Adherence of Staphylococcus aureus in experimental endocarditis and osteomyelitis

Sammanfattning: Staphylococcus aureus has persisted and is now resurging as an important hospital and community pathogen. The microorganism can cause a wide spectrum of diseases ranging from cutaneous infections to more specific infections such as osteomyelitis and infective endocarditis (IE). Infections of bone, or osteomyelitis, remain a clinical challenge. With the difficulty of treatment and the emergence of resistant pathogens in mind, recent research has focused on the pathogenesis and prophylaxis of such infections. Infections of bone may result from hematogenous or from direct introduction of microorganisms. Hence, a susceptible focus and the presence of microorganisms are needed to initiate osteomyelitis. In contrast to osteomyelitis, in which a susceptible focus can be produced by trauma or surgical manipulations of the bone, IE is mainly related to various defects on cardiac valves emanating from clinical conditions such as rheumatic fever, prosthetic valve implants congenital heart disease, and degenerative changes. Such defects cause sterile vegetations that can become colonized with bacteria and lead to IE. In both IE and osteomyelitis, a frequently isolated microorganism is S. aureus and it has been shown that a virulence factor important for colonization is its ability to bind specifically to tissue proteins. The work presented here tries to define the relationship between bacterial adhesion, and bone injury in osteomyelitis and the importance of particular bacterial matrix-protein interaction in IE. Recently, a mutant of S. aureus was constructed containing inactivated genes for the two fibronectin binding proteins and incapable to bind to fibronectin. This mutant and its parent strain were compared in a rat model of catheter induced IE. The rates of infection after one or 24 hours were found to be same for the strains at different challenge doses, indicating that binding in IE is multifactorial. A mutant of S. aureus, containing an insertional inactivation of the gene for the collagen binding protein was documented to differ only with respect to its ability to bind to collagen and was compared with its parent strain in the above mentioned catheter induced endocarditis model. The results showed that the collagen binding strain had significantly outnumbered the mutant in rats that were sacrificed 24 hours post-challenge. Hence, collagen binding of S. aureus is important in the maintenance of experimental IE. Contrary to IE, in which bacterial adherence to host-proteins has been implicated with the initiation of infection, a clear link between the ability of microorganisms to attach to proteins and development of osteomyelitis has not yet been established due to the lack of an appropriate experimental model. We developed a new model of hematogenous osteomyelitis in the rat that mimics human disease in its presentation. The new model involves the surgical manipulation of the tibia or the mandible of the rat and subsequent intravenous inoculation of S. aureus and the assessment of bone samples microbiologically, radiographically, and histopathologically. The two isogenic strains differing in their ability to bind to collagen were used in a modification of our model of hematogenous osteomyelitis. Histopathologic examination revealed that lack of collagen binding of S. aureus does not reduce the microorganism's virulence in experimental osteomyelitis. Using a mutant with an inactivated gene bears the assumption that the phenotypic changes observed are limited to the targeted gene and its protein. We tried to identify the changes caused by the insertional inactivation of the gene for collagen binding in S. aureus strain PHIOO with respect to its parent strain Phillips. We found that PHIOO had a markedly reduced amount of fibronectin binding protein compared to its parent strain and a diminished ability to aggregate in the presence of fibronectin. We concluded that insertional inactivation of a surface protein can cause pleiotropic effects on the phenotype of the mutant and therefore need to be assessed whenever such strains are being applied in an experimental setting. ISBN 91-628-2090-7