*3.5. E*ff*ect of Compound 5 on the Inhibition of IKK Pathway*

*3.4. In Silico Docking with IKKβ* The docking calculations were carried out using the protein structure of IKKβ (The Protein Data Bank code; 4KIK.pdb). Using the Sybyl program, the apo-protein of IKKβ was obtained by removing the original ligand **K252a** contained in 4KIK.pdb. The original ligand K252a was again docked to the apo-protein, to confirm that the flexible docking procedure worked well. Through a flexible docking procedure repeated 30 times, 30 complexes between apo-protein and **K252a** were obtained. Their binding energy ranged from −28.44 to −10.24 kcal/mol and their binding poses were good to be comparable with 4KIK.pdb.The binding pocket of IKKβ was determined using the LigPlot software bind well at the active site of the IKKβ protein (Figure 9B). **Figure 8.** Summary of the overall detailed intermolecular interactions and their contribution to each To validate the in silico docking prediction that IKKβ is a target for compound **5**, we tested the effect of compound **5** on the inhibition of the IKK signaling pathway using a cell-based kinase assay. We confirmed that the phosphorylation of IKKα/β and its downstream targets IκB and p65/RelA was rapidly induced within 10 min and then gradually decreased upon 10 ng/mL TNFα stimulation (Figure 10A). Under this experimental condition, we determined whether compound **5** inhibited the TNFα-induced IKK activity. HCT116 cells were pre-treated with different concentrations of compound **5** (0, 50, 100 µM) for 30 min and then stimulated with 10 ng/mL TNFα for 10 min. We observed

the 30 iterations ranged from −15.92 to −13.09 kcal/mol. The interactions between IKKβ and compound **5** were analyzed using the LigPlot progam. Six residues including Thr23, Val29, Glu61, Met65, Met96, and Ile165 showed the hydrophobic interactions with the ligand and three residues including Gly27, Lys44, and Asp166 formed hydrogen bonds (H bonds) with the ligand (Figure 9A). The binding pocket of compound **5** resided in IKKβ was visualized using the PyMOL program (PyMOL Molecular Graphics System, version 1.0r1, Schrödinger, LLC, Portland, OR, USA). Isoflavone compound **5** in IKKβ exhibited a slightly different binding pattern from those of the original ligand **K252a**. However, both isoflavone **5** and original ligand **K252a** have been shown to

as previously reported [48]. They are composed of 16 residues; 14 residues, namely Leu21, Gly22, Thr23, Val25, Ala42, Lys44, Glu61, Val74, Met96, Tyr98, Glu149, Asn150, Ile165, and Asp166 are *3.4. In Silico Docking with IKKβ*

that pre-treatment with compound **5** dose-dependently reduced phosphorylation of IKKα/β and its downstream target IκB and p65/RelA, induced by TNFα (Figure 10B). These data suggest that compound **5** inhibits TNFα-induced NF-κB activation through the targeting of IKK. (PyMOL Molecular Graphics System, version 1.0r1, Schrödinger, LLC, Portland, OR, USA). Isoflavone compound **5** in IKKβ exhibited a slightly different binding pattern from those of the original ligand **K252a**. However, both isoflavone **5** and original ligand **K252a** have been shown to

*Crystals* **2020**, *10*, x FOR PEER REVIEW 11 of 15

bind well at the active site of the IKKβ protein (Figure 9B).

The binding pocket of compound **5** resided in IKKβ was visualized using the PyMOL program

*Crystals* **2020**, *10*, x FOR PEER REVIEW 10 of 15

**Figure 7.** Two-dimensional fingerprint plots of the most important intermolecular contacts in each molecule **I** and **II**. For **I**: full (**A**) and resolved into O···H (**B**), H···H (**C**), C···H (**D**). For **II**: full (**E**) and

**Figure 8.** Summary of the overall detailed intermolecular interactions and their contribution to each crystal structure **I** and **II**. For **I**; H···H ; 43.5%, O···H; 25.1%, O···C; 6.4%, C···H; 17.8%, C···C; 5.7%, O···O;

The docking calculations were carried out using the protein structure of IKKβ (The Protein Data Bank code; 4KIK.pdb). Using the Sybyl program, the apo-protein of IKKβ was obtained by removing the original ligand **K252a** contained in 4KIK.pdb. The original ligand K252a was again docked to the apo-protein, to confirm that the flexible docking procedure worked well. Through a flexible docking procedure repeated 30 times, 30 complexes between apo-protein and **K252a** were obtained. Their binding energy ranged from −28.44 to −10.24 kcal/mol and their binding poses were good to be comparable with 4KIK.pdb.The binding pocket of IKKβ was determined using the LigPlot software as previously reported [48]. They are composed of 16 residues; 14 residues, namely Leu21, Gly22, Thr23, Val25, Ala42, Lys44, Glu61, Val74, Met96, Tyr98, Glu149, Asn150, Ile165, and Asp166 are involved in hydrophobic interactions, and two residues, Glu97 and Cys99 are involved in hydrogen bonds. Using the three-dimensional structure of compound **5** obtained in this study, docking with apo-protein was performed in the same way as the original ligand. The binding energy generated by the 30 iterations ranged from −15.92 to −13.09 kcal/mol. The interactions between IKKβ and compound **5** were analyzed using the LigPlot progam. Six residues including Thr23, Val29, Glu61,

1.5%, For **II**; H···H ; 42.5%, O···H; 29.1%, O···C; 5.6%, C···H; 16.3%, C···C; 5.7%, O···O; 0.8%.

The overall contribution to the total Hirshfeld surface is illustrated in Figure 8.

resolved into O···H (**F**), H···H (**G**), C···H (**H**).

**Figure 9.** (**A**) The residues participating in the binding sites of the compound **5**-IKKβ complex analyzed by using the LigPlot program. (**B**) Three-dimensional image of the IKKβ and compound **5** complex, where compound **5** is colored in green and the original ligand **K252a** contained in IKKβ is colored in red. TNFα-induced IKK activity. HCT116 cells were pre-treated with different concentrations of compound **5** (0, 50, 100 μM) for 30 min and then stimulated with 10 ng/mL TNFα for 10 min. We observed that pre-treatment with compound **5** dose-dependently reduced phosphorylation of IKKα/β and its downstream target IκB and p65/RelA, induced by TNFα (Figure 10B). These data suggest that compound **5** inhibits TNFα-induced NF-κB activation through the targeting of IKK.

**Figure 10.** Effect of compound **5** on the inhibition of the IKK signaling pathway in HCT116 colon cancer cells. (**A**) HCT116 cells were starved with 0.5% fetal bovine serum for 24 h, followed by stimulation with 10 ng/mL TNFα for the indicated times. Whole-cell lysates were prepared, and Western blotting was performed using the phospho-specific antibody against IKKα/β (Ser176/180), IκBα (Ser32), and p65/RelA NF-κB (Ser536). Glyceraldehyde phosphate dehydrogenase (GAPDH) was used as an internal control. (**B**) HCT116 cells were starved with 0.5% FBS for 24 h, followed by pre-treatment with compound **5** (5 or 20 μM) 30 min before stimulation with 10 ng/mL TNFα. After 10 min, whole-cell lysates were prepared, and Western blotting was performed using the phosphospecific antibody against IKKα/β (Ser176/180), IκBα (Ser32), and p65/RelA NF-κB (Ser536). GAPDH **Figure 10.** Effect of compound **5** on the inhibition of the IKK signaling pathway in HCT116 colon cancer cells. (**A**) HCT116 cells were starved with 0.5% fetal bovine serum for 24 h, followed by stimulation with 10 ng/mL TNFα for the indicated times. Whole-cell lysates were prepared, and Western blotting was performed using the phospho-specific antibody against IKKα/β (Ser176/180), IκBα (Ser32), and p65/RelA NF-κB (Ser536). Glyceraldehyde phosphate dehydrogenase (GAPDH) was used as an internal control. (**B**) HCT116 cells were starved with 0.5% FBS for 24 h, followed by pre-treatment with compound **5** (5 or 20 µM) 30 min before stimulation with 10 ng/mL TNFα. After 10 min, whole-cell lysates were prepared, and Western blotting was performed using the phospho-specific antibody against IKKα/β (Ser176/180), IκBα (Ser32), and p65/RelA NF-κB (Ser536). GAPDH was used as an internal control.

#### was used as an internal control. *3.6. E*ff*ect of Compound 5 on the Inhibition of Clonogenicity of HCT116 Cells*

to proliferate into viable colonies.

*3.6. Effect of Compound 5 on the Inhibition of Clonogenicity of HCT116 Cells* To support the idea that NF-κB inhibition through IKK targeting by compound **5** exerts an antiproliferation effect, we tested the inhibitory activity of compound **5** on the clonogenicity of HCT116 human colon cancer cells. Clonogenic assay is an in vitro non-destructive method known to To support the idea that NF-κB inhibition through IKK targeting by compound **5** exerts an antiproliferation effect, we tested the inhibitory activity of compound **5** on the clonogenicity of HCT116 human colon cancer cells. Clonogenic assay is an in vitro non-destructive method known to reflect the in vivo evaluation of anticancer drugs. Treatment with compound **5** for 7 days resulted in a dose-dependent loss of clonogenicity of HCT116 cells (Figure 11). Its GI<sup>50</sup> value was determined to be

reflect the in vivo evaluation of anticancer drugs. Treatment with compound **5** for 7 days resulted in a dose-dependent loss of clonogenicity of HCT116 cells (Figure 11). Its GI<sup>50</sup> value was determined to 17.2 µM. These data suggest that IKK targeting by compound **5** reduced the individual cell ability to proliferate into viable colonies. *Crystals* **2020**, *10*, x FOR PEER REVIEW 12 of 15

**Figure 11.** Effect of compound **5** on the inhibition of clonogenicity of HCT116 colon cancer cells. **Figure 11.** Effect of compound **5** on the inhibition of clonogenicity of HCT116 colon cancer cells.

NF-κB is constitutively activated in most cancer cells. The inhibition of NF-κB in cancer cells causes the induction of the cell cycle arrest and apoptosis. Therefore, pharmacological NF-κB inhibitors have been widely used for cancer prevention and therapy. The full activation of p65/RelA NF-κB is necessary for the release of the NF-κB complex from IκB. IKK phosphorylates IκB on Serine-32, triggering the proteasome-dependent proteolysis of IκB and releasing NF-κB from IκB. IKK also phosphorylates and degrades the Forkhead transcription factor FOXO3a that is involved in cell cycle arrest and apoptosis [49] and activates insulin receptor substrate (IRS) to impair insulin signaling [50], suggesting that IKK inhibition can exhibit multiple anticancer activities in addition to inhibiting NF-κB. Previous study has demonstrated that the most effective and selective approach for NF-κB inhibition might be offered by the IKK inhibitor [51], suggesting that the selective targeting of IKK is a promising therapeutic strategy for anticancer drug development. In this study, we identified NF-κB is constitutively activated in most cancer cells. The inhibition of NF-κB in cancer cells causes the induction of the cell cycle arrest and apoptosis. Therefore, pharmacological NF-κB inhibitors have been widely used for cancer prevention and therapy. The full activation of p65/RelA NF-κB is necessary for the release of the NF-κB complex from IκB. IKK phosphorylates IκB on Serine-32, triggering the proteasome-dependent proteolysis of IκB and releasing NF-κB from IκB. IKK also phosphorylates and degrades the Forkhead transcription factor FOXO3a that is involved in cell cycle arrest and apoptosis [49] and activates insulin receptor substrate (IRS) to impair insulin signaling [50], suggesting that IKK inhibition can exhibit multiple anticancer activities in addition to inhibiting NF-κB. Previous study has demonstrated that the most effective and selective approach for NF-κB inhibition might be offered by the IKK inhibitor [51], suggesting that the selective targeting of IKK is a promising therapeutic strategy for anticancer drug development. In this study, we identified compound **5** that inhibits the NF-κB signaling pathway through targeting the upstream kinase IKKβ.

compound **5** that inhibits the NF-κB signaling pathway through targeting the upstream kinase IKKβ.
