Exploring the chemical space of influenza neuraminidase inhibitors

Abstract

The fight against the emergence of mutant influenza strains has led to the screening of an increasing number of compounds for inhibitory activity against influenza neuraminidase. This study explores the chemical space of neuraminidase inhibitors (NAIs), which provides an opportunity to obtain further molecular insights regarding the underlying basis of their bioactivity. In particular, a large set of 347 and 175 NAIs against influenza A and B, respectively, was compiled from the literature. Molecular and quantum chemical descriptors were obtained from low-energy conformational structures geometrically optimized at the PM6 level. The bioactivities of NAIs were classified as active or inactive according to their half maximum inhibitory concentration (IC50) value in which IC50 < 1µM and ≥ 10µM were defined as active and inactive compounds, respectively. Interpretable decision rules were derived from a quantitative structure-activity relationship (QSAR) model established using a set of substructure descriptors via decision tree analysis. Univariate analysis, feature importance analysis from decision tree modeling and molecular scaffold analysis were performed on both data sets for discriminating important structural features amongst active and inactive NAIs. Good predictive performance was achieved as deduced from accuracy and Matthews correlation coefficient values in excess of 81% and 0.58, respectively, for both influenza A and B NAIs. Furthermore, molecular docking was employed to investigate the binding modes and their moiety preferences of active NAIs against both influenza A and B neuraminidases. Moreover, novel NAIs with robust binding fitness towards influenza A and B neuraminidase were generated via combinatorial library enumeration and their binding fitness was on par or better than FDA-approved drugs. The results from this study are anticipated to be beneficial for guiding the rational drug design of novel NAIs for treating influenza infections.

Keywords: Chemical space; Combinatorial library enumeration; Data mining; Fragment analysis; Influenza; Molecular docking; Neuraminidase; Neuraminidase inhibitor; QSAR; Scaffold analysis.

Figure 1. Schematic workflow of this study.

Figure 2. Histogram representing the molecular descriptors for NAIs against influenza A.

Note: Active and inactive NAIs are represented with red and blue bars, respectively, whereas their overlapping region are shown in purple.

Figure 3. Histogram representing the molecular descriptors for NAIs against influenza B.

Note: Active and inactive NAIs are represented with red and blue bars, respectively, whereas the purple represents their overlap region.

Figure 4. PCA scores and loadings plots of NAIs against influenza A (A and B, respectively) and B (C and D, respectively).

Active and inactive compounds are represented by green and red circles, respectively, in the scores plots. Important features for rationalizing the active and inactive compounds are highlighted by green and red clusters, respectively. An interactive version is available at https://dx.doi.org/10.6084/m9.figshare.3123136.v1.

Figure 5. Illustration of decision tree model for classifying the activity of NAIs against Influenza types A and B as a function of their substructure fingerprint.

The full descriptive name of the substructure fingerprints are shown for the purpose of clarity while their corresponding acronyms are provided in the text as well as the supplementary data available on figshare at http://dx.doi.org/10.6084/m9.figshare.1612484. It should be noted that “1,3-tautomerizable” and “chiral center specified” correspond to idiosyncratic PaDEL definitions rather than “standard definitions”.

Figure 6. Plots of the descriptor usage derived from the decision tree model.

Descriptors with the largest percentage of descriptor usage is deemed the most important.

Figure 7. Summary of common substructure in active and inactive sets of NAIs against influenza A (A and B, respectively) and B (C and D, respectively).

Number of substructure occurrences are indicated in bracket below the substructure’s rank.

Figure 8. Binding modes of NAIs in active site of influenza A and B neuraminidase are shown in (A) and (B), respectively.

Electrostatic (Elec), hydrogen-bond (Hbond) and van der Waals’ (vdW) interaction sites are indicated by red, blue and orange sphere, respectively. Interacting residues (N2 numbering) of Elec, Hbond and vdW are highlighted in white, cyan and yellow, respectively.

Figure 9. Molecular structures of enumerated ligands against neuraminidase of influenza A (A1–A10) and B (B1–B10) are categorized according to their scaffold types and compared to FDA-approved drugs (e.g., zanamivir, oseltamivir and peramivir) as well as the long-acting laninamivir.

It should be noted that these enumerated ligands passed the decision tree-based post-filter.

Figure 10. Binding pose of enumerated ligands A1 (A) and B1 (B) providing the lowest binding energy against influenza A and B neuraminidase.

The electrostatic potential on the surface of neuraminidase is calculated via APBS and is shown by red, blue and white colors that represents negative, positive and neutral charge, respectively.