биохимия атеросклероза

412 Stewart C. Whitman and Tanya A. Ramsamy

104.Hong S, Wilson MT, Serizawa I, Wu L, Singh N, Naidenko OV, Miura T, Haba T, Scherer DC, Wei J, Kronenberg M, Koezuka Y, Van Kaer L: The natural killer T-cell ligand alpha-galactosylceramide prevents autoimmune diabetes in nonobese diabetic mice. Nat Med 7: 1052–1056, 2001.

105.van der Vliet HJ, Molling JW, von Blomberg BM, Nishi N, Kolgen W, van den Eertwegh AJ, Pinedo HM, Giaccone G, Scheper RJ: The immunoregulatory role of CD1d-restricted natural killer T cells in disease. Clin Immunol 112: 8–23, 2004.

106.Binder CJ, Chang MK, Shaw PX, Miller YI, Hartvigsen K, Dewan A, Witztum JL: Innate and acquired immunity in atherogenesis. Nat Med 8: 1218–1226, 2002.

107.Linton MF, Major AS, Fazio S: Proatherogenic role for NK cells revealed. Arterioscler Thromb Vasc Biol 24: 992–994, 2004.

108.Ostos MA, Recalde D, Zakin MM, Scott-Algara D: Implication of natural killer T cells in atherosclerosis development during a LPS-induced chronic inflammation. FEBS Lett 519: 23–29, 2002.

109.Melian A, Geng YJ, Sukhova GK, Libby P, Porcelli SA: CD1 expression in human atherosclerosis. A potential mechanism for T cell activation by foam cells. Am J Pathol 155: 775–786, 1999.

110.Mendiratta SK, Martin WD, Hong S, Boesteanu A, Joyce S, Van Kaer L: CD1d1 mutant mice are deficient in natural T cells that promptly produce IL-4. Immunity 6: 469–477, 1997.

111.Cui J, Shin T, Kawano T, Sato H, Kondo E, Toura I, Kaneko Y, Koseki H, Kanno M, Taniguchi M: Requirement for Valpha14 NKT cells in IL-12-medi- ated rejection of tumors. Science 278: 1623–1626, 1997.

112.Godfrey DI, Hammond KJ, Poulton LD, Smyth MJ, Baxter AG: NKT cells: facts, functions and fallacies. Immunol Today 21: 573–583, 2000.

113.Whitman SC, Ravisankar P, Daugherty A: Interleukin-18 enhances atherosclerosis in apolipoprotein E(–/–) mice through release of interferon-gamma. Circ Res 90: E34–E38, 2002.

114.Elhage R, Jawien J, Rudling M, Ljunggren HG, Takeda K, Akira S, Bayard F, Hansson K: Reduced atherosclerosis in interleukin-18 deficient apolipoprotein E-knockout mice. Cardiovasc Res 59: 234–240, 2003.

414 Paul K.M. Cheung and Grant N. Pierce

response to endothelial injury remains a guiding framework within which most studies have been conducted. Specifically, the hypothesis can be summarized as:

Injury to the lining of the artery (endothelium) results in protective inflammatory and fibroproliferative responses which, when in excess, becomes the disease process of atherogenesis.

There are several well-studied risk factors of endothelial injury in arteries, such as hypertension [5–7], diabetes [8, 9], obesity [10], high-plasma homocysteine [11–13], and hyperlipidemia [14, 15]. Several comprehensive review articles on these subjects are available [16, 17]. These factors, however, may account for only 50% to 70% of atherosclerotic events in a general population [18]. One less recognized risk factor is the role of infection in stimulating atherogenic events. In recent years, through statistical associations and investigation in animal models, several viral and bacterial pathogens have been identified as potentially atherogenic. Current opinions remains divided over the significance of these infectious agents in the overall, long-term process of lesion and plaque development. This skepticism largely arises from conflicting reports of clinical correlation and the lack of success in treating atherosclerotic patients with the antibiotics. In this article, we shall focus our review on the in vitro and in vivo investigations that link several viral and bacterial pathogens to lesion formation. In particular, we shall discuss the extensive evidence that suggests the mechanism through which Chlamydia pneumoniae, a pathogen that is widely present in the human population, may cause atherosclerosis.

The general development of an atherosclerotic lesion can be summarized as follows. Early stage lesions can occur in young children as an accumulation of foam cells in the subendothelial layer (and above the smooth muscle cell intimal layer) of the arterial environment [3]. These foam cells typically originate from circulating monocytes and T lymphocytes which adhere to the endothelial layer, migrate into the intimal space, and absorb excess lipoproteins [19]. A typical early lesion like this is termed a “fatty streak” and is considered insignificant in affecting blood flow or endothelial function. However, over decades of development, continued foam cell accumulation causes significant narrowing of the arterial lumen. Biochemical interactions among cells in the lesion milieu lead to the heightened expression of extracellular proteases (such as matrix metalloproteinases), chemokines (such as IL-8), and cytokines (such as TNF-α) that results in fibrosis, cell proliferation, cell migration, and accelerated plaque formation. At this stage, the lesion is termed a “plaque.” The development of a plaque has two important outcomes: (i) reduction in the lumen area (flow volume) with dysfunction of the intimal layer including a loss in arterial contractility and (ii) an increased likelihood of autoimmune responses against the lesion and hence the evolution of a lesion into an unstable “fibrotic plaque.” These late stage fibrotic plaques are characterized by an increased proportion of calcified structures

and necrotic cells. Fibrotic plaques can “grow” to occlude the entire blood vessel, resulting in acute myocardial infarction if it happens inside the coronary arteries, or ischemic stroke if it occurs inside cerebral arteries. Destabilization and rupturing of the fibrotic lesions can also occur. Plaque material is released into the lumen of the artery and elicits an immediate thrombotic response, which can result in a blockage of the artery. Figure 19.1 summarizes this process of atherosclerosis.

Infectious Agents and Atherosclerosis

Several viral and bacterial agents are strongly associated with atherogenesis. These include herpesviruses, cytomegalovirus, Helicobacter pylori, Porphyromonas gingivalis, and C. pneumoniae. In the following sections, we shall review current findings that suggest the potential roles of these infectious agents in causing atheroclerosis.

Herpesvirus and Cytomegalovirus

Cytomegalovirus (CMV) and herpesviruses (HPV) have been detected in atherosclerotic arteries [20–25]. Herpes simplex virus type 2 is more frequently observed in coronary artery biopsy specimens than the type 1 virus [26]. A small HIV study involving eight young patients revealed unusually severe coronary lesions that may be associated with the HIV or coexisting CMV and HPV [27]. The CMV genome has also been detected by PCR in arterial samples taken from patients undergoing vascular surgery [28]. CMV-seropositive patients are also associated with more advanced atherosclerotic plaques [28]. Other clinical studies of allograft transplant patients have demonstrated associations between CMV infection and thickening of the intima/endothelial layers [29]. Patients suffering acute myocardial infarction demonstrate a significantly higher prevalence of serum CMV DNA [30].

The important limitation in all of the above studies, however, is that they cannot distinguish the viral infections as causative agents of atherosclerosis from infections that are enhanced by preexisting factors in the lesion milieu. To support the assertion that CMV and HPV can cause atherosclerosis, studies in animal models have been carried out. The Marek’s disease virus (her- pes-type DNA virus) is reported to cause atherosclerosis in chicken, leading to a pathology that is analogous to those observed in human arteries [31]. Interestingly, this atherogenic condition can be prevented by vaccination, supporting a causative role of the virus in atherogenesis. A more definitive test of a causative agent, however, would be to verify that the human virus can also result in similar pathologies in animal models. Data for CMV infection in rats appears to satisfy this criterion. CMV can increase atherosclerotic lesions in the presence of hypercholesterolemia, resembling what is known about low-density lipoprotein (LDL) with atherosclerosis in human [32]. At

the same time, there are in vitro studies that demonstrate the ability of the CMV infection in upregulating the oxidized LDL scavenger receptor CD 36 [33], ICAM-1 [34], VCAM-1 [35], PDGF-β receptor [36], IL-6, and MCP-1 [37]. These molecules are important players in stimulating atherosclerotic lesions and hence this provides important biochemical evidence for the mechanism by which CMV may cause lesions in vivo. Recently, Hillebrands et al. [38] have demonstrated that CMV-infected rats with transplanted aortic allografts demonstrate enhanced atherogenesis if the infection occurs early posttransplantation (1–5 days but not later). This may be relevant to coronary transplant patients who already suffer chronic CMV infections. While this finding supports a role of CMV in stimulating atherogenesis, the authors also cautioned that the conclusion should not be generalized because the effect appeared to be dependent on the combination of host and viral strain [38].

The main weakness in the postulate that CMV and HPV are causative agents in atherosclerosis is the lack of data on treatments that are effective concurrently against both atherosclerosis and the viruses. The presence of studies that fail to find significant correlations between viral replication (HPV and CMV) and plaque development further adds to the controversy. For instance, in the ARIC study which examined 340 patients, it was concluded that vessel wall thickening may not have a significant association with the detection of CMV [39]. In summary, the cases for CMV and HPV as significant atherosclerotic agents are not definitive but do warrant further investigation. Current data seems to suggest that the contribution of these viral infections may be more prominent in certain immune-compromised subjects (e.g., HIV-infected and/or transplant patients). The presence of these viruses in the arterial and atherosclerotic environment of susceptible individuals may further aggravate the already injured endothelial layer with heightened immune responses. This also implies that CMV and HPV infections may not be important atherosclerotic risk factors in the general population. Thus the role of CMV and HPV may resemble more of “environmental” factors that enhance the severity of lesion formation.

Helicobacter pylori

H. pylori (Hpy) is a well-studied gram-negative bacteria. It is now understood that Hpy infection is a major cause of persistent gastritis in adults and children [40, 41]. Given its prevalence and its ability to stimulate inflammatory responses, the possible role of this bacterium in causing atherosclerosis has received significant attention. Several observational studies have demonstrated strong correlations between Hpy infection and coronary heart disease in humans [42, 43]. Hpy infection may enhance atherogenesis in patients with hyperhomocysteinemia [44]. There is an indication that infection by Hpy specifically (and not by CMV nor C. pneumoniae (Cpn)) attenuates coronary vasodilation and induces atherosclerosis [45]. This suggests that Hpy infection may enhance atherogenesis by mechanisms similar to those induced by

418 Paul K.M. Cheung and Grant N. Pierce

hypertension and is acting somewhat differently from CMV or Cpn. Also, there is a recent report that suggests Hpy does not enhance myocardial infarct in lieu with oxidized LDL [46], an observation that again sets it apart from other potential atherogenic agents such as Cpn. A recent clinical investigation of 81 male patients who had an average age 40 years old concluded that Hpy infection is associated with endothelial dysfunction and with elevated levels of C-reactive protein, an inflammatory marker of atherosclerotic disease [47]. There are other similar clinical reports which conclude positive correlations of Hpy with cardiovascular events such as ischemic stroke and coronary heart disease [48–50]. These data provide strong support to the assertion that Hpy is a proatherogenic agent. However, at present, there is a lack of understanding of the biochemical events that demonstrate the causative mechanism between Hpy infection and plaque formation. Various strains of Hpy have been demonstrated to cause endothelial dysfunction and proliferation [51]. Current opinion [50] seems to agree that while this pathogen can elicit significant inflammatory diseases, further studies and in vitro experiments on the molecular effects of the pathogen may help evaluate the value of controlling this pathogen in coronary heart disease patients.

Porphyromonas gingivalis

P. gingivalis (Pgn) and Cpn are probably the most important infectious proatherogenic pathogens known to date. A strong correlation between periodontal disease and atherosclerosis have been reported since 1989 [52–56]. Gingivitis and periodontal diseases are common in the human population and these infections can lead to dental plaque development and destruction of tooth supporting tissues. The usual periodontal pathogens are Pgn,

Actinobacillus actinomycetemcomitans and Bacteroides forsythus. The rDNA of these pathogens have been detected in 18–30% of carotid atheromas [57, 58]. In the National Health and Nutrition Examination Survey (NHANES III, a cross-sectional study) that involves 5564 patients of 40 years and older, after adjustment for age, sex, race, social status, smoking, and diabetes, there was a fourfold increase in the incidence of myocardial infarction in patients with periodontal disease [56]. Further adding weight to this association is the report that cardiovascular risk is related to the severity of periodontal infection [59].

It is particularly noteworthy that longitudinal studies have generated very different conclusions regarding the role of periodontal infections in atherosclerosis. In longitudinal studies, periodontal infection appears weaker (between 1 and 2.5-fold increase in risk) in predicting coronary diseases [60, 61]. Furthermore, two other large longitudinal studies (involving 8000+ patients) failed to find an association, after adjustments for other risk factors, between periodontal disease and atherosclerosis [62, 63]. The apparent discrepancies between cross-sectional and longitudinal studies may reflect the complex multifactorial nature of atherogenesis. The adjustment for other risk

factors that utilize similar inflammatory pathways as do periodontal infections could attenuate, especially over long-term progression of atherosclerosis, the observed significance of periodontal disease. At the same time, it is plausible that the differences between longitudinal and cross-sectional studies may indeed reflect the acute nature of periodontal disease in stimulating cardiovascular incidences. In myocardial infarct and periodontal disease patients, significant inflammatory responses are often observed with high levels of C-reactive proteins and bacteremia [64, 65]. Thus, it is possible that periodontal disease results in atherosclerotic outcomes by causing acute severe immune responses in patients who are already predisposed to atherogenesis in the coronary arteries.

To understand more about the mechanism by which periodontal infection can induce atherosclerotic plaques, two recent investigations studied the molecular mechanisms of lesion aggravation by Pgn. Pgn is receiving special attention over other periodontal disease agents due to a recent report that has identified a serum antibody against Pgn, and not those against A. actinomycetemcomitans, was a strong indicator for coronary heart disease [66]. Lourbakos et al. [67] demonstrated that gingipains (cysteine proteinases) produced by this pathogen are potent coagulation factors. These enzymes increase intracellular calcium and stimulate platelet aggregation at efficiencies similar to that of thrombin. Thus, the presence of Pgn in atheromas such as carotid lesions [57, 58] may increase the likelihood of an arterial blockade. In apo-E null mice, Lalla et al. [68] have demonstrated the acceleration of atherosclerotic plaque formation by Pgn. The infection resulted in an upregulation of IL-6, VCAM-1, and tissue factor (TF) in the animals while a minority of mice (two out of nine) demonstrated the presence of Pgn in the aortic tissue. These findings, combined with other in vitro data that demonstrates the proatherogenic effects of Pgn [69, 70], makes a strong argument for Pgn periodontal disease as a direct inducer of atherosclerotic diseases. However, despite these indirect pieces of evidence implicating the agent in atherogenesis, the causative role of Pgn remains to be proven by one crucial observation: anti-Pgn treatments can reduce atherogenic and cardiovascular events. To this end, therefore, the outcome of the ongoing clinical NIH approved trial of periodontal treatments in several thousand patients with atherosclerotic disease will be an important piece to complete the puzzle that defines the causal relationship between the periodontal infection and atherogenesis (http://www.cscc.unc.edu/pave/).

Chlamydia pneumoniae

In the past 15 years, infection by Cpn and its role in atherosclerosis has been more intensively studied than other infectious agents. This is largely due to findings from various clinical and laboratory studies that strongly implicate its contribution to inflammation, cell proliferation, and plaque formation. However, conflicting reports from animal models and the failure of antibiotics to

420 Paul K.M. Cheung and Grant N. Pierce

reduce atherosclerosis or clinical events associated with the infection have fueled the debate over the atherogenic nature of Cpn. Thus, current opinions remain divided regarding the role of Cpn as a significant agent in human atherosclerosis. In the following sections, we shall provide a concise summary of what is known about the bacteria, its atherogenic effects, and current data that argues for and against its role in plaque formation.

A general serological test for Chlamydia has been available since 1960s (the microimmunofluorescence serologic test) [71]. However, it was not until 1985 that Saikku et al. [72] investigated and distinguished Cpn from Chlamydia psittaci. Subsequent retrospective serology work by Grayston and coworkers [73–78] revealed the presence of this previously undetected species in various outbreaks of pneumonia and acute respiratory disease epidemics from 1950s to 1990s. Thus, Cpn has been an important pathogen in the human population for a considerable time. Cpn is distinguished from the other Chlamydia species in four aspects: (i) Cpn exhibits a unique morphology of a pearshaped inclusion body in a cellular environment [79, 80]; (ii) the Cpn DNA sequence is homologous but discrete from other Chlamydia species [81, 82]; (iii) Cpn serology is highly conserved and only a single serovar has been observed [83–85]; and (iv) Cpn infections cause symptoms in the respiratory system and secondary clinical outcomes are different from other Chlamydia species [86, 87].

During active replication, Cpn resides in membrane-bound vesicles (reticulate bodies) that are metabolically active and susceptible to antibiotic treatments. During unfavorable conditions, the reticulate bodies can relapse into elementary bodies that are resistant to antibiotics but are metabolically inactive. The precise biochemical mechanism through which Cpn persists as a chronic infection in a host cell environment remains unknown. Latency of the bacteria has been observed in both animal and cell culture models [88, 89]. A typical life cycle of Cpn is summarized in Fig. 19.2.

In terms of epidemiology, serological studies have suggested that at age 20, 50% of adults have been infected. Up to 80% of men and 70% of women late in life have been exposed to the pathogen [90, 91]. Considering the nature of a persistent infection and the high frequency of reinfection, it has been suggested that all individuals in a population have been exposed to the pathogen at some point in their life [92].

Symptoms of a Cpn infection are mild and continued coughing and malaise [93]. Treatments with the macrolide class antibiotics (azithromycin, erythromycin, and doxycycline) are effective only if they are administered early postinfection or over a long period of time [94]. Chronic infection is a possible outcome and the pathogen can disseminate to various vital organs via infections of monocytes and macrophages [95, 96]. Secondary complications include sinusitis, pneumonia, myocarditis, and endocarditis [87, 97–99].

The postulate that Cpn infection can cause atherosclerosis is supported by findings in three major areas: Clinical associations, in vitro molecular studies and in vivo infection of animal models. In 1988, Saikku et al. [100] identified

Host cell

Binary Fission/ active metabolism

Dormant elementary bodies resistant to antibiotics inactive metabolism

Exocytosis

Elementary body synthesis

FIGURE19.2. The life cycle of a Chlamydia pneumonia infection. Elementary bodies enter a host cell by endocytosis. The bacteria multiply by binary fission and new reticulate bodies are formed. Reticulate bodies are active, and rely on nutrients and ATP extracted from the host cytoplasm for survival. Under unfavorable conditions, the bacteria may relapse into inactive elementary bodies, which can be resistant to antibiotics. Infection spreads by exocytosis of progeny elementary bodies to neighboring cells.

a significantly higher amount of Cpn antibody in the serum of patients with coronary artery disease. Subsequent work by Thom and coworkers [101–106] have demonstrated strong correlations between the detection of Cpn and coronary artery disease, with a 2–7-fold increase in risk. The presence of Cpn in coronary arterial tissues has been confirmed by electron microscopy, immunocytochemistry, and PCR techniques. Of note, Cpn has been found not only in fatty streak tissues (early stage atherosclerosis) and fibrous plaques (late stage), but also in macrophages and smooth muscle cells [107–109], both of which are important cell types in the composition of plaques. Recovery of cultivable live pathogen from atherosclerotic lesion was performed in 1996 [110]. Cpn infection is more likely to occur in plaques of patients who have undergone balloon angioplasty or stent operations than in native arterial lesions [111]. Unlike findings in the CMV-related atherosclerosis, Cpn is rarely found in nonatherosclerotic tissues [112–114]. Furthermore, Cpn antibody titers from patients strongly correlate with the occurrence of