, 2006). An elegant pathway has been proposed for heme acquisition by S. aureus involving the sequential, and direct, transfer of heme from IsdA, IsdB, and IsdH to IsdC (Mazmanian et al., 2003). This is supported by substantial amounts of in vitro data demonstrating the potential for heme transfer between the proteins (Grigg et al., 2007; Liu

et al., 2008; Muryoi et al., 2008; Villareal et al., 2011). Thus, despite the in vitro capability of heme transfer, the system does not appear from our studies to be a physiological requirement for heme uptake. Such redundancy may allude to other heme acquisition mechanisms. Staphylococcus aureus likely encounters heme-containing proteins during infection through the secretion of hemolysins lysing erythrocytes (Bernheimer et al., 1968). A range of proteins linked via sortase to the cell wall may be involved in heme acquisition as a srtA mutant is unable to use buy NVP-BKM120 heme as iron source (Mazmanian et al.,

2003). Also S. aureus has an array of alternative iron acquisition systems, including two major siderophores (Hammer & Skaar, 2011). The above data do not support a clear role of IsdA, IsdB, and IsdH in iron acquisition by S. aureus. In order to determine their combined function in pathogenesis, the well-established murine model of sepsis was used. The strain find more Newman background was used as this has been the subject of many studies in this model (Palmqvist et al., 2002; Barbagelata et al., 2011). Figure 4 shows the bacterial load in murine kidneys 7 days postinfection.

There is no statistically significant difference (P = 0.484) between Newman and AFH0013 (∆isdABH) strains. Both sets of animals were infected with the same number of cells (1.5 × 107 CFU) determined by serial dilution in PBS and plating on tryptic soy agar. Figure 4 also shows the percentage weight loss of the two groups of mice over the course of the experiment. Interestingly, there is a significant difference in body weight between animals infected with AFH013 and its isogenic parent. At all time points, AFH013-infected animals demonstrate a decrease in the loss of weight. This is the first time that the triple isd mutant has been used in a pathogenesis study. Our PIK3C2G results are potentially at odds with previous studies using single (isdA, isdB, isdH) and a double isdBH mutant, which suggested a role of the gene products in infection (Torres et al., 2006; Cheng et al., 2009; Kim et al., 2010). The differences may be due to the details of the animal models used. This current study highlights the fact that combined IsdA, IsdB, and IsdH do not have an important role in bacterial burden in our model. Of course, all animal models are imperfect as S. aureus has evolved primarily in the human environment. We have previously found that IsdA is required for survival on human skin (Clarke et al., 2007) and nasal colonization in a cotton rat model (Clarke & Foster, 2006).