INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Heart diseases are the most common cause of death in developed countries. Chlamydiae, especially Chlamydia pneumoniae, are highly associated with cardiovascular diseases including atherosclerosis (8, 11, 17, 21, 29-32). C. trachomatis, although rare, is a cause of bacterial endocardial and myocardial diseases (6, 7, 12, 24, 27, 28, 33). Thus far, however, few experimental studies have been carried out to directly investigate the relationship between C. trachomatis and cardiovascular pathology and to examine the protective factors involving host defense against cardiac chlamydial infection. The major reasons underlying the lack of experimental studies are probably the lack of proper animal models for C. trachomatis-induced cardiovascular diseases and the relatively rare observation of C. trachomatis-related cardiovascular diseases documented in epidemiological and clinical studies.

Very recently, however, Bachmaier et al. reported that the injection of chlamydial peptides homologous to murine heart muscle-specific -myosin heavy chain from various chlamydial species induced autoimmune heart disease in mice (2). The authors argued that chlamydia-mediated cardiovascular diseases were induced by antigenic mimicry of heart muscle-specific proteins by chlamydial antigens. Interestingly, the data from that report demonstrated that the peptides from both C. trachomatis and C. pneumoniae exhibited homology to the immunogenic mouse heart muscle-specific -myosin motif and that the peptide from C. trachomatis induced prevalent and severe cardiac inflammation in mice. The study suggests that C. trachomatis infection could be as efficient in inducing cardiovascular pathological changes as C. pneumoniae, at least in certain conditions. A key factor in determining the role of chlamydial infection in cardiovascular diseases might be whether the organism can reach the cardiovascular system during natural infection. We therefore hypothesized that the relative rarity of documentation of C. trachomatis-associated cardiovascular diseases might be at least partially because human C. trachomatis strains normally cause ocular and genital tract infections and thus may not easily disseminate to remote organs such as the heart, as happens with C. pneumoniae, which normally causes pulmonary infection. Indeed, previous studies have clearly shown that pulmonary C. pneumoniae infections normally disseminate to multiple organs (1, 9, 10, 22, 40).

A mouse model of C. trachomatis pneumonia has been established in our and others' laboratories by using the mouse pneumonitis strain of C. trachomatis (MoPn) (35, 37, 38). Unlike other C. trachomatis strains, which normally cause ocular and genital tract infections, MoPn naturally causes murine respiratory tract infection even with a small dose of intranasal inoculator (<100 inclusion-forming units [IFU]). Therefore, in terms of the location of natural infection (the lungs), MoPn is more like C. pneumoniae than it is like other C. trachomatis strains. However, it is not clear whether the respiratory tract MoPn infection can cause systemic dissemination, in particular to the heart, in immunologically competent mice, especially following a short period of infection.

Gamma interferon (IFN-) is a critical cytokine in host defense against pulmonary and genital chlamydial infection in both human and animal studies (14, 18, 23, 26, 34, 36, 39). Epidemiological studies showed that patients with severe sequelae of ocular C. trachomatis infection exhibit significantly lower levels of IFN- production by peripheral blood mononuclear cells (3, 15). In vivo administration of exogenous IFN- promoted the clearance of C. trachomatis infection, while neutralization of endogenous IFN- with anti-IFN- monoclonal antibody (MAb) exacerbated C. trachomatis infection in mice (41, 42). More recently, it was reported that knockout mice deficient in IFN- or IFN- receptor (IFN- KO mice) exhibit multiorgan infection following genital infection with C. trachomatis, although the heart was not examined in theses reported studies (4, 16).

The objective of the present study was to explore whether respiratory tract MoPn can induce systemic (including cardiovascular system) chlamydial infection and, more importantly, whether the infection can cause pathological changes in the cardiovascular system. The results showed that immunocompetent wild-type C57BL/6 mice intranasally infected with MoPn exhibited multiorgan dissemination of chlamydial antigen and a mild but significant inflammatory reaction. In particular, chlamydial antigens were detected in the myocardium (myocytes) and the endothelium of cardiac blood vessels and significant fibrotic reactions were found in myocardial and perivascular areas. Furthermore, experiments with IFN- KO mice showed significant chlamydial growth, massive inflammatory infiltrates in the heart, and large vegetations in the aortic valves following respiratory tract MoPn infection. The data suggest that respiratory tract C. trachomatis infection can cause cardiovascular pathology and that IFN- plays a crucial role in host defense against cardiovascular chlamydial diseases. The data also suggest that murine respiratory tract MoPn infection may be a useful model for the study of cardiovascular diseases induced by chlamydial infection.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Organism. C. trachomatis MoPn was grown in HeLa 229 cells and purified by discontinuous density gradient centrifugation with Renografin (Squibb, Princeton, N.J.) as previously described (37, 38). The infectivity of the stock chlamydial elementary bodies was determined by infection of HeLa 229 cells and enumeration of inclusions that were stained by an anti-chlamydial lipopolysaccharide (LPS) MAb as previously described (37, 38).

Infection of mice. Female C57BL/6 mice were purchased from Charles River Canada (St. Constant, Quebec, Canada). Female homozygote IFN- KO mice (8 to 12 weeks old) with C57BL/6 background (C57BL/6-Ifg<tm1Ts>) were purchased from Jackson Laboratory (Bar Harbor, Maine). All mice were maintained and used in strict accordance with the guidelines issued by the Canadian Council on Animal Care. Mice were kept in a specific-pathogen-free facility for animals at the University of Manitoba with filtered air flow and autoclaved cage, food, and water. The mice were intranasally inoculated with C. trachomatis MoPn in 40 µl of sucrose-phosphate-glutamic acid (SPG) as previously described (37). They were sacrificed on various days following infection to examine chlamydial growth and pathological changes in various organs. To analyze the chlamydial infectivity in various organs, homogenates of these organs were subjected to quantitative culture of chlamydial inclusions by using HeLa 229 cells as previously described (37, 38).

Histopathological and immunohistochemical analysis. Different organs of mice were removed and fixed in 10% buffered formalin and embedded in paraffin. Sections (5 µm) were cut and stained with hematoxylin and eosin (H & E). The structural changes and cellular infiltration in sections were determined by light microscopy. Perivascular fibrotic change and fiber deposition in vegetations were determined by Masson trichrome staining (19). Briefly, tissue sections (5 µm) were deparaffinized and successively stained with Weigert iron hematoxylin, mixed solution of Ponceau, acid facsin and orange G, and fast green. Identified chlamydial inclusions (antigens) in different organs were subjected to immunohistochemical staining with an anti-chlamydial LPS MAb. Briefly, tissue sections (5 µm) were deparaffinized and washed with phosphate-buffered saline (PBS) (5 min each). After blocking with 0.3% H₂O₂-100% methanol for 30 min at room temperature, the sections were washed with PBS and blocked with 2% goat sera for 30 min. Mouse antichlamydial LPS MAb (primary Ab) was added to the tissue sections and incubated overnight at 4°C. After extensive washing with PBS, goat anti-mouse immunoglobulin G (IgG) (secondary Ab) conjugated with horseradish peroxidase was added for 40 min at room temperature. Isotype-matched (IgG) naive Ab was used as control in the staining. Finally, the sections were washed, and substrate (4-chloro-1-naphthol) (Sigma, St. Louis, Mo.) was added. Stained inclusions were visualized by light microscopy. For staining of infiltrating CD4 and CD8 T cells in the tissue sections, an Envision system kit (DAKO Corp., Carpinteria, Calif.) was used. The primary Ab used in staining was either anti-CD4 MAb YTS-191.1 or anti-CD8 MAb YTS 169 (hybridomas kindly provided by H. Waldmann, Cambridge University). The secondary Ab was rabbit anti-rat antibody conjugated with horseradish peroxidase. Staining was performed as specified by the manufacturer and was completed by incubation with 3-amino-9-ethylcarbazole (AEC) substrate chromogen, which results in a precipitate at the antigen sites. Normal rat IgG was used as the negative-control primary Ab and showed no staining in the tested tissues.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Intranasal MoPn infection induces systemic chlamydial infection and inflammation. Our previous studies showed that intranasal infection of C57BL/6 mice with MoPn can readily cause pneumonia, demonstrated by pulmonary inflammatory cellular infiltration and recovery of chlamydial organisms from the lungs (37, 38). We recently examined whether viable organisms can be recovered in nonlung organs of C57BL/6 mice following intranasal infection with 100 to 10,000 IFU of MoPn. The results showed that except for the primarily infected organs (the lungs), no viable (infectious) organisms were recovered from the organs including the liver, kidneys, and heart at 3, 7, 10, and 20 days following intranasal MoPn infection (33a). However, the lack of viable (infectious) organisms in nonlung organs cannot exclude the possibility that small amounts of MoPn are temporarily disseminated to these organs and/or that the disseminated organisms in these organs are inactivated by the host defense mechanism. To test this possibility, we used an immunohistochemical method to detect whether chlamydial antigens exist in the nonlung organs of the mice intranasally infected with MoPn (2,000 IFU). As shown in Fig. 1, in addition to their abundance in the primarily infected organ (the lung), chlamydial antigens were detected in all the organs examined, including the liver, kidneys and heart, of 100% of the infected mice (18 of 18) from four independent experiments in which mouse tissues were examined on day 15 or 20 following intranasal infection with MoPn. Staining with a MAb specific for chlamydial LPS demonstrated irregular shapes of chlamydial antigen accumulation with few areas of typical inclusion-like morphology. Tissues from naive mice without MoPn infection were completely negative for anti-chlamydial LPS MAb staining (data not shown). Similarly, staining with isotype-matched control Ab (normal rat IgG) of the tissues from MoPn-infected mice also showed negative results (data not shown). The existence of chlamydial antigens (inclusions) in multiple organs of the mice intranasally infected with MoPn indicates that chlamydial antigens (organisms) were disseminated during respiratory tract infection. The great irregularity in the shapes of chlamydial antigen accumulation (inclusions) in these nonlung organs suggests the inactivation (inhibition) of infectious organisms by the host defense mechanism, which may partially explain the observed negativity of chlamydial infectivity in these organs. It is also possible that the survival of organisms in these organs was so low that the number of organisms was below the sensitivity of the assay used for testing chlamydial infectivity in the tissue homogenates (the assay can only detect chlamydial infectivity higher than 800 IFU/organ because of the small volume of tissue homogenates which can be placed in each culture well). In association with the existence of chlamydial antigens, significant infiltration of inflammatory cells was also determined in most of these organs, although the extent of inflammation was significantly lower than that in the lungs (Fig. 1E to G). In particular, 100% of the infected mice (18 of 18) showed mild to moderate cardiac inflammation. Notably, although chlamydial antigens (inclusions) were clearly identified in the kidneys, the inflammation in these organs was not very significant (Fig. 1B and F). In aggregate, the results indicate that intranasal inoculation with C. trachomatis MoPn causes not only pneumonia but also systemic dissemination of the infection. In particular, the existence of chlamydial antigens (inclusions) and infiltration of inflammatory cells in cardiac tissues suggest that MoPn pneumonia may cause cardiac pathology.

View larger version (122K): [in this window] [in a new window]   FIG. 1.   Systemic dissemination of chlamydial organisms (antigens) and inflammatory cellular infiltration to multiple organs following intranasal infection with C. trachomatis MoPn. Female C57BL/6 mice (three to five mice/group) were intranasally infected with MoPn (2,000 IFU) and sacrificed on days 15 to 20 following infection. (A to D) Chlamydial inclusions (antigens) were detected by immunohistochemical staining with an anti-chlamydial LPS MAb in the liver (A), kidneys (B), heart (C), and lungs (D). (E to H) Histological structure and inflammatory infiltration (arrows) were examined by H & E staining in the liver (E), kidneys (F), heart (G), and lungs (H). The experiments were repeated four times, and similar results were obtained. Representative histological changes are shown. Magnification, ×400 (A to D) and ×200 (E to H).

MoPn infection induces perivascular inflammation, fibrosis, and myocardial hypertrophy. To further identify whether pulmonary MoPn infection can cause cardiac and vascular pathological changes, we examined the existence of chlamydial antigens and inflammatory reactions in the area within and surrounding cardiac blood vessels. The results showed that all of the infected C57BL/6 mice exhibited chlamydial antigens in the endothelium and/or the smooth muscle layers of cardiac blood vessels (Fig. 2A). The chlamydial antigen accumulation also displayed irregular morphology. Perivascular inflammation with lymphocytes and monocytes/macrophages was detected in 80% of the mice (14 of 18) (Fig. 2B). Staining of inflammatory cells with anti-CD4 and anti-CD8 MAbs showed significant infiltration of CD4 cells in the perivascular inflammation, with few CD8 cells (data not shown). More interestingly, about 60% of C57BL/6 mice with MoPn pneumonia (11 of 18) exhibited perivascular fibrosis in their cardiac blood vessels. The pathological changes in cardiac blood vessels were remarkably similar to that reported by Bachmaier et al., who showed perivascular fibrosis and inflammation induced by C. trachomatis peptides (2). Moreover, 40% of mice (7 of 18) also showed interstitial fiber proliferation in the myocardium (Fig. 3B) and about 20% of mice (4 of 18) showed myocardial hypertrophy and enlargement of nuclei of myocytes (Fig. 3D). The results indicate that respiratory tract C. trachomatis infection can indeed induce cardiovascular pathology, including inflammation and fibrotic changes in cardiac blood vessels and changes in myocardium observed in myocardiopathy.

View larger version (66K): [in this window] [in a new window]   FIG. 2.   Pathological change of cardiac blood vessels following intranasal infection with C. trachomatis MoPn. The mice in Fig. 1 were examined for pathological changes in their cardiac blood vessels. (A) Chlamydial inclusions (antigens) in the walls of blood vessels were detected by immunohistochemical staining with an anti-chlamydial LPS MAb. (B and C) Perivascular cellular infiltrations are shown by H & E staining (B), and the fibrotic changes in perivascular areas were determined by Masson trichrome staining (C). Experiments were repeated four times, and representative pathological changes are shown. Magnification, ×400.

View larger version (136K): [in this window] [in a new window]   FIG. 3.   Fibrotic and degenerative changes in the myocardium following pulmonary C. trachomatis MoPn infection. Mice were examined for fibrosis by using Masson's trichrome staining. (A) Trichrome stain of normal myocardium and blood vessels. (B) Trichrome stain of heart tissue collected from mice intranasally infected with MoPn (Fig. 1), showing fibrotic changes in the myocardium and perivascular areas. (C) H & E staining of normal myocardium. (D) H & E staining of the myocardium of MoPn-infected mice, showing myocardial hypertrophy and enlargement of nuclei. Experiments were repeated four times, and representative pathological changes are shown. Magnification, ×400.

IFN- KO mice show severe myocarditis, endocarditis, and pericarditis. IFN- is inhibitory for chlamydial growth both in vitro and in vivo. We recently found that IFN- KO mice, unlike wild-type controls, showed significant growth of chlamydial organisms in multiple organs, including the lungs, liver, and kidneys, following intranasal infection with MoPn, demonstrated by constant recovery of viable organisms from these organs (33a). In the present study, we further examined chlamydial infectivity in the hearts of intranasally infected IFN- KO mice. As shown in Fig. 4, unlike wild-type controls, IFN- KO mice repeatedly showed chlamydial infectivity in the heart. Immunohistochemical staining with anti-chlamydia LPS MAb also showed chlamydial inclusions in the cardiac tissues of IFN- KO mice (Fig. 5B). The morphology of the inclusions (chlamydial antigen accumulations) in the myocardium of IFN- KO mice was also variable but more closely resembled the shape of chlamydial inclusions seen in in vitro HeLa cell cultures. Histological analysis showed much more profound pathological changes in the heart of IFN- KO mice compared with wild-type mice described above (Fig. 1G). All IFN- KO mice (17 of 17) displayed massive inflammatory infiltration in the interstitial areas of the heart (Fig. 5A). Perivascular cellular infiltration, especially surrounding small blood vessels, was only occasionally found in IFN- KO mice. Interestingly, although all IFN- KO mice showed severe cardiac inflammation, most of them (16 of 17) did not show fibrosis in cardiac blood vessels (data not shown). Moreover, 45% of the IFN- KO mice (4 of 9) whose aortic valves were examined showed changes of endocarditis characterized by aortic valve vegetations (Fig. 5C). In contrast, none of the infected wild-type mice (0 of 7) whose aortic valves were examined showed vegetation. The vegetations in IFN- KO mice were large and were seen only on aortic valves. Vegetations contained mononuclear cells and polymorphonuclear cells admixed with fiber (Fig. 5D). Chlamydial organisms (antigens) were also detected in the vegetations after histoimmunological staining with anti-chlamydial LPS MAb (Fig. 5E). Inflammatory infiltration was also found in pericardial areas of 50% of the IFN- KO mice (9 of 17), with a trend to more severe inflammation in the position close to the root of aorta (Fig. 5F and G) and composed of a significant number of CD4 cells (Fig. 5H). Five naive (uninfected) wild-type and five naive IFN- KO mice were also examined for cardiac inflammation, vegetation, and vascular fibrosis, and none of them showed positive findings. Taken together, the results showed that IFN- KO mice suffered more serious cardiovascular chlamydial infection than did wild-type mice, suggesting that IFN- plays a critical role in preventing massive dissemination of chlamydial infection, thus preventing chlamydial endocarditis and pancarditis.

View larger version (21K): [in this window] [in a new window]   FIG. 4.   Significant growth of C. trachomatis MoPn in the heart following intranasal infection. Wild-type and IFN- KO mice were intranasally infected with MoPn (2,000 IFU) and sacrificed on day 15 postinfection. The homogenates of the cardiac tissues were analyzed for in vivo chlamydial growth as described in Materials and Methods. Each point represents the mean and standard deviation of log10 IFU for five mice. One of two independent experiments with similar results is shown.

View larger version (117K): [in this window] [in a new window]   FIG. 5.   Severe cardiac pathology in IFN- KO mice following intranasal infection with C. trachomatis MoPn. IFN- KO mice were intranasally infected with MoPn (2,000 IFU) and sacrificed on days 15 to 20 postinfection. Photomicrographs of H & E-stained sections show massive interstitial inflammation (A), vegetation on the aortic valve (C), pericardiac inflammation (F), and massive cellular infiltration at the area close to the root of the aorta (G). Immunohistochemical staining with anti-chlamydial LPS MAb shows chlamydial inclusions (antigens) in myocytes (B) and within the vegetation (E). Trichrome staining shows fiber (green) in the aortic vegetation (D). CD4 cell staining shows periaortic infiltration of CD4 cells. (H) The experiments were repeated four times, and similar results were obtained. Representative histological changes are shown. Magnifications, ×400 (A and B) and ×200 (C to H).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, we used a unique C. trachomatis strain (MoPn), which normally causes respiratory tract infection in mice, to explore whether pulmonary C. trachomatis infection can induce cardiovascular pathological changes. The data clearly show that respiratory tract infection with C. trachomatis is able to cause systemic and myocardial chlamydial infection and, probably more importantly, induce fibrotic changes in cardiac blood vessels. Moreover, the present study showed that about 40% of mice exhibited fibrotic changes in the myocardium and that 20% of mice exhibited myocardiopathy, suggesting that C. trachomatis may be an important causative agent of clinical cardiomyopathy. Notably, there are reported clinical cases of chronic cardiomyopathy caused by chlamydial infection (25). The present study, on one hand, demonstrates that C. trachomatis in certain conditions can induce cardiovascular diseases including perivascular fibrosis and, on the other hand, suggests that the intranasal-infection model of MoPn in mice may be useful for investigating cardiovascular diseases caused by chlamydial infection. This study, using a natural murine C. trachomatis infection model, confirmed and extended the very recent finding made by Bachmaier et al. of vascular inflammatory and fibrotic changes caused by the antigens of C. trachomatis (2). Since C. trachomatis infection is common in humans and since the heart is one of the organs infected by the agent, we speculate that the real prevalence of myocarditis and cardiomyopathy may be significantly higher than that previously documented. It should be noted, however, that the cardiovascular pathological changes caused by MoPn infection observed in this study are not the same as the characteristic vascular pathology associated with C. pneumoniae, i.e., arthrosclerosis. Further study is required to examine the role of cardiovascular infection with MoPn following intranasal inoculation of organism in the formation of arthrosclerosis.

Another very interesting finding in this study is that IFN- KO mice intranasally infected with MoPn exhibited endocarditis and pericarditis. Endocarditis caused by C. trachomatis infection has been reported in previous clinical studies (6, 7, 24, 33). Several individual cases of chlamydial endocarditis caused by C. pneumoniae have also been documented in clinical practice over the past decade (5, 13, 20). Thus far, however, no experimental study has been carried out to investigate the mechanism for this endocarditic disorder. In particular, it is not clear why only a few individuals among a large population with C. trachomatis infection suffer chlamydial endocarditis. In the present study, we found that IFN- KO, but not wild-type, mice showed severe aortic vegetation and pancarditis following respiratory tract MoPn infection. The results suggest that chlamydial endocarditis, particularly vegetation formation, occurs only when massive infection exists and also suggest that IFN- is very efficient in preventing the formation of endocarditis and vegetation. The observation that IFN- KO mice suffer remarkably more severe cardiovascular infection but significantly less vascular fibrosis may be due to the ongoing acute inflammation in these mice; thus, healing or fibrotic reaction does not occur, although the possibility that IFN- is involved in the process of fibrotic reaction cannot be excluded.

In conclusion, the study suggests that disseminated C. trachomatis infection can induce cardiovascular diseases including perivascular fibrosis. Severe endocarditis and pericarditis occur when the host fails to control the infection, leading to massive growth of the organism and severe systemic dissemination of the infection. The mouse lung infection model may provide a useful system for studying the mechanism and therapeutic approaches for chlamydia-induced cardiovascular diseases.