Insights into the molecular mechanism underlying CD4-dependency and neutralization sensitivity of HIV-1: a comparative molecular dynamics study on gp120s from isolates with different phenotypes

Abstract

The envelope (Env) of HIV-1 plays critical roles in viral infection and immune evasion. Although structures of prefusion Env have been determined and phenotypes relevant to the CD4 dependency and the neutralization sensitivity for various HIV-1 isolates have been identified, the detailed structural dynamics and energetics underlying these two phenotypes have remained elusive. In this study, two unliganded structural models of gp120, one from the CD4-dependent, neutralization-resistant isolate H061.14 and the other from the CD4-independent, neutralization-sensitive R2 strain, were constructed, and subsequently were subjected to multiple-replica molecular dynamics (MD) simulations followed by free energy landscape (FEL) construction. Comparative analyses of MD trajectories reveal that during simulations R2-gp120 demonstrated larger structural fluctuations/deviations and higher global conformational flexibility than H061.14-gp120. Close comparison of local conformational flexibility shows that some of the structural regions involving direct interactions with gp41 and adjacent gp120 subunits in the context of the closed trimeric Env exhibit significantly higher flexibility in R2-gp120 than in H061.14-gp120, thus likely increasing the probability for R2-Env to open the trimer crown and prime gp41 fusogenic properties without induction by CD4. Collective motions derived from principal component analysis (PCA) reveal that R2-gp120 is prone to spontaneous transition to the neutralization-sensitive CD4-bound state while H061.14-gp120 tends to maintain the neutralization-resistant unliganded state. Finally, comparison between FELs reveals that R2-gp120 has larger conformational entropy, richer conformational diversity, and lower thermostability than H061.14-gp120, thus explaining why R2-gp120 is more structurally unstable and conformationally flexible, and has a higher propensity to transition to the CD4-bound state than H061.14-gp120. The present results reveal that the differences in dynamics and energetics between R2-gp120 and H061.14-gp120 impart Env trimers with distinct capacities to sample different states (i.e., R2-Env samples more readily the open state while H061.14-Env is more inclined to maintain the closed state), thus shedding light on the molecular mechanism underlying the HIV-1 phenotype associated with CD4 dependency/neutralization sensitivity.

Fig. 1. Sequence alignments between the target and the template. (A) and (B) Are sequence alignments used for building structural models of H061.14-gp120 and R2-gp120, respectively. 5FYJ_G represents the template sequence from the crystal structure with PDB ID 5FYJ (chain G). Residues are numbered according to the template sequence. Conserved residues are shaded in light blue. Regular secondary structural elements of the template are numbered according to HXBc2 crystal structures (PDB IDs: 3JWD and 1G9M), with orange arrows and red spirals representing β-strands and α-helices (or 3/10 helices), respectively. The variable regions, i.e., V1/V2, V3, V4, and V5 are indicated above the alignment by blue line segments. The four β-strands in the V1/V2 region, designated A to D, are labelled βA to βD, respectively. V1 and V2 loops are located between βA and βB and between βC and βD, respectively, and the loop connecting βB and βC is labelled L1.

Fig. 2. Structural models of the two near full-length gp120s in the unliganded state. (A) Ribbon representation of the model of H061.14-gp120. (B) Ribbon representation of the model of R2-gp120. (C) Backbone superposition of the models of H061.14-gp120 (orange) and R2-gp120 (green). In (A) and (B), the inner domain, outer domain, and small domain (composed of V1/V2, V3, and bridging-sheet elements β2-β3 and β20-β21) are coloured blue, green, and red, respectively; secondary structural elements in the inner and outer domains are numbered in reference to crystal structures of HXBc2 gp120 core (PDB IDs: 3JWD and 1G9M); the “Greek key”-like V1/V2 in the small domain is composed of four antiparallel β-strands (labelled βA to βD) and three connecting loops: V1 (between βA and βB), L1 (between βB and βC), and V2 (between βC and βD); the two bridging-sheet elements, β2-β3 and β20-β21, arrange in the order β2-β3-β21-β20 where β3 and β21 form a parallel β-sheet, in contrast to the order β3-β2-β21-β20 where β2 and β21 form an anti-parallel β-sheet observed in the CD4-bound state of gp120. The seven-stranded β-sandwich is composed of β3̄, β0, β1, β5, β6, β7, and β25; the layers 1 to 3, which emanate from the β-sandwich, are defined as the excursions between β3̄ and β0, between β1 and β5, and between β7 and β25, respectively.

Fig. 3. Time evolution of the backbone root-mean-square deviation (RMSD) values of the two gp120 structural models with respect to their respective starting structures during multiple-replica MD simulations. (A) RMSD curves of H061.14-gp120. (B) RMSD curves of R2-gp120. RMSD curves calculated from MD replicas 1 to 8 are shown in different colours.

Fig. 4. C_α atom root-mean-square-fluctuation (RMSF) profiles of H061.14-gp120 and R2-gp120 and their RMSF-difference as a function of residue number. C_α RMSF values of H061.14-gp120 (orange line) and R2-gp120 (green line) were calculated from respective joined equilibrium MD trajectories. Residue numbering is according to the template sequence (PDB ID: 5FYJ, chain G). Residues belonging to the inner domain (residues 31–119, 204–255, and 474–510), outer domain (residues 256–298, 330–421,443–473), and small domain (residues 120–203, 299–329, 422–442) are indicated above the horizontal axis by line segments coloured in blue, green, and red, respectively. The average RMSF values are shown as dotted lines. RMSF-difference was obtained by subtracting H061.14-gp120's RMSF values from R2-gp120's values at structurally equivalent residue positions, which were determined from the structure-based multiple sequence alignment shown in ESI Fig. S2. The regions with RMSF-difference greater and less than 0 are shaded in light red and light blue, respectively, and the regions with RMSF-difference greater than 0.06 nm are highlighted in red.

Fig. 5. Eigenvalues of the first 30 eigenvectors (main plot) and cumulative contribution of all eigenvectors to the total mean square fluctuations (inset plot) for the H061.14-gp120 (orange line) and R2-gp120 (green line).

Fig. 6. Porcupine plots showing the largest-amplitude collective motions of the two gp120 structural models. (A)–(C) Modes of collective motions of H061.14-gp120 along the eigenvectors 1–3, respectively. (D)–(F) Modes of collective motions of R2-gp120 along the eigenvectors 1–3, respectively. In these plots, the direction and length of the cone drawn on a C_α atom represent the fluctuating direction and amplitude of this atom, respectively, along the corresponding eigenvector; the inner domain, outer domain, and small domain are coloured blue, green, and red, respectively.

Fig. 7. Free energy landscapes (FELs) of the two gp120 structural models as a function of the projection of the joined equilibrium trajectory onto the first (PC1) and second (PC2) principal components. (A) Two-dimensional FEL of H061.14-gp120. (B) Two-dimensional FEL of R2-gp120. The colour bar represents the relative free-energy level in unit of kJ mol⁻¹.