The identification of metal‐binding ligand residues in metalloproteins using nuclear magnetic resonance spectroscopy

Abstract

Abstract:The identification of metal-binding ligands in metalloproteins is an important step in gaining detailed information regarding the environment of the active site. Traditionally, techniques such as '"Cd-substitution for the active metal followed by isotopefiltered NMR techniques have been used to this end. However, for medium to high molecular weight proteins (>20 kDa), these experiments may not be beneficial due to extensive 'H spectral overlap. Here, we describe an alternative approach, where metalbinding ligands such as histidine and cysteine are specifically I5N backbone labeled, excess EDTA i s added and changes to {'H-''N} HSQC spectra are followed. Under these conditions, the amide groups of all I5N labeled histidine and cysteine residues, which were either ligands or residues close to the active site, were identified unambiguously for metallo-P-lactamase from Bacteroides frugilis.Keywords: metal binding ligands; metallo-P-lactamase; metalloprotein; NMR NMR spectroscopy has been used successfully to characterize both the environment of the metal and its ligands in metalloenzymes and metalloproteins. For example, the substitution of the NMR active metal 'I3Cd, for biologically significant metals such as zinc, iron, and calcium, has allowed both the coordination geometry and ligand types to be determined with great accuracy (Coleman, 1993). The Il3Cd-based NMR experiments have also been used to identify side-chain proton resonances of the common zinc-binding ligands, cysteine and histidine in small protein, and model peptide systems (Frey et al., 1985;South et al., 1989;Blake et al., 1994;Gardner & Coleman, 1994 cation of specific ligands using these methods becomes increasingly difficult at higher molecular weights (>20 kDa), since severe overlap in both the 'H aliphatic and aromatic regions can obscure individual resonances. Here we describe a method for identifying metal-binding ligands in large metalloproteins involving the specific I5N backbone labeling of metal-ligand types (His and Cys), and the addition of EDTA to sequester the metal from the protein. Specific assignment of ligand-binding residues is accomplished by first identifying resonances in I5N HSQC spectra, which are affected by the addition of EDTA and then correlating these to I3C

NMR Characterization of the Metallo-β-lactamase from Bacteroides fragilis and Its Interaction with a Tight-Binding Inhibitor: Role of an Active-Site Loop

“…Since phosphate has a high affinity for zinc ions, it is likely that zinc is sequestered from the catalytic active sites to form insoluble Zn 3 (PO 4 ) 2 , suggesting that the removal of zinc ions from the active site precedes or accompanies aggregation. This is consistent with the observation that when the metal chelator EDTA is added to highly stable NMR samples prepared using the ion exchange protocol, aggregation occurs within hours (27). The stability and tendency of the protein to exist as a monomer are greatly improved in the presence of the inhibitor, presumably due to the lowered solvent accessibility of the zinc site.…”

Mutational Analysis of Metallo-β-lactamase CcrA from Bacteroides fragilis

“…Previous biochemical, kinetic, and structural studies have indicated significant differences among the metallo-β-lactamases (1,11,21,33,34). These differences suggest that one inhibitor may not inhibit all metallo-β-lactamases; in fact, previous studies using nonclinical inhibitors have shown widely different effacacies and interactions of the inhibitors with the different metallo-β-lactamases (12,13,16,17,(35)(36)(37)(38). The published crystal structures of CcrA, L1, and β-lactamase II demonstrate different active site amino acids, and computer modeling studies using these crystal structures as a basis have been used to predict distinct substrate binding models and reaction mechanisms (4,5,9).…”

“…Concha et al reported the crystal structure of CcrA in 1996 (), and since then, several other structures have been reported of the enzyme with competitive inhibitors bound within the active site ( , ). Solution NMR studies have also been reported on CcrA ( , ). Since it has been impossible so far to determine the crystal structure of CcrA with substrate bound in the active site, modeling studies have been used to propose how substrate binds to the enzyme.…”

BindWeb: A web server for ligand binding residue and pocket prediction from protein structures

“…Protein–ligand interactions are indispensable for many biological processes, such as gene expression, signal transduction, and antigen–antibody interaction 1–7 . To explore the interaction mechanisms, experimental methods are applied to resolve complex structures, such as X‐ray, nuclear magnetic resonance spectroscopy, and laser Raman spectroscopy.…”

“…Protein-ligand interactions are indispensable for many biological processes, such as gene expression, signal transduction, and antigen-antibody interaction. [1][2][3][4][5][6][7] To explore the interaction mechanisms, experimental methods are applied to resolve complex structures, such as X-ray, nuclear magnetic resonance spectroscopy, and laser Raman spectroscopy. However, considering the time-consuming and high cost of experimental methods, developing efficient computational methods for binding pocket prediction has become an essential topic in structural bioinformatics.…”