The HIV protease (HIVP) is a prominent example for successful structure-based drug design. Besides its pharmaceutical impact, it is a well-studied system for which, as experimentally evidenced, protonation changes in the active site occur upon ligand binding. Therefore, it serves as an ideal candidate for a case study using our newly developed partial charge model, which was optimized toward the application of Poisson-Boltzmann based pK(a) calculations. The charge model suggests reliably experimentally determined protonation states in the active site of HIVP. Furthermore, we perform pKa calculations for two HIVP complexes with novel types of inhibitors developed and synthesized in our group. For these complexes, no experimental knowledge about the protonation states is given. For one of the compounds, containing a central pyrrolidine ring, the calculations predict that both catalytic aspartates should be deprotonated upon ligand binding.