Disruption of focal adhesion kinase and p53 interaction with small molecule compound R2 reactivated p53 and blocked tumor growth

Abstract

Background: Focal Adhesion Kinase (FAK) is a 125 kDa non-receptor kinase that plays a major role in cancer cell survival and metastasis.

Methods: We performed computer modeling of the p53 peptide containing the site of interaction with FAK, predicted the peptide structure and docked it into the three-dimensional structure of the N-terminal domain of FAK involved in the complex with p53. We screened small molecule compounds that targeted the site of the FAK-p53 interaction and identified compounds (called Roslins, or R compounds) docked in silico to this site.

Results: By different assays in isogenic HCT116p53+/+ and HCT116 p53-/- cells we identified a small molecule compound called Roslin 2 (R2) that bound FAK, disrupted the binding of FAK and p53 and decreased cancer cell viability and clonogenicity in a p53-dependent manner. In addition, dual-luciferase assays demonstrated that the R2 compound increased p53 transcriptional activity that was inhibited by FAK using p21, Mdm-2, and Bax-promoter targets. R2 also caused increased expression of p53 targets: p21, Mdm-2 and Bax proteins. Furthermore, R2 significantly decreased tumor growth, disrupted the complex of FAK and p53, and up-regulated p21 in HCT116 p53+/+ but not in HCT116 p53-/- xenografts in vivo. In addition, R2 sensitized HCT116p53+/+ cells to doxorubicin and 5-fluorouracil.

Conclusions: Thus, disruption of the FAK and p53 interaction with a novel small molecule reactivated p53 in cancer cells in vitro and in vivo and can be effectively used for development of FAK-p53 targeted cancer therapy approaches.

Figure 1

The computer modeling and docking of p53 peptide involved in interaction with FAK and small molecules targeting FAK-p53 interaction. A. The secondary structure of p53 peptide (43–73 aa) predicted with PHYRE (Protein Homology/analogy recognition engine), as described [34]. The 7 amino-acid p53 peptide (65–72 amino acids of p53) found to be involved in interaction with FAK [20] is shown by grey color. B. The docking of the 7 amino acid p53 peptide involved in interaction with FAK inside the crystal structure of FAK-NT (N-terminal domain of FAK). The amino acids of FAK-NT interacting with the 7 amino acid p53 peptide are shown in white color. C. Zoomed image of FAK-NT interaction with the 7 amino acid p53 peptide. The amino-acids of FAK interacting with p53 peptide: R86, V95, W97, R125, I126, R127, L129, F147, Q150, D154, E256, F258, K259, P332, I336 and N339. D. Small molecules targeting FAK-p53 interaction. Screening of NCI small molecule database with DOCK5.1 program identified small molecules (called R compounds) docked into the region of FAK and p53 interaction. The purple color marks small molecule spheres. Peptide is shown by blue color and FAK-NT by green color.

Figure 2

R2 is a lead compound targeting FAK-p53 interaction. A. The viability MTT assay with HCTp53⁺/⁺ and HCTp53^-/^- cells identified small molecules, called R compounds that significantly decreased the viability of HCT116 p53⁺/⁺ cells compared with HCT116p53^-/^- cells. * P<0.05 viability less in HCT116p53⁺/⁺ cells versus HCT116 p53^-/^- cells. B. R2 significantly decreased cancer cell clonogenicity in a p53-dependent manner. The compound R2 decreased clonogenicity in HCT116p53⁺/⁺ cells more significantly than in HCT116p53^-/^- cells. C. The structure of R2 compound. D. The R2 compound decreased cancer cell viability in a p53-and dose-dependent manner. MTT assay with different doses of R2 compound was performed in HCT116p53⁺/⁺ and HCT116p53^-/^- cells. *p<0.05, R2-treated HCT116p53⁺/⁺ versus HCT116 p53^-/^- cells. E, F. R2 compound decreased the viability of cancer cell lines with wild type p53 more efficiently than with mutant p53. MTT assay was performed with different doses of R2 in MCF-7 (wild type p53) (E) and MDA231 (mutant p53) (F) breast cancer cells. * p<0.05 treated with R2 versus untreated cells. G. MTT assay with R2 in pancreatic cancer cell line, Miapaca-2 cells (mutant p53). H. MTT assay with R2 in normal human WI 38-hTERT fibroblasts. The MTT assay was performed as in Figure 2 E, F.

Figure 3

R2 bound to the FAK-N-terminal domain and disrupted interaction of FAK and p53 proteins. A. Upper panel. R2 compound docked into the FAK-NT protein. Lower panel: Zoomed image. The Blue color shows area of interaction. In the R2 compound, the blue color shows nitrogen and the red-oxygen and grey color shows carbon. The amino-acids of FAK-NT involved in interaction with R2 are shown in blue color. Hydrogen bonds are marked by yellow dashed color are between R2 compound and FAK amino-acids, Asp154 and Arg252. B. The R2 compound directly bound FAK-N-terminal domain by Octet Binding assay. Binding is observed with R2 and FAK-NT, but not with the negative control buffer. C. Immunoprecipitation showed that R2 disrupted binding of FAK and p53 proteins. The immunoprecipitatioon of p53 was performed after treatment of HCT116 cells with R2 at 100 μM for 24 h. Then Western blotting was performed with FAK antibody to detect complex of p53 with FAK. The binding was present in untreated cells, but not in R2-treated cells. Plus (+) marked immunoprecipitation; and minus (−) marked no immunoprecipitation. D. Pull-down assay demonstrated that R2 disrupted FAK and p53 complex. Left panel: Recombinant proteins andbaculoviral FAK (marked by arrows). Right panel: Pull-down assay with recombinant GST-p53 and FAK protein demonstrated binding of FAK and p53 proteins. The R2 compound disrupted the binding of FAK and p53 proteins. Upper panel: Western blotting with FAK antibody. Lower panel: Western blotting with GST antibody. E. R2 disrupted the binding of FAK and p53 proteins in a dose-dependent manner, while a negative control compound (A18), which was not targeting FAK-p53 interaction did not. The pull-down assay was performed as in Figure 3D with 1 and 10 μM of R2 and with 10 μM of A18 (negative control).

Figure 4

R2 increased and reactivated p53 transcriptional activity that is inhibited by FAK. A. Reactivation of p53 activity with p21 target by R2. The dual luciferase assay was performed in HCT116 p53−/− cells co-transfected with p53 and p21 promoter either without FAK plasmid without R2 or with 25 microM R2 treatment or with FAK plasmid without and with R2 treatment. The dual luciferase assay was performed as described in Materials and Methods. R2 compound reactivated p53 activity with p21 target inhibited by FAK. B. Reactivation of p53 activity with Mdm-2 target. The same assay as in Figure 4 A was performed with Mdm-2 promoter. R2 reactivated p53 activity with Mdm-2 target that was inhibited by FAK. C. Reactivation of p53 activity with Bax target. The same assay as in Figure 4A, B was performed with Bax promoter. R2 compound re-activated p53 activity with Bax target inhibited by FAK. *p<0.05, p53 activity with FAK versus no FAK, no R2 treatment, Student’s t-test.

Figure 5

R2 induced expression of p53 targets. A. Induction of p53 targets in HCT116 and MCF-7 cells. The HCT116 p53⁺/⁺ cells (left panel) and MCF-7 (right panel) were treated with different doses of R2 and Western blotting was performed with p53, Mdm-2, Bax, PARP-1 and caspase-8 antibodies. R2 induced expression of p53 targets in a dose-dependent manner in HCT116 and MCF-7 cells. The affected proteins by R2 are shown by arrows. The densitometry quantitation was performed with Scion Image software. The protein level was measured and expressed relatively for the beta-actin control, and then normalized to untreated sample, which was equal to one. B. Immunostaining demonstrated that R2 activates p21 and increased nuclear localization of p53 and p21 proteins in HCT116 p53⁺/⁺ cells, but not in p53^-/^- cells. Immunostaining with primary p21 (upper panel) or p53 (lower panel) and with secondary Texas-Red conjugated antibodies was performed on HCT116 p53⁺/⁺ and p53^-/^- cells either untreated or treated with R2. The Phalloidin-FITC stained actin was used to observe cell morphology. R2 increased nuclear p53 and p21 in HCT116p53⁺/⁺ cells treated with R2 in contrast to HCTp53^-/^- cells. C. R2 increased G1 arrest in R2-treated cells. Flow Cytometry analysis was performed as described in Materials and Methods on HCT116 p53⁺/⁺ and p53−/− cells that were either untreated or treated with different doses of R2 for 24 h. R2 increased G1-arrested cells and decreased G-2 arrested cells in p53⁺/⁺ cells but not p53^-/^- cells.

Figure 6

R2 significantly decreased tumor growth and activated p21 in HCT116 p53⁺/⁺ but not in HCT116 53^-/^- tumor xenografts in vivo. A. Upper panel: HCT116 p53⁺/⁺ and HCT116 p53^-/^- cells were injected subcutaneously into the right and left leg flanks respectively. The control untreated mice were injected subcutaneously with 1xPBS. The treated group of mice was injected subcutaneously with 60 mg/kg of R2. In the case of R2-treated HCT116p53⁺/⁺ xenografts tumor volume decreased significantly (Student’s t-test, p<0.05, marked by asterisk), while HCT116p53^-/^- xenograft tumor volume was not decreased in p53^-/^- xenografts. Lower panel: R2 caused activation of p21 in HCT116p53⁺/⁺ tumors, but not in HCT116p53^-/^- xenograft tumors. R2 increased p21 expression and activated caspase-3 in HCT116 p53⁺/⁺ xenografts but not in HCT116 p53^-/^- xenografts. B. R2 disrupted FAK and p53 complex in HCT116p53⁺/⁺ xenografts. R2 disrupted FAK and p53 complex in HCT116 p53⁺/⁺ xenografts. We immunoprecipitated p53 in tumor xenograft samples and performed Western blotting with FAK antibody in untreated and R2-treated tumor xenografts. The complex of FAK and p53 was present in untreated xenografts, while the complex was not detected in R2-treated xenografts. Two representative tumors are shown for each group.

Figure 7

R2 sensitized HCT116 cells to doxorubicin or 5 fluorouracil treatments. A. MTT assay was performed on HCT116p53⁺/⁺ cells (upper panel) or HCT116p53^-/^- (lower panel) either with doxorubicin alone (0.5 μg/ml), with different doses of R2 alone (−dox) or with combination of doxorubicin and R2 together (+dox). The combination of R2 and doxorubicin decreased colon cancer viability in a p53-dependent manner more effectively than each inhibitor alone in HCT116p53⁺/⁺ cells (upper panel), but not in HCT116 p53^-/^- cells (left lower panel). * p<0.05, Student’s t-test R2 plus doxorubicin versus R2 without doxorubicin treatment. B. Western blotting on R2, Doxorubicin and R2 plus Doxorubicin-treated HCT116 cells. Cells were treated for 24 hours either with R2 (1 μM) or Doxorubicin (0.5 μg/ml) or a combination of R2 and Dox. Western blotting demonstrated that R2 increased p21, and Mdm-2 in HCT116p53⁺/⁺ cells and increased p21 was more effective in the case of R2+Dox treatment compared with each agent treatment alone. This effect was not observed in HCT116^-/^- cells. C. Combination of 5-fluoroiracil (5-FU) and R2 increased apoptosis in HCT116 cells more significantly than each agent alone. Treatment of cells with R2 alone (10 μM), 5-FU alone (0.2 mM) or with both inhibitors together was performed on HCT116 p53⁺/⁺ and p53^-/^- cells for 24 h. Apoptosis was analyzed by Flow Cytometry assay. R2 increases apoptosis in HCT116 p53⁺/⁺ cells treated with R2 in combination with 5-FU, but not in p53^-/^- cells. Bars represent the average of apoptosis from two independent experiments ± standard deviations. *p<0.05, Student’s t-test.; R2+5-FU versus Untreated, R2-treated, and 5-FU-treated cells.