Asp120 is not a proton donor in di-zinc Bacillus cereus metallo-b-lactamase

LLARRULL, L.I.; VILA, A.J.

Ann Arbor, Michigan

Conferencia; 12th International Conference on Biological Inorganic Chemistry (ICBIC-12); 2005

Metallo-Beta-lactamases (MBL?s) are enzymes with a zinc binding motif in their active site. The absence of zinc in the active site renders these enzymes inactive towards different antibiotics. At present, there are no clinically useful inhibitors. A rational design of new inhibitors should necessarily rely on a thorough knowledge of their catalytic mechanism, still unveiled. Asp90 is conserved in all known metallo-beta-lactamases. This residue is a zinc ligand in enzymes containing two metal ions, whereas in mono zinc enzymes it is involved in a strong hydrogen bond interaction with the nucleophilic OH-. It has been predicted that Asp90 is essential in the mechanism of mono zinc enzymes, as is the case of the MBL from B. cereus (BcII), either: (1) by defining the appropriate orientation of the attacking nucleophile, or (2) through acid/base catalysis. To evaluate the role of Asp90 in the hydrolysis of beta-lactam antibiotics, we engineered four mutants of the enzyme BcII in this position. All of the mutants showed decreased activity against different beta-lactam antibiotics; with the following trend: D90E > D90N, D90Q > D90S. None of the mutants was totally inactive. Kinetic Isotope Effect experiments with different beta-lactam antibiotics indicate that the rate limiting step is a proton transfer, as in the wild type enzyme, even in the absence of a residue with appropriate acid/base properties. Replacement of Asp90, a ligand to one of the zinc ions in the active site, does not abolish the capacity of the enzyme to bind two equivalents of metal ion. Characterization of the structure of the active sites by means of UV-VIS spectroscopy on the cobalt substituted derivatives indicated that a tetrahedral metal site similar to the one in wild type BcII is conserved in these mutants and indicated the presence of cobalt coordinated to a Cys residue in a slightly distorted second site. These results suggest that proton transfer takes place when Asp90 is replaced by a residue incapable of acid/base catalysis, but less efficiently. Alternatively, a water molecule might be the proton donor, and Asp90 might participate in a hydrogen bond network, similar to the one proposed for the mono-Zn enzyme mechanism, which enhances the rate of proton transfer from the water molecule.