Coumestrol from the national cancer Institute's natural product library is a novel inhibitor of protein kinase CK2

Abstract

Background: Casein kinase 2 (CK2) is involved in various cellular events such as proliferation, apoptosis, and the cell cycle. CK2 overexpression is associated with multiple human cancers and may therefore be a promising target for cancer therapy. To identity novel classes of inhibitors for CK2, we screened a natural product library obtained from National Cancer Institute.

Methods: The quantitative luminescent kinase assay ADP-Glo™ was used to screen CK2 inhibitors from the natural product library. The same assay was used to determine cell-free dose-dependent response of CK2 inhibitors and conduct a kinetic study. Docking was performed to predict the binding patterns of selected CK2 inhibitors. Western blot analysis was used to evaluate Akt phosphorylation specific to CK2 and apoptosis effect. The cell viability assay CellTiter-Glo(®) was used to evaluate the inhibition effects of CK2 inhibitors on cancer cells.

Results: We identified coumestrol as a novel reversible ATP competitive CK2 inhibitor with an IC(50) value of 228 nM. Coumestrol is a plant-derived compound that belongs to the class of phytoestrogens, natural compounds that mimic the biological activity of estrogens. In our study, coumestrol showed high selectivity among 13 kinases. The hydrogen bonds formed between coumestrol and the amino acids in the ATP binding site were first reviewed by a molecular docking study that suggested a possible interaction of coumestrol with the hinge region of ATP site of CK2. In addition, coumestrol inhibited cancer cell growth partially through down-regulation of CK2-specific Akt phosphorylation. Finally, coumestrol exerted strong inhibition effects on the growth of three cancer cell lines.

Conclusion: Our study shows that coumestrol, a novel ATP competitive and cell permeable CK2 inhibitor with submicromolar IC(50), had inhibition effects on the growth of three cancer cell lines and may represent a promising class of CK2 inhibitors.

Figure 1

Inhibition effects of representative natural compounds on CK2 kinase activity. A. Natural compounds from NCI plates were screened for their CK2 inhibition activities at 10 μM using a kinase assay. Relative kinase activity of CK2 is shown on the Y axis as a relative number to the control that did not have inhibitor treatment. Compounds 1-F4 (coumestrol), 1-D6 (curcumin) and 1-F3 (aristolochic acid I) showed more than 50% inhibition on kinase activity at 10 μM. Data represents the average of duplicates and bars indicate standard deviation. 1-X represents compounds from NCI plate 13091250 and 2-X represents compounds from NCI plate 13091251. B. Chemical structures of coumestrol, curcumin and aristolochic acid I. C. IC₅₀ of coumestrol.

Figure 2

Kinetic analysis and reversibility assay of CK2 inhibition by coumestrol. A. Lineweaver-Burk plots of inhibition of CK2 by coumestrol: ● 0 μM, ■ 0.1 μM, ▲ 0.5 uM. Substrate concentration was fixed at 200 μM. The plots illustrate that coumestrol is an ATP competitive CK2 inhibitor. The data represents means of duplicate experiments. B. Reversibility study of coumestrol. Coumestrol was pre-incubated with 100 ug/ml of CK2 for 1 hour and then a kinase assay was performed at a final concentration of coumestrol at 100 uM as described in Methods. CK2 kinase activity is represented as relative CK2 activity to control. Data points represent the average of triplicate experiments and bars indicate standard deviation.

Figure 3

Predicted binding of coumestrol, curcumin and aristolochic acid I in the ATP binding site of CK2. The binding mode of coumestrol (A), curcumin (B), and aristolochic acid 1 (C) in the active site of CK2 was predicted by docking. The three compounds (carbon atoms colored in orange) and an ATP analog, phosphoaminophosphonic acid-adenylate ester (carbon atoms colored in yellow), were overlayed together. The docked pose indicated that hydrogen bonds were formed between coumestrol and CK2. Hydrogen bonds are labeled in green dotted lines. CK2 residues adjacent to coumestrol Glu114, Val116, Lys68, and a conserved water molecule, are shown in line representation along with coumestrol (in A), curcumin (in B), or aristolochic acid 1 (in C) (red represents oxygen, blue represents nitrogen and white represents hydrogen). The rest of the CK2 protein is shown in the flat ribbon. (D). Summary of interactions of coumestrol, curcumin and aristolochic acid I with CK2. Residues that make H-bonds, LibDockScore and IC₅₀ values of the three inhibitors were listed.

Figure 4

Downstream signalling in A549 lung cancer cells treated with coumestrol and inhibition effects of coumestrol on cellular viability in three cancer cell lines. A. Phosphorylated Akt (Ser129), total Akt, and PARP were measured by western blot analysis. B-actin was used as loading control. Expression of pAKT s129 was quantified using ImageJ software and the mean of relative expression level to β-actin or to total AKT was presented (mean ± SD). B. Coumestrol significantly decreased the expression of pAKT s129 in A549 cells (*, p < 0.05, Student t-test). C. Annexin V analysis of apoptosis induced by CK2α siRNA. A549 cancer cells were treated with 100 nM CK2α siRNA and 100 nM control siRNA for 72 h. D, E, F. A549, Jurkat and Hela cells were cultured in the absence and in increasing concentrations of coumestrol (0.1 uM to 100 μM) as indicated. Cellular viability (normalized to DMSO control) was measured after 48 hours using CellTiter-Glo®Luminescent Cell Viability Assay. Data points represent the average of IC₅₀ value of coumestrol in triplet experiments and bars indicate SD.