A computational perspective of influenza a virus targets : neuraminidase and endonuclease.

Abstract

Through the ages the viruses have plagued mankind claiming the lives of millions, pre-dating any advancements in the medicinal sciences. One such pathogenic virus is influenza A, which has been implicated in the 1918-Spanish flu, the 2006-avian flu outbreak and the 2009-swine flu pandemic. It is a highly sophisticated species, alluding efforts to thwart the spread of disease and infection. One of the main reasons influenza has survived this long is simple evolution. Natural mutation within the genome of virions expressed in proteins, enzymes or molecular structure render us unable to predict or take preventative measures against possible infection. Thus, research efforts toward the competitive inhibition of biological pathways that lead to the spread of disease, have become attractive targets. The influenza A virus has a number of chemotherapeutic targets, such as: 1) The surface antigens, hemagglutinin and neuraminidase, 2) RNA-dependent RNA polymerase, and 3) The M2 proton channel. Influenza RNA polymerase is composed of three large segments encoding polymerase acidic protein (PA), polymerase basic protein 1 (PB1) and polymerase basic protein 2 (PB2). The PA protein is an N-terminal domain subunit which contains the endonuclease activity. The influenza virus is incapable of synthesizing a 5’-mRNA cap, so it has adapted a cap-snatching mechanism whereby the PB2 subunit binds to the 5’-end of host mRNA, after which 10-14 nucleotides downstream the PA-subunit (aka PAN) cleaves the strand forming a primer for viral mRNA synthesis which is catalysed by the PB1 subunit. Influenza target identification is based primarily on evidence suggesting sequence conservation of each entity and its selective expression in the virus and not the host. In this thesis two enzymatic targets were investigated, the PA protein of RNA polymerase and neuraminidase. The studies focussed on using computational tools to: 1) provide insight into the mechanism of drug-resistance, 2) describe the conformational structure of the protein in the presence of point mutations and in complex with an inhibitor, 3) determine the essential binding pharmacophoric features to aid the design of new drug therapies.

An array of computational techniques were employed in the studies, such as: molecular dynamics (MD) simulation, structure-based and ligand-based in silico screening, principal component analysis, radius of gyration analysis, binding free energy calculations and solventaccessible surface area analysis. The first study (Chapter 5) determined the mechanism of drug-resistance in influenza A neuraminidase as a consequence of antigenic variations. Two distinct mutations in the enzyme sequence that were investigated are H274Y and I222K. The active site residues of neuraminidase are conserved among the subtypes of influenza A. However, it was discovered that the occurrence of resistance to the drug oseltamivir, in the H1N1 species was different to the H5N1 virus. Although both systems shared a loss in hydrophobicity of the active site, the conformational distortion of the active site pocket distinguished the enzyme of the two viral entities, from one another. The discoveries made in the first study laid the foundation for the second study (Chapter 6), which was based on the in silico design and screen of potential neuraminidase inhibitors. As a result 10 characteristic molecular scaffolds were suggested as potential inhibitors. The pharmacophore design was constructed with consideration to the new conformational structure of the active site pocket. Chapter 7 is the third study of this thesis. The active site pocket enclosing the endonuclease activity of the PA subunit was investigated. Using molecular dynamics simulations and postdynamic analyses, a description of the protein conformation was offered. Subsequently, a pharmacophore was proposed as a potential scaffold to which endonuclease inhibitors may be modelled upon. It is my belief that the impact of the results derived from the above mentioned studies would greatly contribute to the development of new and effective anti-influenza drugs.