Staphylococcus aureus is an opportunistic pathogen colonizing roughly 30% of the human population, causing a wide range of diseases. Due to the emergence of resistant strains and the lack of new antibiotics, novel antimicrobials are of high interest. Phage-derived peptidoglycan hydrolases (PGH) could be used as a treatment against drug-resistant bacterial strains. Fast lysis, high specificity and activity against drug-resistant bacteria are just a few advantages for the use of PGHs as protein therapeutics. A major drawback of protein drugs is the immunogenicity of foreign proteins. T cells play a key upstream role in the activation of the adaptive immune system and therefore, the immune reaction against protein therapeutics. Antigen presenting cells sample the environment and proteolytically process proteins, which are then presented as T cell epitopes on MHCII on the cell surface. This leads to the activation of CD4 T cells inducing the activation and differentiation of other T and B cells and with this the production of highly specific and long-lived anti-drug antibodies. Therefore, deimmunization approaches usually focus on T cell epitope prediction and deletion. This project applies computational tools to predict T cell epitopes and design variants with epitope deleting mutations with low impact on protein activity and structure. Deimmunized variants undergo in vitro activity testing, ex vivo immunogenicity assays and in vivo immunogenicity and efficacy studies in humanized mice.