Influenza A viruses are highly contagious RNA viruses that cause respiratory tract infections in humans and animals. Their non-structural protein NS1, a homodimer of two 230-residue chains, is the main viral factor in counteracting the antiviral defenses of the host cell. Its RNA-binding domain is an obligate dimer that is connected to each of the two effector domains by a highly flexible unstructured linker region of ten amino acids. The flexibility of NS1 is a key property that allows its effector domains and its RNA binding domain to interact with several protein partners or RNAs. The three-dimensional structures of full-length NS1 dimers revealed that the effector domains could adopt three distinct conformations as regards their mutual interactions and their orientation relative to the RNA binding domain (closed, semi-open and open). The origin of this structural polymorphism is currently being investigated and several hypotheses are proposed, among which one posits that it is a strain-specific property. In the present study, we explored through computational molecular modeling the dynamic and flexibility properties of NS1 from three important influenza virus A strains belonging to three distinct subtypes (H1N1, H6N6, H5N1), for which at least one conformation is available in the Protein Data Bank. In order to verify whether NS1 is stable in three forms for the three strains, we constructed homology models if the corresponding forms were not available in the Protein Data Bank. Molecular dynamics simulations were performed in order to predict the stability over time of the three distinct sequence variants of NS1, in each of their three distinct conformations. Our results favor the co-existence of three stable structural forms, regardless of the strain, but also suggest that the length of the linker, along with the presence of specific amino acids, modulate the dynamic properties and the flexibility of NS1.

Influenza A viruses still represent a major health issue, for both humans and animals. One of the main viral proteins of interest to target is the NS1 protein, which counters the host immune response and promotes viral replication. NS1 is a homodimer composed of a dimeric RNA-binding domain (RBD), which is structurally stable and conserved in sequence, and two effector domains that are tethered to the RBD by linker regions. This linker flexibility leads to NS1 polymorphism and can therefore exhibit different forms. Previously, we identified a putative drug-binding site, located in the RBD interface in a crystal structure of NS1. This pocket could be targeted to block RNA binding and inhibit NS1 activities. The objective of the present study is to confirm the presence of this druggable site, whatever the sequence variants, in order to develop a universal therapeutic compound that is insensitive to sequence variations and structural flexibility. Using a set of four NS1 full-length structures, we combined different bioinformatics approaches such as pocket tracking along molecular dynamics simulations, druggability prediction and classification. This protocol successfully confirmed a frequent large binding-site that is highly druggable and shared by different NS1 forms, which is promising for developing a robust NS1-targeted therapy.

PockFlex is a web server designed to analyse pockets across protein structural ensembles and support the reconstruction, characterisation, and prioritisation of recurrent binding site organisations. Applicable to ensembles derived from molecular dynamics simulations, multiple experimental structures, or protein structure predictions, PockFlex detects pockets independently in each conformation, retains those overlapping a user-defined region of interest, and groups them across the ensemble by residue-level similarity. This residue-centred clustering framework identifies recurrent binding site clusters, quantifies residue recurrence and variability, and distinguishes persistent from transient binding site regions across the ensemble. Pocket-level druggability, predicted using the PockDrug workflow, is summarised at the cluster level to support binding site prioritisation under conformational variability while preserving access to individual pocket scores. The web application provides interactive, residue-level insights into pocket organisation, variability, and druggability in structural ensembles. The web server is free and open to all users, without login requirement, at https://pockflex.rpbs.univ-paris-diderot.fr/.