Rheumatology 2000; 39: 821-834
© 2000 British Society for Rheumatology
Reviews |
Advances in our understanding of the bone and joint pathology caused by Staphylococcus aureus infection
Cellular Microbiology Research Group, Division of Surgical Sciences, Eastman Dental Institute, University College London, London, UK
 |
Introduction |
|---|
 Top Introduction The role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
There is growing concern that we are entering a bacteriological dark age in which we will no longer be able to treat bacterial infections with antibiotics. This apocalyptic prophecy is fuelled by the increasing incidence of antibiotic-resistant bacteria, and perhaps the best example of how we are losing the antibiotic battle is provided by Staphylococcus aureus [1], which has been reported recently to have become less susceptible to vancomycin, the antibiotic of last resort [2, 3]. S. aureus causes a range of bone and joint pathologies, such as infections of prostheses, osteomyelitis, septic arthritis and septic bursitis. The reported incidence of infection of prosthetic joints is 1–5%, and although this figure may appear to be low, the consequences of infections are dire [4, 5]. Prosthetic implant infections are usually treated by revision operations because the success rate for treatment without revision, at least for prosthetic hip and knee infections, is less than 10% [5, 6]. After revision operations, the rate of infection of replacement prostheses actually rises. Thus, prosthetic infections place a severe burden on the health service through prolonged hospitalization for antibiotic therapy, prosthesis removal and replacement, and space debridement.
S. aureus accounts for 80–90% of cases of pyogenic osteomyelitis [7]. Current treatment for osteomyelitis involves prolonged antibiotic treatment and, in most cases, the debridement of necrotic bone and soft tissues.
S. aureus septic arthritis arises as the result of haematogenous seeding of joints, the result of joint surgery, from a penetrating wound or from an adjacent site of osteomyelitis [8]. Infection of the joint results in damage to the articular cartilage and synovitis, and can also result in severe subchondral bone destruction and consequently debility. Currently, treatment requires primarily antibiotic therapy; whereas removal of synovial effusions by either needle aspiration, arthroscopy or surgical drainage is a matter of controversy.
Septic bursitis is caused by repeated bursal trauma, and in about 50% of cases the portal of entry is a skin lesion [9]. S. aureus is responsible for between 80 and 90% of cases of septic bursitis [8, 9]. Treatment usually involves intravenous administration of antibiotics.
As described, the current treatment for S. aureus infections of bones and joints relies on the use of antibiotics. With the prospect that these antibiotics are likely to be ineffective within a short period of time, it is essential that we determine the mechanisms by which this organism causes pathology. In this article we provide an overview of our current understanding of how S. aureus infections produce bone and joint pathology, and using this information we suggest how it may be possible to treat or prevent such pathology in the future.
|
The role of virulence factors in S. aureus-induced bone and joint pathology |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
S. aureus is a Gram-positive bacterium that is part of the normal microflora in 20–50% of the population, existing in harmony on the skin and/or mucous membranes of the host. These structures act as barriers to S. aureus and prevent it from gaining access to other body tissues and organs. For S. aureus to cause bone and joint disease, conditions must be met that allow this bacterium to breach the host's epithelial barriers. Host predisposition is the predominant factor in the ability of S. aureus to overcome the epithelial barrier, and thus access is usually via a mechanical breach of the epithelium, such as may occur as the result of a wound or surgical procedure. Once S. aureus has gained access to the inner sanctum, its ability to cause bone and joint pathology depends on: (1) the ability to attach to host tissues and its tissue tropism, (2) the ability to evade host defence systems, and (3) the ability to cause damage to host tissues. Whilst all three of these abilities depend in part on host factors, they are predominantly determined by the various classes of virulence factors produced by S. aureus. These range from cell wall components, such as the capsular polysaccharide and peptidoglycan, to secreted proteins, such as exotoxins. The biological activities of virulence factors have been examined using a number of different techniques ranging from in vitro cell culture systems to animal models, and the development of isogenic mutants of S. aureus (i.e. mutants that are genetically identical except for the deletion of specific genes) has enabled the role of certain of these virulence factors to be examined in experimental models of disease. Some of the major virulence factors will be described.
|
Capsular polysaccharide |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
The production of capsular polysaccharide by S. aureus was reported as early as 1930 [10] and has been the subject of investigation for the last 20 yr [11–13]. About 90% of S. aureus isolates produce capsular polysaccharide [14], which forms a layer on the outer surface of the peptidoglycan in the cell wall [15–17]. S. aureus isolates can produce one of 11 different capsular serotypes; serotypes 5 and 8 are the most common, accounting for about 80% of isolates [14–19]. The capsular polysaccharide has been purported to enhance bacterial virulence by inhibiting phagocytosis [15, 20, 21], or, by sterically hindering C3b deposition on the bacterial cell wall, giving it anti-opsonic/anti-phagocytic properties [11, 22]. S. aureus isolates producing capsular polysaccharide serotypes 1 and 2 have been shown to be virulent in animal models; mutants deficient in capsule production were less virulent [23–25]. Clinical isolates of serotypes 1 and 2 are rare [19]. Whilst serotypes 5 and 8 have been shown to be anti-phagocytic [26], some in vivo studies suggest that these capsular polysaccharides do not affect virulence [27–29]. However, expression of the polysaccharide microcapsule has been shown, in a mouse model of haematogenously spread S. aureus arthritis, to increase virulence [30]. S. epidermidis produces polysaccharides that have been implicated in adhesion to the polymer surfaces of medical devices [31, 32]. However, there are no reports that the capsular polysaccharide of S. aureus has a similar function.
|
Lipoteichoic acid and peptidoglycan |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
Gram-positive bacteria have a thick peptidoglycan layer enveloping the cytoplasmic membrane of the cell. Lipoteichoic acid (LTA), another component of the cell wall, is anchored to the cell membrane through ß-gentiobiosyldiacylglycerol. Both of these components of the cell wall have received much attention as potential virulence factors. LTA induces the secretion of the cytokines interleukin (IL)-1ß and IL-6 [33] and the chemokine IL-8 by monocytes, and stimulates nitric oxide production by macrophages [34, 35]. LTA and peptidoglycan are probably most infamous molecules for activating the alternative complement pathway. Peptidoglycan, like LTA, also induces cytokine production by monocytes [36–38]. Although a role for LTA and peptidoglycan in arthritogenicity has been reported [39, 40], it is thought that the major role played by these components is in Gram-positive sepsis [41, 42].
|
Microbial surface components recognizing adhesive matrix molecules |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
The attachment of S. aureus to host tissues is an essential step in the colonization process, and this bacterium produces a range of surface proteins that enable it to attach to host extracellular matrices. These have been termed microbial surface components recognizing adhesive matrix molecules (MSCRAMMs) [43, 44]. Most of the MSCRAMMs are anchored to the cell wall peptidoglycan and have the ability to bind to a variety of host matrix proteins, as detailed in Table 1
. Because the attachment and adhesion of S. aureus to extracellular matrix components or prostheses is seen as the critical event in infections with this organism, much effort has been devoted to identifying and characterizing the MSCRAMMs. The tropism of S. aureus for various body tissues is thought to be due to the ability of MSCRAMMs to bind to the extracellular matrix molecules found in these tissues, and the specific MSCRAMMs expressed by a particular strain of S. aureus may endow it with a lesser or greater propensity for infecting bone and joints. However, the exact relationship between tissue tropism and the expression of particular MSCRAMMs by S. aureus has not been easy to elucidate. For example, it had been thought that expression of the collagen-binding protein, Cna, by S. aureus would be essential to its tropism for bone, since collagen forms 90% of the organic matrix of this tissue. However, Cna is expressed by only 38–56% of S. aureus isolates associated with bone infection [45–47].
|
The genes encoding the adhesins that bind collagen [48], elastin [49], fibrinogen [50–52] and fibronectin [53, 54] have been cloned and sequenced, enabling studies of the effects of knocking out these genes on the virulence of S. aureus in a number of animal models. Although not all of these models have been of bone and joint infection, they do provide a clue as to the role of the MSCRAMMs in the virulence of S. aureus-induced bone and joint pathology.
As mentioned earlier, not all strains of S. aureus isolated from patients with bone or joint infections express the collagen-binding protein Cna; however, its expression has been shown to be necessary for in vitro binding to cartilage and collagenous substrates [45]. The importance of this protein in experimental septic arthritis has been examined by gene knockout and complementation studies. An isogenic mutant in which the cna gene was disrupted was found to be significantly less virulent than the wild-type parental strain [55]. Furthermore, introduction of the cna gene into an S. aureus strain lacking this gene resulted in increased virulence [55]. Thus, it appears that while the collagen-binding protein is not an essential factor in S. aureus bone and joint pathology, it probably plays a role in determining the severity of the disease.
Most clinical isolates of S. aureus express fibronectin-binding proteins encoded by two contiguous genes, fnbpA and fnbpB. Indwelling foreign body implants become coated in plasma proteins, enabling S. aureus to adhere to the implants, and several studies have indicated that fibronectin is the most important protein in this process [56–58]. The implications of these findings are that the fibronectin-binding proteins, FnbpA and FnbpB, may be important determinants in the infection of prostheses. Isogenic mutants of S. aureus in which the genes fnbpA and fnbpB had been knocked out singly were found not to display reduced adherence to fibronectin-coated surfaces when compared with the parental strain. Double knockouts, in which both fnbpA and fnbpB were disrupted, were completely defective in adherence [59]. As yet there have been few studies on the effects of knocking out the fibronectin-binding proteins on the virulence of S. aureus in vivo. Studies using rat infective endocarditis models have produced conflicting results [60, 61]. A study using an animal model of infections of bone–metallic implants demonstrated that an S. aureus isogenic mutant defective in fibronectin adhesion was considerably less effective in colonization than the parental strain [62]. Although more comprehensive studies need to be performed, the initial data seem to suggest that the fibronectin-binding proteins of S. aureus may be important in targeting this organism to bone and joint prostheses.
Two other MSCRAMMs that are possibly important in foreign body infections are the fibrinogen-binding proteins or clumping factors, ClfA and ClfB. Binding of S. aureus to fibrinogen has been shown to be important in attachment to plasma clots and plastic biomaterials exposed to blood [56, 58]. The ability of S. aureus to bind to fibrinogen-coated biomaterials has been shown to be due to ClfA and ClfB by the use of single- and double-knockout mutants of the genes encoding these proteins [63]. This has led to the suggestion that the fibrinogen-binding proteins may be important in wound and foreign body infections. Further support for this hypothesis has come from a rat endocarditis model that demonstrated the reduced virulence of a S. aureus isogenic mutant in which the clfA gene had been disrupted compared with the wild-type parental strain [64]. As yet, the role of ClfA and ClfB in bone and joint infections has not been examined, but given that these proteins appear to be important in the binding of S. aureus to biomaterials and clots, they may play a role in the pathology of these infections.
Whilst MSCRAMMs that bind to bone sialoprotein, osteopontin, thrombospondin and vitronectin have been identified, the proteins responsible for these binding activities, and their corresponding genes, have as yet not been fully characterized, and so the role of these proteins in S. aureus-induced bone and joint pathology remains to be determined. Some of these binding activities may be attributable to the newly discovered serine–aspartate repeat family of surface proteins [65] (Table 1
). The MSCRAMM that is responsible for binding to bone sialoprotein, a protein found only in bone and dentine, may be of particular relevance to S. aureus infection of bone and joints. Indeed, it has been demonstrated that bone sialoprotein is selectively bound by S. aureus isolates that cause osteomyelitis and/or septic arthritis [66–68]. Ryden et al. [69] have speculated that there may be a link between the higher incidence of haematogenously spread osteomyelitis and septic arthritis in young persons and the fact that newly forming bone is particularly rich in bone sialoprotein.
The MSCRAMMs may prove to be very useful therapeutic targets. It is conceivable that patients receiving orthopaedic implants could be vaccinated with particular MSCRAMMs, or peptide analogues, prior to surgery. Experimental models of mouse mastitis have been used to examine the potential of vaccination with the fibrinogen-binding proteins [70], the collagen-binding protein Cna [71] and the fibronectin-binding protein FnbpA [72]. For each protein, vaccination was found to reduce virulence. Immunization with the fibronectin-binding protein of S. aureus also protected against experimental endocarditis in rats [73]. Whilst studies need to be performed in models of S. aureus bone and joint infections, the potential of vaccination is evident.
|
Coagulase |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
A characteristic feature of most strains of S. aureus, and one used in identification, is that they are coagulase-positive, and it has been suggested that coagulase is an important virulence factor of S. aureus [74, 75]. Coagulase is an extracellular protein that exists in free and bound forms [76, 77] and binds prothrombin in a ratio of 1:1 to form staphylothrombin [78, 79]. This binding reaction catalyses plasma clotting. Such clots may provide protection to the bacterium, which may also become coated with fibrin, thereby inhibiting phagocytosis. The gene for coagulase, coa, has been cloned and sequenced [80]. Studies with knockout mutants of the coa gene have provided conflicting data. Phonimdaeng et al. [81] found that disruption of the coa gene had no effect on virulence in subcutaneous and intramammary infections of mice. Similar results were obtained by Baddour et al. [82] using a rat model of experimental endocarditis. More recently Sawai et al. [83] have demonstrated the importance of coagulase in a murine model of haematogenous pulmonary infection. These findings suggest that the role of coagulase in the virulence of S. aureus is dependent on the infection model used. As yet no studies with models of bone and joint infection have been performed. Since clot formation is an important pathophysiological process in osteomyelitis, it is tempting to speculate that coagulase may be an important virulence factor for S. aureus in this condition.
|
Protein A |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
First described some 40 yr ago by Klaus Jensen, this protein is covalently bound to the peptidoglycan layer of the bacterial cell wall in the same manner as the MSCRAMMs discussed previously. Protein A has been proposed to be an important S. aureus virulence factor involved in allowing the bacterium to evade the host immune system because it binds to the Fc portion of IgG [84]. Such binding would result in the antibodies being orientated in the wrong direction for interaction with phagocytic cells and may actually provide a protective coating. The gene for protein A, spa, has been cloned and the protein expressed [85–88]. S. aureus isogenic mutants in which the spa gene had been disrupted were studied in mouse models of intraperitoneal and subcutaneous infection and were found to be less virulent than the parental wild-type strain [89]. More recently, isogenic mutants with a disruption in the protein A gene have been examined in a murine model of S. aureus arthritis [90]. It was found that the wild-type parental strain produced a greater degree of inflammation, pannus formation and cartilage destruction than the isogenic mutant.
|
Exotoxins |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
S. aureus is often described as highly versatile, in part because of its ability to overcome antibiotics, but another reason is the number and diversity of exotoxins that it produces to counter the host's defence systems. S. aureus produces several families of exotoxins with different biological activities, and within these families there are multiple members [Table 2
]. The most widely recognized group of S. aureus exotoxins is the superantigens, which comprise three distinct groups: toxic shock syndrome toxin-1 (TSST-1), the enterotoxins and the exfoliative toxins. The exfoliative toxins are thought to be the causative agent of scalded skin syndrome [91, 92] and their role as classical superantigens has yet to be verified. The enterotoxins and TSST-1 are classical superantigens; they are characterized by their ability to induce high fever and act as T-cell mitogens [93, 94]. Superantigens bind to the major histocompatibility complex (MHC) class II molecules on antigen-presenting cells, and to T-cells bearing specific Vß receptor subsets. This can result in clonal expansion of T-cells, in T-cell anergy or apoptosis. The binding of superantigens to MHC class II molecules is unique in that binding occurs outside the peptide groove and does not require prior antigen processing [94]. Activation of T-cells and antigen-presenting cells by superantigens results in the hypersecretion of specific cytokines [95, 96]. The role of enterotoxins and TSST-1 in S. aureus-induced bacterial arthritis has been examined. Initial interest in the possibility that superantigens play a role in bacterial arthritis was probably stimulated by the finding that in rheumatoid arthritis (RA) the frequency of T cells bearing certain Vß subsets was greater in affected joints [97], and it was suggested that superantigens may be involved in this condition [97, 98]. However, less credence is now given to the involvement of superantigens in RA. Bremell and Tarkowski [99] demonstrated the preferential induction of septic arthritis and mortality in a rat model of septic arthritis by S. aureus strains producing individual enterotoxins (A to D) or TSST-1, over non-producing strains. Studies using a mouse model comparing a S. aureus isogenic mutant defective in TSST-1 production with the parental strain have also demonstrated that TSST-1 contributes to arthritogenicity [100]. Direct intravenous injection of TSST-1 was able to reactivate bacterial cell wall-induced arthritis in a rat model [101]. Similarly, injection of enterotoxin B into mice induced arthritis [102, 103]. All of these data suggest an important role for superantigens in the pathology of bacterial arthritis. The possibility of targeting superantigens as putative therapeutic targets in bone and joint pathology has not been examined in any detail because, to date, more than 10 different superantigens have been identified and different strains of S. aureus can produce quite different individual superantigens, or combinations of them. Additionally, although the superantigens exert their effects by similar mechanisms their target specificities are different. These factors limit the use of individual superantigens as universal therapeutic targets in arthritic disease.
|
Apart from the superantigens, only three other exotoxins, the
-, ß- and 
-toxins, have been investigated as virulence factors in S. aureus-induced bone pathology. From the nomenclature, one could be forgiven for thinking that these exotoxins are related, but in actual fact they are unrelated and have different mechanisms of action. -Toxin, also known as -haemolysin, is a pore-forming toxin produced by a large percentage of S. aureus strains, and was first identified over 100 yr ago by van de Velde for its ability to lyse rabbit erythrocytes. -Toxin forms an oligomeric pore in the cell membrane [104] of a number of cell types, including keratinocytes [105], T lymphocytes [106], fibroblasts [107], endothelial cells, monocytes and platelets [108]. The pores formed allow the rapid diffusion of ions and molecules smaller than 1000 Da through the cell membrane, and this can activate a number of cell processes, such as arachidonic acid metabolism [109, 110], and the secretion of cytokines [108]. ß-Toxin or ß-haemolysin was first identified in 1935 [111] and is similar to -toxin in that it induces haemolysis of erythrocytes. However, ß-toxin is actually an enzyme, sphingomyelinase C, that cleaves sphingomyelin from the outer leaflet of eukaryotic cell membranes, making them leaky and thus allowing small molecules to exit rapidly through the membrane. In addition, the hydrolysis of sphingomyelin by ß-toxin also gives rise to ceramide, a potent second messenger in mammalian cells, and one that could activate protein kinases and apoptosis [112]. Most strains of S. aureus isolated from humans do not produce ß-toxin, probably because of phage conversion [113], and so this toxin is unlikely to be important in bone and joint pathology. -Toxin, or -haemolysin, is a member of the two-component toxin family of exotoxins that includes the Panton–Valentine leukocidins [114]. The -toxin genetic locus of S. aureus comprises three open reading frames encoding HlgA, HlgB and HlgC [115]. HlgA and HlgC are referred to as S-components and HlgB as the F-component [116]. Active -toxin is composed of one of the S-components and the F-component. Binding of the larger F-component, HlgB, to the eukaryotic cell membrane is a prerequisite for binding of the smaller S-component, HlgA or HlgC. Upon binding of both components to the cell membrane, pores are formed [117]. Membrane permeabilization causes ion efflux that can activate a number of cell types, including lymphocytes [118] and neutrophils [117]. -Toxin is produced by a broad range of clinical isolates of S. aureus [119, 120], making it an important candidate virulence factor. The presence of antibodies to -toxin has been used to diagnose osteomyelitis [121]. All three of these toxins have been examined in a murine septic arthritis model. Isogenic mutants deficient in one toxin or combinations of each of the toxins were examined for their virulence in this model. These studies revealed that both -toxin and -toxin were important virulence determinants [122]. The fact that -toxin and -toxin are present in a high proportion of clinical isolates makes them attractive vaccine candidates. Hyperimmune globulin containing high titres of -toxin antibodies has been shown to be protective against the systemic effects of S. aureus infection in monkeys [123]. Vaccines based on -toxin have been investigated for the treatment of mastitis in cows [124, 125]. Inactive, toxoid forms of -toxin have also been examined as possible vaccines [126, 127]. Use of -toxin for vaccine development, on the other hand, does not seem to offer as much promise because patients produce antibodies to this toxin and there is no evidence to suggest that these antibodies are protective. To date only the above-mentioned exotoxins have been investigated as virulence factors in S. aureus-induced bone and joint pathology. However, it is likely that others are also important. We have described previously a group of surface-associated proteins [SAPs] from S. aureus that have potent bone-destructive activity in a murine calvarial bone resorption assay [128]. Unlike the superantigens, the SAPs appear to have a direct effect on bone cells [128]. The SAPs are a mixture of as yet only partly characterized proteins that have a number of actions on bone, including the ability to inhibit the synthetic activity of the bone-forming cells (osteoblasts), the stimulation of osteoclast recruitment [129] and osteoclast activation [130]. We have also shown that serum from some individuals contains antibodies that are able to neutralize the activities of SAPs [131]. Since these exotoxins have direct actions on bone and bone cells, they may turn out to be important therapeutic targets for the treatment of bone and joint disease caused by S. aureus.
|
Enzymes |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
Two groups of S. aureus enzymes, the proteases and the lipases, have been suggested to be important determinants in evading host defences. The V8 protease of S. aureus cleaves and inactivates IgG antibodies in vitro. This and other proteases may also be able to degrade the antimicrobial peptides produced by host cells [132]. Most strains of S. aureus are lipolytic [133]. The host produces a range of fatty acids and lipids that can attack bacteria in response to infection, and it has been postulated that S. aureus lipases counteract these molecules. In particular, an enzyme termed FAME (fatty acid metabolizing enzyme) has been proposed to degrade bactericidal fatty acids [134]. S. aureus also produces a phosphatidylinositol-specific phospholipase C that has been proposed to be a virulence factor [135, 136]; however, little work has been done on this enzyme. To date no deletion mutants of the proteases or lipases have been studied in animal models, so it is difficult to ascribe to these proteins relevance to the pathology of bone and joint infections. Another enzyme produced by S. aureus that is of interest in the context of bone and joint pathology is hyaluronate lyase [137]. However, once again the biological relevance of this enzyme has not been determined.
|
Regulation of virulence factor production by S. aureus and its effects on bone and joint pathology |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
The genes encoding bacterial virulence factors are extremely tightly controlled, and the control mechanism could be a therapeutic target. The transcription of bacterial genes is regulated by mechanisms including DNA supercoiling, control of operons (tandemly linked genes) and coordinate control of unlinked genes by common regulators. Coordinate control of gene expression is also referred to as global regulation. The responses to many environmental stimuli, such as heat, osmolarity, pH and oxygen tension and starvation, are controlled by global regulators that use a two-component signal transduction pathway [138, 139]. The two-component systems typically consist of an operon that contains a transmembrane sensor protein and a transcription factor or response regulator. On recognizing its specific signal, the sensor protein, or receptor, becomes autophosphorylated. The phosphate is transferred from the sensor to the response regulator, which is then in an activated form. Virulence factor production by S. aureus is growth phase-dependent in liquid culture. Classically the growth of bacteria in liquid culture can be divided into four phases: (1) an initial lag phase in which growth is slow; (2) an exponential phase in which cell growth is rapid, (3) a post-exponential phase in which growth begins to level off, and (4) a stationary phase. Most exotoxins and extracellular enzymes are produced in the post-exponential phase of growth [140], whilst the MSCRAMMs and other cell wall-anchored proteins are produced during the logarithmic phase of growth. The regulation of virulence factor production in S. aureus is controlled by at least two global regulatory loci, the agr (accessory gene regulator) locus and the sar (staphylococcal accessory regulator) locus.
The agr locus was first identified as a site that controlled the expression of several virulence factors when it was disrupted by insertion of the transposon Tn551 [141]. Disruption of the agr locus resulted in the repression of exotoxin production but conversely resulted in overexpression of cell wall-associated proteins such as the MSCRAMMs. The agr locus has now been cloned and sequenced [142, 143] and shown to consist of two divergent transcription units (Fig. 1). The first transcription unit, RNAII, transcribed from the promoter P2, codes for four open reading frames: the response regulator AgrA; the sensor AgrC; a precursor of an octapeptide pheromone AgrD; and the AgrD processor AgrB. The octapeptide pheromone synthesized by AgrD and AgrB accumulates during the post-exponential phase, when it binds to the transmembrane sensor AgrC [144]. The second transcription unit, RNAIII, is transcribed from the promoter P3 and codes for delta haemolysin Hld. Perhaps surprisingly, the transcript RNAIII (and not any of its translation products) is actually the effector molecule of the agr response. The role of agr in the pathogenesis of staphylococcal osteomyelitis [145] and arthritis has been studied [146]. In the murine arthritis model, isogenic mutants disrupted in either the agrA gene or the hld gene were examined. Sixty per cent of mice inoculated with the wild-type strain of S. aureus developed arthritis. However, the isogenic mutants of agrA and hld induced arthritis in only 10 and 30% of the mice, respectively. Similarly, in a rabbit model of osteomyelitis, disruption of the agrA gene of S. aureus reduced the incidence and severity of disease. However, disruption of this gene did not eliminate the ability of S. aureus to colonize bone and cause histopathological evidence of osteomyelitis.
|
The sar locus, like the agr locus, was identified by transposon insertional disruption [147, 148]. Disruption of the sar locus has effects on the growth phase-dependent expression of virulence factors some of which are similar to those of disruption of the agr locus [147]. The sar locus seems to control gene expression both independently of the agr locus [149] and also by binding to the P2 and P3 promoters of the agr locus [150]. The role of the sar locus in the pathogenesis of staphylococcal arthritis has been studied in a murine model [151]. Isogenic mutants disrupted in the sar locus were less virulent than the parental wild type.
Given that the agr and sar loci are global regulators of virulence factor production and that their disruption significantly diminishes the virulence of S. aureus in animal models, they are very attractive targets for therapeutic intervention. We are still at an early stage of understanding the mechanisms by which the sar locus exerts its actions. However, much more is known about the agr locus and its mechanisms of action and ideas have been proposed about possible means of targeting this system. Particular interest has centred on the octapeptide pheromone produced by this locus. The octapeptide pheromone is secreted by S. aureus and acts in an autocrine manner, binding to its transmembrane receptor, AgrC, and activating agr regulation of virulence factor production. An interesting finding was that the peptide pheromones from different strains of S. aureus [152] and indeed from S. epidermidis [153] differed and could inhibit virulence gene expression in heterologous strains. These findings suggest that it may be possible to design peptide therapeutics that target AgrC and thus prevent virulence factor expression. It has also been suggested that vaccines could be developed to target the Agr system of virulence factor regulation [154, 155].
|
S. aureus invasion of host cells: a role in bone and joint pathology? |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
S. aureus has traditionally been regarded as an extracellular pathogen, i.e. it has been thought not to invade mammalian cells. However, it has been known for over 10 yr that S. aureus is internalized by endothelial cells [156–158]. More recently it has been shown that internalization of S. aureus by endothelial cells induces a number of responses, including cytokine production [159], hyperadhesiveness for monocytes and granulocytes [160], and apoptosis [161]. Epithelial cells have also been shown to internalize S. aureus [162, 163]. Internalization of S. aureus by epithelial cells induces changes in gene expression in both the eukaryotic cell [164] and the prokaryotic cell [165]. The findings that S. aureus could be internalized and exist as an intracellular pathogen led us and others to speculate that bone cells may also be capable of taking up this organism. The ability of S. aureus to be internalized by bone cells was first demonstrated using mouse osteoblasts [166]. Since then, work from the authors' laboratory has shown that human osteoblasts internalize S. aureus, and some of the mechanisms involved in this process have been elucidated [167, 168]. All of these studies have been performed using cells cultured in vitro, and it remains to be seen whether the internalization of S. aureus by osteoblasts actually occurs in vivo. If this is the case, it could have very important consequences for the pathology of S. aureus infections of bone and joints. For instance, the internalization of S. aureus by endothelial cells can lead to the appearance of small colony variants [169]. These small colony variants are resistant to antibiotic treatment [170, 171] and also persist within the cells [172]. It is possible that the internalization of S. aureus by osteoblasts could also lead to the appearance of small colony variants, and such variants have been isolated from bone infections [173–175].
|
Host factors important in S. aureus-induced bone and joint pathology |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
The role of host factors in S. aureus-induced bone pathology have been studied in a number of ways, and the availability of transgenic animals in which specific genes have been inactivated has allowed the role of identified host factors to be tested. Host factors identified as being important in S. aureus-induced bone pathology may be therapeutic targets.
S. aureus-induced bone and joint conditions are all associated with an inflammatory response, and a number of studies have investigated the role of inflammatory cytokines in these processes. Tumour necrosis factor (TNF) and IL-1 are the two inflammatory cytokines that have received the most attention. A mouse model of staphylococcal arthritis [176] and a rat model of osteomyelitis [177] both revealed an early increase in the level of transcription of messenger RNA for TNF at infected sites. Although not examined in the osteomyelitis model, increased levels of IL-1ß mRNA transcription were also found in the staphylococcal arthritis model [176], and again these increases occurred at early stages of infection. It has also been found that transgenic mice with a double knockout of TNF and lymphotoxin were resistant to S. aureus septic arthritis [178]. The effects of blocking TNF, using a neutralizing monoclonal antibody, and IL-1, using the IL-1 receptor antagonist (IL-1ra), have been examined in a rabbit model of S. aureus-induced arthritis [179]. This study found that co-injection of anti-TNF and IL-1ra with S. aureus inhibited leukocyte infiltration by 80%. However, when treatment with the inhibitors was given 24 h after infection, leukocyte infiltration was not blocked. Taken as a whole, these studies suggest that TNF and IL-1ß are involved in the early stages of S. aureus-induced bone pathology, and as such may not be good therapeutic targets for the treatment of pre-existing disease.
Interferon (IFN-), a cytokine produced primarily by T cells, has also been examined in mouse models of S. aureus septic arthritis. Mice deficient in the receptor for IFN- displayed more severe arthritis than the wild-type controls [180] and higher mortality early after inoculation with S. aureus. Interestingly, however, mortality at later stages of disease was significantly greater in the wild-type mice than in the IFN- receptor-deficient mice. A follow-up study investigated the effects of supplementation with and neutralization of IFN- during septicaemia and arthritis in the same murine model [181]. Supplementation with IFN- significantly decreased mortality but increased the development of arthritis. Treatment of mice with a neutralizing antibody to IFN- did not decrease mortality significantly, but did reduce the frequency and severity of arthritis. So it seems that IFN- plays a dual role in S. aureus infection: it protects the host against septicaemia but at the same time promotes the development of septic arthritis. Thus, in spite of its significant role in bacterial arthritis, IFN- does not appear to be a useful therapeutic target.
Another T-cell cytokine, IL-4, an anti-inflammatory cytokine, has also been examined in murine models of S. aureus-induced septic arthritis and osteomyelitis. Transgenic mice deficient in IL-4 had a lower frequency of arthritis and lower mortality than wild-type mice [182]. These authors suggested that IL-4 and IL-4-dependent responses promote septic arthritis and sepsis-related mortality by inhibition of bacterial clearance. So one could speculate that therapeutics aimed at blocking IL-4 production may have some value. However, it is important to remember that IL-4 is an anti-inflammatory cytokine and as such its blockade may have other consequences. For instance, one of the characteristic features of osteomyelitis is the abnormal osteoclastic activity at uninjured/uninfected sites, and it has been found that this activity can be eliminated by treatment with IL-4 in a murine model of S. aureus-induced osteomyelitis [183]. Thus, blockade of IL-4 in the case of osteomyelitis would exacerbate pathology.
Nitric oxide is an inflammatory mediator that is also microbicidal, and it has received wide attention. It has been shown that inhibition of nitric oxide synthase aggravates S. aureus septicaemia and septic arthritis [184]. Similarly, mice deficient in the inducible nitric oxide synthase had increased frequency and severity of septicaemia and arthritis [185]. These studies demonstrate that, although nitric oxide is a potent inflammatory modulator, it performs a protective role in S. aureus-induced arthritis.
Finally, complement has been suggested to play an important role in S. aureus infections. The importance of complement has been demonstrated in a mouse model of S. aureus-induced arthritis [186]. Depletion of complement using cobra venom factor aggravated septicaemia and septic arthritis. It was suggested that aggravation occurred as a result of a combination of decreased migration/extravasation of polymorphonuclear cells with impairment of phagocytosis.
|
Conclusions |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
This review touches on the number and complexity of S. aureus factors involved in bone and joint pathology, ranging from structural components such as the capsular polysaccharide, lipoteichoic acid and peptidoglycan to the quintessential exotoxins such as the superantigens. Although picture is complex, we are slowly unravelling the pieces and have established the role that many of these molecules play in the pathology of bone and joint disease. However, much work remains to be done. Our conceptions of the role played by host factors in S. aureus-induced bone and joint pathology are not as clear as those for bacterial factors, and this is an area that requires much more development before it is likely to bear fruit and identify treatment modalities.
Although the future appears bleak for the antibiotic treatment of S. aureus infections, our advances in understanding the mechanisms by which this organism causes bone and joint disease will no doubt lead to the identification of candidate therapeutic targets of both bacterial and host origin. Potential S. aureus vaccine targets, such as the capsular polysaccharide and the -, ß- and -toxins, have already been identified and examined in vivo. More recently other bacterial vaccine targets, such as the MSCRAMMs, have been shown to have promise. Possibly the most interesting therapeutic candidate identified to date is the agr locus, which regulates virulence factor production on a global scale. Possible therapies directed at this system include peptide therapeutics that will block the activation of the Agr system and vaccines directed against the octapeptide pheromone and other proteins that activate this system. Thus, alternatives to antibiotic treatment are likely to be available in the not too distant future.
|
Acknowledgments |
|---|
The authors are grateful to the Arthritis Research Campaign for funding.
|
Notes |
|---|
Correspondence to: S. Nair, Cellular Microbiology Research Group, Eastman Dental Institute, University College London, 256 Gray's Inn Road, London WC1X 8LD, UK.
|
References |
|---|
|
TopIntroductionThe role of virulence...Capsular polysaccharideLipoteichoic acid and...Microbial surface components...CoagulaseProtein AExotoxinsEnzymesRegulation of virulence factor...S. aureus invasion of...Host factors important in...ConclusionsReferences |
|---|
- Lyon BR, Skurray R. Antimicrobial resistance of Staphylococcus aureus: genetic basis. Microbiol Rev1987;51:88–134.
[Free Full Text] - Hiramatsu K, Hanaki H, Ino T, Yabuta K, Oguri T, Tenover FC. Methicillin-resistant Staphylococcus aureus clinical strain with reduced vancomycin susceptibility. J Antimicrob Chemother1997;40:135–6.
[Free Full Text] - Rotun SS, McMath V, Schoonmaker DJ, Maupin PS, Tenover FC, Hill BC, Ackman DM. Staphylococcus aureus with reduced susceptibility to vancomycin isolated from a patient with fatal bacteremia. Emerg Infect Dis1999;5:9–17.[Medline]
- Blackburn WD Jr, Alarcon GS. Prosthetic joint infections: a role for prophylaxis. Arthritis Rheum1991; 34:110–7.[Medline]
- Salvati EA, Robinson RP, Zeno SM, Koslin BL, Brause BD, Wilson PD Jr. Infection rates after 3175 total hip and total knee replacements performed with and without a horizontal unidirectional filtered air-flow system. J Bone Joint Surg1982;64A:525–35.
- Inman RD, Gallegos KV, Brause BD, Redecha PB, Christian CL. Clinical and microbial features of prosthetic joint infection. Am J Med1984;77:47–53.[ISI][Medline]
- Dirschl DR, Almekinders LC. Osteomyelitis: common causes and treatment recommendations. Drugs1993; 45:29.[ISI][Medline]
- Goldenberg DL. Bacterial arthritis. Philadelphia: W.B. Saunders, 1994.
- Norden C. Infectious arthritis. Cambridge: Blackwell Scientific, 1994.
- Gilbert I. Dissociation in an encapsulated staphylococcus. J Bacteriol1930;21:157–60.
- Wilkinson BJ, Peterson PK, Quie PG. Cryptic peptidoglycan and the antiphagocytic effect of the Staphylococcus aureus capsule: model for the antiphagocytic effect of bacterial cell surface polymers. Infect Immun1979; 23:502–8.
[Abstract/Free Full Text] - Peterson PK, Kim Y, Wilkinson BJ, Schmeling D, Michael AF, Quie PG. Dichotomy between opsonization and serum complement activation by encapsulated staphylococci. Infect Immun1978;20:770–5.
[Abstract/Free Full Text] - Karakawa WW, Kane JA. Immunochemical analysis of a Smith-like antigen isolated from two human strains of Staphylococcus aureus. J Immunol1975;115:564–8.
[Abstract/Free Full Text] - Hochkeppel HK, Braun DG, Vischer W, Imm A, Sutter S, Staeubli U, Guggenheim R, Kaplan EL, Boutonnier A, Fournier JM. Serotyping and electron microscopy studies of Staphylococcus aureus clinical isolates with monoclonal antibodies to capsular polysaccharide types 5 and 8. J Clin Microbiol1987;25:526–30.
[Abstract/Free Full Text] - Karakawa W, Vann W. Capsular polysaccharides of Staphylococcus aureus. Semin Infect Dis1982;4:285–93.
- Poutrel B, Boutonnier A, Sutra L, Fournier JM. Prevalence of capsular polysaccharide types 5 and 8 among Staphylococcus aureus isolated from cow, goat, and ewe milk. J Clin Microbiol1988;26:38–40.
[Abstract/Free Full Text] - Sompolinsky D, Samra Z, Karakawa WW, Vann WF, Schneerson R, Malik Z. Encapsulation and capsular types in isolates of Staphylococcus aureus from different sources and relationship to phage types. J Clin Microbiol1985;22:828–34.
[Abstract/Free Full Text] - Albus A, Fournier JM, Wolz C, Boutonnier A, Ranke M, Hoiby N, Hochkeppel H. Staphylococcus aureus capsular types and antibody response to lung infection in patients with cystic fibrosis. J Clin Microbiol1988;26:2505–9.
[Abstract/Free Full Text] - Arbeit RD, Karakawa WW, Vann WF, Robbins JB. Predominance of two newly described capsular polysaccharide types among clinical isolates of Staphylococcus aureus. Diagn Microbiol Infect Dis1984;2:85–91.[ISI][Medline]
- Horwitz MA. Phagocytosis of microorganisms. Rev Infect Dis1982;4:104–23.[ISI][Medline]
- Peterson PK, Wilkinson BJ, Kim Y, Schmeling D, Quie PG. Influence of encapsulation on staphylococcal opsonization and phagocytosis by human polymorphonuclear leukocytes. Infect Immun1978;19:943–9.
[Abstract/Free Full Text] - Wilkinson BJ. Staphylococcal capsules and slime. In: Easmon CSF, Adlam C, editors. Staphylococci and staphylococcal infections. New York: Academic Press, 1983;481–523.
- Melly MA, Duke LJ, Liau DF, Hash JH. Biological properties of the encapsulated Staphylococcus aureus M. Infect Immun1974;10:389–97.
[Abstract/Free Full Text] - Peterson PK, Wilkinson BJ, Kim Y, Schmeling D, Quie PG. Influence of encapsulation on staphylococcal opsonization and phagocytosis by human polymorphonuclear leukocytes. Infect Immun1978;19:943–9.
[Abstract/Free Full Text] - Lee JC, Betley MJ, Hopkins CA, Perez NE, Pier GB. Virulence studies, in mice, of transposon-induced mutants of Staphylococcus aureus differing in capsule size. J Infect Dis1987;156:741–50.[ISI][Medline]
- Karakawa WW, Sutton A, Schneerson R, Karpas A, Vann WF. Capsular antibodies induce type-specific phagocytosis of capsulated Staphylococcus aureus by human polymorphonuclear leukocytes. Infect Immun1988;56:1090–5.
[Abstract/Free Full Text] - Albus A, Arbeit RD, Lee JC. Virulence of Staphylococcus aureus mutants altered in type 5 capsule production. Infect Immun1991;59:1008–14.
[Abstract/Free Full Text] - Baddour LM, Lowrance C, Albus A, Lowrance JH, Anderson SK, Lee JC. Staphylococcus aureus microcapsule expression attenuates bacterial virulence in a rat model of experimental endocarditis. J Infect Dis1992;165:749–53.[Medline]
- Xu S, Arbeit RD, Lee JC. Phagocytic killing of encapsulated and microencapsulated Staphylococcus aureus by human polymorphonuclear leukocytes. Infect Immun1992;60:1358–62.
[Abstract/Free Full Text] - Nilsson IM, Lee JC, Bremell T, Ryden C, Tarkowski A. The role of staphylococcal polysaccharide microcapsule expression in septicemia and septic arthritis. Infect Immun1997;65:4216–21.[Abstract]
- Muller E, Takeda S, Shiro H, Goldmann D, Pier GB. Occurrence of capsular polysaccharide/adhesin among clinical isolates of coagulase-negative staphylococci. J Infect Dis1993;168:1211–8.[ISI][Medline]
- Mack D, Fischer W, Krokotsch A, Leopold K, Hartmann R, Egge H, Laufs R. The intercellular adhesin involved in biofilm accumulation of Staphylococcus epidermidis is a linear beta-1,6-linked glucosaminoglycan: purification and structural analysis. J Bacteriol1996;178:175–83.
[Abstract/Free Full Text] - Bhakdi S, Klonisch T, Nuber P, Fischer W. Stimulation of monokine production by lipoteichoic acids. Infect Immun1991;59:4614–20.
[Abstract/Free Full Text] - Hattor Y, Kasai K, Akimoto K, Thiemermann C. Induction of NO synthesis by lipoteichoic acid from Staphylococcus aureus in J774 macrophages: involvement of a CD14-dependent pathway. Biochem Biophys Res Commun1997;233:375–9.[ISI][Medline]
- Kengatharan M, De Kimpe SJ, Thiemermann C. Analysis of the signal transduction in the induction of nitric oxide synthase by lipoteichoic acid in macrophages. Br J Pharmacol1996;117:1163–70.[ISI]
- Lawrence C, Nauciel C. Production of interleukin-12 by murine macrophages in response to bacterial peptidoglycan. Infect Immun1998;66:4947–9.
[Abstract/Free Full Text] - Weidemann B, Brade H, Rietschel ET, Dziarski R, Bazil V, Kusumoto S et al. Soluble peptidoglycan-induced monokine production can be blocked by anti-CD14 monoclonal antibodies and by lipid A partial structures. Infect Immun1994;62:4709–15.
[Abstract/Free Full Text] - Timmerman CP, Mattsson E, Martinez-Martinez L, De Graaf L, Van Strijp JA, Verbrugh HA et al. Induction of release of tumor necrosis factor from human monocytes by staphylococci and staphylococcal peptidoglycans. Infect Immun1993;61:4167–72.
[Abstract/Free Full Text]