Virtual Screening, Biological Evaluation, and 3D-QSAR Studies of New HIV-1 Entry Inhibitors That Function via the CD4 Primary Receptor

Human immunodeficiency virus type 1 (HIV-1) is responsible for the majority of HIV infections worldwide, and we still lack a cure for this infection. Blocking the interaction of HIV-1 and its primary receptor CD4 is one strategy for identifying new anti-HIV-1 entry inhibitors. Here we report the discovery of a novel ligand that can inhibit HIV-1 entry and infection via CD4. Biological and computational analyses of this inhibitor and its analogs, using bioactivity evaluation, Rule of Five (RO5), comparative molecular field analysis (CoMFA)/comparative molecular similarity index analysis (CoMSIA) models, and three-dimensional quantitative structure-activity relationship (3D-QSAR), singled out compound 3 as a promising lead molecule for the further development of therapeutics targeting HIV-1 entry. Our study demonstrates an effective approach for employing structure-based, rational drug design techniques to identify novel antiviral compounds with interesting biological activities.


Introduction
Acquired immune deficiency syndrome (AIDS) is one of the major diseases of large public health impact. It is caused by HIV, and it represents the final stage of HIV infection [1]. One of the hallmarks of HIV infection is the selective destruction of CD4 T-cells [2]. Treatment for patients infected with HIV is usually antiretroviral therapy (ART), with double or triple drug combinations chosen from approved drugs [3,4]. The viral entry process is one of the most promising targets for the development of new anti-HIV drugs effective in the virus replication cycle for the long-term treatment of patients with AIDS [5,6]. Two types of HIV are recognized, HIV-1, which is responsible for the majority of HIV infections worldwide, and HIV type 2 (HIV-2), which causes the infection endemic in western Africa [7,8]. Here, our work mainly focuses on the worldwide infection caused by HIV-1.
CD4, as the primary receptor of the HIV-1 envelope glycol protein 120 (gp120), is critical for HIV-1 entry into host cells [9,10]. HIV-1 infection is initiated by the binding and attachment of gp120

NSC119915 Displayed Potent Anti-HIV-1 Activity
We performed anti-HIV-1 infection and anti-HIV-1 entry assays by measuring GFP expression and intracellular p24 antigen level. For the anti-HIV-1 infection assays, the CEM-GFP reporter cell line was used, and HIV-1 infection induced a 10-to 1000-fold increase in the relative fluorescence of the cells [50]. Cells treated with 3 for 4-5 days showed decreased GFP fluorescence values in the HIV-1 infected group, with an IC50 value of 10 µ M (Figure 2A). This indicated that 3 could inhibit HIV-1 infection.
We also performed anti-HIV-1 entry assays to investigate the potential suppression by 3 of HIV-1 infection via its entry mechanism. Peripheral blood mononuclear cells (PBMCs) that were infected with HIV-1 (NL4-3 strain) for 24 h showed a decrease in their intracellular p24 antigen levels after treatment with 3, as well as dose-dependent HIV-1 inhibition with an IC50 value of 9 μM ( Figure 2C). The inhibition by a positive control compound AMD3100 at 0.1 μM was 57.6% ( Figure 2B). These data further demonstrated that 3 acted as an HIV-1 entry inhibitor.

NSC119915 Displayed Potent Anti-HIV-1 Activity
We performed anti-HIV-1 infection and anti-HIV-1 entry assays by measuring GFP expression and intracellular p24 antigen level. For the anti-HIV-1 infection assays, the CEM-GFP reporter cell line was used, and HIV-1 infection induced a 10-to 1000-fold increase in the relative fluorescence of the cells [50]. Cells treated with 3 for 4-5 days showed decreased GFP fluorescence values in the HIV-1 infected group, with an IC50 value of 10 µ M (Figure 2A). This indicated that 3 could inhibit HIV-1 infection.
We also performed anti-HIV-1 entry assays to investigate the potential suppression by 3 of HIV-1 infection via its entry mechanism. Peripheral blood mononuclear cells (PBMCs) that were infected with HIV-1 (NL4-3 strain) for 24 h showed a decrease in their intracellular p24 antigen levels after treatment with 3, as well as dose-dependent HIV-1 inhibition with an IC50 value of 9 μM ( Figure 2C). The inhibition by a positive control compound AMD3100 at 0.1 μM was 57.6% ( Figure 2B). These data further demonstrated that 3 acted as an HIV-1 entry inhibitor. 36.86

NSC119915 Displayed Potent Anti-HIV-1 Activity
We performed anti-HIV-1 infection and anti-HIV-1 entry assays by measuring GFP expression and intracellular p24 antigen level. For the anti-HIV-1 infection assays, the CEM-GFP reporter cell line was used, and HIV-1 infection induced a 10-to 1000-fold increase in the relative fluorescence of the cells [50]. Cells treated with 3 for 4-5 days showed decreased GFP fluorescence values in the HIV-1 infected group, with an IC50 value of 10 µ M (Figure 2A). This indicated that 3 could inhibit HIV-1 infection.
We also performed anti-HIV-1 entry assays to investigate the potential suppression by 3 of HIV-1 infection via its entry mechanism. Peripheral blood mononuclear cells (PBMCs) that were infected with HIV-1 (NL4-3 strain) for 24 h showed a decrease in their intracellular p24 antigen levels after treatment with 3, as well as dose-dependent HIV-1 inhibition with an IC50 value of 9 μM ( Figure 2C). The inhibition by a positive control compound AMD3100 at 0.1 μM was 57.6% ( Figure 2B). These data further demonstrated that 3 acted as an HIV-1 entry inhibitor.

NSC119915 Displayed Potent Anti-HIV-1 Activity
We performed anti-HIV-1 infection and anti-HIV-1 entry assays by measuring GFP expression and intracellular p24 antigen level. For the anti-HIV-1 infection assays, the CEM-GFP reporter cell line was used, and HIV-1 infection induced a 10-to 1000-fold increase in the relative fluorescence of the cells [50]. Cells treated with 3 for 4-5 days showed decreased GFP fluorescence values in the HIV-1 infected group, with an IC50 value of 10 µ M (Figure 2A). This indicated that 3 could inhibit HIV-1 infection.
We also performed anti-HIV-1 entry assays to investigate the potential suppression by 3 of HIV-1 infection via its entry mechanism. Peripheral blood mononuclear cells (PBMCs) that were infected with HIV-1 (NL4-3 strain) for 24 h showed a decrease in their intracellular p24 antigen levels after treatment with 3, as well as dose-dependent HIV-1 inhibition with an IC50 value of 9 μM ( Figure 2C). The inhibition by a positive control compound AMD3100 at 0.1 μM was 57.6% ( Figure 2B). These data further demonstrated that 3 acted as an HIV-1 entry inhibitor.

NSC119915 Displayed Potent Anti-HIV-1 Activity
We performed anti-HIV-1 infection and anti-HIV-1 entry assays by measuring GFP expression and intracellular p24 antigen level. For the anti-HIV-1 infection assays, the CEM-GFP reporter cell line was used, and HIV-1 infection induced a 10-to 1000-fold increase in the relative fluorescence of the cells [50]. Cells treated with 3 for 4-5 days showed decreased GFP fluorescence values in the HIV-1 infected group, with an IC50 value of 10 µ M (Figure 2A). This indicated that 3 could inhibit HIV-1 infection.
We also performed anti-HIV-1 entry assays to investigate the potential suppression by 3 of HIV-1 infection via its entry mechanism. Peripheral blood mononuclear cells (PBMCs) that were infected with HIV-1 (NL4-3 strain) for 24 h showed a decrease in their intracellular p24 antigen levels after treatment with 3, as well as dose-dependent HIV-1 inhibition with an IC50 value of 9 μM ( Figure 2C). The inhibition by a positive control compound AMD3100 at 0.1 μM was 57.6% ( Figure 2B). These data further demonstrated that 3 acted as an HIV-1 entry inhibitor. 33.09

NSC119915 Displayed Potent Anti-HIV-1 Activity
We performed anti-HIV-1 infection and anti-HIV-1 entry assays by measuring GFP expression and intracellular p24 antigen level. For the anti-HIV-1 infection assays, the CEM-GFP reporter cell line was used, and HIV-1 infection induced a 10-to 1000-fold increase in the relative fluorescence of the cells [50]. Cells treated with 3 for 4-5 days showed decreased GFP fluorescence values in the HIV-1 infected group, with an IC 50 value of 10 µM (Figure 2A). This indicated that 3 could inhibit HIV-1 infection.
We also performed anti-HIV-1 entry assays to investigate the potential suppression by 3 of HIV-1 infection via its entry mechanism. Peripheral blood mononuclear cells (PBMCs) that were infected with HIV-1 (NL4-3 strain) for 24 h showed a decrease in their intracellular p24 antigen levels after treatment with 3, as well as dose-dependent HIV-1 inhibition with an IC 50 value of 9 µM ( Figure 2C). The inhibition by a positive control compound AMD3100 at 0.1 µM was 57.6% ( Figure 2B). These data further demonstrated that 3 acted as an HIV-1 entry inhibitor.

NSC119915 Showed No Cytotoxicity in PBMCs
We employed a cell viability assay to test the potential cytotoxicity of 3. In parallel with the anti-HIV-1 assay described above, the same densities of PBMCs were seeded and treated with the test compounds for the same duration. 3 showed no cytotoxicity in PBMCs, even when tested at a higher concentration (150 μM) than that used in the competitive CD4 binding assay ( Figure 2D).

Bioactivity Evaluations of Compounds Collected from Structural Similarity Screening
We performed a similarity search study using the Enhanced NCI Database Browser 2.2 to select 40 compounds from a total of 250,250 compounds with a Tanimoto index of above 90%, and obtained them from NCI. The Tanimoto index is a useful tool for 2D fragment-based similarity searching [51]. The competitive CD4 binding assay showed that compounds 17, 18, 19, 20, 21, 22, 23, 24, 25, 27, 28, 29, and 32 had better CD4 competitive binding affinities than other compounds, as their inhibitions at 100 µ M were all above 20% ( Figure 3A). The binding inhibition and compound structures are showed in Table 2 and Figure 3A. Among these compounds, 29 ( Figure 3B) showed the best binding affinity when compared to the other compounds, and its chemical structure was quite similar to that of 3. The binding inhibition of 29 in the anti-human CD4 antibody competitive binding assays was 85.29% at 100 μM, and the IC50 value was 14 μM ( Figure 3C).

NSC119915 Showed No Cytotoxicity in PBMCs
We employed a cell viability assay to test the potential cytotoxicity of 3. In parallel with the anti-HIV-1 assay described above, the same densities of PBMCs were seeded and treated with the test compounds for the same duration. 3 showed no cytotoxicity in PBMCs, even when tested at a higher concentration (150 µM) than that used in the competitive CD4 binding assay ( Figure 2D).

Bioactivity Evaluations of Compounds Collected from Structural Similarity Screening
We performed a similarity search study using the Enhanced NCI Database Browser 2.2 to select 40 compounds from a total of 250,250 compounds with a Tanimoto index of above 90%, and obtained them from NCI. The Tanimoto index is a useful tool for 2D fragment-based similarity searching [51]. The competitive CD4 binding assay showed that compounds 17, 18, 19, 20, 21, 22, 23, 24, 25, 27, 28, 29, and 32 had better CD4 competitive binding affinities than other compounds, as their inhibitions at 100 µM were all above 20% ( Figure 3A). The binding inhibition and compound structures are showed in Table 2 and Figure 3A. Among these compounds, 29 ( Figure 3B) showed the best binding affinity when compared to the other compounds, and its chemical structure was quite similar to that of 3. The binding inhibition of 29 in the anti-human CD4 antibody competitive binding assays was 85.29% at 100 µM, and the IC 50 value was 14 µM ( Figure 3C). We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line.
The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC 50 value of 29 was 2 µM ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µM ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µM are shown in Table 2. confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. Most of the compounds collected from NCI database using the similarity search approach share rtain common substructures with 3. These compounds were classified into four classes for further udies. The common structures of each classification and CD4 binding inhibition at 100 µ M are own in Table 2. effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results nfirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic fects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound r developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share rtain common substructures with 3. These compounds were classified into four classes for further udies. The common structures of each classification and CD4 binding inhibition at 100 µ M are own in Table 2. value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. Table 2. The structures and inhibition data of compounds generated from the NCI similarity search. Class I: alue of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results nfirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic fects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound r developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share rtain common substructures with 3. These compounds were classified into four classes for further udies. The common structures of each classification and CD4 binding inhibition at 100 µ M are own in Table 2. was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. as an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 alue of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results nfirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic fects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound r developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share rtain common substructures with 3. These compounds were classified into four classes for further udies. The common structures of each classification and CD4 binding inhibition at 100 µ M are own in Table 2. was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. ithout drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 as an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 alue of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results nfirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic fects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound r developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share rtain common substructures with 3. These compounds were classified into four classes for further udies. The common structures of each classification and CD4 binding inhibition at 100 µ M are own in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. dose-dependent manner (D), and a CellTiter Blue assay of 29 showed a complete lack of cytotoxicity effect (E).
We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. showed the best binding affinity with an IC50 value of 14 µ M (B,C). 29 inhibited HIV-1 infection in a dose-dependent manner (D), and a CellTiter Blue assay of 29 showed a complete lack of cytotoxicity effect (E).
We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. showed the best binding affinity with an IC50 value of 14 µ M (B,C). 29 inhibited HIV-1 infection in a dose-dependent manner (D), and a CellTiter Blue assay of 29 showed a complete lack of cytotoxicity effect (E).
We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2. showed the best binding affinity with an IC50 value of 14 µ M (B,C). 29 inhibited HIV-1 infection in a dose-dependent manner (D), and a CellTiter Blue assay of 29 showed a complete lack of cytotoxicity effect (E).
We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. he GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group ithout drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 as an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 alue of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results nfirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic fects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound r developing new ligands to inhibit HIV-1 infection via CD4. Most of the compounds collected from NCI database using the similarity search approach share rtain common substructures with 3. These compounds were classified into four classes for further udies. The common structures of each classification and CD4 binding inhibition at 100 µ M are own in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.  We also evaluated the anti-HIV-1 infection activity of 29 using the CEM-GFP reporter cell line. The GFP values were significantly lower for HIV-1 infected groups treated with 29 than the group without drug treatment, and the decrease of the GFP values were dose-dependent, indicating that 29 was an effective HIV-1 inhibitor identified by similarity search using 3 as a lead template. The IC50 value of 29 was 2 µ M ( Figure 3D), and 29 showed no cytotoxicity, as determined by the cell viability in the CEM-GFP cell line, when tested at the highest concentration of 25 µ M ( Figure 3E). These results confirmed that 29 is a potent HIV-1 inhibitor, and that its anti-HIV-1 activity was not due to cytotoxic effects on the host cells. These bioactivity data also provide evidence that 3 is a viable lead compound for developing new ligands to inhibit HIV-1 infection via CD4.
Most of the compounds collected from NCI database using the similarity search approach share certain common substructures with 3. These compounds were classified into four classes for further studies. The common structures of each classification and CD4 binding inhibition at 100 µ M are shown in Table 2.

3D Quantitative Structure-Activity Relationship (3D-QSAR) Analyses
The mechanism of the ligand-receptor interaction was investigated by performing a 3D-QSAR study, which is one of the most powerful approaches for guiding further lead optimization. A total of 27 compounds collected from similarity search based on 3 were divided into training and test sets, containing of 19 and 8 compounds, respectively, for 3D-QSAR studies. Compound 31 was discarded for its bad alignment with the same substructure of other compounds listed in Table 2. The training set was used to generate 3D-QSAR models, while the test set was employed to validate the quality of the models. In the alignment step, all compounds of the training and test sets were well aligned with the same substructure. The alignment result is shown in Figure 4. CoMFA and CoMSIA studies were performed based on the molecular alignment, as described in the methods. CoMFA calculates the steric and electrostatic properties, whereas CoMSIA calculates similarity indices in the space surrounding each of the molecules in the dataset [47,48]. The statistical results generated from the leave-one-out (LOO) cross validation of the CoMFA/CoMSIA models for

3D Quantitative Structure-Activity Relationship (3D-QSAR) Analyses
The mechanism of the ligand-receptor interaction was investigated by performing a 3D-QSAR study, which is one of the most powerful approaches for guiding further lead optimization. A total of 27 compounds collected from similarity search based on 3 were divided into training and test sets, containing of 19 and 8 compounds, respectively, for 3D-QSAR studies. Compound 31 was discarded for its bad alignment with the same substructure of other compounds listed in Table 2. The training set was used to generate 3D-QSAR models, while the test set was employed to validate the quality of the models. In the alignment step, all compounds of the training and test sets were well aligned with the same substructure. The alignment result is shown in Figure 4. CoMFA and CoMSIA studies were performed based on the molecular alignment, as described in the methods. CoMFA calculates the steric and electrostatic properties, whereas CoMSIA calculates similarity indices in the space surrounding each of the molecules in the dataset [47,48]. The statistical results generated from the leave-one-out (LOO) cross validation of the CoMFA/CoMSIA models for 78.62

3D Quantitative Structure-Activity Relationship (3D-QSAR) Analyses
The mechanism of the ligand-receptor interaction was investigated by performing a 3D-QSAR study, which is one of the most powerful approaches for guiding further lead optimization. A total of 27 compounds collected from similarity search based on 3 were divided into training and test sets, containing of 19 and 8 compounds, respectively, for 3D-QSAR studies. Compound 31 was discarded for its bad alignment with the same substructure of other compounds listed in Table 2. The training set was used to generate 3D-QSAR models, while the test set was employed to validate the quality of the models. In the alignment step, all compounds of the training and test sets were well aligned with the same substructure. The alignment result is shown in Figure 4. CoMFA and CoMSIA studies were performed based on the molecular alignment, as described in the methods. CoMFA calculates the steric and electrostatic properties, whereas CoMSIA calculates similarity indices in the space surrounding each of the molecules in the dataset [47,48]. The statistical results generated from the leave-one-out (LOO) cross validation of the CoMFA/CoMSIA models for

3D Quantitative Structure-Activity Relationship (3D-QSAR) Analyses
The mechanism of the ligand-receptor interaction was investigated by performing a 3D-QSAR study, which is one of the most powerful approaches for guiding further lead optimization. A total of 27 compounds collected from similarity search based on 3 were divided into training and test sets, containing of 19 and 8 compounds, respectively, for 3D-QSAR studies. Compound 31 was discarded for its bad alignment with the same substructure of other compounds listed in Table 2. The training set was used to generate 3D-QSAR models, while the test set was employed to validate the quality of the models. In the alignment step, all compounds of the training and test sets were well aligned with the same substructure. The alignment result is shown in Figure 4. CoMFA and CoMSIA studies were performed based on the molecular alignment, as described in the methods. CoMFA calculates the steric and electrostatic properties, whereas CoMSIA calculates similarity indices in the space surrounding each of the molecules in the dataset [47,48]. The statistical results generated from the leave-one-out (LOO) cross validation of the CoMFA/CoMSIA models for 6

3D Quantitative Structure-Activity Relationship (3D-QSAR) Analyses
The mechanism of the ligand-receptor interaction was investigated by performing a 3D-QSAR study, which is one of the most powerful approaches for guiding further lead optimization. A total of 27 compounds collected from similarity search based on 3 were divided into training and test sets, containing of 19 and 8 compounds, respectively, for 3D-QSAR studies. Compound 31 was discarded for its bad alignment with the same substructure of other compounds listed in Table 2. The training set was used to generate 3D-QSAR models, while the test set was employed to validate the quality of the models. In the alignment step, all compounds of the training and test sets were well aligned with the same substructure. The alignment result is shown in Figure 4. CoMFA and CoMSIA studies were performed based on the molecular alignment, as described in the methods. CoMFA calculates the steric and electrostatic properties, whereas CoMSIA calculates similarity indices in the space surrounding each of the molecules in the dataset [47,48]. The statistical results generated from the leave-one-out (LOO) cross validation of the CoMFA/CoMSIA models for 16

3D Quantitative Structure-Activity Relationship (3D-QSAR) Analyses
The mechanism of the ligand-receptor interaction was investigated by performing a 3D-QSAR study, which is one of the most powerful approaches for guiding further lead optimization. A total of 27 compounds collected from similarity search based on 3 were divided into training and test sets, containing of 19 and 8 compounds, respectively, for 3D-QSAR studies. Compound 31 was discarded for its bad alignment with the same substructure of other compounds listed in Table 2. The training set was used to generate 3D-QSAR models, while the test set was employed to validate the quality of the models. In the alignment step, all compounds of the training and test sets were well aligned with the same substructure. The alignment result is shown in Figure 4. The mechanism of the ligand-receptor interaction was investigated by performing a 3D-QSAR study, which is one of the most powerful approaches for guiding further lead optimization. A total of 27 compounds collected from similarity search based on 3 were divided into training and test sets, containing of 19 and 8 compounds, respectively, for 3D-QSAR studies. Compound 31 was discarded for its bad alignment with the same substructure of other compounds listed in Table 2. The training set was used to generate 3D-QSAR models, while the test set was employed to validate the quality of the models. In the alignment step, all compounds of the training and test sets were well aligned with the same substructure. The alignment result is shown in Figure 4.

3D Quantitative Structure-Activity Relationship (3D-QSAR) Analyses
The mechanism of the ligand-receptor interaction was investigated by performing a 3D-QSAR study, which is one of the most powerful approaches for guiding further lead optimization. A total of 27 compounds collected from similarity search based on 3 were divided into training and test sets, containing of 19 and 8 compounds, respectively, for 3D-QSAR studies. Compound 31 was discarded for its bad alignment with the same substructure of other compounds listed in Table 2. The training set was used to generate 3D-QSAR models, while the test set was employed to validate the quality of the models. In the alignment step, all compounds of the training and test sets were well aligned with the same substructure. The alignment result is shown in Figure 4. CoMFA and CoMSIA studies were performed based on the molecular alignment, as described in the methods. CoMFA calculates the steric and electrostatic properties, whereas CoMSIA calculates similarity indices in the space surrounding each of the molecules in the dataset [47,48]. The statistical results generated from the leave-one-out (LOO) cross validation of the CoMFA/CoMSIA models for CoMFA and CoMSIA studies were performed based on the molecular alignment, as described in the methods. CoMFA calculates the steric and electrostatic properties, whereas CoMSIA calculates similarity indices in the space surrounding each of the molecules in the dataset [47,48]. The statistical results generated from the leave-one-out (LOO) cross validation of the CoMFA/CoMSIA models for CD4 binding activity are shown in Table 3, which lists the partial least squares (PLS) statistical results of the CoMFA/CoMSIA with a determined optimum number of components (ONC), cross-validated correlation coefficient (q 2 ), non-cross-validation correlation coefficient (r 2 ), standard error of the estimate (SEE) and the F value. The CoMFA model is derived from three components, showed a q 2 of 0.624, a r 2 of 0.961, a SEE of 0.145 and an F value of 141.287, and the CoMSIA model derived from 4 components, showed a q 2 of 0.732, a r 2 of 0.973, a SEE of 0.124, and an F value of 127.654. In further cross-validation of the CoMFA/CoMSIA models, the CoMFA mean q 2 values of the leave-many-out (LMO 10 and LMO 5 ) cross-validation were 0.630 and 0.561, respectively. The average r 2 and SEE of 100 runs of bootstrapping analysis were 0.969 and 0.122. The CoMSIA mean q 2 values of LMO 10 and LMO 5 cross-validation were 0.736 and 0.685, respectively. The average r 2 and SEE of the 100 runs of bootstrapping analysis were 0.985 and 0.085. The values of the bootstrap r 2 confirmed further that the CoMFA/CoMSIA models in our studies were both good models for 3D-QSAR studies. All these parameters indicated a satisfactory internal predictive ability of the model. For a further evaluation study, the CoMFA/CoMSIA models were employed to predict the activity of the test set, and the result was compared with the experiment data. The experimental and predicted data with the residuals are listed in Table 4, and the relationship between the experimental and predicted activity values of the training set and the test set are depicted in Figure 5. The predicted activity values were in good agreement with the experimental data. The CoMSIA model was better than CoMFA model when compared with the PLS parameters and residuals listed in Tables 3 and 4. Therefore, we chose the CoMSIA model for further discussion of the 3D-QSAR results.
To visualize the fields effects of the CoMSIA model, 3D contour maps in Figure 6 were generated from the SYBYL-X program, the data was transformed using from % inhibition at 100 micromolar concentration, and the transformed data was named as pIn. 3 was selected as a reference and overlaid in the maps, as it is the most active compound in the training and test sets. In the CoMSIA contour maps, various (steric, electrostatic, hydrophobic, hydrogen bond donor and hydrogen bond acceptor) characters of the compounds were collected, and they contributed 22.3%, 10.2%, 23.0%, 22.7%, and 21.8%, respectively, to the interaction of compounds listed in Table 4.  To visualize the fields effects of the CoMSIA model, 3D contour maps in Figure 6 were generated from the SYBYL-X program, the data was transformed using from % inhibition at 100 micromolar concentration, and the transformed data was named as pIn. 3 was selected as a reference and overlaid in the maps, as it is the most active compound in the training and test sets. In the CoMSIA contour maps, various (steric, electrostatic, hydrophobic, hydrogen bond donor and hydrogen bond acceptor) characters of the compounds were collected, and they contributed 22.3%, 10.2%, 23.0%, 22.7%, and 21.8%, respectively, to the interaction of compounds listed in Table 4.   The yellow contour of the CoMSIA steric contour map ( Figure 6A) near positions 6 and 7 of the 3-oxo-3H-xanthen ring shows that these positions are sterically unfavorable regions (i.e., compound 32), and propanoic acid at 3-position of 3-oxo-3H-xanthene ring is also an unfavorable substituent group. This steric contour map agreed with the docking model of 3 shown in Figure 1A, which demonstrated that these positions can interact with residues in the cavity of CD4. These observations indicated that modifications of these compounds with lager substituent groups should be avoided at these positions. The electrostatic contour map of the CoMSIA model ( Figure 6B) showed that the electronegative groups were favored near the 2, 5 positions of the 3-oxo-3H-xanthene ring, and another electronegative group was favored near the carboxyl of the propanoic acid substituent. This was consistent with the increased activity of compounds 24-27, compared with compounds 14-16. The electropositive region was found in the cyclohexa-2,5-dien-1-one region. In Figure 6C, the yellow contour indicates a favorable hydrophobic interaction region near the 3, 4, 6, 7 positions of the 3-oxo-3H-xanthene ring. These contours can explain the decreases in the activity of compounds 32 and 24-27 when compared with compound 3. The white contour showed an unfavorable hydrophobic interaction region near the 4 position of the 3-oxo-3H-xanthene ring. Figure 6D shows the contour map of the hydrogen bond donor/acceptor. The magenta contour represents a desirable hydrogen bond acceptor and it agreed with the interaction model shown in Figure 1D. The propionyloxy group of 3 can form three hydrogen bonds with the Phe26, Asn39, and Gln40 residues of CD4, and Asn39 and Gln40 are both critical residues in the procedure of the interaction between CD4 and gp120. The magenta contour near the 1 and 2 position represents an undesirable hydrogen bond, which agreed with the interaction model for CD4 and gp120 where the H atoms at the 1 and 2 positions did not contribute to forming any hydrogen bond with the receptor.
Based on the analysis of CoMSIA contour maps, our bioassay data agree well with the 3D-QSAR model, and provide useful information for the new activity of predictable compounds that are designed from the template 3. Furthermore, based on the bioassay data of 3 and 29, the contour maps can help design anti-HIV-1 entry inhibitors from the template 3.

Discussion
CD4 is critical in the process of HIV-1 invasion into its host cells, and blocking the interaction of CD4 and gp120 can significantly inhibit HIV-1 entry into host cells [52]. We investigated new HIV-1 The yellow contour of the CoMSIA steric contour map ( Figure 6A) near positions 6 and 7 of the 3-oxo-3H-xanthen ring shows that these positions are sterically unfavorable regions (i.e., compound 32), and propanoic acid at 3-position of 3-oxo-3H-xanthene ring is also an unfavorable substituent group. This steric contour map agreed with the docking model of 3 shown in Figure 1A, which demonstrated that these positions can interact with residues in the cavity of CD4. These observations indicated that modifications of these compounds with lager substituent groups should be avoided at these positions. The electrostatic contour map of the CoMSIA model ( Figure 6B) showed that the electronegative groups were favored near the 2, 5 positions of the 3-oxo-3H-xanthene ring, and another electronegative group was favored near the carboxyl of the propanoic acid substituent. This was consistent with the increased activity of compounds 24-27, compared with compounds 14-16. The electropositive region was found in the cyclohexa-2,5-dien-1-one region. In Figure 6C, the yellow contour indicates a favorable hydrophobic interaction region near the 3, 4, 6, 7 positions of the 3-oxo-3H-xanthene ring. These contours can explain the decreases in the activity of compounds 32 and 24-27 when compared with compound 3. The white contour showed an unfavorable hydrophobic interaction region near the 4 position of the 3-oxo-3H-xanthene ring. Figure 6D shows the contour map of the hydrogen bond donor/acceptor. The magenta contour represents a desirable hydrogen bond acceptor and it agreed with the interaction model shown in Figure 1D. The propionyloxy group of 3 can form three hydrogen bonds with the Phe26, Asn39, and Gln40 residues of CD4, and Asn39 and Gln40 are both critical residues in the procedure of the interaction between CD4 and gp120. The magenta contour near the 1 and 2 position represents an undesirable hydrogen bond, which agreed with the interaction model for CD4 and gp120 where the H atoms at the 1 and 2 positions did not contribute to forming any hydrogen bond with the receptor.
Based on the analysis of CoMSIA contour maps, our bioassay data agree well with the 3D-QSAR model, and provide useful information for the new activity of predictable compounds that are designed from the template 3. Furthermore, based on the bioassay data of 3 and 29, the contour maps can help design anti-HIV-1 entry inhibitors from the template 3.

Discussion
CD4 is critical in the process of HIV-1 invasion into its host cells, and blocking the interaction of CD4 and gp120 can significantly inhibit HIV-1 entry into host cells [52]. We investigated new HIV-1 entry inhibitors via CD4 by employing a structure-based virtual screening approach to dock compounds from the NCI diversity data set to some critical residues of the cavity of CD4. Among the docked top-score compounds generated from this virtual screening study, 3 showed strong competitive binding affinity with CD4, and good HIV-1 entry inhibition activity without showing cytotoxicity. The calculated properties of 3 showed that: (1) there are four hydrogen bond donors; (2) there are six hydrogen bond acceptors; (3) the molecular weight is 316.27 Daltons; (4) the logP is −1.335; and (5) the number of rotatable bonds is three. All five properties of 3 meet the Lipinski's rule of five (i.e., RO5, specifically MWT < 500, logP < 5, H-bond donors < 5, H-bond acceptors < 10, rotatable bond < 10) [53]. Combined with the bioassay data, 3 was found to be a drug-like compound and a viable lead with good bioactivity and no cytotoxicity. Considering the characteristics of its molecular structure, some previous publications have indicated that 3 at a high concentration (100 µM) could inhibit the growth of various cancer cell lines by increasing the intracellular reactive oxygen species (ROS) level [54,55]. Based on our bioassay data, especially our PBMC cell viability assay results, 3 did not show an ability to affect cell growth at the concentrations (≤150 µM) used in our bioassays ( Figure 2D). This finding demonstrated that the ROS mechanism may not be the reason for why 3 acts as an HIV-1 inhibitor. The chemical structure of 3 also suggests that additional structurally similar compounds can be discovered that can act as potent CD4-binding ligands. Our data indicated that 3 could be a promising compound for the development of new HIV-1 entry inhibitors that act via CD4, and that other structurally similar compounds can be discovered based on the chemical structure of 3. According to the data published by Robert J. Fisher et al. [45], 29 could also bind to NC-p7, and contribute to inhibition of HIV-1 infection. Although 3 was regarded as an inactive or less effective (0-49% protection) compound when evaluated in their anti-HIV-1 assay, our data revealed another function of 3 and 29, as these two compounds could also bind to CD4 (the primary receptor of gp120), and may therefore act as potential dual-targeting agents to inhibit HIV-1 infection. Support for this idea was provided by two advanced anti-HIV-1 infection assays: recording the fluorescence value change of infected CEM-GFP cells, and determining the intracellular p24 levels in the PBMCs. The relationship of ligand structure and CD4 binding activity was investigated in more detail with the 3D-QSAR method. Twenty-seven compounds generated by a similarity search with the lead compound 3 were used to build 3D-QSAR models. CoMFA and CoMSIA models obtained based on ligand alignment both had high q 2 and r 2 values (both greater than 0.5) and small standard errors of estimates. The models were also validated by LMO cross-validation, bootstrapping, and activity prediction of the test set. All the validations confirmed the models as being reliable and with a high predictive ability. The chemical interpretation of the contour maps generated by CoMSIA agreed well with the changes in the chemical structures and the bioactivity values of the compounds. We also investigated that all our compounds that showed a greater than 30% level of inhibition, using the PubChem database and online literature searches for any reported bioactivity and targets. The results were complicated, as compounds 3 and 29 had been reported as being active in several bioassays and multitargets; the active results and targets (Hsp70; RAPGEF4; GAPDH; RNASEH1) were fewer for 29 than for 3. However, other compounds had even fewer or no reported bioactivities. Some of these reported compounds were focused on anticancer screening or antimicrobial assays without targeting a specific receptor. Since the pan-assay interference compounds (PAINS) filter was first reported, it has become a common component of the triage process in biological screening [56]. Based on the PAINS filter results, some of our compounds (3, 28 and 29) failed to pass, and other compounds that came from similarity searches and that were used in establishing 3D-QSAR models all passed the filter (Table 5). Recently, a large-scale analysis of PAINS alerts was reported, indicating that computational PAINS filters may inappropriately flag some compound classes, creating the potential for incorrect prediction. In fact, 87 Food and Drug Administration (FDA) approved drugs contained PAINS alerts. The current recommendation is that conclusions should be drawn only after conducting orthogonal experiments. [57]. Based on theory, published large-scale analysis results, and our results reported here, some of our compounds could pass a PAINS filter and show bioactivity based on a compound that failed to pass the filter. This leads us to consider the advantage of our 3D-QSAR models, which could help future studies to avoid failing the PAINS filter, and that could provide higher specificity and lower risk based mainly on class I and class II compounds (Table 2). 4 (anticancer drug screen in mice, fructose-bisphosphate aldolase inhibitor) PASS a The contents in the bioassay column are presented as "reported active bioassay number (bioassays names or functions)", if none was reported, a "NONE" presents. b PAINS was evaluated using the function of the PAINS filter in SYBYL-X. A "PASS" means a compound passes through the PAINS filter.
In conclusion, the results from virtual screening, CD4 competitive binding assays, and HIV-1 entry inhibition assays, together with the 3D-QSAR models, can guide the use of molecular databases for the discovery of new CD4 ligands that inhibit HIV-1 entry. The lead compound 3 is an effective template for the identification of additional HIV-1 entry inhibitors. The mechanistic study using the 3D-QSAR method provided valuable insights into the structure-activity relationship of 3 and its analogs. Our 3D-QSAR models can also be used to predict the activity of new small molecules designed based on 3 prior to biological testing, making these models be a valuable tool for optimizing lead compounds with higher activities.

Molecular Modeling and Virtual Screening
The molecular modeling and virtual screening were performed using the Surflex-Dock module implemented in the SYBYL-X 2.1 program (Tripos Inc., St. Louis, MO, USA). Before the docking procedures, water molecules and other ligands were removed from the crystal structure (PDB: 1GC1 [21]), and energy minimization was performed. The NCI diversity dataset, which contains 1,974 diversity compound structures, was selected for the virtual screening study and prepared using the preparation protocol of Surflex-for-searching in the ligand structure preparation tool. The protomol (an idealized active site ligand to generate putative poses of molecules [58]) was generated, selecting the critical amino acids (residues 29,35,[40][41][42][43][44][45][46][47][48][49] in the interaction of CD4 and gp120 [21,59]. In the docking procedure, the minimization was performed pre-dock and post-dock. A total of 20 poses were generated from each docked ligand. Forty compounds of top-ranking total scores were selected for further bioassays. All compounds were ordered from the Division of Cancer Treatment & Diagnosis (DCTD) of the National Cancer Institute (NCI) (website: https://dtp.cancer.gov/organization/dscb/obtaining/vialed.htm) [60].

PBMC Viability Assay
In addition to the anti-HIV assays described above, the same numbers of PBMCs were seeded and treated with the test compounds for the same periods. After a 24 h incubation at 37 • C, the cell viability was determined using the CellTiter-Blue (Promega Co., Madison, WI, USA) cell viability assay. The fluorescence (560 EX /590 EM ) was recorded using a Synergy II microplate reader.

Compound Structural Similarity Search
For more chemical structures similar to compounds that were identified from the NCI diversity database, we conducted a similarity search study using the enhanced NCI database browser 2.2 (https://cactus.nci.nih.gov/ncidb2.2/) [61]. We obtained 40 more compounds with >90% Tanimoto index (which is superior to the Euclidean distance in 2D-fragment based similarity searching) from the NCI. Subsequently, competitive CD4 binding assays were performed to identify compounds that can bind to CD4.

3D-QSAR Study
The 27 compounds involved in 3D-QSAR study were generated from the structure similarity search listed in Table 2 and the lead compound 3 in Table 1. Molecules that had no bioactivity and did not share a common scaffold were discarded. All ligands' 3D structures were constructed using the Sketch Molecule function in SYBYL-X software. Gasteiger-Hückel charges were employed to calculate the partial atomic charges, and a TRIPOS force field was used for energy minimization with a convergence criterion of 0.05 kcal/mol·Å. The total compounds were divided into a training set and a test set.
Subsequently, molecular alignment was performed to obtain valid and reliable 3D-QSAR models [62]. In this study, 3, which had the most potent inhibitory activity was chosen as the template molecule to perform the molecular alignment by using the database align function in the SYBYL-X program. Based on the molecular alignment, CoMFA/CoMSIA models were developed for the aligned molecular data set. In this study, five molecular fields were calculated: steric, electrostatic, hydrophobic, hydrogen bond donor, and hydrogen bond acceptor. CoMSIA molecular fields were calculated using a sp 3 carbon probe atom carrying a charge of +1.0, and an attenuation factor α of 0.3 using default settings in SYBYL.
To generate statistically significant 3D-QSAR models, a PLS regression was carried out. The LOO cross-validation analysis was performed to determine the ONC and q 2 . After that, non-cross-validation analysis was performed with a threshold column filtering of 2.0 kcal/mol to generate the optimally PLS regression models for CoMFA/CoMSIA. The r 2 , SEE, and F ratio between the variances of experimental and predicted activity values were obtained.
To further evaluate the statistical significance of the derived models, additional cross-validation analyses were performed using groups in the training set. The CoMFA and CoMSIA models were further analyzed by additional rigorous statistical cross-validation, using 10 and five groups in the training set, and each cross-validation process was repeated 25 times. Subsequently, a bootstrapping analysis for 100 runs was performed to measure the bias of the original calculations. In addition, the external predictive ability of the CoMFA/CoMSIA model was assessed by prediction the activity of the test set and compared with the experiment data. After the evaluation of 3D-QSAR model, contour maps of all fields were constructed with default values of 80% favored and 20% disfavored contributions.

Data and Statistical Analysis
Statistical analysis was performed using a one-way ANOVA (GraphPad Software: GraphPad Prism 7 for Windows, GraphPad Software Inc., La Jolla, CA, USA). Average values were expressed as mean ± SD or SEM, n > 3. The results of calcium influxes/effluxes and Western blots were representatives of at least of three independent experiments. A P value less than 0.05 was considered statistically significant. * p value < 0.05, ** p value < 0.005, *** p value < 0.0005.