Accelerated development of genitalChlamydia trachomatisserovar E in McCoy cells grown on microcarrier beads

Microbial Pathogenesis 1996; 20: 31–40

Accelerated development of genital Chlamydia trachomatis serovar E in McCoy cells grown on microcarrier beads Priscilla B. Wyrick,∗ Donald G. Gerbig Jr., Stephen T. Knight, and Jane E. Raulston Department of Microbiology and Immunology, University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599-7290, U.S.A. (Received August 8, 1995; accepted in revised form October 12, 1995)

Wyrick, P. B. (Department of Microbiology and Immunology, University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599-7290, U.S.A.), D. G. Gerbig Jr., S. T. Knight and J. E. Raulston. Accelerated development of genital Chlamydia trachomatis serovar E in McCoy cells grown on microcarrier beads. Microbial Pathogenesis 1996; 20: 31–40. Chlamydia trachomatis serovar E is a major cause of bacterially-acquired sexually transmitted infections. Stock cultures of these obligate intracellular bacteria are often propogated in McCoy cells. We recently reported that greater infectious titers of chlamydiae could be obtained if the McCoy cells were cultured on collagen-coated microcarrier beads versus plastic flasks, although the reason for the difference in efficiency was not clear. This study analyzed the development of C. trachomatis grown in McCoy cells by the two methods. Transmission electron microscopy analysis revealed an accelerated chlamydial development, with maturation of reticulate bodies into elementary bodies sooner in McCoy cells grown on the porous substratum. Comparison of particle counts versus infectivity titers indicated the production of fewer numbers of elementary bodies but which were highly infectious sooner from the infected McCoy cell-microcarrier bead cultures than from duplicate infected McCoy cell cultures grown in plastic tissue culture flasks. 1996 Academic Press Limited

Key words: Chlamydia trachomatis; microcarrier beads; accelerated development cycle; infectivity titration; particle count.

Introduction Members of the Chlamydia trachomatis biovar are responsible for the blinding eye disease trachoma (serovars A–C) and epidemic sexually transmitted infections (serovars D-K). Serovars A–C occur primarily in Asia and Africa and cause an estimated 500 million ocular infections, leaving as many as 9 million people blind.1 Although no estimates are available, a conservative guess is that several hundred millions of persons worldwide have chlamydial sexually transmitted infections. Thus, the magnitude of chlamydia-induced diseases in a given year may approach one billion infected persons.2,3 The target cell in vivo for these obligate intracellular mucosal pathogens is the epithelial cell of the conjunctiva and lining of the genital tract. The initial stage ∗ Author to whom correspondence should be addressed: Priscilla B. Wyrick, Department of Microbiology and Immunology, University of North Carolina School of Medicine, Campus Box #7290, 804 FLOB, Chapel Hill, NC 27599, U.S.A. 0882–4010/96/010031+10 $12.00/0

 1996 Academic Press Limited

32

P. B. Wyrick et al.

of the chlamydial infectious process begins with binding and entry of the infectious form of the organism, the elementary body (EB), to its target host cell. Following internalization, EB-containing endosomes fuse with one another, but not with lysosomes, to maintain a single membrane-bound microcolony in which to develop. Generally, within several hours, at least in in vitro cultured infected cells, the non-metabolically active EB transforms into a metabolically active intracellular form termed reticulate body (RB). Nucleic acid and protein synthesis are triggered and the growing RB divide by binary fission, beginning around 8–12 h postinfection (pi). As the number of RB progeny increases, the membrane-bound microenvironment expands; when it is visible by light microscopy, around 24–36 h, it is termed an inclusion. Depending on the species, as many as 100–500 chlamydiae may be contained per inclusion. By 48–60 h, non-infectious RB begin an asynchronous conversion back to infectious EB and rupture of the inclusion occurs anywhere from 72–96 h pi. RB are unable to bind to and infect epithelial cells; since they are also osmotically fragile, they usually cannot survive in the extracellular milieu. Thus, the infectious process is perpetuated only by EB when they attach to and enter a new epithelial cell. For many years, the traditional method for maintaining isolates of the trachoma agent was in the yolk sac of embryonated eggs. It was not until 1965 that Gordon and Quan4 first reported the growth and adaptation of these organisms in a cultured cell line, specifically in irradiated McCoy cells. Even though infection with the trachoma biovar agents is not invasive but confined strictly to superficial conjunctival or luminal columnar epithelial cells, investigators worldwide have continued to maintain stock cultures of the bacteria in the murine fibroblast cell line. Propagation of these intracellular bacteria is quite challenging. Defining conditions for chlamydial growth requires elucidating the proper conditions for growth of its eukaryotic host cell since chlamydiae direct their metabolism through the commandeered host cell. The health of the host cell is directly related to survival of chlamydiae. We recently reported that growing chlamydiae-infected McCoy cells on microcarrier beads (collagen-coated, dextran matrix-filled spheres) instead of plastic tissue culture flasks resulted in greater yields of infectious chlamydial progeny.5,6 The reason for the increased efficiency was not clear; it was not known if an increased percentage of the particles was infectious or if greater quantities of chlamydial particles were produced per infected McCoy cell. The purpose of this study was to analyze inclusion development and chlamydial infectivity in McCoy cells grown on the Cytodex 3 microcarrier beads and in tissue culturetreated plastic flasks. In the course of these comparative studies in which the only variable parameter was growth of infected cultures on permeable versus nonpermeable substrata, some interesting differences in the morphological characteristics of the inclusions as well as the progression of the chlamydial developmental cycle were observed. Such studies may provide clues for more appropriate designs of experimental cultivation methods such that the data generated are more relevant to the pathogenesis of chlamydiae in vivo.

Results Morphological comparison of infected McCoy cells from microcarrier beads versus flask cultures Duplicate aliquots of a stock inoculum of C. trachomatis serovar E EB were used to infect an equal number of McCoy cells cultivated either in flasks or

C. trachomatis accelerated development

33

on microcarrier beads. The infected cultures, incubated in medium containing cycloheximide, were harvested at 7, 24, 48 and 72 h pi, processed for Epon embedding, and stained thin sections were examined by transmission electron microscopy. The early intracellular events—EB-EB endosome fusion, transformation of EB into RB, and initial RB development—appeared from a morphological view to proceed essentially the same in the two systems. However, by 48 h pi there were noticeable differences in inclusion morphology. Inclusions in infected McCoy cells grown on microcarrier beads contained considerably more EB (Fig. 1C) than inclusions in infected McCoy cells grown in flasks (Fig. 1A and B). In addition, several individual EB were visualized in the cytoplasm, outside the inclusion, in bead-grown McCoy cells (Fig. 1D). A third surprising observation was the apparent difference in intra-inclusion material, presumably glycogen. In the flask-grown cultures, the glycogen had a characteristic dense, granular appearance (Fig. 1E) whereas in the bead-grown cultures, the glycogen deposits were larger and more globular in appearance (Fig. 1F). It should be noted that the composite electron photomicrographs presented are representative of numerous photographs from repeated samples; they are not examples of rare, isolated events. By 72 h pi, morphological differences between chlamydial inclusions in the two infected McCoy cell cultures were even more dramatic. The architecture of inclusions appeared intact in flask-grown McCoy cells and the inclusions contained predominantly RB with the expected few EB and some intermediate forms (Fig. 2A and B). In marked contrast, intact inclusions in bead-grown McCoy cells were not obvious. It was as if there had been gradual disintegration of the inclusion membrane and numerous, mature EB spilled out into the host cell cytoplasm (Fig. 2C and D). Concerned that the much accelerated maturation of RB to EB in the microcarrier bead cultures was more reflective of the faster growing Lymphogranuloma venereum biovar and that cross-contamination of biovars might have occurred in our laboratory, six coded samples of progeny were sent to Dr C. C. Kuo, University of Washington, Seattle, for typing. Dr Kuo confirmed that the chlamydiae purified from the microcarrier bead cultures was, indeed, C. trachomatis serovar E.

Infectivity titers and particle counts Duplicate cultures of equal numbers of McCoy cells propagated either in plastic flasks or on microcarrier beads were infected with a C. trachomatis serovar E EB inoculum calculated to infect 50% of the McCoy cells. At 48, 54, and 60 h pi, the infected fibroblasts were lysed by gentle sonication, and a crude preparation of chlamydiae was obtained by differential centrifugation for (i) infectivity titration on fresh McCoy cells cultured on coverslips and (ii) particle count determination. For the infectivity determination, the number of inclusions per 400 cells was counted and recorded as the percentage of infected cells (Fig. 3). At 48 and 54 h pi, the number of inclusion forming units generated in McCoy cells grown on microcarrier beads was clearly greater than that from the flask-grown cultures. For example, a 1/8 dilution of chlamydial progeny harvested from infected flaskgrown McCoy cells at 48 and 54 h resulted, on passage to fresh McCoy monolayers, in 9 and 28% infected cells, respectively. In contrast, a 1/8 dilution of chlamydial progeny harvested from infected microcarrier bead-grown McCoy cells at 48 and 54 h resulted, on passage, in 62% and 93% infected cells. However, by 60 h pi, the number of inclusion forming units generated from the two infected culture

34

P. B. Wyrick et al.

Fig. 1. Transmission electron photomicrographs of C. trachomatis serovar E infected McCoy cells harvested at 48 h pi. (A and B) Representative inclusions from infected flask-grown cultures contain primarily RB. Bars 1.0 lm. (C and D) Representative inclusions from infected bead-grown cultures contain primarily EB, some of which are visible outside the inclusion in the host cytoplasm (D). Bars 1.0 lm. (E) The appearance of the glycogen in inclusions from flask-grown cultures was more dense and granular than (F) the glycogen in inclusions from bead-grown cultures, which was larger and more globular. Bars 0.1 lm.

C. trachomatis accelerated development

35

Fig. 2. Transmission electron photomicrographs of C. trachomatis serovar E infected McCoy cells harvested at 72 h pi. (A and B) Representative intact inclusions from flask-grown cultures contain a mixture of EB and RB. Bars 1.0 lm. (C and D) Representative infected cell from bead-grown cultures contain almost exclusively EB. The inclusion membrane appears to have disintegrated and numerous, mature EB have spilled into the host cell cytoplasm. Bars 1.0 lm.

systems was essentially the same, resulting in 64–66% infected cells for a 1/8 dilution of progeny. To determine particle counts, duplicate 10 ll drops of the crude preparation of chlamydiae harvested from each culture at the times indicated were dried on a microscope slide, stained with a pool of commercially obtained fluoresceinconjugated monoclonal antibodies directed against the C. trachomatis major outer membrane protein (Syva MicroTrak) and the number of fluorescent chlamydial

P. B. Wyrick et al.

36

100

(a)

% Infectivity

75

50 * 25

0

*

2

*

4

8

* 16

* 32

64

% Infectivity

100

128

(b)

75

50 * 25 * 0

2

4

8

16

* 32

64

128

100

% Infectivity

(c) 75

50

25

0

2

4

16 8 32 Reciprocal dilution

64

128

Fig. 3. Infectivity titers of C. trachomatis serovar E progeny harvested at 48, 54 and 60 h pi from 50% infected McCoy cells grown in plastic flask cultures or in microcarrier bead cultures. (A) Infectivity titers of progeny at 48 h pi from flask-grown cultures (Φ) were much lower than titers of progeny from bead-grown cultures (Ε). (B) Infectivity titers of progeny at 54 h pi from flask grown cultures (Φ) were generally lower than titers of progeny from bead-grown cultures (Ε). (C) Infectivity titers of progeny at 60 h pi from both flask-grown cultures (Φ) and bead-grown cultures (Ε) were comparable. ∗, P<0.01 compared to bead cultures at the same dilution.

particles was counted. The number of chlamydial particles released from flaskgrown McCoy cells gradually increased the longer the incubation times whereas the number of particles released from the bead-grown cultures was relatively stable between 48 and 54 h pi and then increased by 30% at 60 h (Table 1). Adhesion of eukaryotic cells to extracellular matrix components, such as collagen, laminin, fibronectin and vitronectin, triggers outside-to-inside molecular

C. trachomatis accelerated development

37

Table 1 Comparison of particle counts of C. trachomatis serovar E harvested from 50% infected McCoy cells grown in flask cultures versus microcarrier bead cultures Time

Flask

Microcarrier bead

48 h pi 54 h pi 60 h pi

4.6±0.5a,b,c 11.7±1.4d,e 13.0±0.9

7.2±0.6a,b 7.9±1.0 11.0±1.8f

a Counts are expressed as particles of C. trachomatis serovar E per ml (×108). b Particle counts were determined from C. trachomatis infected McCoy cells harvested at 48, 54 and 60 h pi. For each time, 10 ll drops of a crude preparation of chlamydiae harvested from each culture were dried on a glass slide, stained with fluorescein-conjugated monoclonal antibodies directed against the C. trachomatis major outer membrane protein (Syva MicroTrak) and the number of fluorescent particles in 7 representative fields was counted by two methods and averaged. c P<0.01 compared to particle count from bead cultures at this time. d P<0.01 compared to particle count from flask cultures at 48 h pi. e P<0.05 compared to particle count from bead cultures at this time. f P<0.08 compared to particle count from bead cultures at 54 h pi.

information pathways necessary for cell growth and shape, internal cellular architecture, and biochemical signaling for activation/differentiation.7 Collagen coating of the microcarrier beads may serve to induce some signaling pathways and/or may simply provide for better adhesion of eukaryotic cells than the charged molecules added commercially to coat plastic surfaces for tissue culture use. McCoy cells were added to duplicate 25 cm2 plastic tissue culture flasks, one set of which had been previously coated with 0.75 ng of type II and type IV collagen. When the monolayers reached near-confluency, they were each inoculated with a crude EB stock preparation demonstrated to yield a 50% infection and incubated for 54 h. There was no difference in the infectivity titers of EB harvested from the two flask cultures at this time (data not shown), suggesting that merely coating an impermeable surface with collagen is not the contributing factor leading to increased infectivity of chlamydial progeny.

Discussion These studies, encompassing morphological observations with particle counts and infectivity titrations, indicate that development of a genital isolate of C. trachomatis serovar E in McCoy cells grown on microcarrier beads is much accelerated and results in the production of fewer numbers of EB but which are highly infectious sooner than in duplicate infected McCoy cultures grown in plastic tissue culture flasks. The accelerated feature of the developmental cycle was in the conversion of RB to EB. Following these observations, we have begun to harvest EB progeny from infected McCoy cells grown on microcarrier beads around 54 h pi. Although the particle counts from infected cultures at this time were lower, the infectivity titers were considerably higher than at later times. These results have since been confirmed by other members of our laboratory. A number of plausible explanations may account for the differences observed between chlamydiae-infected cells grown on microcarrier beads versus in plastic flasks. The simplest and most obvious explanation for accelerated development

38

P. B. Wyrick et al.

is nutrition. Growth of host cells on a permeable substratum permits nutrient uptake over the entire cell surface, whereas anchorage dependent cells grown on a non-permeable substratum become nutrient restricted. Competition for available energy, metabolites and cofactors may be greater for chlamydiae grown in cells cultured on flasks which would lead to a slower rate of metabolism and retarded chlamydial development. However, the nutrient rich fibroblasts cultured on microcarrier beads apparently provide for a more synchronous chlamydial developmental cycle, more efficient conversion of RB to EB and, in turn, a more infectious yield of EB progeny. A second possibility is that eukaryotic cells grown in bead culture are constantly moving and are likely more aerated than their counterparts in static flask culture. This idea agrees with long term observations in our laboratory that increasing the medium depth above an infected monolayer in flask culture results in decreasing inclusion size. As development progresses, the culture medium bathing chlamydiae-infected cells becomes more acid and a series of asynchronous late developmental cycle events occurs. In some infected cells harvested from bead cultures at 72 h pi, the inclusion membrane appears to have disintegrated, spilling numerous EB into the host cell cytoplasm. A number of potential mechanisms may account for this loss of inclusion membrane and release of chlamydiae. Inclusion membrane degradation, in C. psittaci infected cultures, has been attributed to hydrolytic enzymes released from lysosomes8 and/or activity of a chlamydia-specific proteinase released late in the developmental cycle on maturation of RB to EB.9 In C. trachomatis serovar D infected Hela cells, Todd and Caldwell10 proposed two escape mechanisms: (i) exocytosis of the inclusion with subsequent scar formation over the cavernous opening; and (ii) possibly fusion of the inclusion membrane with the host plasmalemma, opening the contents of the inclusion to the extracellular environment. Cooper et al.11 noted in morphological studies of C. trachomatis serovar E infected human fallopian tube organ cultures that by 72 h pi the inclusion membrane as well as the surface exposed epithelial cell plasma membrane had disintegrated, permitting release of chlamydial progeny at the apical surface. Time-lapse cinematography studies by Patton and Kuo12 revealed that prior to rupture of C. trachomatis serovar E inclusions in infected human amnion cells grown on coverslips, the entire intact inclusion rotated clockwise for approximately 8–10 h. The rupture event then appeared as a rapid, dynamic explosion with release of the majority of chlamydial particles within the first minute. Released EB are exposed to a rapidly deteriorating environment, consisting of acidic conditions and lysed host cell debris. Peeling and Brunham13 demonstrated with kinetic studies the heat lability of C. trachomatis serovar D; the infectivity of serovar D EB, suspended in a protein carrier solution to reduce thermal inactivation, dropped 70–80% in 60 min at 37°C. Whatever the mechanism of EB release, the later EB are harvested, the longer they may have been exposed to a deleterious environment which, in turn, reduces their infectivity. As an alternative, nutrientdepleted infected cells are easily lysed by gentle sonication; rapid harvest of EB just released from their protected inclusion environment seems to preserve their infectivity potential.

Materials and methods Chlamydia. Chlamydia trachomatis E/UW-5/CX, a human urogenital isolate provided to this laboratory by C. C. Kuo and S. P. Wang, University of Washington, Seattle, was used

C. trachomatis accelerated development

39

for the studies reported here. Stock cultures of C. trachomatis serovar E were maintained and titrated in mycoplasma-free McCoy cell monolayers (CRL 1696; ATCC, Rockville, MD, U.S.A.; Moorman et al.14).

Culture of chlamyidiae for morphological comparisons. Four 25 cm2 tissue culture flasks were seeded with McCoy cells and cultured in Hanks buffered Eagles Minimum Essential Medium (MEM; GIBCO, Grand Island, NY, U.S.A.), pH 7.3, supplemented with 10% heat inactivated fetal bovine serum (HyClone, Logan, UT, U.S.A.). The flasks were incubated at 37°C and monitored by phase contrast microscopy until the monolayers were nearly confluent. Cytodex 3 microcarrier beads (Sigma, St. Louis, MO, U.S.A.) were rehydrated in phosphate buffered saline (PBS), pH 7.2 and sterilized by autoclaving according to the manufacturers instructions. The concentration of microcarrier beads was adjusted to 1×104 beads per ml in 125 ml of MEM per spinner bottle, which is equivalent to 100 cm2 of culture surface area. McCoy cells, scraped from a subconfluent 75 cm2 plastic flask, were used to seed the microcarrier beads at a ratio of 50 cells per bead. The beads were kept in suspension by low speed stirring (ca. 15 rpm). McCoy cell monolayer development on the beads was assessed by phase contrast microscopy6 and the medium was changed as required. Subconfluent monolayers from both flask and microcarrier bead cultures were inoculated by adsorption5,6 with C. trachomatis at a titre calculated to infect 65% of the McCoy cells. Following a 2 h incubation at 35°C to permit attachment/entry of chlamydiae, infection proceeded in the presence of cycloheximide. At 7, 24, 48, and 72 h post inoculation (pi), infected McCoy cells from one 25 cm2 flask and 25 ml of microcarrier bead culture were harvested, washed once with PBS, and dislodged from the flask surface with Versene (1:5000 v/v; GIBCO, Grand Island, NY, U.S.A.) or were stripped from the microcarrier beads with 0.12% collagenase. Both cell suspensions were pelleted by low speed centrifugation and fixed with cacodylate-buffered 2% gluteraldehyde containing 0.5% paraformaldehyde (Polysciences, Rydal, PA, U.S.A.) for 1 h at 25°C. The samples were processed as previously described15 and embedded in Epon 812 resin. Ultrathin sections were cut with a diamond knife (Microstar; Huntsville, TX, U.S.A.) on a Riechert UltracutS microtome (Wein, Austria), poststained with 3% aqueous uranyl acetate followed by lead citrate, and examined in a Zeiss transmission electron microscope operating at 60 kv. Determination of infectivity titers and particle counts. McCoy cells were propagated in flask and microcarrier bead cultures as described above with slight modifications for each application. The precise number of McCoy cells forming a subconfluent monolayer in a 25 cm2 culture flask was determined from the average number of cells counted within a calibrated microscope ocular grid (0.36 mm2) at 16× magnification multiplied by the number of ocular grids contained within the surface area of the flask. The monolayers were infected with an inoculum of C. trachomatis calculated to infect 50% of the McCoy cells and incubated in MEM+CX at 35°C. Chlamydiae were harvested from the infected cells at 48, 54 and 60 h pi by rolling glass beads across the monolayer to cause mechanical lysis, followed by sonication in a water bath sonicator set at 1.5 amps for 5 min to ensure complete release of the chlamydiae into the supernatant. The lysate was centrifuged at 500×g to remove eukaryotic debris followed by 10000×g to pellet the chlamydial particles. The final pellet, representing the number of chlamydiae released from 1×106 McCoy cells, was resuspended in 1.0 ml storage buffer (0.2 m phosphate buffer, 0.2 m sucrose, 5 mm glutamine, 2SPG) and frozen at −80°C until infectivity titrations and particle counts could be determined. The progeny from 1×106 infected McCoy cells propagated on Cytodex 3 microcarrier beads was harvested accordingly at 48, 54, and 60 h pi and frozen in 1.0 ml of 2SPG for subsequent infectivity and particle count determinations. Confirmation that each sample taken from the spinner bottles (i) contained 1×106 McCoy cells was obtained by hemocytometer chamber counting and (ii) consisted of 50% infected McCoy cells was obtained by analysis of the number of fluorescent chlamydial inclusions in 1×106 McCoy cells stripped from the beads. The infectivity of C. trachomatis harvested from flask and microcarrier bead cultures at 48, 54, and 60 h pi was determined by (i) formation of inclusions in McCoy cell monolayers, as described in detail by Moorman et al.14 and (ii) the number of chlamydial particles present in the suspension as determined microscopically by the method of Knight et al.16

40

P. B. Wyrick et al.

and confirmed with computer aided morphometry. Briefly, 10 ll drops of the frozen harvested progeny preparations, as well as 10 ll drops of a purified preparation of C. trachomatis serovar E EB of known particle count, were dried on a 28.3 mm2 glass surface and the number of fluorescein-labeled particles observed at 40× within a calibrated microscopic field of view (1.8 mm2) were counted as well as recorded as a digitized image corrected for background, shading, and threshold (Metamorph, Universal Imaging Corp., West Chester PA, U.S.A.). Seven representative fields per sample were counted and recorded, averaged, and the concentration expressed as C. trachomatis particles per ml. All analyses were performed in duplicate or triplicate and the experiments were repeated at least twice on separate occasions.

Statistics. Data were analyzed by two tailed paired t-tests. A P value of less than 0.05 was considered significant. We thank Dr Johnny Carson, Director of the Department of Pediatrics Electron Microscopy Core Facility for the use of the microscope and darkroom facilities. This study was supported by Public Health Service grant AI13446 (to PBW). A portion of this study was presented in Abstract form at the Eighth International Symposium on Human Chlamydial Infections held in Chantilly, France, June 19–24, 1994.

References 1. Schachter J, Dawson CR. Human Chlamydial Infections. Littleton, Massachusetts: PSP Publishing Company, 1978. 2. Kunimoto D, Brunham RC. Human immune response and Chlamydia trachomatis infection. Rev Infect Dis 1985; 7: 665–73. 3. Bavoil PM. Determinants of chlamydial pathogenesis and immunity. In: Miller VL, Kaper JB, Portnoy DA, Isberg RR, eds. Molecular genetics of bacterial pathogenesis. Washington, DC: American Society for Microbiology Press, 1994: 295–308. 4. Gordon F, Quan A. Isolation of the trachoma agent in cell culture. Proc Soc Exp Biol Med 1965; 118: 354–9. 5. Tam JE, Knight ST, Davis CH, Wyrick PB. Eukaryotic cells grown on microcarrier beads offer a cost-efficient way to propagate Chlamydia trachomatis. Biotechniques 1992; 13: 374–8. 6. Schachter J, Wyrick PB. Culture and isolation of Chlamydia trachomatis. In: Clark VL, Bavoil PM, eds. Methods in enzymology: bacterial pathogenesis. B. Interaction of pathogenic bacteria with host cells, Vol. 236. San Diego: Academic Press, 1994: 377–90. 7. Juliano RL, Haskill S. Signal transduction from the extracellular matrix. J Cell Biol 1993; 120: 577–85. 8. Todd WJ, Storz J. Ultrastructural cytochemical evidence for the activation of lysosomes in the cytocidal effect of Chlamydia psittaci. Infect Immun 1975; 12: 638–46. 9. Stokes GV. Proteinase produced by Chlamydia psittaci in L cells. J Bacteriol 1974; 118: 616–20. 10. Todd WJ, Caldwell HD. The interaction of Chlamydia trachomatis with host cells: ultrastructural studies of the mechanism of release of a biovar II strain from HeLa 229 cells. J Infect Dis 1985; 151: 1037–44. 11. Cooper MD, Rapp J, Jeffery-Wiseman C, Barnes RC, Stephens DS. Chlamydia trachomatis infection of human fallopian tube organ cultures. J Gen Microbiol 1990; 136: 1109–15. 12. Patton DL, Kuo C. Cinematographic observations of growth of Chlamydia pneumoniae in primary cultures of monkey conjunctival and human amniotic epithelial cells. In: Bowie WR, Caldwell HD, Jones RP et al. eds. Chlamydial Infections: proceedings of the seventh international symposium on human chlamydial infections. Cambridge: Cambridge University Press, 1990: 20–3. 13. Peeling R, Brunham R. Neutralization of Chlamydia trachomatis: kinetics and stoichiometry. Infect Immun 1991; 59: 2624–30. 14. Moorman D, Sixbey JW, Wyrick PB. Interaction of Chlamydia trachomatis with human genital epithelium in culture. J Gen Microbiol 1986; 132: 1055–67. 15. Wyrick PB, Choong J, Davis CH et al. Entry of Chlamydia trachomatis into polarized human epithelial cells. Infect Immun 1989; 57: 2378–89. 16. Knight ST, Neece VR, Witt DJ. Rapid culture-independent techniques for quantitation of Chlamydia trachomatis elementary bodies. J Microbiol Methods 1989; 10: 255–63.