Differential arthritogenicity of Staphylococcus aureusstrains isolated from biological samples

Staphylococcus aureus is a major cause of bacteremia which frequently leads to infective endocarditis, metastatic abscess formation, toxic shock syndrome, gastroenteritis, pneumonia, osteomyelitis and septic arthritis (SA) [1]. The development of these secondary infections is due to bacterial dissemination from the blood to surrounding tissues and is associated with significantly increased morbidity and mortality [1]. Even though all these secondary infections are severe, SA deserves special attention because it is a rapidly progressive and highly erosive disease of the joints that needs an immediate therapeutical intervention [2, 3]. The most important risk factor for SA is pre-existing joint disease, especially rheumatoid arthritis (RA) and prosthetic joint surgery [2]. The mortality rate in patients with SA is elevated; around 5-20% of adults with this disease may die as a consequence of their systemic infection [3]. However, in RA patients that have S. aureus infections in more than one joint, the mortality risk increases to 50% due to the combination of delayed diagnosis, therapeutic immunosuppression, older age and also the polyarticular involvement [2, 3].

One of the hallmarks of SA is the massive inflammation that anticipates bone destruction. The infection by S. aureus is accompanied by a rapid recruitment of polymorphonuclear granulocytes and activated macrophages that are then followed by T cells [4]. Although monocytes and macrophages are important to clear bacteria, they also play a pivotal role in the destructive inflammation within the joint [5]. The involvement of pro-inflammatory cytokines in the pathogenesis of S. aureus infection has been reported. This bacteria can induce cytokines such as TNF-α, IFN-γ, IL-1, IL-2, and IL-6 [6, 7]. Cytokines released from macrophages as TNF-α, IL-1β and IL-6 have been classically pointed as the major players of the severe inflammation that precedes cartilage and bone destruction in SA [2]. The role of IL-17 in SA is not well established. However, a possible deleterious role is highly supported by many reports in the areas of rheumatoid arthritis and osteoarthritis [8]. IL-17A appears to play a key role in host defense against local S. aureus infections by inducing the production of neutrophil-mobilizing chemokines, colony-stimulating factors, and cytokines [9].

S. aureus strains can produce a number of different components that may contribute to virulence and arthritogenicity, including surface-associated adhesins, capsular polysaccharides, clumping factor A, exoenzymes, and exotoxins [10‐12]. Some of the toxins produced by S. aureus are called superantigens (SAgs) because they are endowed with the ability to activate various T cell clones, independently of their specificity. These SAgs mediate T cell activation in a very distinctive way from conventional antigens. These molecules are able to simultaneously bind to class II molecules, on antigen presenting cells, and to a large T cell population comprising all clones that share certain variable regions in the TCR Vβ chain [13]. They cause fever, hypotension and other acute toxic-shock-like symptoms by inducing the release of pro-inflammatory cytokines, such as IFN-γ, TNF-α, IL-1 and IL-12 [14, 15].

Several studies indicate that experimental staphylococcal arthritis in mice is the best model to study SA because of the striking resemblances between the murine and human immune systems [16, 17]. The characteristics of the murine model closely mirror changes seen in human SA, especially with regard to the high frequency and severity of periarticular bone erosivity [4]. LS-1 strain is able to produce TSST-1 and is the most employed S. aureus strain to trigger experimental SA [5, 16, 17]. The main objective of this work was to compare the arthritogenic potential of various SAg-producing staphylococci. Disease incidence, clinical scores, histopathological alterations and cytokine production were the criteria used to characterize arthritis severity.

Septic arthritis is an infectious disease that affects the joints. Due to its fast evolution, even with prompt therapy, it can cause irreversible joint damage and even death [2]. S. aureus is the most common causative agent of this disease and it has been believed that SAg-production plays a pivotal role in arthritogenicity [2, 15]. The main goal of this work was to compare the arthritogenic potential of SAg-producing S. aureus strains considering disease incidence, clinical score and histopathological alterations. Cytokine production was also evaluated to get some insight into possible differences among the various strains. An initial screening process was performed to evaluate susceptibility of C57BL/6 and BALB/c mice to develop septic arthritis after S. aureus infection. Gender effect was also checked. This preliminary evaluation, that was done only with the ATCC 19095 SEC⁺ strain, indicated that male and female, from both mice strains, lost a significant percentage of weight in the first days of infection. We believe that all animals were similarly infected because weight loss is accepted as a parameter to indicate effective infection [25]. However, only C57BL/6 male mice developed arthritis. This resistance of BALB/c mice to develop experimental arthritis was already described by Bremmel et al. (1991) [17]. Our results were, however, distinct from the more recently published data of Henningsson et al. (2010) [26]. These authors described a high incidence of arthritis in both male and female C57BL/6 mice infected with the LS-1 TSST-1⁺ S. aureus strain. This variable susceptibility could be related, at least partially, to S. aureus strain particularities. C57BL/6 male mice were, therefore, chosen to compare the arthritogenic potential of the different S. aureus strains. These animals were then infected by the retro-orbital plexus and body weight and clinical scores were daily checked until the 14th day when animals were euthanized for histopathological and immunological evaluations. The fact that inoculation of all 5 strains triggered significant weight loss, suggests successful experimental infections as has been proposed by other authors [25].

Arthritis incidence and clinical scores varied among S. aureus strains. ATCC 19095 SEC⁺ and N315 ST5 TSST-1⁺ were the most arthritogenic ones. They triggered earlier symptoms and the highest levels of incidence. The ATCC 19095 SEC⁺ strain was associated with the highest clinical scores. The S-70 TSST-1⁺ infected group presented a delayed clinical manifestation and lower clinical scores whereas the ATCC 51650 TSST-1⁺ strain did not cause any sign of arthritis during 14 days. The ATCC 13565 SEA⁺ provoked death of 85% of the animals after 48 h (not shown). If these differential outcomes in mice can be translated to human SA is a subject that needs further investigation. Many bacterial components are thought to contribute to arthritogenicity [10‐12]. The delayed appearance of arthritis in S-70 TSST-1⁺ infected mice could mean, for example, that this strain is endowed with a smaller number of arthritogenic factors than the other ones. In this scenario, we could think that this strain is being better controlled by the innate immunity. Alternatively, some strains, as seems to be the case of S-70 TSST-1⁺, could be less inflammatory. The level of pro-inflammatory cytokines showed by our results indicates that this less arthritogenic strain induced lower levels of TNF-α, IL-6 and IL-17 than the two more arthritogenic ones.

No direct relationship was found between arthritogenicity and bacterial inoculum. By comparing bacterial inoculum with the degree of arthritis severity, we can conclude that the differential arthritis severity was not associated with distinct bacterial concentrations. We cannot rule out, however, the possibility that these strains present distinct potential to colonize the joints. Unexpectedly, even though the bacterial suspensions from the 5 strains were prepared by the same methodology, they really contained distinct number of viable bacteria. This subject was not further evaluated but it suggests that these preparations could have distinct proportions of dead and alive bacteria. Concerning this differential arthritogenicity among S. aureus strains we would like to highlight two aspects. This is the first demonstration that these strains, that were originally isolated from biological samples, can cause septic arthritis in mice. In addition, these results indicate that superantigenicity was not enough to elicit arthritis. The ATCC 51650 TSST-1⁺ strain, for example, was not able to induce septic arthritis, even though it was used in the adequate range of bacterial concentration [16]. This finding is in accordance with the postulate that induction of arthritis by S. aureus infection is likely elicited by the concerted action of multiple events as activation of T lymphocytes by SAgs, exposure to peptidoglycans/capsular polysaccharides, presence of surface-associated adhesins, clumping factor A and also free bacterial DNA [2, 10‐12]. Even though not clearly demonstrated in the literature, the ability of S. aureus to form biofilms with joint components is being suggested as a possible arthritogenic element [27, 28]. In this sense, it would be very enlightening to compare the ability of these S. aureus strains to form biofilms.

The histopathological analysis of the joints revealed the presence of synovial proliferation, pannus formation and inflammatory infiltrates in the joint cavity. These findings are similar to the most consensual features of SA caused by S. aureus[29]. Cartilage and bone erosion were also present, mainly in joints from animals infected with the ATCC 19095 SEC⁺ strain. This evolution from the inflammatory process to cartilage and bone erosion is very relevant because it mimicries the human situation during SA by S. aureus. In this case, 25-50% of the patients progress to bone destruction and irreversible loss of joint function [30‐32].

TNF-α, IL-6 and IL-17 are being described as some of the most relevant mediators in SA immunopathogenesis [26, 33]. To investigate the possible contribution of cytokines to these histopathological alterations, the production of TNF-α, IFN-γ, IL-6 and IL-17 was simultaneously quantified in spleen cultures stimulated with S. aureus or ConA. Even though all of them can be released during the initial innate immunity, IFN-γ and IL-17 can also be produced during specific immunity by effector Th1 and Th17 cells, respectively. Interestingly, the lowest levels of TNF-α, IL-6 and IL-17 were found in cultures from mice that presented the lowest clinical scores. In addition, differently from IFN-γ, TNF-α and IL-6, IL-17 was not produced by spleen cells from normal mice. The highest production of IL-17 was observed in infections caused by the more arthritogenic strains, i.e., ATCC 19095 SEC⁺ and N315 ST5 TSST-1⁺. The role of IL-17 as a mediator of joint destruction is being elucidated in experimental RA. The intra-articular injection of IL-17 into the knee results in joint inflammation and local damage [32]. This effect has been attributed, at least partially, to IL-17 induction of matrix metalloproteinases and also to its ability to promote osteoclastogenesis [34]. IL-17 also induces production of IL-6 and IL-8 by RA synovial fibroblasts via NF-kB and PI3-kinase/Akt-dependent pathways [35]. Furthermore, this cytokine induces production of chemokines and other pro-inflammatory cytokines such as TNF-α, IL-1β, CXCL1 and CXCL5 [36, 37]. These cytokines affect bone remodeling by stimulating proliferation and differentiation of osteoclast progenitors into mature osteoclasts [38, 39]. Contrasting with this well-established role of IL-17 in RA, the role of IL-17 in S. aureus-induced arthritis is not well understood. To our knowledge, this higher production of IL-17 in mice infected with the most arthritogenic strains is being described for the first time. Although this cytokine has been associated with protection in animals immunized with clumping factor A and also with local host defense during S. aureus-induced arthritis, its arthritogenic contribution in SA is still not disclosed. In this sense, these results support the possibility that higher IL-17 inducer S. aureus strains are endowed with stronger arthritogenic abilities [26, 40].