INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Chlamydia pneumoniae is a pathogen causing respiratory infections such as sinusitis, bronchitis, and pneumonia (12, 13). Chronic respiratory infections with this organism may also develop following acute illness in spite of appropriate antibiotic therapy (15). A potential role for C. pneumoniae in atherosclerosis has been suggested by studies documenting an association between patients with coronary artery disease and increased antibody titers to this organism (30, 38, 39). Furthermore, the presence of C. pneumoniae in atherosclerotic lesions has been shown by different techniques, such as electron microscopy, immunocytochemistry, PCR, and culture (reviewed in references 3, 40, and 43). Additional evidence supporting a relationship between C. pneumoniae and atherosclerosis has come from in vitro experiments showing the ability of this organism to replicate in cells of the vascular wall (i.e., endothelial cells, aortic smooth muscle cells, and macrophages) (9, 10). A causal role for this organism in atherogenesis has recently been strengthened by in vivo studies documenting the ability of C. pneumoniae to induce atheromatous-like lesions in the aortas of infected rabbits (7, 25). Still, the mechanisms by which C. pneumoniae might contribute to the pathogenesis of atherosclerosis remain unclear.

A major inflammatory event that takes place during atherosclerosis is the accumulation of monocytes from the circulation into the arterial intima (37). Monocyte chemotactic protein 1 (MCP-1), a member of the C-C chemokine family, has been proposed to play an important role in the early events of atherogenesis. Recent studies suggest that mice deficient in MCP-1 are less susceptible to experimental atherosclerosis (11, 14). MCP-1 is a 14-kDa glycoprotein secreted by many cells, including endothelial cells and vascular smooth muscle cells, which can be transcriptionally activated by different stimuli, such as tumor necrosis factor alpha, interleukin-1, and lipopolysaccharide (LPS) (2). A variety of signaling mechanisms are involved in the intracellular activation of MCP-1 gene expression by these stimuli, including activation of phospholipase C, generation of diacylglycerol, and activation of protein kinase C and tyrosine kinases (42). Subsequent events include generation of reactive oxygen intermediates and activation of transcription factor NF-B (34). There is evidence that NF-B is a major regulator of the transcriptional activation of MCP-1 in cytokine-activated human endothelial cells (27). In addition, other studies have shown that NF-B is activated during the early stages of atherosclerosis and may function as a point of convergence of the diverse risk factors associated with this disease (4).

Previous studies from this laboratory have shown that C. pneumoniae induces MCP-1 secretion from human endothelial cells and promotes the transendothelial migration of monocytes in vitro (31, 32). Interestingly, infection with Chlamydia trachomatis does not result in stimulation of MCP-1 (32). This divergence between species may reflect the presence of specific features in C. pneumoniae that activate different signaling pathways within human endothelial cells. To provide a better understanding of the underlying mechanisms involved in the C. pneumoniae-dependent activation of MCP-1 in infected endothelial cells, this study investigated the role of NF-B in MCP-1 gene expression.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Chlamydia isolates. C. pneumoniae A-03 (ATCC VR-1452) was isolated from an atheroma of a patient with coronary artery disease (35) and propagated in HEp-2 cell cultures (ATCC CCL-23) as previously described (31). C. trachomatis L2/434 was kindly provided by James B. Mahoney, McMaster University Regional Virology and Chlamydiology Laboratory, Hamilton, Ontario, Canada.

Endothelial cell cultures. Human umbilical vein endothelial cells (HUVEC) (ATCC 1730-CRL) were cultured in 75-cm² flasks and maintained in Ham's F12K medium supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin-amphotericin B (Fungizone) mix, 30 µg of endothelial cell growth supplement per ml, and 100 µg of heparin (Sigma, St. Louis, Mo.) per ml. Prior to infection, HUVEC were transferred into gelatin-coated 24-well plates at 2 × 10⁵ cells/well (MCP-1 experiments) or 6-well plates at 5 × 10⁵ cells/well (NF-B experiments) and incubated overnight at 37°C with 5% CO₂.

Infection protocol. HUVEC monolayers in 24- or 6-well plates were inoculated separately with C. pneumoniae A-03 or C. trachomatis L2 suspended in Ham's medium without antibiotics. Cells grown in 24-well plates received 2 × 10⁵ inclusion-forming units per well, while cells grown in 6-well plates were inoculated with 5 × 10⁵ inclusion-forming units per well, resulting in a multiplicity of infection (MOI) of 1:1 for each case. The inocula of both chlamydial species contained equivalent amounts of LPS as determined by the Limulus amebocyte lysate test for endotoxin (Sigma), and 2-keto-3-deoxyoctonate, measured by the method of Karkhanis et al. (20). Such controls were performed to rule out possible differences in the numbers of total chlamydial particles (i.e., infectious and noninfectious) between C. pneumoniae and C. trachomatis inocula. Following inoculation, infection of HUVEC was followed by centrifugation as described (31). Mock-infected controls were also included, which consisted of HUVEC treated with crude lysates of HEp-2 cells. For the MCP-1 experiments, uninfected, mock-infected, and infected cells were incubated for 1, 2, 4, 8, 12, and 24 h at 37°C in 5% CO₂ before total RNA isolation. For the NF-B experiments, cells were incubated for 30 min before nuclear protein extraction. The response of HUVEC to stimulation with 500 U of human recombinant tumor necrosis factor alpha (TNF-; Promega, Madison, Wis.) per ml was used as a positive control for MCP-1 gene expression and NF-B activation.

RT-PCR. For reverse transcription-PCR (RT-PCR), total RNA from HUVEC cultured in 24-well plates was isolated using RNeasy minikits following the manufacturer's procedures (Qiagen, Santa Clarita, Calif.). The amount of RNA was measured with a spectrophotometer and determined to be 16 to 20 µg. Reverse transcription was performed at 42°C for 20 min in 10 µl of reaction mixture containing 0.2 µg of total RNA, 5 mM MgCl₂, 1× reverse transcription buffer (10 mM Tris-HCl, 50 mM KCl, 0.1% Triton X-100), 1 mM each of the four deoxynucleoside triphosphates (dNTPs), 20 U of recombinant RNasin RNase inhibitor, 15 U of avian myeloblastosis virus reverse transcriptase, and 0.5 µg of oligo(dT)₁₅ (Promega). Reactions were stopped by heating at 99°C and cooling at 4°C for 5 min. Subsequently, 2 µl of cDNA products was amplified in 50 µl of 50 mM KCl-10 mM Tris-HCl-2.5 mM MgCl₂-0.2 mM each dNTP-1 U of AmpliTaq DNA polymerase (Perkin-Elmer, Foster City, Calif.)-0.45 µM human MCP-1 primers. As internal controls for RT-PCR, 2 µl of cDNA products was also amplified in separate reactions containing -actin primers. The concentrations of PCR reagents and primers for -actin were the same as for MCP-1. Primers were purchased from R&D Systems, Minneapolis, Minn. The sequences of the forward and reverse primers for MCP-1 were 5'-CAGCCAGATGCAATCAATGC-3' and 5'-GTGGTCCATGGAATCCTGAA-3', respectively. The sequences of the forward and reverse primers for -actin were 5'-CTACAATGAGCTGCGTGTGG-3' and 5'-AAGGAAGGCTGGAAGAGTGC-3', respectively. The parameters for PCR were as follows: 94°C for 10 min and 25 cycles of 30 s of denaturation at 94°C, 22 s of annealing at 70°C, and 30 s of extension at 72°C. RT-PCR product sizes for MCP-1 and -actin were 198 and 528 bp, respectively. To control for the presence of PCR inhibitors, cDNA samples were also coamplified with MCP-1 or -actin PCR-positive control templates, which generated products of 320 and 528 bp, respectively (R&D Systems). The concentration of template added in each PCR reaction was 5.0 pg/µl.

Densitometric analysis of MCP-1 and -actin RT-PCR products. Densitometry of MCP-1 and -actin RT-PCR products was performed with the AlphaImager 2000 software. Densitometric values of -actin bands were used to standardize the results. Increases in MCP-1 mRNA levels in experimental groups were expressed as the ratio of MCP-1 to -actin RT-PCR products. The numerical data were subjected to analysis of variance followed by the Tukey-Kramer multiple-comparison test. A P value of <0.05 was used to determine statistical significance for all analyses.

Nuclear protein extraction. Nuclear protein extracts were obtained by a modification of the procedures described by Dignam et al. (6). Briefly, HUVEC cultured in six-well plates (total of 3 × 10⁶ cells for each experimental condition) were washed with ice-cold phosphate-buffered saline (PBS), removed by gentle scraping, and centrifuged at 1,200 rpm for 10 min at 4°C. Cell pellets were resuspended in 400 µl of ice-cold buffer A (10 mM HEPES [pH 7.9], 10 mM KCl, 0.1 mM EDTA, 1 mM dithiothreitol, and 0.5 mM phenylmethylsulfonyl fluoride) and allowed to swell on ice for 15 min. Cells were then lysed by addition of 25 µl of 10% Nonidet P-40 with vigorous vortexing for 10 s. Nuclear pellets were collected by microcentrifugation at 14,000 rpm for 30 s at 4°C. Supernatants were removed, and nuclear pellets were resuspended in 50 µl of ice-cold buffer C (20 mM HEPES, 0.4 M NaCl, 1 mM EDTA, 1 mM dithiothreitol, and 0.5 mM phenylmethylsulfonyl fluoride) for 20 min on ice with periodic mixing. Nuclear extracts were centrifuged at 14,500 rpm for 5 min to pellet insoluble material. Supernatants containing nuclear proteins were collected and frozen at 80°C. Protein concentrations were 1 to 2 pg/µl as determined by the bicinchoninic acid assay (Pierce, Rockford, Ill.).

EMSA. For the electrophoretic mobility shift assay (EMSA), a double-stranded oligonucleotide containing the NF-B consensus sequence (underlined) (5'-GTGAGGGGACTTTCCCAGGC-3'; Promega Corp.) was end labeled with [-³²P]ATP (3,000 Ci/mmol at 10 mCi/ml; Amersham Corp., Arlington Heights, Ill.) as described previously (26). Binding reactions were performed at room temperature for 30 min in 15 µl of a mixture containing 5 µg of nuclear protein and 35 fmol (~50,000 cpm) of oligonucleotide in binding buffer [4% glycerol, 1 mM MgCl₂, 0.5 mM EDTA (pH 8.0), 0.5 mM dithiothreitol, 50 mM NaCl, 10 mM Tris (pH 7.6), and 50 µg of poly(dI-dC) (Pharmacia, Piscataway, N.J.) per ml] (26). Antibodies to NF-B p50 (NF-B1), p52 (NF-B2), p65 (RelA), p68 (RelB), and p75 (c-Rel) (Santa Cruz Biotechnology, Santa Cruz, Calif.) were used for supershift assays. Reaction products were separated in 4% nondenaturing polyacrylamide gels and analyzed by autoradiography.

Immunofluorescent staining for NF-B. HUVEC grown on gelatin-coated glass coverslips were infected by centrifugation with C. pneumoniae A-03 or C. trachomatis L2 at an MOI of 1:1. After 30 min of infection, cells were fixed for 5 min with methanol at 20°C. To suppress nonspecific binding of immunoglobulin G (IgG), cells were initially incubated for 20 min with 1% bovine serum albumin (BSA) at room temperature. Incubation with rabbit polyclonal IgG antibody against the p65 component of NF-B (2 µg/ml in PBS with 1% BSA) was carried out for 1 h at room temperature. Monolayers were then washed three times with PBS for 5 min each. This was followed by incubation with fluorescein isothiocyanate (FITC)-conjugated anti-rabbit IgG (5 µg/ml in PBS with 1% BSA) for 45 min. Antibodies were purchased from Santa Cruz Biotechnology. Coverslips with stained cells were mounted in 80% glycerol in PBS and examined by confocal microscopy under an ×1,000 objective.

Treatment of C. pneumoniae-infected HUVEC with the NF-B inhibitor parthenolide. The effects of NF-B inhibition on MCP-1 mRNA production were examined by treatment of infected HUVEC with parthenolide, a compound that has been shown to suppress NF-B activation (16). Prior to infection, HUVEC grown on 24-well plates were preincubated for 1 h with 50 µM of parthenolide (Biomol, Plymouth Meeting, Pa.). Cells were then infected with C. pneumoniae A-03 in medium containing 50 µM parthenolide for 1 h, and expression of MCP-1 mRNA was examined by RT-PCR as described above. In separate experiments, HUVEC grown on six-well plates were preincubated with equivalent concentrations of parthenolide for 1 h, infected with C. pneumoniae A-03, and incubated for an additional 30 min in the presence of the inhibitor prior to examination of NF-B by EMSA.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Expression of MCP-1 mRNA in HUVEC infected with C. pneumoniae or C. trachomatis. The kinetics of MCP-1 mRNA expression were examined by RT-PCR in HUVEC infected with C. pneumoniae or C. trachomatis. As shown in Fig. 1A, expression of MCP-1 mRNA in C. pneumoniae-infected cells was increased as early as 1 h postinfection compared to uninfected and mock-infected controls (198-bp RT-PCR product). MCP-1 gene expression remained elevated between 2 and 8 h of infection and peaked at 12 h postinfection. In contrast, infection of HUVEC with C. trachomatis L2 did not induce MCP-1 mRNA expression at any time point. Induction of MCP-1 gene expression by C. pneumoniae A-03 was comparable to that by 500 U of TNF- per ml. Expression of -actin mRNA, as an internal control for RT-PCR analyses, did not differ significantly among experimental groups throughout the 24-h period of incubation (data not shown).

View larger version (48K): [in this window] [in a new window]   FIG. 1.   Expression of MCP-1 mRNA in HUVEC infected with C. pneumoniae or C. trachomatis. (A) RNA was isolated at 0, 1, 2, 4, 8, 12, and 24 h of incubation, and levels of MCP-1 mRNA were determined by RT-PCR as described in Materials and Methods. (Top panel) HUVEC were incubated with medium alone (uninfected) or crude lysates of HEp-2 cells (mock infected). (Middle panel) HUVEC were infected with C. pneumoniae A-03 at an MOI of 1:1 or treated with 500 U of TNF- per ml as a positive control. (Bottom panel) Infection with C. trachomatis L2. The inoculum was equivalent to that of C. pneumoniae A-03. M, molecular size markers. (B) Levels of MCP-1 mRNA following infection of HUVEC with C. pneumoniae (Cp) or C. trachomatis (Ctr) were measured by plotting the densitometric MCP-1/-actin RT-PCR product ratios. Data points represent the means ± standard errors of the mean of five separate experiments. *, P < 0.01.

Activation of NF-B in HUVEC infected with C. pneumoniae or C. trachomatis. To determine if differences in MCP-1 mRNA synthesis between C. pneumoniae and C. trachomatis were related to activation of the transcription factor NF-B, infected HUVEC were examined by EMSA and immunofluorescence staining following 30 min of incubation. As shown in Fig. 2, infection of HUVEC with C. pneumoniae A-03 resulted in an increase in the nuclear translocation of NF-B compared to mock-infected cells. In contrast, C. trachomatis L2 did not induce NF-B activation above that in mock-infected cells. TNF--induced NF-B activation served as a positive control.

View larger version (77K): [in this window] [in a new window]   FIG. 2.   Translocation of NF-B determined by EMSA in HUVEC infected with C. pneumoniae or C. trachomatis. Mock-infected HUVEC were incubated with crude lysates of HEp-2 cells. Infection with C. pneumoniae A-03 or C. trachomatis L2 was performed at an MOI of 1:1. Treatment of HUVEC with 500 U of TNF- per ml served as a positive control. Nuclear extracts were prepared after 30 min of incubation. EMSAs were performed with a 32P-labeled oligonucleotide containing the NF-B consensus sequence (see Materials and Methods). C. pneumoniae A-03 and TNF- caused an increase in NF-B DNA-binding activities of nuclear protein complexes compared to the levels in mock-infected cells and C. trachomatis L2-infected cells.

View larger version (74K): [in this window] [in a new window]   FIG. 3.   Supershift analysis of NF-B-binding complexes. DNA-binding reactions with nuclear extracts from HUVEC infected with C. pneumoniae A-03 were incubated with 32P-labeled NF-B oligonucleotide in the presence of antibodies (Ab) to NF-B proteins p50 (NF-B1), p52 (NF-B2), p65 (RelA), p68 (RelB), and p75 (c-Rel). The solid arrowhead indicates the NF-B band. Supershifts of p50 and p65 are indicated by the open and solid arrows, respectively.

View larger version (15K): [in this window] [in a new window]   FIG. 4.   Immunofluorescent staining of NF-B p65 in HUVEC infected with C. pneumoniae or C. trachomatis. HUVEC grown on gelatin-coated glass coverslips were infected with C. pneumoniae A-03 or C. trachomatis L2 at an MOI of 1:1. After 30 min of infection, cells were processed for immunofluorescent staining as described in Materials and Methods with rabbit polyclonal antibody against the p65 component of NF-B. Stained cells were immediately examined by confocal microscopy (×1,000). Absence of nuclear NF-B p65 is observed in uninfected, mock-infected, and C. trachomatis L2-infected cells (panels A, B, and D, respectively). Nuclear localization of NF-B is observed in HUVEC infected with C. pneumoniae A-03 (panel C) and to a greater extent in cells treated with 500 U of TNF- per ml for 30 min as a positive control (panel E). Panel F represents a control to rule out possible nonspecific binding of the FITC-conjugated anti-rabbit IgG in C. pneumoniae A-03-infected HUVEC.

Inhibition of NF-B prevents C. pneumoniae-induced MCP-1 gene expression in HUVEC. The relationship of NF-B activation to induction of MCP-1 mRNA expression in C. pneumoniae-infected HUVEC was investigated further with parthenolide, a specific inhibitor of NF-B that prevents the inducible degradation of IB and IB proteins (16). As shown in Fig. 5, C. pneumoniae-induced activation of NF-B at 30 min postinfection was markedly decreased by incubation of cells with 50 µM parthenolide (top panel). Treatment with parthenolide resulted in a similar reduction in TNF--induced NF-B activation. The effects of parthenolide on MCP-1 gene expression in HUVEC infected with C. pneumoniae are also depicted in Fig. 5 (bottom panel). Expression of MCP-1 mRNA induced by C. pneumoniae A-03 at 1 h postinfection was decreased to mock-infected levels with 50 µM parthenolide. These data suggest that NF-B activation is required for induction of MCP-1 gene expression by C. pneumoniae in HUVEC.

View larger version (29K): [in this window] [in a new window]   FIG. 5.   Effects of parthenolide on NF-B activation and MCP-1 gene expression in HUVEC infected with C. pneumoniae. (Top panel) Prior to infection, HUVEC grown on six-well plates were preincubated with 50 µM parthenolide for 1 h. Cells were then infected with C. pneumoniae A-03 at an MOI of 1:1 or treated with 500 U of TNF- per ml in medium containing 0 or 50 µM parthenolide for 30 min before nuclear protein extraction. NF-B activation was examined by EMSA. (Bottom panel) HUVEC grown on 24-well plates were preincubated with 50 µM parthenolide for 1 h. Cells were then infected with C. pneumoniae A-03 at an MOI of 1:1 or treated with 500 U of TNF- per ml in medium containing 0 or 50 µM parthenolide for 1 h before total RNA isolation. Expression of MCP-1 mRNA was examined by RT-PCR.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The present study demonstrates that infection of human endothelial cells with C. pneumoniae results in rapid induction of MCP-1 mRNA expression, an event that appears to be dependent on early activation of NF-B. Stimulation of MCP-1 gene expression occurred in a time-dependent fashion, with increased levels of MCP-1 mRNA observed as early as 1 h postinfection. C. pneumoniae stimulated translocation of NF-B in infected HUVEC at 30 min of incubation, as determined by EMSA as well as immunofluorescent staining of NF-B p65. Supershift analyses revealed that the components of the NF-B complex in C. pneumoniae-infected HUVEC were p50-p65 heterodimers. These results confirm those from a recent study in which p50-p65 heterodimers were identified as the active NF-B components present in smooth muscle cells and endothelial cells infected with C. pneumoniae (5). In our study, a functional role for NF-B in induction of MCP-1 gene expression by C. pneumoniae was demonstrated by experiments using parthenolide, a compound that has been shown to prevent the inducible degradation of IB and IB proteins (16). Treatment with parthenolide suppressed C. pneumoniae-induced NF-B activation and completely abrogated MCP-1 mRNA expression in infected HUVEC.

As opposed to C. pneumoniae A-03, C. trachomatis L2 did not stimulate MCP-1 mRNA synthesis in infected HUVEC, which was consistent with an inability of this species to activate NF-B. These results confirm our earlier findings of a failure of C. trachomatis L2 to stimulate MCP-1 protein production despite significant replication of this organism in HUVEC (32). At the MOIs used in the present study, we have found that average C. trachomatis L2 growth titers in infected HUVEC are 100-fold higher than those of C. pneumoniae, indicating that the differences in endothelial cell activation between the two species are not related to poor replication by C. trachomatis L2. In addition, possible discrepancies in the total numbers of chlamydial particles (i.e., infectious and noninfectious) between C. pneumoniae and C. trachomatis inocula were ruled out by measurements of similar amounts of LPS, as determined by levels of endotoxin and 2-keto-3-deoxyoctonate.

The basis for the differences observed between chlamydial species in the NF-B-mediated activation of MCP-1 in HUVEC reported herein remains unclear. A possible scenario is that C. pneumoniae may possess specific features required for attachment and uptake in endothelial cells that trigger distinct signal transduction components linked to the activation of NF-B. A recent study showed that C. pneumoniae but not C. trachomatis significantly exacerbated atherosclerosis in low-density lipoprotein receptor-deficient mice fed a high-cholesterol diet, suggesting that C. pneumoniae may possess a unique atherogenic property (18). Since NF-B is considered a major regulator of endothelial dysfunction during atherogenesis (4), it is possible that the C. pneumoniae-dependent activation of NF-B plays an important role in the development of atherosclerosis. It should be mentioned, however, that the differences between these two Chlamydia species may also be cell specific, since previous work has shown that C. trachomatis L2 activates NF-B in epithelial cells (36).

Recent experiments performed in vitro with C. pneumoniae have been compatible with the pathological characteristics of atherosclerosis in humans (1, 8-10, 17, 19, 29). In addition, cell culture studies have provided a better understanding of the mechanisms involved in the generation of atherosclerotic lesions described in animal models. Initial reports documenting the susceptibility of endothelial cells, aortic smooth muscle cells, and macrophages to productive infection with C. pneumoniae suggested the ability of this organism to localize and survive within the arterial wall (9, 10). Transfer of C. pneumoniae infection from mononuclear phagocytes to endothelial cells by cell-to-cell spread suggests that dissemination from the respiratory tract to the arteries could occur via circulating monocytes (8). Infection of monocyte-derived macrophages with this organism in the presence of low-density lipoprotein leads to foam cell formation and accumulation of cholesteryl esters (19). In addition, macrophages exposed to C. pneumoniae increase the production of proinflammatory cytokines such as TNF- and interleukin-1 (17), which may contribute to endothelial cell activation and tissue damage. Recently, data have also shown that C. pneumoniae infection induces a procoagulant phenotype in smooth muscle cells and endothelial cells by increasing the expression of tissue factor and plasminogen activator inhibitor 1 (5). Lastly, the development of a persistent state in cell culture with noninfectious but viable C. pneumoniae (1, 29) correlates with the ability of this organism to cause a chronic infection within arterial tissues, an event which may be critical for the maintenance of a long-term inflammatory response.

Elucidation of the bacterial and cellular components involved in the activation of MCP-1 by C. pneumoniae in endothelial cells requires further investigations. Previous data demonstrated a role for a heat-labile component of C. pneumoniae, since UV-inactivated but not heat-inactivated organisms retained the ability to stimulate MCP-1 secretion from HUVEC (31, 32). These results also suggested that chlamydial LPS may not be required for this response. Possible candidates involved in attachment of C. pneumoniae to endothelial cells may include surface-exposed outer membrane proteins such as Omp4 and Omp5 (21). A recent report has shown that tyrosine phosphorylation of endothelial cell proteins occurs within 5 min of C. pneumoniae infection, suggesting that bacterial attachment may be sufficient to trigger an endothelial inflammatory response (24). In addition, C. pneumoniae infection caused rapid phosphorylation of mitogen-activated protein kinase, which could be an important upstream mediator of the NF-B signaling cascade (24, 28). Maintenance of a chronic inflammatory response may involve ongoing activation of endothelial cells by C. pneumoniae antigens. Of interest in the immunopathogenesis of chlamydial infections is heat shock protein 60 (Hsp-60), a delayed-type hypersensitivity antigen implicated in the chronic inflammatory response of trachoma (33). Recent studies have shown that chlamydial Hsp-60 localizes in human atheromas and triggers activation of NF-B in human endothelial cells, smooth muscle cells, and macrophages in vitro, supporting a contribution of this antigen to the pathogenesis of atherosclerosis (22, 23).

In summary, despite the numerous reports implicating C. pneumoniae in atherosclerosis, a causal role remains to be established. Our data suggest that activation of NF-B and the resulting expression of MCP-1 by C. pneumoniae in human endothelial cells may participate in monocyte-macrophage recruitment during the early stages of atherogenesis. These events may also contribute to the exacerbation of a previously established atherosclerotic process. The differences observed between Chlamydia species, documented in this report, suggest that C. pneumoniae may possess specific features that function as triggering mechanisms in the activation of inflammatory mediators involved in atherosclerosis. Analysis of the signaling pathways involved in NF-B activation of MCP-1 by C. pneumoniae may help identify potentially unique intracellular mechanisms utilized by this organism in human endothelial cells. The identification of NF-B as a critical transcription factor in C. pneumoniae-induced MCP-1 gene expression may provide insights into specific therapeutic strategies for the treatment of diseases associated with this organism, including atherosclerosis.