INFECTION AND IMMUNITY, Apr. 2009, p. 1285–1292 0019-9567/09/$08.00⫹0 doi:10.1128/IAI.01062-08 Copyright © 2009, American Society for Microbiology. All Rights Reserved.
Vol. 77, No. 4
Host Complement Regulatory Protein CD59 Is Transported to the Chlamydial Inclusion by a Golgi Apparatus-Independent Pathway䌤 Ayako Hasegawa,1† L. Farah Sogo,1† Ming Tan,2 and Christine Su ¨tterlin1* Department of Developmental and Cell Biology1 and Departments of Microbiology & Molecular Genetics and Medicine,2 University of California, Irvine, Irvine, California 92697-2300 Received 26 August 2008/Returned for modification 30 September 2008/Accepted 13 January 2009
Chlamydia is an obligate intracellular bacterium that grows and replicates inside a cytoplasmic inclusion. We report that a host protein, CD59, which regulates complement function at the surfaces of uninfected cells, can be detected at the membrane of the chlamydial inclusion. This localization to the inclusion membrane was specific for CD59 and not a general feature of other glycosylphosphatidylinositol (GPI)-anchored proteins or representative cell surface proteins. Using differential permeabilization studies, we showed that CD59 is localized to the luminal but not the cytoplasmic face of the inclusion membrane, consistent with membrane association via its GPI anchor. Furthermore, CD59 was present at the inclusion even when we prevented it from associating with membrane microdomains via the GPI anchor or when we inhibited general protein transport to the cell surface, indicating that a conventional Golgi apparatus-dependent trafficking mechanism was not involved. Based on these findings, we propose that selected host proteins are trafficked to the inclusion by a Golgi apparatus-independent pathway during a Chlamydia infection. Chlamydia is a genus of obligate intracellular bacteria that cause a wide range of human diseases. Chlamydia trachomatis is a leading cause of sexually transmitted infection and preventable blindness (48). Chlamydia pneumoniae (also called Chlamydophila pneumoniae) causes community-acquired pneumonia and has been associated with atherosclerotic heart disease (22). Despite the differences in tissue tropism and disease manifestation, all chlamydial infections share fundamental features at the level of an infected cell, including invasion of the host cell and growth and replication of chlamydiae within a membrane-bound cytoplasmic compartment called the chlamydial inclusion (44). The membrane of the inclusion contains host-derived lipids and a limited number of specific proteins. Cholesterol and sphingolipids are transported to the chlamydial inclusion, and acquisition of these host lipids appears to be essential for chlamydial growth (7, 16, 50). Other lipids, such as glycerophospholipids and the multivesicular body-specific lipid LBPA, have also been detected in the inclusion (4, 53). A number of inclusion membrane proteins of chlamydial origin have been described (12, 26, 29, 38, 42, 46), including IncA, which is involved in homotypic fusion of chlamydial inclusions (17). In contrast, only a small number of host proteins have been detected in the inclusion. For instance, the inclusion lacks endosomal and lysosomal proteins, which has been used as strong evidence to show that this compartment is not part of the endocytic pathway (45, 49). A few host proteins, including specific Rab GTPases (12, 42, 43) and the pathogen recognition molecules Toll-like receptor 2 (TLR2) and MyD88 (32), have been found to associate with the cytoplasmic face of the
inclusion membrane. The multivesicular body proteins CD63 and MLN64 have also been detected in the inclusion (4). The mechanisms by which specific host lipids and proteins are transported to or excluded from the chlamydial inclusion are incompletely understood. Hackstadt and colleagues have shown that cholesterol and sphingomyelin are transported to the chlamydial inclusion in Golgi apparatus-derived vesicles (15). Glycerophospholipids, in contrast, reach the inclusion in a vesicle-independent manner (53). A third mechanism of lipid delivery to the inclusion was recently shown to involve lipid droplets (10). In contrast, the only protein trafficking pathway that has been described is the delivery of CD63 to the inclusion via multivesicular bodies (4). In this study, we examined whether host proteins that are sorted and trafficked together with cholesterol and sphingolipids in an uninfected cell are transported to the inclusion during an infection. We focused on glycosylphosphatidylinositol (GPI)-anchored proteins, which associate with cholesteroland sphingolipid-enriched microdomains (51) and are cosorted in the trans-Golgi network for transport to the cell surface (6). We found that CD59, a host protein with a role in regulating complement function (20), localized to the inclusion during a chlamydial infection. CD59 was present in the inclusion membrane and faced the lumen of the inclusion. Unexpectedly, however, transport of CD59 to the inclusion did not involve conventional post-Golgi apparatus trafficking routes of the host cell. Thus, our findings provide support for a Golgi apparatus-independent trafficking pathway by which host proteins are selectively delivered to the chlamydial inclusion membrane.
* Corresponding author. Mailing address: Department of Developmental and Cell Biology, University of California, Irvine, 3128 Natural Sciences I, Irvine, CA 92697-2300. Phone: (949) 824-7140. Fax: (949) 824-4709. E-mail:
[email protected]. † A.H. and L.F.S. contributed equally to this work. 䌤 Published ahead of print on 21 January 2009.
MATERIALS AND METHODS Antibodies and reagents. The following antibodies used in this study were generously provided by the indicated individuals: anti-CD59 (Yusuke Maeda, Osaka University, Japan), polyclonal rabbit antibodies to Chlamydia muridarum elementary bodies (Ellena Peterson, UC Irvine), mouse anti-IncA antibodies (Dan Rockey, Oregon State University), mouse and rabbit polyclonal antibodies
1285
1286
HASEGAWA ET AL.
to C. trachomatis IncA (Guangming Zhong, University of Texas Health Science Center, San Antonio, TX), anti-CD46 (Magdalene So, Oregon Heath and Sciences University), and anti-vesicular stomatitis virus glycoprotein (anti-VSV-G) antibodies (Vivek Malhotra, Center for Genomic Regulation, Barcelona, Spain). Antibodies to hemagglutinin (HA) and FLAG were purchased from Roche Diagnostics Corporation (Indianapolis, IN) and Sigma-Aldrich (St. Louis, MO), respectively. Additional reagents used in this study were obtained from the following sources: fluorochrome-conjugated secondary antibodies and the DNA dye Hoechst 33342 were obtained from Molecular Probes/Invitrogen, Eugene, OR; nocodazole and brefeldin A were obtained from EMD Chemicals, Gibbstown, NJ; and U18666A, LY294002 hydrochloride, and filipin were obtained from Sigma, St. Louis, MO. Cell culture. HeLa cells (ATCC, Manassas, VA) were grown on glass coverslips in 24-well plates in advanced Dulbecco’s modified Eagle’s medium (Invitrogen, Carlsbad, CA) supplemented with 2% fetal bovine serum and 2 mM GlutaMAX-I (Invitrogen). CHO cells stably expressing CD59 and DAF (CHO-K1 3B2A) and CHO-K1 3B2A cells which are deficient in PigL or in both PGAP2 and PGAP3 were generously provided by Yusuke Maeda, Osaka, Japan. These cells were cultured in Ham’s F-12 medium (Invitrogen) supplemented with 10% fetal bovine serum. Expression constructs. FLAG-CD59, HA-CD59, HA-human placental alkaline phosphatase (HA-PLAP), and PigL were obtained from Yusuke Maeda, Osaka University, Japan; GFP-p75NTR was obtained from Enrico Rodriguez Boulan, Cornell Medical School; FLAG-folate receptor was obtained from Chiara Zurzolo, Institute Curie, Paris, France; GFP–VSV-G was obtained from Jennifer Lippincott Schwartz, NIH; and GFP-PKD1-KD and GFP-PKD2-KD were obtained from Charles Yeaman, University of Iowa. Chlamydia infection and transfection. Monolayers of HeLa or CHO cells were infected with Chlamydia trachomatis serovar L2 (L2/434/Bu), biovar LGV, at a multiplicity of infection of 3 by centrifugation at 2,000 rpm (700 ⫻ g) in a Beckman Allegra 6 centrifuge for 1 h at room temperature. After centrifugation, the inoculum was replaced by 500 l of fresh cell culture medium and the monolayers were incubated at 37°C and 5% CO2. In studies where a protein was expressed in the host cells, the expression construct was transfected at 1 h postinfection (hpi) with FuGENE 6 (Roche) according to the manufacturer’s protocol, except for GFP-PKD1-KD and GFP-PKD2-KD, which were transfected at 2 hpi. In studies with pharmacologic inhibitors, host cells were treated from 1 hpi until fixation at 20 hpi with 0.3 g/ml brefeldin A, 0.05 g/ml nocodazole, 100 M LY294002, or 20 M U18666A. In the case of brefeldin A, fresh drug was also added at 13 hpi. Immunofluorescence microscopy. For indirect immunofluorescence assays, infected cells were fixed at 20 hpi in 3.7% paraformaldehyde in phosphate-buffered saline for 30 min at room temperature. Cells were permeabilized and blocked in 0.25% saponin and 2.5% fetal bovine serum for 30 min, followed by staining with primary antibodies for 1 h and with secondary antibodies for 30 min. Host and Chlamydia DNAs were stained with Hoechst 33342 (Molecular Probes/Invitrogen). Cells were imaged with a Zeiss Axiovert 200 M microscope and analyzed with linear adjustments with Zeiss Axiovision software. Confocal images were produced using a Zeiss LSM 510 Meta two-photon laser scanning confocal microscope. The thickness of each image was smaller than 0.9 m. For differential permeabilization studies, paraformaldehyde-fixed cells were permeabilized for 30 min at 0, 0.025, 0.05, 0.1, 0.2, or 0.3% saponin in blocking solution (2.5% fetal bovine serum in phosphate-buffered saline). Chlamydiae were stained with anti-Chlamydia antibodies to assess the degree of permeabilization of the inclusion membrane. Inclusions were visualized with the DNA dye Hoechst 33342, which stains the nucleoids of chlamydiae and host cell nuclei. Cholesterol at the inclusion membrane was visualized by staining fixed Chlamydia-infected cells on coverslips with 0.05% filipin for 60 min at room temperature. To determine the percentage of infected cells with CD59 at the inclusion membrane, infected cells were first identified by staining with a Chlamydiaspecific antibody. These cells were also stained with the anti-CD59 antibody, and the percentage of infected cells with CD59 at the inclusion was determined. At least 100 infected cells were analyzed per experiment. These assays were repeated as three independent experiments, and a mean and standard deviation were calculated. VSV-G transport assay. HeLa cells were transfected with the GFP-VSV-G (tsO45) plasmid (19) in the presence or absence of constructs expressing kinaseinactive forms of protein kinase D isoform 1 (PKD1-KD) or isoform 2 (PKD2KD) (55). The cells were incubated at 37°C for 12 h and then shifted to 40°C for 4 h to accumulate GFP–VSV-G in the endoplasmic reticulum (ER). Cells were released from the temperature block by incubating them at 32°C for 4 h. The
INFECT. IMMUN.
FIG. 1. CD59 localizes to the inclusion of a Chlamydia-infected cell. (A) Endogenous CD59 in Chlamydia-infected HeLa cells was detected at chlamydial inclusions with antibodies to CD59. All inclusions are marked by asterisks, while an inclusion that did not stain with CD59 is labeled with an additional arrow. CD59 at the plasma membrane is marked with two parallel arrows. (B) Transiently expressed FLAG-tagged CD59 was detected with antibodies to the FLAG epitope at the inclusion and the cell surface in transfected infected cells, but only at the cell surface in transfected uninfected cells. FLAGCD59 at the plasma membrane is marked with two parallel arrows. The two untransfected cells in this image indicate the level of nonspecific background staining with this antibody. Bar, 10 m.
monolayer was fixed and stained with a VSV-G-specific antibody to determine VSV-G localization by immunofluorescence microscopy.
RESULTS CD59 localizes to the membrane of the chlamydial inclusion. To investigate if host proteins associated with membrane microdomains are trafficked to the inclusion, we examined the intracellular distribution of the well-studied GPI-anchored protein CD59 during chlamydial infection. We detected CD59 at the chlamydial inclusion by immunofluorescence with CD59-specific antibodies in ⬎50% of infected HeLa cells (Fig. 1A). CD59 was also present on the plasma membrane in infected and uninfected cells (Fig. 1A and B), which is consistent with its known function as a cell surface protein. We detected CD59 at the inclusion at various times in the chlamydial infection, from 16 hpi, when the inclusion was small, to 48 hpi, when the inclusion occupied most of the cytoplasmic space (data not shown). We next examined the specificity of CD59 staining at the chlamydial inclusion. We used antibodies to the FLAG epitope to detect CD59 at the inclusions of infected HeLa cells that transiently expressed FLAG-tagged CD59 (Fig. 1B). We also observed CD59 at the inclusion in another cell line by examining infected CHO cells stably expressing CD59 (Fig. 2A), which is a cell line commonly used for studying GPI-anchored proteins (30). We performed a third specificity study with cells in which the expression of CD59 could be regulated. We used CHO cells with a mutation in the PigL gene, which encodes a critical component of the GPI anchor synthesis machinery (30). Although these cells stably express CD59, the defect in GPI anchor synthesis prevents the transfer of the GPI anchor to CD59, which leads to the degradation of the protein in the ER (data not shown). GPI anchor synthesis can be restored,
VOL. 77, 2009
FIG. 2. Localization of CD59 to the inclusion of CHO cells is specific. (A) CHO cells stably expressing CD59 were infected with Chlamydia and stained with antibodies to CD59 (green) and Chlamydia (red) and with the DNA dye Hoechst 33342 (blue). (B) Similar infections were performed with PigL-deficient CHO cells stably expressing CD59 and transiently transfected with a PigL expression vector (upper panels) and with untransfected PigL-deficient cells in which stably expressed CD59 was degraded due to a defect in GPI anchor synthesis (lower panels). The cells were stained with antibodies to CD59 (green), the C. trachomatis inclusion membrane protein IncA, or Chlamydia (red) and with the DNA dye Hoechst 33342 (blue). The inclusions of cells expressing both PigL and CD59 are marked by arrows, while the inclusions of cells that did not express PigL are marked by asterisks. Bar, 10 m.
however, by transfection of a PigL expression vector (30). When we infected these cells with Chlamydia and transfected them with the PigL vector, we detected staining of the inclusion with CD59-specific antibodies in transfected cells only (Fig. 2B, upper panels). Untransfected PigL-deficient cells, in contrast, did not show any specific CD59 staining (Fig. 2B, lower panels). These studies demonstrate that the signal for CD59 at the inclusion is specific and is not due to nonspecific cross-reactivity of our antibody with the inclusion. We next investigated the exact localization of CD59 at the chlamydial inclusion, focusing on the inclusion membrane because of the established membrane association of CD59 via its GPI anchor. CD59 (Fig. 3A, upper panels) showed a similar distribution to that of IncA, a chlamydial protein that is a known integral component of the inclusion membrane (39). We confirmed these findings by confocal microscopy, which showed that CD59 and IncA both stained a distinct rim around the inclusion (Fig. 3A, lower panels). CD59 also colocalized at the inclusion membrane with filipin, a fluorescent compound that binds to cholesterol, a host molecule that is incorporated into the inclusion membrane and individual chlamydiae (Fig. 3B) (7). In summary, these studies demonstrate that the host protein CD59 localizes to the membrane of the chlamydial inclusion through a process that involves its GPI anchor.
HOST PROTEINS AT THE CHLAMYDIAL INCLUSION
1287
FIG. 3. CD59 localizes to the membrane of the inclusion. (A) HeLa cells infected with C. trachomatis were stained with antibodies to CD59 and IncA (upper panels), with additional images obtained by confocal microscopy (lower panels). In the merged images, CD59 is shown in green, IncA is shown in red, and chlamydial and host cell DNA, stained by the Hoechst dye 33342, is shown in blue. Bar, 10 m. (B) Chlamydia-infected HeLa cells stained with antibodies to CD59 and the fluorescent cholesterol-binding compound filipin. In the merged image, CD59 is shown in green, filipin is shown in blue, and chlamydiae stained with anti-Chlamydia antibodies are shown in red. Inclusions in which CD59 and filipin costained are marked by arrows. Bar, 10 m.
CD59 localizes to the luminal face of the inclusion membrane. Since GPI-anchored proteins are located on the exoplasmic side of cellular membranes, we performed differential permeabilization experiments to determine the topology of CD59 at the inclusion membrane. We first fixed infected cells with paraformaldehyde, which perforates the plasma membrane and allows antibodies to access the cytoplasm. We then treated the fixed cells with 0.1% saponin, which we used to permeabilize the inclusion membrane. If we left out the saponin treatment, the inclusions did not stain with Chlamydiaspecific antibodies, showing that the inclusion membrane was intact after paraformaldehyde fixation (Fig. 4A, upper panels). After treatment with 0.1% saponin, about 75% of inclusions stained with Chlamydia-specific antibodies, demonstrating that these inclusions had been permeabilized (Fig. 4A, lower panels). We detected CD59 only at the inclusions of cells that also stained with Chlamydia-specific antibodies, indicating that this pool of CD59 is sequestered within an intact inclusion and requires permeabilization of the inclusion membrane for exposure to antibodies (compare permeabilized and nonpermeabilized cells in Fig. 4A, lower panels). In contrast, we detected IncA staining with specific antibodies regardless of saponin treatment, showing that IncA is exposed to the cytoplasm (Fig. 4B, upper and lower panels). The proportion of inclusions that was permeabilized, as determined by staining for chlamydiae, was dependent on the saponin concentration, ranging from
1288
HASEGAWA ET AL.
INFECT. IMMUN.
FIG. 5. Transport of CD59 to the chlamydial inclusion is not a general phenomenon of plasma membrane proteins. Confocal images showing Chlamydia-infected HeLa cells stained with specific antibodies to IncA and the HA tag to visualize transiently expressed HA-CD59 (A) and the GPI-anchored protein HA-hPLAP (B). In similar studies, GFP-tagged versions of the apical marker p75NTR (C) and the basolateral marker VSV-G (D) were detected by fluorescence. In the merged images on the right, IncA is shown in red, HA- or GFP-tagged host proteins are shown in green, and DNA, stained by the Hoechst dye 33342, is shown in blue. Bar, 10 m.
FIG. 4. CD59 localizes to the luminal face of the inclusion membrane. Differential permeabilization studies were conducted in which HeLa cells were infected with Chlamydia, fixed with paraformaldehyde, and treated with various concentrations of saponin to permeabilize the inclusion membrane. (A) Epifluorescence images of cells that were either left untreated or treated with 0.1% saponin prior to staining with antibodies against CD59 and Chlamydia. These cells were also stained with the DNA dye Hoechst 33342 to reveal the organization of the host cell nuclei and chlamydial nucleoids. Arrows indicate inclusions that stained for both Chlamydia and CD59, whereas asterisks mark Chlamydia-positive, CD59-negative inclusions. Bar, 10 m. (B) Same as panel A, except that cells were stained with antibodies to CD59 and IncA. (C) Percentages of cells with inclusions that stained with antibodies against Chlamydia or CD59 over a range of saponin concentrations. (D) Percentages of inclusions staining with antibodies against Chlamydia that also stained for CD59 at different saponin concentrations. These assays were repeated as three independent experiments, and means and standard deviations were calculated.
about 20% with 0.03% saponin to about 90% with 0.3% saponin treatment (Fig. 4C). However, the percentage of permeabilized, chlamydia-staining inclusions at which CD59 was detected was fairly constant, at about 50% over this range of
saponin concentrations (Fig. 4D). These topology studies demonstrate that CD59 faces the lumen of the inclusion and is not in contact with the host cell cytosol. Trafficking of CD59 to the inclusion is selective. We next examined if other cell surface proteins could be transported to the inclusion. We expressed selected host proteins as HA- or green fluorescent protein (GFP)-tagged fusion proteins and detected them by confocal microscopy with specific antibodies to the HA epitope or by GFP fluorescence, respectively. Using this approach, we confirmed that HA-tagged CD59 colocalized at the inclusion membrane with IncA (Fig. 5A). In contrast, two other HA-tagged GPI-anchored proteins, human PLAP (Fig. 5B) and folate receptor (data not shown), did not localize to the inclusion membrane and were found only on the surfaces of infected cells (Fig. 5B). Host integral membrane proteins that are transported to the cell surface in a GPI-independent manner were also not detected at the inclusion. These proteins included GFP-tagged versions of p75NTR (Fig. 5C) (25) and VSV-G (Fig. 5D) (19), which localize to the apical and basolateral cell surfaces, respectively, and an apically targeted isoform of CD46 (8; data not shown). In summary, we found that the localization of CD59 to the chlamydial inclusion
VOL. 77, 2009
HOST PROTEINS AT THE CHLAMYDIAL INCLUSION
1289
CD59 transport to the inclusion membrane (Fig. 6B). We observed similar results with kinase-inactive PKD2 (data not shown). From these studies, we concluded that CD59 delivery to the inclusion does not involve microdomains or PKD-dependent post-Golgi apparatus transport routes. To further investigate the mechanism of CD59 transport to the inclusion, we used pharmacologic inhibitors to disrupt Golgi apparatus organization and function or to block protein transport via multivesicular bodies. Neither brefeldin A, which causes the relocation of Golgi membranes into the ER (27), nor nocodazole, which induces the formation of Golgi ministacks at ER exit sites (11), had an effect on CD59 localization to the inclusion membrane (Fig. 7A and B). In addition, inhibitors of phosphatidylinositol 3-kinase activity (LY294002) and cholesterol transport (U18666A), which each prevent multivesicular body formation, did not block CD59 transport to the inclusion and may even have caused a slight increase in CD59 at the inclusion. Both of these compounds reduced the size of the inclusion (data not shown), as previously reported (4). These results suggest that CD59 is transported to the chlamydial inclusion by a pathway that does not require normal Golgi apparatus organization and function or the formation of multivesicular bodies. DISCUSSION FIG. 6. CD59 does not require canonical post-Golgi apparatus transport routes for localization to the inclusion. (A) Wild-type CHO cells (upper panels) and CHO cells deficient in both PGAP2 and PGAP3 (lower panels) were infected with C. trachomatis and stained with antibodies to CD59 and Chlamydia and with the DNA dye Hoechst 33342. Arrows indicate inclusions that stained for both Chlamydia and CD59. Bar, 10 m. (B) Control HeLa cells (upper panels) and HeLa cells transfected with PKD1-KD (lower panels) were infected with C. trachomatis and stained with antibodies to CD59 and PKD (not shown) and with Hoechst 33342. To verify the effect of PKD1-KD on protein transport, VSV-G transport was measured in control HeLa cells transfected with the GFP-VSV-G (tsO45) plasmid, in which it trafficked to the cell surface, as expected (top right image), and in kinase-inactive PKD1-expressing cells, in which it accumulated in the Golgi apparatus (bottom right image). Bar, 10 m.
is specific for this particular host protein and is not a general feature of all GPI-anchored proteins or cell surface proteins. CD59 transport to the chlamydial inclusion does not involve a canonical trafficking route from the Golgi apparatus. We next investigated whether CD59 is transported to the chlamydial inclusion via established post-Golgi apparatus trafficking pathways of the host cell. For example, transport of CD59 from the Golgi apparatus to the apical surface in polarized cells requires an association of its GPI anchor with sphingomyelinand cholesterol-enriched microdomains (6, 13). However, we still detected CD59 at the inclusion membranes of infected PGAP2/PGAP3-deficient cells (Fig. 6A), which have a defect in GPI anchor remodeling that prevents association of GPIanchored proteins with microdomains (28). We also analyzed whether transport to the inclusion involved transport routes to the cell surface that require the activity of the PKD isoforms PKD1 and PKD2 (55). Cells expressing a dominant-negative, kinase-inactive form of PKD1 (PKD1-KD) blocked VSV-G trafficking to the cell surface, as expected, but did not affect
In this study, we show that the complement regulatory protein CD59 is present in the chlamydial inclusion membrane of an infected cell. CD59 is an abundant cell surface protein that protects cells against uncontrolled complement-mediated lysis by regulating the assembly of the membrane attack complex at the terminal step of the complement activation cascade (24). Patients with paroxysmal nocturnal hematuria have erythrocytes that lack CD59 at the cell surface due to a defect in GPI anchor synthesis (41). As a consequence, these erythrocytes are susceptible to destruction by complement, which produces the cardinal features of intravascular hemolysis and anemia observed in these patients. CD59 can be coopted by a microbial pathogen, as shown by incorporation of this protein into human immunodeficiency virus and human cytomegalovirus envelopes, which provides protection to the virus from complement-mediated lysis (47). Serum-resistant strains of Borrelia burgdorferi express a CD59-like molecule on the bacterial surface that inhibits complement (34). However, CD59 has not previously been associated with a chlamydial infection, and complement activation is not considered a major component of the host immune response against Chlamydia (36). We provide several lines of evidence to demonstrate that the signal for CD59 at the inclusion is specific and not due to an artifact, such as antibody cross-reactivity with chlamydial components in the inclusion. (i) We detected endogenous and tagged forms of CD59 at the inclusion with three independent antibodies in two different cell lines. (ii) We did not detect CD59 labeling at the inclusion in PigL-deficient CHO cells in which GPI anchor synthesis was blocked. (iii) We observed CD59 at the inclusion in only about 50% of infected cells, which indicates that our antibody did not cross-react with chlamydiae in the inclusion of the remaining 50% of infected cells. Our results indicate that transport of CD59 to the inclusion is selective and does not occur by the conventional secretory
1290
HASEGAWA ET AL.
FIG. 7. CD59 delivery to the inclusion does not involve the Golgi apparatus or multivesicular bodies. (A) HeLa cells grown on coverslips were treated at 1 hpi with brefeldin A or nocodazole to disrupt the organization and function of the Golgi apparatus or with LY294002 and U18666A to prevent multivesicular body formation. Cells were stained with antibodies to CD59 and IncA. Arrows indicate inclusions that stained for both CD59 and IncA. The effectiveness of brefeldin A and nocodazole treatments was confirmed on parallel coverslips by visualizing disruption of normal Golgi apparatus organization with antibodies to the Golgi marker protein Golgin97. Bar, 10 m. (B) Percentages of IncA-positive cells in which CD59 was detected at the inclusion after treatment with the above pharmacologic inhibitors. An average of 50 cells were counted for each of the three independent experiments, and the mean and standard deviation were calculated. Fisher’s exact test was used to determine P values for brefeldin A (0.2008), nocodazole (0.2210), LY294002 (0.0002), and U18666A (0.0164).
INFECT. IMMUN.
pathway from the ER via the Golgi apparatus. For example, we did not detect other GPI-anchored proteins or well-studied cell surface markers at the inclusion. We also did not detect CD46, another complement regulatory protein that is a cellular receptor for Streptococcus pyogenes and Neisseria spp. (23, 33). Transport of CD59 to the inclusion was brefeldin A insensitive and did not occur via lipid microdomains, PKD-dependent cell surface transport, or multivesicular bodies, showing that these known post-Golgi apparatus trafficking pathways are not involved. Thus, it is unlikely that CD59 is cotransported to the inclusion with cholesterol and sphingomyelin because trafficking of these host lipids to the inclusion is sensitive to brefeldin A and inhibitors of multivesicular body formation (3, 4, 7, 15). CD59, as a cell surface protein, could possibly reach the inclusion by endocytosis rather than by an exocytic delivery route via the Golgi apparatus, although such a transport pathway would have to involve a multivesicular body-independent mechanism. Nonconventional transport routes that allow proteins to be trafficked from the ER without involvement of the Golgi apparatus have been described both for healthy cells and for cells during infection by an intracellular pathogen. For instance, Golgi apparatus-independent transport from the ER to the cell surface has been demonstrated for the transmembrane protein CD45, the GPI-anchored protein F3/contactin, and the yeast protein Ist2p (1, 5, 21). Moreover, several intracellular bacteria that reside within a cytoplasmic compartment acquire host proteins directly from the ER (40). In the case of Legionella pneumophila, ER-derived vesicles have been shown to attach to and fuse with the parasitophorous vacuole (37). Also, Brucella spp. replicate within an ER-derived compartment that is generated by fusion of the Brucella-containing vacuole with the ER (9). However, these mechanisms appear to be nonselective, which contrasts with the specific transport of CD59 to the inclusion membrane that we observed during chlamydial infection. We showed that CD59 has an unusual localization on the luminal but not the cytoplasmic face of the inclusion membrane, which is the first description of a host protein exposed to the lumen of the chlamydial inclusion. This asymmetric placement on one face of a membrane is a unique feature of GPI-anchored proteins, which associate only with the exoplasmic leaflet of a lipid bilayer via their GPI anchor (52). Furthermore, since the lumen of the inclusion is topologically equivalent to the extracellular space, the localization of CD59 to the luminal surface of the inclusion membrane is consistent with its location at the cell surface on the exoplasmic side of the plasma membrane. Thus, the GPI anchor of CD59 appears to be important for its association with the inclusion membrane rather than for sorting and transport to the inclusion. The location of CD59 on the luminal face of the inclusion membrane raises the possibility that it may have physical interactions with chlamydiae inside the inclusion or with chlamydial proteins in the inclusion membrane. CD59 is known to bind bacterial molecules, as it has been shown to serve as a cellular receptor for bacterial lipopolysaccharide (18, 54). In addition, reticulate bodies (RBs), which are the replicating intracellular form of chlamydiae, are in close proximity to the inclusion membrane during mid-stages in the chlamydial developmental cycle, when they are arranged along the periphery
VOL. 77, 2009
HOST PROTEINS AT THE CHLAMYDIAL INCLUSION
of the inclusion (2, 14, 35). Intriguingly, RBs immediately adjacent to the luminal surface of the inclusion membrane have higher expression levels of a putative chlamydial phospholipase D, CT155, than do RBs in the lumen of the inclusion (31). Since phospholipases cleave the lipid anchor of GPIanchored proteins, expression of these enzymes by chlamydiae could release CD59 from the inclusion membrane, which could explain why we were able to detect CD59 in the inclusion membrane in only about 50% of infected cells. In summary, our studies demonstrate that a mammalian cell surface protein with a complement regulatory function can be detected on the chlamydial inclusion inside an infected cell. At this time, it is not known if CD59 has a role in chlamydial pathogenesis or in the host immune response to chlamydial infection. However, the localization of CD59 to the luminal surface of the inclusion membrane provides this host protein with an opportunity to interact with chlamydiae within the confines of the lumen. From a therapeutic angle, the presence of CD59 on the inclusion membrane may be exploitable as a novel delivery method for targeting proteins to the inclusion.
23.
ACKNOWLEDGMENTS
24.
We thank Ellena Peterson, Yusuke Maeda, Vivek Malhotra, Charles Yeaman, Jennifer Lippincott-Schwartz, Enrico Rodriguez-Boulan, Chiara Zurzolo, Maggie So, Guanming Zhong, and Dan Rockey for generously providing us with reagents. We are grateful to Elizabeth DiRusso Case, Johnny Akers, Allan Chen, Eric Cheng, and Kirsten Johnson for critical readings of the manuscript and for helpful advice. M.T. was supported by an NIH Independent Scientist Award (AI 057563).
25.
REFERENCES 1. Baldwin, T. A., and H. L. Ostergaard. 2002. The protein-tyrosine phosphatase CD45 reaches the cell surface via Golgi-dependent and -independent pathways. J. Biol. Chem. 277:50333–50340. 2. Bavoil, P. M., and R. C. Hsia. 1998. Type III secretion in Chlamydia: a case of deja vu? Mol. Microbiol. 28:860–862. 3. Beatty, W. L. 2008. Late endocytic multivesicular bodies intersect the chlamydial inclusion in the absence of CD63. Infect. Immun. 76:2872–2881. 4. Beatty, W. L. 2006. Trafficking from CD63-positive late endocytic multivesicular bodies is essential for intracellular development of Chlamydia trachomatis. J. Cell Sci. 119:350–359. 5. Bonnon, C., L. Goutebroze, N. Denisenko-Nehrbass, J. A. Girault, and C. Faivre-Sarrailh. 2003. The paranodal complex of F3/contactin and caspr/ paranodin traffics to the cell surface via a non-conventional pathway. J. Biol. Chem. 278:48339–48347. 6. Brown, D. A., and J. K. Rose. 1992. Sorting of GPI-anchored proteins to glycolipid-enriched membrane subdomains during transport to the apical cell surface. Cell 68:533–544. 7. Carabeo, R. A., D. J. Mead, and T. Hackstadt. 2003. Golgi-dependent transport of cholesterol to the Chlamydia trachomatis inclusion. Proc. Natl. Acad. Sci. USA 100:6771–6776. 8. Cattaneo, R. 2004. Four viruses, two bacteria, and one receptor: membrane cofactor protein (CD46) as pathogens’ magnet. J. Virol. 78:4385–4388. 9. Celli, J., and J. P. Gorvel. 2004. Organelle robbery: Brucella interactions with the endoplasmic reticulum. Curr. Opin. Microbiol. 7:93–97. 10. Cocchiaro, J. L., Y. Kumar, E. R. Fischer, T. Hackstadt, and R. H. Valdivia. 2008. Cytoplasmic lipid droplets are translocated into the lumen of the Chlamydia trachomatis parasitophorous vacuole. Proc. Natl. Acad. Sci. USA 105:9379–9384. 11. Cole, N. B., N. Sciaky, A. Marotta, J. Song, and J. Lippincott-Schwartz. 1996. Golgi dispersal during microtubule disruption: regeneration of Golgi stacks at peripheral endoplasmic reticulum exit sites. Mol. Biol. Cell 7:631–650. 12. Cortes, C., K. A. Rzomp, A. Tvinnereim, M. A. Scidmore, and B. Wizel. 2007. Chlamydia pneumoniae inclusion membrane protein Cpn0585 interacts with multiple Rab GTPases. Infect. Immun. 75:5586–5596. 13. Delacour, D., and R. Jacob. 2006. Apical protein transport. Cell. Mol. Life Sci. 63:2491–2505. 14. Hackstadt, T., E. R. Fischer, M. A. Scidmore, D. D. Rockey, and R. A. Heinzen. 1997. Origins and functions of the chlamydial inclusion. Trends Microbiol. 5:288–293. 15. Hackstadt, T., D. D. Rockey, R. A. Heinzen, and M. A. Scidmore. 1996.
16.
17. 18. 19.
20.
21. 22.
26. 27.
28. 29. 30.
31.
32.
33. 34.
35. 36. 37. 38.
39. 40.
1291
Chlamydia trachomatis interrupts an exocytic pathway to acquire endogenously synthesized sphingomyelin in transit from the Golgi apparatus to the plasma membrane. EMBO J. 15:964–977. Hackstadt, T., M. A. Scidmore, and D. D. Rockey. 1995. Lipid metabolism in Chlamydia trachomatis-infected cells: directed trafficking of Golgi-derived sphingolipids to the chlamydial inclusion. Proc. Natl. Acad. Sci. USA 92: 4877–4881. Hackstadt, T., M. A. Scidmore-Carlson, E. I. Shaw, and E. R. Fischer. 1999. The Chlamydia trachomatis IncA protein is required for homotypic vesicle fusion. Cell. Microbiol. 1:119–130. Heine, H., A. J. Ulmer, V. T. El-Samalouti, A. Lentschat, and L. Hamann. 2001. Decay-accelerating factor (DAF/CD55) is a functional active element of the LPS receptor complex. J. Endotoxin Res. 7:227–231. Hirschberg, K., C. M. Miller, J. Ellenberg, J. F. Presley, E. D. Siggia, R. D. Phair, and J. Lippincott-Schwartz. 1998. Kinetic analysis of secretory protein traffic and characterization of Golgi to plasma membrane transport intermediates in living cells. J. Cell Biol. 143:1485–1503. Huang, J., D. Gou, C. Zhen, D. Jiang, X. Mao, W. Li, S. Chen, and C. Cai. 2001. Protection of xenogeneic cells from human complement-mediated lysis by the expression of human DAF, CD59 and MCP. FEMS Immunol. Med. Microbiol. 31:203–209. Juschke, C., A. Wachter, B. Schwappach, and M. Seedorf. 2005. SEC18/NSFindependent, protein-sorting pathway from the yeast cortical ER to the plasma membrane. J. Cell Biol. 169:613–622. Kalayoglu, M. V., P. Libby, and G. I. Byrne. 2002. Chlamydia pneumoniae as an emerging risk factor in cardiovascular disease. JAMA 288:2724–2731. Kallstrom, H., M. K. Liszewski, J. P. Atkinson, and A. B. Jonsson. 1997. Membrane cofactor protein (MCP or CD46) is a cellular pilus receptor for pathogenic Neisseria. Mol. Microbiol. 25:639–647. Kim, D. D., and W. C. Song. 2006. Membrane complement regulatory proteins. Clin. Immunol. 118:127–136. Kreitzer, G., J. Schmoranzer, S. H. Low, X. Li, Y. Gan, T. Weimbs, S. M. Simon, and E. Rodriguez-Boulan. 2003. Three-dimensional analysis of postGolgi carrier exocytosis in epithelial cells. Nat. Cell Biol. 5:126–136. Li, Z., C. Chen, D. Chen, Y. Wu, Y. Zhong, and G. Zhong. 2008. Characterization of fifty putative inclusion membrane proteins encoded in the Chlamydia trachomatis genome. Infect. Immun. 76:2746–2757. Lippincott-Schwartz, J., L. C. Yuan, J. S. Bonifacino, and R. D. Klausner. 1989. Rapid redistribution of Golgi proteins into the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell 56:801– 813. Maeda, Y., Y. Tashima, T. Houjou, M. Fujita, T. Yoko-o, Y. Jigami, R. Taguchi, and T. Kinoshita. 2007. Fatty acid remodeling of GPI-anchored proteins is required for their raft association. Mol. Biol. Cell 18:1497–1506. Muller, N., F. Sattelmacher, R. Lugert, and U. Gross. 2008. Characterization and intracellular localization of putative Chlamydia pneumoniae effector proteins. Med. Microbiol. Immunol. 197:387–396. Nakamura, N., N. Inoue, R. Watanabe, M. Takahashi, J. Takeda, V. L. Stevens, and T. Kinoshita. 1997. Expression cloning of PIG-L, a candidate N-acetylglucosaminyl-phosphatidylinositol deacetylase. J. Biol. Chem. 272: 15834–15840. Nelson, D. E., D. D. Crane, L. D. Taylor, D. W. Dorward, M. M. Goheen, and H. D. Caldwell. 2006. Inhibition of chlamydiae by primary alcohols correlates with the strain-specific complement of plasticity zone phospholipase D genes. Infect. Immun. 74:73–80. O’Connell, C. M., I. A. Ionova, A. J. Quayle, A. Visintin, and R. R. Ingalls. 2006. Localization of TLR2 and MyD88 to Chlamydia trachomatis inclusions. Evidence for signaling by intracellular TLR2 during infection with an obligate intracellular pathogen. J. Biol. Chem. 281:1652–1659. Okada, N., M. K. Liszewski, J. P. Atkinson, and M. Caparon. 1995. Membrane cofactor protein (CD46) is a keratinocyte receptor for the M protein of the group A streptococcus. Proc. Natl. Acad. Sci. USA 92:2489–2493. Pausa, M., V. Pellis, M. Cinco, P. G. Giulianini, G. Presani, S. Perticarari, R. Murgia, and F. Tedesco. 2003. Serum-resistant strains of Borrelia burgdorferi evade complement-mediated killing by expressing a CD59-like complement inhibitory molecule. J. Immunol. 170:3214–3222. Peterson, E. M., and L. M. de la Maza. 1988. Chlamydia parasitism: ultrastructural characterization of the interaction between the chlamydial cell envelope and the host cell. J. Bacteriol. 170:1389–1392. Roan, N. R., and M. N. Starnbach. 2008. Immune-mediated control of Chlamydia infection. Cell. Microbiol. 10:9–19. Robinson, C. G., and C. R. Roy. 2006. Attachment and fusion of endoplasmic reticulum with vacuoles containing Legionella pneumophila. Cell. Microbiol. 8:793–805. Rockey, D. D., and D. T. Alzhanov. 2006. Proteins in the chlamydial inclusion membrane, p. 235–254. In P. M. Bavoil and P. B. Wyrick (ed.), Chlamydia: genomics and pathogenesis. Horizon Bioscience, Wymondham, United Kingdom. Rockey, D. D., R. A. Heinzen, and T. Hackstadt. 1995. Cloning and characterization of a Chlamydia psittaci gene coding for a protein localized in the inclusion membrane of infected cells. Mol. Microbiol. 15:617–626. Roy, C. R., S. P. Salcedo, and J. P. Gorvel. 2006. Pathogen-endoplasmic-
1292
41. 42. 43.
44.
45. 46.
47.
HASEGAWA ET AL.
reticulum interactions: in through the out door. Nat. Rev. Immunol. 6:136– 147. Ruiz-Arguelles, A., and L. Llorente. 2007. The role of complement regulatory proteins (CD55 and CD59) in the pathogenesis of autoimmune hemocytopenias. Autoimmun. Rev. 6:155–161. Rzomp, K. A., A. R. Moorhead, and M. A. Scidmore. 2006. The GTPase Rab4 interacts with Chlamydia trachomatis inclusion membrane protein CT229. Infect. Immun. 74:5362–5373. Rzomp, K. A., L. D. Scholtes, B. J. Briggs, G. R. Whittaker, and M. A. Scidmore. 2003. Rab GTPases are recruited to chlamydial inclusions in both a species-dependent and species-independent manner. Infect. Immun. 71: 5855–5870. Scidmore, M. A. 2006. Chlamydial exploitation of host signaling, cytoskeletal, and membrane trafficking pathways, p. 255–295. In P. M. Bavoil and P. B. Wyrick (ed.), Chlamydia: genomics and pathogenesis. Horizon Bioscience, Wymondham, United Kingdom. Scidmore, M. A., E. R. Fischer, and T. Hackstadt. 1996. Sphingolipids and glycoproteins are differentially trafficked to the Chlamydia trachomatis inclusion. J. Cell Biol. 134:363–374. Scidmore-Carlson, M. A., E. I. Shaw, C. A. Dooley, E. R. Fischer, and T. Hackstadt. 1999. Identification and characterization of a Chlamydia trachomatis early operon encoding four novel inclusion membrane proteins. Mol. Microbiol. 33:753–765. Spear, G. T., N. S. Lurain, C. J. Parker, M. Ghassemi, G. H. Payne, and M. Saifuddin. 1995. Host cell-derived complement control proteins CD55 and CD59 are incorporated into the virions of two unrelated enveloped viruses. Human T cell leukemia/lymphoma virus type I (HTLV-I) and human cytomegalovirus (HCMV). J. Immunol. 155:4376–4381.
Editor: R. P. Morrison
INFECT. IMMUN. 48. Stamm, W. E., R. B. Jones, and B. E. Batteiger. 2005. Chlamydia trachomatis (trachoma, perinatal infections, lymphogranuloma venereum, and other genital infections), p. 2239–2255. In G. L. Mandell, J. E. Bennett, and R. Dolin (ed.), Mandell, Douglas, and Bennett’s: principles and practice of infectious diseases, 6th ed. Elsevier Churchill Livingstone, Philadelphia, PA. 49. Taraska, T., D. M. Ward, R. S. Ajioka, P. B. Wyrick, S. R. Davis-Kaplan, C. H. Davis, and J. Kaplan. 1996. The late chlamydial inclusion membrane is not derived from the endocytic pathway and is relatively deficient in host proteins. Infect. Immun. 64:3713–3727. 50. van Ooij, C., L. Kalman, van Ijzendoorn, M. Nishijima, K. Hanada, K. Mostov, and J. N. Engel. 2000. Host cell-derived sphingolipids are required for the intracellular growth of Chlamydia trachomatis. Cell. Microbiol. 2:627–637. 51. Varma, R., and S. Mayor. 1998. GPI-anchored proteins are organized in submicron domains at the cell surface. Nature 394:798–801. 52. Vidugiriene, J., and A. K. Menon. 1994. The GPI anchor of cell-surface proteins is synthesized on the cytoplasmic face of the endoplasmic reticulum. J. Cell Biol. 127:333–341. 53. Wylie, J. L., G. M. Hatch, and G. McClarty. 1997. Host cell phospholipids are trafficked to and then modified by Chlamydia trachomatis. J. Bacteriol. 179:7233–7242. 54. Yamamoto, T., T. Nakane, S. Doi, and T. Osaki. 2003. Lipopolysaccharide signal transduction in oral keratinocytes—involvement of CD59 but not CD14. Cell Signal. 15:861–869. 55. Yeaman, C., M. I. Ayala, J. R. Wright, F. Bard, C. Bossard, A. Ang, Y. Maeda, T. Seufferlein, I. Mellman, W. J. Nelson, and V. Malhotra. 2004. Protein kinase D regulates basolateral membrane protein exit from transGolgi network. Nat. Cell Biol. 6:106–112.
