Small-molecule inhibition of human immunodeficiency virus type 1 infection by virus capsid destabilization

Abstract

Human immunodeficiency virus type 1 (HIV-1) infection is dependent on the proper disassembly of the viral capsid, or "uncoating," in target cells. The HIV-1 capsid consists of a conical multimeric complex of the viral capsid protein (CA) arranged in a hexagonal lattice. Mutations in CA that destabilize the viral capsid result in impaired infection owing to defects in reverse transcription in target cells. We describe here the mechanism of action of a small molecule HIV-1 inhibitor, PF-3450074 (PF74), which targets CA. PF74 acts at an early stage of HIV-1 infection and inhibits reverse transcription in target cells. We show that PF74 binds specifically to HIV-1 particles, and substitutions in CA that confer resistance to the compound prevent binding. A single point mutation in CA that stabilizes the HIV-1 core also conferred strong resistance to the virus without inhibiting compound binding. Treatment of HIV-1 particles or purified cores with PF74 destabilized the viral capsid in vitro. Furthermore, the compound induced the rapid dissolution of the HIV-1 capsid in target cells. PF74 antiviral activity was promoted by binding of the host protein cyclophilin A to the HIV-1 capsid, and PF74 and cyclosporine exhibited mutual antagonism. Our data suggest that PF74 triggers premature HIV-1 uncoating in target cells, thereby mimicking the activity of the retrovirus restriction factor TRIM5α. This study highlights uncoating as a step in the HIV-1 life cycle that is susceptible to small molecule intervention.

FIG. 1.

PF74 selectively inhibits infection by HIV-1. (A) Chemical structure of PF74. (B) Selectivity of PF74 antiviral activity. VSV-G-pseudotyped HIV-1, SIV, and lacZ-transducing MLV vector particles were inoculated on HeLa-P4 indicator cells in the presence of the indicated concentrations of PF74. In the case of HIV-1 and SIV, expression of viral Tat protein in the target cells transactivates an LTR-lacZ reporter, resulting in the expression of β-galactosidase. For the MLV vector particles, transduction of the lacZ-containing vector results in expression of β-galactosidase. Two days after inoculation, the cultures were fixed and stained with X-Gal, and blue cells were enumerated by digital image processing. Shown is the percentage of infection at the corresponding inhibitor concentrations, with error bars representing the standard deviations of the means of the triplicate determinations. The results are representative of four independent experiments.

FIG. 2.

PF74 binds HIV-1 particles specifically and inhibits an early postentry stage of infection. Concentrated wild-type (A) or PF74-resistant mutant (B) HIV-1 particles were incubated with ³H-PF74 for 3 h at room temperature and then subjected to equilibrium sucrose gradient ultracentrifugation. The gradients were fractionated, and the particles in each fraction were pelleted by ultracentrifugation. The radioactivity and CA protein in the pellets were quantified by scintillation counting and antigen-capture ELISA, respectively. (C) PF74 treatment inhibits HIV-1 reverse transcription in target cells. Cells were inoculated with HIV-1 particles in the presence of PF74 (10 μM), and DNA was harvested and assayed by qPCR for late reverse transcription products. EFV: cultures containing HIV-1 with the RT inhibitor Efavirenz (EFV) to serve as a control for contaminating plasmid DNA. Shown are the mean values of triplicate determinations, with error bars depicting one standard deviation. The results in this figure are representative of at least three independent experiments.

FIG. 3.

PF74 destabilizes the HIV-1 capsid. (A) Treatment of HIV-1 with PF74 reduces the level of CA recovered upon purification of HIV-1 cores. Concentrated HIV-1 particles were incubated with or without PF74 then subjected to ultracentrifugation through a layer containing Triton X-100. Fractions were collected from the gradient and analyzed for CA concentrations by ELISA. Shown is the quantity of CA in the fractions of each gradient corresponding to the density of viral cores, expressed as a percentage of the total CA protein in the gradient. (B) Analysis of the effects of 5 and 10 μM PF74 on CA recovery from wild-type and resistant (5Mut) HIV-1 particles. (C) Effects of PF74 on uncoating of HIV-1 cores. Purified wild-type and 5Mut cores were incubated at 37°C in the presence or absence of PF74 (0.25 μM). Wild-type cores were incubated for 30 min, and the 5Mut cores were incubated at 70 min due to their slower uncoating kinetics. After incubation, the extent of uncoating was determined by quantifying the percentage of CA released into soluble form. Shown are the mean values of triplicate parallel determinations from a single experiment. (D) Effect of PF74 on pelletable CA following virus entry into cells. Cultures of CrFK cells were inoculated with VSV-G-pseudotyped HIV-1 in the presence or absence of PF74 (10 μM). Four hours after inoculation, the cells were harvested and lysed, the lysates subjected to ultracentrifugation, and CA in the supernatants and pellets was quantified by ELISA. Shown are the mean values of triplicate parallel determinations, with error bars depicting the standard deviations. All experiments in this figure were performed at least three times, with similar outcomes.

FIG. 4.

Mutations that alter HIV-1 core stability modulate HIV-1 sensitivity to PF74. (A) Wild-type and mutant HIV-1 viruses were titrated on HeLa-P4 cells in the presence of various concentrations of PF74. For each virus, infection at each concentration of PF74 was calculated as a percentage of the corresponding untreated virus infection. The experiment was performed more than three times, with similar results. (B) Binding of ³H-PF74 to the indicated HIV-1 mutant particles was tested. Samples of concentrated virus particles were incubated with the radiolabeled compound, pelleted, resuspended in PBS, and then repelleted through a 20% sucrose cushion. The pelleted particles were dissolved in sodium dodecyl sulfate-containing sample buffer, and the levels of ³H and CA protein were quantified by liquid scintillation counting and ELISA, respectively. The ratios of the radioactive signals to the CA protein were calculated and expressed as a percentage of the wild-type value. Shown are the mean values obtained from averaging the data from at least six independent experiments, with error bars representing one standard deviation.

FIG. 5.

Dependence of PF74 antiviral activity on CypA-CA interactions. (A) Macsynergy plot from analysis of Cs and PF74. The bowl-shaped surface indicates antagonism between the two inhibitors. This experiment was performed twice, with each assay performed in quadruplicate. The two experiments yielded similar outcomes. (B to D) Antiviral activity of PF74 was determined against HIV-1 in the presence or absence of 5 μM Cs (B), in CypA-depleted cells (C), and against the G89V and P90A CA mutants impaired for CypA binding (D). In panel D, R9 represents the wild-type control for G89V, whereas R7 is the control for the P90A mutant. Shown are data from representative experiments; each of the experiments in panels B to D was performed a minimum of thrice. The error bars shown in panels B to D represent the standard deviations of triplicate determinations.