However, grouping patients into clinically severe infections (bacteremia/sepsis, endocarditis, osteomyelitis or severe deep tissue infections) and mild infections (superficial infections and/or deeper wounds lacking clinical signs of severe infection such

as elevated leukocyte count, fever and hyperemia of the affected tissue) revealed significantly higher titers for patients with severe infections (P=0.045). Although Eap appears to play a role in biofilm formation under in vitro conditions (Thompson find more et al., 2010), patients with foreign body-associated infections did not present with higher anti-Eap titers than patients with other types of infections. Comorbidities, age of the patients, onset of disease, the type of acquisition (nosocomial vs. community acquired) and strain susceptibility (methicillin Akt inhibitor resistant vs. methicillin sensitive) were found not to be statistically different. IgM titers were significantly higher in patients compared with healthy controls (Fig. 3a). Additionally, antibody titers were higher for sera sampled within the first 4 weeks after the onset of infections (P=0.045, Table 2) in line with IgM antibodies being

the first immunoglobulins produced upon antigen contact. The group of patients with deep infections revealed higher IgM titers compared with patients with superficial infections, although this did not reach statistical significance (P=0.085). However, in contrast to the results obtained for IgG antibody determination, no significant differences in IgM titers could be detected for the different types

of infections. Selected sera were tested for the presence of rheuma factors, and found to be negative, making cross-reactivity with rheuma factors unlikely. Previous studies indicated a correlation between S. aureus antibodies in serum and the extent of in vitro opsonization and phagocytosis (Dryla et al., 2005b; DNA ligase Verkaik et al., 2009). In our study, the functionality of anti-Eap antibodies was determined using an opsonophagocytosis assay with inert fluorescent beads to rule out the unwanted influence of other S. aureus surface components such as protein A. Incubation of granulocytes and PBMCs with EB, but not with NB, resulted in an increase in granularity and fluorescence, indicating an Eap-stimulated phagocytosis (Fig. 3a). Coupling of the beads with human albumin as an unrelated control protein, on the other hand, had no stimulatory effect on phagocytosis (data not shown). Within the group of PBMCs, only the CD14-positive population of macrophages/monocytes emitted fluorescence. Lymphocytes neither changed significantly in quantity nor emitted any fluorescence. Quantitative analyses revealed that EB, in contrast to NB, were phagocytosed efficiently even without the addition of serum (Fig. 3b).