Computational assay of H7N9 influenza neuraminidase reveals R292K mutation reduces drug binding affinity

Abstract

The emergence of a novel H7N9 avian influenza that infects humans is a serious cause for concern. Of the genome sequences of H7N9 neuraminidase available, one contains a substitution of arginine to lysine at position 292, suggesting a potential for reduced drug binding efficacy. We have performed molecular dynamics simulations of oseltamivir, zanamivir and peramivir bound to H7N9, H7N9-R292K, and a structurally related H11N9 neuraminidase. They show that H7N9 neuraminidase is structurally homologous to H11N9, binding the drugs in identical modes. The simulations reveal that the R292K mutation disrupts drug binding in H7N9 in a comparable manner to that observed experimentally for H11N9-R292K. Absolute binding free energy calculations with the WaterSwap method confirm a reduction in binding affinity. This indicates that the efficacy of antiviral drugs against H7N9-R292K will be reduced. Simulations can assist in predicting disruption of binding caused by mutations in neuraminidase, thereby providing a computational 'assay.'

Figure 1. Schematic showing the approximate spatial location of the active site residues of neuraminidase in relation to the main functional groups of the antiviral compounds (see figure S1 for the chemical structure of the drugs).

Residues that form direct interactions with the drugs are drawn inside the outer circle. The R292 residue is shown in red. Residues are numbered using N2 numbering.

Figure 2. Oseltamivir bound to H7N9-R292K neuraminidase.

This initial computational model was built by point mutating individual residues of H11N9 neuraminidase bound to oseltamivir (PDB structure 2QWK20) to match the sequence of H7N9-R292K. Residues that differ from the reference H11N9 neuraminidase are shown in dark cyan, with the R292K lysine residue highlighted in red. This structure provides the starting point for the H7N9-R292K oseltamivir-bound simulation (see Methods section for more details).

Figure 3. Structure of oseltamivir bound to (a) H7N9 and (b) H7N9-R292K neuraminidase.

In cyan is shown the experimental X-ray structure of oseltamivir bound to (a) H11N9 (2QWK20) and (b) H11N9-R292K (2QWE20), while in white is shown a representative snapshot taken from simulation. The average location of oseltamivir and arginine/lysine during the simulation is shown using isosurfaces, with the volume of space occupied on average during 90% (solid yellow) and 60% (transparent yellow) of the trajectory displayed. This shows that lysine in the H7N9-R292K simulation is located to disrupt binding of the bulky group of oseltamivir, with the bulky group twisted up, out of the bulky group binding pocket. This is in agreement with the X-ray structure (2QWE) of H11N9-R292K.

Figure 4. Simulation structures of zanamivir bound to (a) H11N9, (b) H7N9 and (c) H7N9-R292K neuraminidase.

All figures are rendered from the same viewpoint, showing that the lysine residue in H7N9-R292K hydrogen bonds with the bulky group of zanamivir, shifting the drug up and to the right compared to H11N9 and H7N9. The figures show representative snapshot structures from the simulations, with the average location of zanamivir and arginine/lysine shown using isosurfaces, with the volume of space occupied during 90% (solid yellow) and 60% (transparent yellow) displayed. The graphs show the distance between the arginine/lysine and the carboxylate group (D1) and bulky group (D2) on zanamivir. Distances are shown on a color scale from 2–8 Å, with blue-green values indicating the presence of a hydrogen bond or salt bridge. These show that while arginine interacts directly with both the carboxylate and bulky groups of zanamivir, lysine hydrogen bonds only sporadically with the bulky group.

Figure 5. Structures of peramivir bound to H7N9-R292K neuraminidase, showing representative snapshots and the average structure between (a) 20–40 ns and (b) 40–70 ns.

The average location of peramivir and lysine is shown using isosurfaces, with the volume of space occupied during 90% (solid yellow) and 60% (transparent yellow) of the two time periods of the simulation displayed. The H11N9-R292K X-ray structure (1L7H21) is shown in cyan. This shows that the simulation closely matched the X-ray structural data in the first part of the simulation. After 40 ns, the 150-loop partially opened, leading to a loss of interaction between peramivir and D151/R152, and a change in binding mode of the drug. The shortest distance between the hydrogen bond donors and acceptors of peramivir (guanidinium and hydroxyl groups) to the D151 and R152 residues on the 150-loop are shown on a color-scale from 2–8 Å, with blue-green values indicating the presence of a salt bridge or hydrogen bond. This shows that the interactions between peramivir and the two 150-loop residues were lost at approximately 45 ns. Equivalent figures for peramivir bound to H11N9 and H7N9 neuraminidase are given in figure S8 in supporting information.