点击显示 收起
【关键词】 Small-Molecule
A panel of four CCR5 monoclonal antibodies (mAbs) recognizing different epitopes on CCR5 was examined in CCR5-mediated cell-cell fusion assay, alone or in combination with a variety of small molecule CCR5 antagonists. Although no antagonism was observed between any of the CCR5 inhibitors, surprisingly potent synergy was observed between CCR5 mAbs and antagonists, and the synergistic activity was confirmed in other antiviral assays. Strong synergy was also observed between CCR5 inhibitors and the human immunodeficiency virus (HIV) fusion inhibitor enfuvirtide. There was no synergy observed between small molecule CCR5 inhibitors; however, potent synergy was observed between mAbs recognizing different parts of CCR5. In all synergistic combinations, greater synergy was achieved at higher percent inhibition levels. A negative correlation was found between the degree of synergy between the two classes of CCR5 inhibitors and the ability to compete each other for binding to the receptor. For example, the greatest synergy, observed between the mAb ROAb13 and the small molecule inhibitor maraviroc, did not interfere with binding to CCR5 for either inhibitor, whereas no synergy was found between mAb 45523 and maraviroc, which do compete for binding to CCR5. In addition, in contrast to a recent report, the CCR5 inhibitors tested here were found to inhibit the same stage of HIV entry. Based on the data presented here, we hypothesize that CCR5 inhibitors exert synergistic antiviral actions through a cobinding mechanism.
Highly active antiretroviral therapy using three or more anti-human immunodeficiency virus (HIV) agents from different classes in combination has become the standard treatment regime for patients infected with HIV. This treatment strategy has greatly improved the effectiveness of HIV infection control and the survival of patients with acquired immunodeficiency syndrome patients (Barbaro et al., 2005). However, the emergence of drug resistance and drug-related side effects encourages the development of new classes of anti-HIV drugs. One of the most promising steps in the viral life cycle for intervention is the viral entry process. HIV enters host cells through virus-cell membrane fusion. The first step of viral entry is the high-affinity attachment of the HIV envelope protein gp120 to CD4, followed by specific interaction with a chemokine receptor, CCR5 or CXCR4. The CC-chemokine receptor CCR5 is the major coreceptor for HIV and plays a pivotal role in HIV transmission and pathogenesis (Deng et al., 1996; Dragic et al., 1996). Persons who are CCR5-deficient (32) are essentially protected against infection by HIV-1 in high-risk populations. Persons heterozygous for 32 are not protected against HIV-1 infection but are often long-term nonprogressors (Liu et al., 1996; Eugen-Olsen et al., 1997). Because it is the predominant coreceptor for the majority of the clinical HIV populations yet seems dispensable for human health, CCR5 has become a very attractive target for anti-HIV therapy.
In recent years, many CCR5-targeting small molecules and monoclonal antibodies (mAbs) have been identified that have shown potent antiviral effect both in vitro and in vivo (Palani et al., 2002; Maeda et al., 2004; Watson et al., 2005; Wood and Armour, 2005). We also have identified several novel CCR5 mAbs and small-molecule antagonists with potent antiviral activities (Ji et al., 2006a). All known small molecule CCR5 inhibitors are CCR5 antagonists, which bind in the hydrophobic pocket formed by the seven-transmembrane helices. These CCR5 antagonists compete for binding to the same pocket, although they may interact with different residues in the helices (Dragic et al., 2000; Seibert et al., 2006). Because CCR5 antagonists sit deep in the pocket, it is believed that the small molecule CCR5 antagonists inhibit HIV entry via allosteric mechanism (Watson et al., 2005). Maraviroc is clinically the most advanced CCR5 inhibitor (filed for new drug application in 2007) for HIV therapy, followed by vicriviroc, which is in phase 2b/3 trials.
Because highly active antiretroviral therapy has proven to be much more effective than monotherapy, it is important to assess all new anti-HIV agents in development for potential interactions with other antiretroviral drugs. Although in vitro drug-drug interactions do not necessarily reproduce in vivo, often complicated by pharmacokinetics, it is still necessary to ensure that the new drugs in development do not exert antagonistic interactions with other classes of drugs in vitro. Because CCR5 antibodies and small molecule antagonists both target CCR5, it is particularly important to determine whether they exhibit any antagonistic interactions in antiviral activities. On the other hand, it would be beneficial if the mAbs and antagonists show synergistic interactions.
When performing drug-drug interaction studies, variable conclusions might be inferred depending on the way data are analyzed. Many models and approaches have been described for the assessment of in vitro drug interactions (Prichard and Shipman, 1990; Suhnel, 1990; Chou, 2006). The assumption of no interaction has a central position in these debates, because synergy and antagonism are defined as departures from this. Therefore, when the observed effect is more or less than the effect predicted from the no-interaction theory, synergy or antagonism, respectively, is claimed. Among the various "no-interaction" theories, the Loewe additivity (LA) and Bliss independence (BI) theories are the mostly referenced models (Greco et al., 1995). The LA theory (Loewe, 1953) is based on the assumption that a drug cannot interact with itself, whereas the BI theory (SAS Institute, 1999) is based on the assumption that two drugs act independently with the probabilistic sense of independence. Based on these concepts, various models have been described based on both parametric and nonparametric approaches of these two reference theories. In this study, both the LA-based Greco's model (Greco et al., 1990) and BI-based Prichard's model (Prichard and Shipman, 1990) were used for the analysis of full-range drug-drug interactions between HIV entry inhibitors.
Despite the large numbers of drug combination studies in antiviral assays, there is still heterogeneity regarding the choice of proper assays and data analysis methods. CCR5-mediated cell-cell fusion (CCF) assay is a validated surrogate antiviral assay that yields highly consistent data mainly attributable to the stable effector and target cell lines used in the assay (Ji et al., 2006b). Unlike HIV replication or single-cycle antiviral assays in which the luciferase production takes 2 or more days to peak, the CCF assay needs only overnight incubation (coculture) for cell fusion and sufficient luciferase production to be completed. In the current study, several representative CCR5 mAbs and antagonists were tested in the CCF assay, alone and in combination. Strong synergy was observed between the two novel CCR5 mAbs and CCR5 antagonists. Potent synergistic antiviral effects were also observed between the two novel CCR5 mAbs. In addition, both CCR5 mAbs and antagonists have shown synergistic interactions with the fusion inhibitor enfuvirtide (ENF).
Reagents. All cell culture media and supplements and fetal bovine sera were purchased from Invitrogen (Carlsbad, CA). Human CCR5 mAb 2D7 and phycoerythrin-conjugated goat anti-mouse antibodies were purchased from BD PharMingen (San Diego, CA). CCR5 mAb 45523 was obtained from R&D Systems (Minneapolis, MN). CCR5 antagonists SCH-C (Palani et al., 2002), vicriviroc (VVC; SCH-D) (Strizki et al., 2005), maraviroc (MVC, UK427,857) (Wood and Armour, 2005), 3H-labeled MVC, aplaviroc (APL; GW873140, AK602) (Watson et al., 2005), and Roche CCR5 mAbs ROAb13 and ROAb14 (Ji et al., 2006a) and antagonists ROAT-01, ROAT-02, and ROAT-03 were produced in-house. Fusion inhibitor enfuvirtide (ENF, Fuzeon, T20) was obtained from the batches synthesized at Roche for clinical use.
Fluorescence-Activated Cell Sorting Analysis. CHO-CCR5 cells were harvested and washed twice in phosphate-buffered saline (PBS) containing 0.5% fetal bovine serum (FACS buffer), then resuspended in FACS buffer at 4 x 106 cells/ml. For each reaction, 25 µl of cells (1 x 105) was transferred to a 5-ml tube, 1 µg/ml various primary antibodies and isotype controls were added, and the cells were incubated on ice for 30 to 45 min. Cells were washed three times in FACS buffer and incubated with PE-labeled anti-mouse secondary antibodies for 30 min on ice. At the end of incubation, cells were washed three times and resuspended in 300 µl of FACS buffer and the stained cells were analyzed with a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA)
CCR5-Mediated CCF Assay. CCF assay was performed as described previously (Ji et al., 2006b).
Single-Cycle Antiviral Assay. Pseudotyped NL-Bal viruses were produced by cotransfecting human embryonic kidney 293T cells with VN1 (HIV pNL4-3 genomic construct with its env gene substituted by a luciferase reporter gene) and pcDNA3.1/NL-BAL env [pcDNA3.1 plasmid containing NL-Bal env gene (obtained from Roche Welwyn)]. The supernatant containing pseudotyped viruses was stored at -80°C in aliquots. Reporter cell MAGI-R5 cells were generated by transfecting U373-MAGI-CXCR4CEM cells (National Institutes of Health AIDS Research and Reference Reagent Program, Germantown, MD) with pcDNA3.1Zeo(-) (Invitrogen) construct expressing human CCR5 (Ji et al., 2006a). FuGene 6 (Roche Applied Science) was used for the transfection according to manufacturer's instructions. Stable expression population of CCR5 was enriched by three rounds of FACS by using phycoerythrin-labeled 2D7. For the single-cycle HIV entry assay, test antibodies or compounds were serially diluted in 96-well plates. The equivalent of 1.5 x 105 RLU of viruses and 2.5 x 104 MAGI-R5 cells were added to each well. After 3-day incubation at 37°C, 50 µl of Steady-Glo luciferase assay system (Promega, Rockford, IL) was added, and the assay plates were read on a luminometer (Luminoskan; Thermo Electron Corporation, Waltham, MA). Percentage inhibition curves were generated using the sigmoidal dose-response model with variable slope in Prism software (GraphPad Software, San Diego, CA). For the time course study, MAGI-R5 cells (6 x 104/well) were seeded in 24-well plates overnight. HIV-1 pseudotype viruses were chilled at 4°C for 20 min and added into prechilled MAGI-R5 cells. Spinoculation (spin injection) was performed by spinning at 2000 rpm at 4°C for 1 h. The cells were washed once with ice-cold PBS and then followed by 450 µl of medium at 37°C. At different time points, CCR5 inhibitors at IC90 to IC95 concentrations were added to the cells, in 50 µl of medium containing 0.5% fetal bovine serum. Luciferase activity was measured 48 h after infection, and percentage virus entry for each time point was calculated as (RLU with inhibitor)/(RLU without inhibitor) x 100.
Radioligand Binding Assay. Adherent CHO-CCR5 cells at 90% confluence were detached in freshly made PBS containing 1 mM EDTA. Cells were washed twice in PBS without Ca2+ and Mg2+, and resuspended in ice-cold binding buffer (phenol red-free Ham's F12 medium, pH 7.24, supplemented with freshly made 0.1% bovine serum albumin and 0.1% NaN3). Cells were then plated in 96-well culture plates at 1.5 x 105 cells/well. Serially diluted CCR5 mAbs were added to the cells, followed by addition of 26 nM [3H]MVC. After 2 h of incubation at room temperature with gentle shaking, cells were harvested onto UniFilter GF/C plates (PerkinElmer Life and Analytical Sciences, Boston, MA) using a cell harvester. UniFilter plates were pretreated with PBS containing 0.3% polyethylenimine and 0.2% bovine serum albumin for 30 min before harvest. Filter plates were washed five times with 25 mM HEPES buffer, pH 7.1, containing 500 mM NaCl, 1 mM CaCl2, and 5 mM MgCl2. Plates were dried in 70°C oven for 20 min, 40 µl of scintillation fluid was added, and radioactivity was measured using TopCount NXT (PerkinElmer Life and Analytical Sciences). In all experiments, each data point was assayed in duplicate. Curve-fitting and subsequent data analysis were carried out using Prism software.
Drug Interaction Analyses. In the cell-cell fusion assay, the possibility of either enhanced or reduced efficacy of one entry inhibitor in combination with another entry inhibitor was analyzed using two different models. These two models followed two different additive drug interaction theories: the LA theory and the BI theory.
LA-based model. For the drug interaction models based on the LA theory, the concentrations of the drugs in combination are compared with the concentrations of the drugs alone that produce the same effect. The relationship is described by eq. 1:
(1)
where dA and dB are the concentrations of drugs A and B in combination that elicit a certain effect (e.g., 50% inhibition). DA and DB are the isoeffective concentrations (e.g., IC50) for each drug alone. The concentration response surface approach described by Greco et al. (1990) was used to analyze the data. The fundamental concept from Loewe (1953) underpins this approach. A seven-parameter nonlinear model was fit to all experimental data including percentage inhibitions calculated from replicates for all concentrations of the two drugs alone and in combination from two 384-well plates. The calculation was based on eq. 2:
(2)
where Emax is the maximal response, over background, at 0 drug concentration; IC50A and IC50B are the median inhibitory concentrations of drugs A and B, respectively, that produce 50% of the Emax; mA and mB are the slopes of concentration response curves for the drugs A and B, respectively; DA and DB are the drug concentrations for drugs A and B, respectively, as inputs in the above equation; E is the measured response at the drug concentrations DA and DB, as the output; and is the drug interaction parameter. The above equation was fit to the complete data set from experiment with unweighted lease squares nonlinear regression using SAS (SAS Institute, 1999). The estimates of all seven parameters and their associated asymptotic standard errors and 95% confidence intervals were generated to interpret the results. In addition, the R2, correlation and covariance matrices, and residual plots were checked for goodness of fit for the model.
To interpret the drug interaction from the model fitting, synergy is indicated when the parameter was positive and its 95% confidence interval did not include 0. Antagonism is indicated when was negative and its 95% confidence interval did not include 0. Loewe additivity or no interaction is indicated when the 95% confidence interval of includes 0. Furthermore, the predicted additivity of the drugs combined was calculated by using all estimated parameters of the Greco model, except that is fixed at 0. The deviance between the predicted response surface and the predicted additive surface is interpreted as percentage synergy (positive percentages, if the response surface is above the additive surface), or percentage antagonism (negative percentages, if the response surface is under the additive surface). A three-dimensional graph and a contour plot were generated to examine the magnitude of synergism as well as to determine the range of drug concentrations that produce synergism.
BI-based model. For the drug interaction models based on the BI theory (Bliss, 1939), the estimates of effect of the drugs combined based on the effect of the drugs alone are compared with the observed data from experiment. Its relationship is described by eq. 3:
(3)
where Icomb is the predicted percentage inhibition of the drugs A and B in combination that have no interaction. IA and IB are the observed percentage inhibition of each drug alone. A three-dimensional approach developed by Prichard and Shipman (1990) was used to access the drug interactions. Theoretical additive interactions were calculated from the dose-response curves of the individual drugs based on the Bliss independence equation. For each combination of the two drugs in each plate, the observed percentage inhibitions were subtracted from the theoretical additive percentage inhibition to reveal greater than expected activities. The resulting surface would appear as a horizontal plane at 0% inhibition above the predicted additive surface if the interactions were merely additive. Any peaks above this plane would be indicative of synergy. Likewise, any depression in the plane would indicate antagonism. The 95% confidence intervals around the experimental dose response surface were used to evaluate the data statistically. The total sum of differences between the observed percentage inhibitions and the upper bound of 95% confidence interval of predicted additive percentages is calculated as a statistically significant synergy volume Syn. The total sum of differences between the observed percentage inhibitions and the lower bound of 95% confidence interval of predicted additive percentages is calculated as a statistically significant antagonism volume Ant. In general, the drug interaction is considered weak when the interaction volume is less than 100%. The interaction is considered moderate when the interaction volume is between 100% and 200%. And, the interaction is considered strong when the interaction volume is more than 200%.
For the analysis of drug-drug interaction in the single-cycle assays, percentage inhibition data from three independent experiments were averaged and analyzed for mode of interactions by using the combination index (CI) method as described by Chou and Talalay (1984). CI analysis is a commonly used tool for characterizing drug-drug-interactions. It provides qualitative information on the nature of drug interaction and the extent of drug interaction. CI was calculated according to eq. 4:
(4)
where CA,x and CB,x are the concentrations of drug A and drug B used in combination to achieve x% drug effect. ICx,A and ICx,B, are the concentrations for single agents to achieve the same effect. CI < 1 indicates synergy; CI = 1 indicates additive effects; and CI > 1 indicates antagonism.
Novel CCR5 mAbs and Small Molecule Antagonists with Potent Antiviral Activities. Two of the mouse antihuman CCR5 mAbs ROAb13 and ROAb14 that have been described previously (Ji et al., 2006a) were tested in the CCR5-mediated CCF assay, along with two other CCR5 mAbs, 2D7 and 45523. Three representative CCR5 antagonists discovered at Roche, and three known antagonists were also tested in the CCF assay for IC50 determinations. As shown in Table 1, both novel CCR5 mAbs, ROAb13 and ROAb14, showed strong inhibitory effects in the CCF assay, with IC50 values of 14 and 1.3 nM, respectively. MAb 2D7 also showed potent antiviral activity (IC50 = 4.3 nM), and mAb 45523 exhibited weaker inhibitory effects on cell-cell fusion (IC50 = 23 nM). In contrast, all CCR5 antagonists exhibited low nanomolar or subnanomolar IC50 values (0.4-5 nM) in the CCF assay system.
TABLE 1 Potency of CCR5 inhibitors in CCF assay
Values are given as mean ± S.D. Data were from three independent experiments.
Fig. 1. Synergistic interactions between ROAb14 and MVC in inhibiting cell-cell fusion. A, CCR5 antagonist MVC and mAb ROAb14 were serially diluted starting from 200 and 65 nM, respectively. Various doses of both inhibitors were added to the CCR5-mediated CCF assay either alone or in combination, and percentage inhibition of CCF was calculated for various doses of inhibitors. Cells with a value of less than 50% inhibition are shaded for easier identification of synergy. B, three-dimensional dose-response surface graph. The doses of both inhibitors were plotted against percentage synergy calculated from percentage inhibition data shown in A by using Greco model. Percentage synergy at each 10% increment is shown in different colors. C, all percentage synergy levels identified in the three-dimensional graph are plotted in the two-dimensional contour graph, and their corresponding dose ranges for CCR5 mAb and antagonists are listed in the table (D). Data are from a representative experiment.
Strong Synergy between CCR5 mAb ROAb14 and Antagonist MVC. CCR5 mAbs and CCR5 antagonists were tested in the CCR5-mediated CCF assay in various combinations for the evaluation of potential interactions. CCF assay was chosen as the primary drug-drug interaction tool because it is easy and quick and has been well validated as an effective surrogate antiviral assay for the evaluation of HIV entry inhibitors, especially CCR5 inhibitors (Ji et al., 2006b). In addition, similar drug-drug interaction results were reported in antiviral and CCF assays (Murga et al., 2006). As shown in Fig. 1A, seven-point, half-log dilutions of mAb ROAb14 and 10-point, half-log dilutions of antagonist MVC were tested in the CCF assay, alone or in various dose combinations. The inhibitory effects at each dose point were calculated and indicated as percentage inhibition. A statistically significant synergy exists where the percentage inhibition at a fixed concentration of the drugs is greater than the upper bound of 95% confidence interval of the predicted additive percentage inhibitions of the drugs at the same concentrations dosed alone. This difference, percentage synergy, is defined as a synergy indicator based on Bliss independence theory. Strong synergy is evident between ROAb14 and MVC on cell-cell fusion. For example, when MVC and ROAb14 were added alone, both at 0.27 nM, 13% and 12% of inhibition was obtained, respectively. However, when these two drugs were added together at the same concentration, 42% inhibition was observed. This is 19% greater than the predicted additive 23% inhibition (based on the Bliss independence equation). Furthermore, 16% synergy with 95% confidence was calculated under this dosing combination. Likewise, the percentage synergy with 95% confidence was calculated for all checkerboard dosing points, and a three-dimensional graph was generated that suggested a significant synergy at wide dose ranges for both drugs ROAb14 and MVC (Fig. 1B). A contour plot was also generated with each section of 10% increment shown in different colors (Fig. 1C). The various dose ranges for ROAb14 and MVC to achieve certain percentage synergy were identified from this contour plot and summarized in Fig. 1D. To achieve 20% or greater synergy, the dose ranges for ROAb14 and MVC are 0.09 to 9 nM and 0.06 to 3.1 nM, respectively. The IC50 values for ROAb14 and MVC alone were measured as 1.3 and 0.6 nM, respectively, well within their synergistic dose ranges. The interaction parameter of the fully parametric Greco's model was positive (24.8 ± 2.8), and the 95% confidence interval did not overlap 0, indicating a statistically significant synergy (Table 2). When the interaction was determined based on Bliss independence theory using the Prichard model, a strong synergy was also suggested (Table 2), with a 385% synergy volume (95% Syn). No antagonistic effects were observed.
TABLE 2 Interaction between CCR5 mAbs and antagonists
It can be inferred from Fig. 1A that lower doses of the antibody and antagonist in combination were required than they were when added alone to achieve the same percentage inhibition. For instance, to reach 95% inhibition, 65 and 22.2 nM ROAb14 and MVC, respectively, were required; however, if the drugs were added together, only 0.8 nM ROAb14 and 2.47 nM MVC were required to achieve 96% inhibition. A reduction of 81-fold in ROAb14 dose or 9.8-fold in MVC dose was observed in this case. To obtain information on the dose effects of ROAb14 and MVC in combination at given inhibition levels, isobolograms were generated at 50%, 75%, 90%, and 95% inhibition levels (Fig. 2). A diagonal straight line is expected if only additive effect is observed; an inward curve toward the low doses indicates synergism, and an outward curve indicates antagonism. Significant synergism was observed for ROAb14 and MVC at various inhibition levels. Reduction of 3.3-, 5-, 10-, and 15-fold in the doses normalized by the respective level of potency for the two drugs was obtained at 50, 75, 95, and 95% inhibition levels, respectively.
Fig. 2. Isobolograms of ROAb14-MVC combination. Isobolograms were created as described under Materials and Methods at 50% (A), 75% (B), 90% (C), and 95% (D) inhibition levels. The x- and y-axes are fractions of 1 unit of percentage inhibition concentrations (IC) for ROAb14 and MVC, respectively. Diagonal straight line indicates theoretical additive interactions, and the curve under it or above it indicates synergistic or antagonistic interactions between the two drugs (shown in A). Fold synergy is indicated in parenthesis for each percentage inhibition level.
Strong Synergy between CCR5 mAb ROAb13 and Antagonist MVC. To examine whether the synergy between CCR5 mAb ROAb14 and CCR5 antagonists applies to all CCR5 antibodies, a few other CCR5 mAbs with different recognition sites were then tested in combination with CCR5 antagonists. ROAb13, which binds to the N-terminal end of CCR5 (Zhang et al., 2007), exhibited much greater synergy than ROAb14 when combined with the same CCR5 antagonist MVC (Fig. 3). The 3D graphs generated by using the Greco (Fig. 3A) and Prichard (Fig. 3B) models showed very similar synergy surface, demonstrating that 60% or greater synergy could be obtained by dosing ROAb13 with MVC. The parameter for the ROAb13-MVC combination was calculated using the Greco's model as 662 ± 99 (Table 2), which is much greater than that for the ROAb14-MVC combination (24.8 ± 2.8).
Fig. 3. Three-dimensional drug response surface for the ROAb13-MVC combinations. Percentage synergy obtained from each combination dose was plotted against ROAb13 and MVC doses, based on Greco's (A) and Prichard's (B) mathematics models.
Interaction between Other CCR5 mAbs and Antagonists. Murine CCR5 mAb 2D7, which is known to bind to the N-terminal half of extracellular loop 2 (ECL2) of CCR5, exhibited weak to moderate synergy in combination with CCR5 antagonist MVC and APL. The parameters for 2D7-MVC and 2D7-APL combinations were determined as 13.2 and 2.1, respectively, using Greco's model. These values were much smaller than those for the ROAb13-MVC or ROAb14-MVC combinations (Table 2). Another commercially available anti-CCR5 mAb 45523 that was previously shown to bind multiple exodomains of CCR5 was also investigated for its interactions with CCR5 antagonists. As shown in Table 2, the parameter and Syn for 45523-MVC combination were -0.03 and 3, respectively, suggesting no synergism between 45523 and MVC. It has been demonstrated that the CCR5 antagonist APL completely blocked the binding of mAb 45523 (Maeda et al., 2004). In the current study, interactions between mAb 45523 and various CCR5 antagonists, including APL, were performed. The result revealed that antagonists APL, MVC, and VVC strongly inhibited 45523 binding by 75 to 85% (Fig. 6A). These data suggest that the lack of synergy between mAb 45523 and CCR5 antagonists is probably due to the inhibition of 45523 binding by CCR5 antagonists.
Fig. 6. Effects of CCR5 mAb or antagonists on CCR5 mAb binding. A, CHO-CCR5 cells were preincubated with 50 nM APL, MVC, VVC, or vehicle at room temperature for 1 h, then incubated with fluor-labeled 45523 on ice for 30 min, followed by FACS analysis. Mean fluorescence intensity (MFI) for each treatment was obtained and converted to percent binding by setting vehicle control MFI as 100%. B, CHO-CCR5 cells were preincubated with 10 µg/ml 2D7 or ROAb13 on ice for 1 h, then incubated with fluor-labeled ROAb14 for 30 min, followed by FACS analysis. MFI for each treatment was obtained and converted to percentage binding by setting vehicle control MFI as 100%. Bars with standard errors for A and B are percentage mAb binding data from two or more experiments. C, CHO-CCR5 cells were preincubated with 5 µM ROAT-02 or vehicle at room temperature for 1 h, then with 1 µg/ml fluor-labeled ROAb14 or ROAb13 on ice for 30 min in the presence of ROAT-02, followed by FACS analysis. Bars with S.E. are MFI from two independent experiments.
Several other CCR5 antagonists, including VVC, APL, and novel antagonists ROAT-01, ROAT-02, and ROAT-03, were also tested for their interactions with various antibodies in the CCF assay system. These antagonists are structurally distinct but all exhibited potent antiviral activities. Both the Greco model and the Prichard model were used to analyze the drug interactions for these different combinations, and the results were summarized in Table 2. Among all the CCR5 antagonists, APL exhibited the greatest synergy when in combination with ROAb14 or ROAb13; however, it showed weakest synergy with 2D7.
Fig. 4. In vitro drug interactions between CCR5 mAbs or antagonists. Three-dimensional drug-response surface graphs for ROAb14-ROAb13 (A), ROAb14 -2D7 (B), ROAT-01-ROAT-02 (C), and ROAT-02-APL (D) combinations were created based on Greco's model.
Interaction between CCR5 mAbs. As mentioned above, CCR5 mAbs with different binding epitopes showed different modes of interaction with CCR5 antagonists. Therefore, it is possible that antibodies recognizing different parts of CCR5 may interact with each other differently. In the current study, several CCR5 mAbs were tested for their combinational effects on cell-cell fusion. As shown in Fig. 4A, strong synergy was observed when ROAb14 (binds to ECL2) (Zhang et al., 2007) was combined with the N-terminal binding mAb ROAb13. However, no synergy was observed when ROAb14 was combined with 2D7 (Fig. 4B). Because CCR5 ECL2 is important for both ROAb14 and 2D7 binding, these two antibodies are likely to compete for binding to CCR5. This was confirmed by FACS analysis (Fig. 6B). Preincubation of CHO-CCR5 cells with 2D7 completely blocked the binding of ROAb14 to cell surface CCR5 receptor, whereas preincubation with ROAb13 had no effect on ROAb14 binding.
No Synergy between Small Molecule CCR5 Antagonists. Published literature suggests that all known CCR5 antagonists bind into the same cleft formed by the transmembrane domains of CCR5; as a result, these antagonists bind to CCR5 in a competitive manner (Watson et al., 2005). Therefore, no synergy was expected for the antagonist-antagonist combinations. To prove this hypothesis, three CCR5 antagonists from different structure classes were tested in two-drug combinations in the CCF assay system. As shown in Fig. 4C, no significant synergism or antagonism was observed between antagonists ROAT-01 and ROAT-02 ( = -0.023 ± 0.01). The same result was also observed between antagonist ROAT-02 and APL ( = 0.11 ± 0.06; three-dimensional graph not shown). This observation is consistent with previously published results (Murga et al., 2006).
Fig. 5. CCR5 mAb and antagonist MVC inhibit the same entry stage. A, serially diluted ROAb14 and MVC were added to the single-cycle antiviral assays, alone or in combination at 1:1 ratio. Percentage inhibition curves were generated using data from three independent experiments. B, NL-Bal pseudoviruses were spun onto MAGI-R5 cells at 4°C. After washing, cells were warmed to 37°C, and inhibitors were added at different time points. Percentage virus entry time course curves were plotted, and t values were calculated using one-phase exponential association curve fitting. Data are means of three independent experiments.
Strong Synergy between Fusion Inhibitor ENF and CCR5 Inhibitors. It has been hypothesized that inhibitors targeting at different steps of HIV entry/fusion process may exert synergistic interactions (Tremblay, 2004). And this has been proven to be true for some HIV entry inhibitors, including fusion inhibitor ENF, CCR5 antagonists, CD4 inhibitors, and gp120 inhibitors (Nagashima et al., 2001; Tremblay, 2004; Tremblay et al., 2005). In the current study, we tested interactions between HIV fusion inhibitor ENF and CCR5 mAbs or CCR5 antagonists in the CCR5-mediated CCF assay system. As shown in Table 3, potent synergistic interactions were found between ENF and mAbs or antagonists (Table 3).
TABLE 3 Interaction between ENF and CCR5 inhibitors
Data were from two or more independent experiments.
CCR5 mAbs and Antagonists Inhibit the Same Stage of Viral Entry. Multiple stages may exist between the binding of gp120 to CCR5 and the initiation of gp41 conformational changes. Published data suggest that CCR5 mAb PA14 and small-molecule antagonists may inhibit different stages of HIV entry, and this was used to explain the synergy between the two CCR5 inhibitors (Safarian et al., 2006). We also sought to determine whether this was the case for the mAbs and antagonists used in the current study. For this purpose, the single-cycle antiviral assay was used. First, we verified the synergy between CCR5 mAb ROAb14 and antagonist MVC in the single-cycle assay (Fig. 5A). Because of low throughput of the assay, only the fixed 1:1 ratio combination was performed, and CI was used to evaluate synergy. At the 90% inhibition level, the CI index for ROAb14 and MVC was 0.146; at the 50% inhibition level, the CI index was 0.116. This result indicates strong synergy between the two CCR5 inhibitors (CI < 1 indicates synergy). To determine whether the CCR5 mAbs and antagonists inhibit different stages of HIV entry, time course experiments were performed. NL-BalHIV viruses were spinoculated onto MAGI-R5 cells at 4°C, and free viruses were washed away. Synchronized HIV infection was triggered by adding medium or inhibitors at 37°C. Inhibitors were then added at different time points afterward. As shown in Fig. 5B, similar time course curves for mAbs ROAb14 and 2D7 and small-molecule antagonist MVC were observed, suggesting that they inhibit the same stage of HIV entry. The time required for half-maximal inhibition (t) of viral entry for ROAb14, 2D7, and MVC are 6, 8, and 10 min, respectively. The control fusion inhibitor ENF, which inhibits HIV entry at a later stage, showed a much lagged inhibition time course curve (t = 48 min). Published data suggested that mAb PA14 (the murine form of PRO 140) inhibits later stages than CCR5 antagonists, but our data indicate that mAb 2D7 and ROAb14 inhibit the same stage as antagonists. This discrepancy could be due to the fact that 2D7 and ROAb14 recognize different epitopes from PA14.
CCR5 mAbs and CCR5 Antagonists Can Cobind to the Receptor. To understand the mechanism of the synergy between CCR5 mAbs and CCR5 antagonists, several experiments were performed to investigate the interactions between CCR5 mAbs and CCR5 antagonists at the receptor binding level. When CHO-CCR5 cells were pre-exposed to excess CCR5 antagonist ROAT-02 for 2 h and then to saturating doses of various fluor-labeled CCR5 mAbs ROAb13 or ROAb14, the total binding of these mAbs was found to be the identical to the no-antagonist control samples (Fig. 6C). Considering the long off-rate of ROAT-02 (t1/2off 9 h, our unpublished data), the short mAb incubation time (30 min), and the presence of excess antagonists during mAb incubation, this result suggests that the mAbs can bind to the antagonist-bound CCR5 as efficiently as to the antagonist-free CCR5. These data thus suggest that CCR5 mAbs and CCR5 antagonists can cobind to CCR5.
Fig. 7. Effects of CCR5 antagonists on CCR5 mAb binding. A-D, time course. CHO-CCR5 cells were preincubated with 50 nM APL, MVC, VVC, or vehicle at room temperature for 1 h, then incubated with fluor-labeled CCR5 mAb ROAb14 (A), ROAb13 (B), 2D7 (C), or 45523 (D) on ice for various time points, followed by cell fixation in 2% paraformaldehyde and FACS analysis; or incubated with various doses of fluorescence-labeled CCR5 mAb 45523 (E) or 2D7 (F) on ice for 30 min, followed by FACS analysis. The time-course and dose-response curves for each mAb in the presence of various antagonists were created based on their MFI values.
Effects of CCR5 Antagonists on CCR5 mAb Affinity and On-Rate. The dose responses and on-rates of the CCR5 mAbs were then evaluated in the presence and absence of various antagonists. Based on the time course study data shown in Fig. 7, the on-rate and total binding of ROAb13 and ROAb14 were not affected at all by preincubation with and continued presence of CCR5 antagonist MVC, APL, or VVC. However, the total binding of mAb 2D7 was weakly inhibited by preincubation of CHO-CCR5 cells with antagonist APL, MVC, and VVC. The on-rate of 2D7 was also slightly slowed down by these antagonists (Fig. 7C). The total binding of 45523 was almost completely blocked by the three antagonists mentioned above, with its on-rate significantly reduced (Fig. 7D). The binding of mAb 45523 and 2D7 in the presence of CCR5 antagonists were also tested at multiple concentrations. As shown in Fig. 7E, MVC, APL, and VVC strongly inhibited mAb 45523 binding at all doses. And mAb 2D7 binding was weakly inhibited by the three antagonists at most doses (Fig. 7F). In both cases, APL seems to have exhibited stronger inhibition than VVC and MVC on 2D7 binding. This supports the observation that APL exhibited the weakest synergy with mAb 2D7 among all the antagonists tested (Table 2). In addition, the EC50 of 2D7 was not significantly affected by CCR5 antagonists, suggesting a noncompetition allosteric inhibition.
Effects of CCR5 mAbs on CCR5 Antagonist Binding. To determine whether the binding kinetics of CCR5 antagonists were influenced by the synergistic CCR5 antibodies, the binding of 3H-labeled MVC to CHO cell surface CCR5 was measured at various time points by preincubating cells with saturating amount (30 µg/ml) of CCR5 mAb 2D7, ROAb13, or ROAb14. Averaged values from three independent experiments were shown in Fig. 8. The binding of MVC was not affected by preincubation of CHO-CCR5 cells with ROAb13 at all time points, suggesting that ROAb13 did not change the on-rate or affinity of MVC to CCR5. However, approximately 38% and 67% inhibition of MVC binding was observed when CHO-CCR5 cells were preincubated with ROAb14 or 2D7, respectively.
Fig. 8. Effects of CCR5 mAbs on MVC binding. CHO-CCR5 cells (2 x 105/100 µl) were preincubated with 30 µg/ml various CCR5 mAbs or PBS at room temperature for 1 h, then incubated with 26 nM [3H]MVC. At the end of various time points, cells were washed and the membrane-bound [3H]MVC was measured as described under Materials and Methods. The maximal counts from the control samples were set as 100% binding, and the relative binding for all other samples were calculated and the time course curves were generated based on these relative binding at each time point.
In the current study, strong synergy was observed between CCR5 mAb ROAb14 or ROAb13 and a CCR5 antagonist. It is noteworthy that the synergy between ROAb13 and antagonists (mean parameter = 1187.8) was much greater than that between ROAb14 and antagonists (mean parameter = 36.6) (Table 2). Our previously published data indicate that ROAb13 binds to the N terminus of CCR5, whereas ROAb14 primarily binds to ECL2 (Zhang et al., 2007). CCR5 mAb PRO 140 has also been found to act synergistically with CCR5 antagonists in inhibiting cell-cell or virus-cell fusion (Murga et al., 2006; Safarian et al., 2006). PRO 140 binds to multiple exodomains of CCR5 with epitopes different from ROAb13, ROAb14, and 2D7 (Trkola et al., 2001). However, CCR5 mAb 2D7 showed only weak synergy with antagonists, and mAb 45523 showed no synergy at all with antagonists. Therefore, it seems that not all CCR5 mAbs act synergistically with small-molecule antagonists. The reason for the lack of synergy between 45523 and CCR5 antagonists could be the strong inhibition of 45523 binding by CCR5 antagonists (Fig. 6A).
Because antibodies and antagonists targeting the same receptor showed strong synergistic interactions in inhibiting viral entry, it would be interesting to find out whether antibodies binding to different epitopes can also exert synergistic antiviral actions. Strong synergy was observed between ROAb14 and ROAb13; however, no synergy was observed between ROAb14 and 2D7. Competition binding data indicate that ROAb13 does not compete with ROAb14 for binding, whereas 2D7 strongly competes with ROAb14 (Fig. 6B). These results suggest that antibodies that do not compete for binding to CCR5 may exhibit synergistic interactions, and antibodies that compete for binding may exert additive or antagonistic interactions. This is in agreement with previous findings of synergistic HIV neutralization by mAbs binding to different epitopes on envelope proteins (Laal et al., 1994; Vijh-Warrier et al., 1996), yet additive HIV neutralization by mAbs binding to similar epitopes on gp41 (Zwick et al., 2005). Likewise, CCR5 antagonists also failed to show any synergy in two-drug combinations, possibly because of their competition for the same binding sites on the receptor. In conclusion, it seems that CCR5 inhibitors that do not interfere with each other's binding may exert synergistic antiviral effects. We demonstrated that the observed strong synergy between ROAb14 or ROAb13 and the antagonists is not through enhanced binding of either molecule to the receptor; rather, it is likely to be through synergistic interactions after binding to the receptor. We also showed that greater synergy was obtained at higher CCF inhibition levels (Fig. 2), and the synergy between CCR5 mAb and antagonist negatively correlates with binding inhibition of the inhibitor by its partner (Table 4). Based on these observations, we hypothesize that the synergy between CCR5 inhibitors such as an antibody and a small-molecule antagonist requires the cobinding of both inhibitors to the receptor. Greater synergy is expected when a higher percentage of CCR5 receptors is occupied by both inhibitors. As shown in Fig. 9, the small-molecule antagonists bind to the pocket, whereas the antibodies bind to the surface of CCR5, allowing the cobinding of a CCR5 mAb and a CCR5 small molecule antagonist. Likewise, because the two CCR5 mAbs ROAb13 and ROAb14 bind to different surface areas of CCR5, they do not interfere with each other's binding to CCR5. Under these circumstances, because two different CCR5 inhibitors can bind to CCR5 simultaneously, synergistic antiviral effects were observed. In contrast, because ROAb14 and 2D7 share epitope Lys171/Glu172, they compete for binding to CCR5, and no synergy was observed between these two mAbs.
TABLE 4 Relationship between synergy and binding
Fig. 9. Diagram of CCR5 structure. CCR5 is a seven-transmembrane protein that contains four extracellular domains: the N-terminal end (black), extracellular loop (ECL) 1, 2, and 3 (green); seven transmembrane helices (pink); and four intracellular domains: the C-terminal end and intracellular loop (ICL) 1, 2, and 3 (purple). The epitopes for CCR5 mAbs 2D7, ROAb13, and ROAb14 (Zhang et al., 2007) are shown. The epitope for 2D7 is Lys172/Glu172 (red) in ECL2, the epitope for ROAb14 is Lys171/Glu172 (red) and Trp190 (brown) in ECL2, and the epitope for ROAb13 is the N-terminal eight amino acid residues (blue). The structure of CCR5 antagonist maraviroc (Wood and Armour, 2005) is also shown. Maraviroc binds to the hydrophobic pocket formed by the transmembrane helices.
It is not yet fully understood what conformational changes in HIV envelope proteins and chemokine receptors are required for the fusion to occur and how CCR5 antibodies and antagonists inhibit HIV entry. It is hypothesized that CCR5 may exist in multiple conformational states; therefore, multiple conformational changes may also occur when CCR5 is bound by antibodies or antagonists (Blanpain et al., 2002). Thus, the ultimate inhibitory potency of a CCR5 inhibitor is probably dependent on the combined conformational effects. Two different inhibitors, such as a mAb and an antagonist, could compensate each other on weak inhibitory conformations to result in synergistic inhibition.
ENF, a 36-amino acid peptide mimetic of gp41 heptad repeat 2 and its downstream region, is the only HIV entry inhibitor on the market (Lazzarin, 2005). ENF is a potent fusion inhibitor that exerts antiviral activities by binding to the HR1 trimeric coiled-coil prehairpin intermediate. Interaction of gp120 with coreceptor triggers a series of conformational changes and energy transfers. Factors impeding this energy transfer cascade can slow down the fusion kinetics and thus increase the exposure time of the prehairpin structures to fusion inhibitors. Platt et al. (2005) demonstrated that low numbers of surface CCR5 or the presence of CCR5 antagonist TAK-779 greatly reduced HIV fusion kinetics. This may help explain the strong synergy between ENF and CCR5 inhibitors described in this article.
Synergistic interactions have been found for a variety of drugs, especially antimicrobial, antiviral, and antitumor drugs. Most of these synergistic drugs target different molecules that belong to the same pathway or different pathway but leading to the same biological outcome. However, synergy between drugs that target the same protein molecule has been demonstrated for only a few drugs. One well studied example is the synergy between HIV-1 non-nucleotide reverse transcriptase inhibitors and nucleotide reverse transcriptase inhibitor 3'-azido-2',3'-dideoxythymidine (AZT) (Cruchaga et al., 2005). Synergy between antibodies that target the same proteins were also reported but very rare. Friedman reported that two ErbB-2 antibodies that bind to different epitopes exhibited synergy in inducing receptor internalization (Friedman et al., 2005). Synergy has also been detected for neutralization mAbs recognizing different gp120 or gp41 exodomain epitopes. However, the results were controversial, and the synergistic effects were moderate (Laal et al., 1994; Vijh-Warrier et al., 1996). Synergistic neutralization has been demonstrated most extensively between mAbs binding to the V3 loop or CD4 binding domain of gp120 (Laal et al., 1994; Vijh-Warrier et al., 1996). The mechanism of the synergy between HIV neutralizing antibodies remains unclear. Synergistic antiviral effects between two mAbs, PA12 and 2D7, that bind to the same molecule, CCR5, were also reported (Olson et al., 1999). Here, we reported highly potent synergy between two CCR5 mAbs, ROAb13 and ROAb14; we also reported potent synergy between CCR5 mAb ROAb13 or ROAb14 and CCR5 antagonists.
In vitro drug-drug interactions have been examined for a variety of HIV entry inhibitors, and synergy was found in all combinations between these inhibitors. These inhibitors target viral and host proteins that are involved in the attachment/fusion process, including gp120 inhibitor PRO542, gp41 inhibitor ENF, CD4 inhibitor cyclotriazadisulfonamide, CCR5 mAbs, and antagonists. Synergistic interactions between CCR5 inhibitors and HIV reverse transcriptase or protease inhibitors have also been reported (Tremblay, 2004; Tremblay et al., 2005). Combination antiviral therapy, including a CCR5 mAb, would be highly favorable because antibodies use different administration, uptake, and metabolism routes. Combinations between two CCR5 mAbs that do not compete for binding may also be beneficial; however, combinations between two CCR5 antagonists may be unfavorable. Markedly boosted antiviral effects and potential dose reduction may be achieved if a CCR5 mAb and a CCR5 antagonist are coadministered. Because the synergistic CCR5 mAb and antagonist can cobind to CCR5, another potential benefit of coadministering CCR5 mAbs and antagonists is the raised hurdle for the emergence of HIV-1 mutants resistant to both inhibitors. In addition, because HIV-1 mutants resistant to one of the CCR5 inhibitors are still susceptible to the other class of CCR5 inhibitors (Trkola et al., 2002), coadministration of both classes of CCR5 inhibitors may greatly facilitate the clearance of existing single drug-resistant HIV variants.
Acknowledgements
We acknowledge our medicinal chemists at Roche Palo Alto, especially David Rotstein and Chris Melville, for providing the CCR5 antagonists and the HIV biology group for providing the reagents and technical help for the preparation of pseudotyped NL-BalHIV virus. We also thank National Institutes of Health AIDS Reagents and Reference Program for providing the reagents that were used in this study.
【参考文献】
Barbaro G, Scozzafava A, Mastrolorenzo A, and Supuran CT (2005) Highly active antiretroviral therapy: current state of the art, new agents and their pharmacological interactions useful for improving therapeutic outcome. Curr Pharm Des 11: 1805-1843.
Blanpain C, Vanderwinden JM, Cihak J, Wittamer V, Le Poul E, Issafras H, Stangassinger M, Vassart G, Marullo S, Schlndorff D, et al. (2002) Multiple active states and oligomerization of CCR5 revealed by functional properties of monoclonal antibodies. Mol Biol Cell 13: 723-737.[Abstract/Free Full Text]
Bliss CI (1939) The toxicity of poisons applied jointly. Ann Appl Biol 26: 585-615.
Chou TC (2006) Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacol Rev 58: 621-681.[Abstract/Free Full Text]
Chou TC, and Talalay P (1984) Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 22: 27-55.
Cruchaga C, Odriozola L, Andreola M, Tarrago-Litvak L, and Martinez-Irujo JJ (2005) Inhibition of phosphorolysis catalyzed by HIV-1 reverse transcriptase is responsible for the synergy found in combinations of 3'-azido-3'-deoxythymidine with nonnucleoside inhibitors. Biochemistry 44: 3535-3546.
Deng H, Liu R, Ellmeier W, Choe S, Unutmaz D, Burkhart M, Di Marzio P, Marmon S, Sutton RE, Hill CM, et al. (1996) Identification of a major co-receptor for primary isolates of HIV-1. Nature 381: 661-666.
Dragic T, Litwin V, Allaway GP, Martin SR, Huang Y, Nagashima KA, Cayanan C, Maddon PJ, Koup RA, Moore JP, et al. (1996) HIV-1 entry into CD4+ cells is mediated by the chemokine receptor CC-CKR-5. Nature 381: 667-673.
Dragic T, Trkola A, Thompson DA, Cormier EG, Kajumo FA, Maxwell E, Lin SW, Ying W, Smith SO, Sakmar TP, et al. (2000) A binding pocket for a small molecule inhibitor of HIV-1 entry within the transmembrane helices of CCR5. Proc Natl Acad Sci U S A 97: 5639-5644.[Abstract/Free Full Text]
Eugen-Olsen J, Iversen AK, Garred P, Koppelhus U, Pedersen C, Benfield TL, Sorensen AM, Katzenstein T, Dickmeiss E, Gerstoft J, et al. (1997) Heterozygosity for a deletion in the CKR-5 gene leads to prolonged AIDS-free survival and slower CD4 T-cell decline in a cohort of HIV-seropositive individuals. Aids 11: 305-310.
Friedman LM, Rinon A, Schechter B, Lyass L, Lavi S, Bacus SS, Sela M, and Yarden Y (2005) Synergistic down-regulation of receptor tyrosine kinases by combinations of mAbs: implications for cancer immunotherapy. Proc Natl Acad Sci U S A 102: 1915-1920.[Abstract/Free Full Text]
Greco WR, Bravo G, and Parsons JC (1995) The search for synergy: a critical review from a response surface perspective. Pharmacol Rev 47: 331-385.
Greco WR, Park HS, and Rustum YM (1990) Application of a new approach for the quantitation of drug synergism to the combination of cis-diamminedichloroplatinum and 1-beta-D-arabinofuranosylcytosine. Cancer Res 50: 5318-5327.[Abstract/Free Full Text]
Ji C, Brandt M, Dioszegi M, Jekle A, Schwoerer S, Challand S, Zhang J, Chen Y, Zautke L, Achhammer G, et al. (2006a) Novel CCR5 monoclonal antibodies with potent and broad-spectrum anti-HIV activities. Antiviral Res 74: 125-137.
Ji C, Zhang J, Cammack N, and Sankuratri S (2006b) Development of a novel dual CCR5-dependent and CXCR4-dependent cell-cell fusion assay system with inducible gp160 expression. J Biomol Screen 11: 65-74.[Abstract/Free Full Text]
Laal S, Burda S, Gorny MK, Karwowska S, Buchbinder A, and Zolla-Pazner S (1994) Synergistic neutralization of human immunodeficiency virus type 1 by combinations of human monoclonal antibodies. J Virol 68: 4001-4008.[Abstract/Free Full Text]
Lazzarin A (2005) Enfuvirtide: the first HIV fusion inhibitor. Expert Opin Pharmacother 6: 453-464.
Liu R, Paxton WA, Choe S, Ceradini D, Martin SR, Horuk R, MacDonald ME, Stuhlmann H, Koup RA, and Landau NR (1996) Homozygous defect in HIV-1 coreceptor accounts for resistance of some multiply-exposed individuals to HIV-1 infection. Cell 86: 367-377.
Loewe S (1953) The problem of synergism and antagonism of combined drugs. Arzneimittelforschung 3: 285-290.
Maeda K, Nakata H, Koh Y, Miyakawa T, Ogata H, Takaoka Y, Shibayama S, Sagawa K, Fukushima D, Moravek J, et al. (2004) Spirodiketopiperazine-based CCR5 inhibitor which preserves CC-chemokine/CCR5 interactions and exerts potent activity against R5 human immunodeficiency virus type 1 in vitro. J Virol 78: 8654-8662.[Abstract/Free Full Text]
Murga JD, Franti M, Pevear DC, Maddon PJ, and Olson WC (2006) Potent antiviral synergy between monoclonal antibody and small-molecule CCR5 inhibitors of human immunodeficiency virus type 1. Antimicrob Agents Chemother 50: 3289-3296.[Abstract/Free Full Text]
Nagashima KA, Thompson DA, Rosenfield SI, Maddon PJ, Dragic T, and Olson WC (2001) Human immunodeficiency virus type 1 entry inhibitors PRO 542 and T-20 are potently synergistic in blocking virus-cell and cell-cell fusion. J Infect Dis 183: 1121-1125.
Olson WC, Rabut GE, Nagashima KA, Tran DN, Anselma DJ, Monard SP, Segal JP, Thompson DA, Kajumo F, Guo Y, et al. (1999) Differential inhibition of human immunodeficiency virus type 1 fusion, gp120 binding, and CC-chemokine activity by monoclonal antibodies to CCR5. J Virol 73: 4145-4155.[Abstract/Free Full Text]
Palani A, Shapiro S, Josien H, Bara T, Clader JW, Greenlee WJ, Cox K, Strizki JM, and Baroudy BM (2002) Synthesis, SAR, and biological evaluation of oximino-piperidino-piperidine amides. 1. Orally bioavailable CCR5 receptor antagonists with potent anti-HIV activity. J Med Chem 45: 3143-3160.
Platt EJ, Durnin JP, and Kabat D (2005) Kinetic factors control efficiencies of cell entry, efficacies of entry inhibitors, and mechanisms of adaptation of human immunodeficiency virus. J Virol 79: 4347-4356.[Abstract/Free Full Text]
Prichard MN and Shipman C Jr. (1990) A three-dimensional model to analyze drug-drug interactions. Antiviral Res 14: 181-205.
Safarian D, Carnec X, Tsamis F, Kajumo F, and Dragic T (2006) An anti-CCR5 monoclonal antibody and small molecule CCR5 antagonists synergize by inhibiting different stages of human immunodeficiency virus type 1 entry. Virology 352: 477-484.
SAS Institute (1999) SAS User's Guide: Statistics, SAS Institute, Cary, NC.
Seibert C, Ying W, Gavrilov S, Tsamis F, Kuhmann SE, Palani A, Tagat JR, Clader JW, McCombie SW, Baroudy BM, et al. (2006) Interaction of small molecule inhibitors of HIV-1 entry with CCR5. Virology 349: 41-54.
Strizki JM, Tremblay C, Xu S, Wojcik L, Wagner N, Gonsiorek W, Hipkin RW, Chou CC, Pugliese-Sivo C, Xiao Y, et al. (2005) Discovery and characterization of vicriviroc (SCH 417690), a CCR5 antagonist with potent activity against human immunodeficiency virus type 1. Antimicrob Agents Chemother 49: 4911-4919.[Abstract/Free Full Text]
Suhnel J (1990) Evaluation of synergism or antagonism for the combined action of antiviral agents. Antiviral Res 13: 23-39.
Tremblay C (2004) Effects of HIV-1 entry inhibitors in combination. Curr Pharm Des 10: 1861-1865.
Tremblay CL, Giguel F, Chou TC, Dong H, Takashima K, and Hirsch MS (2005) TAK-652, a novel CCR5 inhibitor, has favourable drug interactions with other antiretrovirals in vitro. Antivir Ther 10: 967-968.
Trkola A, Ketas TJ, Nagashima KA, Zhao L, Cilliers T, Morris L, Moore JP, Maddon PJ, and Olson WC (2001) Potent, broad-spectrum inhibition of human immunodeficiency virus type 1 by the CCR5 monoclonal antibody PRO 140. J Virol 75: 579-588.[Abstract/Free Full Text]
Trkola A, Kuhmann SE, Strizki JM, Maxwell E, Ketas T, Morgan T, Pugach P, Xu S, Wojcik L, Tagat J, et al. (2002) HIV-1 escape from a small molecule, CCR5-specific entry inhibitor does not involve CXCR4 use. Proc Natl Acad Sci U S A 99: 395-400.[Abstract/Free Full Text]
Vijh-Warrier S, Pinter A, Honnen WJ, and Tilley SA (1996) Synergistic neutralization of human immunodeficiency virus type 1 by a chimpanzee monoclonal antibody against the V2 domain of gp120 in combination with monoclonal antibodies against the V3 loop and the CD4-binding site. J Virol 70: 4466-4473.
Watson C, Jenkinson S, Kazmierski W, and Kenakin T (2005) The CCR5 receptorbased mechanism of action of 873140, a potent allosteric noncompetitive HIV entry inhibitor. Mol Pharmacol 67: 1268-1282.[Abstract/Free Full Text]
Wood A, and Armour D (2005) The discovery of the CCR5 receptor antagonist, UK-427,857, a new agent for the treatment of HIV infection and AIDS. Prog Med Chem 43: 239-271.
Zhang J, Rao E, Dioszegi M, Kondru R, Derosier A, Chan E, Schwoerer S, Cammack N, Brandt M, Sankuratri S and Ji C (2007) The second extracellular loop of CCR5 contains the dominant epitopes for highly potent Anti-HIV monoclonal antibodies. Antimicrob Agents Chemother 51: 1386-1397.[Abstract/Free Full Text]
Zwick MB, Jensen R, Church S, Wang M, Stiegler G, Kunert R, Katinger H, and Burton DR (2005) Anti-human immunodeficiency virus type 1 (HIV-1) antibodies 2F5 and 4E10 require surprisingly few crucial residues in the membrane-proximal external region of glycoprotein gp41 to neutralize HIV-1. J Virol 79: 1252-1261.[Abstract/Free Full Text]
作者单位：Department of Viral Diseases (C.J., J.Z., M.D., E.R., A.D., N.C., S.S.) and Genome Informatics (S.C.), Roche Palo Alto, Palo Alto, California; and Pharmaceuticals Division, Roche Penzberg, Penzberg, Germany (M.B.)