OUP user menu

Chlamydia pneumoniae infections in mouse models: relevance for atherosclerosis research

Martijn D. de Kruif, Eric C.M. van Gorp, Tymen T. Keller, Jacobus M. Ossewaarde, Hugo ten Cate
DOI: http://dx.doi.org/10.1016/j.cardiores.2004.09.031 317-327 First published online: 1 February 2005


Mouse models have been frequently used in the study of Chlamydia pneumoniae (also known as Chlamydophila pneumoniae) infections. This gram-negative obligate intracellular bacterium causes respiratory infections, followed by dissemination of the bacterium to various organs throughout the body, including cardiovascular tissues, supporting the current hypothesis of a relationship between C. pneumoniae and atherosclerosis. Recently, clinical trials evaluated the effect of antichlamydial antibiotics on secondary cardiovascular events. Although small studies showed some effect, the large WIZARD study did not confirm these results, and the role of antichlamydial antibiotics in prevention of secondary events was questioned. To address these issues, data obtained from mouse models were systematically reviewed here. C. pneumoniae infections showed atherogenic properties in mice that were reproducible and confirmed by different research groups. However, antibiotic therapy was of limited value in these mouse models. Antibiotic therapy effectively cleared the acute infection, but did not influence the atherogenic properties of C. pneumoniae unless the therapy was started early during the acute infection. The results summarized here may help to better understand the results of the clinical antibiotic trials.

  • Chlamydia pneumoniae
  • Chlamydia psittaci strain TWAR
  • Chlamydophila pneumoniae
  • Mouse models
  • Mice
  • Cardiovascular diseases
  • Atherosclerosis
  • Antibiotics

1. Introduction

Atherosclerosis is considered an inflammatory disease [1]. The precise trigger for inflammation is not known, but several infectious organisms have been suggested to play a role in this process [2]. The most prominent of these organisms is the gram-negative, obligate intracellular bacterium Chlamydia (C.) pneumoniae (officially known as Chlamydophila pneumoniae [3], but more commonly called Chlamydia pneumoniae). C. pneumoniae enters the body as a respiratory pathogen, and causes infections in a wide range of severity, from asymptomatic disease to severe pneumonia [4,5]. The acute infection is sometimes followed by a prolonged period of malaise and coughing, suggesting a chronic reaction to the organism [4,6]. Notably, the effects of C. pneumoniae infection are not limited to the respiratory tract. Cases of myocarditis and endocarditis due to C. pneumoniae have been described [7,8]. In addition, C. pneumoniae infection has been associated with chronic diseases, including atherosclerosis, asthma, arthritis, hypertension and multiple sclerosis in merely epidemiologic studies [4,5,9].

A serologic relationship of C. pneumoniae infection with atherosclerosis was first demonstrated in 1988 [10]. Saikku et al. showed that patients with coronary artery disease had elevated antibody titers against C. pneumoniae. Although this serologic relationship has been disputed [11], it stimulated further research. In atheromatous tissues from human vessels, C. pneumoniae was detected by several in situ methods [12,13]. In vitro data suggested a role for chlamydial heat shock protein 60 (cHSP60) in oxidation of low-density lipoproteins and a role for chlamydial lipopolysaccharide (LPS) in foam cell formation [14,15]. In mice, atherosclerotic lesion development was accelerated by C. pneumoniae infection, and reactive infiltration of inflammatory cells in intimal plaques was demonstrated [16–22]. In rabbits, acceleration of lesion development by C. pneumoniae could be inhibited by antibiotics after infection [23].

This accumulating evidence for a relationship between C. pneumoniae and atherosclerosis led to the hypothesis that antibiotic therapy might influence the course of cardiovascular disease [24]. Initially, a number of small studies showed some beneficial effects of antichlamydial antibiotics in the outcomes of various cardiovascular complications, including myocardial infarction and angina pectoris [25]. However, the WIZARD study, a recently published large randomized trial that included 7747 patients with previous myocardial infarction and elevated C. pneumoniae titers, showed no beneficial effects of a 3-month course of azithromycin on secondary cardiac events after a median follow-up of 14 months [26]. Two other large trials, ACES and PROVE IT [27,28], were recently preliminary presented at the European society of cardiologists congress 2004, which was held in Munich, Germany, from August 28 to September 1st, 2004, and showed similar negative results. The negative results of these studies prompted a lot of questions, related to the pathogenetic involvement of C. pneumoniae in cardiovascular disease, on the one hand, and its susceptibility to antibiotic treatment on the other hand [29]. Many of these questions can be discussed on the basis of studies that have been performed in mouse models, which have been the predominant models for studying atherogenesis during the past several years. Here, we present a systematic overview of C. pneumoniae research in mice, emphasizing its relationship with atherosclerosis.

2. Methods

Citations were retrieved from Pub Med and MEDLINE databases. Using the single terms “Chlamydophila pneumoniae”, “Chlamydia pneumoniae” and “Chlamydia psittaci strain TWAR”, and combinations of these terms with “mouse” or “mice”, or “atherosclerosis”, or “antibiotics”, titles, abstracts and references were scanned for the use of mouse models in C. pneumoniae research. A total of 56 articles using mouse models was retrieved, from 1990 to 2004.

3. Mouse models: general aspects

The first mouse models for C. pneumoniae infection were developed in the beginning of the 1990s, when evidence became clear that C. pneumoniae was a major respiratory pathogen [30–32]. The models used several mouse strains that differed in susceptibility to infection. Swiss Webster mice and NIH/S mice were highly susceptible, followed by C57BL/6 mice, whereas BALB/c mice were least susceptible [31,32]. In addition, C. pneumoniae strains differed in virulence, although they are known to show very little genetic diversity [33]. For each strain, a specific dose was established (Table 1). The most widely used C. pneumoniae strain was AR-39, a strain originally isolated from a patient's throat swab in 1983 and known to be host to a bacteriophage with possible pathogenic properties [34,35]. Other commonly used strains were TW-183, which was the first C. pneumoniae isolate and was isolated from a trachoma patient in Taiwan in 1965 and was propagated afterwards for a long time in a yolk sac [5], and Kajaani 6, which was isolated from pharyngeal swabs in an epidemic of C. pneumoniae infections in military conscripts in 1987 [36].

View this table:
Table 1

Methods in studies using mouse models for the study of C. pneumoniae infection

AuthorsYearMouse strainClpn strainDose × 106 IFUaInoculation (week)bMultiple inoculationscInterval (week)Reinfection (week postinfection)dFollow-up (week)eMain topic
Kishimoto [30]'90ICRTW-1836–73Course of infection
Yang et al. [32]'93Swiss Webster, Icr, BALB/c C, C57Bl/6, C3H/HeN, B6C3F1AR-39304–510, 14 2/715 2/7Course of infection
Kaukoranta-Tolvanen et al. [31]'93NIHS, Swiss Webster, BALB/cK 6, H12, TW-1830.2–26–86Course of infection
Kimura [44]'94ICRTW-1836–753Course of infection
Nakata et al. [45]'94MCH-ICRK 6, IOL-2070.0001–0.13Antibiotics
Kaukoranta-Tolvanen et al. [46]'95NIHSK 616–84, 1014Course of infection
Yang et al. [47]'95Swiss WebsterAR-39304–51 4/7Course of infection
Malinverni et al. [48]'95Swiss WebsterAR 388, AR-391.5–13414 4/7Antibiotics
Masson et al. [49]'95MF1, NIHS, BALB/c, C3H/HeTW-1830.0017–0.0454–64 1/7Antibiotics
Malinverni et al. [50]'95Swiss WebsterAR-391.545 4/7Course of infection
Laitinen et al. [51]'96NIHSK 6166 2/7Course of infection
Peterson et al. [37]'96BALB/c, C57Bl/6TW-1831069 4/7Immunology
Moazed et al. [42]'96Apo E, C57Bl/6AR-393083 ×24Course of infection
Moazed et al. [38]'98C57Bl/6AR-39306–81Course of infection
Penttila et al. [52]'98BALB/cK 616–84–816 6/7Course of infection
Penttila et al. [53]'98BALB/c, NIHS, C57Bl/6K 61–1.56–7612Immunology
Penttila et al. [54]'98BALB/c, nudeK 61–106–85–810 5/7Immunology
Peterson et al. [55]'98Swiss WebsterAR-390.002–204–65/7Immunology
Rottenberg et al. [56]'99Multiple KO miceK 616–108 4/7Immunology
Wolf and Malinverni [57]'99NMRIAR-39503–42 6/7Antibiotics
Hu et al. [16]'99LDL R−/−AR-395–104–59 ×443 4/7Atherosclerosis
Moazed et al. [17]'99Apo E −/−AR-393083 ×110Atherosclerosis
Blessing et al. [43]'00C57Bl/6AR-393083 ×120Atherosclerosis
Liu et al. [18]'00LDL R−/−AR-395–104–512 ×223 4/7Atherosclerosis
Liuba et al. [58]'00Apo E −/−, C57Bl/6IOL-2070.484 ×210Atherosclerosis
Vuola et al. [59]'00C57Bl/6, BALB/cK 616–84 5/7, 910 5/7Immunology
Geng et al. [60]'00BALB/c, 129sv, IFN-y R−/−TW-1830.036–82 2/7Immunology
Rottenberg et al. [61]'00Multiple KO micefK 616–108 4/7Immunology
Pal et al. [62]'00129 sv, Nramp1−/−TW-1830.17–91 3/7Immunology
Erkkila et al. [41]'00NIHS, C57Bl/6K 6, K 71.2 –1.86–83Techniques
Meijer et al. [63]'00C57Bl/6AR-391–4304–63Techniques
Svanholm et al. [64]'00Multiple KO micefK 6110–141Immunology
Murdin et al. [65]'00BALB/cAR-390.05–0.53 1/7Immunology
Bin et al. [66]'00NMRIAR-395049Antibiotics
Rothstein et al. [21]'01Apo E −/−A-035102 ×214Antibiotics
Penttilä et al. [67]'01BALB/cK 6111–131 3/7Immunology
Blessing et al. [19]'01C57Bl/6AR-393083 ×216Atherosclerosis
Aalto-Setälä et al. [68]'01Apo E −/−K 70.1–384 ×3–418Atherosclerosis
Caligiuri et al. [69]'01Apo E −/−, C57Bl/6K 616–81822Atherosclerosis
Burian et al. [70]'01BALB/c0.02–0.046–83 ×7Atherosclerosis
Burnett et al. [20]'01Apo E −/−62 ×210Atherosclerosis
Blessing et al. [71]'02C57Bl/6AR-393083 ×218Atherosclerosis
Ezzahiri et al. [39]'02Apo E3-Leiden −/−TWAR 20436092 ×29, 3236Atherosclerosis
Erkkila et al. [72]'02NIHSK 616–84 2/78 4/7Techniques
Tiirola et al. [73]'02NIHSK 70.58–92 6/7Lipids
Bandholtz et al. [74]'02Multiple KO micefK 616–103Immunology
Airaksinen et al. [75]'03BALB/cK 616–889 3/7Immunology
Burian et al. [76]'03BALB/c, HDC −/−, CD1TW-18316–84 3/7Immunology
Shi et al. [77]'03ICRCWL-0297–9.788 4/7Course of infection
Ezzahiri et al. [40]'03Apo E −/− and LDL R −/−TWAR 20435–50166 ×424Atherosclerosis
Liuba et al. [78]'03Apo E −/−IOL-2070.273x210Atherosclerosis
Chesebro et al. [22]'03C57BL/6, iNOS−/−, eNOS−/−AR-392483 ×218Atherosclerosis
Rothfuchs et al. [79]'04Multiple KO micefK 616–103 4/7Immunology
Mygind et al. [80]'04C57BL/6VR 13100.25–212Immunology
Little et al. [81]'04C57BL/696-410.02–0.041212Amyloid depositions
Törmäkangas et al. [82]'04C57BL/6K 70.683Antibiotics
  • a IFUs: inclusion forming units.

  • b Inoculation: age of mice during first inoculation.

  • c Multiple inoculations: subsequent infections aimed at increasing the chance of establishing a chronic infection; number of inoculations and intervals in weeks.

  • d Reinfection: occasional extra inoculation.

  • e Follow-up: total duration of main experiment in weeks after first infection.

  • f Multiple KO mice: multiple gene knockout mouse strains (C57BL/6 background).

The mice were inoculated intranasally by droplets of 0.5 ml administered onto the nostrils, or sometimes intraperitoneally, when the respiratory tract infection was not the main focus of attention [22,37–40]. Anesthetics used during inoculation differed greatly. One study showed methoxyflurane anesthesia to be superior to CO2 anesthesia, resulting in more homogeneously infected animals [41]. In some models, multiple inoculations were introduced in order to mimic chronic infection; generally 2–4 intranasal inoculations were given at 1- to 2-week intervals [42].

Under normal conditions, mice do not develop atherosclerosis. However, wild-type C57BL/6 mice can develop early atherosclerotic lesions in the aortic root when they are being fed an atherogenic diet [43]. Several genetically modified mouse strains, mostly with a C57BL/6 background, were used, such as LDL receptor knockout mice and Apo E3 Leiden transgenic mice [16,39]. These mice are more prone to develop atherosclerosis than wild-type mice but still need to be fed an atherogenic diet. Apo E knockout mice do not need a special diet. They are characterized by disturbed LDL uptake by the liver and develop atherosclerosis spontaneously from the first stage of macrophage adhesions at the age of 10 weeks to the development of mature atheromas at 24 weeks [42].

4. Course of infection

Infection of mice with C. pneumoniae resulted in a self-limiting pneumonia [30–32,77]. Depending on the dose given, mice displayed symptoms, like dyspnea, weakness and weight loss. The symptoms reached a maximum within 2 to 4 days and rarely lasted longer than 1 week. Histological examination revealed peribronchial and perivascular inflammation, with exudates in bronchi and alveoli in severe cases [32,59,77]. Proliferation of lymphoid tissue and peribronchial fibrosis was observed after 1–2 weeks [31,32,46,63]. The infiltrates consisted of polymorphonuclear cells in the first days, which were gradually replaced by mononuclear cells within two weeks. At 3–6 weeks, no lesions were left in the lungs [31,32,42,49,77].

Isolation of viable C. pneumoniae organisms from the lungs was generally possible up to 4 weeks after infection and occasionally up to 6 weeks [31,32,42,47,49,52,54,66,72]. However, C. pneumoniae antigens and DNA were detected in the lungs for a much longer period, up to 20 weeks after infection [42]. The presence of C. pneumoniae antigens in the lungs was limited to macrophages in alveoli and bronchus-associated lymphoid tissue [63]. These findings suggested a latent persistence of the organisms, which was confirmed by reactivation of pulmonary infection using cortisone-induced immunosuppression experiments [50,51].

When mice were administered multiple inoculations, the histopathological changes lasted much longer, up to 16 weeks [42]. In reinfection experiments, a partial immunity developed, since less organisms were isolated [32,44,46,52]. Despite this clinical improvement, histopathological changes were similar to those in primary infection, except for a more pronounced lymphoid reaction indicative of induced immunological memory [46,52].

C. pneumoniae infection was not limited to the respiratory tract only. Following local infection, spreading of C. pneumoniae to multiple organs throughout the body was detected by PCR and immunohistochemical staining up to 20 weeks [42]. The dissemination was probably mediated by peripheral blood monocytes as they were positive both by PCR and isolation in some studies, whereas blood plasma was negative [38,47]. Transiently raised levels of triglycerides were found in three studies [21,68,73], but three other studies reported to have found no differences in lipid levels [16,18,19].

5. Immunological reaction

Like other chlamydiae, C. pneumoniae challenges the host defense mechanisms in a complicated manner [83]. The intracellular location of the bacterium provides relative protection against the immune system. The bacterium probably hides from detection by down-regulating processing and presentation of its own antigens [84,85] and the bacterium is able to prevent spontaneous apoptosis of its host cell [86,87].

The immunological reaction to C. pneumoniae infection in mice proceeded in two stages, a nonspecific reaction of the innate immune system followed by a specific reaction a few days later. In the specific reaction, CD8+ T lymphocytes played a major role [54,56] (see Fig. 1). The function of these cells was essential for resolution of the infection, although in C. pneumoniae infection, they did not act as perforin mediated cytolytic cells, since mice lacking perforin did not show more severe infection [56]. Probably, the function of CD8+ cells was mediated by their ability to release cytokines, thus stimulating a Th1 response and inhibiting a Th2 response. In addition, the CD8+ cells were important for C. pneumoniae specific memory [54]. In contrast to the CD8+ cells, CD4+ cells contributed to protection only in the later stages of infection. In the first weeks, CD4+ cells even played a potentially harmful role [54,56].

Fig. 1

Immunological reaction to C. pneumoniae infection. The innate immunity, represented by polymorphonuclear cells and macrophages, plays an important role in the first days of infection. Subsequently, CD8+T lymphocytes are predominant, mainly by stimulating the powerful antichlamydial cytokine IFN-γ. Antibody production by B lymphocytes plays a minor role (see also text). Abbreviations: PMN=polymorphonuclear cell, Mϕ=macrophage, CP-IB=C. pneumoniae inclusion body (intracellular), CP-EB=C. pneumoniae elementary body (extracellular).

The cytokine release was characterized by a Th1-type response with elevated levels of IFN- γ, TNFα, IL-12 and IL-10 and unaltered IL4 levels [60]. The most important cytokine was IFN-γ, both in specific and in nonspecific immunity [59]. IFN-γ stimulated several antibacterial mechanisms, including stimulation of inducible nitric oxide oxidase, and stimulated IL-12, which in turn stimulated IFN-γ [56,61].

Antibody production played only a minor role in the neutralization of C. pneumoniae infection [56]. C. pneumoniae specific IgG titers started to rise from 10 to 12 days after infection, reaching a maximum at 4 weeks and gradually disappearing by 16 weeks [31,32,42,72]. IgM production started from 1 week after infection, reaching a maximum in 2–3 weeks, and gradually subsided until 6 weeks [30,31,43].

6. Antichlamydial strategies

A few short-termed antibiotic trials were conducted in mouse models. There were no apparent differences found among doxycycline, azithromycin, erythromycin, amoxicillin-clavunalate, telithromycin and ciprofloxacin treatments, as they were all efficacious in clearing the active infection within 2 weeks [48,49,82]. Sparfloxacin was superior to clarithromycin, tosufloxacin, ofloxacin and minocycline in a model of leukopenic mice [45]. A combination of azithromycin and rifampin treated the infection better than amoxyciline or azithromycin treatment alone [57,66]. After therapy, C. pneumoniae DNA could still be detected by PCR, but viable bacteria could not be isolated from the tissues, which suggested the persistence of nonviable or nonreplicating bacteria [48,82].

A vaccine should theoretically circumvent the problem of persisting bacteria [83]. Partially effective immunizations were reported in C. pneumoniae infection using DNA vaccines encoding for several C. pneumoniae proteins, including MOMP, Omp2, Hsp60 and an ADP/ATP translocase [37,64,65,67,74,75]. Current research focuses on enhancing the power of these proteins as a vaccine by adding adjuvantia or combining proteins to multisubunit vaccines, and on several strategies to find new immunoreactive proteins.

7. Atherosclerosis

An accelerating effect of C. pneumoniae infection on atherogenesis was demonstrated in a considerable number of hyperlipidemic mouse models (Table 2). In wild-type C57BL/6 mice fed an atherogenic diet, the atherosclerotic lesions were significantly larger in infected animals [19,22]. Similar findings were found for genetically modified, hyperlipidemic mice with C57BL/6 backgrounds, such as Apo E knockout mice, LDL receptor knockout mice and Apo E3 Leiden transgenic mice [16–18,20,21]. Furthermore, lesions from infected mice represented more advanced stages of atherosclerosis, showed more unstable plaque phenotypes and attracted T cell infiltration [16,39,40]. Antibiotics did not influence lesion acceleration when administered to mice 2 weeks after the last infection [21]. In two studies, C. pneumoniae did not influence atherogenesis [68,69] (see Fig. 2).

Fig. 2

Scheme summarizing the limitations of antibiotic therapy of C. pneumoniae infection in mice. Antibiotic therapy is effective in clearing the acute active infection in lungs, but a number of organisms are able to survive in the lungs in a persistent state. Furthermore, inhibition of the pro-atherogenic effects of C. pneumoniae is only possible when the antibiotics are given early during the infection. It is not clear what the precise mechanism is of the pro-atherogenic properties of C. pneumoniae in mice, and it is not clear whether antibiotics can affect the presence of C. pneumoniae antigens in atherosclerotic plaques (see also text).

View this table:
Table 2

Effect of C. pneumoniae infection upon atherosclerotic lesions in mice; overview of studies

AuthorYearMiceaGenderAtherogenic dietTime post infection (week)Lesion size accelerationbOther findings
Hu et al. [16]'99LDL R −/−F+43 4/7Yes. 1.5 ×Plaque composition accelerationc
Moazed et al. [17]'99Apo E −/−M10Yes. 1.6 ×
Blessing et al. [43]'00C57BL6M20(No)Inflammatory changes aorta
Liu et al. [18]'00LDL R −/−F+23 4/7Yes. 1.3 ×
Liuba et al. [58]'00Apo E −/−, C57BL6M/F6(No)Endothelial dysfunction
Blessing et al. [19]'01C57BL6M+16Yes. 3.3 ×
Aalto-Setälä et al. [68]'01Apo E −/−M/F±18No
Caligiuri et al. [69]'01Apo E −/−, C57BL6F22No
Burian et al. [70]'01BALBCM7(No)Inflammatory changes aorta
Burnett et al. [20]'01Apo E −/−M/F10Yes. 1.7 ×
Rothstein et al. [21]'01Apo E −/−F14Yes. 1.5 ×No effect of azithromycine
Blessing et al. [71]'02C57BL6M+18(No)Diet started after infection
Ezzahiri et al. [39]'02Apo E3-LeidenM/F+36(No)Plaque composition acceleration
Chesebro et al. [22]'03iNOS −/−, eNOS −/−M+18Yes. 2.6 ×
Ezzahiri et al. [40]'03Apo E −/− LDL R −/−M24(No)Reduced fibrous cap area
Liuba et al. [78]'03Apo E −/−M/F10(No)Endothelial dysfunction
  • a Time postinfection: time after first inoculation (weeks). More details about the inoculation schedules are given in Table 1.

  • b Lesion size acceleration: lesion size of plaques from C. pneumoniae-infected mice compared to lesion size from uninfected mice.

  • c Plaque composition acceleration: atherosclerotic lesions in a more advanced stage at a certain age in C. pneumoniae-infected mice compared to uninfected mice.

The atherogenic properties of C. pneumoniae required multiple inoculations in all cases, and hyperlipidemia was essential. When the atherogenic diet was administered to wild-type mice 2 weeks after infection, no lesion acceleration was seen [71]. Infection at times when advanced atherosclerotic plaques had already formed did not influence lesion size but resulted in more unstable plaque phenotypes [40]. Wild-type mice on a normal diet developed temporary inflammatory lesions in several organs, including the aorta, heart, spleen and liver [42,43,47,70], but none of the aorta inflammatory lesions progressed to atherosclerotic lesions [43,70]. Murine cytomegalovirus infection produced similar inflammatory lesions, and superinfection with C. pneumoniae had an impressive additional effect upon the size of these lesions [70]. However, no such additional effects were found in Apo E knockout mice, although each pathogen on itself accelerated lesion development [20]. Infection with C. pneumoniae resulted in increased endothelial dysfunction and enhanced VCAM-1 expression in Apo E knockout mice, and coinfection with Helicobacter pylori stimulated these effects [78]. In one study, lesion acceleration was found in male mice only, which suggested a gender-dependent effect [20].

8. Discussion

Mouse models provided detailed information about C. pneumoniae infections. Many of these details were rather uncommon to a respiratory tract infection. The histopathological changes in the lungs were considerable, although the clinical disease was rather mild [31,32]. The host defense, characterized by a Th1 response requiring IFN-γ for clearing the bacteria, was only partly successful in controlling the infection [56]. Afterwards, the C. pneumoniae organisms persisted in the lungs in a latent phase that could be reactivated by the administration of cortisone [50,51]. This phenomenon could be related to the capability of IFN-γ to induce persistence of the bacteria. The addition of IFN-γ to C. pneumoniae-infected monocytic cell lines induced a similar kind of persistence of nonreplicating, viable bacteria [88].

Another uncommon finding of a respiratory tract infection is that C. pneumoniae disseminated throughout the body and caused inflammatory lesions in several organs, including the heart and the vessel walls [47]. The dissemination was generally mediated by leukocytes, in which C. pneumoniae has found a way of surviving [38]. Notably, only during a short period following infection, viable organisms were isolated from the organ tissues [42]. Also in human atherosclerotic tissue specimens, isolation of viable organisms generally was a rare event. In these tissue specimens, membrane components of C. pneumoniae were detected by in situ methods, but no inclusions were found that resembled viable organisms [89]. Probably, viable organisms were capable of reaching the atheroma, but did not survive there for long, and left nonviable, potentially pathogenic remnants behind. It is also possible that C. pneumoniae persisted in the atheroma in a latent state as had been found in lung tissue, but no such inclusions were detected in mice. In either case, antibiotic treatment is of limited value, because it aims at viable, replicating bacteria.

The presence of C. pneumoniae in the atheroma seems to be pathogenic. Atherogenesis was accelerated in a number of mouse models, and the results were reproducible and confirmed by different research groups (Table 2). At first, acceleration of atherogenesis was demonstrated in genetically modified mice only [16,17]. These results were difficult to interpret because the genetic modification in itself could have influenced the inflammatory process [90]. Additional evidence came from normal C57BL/6 mice fed an atherogenic diet [19,71]. It became clear that multiple inoculations and a hyperlipidemic environment were essential, and some evidence suggested a gender-dependent effect [43,71]. In two studies, the conclusion was drawn that C. pneumoniae did not accelerate atherogenesis. In one of these studies [69], no reinfection after the first inoculation had been given until 18 weeks after infection, which could explain the lack of association with atherogenesis, arguing that a model of chronic infection is necessary. Furthermore, these investigators used a lower dose of C. pneumoniae [91]. In the other study by Aalto-Setälä et al. [68], Apo E knockout mice on an FVB background were used. These mice are known to develop atherosclerosis more slowly than Apo E knockout mice on C57BL/6 background, which may have influenced the findings [92]. The investigators used C. pneumoniae strain Kajaani 7, which is known to be cleared quickly [41], so it may not have been able to induce chronic infection. On the other hand, this strain was very virulent in acute disease, and many mice died, up to 80% in a group, leaving 1 to 5 mice in a group. The conclusions drawn from this study are therefore difficult to interpret.

The atherogenic properties of C. pneumoniae in mice have also been demonstrated in a few studies in rats, rabbits and pigs, suggesting that the results of these animal models can be extrapolated to human beings [93–95]. The mechanism is not known, but several pathways were proposed for C. pneumoniae-enhanced atherosclerosis [96]. These mechanisms include (1) direct stimulation of monocyte migration to the atheromas [97]; (2) direct interference with intracellular metabolism within the plaque, as C. pneumoniae was able to infect all cell types present in a plaque in vitro, thereby inducing atherogenic processes such as foam cell formation and LDL oxidation [14,15]; (3) the lesions could be complicated by an immunopathogenic role of C. pneumoniae, attracting inflammatory cells causing tissue damage. T cell infiltration which was demonstrated both in murine and human atherosclerotic plaques [39,98]; (4) triggering of an autoimmune reaction to human Hsp60 by antigenic mimicry could be interfering with atherogenesis [14,99]; and (5) apart from the atherogenic properties, C. pneumoniae could complicate atherosclerotic disease further by destabilizing plaques by reducing the fibrous cap area and stimulating matrix degrading metalloproteinases [40], or by enhancing thrombogenicity by stimulating coagulation factors and tissue factor expression [100,101].

Antibiotics were successful for the treatment of the acute pulmonary infection, but had no effect on the atherogenic properties of C. pneumoniae when they were administered 2 weeks after infection [21]. In addition, in rabbits, antibiotics inhibited atherogenesis only slightly when they were administered 2 weeks after infection, but they were much more effective when they were administered within 5 days after infection [23,102]. Furthermore, antibiotics failed to completely eradicate C. pneumoniae from the organ tissues [48,103]. These findings may help to understand the negative results of the human WIZARD study, because the antibiotics in animals were apparently only effective when they were given during the actual infection itself. The acute C. pneumoniae infection in humans frequently passes clinically unnoticed.

C. pneumoniae was one of the first and therefore most frequently studied infectious organisms that showed atherogenic properties in mice. The effect seemed to be specific to C. pneumoniae because C. trachomatis did not influence atherogenesis [104]. However, other organisms that were not clearly related did show atherogenic properties in similar mouse models. These organisms included various bacteria, viruses and protozoa, such as Porphyromonas gingivalis, murine cytomegalovirus, murine γ-herpesvirus-68, Toxoplasma gondii and Trypanosoma cruzi [105–110]. The contribution of these single pathogens is still being researched. Additional atherogenic properties to C. pneumoniae infection were found in two out of three coinfection studies with H. pylori and murine cytomegalovirus [70]. After all, the impact of the pathogen burden as a whole may still be more important than the impact of single pathogens alone.

In conclusion, the evidence for a pathogenic role of C. pneumoniae in atherosclerosis in mouse models seems convincing, because the results were reproducible and confirmed by different research groups. However, the precise mechanisms of this relationship remain unclear, although many theories exist. Temptations to eradicate C. pneumoniae from the body, both by the immune system and by antibiotic treatment, generally encountered many difficulties which could be related to the unusual gram-negative, obligate intracellular characteristics of the bacterium. The value of the currently available antibiotics should not be overestimated, because they were only effective in preventing the long-term atherogenic effects of C. pneumoniae infection when they were given during the acute infection, which is often asymptomatic or aspecific in humans. Therefore, the focus of future research should merely shift back to basic research. The precise mechanisms by which C. pneumoniae is interfering with atherogenesis should be further elucidated to enable more specific strategies to interfere with this process, e.g. by immunosuppressive agents. In addition, the role of other infectious pathogens should be further examined, and especially the impact of the pathogen burden as a whole. The development of an effective vaccine would be the final strategy to prevent chlamydial disease and its complications, but not to prevent the impact of the pathogen burden.


  • Time for primary review 26 days


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
  46. [46]
  47. [47]
  48. [48]
  49. [49]
  50. [50]
  51. [51]
  52. [52]
  53. [53]
  54. [54]
  55. [55]
  56. [56]
  57. [57]
  58. [58]
  59. [59]
  60. [60]
  61. [61]
  62. [62]
  63. [63]
  64. [64]
  65. [65]
  66. [66]
  67. [67]
  68. [68]
  69. [69]
  70. [70]
  71. [71]
  72. [72]
  73. [73]
  74. [74]
  75. [75]
  76. [76]
  77. [77]
  78. [78]
  79. [79]
  80. [80]
  81. [81]
  82. [82]
  83. [83]
  84. [84]
  85. [85]
  86. [86]
  87. [87]
  88. [88]
  89. [89]
  90. [90]
  91. [91]
  92. [92]
  93. [93]
  94. [94]
  95. [95]
  96. [96]
  97. [97]
  98. [98]
  99. [99]
  100. [100]
  101. [101]
  102. [102]
  103. [103]
  104. [104]
  105. [105]
  106. [106]
  107. [107]
  108. [108]
  109. [109]
  110. [110]
View Abstract