Novel Human Immunodeficiency Virus (HIV) - Antimicrobial Agents ...

2 downloads 0 Views 233KB Size Report
Mar 3, 2003 - quinoline-3-carboxylic acid (K-37) was kindly provided by M. Baba ( ... giant cell formation in CEM cell cultures at day 4 postinfection by ...
ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Oct. 2003, p. 3109–3116 0066-4804/03/$08.00⫹0 DOI: 10.1128/AAC.47.10.3109–3116.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Vol. 47, No. 10

Novel Human Immunodeficiency Virus (HIV) Inhibitors That Have a Dual Mode of Anti-HIV Action Miguel Stevens, Christophe Pannecouque, Erik De Clercq, and Jan Balzarini* Rega Institute for Medical Research, Katholieke Universiteit Leuven, B-3000 Leuven, Belgium Received 3 March 2003/Returned for modification 27 May 2003/Accepted 24 June 2003

We have found that novel pyridine oxide derivatives are inhibitors of a wide range of human immunodeficiency virus (HIV) type 1 (HIV-1) and HIV-2 strains in CEM cell cultures. Some of the compounds showed inhibitory activities against recombinant HIV-1 reverse transcriptase (RT), whereas others were totally inactive against this viral protein in vitro. Partial retention of anti-HIV-1 activity against virus strains that contain a variety of mutations characteristic of those for resistance to nonnucleoside RT inhibitors and a lack of inhibitory activity against recombinant HIV-2 RT suggested that these pyridine oxide derivatives possess a mode of antiviral action independent from HIV RT inhibition. Time-of-addition experiments revealed that these pyridine oxide derivatives interact at a postintegration step in the replication cycle of HIV. Furthermore, it was shown that these compounds are active not only in acutely HIV-1-infected cells but also in chronically HIV-infected cells. A dose-dependent inhibition of virus particle release and viral protein expression was observed upon exposure to the pyridine oxide derivatives. Finally, inhibition of HIV-1 long terminal repeatmediated green fluorescence protein expression in quantitative transactivation bioassays indicated that the additional target of action of the pyridine oxide derivatives may be located at the level of HIV gene expression. antiviral activity relationship of a new class of anti-HIV compounds that may have multiple mechanisms of antiviral action (4). Furthermore, it has recently been shown that the HIV-1specific inhibitors within the class of pyridine oxide derivatives were endowed with properties characteristic of those of nonnucleoside RT inhibitors (NNRTIs) (18). Here, we provide convincing evidence that, in addition to HIV-1 RT inhibition, the pyridine oxide derivatives also interfere with viral gene expression after the proviral DNA has been incorporated into the genomes of the target cells.

During the last decade, numerous compounds have been reported to inhibit the replication of human immunodeficiency virus (HIV) in cell culture. Among these, a variety of HIV type 1 (HIV-1) reverse transcriptase (RT) and protease inhibitors and one fusion inhibitor have been formally licensed for clinical use in the treatment of HIV-1 infections (8, 9). Treatment of HIV-infected individuals at present is based on combination therapy with HIV RT and/or protease inhibitors. Since these compounds interact with virus-specific enzymes, the emergence of drug-resistant viruses may not be avoidable during long-term drug treatment. Indeed, multidrug-resistant virus strains have increasingly been reported in patients receiving highly active antiretroviral therapy (19). Therefore, the discovery of new antiviral targets or the development of novel antiviral strategies for anti-HIV treatment is necessary. Several targets in the HIV-1 replication cycle other than RT, protease, and virus entry have been identified as possible intervention sites for antiviral chemotherapy. Among these, viral gene regulation (processes) seems to be very attractive, as it would open the possibility to control HIV-1 replication not only in acutely infected cells but also in chronically infected cells. In this way, inhibitors of HIV gene regulation may have great potential in anti-HIV drug combination therapy because they can force the virus to slow down its replication rate or even shut off replication and establish a dormant state. Furthermore, it may be argued that the use of antivirals targeted at HIV gene expression would result in the development of a lower incidence of drug resistance, as HIV gene regulation requires the interplay of both viral and cellular components (2, 6). Recently, we have reported the identification and structure-

MATERIALS AND METHODS Compounds. Pyridine oxide derivatives (Fig. 1) were synthesized at and supplied by Crompton Corporation (Middlebury, Conn., and Guelph, Ontario, Canada). The thiocarboxanilide NNRTI UC-781 was obtained from W. Brouwer (Crompton Corporation, Guelph, Ontario, Canada). Tenofovir [(R)-9-(2-phosphonylmethoxypropyl)adenine] was a kind gift from N. Bischofberger (Gilead Sciences, Foster City, Calif.). Zidovudine was obtained from D. G. Johns (National Institutes of Health, Bethesda, Md.), and lamivudine was provided by J. Cameron (at that time at Glaxo-Wellcome, Stevenage, United Kingdom). Dextran sulfate (average molecular weight, 5,000) was purchased from Sigma. AMD-3100 was obtained from G. Henson (AnorMED, Langley, British Columbia, Canada). Ritonavir (ABT538) was kindly provided by J. M. Leonard (Abbott Laboratories, Abbott Park, Ill.). The fluoroquinolone derivative 7-(3,4-dehydro4-phenyl-1-piperidinyl)-1,4-dihydro-6-fluoro-1-methyl-8-trifluoromethyl-4-oxoquinoline-3-carboxylic acid (K-37) was kindly provided by M. Baba (Fukushima Medical College, Fukushima, Japan). Cells and viruses. Human lymphocyte CEM cells were obtained from the American Tissue Cell Culture Collection (Manassas, Va.). Cells were maintained in RPMI 1640 medium (Life Technologies, Merelbeke, Belgium) supplemented with 10% heat-inactivated fetal calf serum (FCS; Integro, Zaandam, The Netherlands), 2 mM L-glutamine (Life Technologies), and 0.1% NaHCO3 (Life Technologies) and incubated at 37°C in a humidified CO2-controlled atmosphere. Peripheral blood mononuclear cells were isolated from buffy coats from HIVseronegative donors by using a Ficoll gradient (Lymphoprep; Nycomed Pharma AS Diagnostics, Oslo, Norway); stimulated for 3 days in medium containing phytohemagglutinin (2 ␮g/ml; Sigma Chemical Co., Bornem, Belgium); washed; and resuspended in RPMI 1640 medium supplemented with 2 mM L-glutamine, 15% FCS, and recombinant human interleukin 2 (10 U/ml; Boehringer Mannheim, Darmstadt, Germany). HeLa–CD4–long terminal repeat (LTR)–␤-galac-

* Corresponding author. Mailing address: Rega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, B-3000 Leuven, Belgium. Phone: 0032 16 33 73 41. Fax: 0032 16 33 73 40. E-mail: [email protected]. 3109

3110

STEVENS ET AL.

FIG. 1. Structural formulae of pyridine oxide derivatives. tosidase cells (kindly provided by P. Charneau, Institut Pasteur, Paris, France) and HLtat cells (contributed by B. K. Felber and G. N. Pavlakis, National Institutes of Health) were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% FCS. CEM cells chronically infected with HIV-1 and HIV-2 were obtained after infection of CEM cells with the HIV-1(IIIB) and HIV-2(ROD) strains, respectively. HIV-1(IIIB), HIV-1(MN), and HIV-1(RF) (15) were provided by R. C. Gallo and M. Popovic (at that time at the National Institutes of Health). HIV-1(HE) represents a clinical isolate obtained from a patient with AIDS in Leuven, Belgium. HIV-2(ROD) (5) and HIV-2(EHO) (17) were both provided by L. Montagnier (at that time at the Pasteur Institute, Paris, France). Cytotoxicity and antiviral activity assays. All cytotoxicity and antiviral activity assays were performed in 96-well microtiter plates (Falcon 3072; Becton Dickinson, Paramus, N.J.). Determination of cytotoxicity involved plating of 4 ⫻ 104 CEM cells (100 ␮l) into each well in the presence of a given amount of the test compound (100 ␮l). The cells were allowed to proliferate for 96 h at 37°C in a humidified atmosphere in which the CO2 concentration was controlled. At the end of the incubation period, the cells were counted in a Coulter counter (model ZB; Coulter Electronics Ltd., Harpenden, England). The 50% cytotoxic concentration (CC50) was defined as the concentration of compound that inhibited CEM cell proliferation by 50%. The procedures used to assess the anti-HIV activity in cell culture were based on assessment of inhibition of HIV-induced giant cell formation in CEM cell cultures at day 4 postinfection by microscopic examination. Briefly, CEM cells were suspended at 250,000 cells ml⫺1 in culture medium and infected with HIV at approximately 100 times the 50% cell culture infectious dose (CCID50) per ml. Then, 100 ␮l of the infected cell suspension was added to 200-␮l microtiter plate wells containing 100 ␮l of an appropriate dilution of the test compounds. After 4 days of incubation at 37°C, the cell cultures were examined for syncytium formation. The 50% effective concentration (EC50) was defined as the compound concentration required to inhibit virus-induced syncytium formation by 50%.

ANTIMICROB. AGENTS CHEMOTHER. RT assays. The RT assays with recombinant HIV-1 and HIV-2 RTs were performed as described previously (3). Briefly, the reaction mixture (50 ␮l) contained 50 mM Tris-HCl (pH 7.8), 5 mM dithiothreitol, 300 mM glutathione, 500 ␮M EDTA, 150 mM KCl, 5 mM MgCl2, 1.25 ␮g of bovine serum albumin, a fixed concentration of the labeled substrate dGTP (2 ␮Ci of [3H]dGTP [13.8 ␮M] per assay), a fixed concentration of the template-primer poly(rC) 䡠 oligo(dG) (0.1 mM), 0.06% Triton X-100, 10 ␮l of inhibitor solution (containing various concentrations of the compounds), and 5 ␮l of the purified recombinant HIV RT preparation. The reaction mixtures were incubated at 37°C for 60 min, at which time 200 ␮l of calf thymus DNA in H2O (2 mg ml⫺1) and 1 ml of saturated sodium phosphate buffer (an equimolar concentration in NaH2PO4 and Na2HPO4) in a 5% (vol/vol) aqueous trichloroacetic acid solution were added. The solutions were kept on ice for 30 min, after which the acid-insoluble material was washed and analyzed for radioactivity. Syncytium formation assays. MOLT-4 (clone 8) cells and C8166 cells (106 cells/ml) were cultured in the presence of CEM cells (106 cells/ml) persistently infected with HIV-1(IIIB) or HIV-2(ROD) in microtiter plate wells containing various concentrations of the test compounds. After a 12-h cocultivation period, the number of giant cells (syncytia) was recorded microscopically, as described previously (1). MAGI assay. HeLa–CD4–LTR–␤-galactosidase cells (11), which use Tat protein-induced transactivation of the ␤-galactosidase gene driven by the HIV-1 LTR promoter, were included in our study to evaluate the inhibitory activities of the pyridine oxide derivatives on ␤-galactosidase expression in these monolayer cells and to perform time-of-addition experiments. The multinuclear activation of a galactosidase indicator (MAGI) assay involved plating of 2 ⫻ 104 HeLa– CD4–LTR–␤-galactosidase cells in 200 ␮l of cell culture medium in flat-bottom, 96-well microtiter plates. After an overnight incubation, the medium was removed and replaced by 100 ␮l of virus-containing medium, followed by the addition of the pyridine oxides at various concentrations. Individual blue-stained (␤-galactosidase-expressing) cells or syncytia were counted microscopically after incubation with 5-bromo-4-chloro-3-indolyl-␤-D-galactopyranoside (X-Gal). The time-of-addition experiment was performed by addition of a fixed concentration of compound to a number of wells (containing HIV-exposed cell cultures) corresponding to time zero. After an additional hour of incubation, the monolayer was washed extensively to remove unbound virus and was replaced by 100 ␮l of fresh medium. The wells corresponding to time courses of 0 and 1 h were supplemented with 100 ␮l of test compound at the concentration appropriate for that well. At different time intervals (1.5 to 41.5 h), 100 ␮l of an appropriate concentration of test compound was added to the corresponding wells. At 48 h postinfection, the cells were fixed with a 1% formaldehyde–0.2% glutaraldehyde solution and stained with 4 ␮M potassium ferrocyanide, 4 ␮M potassium ferricyanide, 2 ␮M MgCl2, and 0.4% X-Gal in phosphate-buffered saline. Blue multinuclear cells were counted microscopically. Time-of-addition experiments in CEM cells. CEM cells (5 ⫻ 105 cells/ml) were infected with HIV-2(ROD) at approximately 7,500 times the CCID50 per ml. Following a 2-h adsorption period, the cells were washed three times and incubated at 37°C. Test compounds were added at different times (0, 1, 3, 5, 7, 9, 12, 18, 24, and 36 h) after infection: AMD-3100 at 10 ␮g/ml, lamivudine at 20 ␮g/ml, tenofovir at 200 ␮g/ml, ritonavir at 10 ␮g/ml, K-37 at 2.2 ␮g/ml, and the pyridine oxide derivative JPL-153 at 20 ␮g/ml. Viral p24 antigen production was determined at 72 h postinfection by an HIV-2 p24 enzyme-linked immunosorbent assay (Innogenetics, Ghent, Belgium). Inhibitory activities of pyridine oxide derivatives for virus production in CEM cells chronically infected with HIV. CEM cells chronically infected with HIV1(IIIB) were washed twice (to remove cell-free virus) and resuspended at 120,000 cells/well in fresh medium containing the test compounds at the appropriate concentrations. After 48 h of incubation at 37°C, the supernatants of the cell cultures were collected to determine p24 production, the viral RNA load, and viral infectivity. Viral p24 production was assessed by an HIV-1 p24 enzymelinked immunosorbent assay (NEN, Brussels, Belgium). HIV-1 RNA was quantified by the VERSANT HIV-1 RNA (version 3.0) branched DNA (bDNA) signal amplification nucleic acid probe assay on a bDNA analyzer (System 340; Bayer Corporation, Tarrytown, N.Y.). A standard curve was defined from standards containing known concentrations of ␤-propiolactone-treated virus. The concentrations of HIV-1 RNA in the culture specimens were determined from this standard curve. Finally, virus titers in C8166 cells were assessed by the endpoint dilution method of Reed and Muench (16). The same cell cultures were used for 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) viability staining as described previously (14). Inhibition of HIV-1 transactivation. Tat-dependent transactivation was monitored mainly as described previously (7), with the following modifications. The HeLa-derived HLtat cell line stably expresses the HIV-1 Tat protein. The cells

VOL. 47, 2003

ANTI-HIV ACTIVITIES OF PYRIDINE OXIDES

3111

TABLE 1. Activities of pyridine oxide derivatives against different HIV strains in CEM cell cultures EC50 (␮g/ml)a Compound

JPL-10 JPL-30 JPL-44 JPL-88 JPL-153 a

HIV-1

CC50 (␮g/ml)a

HIV-2

IIIB

RF

MN

HE

ROD

EHO

5.3 ⫾ 2.6 5.9 ⫾ 3.0 8.0 ⫾ 0.8 2.4 ⫾ 0.1 2.0 ⫾ 0.5

7.2 ⫾ 2.6 10 ⫾ 1 9.0 ⫾ 2.0 12 ⫾ 1 10 ⫾ 1

10 ⫾ 1 40 ⫾ 4 9.4 ⫾ 2.0 7.6 ⫾ 0.5 7.6 ⫾ 0.5

11 ⫾ 4 38 ⫾ 2 11 ⫾ 1 5.8 ⫾ 2.9 4.0 ⫾ 1.9

10 ⫾ 1 40 ⫾ 4 8.0 ⫾ 1.6 1.5 ⫾ 0.1 2.0 ⫾ 0.2

14 ⫾ 1 44 ⫾ 5 12 ⫾ 2 2.1 ⫾ 0.4 2.4 ⫾ 0.3

31 ⫾ 6 70 ⫾ 20 30 ⫾ 6 35 ⫾ 8 31 ⫾ 8

Data represent the averages ⫾ standard deviation for at least three independent experiments.

were transfected in 0.7 ml of medium with 8 ␮g of plasmid pHIV-GFPemd DNA by electroporation with an Easyject One electroporator (Cell one, Herstal, Belgium) in a 4-mm cuvette at 200 V, 1,650 ␮F, and infinite resistance plasmid pHIV-GFPemd contains the green fluorescence protein (GFP) gene driven by the HIV-1 LTR promoter. The electroporated cells (70 ⫻ 103/well) were incubated in 96-well microtiter plates for 24 h in the presence of various concentrations of the test compounds. Then, medium was removed by gentle aspiration and the monolayers were washed with PBS. Inhibition of transactivation was measured with a Fluorocount apparatus (Packard) by quantification of GFP reporter gene activity 24 h after transfection. The 50% inhibitory concentration (IC50) was defined as the inhibitor concentration that reduced the level of GFP expression by 50%. The cytotoxicities of the test compounds for the cells were determined in the same cell cultures by the MTT method. The experiments performed to determine the IC50s and CC50s of the test compounds were performed in quadruplicate.

RESULTS Antiviral activity spectra of pyridine oxide derivatives. Various pyridine oxide derivatives inhibited the replication of different strains of HIV-1 (strains IIIB, RF, MN, and HE) and HIV-2 (strains ROD and EHO) in CEM cell cultures, with EC50s ranging from 1.5 to 44 ␮g/ml, depending on the nature of the compound and virus strain examined (Table 1). Among the compounds most inhibitory for both HIV-1 and HIV-2 strains, JPL-88 and JPL-153 exhibited similar toxicity values compared to those for the other pyridine oxide congeners tested (CC50s, 30 to 70 ␮g/ml), resulting in a selectivity index of approximately 20. When the pyridine oxide derivatives were evaluated for their antiviral activities in peripheral blood mononuclear cells, the selectivity index was further reduced as a consequence of a higher toxic effect on these cells (data not shown). We also evaluated HIV-1(IIIB) strains containing a variety of NNRTI-specific mutations in their RTs for their sensitivities to the inhibitory effects of the pyridine oxide derivatives (Table 2). Only minimal decreases (two- to fourfold) in the activities of the pyridine oxide derivatives against the mutant HIV-1 strains compared to those against the wild-type virus strains were detected. JPL-30 showed only a seven- to eightfold decrease in inhibitory potency. The inhibitory activities of these pyridine oxides were very similar against all RT mutant strains tested. These observations suggest that the predominant antiviral activity of the pyridine oxide derivatives must be attributed to the inhibition of a molecular target different from the HIV-1 RT. Inhibitory effects of pyridine oxide derivatives against recombinant HIV-1 and HIV-2 RTs. In order to estimate the contribution of HIV RT as a potential antiviral target of the pyridine oxide derivatives, the compounds were evaluated for their inhibitory effects on recombinant HIV-1 and HIV-2 RTs.

Previously, it was shown that the most active HIV-1-specific inhibitors (JPL-133 and JPL-71) within the class of pyridine oxide derivatives inhibited HIV-1 RT, with IC50s of 2.4 and 5.2 ␮g/ml, respectively (18). JPL-10, JPL-30, and JPL-44 marginally inhibited the HIV-1 RT, with IC50s of 107, 94, and 186 ␮g/ ml, respectively, whereas JPL-88 and JPL-153 were not found to inhibit HIV-1 RT at concentrations as high as 500 ␮g/ml. Also, none of the compounds inhibited HIV-2 RT at 500 ␮g/ml. Syncytium formation in cocultures of CEM cells persistently infected with HIV-1 and HIV-2 and uninfected CD4ⴙ cells. The pyridine oxide derivatives (i.e., JPL-10, JPL-44, and JPL88) failed to inhibit HIV-induced syncytium formation on cocultivation of CEM cells chronically infected with HIV-1and HIV-2 and uninfected MOLT-4 (clone 8) and C8166 cells at concentrations as high as 20 ␮g/ml under experimental conditions in which dextran sulfate and AMD-3100 effectively suppressed syncytium formation between virus-infected and uninfected cells (data not shown). Effects of pyridine oxide derivatives on ␤-galactosidase production in HeLa–CD4–LTR–␤-galactosidase cell cultures. The MAGI assay was used to determine whether pyridine oxide derivatives interfered with steps in the viral infection cycle from virus adsorption to HIV gene transcription. Inhibition of HIV-1(IIIB) and HIV-2(ROD) replication in HeLa–CD4– LTR–␤-galactosidase cells was measured by estimation of the number of blue-stained syncytia in drug-treated cell cultures compared to the number in virus-infected control cell cultures. Specific HIV RT inhibitors, such as the nucleotide RT inhibitor tenofovir (IC50 ⫽ 0.8 ␮g/ml) and the NNRTIs UC-781 (IC50 ⫽ 0.001 ␮g/ml) and JPL-133 (IC50 ⫽ 0.1 ␮g/ml), all

TABLE 2. Sensitivities of mutant HIV-1(IIIB) strains to various pyridine oxide analogues in CEM cell cultures Compound

JPL-10 JPL-30 JPL-44 JPL-88 JPL-153

EC50 (␮g/ml)a Wild type

Leu100Ile

Lys103Asnb

Glu138Lysb

Tyr181Cysb

5.3 ⫾ 2.6 5.9 ⫾ 3.0 8.0 ⫾ 0.8 2.4 ⫾ 0.1 2.0 ⫾ 0.5

10 ⫾ 2 40 ⫾ 8 10 ⫾ 2 4.5 ⫾ 2.0 4.6 ⫾ 2.6

11 ⫾ 5 44 ⫾ 4 9.0 ⫾ 1.2 11 ⫾ 5 9.0 ⫾ 1.4

12 ⫾ 4 48 ⫾ 24 15 ⫾ 6 10 ⫾ 3 9.5 ⫾ 0.8

20 ⫾ 6 50 ⫾ 4 20 ⫾ 7 10 ⫾ 2 9.4 ⫾ 0.8

a Data represent the averages ⫾ standard deviation for at least three independent experiments. b HIV-1 (IIIB) strains that emerged upon exposure to a variety of pyridine oxide derivatives are well described elsewhere (18). The Lys103Asn and the Tyr181Cys RT mutants were selected after exposure to JPL-71, whereas the Glu138Lys RT mutation emerged after treatment of HIV-1 (IIIB)-infected CEM cell cultures with JPL-74.

3112

STEVENS ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 2. Time-of-addition experiment with HeLa–CD4–LTR–␤-galactosidase cells (MAGI assay) in which the test compounds were added at different times after infection with HIV-1(IIIB) (A) or HIV-2(ROD) (B).

markedly inhibited lacZ expression and, thus, the appearance of blue-stained syncytia after infection with HIV-1(IIIB). The pyridine oxide derivatives JPL-32, JPL-58, and JPL-88 inhibited both HIV-1(IIIB)- and HIV-2(ROD)-directed lacZ expression, with IC50s of 0.5, 4, and 6 ␮g/ml, respectively, for HIV-1 and 2, 12, and 20 ␮g/ml, respectively, for HIV-2. The

fluoroquinoline K-37, known as an HIV transactivation inhibitor (13), prevented the appearance of blue-stained syncytium formation, with an IC50 as low as 0.02 ␮g/ml. Next, a time-of-addition experiment with HeLa–CD4–LTR– ␤-galactosidase cells was performed to determine the possible step(s) in the HIV replication cycle that is inhibited by the

VOL. 47, 2003

ANTI-HIV ACTIVITIES OF PYRIDINE OXIDES

3113

FIG. 3. Time-of-addition experiment. CEM cell cultures were infected with HIV-2(ROD) at a multiplicity of infection of more than 7,500 times the CCID50 per milliliter, and the test compounds were added at different times postinfection. Viral p24 antigen production was determined 72 h after infection.

pyridine oxide compounds (Fig. 2A and B). Dextran sulfate was found to interact at a very early step (virus adsorption) in the HIV-1(IIIB) and HIV-2(ROD) infection cycle and lost its protective effect when its administration was delayed for a few hours after infection. The nucleoside RT inhibitors lamivudine and zidovudine and the nucleotide RT inhibitor tenofovir retained their full inhibitory activities when drug addition was delayed for no longer than 5 h after infection. The addition of the NNRTIs UC-781 and JPL-133 could be delayed for an additional 2 h (up to ⬃7 h) before these drugs lost their antiviral efficacies (Fig. 2A). In sharp contrast, the pyridine oxide derivatives JPL-32, JPL-58, and JPL-88 clearly interacted with a later step in the HIV-1 and HIV-2 replication cycle; and their administration to the virus-infected cell cultures could be delayed for at least 15 h postinfection without any loss of antiviral potency in the MAGI assay. In this assay, the pyridine oxide derivatives showed the same pattern of virus inhibition as the HIV transactivation inhibitor K-37. Time-of-addition experiments in HIV-2-infected CEM cell cultures. Time-of-addition experiments, in which the test compounds were added at different time points after infection of the CEM cell cultures with HIV-2(ROD), were also carried out by measurement of p24 antigen production as parameter of viral replication. These experiments revealed that the antiviral potential of JPL-153 in the virus-infected cell cultures could be preserved when drug addition was delayed for at least 15 h after infection (Fig. 3). Longer delay of drug addition resulted in diminished suppression of virus production compared to that for the untreated control. This is in contrast to the RT inhibitors, such as lamivudine and tenofovir, which lost their antiviral effects when they were added later than 4 h postin-

fection. These observations indicate that JPL-153 may likely interact at a postintegration step. These results also confirm our findings obtained in the MAGI assays, which also pointed to a postintegration step in the HIV replication cycle as the possible antiviral target of the pyridine oxide derivatives. Inhibitory effects of pyridine oxide derivatives on HIV production in CEM cells chronically infected with HIV. We evaluated the pyridine oxide analogues JPL-153 and JPL-88 for their inhibitory effects on virus production by CEM cells chronically infected with HIV-1(IIIB) (Fig. 4). JPL-58 was also included in this study because its CC50s and EC50s for HIV1(IIIB) in CEM cell cultures were in the same range as those of JPL-153 and JPL-88 (24 and 3.8 ␮g/ml, respectively). Cell supernatants were subjected to analysis of three different parameters for viral infection (p24 production, viral RNA load determination, and virus titrations in C8166 cells by the endpoint dilution method of Reed and Muench [16]) at 48 h after drug administration. For all three compounds concentrationdependent inhibition of p24 production, viral RNA load, and release of infectious virus particles in the supernatants of chronically infected CEM cell cultures was observed. JPL-153, JPL-58, and JPL-88 reduced the HIV-1 progeny yield by 50% at 10, 15 and 17 ␮g/ml, respectively, whereas p24 and viral RNA production in the cell culture supernatants were inhibited by 50% at drug concentrations ranging from 20 to 40 ␮g/ml. JPL-133, which was previously found to be a pyridine oxide derivative that, like UC-781, behaved as an HIV-1-specific NNRTI, was included in this experiment as a negative control. In the presence of this compound no inhibition of p24 production or decrease in the viral RNA load at concentrations up to 40 ␮g/ml was observed. The latter concentration of the

3114

STEVENS ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 4. Inhibitory effects of pyridine oxides on the release of infectious HIV-1 progeny (virus yield) (bars), p24 production (■), and RNA viral load (Œ) in the supernatants of chronically HIV-1(IIIB)infected CEM cells. The cytotoxicities of the test compounds were determined by the MTT viability staining method (}).

tion inhibitor K-37 inhibited Tat-dependent GFP expression by 50% at 0.3 ␮g/ml, a concentration that was about 10 times lower than its CC50. DISCUSSION

drug was approximately 800-fold higher than its EC50. At concentrations greater than 40 ␮g/ml, all compounds became cytotoxic after 2 days of incubation. Inhibition of Tat-mediated HIV-1 transactivation. A GFP transactivation assay allowed us to quantify HIV-1 LTR transactivation by Tat in HLtat cells after transient transfection with pHIV-GFPemd. As shown in Fig. 5, the pyridine oxide derivatives inhibited Tat-mediated reporter gene expression from the transiently transfected HLtat cells in a dose-dependent manner. JPL-32 at 1.6 ␮g/ml inhibited HIV-1 LTR-mediated GFP expression by 50% under conditions in which cell viability was fully preserved. On the other hand, JPL-58 and JPL-153 proved to be less potent in this reporter assay, with inhibitory actions that were 10-fold higher than that of JPL-32, whereas JPL-88 at concentrations below 80 ␮g/ml did not cause any marked decrease in GFP expression. As expected, the NNRTI JPL-133 was entirely inactive in this assay at concentrations as high as its CC50. In contrast, the well-established transactiva-

Among the new class of pyridine oxide anti-HIV agents, several pyridine oxide derivatives selectively inhibited the replication of HIV-1 (i.e., JPL-133), whereas other compounds (i.e., JPL-88 and JPL-153) were able to inhibit the replication of both HIV-1 and HIV-2. In this study we focused on investigation of the molecular mechanism of antiviral action of the latter type of pyridine oxide compounds. The compounds that were active against both HIV-1 and HIV-2 were, in general, more cytotoxic than the HIV-1-specific pyridine oxides (i.e., JPL-133). The CC50s of the most selective compounds ranged between 30 and 70 ␮g/ml in CEM cell cultures, with the EC50s for virus replication being about 15 to 20 times lower. These pyridine oxides were evaluated for their inhibitory effects on recombinant HIV-1 RT in vitro and against a variety of mutant HIV-1(IIIB) strains in cell culture to determine the potential contribution of the NNRTI effect to the antiviral potencies of this group of HIV inhibitors. In contrast to the NNRTI JPL-133 (concentration required for anti-HIV-1 RT activity, 2.4 ␮g/ml), only JPL-10, JPL-30, and JPL-44 were able to inhibit HIV-1 RT to some extent, whereas JPL-88 and JPL-153 showed absolutely no inhibitory activity against HIV-1 RT at concentrations as high as 500 ␮g/ml. Furthermore, the EC50s of JPL-10, JPL-30, and JPL-44 for HIV-2 were comparable to the EC50s for the mutant virus strains, whereas the EC50s of JPL-88 and JPL-153 for HIV-2 were even lower than the EC50s for the HIV-1 mutants. Consequently, these observations indicate that a potential “NNRTI” effect is of less importance as a contributor to the antiviral

VOL. 47, 2003

ANTI-HIV ACTIVITIES OF PYRIDINE OXIDES

3115

FIG. 5. Effects of pyridine oxide derivatives on the expression of GFP (bars) in HLtat cells. Cytotoxicity was determined in the same cell cultures by a tetrazolium-based (MTT) cell viability assay (}).

potencies of these pyridine oxide derivatives. In particular, JPL-88 and JPL-153 may be considered members of the novel class of pyridine oxide derivatives that inhibit both HIV-1 and HIV-2 replication through a mechanism of action that is different from that of the NNRTIs. Several observations point to an interaction of the pyridine

oxide derivatives at a postintegration step in the HIV replication cycle. The pyridine oxides proved to be inhibitory for HIV-1(IIIB) and HIV-2(ROD) in the MAGI assays, with JPL-32 being the most potent inhibitor of lacZ expression, followed by JPL-58 and JPL-88. Remarkably, this spectrum of activity is just the opposite of that found in CEM cell cultures, in which JPL-88 proved to be more active than JPL-58 and in which JPL-32 even exhibited poor, if any, antiviral potency. A time-of-addition experiment carried out by the MAGI assay with both HIV-1 and HIV-2 strains revealed that addition of the pyridine oxides JPL-32, JPL-58, and JPL-88 to the HIVinfected cell cultures could be delayed much longer (⬃15 h) than the time of delay of addition of the HIV RT inhibitors before a loss of antiviral activity. Therefore, these compounds must interact at a postintegration step in the replication cycle of the virus. Furthermore, the data in Fig. 2B also show that JPL-32 inhibited HIV-2 production to the same extent that the known transactivation inhibitor K-37 did. This conclusion was further supported by the time-of-addition experiments with CEM cell cultures infected with HIV-2(ROD). An equally potent member of this new class of pyridine oxide derivatives, JPL-153, still inhibited p24 production until integration of HIV-2 was fully completed. After this crucial step, p24 was released in a time-dependent manner, as seen in the MAGI assay system. These results clearly indicate that the second target of the pyridine oxide derivatives must be located beyond the integration process. Also, in agreement with these data, prevention of the appearance of HIV-1(IIIB) or HIV-2(ROD)

3116

STEVENS ET AL.

proviral DNA in CEM cells in the presence of the pyridine oxide derivatives was never observed (data not shown). The pyridine oxide derivatives not only inhibited acute HIV-1 and HIV-2 infections but also inhibited p24 production in cells chronically infected with HIV. All three parameters, i.e., p24 production, viral RNA release, and viral infectivity, were evaluated; and JPL-153, JPL-58, and JPL-88 caused similar and concentration-dependent decreases in these parameters. Thus, not only were p24 antigen production and the number of viral RNA copies in the cell culture supernatants decreased, but as a result, viral infectivity also became suppressed following exposure to the pyridine oxides. Furthermore, in a recently developed HIV transactivation assay with the GFP reporter gene linked to the HIV-1 LTR promoter for detection and quantification of transactivation, JPL-32 at 2.5 ␮g/ml caused a major decrease in GFP expression, whereas cell viability was still intact. The same dose-dependent reduction in GFP expression was seen for JPL-58 and JPL-153, but within a concentration range that was 10-fold higher and with a potency that was 2-fold lower. JPL-88, however, showed less or no activity at concentrations below 80 ␮g/ml, an observation that could probably be attributed to the lower antiviral activity of this compound in HeLa cells compared to that in CEM cells. This is in sharp contrast to the findings for JPL-32, which proved to be more active in HeLa cells than in CEM cells, a cell line in which this compound showed poor, if any, antiviral activity. It is quite remarkable and important that the compounds in this new class of pyridine oxide derivatives had different inhibition profiles in CEM cells and HeLa cells. Some compounds preferentially inhibited HIV in CEM cells (e.g., JPL-88 and JPL-153), whereas other compounds showed better inhibition profiles in HeLa cells (e.g., JPL-32). Therefore, we preferentially used JPL-88 and JPL-153 in experiments with CEM cells, whereas we preferentially used JPL-32 in assays with HeLa cells. When all the data are taken together, it can be concluded that these pyridine oxide derivatives constitute a new class of HIV transactivation inhibitors. As is already known, HIV gene expression involves a complex interplay of viral and multiple cellular proteins (12). Initially, several cellular factors activate HIV transcription, resulting in the expression of Tat (10). The Tat protein subsequently interacts with a secondary RNA stem-loop structure, the transactivation-responsive element, transcribed from the LTR. As a result, the rate of transcriptional elongation is increased by inducing hyperphosphorylation at the carboxy-terminal domain of RNA polymerase II. Further experiments are ongoing in order to determine whether a viral (i.e., Tat) or cellular (i.e., NF-␬B or pTEFb) protein is targeted by the pyridine oxide derivatives in the complicated context of the HIV transactivation process. In summary, the pyridine oxides represent a class of novel HIV inhibitors possessing a dual mode of anti-HIV action. On the one hand, the compounds may or may not act as typical NNRTIs (depending on the nature of the compounds); on the other hand, they may, additionally or alternatively, interact with a postintegration step. Our findings indicate that the second target of the pyridine oxide derivatives may be located at the level of HIV gene expression. This unique feature, as well as the wealth of chemical modifications that may be introduced into this series of molecules, makes this class of compounds of

ANTIMICROB. AGENTS CHEMOTHER.

potential interest as a new lead in the development of candidate drugs for anti-HIV chemotherapy. ACKNOWLEDGMENTS We thank John Lacadie and James Pierce (Crompton Corporation, Middlebury, Conn.) for supplying the compounds used in this study. We are also grateful to David Stammers for providing the recombinant HIV-2 RT and Dirk Daelemans for supplying the pHIV-GFPemd construct. These investigations were supported by grants from the European Commission (QLRT 2000-00291, QLRT 2001-01311, and the Rene Descartes 2001 Prize HPAW-2002-90001), the Belgian Geconcerteerde Onderzoeksacties (project GOA-00/12), and the Fonds voor Wetenschappelijk Onderzoek Vlaanderen (project G.0104.98). REFERENCES 1. Baba, M., D. Schols, R. Pauwels, H. Nakashima, and E. De Clercq. 1990. Sulfated polysaccharides as potent inhibitors of HIV-induced syncytium formation: a new strategy toward AIDS chemotherapy. J. Acquir. Immune Defic. Syndr. 3:493–499. 2. Baba, M. 1997. Cellular factors as alternative targets for inhibition of HIV-1. Antivir. Res. 33:141–152. 3. Balzarini, J., M.-J. Pe´rez-Pe´rez, A. San-Fe´lix, M.-J. Camarasa, I. C. Bathurst, P. J. Barr, and E. De Clercq. 1992. Kinetics of inhibition of human immunodeficiency virus type 1 (HIV-1) reverse transcriptase by a novel HIV-1-specific nucleoside analogue [2⬘,5⬘-bis-O-(tert-butyldimethylsilyl)-␤D-ribofuranosyl]-3⬘-spiro-5⬙-(4⬙-amino-1⬙,2⬙-oxathiole-2⬙,2⬙-dioxide)thymine (TSAO-T). J. Biol. Chem. 267:11831–11838. 4. Balzarini, J., M. Stevens, G. Andrei, R. Snoeck, R. Strunk, J. B. Pierce, J. A. Lacadie, E. De Clercq, and C. Pannecouque. 2002. Pyridine oxide derivatives: structure-activity relationship for inhibition of human immunodeficiency virus and cytomegalovirus replication in cell culture. Helv. Chim. Acta 85:2961–2974. 5. Barre´-Sinoussi, F., J. C. Chermann, F. Rey, M. T. Nugeyre, S. Chamaret, J. Gruest, C. Dauguet, C. Axler-Blin, F. Ve´zinet-Brun, C. Rouzioux, W. Rozenbaum, and L. Montagnier. 1983. Isolation of a T-lymphotropic retrovirus from a patient at risk for AIDS. Science 220:868–871. 6. Daelemans, D., A.-M. Vandamme, and E. De Clercq. 1999. Human immunodeficiency virus gene regulation as a target for antiviral chemotherapy. Antivir. Chem. Chemother. 10:1–14. 7. Daelemans, D., E. De Clercq, and A.-M. Vandamme. 2001. A quantitative GFP-based bioassay for the detection of HIV-1 Tat transactivation inhibitors. J. Virol. Methods 96:183–188. 8. De Clercq, E. 2001. New developments in anti-HIV chemotherapy. Pure Appl. Chem. 73:55–66. 9. De Clercq, E. 2002. New anti-HIV agents and targets. Med. Res. Rev. 22:531–565. 10. Jones, K. A., and B. M. Peterlin. 1994. Control of RNA initiation and elongation at the HIV-1 promotor. Annu. Rev. Biochem. 63:717–743. 11. Kimpton, J., and M. Emerman. 1992. Detection of replication-competent and pseudotyped human immunodeficiency virus with a sensitive cell line on the basis of activation of an integrated ␤-galactosidase gene. J. Virol. 66: 2232–2239. 12. Kingsman, S., and A. J. Kingsman. 1996. The regulation of human immunodeficiency virus type-1 gene expression. Eur. J. Biochem. 240:491–507. 13. Okamoto, H., T. Cujec, M. Okamoto, B. M. Perlin, M. Baba, and T. Okamoto. 2000. Inhibition of the RNA-dependent transactivation and replication of human immunodeficiency virus type 1 by a fluoroquinoline derivative K-37. Virology 272:402–408. 14. Pauwels, R., J. Balzarini, M. Baba, R. Snoeck, D. Schols, P. Herdewijn, J. Desmyter, and E. De Clercq. 1988. Rapid and automated tetrazolium-based colorimetric assay for the detection of anti-HIV compounds. J. Virol. Methods 20:309–321. 15. Popovic, M., M. G. Sarngadharan, E. Read, and R. C. Gallo. 1984. Detection, isolation and continuous production of cytopathic retrovirus (HTLVIII) from patients with AIDS and pre-AIDS. Science 224:497–500. 16. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty percent endpoints. Am. J. Hyg. 27:493–497. 17. Rey, M.-A., B. Kurst, A. G. Laurent, D. Gue´tard, L. Montagnier, and A. G. Hovanessian. 1989. Characterization of an HIV-2-related virus with a smaller sized extra-cellular envelope glycoprotein. Virology 173:258–267. 18. Stevens M., C. Pannecouque, E. De Clercq, and J. Balzarini. 2003. Inhibition of human immunodeficiency virus by a new class of pyridine oxide derivatives. Antimicrob. Agents Chemother. 47:2951–2957. 19. Vandamme A.-M., K. Van Vaerenbergh, and E. De Clercq. 1998. Anti-human immunodeficiency virus drug combination strategies. Antivir. Chem. Chemother. 9:187–203.