Role of complex carbohydrates in human immunodeficiency virus type 1 infection and resistance to antibody neutralization

Abstract

Complex N-glycans flank the receptor binding sites of the outer domain of HIV-1 gp120, ostensibly forming a protective "fence" against antibodies. Here, we investigated the effects of rebuilding this fence with smaller glycoforms by expressing HIV-1 pseudovirions from a primary isolate in a human cell line lacking N-acetylglucosamine transferase I (GnTI), the enzyme that initiates the conversion of oligomannose N-glycans into complex N-glycans. Thus, complex glycans, including those that surround the receptor binding sites, are replaced by fully trimmed oligomannose stumps. Conversely, the untrimmed oligomannoses of the silent domain of gp120 are likely to remain unchanged. For comparison, we produced a mutant virus lacking a complex N-glycan of the V3 loop (N301Q). Both variants exhibited increased sensitivities to V3 loop-specific monoclonal antibodies (MAbs) and soluble CD4. The N301Q virus was also sensitive to "nonneutralizing" MAbs targeting the primary and secondary receptor binding sites. Endoglycosidase H treatment resulted in the removal of outer domain glycans from the GnTI- but not the parent Env trimers, and this was associated with a rapid and complete loss in infectivity. Nevertheless, the glycan-depleted trimers could still bind to soluble receptor and coreceptor analogs, suggesting a block in post-receptor binding conformational changes necessary for fusion. Collectively, our data show that the antennae of complex N-glycans serve to protect the V3 loop and CD4 binding site, while N-glycan stems regulate native trimer conformation, such that their removal can lead to global changes in neutralization sensitivity and, in extreme cases, an inability to complete the conformational rearrangements necessary for infection.

FIG. 1.

Glycan biosynthesis and distribution on gp120 and gp41. (A) Putative carbohydrate modifications are shown on gp120 and gp41 secondary structures, based on various published works (26, 42, 63, 74, 119, 128). The gp120 outer domain is indicated, as are residues that form the SOS gp120-gp41 disulfide bridge. The outer domain is divided into neutralizing and silent faces. Symbols distinguish complex, oligomannose, and unknown glycans. Generally, the complex glycans of the outer domain line the receptor binding sites of the neutralizing face, while the oligomannose glycans of the outer domain protect the silent domain (105). Asterisks denote sequons that are unlikely to be utilized, including position 139 (42), position 189 (26, 42), position 406 (42, 74), and position 637 (42). Glycans shown in gray indicate when sequon clustering may lead to some remaining unused, e.g., positions 156 and 160 (42, 119), positions 386, 392, and 397 (42), and positions 611 and 616 (42). There is also uncertainty regarding some glycan identities: glycans at positions 188, 355, 397, and 448 are not classified as predominantly complex or oligomannose (26, 42, 63, 128). The number of mannose moieties on oligomannose glycans can vary, as can the number of antennae and sialic acids on complex glycans (77). The glycan at position 301 appears to be predominantly a tetra-antennary complex glycan, as is the glycan at position 88, while most other complex glycans are biantennary (26, 128). (B) Schematic of essential steps of glycan biosynthesis from the Man₉GlcNAc₂ precursor to a mature multiantennary complex glycan. Mannosidase I progressively removes mannose moieties from the precursor, in a process that can be inhibited by the drug kifunensine. GnTI then transfers a GlcNAc moiety to the D1 arm of the resulting Man₅GlcNAc₂ intermediate, creating a hybrid glycan. Mannose trimming of the D2 and D3 arms then allows additional GlcNAc moieties to be added by a series of GnT family enzymes to form multiantennary complexes. This process can be inhibited by swainsonine. The antennae are ultimately capped and decorated by galactose and sialic acid. Hybrid and complex glycans are usually fucosylated at the basal GlcNAc, rendering them resistant to endo H digestion. However, NgF is able to remove all types of glycan.

FIG. 2.

Comparison of parent and GnTI- virus infectivity and expression. (A) The infectivities of parent and GnTI- viruses were measured using CF2.CD4.CCR5 target cells. (B) The relative expression levels of parent and GnTI- VLPs were measured by titrating concentrated preparations of each by BN-PAGE and then detecting Env by Western blotting.

FIG. 3.

Neutralization sensitivities of parent, GnTI-, and N301Q viruses. The neutralization activities of a panel of MAbs, 4D-sCD4, and HIV-1-infected donor plasmas were assayed against parent, GnTI-, and N301Q viruses all bearing the SOS mutation. Each virus is color coded and indicated by the symbol shown. Representative data from duplicate titrations are shown.

FIG. 4.

Relative affinity of soluble CD4 for parent and GnTI- trimers. Concentrated viruses were mixed with graded concentrations of sCD4 for 1 h. Samples were then resolved by BN-PAGE.

FIG. 5.

Effect of neuraminidase treatment on neutralization sensitivity. The JR-FL SOS parent virus transfection supernatant was concentrated by centrifugation and split into two batches, of which one was treated with neuraminidase and the other mock treated for 1 h at 37°C. Viruses were then diluted in culture medium and tested for their infectivity (A) and for their sensitivity to various MAbs and sCD4 (B), as indicated.

FIG. 6.

Effect of endo H on Env trimers. VLPs were incubated with or without endo H for the times indicated and then pelleted, washed, and resolved by BN-PAGE and Western blotting.

FIG. 7.

Effects of endo H on viral infectivity. Viruses were incubated for various times at 37°C in the presence or absence of 200 U of endo H, as indicated. Particles were then pelleted in microcentrifuge tubes and incubated with CF2.CD4.CCR5 cells. Infection was then later determined by a luciferase readout and was plotted as a percentage of the infectivity exhibited by the untreated parent virus at time zero.

FIG. 8.

Effect of endo H on Env trimer-ligand binding. We evaluated the binding of various ligands to parent and GnTI- trimers with or without endo H treatment, as indicated. Ligands (each at 30 μg/ml) included MAbs b12, 15e, 4E10, 2F5, E51, and 39F, and 2D-sCD4, scFv X5, and Fab 58.2 were also evaluated. After incubation with ligand, VLPs were then pelleted, washed, and resolved by BN-PAGE and Western blot assays.

FIG. 9.

Models of Env trimers carrying different glycan cargoes. (A) Parent trimer decorated with a full complement of complex and oligomannose glycans. All complex glycans are shown in light orange, while oligomannose glycans are in slate blue. (B) GnTI- trimer in which complex glycans have been replaced by Man₅GlcNac₂. (C) GnTI- trimer in which oligomannose glycans of the neutralizing face have been replaced by a single GlcNac stump to simulate the effect of endo H in removing the susceptible Man₅GlcNac₂/oligomannose mannose glycans. In all three panels, gp120 (protein) protomers (based on the HXBc2 isolate) are shown as molecular surface representations and are primarily colored red. Colors for gp120 features are as follows: the D368 residue that is essential for binding both sCD4 and b12 is shown in dark blue, the b12 binding site is yellow, V1-V2 stems are pink, V3 stems are white, and V4 stems are green. The electron density map from reference is shown as a mesh at 2 σ (2 standard deviations). The putative complex and oligomannose glycans are based on the findings reported in reference . The particular glycans were chosen probabilistically at each position, based on references and . The complex carbohydrates include tetraantennary, triantennary, and biantennary structures and largely correspond to the structures in reference . We considered the possibility of constructing V3-containing trimer models to illustrate the large complex glycan at residue N301 and the possible effects of its replacement or removal. A major caveat, however, is that the only V3 loop-containing gp120 structure presently available is in a complex with sCD4 and X5 (47). Docking this structure into the unliganded density of the trimer produces a trimer model with a highly exposed V3 loop (see Fig. 2F in reference 69), but that cannot be the case on unliganded primary isolates, because V3 antibodies do not tend to neutralize those isolates (11).