Delivery of azithromycin to Chlamydia trachomatis-infected polarized ...

JAC

Journal of Antimicrobial Chemotherapy (1997) 39, 623–630

Delivery of azithromycin to Chlamydia trachomatis-infected polarized human endometrial epithelial cells by polymorphonuclear leucocytes T. R. Paul, S. T. Knight, J. E. Raulston and P. B. Wyrick* Department of Microbiology and Immunology, University of North Carolina School of Medicine, CB 7290, 804 FLOB, Chapel Hill, NC 27599-7290, USA An in-vitro model was designed to evaluate whether polymorphonuclear leucocytes (PMN) loaded with azithromycin could migrate and deliver the antibiotic in a bioactive form to chlamydia inclusions in polarized human endometrial epithelial (HEC-1B) cells infected with Chlamydia trachomatis. PMN chemotaxis through the extracellular matrix and between infected epithelial cells was readily observed if the HEC-1B cells had been infected with chlamydiae for 36 or 48 h. Inclusions in infected epithelial cells exposed to PMN loaded with azithromycin were initially distinguished by deformed reticulate bodies and an excessive amount of chlamydial outer membrane vesicles. As the amount of PMN-delivered antibiotic increased, chlamydial inclusions were filled with large cell envelope ‘ghosts’ which were the remnants of lysed reticulate bodies. The lethal effect of azithromycin was confirmed by a reduction in the viability of infectious progeny. Our results demonstrate that the damage to chlamydiae was due to transport and delivery of azithromycin by PMN to infected genital epithelial cells. When infected HEC-1B cells were exposed to PMN not loaded with the antibiotic, chlamydial morphology was not obviously affected yet few viable progeny could be recovered. In this case, PMN-induced damage to host epithelial cells probably interrupted chlamydial nutrient acquisition and subsequent maturation and formation of infectious progeny.

Introduction Chlamydia trachomatis is one of the most common aetiological agents of sexually transmitted disease in the world. In the USA more than four million cases of chlamydial genital infection occur annually. In women, the initial target for infection by genital serovariants (serovars D–K) is the columnar epithelium lining of the endocervical canal. The chlamydiae may spread canicularly to the endometrium, the fallopian tubes and into the peritoneal cavity giving rise to a spectrum of immunopathological sequelae which include pelvic inflammatory disease, tubal damage, ectopic pregnancy and infertility.1 Genital infection of humans with C. trachomatis sometimes elicits an inflammatory response as manifested by infiltration of intraepithelial and intraluminal polymorphonuclear leucocytes (PMN) to sites of infection and focal loss of glandular epithelial cells.2 A strong correlation has been established between genital infection with C.

trachomatis and the presence of PMN in male urethral smears and in endocervical mucus.3,4 The propensity for PMN to concentrate in areas of infection has been exploited as a means of delivering antimicrobial agents to the target site of infection. Azithromycin, an azalide antibiotic, has been shown by several researchers to be concentrated and retained for long periods inside neutrophils, macrophages and fibroblasts. 5–8 Following internalization, azithromycin accumulation is mediated by the two basic amine groups in its structure which promote greater protonation and subsequent trapping within lysosomes.5 Collectively, these studies support the potential of neutrophils to serve as efficient vehicles for the transport and delivery of azithromycin in an active form to infection sites.5,9 This mechanism generates high antibiotic concentrations sustained over extended periods in patient tissue10 and long half-lives in tissue and serum.11 Such attributes may partially explain the enhanced efficacy of azithromycin against intracellular pathogens such as Listeria monocytogenes,11

*Corresponding author

623 © 1997 The British Society for Antimicrobial Chemotherapy

T. R. Paul et al. Legionella micdadei12 and C. trachomatis.13,14 In particular, clinical trials in patients with chlamydial urethritis or cervicitis indicate that a single, 1 g oral dose of azithromycin provides equivalent success to the standard, twicedaily, 100 mg dosages of doxycycline given for 7 days.15 Details of how Chlamydia interacts with genital epithelia to evoke recruitment of PMN are not known. Are PMN recruited to infected tissues by complement-fixing chlamydial antigens or mediated by chemokines secreted as a consequence of bacterial–epithelial cell interactions, or both? As a starting point to address these questions, we developed an in-vitro model to study PMN chemotaxis to polarized human endometrial epithelial (HEC-1B) cells infected with C. trachomatis serovar E. In order to evaluate the feasibility of our model, it was necessary to demonstrate that PMN migration had occurred. The approach taken was to load isolated neutrophils with azithromycin followed by incubating them with the infected polarized monolayer, the rationale being that PMN loaded with antibiotic would migrate and deliver azithromycin to infected cells. The effects of the released antibiotic on the structural integrity and morphology of chlamydiae were examined using light and transmission electron microscopy.

Materials and methods Growth of chlamydiae C. trachomatis E/UW-5/CX, a human urogenital isolate, was obtained from C. C. Kuo and S. P. Wang, University of Washington, Seattle, WA, USA. Growth of C. trachomatis serovar E in McCoy cells by the microcarrier bead technique16 and subsequent harvest and titration of the elementary bodies were performed as previously described by Moorman et al.17 and Wyrick et al.18

Epithelial cells A human endometrial cell line, HEC-1B (HTB-113, American Type Culture Collection, Rockville, MD, USA), originally obtained from a patient with endometrial carcinoma, was used in this study. HEC-1B cells were shown to be free of mycoplasma contamination and were grown in a polarized manner as previously described by Wyrick et al.19 Epithelial cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal calf serum (Hyclone, Logan, UT, USA) and 10 mM HEPES, pH 7.3 (DMEM).

Azithromycin Azithromycin, obtained from Pfizer Laboratories (Groton, CT, USA), was prepared as a sterile 1 g/L stock solution in phosphate buffer, pH 6.0. Dilutions were made in DMEM. Antibiotic solutions were stored at 4°C in the dark.

Isolation of PMN and loading with azithromycin Human PMN were obtained from a single donor and isolated from heparinized peripheral blood by a one-step Ficoll-Hypaque separation procedure. The leucocytes were washed twice in phosphate-buffered saline, pH 7.4 and then resuspended in DMEM supplemented with 5% human serum (inactivated by heating at 56°C for 30 min) to yield a final cell suspension of 1 107 PMN/mL. Cells were tested for viability by the Trypan Blue dye exclusion technique and found to be 95% viable. PMN were incubated with 25 mg/L azithromycin for 1 h at 35°C to ensure that optimal concentrations of antibiotic were available for uptake. Azithromycin-loaded PMN were then washed three times in DMEM supplemented with 5% heatinactivated human serum to remove excess antibiotic and resuspended in fresh medium. The third DMEM wash was saved and analysed for the presence of residual azithromycin.

PMN model system PMN migration chambers were constructed by cementing a nylon screen (8 mm diameter and 8 m porosity; BioDesign, Carmel, NY, USA) between two Plexiglass rings (6 mm diameter and 3 mm height) using a Cyanoacrylate adhesive (Pacer Technology, CA, USA). Following sterilization of the chambers by UV irradiation for 2–12 h, the nylon screen was coated with extracellular matrix (ECM; BioTechnologies Inc., Staughton, MA, USA) which was allowed to polymerize to form a gel on both sides of the nylon screen. The upper chamber was seeded with approximately 5 104 HEC-1B cells/mL and grown in DMEM followed by incubation in 5% CO2 at 35°C for 5 days or until cell monolayers became confluent and polarized. Polarized HEC-1B cell monolayers were inoculated with 50 L of C. trachomatis serovar E EB stock at a dilution demonstrated to yield a 50% infection, and the cultures were incubated for 2 h at 35°C. Non-adherent bacteria were removed by washing the monolayer twice in complete medium and the infected monolayers were then replenished with fresh medium and re-incubated at 35°C. At 24, 36 or 48 h post-infection, medium was removed from the culture chambers and the chambers were inverted. The lower chambers were filled with 1% agarose (SeaPlaque, FMC BioProducts, Rockland, MD, USA), prepared in DMEM containing 5% heat-inactivated human serum, and allowed to solidify for 5 min. A 3 mm diameter well was excised from the centre of the agarose layer, and 10 L of the PMN suspension (1 106 PMN/10 L), previously incubated in the presence or absence of azithromycin, was added to each agarose well. A drop of fresh agarose was applied to the well to prevent drying and the chambers were re-incubated at 35°C for an additional 1.5, 3, 5 or 12 h to allow for PMN migration (Figure 1a).

624

Delivery of azithromycin to C. trachomatis by PMN

(a)

(b)

Figure 1. The co-culture model system. (a) Cross-section of a schematic representation of the co-culture model system. (b) Orientation of sample strips for sequential sectioning. Broken lines indicate sectioning regions.

Viability assay for chlamydiae exposed to azithromycin In order to determine whether azithromycin had a bactericidal effect upon chlamydial progeny, HEC-1B cells infected with chlamydiae for 36 h were exposed for 19 h to PMN loaded with azithromycin, PMN not loaded with antibiotic, or azithromycin alone in the absence of PMN. Infected HEC-1B cells were harvested from the PMN migration chambers by solubilizing the ECM with 10 U/cm2 Dispase solution (Collaborative Biomedical Products, Bedford, MA, USA) and washed gently three times in DMEM by centrifugation at 500g for 10 min. The resulting pellet was resuspended in 1 mL DMEM, the epithelial cells were disrupted by sonication and the lysate was subjected to centrifugation to remove large cell debris. The supernatant was centrifuged at 15,000g for 10 min to harvest the chlamydiae, and the pellet was resuspended in 0.1 mL DMEM. Serial two-fold dilutions of chlamydial progeny were adsorbed on to fresh, subconfluent HEC-1B monolayers grown on glass coverslips and then incubated for 48 h. Infected monolayers were fixed with cold absolute methanol, stained with a pool of fluorescein isothiocyanate (FITC)-labelled monoclonal antibodies directed against C. trachomatis major outer membrane protein (MOMP) (Syva Microtrak, San Jose, CA, USA) and examined for the presence of fluorescent chlamydial inclusions as a measure of chlamydial viability.

paraformaldehyde–0.25% glutaraldehyde in 0.1 M Sorenson’s buffer for 18 h at 4°C. Samples were washed three times in buffer and then dehydrated in 70% and 90% methanol for 1 h each at 4°C. Infiltration was carried out at 4°C in 90% methanol–JB-4 resin (PolySciences Inc., Warrington, PA, USA) at a ratio of 1:1 and 1:2 for 1 h each, followed by 100% JB-4 resin for 1 h and finally in fresh resin for 18 h. Infiltrated samples were embedded in fresh resin and polymerized for 90 min at room temperature. Thick (2–5 m) sections were cut on a Reichert UM-2 ultramicrotome (Leica Inc., Deerfield, IL, USA) and stained with Toluidine Blue for morphological examination.

Transmission electron microscopy Sample processing for embedding in Epon–Araldite resin was performed as described previously by Wyrick et al. 20 Before embedding in the plastic resin, the co-culture chambers were cut into four strips (6 mm long and 1.5 mm wide). Each strip was positioned so that one of the two ends that represented the periphery of the monolayer was located at the tip of the embedding block. This sample orientation facilitated a sequential examination of the morphology and structure of the sample from the perimeter of the monolayer inward to the centre region directly above the PMN well (Figure 1b).

Results

Light microscopy

PMN migration

Infected and uninfected samples were washed twice in 0.1 M Sorenson’s buffer, pH 7.2, prior to fixation in 2%

Uninfected HEC-1B cells, or epithelial cells infected with C. trachomatis for 24 h (Figure 2a) failed to evoke PMN

625

T. R. Paul et al.

ID ONLY See Film supplied

Figure 2. Light photomicrographs of polarized HEC-1B cells infected with C. trachomatis for 24 or 36 h and then exposed to PMN for 1.5–5 h. (a) Cells infected for 24 h. Note the absence of PMN migration. (b) Cells infected for 36 h then exposed to PMN for 1.5 h. Note migration of PMN through the ECM and contact with the basolateral domain of the polarized epithelial monolayer. (c) Cells infected for 36 h then exposed to PMN for 5 h. Note extensive PMN migration through the epithelial cell layer and out on to the apical cell surface. Arrowheads indicate inclusions and arrows denote PMN. Magnification for all panels are 1400.

chemotaxis. However, when HEC-1B cells were infected with chlamydiae for 36 or 48 h, PMN migration occurred. Within 1.5 h of addition of PMN to the well, migration was apparent by a directional movement of PMN out of the well, through the ECM layer, and towards the basolateral domain of the epithelial cells (Figure 2b). During migration through the epithelial layer, PMN were in close proximity to infected cells containing chlamydial inclusions. Extensive intraepithelial migration culminated in the presence of PMN on the apical surface of the polarized epithelial cells (Figure 2c). The magnitude of these events was enhanced during prolonged (3–12 h) exposure to PMN. Control experiments using PMN alone revealed that PMN migration was not affected by loading PMN with azithromycin (data not shown).

Effects of PMN migration on epithelial cells and chlamydiae There were minimal alterations in the structure and morphology of chlamydiae in epithelial cells exposed to PMN not loaded with azithromycin during the incubation periods studied. However, PMN appeared to induce large, ellipsoidal, electron-translucent vacuoles within the epithelial cell cytoplasm (data not shown). More vacuoles were observed in epithelial cells located adjacent to the PMN well than in epithelial cells at a distance from the well. Moreover, HEC-1B cell integrity deteriorated with prolonged exposure to PMN, as demonstrated by abnormally large intracellular and intercellular vacuoles, as well as some cell lysis.

Effects of azithromycin on host cells and chlamydiae The morphology of chlamydiae in 36 or 48 h infected epithelial cells exposed to chemotactic PMN loaded with azithromycin was dependent on the location of the infected cells in the antibiotic concentration gradient. Chlamydiae in inclusions in infected epithelial cells at the periphery of the culture chamber, the zone devoid of PMN, consisted of intact elementary bodies, reticulate bodies and intermediate bodies which were similar in morphology to their counterparts in control infected epithelial cells (Figure 3a). Infected epithelial cells in the middle zone of the chamber were exposed to gradually increasing numbers of migrating PMN. Inclusions in these cells were marked by a slight reduction in reticulate bodies and an increase in the presence of outer membrane vesicles (Figure 3b). Interestingly, there was a propensity for chlamydiae and their outer membrane vesicles to associate with the inclusion membrane. The most fascinating aspect of this interaction was the manner in which the outer membrane vesicles and cell envelope ‘ghosts’ appeared to exit the inclusion and become concentrated in extra-inclusion vesicles which were located in the host cell cytoplasm (Figure 3c). This phenomenon was also observed, though to a lesser extent, in infected cells in the absence of azithromycin. The most profound effect of azithromycin was seen in epithelial cells located in close proximity to and positioned directly above the PMN well. Inclusions in these cells contained reticulate bodies that displayed a spectrum of antibiotic-induced effects which included deformed reticulate bodies containing conspicuous, wavy membranes, reticu-

626



Figure 3. Transmission electron photomicrographs of polarized HEC-1B cells infected with C. trachomatis for 36 h and exposed to PMN loaded with azithromycin. (a) An inclusion in infected epithelial cells at the periphery of the culture chamber containing elementary bodies, intermediate bodies and reticulate bodies. (b) An inclusion in infected epithelial cells in the middle zone of the culture chamber containing few reticulate bodies, some elementary bodies and numerous outer membrane vesicles. (c and d) Inclusions in infected epithelial cells located directly above the PMN well contain primarily damaged reticulate bodies and cell envelope ‘ghosts’. In panel c, note the outer membrane vesicles and cell envelope ‘ghosts’ inside extra-inclusion vesicles (arrowheads). Magnification for all panels is 10,000.

late bodies lacking most of their cytoplasmic contents, and large cell envelope ‘ghosts’ which were the remnants of lysed reticulate bodies (Figure 3c and d). In contrast, the morphology of elementary bodies and intermediate forms revealed minimal structural effects from antibiotic action.

Activity of azithromycin HEC-1B cells infected with chlamydiae for 36 h and then exposed to azithromycin loaded PMN for 19 h resulted in a

reduction in the viability (86%) of the chlamydial progeny upon subpassage to fresh HEC-1B cells (Table). This reduction in infectious chlamydial progeny was comparable (99%) to that observed when infected HEC-1B cells were exposed for 19 h to azithromycin alone added to the tissue culture medium. To confirm that the loss of chlamydial viability was due to delivery of azithromycin to infected HEC-1B cells by migrating PMN rather than leakage of azithromycin out of PMN into the well fluid and subsequent diffusion, the DMEM washes from azithro-

627

T. R. Paul et al. Table. Viability of progeny obtained after subpassage from HEC-1B cells infected with C. trachomatis serovar E and exposed to azithromycin Sample added to culture chamber well

Inhibition of chlamydial inclusions (%)

DMEM alone, no azithromycin

0

Azithromycin added exogenously to apical side of chamber

99

PMN loaded with azithromycin

86

Wash fluid from PMN loaded with azithromycin

0

Wash fluid from glass beads exposed to azithromycin

0

Azithromycin MBC90 (0.5 mg/L)

93

PMN not loaded with azithromycin

70

mycin-loaded PMN were added to the wells of the infected culture chambers. Following incubation for 19 h the chlamydial progeny were harvested and assessed for viability on subpassage. There was no loss of chlamydial viability under these circumstances. In the same vein, sterile glass beads (1 106; 75–150 m diameter) were exposed to 25 mg/L azithromycin for 1 h at 35°C and then washed three times in DMEM-supplemented medium. The third wash fluid was added to the wells of infected culture chambers for 19 h. The chlamydiae retained viability on subpassage indicating residual azithromycin was not present for diffusion into the infected HEC-1B monolayers. It should be noted there was also a reduction in viability of chlamydial progeny obtained from infected HEC-1B cells exposed to migrating PMN not loaded with azithromycin (Table). In this case, the considerable PMN-induced damage to host epithelial cells probably deprived metabolically active reticulate bodies of essential metabolites and energy. As a consequence, maturation of reticulate bodies to infectious elementary bodies was probably interrupted and thus, the number of viable progeny was also reduced.

Discussion These studies have shown that an in-vitro cell culture system designed to mimic as closely as possible the endometrial mucosa in vivo can serve as a model for (i) dissecting the initial parameters signalling the inflammatory response to epithelial cells infected by a sexually transmitted disease pathogen, (ii) examining antibiotic delivery by biologically active PMN undergoing chemotaxis towards infected epithelial sites and (iii) analysing the efficacy of the PMN-delivered antibiotic on bacterial

pathogens residing in polarized epithelial cells. The fact that PMN did not migrate towards endometrial epithelial cells infected with chlamydiae for 24 h but did migrate towards cells infected for 36 h suggests that discriminatory signals are manifested in the in-vitro system. Several investigators have reported the appearance of chlamydia-specific LPS on the surface of chlamydiainfected host cells prior to rupture of the mature chlamydial inclusion.21,22 Perhaps LPS-containing outer membrane blebs in pinched-off inclusion membrane vesicles (Figure 3c) traffic to and fuse with the epithelial plasmalemma. This exposed complement-fixing molecule, on generation of C5a, could serve as one source of chemotactic signal. Wilde et al.23 have suggested that the accumulation of LPS in the eukaryotic membrane decreases membrane fluidity, as assessed by spin probe electron spin resonance spectroscopy. In an elegant study by Raulston14 on the pharmacokinetics of azithromycin in polarized human endometrial epithelial cells infected with C. trachomatis, there was a decrease in azithromycin uptake by the infected HEC-1B cells between 24 and 48 h post-inoculation. It was not clear whether the reduction in azithromycin internalization was due to altered membrane fluidity of the infected epithelial cell or altered transport kinetics due to energy parasitism by chlamydiae. However, it is clear from the morphology and viability studies reported here that azithromycin, delivered by chemotactic PMN, does enter 36 h infected genital epithelial cells in concentrations sufficient to kill metabolically active chlamydiae. Chlamydiae in inclusions in infected endometrial epithelial cells exposed to PMN-delivered azithromycin displayed a spectrum of antibiotic-induced effects which, from a morphological perspective, appeared specific to reticulate bodies. These effects ranged from perturbation of the

628



Figure 4. Viability of chlamydiae obtained from infected epithelial cells exposed to PMN-delivered azithromycin. Chlamydiae were harvested from epithelial cells in the co-culture chambers, subpassaged to fresh HEC-1B cells and examined by fluorescence microscopy for chlamydial inclusions. (a) Inclusions formed by chlamydial progeny harvested from control infected epithelial cells not exposed to PMN nor antibiotic. (b) Inclusions formed by chlamydial progeny harvested from infected epithelial cells exposed to PMN not loaded with azithromycin. (c) Inclusions formed by chlamydial progeny harvested from infected epithelial cells exposed to azithromycin-loaded PMN for 19 h. Note the decrease in inclusion population and the abnormally reduced inclusion size. Magnification for all panels is 1900.

outer membrane which, in turn, yielded a proliferation of outer membrane vesicles to lysis of reticulate bodies. Similar morphological abnormalities were observed by Wyrick et al.13 and Raulston14 when C. trachomatis-infected polarized HEC-1B cells were exposed to culture medium containing varying concentrations of exogenously added azithromycin, and by Patton et al.24 in azithromycinexposed human amniotic cells infected with the same serovar of C. trachomatis. The reasons for the multiple morphological effects reported in this study are probably two-fold. First, they are a consequence of exposure of the infected epithelial cells to a gradient of azithromycin concentrations. The most rapid deleterious effects of antibiotic action were visualized in inclusions in infected cells in close proximity to the well containing the greatest number of azithromycin loaded PMN, whereas chlamydiae showing minimal damage were in peripheral epithelial cells exposed to fewer migrating PMN. Second, the asynchronous nature of the chlamydial development cycle would also influence the nature of the azithromycin-induced effects on chlamydiae. The excellent studies of Engel25 clearly showed the activity of azithromycin in chlamydial protein synthesis was rapid, occurring within 5 min. This would explain why the most obvious morphological changes occur in metabolically active reticulate bodies. However, with a generation time of approximately 3 h, there is a considerable delay before reticulate body lysis, leaving behind residual cell envelope ‘ghosts’. The combination of both slow and asynchronous growth of chlamydiae underscores the use of the polarized cell system for evaluating the efficacy of

azithromycin towards chlamydiae because polarized HEC1B cells concentrate greater amounts of azithromycin and retain the antibiotic for longer periods of time than their non-polarized counterparts.13,14

Acknowledgements This study was supported by grants from Pfizer Inc. (95-S-0572) and the National Institutes of Health, National Institute of Allergy and Infectious Diseases (NIAID) grant 2 U19 AI31496 to the North Carolina STD Cooperative Research Center. The polarized model system was developed under NIAID grant AI13446.

References 1. Paavonen, J. & Wolner-Hanssen, P. (1989). Chlamydia tracho matis: a major threat to reproduction. Human Reproduction 4, 111–24. 2. Kiviat, N. B., Paavonen, J. A., Brockway, J., Critchlow, C. W., Brunham, R. C., Stevens, C. E. et al. (1985). Cytological manifestations of cervical and vaginal infections. 1. Epithelial and inflammatory cellular changes. Journal of the American Medical Association 253, 989–96. 3. Arya, O. P., Mallinson, H., Andrews, B. E. & Sillis, M. (1984). Diagnosis of urethritis: role of polymorphonuclear leukocyte counts in Gram-stained urethral smears. Sexually Transmitted Diseases 11, 10–7. 4. Brunham, R. C., Paavonen, J., Stevens, C. E., Kiviat, N., Kuo, C. C., Critchlow, C. W. et al. (1984). Mucopurulent cervicitis—the

629

T. R. Paul et al. ignored counterpart in women of urethritis in men. New England Journal of Medicine 311, 1–6.

urethritis and cervicitis. New England Journal of Medicine 327, 921–5.

5. Gladue, R. P., Bright, G. M., Isaacson, R. E. & Newborg, M. F. (1989). In vitro and in vivo uptake of azithromycin (CP-62,993) by phagocytic cells: possible mechanism of delivery and release at sites of infection. Antimicrobial Agents and Chemotherapy 33, 277–82.

16. Tam, J. E., Knight, S. T., Davis, C. H. & Wyrick, P. B. (1992). Eukaryotic cells grown on microcarrier beads offer a cost-efficient way to propagate Chlamydia trachomatis. Biotechniques 13, 374–8.

6. Gladue, R. P. & Snider, M. E. (1990). Intracellular accumulation of azithromycin by cultured human fibroblasts. Antimicrobial Agents and Chemotherapy 34, 1056–60. 7. Meyer, A. P., Bril-Bazuin, C., Mattie, H. & van den Broek, P. J. (1993). Uptake of azithromycin by human monocytes and enhanced intracellular antibacterial activity against Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 37, 2318–22. 8. Panteix, G., Guillaumond, B., Harf, R., Desbos, A., Sapin, V., Leclercq, M. et al. (1993). In-vitro concentration of azithromycin in human phagocytic cells. Journal of Antimicrobial Chemotherapy 31, Suppl. E, 1–4. 9. Frank, M. O., Sullivan, G. W., Carper, H. T. & Mandell, G. L. (1992). In vitro demonstration of transport and delivery of antibiotics by polymorphonuclear leukocytes. Antimicrobial Agents and Chemotherapy 36, 2584–8. 10. Foulds, G., Shepard, R. M. & Johnson, R. B. (1990). The pharmacokinetics of azithromycin in human serum and tissues. Journal of Antimicrobial Chemotherapy 25, Suppl. A, 73–82. 11. Girard, A. E., Girard, D., English, A. R., Gootz, T. D., Cimochowski, C. R., Faiella, J. A. et al. (1987). Pharmacokinetic and in vivo studies with azithromycin (CP-62,993), a new macrolide with an extended half-life and excellent tissue distribution. Antimicrobial Agents and Chemotherapy 31, 1948–54. 12. Donowitz, G. R. & Earnhardt, K. I. (1993). Azithromycin inhibition of intracellular Legionella micdadei. Antimicrobial Agents and Chemotherapy 37, 2261–4. 13. Wyrick, P. B., Davis, C. H., Knight, S. T. & Choong, J. (1993a). In-vitro activity of azithromycin on Chlamydia trachomatis infected, polarized human endometrial epithelial cells. Journal of Antimicrobial Chemotherapy 31, 139–50. 14. Raulston, J. E. (1994). Pharmacokinetics of azithromycin and erythromycin in human endometrial epithelial cells and in cells infected with Chlamydia trachomatis. Journal of Antimicrobial Chemotherapy 34, 765–76. 15. Martin, D. H., Mroczkowski, T. F., Dalu, Z. A., McCarty, J., Jones, R. B., Hopkins, S. J. et al. (1992). A controlled trial of a single dose of azithromycin for the treatment of chlamydial

17. Moorman, D. R., Sixbey, J. W. & Wyrick, P. B. (1986). Interaction of Chlamydia trachomatis with human genital epithelium in culture. Journal of General Microbiology 132, 1055–67. 18. Wyrick, P. B., Gerbig, D. G., Knight, S. T. & Raulston, J. E. (1996). Accelerated development of genital Chlamydia trachomatis serovar E in McCoy cells grown on microcarrier beads. Microbiol Pathogenesis 20, 31–40. 19. Wyrick, P. B., Davis, C. H., Knight, S. T., Choong, J., Raulston, J. E. & Schramm, N. (1993). An in vitro human epithelial cell culture system for studying the pathogenesis of Chlamydia trachomatis. Sexually Transmitted Diseases 20, 248–56. 20. Wyrick, P. B., Choong, J., Davis, C. H., Knight, S. T., Royal, M. O., Maslow, A. S. et al. (1989). Entry of genital Chlamydia trachomatis into polarized human epithelial cells. Infection and Immunity 57, 2378–89. 21. Karimi, S. T., Schloemer, R. H. & Wilde, C. E. (1989). Accumulation of chlamydial lipopolysaccharide antigen in plasma membranes of infected cells. Infection and Immunity 57, 1780–5. 22. Wyrick, P. B., Choong, J., Knight, S. T., Goyeau, D., Stuart, E. S. & MacDonald, A. B. (1994). Chlamydia trachomatis antigens on the surface of infected human endometrial epithelial cells. Immunology and Infectious Diseases 4, 131–41. 23. Wilde, C. E., Karimi, S. T. & Haak, R. A. (1986). Cell surface alterations during chlamydial infection. In Microbiology—1986 (Leive, L., Bonventre, P. F., Morello, J. A., Silver, S. D, & Wu, H. C., Eds), pp. 96–8. American Society for Microbiology, Washington, DC. 24. Patton, D. L., Wang, S.-K. & Kuo, C.-C. (1995). The activity of azithromycin on the infectivity of Chlamydia trachomatis in human amniotic cells. Journal of Antimicrobial Chemotherapy 36, 951–9. 25. Engel, J. N. (1992). Azithromycin-induced block of elementary body formation in Chlamydia trachomatis. Antimicrobial Agents and Chemotherapy 36, 2304–9.

Received 23 August 1996; returned 21 October 1996; revised 7 November 1996; accepted 17 December 1996

630