Staphylococcus aureus Escapes More Efficiently from the Phagosome of a Cystic Fibrosis Bronchial Epithelial Cell Line than from Its Normal C
http://www.100md.com
感染与免疫杂志 2006年第5期
Department of Microbiology and Immunology, Dartmouth Medical School, Hanover, New Hampshire 03755
ABSTRACT
Staphylococcus aureus is frequently the initial bacterium isolated from young cystic fibrosis (CF) patients, and yet its role in CF disease progression has not been determined. Recent data from our lab demonstrates that S. aureus can invade and replicate within the CF tracheal epithelial cell line (CFT-1). Here we describe the finding that the fate of internalized S. aureus in CFT-1 cells differs from its complemented counterpart (LCFSN). S. aureus strain RN6390 was able to replicate within the mutant CFT-1 cells after invasion but not in the complemented LCFSN cells. At 1 h postinvasion, S. aureus containing vesicles within both cell lines acquired vacuolar-ATPase, lysosomal markers LAMP 1 and 2, and the lysomotrophic dye LysoTracker to a similar degree. However, at 4 h postinvasion, the percentage of S. aureus within CFT-1 cells associated with these markers decreased significantly compared to LCFSN, where the association approached 100%. Transmission electron microscopic analysis revealed that the majority of bacteria within CFT-1 cells were free in the cytosol at 4 h after invasion, whereas most S. aureus bacteria internalized by LCFSN cells remained within vesicles. These results demonstrate a fundamental difference in the fate of live S. aureus after invasion of CFT-1 versus LCFSN cell lines and may explain the propensity of S. aureus to cause chronic lung infection in CF patients.
INTRODUCTION
Cystic fibrosis (CF) is the most common lethal autosomal-recessive disorder, affecting 1 in 3,500 newborns each year in the United States (6). Patients with CF are susceptible to respiratory tract infections at an early age, with the most frequent initial isolate being Staphylococcus aureus (4, 14, 18). Infections with S. aureus usually precede chronic colonization of the respiratory tract by the opportunistic pathogen Pseudomonas aeruginosa and continue into adulthood, when 51% of patients become culture positive for S. aureus. Administration of antibiotics is often ineffective at eradicating S. aureus from the CF lung and generally results in relapsing infections that are clonal in nature (13, 19). Although the role of S. aureus in the disease progression of CF remains unknown (6, 17), the clinical observation of rapid recolonization soon after treatment suggests that S. aureus may persist inside host cells of the lung (23), possibly explaining the propensity for recurrent S. aureus infection in CF patients.
Bacteria internalized by nonphagocytic host cells are generally processed in a membrane-bound vesicle that matures through a series of fusion steps to yield a fully mature phagolysosome characterized by a low pH and containing lysosomal hydrolases. Recent studies suggest that a growing number of bacterial pathogens have evolved to circumvent the normal endocytic trafficking pathways within the host cell. Phagocytosed Mycobacterium tuberculosis arrests maturation of its endosomal compartment by inhibiting the acquisition of early endosomal antigen 1, a protein involved in maturation of early endosomes into late endosomes, thus avoiding degradation within the lysosomal compartment (11, 32). Listeria monocytogenes, in contrast, escapes from the phagosomal vesicle minutes after internalization by secreting the pore-forming toxin listeriolysin O (12). Some bacterial species, such as Coxiella burnetii, can withstand the harsh lysosomal environment and replicate within the lysosomal compartment (5). In the case of S. aureus, which is classically considered as an extracellular pathogen, accumulating experimental evidence suggests that S. aureus can act as an intracellular pathogen, capable of invading and surviving within a broad range of nonphagocytic cells, including enterocytes, osteocytes, and endothelial and epithelial cells (2, 16, 20, 24, 26). Recent data from our lab demonstrated that S. aureus can invade and replicate within a CF tracheal (CFT-1) epithelial cell line derived from a CF patient homozygous for the F508 mutation of CFTR (20). However, the specific pathway by which S. aureus subverts host cell trafficking is not known. Understanding the trafficking of pathogens such as S. aureus within host cells will help us understand the complex interactions between pathogen and hosts.
Accordingly, the objective of the present study was to compare the intracellular trafficking of S. aureus between the CF tracheal epithelial cell line (CFT-1) and the complemented cell line (LCFSN). We found that live S. aureus, once internalized into CFT-1 cells, resided within a vesicle with characteristics of a late endosome prior to its escape into the host cytosol. In contrast, S. aureus that invaded LCFSN cells remained bound within vesicles that retained late endosome markers, leading to inhibition of replication and progressive destruction by the endosomal contents. These data suggest that the F508 mutation in CFTR is associated with a defect in the ability of epithelial cells to control endosomal escape and also a failure to degrade internalized S. aureus effectively, thus allowing for intracellular bacterial replication and subsequent tissue damage in the lung. This finding regarding the interactions of S. aureus with CF airway epithelial cells implies that S. aureus might play a more substantial role in the progression of lung disease in CF patients than what was originally suspected.
MATERIALS AND METHODS
Bacterial strains and growth conditions. For invasion experiments, S. aureus strain RN6390 containing pALC1743, a pSK236-derived shuttle vector containing the S. aureus RNAIII promoter driving the expression of GFPuvr, was grown overnight 37°C in 5 ml of Trypticase soy broth with 10 μg of chloramphenicol/ml (20). Bacillus subtilis strain 168 was grow overnight at 37°C with shaking in Luria-Bertani (LB) broth. The cultures were then diluted 1:100 in 10 ml or respective media and grown at 37°C to an optical density at 650 nm (OD650) of 1.2 (late log phase). L. monocytogenes strain EGD was grown in brain heart infusion overnight shaking at 30°C. Bacteria were washed two times in phosphate-buffered saline (PBS) and resuspended in the invasion medium Dulbecco modified Eagle medium-F-12 (50:50; DMEM/F12; Cellgro) supplemented with 2 mM L-glutamine and 1% fetal bovine serum (Cellgro) at an OD650 of 0.4 (4 x 108 CFU). Bacteria were diluted to achieve a multiplicity of infection (MOI) of 10:1 (bacteria to epithelial cell) in invasion medium. Dilutions of S. aureus were then subjected to sonication for 15 s using a microtip (Branson model LS75) to break up bacterial clusters prior to use in invasion experiments. For the heat-killed control, S. aureus bacteria in invasion media were incubated at 55°C for 1 h.
Cell culture and growth conditions. Two established cystic fibrosis tracheal epithelial cell lines (CFT-1 and LCFSN) were used in the present study. CFT-1 is a papillomavirus-immortalized tracheal epithelial cell line derived from a CF tissue donor homozygous for the F508 mutation of the CFTR gene. The complemented cell line, LCFSN, contains a retroviral vector expressing the wild-type CFTR gene (33). The cell lines were maintained in medium containing DMEM/F12 supplemented with 10% fetal bovine serum and the following supplements (from Sigma unless otherwise noted): 5 μg of insulin/ml, 3.7 μg of endothelial cell growth supplement/ml, 25 ng of epidermal growth factor/ml, 3 x 10–8 M triiodothyronine, 10–6 M hydrocortisone, 5 μg of transferrin/ml, and 10 ng of cholera toxin/ml. For routine passage, the medium containing an antibiotic-antimycotic solution with 100 U of penicillin G/ml, 25 μg of amphotericin B/ml, and 100 μg of streptomycin (Cellgro)/ml was used. Neomycin (150 μg/ml) was added to LCFSN cultures to maintain the CFTR-expressing retroviral vector.
For invasion experiments, epithelial cells (105 cells per well) were seeded onto 24-well plates (Costar) in supplemented DMEM/F12 medium without antibiotics and antimycotics and allowed to grow at 37°C in 5% CO2 for 3 days until confluent (roughly 2 x 105 cells per well). One hour prior to the experiments, cells were washed three times with PBS (Cellgro), followed by the addition of 1 ml of invasion media. At time zero, 500 μl of invasion medium containing bacteria was added at an MOI of 10:1 to each well, followed by incubation for 30 min at 37°C in 5% CO2. Cells were then washed three times with PBS, and 500 μl of invasion media containing 5 μg of lysostaphin (AMBI, Inc.)/ml was added to each well to lyse extracellular and adherent S. aureus. The absence of extracellular bacteria was then confirmed by CFU enumeration on agar plates. For intracellular replication experiments, wells were washed at the desired times three times with PBS, followed by incubation with 500 μl of 0.25% trypsin-0.1% EDTA (Cellgro) at 37°C in 5% CO2. After detachment, the cells were transferred to 500 μl of 0.025% Triton X-100 in H2O, sonicated to release intracellular bacteria, and then plated on Trypticase soy agar for determination of the CFU. For invasion assays with B. subtilis and L. monocytogenes, nonadherent bacteria, after a 1-h incubation, were removed by three washes with PBS, followed by the addition of invasion media containing gentamicin (50 μg/ml) to kill extracellular bacteria (15, 22). CFU enumeration of B. subtilis and L. monocytogenes was done on LB agar plates and brain heart infusion agar plates, respectively.
Immunofluorescence staining and confocal microscopy. Invasion of epithelial cells was done as described above with live or heat-killed S. aureus. At 1, 2, and 4 h postinvasion, monolayers were washed three times with PBS and fixed with 2% paraformaldehyde in PBS at room temperature overnight. For immunofluorescence staining, fixed cells were washed three times with PBS and then immersed in permeabilization buffer (PB) containing PBS, 1% bovine serum albumin, 0.1% sodium azide, and 0.1% saponin for 30 min at room temperature. Antibody solutions were diluted in PB at a 1/100 dilution for LAMP-1 and LAMP-2 antibodies (DSHB, University of Iowa) or a 1/1,000 dilution for anti-V-ATPase antibody (a gift from S. Sato), added to coverslips, and incubated for 1 h at 37°C in 5% CO2. Coverslips were washed three times in PB without saponin and then incubated for 1 h with a 1/40 dilution of goat anti-mouse antibody (Fab')2 conjugated to Alexa Fluor 555 (Molecular Probes). After three washes with PBS, both bacteria and epithelial cell DNA were stained with 1 μM To-Pro-3 (Molecular Probes) for 10 min at room temperature in PBS. Coverslips were then washed three times in PBS, treated with SloFade (Molecular Probes), mounted on glass slides, and examined on a Olympus IX-70 confocal microscope. Images were captured by using Flouview software (Olympus). For quantitative analysis, more than 15 fields of view were examined per slide, and for each condition more than 100 events were analyzed.
Assessment of vesicular pH. To evaluate whether S. aureus-containing vesicles are exposed to the acidic environment of the maturing lysosome, we utilized the lysomotrophic LysoTracker (LyT) DND-99 (Molecular Probes). Briefly, cell lines were grown on glass coverslips in 24-well plates until confluent and then used for subsequent invasion assays with live or heat-killed bacteria as described above. Thirty minutes prior to the time point, the cells were washed and incubated with 1 μM LysoTracker/ml, washed three times PBS, and then fixed with 2% paraformaldehyde. Coverslips were then stained with 1 μM To-Pro-3 (Molecular Probes) and mounted on glass slides for confocal analysis.
EM of S. aureus-infected cells. CFT-1 and LCFSN were grown in six-well plates (Costar) until confluent. Live and heat-killed bacteria were prepared as described above and added to epithelial cells at an MOI of 100:1 to increase the frequency of bacterial internalization for quantitative analysis. The six-well plates were then spun at 200 x g for 10 min at 4°C to maximize the number of bacteria adhering to the cell surface. After incubation for 10 min at 37°C in 5% CO2, the plates were washed three times with PBS and then incubated in invasion medium with 10 μg of lysostaphin/ml. At 1, 2, and 4 h after infection, the cells were washed three times with PBS and fixed with 2% glutaraldehyde-1% paraformaldehyde in 0.1 M sodium cacodylate (pH 7.4) at room temperature. The monolayers were then scraped from the dish and resuspended in 2% glutaraldehyde-1% paraformaldehyde in 0.1 M sodium cacodylate (pH 7.4) overnight at 4°C. Loose pellets were then postfixed in 1% OsO4 in 0.1 M sodium cacodylate (pH 7.4) for 1 h at room temperature, rinsed, blocked, and stained with 2% aqueous uranyl acetate for 30 min at room temperature in the dark. The samples were then dehydrated, immersed in LX112-propylene oxide, and allowed to polymerize at 60°C for 48 h. Sections were cut and stained with uranyl acetate in methanol for 5 min. Samples were examined by using a JEOL 2000FX electron microscope at 100 kV. For quantification, bacteria were considered within a vesicle when a single continuous membrane was visible by electron microscopy (EM).
Statistical analysis. For statistical analysis, the Student t test was performed using JMP 5.0.1.2 (2003). A P value of <0.05 was considered significant. The data for Fig. 2 were log transformed to equalize variance.
RESULTS
Intracellular replication of S. aureus occurs within CFT-1 cell line but not within the complemented cell line LCFSN. Previous work in our laboratory showed that S. aureus strain RN6390 is capable of invading and replicating within CFT-1 cells prior to inducing apoptosis (20). In light of these findings, we wanted to determine whether the ability of S. aureus to replicate intracellularly was due to the F508 mutation of the CFTR gene by comparing bacterial replication in CFT-1 and LCFSN cells. Monolayers of LCFSN or CFT-1 cells were infected with sonicated S. aureus at an MOI of 10:1 in 24-well plates, and the numbers of intracellular bacteria were monitored over time as described in Materials and Methods. The average number of intracellular S. aureus within CFT-1 cells was found to increase from 0.3 to 3 bacteria per epithelial cell over a 9-h period, with an intracellular doubling time of 2 h (Fig. 1A). We were unable to assess CFU for CFT-1 cells beyond the 10-h time point due to detachment of the monolayers, probably due to apoptosis as previously described (20). In contrast, the number of bacteria in LCFSN cells remained relatively constant at 0.03 bacteria per cell over the 10-h period, showing no signs of apoptosis.
Further evidence of bacterial replication was observed via direct microscopic analysis (Fig. 1B). Sonication of the bacterial sample prior to invasion experiments has generally resulted in one to two bacteria per cluster. At 1 h postinvasion, the size of the cluster did not change (Fig. 1B), whereas at 4 h in the CFT-1 cells the majority of the bacterial clusters were in groups of 4 to 10, suggesting that bacterial replication had occurred (Fig. 1B). This was observed infrequently with live bacteria within LCFSN cells and not at all with heat-killed bacteria, as expected (Fig. 1B). These data showed that S. aureus internalized into CFT-1 cells were able to replicate, whereas bacteria that were inside the complemented cell line LCFSN could not.
Adherence and internalization of S. aureus. The ability of S. aureus to replicate in the CFT-1 but not in the LCFSN cell line prompted us to examine whether adherence and internalization differed between the two cell lines. Incubation of bacteria with epithelial cells at an MOI of 10:1 for 30 min resulted in equal numbers of bacteria adherent to CFT-1 and LCFSN cell lines, with 10% of inoculated bacteria adhering to the monolayers (approximately 1 bacterium bound per epithelial cell) (Fig. 1C). This observation suggests that S. aureus is probably not binding directly to CFTR present on the epithelial cell surface.
In internalization assays, CFT-1 cells were found to internalize more than 50% of the adherent bacteria over a 2-h period compared to the LCFSN cells, which internalized only 5% (Fig. 1D). These findings demonstrated an increased rate of internalization of S. aureus in the CFT-1 cells compared to LCFSN cells.
To determine whether bacterial replication within CFT-1 cells was S. aureus specific, we compared the internalization and replication of S. aureus to the well-characterized intracellular pathogen (L. monocytogenes) and a nonpathogenic bacterium (Bacillus subtilis) between these two cell lines. L. monocytogenes has been shown to invade and replicate within a broad range of nonphagocytic cell lines, including hepatocytes and epithelial and endothelial cell lines (10, 21, 28). Although less efficient at being internalized into both cell lines, L. monocytogenes could replicate in both cell lines, as opposed to S. aureus, which was only able to replicate within the CFT-1 cell line (Fig. 2). More importantly, B. subtilis, a noninvasive bacterium was not able to replicate within either of the two cell lines.
Acquisition of late endosomal markers LAMP-1 and LAMP-2. Recent EM data from our lab showed that S. aureus was internalized into a membrane-bound vesicle prior to intracellular replication in CFT-1 cells (20). To examine the fate of S. aureus within this vesicle, we analyzed the acquisition of endosomal markers LAMP-1 and LAMP-2, proteins present in the membrane of late endosomes and lysosomes (9). In both CFT-1 and LCFSN cell lines at 1 h postinvasion, approximately 80% of the S. aureus-containing vesicles were associated with LAMP-1 and LAMP-2, as shown by the presence of LAMP-containing membranes (green circles) surrounding the bacteria (arrow) (Fig. 3A to D). At 2 and 4 h postinvasion, S. aureus within LCFSN cells maintained its association with LAMP-1 and LAMP-2, whereas this association with LAMP-1 and LAMP-2 in CFT-1 cells significantly dropped to 30 and 20%, respectively (Fig. 3E and F).
To determine whether internalized bacteria were within fully formed vesicles, we acquired a z-series of images at 1, 2, and 4 h postinternalization, using the vesicular marker LAMP-2. At 4 h postinvasion, live S. aureus within LCFSN cells (Fig. 4A) resided within a vesicle that stained entirely with the LAMP-2 marker (arrows). In contrast, at 4 h postinvasion, most bacteria within CFT-1 cells appeared as bacterial clusters that were devoid of LAMP-2-associated staining (Fig. 4B, arrowhead) whereas only a small percentage remained within LAMP-2 vesicles (arrow). As a control, heat-killed bacteria internalized into both cell lines acquired both LAMP-1 and LAMP-2 and maintained their association with these markers over the entire course of the experiment, as would be expected for vesicles destined to fuse with lysosomes (Fig. 3E and F, open symbols). Together, these data suggest an anomaly in the trafficking pathway of S. aureus within CFT-1 cells compared to LCFSN.
Acidification of the S. aureus containing vesicle. Maturation of an endocytic vesicle is normally accompanied by a steady decrease in luminal pH (from pH of <6.2 in early endosomes to a pH of <5.5 in lysosomes). The ability of the vesicle to acidify is essential for the optimal activity of hydrolytic enzymes found within the lysosomal compartment. To determine whether S. aureus resides in an acidic vesicle after internalization, we stained infected cells with LysoTracker DND 99, a weak lysomotrophic base that accumulates and fluoresces within acidic vesicles. Approximately 80% of heat-killed bacteria were within acidic vesicles of CFT-1 cells at 1 h postinvasion and approached 100% after 4 h (Fig. 5). Heat-killed bacteria internalized into LCFSN cells were uniformly found to colocalize with LysoTracker-positive vesicles, whereas LCFSN infected with live bacteria were associated with a delayed acidification at 1 h, but by 4 h more than 85% of the bacteria were found to be within acidic vesicles (Fig. 5C). In contrast, only 40% of the live internalized bacteria in CFT-1 cells were associated with acidic vesicles by 1 h, and this percentage significantly decreased to less than 20% after 4 h (Fig. 5C). These data suggest that a majority of live S. aureus bacteria within CFT-1 cells are not within acidic vesicles at 4 h postinvasion. Additional experiments with the lysomotrophic agent acridine orange revealed a similar trend (data not shown).
Association with V-ATPase. A decrease in the luminal pH of endocytic vesicles is mediated by V-ATPase, part of an integral membrane complex that pumps hydrogen ions into the vesicular lumen (25). Using an antibody against V-ATPase, we were able to quantify the percentage of S. aureus-containing vesicles associated with V-ATPase (Fig. 6). Approximately 80% of live bacteria within CFT-1 cells were associated with V-ATPase at 1 h after invasion, but by 4 h the number decreased to 65%. However, the data did not reach statistical significance compared to those obtained at 1 h postinvasion. Nevertheless, we saw a similar trend in repeating this experiment. In contrast, 90% of the live S. aureus cells within LCFSN cells were associated with V-ATPase after 1 h and maintained this association over 4 h. Similarly, 80 and 90% of heat-killed S. aureus cells that were internalized into CFT-1 and LCFSN cells, respectively, were associated with V-ATPase after 1 h of invasion, and this association persisted over 4 h (Fig. 6C).
EM analysis of infected cells. The relative absence of lysosomal marker in vesicles containing live internalized bacteria within CFT-1 cells at 2 h after invasion (Fig. 3E and F) implies that S. aureus either altered the acquisition of endosomal markers or may escape into the host cytosol. To narrow the possibility, we examined monolayers of CFT-1 and LCFSN cells infected with S. aureus at an MOI of 100:1 by EM (Fig. 7). Quantitative EM analysis of more than 100 bacteria per sample revealed that at 1 h postinvasion a majority of live bacteria were still within membrane-bound vesicles in both cell lines (Fig. 7D). After 4 h, evidence of membrane degradation and bacterial escape into the host cytosol was seen in CFT-1 cells infected with live bacteria (Fig. 7A). Between 2 and 4 h, the percentage of bacteria within vesicles in CFT-1 cells infected with live bacteria significantly decreased from 77 to 18%, whereas heat-killed bacteria internalized by CFT-1 cells remained bound within vesicles after 4 h (Fig. 7C and D). In contrast, the percentage of live bacteria within vesicles in infected LCFSN cells dropped from 87 to 55% over the 4-h incubation period. Thus, a majority of live S. aureus internalized by CFT-1 cells escaped by 4 h after infection into the cytoplasm, where they are protected from the hydrolases found within the acidic lumen of the mature lysosome.
A number of recent reports have shown that the cellular process of autophagy is important for the immune detection and clearance of intracellular bacteria that reside within the host cell cytosol (8). As part of the process, cytoplasmic bacteria are engulfed into double-membrane-bound vesicles that are targeted for fusion with the lysosome. In our EM experiments, we did not detect any evidence of autophagy occurring in either of the cell lines as a result of bacterial internalization. Therefore, it is unlikely that autophagy plays a major role in controlling S. aureus replication within either the LCFSN or the CFT-1 cell line.
DISCUSSION
Mutations in CFTR, particularly the F508 mutation, compromise the innate immunity of the lung and predispose CF patients to chronic bacterial infections (1). As a result, chronic inflammation of the lung results in tissue damage and reduced pulmonary function, culminating in death. Previously, we have shown that S. aureus can replicate within pulmonary epithelial cells derived from a patient with a homozygous mutation (F508/F508) in CFTR (20). Persistence in an intracellular location would protect S. aureus from antibiotics and the host innate immune response, thus facilitating rapid recolonization and recurrent infections that are common after termination of antibiotic therapy (17).
The role of CFTR in the clearance of bacteria from the lung is controversial (7, 29). Our studies demonstrated that similar numbers of S. aureus bacteria adhered to the surface of both CFT-1 and the complemented cell line LCFSN (Fig. 1F), suggesting that the adherence of S. aureus to epithelial cell lines is not influenced by the F508 mutation in CFTR. Indeed, we have found that S. aureus showed reduced adhesion to both cell lines in the presence of soluble fibronectin, thus implicating fibronectin as the major ligand for bacterial binding (data not shown). This was also corroborated by reports from other investigators that fibronectin-binding protein is probably the bacterial adhesin that interacts with cell-surface bound fibronectin (27, 31).
Most bacteria internalized by epithelial cells enter the endocytic pathway, followed by fusion with the lysosomal compartment and subsequent degradation by lysosomal hydrolases. However, a number of intracellular pathogens have been shown to alter the host cell endocytic trafficking to avoid fusion with the lysosomal compartment. We have shown that after internalization S. aureus enters the endocytic pathway of CFT-1 or LCFSN cells and, within 1 h, acquired markers indicative of a late endosomal/lysosomal compartment. Thereafter, the fate of live internalized S. aureus diverges significantly between the two cell lines. In CFT-1 cells, S. aureus escapes into the host cell cytosol and thus avoids fusion and prolonged exposure to the acidic pH and hydrolytic enzymes within the lysosome. The sequential decrease in the number of bacteria associated with LysoTracker in CFT-1 cells at 1 h after infection (Fig. 5), followed by a decrease in integral membrane proteins such as LAMP-1, suggests that the integrity of the vesicles is compromised early or that there is a defect in endosomal/lysosomal maturation. EM showed that S. aureus bacteria are no longer associated with vesicular membranes within infected CFT-1 cells at 2 h postinfection, thus suggesting that the membrane was probably degraded to enable escape. This contrasts with heat-killed bacteria, which remained within membrane-bound vesicles and are subsequently degraded. Once free in the cytoplasm, replication proceeded more efficiently (Fig. 1A), providing a possible explanation for the persistence of S. aureus within CFT-1 cells but not in the complemented LCFSN cell line.
Although most of the endosomal escape events in CFT-1 cells, as defined by EM, occurred by 2 h after infection, the alteration in endosomal pH in CFT-1 cells compared to LCFSN happened even at 1 h, as demonstrated by LysoTracker studies (Fig. 5). This finding, coupled with a lack of similar observation in CFT-1 cells infected with heat-killed bacteria, suggests that S. aureus, in a proper host environment (i.e., within CFT-1 but not LCFSN), likely affects normal maturation of vesicles, including a relative increase of endosomal pH in CFT-1 cells. However, this defect in vesicular trafficking cannot be attributed to the CFTR mutation alone since active lysosomal hydrolases are delivered to vesicles containing dead but fluorescent bacteria in both CFT-1 and LCFSN cells, thus resulting in progressive loss of fluorescence over time in both cell lines (data not shown). We also investigated whether an increase in endosomal pH in CFT-1 would facilitate bacterial replication. However, this did not seem to be the case, since neutralization of vesicular pH by preincubating epithelial cells in the presence of bafilomycin A (V-ATPase inhibitor) or the weak base NHCl4 has no affect on the ability of S. aureus to replicate in either cell line (unpublished observation). This suggests that the ability of S. aureus to replicate within CFT-1 cell line is independent of the host vesicular pH and that exposure to an acidic environment is not required for virulence gene induction prior to escape into the host cell cytosol.
Our studies with L. monocytogenes also provided us with a close approximation of endosomal pH within CFT-1 cells. It has been shown that activation of listeriolysin O (LLO), a protein that functions optimally at pH 5.5 to 6, is required for endosomal escape (3). The fact that L. monocytogenes is able to replicate in both CFT-1 and LCFSN cell lines suggests that the pH of endocytic vesicles must be at least within the optimal pH required for LLO function (Fig. 2).
We have shown in the present study that S. aureus could replicate within the CFT-1 cell line but not in the complemented cell line LCFSN for up to 10 h postinfection (Fig. 1A). Further evidence of bacterial replication was also confirmed by microscopic analysis, with bacterial clusters of 1 to 2 and 4 to 10 microorganisms after 1 and 4 h in CFT-1 cells, respectively (Fig. 1B). In contrast to CFT-1 cells, live bacteria internalized by LCFSN cells did not cause any gross morphological changes to host cells over the 10-h period of the experiment. Whether the bacteria within LCFSN cells are truly static or reach an equilibrium between replication and death remains to be defined.
The ability of S. aureus to replicate in CFT-1 cells appears to be specific because the nonpathogen B. subtilis, while able to enter CFT-1 cells (Fig. 2), failed to replicate in these cells. As a positive control, the well-studied intracellular pathogen L. monocytogenes was able to invade and replicate both CFT-1 and LCFSN cells effectively. This contrasts with S. aureus, which replicates only in CFT-1 cells and not in LCFSN cells. Thus, the ability to replicate within the CFT-1 cell line but not the complemented cell line LCFSN seems to be specific for S. aureus and not just a generalized phenomenon applicable to a random assortment of bacteria.
Based on studies of Listeria LLO, we speculate that a similar hemolysin may be responsible for endosomal escape in S. aureus. However, in contrast to Listeria, S. aureus has an arsenal of membrane pore-forming toxins including , , , and toxins, leukotoxins (PVL toxin), and phospholipase. Although the toxin of S. aureus may seem to be a reasonable candidate based on work in bovine mammary epithelial cells (30), it is possible that these toxins may be host cell specific, as has been the case with leukotoxins. We are currently in the process of investigating the role of these toxins in mediating endosomal escape in CFT-1 cells.
Predicated upon our data, we have a model whereby S. aureus enters the normal endocytic/degradative pathway within both CFT-1 and LCFSN cells and, at some point after acquisition of normal endocytic markers such as LAMP-1, the bacteria within the CFT-1 cell escape into the cytoplasm of the host cell, thus avoiding degradation within the lysosomal compartment (Fig. 8). This difference in endosomal trafficking between the CFT-1 and LCFSN cell lines may well account for the ability of S. aureus to replicate within the CFT-1 cell line and not the LCFSN cell line. This ability of S. aureus to escape into the host cell cytosol only in CFT-1 cells, but not in the complemented counterpart, may conceivably explain the persistence of S. aureus in CF patients. Understanding the underlying mechanisms of escape and replication will have important bearings on the therapeutic treatment of S. aureus infections in CF patients.
ACKNOWLEDGMENTS
We thank Louisa Howard of the Rippel Electron Microscope Facility for EM work; the Developmental Hybridoma Studies Bank, University of Iowa, for LAMP-1 and LAMP-2 antibodies; B. Kahl for initial work in this system; the Cystic Fibrosis Core facility at Dartmouth Medical School; Dan Portnoy for the L. monocytogenes strain; and S. Sato for the anti-Vacuolar-ATPase antibody (OSW2).
This study was supported by a research grant from the CF Foundation.
REFERENCES
1. Bals, R., D. J. Weiner, and J. M. Wilson. 1999. The innate immune system in cystic fibrosis lung disease. J. Clin. Investig. 103:303-307.
2. Bayles, K. W., C. A. Wesson, L. E. Liou, L. K. Fox, G. A. Bohach, and W. R. Trumble. 1998. Intracellular Staphylococcus aureus escapes the endosome and induces apoptosis in epithelial cells. Infect. Immun. 66:336-342.
3. Beauregard, K. E., K. D. Lee, R. J. Collier, and J. A. Swanson. 1997. pH-dependent perforation of macrophage phagosomes by listeriolysin O from Listeria monocytogenes. J. Exp. Med. 186:1159-1163.
4. Burns, J. L., J. Emerson, J. R. Stapp, D. L. Yim, J. Krzewinski, L. Louden, B. W. Ramsey, and C. R. Clausen. 1998. Microbiology of sputum from patients at cystic fibrosis centers in the United States. Clin. Infect. Dis. 27:158-163.
5. Burton, P. R., N. Kordova, and D. Paretsky. 1971. Electron microscopic studies of the rickettsia Coxiella burnetii: entry, lysosomal response, and fate of rickettsial DNA in L-cells. Can. J. Microbiol. 17:143-150.
6. Cystic Fibrosis Foundation. 2003. Patient registry annual report. Cystic Fibrosis Foundation, Washington, D.C.
7. Darling, K. E., A. Dewar, and T. J. Evans. 2004. Role of the cystic fibrosis transmembrane conductance regulator in internalization of Pseudomonas aeruginosa by polarized respiratory epithelial cells. Cell Microbiol. 6:521-533.
8. Deretic, V. 2005. Autophagy in innate and adaptive immunity. Trends Immunol. 26:523-528.
9. Desjardins, M., J. E. Celis, G. van Meer, H. Dieplinger, A. Jahraus, G. Griffiths, and L. A. Huber. 1994. Molecular characterization of phagosomes. J. Biol. Chem. 269:32194-32200.
10. Dramsi, S., I. Biswas, E. Maguin, L. Braun, P. Mastroeni, and P. Cossart. 1995. Entry of Listeria monocytogenes into hepatocytes requires expression of inIB, a surface protein of the internalin multigene family. Mol. Microbiol. 16:251-261.
11. Fratti, R. A., J. M. Backer, J. Gruenberg, S. Corvera, and V. Deretic. 2001. Role of phosphatidylinositol 3-kinase and Rab5 effectors in phagosomal biogenesis and mycobacterial phagosome maturation arrest. J. Cell Biol. 154:631-644.
12. Gaillard, J. L., P. Berche, J. Mounier, S. Richard, and P. Sansonetti. 1987. In vitro model of penetration and intracellular growth of Listeria monocytogenes in the human enterocyte-like cell line Caco-2. Infect. Immun. 55:2822-2829.
13. Goering, R. V., A. Bauernfeind, W. Lenz, and B. Przyklenk. 1990. Staphylococcus aureus in patients with cystic fibrosis: an epidemiological analysis using a combination of traditional and molecular methods. Infection 18:57-60.
14. Govan, J. R., and J. W. Nelson. 1992. Microbiology of lung infection in cystic fibrosis. Br. Med. Bull. 48:912-930.
15. Havell, E. A. 1986. Synthesis and secretion of interferon by murine fibroblasts in response to intracellular Listeria monocytogenes. Infect. Immun. 54:787-792.
16. Hess, D. J., M. J. Henry-Stanley, E. A. Erickson, and C. L. Wells. 2003. Intracellular survival of Staphylococcus aureus within cultured enterocytes. J. Surg. Res. 114:42-49.
17. Hoiby, N. 1995. Microbiology of cystic fibrosis. Chapman and Hall, Ltd., London, United Kingdom.
18. Hutchison, M. L., and J. R. Govan. 1999. Pathogenicity of microbes associated with cystic fibrosis. Microbes Infect. 1:1005-1014.
19. Kahl, B., M. Herrmann, A. S. Everding, H. G. Koch, K. Becker, E. Harms, R. A. Proctor, and G. Peters. 1998. Persistent infection with small colony variant strains of Staphylococcus aureus in patients with cystic fibrosis. J. Infect. Dis. 177:1023-1029.
20. Kahl, B. C., M. Goulian, W. van Wamel, M. Herrmann, S. M. Simon, G. Kaplan, G. Peters, and A. L. Cheung. 2000. Staphylococcus aureus RN6390 replicates and induces apoptosis in a pulmonary epithelial cell line. Infect. Immun. 68:5385-5392.
21. Karunasagar, I., B. Senghaas, G. Krohne, and W. Goebel. 1994. Ultrastructural study of Listeria monocytogenes entry into cultured human colonic epithelial cells. Infect. Immun. 62:3554-3558.
22. Lin, J. T., M. B. Connelly, C. Amolo, S. Otani, and D. S. Yaver. 2005. Global transcriptional response of Bacillus subtilis to treatment with subinhibitory concentrations of antibiotics that inhibit protein synthesis. Antimicrob. Agents Chemother. 49:1915-1926.
23. Lowy, F. D. 2000. Is Staphylococcus aureus an intracellular pathogen Trends Microbiol. 8:341-343.
24. Lowy, F. D. 1998. Staphylococcus aureus infections. N. Engl. J. Med. 339:520-532.
25. Lukacs, G. L., O. D. Rotstein, and S. Grinstein. 1990. Phagosomal acidification is mediated by a vacuolar-type H+-ATPase in murine macrophages. J. Biol. Chem. 265:21099-21107.
26. Menzies, B. E., and I. Kourteva. 1998. Internalization of Staphylococcus aureus by endothelial cells induces apoptosis. Infect. Immun. 66:5994-5998.
27. Mongodin, E., O. Bajolet, J. Cutrona, N. Bonnet, F. Dupuit, E. Puchelle, and S. de Bentzmann. 2002. Fibronectin-binding proteins of Staphylococcus aureus are involved in adherence to human airway epithelium. Infect. Immun. 70:620-630.
28. Parida, S. K., E. Domann, M. Rohde, S. Muller, A. Darji, T. Hain, J. Wehland, and T. Chakraborty. 1998. Internalin B is essential for adhesion and mediates the invasion of Listeria monocytogenes into human endothelial cells. Mol. Microbiol. 28:81-93.
29. Pier, G. B. 2000. Role of the cystic fibrosis transmembrane conductance regulator in innate immunity to Pseudomonas aeruginosa infections. Proc. Natl. Acad. Sci. USA 97:8822-8828.
30. Shompole, S., K. T. Henon, L. E. Liou, K. Dziewanowska, G. A. Bohach, and K. W. Bayles. 2003. Biphasic intracellular expression of Staphylococcus aureus virulence factors and evidence for Agr-mediated diffusion sensing. Mol. Microbiol. 49:919-927.
31. Sinha, B., P. Francois, Y. A. Que, M. Hussain, C. Heilmann, P. Moreillon, D. Lew, K. H. Krause, G. Peters, and M. Herrmann. 2000. Heterologously expressed Staphylococcus aureus fibronectin-binding proteins are sufficient for invasion of host cells. Infect. Immun. 68:6871-6878.
32. Vergne, I., J. Chua, and V. Deretic. 2003. Tuberculosis toxin blocking phagosome maturation inhibits a novel Ca2+/calmodulin-PI3K hVPS34 cascade. J. Exp. Med. 198:653-659.
33. Yankaskas, J. R., J. E. Haizlip, M. Conrad, D. Koval, E. Lazarowski, A. M. Paradiso, C. A. Rinehart, Jr., B. Sarkadi, R. Schlegel, and R. C. Boucher. 1993. Papillomavirus immortalized tracheal epithelial cells retain a well-differentiated phenotype. Am. J. Physiol. 264:C1219-C1230.(Todd M. Jarry and Ambrose)
ABSTRACT
Staphylococcus aureus is frequently the initial bacterium isolated from young cystic fibrosis (CF) patients, and yet its role in CF disease progression has not been determined. Recent data from our lab demonstrates that S. aureus can invade and replicate within the CF tracheal epithelial cell line (CFT-1). Here we describe the finding that the fate of internalized S. aureus in CFT-1 cells differs from its complemented counterpart (LCFSN). S. aureus strain RN6390 was able to replicate within the mutant CFT-1 cells after invasion but not in the complemented LCFSN cells. At 1 h postinvasion, S. aureus containing vesicles within both cell lines acquired vacuolar-ATPase, lysosomal markers LAMP 1 and 2, and the lysomotrophic dye LysoTracker to a similar degree. However, at 4 h postinvasion, the percentage of S. aureus within CFT-1 cells associated with these markers decreased significantly compared to LCFSN, where the association approached 100%. Transmission electron microscopic analysis revealed that the majority of bacteria within CFT-1 cells were free in the cytosol at 4 h after invasion, whereas most S. aureus bacteria internalized by LCFSN cells remained within vesicles. These results demonstrate a fundamental difference in the fate of live S. aureus after invasion of CFT-1 versus LCFSN cell lines and may explain the propensity of S. aureus to cause chronic lung infection in CF patients.
INTRODUCTION
Cystic fibrosis (CF) is the most common lethal autosomal-recessive disorder, affecting 1 in 3,500 newborns each year in the United States (6). Patients with CF are susceptible to respiratory tract infections at an early age, with the most frequent initial isolate being Staphylococcus aureus (4, 14, 18). Infections with S. aureus usually precede chronic colonization of the respiratory tract by the opportunistic pathogen Pseudomonas aeruginosa and continue into adulthood, when 51% of patients become culture positive for S. aureus. Administration of antibiotics is often ineffective at eradicating S. aureus from the CF lung and generally results in relapsing infections that are clonal in nature (13, 19). Although the role of S. aureus in the disease progression of CF remains unknown (6, 17), the clinical observation of rapid recolonization soon after treatment suggests that S. aureus may persist inside host cells of the lung (23), possibly explaining the propensity for recurrent S. aureus infection in CF patients.
Bacteria internalized by nonphagocytic host cells are generally processed in a membrane-bound vesicle that matures through a series of fusion steps to yield a fully mature phagolysosome characterized by a low pH and containing lysosomal hydrolases. Recent studies suggest that a growing number of bacterial pathogens have evolved to circumvent the normal endocytic trafficking pathways within the host cell. Phagocytosed Mycobacterium tuberculosis arrests maturation of its endosomal compartment by inhibiting the acquisition of early endosomal antigen 1, a protein involved in maturation of early endosomes into late endosomes, thus avoiding degradation within the lysosomal compartment (11, 32). Listeria monocytogenes, in contrast, escapes from the phagosomal vesicle minutes after internalization by secreting the pore-forming toxin listeriolysin O (12). Some bacterial species, such as Coxiella burnetii, can withstand the harsh lysosomal environment and replicate within the lysosomal compartment (5). In the case of S. aureus, which is classically considered as an extracellular pathogen, accumulating experimental evidence suggests that S. aureus can act as an intracellular pathogen, capable of invading and surviving within a broad range of nonphagocytic cells, including enterocytes, osteocytes, and endothelial and epithelial cells (2, 16, 20, 24, 26). Recent data from our lab demonstrated that S. aureus can invade and replicate within a CF tracheal (CFT-1) epithelial cell line derived from a CF patient homozygous for the F508 mutation of CFTR (20). However, the specific pathway by which S. aureus subverts host cell trafficking is not known. Understanding the trafficking of pathogens such as S. aureus within host cells will help us understand the complex interactions between pathogen and hosts.
Accordingly, the objective of the present study was to compare the intracellular trafficking of S. aureus between the CF tracheal epithelial cell line (CFT-1) and the complemented cell line (LCFSN). We found that live S. aureus, once internalized into CFT-1 cells, resided within a vesicle with characteristics of a late endosome prior to its escape into the host cytosol. In contrast, S. aureus that invaded LCFSN cells remained bound within vesicles that retained late endosome markers, leading to inhibition of replication and progressive destruction by the endosomal contents. These data suggest that the F508 mutation in CFTR is associated with a defect in the ability of epithelial cells to control endosomal escape and also a failure to degrade internalized S. aureus effectively, thus allowing for intracellular bacterial replication and subsequent tissue damage in the lung. This finding regarding the interactions of S. aureus with CF airway epithelial cells implies that S. aureus might play a more substantial role in the progression of lung disease in CF patients than what was originally suspected.
MATERIALS AND METHODS
Bacterial strains and growth conditions. For invasion experiments, S. aureus strain RN6390 containing pALC1743, a pSK236-derived shuttle vector containing the S. aureus RNAIII promoter driving the expression of GFPuvr, was grown overnight 37°C in 5 ml of Trypticase soy broth with 10 μg of chloramphenicol/ml (20). Bacillus subtilis strain 168 was grow overnight at 37°C with shaking in Luria-Bertani (LB) broth. The cultures were then diluted 1:100 in 10 ml or respective media and grown at 37°C to an optical density at 650 nm (OD650) of 1.2 (late log phase). L. monocytogenes strain EGD was grown in brain heart infusion overnight shaking at 30°C. Bacteria were washed two times in phosphate-buffered saline (PBS) and resuspended in the invasion medium Dulbecco modified Eagle medium-F-12 (50:50; DMEM/F12; Cellgro) supplemented with 2 mM L-glutamine and 1% fetal bovine serum (Cellgro) at an OD650 of 0.4 (4 x 108 CFU). Bacteria were diluted to achieve a multiplicity of infection (MOI) of 10:1 (bacteria to epithelial cell) in invasion medium. Dilutions of S. aureus were then subjected to sonication for 15 s using a microtip (Branson model LS75) to break up bacterial clusters prior to use in invasion experiments. For the heat-killed control, S. aureus bacteria in invasion media were incubated at 55°C for 1 h.
Cell culture and growth conditions. Two established cystic fibrosis tracheal epithelial cell lines (CFT-1 and LCFSN) were used in the present study. CFT-1 is a papillomavirus-immortalized tracheal epithelial cell line derived from a CF tissue donor homozygous for the F508 mutation of the CFTR gene. The complemented cell line, LCFSN, contains a retroviral vector expressing the wild-type CFTR gene (33). The cell lines were maintained in medium containing DMEM/F12 supplemented with 10% fetal bovine serum and the following supplements (from Sigma unless otherwise noted): 5 μg of insulin/ml, 3.7 μg of endothelial cell growth supplement/ml, 25 ng of epidermal growth factor/ml, 3 x 10–8 M triiodothyronine, 10–6 M hydrocortisone, 5 μg of transferrin/ml, and 10 ng of cholera toxin/ml. For routine passage, the medium containing an antibiotic-antimycotic solution with 100 U of penicillin G/ml, 25 μg of amphotericin B/ml, and 100 μg of streptomycin (Cellgro)/ml was used. Neomycin (150 μg/ml) was added to LCFSN cultures to maintain the CFTR-expressing retroviral vector.
For invasion experiments, epithelial cells (105 cells per well) were seeded onto 24-well plates (Costar) in supplemented DMEM/F12 medium without antibiotics and antimycotics and allowed to grow at 37°C in 5% CO2 for 3 days until confluent (roughly 2 x 105 cells per well). One hour prior to the experiments, cells were washed three times with PBS (Cellgro), followed by the addition of 1 ml of invasion media. At time zero, 500 μl of invasion medium containing bacteria was added at an MOI of 10:1 to each well, followed by incubation for 30 min at 37°C in 5% CO2. Cells were then washed three times with PBS, and 500 μl of invasion media containing 5 μg of lysostaphin (AMBI, Inc.)/ml was added to each well to lyse extracellular and adherent S. aureus. The absence of extracellular bacteria was then confirmed by CFU enumeration on agar plates. For intracellular replication experiments, wells were washed at the desired times three times with PBS, followed by incubation with 500 μl of 0.25% trypsin-0.1% EDTA (Cellgro) at 37°C in 5% CO2. After detachment, the cells were transferred to 500 μl of 0.025% Triton X-100 in H2O, sonicated to release intracellular bacteria, and then plated on Trypticase soy agar for determination of the CFU. For invasion assays with B. subtilis and L. monocytogenes, nonadherent bacteria, after a 1-h incubation, were removed by three washes with PBS, followed by the addition of invasion media containing gentamicin (50 μg/ml) to kill extracellular bacteria (15, 22). CFU enumeration of B. subtilis and L. monocytogenes was done on LB agar plates and brain heart infusion agar plates, respectively.
Immunofluorescence staining and confocal microscopy. Invasion of epithelial cells was done as described above with live or heat-killed S. aureus. At 1, 2, and 4 h postinvasion, monolayers were washed three times with PBS and fixed with 2% paraformaldehyde in PBS at room temperature overnight. For immunofluorescence staining, fixed cells were washed three times with PBS and then immersed in permeabilization buffer (PB) containing PBS, 1% bovine serum albumin, 0.1% sodium azide, and 0.1% saponin for 30 min at room temperature. Antibody solutions were diluted in PB at a 1/100 dilution for LAMP-1 and LAMP-2 antibodies (DSHB, University of Iowa) or a 1/1,000 dilution for anti-V-ATPase antibody (a gift from S. Sato), added to coverslips, and incubated for 1 h at 37°C in 5% CO2. Coverslips were washed three times in PB without saponin and then incubated for 1 h with a 1/40 dilution of goat anti-mouse antibody (Fab')2 conjugated to Alexa Fluor 555 (Molecular Probes). After three washes with PBS, both bacteria and epithelial cell DNA were stained with 1 μM To-Pro-3 (Molecular Probes) for 10 min at room temperature in PBS. Coverslips were then washed three times in PBS, treated with SloFade (Molecular Probes), mounted on glass slides, and examined on a Olympus IX-70 confocal microscope. Images were captured by using Flouview software (Olympus). For quantitative analysis, more than 15 fields of view were examined per slide, and for each condition more than 100 events were analyzed.
Assessment of vesicular pH. To evaluate whether S. aureus-containing vesicles are exposed to the acidic environment of the maturing lysosome, we utilized the lysomotrophic LysoTracker (LyT) DND-99 (Molecular Probes). Briefly, cell lines were grown on glass coverslips in 24-well plates until confluent and then used for subsequent invasion assays with live or heat-killed bacteria as described above. Thirty minutes prior to the time point, the cells were washed and incubated with 1 μM LysoTracker/ml, washed three times PBS, and then fixed with 2% paraformaldehyde. Coverslips were then stained with 1 μM To-Pro-3 (Molecular Probes) and mounted on glass slides for confocal analysis.
EM of S. aureus-infected cells. CFT-1 and LCFSN were grown in six-well plates (Costar) until confluent. Live and heat-killed bacteria were prepared as described above and added to epithelial cells at an MOI of 100:1 to increase the frequency of bacterial internalization for quantitative analysis. The six-well plates were then spun at 200 x g for 10 min at 4°C to maximize the number of bacteria adhering to the cell surface. After incubation for 10 min at 37°C in 5% CO2, the plates were washed three times with PBS and then incubated in invasion medium with 10 μg of lysostaphin/ml. At 1, 2, and 4 h after infection, the cells were washed three times with PBS and fixed with 2% glutaraldehyde-1% paraformaldehyde in 0.1 M sodium cacodylate (pH 7.4) at room temperature. The monolayers were then scraped from the dish and resuspended in 2% glutaraldehyde-1% paraformaldehyde in 0.1 M sodium cacodylate (pH 7.4) overnight at 4°C. Loose pellets were then postfixed in 1% OsO4 in 0.1 M sodium cacodylate (pH 7.4) for 1 h at room temperature, rinsed, blocked, and stained with 2% aqueous uranyl acetate for 30 min at room temperature in the dark. The samples were then dehydrated, immersed in LX112-propylene oxide, and allowed to polymerize at 60°C for 48 h. Sections were cut and stained with uranyl acetate in methanol for 5 min. Samples were examined by using a JEOL 2000FX electron microscope at 100 kV. For quantification, bacteria were considered within a vesicle when a single continuous membrane was visible by electron microscopy (EM).
Statistical analysis. For statistical analysis, the Student t test was performed using JMP 5.0.1.2 (2003). A P value of <0.05 was considered significant. The data for Fig. 2 were log transformed to equalize variance.
RESULTS
Intracellular replication of S. aureus occurs within CFT-1 cell line but not within the complemented cell line LCFSN. Previous work in our laboratory showed that S. aureus strain RN6390 is capable of invading and replicating within CFT-1 cells prior to inducing apoptosis (20). In light of these findings, we wanted to determine whether the ability of S. aureus to replicate intracellularly was due to the F508 mutation of the CFTR gene by comparing bacterial replication in CFT-1 and LCFSN cells. Monolayers of LCFSN or CFT-1 cells were infected with sonicated S. aureus at an MOI of 10:1 in 24-well plates, and the numbers of intracellular bacteria were monitored over time as described in Materials and Methods. The average number of intracellular S. aureus within CFT-1 cells was found to increase from 0.3 to 3 bacteria per epithelial cell over a 9-h period, with an intracellular doubling time of 2 h (Fig. 1A). We were unable to assess CFU for CFT-1 cells beyond the 10-h time point due to detachment of the monolayers, probably due to apoptosis as previously described (20). In contrast, the number of bacteria in LCFSN cells remained relatively constant at 0.03 bacteria per cell over the 10-h period, showing no signs of apoptosis.
Further evidence of bacterial replication was observed via direct microscopic analysis (Fig. 1B). Sonication of the bacterial sample prior to invasion experiments has generally resulted in one to two bacteria per cluster. At 1 h postinvasion, the size of the cluster did not change (Fig. 1B), whereas at 4 h in the CFT-1 cells the majority of the bacterial clusters were in groups of 4 to 10, suggesting that bacterial replication had occurred (Fig. 1B). This was observed infrequently with live bacteria within LCFSN cells and not at all with heat-killed bacteria, as expected (Fig. 1B). These data showed that S. aureus internalized into CFT-1 cells were able to replicate, whereas bacteria that were inside the complemented cell line LCFSN could not.
Adherence and internalization of S. aureus. The ability of S. aureus to replicate in the CFT-1 but not in the LCFSN cell line prompted us to examine whether adherence and internalization differed between the two cell lines. Incubation of bacteria with epithelial cells at an MOI of 10:1 for 30 min resulted in equal numbers of bacteria adherent to CFT-1 and LCFSN cell lines, with 10% of inoculated bacteria adhering to the monolayers (approximately 1 bacterium bound per epithelial cell) (Fig. 1C). This observation suggests that S. aureus is probably not binding directly to CFTR present on the epithelial cell surface.
In internalization assays, CFT-1 cells were found to internalize more than 50% of the adherent bacteria over a 2-h period compared to the LCFSN cells, which internalized only 5% (Fig. 1D). These findings demonstrated an increased rate of internalization of S. aureus in the CFT-1 cells compared to LCFSN cells.
To determine whether bacterial replication within CFT-1 cells was S. aureus specific, we compared the internalization and replication of S. aureus to the well-characterized intracellular pathogen (L. monocytogenes) and a nonpathogenic bacterium (Bacillus subtilis) between these two cell lines. L. monocytogenes has been shown to invade and replicate within a broad range of nonphagocytic cell lines, including hepatocytes and epithelial and endothelial cell lines (10, 21, 28). Although less efficient at being internalized into both cell lines, L. monocytogenes could replicate in both cell lines, as opposed to S. aureus, which was only able to replicate within the CFT-1 cell line (Fig. 2). More importantly, B. subtilis, a noninvasive bacterium was not able to replicate within either of the two cell lines.
Acquisition of late endosomal markers LAMP-1 and LAMP-2. Recent EM data from our lab showed that S. aureus was internalized into a membrane-bound vesicle prior to intracellular replication in CFT-1 cells (20). To examine the fate of S. aureus within this vesicle, we analyzed the acquisition of endosomal markers LAMP-1 and LAMP-2, proteins present in the membrane of late endosomes and lysosomes (9). In both CFT-1 and LCFSN cell lines at 1 h postinvasion, approximately 80% of the S. aureus-containing vesicles were associated with LAMP-1 and LAMP-2, as shown by the presence of LAMP-containing membranes (green circles) surrounding the bacteria (arrow) (Fig. 3A to D). At 2 and 4 h postinvasion, S. aureus within LCFSN cells maintained its association with LAMP-1 and LAMP-2, whereas this association with LAMP-1 and LAMP-2 in CFT-1 cells significantly dropped to 30 and 20%, respectively (Fig. 3E and F).
To determine whether internalized bacteria were within fully formed vesicles, we acquired a z-series of images at 1, 2, and 4 h postinternalization, using the vesicular marker LAMP-2. At 4 h postinvasion, live S. aureus within LCFSN cells (Fig. 4A) resided within a vesicle that stained entirely with the LAMP-2 marker (arrows). In contrast, at 4 h postinvasion, most bacteria within CFT-1 cells appeared as bacterial clusters that were devoid of LAMP-2-associated staining (Fig. 4B, arrowhead) whereas only a small percentage remained within LAMP-2 vesicles (arrow). As a control, heat-killed bacteria internalized into both cell lines acquired both LAMP-1 and LAMP-2 and maintained their association with these markers over the entire course of the experiment, as would be expected for vesicles destined to fuse with lysosomes (Fig. 3E and F, open symbols). Together, these data suggest an anomaly in the trafficking pathway of S. aureus within CFT-1 cells compared to LCFSN.
Acidification of the S. aureus containing vesicle. Maturation of an endocytic vesicle is normally accompanied by a steady decrease in luminal pH (from pH of <6.2 in early endosomes to a pH of <5.5 in lysosomes). The ability of the vesicle to acidify is essential for the optimal activity of hydrolytic enzymes found within the lysosomal compartment. To determine whether S. aureus resides in an acidic vesicle after internalization, we stained infected cells with LysoTracker DND 99, a weak lysomotrophic base that accumulates and fluoresces within acidic vesicles. Approximately 80% of heat-killed bacteria were within acidic vesicles of CFT-1 cells at 1 h postinvasion and approached 100% after 4 h (Fig. 5). Heat-killed bacteria internalized into LCFSN cells were uniformly found to colocalize with LysoTracker-positive vesicles, whereas LCFSN infected with live bacteria were associated with a delayed acidification at 1 h, but by 4 h more than 85% of the bacteria were found to be within acidic vesicles (Fig. 5C). In contrast, only 40% of the live internalized bacteria in CFT-1 cells were associated with acidic vesicles by 1 h, and this percentage significantly decreased to less than 20% after 4 h (Fig. 5C). These data suggest that a majority of live S. aureus bacteria within CFT-1 cells are not within acidic vesicles at 4 h postinvasion. Additional experiments with the lysomotrophic agent acridine orange revealed a similar trend (data not shown).
Association with V-ATPase. A decrease in the luminal pH of endocytic vesicles is mediated by V-ATPase, part of an integral membrane complex that pumps hydrogen ions into the vesicular lumen (25). Using an antibody against V-ATPase, we were able to quantify the percentage of S. aureus-containing vesicles associated with V-ATPase (Fig. 6). Approximately 80% of live bacteria within CFT-1 cells were associated with V-ATPase at 1 h after invasion, but by 4 h the number decreased to 65%. However, the data did not reach statistical significance compared to those obtained at 1 h postinvasion. Nevertheless, we saw a similar trend in repeating this experiment. In contrast, 90% of the live S. aureus cells within LCFSN cells were associated with V-ATPase after 1 h and maintained this association over 4 h. Similarly, 80 and 90% of heat-killed S. aureus cells that were internalized into CFT-1 and LCFSN cells, respectively, were associated with V-ATPase after 1 h of invasion, and this association persisted over 4 h (Fig. 6C).
EM analysis of infected cells. The relative absence of lysosomal marker in vesicles containing live internalized bacteria within CFT-1 cells at 2 h after invasion (Fig. 3E and F) implies that S. aureus either altered the acquisition of endosomal markers or may escape into the host cytosol. To narrow the possibility, we examined monolayers of CFT-1 and LCFSN cells infected with S. aureus at an MOI of 100:1 by EM (Fig. 7). Quantitative EM analysis of more than 100 bacteria per sample revealed that at 1 h postinvasion a majority of live bacteria were still within membrane-bound vesicles in both cell lines (Fig. 7D). After 4 h, evidence of membrane degradation and bacterial escape into the host cytosol was seen in CFT-1 cells infected with live bacteria (Fig. 7A). Between 2 and 4 h, the percentage of bacteria within vesicles in CFT-1 cells infected with live bacteria significantly decreased from 77 to 18%, whereas heat-killed bacteria internalized by CFT-1 cells remained bound within vesicles after 4 h (Fig. 7C and D). In contrast, the percentage of live bacteria within vesicles in infected LCFSN cells dropped from 87 to 55% over the 4-h incubation period. Thus, a majority of live S. aureus internalized by CFT-1 cells escaped by 4 h after infection into the cytoplasm, where they are protected from the hydrolases found within the acidic lumen of the mature lysosome.
A number of recent reports have shown that the cellular process of autophagy is important for the immune detection and clearance of intracellular bacteria that reside within the host cell cytosol (8). As part of the process, cytoplasmic bacteria are engulfed into double-membrane-bound vesicles that are targeted for fusion with the lysosome. In our EM experiments, we did not detect any evidence of autophagy occurring in either of the cell lines as a result of bacterial internalization. Therefore, it is unlikely that autophagy plays a major role in controlling S. aureus replication within either the LCFSN or the CFT-1 cell line.
DISCUSSION
Mutations in CFTR, particularly the F508 mutation, compromise the innate immunity of the lung and predispose CF patients to chronic bacterial infections (1). As a result, chronic inflammation of the lung results in tissue damage and reduced pulmonary function, culminating in death. Previously, we have shown that S. aureus can replicate within pulmonary epithelial cells derived from a patient with a homozygous mutation (F508/F508) in CFTR (20). Persistence in an intracellular location would protect S. aureus from antibiotics and the host innate immune response, thus facilitating rapid recolonization and recurrent infections that are common after termination of antibiotic therapy (17).
The role of CFTR in the clearance of bacteria from the lung is controversial (7, 29). Our studies demonstrated that similar numbers of S. aureus bacteria adhered to the surface of both CFT-1 and the complemented cell line LCFSN (Fig. 1F), suggesting that the adherence of S. aureus to epithelial cell lines is not influenced by the F508 mutation in CFTR. Indeed, we have found that S. aureus showed reduced adhesion to both cell lines in the presence of soluble fibronectin, thus implicating fibronectin as the major ligand for bacterial binding (data not shown). This was also corroborated by reports from other investigators that fibronectin-binding protein is probably the bacterial adhesin that interacts with cell-surface bound fibronectin (27, 31).
Most bacteria internalized by epithelial cells enter the endocytic pathway, followed by fusion with the lysosomal compartment and subsequent degradation by lysosomal hydrolases. However, a number of intracellular pathogens have been shown to alter the host cell endocytic trafficking to avoid fusion with the lysosomal compartment. We have shown that after internalization S. aureus enters the endocytic pathway of CFT-1 or LCFSN cells and, within 1 h, acquired markers indicative of a late endosomal/lysosomal compartment. Thereafter, the fate of live internalized S. aureus diverges significantly between the two cell lines. In CFT-1 cells, S. aureus escapes into the host cell cytosol and thus avoids fusion and prolonged exposure to the acidic pH and hydrolytic enzymes within the lysosome. The sequential decrease in the number of bacteria associated with LysoTracker in CFT-1 cells at 1 h after infection (Fig. 5), followed by a decrease in integral membrane proteins such as LAMP-1, suggests that the integrity of the vesicles is compromised early or that there is a defect in endosomal/lysosomal maturation. EM showed that S. aureus bacteria are no longer associated with vesicular membranes within infected CFT-1 cells at 2 h postinfection, thus suggesting that the membrane was probably degraded to enable escape. This contrasts with heat-killed bacteria, which remained within membrane-bound vesicles and are subsequently degraded. Once free in the cytoplasm, replication proceeded more efficiently (Fig. 1A), providing a possible explanation for the persistence of S. aureus within CFT-1 cells but not in the complemented LCFSN cell line.
Although most of the endosomal escape events in CFT-1 cells, as defined by EM, occurred by 2 h after infection, the alteration in endosomal pH in CFT-1 cells compared to LCFSN happened even at 1 h, as demonstrated by LysoTracker studies (Fig. 5). This finding, coupled with a lack of similar observation in CFT-1 cells infected with heat-killed bacteria, suggests that S. aureus, in a proper host environment (i.e., within CFT-1 but not LCFSN), likely affects normal maturation of vesicles, including a relative increase of endosomal pH in CFT-1 cells. However, this defect in vesicular trafficking cannot be attributed to the CFTR mutation alone since active lysosomal hydrolases are delivered to vesicles containing dead but fluorescent bacteria in both CFT-1 and LCFSN cells, thus resulting in progressive loss of fluorescence over time in both cell lines (data not shown). We also investigated whether an increase in endosomal pH in CFT-1 would facilitate bacterial replication. However, this did not seem to be the case, since neutralization of vesicular pH by preincubating epithelial cells in the presence of bafilomycin A (V-ATPase inhibitor) or the weak base NHCl4 has no affect on the ability of S. aureus to replicate in either cell line (unpublished observation). This suggests that the ability of S. aureus to replicate within CFT-1 cell line is independent of the host vesicular pH and that exposure to an acidic environment is not required for virulence gene induction prior to escape into the host cell cytosol.
Our studies with L. monocytogenes also provided us with a close approximation of endosomal pH within CFT-1 cells. It has been shown that activation of listeriolysin O (LLO), a protein that functions optimally at pH 5.5 to 6, is required for endosomal escape (3). The fact that L. monocytogenes is able to replicate in both CFT-1 and LCFSN cell lines suggests that the pH of endocytic vesicles must be at least within the optimal pH required for LLO function (Fig. 2).
We have shown in the present study that S. aureus could replicate within the CFT-1 cell line but not in the complemented cell line LCFSN for up to 10 h postinfection (Fig. 1A). Further evidence of bacterial replication was also confirmed by microscopic analysis, with bacterial clusters of 1 to 2 and 4 to 10 microorganisms after 1 and 4 h in CFT-1 cells, respectively (Fig. 1B). In contrast to CFT-1 cells, live bacteria internalized by LCFSN cells did not cause any gross morphological changes to host cells over the 10-h period of the experiment. Whether the bacteria within LCFSN cells are truly static or reach an equilibrium between replication and death remains to be defined.
The ability of S. aureus to replicate in CFT-1 cells appears to be specific because the nonpathogen B. subtilis, while able to enter CFT-1 cells (Fig. 2), failed to replicate in these cells. As a positive control, the well-studied intracellular pathogen L. monocytogenes was able to invade and replicate both CFT-1 and LCFSN cells effectively. This contrasts with S. aureus, which replicates only in CFT-1 cells and not in LCFSN cells. Thus, the ability to replicate within the CFT-1 cell line but not the complemented cell line LCFSN seems to be specific for S. aureus and not just a generalized phenomenon applicable to a random assortment of bacteria.
Based on studies of Listeria LLO, we speculate that a similar hemolysin may be responsible for endosomal escape in S. aureus. However, in contrast to Listeria, S. aureus has an arsenal of membrane pore-forming toxins including , , , and toxins, leukotoxins (PVL toxin), and phospholipase. Although the toxin of S. aureus may seem to be a reasonable candidate based on work in bovine mammary epithelial cells (30), it is possible that these toxins may be host cell specific, as has been the case with leukotoxins. We are currently in the process of investigating the role of these toxins in mediating endosomal escape in CFT-1 cells.
Predicated upon our data, we have a model whereby S. aureus enters the normal endocytic/degradative pathway within both CFT-1 and LCFSN cells and, at some point after acquisition of normal endocytic markers such as LAMP-1, the bacteria within the CFT-1 cell escape into the cytoplasm of the host cell, thus avoiding degradation within the lysosomal compartment (Fig. 8). This difference in endosomal trafficking between the CFT-1 and LCFSN cell lines may well account for the ability of S. aureus to replicate within the CFT-1 cell line and not the LCFSN cell line. This ability of S. aureus to escape into the host cell cytosol only in CFT-1 cells, but not in the complemented counterpart, may conceivably explain the persistence of S. aureus in CF patients. Understanding the underlying mechanisms of escape and replication will have important bearings on the therapeutic treatment of S. aureus infections in CF patients.
ACKNOWLEDGMENTS
We thank Louisa Howard of the Rippel Electron Microscope Facility for EM work; the Developmental Hybridoma Studies Bank, University of Iowa, for LAMP-1 and LAMP-2 antibodies; B. Kahl for initial work in this system; the Cystic Fibrosis Core facility at Dartmouth Medical School; Dan Portnoy for the L. monocytogenes strain; and S. Sato for the anti-Vacuolar-ATPase antibody (OSW2).
This study was supported by a research grant from the CF Foundation.
REFERENCES
1. Bals, R., D. J. Weiner, and J. M. Wilson. 1999. The innate immune system in cystic fibrosis lung disease. J. Clin. Investig. 103:303-307.
2. Bayles, K. W., C. A. Wesson, L. E. Liou, L. K. Fox, G. A. Bohach, and W. R. Trumble. 1998. Intracellular Staphylococcus aureus escapes the endosome and induces apoptosis in epithelial cells. Infect. Immun. 66:336-342.
3. Beauregard, K. E., K. D. Lee, R. J. Collier, and J. A. Swanson. 1997. pH-dependent perforation of macrophage phagosomes by listeriolysin O from Listeria monocytogenes. J. Exp. Med. 186:1159-1163.
4. Burns, J. L., J. Emerson, J. R. Stapp, D. L. Yim, J. Krzewinski, L. Louden, B. W. Ramsey, and C. R. Clausen. 1998. Microbiology of sputum from patients at cystic fibrosis centers in the United States. Clin. Infect. Dis. 27:158-163.
5. Burton, P. R., N. Kordova, and D. Paretsky. 1971. Electron microscopic studies of the rickettsia Coxiella burnetii: entry, lysosomal response, and fate of rickettsial DNA in L-cells. Can. J. Microbiol. 17:143-150.
6. Cystic Fibrosis Foundation. 2003. Patient registry annual report. Cystic Fibrosis Foundation, Washington, D.C.
7. Darling, K. E., A. Dewar, and T. J. Evans. 2004. Role of the cystic fibrosis transmembrane conductance regulator in internalization of Pseudomonas aeruginosa by polarized respiratory epithelial cells. Cell Microbiol. 6:521-533.
8. Deretic, V. 2005. Autophagy in innate and adaptive immunity. Trends Immunol. 26:523-528.
9. Desjardins, M., J. E. Celis, G. van Meer, H. Dieplinger, A. Jahraus, G. Griffiths, and L. A. Huber. 1994. Molecular characterization of phagosomes. J. Biol. Chem. 269:32194-32200.
10. Dramsi, S., I. Biswas, E. Maguin, L. Braun, P. Mastroeni, and P. Cossart. 1995. Entry of Listeria monocytogenes into hepatocytes requires expression of inIB, a surface protein of the internalin multigene family. Mol. Microbiol. 16:251-261.
11. Fratti, R. A., J. M. Backer, J. Gruenberg, S. Corvera, and V. Deretic. 2001. Role of phosphatidylinositol 3-kinase and Rab5 effectors in phagosomal biogenesis and mycobacterial phagosome maturation arrest. J. Cell Biol. 154:631-644.
12. Gaillard, J. L., P. Berche, J. Mounier, S. Richard, and P. Sansonetti. 1987. In vitro model of penetration and intracellular growth of Listeria monocytogenes in the human enterocyte-like cell line Caco-2. Infect. Immun. 55:2822-2829.
13. Goering, R. V., A. Bauernfeind, W. Lenz, and B. Przyklenk. 1990. Staphylococcus aureus in patients with cystic fibrosis: an epidemiological analysis using a combination of traditional and molecular methods. Infection 18:57-60.
14. Govan, J. R., and J. W. Nelson. 1992. Microbiology of lung infection in cystic fibrosis. Br. Med. Bull. 48:912-930.
15. Havell, E. A. 1986. Synthesis and secretion of interferon by murine fibroblasts in response to intracellular Listeria monocytogenes. Infect. Immun. 54:787-792.
16. Hess, D. J., M. J. Henry-Stanley, E. A. Erickson, and C. L. Wells. 2003. Intracellular survival of Staphylococcus aureus within cultured enterocytes. J. Surg. Res. 114:42-49.
17. Hoiby, N. 1995. Microbiology of cystic fibrosis. Chapman and Hall, Ltd., London, United Kingdom.
18. Hutchison, M. L., and J. R. Govan. 1999. Pathogenicity of microbes associated with cystic fibrosis. Microbes Infect. 1:1005-1014.
19. Kahl, B., M. Herrmann, A. S. Everding, H. G. Koch, K. Becker, E. Harms, R. A. Proctor, and G. Peters. 1998. Persistent infection with small colony variant strains of Staphylococcus aureus in patients with cystic fibrosis. J. Infect. Dis. 177:1023-1029.
20. Kahl, B. C., M. Goulian, W. van Wamel, M. Herrmann, S. M. Simon, G. Kaplan, G. Peters, and A. L. Cheung. 2000. Staphylococcus aureus RN6390 replicates and induces apoptosis in a pulmonary epithelial cell line. Infect. Immun. 68:5385-5392.
21. Karunasagar, I., B. Senghaas, G. Krohne, and W. Goebel. 1994. Ultrastructural study of Listeria monocytogenes entry into cultured human colonic epithelial cells. Infect. Immun. 62:3554-3558.
22. Lin, J. T., M. B. Connelly, C. Amolo, S. Otani, and D. S. Yaver. 2005. Global transcriptional response of Bacillus subtilis to treatment with subinhibitory concentrations of antibiotics that inhibit protein synthesis. Antimicrob. Agents Chemother. 49:1915-1926.
23. Lowy, F. D. 2000. Is Staphylococcus aureus an intracellular pathogen Trends Microbiol. 8:341-343.
24. Lowy, F. D. 1998. Staphylococcus aureus infections. N. Engl. J. Med. 339:520-532.
25. Lukacs, G. L., O. D. Rotstein, and S. Grinstein. 1990. Phagosomal acidification is mediated by a vacuolar-type H+-ATPase in murine macrophages. J. Biol. Chem. 265:21099-21107.
26. Menzies, B. E., and I. Kourteva. 1998. Internalization of Staphylococcus aureus by endothelial cells induces apoptosis. Infect. Immun. 66:5994-5998.
27. Mongodin, E., O. Bajolet, J. Cutrona, N. Bonnet, F. Dupuit, E. Puchelle, and S. de Bentzmann. 2002. Fibronectin-binding proteins of Staphylococcus aureus are involved in adherence to human airway epithelium. Infect. Immun. 70:620-630.
28. Parida, S. K., E. Domann, M. Rohde, S. Muller, A. Darji, T. Hain, J. Wehland, and T. Chakraborty. 1998. Internalin B is essential for adhesion and mediates the invasion of Listeria monocytogenes into human endothelial cells. Mol. Microbiol. 28:81-93.
29. Pier, G. B. 2000. Role of the cystic fibrosis transmembrane conductance regulator in innate immunity to Pseudomonas aeruginosa infections. Proc. Natl. Acad. Sci. USA 97:8822-8828.
30. Shompole, S., K. T. Henon, L. E. Liou, K. Dziewanowska, G. A. Bohach, and K. W. Bayles. 2003. Biphasic intracellular expression of Staphylococcus aureus virulence factors and evidence for Agr-mediated diffusion sensing. Mol. Microbiol. 49:919-927.
31. Sinha, B., P. Francois, Y. A. Que, M. Hussain, C. Heilmann, P. Moreillon, D. Lew, K. H. Krause, G. Peters, and M. Herrmann. 2000. Heterologously expressed Staphylococcus aureus fibronectin-binding proteins are sufficient for invasion of host cells. Infect. Immun. 68:6871-6878.
32. Vergne, I., J. Chua, and V. Deretic. 2003. Tuberculosis toxin blocking phagosome maturation inhibits a novel Ca2+/calmodulin-PI3K hVPS34 cascade. J. Exp. Med. 198:653-659.
33. Yankaskas, J. R., J. E. Haizlip, M. Conrad, D. Koval, E. Lazarowski, A. M. Paradiso, C. A. Rinehart, Jr., B. Sarkadi, R. Schlegel, and R. C. Boucher. 1993. Papillomavirus immortalized tracheal epithelial cells retain a well-differentiated phenotype. Am. J. Physiol. 264:C1219-C1230.(Todd M. Jarry and Ambrose)