The desire to inhibit zinc-dependent matrix metalloproteinases (MMPs) has, over the course of the last 30 years, led to the development of a plethora of MMP inhibitors that bind directly to the active-site metal. With one exception, all of these drugs have failed in clinical trials, due to many factors, including an apparent lack of specificity for MMPs. To address the question of whether these inhibitors are selective for MMPs in a biological setting, a cell-based screening method is presented to compare the relative activities of zinc, heme iron, and non-heme iron enzymes in the presence of these compounds using the RAW264.7 macrophage cell line. We screened nine different zinc-binding groups (ZBGs), four established MMP inhibitors (MMPis), and two novel MMP inhibitors developed in our laboratory to determine their selectivities against five different metalloenzymes. Using this model, we identified two nitrogen donor compounds--2,2'-dipyridylamine (DPA) and triazacyclononane (TACN)--as the most selective ZBGs for zinc metalloenzyme inhibitor development. We also demonstrated that the model could predict known nonspecific interactions of some of the most commonly used MMPis, and could also give cross-reactivity information for newly developed MMPis. This work demonstrates the utility of cell-based assays in both the design and the screening of novel metalloenzyme inhibitors.