_c and activation of PLD. Furthermore, ATP_e-stimulated bactericidal activity is tightly coupled to [Ca²⁺]_c-dependent promotion of P-L fusion. Thus, we hypothesize that activation of PLD may also be required for maturation of the M. tuberculosis-containing phagosomes to phagolysosomes. To test this hypothesis, levels of P-L fusion were determined in Mφ in which PLD activity was inhibited by 2,3-DPG, ethanol, or MAPTAM/EGTA. In parallel experiments, the viability of intracellular M. tuberculosis was determined to assess its correlation with the extent of P-L fusion. We have previously reported the dose-dependent inhibition of ATP_e-stimulated PLD activity by 2,3-DPG and ethanol (6). In the current experiments, specific concentrations of these inhibitors were selected that either maximally inhibited ATP_e-induced PLD activity (5 mM 2,3-DPG, 1% ethanol) or inhibited PLD activity to the same extent as 25 μM MAPTAM/3 mM EGTA (2 mM 2,3-DPG, 0.5% ethanol), i.e., ∼40% of the maximum (Fig. 4⇑B).

Inhibition of ATP_e-stimulated PLD activity (Fig. 5⇓A) was accompanied by concordant reductions in P-L fusion (Fig. 5⇓B) and inversely proportional increases in M. tuberculosis viability (Fig. 5⇓C). These data support the hypothesis that ATP_e-induced stimulation of PLD activity is coupled to promotion of P-L fusion and reductions in intracellular viability of M. tuberculosis. Of note, incomplete inhibition of PLD activity by 2 mM 2,3-DPG or 0.5% ethanol (to the level of PLD activity present in MAPTAM-treated Mφ) resulted in partial inhibition of P-L fusion (Fig. 5⇓B) and partial restoration of mycobacterial viability (Fig. 5⇓C). These results contrast with the complete inhibition of ATP_e-stimulated P-L fusion and mycobacterial killing in MAPTAM-treated Mφ, in which [Ca²⁺]_c was completely inhibited. Taken together, these data support the hypothesis that increases in PLD activity and [Ca²⁺]_c make distinct and experimentally separable contributions to ATP-induced P-L fusion and mycobacterial killing.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

[Ca²⁺]_c and PLD make distinct contributions to P-L fusion and reduction of the intracellular survival of M. tuberculosis. A, [³H]Oleic acid-labeled Mφ were infected with H37Rv M. tuberculosis for 24 h, then incubated under the following conditions: buffer (Ctr and ATP), 25 μM MAPTAM/3 mM EGTA (M); 2 mM (D2) or 5 mM (D5) 2,3-DPG; or 0.5% (E.5) or 0.1% (E1) ethanol. PLD activity was determined 15 min after addition of buffer (Ctr) or 3 mM ATP (all samples except Ctr) via quantitation of [³H]PEt. Results were normalized for total ³H cpm in phospholipid (mean ± SEM). B, Mφ infected with H37Rv M. tuberculosis were incubated under conditions identical with those in A. Thirty minutes after addition of buffer or ATP, cells were fixed, permeabilized, and stained for lysosomal markers and M. tuberculosis. Results are expressed as the percentage of phagosomes that were positive for all four lysosomal markers (mean ± range). C, Mφ were infected with H37Rv M. tuberculosis and incubated, in parallel, with samples described in A and B above. Following addition of ATP or buffer control, samples were incubated for an additional 24 h at 37°C, followed by determination of the viability of intracellular M. tuberculosis via the Bactec method (mean ± SEM). All data are from five identical experiments.

Note added in proof.

During revision of this manuscript, Lammas and colleagues (8) reported, in contrast to their earlier findings, that ATP-induced killing of BCG is Ca²⁺ dependent and is associated with phagosomal acidification (46).