Abstract The use of animal models to study the virulence of mutant bacteria is a staple of bacterial pathogenesis research. However, in most instances we are studying the ability of a mutant to establish an infection. If the mutant is defective, it is cleared, thereby preventing us from assessing the role of that virulence factor later on in the infection. This is important, since patients present with an ongoing infection, not at the time of first contact with the pathogen. We have developed a recombinase based system to remove virulence factors after infection. The system is stably off for days in vitro, but can be induced to efficiently remove the virulence gene of choice. Here we propose to test this system in a mouse model of Pseudomonas aeruginosa corneal infection. The candidate virulence factor we will examine is the type III secretion system (T3SS), a molecular syringe that is absolutely critical for the ability of P. aeruginosa to establish an infection in the first place. In Aim 1, we will examine the stability of the recombinase “off” state in vivo, as well as the effect of removing the T3SS at later stages on the ability of P. aeruginosa to maintain the infection. In Aim 2, we will examine the effect of removing the T3SS on immune cell recruitment and cytokine production. We will also develop a recombinase-based fusion system, that is designed to generate b-lactamase fusions to effector proteins. This system will allow us to examine which cells are targeted by the T3SS in an established infection. Taken together, we are proposing to develop a set of tools to explore, for the first time, the contribution of P. aeruginosa virulence genes after the infection has taken root. Beyond the novel insights into P. aeruginosa pathogenesis, the tools we will develop will can be applied to any number of systems, such as biofilm formation or establishment of mixed microbial communities, that proceed through an initial establishment phase to the mature structure.