|Home | About | Journals | Submit | Contact Us | Français|
We previously identified a small-molecule anti-human immunodeficiency virus type 1 (anti-HIV-1) compound, ADS-J1, using a computer-aided molecular docking technique for primary screening and a sandwich enzyme-linked immunosorbent assay (ELISA) as a secondary screening method. In the present study, we demonstrated that ADS-J1 is an HIV-1 entry inhibitor, as determined by a time-of-addition assay and an HIV-1-mediated cell fusion assay. Further mechanism studies confirmed that ADS-J1 does not block gp120-CD4 binding and exhibits a marginal interaction with the HIV-1 coreceptor CXCR4. However, ADS-J1 inhibited the fusion-active gp41 core formation mimicked by peptides derived from the viral gp41 N-terminal heptad repeat (NHR) and C-terminal heptad repeat (CHR), as determined by ELISA, native polyacrylamide gel electrophoresis, and circular dichroism analysis. Moreover, using a surface plasmon resonance assay, we found that ADS-J1 could bind directly to IQN17, a trimeric peptide containing the gp41 pocket region, resulting in the conformational change of IQN17 and the blockage of its interaction with a short D peptide, PIE7. The positively charged residue (K574) located in the gp41 pocket region is critical for the binding of ADS-J1 to NHR. These results suggest that ADS-J1 may bind to the viral gp41 NHR region through its hydrophobic and ionic interactions with the hydrophobic and positively charged resides located in the pocket region, subsequently blocking the association between the gp41 NHR and CHR regions to form the fusion-active gp41 core, thereby inhibiting HIV-1-mediated membrane fusion and virus entry.
Human immunodeficiency virus type 1 (HIV-1) enters target cells by binding its gp120 envelope glycoprotein (Env) surface subunit to CD4 and to a chemokine receptor (typically, CXCR4 or CCR5). The gp41 Env transmembrane subunit then changes conformation, resulting in the fusion of the viral and cellular membranes (3, 6, 18, 41). Therefore, HIV-1 gp41 plays a crucial role in the early steps of viral entry into target cells and may serve as an important target for the development of HIV-1 entry inhibitors.
The gp41 ectodomain (extracellular domain) contains three major functional regions: the fusion peptide, the N-terminal heptad repeat (NHR), and the C-terminal heptad repeat (CHR) (Fig. (Fig.1A).1A). Peptides derived from the NHR and CHR regions of gp41, designated NHR and CHR peptides, respectively, have potent activities against HIV-1 infection (22, 52, 53). One of the CHR peptides, T-20 (brand name, Fuzeon) was licensed by the U.S. FDA for use for the treatment of patients with HIV-1 infection and AIDS, especially those infected by virus resistant to the current antiretroviral drugs (26, 53). However, the application of T-20 was constrained due to its lack of oral bioavailability and high production cost. Therefore, the development of small-molecule HIV fusion inhibitors with oral availability and low production costs is urgently needed.
In the course of studying the mechanism by which CHR peptides inhibit HIV-1 fusion, it was demonstrated that when the gp41 NHR and CHR peptides are mixed at equimolar concentrations, they form a stable α-helical trimer of antiparallel heterodimers representing the fusion-active gp41 core (36). Crystallographic analysis has revealed that this is a six-stranded α-helical bundle (6-HB), in which three N helices associate to form the central trimeric coiled coil and three C helices pack obliquely in an antiparallel manner into the highly conserved hydrophobic grooves on the surface of this coiled coil (5, 46, 48). The C helix interacts with the N helix mainly through the hydrophobic residues in the grooves on the surface of the central coiled-coil trimer. Each of the grooves on the surface of the N-helix trimer has a deep pocket that accommodates three conserved hydrophobic residues in the gp41 CHR region (Fig. (Fig.1B)1B) (5), suggesting that this pocket is an attractive target when new antiviral compounds that prevent early fusion events are being designed (4, 5). However, this hydrophobic pocket in the 6-HB core is covered by the CHR peptide and cannot be used to determine the binding affinity of a compound. The NHR peptides cannot form a soluble N trimer since it has a tendency to aggregate in solution. To address this problem, Eckert et al. (12) constructed a hybrid molecule, IQN17 (Fig. (Fig.1C),1C), by conjugating the GCN4 sequence (IQ) with a short NHR peptide (N17) involved in the formation of the gp41 hydrophobic pocket. As a consequence, IQN17 is soluble and can present the hydrophobic pocket of gp41. Using IQN17, Welch et al. (49) identified a short circular anti-HIV-1 peptide consisting of d-amino acids, designated PIE7, which specifically binds to the pocket presented on IQN17. Using the gp41 pocket as a target structure, we previously applied a computer-aided molecular docking program for the primary screening of the database of a chemical library consisting of 20,000 compound structures. We selected 0.1% of the compounds with the highest docking scores for further screening by a sandwich enzyme-linked immunosorbent assay (ELISA) (23) using a monoclonal antibody (MAb), NC-1, which specifically recognizes the fusion-active gp41 6-HB core structure (20). We identified a compound, denoted ADS-J1 (Fig. (Fig.1D)1D) (8, 19), which inhibits HIV-1 fusion, possibly by binding into the gp41 pocket and interfering with the formation of the gp41 trimeric coiled-coil domain. In the present study, we provide further evidence that demonstrates that ADS-J1 interacts with gp41 in the fusion-intermediate conformation through binding to the gp41 hydrophobic pocket. ADS-J1 blocks gp41 6-HB formation, thus inhibiting the fusion between the viral and the target cell membranes. Therefore, ADS-J1 can be used as a lead compound for the design of novel HIV-1 entry inhibitors with improved potency and reduced cytotoxicity.
MT-2 cells, HIV-1IIIB chronically infected H9 (H9/HIV-1IIIB) cells, U373-MAGI-CXCR4CEM cells, HIV-1IIIB, an anti-p24 MAb (MAb 183-12H-5C), HIV immunoglobulin, and AMD-3100 were obtained from the NIH AIDS Research and Reference Reagent Program (contributors, D. Richman, E. Emerman, R. Gallo, B. Chesebro, H. Chen, and L. Barbosa). All cells were maintained in Dulbecco's modified Eagle's medium or RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) and antibiotics (100 U/ml of penicillin, 100 μg/ml of streptomycin). Peptides N36, C34 (5, 36), N36K574D, C34D632K, IQN17 (12), T22, IQ, and biotinylated IQN17 were synthesized by a standard solid-phase 9-fluorenylmethoxy carbonyl method. Biotin-PIE7 and scrambled biotin-PIE7 were synthesized with d-amino acids and oxidized as described previously (12, 49). The peptides were purified to homogeneity by high-performance liquid chromatography (HPLC). The identities of the purified peptides were confirmed by laser desorption mass spectrometry (PerSeptive Biosystems). The sequences of the peptides are listed in Table Table1.1. Rabbit antisera directed against the mixture of N36 and C34 and against IQN17 were prepared as described previously (20). Mouse MAb NC-1 specific for gp41 6-HB was prepared and characterized as described previously (20). Rabbit immunoglobulin G (IgG) and mouse IgG were purified by using protein A/G beads (Pierce, Rockford, IL). Mouse MAb 12G5 specific for CXCR4 was purchased from R&D Systems (Minneapolis, MN). Recombinant soluble CD4 (sCD4) and gp120 were obtained from Immunodiagnostics Inc. (Woburn, MA). Chloropeptin was a generous gift from Satoshi Omura and Haruo Tanaka of the Kitasato Institute, Tokyo, Japan. ADS-J1 was purified by HPLC from the crude material provided by Ciba Specialty Chemicals Corp. (High Point, NC).
The inhibitory activities of the compounds against infections by laboratory-adapted HIV-1 strains were determined as described previously (21). In brief, 1 × 104 MT-2 cells were infected with HIV-1 at 100 50% tissue culture infective doses in 200 μl of RPMI 1640 medium containing 10% FBS in the presence or the absence of compounds at graded concentrations overnight. The culture supernatants were replaced by fresh medium. For the time-of-addition assay, MT-2 cells were incubated with HIV-1IIIB at 37°C for 0, 1, 2, 3, 4, 6, and 8 h before addition of ADS-J1 (5 μM). Zidovudine (AZT) was used as a control. After culture for another 2 h, the cells were washed to remove the free virus and compounds. Fresh medium without the test compounds was added. On the fourth day postinfection, 100 μl of the culture supernatants was collected from each well; mixed with an equal volume of 5% Triton X-100; and assayed for the p24 antigen, which was quantitated by ELISA (55). Briefly, the wells of polystyrene plates (Immulon 1B; Dynex Technology, Chantilly, VA) were coated with HIV immunoglobulin in 0.85 M carbonate-bicarbonate buffer (pH 9.6) at 4°C overnight, followed by washes with 0.01 M phosphate-buffered saline (PBS) containing 0.05% Tween 20 buffer (PBS-T) and blocking with PBS containing 1% dry fat-free milk (Bio-Rad Inc., Hercules, CA). Virus lysates were added to the wells, and the plates were incubated at 37°C for 1 h. After extensive washes, anti-p24 MAb (MAb 183-12H-5C), biotin-labeled anti-mouse IgG1 (Santa Cruz Biotech, Santa Cruz, CA), streptavidin-conjugated horseradish peroxidase (SA-HRP), and the substrate tetramethylbenzidine (TMB) were sequentially added. The reaction was terminated by addition of 1 N H2SO4. The absorbance at 450 nm was recorded in an ELISA reader (Ultra 384; Tecan). Recombinant protein p24 (US Biological, Swampscott, MA) was included to establish a standard dose-response curve.
A dye transfer assay was used for the detection of HIV-1-mediated cell-cell fusion, as described previously (22, 25, 33). H9/HIV-1IIIB cells were labeled with a fluorescent reagent, calcein-acetylmethyl ester (Molecular Probes, Inc., Eugene, OR), and were then incubated with MT-2 cells (ratio, 1:5) in 96-well plates at 37°C for 2 h in the presence or the absence of the compounds tested. The fused and unfused calcein-acetylmethyl ester-labeled HIV-1-infected cells were counted under an inverted fluorescence microscope (Zeiss, Germany) with an eyepiece micrometer disc. The percent inhibition of cell-cell fusion was calculated as described previously (22).
The interaction between gp120 and CD4 in the presence or the absence of inhibitors was determined as described previously (56). Briefly, the wells of polystyrene plates were coated with 100 μl of sheep anti-gp120 antibody D7324 (Cliniqa, Fallbrook, CA) at 2 μg/ml in 0.85 M carbonate-bicarbonate buffer (pH 9.6) at 4°C overnight and were blocked with 1% dry fat-free milk in PBS at 37°C for 1 h. One hundred microliters of recombinant gp120 at 0.5 μg/ml was added, and the plates were incubated at 37°C for 1 h, followed by three washes with PBS-T. Recombinant sCD4 at 0.25 μg/ml was added in the presence of a compound, and the plates were incubated at 37°C for 1 h. After three washes, rabbit anti-sCD4 IgG (0.25 μg/ml, 100 μl/well) was added and the mixture was incubated at 37°C for 1 h. The binding of rabbit anti-sCD4 IgG was determined by sequential addition of biotinylated goat-anti-rabbit IgG, SA-HRP, and TMB. After the reaction was terminated, the absorbance at 450 nm was recorded in an ELISA reader (Tecan).
To determine whether ADS-J1 interacts with the CXCR4 HIV-1 coreceptor, a cell-based ELISA was carried out as described previously (55). Briefly, U373-MAGI-CXCR4CEM cells in Dulbecco's modified Eagle's medium containing 10% FBS were seeded into 96-well plates and cultured in a monolayer at 37°C overnight. The cells were fixed with 5% formaldehyde in 0.01 M PBS for 15 min at room temperature. The plates were washed twice with PBS-T and blocked with 1% nonfat dry milk in 0.01 M PBS (pH 7.2) for 1 h at 37°C. ADS-J1 at graded concentrations was added to cells, followed by incubation at 37°C for 30 min. MAb 12G5 and the isotype IgG2a control were then added to the cells. After incubation at 37°C for 1 h, the unbound antibodies were removed by washing the plates three times with PBS-T. Biotin-labeled goat-anti-mouse IgG (Boehringer Mannheim, Indianapolis, IL), SA-HRP, and TMB were then sequentially added. The absorbance at 450 nm was recorded in an ELISA reader. Each sample was tested in triplicate.
A previously described (23) sandwich ELISA was used to test whether ADS-J1 inhibits gp41 6-HB formation. Briefly, peptide N36 (2 μM) was preincubated with ADS-J1 at graded concentrations at 37°C for 30 min, followed by addition of C34 (2 μM). In the control experiment, N36 was preincubated with C34 at 37°C for 30 min, followed by addition of ADS-J1 at graded concentrations. After incubation at 37°C for 30 min, the mixture was added to the wells of a 96-well polystyrene plate that had been precoated with IgG (10 μg/ml) purified from rabbit antisera directed against gp41 6-HB. MAb NC-1, biotin-labeled goat-anti-mouse IgG, SA-HRP, and TMB were then sequentially added. The absorbance at 450 nm was read, and the percent inhibition by the compounds was calculated as described previously (23). All the samples were tested in triplicate.
An ELISA was established for measurement of the binding of a d-peptide, biotinylated PIE7, to IQN17 (10). In brief, the wells of a 96-well polystyrene plate were coated with IgG (10 μg/ml) purified from rabbit antisera directed against IQN17. Peptide IQN17 (120 μM) was incubated with ADS-J1 at graded concentrations at 37°C for 30 min before addition of the peptide biotin-PIE7 (10 μM). After incubation at 37°C for 30 min, the mixture was added to the plate, followed by incubation at 37°C for 60 min. After extensive washes, the biotin-PIE7 bound to the IQN17 was quantitated by the sequential addition of SA-HRP and the substrate TMB. The absorbance at 450 nm was read with an ELISA reader. The percent inhibition by the compounds was calculated as described above. Each sample was tested in triplicate.
Precast Tris-glycine gels (18%) and a Novex X-Cell II minicell (Invitrogen, Carlsbad, CA) were used for native polyacrylamide gel electrophoresis (N-PAGE). Peptide N36 was incubated with PBS or a compound at the indicated concentrations at 37°C for 30 min before addition of C34 (final concentration of N36 and C34, 40 μM). After incubation at 37°C for 30 min, the sample was mixed with Tris-glycine native sample buffer (Invitrogen, Carlsbad, CA) at a ratio of 1:1 and the mixture was then loaded onto a precast gel (10 by 1.0 cm; 25 μl in each well). Gel electrophoresis was carried out at a constant voltage of 125 V and room temperature for 2 h. The gel was then stained with Coomassie blue and imaged with a FluorChem 8800 imaging system (Alpha Innotech Corp., San Leandro, CA).
The interaction between ADS-J1 and IQN17 was characterized by employing a surface plasmon resonance (SPR) assay, as described previously (17, 34). Briefly, biotin-IQN17 (2 μg/ml) was immobilized onto the streptavidin CM5 sensor chip by the amine coupling protocol, and the unreacted sites were blocked with 1 M Tris-HCl (pH 8.5). The kinetics of the binding of ADS-J1 to the immobilized IQN17 were determined from the dose-dependent binding of ADS-J1 to IQN17. The association reaction was initiated by injecting various concentrations of ADS-J1 at a flow rate of 5 μl/min. The dissociation reaction was performed by washing the chip with PBS. At the end of each cycle, the sensor chip surface was regenerated with 0.1 M glycine-HCl (pH 2.5) for 30 s. The resulting sensograms (plots of the changes in the numbers of response units on the surface as a function of time) were analyzed with BIAeval (version 3.0) software. The curves were fitted by using a 1:1 binding model or a conformational change model. AMD-3100 (an HIV entry inhibitor targeting the CXCR4 coreceptor), biotin-PIE7, and scrambled biotin-PIE7 were used as controls for ADS-J1, while peptide IQ was included as a control for IQN17.
Circular dichroism (CD) spectroscopy was performed as described previously (30). Briefly, N36, C34, and ADS-J1 were dissolved in PBS (pH 7.2). N36 was incubated with ADS-J1 or PBS at 37°C for 30 min, followed by addition of C34. In the control, N36 was incubated with C34 at 37°C for 30 min before addition of ADS-J1 or PBS. The final concentrations of each peptide (N36 or C34) and ADS-J1 were 10 and 40 μM, respectively. After further incubation at 37°C for 30 min, the samples were cooled to room temperature. The CD spectra of isolated peptide N36 or C34 and their complexes were acquired on a CD spectropolarimeter (model J-715; Jasco Inc., Japan) at room temperature by using a 5.0-nm bandwidth, a 0.1-nm resolution, a 0.1-cm path length, a 4.0-s response time, and a 50-nm/min scanning speed. The spectra were corrected by subtraction of a blank corresponding to the solvent. The α-helical content was calculated from the CD signal by dividing the mean residue ellipticity at 222 nm by the value expected for 100% helix formation (−33,000 degrees cm2 dmol−1), as described previously (4, 35). The CD spectra of IQN17 and IQN17 preincubated with ADS-J1 were also measured as described above.
ADS-J1 was originally identified from a database of 20,000 compounds from ComGenex, Inc., Budapest, Hungary, by a virtual screening technique with the DOCK suite of software (8). We further analyzed the potential interaction of ADS-J1 with the internal trimeric coiled coil formed by N36 mutants, in which positively charged residue K574 was replaced by a negatively charged residue. Glide (version 2.5) software from Schrödinger (Portland, OR) was used for docking in the flexible (ligand) docking mode. The initial three-dimensional coordinates of the compounds were generated by using Concord software (Tripos Associates). The ligands were prepared by following a series of steps available within Glide software, which includes addition of hydrogens to the ligands, followed by assignment of the appropriate ionization state of each ligand by using the ionizer option. Since ADS-J1 has acid moieties, it is expected to be ionized at physiological pH. The hydrophobic pocket in the gp41 core X-ray structure (Protein Data Bank code, 1aik) was used as the target site for docking simulations (5). The protein was prepared through a series of steps, as described in the Glide software operating manual (Schrödinger). In brief, one of the C helices and the structural waters were removed from 6-HB to expose the hydrophobic pocket. The modified core structure was optimized by restrained minimization in the Macromodel program with an MMFF94s force field. The minimized structures of the ligands and the target protein were subsequently used in the docking simulation. The conformational flexibility of the ligands was considered by exhaustive conformational search within the Glide software augmented by a heuristic screen that removes conformations which are not suitable for receptor binding or which have a long-range hydrogen bond, whereas the protein conformations were kept fixed. The Glide software performs an exhaustive systematic search of the conformational space and uses a series of hierarchical filters to locate the possible position of the ligand (termed “pose”) in the receptor binding site during the docking simulations. Flexible Monte Carlo sampling and minimization were used to identify the best ligand poses for scoring. Schrödinger's proprietary scoring function, GlideScore, was used in the docking experiment.
Using a time-of-addition assay, we demonstrated that the inhibitory activity of ADS-J1 on p24 production was significantly decreased if ADS-J1 was added to the virus-cell mixture at 1 h postinfection, while that of AZT had no significant change even when it was added at 8 h postinfection (Fig. (Fig.2A),2A), suggesting that ADS-J1 inhibits HIV-1 replication by targeting the HIV-1 early entry stage.
A cell-cell fusion assay was carried out to determine whether ADS-J1 binds to HIV-1-infected cells or uninfected target cells. ADS-J1 at 10 μM almost completely inhibited fusion between H9/HIV-1IIIB and MT-2 cells (Fig. (Fig.2B,2B, bar I). If ADS-J1 was incubated with MT-2 cells at 37°C for 30 min, followed by two washes to remove the unbound ADS-J1 and then addition of H9/HIV-1IIIB cells, ADS-J1 had no inhibitory activity on cell fusion (Fig. (Fig.2B,2B, bar II). However, if ADS-J1 was preincubated with H9/HIV-1IIIB cells and then removed by washes before addition of MT-2 cells, ADS-J1 had partial (about 50%) inhibition (Fig. (Fig.2B,2B, bar III). If ADS-J1 and sCD4 were preincubated with H9/HIV-1IIIB cells and were removed by washes before addition of MT-2 cells, the inhibitory activity of ADS-J1 could reach about 80% (Fig. (Fig.2B,2B, bar IV). These results suggest that ADS-J1 does not bind to the uninfected target cells but, rather, to the HIV-1-infected cells. Since sCD4 could enhance the exposure of gp41 (9), it is also possible that ADS-J1 interacts with the viral Env, particularly gp41, presented on H9/HIV-1IIIB cells.
The binding of the gp120 HIV-1 Env surface unit to CD4 and then to coreceptor CXCR4 are the first two critical steps of virus entry into target cells expressing CD4 and CXCR4. To determine whether ADS-J1 blocks gp120-CD4 binding, an sCD4-based ELISA was performed with chloropeptin, a gp120-CD4 binding inhibitor with potent anti-HIV-1 activity (37), and C34 as the positive and negative controls, respectively. The results indicated that chloropeptin markedly inhibited the binding of sCD4 to gp120 at 10 μg/ml, while ADS-J1 and C34 at the same concentration had no significant inhibition of the interaction between sCD4 and gp120 (Fig. (Fig.3A),3A), suggesting that ADS-J1 is not targeted to the gp120-CD4 binding step.
To determine whether ADS-J1 binds to the HIV-1 coreceptor CXCR4, a cell-based ELISA was established by using anti-CXCR4 MAb 12G5, which specifically recognizes CXCR4 and blocks the infection of CXCR4+ cells by HIV-1 strains (38), and U373-MAGI-CXCR4CEM, the CXCR4-expressing cells (55). T-22, a potent peptidic CXCR4 antagonist (42), and C34 were included as the positive and negative controls, respectively. As shown in Fig. Fig.3B,3B, T-22 at 25 μM significantly inhibited MAb 12G5 binding to a CXCR4-expressing cell line, while ADS-J1 and C34 at the same concentration exhibited marginal inhibitory activities (18% and 8% inhibition, respectively), suggesting that ADS-J1 may have a weak interaction with the HIV-1 coreceptor, which may contribute to part of its anti-HIV-1 activity through the potential gp120-specfic mechanism (1).
The formation of the fusion-active 6-HB core is a critical step in gp41-mediated membrane fusion. In the present study, we investigated whether ADS-J1 blocked gp41 fusion-active core formation, a critical conformational change during the fusion of HIV-1 with the target cell. A model of the gp41 core was established by mixing NHR and CHR peptides N36 and C34, respectively, at equal molar concentrations. This model gp41 core structure could be detected by a sandwich ELISA with a conformation-specific MAb, MAb NC-1 (20, 23). As shown in Fig. Fig.4A,4A, when ADS-J1 was incubated with N36 before addition of C34, ADS-J1 inhibited 6-HB formation in a dose-dependent manner, with the concentration of inhibitor that caused 50% inhibition (IC50) being 3.36 μM, consistent with the level of inhibition of ADS-J1 on HIV-1-mediated cell-cell fusion (8). However, there was no inhibition if ADS-J1 was added after the association of N36 and C34. These results suggest that ADS-J1 may block the gp41 core but that it cannot disrupt the preformed gp41 core.
Subsequently, we used a convenient N-PAGE assay (32) to further verify the ADS-J1 inhibition of 6-HB formation between N36 and C34. Since N36 has a net positive charge, it can migrate up and show no band in the gel under N-PAGE conditions. The isolated peptide, C34, showed a single band (Fig. (Fig.4B,4B, lane 1). The mixture of N36 and C34 showed two bands (Fig. (Fig.4B,4B, lanes 2 and 3). The lower band is located at the same position as the band for C34, and the upper band corresponds to the band for 6-HB, as confirmed by Western blotting with MAb NC-1 (Fig. (Fig.4B,4B, lane 8). When ADS-J1 (40 μM) was preincubated with N36 before addition of C34, the upper band disappeared (Fig. (Fig.4B,4B, lane 4). However, this band appeared again when ADS-J1 was added after the complex of N36 and C34 was preformed (Fig. (Fig.4B,4B, lane 5). These results suggest that ADS-J1 interferes with gp41 6-HB formation but that it cannot disrupt 6-HB once it forms. AZT, regardless of whether it was added before or after N36 and C34 were mixed, could not inhibit 6-HB formation (Fig. (Fig.4B,4B, lanes 6 and 7). We further confirmed the dose dependence of ADS-J1 in inhibiting 6-HB formation by N-PAGE (Fig. (Fig.4C);4C); however, the IC50 was about 17-fold higher than that resulting from the sandwich ELISA (Fig. (Fig.4A).4A). This is understandable because the amount of the N36 and C34 peptides used in the N-PAGE assay was about 20-fold greater than the amount used in the sandwich ELISA in order to show a clear 6-HB band in the gel.
Subsequently, we determined whether ADS-J1 interferes with the α-helical structure of the gp41 coiled-coil domain. Previous studies demonstrated that the individual NHR peptides have a tendency to aggregate and that CHR peptides have a random coil structure in aqueous solution. However, the mixture of the NHR and CHR peptides shows a typical α-helical conformation, as measured by CD spectroscopy (35), suggesting that the interaction between the NHR and CHR peptides results in the change of their secondary structure to an α-helical coiled-coil conformation. Using a CD spectrometer, we confirmed these results and demonstrated that ADS-J1 significantly interfered with the interaction between N36 and C34 since the α-helical content of 6-HB was significantly reduced when ADS-J1 was mixed with N36 before addition of C34 (Fig. (Fig.5A).5A). However, ADS-J1 could not alter the α-helical structure of the preformed N36-C34 complex (Fig. (Fig.5B),5B), confirming that ADS-J1 interferes with the interaction between the NHR and CHR peptides to form an α-helical complex.
Welch et al. (49) previously demonstrated that a short circular d-amino acid peptide, PIE7 (Table (Table1),1), could specifically bind to the pocket presented on IQN17 (12) and inhibit HIV-1 infection. We confirmed this result by showing that the peptide PIE7 bound to IQN17 in a dose-dependent manner (data not shown). Using an SPR assay, we found that ADS-J1 significantly bound to IQN17 (Fig. (Fig.1C1C and Table Table1),1), which contains the gp41 hydrophobic pocket (Fig. (Fig.6A).6A). However, it did not bind to IQ, which lacks the N17 pocket sequence (Fig. (Fig.6B).6B). The control compound, AMD-3100 (a CXCR4 antagonist), exhibited no binding to IQN17 (Fig. (Fig.6A).6A). We then investigated whether ADS-J1 binds to the gp41 hydrophobic pocket using a competition assay with PIE7. As shown in the top curve in Fig. Fig.6C,6C, IQN17 significantly bound to the biotinylated PIE7 that was mobilized on the SA sensor chip, while IQN17 could not bind to the scrambled biotin-PIE7 (Fig. (Fig.6C,6C, dotted line, and Table Table1),1), confirming the specificity of this assay. ADS-J1-treated IQN17 lost its activity of binding to biotin-PIE7 (Fig. (Fig.6C,6C, bottom curve). In the ELISA, we also demonstrated that ADS-J1 significantly inhibited biotinylated PIE7 binding to IQN17 at an IC50 of 0.68 μM, while the control compound, AMD-3100, had no effect on the binding of biotin-PIE7 to IQN17 (Fig. (Fig.6D).6D). These results imply that ADS-J1 may interact with the hydrophobic pocket in the gp41 central trimer and block the interaction between the viral gp41 NHR and CHR regions, which impedes the formation of 6-HB and which finally results in the inhibition of HIV-1 entry and replication.
To further characterize the interaction between ADS-J1 and IQN17, the dose-dependent binding of ADS-J1 to IQN17 was examined by a real-time plasmon resonance assay. For this experiment, biotin-labeled IQN17 was immobilized on the SA sensor chip and graded concentrations of ADS-J1 were injected onto the IQN17-immobilized surface. The binding data were analyzed by using BIAeval software (version 3.0). When the data were fitted to the 1:1 binding model, it was found that ADS-J1 bound to IQN17 with a binding affinity of ~4.3 × 10−7 M, which is similar to the level of ADS-J1 required for the inhibition of biotin-PIE7 binding to IQN17, as measured by ELISA (Fig. (Fig.6A);6A); the association constant was 3.3 × 103 ms−1, and the dissociation constant was 1.4 × 10−3 s−1. However, the binding curves for ADS-J1 did not fit with a 1:1 binding model (Fig. (Fig.7A)7A) but, rather, fit perfectly with a conformational change model (Fig. (Fig.7B),7B), strongly suggesting a conformational change of IQN17 upon ADS-J1 binding. To further confirm the conformational change, we examined the secondary structure of IQN17 employing CD spectroscopic analysis. A typical α-helical structure of IQN17 was revealed, as indicated by two negative peaks at 208 nm and 222 nm. However, when IQN17 was incubated with ADS-J1, the negative peak at 222 nm was significantly reduced, reflecting a reduced α helicity of IQN17 after ADS-J1 binding (Fig. (Fig.7C).7C). All these findings strongly suggest that the binding of ADS-J1 induces a conformational change in IQN17.
Our previous studies have shown that although C34 with a mutation in residue 632 (C34D632K) can form 6-HB with N36 and its mutant version on residue 574 (N36K574D), the helical content and stability of the complex are much lower than those formed between the complex of N36 and C34 (16). In the present study, we tested the inhibitory activity of ADS-J1 on these 6-HBs. Interestingly, ADS-J1 could block the 6-HB formed by N36 and C34D632K (Fig. (Fig.8A),8A), but it was unable to inhibit the 6-HB formed by N36K574D and C34D632K (Fig. (Fig.8B).8B). Taken together, these results suggest that ADS-J1 binds to the NHR of gp41 and that the K574 in the pocket region is the key residue for the inhibitory activity of ADS-J1 on 6-HB formation.
We previously identified a small-molecule HIV-1 fusion inhibitor, ADS-J1, from a computer-aided gp41 pocket-based virtual screening and an ELISA screening using a gp41 6-HB core-specific MAb, MAb NC-1 (8, 23). To determine the binding site of ADS-J1, we first conducted experiments to induce drug resistance using ADS-J1, since the development of viral variants with drug resistance is a classic approach to elucidation of the drug target site (39). Unfortunately, however, we failed to identify an ADS-J1-resistant virus after the passage of HIV-1IIIB in MT-2 cells in the presence of ADS-J1 at two- to fourfold the IC50 for more than 6 months (data not shown). Later, we provided ADS-J1 to José A. Esté at the Universitat Autónoma de Barcelona, who was interested in developing isolates resistant to ADS-J1. Considering that ADS-J1 is a polyanionic compound that may target gp41, Esté and colleagues first compared the activity of ADS-J1 against an HIV-1 mutant that is resistant to AR177, a polyanionic oligonucleotide with a primary target in gp120 (13), and mutants resistant to T-20 and C34, both of which are CHR peptides with defined target sites in gp41 (39). Surprisingly, ADS-J1 was active against all these variants, suggesting that ADS-J1 may target sites different from those targeted by polyanions and CHR peptides (1). They then used AR177-resistant strain NL4-3 as a parent virus for the development of ADS-J1 resistance. After passage of this virus in MT-4 cells for 230 to 270 days in the presence of ADS-J1 at 0.5- to 5-fold the IC50, they obtained four strains that were 4.5- to 9-fold less sensitive to ADS-J1. Analysis of these variants revealed mutations located in both gp120 and gp41. Some of the mutations located in the V3 loop of gp120 resulted in the reduction of a net positive charge in this highly variable region. They therefore concluded that the anti-HIV activity of ADS-J1 targets the HIV-1 gp120 by assuming that there is an association between ADS-J1 resistance and mutations in the V3 loop (1). However, that association has not been verified by constructing NL4-3 variants bearing only those mutations in the V3 loop and using these mutant viruses to test the antiviral activity of ADS-J1. The mutations generated after serial passages of the virus in the culture for more than 8 months may not be specifically associated with the resistance to ADS-J1. Indeed, these four strains also exhibited similar (two- to eightfold) resistance to control peptide C34 (1), whose target site is located in the gp41 NHR region (30, 39). Dwyer et al. (11) have conducted a passaging experiment using T-2544, a pocket-binding sequence-containing peptide with an increased helical structure that can form a highly stable 6-HB with an NHR peptide and using T-20 as a control. They could induce a mutant virus with high-level resistance (81-fold) to T-20 after passage of the virus in the presence of T-20 for 38 days. However, they failed to get a variant with decreased sensitivity to T-2544, even after more than 70 days in culture. After extending the passaging experiment for 227 days, they obtained one strain with weak resistance (8.3-fold) to T-2544, while the related mutation sites could not be defined (11). This finding suggests that viruses may have more difficulty developing resistance to the CHR peptides with the pocket-binding sequence (e.g., T-2544) than to peptides without this sequence (e.g., T-20).
The fact that ADS-J1 is effective against T-20- and C34-resistant strains could not exclude the possibility that ADS-J1 targets gp41. Unlike ADS-J1 and other small-molecule HIV-1 entry inhibitors that interact with the target located in a compact binding motif, CHR peptides T-20 and C34 may interact with multiple binding sites. For example, T-20 has a primary binding site in the region from residues 36 to 45 (GIVQQQNNLL) of the gp41 NHR domain and a secondary binding site in the target cell membrane, while C34 has a primary binding site in the pocket-forming region and a secondary binding site in the region from residues 36 to 45 of the gp41 NHR domain (29, 30, 43, 44, 50, 51). Viruses with mutations in the region from residues 36 to 45 of NHR are highly resistant to T-20 but are only moderately resistant to C34 and other CHR peptides containing the pocket-binding domain (2, 26, 40, 45, 47, 54). The pocket-forming sequence in the NHR domain, which is critical for the interaction between the NHR and the CHR regions and for the stability of the 6-HB core, is highly conservative (4). Mutation of the residues in this region, especially those involved in a direct interaction with those in the CHR, may lead to the disruption of the 6-HB core, resulting in the inability of the virus to enter the target cell.
In the present study, we applied an alternative approach, a combination of biophysical and virological techniques, to revisit the mechanism of action of ADS-J1. Here we provide abundant evidence to show that ADS-J1 interacts with the gp41 NHR domain, particularly the pocket-forming sequence, and blocks fusion-active 6-HB core formation, resulting in the inhibition of HIV-1 fusion with the target cell. First, ADS-J1 inhibits HIV-1 fusion by targeting the viral Env, possibly gp41, as determined by time-of-addition and time-of-removal assays. ADS-J1 was effective in inhibiting HIV-1 replication when it was added to a cell culture prior to or simultaneously with the addition of the virus or virus-infected cells, but became ineffective if it was added at 2 h postinfection. ADS-J1 inhibited HIV-1-mediated cell fusion by interacting with the HIV-1-infected cells rather than with the target cells. Furthermore, ADS-J1 did not block gp120-CD4 binding and exhibited a marginal interaction with the HIV-1 coreceptor CXCR4, suggesting that ADS-J1 may not target the HIV-1 gp120 or the chemokine coreceptor.
Second, ADS-J1 binds to the gp41 NHR domain, particularly the pocket-forming region. Using ELISA and SPR assays to determine whether ADS-J1 binds to an NHR peptide, we found that ADS-J1 significantly bound to IQN17 and blocked the binding of PIE7 to the hydrophobic pocket presented on IQN17. SPR and CD spectroscopic analyses have demonstrated that the binding of ADS-J1 to IQN17 fits with a conformational change model rather than a 1:1 binding model and results in distortion of the α helicity of IQN17, suggest that the binding of ADS-J1 induces a conformational change in IQN17. Notably, a number of ligand-receptor interactions result in conformational changes, which can be detected by SPR and CD spectroscopy. For example, Greenfield et al. (14) first showed that transforming growth factor α binding to the epidermal growth factor receptor results in a conformation change in the receptor, as demonstrated by CD analysis. Later, De et al. (7) reported that the data from SPR analysis of the transforming growth factor-α-epidermal growth factor receptor interaction did not fit a simple binding model but, rather, fitted very well a conformational change model. This may explain why the binding of a small molecule such as ADS-J1 or NB-2 (24) to the gp41 pocket can block the interaction between two protein fragments (NHR and CHR) to form 6-HB. This is possibly because the distorted gp41 NHR structure induced by the binding of the compound may not form an α-helical complex with the CHR domain.
Third, ADS-J1 specifically interacts with the K574 positively charged residue and a cluster of hydrophobic residues in the pocket-forming region of NHR. We have previously showed that the lysine at residue 574 (K574) in the pocket region may interact with the aspartic acid at residue 632 (D632) in the C helix of gp41 to form a salt bridge (Fig. (Fig.1A),1A), which is critical for the stabilization of the 6-HB formed between the viral gp41 NHR and CHR (15, 16, 19). This positively charged K574 residue in the NHR pocket is important for the binding of small-molecule HIV-1 fusion inhibitors with a negatively charged group, such as NB-2 and NB-64 (24). An in-depth computer-aided modeling analysis of the pattern of the interaction of ADS-J1 in the pocket and surrounding region revealed that ADS-J1 was positioned in such a way that its hydrophobic groups (phenyl and naphthalene) were interacting with hydrophobic residues (L568, V570, W571, and L576) in the pocket (8). One of the negatively charged sulfonic acid groups was in close proximity to a positively charged residue (K574) near the pocket, suggesting that these two oppositely charged groups might form a salt bridge. If positively charged amino acid K574 in the NHR peptide was replaced with a negatively charged residue (D574), the negatively charged residue in the CHR region, D632, might not be able to interact with D574 in the NHR region to form a stable fusion-active core (Fig. 9A and B). The virus mutant in which the K574 in gp41 was replaced by an aspartic acid (D574) completely lost its fusion activity (data not shown). The replacement of K574 with a negatively charged residue, such as aspartic acid (D), may result in the loss of its interaction with one of the sulfonic acids in ADS-J1 (Fig. 9C and D). Indeed, ADS-J1 could block 6-HB formation between N36 and C34D632K but was ineffective in inhibiting the formation of 6-HB by N36K574D and C34D632K (Fig. (Fig.88).
Fourth, ADS-J1 is highly effective in blocking gp41 6-HB core formation. The association between the gp41 NHR and CHR domains with the formation of the fusion-active 6-HB core is a critical step during virus-cell fusion (6), and a molecule that can block 6-HB formation may have HIV-1 fusion-inhibitory activity (28, 31). We have previously demonstrated that ADS-J1 can block 6-HB formation between N36 and C34, as shown by a sandwich ELISA and a direct ELISA with 6-HB-specific MAb NC-1, and by fluorescence-linked immunosorbent assay, N-PAGE, fluorescent N-PAGE, and size-exclusion HPLC (23, 27, 28, 32). In this study, we further confirmed that ADS-J1 significantly interfered with the interaction between the NHR and CHR peptides and inhibited the gp41 fusion-active 6-HB core formation.
Taken together, our results suggest that ADS-J1 is an HIV-1 fusion inhibitor targeting gp41 with a binding site located in the NHR pocket region, although we could not exclude the possibility that it might also bind to the V3 loop in gp120 through its sulfonate groups when it is used at a high concentration. The disadvantage of using ADS-J1 as a lead compound for the development of anti-HIV drugs is that it contains several sulfonic acid groups. However, several organic compounds with similar structures (e.g., suramin and FP21399) are being developed as antitumor and anti-HIV drug candidates in clinical trials (10), suggesting that ADS-J1 can still be used as a lead compound for the design of small-molecule HIV entry inhibitors targeting gp41 as a new class of anti-HIV drugs.
We thank Ciba Specialty Chemical Corp. for providing ADS-J1-containing crude materials and Satoshi Omura and Haruo Tanaka of the Katasato Institute, Tokyo, Japan, for providing chloropeptin. We are grateful to Qian Zhao and Louise Bayer-Chatenet for excellent technical assistance.
This work was supported by the Natural Science Foundation of China (grants 30672496, 30729001, and U0832001) to S.L. and U.S. NIH grant (RO1 AI46221) to S.J.
Published ahead of print on 28 September 2009.