Regulation of Chlamydial Infection by Host Autophagy and Vacuolar ...

INFECTION AND IMMUNITY, Oct. 2011, p. 4019–4028 0019-9567/11/$12.00 doi:10.1128/IAI.05308-11 Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Vol. 79, No. 10

Regulation of Chlamydial Infection by Host Autophagy and Vacuolar ATPase-Bearing Organelles䌤 Muhammad Yasir,§† Niseema D. Pachikara,§‡ Xiaofeng Bao,§ Zui Pan, and Huizhou Fan* Department of Physiology and Biophysics, Robert Wood Johnson Medical School, 683 Hoes Lane, Piscataway, New Jersey 08854 Received 27 April 2011/Returned for modification 26 May 2011/Accepted 20 July 2011

As arguably the most successful parasite, Chlamydia is an obligate intracellular bacterium replicating inside a vacuole of eukaryotic host cells. The chlamydial vacuole does not fuse with the defense cell organelle lysosome. We previously showed that chlamydial infection increases markers of autophagy, an innate antimicrobial activity requiring lysosomal function. However, the work presented here demonstrates that p62, an autophagy protein that is degraded in lysosomes, either remained unchanged or increased in chlamydiainfected human epithelial, mouse fibroblast, and mouse macrophage cell lines. In addition, the activities of three lysosomal enzymes analyzed were diminished in chlamydia-infected macrophages. Bafilomycin A1 (BafA), a specific inhibitor of vacuolar ATPase (vATPase) required for lysosomal function, increased the growth of the human pathogen Chlamydia trachomatis (L2) in wild-type murine fibroblasts and macrophages but inhibited growth in the autophagy-deficient ATG5ⴚ/ⴚ fibroblasts. BafA exhibited only slight inhibition or no effect on L2 growth in multiple human genital epithelial cell lines. In contrast to L2, the mouse pathogen Chlamydia muridarum (MoPn) was consistently inhibited by BafA in all cell lines examined, regardless of species origin and autophagy status. Finally, L2 but not MoPn grew more efficiently in the ATG5ⴚ/ⴚ cells than in wild-type cells. These results suggest that there are two types of vATPase-bearing organelles that regulate chlamydial infection: one supports chlamydial infection, while the other plays a defensive role through autophagy when cells are artificially infected with certain chlamydiae that have not been adapted to the host species.

tary body (EB) to a eukaryotic host cell. The EB is taken into a vacuole inside the host cell as a result of endocytosis. The EB-containing vacuole, called an inclusion, is delivered to a perinuclear region. In the inclusion, the EB differentiates into the proliferative but noninfectious reticulate body (RB). As they accumulate inside the inclusion, RBs progressively reorganize back to EBs that are released from the host cell at the end of the developmental cycle. Understanding of the interaction between chlamydiae and their host cells remains incomplete. It is commonly accepted, based on studies first with Chlamydia psittaci (17, 18), an avian pathogen that accidentally infects humans, and later with other species, including human pathogens C. trachomatis (26, 38, 39, 42) and C. pneumoniae (3), that in epithelial cells, the primary target of chlamydiae, and also in fibroblasts, the chlamydial inclusion does not fuse with the lysosome, although invading EBs are degraded mostly in the lysosomes of blood monocytes and in neutrophils (44–46). In contrast to chlamydiae, a large number of other pathogens are taken into the lysosome and readily degraded by lysosomal enzymes in both phagocytes and nonprofessional phagocytes, including epithelial cells (29). The general antimicrobial activity of the lysosome depends on macroautophagy, frequently referred to as autophagy (for a review, see references 16 and 29). In addition to infection, various signals, including starvation, growth factor deprivation, and energy depletion, induce autophagy. In response to these stimuli, the protein kinase mammalian target of rapamycin is inhibited, leading to the association of a protein complex containing several ATG proteins, encoded by autophagy-related genes, as well as proteins encoded by other genes, with lipid membranes originating from various organelles. With the re-

Chlamydiae are obligate intracellular bacteria consisting of multiple species (23). Chlamydia trachomatis and Chlamydia pneumoniae are the two species that naturally infect humans. C. trachomatis is the most prevalent sexually transmitted bacterial pathogen worldwide (37). Since urogenital chlamydial infection is frequently asymptomatic, most infected people do not seek medical treatment. However, a substantial proportion of untreated cases develop long-term complications, including infertility and pelvic inflammatory disease. C. trachomatis also causes conjunctivitis and, even to this day, is a major cause of preventable blindness in the developing world (37). C. pneumoniae is a common respiratory pathogen that is also considered a cofactor of atherosclerosis (10) and neurodegenerative diseases (6). Among the nonhuman chlamydial species, Chlamydia muridarum is especially a useful organism because of its ability to model human chlamydial infections in mice (11, 13, 14). Chlamydiae have a unique developmental cycle consisting of two distinct cellular forms (1, 31). The cycle is initiated by binding of the infectious but metabolically quiescent elemen-

* Corresponding author. Mailing address: Department of Physiology and Biophysics, Robert Wood Johnson Medical School, University of Medicine and Dentistry of New Jersey, 683 Hoes Lane, Piscataway, NJ 08854. Phone: (732) 235-4607. Fax: (732) 235-5038. E-mail: fanhu @umdnj.edu. § These authors made equal contributions to the research. † Present address: Department of Biotechnology, Quaid-i-Azam University, Islamabad, Pakistan. ‡ Present address: Malaria Research Group, International Centre for Genetic Engineering and Biotechnology, New Delhi, India. 䌤 Published ahead of print on 1 August 2011. 4019

4020

YASIR ET AL.

cruitment of LC3-II, which is derived from the cytoplasmic protein LC3-I through posttranslational modifications, the protein complex-bearing membranes are elongated to yield isolation membranes. As the isolation membranes elongate, they wrap cell organelles, microbes, or microbe-containing vacuoles, forming autophagosomes, which are characterized by double membranes. The autophagosomes undergo sequential fusion with endosomes and lysosomes, resulting in the formation of autolysosome, in which the cytoplasmic cargos are degraded by lysosomal enzymes, leading to the regeneration of free amino acids, lipids, and nucleotides, including ATP, and the killing of pathogens (16, 29). In order to survive and grow inside host cells, many microbes have developed strategies to inhibit autophagy (16, 29). For example, Mycobacterium tuberculosis modifies its residential vacuole membrane; as a result, the vacuoles cannot fuse with autophagosomes (24). On the other hand, Coxiella burnetii, the causative agent of Q fever, has adapted the autolysosome as a growth niche (25). Whether and how autophagy regulates chlamydial infection remains unknown. We have reported that autophagy deficiency, resulting from ATG5 gene knockout, had no detectable effect on L2 growth in mouse embryonic fibroblasts (MEFs) (34). Nevertheless, chlamydial infection appears to have an effect on autophagy, as judged by significant elevations in the LC3-II expression level, and colocalization of LC3 with the lysosomal marker LAMP1 (34). To further assess autophagic activity in chlamydia-infected cells, we have now determined the expression level of p62, an autophagy flux marker (35), and lysosomal enzyme activities in chlamydia-infected cells. These analyses suggest a possible blockage of autophagy completion in lysosomes of infected cells and further led us to determine the effect of BafA, an inhibitor of vacuolar ATPase (vATPase) (8), which is required for the function of the lysosome and some other cell organelles, on chlamydial infection. Our findings indicate the existence of at least two types of vATPasebearing vesicles which play opposing roles in chlamydial infection. Whereas one supports chlamydial growth in permissive cells, the other plays a defensive role through autophagy in cells experimentally infected with chlamydiae that have not been adapted to the host species. (This work was presented to the 5th Chlamydia Basic Research Society Biennial Meeting held in Redondo Beach, CA, in March 2011.) MATERIALS AND METHODS Chemicals and antibodies. Dulbecco’s modified Eagle’s medium (DMEM), L-glutamine (200 mM) solution, MEM amino acids solution, MEM nonessential amino acid solution without glutamine, fetal bovine serum (FBS), Hanks buffered salt solution (HBSS), BafA, rapamycin, Ala-Ala-Phe-7-amido-4-methylcoumarin (AAFAMC, substrate of tripeptidyl peptidase-1 [TPP]), 4-methylumbelliferyl ␤-D-glucuronide hydrate (MuGl, substrate of lysosomal ␤-glucuronidase), 4-methylumbelliferyl ␤-D-galactopyranoside (MuGa, substrate of lysosomal ␤-galactosidase), polyclonal rabbit anti-LC3, and all secondary antibodies used in this study were purchased from Sigma-Aldrich. A mouse monoclonal antibody recognizing the major outer membrane protein (MOMP) of C. trachomatis L2 (L2) and C. muridarum (12), a mouse monoclonal antibody for the genus-specific lipopolysaccharide (LPS) of Chlamydia (43), polyclonal rabbit anti-L2, and polyclonal rabbit anti-C. muridarum were generous gifts from Guangming Zhong (University of Texas Health Sciences Center at San Antonio). Polyclonal guinea pig anti-p62 was purchased from American Research Products. Monoclonal mouse anti-vATPase subunit H and horseradish peroxidase-conjugated mono-

INFECT. IMMUN. clonal mouse anti-␤-actin were purchased from Santa Cruz Biotechnology. The ECL and ECL Plus kits were purchased from GE Healthcare. RPMI 1640 cell culture medium, kerotinocyte culture medium, and phosphate-buffered saline containing calcium and magnesium (PBS) were purchased from Invitrogen. [3H]L-valine (2 Ci/mmol) was purchased from Moravek Biochemicals. Cell lifters were purchased from Corning. Cell lines and culture conditions. Human cervical carcinoma HeLa, ectocervical epithelial Ect1/E6E7 (Ect1), endocervical epithelial End1/E6E7 (End1), vaginal epithelial VK2/E6E7 (VK2) (19), and mouse macrophage Raw264.7 cells (RAW) were purchased from the American Type Culture Collection (ATCC, Manassas, VA). Wild-type (ATG5⫹/⫹) mouse embryonic fibroblasts (MEFs) and their isogenic variant autophagy-deficient Atg5⫺/⫺ MEFs (28) were generous gifts from Noboru Mizushima (Tokyo Metropolitan Institute of Medical Science). HeLa and the MEFs were cultured with DMEM supplemented with 10% FBS (Sigma-Aldrich) and 10 ␮g/ml gentamicin. RAW cells were cultured with the same medium but with only 5% FBS. Ect1, End1, and VK2 cells were cultured with serum-free keratinocyte medium using bovine pituitary extract and epithelial growth factor as growth factors (Invitrogen). The keratinocyte medium was supplemented with extra CaCl2 (final concentration, 0.4 mM) (19). All cell lines were cultured as adherent cells at 37°C in a humidified tissue culture chamber containing 5% CO2 and 95% air. Chlamydiae and culture conditions. Strain 434/bu of C. trachomatis serovar L2, an etiological agent of human lymphogranuloma venereum, and strain Nigg II of C. muridarum, previously referred to as C. trachomatis mouse pneumonitis (MoPn), were purchased from ATCC. EB stocks were prepared using HeLa cells as the host as previously described (5). For experiments presented in this work, unless indicated otherwise, cells were seeded on 24-well plates at the density that resulted in 50 to 60% confluence after overnight incubation. To infect the cells, the culture medium was replaced with fresh medium containing EBs. For Western blot analyses, the multiplicity of infection (MOI) for all cell line studies except that of RAW cells was 3 inclusion-forming units (IFU) per cell, which resulted in visible inclusions in ⬃90% of cells at 24 h; for RAW cells, the MOI was 10 IFU per cell for L2 and 5 for MoPn. For experiments determining the effect of BafA on chlamydial growth and immunofluorescence confocal microscopy, the MOI was 1 IFU per cell unless indicated otherwise for all lines except RAW cells; for RAW cells, the MOI was 3 for L2 and 2 for MoPn. After infection at 37°C for 2 h, cells were washed once to remove free EBs and cultured with fresh medium until harvest or further manipulation. For experiments determining the effects of extra amino acids on the inhibition of L2 growth by BafA, the culture medium was supplemented with L-glutamine solution (3% volume of the culture medium), MEM amino acid solution (3%), and MEM nonessential amino acid solution (6%) after free EBs were removed. BafA treatment. BafA was added to the cultures (final BafA concentration, 50 nM) either 2 h prior to or 4 h after inoculation. The culture plates were returned to the incubator. At the end of a 2-h treatment period, the culture medium was replaced with fresh medium free of the inhibitor. Determination of EB production. At 24 h after inoculation, cells were harvested in sucrose-phosphate-glutamate (SPG) buffer with the aid of a cell lifter and were disrupted by sonication to release chlamydiae. The harvests were 1:10 serially diluted and added to HeLa cell monolayers grown on coverslips. After 24 h of incubation, chlamydial inclusions were fixed with methanol prechilled to 4°C and sequentially stained with the monoclonal anti-chlamydial LPS and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse immunoglobulin G (IgG). Inclusions were enumerated under an Olympus IX51 fluorescence microscope (5). Western blotting. Cell lysate preparation, electrophoresis, protein transfer, membrane blocking, antibody reaction, signal visualization, membrane stripping, and reprobing were performed as described previously (34). The LC3 proteins (LC3-I and LC3-II), p62, and MOMP were detected by using appropriate primary antibodies and secondary antibodies described above. LC3 and p62 bands were visualized using the ECL Plus kit, whereas MOMP and actin bands were detected with the ECL kit. Determination of lysosomal enzyme activities. Protocols for preparation of cell extracts and assays for lysosomal ␤-galactosidase, ␤-glucuronidase, and TPP activities were based on those previously described (40). Cells were seeded on 12-well plates and infected with either L2 or MoPn. At 24 h after inoculation, cells were washed twice with ice cold 0.15 M NaCl. Then, 1 ml of lysis buffer (1% NP-40, 0.15 M NaCl, 10 mM Tris, pH 7.5) was added to each well, and the plates were kept at 4°C for 1 h on a rocking platform. The resulting lysates were centrifuged for 30 min at 20,000 ⫻ g at 4°C, and the supernatants containing the enzymes were collected. AAFAMC (50 mM), MuGl (100 mM), and MuGa (100 mM), substrates for lysosomal TPP, ␤-glucuronidase, and ␤-galactosidase, respectively, were prepared as 100⫻ stocks in dimethyl sulfoxide. Solutions (1⫻)

VOL. 79, 2011

AUTOPHAGY AND VACUOLAR ATPase IN CHLAMYDIAL INFECTION

4021

were prepared immediately before the assay in substrate buffer (0.1 M sodium acetate, 0.15 M NaCl, 0.1% Triton X-100, pH 4.5 [for AAFAMC] and pH 4.0 [for MuGl and MuGa]). The reactions were initiated by adding 40 ␮l of the 1⫻ substrate solution to 10 ␮l of undiluted, 2-, 4-, and 8-fold diluted cell extract in duplicate 96-well plates. A solution of 0.1% Triton X-100 and 0.15 M NaCl was used to dilute the cell extract. Plates were incubated at 37°C on a shaking platform for 1 h, and the reactions were terminated by adding 100 ␮l of 0.1 M monochloroacetic acid-0.13 M NaOH-0.1 M acetic acid, pH 4.5, to the TTP reaction or 0.5 M glycine, pH 4.0, for the other two assays. As controls, solubilizing buffer was utilized in place of the cell lysates for each of the enzyme assays. The activities of the control samples were subtracted from those of the actual samples. The activities of the reaction mixtures were measured by utilizing a CytoFluor II (PerSeptive Biosystems, Framingham, MA) fluorescence multiwell plate reader with excitation at 360 nm and emission at 460 nm. Enzyme activities that were linear with respect to the input samples were used to determine the activities, which were normalized using the protein concentration of each sample. A Pierce BCA protein assay kit was employed to determine the protein concentrations. Immunofluorescence confocal microscopy. Cells were seeded onto coverslips in 24-well-plates (5). Infection and treatment were the same as described above. Fixation and staining with primary and appropriate secondary antibodies were performed at room temperature for 1 h each, followed by three washes of 5 min each. Images were acquired with a Zeiss LSM510 META confocal microscope (34). Statistical analysis. A two-sided t test was performed on Microsoft Excel to analyze densitometry data and EB titers. A significant difference was defined as a P value of ⬍0.05.

RESULTS The level of p62 remains unchanged in L2-infected HeLa cells and MEFs. Previously, we have shown that L2 infection elevates the amount of LC3-II that is associated with isolation membranes and autophagosomes. The infection also increases LC3 punctate distribution. In addition, the infection enhances the colocalization of LC3 with the lysosomal marker LAMP1 (34). To determine if the increase in LC3 colocalization with LAMP1 translates to a higher level of autophagy completion, we compared the expression level of the autophagy adaptor protein p62 in chlamydia-infected cells and uninfected cells. Since p62 is recruited to the autophagosomes and subsequently degraded in the autolysosome, the expression level of p62 decreases as a result of increased autophagic activity (35). The analysis was carried out with the human cervical carcinoma HeLa cells because the cervical epithelium is a primary target of C. trachomatis. We also employed two isogenic MEF cell lines, the autophagy-proficient wild-type cells (i.e., ATG5⫹/⫹) and autophagy-deficient ATG5⫺/⫺ cells, since our previous work had shown similar changes in the expression of LC3-I and LC3-II, microscopic distribution pattern of LC3 (diffuse versus punctual), and LC3-LAMP1 colocalization in HeLa and ATG5⫹/⫹ cells after chlamydial infection (34). The chlamydial infection status in the cells was demonstrated by the detection of MOMP using Western blotting. As expected, LC3-II was undetectable in both infected and uninfected ATG5⫺/⫺ cells (Fig. 1A). Consistent with our previous findings, there were concurrent decreases in the expression level of LC3-I and elevations in that of LC3-II in both HeLa and ATG5⫹/⫹ MEF cells following chlamydial infection (Fig. 1A and B). However, there were no accompanying decreases in p62, suggesting that chlamydial infection does not upregulate autophagy completion in epithelial cells and MEFs. L2 or MoPn infection increases p62 but decreases lysosomal enzyme activities in murine macrophages. As essential components of the immune system, macrophages play a pivotal

FIG. 1. p62 remains unchanged in L2-infected HeLa and wild-type MEFs despite increased LC3-II. Uninfected or L2-infected cells were harvested 24 h after inoculation. (A) Protein bands detected by Western blotting; (B) LC3-II/LC3-I densitometric ratios; (C) p62 densitometric data (normalized by ␤-actin densitometry values). Values shown represent means ⫾ standard deviations of results of three experiments. Double asterisks signify statistically highly significant increases in the LC3-II/LC3-I ratio in infected cells compared to control uninfected cells (P ⬍ 0.01).

role in defense against microbial infection. We used the RAW cell line, which is broadly used for studying the interaction between host and microbes, including chlamydiae (7, 27, 36), to investigate the regulation of autophagy in macrophages during chlamydial infection. Since it has been shown previously that the mouse macrophage cell line supports the murine pathogen MoPn more efficiently than the human pathogen L2 (36), we included both organisms in these experiments. Compared to control uninfected cells, RAW cells infected with either organism had significantly increased LC3-II/LC3-I ratios (Fig. 2A and B) and strongly elevated p62 levels (Fig. 2A and C), suggesting that chlamydial infection may cause inhibition of autophagy completion. Since autophagy completion takes

4022

YASIR ET AL.

INFECT. IMMUN.

FIG. 2. Increased LC3 protein and p62 levels (A to C) and decreased lysosomal enzyme activities (D) in RAW cells infected with either L2 or MoPn. Western blotting was performed as in Fig. 1. See Materials and Methods for details of lysosomal enzyme assays. Data are averages ⫾ standard deviations of three experiments. Double asterisks signify statistically highly significant changes compared to control cells (P ⬍ 0.01).

place in autolysosomes (21), we determined if chlamydial infection affects lysosomal enzyme activity. Interestingly, the activities of the three lysosome enzymes analyzed, ␤-galactosidase, ␤-glucuronidase, and TPP, all displayed a nearly 50% decline in RAW cells infected with either chlamydial strain compared to that in control uninfected cells (Fig. 2D). Together, the increases in the p62 level and the decreases in lysosomal enzyme activities in chlamydia-infected RAW cells suggest that the function of lysosomes in macrophages is inhibited as a result of chlamydial infection. In murine cells BafA stimulates L2 growth in an autophagydependent manner and inhibits MoPn growth irrespective of autophagic status. To infer if chlamydial infection-induced lysosomal inhibition plays a role in chlamydial infection, we determined the effect of BafA, which covalently binds to and thus specifically inhibits the vATPase that is required for the function of lysosomes and other organelles (20, 22, 32, 41), on chlamydial infection. Although various effects of BafA on L2 and MoPn have been documented in the literature (4, 15, 26, 38), it has been difficult to interpret the discrepancy because

previous studies used different host cell lines and none compared the effects of BafA on the two organisms. We used a low BafA concentration (50 nM) to treat cells for a brief period (2 h) to avoid cytotoxicity. Interestingly, pretreatment with the compound prior to infection resulted in ⬃70% higher production of L2 EBs (Fig. 3A, upper panel), suggesting that lysosomes and/or other vATPase-expressing nonlysosomal organelles defend against the human pathogen. A 2-h treatment with BafA starting at 4 h after inoculation had the same stimulatory effect on L2 infection as pretreatment (Fig. 3A, upper panel), arguing that the inhibition of L2 EB production by vATPasebearing vacuoles occurs during RB replication and/or RBto-EB transition. Compared to L2, BafA suppressed MoPn infection in the mouse macrophage cell line; either treatment scheme resulted in ⬃70% less MoPn growth from RAW cells (Fig. 3A, lower panel). These results suggest that vATPasebearing organelles play a supportive role in MoPn infection in murine macrophages. We next assessed if vATPase-containing vacuoles play the same roles in MEFs as in RAW cells in the regulation of L2

VOL. 79, 2011


4023

FIG. 4. Effects of additional amino acids on L2 growth inhibition by BafA in ATG5⫺/⫺ cells. BafA treatment was for 2 h prior to EB inoculation. After free EBs were removed, cells were cultured with regular full culture medium (DMEM) or the medium supplemented with extra amino acids (DMEM⫹AA). The number of EBs produced at 24 h after inoculation was determined. Data are averages ⫾ standard deviations of results of triplicate experiments. Single and double asterisks signify statistically significant changes in BafA-treated cells compared to control cells receiving no BafA treatment (P ⬍ 0.05 and 0.01, respectively).

FIG. 3. Chlamydial strain- and host cell autophagy-dependent effect of BafA on L2 and MoPn infection. BafA treatment was for 2 h either prior to (⫺2/0 h) or 4 to 6 h after (4/6 h) EB inoculation. The number of EBs produced at 24 h after inoculation was determined. Data are averages ⫾ standard deviations of results of quadruplicate experiments. Single and double asterisks signify statistically significant changes in BafA-treated cells compared to control untreated cells (P ⬍ 0.05 and 0.01, respectively).

and MoPn infections and, further, if and how the regulatory activity is related to autophagy by employing ATG5⫹/⫹ and ATG5⫺/⫺ MEFs. In the absence of BafA, the ATG5⫺/⫺ cells produced ⬃3-fold more L2 EBs than the ATG5⫹/⫹ cells (Fig. 3B, upper panel), suggesting that autophagy serves as an innate defense mechanism against L2 infection, consistent with the role of autophagy in most microbial infections (29). Both BafA treatment schemes increased L2 growth in the ATG5⫹/⫹ cells (Fig. 3B, upper panel), suggesting that vATPase-bearing organelles in MEFs also play a defense role against L2 infection as in macrophages. Interestingly, the treatments suppressed L2 EB production by ⬃40% in the ATG5⫺/⫺ cells (Fig. 3B, upper panel). A failure of BafA to increase L2 growth in the ATG5⫺/⫺ cells would suggest that vATPase-bearing vacuoles defend against L2 infection through an autophagy-dependent mechanism; further growth inhibition by BafA indicates that some vATPase-bearing organelles may support L2 infection. In contrast to L2, MoPn grew equally efficiently in ATG5⫹/⫹ and ATG5⫺/⫺ cells (Fig. 3B, lower panel), suggesting that autophagy is an ineffective defense tool against MoPn in MEFs or murine cells in general. Similar to findings from RAW cells (Fig. 3A, lower panel), BafA significantly inhibited MoPn

growth in both ATG5⫹/⫹ and ATG5⫺/⫺ cells (Fig. 3B, lower panel). These data suggest that vATPase-bearing organelles mainly play a supportive role in MoPn infection in both macrophages and fibroblasts in an autophagy-independent manner. Extra amino acids fail to reverse L2 growth inhibition by BafA in ATG5ⴚ/ⴚ cells. By degrading cellular components, autophagy serves as an important mechanism for amino acid regeneration (16, 29). We reason that the differential effects of BafA on L2 growth in the MEFs (i.e., increasing EB production in ATG5⫹/⫹ cells and decreasing it in ATG5⫺/⫺ cells) may be due to a more complete block in protein degradation leading to a more severe reduction in amino acid pools in the ATG5⫺/⫺ cells. Therefore, we determined if extra amino acids would alleviate the inhibitory effect of BafA in the ATG5⫺/⫺ cells by supplementing the culture medium with both essential and nonessential amino acids. Since previous studies have shown that L2 growth can be inhibited by an unbalanced supply of amino acids that share a transporter (2, 9), we added all 20 amino acids according to their ratios in the animal cell culture medium MEM (www.safcbiosciences). The supplementations would result in an average 2.5-fold increase in amino acid concentrations with the exception of alanine, asparagine, aspartic acid, glutamic acid, and proline, which are not present in DMEM. Even though the supplemental amino acids were thought to be “in balance,” they still exhibited inhibitory effects on L2 growth (Fig. 4). The growth inhibition by the extra amino acids was independent of autophagy, since it occurred in both ATG5⫹/⫹ and ATG5⫺/⫺ cells. Evidently, the trend that BafA affects chlamydial growth in the presence of the additional amino acids remained the same in their absence. Thus, the inhibition of L2 growth by BafA in ATG5⫺/⫺ cells is un-

4024

YASIR ET AL.

likely to be due to a more severe amino acid pool reduction following chlamydial infection. Inhibition of L2 growth by autophagy in MEF is inoculating dose dependent. As discussed above, overall, data presented in Fig. 3B suggest that in MEFs autophagy is a deterring factor for L2 but not MoPn. Nevertheless, the finding of enhanced L2 growth in the ATG5⫺/⫺ cells, compared to that in ATG5⫹/⫹ cells, was unexpected, since we had previously reported that the two MEF cell lines supported the human pathogen infection equally efficiently (34). The discrepancy was traced to the fact that we had reduced the MOI from 3 IFU per cell in the previous study to 1.0 IFU per cell for the experiments in Fig. 3B. Accordingly, when we infected the cells with the two inoculating doses in parallel, an increase in EB production from the ATG5⫺/⫺ cells, compared to that from the ATG5⫹/⫹ cells, was detected only with an MOI of 1.0 and not with an MOI of 3.0 (Fig. 5A). Immunofluorescence microscopy showed an apparent higher inclusion-forming efficiency in the ATG5⫺/⫺ cells at an MOI of 1; thus, inclusions were detected in more than 50% of ATG5⫺/⫺ cells but in only ⬃30% of ATG5⫹/⫹ cells (Fig. 5B). At an MOI of 3, inclusions were found in ⬃90% of ATG5⫹/⫹ and ATG5⫺/⫺ cells (Fig. 5C). These observations indicate that in murine fibroblasts, autophagy inhibits L2 inclusion formation; increased inoculating doses can overcome the inhibition. BafA inhibits MoPn infection but has weak or no effect on L2 in human epithelial cells. The stimulation of L2 and inhibition of MoPn by BafA in wild-type murine cells and the differential effect of ATG5 deficiency on L2 and MoPn growth (Fig. 3) suggest that there has been host-species adaptation with regard to the interaction between invading chlamydiae and host defense mediated by vATPase-bearing organelles and/or autophagy. This prompted us to determine if BafA would stimulate MoPn infection but inhibit L2 growth in human cells. Nevertheless, BafA suppressed the growth of both organisms in HeLa cells. The inhibitory effect of BafA on MoPn growth is reproducibly stronger than that on the human chlamydia (Fig. 6A). We next expanded the test to three additional human epithelial cell lines, VK2, Ect1, and End1. The VK2 cell line was established from the normal vaginal mucosal tissue taken from a premenopausal woman undergoing anterior-posterior vaginal repair surgery; the ectocervical Ect1 and endocervical End1 cell lines were established from normal epithelial tissues taken from a premenopausal woman undergoing hysterectomy for endometriosis (21). The effects of BafA on the growth of the two organisms in these cell lines (Fig. 6B to D) are overall similar to those observed in HeLa cells. Thus, it exhibited moderate inhibitory effects on MoPn infection but had minor or no inhibitory effects on L2 growth. These results indicate that vATPase-carrying vacuoles are more important for supporting the growth of MoPn, compared to that of L2, in human vaginocervical epithelia. Lack of chlamydia/vATPase colocalization. To understand how vATPase-bearing vacuoles regulate chlamydial infection, we performed immunofluorescence confocal microscopy for vATPase and chlamydiae in RAW cells, ATG5⫹/⫹ MEFs, ATG5⫺/⫺ MEFs, and HeLa cells infected with either L2 or MoPn. Since there was no noticeable difference between the two chlamydial species and among the cell lines studied, only

INFECT. IMMUN.

FIG. 5. Reversal of autophagy-mediated L2 growth inhibition by increase in inoculating dose. Wild-type and ATG5-null MEFs were infected at an MOI of 1 or 3 IFU per cell. (A) EB production was determined at 24 h after inoculation. Data are averages ⫾ standard deviations of results of quadruplicate experiments. Double asterisks signify a statistically highly significant increase in EB production by the ATG5⫺/⫺ cells compared to ATG5⫹/⫹ cells at 1 MOI (P ⬍ 0.01). (B) Cells infected with L2 (MOI ⫽ 1) were fixed at 24 h after inoculation and stained with mouse anti-chlamydial LPS and FITC-conjugated donkey anti-mouse IgG. Images were obtained following counterstaining with Evans blue. (C) Phase-contrast images of live cells were obtained at 32 h after inoculation (MOI ⫽ 3).

representative images from L2-infected ATG5⫹/⫹ cells are shown in Fig. 7. There was no obvious colocalization between chlamydiae and vATPase 2 h after inoculation (Fig. 7); BafA pretreatment did not have a noticeable effect on this staining pattern (Fig. 7). We also performed costaining for vATPase and IncA, a chlamydial protein distributed to the inclusion membrane 20 h after inoculation. We did not observe evidence for localization of vATPase on the L2 inclusion membrane, which is consistent with previously published data (26). Due to the lack of an

VOL. 79, 2011


4025

FIG. 6. BafA inhibits MoPn growth but has weak or no effect on L2 growth in human epithelial cells. Experiments were carried out as done for Fig. 3. Data are averages ⫾ standard deviations of results of triplicate experiments. Single and double asterisks signify statistically significant decreases in EB production by BafA-treated cells compared to control untreated cells (P ⬍ 0.05 and 0.01, respectively).

antibody for an MoPn inclusion membrane protein, we were not able to directly address if vATPase is distributed to the inclusion membrane of MoPn. Therefore, we used a polyclonal antibody raised against the whole MoPn organism to localize its inclusion at 20 h after inoculation. We did not observe enrichment of vATPase signals around the chlamydial inclusion (data not shown), suggesting the unlikelihood of distribution of vATPase to the inclusion membrane. Taken together, the immunofluorescence confocal microscopy data suggest that the chlamydial inclusion remains unassociated with vATPasebearing organelles from 2 to 20 h after infection (likely through the entire developmental cycle) and that they regulate L2 and MoPn infections indirectly. p62 does not bind to chlamydiae. The inhibition of a number of facultative intracellular bacteria by autophagy is due to the recognition of pathogens by p62 (16). As a result of p62 binding, the pathogens are directed to the autophagy machinery. Since our growth data obtained from the ATG5⫹/⫹ and ATG5⫺/⫺ MEFs (Fig. 4 and 5) suggest that autophagy suppresses L2 infection in murine cells, we determined if p62 binds to L2 in these MEFs. However, we did not detect apparent colocalization between L2 organisms with p62 in either ATG5⫹/⫹ cells (Fig. 8) or ATG5⫺/⫺ MEF cells (data not shown). Therefore, it appears unlikely that inhibition of L2 growth by autophagy in murine fibroblasts is mediated by p62. However, in addition to p62, other adaptor proteins can also

bind to microbes (16). It remains to be determined if those other proteins recognize chlamydiae. DISCUSSION Work in the last several years has established autophagy as a general innate defense mechanism against microbial pathogens. In response, pathogens have developed an array of strategies to counter this defense tool (16, 29). At one extreme, the Q-fever pathogen Coxiella burnetii has even adapted to the autolysosome as its growth niche (25). Our work reported here suggests that the role that autophagy plays in chlamydial infection may depend on both the host and pathogen species. Thus, in MEFs, an inhibitory role for autophagy in chlamydial growth is detectable only with L2 and not with MoPn that has highly adapted to the intracellular environment of murine cells. Whereas a previous study has shown that autophagy is required for the inhibition of L2 growth by gamma interferon, our observation of a decreased EB production in ATG5⫹/⫹ cells compared to that in ATG5⫺/⫺ cells in the absence of exogenous gamma interferon suggests that autophagy guards against the human pathogen before the protective cytokine is produced by infected host cells. Even so, a balance between autophagy and L2 seems delicate; as a result, the inhibitory activity of autophagy can be overcome by merely a 3-fold increase in the inoculating dose.

4026

YASIR ET AL.

INFECT. IMMUN.

FIG. 8. p62 does not bind to L2. ATG5⫹/⫹ MEF cells were infected with L2, fixed at 2 h after inoculation, and stained with guinea pig anti-p62, rabbit anti-L2, and secondary antibodies (FITC-conjugated donkey anti-guinea pig IgG and rhodamine-conjugated sheep antirabbit IgG).

FIG. 7. Lack of chlamydia/vATPase colocalization. Cells were infected with L2, fixed at 2 h after inoculation, and stained with mouse anti-vATPase subunit H and rabbit anti-L2 and secondary antibodies (FITC-conjugated sheep anti-mouse IgG and rhodamine-conjugated sheep anti-rabbit IgG).

Whereas autophagy inhibits L2 infection in MEFs, it is not clear if the same is true in human cells or if autophagy affects MoPn growth in human cells. Host species-dependent inhibition of chlamydial infection has been noted previously. For example, gamma interferon is a strong growth inhibitor of C. muridarum in human cells, but the cytokine has a much weakened effect when the organism is cultured in murine cells, the pathogen’s natural host cells (36). The inhibition of L2 growth

by BafA in human cells, which contrasts with the compound’s stimulatory effect in wild-type murine cells, does cast doubt on a strong role for autophagy in defense against L2 infection in human cells. Nevertheless, one should not equate the effect of BafA to autophagy inhibition as discussed below. Thus, if and how human cells use autophagy to regulate the growth of their natural pathogen C. trachomatis and that of an artificially infective MoPn requires further examination. By covalently binding to the vATPase, BafA is a potent and specific inhibitor of the vATPase, which is required for maintaining the acidic luminal pH in lysosomes (32) and endosomes (20). Since these organelles are essential for autophagy completion, BafA has broadly been used as an autophagy inhibitor for numerous studies. Nevertheless, as the compound exhibits effects on chlamydial infection in the ATG5⫺/⫺ cells as well as in wild-type cells, one has to also consider the involvement of an autophagy-independent mechanism(s). It is intriguing that BafA has opposite effects on L2 infection in the isogenic cell lines. Whereas it stimulates L2 growth in wild-type cells, it inhibits it in the ATG5⫺/⫺ cells. We speculate that the growth stimulation is due to the compound’s inhibitory effect on autophagy, while the growth inhibition is through its targeting of vATPase-bearing vacuoles that support chlamydial infection. In each of the cell lines examined, the effect of BafA administered prior to infection is qualitatively the same as that of treatment started 4 h after inoculation, suggesting that both the supportive and inhibitory mechanisms function during RB replication and/or RB-to-EB differentiation. It appears that the supportive role is more important for C. muridarum than for C. trachomatis, since BafA consistently shows stronger in-

VOL. 79, 2011


hibitory effects on MoPn than on L2 grown in all cell lines tested, including the ATG5⫺/⫺ cells. The identity of the vATPase-bearing organelles that support chlamydial growth remains to be defined. Very recently, Ouellette et al. demonstrated that BafA is a growth inhibitor of both L2 and C. pneumoniae in human HEp-2 cells, with C. pneumoniae being particularly susceptible to the inhibitor (33). Since cycloheximide, a general inhibitor of host protein synthesis, relieved the inhibitory effect of BafA, they proposed that lysosomes serve as a critical source of oligopeptides and amino acids, and BafA exerts its effect by indirectly disabling protein degradation in lysosomes (33). However, our observation of the inability of extra amino acids to achieve the same effect on the inhibition of chlamydial growth by BafA as cycloheximide does not support this proposition. In addition, the work of Ouellette et al. did not exclude the role of nonlysosomal ATPase-bearing organelles in the growth inhibition. It is known that the chlamydial inclusion interacts with exosomes but not endosomes. Interestingly, vATPase has also been discovered in the membranes of exosomes and has been found to mediate the secretion from these vacuoles (30). Therefore, it is plausible that BafA inhibits chlamydial growth in part by targeting exosomes that interact with chlamydial inclusions. The lack of colocalization of vATPase with chlamydiae or chlamydial inclusions suggests that the interaction between inclusions and vATPase-bearing organelles, including vATPase-bearing exosomes, is likely to be indirect. It is still not certain how chlamydial infection influences autophagy activity. Whereas the level of LC3-II is increased following chlamydial infection, p62, which is degraded by autophagy, remains unchanged in HeLa and MEFs and is increased in RAW cells. Thus, a blockage of autophagy flux or at least a failure to upregulate autophagy completion may be a contributing factor to the increases in LC3-II levels. It is interesting that all three lysosomal enzymes analyzed showed significantly lower activities in chlamydia-infected RAW cells. There is no clear mechanism explaining why their activities are reduced. During infection, chlamydiae secrete CPAF and other proteases to the host cytoplasm, degrading specific host proteins. However, it is hard to envision how such secreted proteases can access and degrade lysosomal enzymes. It is possible that the biosynthesis of these enzymes is reduced, either as a part of general reduction of protein synthesis due to the shrinkage of amino acid pools caused by RB replication or as a result of specific signaling changes leading to decreased expression. Regardless of the mechanism, if the reductions of enzyme activities are sufficient to decrease lysosomal function in cells, a problem in autophagy completion as well as general degradation in the lysosomes that is independent of autophagy will arise. In summary, our findings presented in this report suggest a delicate balance between Chlamydia and its required host cell. Whereas autophagy may serve as an innate defense mechanism against chlamydiae, the parasites are able to block autophagy completion. Such interaction may have contributed to hostspecies adaptation. In addition, there appear to be at least two types of vATPase-bearing organelles that regulate chlamydial infection: while one defends against chlamydiae, possibly through autophagy, the other supports chlamydial growth. It is important to point out that such interpretation relies on a current and

4027

common belief that at low concentrations, BafA targets only the vATPase in the host cell. Therefore, it is still possible that the compound may have an additional target(s) in the host cell and/or even in the parasite. ACKNOWLEDGMENTS This work was supported by a grant from the National Institutes of Health (AI071954) to H.F. We thank Guangming Zhong for anti-LPS, anti-MOMP, anti-L2, and anti-MoPn, Ted Hackstadt for anti-IncA, Noboru Mizushima for the ATG MEFs, and Peter Lobel’s team for assistance with lysosomal enzyme assays. REFERENCES 1. Abdelrahman, Y. M., and R. J. Belland. 2005. The chlamydial developmental cycle. FEMS Microbiol. Rev. 29:949–959. 2. Al-Younes, H. M., V. Brinkmann, and T. F. Meyer. 2004. Interaction of Chlamydia trachomatis serovar L2 with the host autophagic pathway. Infect. Immun. 72:4751–4762. 3. Al-Younes, H. M., T. Rudel, and T. F. Meyer. 1999. Characterization and intracellular trafficking pattern of vacuoles containing Chlamydia pneumoniae in human epithelial cells. Cell. Microbiol. 1:237–247. 4. Al-Zeer, M. A., H. M. Al-Younes, P. R. Braun, J. Zerrahn, and T. F. Meyer. 2009. IFN-gamma-inducible Irga6 mediates host resistance against Chlamydia trachomatis via autophagy. PLoS One 4:e4588. 5. Balakrishnan, A., et al. 2006. Metalloprotease inhibitors GM6001 and TAPI-0 inhibit the obligate intracellular human pathogen Chlamydia trachomatis by targeting peptide deformylase of the bacterium. J. Biol. Chem. 281:16691–16699. 6. Balin, B. J., et al. 2008. Chlamydophila pneumoniae and the etiology of late-onset Alzheimer’s disease. J. Alzheimers Dis. 13:371–380. 7. Bea, F., et al. 2003. Chlamydia pneumoniae induces tissue factor expression in mouse macrophages via activation of Egr-1 and the MEK-ERK1/2 pathway. Circ. Res. 92:394–401. 8. Bowman, E. J., A. Siebers, and K. Altendorf. 1988. Bafilomycins: a class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells. Proc. Natl. Acad. Sci. U. S. A. 85:7972–7976. 9. Braun, P. R., et al. 2008. Competitive inhibition of amino acid uptake suppresses chlamydial growth: involvement of the chlamydial amino acid transporter BrnQ. J. Bacteriol. 190:1822–1830. 10. Campbell, L. A., and C. C. Kuo. 2004. Chlamydia pneumoniae—an infectious risk factor for atherosclerosis? Nat. Rev. Microbiol. 2:23–32. 11. Chen, L., et al. 2010. Mice deficient in MyD88 develop a Th2-dominant response and severe pathology in the upper genital tract following Chlamydia muridarum infection. J. Immunol. 184:2602–2610. 12. Cheng, W., et al. 2008. Intracellular interleukin-1alpha mediates interleukin-8 production induced by Chlamydia trachomatis infection via a mechanism independent of type I interleukin-1 receptor. Infect. Immun. 76:942– 951. 13. Cotter, T. W., G. S. Miranpuri, K. H. Ramsey, C. E. Poulsen, and G. I. Byrne. 1997. Reactivation of chlamydial genital tract infection in mice. Infect. Immun. 65:2067–2073. 14. de la Maza, L., S. Pal, A. Khamesipour, and E. Peterson. 1994. Intravaginal inoculation of mice with the Chlamydia trachomatis mouse pneumonitis biovar results in infertility. Infect. Immun. 62:2094–2097. 15. Derbigny, W. A., S. C. Hong, M. S. Kerr, M. Temkit, and R. M. Johnson. 2007. Chlamydia muridarum infection elicits a beta interferon response in murine oviduct epithelial cells dependent on interferon regulatory factor 3 and TRIF. Infect. Immun. 75:1280–1290. 16. Deretic, V. 2011. Autophagy in immunity and cell-autonomous defense against intracellular microbes. Immunol. Rev. 240:92–104. 17. Eissenberg, L. G., and P. B. Wyrick. 1981. Inhibition of phagolysosome fusion is localized to Chlamydia psittaci-laden vacuoles. Infect. Immun. 32: 889–896. 18. Eissenberg, L. G., P. B. Wyrick, C. H. Davis, and J. W. Rumpp. 1983. Chlamydia psittaci elementary body envelopes: ingestion and inhibition of phagolysosome fusion. Infect. Immun. 40:741–751. 19. Fichorova, R. N., J. G. Rheinwald, and D. J. Anderson. 1997. Generation of papillomavirus-immortalized cell lines from normal human ectocervical, endocervical, and vaginal epithelium that maintain expression of tissue-specific differentiation proteins. Biol. Reprod. 57:847–855. 20. Forgac, M. 1999. Structure and properties of the clathrin-coated vesicle and yeast vacuolar V-ATPases. J. Bioenerg. Biomembr. 31:57–65. 21. Fuertes, G., J. J. Martin De Llano, A. Villarroya, A. J. Rivett, and E. Knecht. 2003. Changes in the proteolytic activities of proteasomes and lysosomes in human fibroblasts produced by serum withdrawal, amino-acid deprivation and confluent conditions. Biochem. J. 375:75–86. 22. Gatti, J. L., S. Metayer, M. Belghazi, F. Dacheux, and J. L. Dacheux. 2005.

4028

23.

24.

25.

26.

27.

28. 29. 30.

31. 32.

33.

34.

YASIR ET AL.

Identification, proteomic profiling, and origin of ram epididymal fluid exosome-like vesicles. Biol. Reprod. 72:1452–1465. Greub, G. 2010. International Committee on Systematics of Prokaryotes. Subcommittee on the taxonomy of the Chlamydiae: minutes of the closed meeting, 21 June 2010, Hof bei Salzburg, Austria. Int. J. Syst. Evol. Microbiol. 60:2694. Gutierrez, M. G., et al. 2004. Autophagy is a defense mechanism inhibiting BCG and Mycobacterium tuberculosis survival in infected macrophages. Cell 119:753–766. Gutierrez, M. G., et al. 2005. Autophagy induction favours the generation and maturation of the Coxiella-replicative vacuoles. Cell. Microbiol. 7:981– 993. Heinzen, R. A., M. A. Scidmore, D. D. Rockey, and T. Hackstadt. 1996. Differential interaction with endocytic and exocytic pathways distinguish parasitophorous vacuoles of Coxiella burnetii and Chlamydia trachomatis. Infect. Immun. 64:796–809. Kalayoglu, M. V., and G. I. Byrne. 1998. A Chlamydia pneumoniae component that induces macrophage foam cell formation is chlamydial lipopolysaccharide. Infect. Immun. 66:5067–5072. Kuma, A., et al. 2004. The role of autophagy during the early neonatal starvation period. Nature 432:1032–1036. Levine, B., N. Mizushima, and H. W. Virgin. 2011. Autophagy in immunity and inflammation. Nature 469:323–335. Lie´geois, S., A. Benedetto, J.-M. Garnier, Y. Schwab, and M. Labouesse. 2006. The V0-ATPase mediates apical secretion of exosomes containing Hedgehog-related proteins in Caenorhabditis elegans. J. Cell Biol. 173:949– 961. Moulder, J. W. 1991. Interaction of chlamydiae and host cells in vitro. Microbiol. Rev. 55:143–190. Oda, K., Y. Nishimura, Y. Ikehara, and K. Kato. 1991. Bafilomycin A1 inhibits the targeting of lysosomal acid hydrolases in cultured hepatocytes. Biochem. Biophys. Res. Commun. 178:369–377. Ouellette, S. P., F. C. Dorsey, S. Moshiach, J. L. Cleveland, and R. A. Carabeo. 2011. Chlamydia species-dependent differences in the growth requirement for lysosomes. PLoS One 6:e16783. Pachikara, N., H. Zhang, Z. Pan, S. Jin, and H. Fan. 2009. Productive Chlamydia trachomatis lymphogranuloma venereum 434 infection in cells with augmented or inactivated autophagic activities. FEMS Microbiol. Lett. 292:240–249.

Editor: J. N. Weiser

INFECT. IMMUN. 35. Pankiv, S., et al. 2007. p62/SQSTM1 binds directly to Atg8/LC3 to facilitate degradation of ubiquitinated protein aggregates by autophagy. J. Biol. Chem. 282:24131–24145. 36. Roshick, C., H. Wood, H. D. Caldwell, and G. McClarty. 2006. Comparison of gamma interferon-mediated antichlamydial defense mechanisms in human and mouse cells. Infect. Immun. 74:225–238. 37. Schachter, J. 1999. Infection and disease epidemiology. p. 139–169. In R. S. Stephens (ed.), Chlamydia: intracellular biology, pathogenesis, and immunity. ASM Press, Washington, DC. 38. Schramm, N., C. R. Bagnell, and P. B. Wyrick. 1996. Vesicles containing Chlamydia trachomatis serovar L2 remain above pH 6 within HEC-1B cells. Infect. Immun. 64:1208–1214. 39. Scidmore, M. A., E. R. Fischer, and T. Hackstadt. 2003. Restricted fusion of Chlamydia trachomatis vesicles with endocytic compartments during the initial stages of infection. Infect. Immun. 71:973–984. 40. Sohar, I., D. E. Sleat, M. Jadot, and P. Lobel. 1999. Biochemical characterization of a lysosomal protease deficient in classical late infantile neuronal ceroid lipofuscinosis (LINCL) and development of an enzyme-based assay for diagnosis and exclusion of LINCL in human specimens and animal models. J. Neurochem. 73:700–711. 41. Toyomura, T., et al. 2003. From lysosomes to the plasma membrane: localization of vacuolar-type H⫹-ATPase with the a3 isoform during osteoclast differentiation. J. Biol. Chem. 278:22023–22030. 42. van Ooij, C., G. Apodaca, and J. Engel. 1997. Characterization of the Chlamydia trachomatis vacuole and its interaction with the host endocytic pathway in HeLa cells. Infect. Immun. 65:758–766. 43. Wang, J., et al. 2010. A genome-wide profiling of the humoral immune response to Chlamydia trachomatis infection reveals vaccine candidate antigens expressed in humans. J. Immunol. 185:1670–1680. 44. Wolf, K., E. Fischer, and T. Hackstadt. 2005. Degradation of Chlamydia pneumoniae by peripheral blood monocytic cells. Infect. Immun. 73:4560– 4570. 45. Yong, E. C., E. Y. Chi, W. J. Chen, and C. C. Kuo. 1986. Degradation of Chlamydia trachomatis in human polymorphonuclear leukocytes: an ultrastructural study of peroxidase-positive phagolysosomes. Infect. Immun. 53: 427–431. 46. Yong, E. C., E. Y. Chi, and C. C. Kuo. 1987. Differential antimicrobial activity of human mononuclear phagocytes against the human biovars of Chlamydia trachomatis. J. Immunol. 139:1297–1302.