Molecular Dynamics Investigation of the Influenza Hemagglutinin Conformational Changes in Acidic pH

Abstract

The surface protein hemagglutinin (HA) of the influenza virus plays a pivotal role in facilitating viral infection by binding to sialic acid receptors on host cells. Its conformational state is pH-sensitive, impacting its receptor-binding ability and evasion of the host immune response. In this study, we conducted extensive equilibrium microsecond-level all-atom molecular dynamics (MD) simulations of the HA protein to explore the influence of low pH on its conformational dynamics. Specifically, we investigated the impact of protonation on conserved histidine residues (H106₂) located in the hinge region of HA2. Our analysis encompassed comparisons between nonprotonated (NP), partially protonated (1P, 2P), and fully protonated (3P) conditions. Our findings reveal substantial pH-dependent conformational alterations in the HA protein, affecting its receptor-binding capability and immune evasion potential. Notably, the nonprotonated form exhibits greater stability compared to protonated states. Conformational shifts in the central helices of HA2 involve outward movement, counterclockwise rotation of protonated helices, and fusion peptide release in protonated systems. Disruption of hydrogen bonds between the fusion peptide and central helices of HA2 drives this release. Moreover, HA1 separation is more likely in the fully protonated system (3P) compared to nonprotonated systems (NP), underscoring the influence of protonation. These insights shed light on influenza virus infection mechanisms and may inform the development of novel antiviral drugs targeting HA protein and pH-responsive drug delivery systems for influenza.

Figure 1

(A) HA structure showing histidine residues (H106₂, H142₂, H159₂ and H184₁) thought to play a key role in pH-induced conformational changes. (B) A representation of influenza HA protein-mediated membrane fusion: At low pH, the HA1 domains dissociate from the HA2 domains, exposing the fusion peptide. Histidine 106₂ (H106₂), shown in yellow, is located in the hinge region of HA2.

Figure 2

Cα RMSD time series of nonprotonated (NP) and fully protonated (3P) systems of HA (A, B), HA1 (C, D), and HA2 (E, F). Each NP and 3P system simulation was run three times for 2.4 μs each.

Figure 3

S4 Cα RMSD distributions. Probability distribution of S4 helices overall Cα RMSD values in nonprotonated (NP) (A) and fully protonated (3P) (B) systems during the final half of simulations (from 1.2 to 2.4 μs). The overall RMSD refers to aligning the entire protein and calculating the RMSD for the S4 domain specifically, providing insight into the flexibility of the S4 helices relative to the whole protein structure. (C, D) Cartoon representations of the NP and 3P system from the first and last frames of the simulation. The protonated histidine residues (H106₂) are shown with the yellow color, while the S4 helices are represented in blue.

Figure 4

Water count probability distribution. (A) Probability distribution of the number of water molecules among three S4 helices during the final half of simulations (from 1.2 to 2.4 μs), comparing nonprotonated (NP) and fully protonated (3P) systems, based on three independent sets of simulations per condition. (B) Cartoon representation of the 3P system, showing the waters within the three S4 helices in the first and last frames of the simulation. Protonated histidines (H106₂) and water molecules are shown in yellow and gray, respectively.

Figure 5

Average tilt and rotation angles. (A) The average tilt angle of the S4 helices for the last 1.2 μs of simulations in both nonprotonated (NP, blue) and fully protonated systems (3P, red), each repeated three times. (B) Definition of S4 tilt angle that is measured with respect to the entire long helices of HA2. (C) The average rotation angle of the long helices, including S3 and S4, for the last 1.2 μs in both nonprotonated (NP, blue) and fully protonated systems (3P, red), each repeated three times. (D) Definition of long helix rotation. The rotation of each helix is calculated with respect to its crystal structure.

Figure 6

(A–C) Minimum distance between H106₂ (HA2) and its neighboring D31₁ (HA1). (D) Graphical representation of H106₂ and D31₁ in different monomer. (E–G) Minimum distance between H106₂ and E103₂, both located on the same monomer as HA2.

Figure 7

Projections of principal components (PCs) 1 and 2 depict the conformational landscape of the fusion peptide across nonprotonated (NP), partially protonated (1P, 2P), and fully protonated (3P) systems, as revealed by PCA. The first row represents nonprotonated segments in each system, while the second row displays protonated segments. Each color corresponds to a specific system. For comparison of the major modes of variation between protonated and nonprotonated conditions, PCA was performed separately for protonated and nonprotonated segments. This approach highlights the impact of histidine protonation on the conformational landscape of the fusion peptide.

Figure 8

Inter- and intrahydrogen bond distances between FP residues (L2₂, F3₂) and S4 residues (S113₂, D109₂, K117₂, and D112₂). (A–C and G–I) Intermonomer hydrogen bonds between FP and S4 residues, indicating they are located in different monomers. (D–F and J–L) Intramonomer hydrogen bonds between FP and S4 residues, signifying they are located within the same monomer.

Figure 9

Average and standard deviation of penetrated water molecules and center of mass distances within each HA1 domain. (A) Number of water molecules between HA1 and HA2 in nonprotonated (NP) and fully protonated (3P) systems for each repeat, depicting the last 1.2 μs of the simulations. (B) Graphic representation of the first and last frames of 3P system, with water molecules surrounding HA1 (red) and HA2 (blue) depicted in gray. (C) Measurement of the distance between the center of mass of the head and tail of HA1 for each repeat, focusing on the last 1.2 μs of the simulations. (D) Molecular image showing the center of mass distance considered between the head and tail of each HA1.

Figure 10

Comparing the number of water molecules surrounding the side chains of H17₁ (A) and E35₁ (B) in the HA1 domain, highlighting increased hydration in the fully protonated systems (3P). Data from all three repeats for each system are included in these plots, focusing on the last 1.2 μs of the simulations.