The Staphylococcus aureus pathogenicity islands (SaPIs) are the prototypical members of an extremely common and recently identified family of mobile genetic elements, the phage-inducible chromosomal islands (PICIs) (Penades & Christie, 2015). SaPIs are clinically relevant because they carry and disseminate superantigen genes, for example those for toxic shock toxin and enterotoxin B. In absence of a helper phage they reside passively in the host chromosome, under the control of SaPI-coded master repressor, Stl, a DNA-binding protein. Our previous work indicated that SaPI de-repression is effected by specific, non-essential phage proteins that bind to Stl, disrupting the Stl-DNA complex and thereby initiating the excision-replication-packaging (ERP) cycle of the islands (Tormo-Más et al., 2010). Different SaPIs encode different Stl repressors, so each SaPI requires a different non-essential phage protein for its de-repression. Recently, we published data showing how SaPIbov1 is de-repressed by the trimeric dUTPases (Tormo-Más et al., 2010). Here we demonstrate that SaPIbov1 is also de-repressed by the structurally unrelated dimeric dUTPases. We hypothesise that the SaPIs have evolved from targeting specific proteins to targeting phage mechanisms through divergence of the SaPI repressor. The fact that the Stl repressors interact with structurally unrelated proteins performing the same function make this strategy unique in nature and extremely effective. The processes targeted by the SaPIs are extremely well conserved in the staphylococcal phages, so the fact that SaPIs target different versions of the proteins involved in the same biological processes ensures the transferability of these elements. This highlights SaPIs as one of the most fascinating and effective subcellular parasites.