Antiinfluenza Drugs

One of the innovative examples on biostructure-based drug design was the discovery of the antiinfluenza drug zanamivir (Relenza). Together with the subsequent discovery of the orally active anti-influenza agent oseltamivir (Tamiflu) they now constitute a classical example on a successful drug design story.

Two glycoproteins, namely hemagglutinin and neuraminidase, present on the surface of the virus are important for the replication cycle of the virus. The enzyme neuraminidase, or sialidase, is a gly-cohydrolase responsible for cleavage of the terminal sialic acid residues from carbohydrate moieties on the surface of the host cells and the influenza virus envelope. This process facilitates the release of newly formed virus from the surface of an infected cell. Inhibition of neuraminidase will leave uncleaved sialic acid residues that may bind to viral hemagglutinin, causing viral aggregation and thereby a reduction of the amount of virus that may infect other cells.

Thus, selective inhibitors of the enzyme neuraminidase are potential drugs against influenza. The determination of a 3D structure of neuraminidase by x-ray crystallography in 1991 yielded a breakthrough in the discovery of neuraminidase inhibitors being sialic acid analogous by biostructure-based methods.

It was shown that neuraminidase forms a homotetramer and that each monomer contains 6 four-stranded antiparallel beta-sheets arranged as blades on a propeller. The active site was identified from complexes with inhibitors of the enzyme and is located in a deep cavity on the surface of the enzyme (Figure 2.4). The active site is primarily formed by charged and polar residues, reflecting that the substrate is also a polar compound.

The mechanism for hydrolysis of glycoconjugates (Figure 2.5A) by neuraminidase yielding sialic acid (Figure 2.5C) is proposed to proceed via a flat oxonium cation transition state (Figure 2.5B).

(B)

FIGURE 2.4 3D structure of influenza virus neuraminidase (pdb-code 2QWK). (A) Homotetramer shown as ribbon representation. Four inhibitor molecules bound to the enzyme are shown as space-filling models. (B) Ribbon representation of one monomer colored continuously from blue (N-terminus) to red (C-terminus). An inhibitor is shown as a stick model. (C) Surface representation of monomeric neuraminidase with an inhibitor occupying the active site.

RO CO2H

RO CO2H

OH O

OH

OH HN O

HO CO2H

CO2H

CO2H

N NH

OH O

CO2H

OH O

N NH

CO2H

FIGURE 2.5 The biostructure-based design process leading to of zanamivir (Relenza): chemical structure of the glycoconjugates (R = sugar) (A), the proposed transition state (B), sialic acid (C), DANA (D), and zanamivir (E).

HO CO2H

Sialic acid is also a weak inhibitor of neuraminidase (IC50 ~ 1 mM) and binds to the enzyme in a half-boat conformation. The weak, nonselective inhibitor 2-deoxy-2,3-didehydro-D-N-acetyl-neuraminic acid (Neu5Ac2en, DANA, Figure 2.5D) was developed as a sialic acid analog resembling the transition state of the enzymatic process (IC50 = 1-10 |M). Based on the structure of the neuraminidase-sialic acid complex, the complex with DANA and computational studies using the program GRID, which determines favorable binding sites for probes resembling various functional groups, it could be shown that replacement of the hydroxy group at the 4-position on DANA by an amino or a guanidinyl group would enhance binding. It was predicted that a substantial increase in binding could be obtained by introducing an amino group because a salt bridge was formed to a negatively charged side chain in the enzyme. Further replacement of the amino group with the more basic guanidinyl group was as anticipated yielding an even tighter binding inhibitor due to the formation of salt bridges to two negatively charged side chains in the enzyme. Both the predicted increase in affinity and the binding mode were subsequently confirmed experimentally. The 4-guanidinyl analog (Figure 2.5E) was named zanamivir and was the first neuraminidase inhibitor approved for treatment of influenza in humans. It is marketed under the trade name Relenza.

The low oral bioavailability and rapid excretion of zanamivir clearly showed that further improvements were needed in order to obtain a successful anti-influenza drug. Based on the 3D structures of several neuraminidase-inhibitor complexes and computer-based studies, the characteristics of the active site and thereby the properties of the optimal inhibitor were deduced. Replacing the pyra-nose ring with a benzene ring reduced the affinity, showing that the half-boat conformation of the six-membered ring was important for obtaining a proper orientation of the substituents. Inhibitors with the pyranose ring replaced by a carbocyclic ring system showed promising affinities, and by introducing more lipophilic substituents compounds with significantly better oral bioavailability were obtained.

In the final active drug candidate GS4071 (IC50 ~ 1 nM) the carbocyclic ring adopts the half-boat conformation and the polar substituents are all involved in hydrogen bonding to polar residues in the neuraminidase active site (Figure 2.6). The ethyl ester was named oseltamivir and its phosphate salt (a prodrug, see Chapter 9) is marketed under the trade name Tamiflu.

FIGURE 2.6 Left: Chemical structures of oseltamivir (A) (oseltamivir phosphate = Tamiflu) and the active component of Tamiflu (GS4071) (B) formed by enzymatic cleavage. Right: Hydrogen-bonding network between (B) and neuraminidase (pdb-code 2QWK).

FIGURE 2.6 Left: Chemical structures of oseltamivir (A) (oseltamivir phosphate = Tamiflu) and the active component of Tamiflu (GS4071) (B) formed by enzymatic cleavage. Right: Hydrogen-bonding network between (B) and neuraminidase (pdb-code 2QWK).

Was this article helpful?

0 0

Post a comment