The subversion of microbicidal functions of macrophages by intracellular pathogens is critical for their survival and pathogenicity. The replication of Coxiella burnetii, the agent of Q fever, in acidic phagolysosomes of nonphagocytic cells has been considered as a paradigm of intracellular life of bacteria. We show in this study that C. burnetii survival in THP-1 monocytes was not related to phagosomal pH because bacterial vacuoles were acidic independently of C. burnetii virulence. In contrast, virulent C. burnetii escapes killing in resting THP-1 cells by preventing phagosome maturation. Indeed, C. burnetii vacuoles did not fuse with lysosomes because they were devoid of cathepsin D, and did not accumulate lysosomal trackers; the acquisition of markers of late endosomes and late endosomes-early lysosomes was conserved. In contrast, avirulent variants of C. burnetii were eliminated by monocytes and their vacuoles accumulated late endosomal and lysosomal markers. The fate of virulent C. burnetii in THP-1 monocytes depends on cell activation. Monocyte activation by IFN-γ restored C. burnetii killing and phagosome maturation as assessed by colocalization of C. burnetii with active cathepsin D. In addition, when IFN-γ was added before cell infection, it was able to stimulate C. burnetii killing but it also induced vacuolar alkalinization. These findings suggest that IFN-γ mediates C. burnetii killing via two distinct mechanisms, phagosome maturation, and phagosome alkalinization. Thus, the tuning of vacuole biogenesis is likely a key part of C. burnetii survival and the pathophysiology of Q fever.

Coxiellaburnetii, an obligate intracellular Gram-negative bacterium classified in the γ subdivision of proteobacteria, is the agent of Q fever (1). Whereas acute Q fever is controlled by cell-mediated immunity, this latter is defective in chronic Q fever (2). The establishment of C. burnetii infection is based on a specific strategy of invasion of monocytes/macrophages. Virulent organisms are poorly internalized by macrophages and their uptake requires the engagement of αvβ3 integrin; avirulent variants are efficiently internalized through αvβ3 integrin and complement receptor type 3. The selective use of phagocytic receptors is an active process based on interference with complement receptor type 3-mediated phagocytosis (3). Once internalized, C. burnetii survives and replicates in an acidic environment (4, 5), which is needed for bacterial metabolism (6, 7). This low pH also accounts for the relative inefficiency of antibiotics toward C. burnetii (1). The association of acidic pH and phagolysosomal features has led most authors to consider the intracellular life of C. burnetii as a paradigm of intracellular survival without alteration of intracellular traffic (5). However, most of these studies were performed with avirulent C. burnetii (8, 9). In addition, these studies used fibroblasts and murine macrophage-like cells, in which virulent and avirulent C. burnetii replicate (5), whereas only virulent C. burnetii survives in human monocytes/macrophages (3).

The fate of intracellular microorganisms including C. burnetii depends on the microbicidal properties of macrophages and their regulation by cytokines. A defective killing of C. burnetii was found in monocytes from patients with chronic Q fever (10), which partly results from IL-10-mediated impairment of macrophage microbicidal activity. Indeed, IL-10 elicits the replication of C. burnetii in resting monocytes, and neutralizing anti-IL-10 Abs restore microbicidal activity against C. burnetii in patients with chronic Q fever (11). In contrast, IFN-γ, known to stimulate the microbicidal activity of macrophages, triggers C. burnetii killing in THP-1 monocytes (12). The ability of IFN-γ to stimulate the microbicidal activity of macrophages has been related to oxygen-dependent mechanisms (13), but reactive oxygen intermediates are not involved in the killing of C. burnetii (12). As IFN-γ-induced killing of Listeria monocytogenes and Mycobacterium avium has been associated with the modulation of phagosome maturation (14, 15, 16), we hypothesized that such mechanisms may be involved in C. burnetii killing.

We show in this study that the survival of C. burnetii in THP-1 monocytes is associated with altered phagosome maturation. C. burnetii organisms are present in phagosomes that acquire markers of late endosomes and late endosomes-early lysosomes but not the lysosomal enzyme cathepsin D. The survival of C. burnetii depends on the activation of THP-1 cells. Indeed, IFN-γ induces C. burnetii killing and restores phagolysosomal fusion. This study also provides evidence that IFN-γ-induced killing of C. burnetii involves two distinct mechanisms, phagosome maturation and late phagosome alkalinization.

THP-1 monocytic cells were cultured as previously described (12). Cells (5 × 104 cells/assay) were seeded on 12-mm round coverslips in flat-bottom 24-well plates (Nunc, Roskilde, Denmark) and were treated with 10 ng/ml PMA (Sigma-Aldrich, St. Louis, MO) to become adherent. After 24 h at 37°C, cells were washed three times in antibiotic-free RPMI 1640 supplemented with 10% FCS and 2 mM l-glutamine (Invitrogen, Eragny France). PBMC were isolated from healthy volunteers on Ficoll gradient (MSL, Eurobio, Les Ullis, France), and monocytes were purified by adherence on glass Labtek chamber/slides (Miles, Naperville IL), as previously described (11). Nonadherent cells were removed by washing, and remaining cells were cultured for 3 days in RPMI 1640 supplemented with 10% FCS and 2 mM l-glutamine. Virulent and avirulent C. burnetii organisms (Nine Mile strain, ATCC VR-615; American Type Culture Collection, Manassas, VA) were obtained as previously described (3). In brief, virulent organisms were isolated from infected mice and cultured in L929 cells for two passages whereas avirulent variants were cultured in L929 cells by repeated passages. Two other virulent strains of C. burnetii, Priscilla and Q212, isolated from an infected goat and a patient with acute Q fever, respectively, were cultured like the virulent Nine Mile strain. Bacteria were layered on 25–45% linear Renograffin gradient, and the gradients were spun down. Purified bacteria were then collected, washed, and suspended in HBSS before being stored at −80°C. The number of bacteria was determined by Gimenez staining. The viability of C. burnetii was assessed using the LIVE/DEAD BacLight bacterial viability kit (Molecular Probes, Eugene, OR) as recommended by the manufacturer. Briefly, the C. burnetii suspension was incubated with SYTO 9 stain and propidium iodide, and examined with a fluorescence microscope. Results are expressed as the ratio of viable bacteria and the total number of bacteria. Only C. burnetii preparations containing >90% of viable organisms were used. Heat-killed virulent organisms were obtained by heating the bacterial suspension at 100°C for 1 h, and were stored at −80°C.

THP-1 cells and monocytes from healthy donors were incubated with C. burnetii in antibiotic-free RPMI 1640 containing 10% FCS. After 24 h at 37°C, cells were washed to remove free bacteria (this time was designated as day 0): this procedure was sufficient to remove noninternalized and loosely attached organisms (3). Monocytic cells were again cultured for different periods. In some experiments, THP-1 cells were incubated with human rIFN-γ (R&D Systems, Abingdon, U.K.) for 16 h, and then infected with C. burnetii. After 24 h, cells were washed to remove free bacteria (corresponding to day 0), and were again incubated with IFN-γ. Alternatively, IFN-γ was added to infected cells and the same procedure was used. Intracellular bacteria were revealed by indirect immunofluorescence. Briefly, cell preparations were fixed with 1% formaldehyde, incubated with human Abs to C. burnetii (purified IgG from patients with Q fever endocarditis, 1/4000 dilution) in the presence or the absence of 0.1 mg/ml lysophosphatidylcholine, washed, and incubated with a 1/200 dilution of FITC-conjugated F(ab′)2 anti-human IgG Abs (Beckman Coulter, Roissy, France). Results are expressed as an infection index, which is the product of the mean number of bacteria per infected cell and the percentage of infected cells × 100 (3). The viability of intracellular bacteria was assessed using the bacterial viability kit. The infected cells were homogenized in water and vigorously mixed. The cell lysate was centrifuged at 8000 × g for 10 min, and pelleted bacteria were collected. The combination of SYTO 9 stain and propidium iodide was added to the bacterial suspension and the fluorescence of organisms was observed. Results are expressed in percentage of live bacteria.

The phagosome acidification was analyzed using DM-NERF dextran (molecular mass, 10 kDa; Molecular Probes), a fluorescent probe of phagosomal pH (17). THP-1 cells were incubated with 20 μg/ml DM-NERF dextran and C. burnetii, or latex beads (0.8 μm; Sigma-Aldrich) as control, for 24 h. In some experiments, infected cells loaded with DM-NERF dextran were incubated with 10 nM bafilomycin A1 (Sigma-Aldrich), a specific inhibitor of vacuolar proton ATPase (V-H+-ATPase)3 (18) for 2 h. Bacteria were revealed by human Abs to C. burnetii and Texas Red-conjugated F(ab′)2 anti-human IgG Abs (Beckman Coulter) used at a 1/100 dilution. The intraphagosomal pH was measured by ratiometric analysis of fluorescence intensities of DM-NERF dextran (excitation, 490/440 nm; emission, 530 nm). Infected cells were incubated with buffer solutions with graded pH (4.0, 5.0, 6.0, 7.0, and 7.4) in the absence or the presence of 10 μM monensin (Sigma-Aldrich), which equilibrated the intravacuolar pH with extracellular pH (19). After 1 h, fluorescence was recovered using a microplate fluorescent reader (Fisher Scientific, Elancourt, France). The mean pH value of the samples was calculated using a reference pH curve.

Bacterial trafficking was studied by immunofluorescence as follows (20). THP-1 cells and monocytes were infected by C. burnetii (200 virulent bacteria or 25 avirulent bacteria per cell) for 4 h (considered as h 0), washed to remove free organisms, and incubated for additional periods. Cell preparations were then fixed in 3% paraformaldehyde for 20 min. After washing, cells were incubated with ammonium chloride to neutralize free aldehydes and were permeabilized by PBS containing 0.1% saponin and 10% horse serum for 30 min. Human Abs specific for C. burnetii were used at a 1/4000 dilution. The Abs to intracellular markers were: rabbit anti-lysosome-associated membrane protein-1 (Lamp-1) Abs (a gift from Dr. M. Fukuda, The Burnham Institute, La Jolla, CA) used at a 1/1000 dilution, anti-cation-independent mannose-6-phosphate R (M6PR) Abs (a gift from Dr. B. Hoflack, Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany) used at a 1/500 dilution, anti-Rab7 Abs (a gift from Dr. M. Zerial, Max Planck Institute of Molecular Cell Biology and Genetics) used at a 1/200 dilution, anti-cathepsin D Abs (a gift from Dr. S. Kornfeld, Washington University School of Medicine, St. Louis, MO) used at a 1/1000 dilution; mouse anti-CD63 mAbs (BD Biosciences, le Pont de Claix, France) used at a 1/1000 dilution, and anti-V-H+-ATPase mAbs (Chemicon International, Temecula, CA) used at a 1:500 dilution. Primary Abs were added to cell preparations in PBS containing 0.1% saponin and 5% horse serum for 30 min. After being washed, monocytic cells were incubated with fluorescent secondary Abs in 0.1% saponin. Bacteria were revealed by Texas Red-conjugated F(ab′)2 anti-human IgG Abs and intracellular markers by FITC-conjugated F(ab′)2 anti-rabbit or anti-mouse IgG Abs (Beckman Coulter), both Abs being used at a 1:100 dilution. The colocalization of bacteria with intracellular markers was examined with a laser scanning confocal fluorescence microscope (Leica TCS 4D; Heidelberg, Germany). Optical sections of images were collected at 0.5-μm intervals and analyzed using Adobe Photoshop V5.5 software (Mountain View, CA). C. burnetii phagosomes were scored as positive for soluble markers when fluorescence was observed in the phagosome lumen; for membrane markers, phagosomes were scored as positive when a fluorescence ring surrounded organisms. About 30 C. burnetii-containing vacuoles were scored per coverslip, and at least three distinct experiments were performed per condition. Results are expressed as the percentage of phagosomes expressing intracellular markers.

THP-1 cells were infected with C. burnetii for 4 h. The lysosomotropic probe neutral red (Molecular Probes) was added at 5 μg/ml to infected cells 2 h before the end of the infection time. After being washed, cells were fixed in 3% paraformaldehyde. Bacteria were revealed by indirect immunofluorescence with FITC-conjugated secondary Abs, and neutral red was observed with excitation and emission filters for Texas Red. The cathepsin D-sensitive near-infrared fluorescence (NIRF) probe was prepared as previously described (21, 22). It was conjugated with FITC to monitor probe internalization and with a Cy5.5 marker that became fluorescent in the near-infrared spectrum after cathepsin D activation. NIRF at 0.1 μM was added to cells 1 h before the end of the infection time. THP-1 cells were then washed to remove noninternalized organisms and the free NIRF probe and were fixed with methanol at −20°C for 5 min. Bacteria were revealed by indirect immunofluorescence with Texas Red-conjugated secondary Abs. The colocalization of bacteria with the NIRF probe was examined with the laser scanning confocal fluorescence microscope equipped with appropriate excitation and emission filters for FITC, Texas Red, and Cy5.5. Images were analyzed using Adobe Photoshop version 5.5 software.

Results, given as mean ± SE, were compared with Student’s t test. Differences were considered significant when p < 0.05.

As suspended THP-1 cells are not suitable tools for studying intracellular traffic of C. burnetii, they were treated by PMA to induce their spreading and adhesion, and then incubated with C. burnetii. Adding virulent organisms (200:1 bacterium-cell ratio) to adherent THP-1 cells for 24 h led to the infection of 85 ± 6% of cells with 1–2 bacteria per cell. Cellular infection slightly decreased from days 0 to 3 by 17% and then steadily reached the initial value of infection after 5 days (Fig. 1,A). Beyond this time, a decrease in viability of THP-1 cells impaired the determination of cellular infection. The changes in bacterial number correlated with bacterial viability, which was assessed under the same experimental conditions. The percentage of live virulent organisms did not vary over 5 days (Fig. 1,B). In contrast, avirulent variants of C. burnetii were eliminated by adherent THP-1 cells. As they were more efficiently internalized than virulent organisms, they were added to THP-1 cells at a bacterium-cell ratio of 25:1. At day 0, 84 ± 8% of cells were infected with 1 or 2 bacteria, and cellular infection decreased by 42% at day 1 and slowly went down to reach 84% inhibition at day 5 (Fig. 1,A). The viability of avirulent organisms rapidly decreased during the course of experiments because it was diminished by 35% at day 0 and 84% at day 5 (Fig. 1 B). These results indicate that only virulent C. burnetii survived in PMA-treated THP-1 cells.

As previous reports suggested that C. burnetii replicates in acidic compartments (5), we investigated the relationship between acidic pH and C. burnetii survival in adherent THP-1 cells. The intravacuolar pH was determined by ratiometric analysis of fluorescence intensities of pH-sensitive DM-NERF dextran. In uninfected THP-1 cells, the pH of phagosomes containing latex beads was acidic. After 24 h of incubation of THP-1 cells with C. burnetii, intravacuolar pH was 5.0 ± 0.1 for virulent organisms and 5.2 ± 0.1 for avirulent organisms (Table I). The acidic pH of C. burnetii vacuoles results from the acquisition of V-H+-ATPase, known to acidify phagosomes. The percentage of vacuoles that accumulated V-H+-ATPase was 58 ± 11% at h 0, and it progressively increased to 80 ± 8% at h 72. V-H+-ATPase also colocalized with avirulent C. burnetii: 72 ± 10 and 88 ± 8% of vacuoles containing avirulent organisms colocalized with V-H+-ATPase at hours 0 and 72, respectively (Fig. 2). The V-H+-ATPase was functional as demonstrated by using bafilomycin A1, a specific inhibitor of V-H+-ATPase. In the presence of bafilomycin A1, the pH of vacuoles containing virulent or avirulent C. burnetii was significantly (p < 0.02) higher than that in the absence of bafilomycin A1 (Table I). Hence, C. burnetii is present in vacuoles that are acidified by V-H+-ATPase, independently of organism virulence.

As the acidic pH of C. burnetii vacuoles cannot account for the survival of virulent organisms in THP-1 cells, we suggested that the dynamics of vacuoles containing virulent organisms is distinct from that of avirulent organisms. The intracellular traffic of C. burnetii vacuoles was studied by measurement of organism colocalization with the lysosomal protease cathepsin D. Cathepsin D did not accumulate in vacuoles containing virulent C. burnetii (Fig. 3,A). At h 0, only 10 ± 5% of virulent C. burnetii colocalized with cathepsin D, but cathepsin D appeared in the lumen of 42 ± 6% of vacuoles containing avirulent organisms (Fig. 3,A). It is noteworthy that the amount of infection with avirulent C. burnetii remained higher than in cells infected with virulent organisms as a consequence of distinct phagocytosis efficiency. The lack of cathepsin D colocalization with C. burnetii was not due to its delayed acquisition. Indeed, the percentage of vacuoles containing virulent C. burnetii that colocalized with cathepsin D did not exceed 20% whatever the postinfection time, but it steadily increased in cells infected with avirulent variants, reaching 88 ± 7% after 96 h (Fig. 3,B). To confirm that defective acquisition of cathepsin D corresponds to impaired phagosome-lysosome fusion, two probes that accumulated in the lysosomal compartment were used. First, whereas the NIRF probe did not colocalize with virulent organisms, it accumulated in phagosomes containing avirulent organisms (Fig. 4,A). Second, the lysosomotropic probe neutral red colocalized only with phagosomes containing avirulent variants of C. burnetii (Fig. 4 B). Defective phagosome-lysosome fusion was not strain-dependent. Indeed, the percentage of phagosomes containing organisms from Priscilla and Q212 strains that colocalized with cathepsin D was 10 ± 3% and 15 ± 4%, respectively (data not shown). In addition, when virulent C. burnetii organisms were heat-killed, they regained the ability to colocalize with cathepsin D (33 ± 5% of positive vacuoles at day 0 and 80 ± 9% of positive vacuoles after 96 h). Hence, defective phagosome-lysosome fusion was related to C. burnetii virulence.

To determine whether impaired access of C. burnetii vacuoles to late endosomes accounts for defective acquisition of cathepsin D, we studied C. burnetii colocalization with markers of late endosomes-early lysosomes, Lamp-1 and CD63. The Lamp-1 fluorescence appeared as a ring surrounding the organisms, and 35 ± 12% of phagosomes containing virulent C. burnetii colocalized with Lamp-1 at h 0. The percentage progressively increased, and all bacteria were colocalized with Lamp-1 at h 72 (Fig. 2). The colocalization of avirulent C. burnetii with Lamp-1 was similar to that of virulent organisms. In addition, C. burnetii colocalized with CD63: the colocalization was high at h 0 (55 ± 13% for virulent organisms and 62 ± 15% for avirulent organisms), and it remained elevated during the incubation time (Fig. 2). M6PR is a marker of late endosomes that is transiently acquired by phagosomes containing inert particles. After 2 h, 24 ± 17% of phagosomes containing virulent C. burnetii and 27 ± 10% of phagosomes containing avirulent variants expressed M6PR. Although the percentage of vacuoles containing C. burnetii organisms that expressed M6PR remained low, it was similar in cells infected with virulent or avirulent organisms (Fig. 2). Hence, the acquisition of Lamp-1, CD63, and M6PR by C. burnetii phagosomes was not related to bacterial virulence. Distinct results were obtained with Rab7, a small GTPase involved in phagosome maturation (23). Rab7 was diffusively stained within the cytoplasm and its fluorescence was concentrated around bacteria (Fig. 5,A). Only 26 ± 9% of vacuoles containing virulent organisms colocalized with Rab7 at h 0; the percentage remained constant until 96 h despite a moderate increase (38 ± 5%) at h 2. In contrast, 65 ± 7% of vacuoles containing avirulent organisms were colocalized with Rab7 at h 0, and this percentage steadily decreased to values similar to those of virulent organisms after 96 h (Fig. 5 B). Hence, vacuoles containing virulent C. burnetii partially acquired Rab7. Taken together, these data suggest that vacuoles containing virulent C. burnetii follow the endosomal pathway, as revealed by the acquisition of V-H+-ATPase, Lamp-1, and CD63, but partially acquire M6PR and Rab7.

Because IFN-γ induces the killing of virulent C. burnetii in THP-1 cells (12), we wondered whether IFN-γ also affects the maturation of C. burnetii vacuoles. The addition of IFN-γ (at 200 U/ml) to adherent THP-1 cells before their infection decreased the viability of virulent C. burnetii from 92 ± 8% at day 0 to 16 ± 5% after 2 days (Fig. 6). The induction of C. burnetii killing was dose-dependent: a concentration of 50 U/ml IFN-γ was sufficient to reduce C. burnetii viability (30 and 40% inhibition at days 1 and 2, respectively), and maximum killing (85% inhibition) was obtained with 200 U/ml IFN-γ. We investigated the effect of IFN-γ on intravacuolar pH and colocalization of C. burnetii with endosome/lysosome markers. First, IFN-γ significantly (p < 0.05) raised the pH of vacuoles containing virulent C. burnetii to 6.2 ± 0.3, equivalent to pH values obtained by treating monocytes with bafilomycin A1 (Table I). IFN-γ exerted the same effect on the pH of vacuoles containing avirulent organisms. The alkalinization of bacterial vacuoles occurred after 24 h, suggesting that it was a relatively late event. Second, in the presence of 200 U/ml IFN-γ, the percentage of vacuoles containing virulent C. burnetii that colocalized with cathepsin D was 62 ± 6% at h 0 and 80 ± 7% after h 24, whereas it never exceeded 20% in the absence of IFN-γ (Fig. 6). IFN-γ-mediated restoration of bacterial colocalization with cathepsin D was observed with concentrations of IFN-γ similar to those required for bacterial killing (data not shown). However, the cathepsin D that colocalized with virulent organisms was not active. Indeed, the NIRF probe colocalized with virulent organisms as demonstrated by FITC fluorescence, but cathepsin D was inactive as shown by the lack of Cy5.5 fluorescence (Fig. 7). In contrast, the ability of IFN-γ to stimulate the maturation of C. burnetii phagosomes involves Rab7. Indeed, IFN-γ increased the colocalization of virulent C. burnetii with Rab7 as compared with untreated cells (Fig. 6). The percentage of phagosomes that expressed Rab7 was 53 ± 8% at h 0, and it increased to 83 ± 9% at h 2. After 24 h, 45 ± 10% of phagosomes still expressed Rab7. Taken together, these results show that the pretreatment of THP-1 cells by IFN-γ, which induces C. burnetii killing, improves the access of C. burnetii phagosomes to some endosomal markers without leading to complete maturation in phagolysosomes.

We wondered whether IFN-γ-induced C. burnetii killing results from the acquisition of endosomal/lysosomal markers or the alkalinization of bacterial vacuoles. To discriminate between these two hypotheses, THP-1 cells were infected with C. burnetii, thus providing vacuoles containing live organisms, and then were treated with 200 U/ml IFN-γ. This treatment reduced C. burnetii viability after 24 and 48 h (72 ± 9 and 38 ± 3% of viable bacteria, respectively), thus confirming the microbicidal effect of IFN-γ administered before C. burnetii infection (Fig. 6). Adding IFN-γ to infected cells increased the colocalization of C. burnetii with cathepsin D. Indeed, 42 ± 14 and 55 ± 12% of C. burnetii vacuoles colocalized with cathepsin D after 8 and 24 h, respectively (Fig. 6). Vacuolar cathepsin D was active as demonstrated by the Cy5.5 fluorescence of the NIRF probe (Fig. 7). In addition, IFN-γ treatment of infected cells restored the colocalization of Rab7 with C. burnetii phagosomes as did IFN-γ pretreatment (Fig. 6). In contrast to the effect of IFN-γ pretreatment of THP-1 cells, the addition of IFN-γ to C. burnetii-infected cells did not affect the vacuolar pH (Table I). Thus, the effects of IFN-γ on cathepsin D acquisition and vacuolar pH are likely distinct.

To extend the findings we reported in THP-1 cells to circulating monocytes, isolated monocytes were cultured for 3 days to increase their spreading without inducing their maturation into macrophages. This procedure was required to visualize bacterial phagosomes with confocal microscopy. First, we measured the viability of C. burnetii in monocytes. The viability of avirulent organisms decreased by 85% after 3 days of culture (Table II) and was residual after 5 days (data not shown). In contrast, the viability of virulent organisms remained constant during the 5 days of culture. Hence, only virulent C. burnetii survived in circulating monocytes, confirming previous results (3, 11). Second, we assessed the colocalization of C. burnetii organisms with two markers of phagosome maturation, Lamp-1 and cathepsin D. The percentage of vacuoles that colocalized with Lamp-1 was high in monocytes infected with virulent or avirulent C. burnetii at day 0, and all phagosomes had acquired Lamp-1 at day 3 (Table II). In contrast, the pattern of cathepsin D colocalization with C. burnetii was different. At day 0, one-third of vacuoles acquired cathepsin D in monocytes infected with avirulent C. burnetii, and almost all vacuoles were positive for cathepsin D at day 3. In monocytes infected with virulent organisms, 26 ± 7% of vacuoles colocalized with cathepsin D at day 0, and this percentage remained low even at day 3 (Table II). Taken together, these results show that the survival of virulent C. burnetii in circulating monocytes is associated with impaired acquisition of cathepsin D.

In this paper, we show that C. burnetii escapes killing in THP-1 monocytic cells and circulating monocytes by preventing phagosomal maturation. This finding partly questions the C. burnetii paradigm in which C. burnetii replicates in the cellular compartment displaying phagolysosomal features (5, 24). The first characteristic reported for C. burnetii was that acidic pH is required for bacterial metabolism (6, 7). In agreement with previous reports on avirulent organisms (8, 9), we found C. burnetii in acidic vacuoles independently of bacterial virulence. Hence, phagosome pH cannot account for the survival of virulent C. burnetii in human monocytes. The second characteristic reported for the C. burnetii vacuole is its ability to fuse with different intracellular compartments, including lysosomes. Hence, vacuoles that enclose C. burnetii can become large, contain numerous organisms, and fuse with vacuoles containing Leishmania amazonensis or M. avium (25, 26). The conclusions of these studies are limited by their use of avirulent organisms and cells in which C. burnetii replication was independent of bacterial virulence. We show that vacuoles containing virulent C. burnetii did not acquire cathepsin D, a lysosomal hydrolase. The impairment of cathepsin D accumulation reflects defective phagolysosomal fusion because lysosomal trackers did not accumulate in C. burnetii phagosomes. The lack of colocalization of C. burnetii and cathepsin D was not due to delayed acquisition of cathepsin D by C. burnetii vacuoles, which disagrees with a recent paper in which virulent C. burnetii delays phagolysosomal fusion in J774 cells (27), but these murine macrophage-like cells allow the replication of both virulent and avirulent organisms. In blood monocytes, in which only virulent C. burnetii organisms survive (Refs. 3 and 11 and our results), bacterial phagosomes were unable to fuse with lysosomes. This was specific of bacterial virulence because cathepsin D accumulated within vacuoles containing avirulent C. burnetii. In addition, two other virulent strains of C. burnetii exhibit similar impairment of phagosome-lysosome fusion. The lack of cathepsin D colocalization with C. burnetii did not result from impaired interactions of bacterial vacuoles with the endocytic pathway. Hence, markers of late endosomes and early lysosomes such as Lamp-1, CD63, V-H+-ATPase were acquired by C. burnetii vacuoles independently of bacterial virulence. Thus, the survival of C. burnetii in monocytic cells is associated with altered phagosomal maturation, which is reminiscent of the escape mechanism used by Salmonella enterica. Indeed, both types of pathogen-containing vacuoles acquire markers of late endosomes-early lysosomes such as Lamp-1, but are devoid of lysosomal enzymes (28, 29). However, the molecular mechanisms involved in the control of phagosome maturation are likely different. C. burnetii vacuoles partly acquired the late-endosomal GTPase Rab7, while Salmonella vacuoles recruit Lamp-1 in a Rab7-dependent manner (30). It is likely that the ability of Rab7 to regulate vesicle traffic in late endocytosis (31, 32) is altered in C. burnetii infection. This hypothesis is strengthened by the finding that C. burnetii had no effect on the early acquisition of EEA1, a marker of early endosomes (data not shown). The strategy of C. burnetii survival in human monocytes is likely based on interference with Rab7 that controls transport to endocytic degradative compartments, leading to the formation of a vacuole unable to fuse with lysosomes.

IFN-γ stimulated the killing of C. burnetii by THP-1 cells and affected the maturation of C. burnetii vacuoles as assessed by the acquisition of cathepsin D. IFN-γ likely affects cathepsin D acquisition by distinct mechanisms. IFN-γ pretreatment of THP-1 cells induced the accumulation of inactive cathepsin D by C. burnetii vacuoles, which is reminiscent of the results of Ullrich et al. (33), who found that M. avium phagosomes acquire an inactive form of cathepsin D. IFN-γ also stimulated the alkalinization of C. burnetii vacuoles. Vacuole alkalinization did not result from the exclusion of V-H+-ATPase because the colocalization of V-H+-ATPase and C. burnetii was similar in THP-1 cells treated or not with IFN-γ. This finding is surprising because IFN-γ has been reported to lower the pH of M. avium vacuoles through the accumulation of V-H+-ATPase (14) and to impair the interaction of phagosomes with late endosomes and lysosomes without interfering with acidification (16). In contrast, adding IFN-γ to infected THP-1 cells stimulated the acquisition of active cathepsin D but it had no effect on vacuolar pH. Thus, it is likely that IFN-γ-induced phagolysosomal fusion and vacuolar alkalinization play different roles in C. burnetii killing. These results have pathophysiological consequences. Resting monocytic cells are unable to kill virulent C. burnetii but cannot support bacterial replication. This latter is only achieved when monocytes are specifically deactivated by IL-10 (11). IL-10 does not modify the traffic of C. burnetii vacuoles in monocytes (our unpublished data). In contrast, the activation of monocytic cells by IFN-γ reprogrammed them to be microbicidal against C. burnetii through phagosomal maturation. As IFN-γ is associated with the cure of C. burnetii infections (2), it is likely that the restoration of phagosome-lysosome fusion is critical for the control of Q fever.

The survival of C. burnetii into THP-1 cells and monocytes is associated with altered phagosome maturation. The activation of these professional phagocytes by IFN-γ leads to C. burnetii killing and restores phagosomal maturation. We propose two potential mechanisms for IFN-γ-induced killing of C. burnetii: an early mechanism based on phagosome maturation and a late mechanism involving modulation of vacuolar pH. Therapeutic elimination of C. burnetii in Q fever might benefit from exploring these two parameters of bacterial killing.

We thank M. Fukuda, B. Hoflack, M. Zerial, and S. Kornfeld for their generous gift of Abs, and J. Pizzaro, M. Barrad, and S. Méresse for expert support.

1

This work was supported by the Programme de Recherche en Microbiologie Fondamentale et Maladies Infectieuses et Parasitaires.

3

Abbreviations used in this paper: V-H+-ATPase, vacuolar proton ATPase; Lamp, lysosome-associated membrane protein; M6PR, cation-independent mannose-6-phosphate R; NIRF, near-infrared fluorescence.

1
Maurin, M., D. Raoult.
1999
. Q fever.
Clin. Microbiol. Rev.
12
:
518
2
Mege, J. L., M. Maurin, C. Capo, D. Raoult.
1997
. Coxiella burnetii: the “query” fever bacterium: a model of immune subversion by a strictly intracellular microorganism.
FEMS Microbiol. Rev.
19
:
209
3
Capo, C., F. P. Lindberg, S. Meconi, Y. Zaffran, G. Tardei, E. J. Brown, D. Raoult, J. L. Mege.
1999
. Subversion of monocyte functions by Coxiella burnetii: impairment of the cross-talk between αvβ3 integrin and CR3.
J. Immunol.
163
:
6078
4
Akporiaye, E. T., J. D. Rowatt, A. A. Aragon, O. G. Baca.
1983
. Lysosomal response of a murine macrophage-like cell line persistently infected with Coxiella burnetii.
Infect. Immun.
40
:
1155
5
Baca, O. G., Y. P. Li, H. Kumar.
1994
. Survival of the Q fever agent Coxiella burnetii in the phagolysosome.
Trends Microbiol.
2
:
476
6
Chen, S. Y., M. Vodkin, H. A. Thompson, J. C. Williams.
1990
. Isolated Coxiella burnetii synthesizes DNA during acid activation in the absence of host cells.
J. Gen. Microbiol.
136
:
89
7
Hackstadt, T., J. C. Williams.
1981
. Biochemical stratagem for obligate parasitism of eukaryotic cells by Coxiella burnetii.
Proc. Natl. Acad. Sci. USA
78
:
3240
8
Heinzen, R. A., M. A. Scidmore, D. D. Rockey, T. Hackstadt.
1996
. Differential interaction with endocytic and exocytic pathways distinguish parasitophorous vacuoles of Coxiella burnetii and Chlamydia trachomatis.
Infect. Immun.
64
:
796
9
Maurin, M., A. M. Benoliel, P. Bongrand, D. Raoult.
1992
. Phagolysosomes of Coxiella burnetii-infected cell lines maintain an acidic pH during persistent infection.
Infect. Immun.
60
:
5013
10
Dellacasagrande, J., E. Ghigo, C. Capo, D. Raoult, J. L. Mege.
2000
. Coxiella burnetii survives in monocytes from patients with Q fever endocarditis: involvement of tumor necrosis factor.
Infect. Immun.
68
:
160
11
Ghigo, E., C. Capo, D. Raoult, J. L. Mege.
2001
. Interleukin-10 stimulates Coxiella burnetii replication in human monocytes through tumor necrosis factor down-modulation: role in microbicidal defect of Q fever.
Infect. Immun.
69
:
2345
12
Dellacasagrande, J., C. Capo, D. Raoult, J. L. Mege.
1999
. IFN-γ-mediated control of Coxiella burnetii survival in monocytes: the role of cell apoptosis and TNF.
J. Immunol.
162
:
2259
13
Boehm, U., T. Klamp, M. Groot, J. C. Howard.
1997
. Cellular responses to interferon-γ.
Annu. Rev. Immunol.
15
:
749
14
Schaible, U. E., S. Sturgill-Koszycki, P. H. Schlesinger, D. G. Russell.
1998
. Cytokine activation leads to acidification and increases maturation of Mycobacterium avium-containing phagosomes in murine macrophages.
J. Immunol.
160
:
1290
15
Via, L. E., R. A. Fratti, M. McFalone, E. Pagan-Ramos, D. Deretic, V. Deretic.
1998
. Effects of cytokines on mycobacterial phagosome maturation.
J. Cell Sci.
111
:
897
16
Tsang, A. W., K. Oestergaard, J. T. Myers, J. A. Swanson.
2000
. Altered membrane trafficking in activated bone marrow-derived macrophages.
J. Leukocyte Biol.
68
:
487
17
Rathman, M., M. D. Sjaastad, S. Falkow.
1996
. Acidification of phagosomes containing Salmonella typhimurium in murine macrophages.
Infect. Immun.
64
:
2765
18
Drose, S., K. Altendorf.
1997
. Bafilomycins and concanamycins as inhibitors of V-ATPases and P-ATPases.
J. Exp. Biol.
200
:
1
19
Lukacs, G. L., O. D. Rotstein, S. Grinstein.
1990
. Phagosomal acidification is mediated by a vacuolar-type H+-ATPase in murine macrophages.
J. Biol. Chem.
265
:
21099
20
Pizarro-Cerda, J., E. Moreno, V. Sanguedolce, J. L. Mege, J. P. Gorvel.
1998
. Virulent Brucella abortus prevents lysosome fusion and is distributed within autophagosome-like compartments.
Infect. Immun.
66
:
2387
21
Tung, C. H., S. Bredow, U. Mahmood, R. Weissleder.
1999
. Preparation of a cathepsin D sensitive near-infrared fluorescence probe for imaging.
Bioconjugate Chem.
10
:
892
22
Ghigo, E., C. Capo, M. Aurouze, C. H. Tung, J. P. Gorvel, D. Raoult, J. L. Mege.
2002
. Survival of Tropheryma whippelii, the agent of Whipple’s disease, requires phagosome acidification.
Infect. Immun.
70
:
1501
23
Desjardins, M., L. A. Huber, R. G. Parton, G. Griffiths.
1994
. Biogenesis of phagolysosomes proceeds through a sequential series of interactions with the endocytic apparatus.
J. Cell Biol.
124
:
677
24
Sinai, A. P. 2000. Life on the inside: microbial strategies for intracellular survival and persistence. In Persistent bacterial infections. J. P. Nataro, M. J. Blaser, and S. Cunningham-Rundles, eds. ASM Press, Washington, D.C., p. 31.
25
de Chastellier, C., M. Thibon, M. Rabinovitch.
1999
. Construction of chimeric phagosomes that shelter Mycobacterium avium and Coxiella burnetii (phase II) in doubly infected mouse macrophages: an ultrastructural study.
Eur. J. Cell Biol.
78
:
580
26
Veras, P. S., C. de Chastellier, M. F. Moreau, V. Villiers, M. Thibon, D. Mattei, M. Rabinovitch.
1994
. Fusion between large phagocytic vesicles: targeting of yeast and other particulates to phagolysosomes that shelter the bacterium Coxiella burnetii or the protozoan Leishmania amazonensis in Chinese hamster ovary cells.
J. Cell Sci.
107
:
3065
27
Howe, D., L. P. Mallavia.
2000
. Coxiella burnetii exhibits morphological change and delays phagolysosomal fusion after internalization by J774A.1 cells.
Infect. Immun.
68
:
3815
28
Garcia-del-Portillo, F., B. B. Finlay.
1995
. Targeting of Salmonella typhimurium to vesicles containing lysosomal membrane glycoproteins bypasses compartments with mannose 6-phosphate receptors.
J. Cell Biol.
129
:
81
29
Steele-Mortimer, S., B. H. Méresse, J. P. Gorvel Toh, B. B. Finlay.
1999
. Biogenesis of Salmonella typhimurium-containing vacuoles in epithelial cells involves interactions with the early endocytic pathway.
Cell Microbiol.
1
:
33
30
Méresse, S., O. Steele-Mortimer, B. B. Finlay, J. P. Gorvel.
1999
. The rab7 GTPase controls the maturation of Salmonella typhimurium-containing vacuoles in HeLa cells.
EMBO J.
18
:
4394
31
Cantalupo, G., P. Alifano, V. Roberti, C. B. Bruni, C. Bucci.
2001
. Rab-interacting lysosomal protein (RILP): the Rab7 effector required for transport to lysosomes.
EMBO J.
20
:
683
32
Méresse, S., J. P. Gorvel, P. Chavrier.
1995
. The rab7 GTPase resides on a vesicular compartment connected to lysosomes.
J. Cell Sci.
108
:
3349
33
Ullrich, H. J., W. L. Beatty, D. G. Russell.
2000
. Interaction of Mycobacterium avium-containing phagosomes with the antigen presentation pathway.
J. Immunol.
165
:
6073