Metallo-beta-lactamases are enzymes capable of hydrolyzing all known classes of beta-lactam antibiotics, rendering them ineffective. The design of inhibitors active against all classes of metallo-beta-lactamases has been hampered by the heterogeneity in metal content in the active site and the existence of two different mononuclear forms. BcII is a B1 metallo-beta-lactamase which is found in both mononuclear and dinuclear forms. Despite very elegant studies, there is still controversy on the nature of the active Bcll species. We carried out a non-steady-state study of the hydrolysis of penicillin G catalyzed by Co(II)-substituted BcII, and we followed the modifications occurring at the active site of the enzyme. Working at different metal/enzyme ratios we demonstrate that both mono-Co(II) and di-Co(II) BcII are active metallo-beta-lactamases. Besides, we here present evidence that during penicillin G hydrolysis catalyzed by mono-Co(II) Bcll the metal is localized in the DCH site (the Zn2 site in B1 enzymes). These conclusions allow us to propose that both in mono-Co(II) and di-Co(II) BcII the substrate is bound to the enzyme through interactions with the Co(II) ion localized in the DCH site. The finding that the DCH site is able to give rise to an active lactamase suggests that the Zn2 site is a common feature to all subclasses of metallo-beta-lactamases and would play a similar role. This proposal provides a starting point for the design of inhibitors based on transition-state analogs, which might be effective against all M beta Ls.