Glycosyl-Phosphatidylinositol-Anchored Anti-HIV Env Single-Chain Variable Fragments Interfere with HIV-1 Env Processing and Viral Infectivity

Abstract

In previous studies, we demonstrated that single-chain variable fragments (scFvs) from anti-human immunodeficiency virus (HIV) Env monoclonal antibodies act as entry inhibitors when tethered to the surface of target cells by a glycosyl-phosphatidylinositol (GPI) anchor. Interestingly, even if a virus escapes inhibition at entry, its replication is ultimately controlled. We hypothesized that in addition to functioning as entry inhibitors, anti-HIV GPI-scFvs may also interact with Env in an infected cell, thereby interfering with the infectivity of newly produced virions. Here, we show that expression of the anti-HIV Env GPI-scFvs in virus-producing cells reduced the release of HIV from cells 5- to 22-fold, and infectivity of the virions that were released was inhibited by 74% to 99%. Additionally, anti-HIV Env GPI-scFv X5 inhibited virion production and infectivity after latency reactivation and blocked transmitter/founder virus production and infectivity in primary CD4⁺ T cells. In contrast, simian immunodeficiency virus (SIV) production and infectivity were not affected by the anti-HIV Env GPI-scFvs. Loss of infectivity of HIV was associated with a reduction in the amount of virion-associated Env gp120. Interestingly, an analysis of Env expression in cell lysates demonstrated that the anti-Env GPI-scFvs interfered with processing of Env gp160 precursors in cells. These data indicate that GPI-scFvs can inhibit Env processing and function, thereby restricting production and infectivity of newly synthesized HIV. Anti-Env GPI-scFvs therefore appear to be unique anti-HIV molecules as they derive their potent inhibitory activity by interfering with both early (receptor binding/entry) and late (Env processing and incorporation into virions) stages of the HIV life cycle.IMPORTANCE The restoration of immune function and persistence of CD4⁺ T cells in HIV-1-infected individuals without antiretroviral therapy requires a way to increase resistance of CD4⁺ T cells to infection by both R5- and X4-tropic HIV-1. Previously, we reported that anchoring anti-HIV-1 single-chain variable fragments (scFvs) via glycosyl-phosphatidylinositol (GPI) to the surface of permissive cells conferred a high level of resistance to HIV-1 variants at the level of entry. Here, we report that anti-HIV GPI-scFvs also derive their potent antiviral activity in part by blocking HIV production and Env processing, which consequently inhibits viral infectivity even in primary infection models. Thus, we conclude that GPI-anchored anti-HIV scFvs derive their potent blocking activity of HIV replication by interfering with successive stages of the viral life cycle. They may be effectively used in genetic intervention of HIV-1 infection.

Keywords: antiviral agents; entry inhibitor; envelope protein; human immunodeficiency virus; infectivity; neutralizing antibodies; scFv.

FIG 1

Expression levels of GPI-scFv constructs after cotransfection. 293T cells were transfected with GPI-scFv constructs and harvested, and GPI-positive cells were quantified by staining for the His-tagged hinge region by flow cytometry. SSC, side scatter.

FIG 2

Production of virions is significantly decreased by anti-HIV Env GPI-scFvs. 293T cells were cotransfected with GPI-scFvs and HIV clones NL4-3 (A) and AD8 (B). Supernatants were collected, and titers were determined by p24 ELISA. Histograms show means ± standard deviations (n = 3). Statistical comparisons were done by ANOVA (**, P < 0.01, ***, P < 0.001; ****, P < 0.0001). All results shown are representative of three independent experiments.

FIG 3

Anti-HIV Env GPI-scFvs significantly reduce viral infectivity. Equal amounts of HIV p24 collected from cotransfections of NL4-3 (A) and AD8 (B) were used to infect TZM-bl cells. Histograms show means ± standard deviations (n = 3). Statistical comparisons were done by ANOVA (*, P < 0.05; **, P < 0.01, ***, P < 0.001; ****, P < 0.0001). All results shown are representative of three independent experiments.

FIG 4

Production and infectivity of SIV is unaffected by anti-HIV Env GPI-scFvs. (A) Quantification of SIV produced in the presence or absence of GPI-scFvs using an SIV p27 ELISA. (B) Relative infection levels with SIV derived in the presence or absence of the indicated GPI-scFvs. Histograms show means ± standard errors of the means (n = 3). All results shown are representative of three independent experiments.

FIG 5

Processing of Env gp160 precursors in cells is inhibited by anti-HIV Env GPI-scFvs. (A) Western blot analysis of cell lysates for gp160, gp120, p24, and actin. (B) ImageJ was used to quantify the band intensity ratio of gp120 to gp160 in cell lysates. Data shown are the averages ± standard deviations of three experiments. ***, P < 0.001; NS, not significant.

FIG 6

HIV Env gp120 associated with virions released from cells is reduced by anti-HIV Env GPI-scFvs. (A) Western blot analysis of supernatants for gp160, gp120, and p24. (B) ImageJ was used to quantify the band intensity of gp120 relative to that of p24 in virion lysates. Data shown are averages ± standard deviations of three independent experiments. *, P < 0.05; **, P < 0.01, ***, P < 0.001; ****, P < 0.0001.

FIG 7

Production and infectivity of transmitter/founder virions is significantly decreased by anti-HIV Env GPI-scFvs. (A) Quantification of virus in cell supernatants. (B) Relative infectivity of T/F HIV-1 clones produced in the presence or absence of GPI-scFvs. 293T cells were cotransfected with GPI-scFvs and transmitter/founder HIV-1 clones. The amount of virus produced was determined by HIV p24 ELISA, and equal amounts of p24 from each culture were used to determine viral infectivity on TZM-bl cells. Histograms show means ± standard deviations (n = 3). Statistical comparisons were done by ANOVA (*, P < 0.05; **, P < 0.01, ***, P < 0.001; ****, P < 0.0001; NS, not significant). All results shown are representative of three independent experiments.

FIG 8

GPI-X5 transduction of infected primary CD4⁺ T cells inhibits production of transmitter/founder HIV-1. (A) Expression levels of GPI-scFvs on infected CD4⁺ T cells. (B) HIV-1 p24 production. Primary CD4⁺ T cells isolated from PBMCs were infected with three different HIV-1 transmitter/founder strains, AD17, RHPA, and THRO, at an MOI of 0.1 for 4 h. Infected CD4⁺ T cells were then transduced with anti-HIV Env GPI-X5, the GPI-AB65 negative control, or a mock control. Supernatants were collected, and the amounts of HIV-1 were determined using p24 ELISA at 14 days postinfection. Histograms show means ± standard deviations (n = 3). Statistical comparisons were done by ANOVA (****, P < 0.0001; NS, not significant). All results shown are representative of three independent experiments.

FIG 9

Production and infectivity of HIV-1 induced from latency are inhibited by anti-HIV Env GPI-X5. (A) Expression levels of GPI-scFv constructs after treatment with TNF-α. (B) Quantification of HIV-1 in cell supernatants by HIV p24 ELISA. (C) Relative infectivity of HIV-1 produced in the presence or absence of anti-HIV Env GPI-X5 or the GPI-AB65 negative control. The ACH-2 cell line was transduced with GPI-scFvs and treated with TNF-α to induce production of HIV-1 LAI. Cells were harvested, and GPI-positive cells were quantified by staining for the His-tagged hinge region and analyzed by flow cytometry. Equal HIV p24 was used to infect TZM-bl cells. Histograms show means ± standard deviations (n = 3). Statistical comparisons were done by ANOVA (*, P < 0.05; **, P < 0.01, ****, P < 0.0001; NS, not significant). All results shown are representative of three independent experiments.