Second site escape of a T20-dependent HIV-1 variant by a single amino acid change in the CD4 binding region of the envelope glycoprotein

Abstract

Background: We previously described the selection of a T20-dependent human immunodeficiency virus type-1 (HIV-1) variant in a patient on T20 therapy. The fusion inhibitor T20 targets the viral envelope (Env) protein by blocking a conformational switch that is critical for viral entry into the host cell. T20-dependent viral entry is the result of 2 mutations in Env (GIA-SKY), creating a protein that undergoes a premature conformational switch, and the presence of T20 prevents this premature switch and rescues viral entry. In the present study, we performed 6 independent evolution experiments with the T20-dependent HIV-1 variant in the absence of T20, with the aim to identify second site compensatory changes, which may provide new mechanistic insights into Env function and the T20-dependence mechanism.

Results: Escape variants with improved replication capacity appeared within 42 days in 5 evolution cultures. Strikingly, 3 cultures revealed the same single amino acid change in the CD4 binding region of Env (glycine at position 431 substituted for arginine: G431R). This mutation was sufficient to abolish the T20-dependence phenotype and restore viral replication in the absence of T20. The GIA-SKY-G431R escape variant produces an Env protein that exhibits reduced syncytia formation and reduced cell-cell fusion activity. The escape variant was more sensitive to an antibody acting on an early gp41 intermediate, suggesting that the G431R mutation helps preserve a pre-fusion Env conformation, similar to T20 action. The escape variant was also less sensitive to soluble CD4, suggesting a reduced CD4 receptor affinity.

Conclusion: The forced evolution experiments indicate that the premature conformational switch of the T20-dependent HIV-1 Env variant (GIA-SKY) can be corrected by a second site mutation in Env (GIA-SKY-G431R) that affects the interaction with the CD4 receptor.

Figure 1

Evolution of the T20-dependent HIV-1 variant. (A) Replication of wild-type (GIV-SNY), T20-dependent (GIA-SKY) and HIV-1 variants obtained in evolution cultures in the absence of T20. Bars represent CA-p24 values at day 4 post-infection, which represent the relative differences in replication capacity. Infection of fresh SupT1 cells was performed with an equal amount of the viral stocks obtained at day 42 of evolution. (B) Summary of observed mutations, replication capacity and syncytia formation of the six evolution cultures after 42 days of evolution. Replication without T20 and syncytia formation represent relative differences observed in infection experiment displayed in 1A (-, no replication or syncytia; ++++, high replication or all cells involved in syncytia). (C) Schematic of mutations in the Env protein. The complete Env gene is shown (not to scale) with the location of the original GIA-SKY mutations and the second site changes selected in the evolution cultures. The G431R change that was selected in multiple cultures is marked in bold. Both gp160 and gp41 HxB2 numbering references are included.

Figure 2

Replication of the T20-dependent GIA-SKY mutant and the GIA-SKY-G431R revertant virus. Molecular clones were transfected in SupT1 cells that were cultured with (100 ng/ml) and without T20. Virus replication curves were made over a 7-day period. Closed circles represent the original GIA-SKY mutant virus and open circles the GIA-SKY-G431R revertant. This is a representative experiment, similar results were observed in 3 repeated experiments including results in figure 4, top.

Figure 3

Cell-cell fusion assay with the GIA-SKY mutant and the GIA-SKY-G431R revertant. SupT1 cells were transfected with the HIV-1 pLAI constructs indicated below the X-axis. One day later, transfected cells were mixed with SupT1 cells containing a Tat-responsive LTR-luciferase reporter gene construct with (20 ng/ml) or without T20. After 24 hours, formation of syncytia was analysed by light microscopy (-, no syncytia; ++++, all cells involved in syncytia) and quantitated by measurement of luciferase activity in cell extracts.

Figure 4

Replication of G431-mutated viruses. SupT1 cells were transfected with the T20-dependent (GIA-SKY, top panel) and the wild-type (GIV-SNY, lower panel) with variation at position 431 as indicated. Replication was measured both in the presence (100 ng/ml) and absence of T20. Virus replication curves were made over a 12-day period. Syncytia formation at day 12 was recorded (-, no syncytia; ++++, all cells involved in syncytia). Position 431 variation in the GIA-SKY T20-dependent virus (top) and the wild-type GIV-SNY virus (bottom): wild-type 431G (closed circles), G431R (open circles), G431E (dashes) and G431A (open triangles, in wild-type only).

Figure 5

Cell-cell fusion assay with the G431-mutated Env variants. Assay of wild-type (GIV-SNY) and T20-dependent (GIA-SKY) viruses with position 431 variation as indicated, in the presence (20 ng/ml) and absence of T20. SupT1 cells were transfected with the mutants indicated below the X-axis. One day later, transfected cells were mixed with SupT1 cells containing a Tat-responsive LTR-luciferase reporter gene construct with (20 ng/ml) or without T20. After 24 hours, formation of syncytia was analysed by light microscopy (-, no syncytia; ++++, all cells involved in syncytia) and quantitated by measurement of luciferase activity in cell extracts.

Figure 6

Sensitivity of wild-type GIV-SNY and GIA-SKY-G431R revertant viruses to sCD4 and D5-IgG1 gp41 antibody. Bars represent CA-p24 values at day 6 post-transfection, which represent the relative differences in replication capacity in the presence of the sCD4 inhibitor and the D5-IgG1 antibody for the wild-type virus compared to the revertant virus. This is a representative experiment, similar results were observed in 2 independent experiments.