ABSTRACT
The chlamydial inclusion occupies a unique niche within the eukaryotic cell that does not interact with endocytic compartments but instead is fusogenic with a subset of sphingomyelin-containing exocytic vesicles. The Chlamydia trachomatis inclusion acquires these distinctive properties by as early as 2 h postinfection as demonstrated by the ability to acquire sphingomyelin, endogenously synthesized from 6{N-[(7-nitrobenzo-2-oxa-1,3-diazol-4-yl)amino]caproylsphingosine} (C6-NBD-ceramide). The molecular mechanisms involved in transformation of the properties and cellular interactions of the inclusion are unknown except that they require early chlamydial transcription and translation. Although the properties of the inclusion are established by 2 h postinfection, the degree of interaction with endocytic pathways during the brief interval before fusogenicity with an exocytic pathway is established is unknown. Using a combination of confocal and electron microscopy to localize endocytic and lysosomal markers in C. trachomatis infected cells during the early stages of infection, we demonstrate a lack of these markers within the inclusion membrane or lumen of the inclusion to conclude that the nascent chlamydial inclusion is minimally interactive with endosomal compartments during this interval early in infection. Even when prevented from modifying the properties of the inclusion by incubation in the presence of protein synthesis inhibitors, vesicles containing elementary bodies are very slow to acquire lysosomal characteristics. These results imply a two-stage mechanism for chlamydial avoidance of lysosomal fusion: (i) an initial phase of delayed maturation to lysosomes due to an intrinsic property of elementary bodies and (ii) an active modification of the vesicular interactions of the inclusion requiring chlamydial protein synthesis.
Chlamydia trachomatis is the etiological agent of several significant human diseases, including trachoma, the leading cause of infectious blindness worldwide (31). It is also the most common cause of sexually transmitted disease in the United States (4, 24). Chlamydiae are bacterial obligate intracellular parasites that undergo a unique developmental cycle consisting of functionally and morphologically distinct cell types adapted for extracellular survival and intracellular multiplication. The metabolically inactive extracellular forms are termed elementary bodies (EBs), and the replicating, but noninfectious, forms are called reticulate bodies (20).
The chlamydial developmental cycle takes place within a vacuole, termed an inclusion, that does not fuse with lysosomes. Inhibition of lysosomal fusion is not due to a general inhibition of lysosomal function in chlamydia-infected cells but is limited to the chlamydial inclusion (8, 16). The cellular origins of the chlamydial inclusion and its vesicular interactions are only now being elucidated. Markers for fluid-phase endocytosis are not trafficked to the chlamydial inclusion, nor are markers for early or late endosomes localized to the mature chlamydial inclusion membrane (16, 25, 30). Unlike many intracellular parasites that inhabit vesicles with characteristics of distinct compartments in the endosomal pathway (6, 7, 12, 19, 23, 28), the chlamydial inclusion appears to be disconnected from cellular pathways involved in the maturation of endosomes to lysosomes. Although the chlamydial inclusion does not appear to interact with endocytic vesicles, it receives sphingolipids by fusion with exocytic vesicles in transit from the trans-Golgi network to the plasma membrane (14, 15). Interaction with this pathway is common to all species of chlamydiae (22, 34). Sequestration of chlamydiae within a vesicle that intersects an exocytic pathway has been hypothesized to provide a unique intracellular niche in which the chlamydial inclusion is not perceived by the host cell as a vesicle destined to fuse with lysosomes (13).
Establishment of the appropriate vacuolar properties is dependent upon chlamydial modification of the vesicle and appears to be less dependent upon the route of internalization (26). Within a few hours following endocytosis, the properties of the nascent chlamydial inclusion change dramatically. Endocytosed EBs are rapidly transported to the peri-Golgi region of the host cell and become fusogenic with sphingomyelin-containing vesicles from an exocytic pathway, as evidenced by the acquisition of fluorescent sphingomyelin endogenously synthesized from 6{N-[(7-nitrobenzo-2-oxa-1,3-diazol-4-yl)amino]caproylsphingosine} (C6-NBD-ceramide) (14). Transport to the Golgi region and acquisition of fluorescent sphingomyelin are dependent on early chlamydial protein synthesis, suggesting that chlamydiae actively modify the inclusion membrane in order to intersect an exocytic pathway (26). In the absence of chlamydial protein synthesis, the endocytosed EBs are ultimately degraded within lysosomes. Although C. trachomatis becomes interactive with an exocytic pathway by 2 h postinfection, it remains unclear whether the nascent inclusion interacts with endocytic compartments before this change in properties occurs. This study was undertaken in an attempt to identify any interactions of the nascent chlamydial inclusion with host endocytic compartments over the interval between endocytosis and chlamydial modification of the vesicle.
MATERIALS AND METHODS
Cell culture and organisms.Monolayer cultures of HeLa 229 epithelial cells (CCL 1.2; American Type Culture Collection) were grown in RPMI 1640 (GIBCO BRL, Gaithersburg, Md.) supplemented with 10% fetal bovine serum and 10 μg of gentamicin (Whittaker Bioproducts, Walkerville, Md.) per ml at 37°C in an atmosphere of 5% CO2-95% humidified air. C. trachomatis LGV-434, serotype L2, was grown in HeLa 229 cells, and infections were carried out as described previously (3, 15). Yersinia pseudotuberculosis (10) was grown in LB broth overnight at room temperature. Salmonella enterica serovar Typhimurium was kindly provided by O. Steele-Mortimer.
Reagents.C6-NBD-ceramide was obtained from Molecular Probes (Eugene, Oreg.). Human transferrin conjugated to horseradish peroxidase (Tf-HRP) was purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, Pa.). Anti-Tf receptor (anti-TfR) monoclonal antibody (B3/25) was purchased from Boehringer Mannheim Biochemicals (Indianapolis, Ind.). Monoclonal antibody H4A3 to the lysosomal glycoprotein LAMP1 was obtained from the Developmental Studies Hybridoma Bank (Iowa City, Iowa), and monoclonal antibody 2G11 directed against the bovine cation-independent mannose-6-phosphate receptor (M6PR) was generously provided by Suzanne Pfeffer, Department of Biochemistry, Stanford University, Stanford, Calif..
Tf-HRP.Analysis of Tf-HRP trafficking was performed as described by Willingham et al. (33). Specifically, HeLa cells grown on Thermanox coverslips (Nunc, Inc., Naperville, Ill.) were infected with C. trachomatis at a multiplicity of approximately 50 for 30 min at 37°C. Excess EBs were removed by washing with 80 μg of heparin per ml, and at the indicated times postinfection, cells were rinsed three times with Hanks balanced salt solution (HBSS) and incubated for 15 min at 37°C in the presence of 50 μg of human Tf-HRP per ml in serum-free Dulbecco’s modified Eagle medium. The cultures were rinsed three times with HBSS and fixed for 60 min at room temperature in 2% glutaraldehyde in 50 mM sodium phosphate-150 mM NaCl, pH 7.4. The coverslips were rinsed three times in phosphate-buffered saline (PBS) and incubated with ImmunoPure Metal Enhanced DAB (diaminobenzidine) substrate (Pierce Chemical Co.) for 1 h at room temperature. The coverslips were then washed three times with 50 mM Tris-HCl in 7.5% sucrose (pH 7.4) and fixed for an additional 30 min in 4% paraformaldehyde-2.5% glutaraldehyde in 100 mM sodium cacodylate buffer (pH 7.4). Cells were postfixed for 30 min in 1% OsO4-1% K3Fe(CN)6, washed with H2O, dehydrated in a graded ethanol series, and embedded in Spurr's resin. Thin sections were cut with an RMC MT-7000 ultramicrotome (Boeckeler, Tucson, Ariz.), stained with 1% uranyl acetate (aqueous) and Reynold's lead citrate, and observed at 80 kV on a Philips CM-10 electron microscope (FEI, Hillsboro, Oreg.).
Immunohistochemistry.Cells were grown and infected as described above. At the indicated time points, cells were fixed with PLP fixative (2) for 2 h at room temperature, rinsed two times with PBS, and then permeabilized with 0.01% saponin in PBS for 5 min at room temperature. Fixed cells were incubated with monoclonal antibodies diluted in 0.01% saponin in PBS for 1 h. After the cells were washed two times with PBS, FAb sheep anti-mouse immunoglobulin G (Jackson ImmunoResearch Laboratories, Inc., West Grove, Pa.) diluted in 0.01% saponin in PBS was added, and coverslips were incubated for an additional hour at room temperature. The coverslips were then washed three times in PBS and fixed in 1.5% glutaraldehyde in 0.1 M Na cacodylate (pH 7.4)-5% sucrose for 1 h at room temperature. Coverslips were incubated with ImmunoPure Metal Enhanced DAB substrate (Pierce Chemical Co.) and processed for electron microscopy as described above.
Plasma membrane lipids.HeLa 229 cells on Thermanox coverslips were labeled with dipalmitoylphosphatidylethanolamine (DPPE)-nanogold (Nanoprobes, Inc., Stonybrook, N.Y.) according to the manufacturer's instructions prior to infection with C. trachomatis L2. At 30 min postinfection, the cultures were fixed with 4% paraformaldehyde-2.5% glutaraldehyde. The samples were then washed three times with water, and the nanogold was silver enhanced for 4 min with HQ Silver Reagents (Nanoprobes, Inc.) and processed for electron microscopy as described above.
Plasma membrane biotinylation.HeLa 229 cells on Thermanox coverslips were rinsed three times with PBS and surface labeled by incubation for 30 min at room temperature in 0.5 mg of PBS EZ-Link sulfo-N-hydroxysuccinimidyl (NHS)-LC-biotin (Pierce Chemical Co.). The cultures were rinsed three times with PBS prior to infection with C. trachomatis L2 in HBSS for 30 min at 37°C. The medium was replaced with prewarmed RPMI 1640 with 10% fetal calf serum and incubated for an additional 3 h. The cultures were then fixed with PLP fixative for 1.5 h at room temperature, washed twice with PBS, and permeabilized for 5 min at room temperature with 0.01% saponin in PBS. Streptavidin-nanogold (Nanoprobes, Inc) in PBS plus 0.01% saponin was added and incubated at 4°C overnight. The coverslips were then fixed in 4% paraformaldehyde-2.5% glutaraldehyde, silver enhanced, and processed for transmission electron microscopy.
Immunofluorescence and microscopy.Cells grown on 12-mm-diameter coverslips (no. 1 thickness) in 24-well plates were fixed and permeabilized in 100% methanol for 10 min at room temperature. The fixed cells were washed three times with PBS before and after sequential addition of the primary and secondary antibody. All incubations were carried out at room temperature for 60 min. The coverslips were then mounted onto glass slides by using Vectashield (Vector) and viewed. Laser confocal microscopy was performed with a confocal imaging system (MRC-1000; Bio-Rad Laboratories, Hercules, Calif.) equipped with a krypton-argon laser (Bio-Rad Laboratories) on an inverted microscope (Carl Zeiss, Inc., Thornwood, N.Y.) with a 63× objective. Fluorescence micrographs were taken on an FXA photomicroscope (Nikon Inc., Garden City, N.Y.) with either a 20× or a 60× planapochromat objective. Photomicrographs were obtained with T-Max ASA 400 film. Both confocal and immunofluorescent images were processed with Adobe Photoshop 4.0 (Adobe Systems, Inc., Mountain View, Calif.).
C6-NBD-ceramide labeling.C6-NBD-ceramide was complexed with 0.034% defatted bovine serum albumin (DFBSA) in Dulbecco's minimal essential medium (MEM) as described previously (15) to yield complexes of approximately 5 mM in both DFBSA and C6-NBD-ceramide. Mock- and C. trachomatis-infected HeLa cells were incubated with the DFBSA-NBD-ceramide complex at 4°C for 30 min, washed with 10 mM HEPES-buffered calcium- and magnesium-free Puck's saline (pH 7.4), and incubated for various times in MEM plus 0.34% DFBSA to “back exchange” excess probe from the plasma membrane. Cultures on coverslips were rinsed in 10 mM HEPES-buffered calcium- and magnesium-free Puck's saline (pH 7.4) prior to mounting for fluorescence microscopy. For all experiments involving C6-NBD-ceramide trafficking in the presence of inhibitors, monolayers were infected with C. trachomatis for 30 min at 4°C and rinsed with HBSS, and the medium replaced with prewarmed MEM-10 (MEM + 10% fetal bovine serum) with or without chloramphenicol (200 μg/ml). Following incubation at 37°C for the desired interval, the cultures were labeled with C6-NBD-ceramide for 10 min in the presence of inhibitor, rinsed with HBSS, and back exchanged for 1 h in MEM plus 0.34% DFBSA in the presence of inhibitor.
RESULTS
Plasma membrane derivation of the nascent chlamydial inclusion membrane.To confirm the plasma membrane origin of the vesicles within which chlamydiae are endocytosed, HeLa 229 cells were surface labeled with DPPE-nanogold as a lipid marker or with NHS-biotin for host cell surface proteins. The association of plasma membrane markers was examined by transmission electron microscopy at 30 min postinfection for the DPPE-nanogold and 3 h postinfection for the NHS-biotin (Fig. 1). Both lipid and protein markers are present on the nascent inclusion membrane, which is tightly juxtaposed to the internalized EB and remains so until multiplication is initiated.
Host plasma membrane lipids and proteins are constituents of the endosomal vesicle enveloping internalized EBs. (A) HeLa 229 cell plasma membranes were labeled with DPPE-nanogold prior to infection with C. trachomatis L2 and processed for transmission electron microscopy at 30 min postinfection. (B) The plasma membrane of HeLa cells was labeled with NHS-biotin prior to infection with C. trachomatis L2, and biotinylated proteins were detected with avidin-nanogold at 3 h postinfection. Electron-dense deposits after silver enhancement are visible in the nascent inclusion membrane. Arrowheads indicate intracellular EBs. Bars, 0.5 μm.
Characterization of interactions with recycling endosomes.In initial experiments to determine if the nascent chlamydial inclusion interacted with early endosomes, Tf-HRP was added simultaneously with the chlamydial inoculum such that it was present during the internalization of EBs. In those experiments, a small percentage of EBs were observed in vesicles containing Tf-HRP when the cultures were examined immediately after internalization (Fig. 2). Note that the distribution of the Tf-HRP gives the appearance of budding off the developing endosome even as the endosome is forming. The structure of the Tf-HRP-containing vesicles budding off at the base of the forming endosome is very similar to that of coated vesicles described by Reynolds and Pearce (21). Once EBs are internalized, Tf-HRP is rarely detected in EB-containing endosomes. Because of the small percentage of EBs observed in vesicles containing Tf, even at very short intervals postentry, it is difficult to determine if this represents the characteristic rapid recycling of Tf from early endosomes or whether only a small proportion of EBs enter coincidentally with Tf.
Electron micrographs demonstrating the presence of Tf-HRP associated with C. trachomatis EBs at the point of entry. (A) The Tf-HRP distribution gives the appearance of budding off of the developing endosome even as the endosome is forming (white arrowheads). (B) Internalized EBs (black arrowheads) do not display the marker.
To determine whether the nascent chlamydial inclusion is fusogenic with early endosomes after internalization, EBs were allowed to attach and enter HeLa cells, surface-associated EBs were removed by a brief heparin treatment, and Tf-HRP was added to viable infected cells at various times postinfection. Immediately after infection and at 1 and 2 h postinfection, cells were incubated with Tf-HRP for 15 min and processed for electron microscopy (Fig. 3). Reaction product is not associated with EB-containing vesicles, nor are early endosomes observed adjacent to individual membrane-bound chlamydiae until after 1 h postinfection. The nascent inclusion thus appears to be nonfusogenic with recycling endosomes. By 2 h postinfection, however, tubular endosomes containing Tf-HRP are observed closely associated with individual membrane-bound chlamydiae. As has been previously described for the mature chlamydial inclusion (1, 25), reaction product is not observed within the lumen of the inclusion or associated with the inclusion membrane itself. Instead, tubular endosomes containing Tf-HRP are closely associated with these nascent inclusions but do not fuse with the inclusion membrane. Incubation in the presence of chloramphenicol, which inhibits the synthesis of chlamydial early gene products necessary to establish fusogenicity with sphingomyelin-containing exocytic vesicles (26), also prevents the association of early inclusions with tubular endosomes. The concentration of transferrin-containing vesicles in close apposition to the chlamydial inclusion thus appears to require chlamydial modification of the vesicle and to be conferred concomitantly with initiation of the ability to fuse with sphingomyelin-containing exocytic vesicles.
Association of nascent chlamydial inclusions with recycling endosomes at various times early in infection. C. trachomatis EBs were allowed to attach and enter before incubation with Tf-HRP. Electron-dense reaction product is visible within tubular endosomes. At 0 (A) and 1 (B) h postinfection, no association with recycling endosomes is evident. By 2 h postinfection (C), tubular endosomes (arrowheads) are tightly juxtaposed to the nascent inclusion, although no fusion with these vesicles is evident. Inhibition of chlamydial protein synthesis by chloramphenicol prevents the accretion of recycling endosomes near the EBs (D). Bar, 0.5 μm
Early endosomal antigen 1.The nascent inclusions were also stained for EEA1, another marker for early endosomes. No association with EEA1 was observed over a 1-h interval postinfection (Fig. 4). In contrast, distinct circumferential staining was observed with S. enterica serovar Typhimurium as a positive control (29).
Association of nascent chlamydial inclusions with EEA1 at various time postinfection. (A) HeLa 229 cells were infected with C. trachomatis L2 and, at various times postinfection (in minutes), costained for EBs and EEA1 (shown as green and red, respectively, in the merged image). (B) S. enterica serovar Typhimurium was used as a positive control, demonstrating circumferential staining with anti-EEA1 at 15 min postinfection. The corresponding Nomarski differential interference contrast (DIC) image is also shown. Arrowheads indicate the Salmonella-containing vacuole.
Analysis of the early chlamydial inclusion for endosomal and lysosomal markers.To further determine if vesicles containing recently endocytosed EBs display markers of early or late endosomes or lysosomes, cells were infected at a high multiplicity of infection and stained at intervals up to 4 h postinfection for TfR, cation-independent M6PR, and the lysosomal glycoprotein LAMP1 for localization by confocal microscopy. At no point were any of these markers detected in the nascent inclusion membrane (data not shown). By 2 to 4 h postinfection, the EBs are clustered at the peri-Golgi of the host cell in a region in which vesicles bearing the endosomal markers TfR and M6PR are also concentrated. Because of the close juxtaposition of the endocytic vesicles with the chlamydia-containing vesicles, we confirmed the absence of endocytic markers in the inclusion membrane by immunoelectron microscopy with HRP-conjugated secondary antibodies. At 4 h postinfection, a time at which the chlamydiae are concentrated in proximity to the TfR and M6PR, no labeling of the inclusion membranes was detected (Fig. 5).
Immunoperoxidase staining of endocytosed C. trachomatis EBs at 4 h postinfection visualized by electron microscopy. (A) TfR. (B) M6PR. (C) LAMP1. (D) Negative control in which the primary antibody was anti-TfR but the secondary antibody conjugated to HRP was omitted. Arrowheads identify several EBs in each panel. Bar, 0.2 μm.
Disruption of endosomal distribution at high multiplicities of infection.When chlamydiae are internalized at high multiplicities of infection and the distribution of TfR is analyzed, there is a distinct concentration of TfR in the vicinity of the chlamydiae. This atypical distribution of TfR coincides temporally with the trafficking of EBs to the peri-Golgi region. EB trafficking early in infection is most easily visualized at high multiplicities of infection. To determine whether the atypical concentration of TfR that we observed was due the high multiplicities used in these experiments, cells were infected with increasing multiplicities of infection (approximately 1 to 100 EBs/cell), and the distribution of TfR was determined by immunofluorescence microscopy (Fig. 6). A minimal effect on the distribution of TfR vesicles is observed at low multiplicities of infection. However, at high multiplicities of infection, the convergence of recycling endosomes with EBs is markedly enhanced. Infection with high multiplicities of EBs is not, in itself, sufficient to cause the aggregation of TfR, since treatment of cells with chloramphenicol to prevent trafficking of EBs to the peri-Golgi region also prevents aggregation of TfR (data not shown).
Effect of multiplicity of infection on the redistribution of TfR. HeLa cells were infected at multiplicities of infection of 0 (uninfected control) (A and D), 10 (B and E), or 100 (C and F) EBs/cell and simultaneously stained by for localization of EBs with a fluorescein isothiocyanate-conjugated anti-rabbit secondary antibody (A, B, and C) and for TfR with a Texas Red-conjugated anti-mouse secondary antibody (D, E, and F) at 4 h postinfection. Arrows in panels D and E indicate typical organization of early endosomes expressing TfR concentrated in a peri-Golgi region. Arrowheads in panel F identify concentrations of TfR in the same regions as aggregates of early inclusions. Bar, 10 μm.
Rate of delivery of inhibited EBs to lysosomes.Even when prevented from modifying the vesicle by inhibition of protein synthesis, vesicles containing EBs appear to be very slow to mature to lysosomes (11, 26). We determined rates of lysosomal fusion with vesicles containing chloramphenicol-inhibited C. trachomatis EBs in comparison to Y. pseudotuberculosis (Fig. 7). Whereas over 50% of endocytosed Y. pseudotuberculosis was observed in LAMP1-positive vesicles by 30 min postinfection and over 80% was observed in LAMP1-positive vesicles by 1 h, vesicles containing C. trachomatis EBs were very slow to acquire this marker. Acquisition of the LAMP1 marker by vesicles containing chloramphenicol-inhibited EBs showed a half-time of approximately 12 to 24 h.
Rate of lysosomal fusion with chloramphenicol-inhibited EBs. HeLa cells were infected with C. trachomatis L2 EBs or Y. pseudotuberculosis and at various times postinfection were fixed and stained for immunoelectron microscopy with an anti-LAMP1 monoclonal antibody as a lysosomal marker. (A) Immunoelectron microscopy of chloramphenicol-inhibited C. trachomatis L2 EBs and Y. pseudotuberculosis stained for LAMP1. Arrows in left panel identify EBs staining positive for LAMP1; arrowheads indicate EBs negative for LAMP1. (B) Analysis of LAMP1-positive vacuoles containing chloramphenicol-inhibited C. trachomatis L2 EBs and Y. pseudotuberculosis. A minimum of 100 bacteria were scored at each time point, and the percentage showing evidence of the lysosomal marker was plotted. C. trachomatis permitted to replicate normally (no antibiotic) showed no lysosomal marker when examined at either 12 or 24 h postinfection (not shown).
Requirement for de novo chlamydial protein synthesis to regulate vesicle fusion.By 2 h postinfection, internalized EBs modify the endocytic vesicle to establish a functional interaction with an exocytic pathway which transports endogenously synthesized sphingomyelin from the Golgi apparatus to the plasma membrane (14). This initiation of fusion with sphingomyelin-containing exocytic vesicles occurs before the EBs have visibly increased in size to that of reticulate bodies and requires chlamydial transcription and translation (26). To confirm the requirement for chlamydial protein synthesis and test the stability of the interaction with that pathway, C. trachomatis EBs were permitted to enter and initiate development for 2 h and then were inhibited by incubation in the presence of chloramphenicol for an additional 6 h. As shown in Fig. 8, once the functional interaction with this exocytic pathway is established, the inclusion continues to interact with exocytic vesicles as demonstrated by the acquisition of fluorescent derivatives of C6-NBD-ceramide. Thus, once the chlamydial gene products that control vesicle fusion are expressed and the properties of the vesicle are established, functional interaction with this exocytic pathway continues for at least several hours after inhibition of chlamydial protein synthesis.
Fusion of C. trachomatis inclusions with sphingomyelin-containing vesicles requires chlamydial early protein synthesis. HeLa 229 cells were infected at a high multiplicity of infection with chloramphenicol (CAP) present over the intervals indicated. At 8 h postinfection, the cultures were either labeled with C6-NBD-ceramide (A, B, and C) or fixed and stained by immunofluorescence with an anti-L2 EB antiserum (D, E, and F). When incubated for the entire times in the presence of chloramphenicol, EBs do not acquire the ability to obtain sphingomyelin from the host cell and remain dispersed throughout the cytoplasm (C and F). Those cultures permitted to synthesize protein normally for 2 h prior to the addition of chloramphenicol become fusogenic with sphingomyelin-containing vesicles and are trafficked normally to the peri-Golgi region of the host cell. Arrowheads in panels A and B identify peri-Golgi-localized aggregates that have incorporated NBD-sphingomyelin, derived from the metabolism of NBD-ceramide (15) and detected by fluorescence microscopy. Bar, 10 μm.
The chlamydial polypeptides governing vesicular interactions are not stable, however, since extended incubation in the presence of chloramphenicol ultimately results in a cessation of sphingolipid acquisition by chlamydiae (Fig. 9). If C. trachomatis-infected cells are treated with chloramphenicol at 18 h postinfection and incubation is continued for an additional 24 h before C6-NBD-ceramide labeling, the chlamydiae within antibiotic-treated cultures exhibit a markedly diminished ability to acquire sphingomyelin from the host cell in comparison to untreated cultures. Staining of the Golgi apparatus is not affected. The chlamydial early gene products that control vesicle fusion thus appear to turn over and must be replenished to maintain the properties of the inclusion.
Fusion of C. trachomatis inclusions with sphingomyelin-containing vesicles requires continued protein synthesis. (A and B) HeLa 229 cells infected with C. trachomatis L2 were treated at 18 h postinfection with chloramphenicol (CAP) (B) (200 μg/ml) or mock treated (A) and incubated for an additional 24 h before being labeled with C6-NBD-ceramide. (C) A Nomarski image of the infected cell in panel B, showing the intact inclusion. Arrowheads indicate the location of the inclusion. Bar, 10 μm.
Lysosomal fusion does not correlate with loss of ability to acquire sphingomyelin.In experiments similar to those described above, C. trachomatis EBs were used to infect susceptible HeLa cell monolayers and the cultures were allowed to develop for 18 h. Chloramphenicol was then added, and the cultures were fixed and stained for the late endosomal-lysosomal marker LAMP1 at 24-h intervals for an additional 72 h (Fig. 10). Although morphological changes were apparent in the chlamydial developmental forms such that they appeared to be degenerating, no evidence of lysosomal marker in the inclusion membrane was detected. The delayed fusion of chloramphenicol-inhibited C. trachomatis inclusions is similar to that observed with C. pneumoniae (1).
Acquisition of lysosomal markers after inhibition of chlamydial protein synthesis does not occur concurrently with loss of fusogenicity with sphingomyelin-containing vesicles. HeLa 229 cells infected with C. trachomatis L2 were treated at 18 h postinfection with chloramphenicol (200 μg/ml) and incubated in the continuous presence of the inhibitor. Cultures were costained for C. trachomatis (αEB) or LAMP1 at the times indicated after addition of chloramphenicol. Corresponding Nomarski differential interference contrast (DIC) images are also shown. Arrowheads indicate the location of the inclusions.
DISCUSSION
Based upon the absence of the lysosomal enzyme acid phosphatase, the chlamydial inclusion has for some time been considered nonlysosomal in character (11, 17, 36). More recently, these observations have been confirmed and extended by the application of probes for various markers characterizing discrete steps in the pathway of maturation of endosomes to lysosomes (16, 25, 30, 32). Application of these contemporary probes to description of the mature chlamydial inclusion has led to the recognition that the inclusion is not fusogenic with endosomal or lysosomal compartments. The mature chlamydial inclusion is instead fusogenic with a subset of exocytic vesicles identified on the basis of their delivery of NBD-sphingomyelin, endogenously synthesized from C6-NBD-ceramide, to the chlamydial inclusion. Interaction with this pathway of sphingomyelin export is initiated by 2 h postinfection by a process that requires chamydial transcription and translation (14, 26). The properties of the vesicle in the interval between endocytosis and initiation of fusion with sphingomyelin-containing vesicles was examined here. Based upon the inability to detect delivery of Tf to the inclusion, the absence of TfR and M6PR, and a delayed maturation of vesicles containing chloramphenicol-inhibited EBs to LAMP1 positivity, it appears that the nascent inclusion is not fusogenic to any significant degree with compartments of the endocytic pathway. The collective results suggest two distinct mechanisms for chlamydial avoidance of lysosomal fusion: (i) an initial delay in acquisition of lysosomal markers that is reflected in restricted interactions with endosomal compartments and appears to be mediated by EB components, since protein synthesis is not required, and (ii) an active modification of the inclusion membrane such that it becomes fusogenic with sphingomyelin-containing vesicles and remains nonfusogenic with lysosomes throughout the remainder of the developmental cycle.
The coalescence of early endosomal markers in close apposition to the chlamydial inclusion membrane that we observed is consistent with the observations of van Ooij et al. (32), although there are qualitative differences that we were able to resolve by electron microscopy. At an ultrastructural level, we observe no evidence of fusion with endosomal compartments bearing Tf, TfR, or M6PR. The concentration of recycling endosomes adjacent to the mature inclusion membrane (25) is apparent by 2 to 4 h postinfection in the immediate vicinity of the endocytosed EBs, requires chlamydial early protein synthesis, and is most readily distinguished at high multiplicities of infection. Because transport and concentration of endocytosed EBs to a perinuclear location require chlamydial protein synthesis (26) and likely involve host cytoskeletal components (5, 18), it is probable that the concentration of endosomes observed near the chlamydial inclusion is due to redistribution of cytoskeletal components to effect a redistribution of vesicles involved in receptor recycling. The association of tubular endosomes containing Tf with early inclusions is similar to that observed with 18-h inclusions (25) in that there is a very close association of these vesicles with the inclusion but no evidence of Tf in the inclusion membrane itself or within the lumen of the inclusion. We conclude that, although chlamydial infection may result in a morphological redistribution of recycling endosomes in infected cells, the inclusion is not fusogenic with the endocytic compartment even at very early times postinfection.
The necessity for chlamydial modification of the inclusion membrane to establish fusogenicity with sphingomyelin-containing vesicles was confirmed here by inhibition of chlamydial protein synthesis at various times postinfection. If EBs are given an initial 2-h interval in which they synthesize protein and modify the cellular interactions of the inclusion, the inclusions are recognized and trafficked by the host cell to a perinuclear location and become fusogenic with sphingomyelin-containing exocytic vesicles. This interaction remains active for at least 6 h following inhibition of chlamydial protein synthesis, indicating that once interactions with this pathway are established, they remain active for some period of time. These interactions do not remain effective indefinitely, however, since prolonged incubation in the presence of chlamydial protein synthesis inhibitors results in an eventual loss of ability to acquire sphingomyelin from the host cell. These results imply that the chlamydial proteins mediating fusogenicity with sphingomyelin-containing vesicles turn over or are otherwise unstable and must be continually replenished by the chlamydiae. The implication is that specific interactions and active transport to the inclusion are required.
Our model of the chlamydial developmental cycle hypothesizes a sequence of events whereby the chlamydiae are sequentially triggered to transcribe a cascade of early, mid-cycle, and late genes (27) in response to a changing series of intracellular microenvironments as the inclusion matures. We envision this maturation occurring through a combination of host cell activities and modulation of those activities by the chlamydiae. The developmental cycle is initiated as the metabolically quiescent EBs are endocytosed by the host cell, where they are tightly bound within a membranous vesicle. Within this vesicle, unknown environmental signals trigger an initial round of chlamydial transcription. One or more of these early gene products modifies the interactions with the host cell as evidenced by the initiation of fusion with sphingomyelin-containing exocytic vesicles. A consequence of this change in fusion competence of the initial endocytic vesicle is predicted to be a change in the microenvironment of the inclusion as the inclusion begins to fuse with sphingomyelin-containing exocytic vesicles. Fusion with this class of vesicles deliver both host lipids (14, 35) and, presumably, the luminal contents of these vesicles to the inclusion such that the resultant changes in the inclusion environment may provide the signal to trigger subsequent sets of developmentally expressed genes.
The results presented here suggest that even very early in infection there is minimal interaction of the nascent chlamydial inclusion with endocytic vesicles. Such a finding is consistent with the delayed lysosomal fusion of vesicles containing even chloramphenicol-inhibited EBs (11). Indeed, even chlamydial cell wall fractions display a similar, slow acquisition of lysosomal markers, suggesting that structural components of the chlamydial cell wall may mediate this activity (9). Delayed maturation of the endocytic vesicle may be expected to give EBs a survival advantage within the host cell until the chlamydiae actively modify their intracellular environment by means that require early protein synthesis.
ACKNOWLEDGMENTS
We thank H. Caldwell, K. Fields, R. Carabeo, S. Grieshaber, N. Grieshaber, D. Clifton, and K. Wolf for review of the manuscript. We also thank J. Sager for technical assistance.
FOOTNOTES
- Received 9 August 2002.
- Returned for modification 10 October 2002.
- Accepted 4 November 2002.
- Copyright © 2003 American Society for Microbiology
