J Dent Res 85(2):106-121, 2006
© 2006 International and American Associations for Dental Research
REVIEW
CRITICAL REVIEWS IN ORAL BIOLOGY & MEDICINE |
Innate Immune Signaling and Porphyromonas gingivalis-accelerated Atherosclerosis
F.C. Gibson, III1,
H. Yumoto1,2,
Y. Takahashi1,3,
H.-H. Chou1,4, and
C.A. Genco1,5,6,*
1 Department of Medicine, Section of Infectious Diseases, and
6 Department of Microbiology, Boston University School of Medicine, Evans Biomedical Research Center, 650 Albany Street, Room 637, Boston, MA 02118, USA;
2 Department of Conservative Dentistry, Tokushima University School of Dentistry, Tokushima, Japan;
3 Department of Oral Microbiology, Kanagawa Dental College, 82 Inaokoa-cho, Yokosuka 238-8580, Japan;
4 School of Dentistry, College of Oral Medicine, Taipei Medical University, Taipei, Taiwan; and
5 Department of Periodontology and Oral Biology, Goldman School of Dental Medicine, Boston University Medical Center, Boston, MA, USA
* corresponding author, caroline.genco{at}bmc.org
 |
ABSTRACT
|
|---|
Periodontal diseases are a group of diseases that lead to erosion of the hard and soft tissues of the periodontium, which, in severe cases, can result in tooth loss. Anecdotal clinical observations have suggested that poor oral health may be associated with poor systemic health; however, only recently have appropriate epidemiological studies been initiated, with defined clinical endpoints of periodontal disease, to address the association of periodontal disease with increased risk for cardiovascular and cerebrovascular disease. Although conflicting reports exist, these epidemiological studies support this connection. Paralleling these epidemiological studies, emerging basic scientific studies also support that infection may represent a risk factor for atherosclerosis. With P. gingivalis as a model pathogen, in vitro studies support that this organism can activate host innate immune responses associated with atherosclerosis, and in vivo studies demonstrate that this organism can accelerate atheroma deposition in animal models. In this review, we focus primarily on the basic scientific studies performed to date which support that infection with bacteria, most notably P. gingivalis, accelerates atherosclerosis. Furthermore, we attempt to bring together these studies to provide an up-to-date framework of emerging theories into the mechanisms underlying periodontal disease and increased risk for atherosclerosis, as well as identify intervention strategies to reduce the incidence of periodontal disease in humans, in an attempt to decrease risk for systemic complications of periodontal disease such as atherosclerotic cardiovascular disease.
KEY WORDS: Porphyromonas gingivalis • periodontal disease • Toll-like receptors • endothelium • innate immunity
Abbreviations: ApoE = apolipoprotein E • CAM = cell adhesion molecule • CMV = cytomegalovirus • HAEC = human aortic endothelial cell • HSP = heat-shock protein • HUVEC = human umbilical vein endothelial cell • ICAM-1 = intracellular adhesion molecule-1 • IFN-
= interferon- • IgA = immunoglobulin A • IL = interleukin • LPS = lipopoly-saccharide • LDLR = low-density lipoprotein receptor • MCP-1 = macrophage chemotactic protein-1 • MOI = multiplicity of infection • PAMPs = pathogen-associated microbial products • PGE2 = prostaglandin E2 • RANTES = Regulated on Activation, Normal T Expressed and Secreted • SMC = smooth-muscle cells • TLR = toll-like receptor • TNF-
= tumor necrosis factor- • VCAM-1 = vascular cell adhesion molecule-1
|
INTRODUCTION
|
|---|
Periodontal disease, a complex chronic inflammatory disease that affects the periodontium, is initiated by bacteria, with incipient erosion of the attachment apparatus of the tooth. It has been reported that more than 100 million people in the US may possess measurable periodontal bone loss, making this disease one of the most common chronic infectious diseases of humans (Slade and Beck, 1999). Despite identification of over 500 different bacterial species in the oral cavity, only a relative few organisms are linked to periodontal disease; however, Porphyromonas gingivalis is the most common organism linked to adult forms of periodontal disease. Over the past 10 years, mounting evidence has accumulated supporting a role for periodontal disease and infection, with P. gingivalis as a potential risk factor for several systemic diseases, including diabetes, pre-term birth, heart disease, and atherosclerosis (Genco, 1996; Morrison et al., 1999; Dasanayake et al., 2003).
Atherosclerosis, formerly considered a condition associated with elevated circulating lipids with arterial vessel lipid accumulation, actually involves an ongoing inflammatory response. In humans, the inflammatory reactions within coronary atherosclerotic plaques are increasingly thought to be crucial determinants of the clinical course of patients with coronary artery diseases. Since numerous reviews exist detailing the development of atherosclerosis (Stary et al., 1994, 1995; Ross, 1999; Libby et al., 2002), we will not discuss the mechanisms underlying the development of atherosclerosis, other than those potentially relevant to infection-associated atherosclerosis. In various animal models of atherosclerosis, inflammation occurs simultaneously with incipient lipid accumulation in the artery wall. The stimuli that initiate and sustain the inflammatory process, however, have not been fully identified. Modified lipoproteins and local or distant infections have been proposed to contribute to the inflammatory process in atherosclerosis (Epstein et al., 1999a; Espinola-Klein et al., 2002; Libby et al., 2002). Evidence in humans suggesting that infection with P. gingivalis and periodontal disease predisposes to atherosclerosis is derived from studies demonstrating that periodontal disease pathogens reside in the walls of atherosclerotic vessels, and from sero-epidemiological studies demonstrating an association between the pathogen-specific antibodies and atherosclerosis.
The goal of this review is to discuss specifically the association between periodontal disease and the increased risk for heart disease and atherosclerosis, with particular emphasis on the putative mechanisms linking P. gingivalis infection to the acceleration of atherosclerosis. Using P. gingivalis as a model organism to study pathogen-accelerated atherosclerosis, we have demonstrated, in recent studies, that invasive bacteria are required for the acceleration of atherosclerosis, and that there is an innate immune response directed to invasive P. gingivalis, as demonstrated by the increased expression of innate immune receptors in the aorta of hyperlipidemic mice following oral challenge with P. gingivalis. We have also demonstrated that invasive P. gingivalis up-regulates the expression of chemokines and cell adhesion molecules in endothelial cells and macrophages in vitro. Furthermore, invasive P. gingivalis also up-regulates the expression of innate immune receptors on the surfaces of both endothelial cells and macrophages. Collectively, these studies have begun to define the role of specific innate immune signaling molecules in response to invasive bacterial infection, and to correlate these responses to putative mechanisms involved in microbial accelerated atherosclerosis.
|
PERIODONTAL DISEASE AND P. gingivalis
|
|---|
Periodontal diseases comprise a group of inflammatory diseases of the gingiva and supporting structures of the periodontium. P. gingivalis, a Gram-negative anaerobe, has been long considered an important pathogen associated with human periodontal disease. Currently, the underlying theme of understanding is that this organism, along with the host immune response, is critical to the destruction of the supporting structures of the teeth (Holt et al., 1988). During periodontal health, the tissues adjacent to and underneath the gingival epithelium commonly possess a modest accumulation of neutrophils that are believed to be important in clearing any transient bacteria that gain access to these tissues. However, in patients in the acute or active stage of periodontal disease, the periodontium presents with a neutrophilic cellular infiltrate that switches to a predominating monocytic and lymphocytic cellular infiltrate in chronic lesions. An understanding of the complex cellular interactions that occur during periodontal disease is critical to the definition of a mechanistic approach to the determination of the bacterial factors that might be responsible for this response. Gingival crevicular fluid obtained from periodontal disease sites have high levels of IL-1ß, IL-8, and IL-10 and the chemokine RANTES (Gamonal et al., 2000), plus IL-6, transforming growth factor, PGE2, IL-2, TNF-, and interferon (IFN)- (Salvi et al., 1998). The mechanisms by which P. gingivalis stimulate cytokine and chemokine production are not well-known, but recent in vitro studies have been performed with P. gingivalis as well as with purified antigens from this organism. Results from these studies have collectively determined that the host cell type, the number of bacteria, or the amount of antigen being tested is critical to the reported observations. While a large number of different bacterial species exist in the oral cavity, it is now recognized that large numbers of bacteria (bacterial load) do not necessarily result in the biological progression from health to periodontal disease. Rather, the establishment and growth of a very few bacterial species from among those resident in the subgingival niche are apparently periodontopathic (Holt et al., 1999), including Fusobacterium nucleatum, Bacteroides forsythus, Prevotella intermedia, Treponema denticola, and P. gingivalis. P. gingivalis is essentially absent during periodontal health but, during disease progression to periodontitis, can reach a very significant percentage of the pathogenic microbiota.
Oral epithelium—more specifically, the sulcular and junctional epithelium—represents one of the initial host barriers to P. gingivalis when this organism is present in the gingival sulcus. Several studies have begun to characterize the host response of oral epithelial cells to P. gingivalis. Challenge of oral epithelial cells with P. gingivalis elicits a TNF- and IL-1ß response. Additional studies have demonstrated that these cells also express cell adhesion molecules on their surface in response to P. gingivalis antigens and include ICAM-1 and VCAM-1 (Wang et al., 1999). Interestingly, work by Darveau et al.(1998) has demonstrated that gingival epithelial cells challenged with P. gingivalis LPS fail to produce IL-8; furthermore, P. gingivalis LPS stimulation functions as a potent inhibitor of subsequent E. coli LPS stimulation of IL-8. In addition to these antagonistic properties, P. gingivalis infection of gingival epithelial cells can inhibit IL-8 production in response to other bacteria present in dental plaque, including F. nucleatum (Darveau et al., 1998). This mechanism has been termed ’localized chemokine paralysis’. The nature of this inhibition is not well-understood, but may relate to the processing of P. gingivalis LPS by gingival epithelial cells. Thus, along with the ability of P. gingivalis cysteine proteinases (gingipains) to digest cytokines and chemokines, the antagonistic properties of P. gingivalis LPS and localized chemokine paralysis are all attractive mechanisms by which P. gingivalis may be able to circumvent the host response.
|
BIOLOGY OF ATHEROSCLEROSIS
|
|---|
Recent advances have established a fundamental role for inflammation in mediating all stages of atherosclerosis, from initiation through progression and, ultimately, the thrombotic complications of atherosclerosis (Libby et al., 2002). Blood leukocytes, mediators of host defense and inflammation, localize in the earliest lesions of atherosclerosis in both experimental animal models and in humans. The normal endothelium does not generally support binding of white blood cells. However, it has been demonstrated, in animal models, that early after initiation of an atherogenic diet, portions of arterial endothelial cells begin to express, on their surface, selective adhesion molecules that bind to various classes of leukocytes. Once adherent to the endothelium, the leukocytes penetrate the intima. Following residency in the arterial wall, the blood-derived inflammatory cells participate in and perpetuate a local inflammatory response. Inflammatory processes not only promote initiation and evolution of atheroma, but also contribute decisively to precipitating acute thrombotic complications of atheroma (Libby et al., 2002).
Injury to the vessel wall and the associated inflammatory response to injury are now generally recognized as the essential components of atherogenesis. The triggers that initiate and sustain the inflammatory process, however, have not been definitively identified. Among the candidate triggers are oxidized LDL (ox-LDL) and heat-shock proteins (HSPs). These components of the atheroma are believed by some investigators to elicit an inflammatory response (Epstein et al., 1999a). Patients with cardiovascular disease can develop antibodies to ox-LDL and HSPs, and, some studies, although controversial, suggest that these antibodies may play a role in causing auto-immune-induced damage to the vessel wall (Epstein et al., 2000). Another candidate trigger of both inflammatory and auto-immune responses leading to the initiation and/or acceleration of atherosclerosis is infection.
|
INFECTION AND ATHEROSCLEROSIS
|
|---|
As many as 50% of patients with atherosclerosis lack currently identified risk factors, an observation indicting that additional factors predisposing humans to atherosclerosis are as yet undetected (Vita and Loscalzo, 2002). Inflammation in the arterial vessel wall is considered to play an important role in the pathogenesis of atherosclerosis (Ross, 1999; Libby et al., 2002). Likewise, in a variety of animal models of atherosclerosis, signs of inflammation occur hand-in-hand with lipid accumulation in the artery wall. Modified lipoproteins and local or distant infections have been proposed to contribute to the inflammatory process in atherosclerosis (Epstein et al., 1999a; Espinola-Klein et al., 2002; Libby et al., 2002). Accumulating evidence has implicated specific infectious agents—including cytomegalovirus (CMV), C. pneumoniae, H. pylori, Herpes simplex virus types 1 and 2, T. cruzi, and P. gingivalis—in the progression of atherosclerosis. The observed association of infection and atherosclerosis is based primarily on epidemiological data, as well as on emerging experimental studies where animal models with defined microbial pathogens were used (Beck et al., 1998; Epstein et al., 1999b; Liuba et al., 2003; Spence and Norris, 2003; Gibson et al., 2004). Several of these micro-organisms are found in atherosclerotic lesions and can aggravate atherosclerosis in experimental models (Beck et al., 1996; Epstein et al., 1999b; Scannapieco and Genco, 1999; Liu et al., 2000; Rothstein et al., 2001; Li et al., 2002; Lalla et al., 2003; Gibson et al., 2004).
|
Chlamydia pneumoniae, CYTOMEGALOVIRUS (CMV), AND Helicobacter pylori
|
|---|
C. pneumoniae has been implicated, by serological and pathological studies, in the pathogenesis of coronary artery disease. C. pneumoniae infection might contribute to early atherogenesis, which might be associated with chronic inflammation and atherosclerosis at an early stage, even before clinical events occur (Tasaki et al., 2003). Several epidemiological studies have also reported on a possible association of various forms of vascular disease with the presence and titer of CMV antibodies (Grattan et al., 1989; Blum et al., 1998; Zhou et al., 1999; Lunardi et al., 2000). Other studies show the presence of the virus, viral antigens, or nucleic acid in atherosclerotic lesions (Melnick et al., 1983). Studies in animal models and cell-culture studies present attractive mechanisms by which CMV may play a role in atherogenesis. Several epidemiological studies have also reported on the association of H. pylori antibody titers and risk for coronary heart disease and stroke (Kahan et al., 2000; Grau et al., 2001). One recent study suggests that virulent strains of H. pylori may induce systemic inflammation, whereas avirulent strains do not and thus would not be associated with accelerated atherosclerosis (Pietroiusti et al., 2002).
|
MULTIPLE PATHOGENS/INFECTIONS
|
|---|
In addition to individual pathogens, the extent of atherosclerosis and the prognosis of patients with atherosclerosis also seem to be correlated with the number of infections to which an individual has been exposed. In a prospective study, the effects of 8 pathogens and the aggregate pathogen burden on the progression of carotid atherosclerosis were evaluated in 504 patients (Espinola-Klein et al., 2002). Elevated IgA antibodies against C. pneumoniae and IgG antibodies to Epstein-Barr virus (EBV) and HSV were associated with the progression of atherosclerosis. Infectious burden was also significantly associated with progression of atherosclerosis. The authors concluded, from this study, that the number of infectious pathogens to which an individual has been exposed contributes to the progression of carotid atherosclerosis (Espinola-Klein et al., 2002). A more recent study also demonstrated a strong association between pathogen burden and cardiovascular disease, independent of classic risk factors (Georges et al., 2003). These authors suggested that the pathogen burden could also be a predictor of coronary complications.
|
SPECIFICITY OF PATHOGEN STIMULATION TO THE INDUCTION OF ATHEROSCLEROSIS
|
|---|
Screening of human atheroma has revealed that several pathogens are frequently detected in these tissues. Interestingly, the majority of organisms reported to be in these atherosclerotic plaques are responsible not for acute infections, but rather for chronic infections. Despite these findings, limited work has been performed to address the specificity of pathogen exposure to the development of atherosclerosis. Using hyperlipidemic mice, Hu et al. observed that C. pneumoniae strain AR39 and C. trachomatis strain MoPn displayed differences in ability to accelerate atheroma deposition, since C. pneumonia AR39, but not C. trachomatis MoPn, accelerated atherosclerosis (Hu et al., 1999). Interestingly, despite differences in abilities to stimulate atherosclerotic plaque deposition, both strains were detected in the aorta of challenged mice (Hu et al., 1999). This seminal work supports the hypothesis that the host response to infection is insufficient to accelerate atherosclerosis, but that there is specificity to infection-accelerated atherosclerosis.
|
P. gingivalis INFECTION AND ATHEROSCLEROSIS
|
|---|
There is evidence to support the relationship between human periodontal disease and an increased risk for acute myocardial infarction (Beck et al., 1996; Scannapieco and Genco, 1999; Haraszthy et al., 2000). Case-control studies have demonstrated a significant correlation between cardiovascular disease and periodontal disease after adjustment for cholesterol, smoking, hypertension, social class, and body mass index (Beck et al., 1996; Scannapieco and Genco, 1999; Wu et al., 2000). While some reports have not confirmed this association, this may be due to the fact that self-reported periodontal disease was used for the analysis of the patient population (Howell et al., 2001). Several bacteria associated with periodontal disease have been detected in atherosclerotic plaque (Haraszthy et al., 2000). The primary bacterium associated with adult periodontal disease, P. gingivalis, has also been identified, by PCR and fluorescence in situ hybridization, in atheromatous plaques of two patients suffering from atherosclerosis, suggesting that these micro-organisms might be metabolically active within the atherosclerotic lesions (Cavrini et al., 2005). This is the first report supporting that metabolically active oral pathogens are present in human atheroma; however, confirmation of this observation in a larger set of patients is required, and, ultimately, bacterial culture will need to be demonstrated. Finally, serum antibodies to P. gingivalis and A. actinomycetemcomitans have also been associated with coronary heart disease (Pussinen et al., 2003).
It has been suggested that periodontal disease can lead to low-level bacteremia, an elevated white cell count, and systemic endotoxemias, which could affect endothelial integrity, metabolism of plasma lipoproteins, blood coagulation, and platelet function. Furthermore, it is well-established that infection with P. gingivalis induces local inflammation. The induction of this inflammatory response can lead to gingival ulceration and local vascular changes, which have the potential to increase the incidence and severity of transient bacteremias. Several studies have also demonstrated that patients with periodontal disease have elevated levels of systemic inflammatory mediators. Extensive periodontal disease has been associated with increased levels of C-reactive protein (CRP) (Slade et al., 2003). Moderately elevated CRP is a systemic marker of inflammation and a documented risk factor for cardiovascular disease (Teragawa et al., 2004). In a separate study, control of periodontal disease achieved with non-surgical periodontal therapy significantly decreased serum IL-6 and CRP levels (D’Aiuto et al., 2004). Recent experimental studies support that local inflammation within the artery wall may also contribute to the acceleration of atherosclerosis in response to infection with P. gingivalis. Indeed, the focus of current studies is to define the precise molecular mechanisms by which P. gingivalis infection contributes to the progression of atherosclerosis, and the links among lipids and innate immune and inflammatory responses.
|
INNATE IMMUNITY AND ATHEROSCLEROSIS
|
|---|
Innate immunity consists of the inherent immune mechanisms of a host to prevent or control an infectious challenge and is typically characterized by the activation and recruitment of monocytes and macrophages. Classically, the innate immune system was defined, in part, by cell adhesion molecules and cytokine and chemokine expression, as well as by activation of the complement system and other receptors (Janeway and Medzhitov, 2002). Activation of monocytes and macrophages and other cells, such as endothelial cells, represents an important initial step in the cascade of events leading to host response in acute and chronic inflammatory diseases. Furthermore, the inflammatory nature of cardiovascular disease and the development of atherosclerosis is well-established (Ross, 1999). The immune system is capable of making qualitatively distinct responses to different microbial infections, and recent advances are starting to reveal how it manages this complex task. Genome-encoded innate immune systems target structurally conserved pathogen-associated microbial products (PAMPs), thereby allowing immediate and, in most cases, sufficient responses to eliminate invading pathogens (Vasselon and Detmers, 2002). The recently described Toll-like receptors (TLRs) recognize a specific set of PAMPs and appear to play a key role in detecting microbes and initiating inflammatory responses (Takeda and Akira, 2005). Ligation of these receptors initiates signal transduction pathways that lead to activation of nuclear factor-
B (NF-B) and subsequent expression of a wide array of inflammatory genes (Imler and Hoffmann, 2001; Underhill and Ozinsky, 2002). The best-studied of the TLRs are TLR-2 and TLR-4. As the receptor for Gram-negative enterobacterial LPS, TLR-4 is the best-characterized member of the TLR family. Enterobacterial LPS is bound in serum by LPS-binding protein (LBP), which delivers LPS to CD14, a protein that exists both in soluble form and as a glycosylphosphatidylinositol (GPI)-linked outer membrane protein (Fenton and Golenbock, 1998). CD14 physically associates with a complex including TLR-4 and an extracellular accessory protein, MD-2. Each component of this complex is required for efficient LPS-induced signaling. Several other bacterial proteins have been suggested to activate immune cells through TLR-4, including teichuronic acid, fimbriae, and HSP60 from both human and microbial sources (Beutler, 2002; Hajishengallis et al., 2002; Underhill and Ozinsky, 2002).
An array of molecules has been reported to activate innate immune responses through TLR-2, including bacterial lipopeptides, peptidoglycan, and zymosan (Asai et al., 2001; Henneke et al., 2001; Thoma-Uszynski et al., 2001; Massari et al., 2002). In addition, certain structural variants of LPS are detected by TLR-2, including LPS from P. gingivalis (Bainbridge and Darveau, 2001; Hirschfeld et al., 2001; Pulendran et al., 2001; Kusumoto et al., 2004; Zhou et al., 2005). Some of the broad ligand specificity attributed to TLR-2 may be accounted for by the observation that TLR-2 must dimerize with other TLRs to detect ligands and induce signaling. Whereas several microbial products appear capable of stimulating inflammatory responses through TLR-2 and TLR-4, to date, only single microbial targets have been identified for TLR-3, TLR-5, and TLR-9. TLR-5 recognizes bacterial flagellin. Recently, TLR-9 was identified as the receptor for non-methylated CpG DNA. TLR-3 recognizes double-stranded RNA.
|
TOLL-LIKE RECEPTORS AND ATHEROSCLEROSIS
|
|---|
There is increasing evidence for differential responses in cells activated with different TLR stimuli, indicating that the repertoire of TLRs that detect a pathogen may coordinate a response tailored for defense against a class of organism (Muzio et al., 2000a,b; Visintin et al., 2001). Importantly, recent reports indicate that expression of TLRs is enhanced in atherosclerotic lesions. Dybdahl et al.(2002) reported on the expression of monocyte TLR-2 and TLR-4 following coronary artery bypass grafting in humans. Likewise, Frantz et al.(1999) recently found that cardiac myocytes constitutively express TLR-4, and that this expression is up-regulated in the hearts of humans with cardiovascular disease. TLR-1, TLR-2, and TLR-4 have been reported to be markedly augmented in human atherosclerotic lesions, and expression occurred preferentially by endothelial cells and macrophages (Edfeldt et al., 2002). Another report has demonstrated TLR-2 expression within atherosclerotic plaques in humans (Laman et al., 2002). Xu et al.(2001) recently demonstrated that TLR-4 is expressed in lipid-rich, macrophage-infiltrated atherosclerotic lesions of mice and humans, and that TLR-4 mRNA in cultured macrophages is up-regulated by ox-LDL, but not by native LDL. These findings raise the possibility that enhanced TLR expression may play a role in inflammation in atherosclerosis. Furthermore, the findings of increased expression of TLRs, specifically TLR-4 induced by ox-LDL, suggest a potential mechanism for the synergistic effects of hypercholesterolemia and infection in the acceleration of atherosclerosis observed in experimental models and human epidemiological observations.
A recent study suggests that polymorphisms in the human TLR-4 gene, which attenuates receptor signaling and diminishes the inflammatory response to Gram-negative pathogens, are associated with low levels of certain circulating mediators of inflammation and a decreased risk for atherosclerosis in humans (Kiechl et al., 2002). Likewise, a separate study found that the TLR-4 polymorphism was associated with the risk of cardiovascular events among men with documented coronary artery disease (Boekholdt et al., 2003). Another study reported that this TLR-4 polymorphism does not influence the predisposition for and progression of coronary artery disease (Yang et al., 2003). As discussed by the authors of this study, however, the lack of association does not exclude the involvement of the TLR-4 gene in atherosclerotic plaque formation or plaque stability, as opposed to vessel stenosis. Results from epidemiological studies illustrate how important it is to perform well-controlled animal studies to address the specific involvement of defined innate signaling molecules in the progression of atherosclerosis.
|
ANIMAL MODELS FOR THE STUDY OF ATHEROMA PROGRESSION
|
|---|
Numerous animal models—including those developed for the rat (Herrera et al., 2003), the rabbit (Jain et al., 2003), and mice (Paigen et al., 1987)—have been used for the study of the mechanisms that underlie the progression of atherosclerosis. The studies performed to date include investigation of classically defined risk factors, such as high dietary fat consumption (Lutgens et al., 1999), as well as the impact of innate immune responses in this process (Boring et al., 1998; Collins et al., 2000; Bjorkbacka et al., 2004; Michelsen et al., 2004). The two most widely utilized murine models for the study of atherosclerosis are the apolipoprotein E knockout (ApoE–/–) mice and, to a lesser extent, the low-density lipoprotein receptor knockout (LDLR–/–) mouse (Plump et al., 1992; Ishibashi et al., 1993). Utilization of these animal models for study of the progression of pathogen-accelerated atherosclerosis has been pivotal for the dissection of the mechanisms underlying specific pathogens, the infections they cause, and their effect on the acceleration of atheroma and has been proven useful for several organisms, including P. gingivalis (Epstein et al., 1999b; Moazed et al., 1999; Li et al., 2002; Lalla et al., 2003; Gibson et al., 2004; Spence and Norris, 2003). Additionally, breeding of these genetically hyperlipidemic mice with other defined knockout mice has previously demonstrated functional roles of innate immune molecules in the acceleration of atherosclerosis (Gupta et al., 1997; Bourdillon et al., 2000; Collins et al., 2000; Kirii et al., 2003; Branen et al., 2004), and is an excellent strategy to continue to unravel the mechanisms underlying pathogen-accelerated atherosclerosis.
Emerging animal studies with defined genetic mutants are beginning to demonstrate the importance of TLRs and TLR adapter molecules in the progression of atherosclerosis. Bjorkbacka et al.(2004) reported that MyD88, one of several known adapter molecules for TLR-mediated signal transduction, plays a significant role in atherosclerotic plaque deposition in ApoE-deficient mice. In these studies, MyD88/ApoE double-knockout mice developed ~ 50% less atherosclerotic plaque as compared with ApoE-deficient mice placed on similar high-fat diets (Bjorkbacka et al., 2004). The importance of MyD88 in atherosclerosis was confirmed by Michelsen et al.(2004), and these observations were extended by demonstrations that, in addition to MyD88, TLR4/ApoE double-knockout mice developed significantly less plaque as compared with ApoE-deficient mice. These studies experimentally link the TLR-specific innate immune response (specifically TLR4) and the TLR adapter molecule MyD88 in the progression of atherosclerosis. Future studies will be needed to determine if these observations translate to the progression of atherosclerosis in humans.
|
INFECTION WITH P. gingivalis ACCELERATES ATHEROSCLEROSIS IN AN ApoE MOUSE MODEL
|
|---|
As discussed above, studies performed in humans have recently demonstrated that carotid atheromatous tissue obtained from patients with both periodontal disease and cardiovascular disease possesses P. gingivalis-specific DNA (Haraszthy et al., 2000) and rRNA (Cavrini et al., 2005). These studies suggest that P. gingivalis present in the oral cavity gained access to the vasculature, and that either a bacteremia, or a bacteremia followed by invasion of the vascular endothelium, was responsible for localizing P. gingivalis at this site. These initial studies have been supported by recent studies in an ApoE mouse model of atherosclerosis. Li et al.(2002) initially reported that mice heterozygous at the apoE allele and injected with P. gingivalis via the tail vein were susceptible to accelerated atherosclerosis. Lalla et al.(2003) reported that mice homozygous at ApoE–/–, and were orally challenged with P. gingivalis, were susceptible to accelerated atherosclerosis. Our group has recently demonstrated that only invasive P. gingivalis can accelerate atherosclerosis in the ApoE–/– model (Gibson et al., 2004). In these studies, we observed that ApoE–/–mice challenged with an invasive fimbriate P. gingivalis strain exhibited significantly more atherosclerotic plaque on the intimal surface of the aortic arch, as compared with unchallenged ApoE–/– mice, or with mice challenged with a non-invasive non-fimbriate P. gingivalis mutant (Gibson et al., 2004). Interestingly, we found that oral infection with both the wild-type and the non-invasive mutant resulted in bacteremia and localization in the aortic tissue; however, only the invasive P. gingivalis strain was demonstrated to induce the up-regulation of the innate immune receptors TLR2 and TLR4 (Gibson et al., 2004). Importantly, we also demonstrated that atherosclerosis and innate immune activation are detectable shortly after bacterial infection, and these can be significantly prevented by immunization (Gibson et al., 2004). Taken together, these results indicate that early innate immune activation locally in the aortic arch, in response to infectious challenge, is associated with pathogen-accelerated atherosclerosis and suggest that immunization rather than antibiotic therapy may be required to control pathogen-accelerated atherosclerosis.
|
PUTATIVE MODELS BY WHICH PATHOGENS ACCELERATE ATHEROSCLEROSIS
|
|---|
Results from animal studies linking infection and inflammation to atherosclerosis have enabled investigators to begin to examine putative models for pathogen-accelerated atherosclerosis. While the hypothesis that infection may be associated with atherosclerosis is not novel, having been suggested over 100 years ago by Hektoen (1896), only recently have detailed studies been initiated to test it. To date, four working models have emerged to link mechanisms governing pathogen-accelerated atherosclerosis: (1) direct invasion of the vascular endothelium, (2) immunological sounding, (3) pathogen trafficking, and (4) auto-immunity (Fig. 1
). It is important to note that one or a combination of these different mechanisms may collectively contribute to pathogen-accelerated atherosclerois.
Direct invasion of and/or replication of a pathogen within endothelial cells has been demonstrated for several organisms, including C. pneumoniae, CMV, and P. gingivalis (Span et al., 1989; Kaukoranta-Tolvanen et al., 1994; Deshpande et al., 1998b) (Fig. 2). Invasion strategies of microbes vary; however, bacterial attachment via fimbriae appears to play a primary role in the subsequent infection of cells, including cells of endothelial origin (Deshpande et al., 1998a). Previously, our group has demonstrated that fimbriae play a central role in the attachment and invasion of endothelial cells by P. gingivalis. Invasive wild-type P. gingivalis readily adhere to and invade endothelial cells (Deshpande et al., 1998b; Progulske-Fox et al., 1999), and invasion stimulates endothelial cells to express elevated levels of cell adhesion molecules and chemokines (Khlgatian et al., 2002; Nassar et al., 2002). Toward this end, a major fimbriae-deficient P. gingivalis mutant (FimA-) failed to adhere to and subsequently invade endothelial cells, and did not elicit cell adhesion molecule or chemokine expression (Khlgatian et al., 2002; Nassar et al., 2002). These results demonstrate that P. gingivalis invasion of endothelial cells is dependent, at least in part, on fimbriae expression, and that endothelial cells challenged with invasive P. gingivalis express markers consistent with activated endothelium present in atheroma. Recently, it has been demonstrated that patients with periodontal disease are at greater risk for developing vascular dysfunction (Amar et al., 2003). Although, in this study, there was no attempt to culture bacteria from atheromatous tissues, when these in vivo observations are combined with in vitro data demonstrating P. gingivalis infection of endothelial cells, it is intriguing to extend these data to suggest that oral pathogens might directly affect vascular endothelium in a manner consistent with increased cardiovascular risk.
View larger version (33K):
[in this window]
[in a new window]
|
Figure 2. Direct invasion of vascular endothelium. (A) In this model, invasion of the vascular endothelium by pathogenic bacteria such as P. gingivalis (red circles) results in the induction of a local inflammatory response, defined by the expression of cell adhesion molecules (CAMs; green trapezoid), Toll-like receptor (TLRs; blue triangle), chemokines, and cytokines. These inflammatory molecules have all demonstrated significant roles in the initiation and/or acceleration of atherosclerosis. The ability of P. gingivalis to stimulate host endothelial cell activation, both in vitro and in vivo, is a function of surface-expressed major fimbriae. P. gingivalis that does not possess fimbriae (fimA) fails to enter endothelial cells efficiently, whereas those organisms that possess fimbriae (wild-type, WT) readily enter these cells (Deshpande et al., 1998b; Khlgatian et al., 2002; Nassar et al., 2002). Following uptake, P. gingivalis-infected endothelial cells, possibly via a receptor-mediated signaling event, activate gene transcription and stimulate these cells to produce a variety of innate immune markers, including CAMs (ICAM-1, VCAM-1), TLRs (TLR-2, TLR-4), pro-inflammatory cytokines (TNF-, IL-1ß), and chemokines (MCP-1 and IL-8). These mediators are believed to be involved in the immunological switch of endothelial cells from a normal anti-thrombotic to a pro-thrombotic state. (B) Following P. gingivalis invasion/activation of vascular endothelial cells, these cells recruit monocytes, and, in the presence of elevated circulating lipids such as ox-LDL, atheroma forms.
|
|
The second hypothesis by which pathogens may accelerate atherosclerosis is immunological sounding (Epstein et al., 1999a) (Fig. 3). To the best of our knowledge, there is no direct clinical or experimental evidence linking immunological sounding to the acceleration of atherosclerosis. Thus, this is a hypothetical mechanism by which the inflammatory response to extravascular infection could exacerbate vascular inflammation via secreted cytokines and/or chemokines that ultimately modulate atherosclerosis. In this model, local disease, such as Chlamydia pneumonia, or periodontal disease signals systemic changes in the host inflammatory response via molecules secreted from the site of infection, including acute phase mediators, cytokines, and chemokines. Indeed, it is well-established that atherosclerosis is an inflammatory disease (Ross, 1999), and it is feasible that persistent local infections, such as those in the oral cavity, could promote atherosclerosis via chronic up-regulation of inflammatory cascades (Epstein et al., 1999a). Oral infection with P. gingivalis and the subsequent host response to this organism in the oral cavity are known both to present with inflammatory mediators, as discussed above, and to include C-reactive protein (Noack et al., 2001), IL-1ß (Engebretson et al., 2002), TNF- (Roberts et al., 1997), and IL-6 (Geivelis et al., 1993), as well as a vast array of chemokines, including IL-8 (Tsai et al., 1995) and MCP-1 (Hanazawa et al., 1993). Furthermore, with high-fat dietary models of atherosclerotic plaque development/progression in the absence of infection, it has been well-documented that animals deficient in IL-1ß (Kirii et al., 2003), INF- (Gupta et al., 1997), and chemokines such as MCP-1 (Gosling et al., 1999) deposit less atherosclerotic plaque compared with wild-type animals.
View larger version (34K):
[in this window]
[in a new window]
|
Figure 3. Infection-induced stimulation of accelerated atherosclerosis by immunological sounding. Persistent local infection, such as oral infections by P. gingivalis, may promote atherosclerosis via chronic up-regulation of inflammatory cascades involving TNF-, IL-1, IFN, IL-8, MCP-1, and CRP. These cytokines, chemokines, and acute phase mediators could be shed into the vasculature from a focus of P. gingivalis infection in the periodontium. Once in the circulation, these mediators may subsequently activate vascular endothelial cells in a manner that shifts them from a normally anti-thrombotic state to one expressing high levels of inflammatory mediators, including CAMs (blue triangles) and TLRs (green trapezoid), that become pro-thrombotic. This activated endothelium would likely be a site for subsequent atheroma formation, independent of direct pathogen involvement at this site. This further immunological activation results in the recruitment of monocytes, as well as the stimulation, migration, and proliferation of smooth-muscle cells that, together with elevated levels of circulating lipids such as ox-LDL, ultimately results in acceleration of the atheroma.
|
|
Chronic local infections such as periodontitis, despite evidence of a specific inflammatory response to clear the pathogen, are not readily cleared by the host. Thus, perpetuation of these infections may result in low-level, long-term, "smoldering" immunological activation that could result in chronically stimulated cytokine and chemokine production. Since it is known that cytokines and chemokines play a role in the activation of endothelium, SMCs, and the monocytes associated with developing atheroma (Ross, 1999), it is feasible that chronic periodontal infections generate an environment, in the aorta and arterial vasculature, conducive to atheroma development via soluble mediators secreted into the circulation from the oral cavity. The activation of vascular endothelium in response to inflammatory cytokines and chemokines would shift the balance of the vascular endothelium from a "healthy", normal, anti-thrombotic environment to a "diseased", pro-thrombotic environment (Vita and Loscalzo, 2002), and would up-regulate cell adhesion molecule expression that could subsequently participate in the localization of leukocytes (Springer, 1990). To this end, it has been demonstrated, by several groups, that endothelial cells cultured with P. gingivalis, or P. gingivalis outer membrane vesicles, express cell adhesion molecules that could play a role in the establishment and/or acceleration of atheroma (Srisatjaluk et al., 1999; Khlgatian et al., 2002).
In addition to the localization of inflammatory cells to activated endothelium expressing cell adhesion molecules, endothelium could perpetuate the inflammatory lesion by production of soluble mediators that could further stimulate both the migration and proliferation of SMCs (Libby, 2002). Indeed, we previously reported that HUVEC cells cultured with P. gingivalis produced IL-8 and MCP-1 (Nassar et al., 2002). Moreover, it was demonstrated, in that study, that production of these chemokines by HUVEC was dependent on fimbriate organisms, since a P. gingivalis fimA mutant failed to induce chemokine production from HUVEC (Nassar et al., 2002). In addition, Khlgatian et al.(2002) demonstrated that HUVEC cultured with wild-type P. gingivalis, but not a fimA-deficient mutant, elicited surface expression of selectins, ICAM-1, and VCAM-1.
To date, it has not been demonstrated experimentally that immunological sounding from an extravascular infection is a mechanism by which local infection can aggravate systemic diseases such as atherosclerosis. Despite this, intranasal challenge of mice with Chlamydia (Hu et al., 1999), as well as oral challenge of mice with P. gingivalis (Lalla et al., 2003), as discussed above, present with evidence of local infection and aggravation of atherosclerosis. In each of these studies, there was evidence of pathogen localization in the site pre-disposed for atheroma development, thereby failing to support immunological sounding; however, cultivation of organisms from these sites has remained elusive. Analysis of these data supports the hypothesis that infection-accelerated atherosclerosis may be limited to those pathogens which gain access to the vasculature, interact with vascular endothelium, and activate these cells in a manner that accelerates atheroma deposition.
Although the data have many interpretations, it could be argued that inflammatory mediators, released at the site of local infection, initiate the activation of vascular endothelium, and then microbial antigens or DNA released from these pathogens could reach the circulation and bind to endothelial cells or localize with the developing atheroma. In culture, some strains of P. gingivalis produce vesicles (Srisatjaluk et al., 1999), and these vesicles might be released into the circulation, thus serving as a source of P. gingivalis antigens. Thus, it is reasonable to consider that chronic shedding of bacterial components into the vasculature may be capable of accelerating atherosclerosis. Human studies suggest that, shortly after dental treatment, P. gingivalis bacteremia is evident (Messini et al., 1999), and analysis of recent data from the study of mice suggests that P. gingivalis enters the vasculature soon after oral challenge (Gibson et al., 2004). It is feasible that, upon entering this immunologically privileged site, the host mounts a response to this organism while it is in the blood, independent of pathogen-endothelium interactions. It is well-documented that both monocytes and PMNs are activated and secrete pro-inflammatory cytokines and chemokines when cultured with P. gingivalis (Ishikawa et al., 1997; Landi et al., 1997). Thus, the immunological response to blood-borne infection may be sufficient to activate the host immunologically in a manner that promotes atherosclerosis. Future studies are needed to determine the impact of chronic local infection, the inflammatory response to these infections, and its relationship with the acceleration of atherosclerosis.
Conceptually similar to the second model, the third model for assessing the role of infection as a risk factor for pathogen-accelerated atherosclerosis is via trafficking of pathogens from the local site of infection (such as the oral cavity) to the developing atheroma via inflammatory cells (Fig. 4). In this model, as a result of the tissue damage caused by the infection and subsequent inflammatory response, the inflammatory cells present at the local infection ingest the pathogens, and, upon re-emergence of these pathogen-laden inflammatory cells into the vasculature, subsequent localization at the site of developing atheroma may occur. There is a paucity of experimental data to support the hypothesis that pathogen trafficking is valid for the study of atherosclerosis development; however, in other diseases, transport of an infectious agent inside an inflammatory cell from one anatomical site to another has been reported. HIV infection of macrophages has been suggested to be an important mechanism in facilitating HIV infection of the CNS, presumably via HIV infection of monocytes and/or lymphocytes, followed by subsequent t
TOP
ABSTRACT
INTRODUCTION
