Bioinformatic and biochemical evidence for the identification of the type III secretion system needle protein of Chlamydia trachomatis.

Chlamydia spp. express a functional type III secretion system (T3SS) necessary for pathogenesis and intracellular growth. However, certain essential components of the secretion apparatus have diverged to such a degree as to preclude their identification by standard homology searches of primary protein sequences. One example is the needle subunit protein. Electron micrographs indicate that chlamydiae possess needle filaments, and yet database searches fail to identify a SctF homologue. We used a bioinformatics approach to identify a likely needle subunit protein for Chlamydia. Experimental evidence indicates that this protein, designated CdsF, has properties consistent with it being the major needle subunit protein. CdsF is concentrated in the outer membrane of elementary bodies and is surface exposed as a component of an extracellular needle-like projection. During infection CdsF is detectable by indirect immunofluorescence in the inclusion membrane with a punctuate distribution adjacent to membrane-associated reticulate bodies. Biochemical cross-linking studies revealed that, like other SctF proteins, CdsF is able to polymerize into multisubunit complexes. Furthermore, we identified two chaperones for CdsF, termed CdsE and CdsG, which have many characteristics of the Pseudomonas spp. needle chaperones PscE and PscG, respectively. In aggregate, our data are consistent with CdsF representing at least one component of the extended Chlamydia T3SS injectisome. The identification of this secretion system component is essential for studies involving ectopic reconstitution of the Chlamydia T3SS. Moreover, we anticipate that CdsF could serve as an efficacious target for anti-Chlamydia neutralizing antibodies.

Another putative heat-shock gene and aminoacyl-tRNA synthetase gene are located upstream from the grpE-like and dnaK-like genes in Chlamydia trachomatis.

The 4.1-kb sequence of genomic DNA located upstream from the Chlamydia trachomatis grpE-like and dnaK-like heat shock (HS) genes was determined. Another putative HS gene was located just 5' to grpE along with an inverted repeat (IR) sequence proposed to be involved in HS regulation. The overall organization of this locus in Chlamydia resembles that of Bacillus subtilis, rather than Escherichia coli. Two other open reading frames (ORFs) were found in the sequence, one of which has homology to aminoacyl-tRNA synthetases. The other ORF has no significant homology to reported genes. We also examined the codon usage bias for these newly identified chlamydial ORFs and for previously reported chlamydial genes, and found them to be different from E. coli.

Eukaryotic cells grown on microcarrier beads offer a cost-efficient way to propagate Chlamydia trachomatis.

The use of microcarrier cell culture as a method for the in vitro propagation of the obligate intracellular bacterial parasite, Chlamydia trachomatis, is described. The microcarrier beads proved to be a more cost-effective means to propagate C. trachomatis than traditional tissue culture flasks or roller bottles without sacrificing yields or infectivity. In addition, microcarrier cell culture was found to be a much simpler technique to study the intracellular development of these bacteria.

Intracellular survival by Chlamydia.

Chlamydiae are obligate intracellular bacterial pathogens whose entry into mucosal epithelial cells is required for intracellular survival and subsequent growth. After a seemingly stealthy entry, chlamydiae quickly modify their vacuole (i) for exit from the endosomal pathway to the exocytic pathway and (ii) to permit fusion with intercepted endoplasmic reticulum- and Golgi-derived vesicles carrying glycerophospholipids and sphingolipids for chlamydiae-containing vacuole membrane expansion. Chlamydiae possess novel hollow proteinaceous structures, termed projections, which they use to pierce the inclusion membrane, possibly to acquire from the epithelial cytoplasm nutrients they cannot synthesize; whether or not these truncated flagellar-like structures serve a dual exchange function for secretion of molecules to programme host cell signalling is unknown. Despite the accumulation of some 500-1000 progeny in the enormously enlarged inclusion, host cell function is surprisingly little disrupted, and progeny escape can be unobtrusive. This elegant adaptive pathogen strategy, which leads to silent, chronic human infection, is fascinating from a cellular microbiology perspective.

Cytoskeletal requirements in Chlamydia trachomatis infection of host cells.

Infection of genital epithelial cells by the closely related sexually transmitted pathogens Chlamydia trachomatis serovars E and L2 results in different clinical disease manifestations. Following entry into target host cells, individual vesicles containing chlamydiae fuse with one another to form one large inclusion. At the cellular level, the only obvious difference between these serovars is the time until inclusion maturation, which is 48 h for the invasive serovar L2 and 72 h for serovar E. To begin to define the intracellular events of these pathogens, the effect of cytoskeletal disruption on early endosome fusion and inclusion development in epithelial (HEC-1B) and fibroblast (McCoy) cells was analyzed by fluorescence microscopy. Disruption of microfilaments with cytochalasin D markedly reduced serovar E, but not serovar L2, infection of both cell lines. Conversely, microfilament as well as microtubule disruption, with colchicine or nocodazole, had no effect on serovar E inclusion development but resulted in the formation of multiple serovar L2 inclusions per cell during early and mid-development. Later in serovar L2 inclusion development (> 36 h postinfection), vesicles containing chlamydiae fused to form one large inclusion in the absence of an intact cytoskeleton. These results imply that (i) C. trachomatis serovar E may utilize a different pathway for uptake and development from serovar L2; (ii) these differences are consistent in both epithelial cells and fibroblasts; and (iii) the cytoskeleton plays a unique role in the infection of host cells by these two genital pathogens.

Ultrastructural study of mode of entry of Chlamydia psittaci into L-929 cells.

The entry of Chlamydia psittaci into L-929 cells was studied morphologically by transmission electron microscopy and quantitatively by a method that discriminates between attachment and uptake. Upon adsorption of 3H-labeled elementary bodies (EBs) to host cells at 4 degrees C, the EBs bound efficiently to the L-cell surface. Binding reached an equilibrium level of 55% in 3 h. Ultrastructural analysis revealed that EBs were bound preferentially to the tips and sides of microvilli at this temperature. The EBs were also observed in coated pits located at the bases of microvilli and along smooth surfaces of the host cell. No internalization was observed at 4 degrees C. When cells with prebound 3H-labeled EBs were warmed to 37 degrees C, the EBs rapidly became resistant to proteinase K removal (half time = 5 min), indicating ingested chlamydiae. At 37 degrees C, the EBs were internalized within tightly bound vesicles surrounded by an electron-dense coat of fibrillar material. EBs were also present in smooth-surfaced pits and vesicles of the host cell. Using alpha 2-macroglobulin coupled to colloidal gold (a known marker for receptor-mediated endocytosis), we observed that the entry of EBs into cells via coated pits was identical in appearance to the internalization of alpha 2-macroglobulin. Also, when the two ligands were mixed together, they could be seen within the same coated pits and were cointernalized within endocytic vesicles of the host cell. These results suggest that C. psittaci can enter nonprofessional phagocytic cells by a pathway which is similar to that of receptor-mediated endocytosis of many physiologically important macromolecules, bacterial toxins, and viruses.

Differences in the association of Chlamydia trachomatis serovar E and serovar L2 with epithelial cells in vitro may reflect biological differences in vivo.

Chlamydia trachomatis serovar E is one of the most common bacterial sexually transmitted pathogens. Since it is an obligate intracellular bacterium, efficient colonization of genital mucosal epithelial cells is crucial to the infectious process. Serovar E elementary bodies (EB) metabolically radiolabeled with 35S-Cys-Met and harvested from microcarrier bead cultures, which significantly improves the infectious EB-to-particle ratio, provided a more accurate picture of the parameters of attachment of EB to human endometrial epithelial cells (HEC-1B) than did less infectious 14C-EB harvested from flask cultures. Binding of serovar E EB was (i) equivalent at 35 and 4 degrees C, (ii) decreased by preexposure of EB to heat or the topical microbicide C31G, (iii) comparable among common eukaryotic cell lines (HeLa, McCoy), and (iv) significantly increased to the apical surfaces of polarized cells versus nonpolarized cells. In parallel experiments with C. trachomatis serovar L2, serovar E attachment was not affected by heparin or heparan sulfate whereas these glucosaminoglycans dramatically reduced serovar L2 attachment. These data were confirmed by competitive inhibition of serovar E binding and infectivity by excess unlabeled live and UV-inactivated serovar E EB but not by excess serovar L2 EB. The noninvasive serovar E strains in the lumen of the genital tract enter and exit the apical domains of target columnar epithelial cells to spread canalicularly in an ascending fashion from the lower to the upper genital tract. In contrast, the invasive serovar L2 strains are primarily submucosal pathogens and likely use the glucosaminoglycans concentrated in the extracellular matrix to colonize the basolateral domains of mucosal epithelia to perpetuate the infectious process.

Interaction of Chlamydia psittaci reticulate bodies with mouse peritoneal macrophages.

Noninfectious reticulate bodies of Chlamydia psittaci are readily phagocytized by thioglycolate-elicited mouse peritoneal macrophages in monolayer culture. The internalized reticulate bodies are rapidly destroyed as indicated by a 60 to 70% decrease in trichloroacetic acid-precipitable radioisotopic counts in the macrophage pellet by 10 h and a concomitant increase of the trichloroacetic acid-soluble radiolabeled chlamydial nucleic acid in the cytoplasm. This intracellular destruction of reticulate bodies in macrophages is independent of the multiplicity of infection. Reticulate bodies at a high multiplicity of infection, up to 1,000:1, are also incapable of inducing immediate cytotoxicity in macrophages as evidenced by the lack of early release of the host cell-soluble cytoplasmic enzyme lactic dehydrogenase. Thus, it appears that the virulence factors for (i) initiation or maintenance of intracellular survival via circumvention of phagolysosome formation and (ii) host cell damage are either missing or not expressed by the RB form of this bacterium.

Growth of Chlamydia psittaci in macrophages.

Survival and growth of L-cell-cultivated Chlamydia psittaci occurred in mouse macrophages in vitro. Two major factors governing the intracellular fate of chlamydiae in macrophages are: (i) the multiplicity of infection (MOI), i.e., the elementary body (EB)-to-macrophage ratio, and (ii) the state of the EB. At a low MOI (1:1) survival and growth of live, untreated chlamydiae were optimal. The chlamydiae were internalized in macrophages within 30 to 40 min. EB proceeded to differentiate into reticulate bodies, which underwent multiplication and further matured into infectious EB in the professional phagocytic cells. In contrast, at a high MOI (100:1), survival of untreated chlamydiae was greatly reduced as a result of immediate damage to the macrophages. eb that were pretreated with heat (56 degrees C for 10 to 30 min) or coated with homologous antibody were rapidly destroyed in macrophage phagolysosomes. Fusion of ferritin-labeled lysosomes with heat-treated or opsonized EB-laden phagosomes occurred in 2 to 4 h, resulting in transfer of the ferritin marker into phagolysosomes.

Modulation of the host immune response as a result of Chlamydia psittaci infection.

After intraperitoneal injection of mice with infectious, inactivated, or envelope preparations of the elementary body of Chlamydia psittaci, lymphocyte transformation of spleen cells to the mitogens concanavalin A, phytohemagglutinin, and lipopolysaccharide was significantly reduced 1 and 2 weeks postinjection. Lymphocyte response returned to the control values by 4 weeks. Similarly, transformation of cells by chlamydial antigen was not detected until 4 weeks postinjection. Injection of the noninfectious intracellular reticulate body, in contrast, had little effect on transformation of cells to concanavalin A. When control spleen cells were incubated with infectious or inactivated elementary bodies in vitro, response to all three mitogens was also reduced. The sooner the organisms were added after the addition of mitogen, the greater the reduction in transformation. Incubation with elementary body envelopes and reticulate bodies had no effect on lymphocyte transformation of the spleen cells to concanavalin A. The relationship between the observed ability to reduce the response in the in vitro assay of lymphocyte transformation and the actual in vivo establishment of infection is discussed.

Inhibition of phagolysosome fusion is localized to Chlamydia psittaci-laden vacuoles.

Intracellular survival of Chlamydia psittaci is in part dependent on the ability of the organism to thwart phagolysosome formation. Circumvention of phagolysosome fusion could be either localized to chlamydia-laden vacuoles or generalized to all phagosomes in the host cell. To determine which of these modes is in operation the ability of chlamydia elementary and reticulate bodies to protect Saccharomyces cerevisiae from degradation in macrophage phagolysosomes was examined via acridine orange and Giemsa staining. No statistically significant difference was evident between the amount of fusion observed in coinfected macrophages and those infected with yeast cells alone. This was ot dependent on some unique interaction between the chlamydia and the yeast cells since viable count studies to determine the protection of a second organism, Escherichia coli, also failed to show significantly different amounts of inactivation of the bacteria by macrophages in the presence of C. psittaci. Therefore, the inhibition of phagolysosome fusion is localized to chlamydia-laden phagosomes.

Filterability of streptococcal L-forms.

The filterability of the broth-grown stable L-form derived from Streptococcus faecium F24 was tested by filtration under the influence of varying amounts of applied pressure. A decrease in the pore size of the filter resulted in a corresponding decrease in viable count, but no major effect was noted due to the different pressures applied. Serial filtration of a deoxyribonuclease-treated L-form culture in the mid-logarithmic phase of growth resulted in recovery of viable L-forms from the 0.45-mum filtrate but not from the 0.22-mum filtrate. It is possible that disruption of the L-form bodies with release of small viable elements had occurred. Protoplasts, diluted in an osmotic stabilizer, were filtered similarly; L-forms could be grown from the filtrate passing through the filters of 0.45 mum or greater. Filtration of the parent streptococci gave rise to streptococcal colonies from the 1.2-mum filtrate only.

Growth of streptococcal protoplasts and L-colonies on membrane filters.

Membrane filters (Millipore Corp.; pore sizes 1.2 to 0.22 mum) were placed on the surface of L-phase growth medium solidified with agar. The filter and the surrounding medium were inoculated with either protoplasts or stable broth-grown L-phase variants obtained from Streptococcus faecium strain F24. The L-phase inoculum gave rise to viable L-colonies on the filters and on the medium, whereas protoplasts gave colony formation only on the medium. However, when the Millipore filters were covered by a layer of solid L-phase medium, 75 mum or greater in depth, before inoculation with protoplasts, colony formation resulted but with atypical morphology. In contrast, inoculation of protoplasts on Nuclepore and Sartorius membrane filters did give rise to L-colonies on the surface and underneath the filters after 2 days of incubation at 37 C. Submicroscopic, viable L-phase elements produced during colony formation were capable of passing through membrane filters with pore channels as small as 0.22 mum; these elements required transfer from underneath the filters to fresh agar medium in order to develop into L-phase colonies. Membrane filters were also placed on the surface of L-phase growth medium solidified with gelatin. Inoculation of the filters and surrounding medium with a lysozyme-prepared protoplast suspension gave rise to streptococci on the surface of the filters and on the medium. However, inoculation with the stable broth-grown L-phase variants gave rise to atypical colonies on the medium and only small patches of abortive growth on the filters.

Isolation and characterization of cell wall-defective variants of Bacillus subtilis and Bacillus licheniformis.

Quantitative mass conversion of intact bacterial cells of Bacillus subtilis and B. licheniformis to L-phase variants has been effected after lysozyme treatment. After subculture of the unstable L-phase variants for several months in the presence of methicillin, stable L-phase variants were obtained which grew in the absence of the antibiotic and were then unable to revert to the classical bacterial phase under conditions which gave rise to mass reversion of the protoplasts and unstable L-variants. These stable L-phase variants, which retained many of the physiological properties of the bacterium from which they were derived, were capable of growing exponentially and multiplying in liquid medium. Their morphology and apparent modes of reproduction were consistent with that described for other L-phase variants. The morphological events, as monitored by the electron microscope, of the reversion to the intact bacterial phase of an unstable L-phase variant of B. licheniformis are described.

Response of C3H/HeJ and C3H/HeN mice and their peritoneal macrophages to the toxicity of Chlamydia psittaci elementary bodies.

Intravenous injection of toxic doses of Chlamydia psittaci elementary bodies into endotoxin-responsive C3H/HeN mice or endotoxin-nonresponsive C3H/HeJ mice resulted in essentially identical time intervals to death. Inoculation of monolayer cultures of thioglycolate-stimulated peritoneal macrophages from the two strains of mice with 250 elementary bodies per macrophage resulted in immediate host cell toxicity, although the C3H/HeJ macrophages were somewhat less sensitive to elementary body toxicity than were the C3H/HeN macrophages.

Elementary body envelopes from Chlamydia psittaci can induce immediate cytotoxicity in resident mouse macrophages and L-cells.

Isolated, purified Chlamydia psittaci elementary body envelopes at a high multiplicity of infection (1,000:1) are capable of inducing immediate cytotoxicity in resident mouse macrophages and 929 L-cells.

Chlamydial infection of polarized HeLa cells induces PMN chemotaxis but the cytokine profile varies between disseminating and non-disseminating strains.

While genital infections caused by Chlamydia trachomatis are generally asymptomatic, the density and pattern of inflammation varies considerably. The purpose of this study was to try to dissect the signalling in chlamydiae-infected epithelial cells that triggers innate responses and regulates polymorphonuclear neutrophil (PMN) chemotaxis. Polarized endocervical epithelial HeLa cells, grown in commercial inserts, were inoculated either with the non-disseminating (luminal) serovar E or the disseminating serovar L2. At 12-48 h after infection, the chambers were used in a quantitative chemotaxis assay, and cytokine production by infected cells was examined using cDNA microarray technology and confirmed by enzyme-linked immunosorbent assay (ELISA). Infection of HeLa cells with C. trachomatis E or L2 induced a strong and similar PMN chemotactic response, but larger amounts of interleukin (IL)-8 and IL-11 were released after infection with serovar L2. IL-6 was also produced in modest amounts after infection with either strain, but no IL-1alpha or tumour necrosis factor (TNF)-alpha was detected in any of the culture supernatants tested. IL-11 did not appear to influence the PMN response to chlamydial infection, but secretion of large amounts of this anti-inflammatory cytokine, mainly active on macrophages, in the very early stages of the infection may allow C. trachomatis to escape some innate defences to establish infection.

Chlamydia trachomatis does not bind to alpha beta 1 integrins to colonize a human endometrial epithelial cell line cultured in vitro.

Chlamydia trachomatis is the leading cause of bacterially acquired sexually transmitted diseases in the United States and Europe. As an obligate intracellular pathogen, this bacterium must invade epithelial cells in order to survive and grow. Thus, multiple strategies probably exist for initial binding of chlamydiae to their target cells. Since a variety of bacteria have exploited integrins to colonize tissues, and a precedent existed for the involvement of extracellular matrix components in chlamydial attachment, this study first analyzed, by flow cytometry, integrins expressed by the human endometrial epithelial cell line HEC-1B. The genital cells were then exposed to monoclonal antibodies directed against those integrins and assayed for chlamydial attachment and inclusion development. Monoclonal antibodies bound to the alpha and/or beta 1 subunit of classic integrin receptors displayed by HEC-1B cells were not able to prevent colonization and infection of the epithelial cells by a genital isolate of C. trachomatis.