Characterization of HIV-1 entry inhibitors with broad activity against R5 and X4 viral strains

Abstract

Background: Combined antiretroviral therapy has drastically reduced mortality and morbidity of HIV-infected individuals. Nevertheless long-term toxicity and appearance of viral resistance hampers the prolonged effectiveness of combination therapy, requiring a continuous input of drugs to replace those utilized in combination regimens. We here investigated the anti-HIV activity of novel derivatives of the suradista chemical class.

Methods: Compounds were tested on acute HIV-1 infection of activated peripheral blood mononuclear cells. HIV production was monitored by enzyme-linked immunosorbent assay measuring the protein p24 released in culture supernatants. Fusion assays were carried out to study the mechanism of action of these compounds. A modified version of a previously established recombinant vaccinia virus-based assay was used measuring activation of a reporter gene upon fusion of two distinct cell populations. Flow cytometry was performed in competition assays for the binding of several antibodies targeting different sites of the viral envelope glycoprotein gp120, or the receptor CD4, or the coreceptors CXCR4 and CCR5.

Results: Four compounds inhibited replication of a prototypic R5 (BaL) and X4 (IIIB) laboratory-adapted HIV-1 strain at low micromolar concentrations, in the absence of cytotoxicity. Approximately a ten fold greater activity was achieved against the X4 as compared to the R5 strain. The compounds blocked X4 and R5 HIV-1 fusion, a step of viral entry. This activity appeared specific for HIV-1, as entry of human herpesvirus 6 (HHV-6) and influenza virus was not substantially affected. Further investigation of the inhibitory mechanism revealed that these new molecules target the viral envelope, rather than the coreceptors, as previously shown for a congener of the same class characterized by a long plasmatic half-life. Indeed ND-4043, the most active compound, specifically competed with binding of monoclonal antibodies against the CD4-binding site (CD4-BS) and coreceptor-binding site (CoR-BS) of gp120. These compounds displayed broad anti-HIV activity, as they inhibited various primary R5, X4 and, importantly, dualtropic R5X4 HIV-1 isolates. Of the four derivatives tested, the dimeric compounds were consistently more potent than the monomeric ones.

Conclusions: Given their unique features, these molecules represent promising candidates for further development and exploitation as anti-HIV therapeutics.

Figure 1

Effect of the six compounds on HIV-1 replication (left panel X4, right panel R5) measured by the p24 assay on day 7 culture supernatants. Infections were carried out in triplicate cultures in complete medium plus 10% of heat-inactivated FCS. Data are reported as percent of control which are PBMC in the absence of any compound, but pre-incubated with DMSO at 0.2%, corresponding to the concentration present in 30 μM of the compound. Values refer to mean ± standard deviation (SD) of 4 independent experiments each with triplicate cultures.

Figure 2

Comparison of maraviroc, AMD3100 and ND-4043 on inhibition of HIV-1 replication (left panel X4, right panel R5). The antigen p24, expressed as ng/ml, is measured at day 7 post infection and data are reported as mean ± SD of three replicates. Infections were carried out in complete medium plus 10% of heat-inactivated FCS. Control PBMC were incubated with DMSO at 0.2% corresponding to the concentration present in 30 μM of ND-4043. The tested concentrations of compounds (ten fold dilutions) are shown in the figure.

Figure 3

Effect of different concentrations of the six compounds plus AMD3100 and maraviroc on HIV-1-mediated fusion (A) and cytotoxicity (B). Fusion assays were carried out in complete medium plus 2.5% of heat-inactivated FCS with SupT1-IIIB (X4) or PM1-BaL (R5) as donor cells and TZM-bl as target cells (left panel X4, right panel R5). Viral fusion data, expressed as OD, are presented as mean ± SD of three replicates on the x axes, whereas concentrations of tested compounds are reported on the y axes. B. MTT assay on day 4 cultures of TZM-bl not infected with vaccinia virus, in the presence or absence of different concentrations of the six compounds. Data are expressed as OD and represent mean ± SD of three replicates. Control TZM-bl were incubated with DMSO at 0.2% corresponding to the concentration present in 30 μM of the compounds.

Figure 4

Differential effect of ND-4043 on donors and targets in HIV-1-mediated fusion (upper panel X4, lower panel R5) entry. The fusion assay was carried out in complete medium plus 2.5% of heat-inactivated FCS. Both donor and target were 3T3 cells: as donors they were infected with vaccinia virus expressing X4 (vCB41) or R5 (vCB43) HIV-1 envelope, as targets they were stably transfected with CD4 plus CXCR4/CCR5. Ctrl = control is the mixture of the donor and target populations of 3T3 cells in the absence of ND-4043. Three incubation conditions were compared in the same experiment. Viral fusion is expressed as OD and reported data are mean ± SD of three replicates. The compound ND-4043 is tested at 30, 10, 3 and 1 μM.

Figure 5

Flow cytometric analyses in 3T3 cells expressing Env (A) or CXCR4 and CCR5 (B). Three concentrations (30, 10 and 3 μM) of the ND-4043 compound were used to compete with binding of anti HIV envelope (A) or anti coreceptor (B) moAbs. Incubation with ND-4043 was carried out in PBS plus 2% of heat-inactivated FCS. DMSO refers to control cells stained with the moAbs but pre-incubated only with DMSO at 0.2%, corresponding to the concentration present in 30 μM of the compound; uns: unstained cells. A color-coded legend is displayed within the figure.

Figure 6

Flow cytometric analyses in uninfected PM1 cells. ND-4043 is used to compete with binding of 4 anti CXCR4 (A) and 6 anti CD4 (B) moAbs. Incubation with ND-4043 was carried out in PBS plus 2% (A) or 1% (B) of heat-inactivated FCS. A. ND-4043 was used at three concentrations (30, 10 and 3 μM) and also AMD3100 (3, 0.3 and 0.03 μM). B. ND-4043 was used at 30 μM. DMSO refers to control cells stained with the moAbs but pre-incubated only with DMSO at 0.2%, corresponding to the concentration present in 30 μM of the compound; uns: unstained cells. A color coded legend is displayed within the figure.

Figure 7

Effect of compounds on entry of different viruses. A. influenza virus (flu) VLP entry in MDCK cells. ND-4041, ND-4043 and ND-4045 were tested at three concentrations; bafilomycin (100 nM) was used as positive control of flu entry inhibition. Data are expressed as percentage of control (cells in the absence of compounds) and represent mean ± SD of three replicates of two independent experiments. B. HHV-6 virus-mediated fusion of HSB-2 and TZM-bl cells. Viral entry is expressed as OD and data are mean ± SD of three replicates. ND-4043 was tested at three concentrations and the anti CD46 moAb (10, 1, 0.1 and 0.01μg/ml) was used as positive control of entry inhibition. Data are representative of two independent experiments.

Figure 8

Breadth of anti-HIV activity by the active compounds. Fusion assays were performed with target TZM-bl cells plus donor PM1 cells chronically infected with 5 X4 (A), 5 R5 (B) and 2 R5X4 viruses (C). Each panel represents a single experiment carried out with three replicates. Fusion assays were performed in complete medium plus 2.5% of heat-inactivated FCS. The four active compounds have been tested at the indicated concentrations. Viral entry is expressed as OD/min and data are reported as mean ± SD of the three replicates. C. The p24 assay, expressed as ng/ml, was performed with day 7 supernatants from PBMC infected with the three indicated R5X4 viruses and only ND-4043 was tested (30, 10, 3 μM). Values refer to mean ± SD of three replicates.