NDM-1 Zn1-binding residue His116 plays critical roles in antibiotic hydrolysis

Abstract

Bacteria expressing NDM-1 have been labeled as superbugs because it confers upon them resistance to a broad range of β-lactam antibiotics. The enzyme has a di‑zinc active centre, with the Zn2 site extensively studied. The roles of active-site Zn1 ligand residues are, however, still not fully understood. We carried out structure-function studies using the mutants, H116A, H116N, and H116Q. Zinc content analysis showed that Zn1 binding was weakened by 40 to 60% in the H116 mutants. The enzymatic-activity studies showed that the lower hydrolysis rates were mainly caused by their weaker substrate binding. The catalytic efficiency (kcat/Km) of the mutants followed the order: WT > > H116Q (decreased by 4–20 fold) > H116A (decreased by 20–700 fold) ≥ H116N (decreased by 6–800 fold). The maximum effect was observed on H116N against penicillin G, whereas ampicillin was not hydrolyzed at all. The fold-increase of Km values, which informs the weakening of substrate binding, were: H116A by 5–45 fold; H116N by 6–100 fold; H116Q by 2–10 fold. Molecular dynamics simulations suggested that the Zn1 site mutations affected the positions of Zn2 and the bridging hydroxide, by 0.8 to 1.2 Å, with the largest changes of ~1.5 Å observed on Zn2 ligand C221. A native hydrogen bond between H118 and D236 was disrupted in the H116N and H116Q mutants, which led to increased flexibility of loop 10. Consequently, residue N233 was no longer maintained at an optimal position for substrate binding. H116 connected loop 7 across Zn1 to loop 10, thereby contributed to the overall integrity. This work revealed that the H116-Zn1 interaction plays a critical role in defining the substrate-binding site. From these results, it can be inferred that inhibition strategies targeting the zinc ions may be a new direction for drug development.