Bacterial Toxins in Musculoskeletal infections

Abstract

Musculoskeletal infections (MSKIs) remain a major health burden in orthopaedics. Bacterial toxins are foundational to pathogenesis in MSKI, but poorly understood by the community of providers that care for patients with MSKI, inducing an international group of microbiologists, infectious diseases specialists, orthopaedic surgeons and biofilm scientists to review the literature in this field to identify key topics and compile the current knowledge on the role of toxins in MSKI, with the goal of illuminating potential impact on biofilm formation and dispersal as well as therapeutic strategies.

The group concluded that further research is needed to maximize our understanding of the effect of toxins on MSKIs, including:

(i) further research to identify the roles of bacterial toxins in MSKIs,

(ii) establish the understanding of the importance of environmental and host factors and in vivo expression of toxins throughout the course of an infection,

(iii) establish the principles of drug-ability of anti-toxins as antimicrobial agents in MSKIs,

(iv) have well-defined metrics of success for anti-toxins as anti-infective drugs,

(v) design a cocktail of anti-toxins against specific pathogens to (a) inhibit biofilm formation and (b) inhibit toxin release.

The applicability of anti-toxins as potential antimicrobials in the era of rising antibiotic resistance could meet the needs of day-to-day clinicians.

Introduction

Musculoskeletal infections (MSKIs) remain a major health burden in orthopaedics ¹. Microbial virulence factors can play a major role in the pathogenies of MSKIs. Among these, toxins are imperative virulence factors. They can be defined as poisonous substances/ molecules that are produced by organisms that function autonomously and interfere with or directly harm the host. From the clinical perspective, bacterial toxins (exotoxins and endotoxins) are generally classified into three main categories:

1. Receptor-dependent or receptor-independent membrane-damaging toxins e.g. cytolytic toxins such as Staphylococcus aureus α-toxin and Panton-Valentine-Leukocidin (PVL)

2. Non-membrane-damaging toxins that interfere with receptor function e.g. toxic shock syndrome toxin (TSST) and a variety of enterotoxins

3. Secreted enzymes. These are enzymes that degrade host molecules or alter host metabolic pathways e.g. protease aureolysin

One strain of bacteria can produce a multitude of toxins simultaneously. Exotoxins are usually proteins, such as bacterial proteases and hyaluronidases that can degrade extracellular matrix or disrupt the proteins that hold epithelial cells together. The most common tissue reaction to a MSKI is suppurative inflammation attracting neutrophils (polymorphonuclear leukocytes, PMN) to the site, and experimental studies with biomaterials have demonstrated intense acute inflammation caused by residual toxins. With respect to osteomyelitis, the morphologic effects of bacterial toxins are essentially those of the infection itself. With respect to periprosthetic infection, numerous studies have evaluated the tissue concentration and distribution of acute inflammatory cells thought to be predictive of bacterial infection.

Other toxins have systemic effects and can induce the release of pro-inflammatory cytokines, including TNF-alpha, IL-6 and IL-1 from cells of the innate immune system, including neutrophils and macrophages. It can also induce differential activation of Th1 and Th2 T-lymphocytes. On one hand, this host response can be beneficial, potentially facilitating bacterial clearance by the host. On the other hand, it can also induce TNF, IL-1 and IL-12 release, potentially culminating in disseminated intravascular coagulation, or septic shock. Many bacterial toxins contribute to the molecular machinery employed in biofilm formation and/or dispersal of microorganisms from mature biofilm.. However, bacteria can possess other non-toxin virulence factors that interplay with toxins and biofilm formation, is a crucial aspect of MSKIs. Staphylococci the most common causative species of MSKIs provide instructive examples of this interplay and hence a large section of this review is devoted to these organisms

Although, bacterial toxins are foundational to pathogenesis in MSKI, they are poorly understood by the community of providers that care for patients with MSKI, inducing an international group of microbiologists, infection disease, orthopaedic and biofilm experts to review the literature in this field and compile the current knowledge on the role of toxins in MSKI, with the goal of illuminating potential impact on biofilm formation, dispersal as well as therapeutic strategies for MSKIs.

Methodology

A biofilm symposium was held at the Hospital for Special Surgery in New York, NY, USA on August 1^st 2019, and the contents of the discussion on bacterial toxins in MSKI was discussed with the Editor-in-Chief of the Journal of Orthopaedic Research. Following a presentation on the topic (by KS) in the symposium, experts who attended the meeting were invited to participate and come up with key topics related to this area and contribute to a review manuscript. The willing participants (co-authors) identified topics related areas of interest and unknowns in the field of toxins and MSKIs. Contributors provided draft topics to the lead author (KS), a manuscript was drafted and sent back to all authors for further comments, suggestions and agreement was obtained on the final version of the article before submission.

This work focuses mainly on toxins that play important roles in pathogenesis and present treatment opportunities. Additionally as Staphylococci, the most common causative species of MSKIs, provide instructive examples of toxin-biofilm interplay and a large section of this review is devoted to these organisms.

Results

The following key topics were selected and reviewed:

1). Staphylococcal species

1.A. Staphylococcal Toxins and MSKIs

Staphylococcus aureus [methicillin-sensitive (MSSA); methicillin-resistant (MRSA) S. aureus] and coagulase-negative staphylococci (CoNS) belong to the most common causes of MSKIs. Although S. aureus infection can be associated with florid inflammation and CoNS with a low-grade presentation, the organisms can reveal a converse clinical pattern. Moreover, S. aureus can still be isolated in reactivated osteomyelitis after decades of remission ².

There are a multitude of S. aureus virulence factors with activities such as degradation of antimicrobial peptides and inhibition of complement mediated destruction. Virulence factors include the microbial surface components recognizing adhesive matrix molecules (MSCRAMMs) that allow interactions with proteins such as collagen, fibronectin, and fibrinogen, and are integral to biofilm formation along with staphylococcal autolysins. These proteins perform critical roles in their interaction with the extracellular matrix (ECM) and serum proteins by either coating implants or contributing to aggregation in the synovial fluid ³ to result in prosthesis- or tissue-adherent biofilms or floating biofilms formed on a fibrous proteinaceous matrix ⁴.

Nearly all strains of S. aureus and S. epidermidis secrete the four hemolysins (alpha, beta, gamma and delta), lipases, proteases, hyaluronidase, nucleases, and collagenase. The main functions of these toxins, in MSKI, are to break down host tissue and provide nutrients for bacterial survival and growth (table 1) ^5–18.

Table 1:

some Staphylococcal toxins and impact on bone and joint infections

Toxin	Action in relation to bone and joint infections	Comments*
α-hemolysin	Osteocyte necrosis	A small β-barrel pore-forming cytotoxin, is the most well studied member of staphylococcal hemolysins. It lyses red blood cells and leukocytes, but not neutrophils via binding to its proteinaceous receptor; a disintegrin and metalloproteinase (ADAM10). It has been shown to contribute to the quiescent and latent nature of recurrent osteomyelitis.
Panton-Valentine-Leukocidin (PVL)	Infection persistence and spread	A pore forming toxin, its production is associated with rapid and extensive spread of osteomyelitis in murine models. A study focused on the molecular characterization of a S. aureus strain isolated from an implant-related knee infection retrospectively showed by genotyping that the strain was positive for the PVL toxin gene. This was the first report of a PVL-positive S. aureus strain associated with an orthopaedic implant infection.
S. aureus toxic shock syndrome toxin 1 (TSST-1)	Osteoclastogenesis	A super antigen, is not cytotoxic to osteoclasts, but it has been shown to increase bone resorption through stimulation of osteoclasts and contributes to the loss of bone strength due to matrix resorption.
Phenol soluble modulins (PSMs)	Osteoblast cytotoxicity and biofilm dispersal	In S. epidermidis, the identification and characterization of the PSMs brought insight to the toxin-pathogenesis of this organism. PSMs are positively regulated by the global regulatory quorum sensing system, accessory gene regulator (agr). PSMs are α-helical peptides that serve as cytotoxins and play a critical role in biofilm detachment and regrowth, allowing bacteria-laden biofilm to disseminate to distant sites11. PSMβ peptides are the primary PSMs produced by S. epidermidis, are expressed at high levels during biofilm formation, and have been found to be crucial for the detachment of biofilm clusters in vitro and dissemination from biofilms on indwelling devices in vivo, as demonstrated in a mouse model of indwelling device-associated infection 12. PSMs also provide biofilm with the ability to expand and stabilize on implanted devices, such as orthopaedic implants, through the production of amyloid fibers. S. epidermidis-derived PSMδ is cytolytic against neutrophils in vitro. Additionally, some S. epidermidis strains have been shown to produce PSM-mec, which has cytolytic potential against neutrophils in vitro and its presence has been associated with decreased bacterial clearance and higher mortality rates in a murine model of sepsis. The role of PSM in vivo needs to be further studied as this may open potential therapeutic strategies to treat MSKIs.
Canonical coagulase (Coa)	Inhibits osteoblast function	Together with PVL toxin and microbial surface components recognizing adhesive matrix molecules (MSCRAMMs), Coa inhibits osteoblast proliferation, induce osteoblast apotosis, prohibits bone formation and mineralization and triggeres osteoclast-induced bone resorption and bone destruction by upregulated RANK-L expression.
Aureolysin	Osteoblast necrosis	A S. aureus protease that cleaves a number of proteins and inactivates PSMs which have a major impact on the pathogenesis of osteomyelitis.

^*Adopted from reference 5–18

1.B. Factors associated with the aggregation of S. aureus in synovial fluid

Biofilm-associated MSKI is a clinically challenging ¹⁹, the biofilm that is formed in synovial fluid (rich in serum and ECM proteins) is unlike the biofilm observed in vitro. It is defined by the binding of ECM/serum proteins by bacteria to form fibrinogen-containing biofilms/aggregates ²⁰. Aggregated bacterial biofilm that forms with S. aureus and other bacterial strains in multiple types of synovial fluid (bovine, porcine, equine, human; normal, diseased). As fibrin[ogen] is abundant in diseased joints ²¹, it appears to be the key component of the aggregate. Pre-digestion of the synovial fluid with plasmin blocks S. aureus aggregation, ²⁰ and treatment of existing aggregates by tissue plasminogen activator (TPA) seems to disperse aggregates ²², suggesting that interactions with serum proteins such as fibrinogen are important for aggregation. In fact, when the bacterial proteins responsible for binding to the serum proteins fibrinogen, albumin, and fibronectin (fibronectin binding protein A and B (FnbA and FnbB) and clumping factor A and B (ClfA and ClfB)) are inactivated due to gene insertions, the ability of MRSA to form aggregates in synovial fluid is attenuated or eliminated ²³.

When Staphylococci reach a critical mass at an infection site, cell-to-cell communication via peptide signaling can activate a series of cellular processes by a mechanism known as “quorum sensing”. Mutations in the MSCRAMMs can moderate the ability of S. aureus to cause disease infection ²⁴. The accessory gene regulator (agr) system controls MSCRAMM expression and regulates the quorum sensing system along with the P2 and P3 promoters. In turn, they regulate a variety of cell surface and secreted proteins, notably the PSMs that are critical to the aggregatory phenotype²³. In synovial fluid, agr and its downstream effectors the P2 and P3 promoters, as well as PSMα and β show reduced expression ²³, showing that interactions with the synovial fluid actively modify the phenotype of the bacteria. Even in synovial fluid, aggregation can be reduced by over expression of the surfactant-like PSMα or PSMβ in MRSA USA300, suggesting that this agr-mediated suppression is critical in aggregation. Interestingly, preclinical studies have confirmed that agr is required for S. aureus emigration from implant-associated biofilm, but is not required for colonization of osteocytic-canalicular networks during chronic osteomyelitis ²⁵.

Because aggregation of S. aureus within synovial fluid alters both quorum sensing and virulence factor/toxin production, it seems likely that the MRSA/MSSA aggregate represents a less virulent infection. With bacterial release from the aggregate, the polysaccharide intracellular adhesin (PIA) coating of the aggregate is released, and there should be a marked increase in virulence factor/ toxin expression—with an accompanying increase in antibiotic sensitivity. Thus, disaggregation resulting in increased agr-mediated gene expression of the deleterious virulence factors/toxins restores antibiotic susceptibility and by implication, the possibility of bacterial eradication.

1.C. Biofilm and exotoxin interactions

S. aureus has an inverse relationship between biofilm formation and exotoxins production/ secretion and other virulence factors. There is evidence for both direct and inverse relationships that potentially allows S. aureus to fine-tune its survival in the host. S. aureus biofilm can persist in multiple phenotypes by which it can evade detection by the host and clinical microbiology or attack the host during episodes of proliferation.

These phenotypes have been associated with phases of infection severity. Five distinct stages in the development of biofilm formation by S. aureus have been identified (figure 1) ²⁶:

Figure 1:

Current paradigm of establishment of Staphylococcus aureus Peri – prosthetic joint infection (modified from reference²⁴)

1. attachment of cells to a surface

2. multiplication of these cells into a thin layer and production of an extracellular polymeric substance (EPS), or slime, matrix consisting of proteins and extracellular DNA (eDNA) which holds the cells together and anchors the developing biofilm to the surface

3. exodus in which some cells are released through degradation of the eDNA by enzymatic extracellular nuclease activity

4. maturation into larger microcolonies which triggers the density dependent agr quorum sensing system to begin to produce proteases which break down the proteinaceous component of the EPS and PSMs (which have a surfactant effect) that leads to the final phase

5. dispersal in which cells and aggregates are released from the biofilm into the surrounding fluid.

Suppression of PSMs has been associated with the ability of S. aureus to form large suspended aggregates in synovial fluid, with an antibiotic or antimicrobial tolerant biofilm-like phenotype ²³.

The staphylococcal accessory regulator (SarA) protein, has similarly been shown to regulate the production of secreted virulence factors such as α-toxin while repressing biofilm formation ²⁷. Interestingly the saePQRS regulatory system, a third system that regulates biofilm formation and toxin production (including α-toxin) has been shown to promote biofilm formation by limiting the production of extracellular proteases that degrade the biofilm EPS ²⁸.

Although the above mentioned 5-stage model suggests a linear timeline of development, it is not clear how long each stage may persist in Periprosthetic joint infections (PJI). It is possible that the biofilm can cycle through the more symptomatically quiescent stages of multiplication to the more aggressive “flare-ups” caused by the dispersal stage. Such cycling could explain anecdotal evidence that suggests that the biofilm can reside in the quiescent stages for months or even years before manifesting as an initial or recurrent severe localized or systemic infection ².

S. aureus α-toxin (table 1) contributes to the formation of biofilms on mucosal surfaces by disrupting the host epithelium ²⁹.β-toxin of S. aureus is a neutral sphingomyelinase (SMase). β-toxin plays a key role in the establishment of staphylococcal biofilms. However, S. aureus biofilms can be made by many strains that do not produce β-toxin, requiring other factors dependent upon the specific strain. These factors include cell surface virulence factors, extracellular polysaccharides, other proteins, and extracellular DNA, all pointing to a molecular mechanism for forming the structural framework of many staphylococcal biofilms.

1.D. Influence of the environment and surfaces on the production of staphylococcal exotoxins

The detection of multiple toxins (HlgB, LukD/E and α-toxin) in biofilms of multiple strains on Leiden human epidermal models (LEMs), but not on polystyrene surfaces (PS), indicates surface specific protein expression. This suggests that currently used biofilm models, might not adequately reflect biofilm formation on more complex surfaces, such as the human skin. The detection of S. aureus toxins, most notably α-toxin, in biofilms on LEMs is in agreement with their well-established roles in the pathogenesis of skin infections. The cytolytic pore-forming α-toxin lyses human cells, interferes with the innate and adaptive immune responses in a murine skin infection model, and is essential for biofilm development on mucosal surfaces. In human skin, the filaggrin protein may inhibit α-toxin’s cytotoxicity by its ability to regulate the secretion of SMase. Other toxins detected including gamma-haemolysin component B precursor (HlgB) and the leukocidins D and E (LukD and LukE), have also been associated with S. aureus skin colonization and infection. Regulation of Hla and other genes for S. aureus virulence factors is influenced by many factors, including the agr, RNAIII, downstream transcription factors Rot, SarA and SarS and Sae. Therefore, it may well be that the quorum sensing system of AgrA/RNAIII of S. aureus is activated differently upon interaction of S. aureus with either LEMs or PS, leading to up- or down-regulation of specific genes depending on the surface ³⁰.

S. aureus has intricate layers of regulation that control the release of virulence factors. The agr regulatory system controls a myriad of secreted toxins, and identifying the exact agents responsible for dispersal has represented a challenge. Early studies suggested that surfactant properties of delta-toxin could be responsible considering that many bacterially produced surfactants have known anti-biofilm effects, as it became evident that delta-toxin was part of a larger family of PSM peptides. Numerous environmental and metabolic factors such as pH, reactive oxygen species, and nutrient availability, can modulate agr quorum sensing in S. aureus. The mechanisms by which these factors impact quorum sensing are diverse, but can be organized roughly into three broad categories: (i) suppression of quorum sensing through the action of reactive oxygen species, generated by innate immune cells on components of the agr system; (ii) transcriptional up- or down- regulation of agr expression as a consequence of various non-agr regulatory proteins binding within the agr P2-P3 promoter region; and (iii) the up- or down-regulation of agr expression by environmental factors for which the mechanism of action is yet to be identified ³¹.

1.E. Understanding Bacterial Toxins, Humoral Immune Responses in MSKIs: potential diagnostic and therapeutic strategies?

S. aureus’ repertoire of toxins and proteases targets elements of the innate and adaptive immune responses. For example, a recent report described the priming of regulatory T-cells initiated by the action of PSMs on human dendritic cells ³²; this is of importance in MSKIs. PSMs have been shown in murine and human models to modulate human monocyte-derived dendritic cells to prime regulatory T cells, thus muting the local adaptive immune response and increasing tolerance to the bacteria ³². One possible mechanism for this phenomenon is the ability of PSMs to block the p38-CREB pathway in dendritic cells, impairing cytokine production, and therefore leading to regulatory T cell priming ³³.

Humans express IgG antibodies against numerous S. aureus antigens in their serum. During acute MSKIs, antibody levels for specific antigens increase 2 to 10 fold ³⁴. Most of the antibodies are specific for the secreted products/ enzymes displayed on the cell wall ³⁵. Further, these antibodies are maintained at relatively high levels even in non-infected patients, but it is not clear that they provide protection against new infections. How and if patients clear S. aureus infections, and the role played by circulating antibodies, remain topics of considerable debate, but there are three practical roles for antibodies in patient care:

1. Utility of antibodies as diagnostic biomarkers. Antibody levels do increase in new infections, and that increase can be utilized as a measure of ongoing infection. In MSKIs, serum antibody levels for particular antigens, especially Iron-regulated surface determinant system-B (IsdB), rise substantially providing a possible diagnostic tool ³⁶. However, the utility of this approach is limited by: a) the huge range of antibody levels in non-infected patients; and b) the fact that, once elevated, anti-S. aureus antibody levels decline slowly over months. Similar diagnostic power can be obtained using the recently stimulated circulating plasma cell population, antibody-secreting cells (ASC), which can be cultured in vitro to express their antibodies into a cell culture fluid termed medium enriched for newly synthesized antibodies (MENSA) ^37–39. For diagnostic purposes, MENSA has a major advantage, because there is no pre-existing antibody from prior exposures, so the analytic background is practically zero.

2. Utility of antibodies as prognostic tools. The presence of ASC in the circulation is a measure of the ongoing immune response. ASC actually emerge earlier than the rise of antibody levels in the serum. They decline once the pathogen is defeated making them potential tools for monitoring response to therapy.

3. Utility of antibodies as therapeutic agents. It is uncertain if and why the abundant anti-S. aureus antibodies in most patients’ sera provide so little protection against infection. Numerous investigators have brought forward new candidate antibodies for passive immunological protection and antigens for active vaccines. Each has demonstrated potential in animal models, but none has exhibited utility in human trials. However, antibody-based therapeutics continue to be pursued. The identification of ADAM10 as a cellular receptor for α-toxin has provided keen insight on the biology of the toxin’s action during disease pathogenesis, demonstrating the molecular mechanisms by which the toxin causes tissue barrier disruption of host interfaces lined by epithelial or endothelial cells. The use of anti-α-toxin has been shown to induce passive immunity against pneumonia in a mouse model ⁴⁰. In addition, the use of small molecular ADAM10 inhibitors have shown promise in experimental models in sepsis ⁴¹.

In MSKIs, the potential for biofilm disruption with antitoxin is being studied ⁴². Testing of the 4-C Staph vaccine demonstrated strong antibody response and reduction in S. aureus bacterial load in a murine model of septic arthritis ⁴³. It is worth mentioning that the clinical trial for the staphylococcal vaccine from Pfizer had to be stopped, because of the low statistical probability for the study to meet primary efficacy objectives and not safety issues according to the conclusion of the Data Monitoring Committee.

Furthermore, animal studies has suggested that the agr system is important in MSKIs ⁴⁴. Conversely, Vuong et.al,. demonstrated that an agr-deficient strain of S. epidermidis showed enhanced biofilm development in an infected implant rabbit model ⁴⁵. They also studied 53 S. epidermidis strains isolated from patients with PJIs and 21 strains isolated from the skin of healthy individuals. Strains that showed a non-functioning quorum sensing system, suggesting a non-functioning agr system, were found in 36% of PJIs and only in ~ 5% of the healthy samples. Valour et.al., performed a similar evaluation of S. aureus isolates obtained from acute and chronic cases of human MSKIs. Isolates were screend for delta-toxin production, which is a product of the RNA III transcript that is up-regulated by the agr system, and the authors found a lack of delta-toxin production in ~26% of chronic infections compared with ~11% of acute infections. No difference was noted in the virulence gene distribution or biofilm formation between the acute and chronic groups ⁴⁶. Such data potentially highlight the difference between in-vitro and in-vivo studies. Alternatively, it may simply highlight the changing necessity of various factors/systems in the establishment of biofilms. The latter idea has particular therapeutic implications e.g., treatment with RNA III-inhibiting-peptide may inhibit/prevent biofilm formation in the early stages of staphylococcal infection, but not in chronic infections.

There are also monoclonal antibodies that, by neutralizing specific targets such as α-hemolysin ⁴⁷ or the cell wall enzyme glucosaminidase ⁴⁸, may augment host defences. Combinations of antigens or monoclonal antibodies have been sought in frank recognition of the ability of S. aureus to shift its antigenic repertoire ⁴⁹. Finally, there is the hope of introducing a new class of recognition molecules called centyrins that can be selected from variants built on an immunoglobulin-like domain of the human protein tenascin-C ⁵⁰.

2. Streptococcus species

Beta-haemolytic streptococci are the most commonly described streptococcal species in MSKIs. While group A Streptococcus (GAS) or S. pyogenes are prevalent in children, group B Streptococcus (GBS, S. agalactiae) play an increasing role in elderly/ adults with comorbidities. In the past decades, a growing incidence of MSKIs due to group C and G Streptococcus (GCGS, S. dysgalactiae sup. equisimilis), in the elderly, has been noted. However, molecular investigations have not yet identified a specific factor that could explain the arthritogenicity of GCGS.

GAS and GCGS have several virulence determinants in common, including Streptolysin O and S, Streptococcal pyrogenic exotoxins [Spe] and M protein. M proteins and pili-like structures play a role in adherence to host cells and biofilm formation. M protein, SpeA and SpeB induce human mononuclear cells to produce TNF-α, IL-1 and IL-6. Thereby, they trigger a severe inflammatory response, eventually leading to the destruction of chondrocytes and osteoblasts. In a mouse model, GAS infection of osteoblasts induces receptor activator of NF-kappa-B-ligand (RANKL) expression ⁵¹ that contributes to bone destruction. GBS consists of several virulence factors that are involved in the pathogenesis of MSKIs; β-hemolysin/cytolysin (β-h/c) toxin is associated with pigment expression. It is hemolytic and cytolytic due to its pore-forming capacities ^52–54 and triggers cytokine release. Hyperhemolytic GBS strains are considered to be frequently, but not uniformly more pathogenic compared to nonhemolytic strains ^55–57.

3. Cutibacterium acnes

Cutibacterium acnes (C. acnes) [formerly Propionibacterium acnes] is often an indolent cause of MSKIs. There is a possibility that pore-forming toxin, like the α-hemolysin of S. aureus, contributes to the pathogenesis and/ or symptomatology of chronic C. acnes vertebral disc infections. Hyperhemolysis on blood agar has been proposed as a microbiological marker for MSKIs ^58–61.

4. Clostridium perfringens

Gas gangrene (Clostridial myonecrosis) is a life-threatening infection, typically, but not exclusively caused by Clostridium perfringens. Among the toxins that are produced by C. perfringens, the alpha and theta-toxins are known to be major virulence factors. The α-toxin is haemolytic and has phospholipase C and SMase activities. It inhibits the differentiation of neutrophilic granulocyte cells and impairs erythropoiesis. The theta-toxin, also known as perfringolysin O, is a member of the cholesterol-dependent cytolysin family and is pore-forming. MSKIs caused by Clostridium spp. are rare, but include septic arthritis, PJI, postoperative wound infections and spondylodiscitis.

In comparison to gas gangrene that is characterized by the destruction of muscle accompanied with organ failure and shock, clostridial arthritis presents with typical clinical features of pyogenic joint infection. Radiological signs of intra-articular gas are infrequently reported. In fracture-related infections caused by Clostridium spp., there is often involvement of multiple organisms (i.e.; polymicrobial infections) and these patients have a higher likelihood of complications, including delayed wound healing or development of a septic non-union.

Despite the difference in clinical presentation between gas gangrene and osteoarticular infection, there are pathogenic similarities in these clostridial infections, mainly triggered by their toxins. In an experimental model, intramuscular injection of a clostridial toxin preparation containing both phospholipase C and theta-toxin decreased the blood flow rapidly and irreversibly via formation of intravascular aggregates that obstructed the blood vessels. The toxins worked synergistically for bacterial survival and the spread of disease in gas gangrene ⁶². The perfusion deficits led to an anaerobic environment, which was an ideal niche for bacterial persistence. This process possibly prevents bone healing and contributes to tissue necrosis, and eventually to the development of a non-union in fractured bones. However, the effect of C. perfingens toxins on osteoblasts and chondrocytes have not yet been reported. The absence of tissue inflammatory response is another characteristic observed in clostridial infections ⁶³. In addition to the hindered blood flow which impairs the movement of inflammatory cells to the infected tissue, alpha and theta-toxin also destroy leukocytes due to their cytotoxic properties ⁶⁴.

5. Kingella kingae

Kingella kingae is an increasingly identified cause of MSKIs typically in children less than 4 years of age. The true incidence of paediatric K. kingae septic arthritis is unknown, as the clinical presentation may mimic transient synovitis or rheumatologic conditions, and the organism is often not detected by routine culture techniques. While pili play a role in adherence to host cells, K. kingae ’s RTX toxin may help the invasion to skeletal tissues and has a wide-spectrum of cytotoxic effects, including to synovial cells ⁶⁵. Additionally, marked reduction in the circulating white blood cells could be a strategy to guarantee K. kingae’s subsistence in the host’s bloodstream, skeletal tissues, and endocardium. Disruption of the K. kingae RTX locus resulted in a loss of cytotoxicity for respiratory epithelial, synovial, and macrophage cell lines, and the absence of RTX toxin in species like K. denitrificans and K. oralis rendered them less virulent ⁶⁶.

6. Pasteurella multocida

Pasteurella multocida toxin (PMT) is a potent toxin involved in bone destruction. Experiments in animal models demonstrated that PMT manipulate host cell signaling cascades resulting in the activation of mitogenic pathways and the production of cytokines for many cell types; leading to excessive bone degradation.

PMT-mediated bone damage is independent of RANKL, a cytokine thought to be crucial for triggering osteoclastogenesis. When RANKL plus osteoprotegerin are added to PMT-treated macrophages, no effect on PMT-induced osteoclast formation is observed ⁶⁷. PMT mediates the release of osteoclastogenic cytokines such as IL-6 and TNF-α, but not IL-1, which supports the differentiation process. An in-vitro study using etanercept (an anti-TNF compound) strongly reduced PMT-induced osteoclast formation. Similarly blocking IL-6 also led to a reduction in PMT-induced osteoclast formation in macrophages ^68–70, these may be used as potential therapeutic strategies in the management of P. multocida osteomyelitis.

Conclusion and future directions

We have presented here a number of topics in a thought-provoking manner related to bacterial toxins and their roles in MSKIs. However, the role of toxins, other virulence factors and host immunological factors in MSKIs and for other organisms (e.g. Salmonella spp., Mycobacteria spp., Neisseria gonorrhoea, Borrelia burgdorferi) have not been discussed in this review. There are many unknowns on the effect of toxins on osteocytes, chondrocytes, biofilm formation and MSKIs and persistent infections. Identification of the interaction between toxins and antitoxins at the molecular level coupled with the interaction of human immune responses will be critical to the understanding of the impact in pathogenesis and the production of next-generation antimicrobial molecules, anti-toxins and vaccines to manage MSKIs.

Further studies, to maximize our understanding of the effect of toxins on MSKIs, are needed including:

1. research to identify the roles of bacterial toxins in MSKIs,

2. establish the understanding of environmental and host factors and in vivo expression of toxins throughout the course of an infection,

3. establish the principles of drug-ability of antitoxins as antimicrobial agents in MSKIs,

4. have well-defined metrics of success for anti-toxins as anti-infective drugs,

5. design cocktail(s) of anti-toxins against specific pathogens to (a) inhibit biofilm formation and (b) inhibit toxin release. This may also reduce or eliminate impact on selection pressure or the microbiome compared to traditional antibiotics.

The applicability of anti-toxins as potential antimicrobials in the era of rising antibiotic resistance could meet the needs of day-to-day clinicians.

Glossary

ADAM10

A disintegrin and metalloproteinase

agr

Accessory gene regulator

ASC

Antibody-secreting cells

Clf

Clumping factor

ECM

Extracellular matrix

eDNA

Extracellular DNA

EPS

Extracellular polymeric substance

FnbA or B

Fibronectin binding protein A or B

GAS

Group A Streptococcus

GBS

Group B Streptococcus

GCGS

Group C and G Streptococcus

GGS

Group G Streptococcus

Hla

Alpha-hemolysin

IsdB

Iron-regulated surface determinant system-B

LEMs

Leiden human epidermal models

Luk

Leukocidins

MENSA

Medium enriched for newly synthesized antibodies

MSCRAMMs

Microbial surface components recognizing adhesive matrix molecules

P2 or P3

Promotors

p38-CREB

p38-cAMP response element binding

PIA

Polysaccharide intracellular adhesion

PMT

Pasteurella multocida toxin

PSMs

Phenol soluble modulins

RANKL

Receptor activator of NF-kappaB ligand

RTX

Repeat-in-toxin

SaePQRS

S. aureus exoprotein regulatory system

SarA

Staphylococcal accessory regulator

SMase

phingomyelinase

Spe

Streptococcal pyrogenic exotoxins

TPA

Tissue plasminogen activator

TSST-1

Toxic shock syndrome toxin 1