FAK - Oxford Academic - Oxford University Press

Carcinogenesis vol.33 no.5 pp.1004–1013, 2012 doi:10.1093/carcin/bgs120 Advance Access publication March 7, 2012

A small molecule focal adhesion kinase (FAK) inhibitor, targeting Y397 site: 1-(2-hydroxyethyl) -3, 5, 7-triaza-1-azoniatricyclo [3.3.1.13,7]decane; bromide effectively inhibits FAK autophosphorylation activity and decreases cancer cell viability, clonogenicity and tumor growth in vivo Vita M.Golubovskaya, Sheila Figel1, Baotran T.Ho, Christopher P.Johnson, Michael Yemma, Grace Huang, Min Zheng2, Carl Nyberg2, Andrew Magis3, David A.Ostrov3, Irwin H.Gelman1 and William G.Cance Department of Surgical Oncology, 1Department of Cancer Genetics, Roswell Park Cancer Institute, Buffalo, NY 14263, USA, 2Department of Pathology and Laboratory Medicine, UF Shands Cancer Center, University of Florida, Gainesville, FL, USA and 3Department of Pathology and Laboratory Medicine, University of Florida, Gainesville, FL, USA To whom correspondence should be addressed. Tel: þ1 716 845 3920; Fax: þ1 716 845 3820; Email: [email protected] Correspondence may also be addressed to William G.Cance. Tel: þ1 716 845 3434; Fax: þ1 716 845 3434; Email: [email protected]

Focal adhesion kinase (FAK) is a protein tyrosine kinase that is overexpressed in most solid types of tumors and plays an important role in the survival signaling. Recently, we have developed a novel computer modeling combined with a functional assay approach to target the main autophosphorylation site of FAK (Y397). Using these approaches, we identified 1-(2-hydroxyethyl)-3, 5, 7-triaza-1azoniatricyclo [3.3.1.13,7]decane; bromide, called Y11, a small molecule inhibitor targeting Y397 site of FAK. Y11 significantly and specifically decreased FAK autophosphorylation, directly bound to the N-terminal domain of FAK. In addition, Y11 decreased Y397-FAK autophosphorylation, inhibited viability and clonogenicity of colon SW620 and breast BT474 cancer cells and increased detachment and apoptosis in vitro. Moreover, Y11 significantly decreased tumor growth in the colon cancer cell mouse xenograft model. Finally, tumors from the Y11-treated mice demonstrated decreased Y397-FAK autophosphorylation and activation of poly (ADP ribose) polymerase and caspase-3. Thus, targeting the major autophosphorylation site of FAK with Y11 inhibitor is critical for development of cancer therapeutics and carcinogenesis field.

Introduction Focal adhesion kinase (FAK) localizes to focal adhesions and is involved in cellular functions such as motility, adhesion, survival, proliferation, invasion, metastasis and angiogenesis (1). FAK tyrosine Y397 autophosphorylation increases in response to integrin clustering, growth factor and angiogenesis receptor-mediated signaling (2). Tyrosine 397 site is the main autophosphorylation site of FAK leading to activation of its intrinsic kinase function as well as downstream signaling players (3) and providing a high-affinity binding site for the SH2 domain of Src family kinases (4). The interaction of Y397-activated FAK and Src causes tyrosine phosphorylation of multiple sites in FAK (-576, -577, -925), as well as other signaling molecules such as paxillin, p130CAS, leading to cytoskeletal and morphological changes associated with activation of downstream signaling (5). Other proteins such as PI3 kinase, Nck-2, Grb-7 and Shc also bind Y397 site of FAK. Thus, Y397 site is one of the main phosphorylation sites that activate downstream cellular signaling. FAK has been proposed recently to be a target for anticancer therapy (6,7). Several approaches to inhibit FAK have been used, including Abbreviations: FAK, focal adhesion kinase; NCI, National Cancer Institute; PARP, poly (ADP ribose) polymerase.

antisense oligonucleotides (8), dominant-negative C-terminal domain of FAK, FAK-CD (called also FRNK) (9,10) and FAK small interfering RNA (11,12), causing decreased cancer cell viability, growth or apoptosis. Recently, several small molecule inhibitors of FAK that target the ATP-binding site and block FAK kinase activity have been developed and reported—one by Novartis: NVP-TAE226 (13,14), two by Pfizer: PF-573 228 (15) and PF-562 271 (16). TAE226 inhibited glioma and ovarian tumor growth in vivo (14,17), but also inhibited IGF-1R kinase, and has been abandoned due to such off-target effects (14). The efficacy of the PF-573 228 on tumor growth in vivo has not been reported, and it did not inhibit cell growth and survival in vitro (15). PF-573 271 inhibitor blocked activity of FAK and its homologous kinase PYK-2 and decreased tumor growth in mouse xenograft models and has been tested in clinical trials (16,18). Another inhibitor, PND-1186, targeting ATP-binding site has been reported by Walsh et al. (19). All of these inhibitors effectively blocked Y397FAK phosphorylation, but development of these inhibitors has been complicated by the fact that the ATP-binding site shares consensus sequences and structural domains across many different tyrosine kinases, making it less suitable for clinical testing due to off-target effects, as seen with the Novartis and Pfizer inhibitors. Since the Y397 site is a critical site for FAK activation and its survival function, we pioneered a different approach to target the kinase activity of FAK. Using computer modeling, we targeted the Y397 site on the crystal structure of FAK and in silico screened compounds for their ability to bind at this site performed computer modeling approach, as described in (20). This approach allowed us to specifically target Y397 site of FAK and find potential small molecule drugs that inhibited FAK function. We identified the first FAK allosteric inhibitor, 1,2,4,5-benzenetetraamine tetrachloride, called Y15 that targets Y397 site, which specifically decreased Y397 phosphorylation of FAK in vitro, inhibited cancer cell viability in vitro and blocked breast, pancreatic and neuroblastoma tumor growth in vivo (21–23). This report identifies a novel small molecule inhibitor of FAK that targets the Y397 site, 1-(2-hydroxyethyl) -3, 5, 7-triaza-1-azoniatricyclo [3.3.1.13,7]decane; bromide (called Y11) found by this computer modeling screening of .140 000 compounds from the National Cancer Institute (NCI) database of small molecule compounds and combining our findings with functional cellular assays (21). Y11 effectively and specifically blocked autophosphorylation kinase activity of FAK and directly bound to the FAK-N-terminal domain. In addition, Y11 decreased Y397 phosphorylation in breast cancer BT474 and colon cancer SW620 cells and decreased cancer cell viability and clonogenicity. Y11 induced detachment and apoptosis in SW620 cells in a dose-dependent manner and significantly decreased SW620 tumor growth in the mouse xenograft model and demonstrated decreased Y397-FAK phosphorylation compared with untreated control tumors. Thus, Y11 is a novel inhibitor of FAK that can be tested for future FAK-targeted cancer therapeutics. Materials and methods Cell lines The SW620 colon cancer cells were maintained in McCoy’s 5A plus 10% fetal bovine serum medium. BT474 breast carcinoma cells were maintained in RPMI1640 medium supplemented with 10% fetal bovine serum, 5 lg/ml insulin and 1 lg/ml penicillin/streptomycin. The normal human WI38-TERT cells were maintained according to American Tissue Culture Collection protocol. Antibodies Antiphospho-Tyr397-FAK antibody was obtained from Biosource Inc. Anti-FAK (4.47) antibody, caspase-3 and poly (ADP ribose) polymerase (PARP) antibodies were obtained from Upstate Biotechnology, Inc. Monoclonal anti-b-actin antibodies were obtained from Sigma.

Ó The Author 2012. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]

1004

FAK autophosphorylation inhibitor

Fig. 1. Y11 compound is docked into Y397 site of FAK and decreases cancer cell viability. (A) 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyl tetrazolium bromide assay with 35 small molecule compounds, described in (21) in MCF-7 breast cancer cell lines, shows that Y11 effectively decreases colon cancer viability as well as control Y15 inhibitor. (B) The structure, formula and a chemical name of Y11 compound. (C) The docking by computer modeling of Y11 into the Y397 site region of FAK.

Protein isolation The His-tagged FAK-N-terminal domain (1-422 aa) was obtained by PCR and subcloned into the PET200 vector (Invitrogen). The Invitrogen Champion pET Directional TOPO Expression Kit was used for protein expression and purification. The baculoviral FAK protein for in vitro kinase assay was isolated, as described in (24). Structure-based molecular docking of NCI database small molecule compounds The crystal structure of the FAK FERM domain from the Protein Data Bank (25) was used for docking of FAK inhibitors. We used a structure-based approach that included molecular docking with functional testing. More than 140 000 small molecule compounds with drug-like characteristics (following the Lipinski rules) were docked into the N-terminal domain of FAK crystal structure in 100 different three-dimension orientations using DOCK 5.1 program. Small molecule inhibitor compounds The 35 small molecule compounds detected by the DOCK program and that best fit into the Y397 site of FAK were ordered from the NCI, Developmental Therapeutics Program (DTP). The structures of 35 small molecule compounds were described in (21). Each compound was dissolved in dimethyl sulfoxide at concentration of 25 mM and stored at 20°C. The Y11 compound: 1-(2-hydroxyethyl) -3, 5, 7-triaza-1-azoniatricyclo [3.3.1.13,7] decane; bromide was synthesized by Alchem Laboratories Corporation (www. alchem.com). The FAK inhibitor, PF-573 228, was obtained from Pfizer Inc. (15). The FAK inhibitor, TAE226, was obtained from Novartis Inc. (13). The Y15 1,2,4,5-benzenetetraamine tetrachloride inhibitor (purity 98.1%) was obtained from Sigma.

Western blotting Cells or tumor samples were collected, washed with cold 1 phosphate-buffered saline and lysed on ice for 30 min in a buffer containing 150 mM NaCl, 50 mM Tris–HCl (pH 7.5), 0.5% NaDOC, 1% Triton-X, 0.1% sodium dodecyl sulfate, 5 mM ethylenediaminetetraacetic acid, 1 mM NaVO3, 50 mM NaF, 10% glycerol and protease inhibitors: 10 lg/ml phenylmethylsulfonyl fluoride, 10 lg/ml leupeptin and 1 lg/ml aprotinin. The lysates were centrifuged at 11 000 r.p.m. for 30 min at 4°C and protein concentrations were determined using a Bio-Rad Kit. The 20 lg of protein extract was boiled for 5 min and loaded on polyacrylamide gels (Bio-Rad, Inc.). Then transfer was performed to membranes and western blotting was used for analysis according to standard protocol. The western blotting membranes were analyzed with chemiluminescence agent. Octet binding The binding of Y11 and FAK-N-terminal domain was performed by Forte´Bio Inc. company (www.fortebio.com). The human purified His-tagged FAK-N-terminal domain was biotinylated using NHS-PEO4-biotin (Pierce). Superstreptavidin biosensors (Forte´Bio Inc., Menlo Park, CA) were coated in a solution containing 1 lM of biotinylated protein for 4 h at 25°C. A duplicate set of sensors was incubated in an assay buffer (1 kinetics buffer of ForteBio Inc.) with 5% dimethyl sulfoxide 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 min at 25°C. A negative control of 5% dimethyl sulfoxide was also used. Binding of samples (500 lM) to coated and uncoated reference sensors was measured over 120 s. Data analysis on the Forte´Bio Octet RED instrument were performed using a double reference subtraction (sample and sensor references) in the Forte´Bio data analysis software. The analysis accounts for non-specific binding, background and signal drift and minimizes well based and sensor variability.

1005

V.M.Golubovskaya et al.

Fig. 2. Y11 directly decreases FAK autophosphorylation activity by in vitro kinase assay and directly binds FAK-N-terminal domain. (A) Isolated purified baculoviral FAK used for in vitro kinase assay. MW, molecular weight. (B) Y11 decreases FAK autophosphorylation in a dose-dependent manner. In vitro kinase assay has been performed as described in Materials and methods. The Pfizer inhibitor, PF573228, is used as a positive control. (C) Y11 effectively decreases FAK autophosphorylation as well as control Y15. In vitro kinase assay was performed with purified FAK and 0.1 and 1 lM of Y11 and control Y15, as described in Materials and methods. Y11 effectively decreased autophoshorylation activity of FAK as well as Y15. (D) Y11 compound directly binds the N-terminal domain of FAK. Octet binding of Y11 and FAK-N-terminal domain has been performed as described in Materials and methods. Y11 binds to the N-terminal domain compared with the negative control buffer. (E) Y11 specifically inhibits FAK autophosphorylation kinase activity. The Kinase Profiler with different kinases has been performed by Millipore Kinase Profiler Service, as described in Materials and methods. Y11 inhibits FAK autophosphorylation activity but does not inhibit other kinase activities, including homologous Pyk-2 activity at 1 lM dose.

Immunostaining and immunohistochemistry staining Immunostaining was performed with FAK, FAK-Y397 and caspase-3 antibodies, as described in (10). Immunohistochemistry staining was performed on paraffinembedded tumor samples, as described in (26).

Detachment The detached and attached cells were counted using a hemacytometer. The percent of detachment was calculated by dividing the number of detached cells by the total number cells.

Kinase profiler screening To test specificity of FAK inhibitor, Kinase Profiler Service (Millipore) was used with different available kinases (http://www.millipore.com/drugdiscovery/dd3/ KinaseProfiler). The screening was performed with 1 lM of compound and 10 lM ATP and kinase substrates according to Millipore’s protocol.

Apoptosis Detached and attached cells were collected and fixed in 3.7% formaldehyde for the apoptosis assay. The apoptosis was detected with Hoechst 33342 staining, that produced very similar results with TUNEL assay, as described in (27). The fixed cells were stained with Hoechst 33342. The apoptotic cells with fragmented nuclei were counted in several fields using Zeiss fluorescent microscope under 4#,6-diamidino-2-phenylindole filter. The percent of apoptotic cells was calculated as a ratio of apoptotic cells divided by the total number of cells. For each experiment, 300 cells per treatment were counted from three independent fields. In addition, apoptosis was determined using PE-Annexin V and 7-AAD (7-amino-actinomycin) staining with PE-Annexin V Apoptosis Detection kit from BD Pharmingen according to the manufacturer’s protocol. The stained cells were analyzed by flow cytometry analysis in the Flow Cytometry Facility (Roswell Park Cancer Institute). The images were prepared with FACS DiVa software.

Cell viability Cells were treated with small molecule compounds in 96-well plates for 24 h at different concentrations of small molecule compounds. Novartis inhibitor TAE226 was used as a reference control in 3-(4,5-dimethylthiazole-2-yl)-2,5-bdiphenyl tetrazolium bromide assay. The 10 ll of 3-(4,5-dimethylthiazol-2-yl)-5-(3carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium compound from Promega 96-well viability kit was added, and the cells were incubated at 37°C for 1–2 h. To determine cell viability, the optical density at 490 nm (OD 490 nm) of cells on 96-well plate was analyzed with a microplate reader. All experiments were performed in triplicate. Clonogenicity The 500–1000 cells were plated on six-well plates and incubated at 37°C for 1–2 weeks. Then, cells were fixed in 25% methanol and stained with crystal violet and colonies were visualized and counted.

1006

In vitro kinase assay Ten microCi of [c-32P]-ATP with 0.1 lg of purified baculoviral FAK protein, described in (24), were incubated in kinase buffer with 10 lCi of [c-32P]-ATP. The kinase reaction was performed for 30 min at 30°C and stopped by addition of 2 Laemmli buffer. Proteins were separated on a sodium dodecyl


1.4

1.4

1.2 *

1

1

Viability (fold)

0.8 0.6

*

0.8 0.6

*

*

Y11 Y15

0.4

*

0.4

0.2

0.2

SW620 cells

m ic ro M

ic ro M

10

10

m ic ro M

m TA E2 26

BT474 cells

8

4

m ic ro M

m ic ro M 2

1 microM 10 microM 20 microM 50 microM

ic ro M

DMSO

Un t

Blank Untreated

Im

0

SO

re at ed

0 DM

Viability, (OD 490 nm)

1.2

Viability (OD 490 nm)

2.5

2

1.5

1

0.5

0 No cells

Untreated

DMSO

1 uM

10 uM

20 uM

50 uM

WI38-TERT cells

Fig. 3. Y11 significantly decreases cancer cell viability in a dose-dependent manner. The cells were treated with Y11 for 24 h and 3-(4,5-dimethylthiazole2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay was performed to detect viability. (A) The MTT assay was performed with different doses of Y11 with breast cancer BT474 cells. (B) The MTT assay has been performed with different doses of Y11 and Y15 with colon cancer SW620 cells. TAE226 from Novartis was used as a positive control. Bars show averages of viability þ/ standard deviations. P , 0.05, Student’s t-test. (C) The MTT assay has been performed with different doses of Y11 with normal human fibroblasts, WI38-TERT. Y11 did not significantly affected viability of normal fibroblasts. sulfate-10% polyacrylamide gel electrophoresis gel, and the phosphorylated FAK was visualized by autoradiography. Tumor growth in nude mice Mice were maintained in the animal facility and all experiments were performed in compliance with IACUC protocol approved by Roswell Park Animal Care Committee. Female nude mice were purchased 6 weeks old and used for experiments. The 1 million of SW620 cells were injected into mice subcutaneously. Next day after injection, the drug was introduced by interperitoneal (i. p.) injection at 125 mg/kg dose daily 5 days/week for 4 weeks. Tumor volume was calculated in mm3 using the following formula: tumor volume 5 (width)2 length/2. Tumors weight was determined in grams at the end of experiment. Tumor samples were collected for western blotting and for pathology immunohistochemical analysis. Statistical analyses To determine significance between the tested groups, Student’s t-test was used. The difference between untreated and treated groups with P , 0.05 was considered significant.

Results Y11 compound targets the Y397 site of FAK and decreases cancer cell viability Using computer modeling approach, we docked .140 000 small molecule compounds with known three-dimensional structures into the structural pocket of crystal structure of FAK, containing the Y397 site (25). The computer modeling approach combined the NCI database with improved molecular docking and scoring algorithms of the DOCK 5.1 program (20). Each of 140 000 compounds with known three-dimensional structure was docked in 100 different orientations

into the FERM domain of FAK. We ordered from NCI the 35 compounds out of .140 000 compounds with the highest scores of interaction with Y397-FAK and tested their effect on cancer cell viability by 3-(4,5-dimethylthiazole-2-yl)-2,5-bdiphenyl tetrazolium bromide assays on different cancer cell lines: breast, colon, melanoma and lung cancer cell lines, described in (21). Among these 35 compounds, Y11 effectively decreased viability in most of the cancer cell lines (lung cancer: A549; breast cancer: T47D, BT474, MCF-7; melanoma C8161; colon cancer HT29), which was comparable with previously characterized Y15 compound (21). Figure 1A shows that Y11 effectively reduced the viability of the MCF-7 breast cancer cell line and gave similar results with the other cancer cell lines we tested (data not shown). The name of Y11 compound is 1-(2-hydroxyethyl) -3, 5, 7-triaza-1-azoniatricyclo [3.3.1.13,7]decane, bromide, and the structure is shown in Figure 1B. Figure 1C shows this Y11 compound docked into the Y397 site of FAK. Thus, Y11 targeted the Y397 site and effectively decreased cancer cell viability. Subsequently, we synthesized Y11 compound and used it for the next functional and biochemical studies. Y11 directly blocks FAK autophosphorylation activity in a dosedependent manner To test whether Y11 directly inhibits FAK autophosphorylation, we performed in vitro kinase assays with recombinant baculoviral-purified FAK protein, shown in Figure 2A and described in (24). The in vitro kinase assay with FAK, c-ATP and Y11 showed a significant dosedependent decrease of FAK autophosphorylation activity at nanomolar doses (Figure 2B). The amount of Y11 needed to inhibit .50% of FAK autophosphorylation activity (IC50) in this assay was about 50 nM. We used the PF-573 228 (Pfizer) FAK inhibitor as a reference control, and it

1007


Fig. 4. Y11 decreases cancer cell clonogenicity, decreases Y397-FAK in a dose-dependent manner and changes cell morphology. (A and B) Y11 decreases clonogenicity in a dose-dependent manner in SW620 colon and BT474 cancer cells. The clonogenicity assay has been performed with different doses of Y11 in SW620 cells (A) and BT474 (B) for 2 weeks, as described in Materials and methods. Y11 significantly decreases clonogenicity at 10–20 lM dose in SW620 cells (A) and completely decreases clonogenicity in BT474 cells at 10 lM dose (B). (C, D) Y11 decreases Y397-FAK in a dose-dependent manner in SW620 and BT474 cells. Western blotting with Y397-FAK antibody was performed in SW620 (C) and BT474 (D) cells treated with different doses of Y11. Then, western blotting has been performed with FAK and b-actin antibodies. Y11 decreases Y397-FAK in a dose-dependent manner in SW620 cells and in BT474 cells. Y11 significantly decreased Y397-FAK but not the total FAK. (E) Y11 decreased total FAK phosphorylation. The BT474 cells were treated with 10 lM of Y11 for 24 h. FAK was immunoprecipitated and western blotting was performed with phosphotyrosine antibody and then with total FAK antibody. Y11 decreased total FAK phosphorylation. (F, G) Y11 decreases Y397-FAK in a dose-dependent manner and changes cell morphology. Immunostaining of Y397-FAK and FAK was performed in Y11-treated SW620 (F) and BT474 (G) cells with Y397-FAK (left panels) and FAK (right panels) antibodies. Y11 decreased Y397-FAK, displaced FAK from focal adhesions and caused cell rounding in both cell lines. Actin was stained with Phalloidin-FITC. Y11 compound causes FAK displacement from focal adhesions, cell rounding and decrease of actin stress fibers.

effectively blocked FAK autophosphorylation activity at 1 lM. We performed in vitro kinase assay with different doses Y11 and control Y15 inhibitor (Figure 2C). Y11 decreased FAK autophosphorylation activity in a dose-dependent manner as well as control FAK inhibitor Y15 (Figure 2C). Thus, Y11 is a direct and effective inhibitor of FAK autophosphorylation.

domain that contains the Y397 site (Figure 2E). Thus, Y11 is a specific inhibitor of FAK autophosphorylation.

Y11 directly binds to the N-terminal domain of FAK To test whether Y11 compound bound to the N-terminal domain of FAK, containing the Y397 site, we performed an Octet binding assay, using biotinylated FAK-N-terminal domain on superstreptavidin sensors and the Y11 compound (Figure 2D). Y11 directly bound the FAK-N-terminal domain at significantly higher levels above buffer alone (Figure 2D). This demonstrated that Y11 directly bound to the N-terminal domain of FAK.

Y11 decreases viability of cancer cells in a dose-dependent manner To test the effect of the Y11 compound on cancer cell viability in a dosedependent manner, we treated BT474 breast cancer cells and SW620 colon cancer cells with different doses of Y11 (Figure 3A and B). Y11 significantly decreased viability of BT474 cells in a dose-dependent manner (Figure 3A). The same was observed in SW620 colon cancer cells, where Y11 effectively and significantly decreased viability starting at 8 lM, which was comparable with control inhibitors Y15 and TAE226 (Figure 3B). To test the effect of the Y11 compound on normal human cells, we used WI38-TERT human fibroblasts. Y11 did not significantly decrease viability of the human fibroblasts (Figure 3C). Thus, Y11 decreases cancer cell viability in a dose-dependent manner.

Y11 is a specific inhibitor of FAK autophosphorylation To determine whether Y11 compound is a specific inhibitor of FAK autophosphorylation, we tested its effect on other kinase activities, using the Millipore Kinase Profiler System (Figure 2E). Y11 significantly decreased FAK autophosphorylation activity at 1 lM, but it did not affect the kinase activities of other commercially available kinases: c-Src, EGFR, Pyk-2, Flt-4 (VEGFR-3), IGF-1R, Met, PDGFR-alpha, PI3 Kinase (Figure 2E). Y11 also did not affect the kinase activity of the FAK kinase domain (411–686 aa), which lacks the N-terminal

Y11 decreases clonogenicity in cancer cells To test the effect of Y11 compound on colony formation or clonogenicity, we treated SW620 and BT474 cells with different doses of Y11 and performed clonogenicity assays (Figure 4A and B). Y11 significantly decreased clonogenicity at 10–20 lM of Y11 in SW620 cells (Figure 4A). The same decrease of clonogenicity was observed in BT474 cells, where no colonies were observed at 10 lM dose (Figure 4B). Thus, Y11 significantly decreased clonogenicity in SW620 and BT474 cancer cells.

1008


Fig. 4. Continued.

Y11 decreases Y397-FAK autophosphorylation and changes cell morphology in cancer cells To test the effect of Y11 compound on the autophosphorylation of FAK at its Y397 site in cancer cells, we treated SW620 and BT474 cells with different doses of Y11 and performed western blotting with Y397-FAK antibody (Figure 4C and D). Y11 inhibitor significantly decreased Y397-FAK autophosphorylation starting at 4 lM dose in SW620 cells, whereas it did not decrease total FAK protein levels (Figure 4C). Y11 also significantly decreased Y397-FAK phosphorylation in BT474 cells, starting at 0.1 lM (Figure 4D). To test the effect of Y11 on total FAK phosphorylation in BT474 cells, we immunoprecipitated FAK and performed western blotting with phosphotyrosine antibody (Figure 4E). Y11 significantly decreased total FAK phosphorylation (Figure 4E). To check the effect of Y11 on Y397-FAK and FAK levels and localization and cell morphology, we performed immunostaining of Y397-FAK and FAK in untreated and Y11-treated SW620 (Figure 4F) and BT474 cells (Figure 4G). We found that the levels of Y397-FAK and total FAK localized in focal adhesions in untreated cells were decreased by Y11 treatment (Figure 4F, left panel and right panels, respectively). Y11 also induced remodeling of the actin-based cytoskeleton, resulting in cell rounding of SW620 cells (Figure 4F). Immunostaining of Y397-FAK in Y11-treated BT474 cells demonstrated decreased Y397-FAK (Figure 4G), as well as displacement of Y397-FAK (Figure 4G, left panels) and total FAK (Figure 4G, right panels) from focal adhesions accompanied with cell rounding and decreased actin stress fibers (Figure 4G). Thus, Y11 decreased Y397-FAK phosphorylation in both SW620 and BT474 cells in a dose-dependent manner, decreased Y397-FAK and total FAK in focal adhesions and induced cytoskeletal remodeling and cell rounding in cancer cells.

Y11 increases detachment, cell rounding and apoptosis in SW620 cancer cells Since Y11 caused cell rounding, we tested the effect of Y11 on cell detachment and apoptosis in SW620 cells. Y11 caused a dose-dependent detachment with significant increase at 50–100 lM (Figure 5A). Y11 also caused a significant increase in cell rounding at 10 lM dose (Figure 5B). To test the effect of Y11 on apoptosis, we performed Hoechst staining and found that Y11 caused a significant increase in apoptosis at the 50 lM dose (Figure 5C, upper panel). The apoptotic Hoechst-stained fragmented nuclei are shown in Figure 5C, lower panel. The representative apoptotic fragmented nuclei are marked by white arrows. In addition, we treated SW620 cells with 50 lM of Y11 for 48 h and performed flow cytometry analysis of Annexinpositive cells after Annexin V/7-AAD staining. We detected significantly increased percent of apoptotic Annexin V-positive cells versus the untreated cells (Figure 5D) that confirmed the Hoechst staining data. The increased apoptosis in Y11-treated cells was also demonstrated by western blotting with PARP antibody (Figure 5E). The SW620 cells treated with Y11 decreased PARP at 50 lM that is consistent with increased percent of Annexin V-positive and Hoechst-stained apoptotic cells with fragmented nuclei. Thus, Y11 compound increased detachment, cell rounding and apoptosis in SW620 cells. Y11 compound inhibits colon tumor growth in vivo To test the effect of Y11 compound on tumor growth in vivo, we used SW620 colon cancer cells in a xenograft mouse model (Figure 6). Y11 dosed by i. p. injection at 125 mg/kg and found that it significantly decreased tumor growth of SW620 xenograft mice tumors compared with the vehicle (phosphate-buffered saline) control group

1009


Fig. 5. Y11 causes detachment, cell rounding and apoptosis in a dose-dependent manner in SW620 colon cancer cells. (A) Left panel: SW620 cells were treated with different dose of Y11 for 48 h and detachment assay has been performed, as described in Materials and methods. (B) Y11 compound caused cell rounding. Actin staining was performed with Phalloidin-FITC to detect changed cell morphology. Y11 causes increase in cancer cell rounding. The morphology of Y11 rounded cells is shown in the left panel. The rounded cells were counted and percent of rounding is shown in the right panel. P , 0.05, Student’s t-test. (C) Y11 increases apoptosis in a dosedependent manner. The cells were treated with different doses of Y11 for 48 h and apoptosis was detected by Hoechst staining, as described in Materials and methods. More than 100 nuclei from three independent fields in three independent experiments were analyzed. Upper panel: Hoechst staining shows dose-dependent increase of apoptotic SW620 cells. Lower panel: Hoechst staining has been performed to detect apoptosis. The representative apoptotic fragmented nuclei are marked with white arrows. Y11 causes increased apoptosis at 50 lM dose. P , 0.05, Student’s t-test. (D) Y11 increases percent of apoptotic Annexin V-positive cells. Upper panel: The SW620 cells were treated with 50 lM Y11 for 48 h. Y11 was added every 24 h. The apoptotic cells were detected by staining with Annexin V and 7-AAD and analyzed by fluorescence-activated cell sorting analysis in two independent experiments. The representative dot-plot image is shown. The horizontal and vertical axes represent labeling with Annexin V and 7-AAD, respectively. The Q3 (lower left) quadrant of the dot-plot graph represents viable non-apoptotic cells. The Q1 (upper left) contains cells that are 7-AAD but Annexin V negative. Q4 (lower right) quadrant represents early apoptotic cells that bind Annexin V, and Q2 (upper right) quadrant represents late apoptotic cells that are Annexin V and 7-AAD-positive with loss of the membrane integrity. The percent of Annexin V-positive cells is increased in Y11-treated cells. Lower panel: Y11 significantly increases percent of Annexin V-positive apoptotic cells compared with untreated cells. The bars represent Annexin V-positive cells from two independent experiments. P , 0.05, Student’s t-test. (E) Y11 downregulates PARP in 620 colon cancer cells. SW610 cells were treated with different doses of Y11 for 48 h and western blotting with PARP antibody was performed. Y11 decreases PARP at 50 lM dose.

1010


Fig. 5. Continued.

(Figure 6A). Y11 significantly decreased tumor weights compared with controls (Figure 6B). In addition, Y11 decreased Y397-FAK and uncleaved PARP levels in tumor xenografts compared with controls (Figure 6C). Immunohistochemical staining demonstrated that Y11 decreased Y397-FAK phosphorylation and increased caspase-3 staining in tumor xenografts compared with control group (Figure 6C). Thus, Y11 suppressed tumor growth and decreased tumor weight in the SW620 colon xenograft model, accompanied with decreased phosphorylation of Y397-FAK and activation of PARP and caspase-3. Discussion Thus, in this report, we identified a novel small molecule inhibitor of FAK, targeting the Y397 site, 1-(2-hydroxyethyl) -3,

5, 7-triaza-1-azoniatricyclo [3.3.1.13,7]decane; bromide (called Y11) that was found by computer modeling approach and by screening of .140 000 small molecule compounds from the NCI database and by functional cellular assays (21). We demonstrate that Y11 effectively and specifically blocks autophosphorylation kinase activity of FAK without affecting kinase activities of other enzymes. It directly binds to the N-terminal domain of FAK. In addition, Y11 decreased Y397-FAK phosphorylation in BT474 breast and SW620 colon cancer cells. Y11 decreased cancer cell viability and clonogenicity in a dose-dependent manner. Y11 induced detachment and apoptosis in SW620 cells in a dose-dependent manner. Moreover, Y11 significantly decreased colon cancer SW620 tumor growth in mouse xenograft model, accompanied with decreased Y397-FAK and activation of PARP and caspase-3 in Y11-treated tumors. This is the first report on novel small molecule

1011


Fig. 6. Y11 decreases SW620 colon cancer xenograft tumor growth. (A) SW620 cells were injected into nude mice subcutaneously. Y11 has been given at 125 mg/kg to the nude mice i. p. daily 5 days/week. Five untreated (treated with vehicle, 1 phosphate-buffered saline) and five Y11-treated mice were used for the experiment. Tumor sizes have been measured with calipers and tumor volume has been calculated, as described in Materials and methods. Y11 significantly decreased tumor growth. P , 0.05, Student’s t-test. (B) Y11 significantly decreases SW620 tumor xenograft weight. The weights of tumors from untreated and Y11-treated mice were measured at day 36. Y11 significantly decreased tumor weight. Bars show mean þ/ standard errors. P , 0.05, Student’s t-test. (C) Y11 decreased Y397-FAK and activates PARP in tumor xenografts. Tumor xenografts from untreated and Y11-treated mice were analyzed by western blotting with Y397-FAK, FAK, PARP and b-actin antibodies. Y11 significantly decreased Y397-FAK, FAK and activated PARP in tumor samples. The image is a composite from the same film. (D) Y11 decreases Y397-FAK and activates caspase-3 in tumor xenograft by immunohistochemical staining. Immunostaining was performed on Y11-treated and untreated tumor samples with Y397-FAK, FAK and caspase-3 antibodies. Immunostaining demonstrates decrease of Y397-FAK and activation of caspase-3 in Y11-treated xenograft tumors.

inhibitor of FAK that effectively blocks colon cancer cell viability, clonogenicity and in vivo tumor growth. Thus, the novel FAK inhibitor Y11 targets Y397 site and can be used for future FAK-targeted cancer therapeutics. This report validates the computer modeling structure-based approach together with biochemical functional assays to find novel inhibitors of FAK and identifies novel small molecule FAK inhibitor, Y11. This approach shows that targeting the autophosphorylation site of FAK, Y397-FAK, is an affective way to inhibit intracellular signaling, decrease phosphorylation of FAK and block cancer viability, clonogenicity and tumor growth. The inhibitors that target ATP-binding site can have non-specific effect due to conservative structure of FAK kinase domain, as has been shown with TAE226 inhibitor (14) that also inhibited Pyk-2, IGF-1R, ALK and c-Met. The Pfizer’s inhibitor PF-573 228 did not have effect on cell viability and an effect on tumor growth was not demonstrated. The PF-562 271 inhibitor inhibited both FAK and Pyk-2 kinase activity. Another FAK inhibitor PND-1186 that targeted ATP-binding site also inhibited FLT-3 even more effectively than FAK at 0.1 lM and also significantly inhibited ACK1, CDK2, IR, Aurora-A, Lck (28) at 1 lM dose, which is consistent with the conservative nature of kinase domains. Thus, our approach to target the active site of FAK, Y397, is critical for

1012

development of allosteric FAK inhibitors. Targeting the other phosphorylation sites of FAK Y861, Y925, Y407 can be applied by similar approach. Y11 inhibitor has a simple structure with a chemical name 1-(2-hydroxyethyl) -3, 5, 7-triaza-1-azoniatricyclo [3.3.1.13,7]decane; bromide that satisfies Lipinsky rules and has high efficacy in vivo in xenograft model. It is a specific FAK inhibitor that did not affect eight other kinases at 1 lM dose: PI3 kinase, c-Src, IGFR-1R, Flt-4, c-Met, PDGFR, homologous Pyk-2. Thus, targeting Y397 site of FAK is highly specific and has promises for future development of specific allosteric inhibitors. This report identifies novel inhibitor Y11 with a different chemical structure compared with the previously identified Y15 compound that was shown to inhibit breast, neuroblastoma and pancreatic tumor growth (21–23). The Y11 compound has higher docking scores with the N-terminal domain of FAK based on the van der Waals and electrostatic properties than the previously identified Y15, as determined by computer modeling approach and binding of Y11 to the N-terminal domain of FAK was detected for the first time by the sensitive Octet binding assay. Thus, Y11 directly binds the FAK-N-terminal domain, containing Y397 site, it blocks FAK autophosphorylation activity as effectively as Y15 and significantly blocks cancer cell viability,


clonogenicity and colon tumor growth in vivo. This is the first report to show that the novel small molecule inhibitor Y11 blocks tumor growth of highly metastatic SW620 colon cancer cell line. Future compound optimization, structure-activity relationship analyses of chemical derivatives of Y11 and Y15 compounds (21), toxicology, stability, pharmacodynamic and pharmacokinetic studies will reveal the best inhibitor for future clinical trials in patients. The study demonstrates that Y11 blocks highly metastatic colon cancer tumor growth in vivo. The study provides a novel autophosphorylation inhibitor of FAK that can be used for treating tumors, resistant to other FAK inhibitors or in combination with the inhibitors and chemotherapy. We treated mice next day after colon cancer cell injection to mimic adjuvant chemotherapy conditions in oncology research. The future studies will be performed with Y11 treatment started on established tumors or started before tumor cell injection in xenograft model for cancer prevention studies. This study provides a basis for these studies and is critical for the field of carcinogenesis and for development of anticancer therapeutics. Thus, we isolated and characterized a novel small molecule inhibitor of FAK autophosphorylation, Y11, that directly binds FAK, specifically inhibits it autophosphorylation activity and has an effective inhibiting effect on cancer cell viability, clonogenicity and colon cancer tumor growth. This study demonstrates a novel inhibitor of FAK that has a high impact on development of FAK-targeted therapies. Funding National Institutes of Health (CA65910 to W.G.C.); Susan Komen for the Cure (BCTR to V.M.G.); Roswell Park Cancer Institute and National Cancer Institute (P30 CA016056). Acknowledgements We would like to thank Kristiana Ferguson for excellent technical help. We acknowledge Dr Yian Zhai from Alchem Laboratories for synthesis of Y11 compound. We would like to acknowledge Dr Carl Morrison and Pathology Resource Network in the Department of Pathology and Laboratory Medicine (Roswell Park Cancer Institute) for immunohistochemical analysis of tumor samples. We would like to thank Dr Jones Craig for flow cytometry analysis of apoptosis assay services and the Flow and Image Cytometry Core facility at the Roswell Park Cancer Institute. Conflict of Interest Statement: None declared.

References 1. Schaller,M.D. et al. (1994) Focal adhesion kinase and associated proteins. Curr. Opin. Cell Biol., 6, 705–710. 2. Cance,W.G. et al. (2008) Focal adhesion kinase versus p53: apoptosis or survival? Sci. Signal., 1, pe22. 3. Hanks,S.K. et al. (2003) Focal adhesion kinase signaling activities and their implications in the control of cell survival and motility. Front. Biosci., 8, 982–996. 4. Schaller,M.D. et al. (1994) Autophosphorylation of the focal adhesion kinase, pp125FAK, directs SH2-dependent binding of pp60src. Mol. Cell. Biol., 14, 1680–1688. 5. Hanks,S.K. et al. (1997) Signaling through focal adhesion kinase. Bioessays, 19, 137–145.

6. Frame,M.C. et al. (2010) The FERM domain: organizing the structure and function of FAK. Nat. Rev. Mol. Cell Biol., 11, 802–814. 7. McLean,G.W. et al. (2003) Focal adhesion kinase as a potential target in oncology. Expert Opin. Pharmacother., 4, 227–234. 8. Smith,C.S. et al. (2005) Effect of focal adhesion kinase (FAK) downregulation with FAK antisense oligonucleotides and 5-fluorouracil on the viability of melanoma cell lines. Melanoma Res., 15, 357–362. 9. Xu,L.-h et al. (2000) The focal adhesion kinase suppresses transformationassociated, anchorage-Independent apoptosis in human breast cancer cells. J. Biol. Chem., 275, 30597–30604. 10. Golubovskaya,V. et al. (2002) Dual inhibition of focal adhesion kinase and epidermal growth factor receptor pathways cooperatively induces death receptor-mediated apoptosis in human breast cancer cells. J. Biol. Chem., 277, 38978–38987. 11. Halder,J. et al. (2005) Focal adhesion kinase silencing augments docetaxel-mediated apoptosis in ovarian cancer cells. Clin. Cancer Res., 11, 8829–8836. 12. Beierle,E.A. et al. (2007) N-MYC regulates focal adhesion kinase expression in human neuroblastoma. J. Biol. Chem., 282, 12503–12516. 13. Golubovskaya,V.M. et al. (2008) TAE226-induced apoptosis in breast cancer cells with overexpressed Src or EGFR. Mol. Carcinog., 47, 222–234. 14. Liu,T.J. et al. (2007) Inhibition of both focal adhesion kinase and insulinlike growth factor-I receptor kinase suppresses glioma proliferation in vitro and in vivo. Mol. Cancer Ther., 6, 1357–1367. 15. Slack-Davis,J.K. et al. (2007) Cellular characterization of a novel focal adhesion kinase inhibitor. J. Biol. Chem., 282, 14845–14852. 16. Roberts,W.G. et al. (2008) Antitumor activity and pharmacology of a selective focal adhesion kinase inhibitor, PF-562,271. Cancer Res., 68, 1935–1944. 17. Halder,J. et al. (2007) Therapeutic efficacy of a novel focal adhesion kinase inhibitor TAE226 in ovarian carcinoma. Cancer Res., 67, 10976–10983. 18. Schultze,A. et al. (2010) Therapeutic potential and limitations of new FAK inhibitors in the treatment of cancer. Expert. Opin. Investig. Drugs, 19, 777–788. 19. Walsh,C. et al. (2010) Oral delivery of PND-1186 FAK inhibitor decreases tumor growth and spontaneous breast to lung metastasis in pre-clinical models. Cancer Biol. Ther., 9, 778–790. 20. Sandberg,E.M. et al. (2005) Identification of 1,2,3,4,5,6-hexabromocyclohexane as a small molecule inhibitor of jak2 tyrosine kinase autophosphorylation [correction of autophophorylation]. J. Med. Chem., 48, 2526–2533. 21. Golubovskaya,V.M. et al. (2008) A small molecule inhibitor, 1,2,4,5benzenetetraamine tetrahydrochloride, targeting the Y397 site of focal adhesion kinase decreases tumor growth. J. Med. Chem., 51, 7405–7416. 22. Hochwald,S.N. et al. (2009) A novel small molecule inhibitor of FAK decreases growth of human pancreatic cancer. Cell Cycle, 8, 2435–2443. 23. Beierle,E.A. et al. (2010) Inhibition of focal adhesion kinase decreases tumor growth in human neuroblastoma. Cell Cycle, 9, 1005–1015. 24. Golubovskaya,V.M. et al. (2005) Direct interaction of the N-terminal domain of focal adhesion kinase with the N-terminal transactivation domain of p53. J. Biol. Chem., 280, 25008–25021. 25. Ceccarelli,D.F. et al. (2006) Crystal structure of the FERM domain of focal adhesion kinase. J. Biol. Chem., 281, 252–259. 26. Golubovskaya,V.M. et al. (2008) p53 regulates FAK expression in human tumor cells. Mol. Carcinog., 47, 373–382. 27. Golubovskaya,V.M. et al. (2003) Simultaneous inhibition of focal adhesion kinase and SRC enhances detachment and apoptosis in colon cancer cell lines. Mol. Cancer Res., 1, 755–764. 28. Tanjoni,I. et al. (2010) PND-1186 FAK inhibitor selectively promotes tumor cell apoptosis in three-dimensional environments. Cancer Biol. Ther., 9, 764–777. Received November 29, 2011; revised February 7, 2012; accepted March 3, 2012

1013