Primary screening assay in 5BD.1 cells. 5BD.1 cells constitutively producing HIV-like particles in a Rev-dependent manner were used to test the initial 40,000 heterocyclic compounds. p24 production was measured by ELISA to compare viral particle production from cells exposed to compound with production from those exposed only to medium and DMSO (control). Of the compounds 5,120 are shown in this figure. On average, the p24 value for the control (100%) was 23 ± 1 ng/ml. The points below the 50% cutoff line indicate compounds that were selected for further testing. One hundred ninety-two of the original 40,000 compounds were selected in this way and analyzed further in dose-response and toxicity assays.

Efficacy and toxicity dose-response assays for 103833 and 104366 in 5BD.1 cells. All assays were performed in triplicate, and the error bars represent standard deviations. (A) Six concentrations of 103833 were tested in 5BD.1 cells for inhibition of p24 and cellular toxicity compared to control. The EC₅₀ calculated for the assay was 0.9 μM; the TC₅₀ was >30 μM. The value of 100% represents 23.3 ± 0.4 ng/ml for the p24 curve and an optical density (A₄₉₀) of 0.953 ± 0.3. ▪, p24; ○, toxicity. (B) Six concentrations of 104366 were tested in 5BD.1 cells for inhibition of p24 and cellular toxicity compared to control. The EC₅₀ calculated for the assay was 2.5 μM; the TC₅₀ was >30 μM. The value of 100% represents 22.3 ± 0.3 ng/ml for the p24 curve and an optical density (A₄₉₀) of 1.01 ± 0.13. ▪, p24; ○, toxicity. (C) Six concentrations of 103833 and 104366 were tested in MT-4 cells for cellular toxicity compared to control. The TC₅₀ calculated for 103833 was 32 μM. The TC₅₀ calculated for 104366 was 28 μM. The value of 100% represents an optical density (A₄₉₀) of 0.745 ± 0.16 for 103833 and 0.706 ± 0.58 for 104366. p24 was quantified by ELISA, and toxicity was measured using the MTS assay as described in Materials and Methods. The values obtained are plotted as percentages of the values obtained from cultures with no added drug.

Dose-response assays for 103833 and 104366 with a Rev/RRE-dependent construct and a CTE-dependent construct. 293T cells were transfected with either pCMVGagPol-RRE and pCMV Rev (▪) or pCMVGagPol-CTE (○). Compounds were added at the indicated concentrations. Medium was collected after 72 h, and p24 was measured. Data are plotted as percentages of p24 values obtained from cultures with no added drug. The value of 100% represents 7,106 ± 3 pg/ml for the CTE and 545 ± 33 pg/ml p24 for the Rev/RRE curves in the 103833 experiment (A) and 3,704 ± 454 pg/ml for the CTE and 405 ± 42 pg/ml for the Rev/RRE curves in the 104366 experiment (B). Assays were performed in triplicate, and the standard deviations are shown.

Western blot analysis of cells treated with 103833 or 104366. (A) U1 cells were treated with TNF-α (lanes I) or not treated with TNF-α (lanes U) and 5 μM of 104366 or 10 μM of 103833. After 24 h cells were harvested and subjected to Western blot analysis for the indicated proteins using specific monoclonal antibodies. Each band was quantified and normalized using the β-tubulin band as a normalization control. The results of the quantitation are shown in the lower part of the panel. (B) 293T cells (2 × 10⁶) were transfected with 5 μg of pNL4-3 (lanes +) with or without the addition of 10 μM 103833 or 8 μM 104366. At 48 h posttransfection the cells were harvested and analyzed by Western blotting as described for the U1 cells in panel A. Untransfected cells were also analyzed as a control (lanes −).

Dose-response assay for viral replication in PBMCs. PHA-stimulated PBMCs were infected with the primary HIV-1 Brazilian isolate 93BR021 at a multiplicity of infection of 0.1. The indicated concentrations of 103833, 104366, or AZT were added at the time of infection. After 7 days of culture at 37°C, supernatant was collected and assayed for reverse transcriptase, as a measure of viral yield. Toxicity was also measured after 7 days by collecting the cells and performing the MTS assay. Assays were performed in triplicate, and the standard deviations are shown. (A) 103833. The value of 100% represents 12,475 ± 810 cpm for the reverse transcriptase curve and an optical density (A₄₉₀) of 0.832 ± 0.012 for the toxicity curve. (B) 104366. The value of 100% represents 15,771 ± 1,273 cpm for the reverse transcriptase curve and an optical density (A₄₉₀) of 0.808 ± 0.060 for the toxicity curve. (C) AZT control. The value of 100% represents 18,477 ± 2,443 cpm for the reverse transcriptase curve and an optical density (A₄₉₀) of 0.800 ± 0.035 for the toxicity curve. ▪, reverse transcriptase; ○, toxicity.

Structures of compounds selected for further study. The structures of each of the compounds that gave EC₅₀s lower than 30 μM in the 5BD.1 cell assay, with at least a 1-log difference in toxicity, are shown.

Dose-response assays for 103833 and 104366 comparing their effects on Rev-dependent and Rev-independent luciferases. HeLa cells containing stably integrated expression plasmids for Renilla and firefly luciferases were treated with the indicated concentrations of compound, as described in the text, and luciferase activity was measured after 24 h of treatment. The Renilla luciferase is expressed in a Rev-dependent context, while the firefly luciferase is expressed in a Rev-independent fashion. Data are plotted as percentages of luciferase values obtained from cultures with no added drug. The value of 100% for the Renilla luciferase represents 0.122 ± 0.007, 0.150 ± 0.012, and 0.132 ± 0.006 light units for the 103833 (A), 104366 (B), and leptomycin B (C) experiments. The value of 100% for the firefly luciferase represents 16.57 ± 0.15, 15.03 ± 0.3, and 14.34 ± 0.21 light units for the 103833 (A), 104366 (B), and leptomycin B (C) experiments. Assays were performed in triplicate, and the standard deviations are shown. ▪, Renilla luciferase; ○, firefly luciferase.

Dose-response assay for 103833 and 104366 in U1 cells. U1 cells were induced with TNF-α and tested for viral replication in the presence of the compounds at the indicated concentrations. Virus release from cells was measured as reverse transcriptase in the medium. Toxicity was measured with MTS dye for cell viability as described in Materials and Methods. The values obtained are plotted as percentages of the values obtained from cultures with no added drug. Assays were performed in triplicate, and the standard deviations are shown. (A) 103833. The value of 100% represents 972 ± 97 cpm for the reverse transcriptase curve and an optical density (A₄₉₀) of 1.14 ± 0.08 for the toxicity curve. (B) 104366. The value of 100% represents 1,182 ± 89 cpm for the reverse transcriptase curve and an optical density (A₄₉₀) of 1.01 ± 0.05 for the toxicity curve. (C) Temacrazine control. The value of 100% represents 959 ± 17 cpm for the reverse transcriptase curve and an optical density (A₄₉₀) of 1.08 ± 0.15 for the toxicity curve. ▪, reverse transcriptase; ○, toxicity.

Dose-response assay for in vitro Rev-RRE binding. An EMSA was performed by incubating radiolabeled RRE probe and Rev protein expressed and purified from bacteria plus or minus the test compounds at the indicated concentrations on ice for 10 min. The resulting complexes were analyzed by electrophoresis on a native polyacrylamide gel. The gel was dried and examined using a phosphorimager. The lane marked “pr” contained the radiolabeled probe without added Rev protein.

Virus growth inhibition assay. SupT1 cells (5 × 10⁶) were infected with HIV-1 NL4-3 virus (50 ng of p24) in the presence of compound, and the cultures were passaged for 30 to 31 days by removing three-fifths of the culture on the indicated days and replacing it with fresh medium plus compound. Virus yield was determined by measuring p24 in the medium using an ELISA. Assays were performed in duplicate. (A) Compound 104366. ♦, no compound; ▴, 0.8 μM; •, 8 μM. (B) Compound 103833. ♦, no compound; ×, 1 μM; +, 10 μM.