Invasion of host cells by S. aureus is likely to play an important role during colonisation and infection. Studies have shown S. aureus invasion to occur via an interaction between its fibronectin binding proteins (FnBPs) and the host integrin a5~J utilising fibronectin (Fn) as a bridging molecule. The reports of S. aureus invasion vary in a number of ways and a number of factors are likely to influence invasion. The aim of this study was to investigate factors involved in determining S. aureus invasion of epithelial cells. The presence of a stable sub-population with enhanced invasive capability has previously been identified for the oral bacterium Porphyromonas gingivalis. In this study consecutive antibiotic protection assays were used in an attempt to determine whether such a sub-population structure exists in S. aureus cultures. Initial findings indicated that S. aureus cultures may contain an invasive sub-population but that this was transient and not stable. This was later attributed to technical aspects of the assay, in particular variation in multiplicity of infection (MOl) between the first and second rounds of invasion. When a MOl of 100 was maintained throughout, S. aureus cultures did not appear to contain invasive sub-populations. This work also highlighted a number of other methodological factors in the antibiotic protection assay, which is most commonly used, including intracellular growth, growth phase of bacteria, MOl and host cell confluence. S. aureus invasion is highly variable between strains and host cell lines. Expression and availability of FnBPs and a5~J are likely to be factors in determining the level of invasion. However, invasion did not correlate completely with levels of Fn-binding and expression of FnBPs, indicating additional factors are involved. The multi functional adhesin Iron-regulated surface determinant protein A (lsdA), was identified as a potential invasin due to its ability to bind Fn and its expression under conditions expected to pertain in vivo. Despite binding Fn, IsdA was not found to facilitate invasion. However, growth of S. aureus in minimal media produced slightly and reproducibly enhanced invasion and may therefore influence expression of other factors involved in facilitating invasion. The level of expression of a5~1 was found to be comparable in an oral, nasal and skin cell line and is therefore unlikely to be responsible for variation in S. aureus invasion. Invasion was enhanced in all cell lines when carried out in suspension; conditions that would allow access to a5~1 molecules even if they had been distributed unevenly on the cell surface when growing on polystyrene. Such polarity, however, did not account for the variation seen in invasion between cell types though further work showed that the nasal epithelial cells exibited significantly reduced binding to Fn. Since Fn acts as a bridging molecule between a5~1 and S. aureus, presence of non-active integrins would explain the difference in invasion seen. In summary, this work confirms Fn-binding capability by host cells is important in bacterial adhesion and invasion of cells but shows that the traditional proteins involved in Fn binding (e.g. FnBP) do not explain fully the different invasive behaviours of S. aureus strains. There is, therefore, redundancy in this functionality. II