Characterization of the steric defense of the HIV-1 gp41 N-trimer region

Abstract

During viral entry, HIV gp41 adopts a transient conformation called the "prehairpin intermediate" in which a highly conserved therapeutic target, the N-trimer, is exposed. Despite extensive discovery efforts, potent and broadly neutralizing antibodies that target the N-trimer are elusive. We previously demonstrated the N-trimer is protected by a steric block that prevents large proteins, such as antibodies, from accessing it. Here we further characterize the steric block and identify its source. To study the N-trimer steric accessibility, we produced two sets of C-peptide inhibitors (a potent inhibitor targeting the N-trimer) fused to cargo proteins of increasing size facing either the virus or cell side of the prehairpin intermediate. Both bulky inhibitor sets show a steric block, but the effect is more pronounced with virus-side cargo. Additionally, both sets maintain their potencies in a modified entry assay that removes possible sources of target cell steric hindrance. These results implicate a viral source, likely gp120, as the primary component of the steric block. In addition, we studied the steric accessibility of the "pocket" region of the N-trimer, a highly attractive drug and vaccine target. We demonstrated a pocket-specific antibody, D5, is more potent as an scFv than as a full-length IgG, suggesting the N-trimer steric restriction extends to the pocket. This characterization will facilitate the design of sterically restricted antigens that mimic the steric environment of the N-trimer in the prehairpin intermediate and are capable of inducing potent and broadly neutralizing antibodies that circumvent the N-trimer steric block.

Figure 1.

Exploring the geometry of the steric block of an HIV membrane fusion intermediate. (A) Depicted is the model of Env-mediated HIV membrane fusion in which conformational changes in Env are induced by CD4 and coreceptor binding leading to fusion between the viral and host cell membranes. In the diagram are gp120 (yellow), gp41 (light blue), and important regions of gp41: fusion peptide (red), N-peptide region (gray), disulphide-loop (orange), C-peptide region (blue), and transmembrane domain (purple). Also shown is an exogenous C-peptide binding to the prehairpin intermediate and inhibiting formation of the trimer of hairpins. gp120 is shown bound to the disulphide loop and C-peptide regions of the gp41 prehairpin intermediate. (B) The prehairpin intermediate is shown with an exposed N-trimer (gray), including the conserved pocket region. Also shown are the two sets of C-peptide–cargo fusion inhibitors analyzed in this study. On the top are the “cell-side” cargo inhibitors, and on the bottom are the “virus-side” cargo inhibitors. The N and C termini of the C-peptide inhibitors are indicated. The ubiquitin (Ub) cargo is depicted in light green, while maltose-binding protein (MBP) is shown in darker green. In this study we compare the steric block (indicated by the curved lines) faced by the cell-side inhibitors to that faced by the virus-side inhibitors as well as the steric restriction of the pocket region.

Figure 2.

Inhibitory activity of cell-side and virus-side cargo inhibitors in standard infectivity assay. Representative inhibition curves for cell-side (A) and virus-side (B) cargo inhibitors using HXB2 pseudovirions. (C) Calculated IC₅₀ ratios for cell-side and virus-side cargo inhibitors relative to the control, C37. Data represent averages of quadruplicate measurements from at least two independent assays. Error bars represent the standard error of the mean (SEM).

Figure 3.

Inhibitory activity of the pocket-specific neutralizing antibody, D5. Representative inhibition curves for D5 scFv and IgG using HXB2 pseudovirions in the standard viral infectivity assay. Error bars represent the standard error of the mean (SEM).

Figure 4.

Inhibitory activity of cell-side and virus-side cargo inhibitors in sCD4-assisted infectivity assay. Representative inhibition curves for cell-side (A) and virus-side (B) cargo inhibitors against JRFL. (C) Calculated IC₅₀ ratios for cell-side and virus-side cargo inhibitors relative to the control, C37. Data represent averages of quadruplicate measurements from at least two independent assays. Error bars represent the standard error of the mean (SEM).

Figure 5.

Binding of cell-side cargo inhibitors to IZN36 measured by SPR. Binding responses (dots) were normalized to maximal response to facilitate comparison. Fits to the interaction model are shown as solid lines. Proteins were injected at four concentrations in a threefold dilution series starting at 11.1 nM (highest concentration data shown).