Staphylococcus aureus is an opportunistic pathogen causing high morbidity and mortality. Since multi-drug resistant S. aureus lineages are nowadays omnipresent, alternative tools for preventive or therapeutic interventions, like immunotherapy, are urgently needed. However, there are currently no vaccines against S. aureus. Surface-exposed and secreted proteins are regarded as potential targets for immunization against S. aureus infections. Yet, many potential staphylococcal antigens of this category do not elicit protective immune responses. To obtain a better understanding of this problem, we compared the binding of serum IgGs from healthy human volunteers, highly S. aureus-colonized patients with the genetic blistering disease epidermolysis bullosa (EB), or immunized mice to the purified S. aureus peptidoglycan hydrolases Sle1, Aly and LytM and their different domains. The results show that the most abundant serum IgGs from humans and immunized mice target the cell wall-binding domain of Sle1, and the catalytic domains of Aly and LytM. Interestingly, in a murine infection model, these particular IgGs were not protective against S. aureus bacteremia. In contrast, relatively less abundant IgGs against the catalytic domain of Sle1 and the N-terminal domains of Aly and LytM were almost exclusively detected in sera from EB patients and healthy volunteers. These latter IgGs may contribute to the protection against staphylococcal infections, as previous studies suggest that serum IgGs protect EB patients against severe S. aureus infection. Together, these observations focus attention on the use of particular protein domains for vaccination to direct potentially protective immune responses towards the most promising epitopes within staphylococcal antigens.