Vanderbilt University



Discoveries Featured

Dissecting the Molecular Pathogenesis of Bone Infection by Staphylococcus aureus

By: Carol A. Rouzer, VICB Communications
Published: July 2, 2013

A new model of S. aureus osteomyelitis reveals a protease that modulates the levels of multiple bacterial virulence factors.

Capable of colonizing nearly every organ of the body, Staphylococcus aureus (Figure 1) causes a wide range of infections. One of the most difficult to treat is infection of the bone, osteomyelitis, which often requires multiple surgical debridements accompanied by a protracted course of antibiotics. The extensive destruction of bone that occurs during S. aureus osteomyelitis compromises the circulation to the site of infection, hindering the delivery of antibiotics to the site. Consequences of these infections can be severe, including septicemia, venous thrombosis, and pathological fractures. To more effectively combat this dangerous S. aureus infection requires a better understanding of the mechanisms by which the bacteria invade and destroy bone, leading Vanderbilt Institute of Chemical Biology (VICB) member Eric Skaar and his laboratory to develop a new mouse model of osteomyelitis. They now report the discovery of a key enzyme that regulates bacterial virulence and bone remodeling during infection by S. aureus [J. E. Cassat, et al. (2013) Cell Host & Microbe, 13, 759].

Figure 1. Electron micrograph of S. aureus bacteria (gold) escaping from a white blood cell (blue). Figure is reproduced from NIAID and is in the public domain.

To establish a model of osteomyelitis in mice, the Skaar lab created a 1 mm defect in the femur of each anesthetized mouse, and then injected S. aureus bacteria through the defect into the bone marrow cavity. Progress of the infection was confirmed by monitoring bacterial cell count in the bones, which peaked at 4 days after injection. By 14 days after injection, the infected bones exhibited marked destruction of bone at the injection site, with reactive new bone formation at the periphery. Quantification by microcomputed tomography imaging (Figure 2) indicated a loss of 10 to 20% of cortical bone near the focus of infection and new bone growth equivalent to 30 to 50% of the original cortical bone volume. These findings were confirmed by histologic examination, which also revealed numerous S. aureus abscesses in the bone marrow.

Figure 2. A Staphylococcus aureus-infected murine femur (left, gray) was subjected to microCT imaging analysis. Volumetric renderings of new bone (green) and cortical bone destruction (yellow) were created to quantify pathogen-induced changes in bone remodeling. Right: a concatenated image showing microCT renderings overlying the infected murine femur. Image kindly provided by Eric Skaar, copyright 2013.

One contributing factor to the extensive bone remodeling observed in S. aureus infection is the bacteria’s ability to invade and persist within osteoblasts, the cells primarily responsible for new bone formation. Infected osteoblasts secrete inflammatory cytokines and RANK-L (receptor activator of NF-κB ligand), which activates bone-destroying osteoclasts. The infection eventually kills the osteoblasts, so the result is an imbalance between the numbers of osteoclasts and osteoblasts in the bone, resulting in net bone destruction. To better define the factors responsible for the pathologic changes in S. aureus-infected bone, the Skaar lab investigated the ability of bacteria carrying mutations in the agr and sae loci (Δagr/sae) to establish osteomyelitis in their model. The agr and sae loci code for proteins that regulate the production of exoproteins by the bacteria. As expected, the double mutants exhibited marked reductions in exoprotein production. They also exhibited marked reductions in their ability to infect bone, as indicated by less bone destruction, less new bone formation, and lower bacterial burdens.

These initial results indicated that exoprotein production controlled by the Agr and/or Sae systems was important for S. aureus-dependent bone infection and remodeling. Further studies with bacteria bearing mutations at only one of the loci indicated that Sae was the more important of the two regulators. To better understand the role of Sae-modulated exoproteins in osteomyelitis, the investigators explored the effects of exoproteins present in supernatants from bacterial cultures on the viability of osteoblasts from mice (MC3T3 cells) and humans (Saos-2 cells). The results showed that concentrated supernatants from wild-type (WT) S. aureus cultures were highly toxic to the cells, while those from Δsae bacteria were nontoxic. These findings were consistent with the reduced infectivity of Δsae bacteria. However, a surprising finding was that Δsae supernatants that had not been concentrated were toxic. This suggested that Δsae bacteria produce cytotoxic species, but that these are subject to degradation during the concentration process.

To further explore the basis for the cytotoxicity results, the investigators performed multidimensional protein identification technology (MudPIT) analyses on the concentrated WT and Δsae supernatants. The results revealed significant reductions in the levels of 49 proteins in Δsae as compared to WT concentrated supernatants. These included virulence factors, cytotoxins, immunomodulatory molecules, and exoenzymes. Increased levels of 31 proteins were also observed, with the most notable change (8-fold) in aureolysin, a secreted protease. This finding led to the hypothesis that high levels of aureolysin secreted by Δsae cells lead to degradation of a large number of virulence factors, resulting in reduced cytotoxicity of the concentrated Δsae supernatants. Consistent with this hypothesis, MudPIT analysis of concentrated supernatants from cells lacking both sae and the gene for aureolysin (Δsae/aur) showed increased levels of 230 proteins and decreases in 51 proteins when compared to WT supernatants. These results indicate that the major loss of proteins in the concentrated Δsae supernatants occurred through an aureolysin-dependent mechanism.

To determine if aureolysin contributes to the reduced infectivity of Δsae bacteria in vivo, the Skaar lab tested the ability of Δsae/aur mutant bacteria to establish osteomyelitis in their mouse model. They found that compared to the Δsae strain, Δsae/aur bacteria cause greater levels of bone destruction, though they are not as destructive as WT S. aureus. Deletion of aur did not restore the low bacterial count observed in Δsae-infected as compared to WT-infected bones. The results confirm that aureolysin plays a role in bone remodeling observed in S. aureus infection, though other factors must control infectivity.

A comparison of the exoproteins in the concentrated supernatants of WT, Δsae, and Δsae/aur S. aureus indicated the presence of alpha-type phenol-soluble modulins (PSMs) in samples from WT and Δsae/aur, while these peptides were either undetectable or present in very low amounts in Δsae bacterial supernatants. PSMs, which are lytic and chemotactic for neutrophils, contribute to S. aureus virulence. Concentrated supernatants from bacteria genetically deficient in PSMs were not toxic to human or mouse osteoblasts, whereas the purified PSM peptides were cytotoxic. Consistently, PSM-deficient S. aureus demonstrated reduced bone destruction, compared to WT bacteria, in the in vivo osteomyelitis model.

Together the results indicate that the protease aureolysin plays a key role in regulating levels of exoprotein virulence factors, such as PSMs, that mediate the extensive bone remodeling observed in S. aureus osteomyelitis. The investigators note that aureolysin activity is dependent on zinc, which is often present in reduced concentrations at the site of bacterial infection. Inactivation of aureolysin through zinc ion deficiency would be expected to reduce degradation of virulence factors such as PSMs, thereby increasing the destructive nature of the S. aureus infection. Clearly, further work is needed to completely characterize the pathogenesis of osteomyelitis, but the availability of the new mouse model reported by the Skaar lab promises to greatly facilitate this effort.























Vanderbilt University School of Medicine | Vanderbilt University Medical Center | Vanderbilt University | Eskind Biomedical Library

The Vanderbilt Institute of Chemical Biology 896 Preston Building, Nashville, TN 37232-6304 866.303 VICB (8422) fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2013 by Vanderbilt University Medical Center