A Small-molecule Inhibitor, 5′-O-Tritylthymidine, targets FAK and Mdm-2 Interaction, and Blocks Breast and Colon Tumorigenesis in vivo

Abstract

Focal Adhesion Kinase (FAK) is overexpressed in many types of tumors and plays an important role in survival. We developed a novel approach, targeting FAK-protein interactions by computer modeling and screening of NCI small molecule drug database. In this report we targeted FAK and Mdm-2 protein interaction to decrease tumor growth. By macromolecular modeling we found a model of FAK and Mdm-2 interaction and performed screening of >200,000 small molecule compounds from NCI database with drug-like characteristics, targeting the FAK-Mdm-2 interaction. We identified 5′-O-Tritylthymidine, called M13 compound that significantly decreased viability in different cancer cells. M13 was docked into the pocket of FAK and Mdm-2 interaction and was directly bound to the FAK-N terminal domain by ForteBio Octet assay. In addition, M13 compound affected FAK and Mdm-2 levels and decreased complex of FAK and Mdm-2 proteins in breast and colon cancer cells. M13 re-activated p53 activity inhibited by FAK with Mdm-2 promoter. M13 decreased viability, clonogenicity, increased detachment and apoptosis in a dose-dependent manner in BT474 breast and in HCT116 colon cancer cells in vitro. M13 decreased FAK, activated p53 and caspase-8 in both cell lines. In addition, M13 decreased breast and colon tumor growth in vivo. M13 activated p53 and decreased FAK in tumor samples consistent with decreased tumor growth. The data demonstrate a novel approach for targeting FAK and Mdm-2 protein interaction, provide a model of FAK and Mdm-2 interaction, identify M13 compound targeting this interaction and decreasing tumor growth that is critical for future targeted therapeutics.

INTRODUCTION

Focal adhesion kinase (FAK) is a 125 kDa non-receptor tyrosine kinase that is localized at focal adhesions, the contact sites between cells and extracellular matrix [1]. FAK becomes tyrosine phosphorylated in response to a number of stimuli, such as mitogen agents, integrin clustering or growth factor receptor signaling [2]. FAK is involved in the regulation of different cellular processes, such as proliferation, survival, spreading, adhesion, motility and angiogenesis [3].

FAK protein structure includes three major domains [3]: the N-terminal (FERM) domain with a primary tyrosine-397 autophosphorylation site, directly interacting with the SH2 domain of Src and with PI-3 kinase; a central kinase domain with Tyr-576/577 sites, two major sites of phosphorylation; and a C-terminal domain with two proline-rich segments and a focal adhesion targeting subdomain (FAT) that binds paxillin [4], talin [5], VEGFR-3 [6] and other proteins [3]. The N-terminal domain of FAK associates with the growth factor receprors, such as EGFR, PDGFR, and c-Met receptors [7–9], and with the death-receptor complex binding protein, RIP [10].

Recently we have demonstrated a direct interaction of p53 with the N-terminal domain of FAK [11]. We performed phage display and mapping analysis and showed that the FAK-N-terminal domain binds to the N-terminal domain of p53 (from 1 to 92 a.a) [11]. Then we identified the 7 amino-acid region (65–71 a.a.) of the N-terminal proline-rich region of p53 that binds to the N-terminal domain of FAK [12]. Recently, the direct interaction of the N-terminal FERM domain of FAK and p53 proteins has been shown and also FAK interaction with Mdm-2 protein has been demonstrated to down-regulate p53 [13]. We proposed a model that FAK sequesters p53 from pro-apoptotic signaling and that FAK and p53 can shuttle between the cytoplasm and nucleus to mediate survival signaling in cells [14], [15, 16]. In addition, recent review demonstrated that FERM domains of other proteins in addition to FAK contain nuclear localization (NLS) and nuclear export signals (NES) that can mediate signal transfer from the membrane to the nucleus suggesting a general mechanism in regulating cross-talk between cell border, cytoplasm and the nucleus [17].

In this report, we performed computer modeling approach to model novel FAK and Mdm-2 interaction, based on the crystal structures of both proteins from the NCBI protein database [18] and by computer macromolecular modeling we found the optimal complex of FAK and Mdm-2 proteins. In addition, we performed immunoprecipitation and demonstrate a complex of FAK and Mdm-2 in breast BT474 cancer and colon HCT116 cancer cells. We performed in silico screening of NCI database of >200,000 small molecule compounds to dock them into the FAK-Mdm-2 complex and identified 24 optimal compounds, called M-compounds, that target this complex. We performed MTT assay on many cancer cell lines (breast, pancreatic, colon and melanoma) and found that 5′-O-Tritylthymidine (called M13 compound) decreased maximally viability in several cancer cell lines. We found that M13 increased detachment and apoptosis in a dose-dependent manner in BT474 breast and HCT116 colon cancer cells, accompanied with down-regulation of FAK, activation of p53 and caspase-8 proteins. Moreover, M13 was able to bind to FAK-NT protein, affected Mdm-2 and FAK protein levels, resulting in decreased complex of FAK and Mdm-2 in BT474 breast and HCT116 colon cancer cells. Moreover, M13 re-activated p53 activity, blocked by FAK in a dual-luciferase assay with Mdm-2 promoter construct. The M13 compound decreased tumor growth in BT474 and HCT116 colon xenografts in vivo. The data show that targeting FAK and Mdm-2 protein interaction can be a novel approach in therapy.

MATERIALS AND METHODS

Cell Lines

BT474 breast carcinoma cells were maintained in RPMI1640 medium supplemented with 10% fetal bovine serum (FBS), 5 µg/ml insulin, and 1 µg/ml penicillin/streptomycin. The MCF-7 cell line was obtained from ATCC and maintained according to the manufacturer’s protocol. HCT116p53^−/− and HCT116p53^+/+ colon cancer cells were maintained in McCoy’s5A medium with 10% FBS.

Antibodies

Anti-FAK monoclonal FAK (4.47) antibody was purchase from Upstate Biotechnology, phospho-Y397-FAK was from Biosource Inc. Monoclonal anti-β-actin was obtained from Sigma. Mdm-2 antibody and caspase-8 antibodies were obtained from Cell Signaling. Anti-p53 antibody, (Ab-6, clone DO-1) was obtained from Oncogen Research Inc.

Proteins

The FAK-NT (1–422 aa) sequence was generated by PCR and subcloned into the pET200 vector (Invitrogen). The His-tagged FAK-NT protein was isolated according to the instructions with the Ni-NTA Purification system kit (Invitrogen). Protein was purified, checked by Coomasie staining and Western blotting with FAK 4.47 antibody and used for the Octet Binding assay.

Macromolecular Computer Modeling and Docking of Small Molecule Compounds

We used a structure-based approach combining macromolecular docking of protein-protein interaction, molecular docking of small molecule compounds with functional testing. First, the crystal structure of FAK, N-terminal FERM domain (PDB ID:2AL6), reported in [18] and NMR and crystal structures of Mdm-2 protein from the Protein Database (NCBI) were used for macromolecular docking and computer modeling of the interaction. To model the FAK-NT-Mdm-2 interaction, the DOT software (http://www.sdsc.edu/CCMS/ DOT/) was used that analyzed >10,000 possible orientations of this interaction, based on the scores of the resulting interfaces using electrostatics, van der Waals, and desolvation energies. The model with the highest scoring of FAK-NT and Mdm-2 interaction has been generated that included primarily amino-acids from F3 lobe (254–352 aa), reported recently to interact with FAK [13]. Then more than 2000,000 small-molecule inhibitors following the Lipinski rules were docked into the pocket of the N-terminal domain of FAK and Mdm-2 interaction in 100 different orientations using DOCK5.1 program. The spheres describing the target pocket of FAK-Mdm-2 were created using DOCK 5.1 suite program SPHGEN. Docking calculations were performed on the University of Florida High Performance Computing supercomputing cluster using16 processors (http://hpc.ufl.edu).

Docking Calculations

The docking calculations were performed with the DOCK 5.1. program, using a clique-matching algorithm to orient small molecule structures with sets of spheres that target the FAK-Mdm-2 interaction. Orientations were optimized using a simplex minimization algorithm, 100 orientations were created for each small molecule in the target site that were scored using the program grid-based scoring function. The hydrogen atoms and partial charges files were created by the SYBDB program.

Small-Molecule Compounds

The top compounds that were detected by the DOCK5.1 program to best fit into FAK-Mdm-2 pocket were ordered from the NCI/DTP database free of charge. Each compound was solubilized in water at concentration of 25 mM. The The M13 compound (purity 98%) was ordered from Sigma for biochemical analyses in vitro and for in vivo mice studies.

Octet RED Binding

The binding was performed by ForteBio Inc. company (www.fortebio.com). The human FAK-N-terminal domain protein was biotinylated using NHS-PEO4-biotin (Pierce). Super-streptavidin (SSA) biosensors (FortéBio Inc., Menlo Park, CA) were coated in a solution containing 1 µM of biotinylated protein for 4 hours at 25°C. A duplicate set of sensors were incubated in an assay buffer (1× kinetics buffer of ForteBio Inc.) with 5% DMSO without protein for use as a background binding control. Both sets of sensors were blocked with a solution of 10 mg/ml Biocytin for 5 minutes at 25°C. A negative control of 5% DMSO was also used. Binding of compound samples (500 µM) to coated and uncoated reference sensors was measured over 120 seconds. Data analysis on the FortéBio Octet RED instrument was performed using a double reference subtraction (sample and sensor references) in the FortéBio data analysis software. The analysis accounts for non-specific binding, background, and signal drift and minimizes well based and sensor variability.

Cell Viability Assay

The cells were plated on a a 96 well plate and were treated with the small molecule compounds at different concentrations for 24 hours. The 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium compound from Promega Viability kit (Madison, IL) was added, and the cells were incubated at 37C for 1–2 hours. The optical density on 96-plate was analyzed at 490 nm to determine cell viability.

Dual Luciferase Assay

For dual luciferase assay, 2×10⁵ cells were plated on 6-well plates, cultured overnight, and co-transfected with the Mdm-2 promoter in the pGL2 or pGL3-luciferase containing plasmids (1µg/well) and pPRL-TK plasmid, containing the herpes simplex virus thymidine kinase promoter encoding Renilla luciferase (0.1 µg/well)using Lipofectamine (Invitrogen) transfection agent according to the manufacture’s protocol. The pRL-TK control encoding Renilla luciferase vector was used for normalization of luciferase activity due to constitutive expression in a variety of cell types (Promega). For all experiments, cells were cultured for 48 hours after transfection and lysed with the 1× Passive Lysis Buffer (Promega). Lysates were analyzed using a Dual-Luciferase Reporter Assay System kit (Promega). Luminescence was measured on the luminometer. HCT116 p53^−/− cells were co-transfected with the above plasmids and p53 in the presence or absence of FAK plasmids and tested either without or with the addition of the small molecule compound added at 25 microM for 24 h. All experiments were performed at least three times.

Western Blotting

Cells or homogenized tumor samples were washed twice with cold 1×PBS and lysed on ice for 30 minutes in a buffer containing: 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Triton-X, 0.5% NaDOC, 0.1% SDS, 5mM EDTA, 50 mM NaF, 1mM NaVO3, 10% glycerol and protease inhibitors: 10 µg/ml leupeptin, 10 µg/ml PMSF and 1 µg/ml aprotinin. The lysates were cleared by centrifugation at 10,000 rpm for 30 minutes at 4°C. Protein concentrations were determined using a Bio-Rad Kit. The boiled samples were loaded on Ready SDS-10% PAGE gels (Bio Rad, Inc) and used for Western blot analysis with the protein-specific antibody. Immunoblots were developed with chemiluminescence Renaissance reagent (NEN Life Science Products, Inc). For quantification, densitometry of protein bands was performed with NIH Scion Image software.

Immunoprecipitation

Immunoprecipitation was performed according to the standard protocol. In brief, the pre-cleared lysates with equal amount of protein were incubated with 1 µg of primary antibody and 30 µl A/G agarose beads overnight at 4°C. The precipitates were washed with lysis buffer three times and re-suspended in 2×Laemmli buffer. The boiled samples were used for Western blotting, as described above.

Detachment Assay

Cells were plated with and without inhibitors for 24 hours, and detached and attached cells were counted in a hemocytometer. We calculated the percent of detachment by dividing the number of detached cells by the total number of cells.

Clonogenicity Assay

The 500–1000 cells were plated on 6 well plates for 1–2 weeks. Then cells were fixed in 25% methanol and stained with Crystal Violet and colonies were visualized and counted.

Apoptosis Assay

Detached cells were collected and fixed in 3.7% formaldehyde in 1×PBS solution for the apoptosis assay. Detection of apoptosis was done with Hoechst 33342 staining. The percent of apoptotic cells was calculated as a ratio of apoptotic detached cells divided by the total number of cells in three independent experiments in several fields with the fluorescent microscope. For each experiment 300 cells per treatment were counted.

Tumor Growth in Nude Mice in vivo

Female nude mice, 6 weeks old, were purchased from Harlan Laboratory. The mice were maintained in the animal facility and all experiments were performed in compliance with NIH animaluse guidelines and IACUC protocol approved by the Animal Care Committee. BT474 cells (2×10⁶ cells/injection) were injected subcutaneously into the mice. The HCT116p53^−/− and HCT116p53^+/+ cells were injected subcutaneously into the left and right side of the same mice to decrease variations. In preliminary experiment different doses of the compound were introduced into the mice, and 30–50 mg/kg was chosen as optimal, non-toxic doses. The day after injection, the M13 compound was introduced by IP injection daily 5 days/week for 3 weeks. At the end of experiment, tumor weight and volume was determined.

Statistical Analyses

Student’s t test was performed to determine significance. The difference between treated and untreated samples with P<0.05 was considered significant.

RESULTS

A Model of FAK-N-terminal Domain and Mdm-2 Interaction and Small-Molecule Inhibitors Targeting this Interaction are Generated

The model with the highest scoring of the FAK N-terminal domain, FAK-NT and Mdm-2 macromolecular interaction has been created that included primarily amino-acids from F3 lobe (254–352 aa), reported recently to interact with FAK [13]. The model of the FAK-N-terminal FERM domain and Mdm-2 interaction is shown on Fig. (1A). The amino-acids of FAK and Mdm-2 proteins participating in the formation of hydrogen bonds and salt bridges were determined as described in Materials and methods and are presented in Table 1.

Fig. (1).

A. Macromolecular computer modeling defined the complex of FAK-NT and Mdm-2. FAK-NT protein is on the left, Mdm-2 protein on the right, shown by arrows. B. Zoomed image of FAK N-terminal (FERM) domain, reported in [18] with one of the compounds, targeting FAK-Mdm-2 interaction.

Table 1.

The Amino-Acids of FAK-NT Domain and Mdm-2 Proteins that Participate in Hydrogen Bonds and Salt Bridges.

Amino-acids Forming Hydrogen Bonds
Number	Mdm-2	Atom	FAK	Atom
1	CYS2	[N]	TYR282	[OH]
2	MET6	[N]	THR284	[O]
3	THR16	[N]	ASP285	[OD2]
4	THR16	[OG1]	ASP285	[OD1]
5	THR16	[OG1]	THR284	[OG1]
6	SER22	[OG]	GLY287	[O]
7	THR49	[N]	GLU277	[OE2]
8	THR49	[OG1]	GLU277	[OE2]
9	LYS 51	[NZ]	ASP 254	[OD1]
10	SER 115	[OG]	ARG 229	[O]
11	ASP 46	[OD2]	GLN 136	[N]
12	ASP 117	[OD2]	ARG 229	[NE]
13	GLU25	[OE1]	ARG252	[NH1]
14	ILE19	[O]	ASN289	[ND2]
15	VAL41	[O]	ASN297	[ND2]

Amino-acids Forming Salt Bridges
Number	Mdm-2	Atom	FAK	Atom
1	LYS 51	[NZ]	ASP 254	[OD1]
2	ASP 117	[OD2]	ARG 229	[NE]
3	ASP 117	[OD1]	ARG 229	[NE]
4	ASP 117	[OD2]	ARG 229	[NH2]
5	GLU 25	[OE1]	ARG 252	[NH1]
6	GLU 25	[OE2]	ARG 252	[NH1]

Then more than 200,000 small-molecule inhibitors from NCI (National Cancer Institute) database following the Lipinski rules were docked into the pocket of the N-terminal domain of FAK and Mdm-2 interaction in 100 different orientations using DOCK5.1 program. The 24 compounds (called M-compounds) with the highest scores were ordered from NCI database. The structures of M compounds are shown in Table 2. The example of one of the compounds, docked in the FAK-NT pocket involved in interaction with Mdm-2 is shown in Fig. (1B).

Table 2.

Top Scoring Compounds Targeting FAK and Mdm-2 Complex

Compound No	Compound Label	NSC No	Formula	Molecular Weight	Structure
1	M1	21371	C8H10O8	234
2	M2	32237	C14H18BrN4O	338
3	M3	35024	C14H19IN5O	400
4	M4	35450	C20H23N4O	335
5	M5	618449	C20H16ClN5O5	440
6	M6	45202	C18H16N4O7S2	464
7	M7	45588	C21H15N5O5S	449
8	M8	46713	C10H15N2O8P	322
9	M9	47065	C12H9ClN4O6	341
10	M10	47096	C15H22N2O5	310
11	M11	51135	C12H14O4	222
12	M12	72381	C26H23ClN6O2	487
13	M13	75113	C29H28N2O5	484
14	M14	76150	C25H43N3O5	466
15	M15	89296	C4H10NO6P	199
16	M16	117523	C16H18N4O4S	362
17	M17	134514	C10H19ClN4O4	295
18	M18	159456	C19H14O6	338
19	M19	305483	C22H20N2O8	440
20	M20	310777	C17H23N3O3	317
21	M21	332639	C8H12N2O7	248
22	M22	344553	C12H16N3O7PS	377
23	M23	351358	C20H32N6O6	452
24	M24	617076	C20H20N4O4	380

The Small Molecule Compound: 5′-O-Tritylthymidine or M13 Compound Significantly Decreased Viability of Most Cancer Cells in vitro

We performed MTT assay with 24 compounds (called M compounds) and detected that among all tested compounds, M13 compound was the most effective in decreasing viability of different cancer cells, including breast, melanoma, colon and pancreatic cancers. Representative viability assay on breast MCF-7 cancer cells is shown on (Fig. (2A), upper panel) and on BT474 breast cancer cell line (Fig. 2A, lower panel). Several other compounds, such as M24 also efficiently decreased viability in cancer cells (Fig. 2A). Since M13 compound had the highest inhibiting effect on viability in many cancer cells and was commercially available, we ordered this compound and used for the study.

Fig. (2).

A. The effect of compounds targeting FAK and Mdm-2 interaction on viability of breast cancer cells. The 24 compounds, called M-compounds were added to the cancer cells for 24 h at 100 µM dose and MTT viability assay was performed, as described in Materials and Methods. Upper panel: MCF-7 breast cancer cells; Lower panel: BT474 breast cancer cells. Bars show averages +/− standard deviations of three independent experiments. Compound M13 decreased maximally cancer cell viability in many cancer cell lines tested. 2 B, C. Structure of M13 that targets FAK and Mdm-2 and binds FAK-NT protein. B. Chemical name and structure are shown. C. Upper panel: Close-up view of M13 compound in the pocket of FAK and Mdm-2 interaction. Lower panel: Zoomed image of boxed area from the upper panel. D. Binding of M13 to FAK-NT domain by label-free ForteBio Octet 384 System assay. The binding curve of M13 compound (500 microM) with biotynilated FAK-NT protein was obtained by ForteBio OctetRED384 system (Materials and Methods) using protein biosensors. The real-time kinetic characterization of binding of FAK-NT protein with M13 compound and with negative control buffer is shown. X-axis, time in seconds; Y-axis, binding in nm. M13 directly binds FAK-NT protein. The kinetic dissociation constant, KD of binding was equal to 1.57 mM.

M13 Compound Targets FAK and Mdm-2 Interaction and Directly Binds the N-terminal Domain of FAK

The chemical name of the M13 compound (NCI number: 75113) is 5′-O-Tritylthymidine, chemical formula is C29H28N2O5. The chemical structure of M13 is shown on Fig. (2B). The M13 compound was effectively docked into the pocket of FAK and Mdm-2 interaction and is shown on Fig. (2C), upper panel and zoomed in lower panel.

To detect direct binding of the M13 compound to the N-terminal domain of FAK, we isolated human N-terminal domain of FAK and performed Real-time binding assay with M13 compound using label-free biomolecular binding ForteBioOctet Red384 system (Materials and Methods) (Fig. 2D). The Octet assay demonstrates that M13 directly binds to the FAK-NT protein, but not to the buffer control (Fig. 2D). Thus, M13 targets FAK-NT and Mdm-2 interaction and directly binds to the FAK-NT protein.

M13 Compound Decreased Viability of BT474 Breast Cancer Cells, and Increased Detachment and Apoptosis in a Dose-dependent Manner in BT474 Cancer Cells

To study dose-dependent effect of M13 compound on viability and clonogenicity we performed MTT assay with BT474 cells. We showed that M13 significantly decreased viability in the cells at doses > 10 µM (Fig. 3A). M13 also increased detachment (Fig. 3B, left panel) and apoptosis in a dose-dependent manner (Fig. 3B, right panel). The apoptotic Hoechst stained nuclei of M13-treated BT474 cells are shown on Fig. (3C). Thus, M13 decreased viability and increased detachment and apoptosis in a dose-dependent manner in BT474 cells.

Fig. (3). M13 compound inhibits viability, induces detachment and apoptosis in BT474 breast cancer cells.

A. Breast cancer cells were treated with different doses of M13 compound for 24 h, and MTT assay was performed to test the effect on cell viability. Bars show averages +/− standard deviations. B. Left panel: Compound M13 causes dose-dependent detachment in BT474 cells. BT474 cells were treated with different doses of M13 compound and detachment was determined on a hemacytometer, as described in Materials and Methods. Compound M13 significantly decreased cell detachment. Right panel: Apoptosis was analysed by Hoechst staining in BT474 cells treated with different doses of M13. The M13 compound increased apoptosis in a dose-dependent manner in BT474 cells. Bars show averages +/− standard deviations of three independent experiments. * P< 0.05 versus untreated cells, Student’s t-test. C. Hoechst staining of M13-treated BT474 cells. BT474 cells were treated as described above, and nuclei were stained with Hoechst. Apoptotic nuclei stained with Hoechst in BT474 cells treated with 50 µM M13 are shown with arrows.

M13 Compound Increased Mdm-2 Protein Level, Decreased FAK and Activated Caspase-8 in BT474 Cells

To demonstrate the effect of M13 on FAK, Mdm-2 and p53 protein expression, we treated BT474 cells with different doses of M13 and performed Western blotting with anti-FAK, Y397-FAK, Mdm-2, p53 and caspase-8 antibodies (Fig. 4). At 10 µM dose of M13 increased p53 levels that resulted in increased Mdm-2 levels. At high doses (50–100 µM) M13 decreased FAK and Y397-FAK and activated caspase-8 that is consistent with decreased viability and increased detachment and apoptosis, demonstrated in Fig. (3). Thus, M13 increased apoptosis in BT474 cells, consistent with up-regulated activity of p53, increased apoptotic Mdm-2, down-regulation of FAK and activation of caspase-8.

Fig. (4). M13 treatment down-regulates FAK, activates p53, up-regulates Mdm-2 and activates caspase-8 in BT474 cells.

BT474 cells were treated with different doses of M13 compound, and Western blotting was performed with Y397-FAK antibody, main autophosphorylation site of FAK, with total FAK antibody, Mdm-2, p53, caspase-8 to detect level of these proteins. Western blotting with β-Actin was used for control of equal loading. At high doses M13 compound decreased FAK, activated Mdm-2 and caspase-8 proteins.

FAK and Mdm-2 Proteins Interact in BT474 Breast and HCT116 Colon Cancer Cells and M13 Compound Affects FAK and Mdm-2 and Decreases Its Complex in Both Cell Lines

To detect that M13 affect FAK and Mdm-2 protein complex, we immunoprecipitated FAK in BT474 cells and performed Western blotting with Mdm-2 antibody. We demonstrate a complex of FAK with Mdm-2 proteins in BT474 cells, while no complex is present in control immunoprecipitation without antibody (Fig. 5). When we treated BT474 cells with 100 µM dose of M13, Mdm-2 level was increased, but FAK protein level was significantly decreased that resulted in decreased complex of FAK and Mdm-2 proteins (Fig. 5A). Thus, FAK and Mdm-2 proteins interacted in BT474 cells and M13 caused decreased FAK and Mdm-2 complex. The same result was observed in HCT116 colon cancer cells. We performed immunoprecipitation of FAK and performed Western blotting with Mdm-2 antibody (Fig. 5B). We detected a complex of FAK and Mdm-2 in HCT116 colon cancer cells (Fig. 5B), and at high dose (100 µM) M13 caused significantly decreased FAK and Mdm-2 proteins and its complex formation (Fig. 5B). Thus, FAK and Mdm-2 interacted in BT474 breast and in HCT116 colon cancer cells and high dose of M13 decreased FAK and Mdm-2 protein and its binding in both cell lines.

Fig. (5). The effect of M13 on FAK and Mdm-2 protein complex in breast BT474 and colon HCT116 cancer cells.

A. Immunoprecipitation of FAK with anti-FAK antibody and Western blotting with Mdm-2 antibody detected a complex of FAK and Mdm-2 in BT474 breast cancer cells. Left panels: No precipitation, Right panel: IP:FAK, immunoprecipitatiopn with FAK. M13 increased level of Mdm-2 and decreased FAK protein level at 100 µM dose that resulted in decreased complex of FAK and Mdm-2 proteins. B. Immunoprecipitation of Mdm-2 and Western with FAK antibody detected a complex of FAK and Mdm-2 in HCT116 colon cancer cells. At 100 µM, M13 decreased both FAK and Mdm-2 and no complex of FAK and Mdm-2 proteins was detected.

M13 Compound Decreased Clonogenicity of HCT116 Cells in a p53-dependent Manner, Re-activated p53 Luciferase Activity with Mdm-2 Promoter and Increased Detachment and Apoptosis in HCT116 Cells

To test the in vitro effect of M13 on HCT116 cells, we performed clonogenicity assay with HCT116 p53^+/+ and HCT116 p53^−/− cells (Fig. 6A). M13 decreased clonogenicity of HCT116 cells in a dose-dependent manner and it decreased clonogenicity more significantly (p<0.05) in HCT116p53^+/+ than in HCT116p53^{− /−} cells (Fig. 6A) Thus, the effect of M13 on the clonogenicity in HCT116 cells is p53-dependent.

Fig. (6). M13 decreased clonogencity, activated p53 activity with Mdm-2 promoter target, increased detachment and apoptosis and activated caspase-8 in HCT116 colon cancer cells.

A. Clonogenicity assay with untreated and M13-treated at 1 and 10 µM dose was performed on HCT116p53^+/+ and HCT116p53^−/− cells. M13 decreased clonogencity in a dose-dependent manner more effectively in HCT116 p53^+/+ than in p53^−/− cells. The difference in M13- treated cells between p53^−/− and p53^+/+ cells at 10 µM was significant in 3 independent experiments, p<0.05. B. M13 re-activated p53 transcriptional activity in a dual-luciferase assay with Mdm-2 promoter construct. HCT116 p53−/− cells were co-transfected with p53, FAK, and Mdm-2 promoter plasmids, and dual-luciferase assay was performed in the presence or absence of M13 (25 microM). FAK inhibited activity of p53 with Mdm-2 promoter construct, while M13 re-activated the transcriptional activity of p53. The experiment was repeated 3 times. P<0.05, FAK-dependent decrease of p53 activity. The reactivation of p53 increased in a dose-dependent manner (not shown). C. Detachment and apoptosis in M13-treated HCT116 cells. M13 increased detachment and apoptosis in HCT116 cells in a dose-dependent manner. D. At high doses M13 decreased Y397-FAK, FAK and activated p53 and caspase-8 in HCT116 cells. Western blotting was performed in M13-treated cells with Y397-FAK, FAK, Mdm-2, p53 and caspase-8 antibodies. Western blotting with Beta-Actin antibody used as an equal loading control.

We have shown that FAK inhibited p53 transcriptional activity through direct binding of the N-terminal domain of FAK with p53 protein [19]. We co-transfected p53 vector with the FAK plasmid into HCT116 p53^−/− cells with Mdm-2 promoter target and performed dual-luciferase assay either without M13 or with M13 small molecular compound (Fig. 6B). M13 increased Mdm-2 promoter activity in a dose-dependent manner (not shown). We show that FAK inhibited p53 transcriptional activity with the Mdm-2 promoter, while M13 compound re-activated p53 activity. Thus, M13 decreases clonogenicity of HCT116 p53 cells in a p53-dependent manner and re-activates p53 activity with the Mdm-2 promoter.

In addition, M13 efficiently increased detachment and apoptosis in HCT116 colon cancer cells in a dose-dependent manner (Fig. 6C), resulting in increased level of p53 and decreased Y397-FAK and total FAK at high dose. At 50 µM dose, M13 activated of caspase-8, decreased uncleaved form of caspase-8 (Fig. 6D) consistent with increased detachment and apoptosis of these cells. The M13 did not decrease Y397-FAK and did not activate caspase-8 at 50 µM dose in HCT116p53−/− cells (not shown) that supports p53-dependent decrease in clonogenicity.

M13 Compound Significantly Decreased Breast and Colon Tumor Growth in vivo

To detect the in vivo effect of M13 on tumorigenesis in vivo, we injected breast cancer BT474 cells into the mice to generate xenograft tumors. M13 was used at 30 mg/kg and 50 mg/kg doses. M13 effectively decreased breast tumor growth in vivo in a dose-dependent manner. M13 significantly decreased BT474 breast tumorigenesis in vivo at 30–50 mg/kg (Fig. 7A).

Fig. (7). M13 decreased BT474 breast and HCT116 colon tumor growth in vivo.

A. M13 decreased breast tumor growth in a dose-dependent manner.BT474 cells were subcutaneously injected into mice, and M13 at doses 30 mg/kg and 50 mg/kg were added daily 5 days per week (n=5). Five control mice were injected with vehicle alone and used as a control group (marked as untreated with M13, untreated group). Tumor volume was measured with calipers, as described in Materials and Methods. Bars show averages +/− standard deviations. M13 compound significantly decreased BT 474 tumor volume in a dose-dependent manner. * p< 0.05, Student’s t-test. B. M13 significantly decreases HCT116 colon tumor growth in vivo. Left panel: HCT116 p53^+/+ cells were subcutaneously injected into mice, and M13 at doses 40 mg/kg were added daily 5 days per week. Five untreated mice were used as a control group. Tumor volume was measured with calipers, as described in Materials and Methods. Bars show averages +/− standard deviations. M13 compound significantly decreased HCT116 tumor volume. * p< 0.05, Student’s t-test. Right panel: FAK is down-regulated and p53 is up-regulated in tumors of M13-treated mice. Cell lysates were prepared from tumors of untreated and M13-treated mice. Western blotting was performed with FAK and p53 antibody. M13 activated p53 and decreased FAK in tumor samples. Western blotting with β-Actin was used for control. Two tumors from each group are shown: T1, T2, tumor one and tumor two.

To test the effect of M13 on colon tumor growth, we injected HCTp53^+/+ and HCT116p53^−/− into the left and right sides of the same mice and treated with M13 compound. The M13 compound (40 mg/kg) significantly decreased HCT116 colon tumor growth in vivo (Fig. 7B, left panel), while the effect of M13 on HCT116p53^−/− tumors was not significant (not shown). We analyzed tumor tissues of tumors from mice treated with vehicle alone (untreated) and treated with M13 of HCT116 tumors by Western blotting with p53 and FAK antibodies (Fig. 7B, lower panel). M13 increased p53 level and decreased FAK protein levels in tumor samples (Fig. 7B, right panel) that is consistent with increased activity of p53 and increased apoptosis in vitro. Thus, M13 activated p53, down-regulated FAK signaling and effectively decreased breast and colon tumor growth in vivo.

DISCUSSION

Thus, based on the known crystal structures of FAK and Mdm- 2 proteins we demonstrated by macromolecular computer modeling the optimal complex of FAK and Mdm-2. Based on this data, we docked >200,000 NCI compounds in 100 different orientation into the pocket of FAK and Mdm-2 interaction. We found and ordered 24 compounds with highest docking scores from NCI library and screened different cancer cell lines, including colon, breast, pancreatic and melanoma. Among 24 screened compounds, M13 compound that targeted FAK-Mdm-2 interaction and decreased the most cancer cell viability. We found that M13 compound dose-dependently decreased viability, and increased detachment and apoptosis in both BT474 breast cancer cells. Computer molecular docking demonstrated that M13 targeted FAK and Mdm-2 interaction and Octet ForteBio system demonstrated that M13 directly interacted with FAK protein. In addition, M13 increased p53 level and activity, demonstrated by increased Mdm-2 in BT474 cancer cells. Moreover, M13 caused decreased FAK levels in both BT474 and HCT116 colon cancer cells, decreased complex of FAK and Mdm-2 and activated caspase-8 in BT474 cells and HCT116 colon cancer cells. M13 decreased clonogenicity in HCT116 cells in a p53-dependent manner, increased p53 transcriptional activity with Mdm-2 promoter target and increased detachment and apoptosis in HCT116 colon cancer cells in vitro. In addition, M13 decreased breast and colon tumor growth in vivo. In M13-treated tumors p53 protein was increased and FAK was significantly decreased that is consistent with decreased tumorigenesis in vivo.

We identified M13 compound (5′-O-Tritylthymidine) as one of the small molecule compounds that effectively decreased viability of most cancer cell lines. In addition, it induced dose-dependent apoptosis and blocked breast and colon tumor growth in vivo. The data are consistent with report of [20], where authors found that 5’-O-thritylthymidine (called Kin6), involved in angiogenesis and tumor invasiveness, decreased endothelial cell proliferation and viability and angiogenesis. The compound also inhibited dose-dependently endothelial cell migration by unknown mechanism [20]. The authors did not observe significant effect on angiogenesis, while observed effective inhibition of cell proliferation that is consistent with targeting FAK and Mdm-2 interaction, increase of p53 activity and apoptosis. Thus, one of the mechanisms of decreased viability in endothelial cells can be through targeting of FAK and Mdm-2 interaction, resulting in activation of p53 and down-regulation of FAK, a known player of cell motility, metastasis and survival [2, 15]. Our report for the first time demonstrates the effective effect of M13, 5’-O-thritylthymidine, on decreased tumorigenesis in vivo.

The data demonstrate a novel approach for targeting FAK and Mdm-2 protein interaction that can provide future effective therapeutics. In the recent report of Lim et al., [13], the authors demonstrated that FAK and Mdm-2 interaction and showed that Mdm-2 interacted more strong with F 3 lobe (254–352 aa) of FAK than with F1 and F2. It is well known that hydrogen bonds and salt bridges are critical for protein-protein interaction [21]. The macromolecular docking demonstrated the best complex of FAK and Md-2 proteins that included 15 hydrohen bonds. The amino acids that form hydrogen bonds are shown in Table 1. Importantly, 11 out 15 (>73 %) of hydgrogen bonds in FAK-NT domain were from the lobe F3, validating our modeling data, and consistent with the data on higher binding affinity of F3 FAK FERM lobe with Mdm-2 protein [13]. In addition, 3 out of 6 (50%) amino acids forming salt bridges, shown in Table 1 were adjacent to the start of F3 lobe (254 amino acid). The Mdm-2 amino acids that are involved in interaction with FAK were not reported, the macromolecular modeling approach show that 12 of 15 (80%) amino acids forming hydrogen bonds with FAK are located in 1–50 amino acid region of Mdm-2 and all of the amino acids forming hydrogen bonds and salt bridges are located in the area of 1–117 amino acids that is the same area, where p53-binding domain in Mdm-2 protein has been reported [22]. Thus, M13 compound targeting FAK and Mdm-2 complex can also affect binding of Mdm-2 and p53.

We observed increased level of p53 at 10–50 µM doses of M13 in BT474 and HCT116 cancer cells lines, and increased activity of p53 resulting in decreased FAK protein level in both cell lines. It is known that p53 can inhibit FAK promoter activity [14, 23, 24] and decrease FAK protein level [14]. The M13 could affect conformation of FAK-Mdm-2 complex, leading to increased p53 stability that is consistent with data of Lim et al., where authors show that FAK facilitated p53 and Mdm-2 association, leading to p53 ubiquitination and degradation [13]. We observed association of FAK and Mdm-2 proteins that supports data of [13]. In addition, M13 affected both targets FAK and Mdm-2 proteins. M13 increased level of Mdm-2 in BT474 cells, that can be explained by activation of p53 activity, as Mdm-2 is a known transcriptional p53 target, activated by p53. In fact, when we overexpressed FAK in HCT116 p53^−/− cells with p53 and Mdm-2 promoter, we found that M13 re-activated p53 transcriptional activity that was inhibited by FAK. This was not observed in HCT116 colon cancer cells, indicating cell type specific differences. The increased p53 and decreased FAK was enough to cause apoptosis and decrease tumorigenesis in vivo. In fact, HCT116 tumors treated with M13 expressed increased p53 and decrease of FAK.

It is known that Nutlin 1 is the small molecule compound that targets Mdm-2 and p53 interaction [25]. It will be interesting to explore in future combination of Nutlins targeting Mdm-2 and p53 interaction and activating p53 with small molecule compounds targeting FAK and Mdm-2 interaction, demonstrated in this report and FAK and p53-targeting compounds. We also performed modeling of FAK and p53 complex and discovered several P-compounds, targeting and disrupting this interaction that can be important for therapy.

Thus, computer modeling approach with macromolecular modeling can be efficient approach for anti-cancer therapy. In this report, we demonstrate that targeting Mdm-2 and FAK proteins is important for decreased cancer cell viability and induction of apoptosis and inhibition of tumorigenesis in vivo that can be effective for future targeted FAK-Mdm-2-p53 therapy. In addition, this reveal novel functions of FAK and Mdm-2 in the cells that is one of the examples of FAK FERM domain interaction reported in [17].

ABBREVIATIONS

FAK

Focal Adhesion Kinase

Mdm-2

Murine double minute 2