A V3 loop-dependent gp120 element disrupted by CD4 binding stabilizes the human immunodeficiency virus envelope glycoprotein trimer

Abstract

Human immunodeficiency virus (HIV-1) entry into cells is mediated by a trimeric complex consisting of noncovalently associated gp120 (exterior) and gp41 (transmembrane) envelope glycoproteins. The binding of gp120 to receptors on the target cell alters the gp120-gp41 relationship and activates the membrane-fusing capacity of gp41. Interaction of gp120 with the primary receptor, CD4, results in the exposure of the gp120 third variable (V3) loop, which contributes to binding the CCR5 or CXCR4 chemokine receptors. We show here that insertions in the V3 stem or polar substitutions in a conserved hydrophobic patch near the V3 tip result in decreased gp120-gp41 association (in the unliganded state) and decreased chemokine receptor binding (in the CD4-bound state). Subunit association and syncytium-forming ability of the envelope glycoproteins from primary HIV-1 isolates were disrupted more by V3 changes than those of laboratory-adapted HIV-1 envelope glycoproteins. Changes in the gp120 beta2, beta19, beta20, and beta21 strands, which evidence suggests are proximal to the V3 loop in unliganded gp120, also resulted in decreased gp120-gp41 association. Thus, a gp120 element composed of the V3 loop and adjacent beta strands contributes to quaternary interactions that stabilize the unliganded trimer. CD4 binding dismantles this element, altering the gp120-gp41 relationship and rendering the hydrophobic patch in the V3 tip available for chemokine receptor binding.

FIG. 1.

V3 loop insertion mutants. The V3 loop sequences of the wild-type (wt) and insertion mutant envelope glycoproteins from the indicated HIV-1 strains are aligned. The numbering of the gp120 residues corresponds to that of the HXBc2 prototype, according to current convention (41). The corresponding segments of the V3 loop structure in the CD4-bound state (38) are shown above the alignment. The inserted glycine residues are shown in lower case.

FIG. 2.

Syncytium-forming activity and ability of V3 insertion mutants to support HIV-1 infection. (A) 293T cells expressing the wild-type (wt) or the indicated HIV-1 ADA envelope glycoprotein mutants were cocultivated with Cf2Th-CD4/CCR5 cells and syncytium formation measured as described in Materials and Methods. The negative control cells were transfected with a plasmid expressing the HIV-1 HXBc2 envelope glycoproteins. The syncytium-forming activity of the mutants is reported relative to that seen for the wild-type HIV-1 ADA envelope glycoproteins. Means and standard deviations derived from four replicate assays are shown. (B) The relative syncytium-forming activities of the mutant 1 and 2 envelope glycoproteins are shown for each of the indicated HIV-1 strains. Cf2Th-CD4/CXCR4 target cells were used for the HXBc2, MN27 and 89.6 envelope glycoproteins, and Cf2Th-CD4/CCR5 target cells were used for the YU2 and 89.6 envelope glycoproteins. The experiments were conducted as described in panel A. The means and standard deviations derived from four replicate assays are shown. (C and D) The wild-type (wt) or mutant (m1, m2, etc.) envelope glycoproteins from the indicated HIV-1 strain were assessed for the ability to support the infection of recombinant HIV-1 expressing luciferase, using Cf2Th-CD4/CCR5 target cells (for the ADA, YU2, and 89.6 HIV-1 envelope glycoproteins) or Cf2Th-CD4/CXCR4 target cells (for the HXBc2 and 89.6 HIV-1 envelope glycoproteins). The infectivity of the mutants relative to that seen for the respective wild-type envelope glycoproteins is shown. The experiments were repeated with comparable results.

FIG. 3.

Processing and subunit association of HIV-1 envelope glycoprotein variants. (A to D) 293T cells expressing the wild-type (wt) or mutant envelope glycoproteins from the indicated HIV-1 strains were radiolabeled. The cells were pelleted and lysed. The radiolabeled cell lysates and media were precipitated by a mixture of sera from HIV-1-infected individuals. The envelope glycoproteins precipitated from the cell lysates and media were analyzed by SDS-PAGE. The gp160 and gp120 envelope glycoproteins are indicated.

FIG. 4.

Binding of ligands to HIV-1 envelope glycoprotein variants. (A) 293T cells expressing the wild-type (wt) or mutant (m1 to m9) HIV-1 ADA envelope glycoproteins were radiolabeled. The cell supernatants were precipitated by a mixture of sera from HIV-1-infected individuals (PS), by CD4-Ig, or by the indicated monoclonal antibody. The gp120 glycoproteins precipitated from the media were analyzed by SDS-PAGE and autoradiography. (B and C) Radiolabeled wild-type (wt) or mutant gp120 envelope glycoproteins from the indicated HIV-1 strains were incubated at room temperature for 1 h with Cf2Th-CCR5 cells in the presence of sCD4. The cells were washed and lysed, and the bound gp120 was precipitated by a mixture of sera from HIV-1-infected individuals. The bound gp120 glycoprotein was analyzed by SDS-PAGE and autoradiography.

FIG. 5.

Syncytium-forming ability of V3 insertion mutants with target cells expressing a CCR5 mutant. (A to C) 293T cells expressing the wild-type (wt) or V3 mutant envelope glycoproteins from the indicated HIV-1 strain were cocultivated with Cf2Th cells expressing CD4 and either wild-type (wt) CCR5 (white) or the CCR5-GG mutant (gray). CCR5-GG has two glycines inserted after residue 18, thus extending the N terminus. Syncytium-forming ability was measured as described in Materials and Methods. In panels A and B, the syncytium-forming abilities of the envelope glycoproteins are all normalized to that of the wild-type HIV-1 envelope glycoproteins on target cells expressing CD4 and wild-type CCR5. In panel C, the syncytium-forming abilities of the mutant envelope glycoproteins are normalized to that of the wild-type envelope glycoproteins on target cells expressing CD4 and the corresponding CCR5 variant. The means and standard deviations derived from three experiments are shown. The delta V3 envelope glycoprotein is missing the gp120 V3 loop and is a fusion-defective negative control (81). 293T cells transfected with the empty pcDNA vector serve as an additional negative control.

FIG. 6.

Changes in HIV-1 gp120 resulting in decreased association with gp41. A ribbon diagram of the unliganded simian immunodeficiency virus (SIV) gp120 core structure (5) is shown in the left column. In the three columns on the right, the HIV-1 gp120 glycoprotein is shown in the different conformations observed in available crystal structures, complexed with the Fab fragments of the b12 or F105 neutralizing antibodies (6, 87), or with two-domain CD4 (38). The gp120 structures are viewed from the perspective seen by the Fab or CD4 proteins; the outer domains of the gp120 cores are aligned. The trimeric axis of the envelope glycoprotein complex is located on the left side of each structure, in approximately the vertical orientation (44, 48). The gp120 core in the b12-bound structure was modified to predispose the protein to assume the CD4-bound conformation (87). In the top row, the HIV-1 domains are colored as follows: outer domain (yellow), inner domain (red) and bridging sheet components (blue for the β20-β21 loop and green for the β2-β3 V1/V2 stem). In the cases where the V3 loop structure was not determined, the position of the V3 base is indicated. In the CD4-bound structure, the elements of the V3 loop are labeled. The gp120 beta strands (defined in the CD4-bound structure [43]) relevant to the present study are also labeled. In the bottom row, the gp120 residues are colored according to the gp120-gp41 association index (red, association index < 0.5; green, association index ≥ 0.5) observed upon mutagenesis of the HIV-1YU2 and HXBc2 gp120 glycoproteins (, , , ; the present study; Finzi et al., unpublished). In the CD4-bound gp120 structure, the three hydrophobic residues in the tip of the V3 loop that were implicated in gp120-gp41 association are labeled.

FIG. 7.

The hydrophobic V3 patch. (A) The V3 variable loops of the primate immunodeficiency viruses are aligned. The degree of variation in residues 307, 309, and 317 in the HIV-1 gp120 V3 tip is shown (42). The percentage of sequences in each virus group that contain the indicated residue is shown as a superscript to the right of the single-letter amino acid designation. Amino acid variants that are found in less than 1% of the sequences surveyed are not listed. (B to D) 293T cells expressing the wild-type (wt) or mutant HIV-1 YU2 envelope glycoproteins were radiolabeled. The cells were pelleted and lysed. The radiolabeled cell lysates and media were precipitated by a mixture of sera from HIV-1-infected individuals. The envelope glycoproteins precipitated from the cell lysates and medium were analyzed by SDS-PAGE. The gp160 and gp120 envelope glycoproteins are indicated.

FIG. 8.

HIV-1 gp120 residues contributing to recognition by the G3-299 antibody. The ribbon diagrams of the F105-bound and CD4-bound gp120 core plus V3 loop (6, 38) are shown from the same perspective as in Fig. 6. The trimeric axis of the HIV-1 envelope glycoproteins is located on the left side of each structure, in approximately the vertical orientation (44, 48). In the F105-bound gp120 structure, the V3 loop is disordered (6). The gp120 residues in which alterations resulted in decreased recognition by the G3-299 antibody (less than 50% of the amount of the recognition observed for the wild-type HIV-1 YU2 gp120) are colored magenta (; the present study). In the F105-bound gp120, residues that contact the F105 antibody (<4 Å) are colored blue (6). Serine 375 also contacts the F105 antibody (6).

FIG. 9.

The gp120 elements implicated in subunit association in the unliganded HIV-1 envelope glycoprotein trimer. The CD4-bound structure of the HIV-1 gp120 core (43) was fitted into the density derived from cryo-electron tomograms of the unliganded HIV-1 envelope glycoprotein trimer (48). The V3 base is colored magenta, and the other gp120 residues are colored according to the effects of changes on gp120-gp41 association, as in Fig. 6 (bottom row). In the upper part of the figure, the viral membrane is at the top and the target cell at the bottom. The lower part of the figure shows the HIV-1 envelope glycoprotein trimers from the perspective of the target cell. The positions of the V3 loops (magenta) were approximated by extending the two beta strands, β12 and β13, that lead into the V3 loop strands (38, 43, 44). The positions of the V1/V2 stems are indicated.

FIG. 10.

Model of the conformational changes in the HIV-1 envelope glycoprotein induced by CD4 binding. One of the three subunits of the HIV-1 envelope glycoprotein trimer is depicted, oriented so that the viral membrane is at the top of the picture and the trimer axis is vertical. The gp120 outer domain (OD) is yellow. The HIV-1 gp120 inner domain consists of a β-sandwich (red) and loops that form three topological layers (layer 1 [magenta], layer 2 [green], and layer 3 [yellow]). The β-sandwich and the gp120 N and C termini (cyan) are the major gp120 elements that mediate interaction with the gp41 ectodomain (32, 55, 84). Layers 1 and 2, as well as the β20-β21 loop (blue) and V3 loop (purple), all contribute to stabilizing the interaction of gp120 with gp41 in the unliganded state (left figure). CD4 binding results in the apposition of layer 1 and layer 2; the formation of the bridging sheet from the β2, β3, β20, and β21 strands; and the projection of the V3 loop away from the gp120 core (, , , ; Finzi et al., unpublished). This rearrangement of gp120 slows the off-rate of CD4 (Finzi et al., unpublished), promotes chemokine receptor binding (78), and allows the gp41 ectodomain to undergo additional conformational changes necessary for HIV-1 entry.