Interaction with CD4 and antibodies to CD4-induced epitopes of the envelope gp120 from a microglial cell-adapted human immunodeficiency virus type 1 isolate

Abstract

We previously showed that the envelope glycoprotein from an in vitro microglia-adapted human immunodeficiency virus type 1 isolate (HIV-1(Bori-15)) is able to use lower levels of CD4 for infection and demonstrates greater exposure of the CD4-induced epitope recognized by the 17b monoclonal antibody than the envelope of its parental, peripheral isolate (HIV-1(Bori)). We investigated whether these phenotypic changes were related to a different interaction of their soluble monomeric gp120 proteins with CD4 or 17b. Equilibrium binding analyses showed no difference between Bori and Bori-15 gp120s. However, kinetic analysis of surface plasmon resonance-based, real-time binding experiments showed that while both proteins have similar association rates, Bori-15 gp120 has a statistically significant, 3-fold-lower dissociation rate from immobilized CD4 than Bori and a statistically significant, 14-fold-lower dissociation rate from 17b than Bori in the absence of soluble CD4. In addition, using the sensitivity to inhibition by anti-CD4 antibodies as a surrogate for CD4:trimeric envelope interaction, we found that Bori-15 envelope-pseudotyped viruses were significantly less sensitive than Bori pseudotypes, with four- to sixfold-higher 50% inhibitory concentration values for the three anti-CD4 antibodies tested. These differences, though small, suggest that adaptation to microglia correlates with the generation of a gp120 that forms a more stable interaction with CD4. Nonetheless, the observation of limited binding changes leaves open the possibility that HIV-1 adaptation to microglia and HIV-associated dementia may be related not only to diminished CD4 dependence but also to changes in other molecular factors involved in the infection process.

FIG. 1.

Binding of gp120s to CD4-pseudotyped MLV particles. 293T cells were cotransfected with an MLV gag/pol construct (pcGP) and CD4, and culture supernatants containing MLV-CD4 particles were concentrated by ultracentrifugation through 20% sucrose. MLV-CD4 particles were immobilized, and binding of increasing concentrations of Bori (○) and Bori-15 (•) gp120 proteins was detected with an anti-gp120 D7324 sheep polyclonal Ab. Both Bori and Bori-15 gp120s showed a similar equilibrium binding to CD4 incorporated into MLV particles. The average from duplicate wells of a representative experiment is shown.

FIG. 2.

Kinetic analysis of gp120 binding to CD4 by SPR. Representative binding curves of Bori (A) and Bori-15 (B) gp120s to immobilized CD4 at 37°C are shown (blue, 500 nM; green, 250 nM; cyan, 125 nM; red, 62.5 nM; and magenta, 31.25 nM) with the response measured in arbitrary resonance units (RU). The symbols are the experimental data, while the lines are the local fit to a 1:1 binding model, with the residual plot shown below. The association phase of the sensorgrams is also shown as derivative (dR/dt) versus response for Bori (C) and Bori-15 (D). The slopes of the linear regression analysis for each concentration are used in the −k_obs versus concentration plot (E), where the slopes of the linear regression analysis for Bori (blue diamonds) and Bori-15 (pink squares) correspond to the apparent on rates of the interaction. (F) Plot of ln(R₀/R) versus time of the 500-nM concentration of Bori (blue diamonds) and Bori-15 (pink squares), where R₀ is the starting point of dissociation and the slope of the line gives k_off.

FIG. 3.

Kinetic analysis of gp120 binding to 17b MAb by SPR. Representative binding curves of Bori (A) and Bori-15 (B) gp120s to immobilized 17b MAb at 37°C are shown (blue, 400 nM; green, 200 nM; cyan, 100 nM; red, 50 nM; and magenta, 25 nM) with the response measured in arbitrary resonance units (RU). The symbols are the experimental data, while the lines are the local fit to a 1:1 binding model, with the residual plot shown below. The association phase of the sensorgrams is also shown as derivative (dR/dt) versus response for Bori (C) and Bori-15 (D). The slopes of the linear regression analysis for each concentration are used in the −k_obs versus concentration plot (E), where the slopes of the linear regression analysis for Bori (blue diamonds) and Bori-15 (pink squares) correspond to the apparent on-rates of the interaction. (F) ln(R₀/R) versus time plot of the 400 nM concentration of Bori (blue diamonds) and Bori-15 (pink squares), where R₀ is the starting point of dissociation and the slope of the line gives k_off.

FIG. 4.

Inhibition of infection of pseudotypes by TAK-779. Bori and Bori-15 envelope-pseudotyped luciferase viruses were used in single-round infection assays in the presence of increasing concentrations of CCR5 inhibitor TAK-779. U87 astroglioma cells transiently transfected for the expression of high (top) or low (bottom) levels of CCR5 expression were used as target cells. Bori (○) and Bori-15 (•) pseudotypes were more sensitive to TAK-779 inhibition when target cells expressed low levels of CCR5, but we did not observe any difference in sensitivity between them regardless of CCR5 levels on target cells. Results are expressed as percentages of infection with respect to no treatment (100%), and the average of 8 to 12 independent experiments performed in triplicate is shown.

FIG. 5.

Inhibition of pseudotype infection by anti-CD4 MAbs. Bori and Bori-15 envelope-pseudotyped luciferase viruses were used in single-round infection assays in the presence of increasing concentrations of anti-CD4 MAbs RPA-T4, SK3 and #21, or mouse IgG as a negative control. HOS cells stably expressing CD4 and CCR5 were used as target cells. Results are expressed as percentages of luciferase activity with respect to that observed in the absence of antibody (considered to be 100%), and shown as means ± standard errors from six to eight independent experiments performed in quadruplicate. Infection by Bori-15 (•) pseudotypes was less sensitive to inhibition by anti-CD4 MAbs than infection by Bori (○) pseudotypes. Asterisks denote those Ab concentrations in which the difference between the percentages of infection of Bori and Bori-15 pseudotypes were statistically significant by Student's t test (*, P < 0.05; **, P < 0.01).