HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 280, Issue 26, 25008-25021, July 1, 2005

This Article Abstract Full Text (PDF) All Versions of this Article: 280/26/25008    most recent M414172200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Golubovskaya, V. M. Articles by Cance, W. G. Search for Related Content PubMed PubMed Citation Articles by Golubovskaya, V. M. Articles by Cance, W. G. Social Bookmarking                      What's this?

Direct Interaction of the N-terminal Domain of Focal Adhesion Kinase with the N-terminal Transactivation Domain of p53^*

Vita M. Golubovskaya, Richard Finch, and William G. Cance¶

From the Departments of Surgery and Biochemistry and Molecular Biology, University of Florida, School of Medicine, Gainesville, Florida 32610

Received for publication, December 16, 2004 , and in revised form, April 21, 2005.

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Focal adhesion kinase (FAK) is a nonreceptor kinase that is overexpressed in many types of tumors and associates with multiple cell surface receptors and intracellular signaling proteins through which it can play an important role in survival signaling. A link between FAK and p53 in survival signaling has been reported, although the molecular basis of these events has not been described. In the present study, we report that FAK physically and specifically interacts with p53 as demonstrated by pull-down, immunoprecipitation, and co-localization analyses. Using different constructs of N-terminal, central, and C-terminal fragments of FAK and p53 proteins, we determined that the N-terminal fragment of FAK directly interacts with the N-terminal transactivation domain of p53. Inhibition of p53 with small interfering p53 RNA resulted in a decreased complex of FAK and p53 proteins in 293 cells, and induction of p53 with doxorubicin in normal human fibroblasts caused an increase of FAK and p53 interaction. Introduction of the FAK plasmid into p53-null SAOS-2 cells was able to rescue these cells from apoptosis induced by expression of wild type p53. In HCT 116 colon cancer cells, co-transfection of FAK plasmid with p21, MDM-2, and BAX luciferase plasmids resulted in significant inhibition of p53-responsive luciferase activities, demonstrating that FAK can reduce transcriptional activity of p53. The results of the FAK and p53 interaction study strongly support the conclusion that FAK can suppress p53-mediated apoptosis and inhibit transcriptional activity of p53. This provides a novel mechanism for FAK-p53-mediated survival/apoptotic signaling.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
The focal adhesion kinase (FAK)¹ is a nonreceptor cytoplasmic 125-kDa protein-tyrosine kinase that controls a number of cellular signaling pathways, including proliferation, cell spreading, motility, and survival (1-4). The FAK gene was first isolated from chicken (1) and mouse (5) fibroblasts and then from human T and B lymphocytes (6) and human sarcoma samples (7). The FAK gene is located on chromosome 8q24 (8, 9) in humans and on chromosome 15 in mice (8).

FAK is overexpressed in many types of tumors (7, 10-14), and it was suggested that FAK up-regulation occurred in early stages of tumorigenesis (13, 15). FAK mRNA levels have been shown to be increased in premalignant adenomatous tissues and invasive and metastatic tumors (7, 15). Increased FAK mRNA expression was demonstrated in adenomatous tissues, invasive tumors, and metastatic tumors (7, 16). Recently, real time PCR analysis of matched samples of normal colon mucosa, colorectal carcinoma, and liver metastases demonstrated increased FAK mRNA and protein levels in tumor and metastatic tissues versus normal tissues (16). Cloning and characterization of the FAK promoter demonstrated NF-B and p53 transcription factor binding sites that had potential importance for regulation of FAK transcription (17).

FAK was originally isolated as a tyrosine phosphorylated protein in v-Src-transformed chicken embryo fibroblasts (1, 18). The structure of the FAK protein includes a N-terminal domain with a primary autophosphorylation site, tyrosine 397, that directly interacts with the Src homology 2 domain (19) and PI-3 kinase (2), a central catalytic domain with two major sites of phosphorylation, Tyr-576/577, and a C-terminal domain containing two proline-rich segments and a focal adhesion-targeting subdomain that binds paxillin, talin, and other proteins (2, 20). The N-terminal domain of FAK associates with the epidermal growth factor receptor and platelet-derived growth factor receptor (21, 22), with cytoplasmic tails of integrins (23) and with death receptor complex-binding protein, RIP (24). Recently, the N-terminal domain of FAK had been demonstrated in a complex with a protein inhibitor of activated STAT1 (PIAS1) (25). PIAS1 caused sumoylation of FAK and increased its autophosphorylation activity, suggesting a novel FAK role in signaling between the plasma membrane and the nucleus (25).

The C-terminal noncatalytic domain of FAK, FAK-related nonkinase (p41/p43), is autonomously expressed in chicken embryo cells (26), initiated from an alternative promoter and start site residing within an intron (27). Expression of FAK-related nonkinase caused dephosphorylation of FAK at Tyr-397 (28) and blocked FAK-mediated fibroblast migration (29). In human cells, we have shown that an analogous C-terminal FAK fragment construct that we call FAK-CD can also regulate FAK function (14, 30). Inhibition of FAK expression with anti-sense oligonucleotides to FAK or overexpression of the FAK C-terminal domain led to cell rounding, detachment, reduction of invasion, and apoptosis (20, 31-35). In addition, FAK has been shown to suppress both transformation-associated apoptosis (30) and anoikis (detachment-induced apoptosis) of epithelial cells (36), suggesting that one function of FAK is to promote survival in cells subjected to apoptotic signals. Consistent with this hypothesis, constitutively active forms of FAK prevented anoikis and stimulated transformation of epithelial cells, resulting in anchorage-independent growth and tumor formation in nude mice (36).

The functional link between FAK and tumor suppressor gene p53 was reported first by Ilic et al. (33). The authors demonstrated that p53 controls survival signals from the extracellular matrix transduced by FAK in anchorage-dependent cells (33). Recently, a link between p53-mediated anoikis and FAK was demonstrated in squamous cell carcinoma cells (37). In addition, immunohistochemical analysis of 115 endometrial carcinoma samples demonstrated a correlation between FAK and p53 overexpression (38). However, the mechanism of the p53-FAK-mediated role in anoikis and tumorigenesis has not been described.

We have based our current study on the indirect link of FAK and p53 that has been reported (17, 33, 37-39). We have demonstrated that FAK directly and physically interacts with p53 in vitro and in vivo. We performed mapping analysis and have shown that the FAK-N-terminal domain binds the transactivation domain of p53. Binding of FAK and p53 was shown in different 293, HCT116, and BT474 cancer cell lines and in normal human fibroblasts. In addition, knockdown of p53 expression in 293 cells, treated with small interfering p53 siRNA, caused decreased binding of FAK and p53 proteins. In contrast, increased expression of p53 with doxorubicin treatment resulted in elevation of the FAK-p53 complex in normal human fibroblasts. In addition, we have shown that overexpression of FAK inhibited p53-induced apoptosis in SAOS-2 cells and decreased p53-mediated activation of p21 and MDM-2 luciferase constructs in HCT116 p53⁺/⁺ cells. Moreover, co-expression of FAK with p53 in p53-deficient HCT116 p53^-/^- and p53-null SAOS-2 cells significantly inhibited p53-mediated activation of a BAX-luciferase reporter. Thus, these studies strongly support a survival function of FAK and demonstrate that the FAK and p53 interaction can affect p53 transcriptional and apoptotic activity and play role in cancer cell survival signaling.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Cells, Cell Culture, and Transfections—BT474 breast carcinoma cells, described in Refs. 22 and 30, were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, 5 µg/ml insulin, and 1 µg/ml penicillin/streptomycin. Human embryonic kidney HEK293 cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. The human normal foreskin fibroblast cell line, NHF-1 (40), was kindly provided by Dr. William Kaufmann (University of North Carolina) and was grown in minimal essential medium supplemented with 10% fetal bovine serum, L-glutamine, and 1 µg/ml penicillin/streptomycin. Human colon carcinoma cell lines, HCT 116, p53⁺/⁺wt and p53^-/^- were kindly provided by Dr. Bert Vogelstein (The Johns Hopkins University, Baltimore, MD). HCT116 cells were maintained in McCoy's 5A medium with 10% fetal bovine serum and 1 µg/ml penicillin/streptomycin. SAOS-2 human osteogenic sarcoma p53-null cell line was kindly provided by Dr. Daiqing Liao (University of Florida, Gainesville, FL) and was maintained in McCoy's 5A medium containing 15% fetal bovine serum and 1 µg/ml penicillin/streptomycin. FAK^-/^-p53^-/^- and FAK⁺/⁺p53^-/- mouse embryo fibroblasts (MEFs) were obtained from ATCC and were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1 µg/ml penicillin/streptomycin. Transfections were done with Lipofectamine (Invitrogen) following the manufacturer's instructions. Equal amounts of DNA were used for each transfection.

DNA Constructs—The N-, kinase/central and C-terminal parts of FAK and p53 were generated by PCR with the gene-specific primers (available upon request). Amplified by PCR, N-transactivation domain (aa 1-92), central DNA binding domain, C (aa 102-292), N+C (aa 1-292), or tetramerization terminal domain, T (aa 325-393) fragments of p53 or FAK fragments (N-terminal (aa 1-423), FAK-NT (aa 1-205), and FAK-NT (aa 206-422) domains, kinase domain (aa 416-676), and C-terminal domain (aa 677-1052)) were cloned into the pGEX-4T1 GST vector (Amersham Biosciences). All sequences were analyzed by automatic sequencing (University of Florida Sequencing Facility). The FAK construct with the deleted (aa 1-416) FAK-NT fragment, -FAK-NT-pcDNA3.1-His-Xpress-tagged plasmid, -FAK-NT, was kindly provided by Dr. E. Kurenova (University of Florida, Gainesville, FL). The p21-pGL3 plasmid that contains a fragment from the p21/CIP/WAF gene promoter with two wild type p53 sites (41, 42) and p21-mut-pGL3 luciferase constructs with mutant p53 binding sites, described in Ref. 41, were kindly provided by Dr. D. Liao (University of Florida). The HA-tagged-FAK-pcDNA3 plasmid, containing full-length HA-tagged FAK cDNA, was described before (43). The Mdm-2pGL2, luciferase construct with the p53 binding sites from the Mdm-2 gene and the Bax-pGL3 luciferase construct with p53 binding site in the Bax gene promoter, described in Ref. 44, were kindly provided by Dr. D. Liao. The GFP-FAK plasmid was a kind gift of Dr. J. Guan (Cornell University, Ithaca, NY). The pFastBacHT plasmids with full-length FAK and proline-rich tyrosine kinase-2 (PYK-2) cDNA were kindly provided by Dr. Michael Schaller (University of North Carolina, Chapel Hill, NC).

Antibodies and Reagents—Monoclonal anti-FAK (4.47) and clone-77 monoclonal antibodies specific to the N-terminal domain of FAK were obtained from Upstate Biotechnology, Inc., and BD Transduction Laboratories, respectively. Polyclonal antibody for the C-terminal domain of FAK, C-20, was obtained from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). BC2, a polyclonal antiserum raised against FAK catalytic domain and the end of the N-terminal domain of FAK (45) was kindly provided by Dr. Michael Schaller (University of North Carolina). Monoclonal anti-GFP antibody was obtained from Clontech BD Biosciences. Polyclonal calcium-dependent tyrosine kinase (CADTK)/PYK-2 antibody (CADTK 91 antiserum) was kindly provided by Dr. H. Shelton Earp (Lineberger Comprehensive Cancer Center, University of North Carolina). Monoclonal anti-p73-/ Ab 4 antibody was obtained from LABVISION Corp. Monoclonal anti-vinculin, -actin, GST antibodies were obtained from Sigma. Monoclonal anti-p53 antibody Ab-6, clone DO-1, was used from Oncogene Research Products Inc. Monoclonal anti-glyceraldehyde-3-phosphate dehydrogenase antibody was purchased from Advanced ImmunoChemical Inc. Polyclonal anti-Histone H3 antibody was obtained from Cell Signaling Inc.

A recombinant human p53 protein (wild type) expressed in the baculovirus expression vector system was obtained from Pharmingen BD. A recombinant human purified p73- protein was obtained from American Proteomics Inc.). Thrombin was obtained from Amersham Biosciences. HA tag antibody was used from Cell Signaling Inc. Xpress tag antibody was obtained from Invitrogen. Doxorubicin was purchased from Calbiochem and used at a concentration of 0.5 µg/ml for 24 h, as described (46). Recombinant purified GFP expressed in the baculovirus expression vector system was kindly provided by Dr. S. Haskill. p53 siRNA (SMARTpool p53 reagent) and control siRNA (siCONTROL non-targeting siRNA) were obtained from Dharmacon Inc.

Immunoprecipitation and Western Blotting—Cells were washed twice with cold 1x PBS and lysed on ice for 30 min in a buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Triton-X, 0.5% NaDOC, 0.1% SDS, 5 mM EDTA, 50 mM NaF, 1 mM NaVO₃, 10% glycerol, and protease inhibitors: 10 µg/ml leupeptin, 10 µg/ml phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin. The lysates were cleared by centrifugation at 10,000 rpm for 30 min at 4 °C. Protein concentrations were determined using Bio-Rad kit. The cleared lysates with equivalent amount of protein were incubated with 5 µl of primary antibody for 1 h at 4 °C and 25 µl of protein A/G-agarose beads (Oncogene Research Products Inc.). The precipitates were washed with the lysis buffer three times and resuspended in 30 µl of 2x Laemmli buffer. The boiled samples were loaded on Ready SDS-10% polyacrylamide gels (Bio-Rad) and used for Western blot analysis with the protein-specific antibody. The blots were stripped in a stripping solution (Bio-Rad) at 37 °C for 15 min and then reprobed with the primary antibody for checking equal loading of proteins. Immunoblots were developed with chemiluminescence Renaissance reagent (PerkinElmer Life Sciences).

Immunostaining and Confocal Microscopy—Cells were fixed in 4% paraformaldehyde in 1x PBS for 10 min and permeabilized with 0.2% Triton X-100 for 5 min on ice. Cells were blocked with 25% normal goat serum in 1x PBS for 30 min, washed in 1x PBS, and incubated with primary antibody diluted 1:200 in 25% goat serum in 1x PBS. Cells were washed in 1x PBS three times and a secondary rhodamine (TRITC)-conjugated antibody (1:400 dilution in 25% goat serum) was applied to the coverslip. After washing three times with 1x PBS, cells were stained with Hoechst 33342 for detecting nuclei. After mounting, cells were examined under a fluorescent Zeiss microscope. Confocal microscopy was performed on a confocal laser inverted Leica microscope. Images were captured and merged with Leica CS software.

Subcellular Fractionation—Cells were washed with 1x PBS and resuspended in ice-cold hypotonic buffer (10 mM Tris-HCl, pH 7.4, 10 mM KCl, 1.5 mM MgCl₂, 10 µg/ml leupeptin, 10 µg/ml phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, 1 mM dithiothreitol, 0.5% Nonidet P-40). Samples were centrifuged at 1000 rpm for 10 min at 4 °C, and the supernatants were collected as cytoplasmic fractions. Pellets containing cell nuclei were washed once with hypotonic buffer and then extracted with high salt lysis buffer (50 mM Tris-HCl, pH 8.0, 400 mM NaCl, 1% Nonidet P-40, 1% sodium deoxycholate, 0.025% SDS, 10 µg/ml leupeptin, 10 µg/ml phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin), followed by centrifugation at 10,000 rpm at 4 °C. The supernatant was the nuclear fraction. Equal protein amounts were loaded on the Ready SDS-10% polyacrylamide gels (Bio-Rad) and used for Western blot analysis with anti-FAK, anti-p53, anti-glyceraldehyde-3-phosphate dehydrogenase (cytoplasmic fraction marker) and anti-histone H3 (nuclear fraction marker) antibodies.

Expression of Recombinant GST Fusion Proteins—GST fusion proteins containing different domains of FAK and p53 were engineered by PCR. The fusion proteins were expressed in Escherichia coli bacteria by incubation with 0.2 mM isopropyl -D-galactopyranoside for 6 h at 37 °C. The bacteria were lysed by sonication, and the fusion proteins were purified with glutathione-agarose beads. The purified human FAK-NT protein was isolated from FAK-NT-GST fusion protein by thrombin cleavage according to manufacturer's protocol (Amersham Biosciences).

Expression and Isolation of Baculovirus-expressed Proteins—PFastBac HT donor plasmids with the full-length FAK and PYK-2 cDNA were transformed into DH10Bac competent cells (Invitrogen). After transformation and transposition of the FAK DNA and PYK-2 DNA into a bacmid DNA, colonies containing recombinant bacmids were identified by disruption of the LacZ gene inside the bacmid DNA. The isolated recombinant bacmid DNAs from white LacZ^- colonies were used for transfection of Sf9 insect cells. After three rounds of viral amplification, a recombinant FAK baculovirus and a PYK-2 baculovirus with a high titer (>1 x 10⁸ plaque-forming units/ml) were used for infection of High Five insect cells (Invitrogen). The His₆-tagged FAK and PYK-2 full-length proteins were isolated from High five, H5 insect cells with the Bac-to-Bac baculovirus system buffers on Ni²⁺-nitrilotriacetic acid resin (Qiagen Inc.) columns following the manufacturer's protocol (Invitrogen Inc.). In brief, High five insect cell extract pellets were resuspended in the lysis buffer, containing 50 mM Tris-HCl, pH 8.5, 10 mM 2-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 1% Nonidet P-40 at 4 °C. After centrifugation step, the supernatants were loaded on Ni²⁺-nitrilotriacetic acid resin column, pre-equilibrated with the washing buffer, containing 20 mM Tris-HCl, pH 8.0, 500 mM KCl, 20 mM imidazole, 10 mM mercaptoethanol, 10% glycerol. After washing steps, the proteins were eluted with the buffer, containing 20 mM Tris-HCl, pH 8.5, 100 mM KCl, 100 mM imidazole, 10 mM 2-mercaptoethanol, 10% glycerol at 4 °C. The collected proteins were purified with slide-A-Lyzer 7K dialysis Cassettes (Pierce) in 1x PBS for 24 h at 4 °C. Isolated His-tagged FAK and PYK-2 proteins were analyzed by Western blot with anti-His and anti-FAK and anti-PYK-2 antibodies.

Pull-down Assay—For the pull-down binding assay, cell lysates (0.5 mg) or purified recombinant baculoviral protein (0.1 µg) were precleared with GST immobilized on glutathione-agarose beads by rocking for 1 h at 4 °C. The washed precleared lysates were incubated with 2-4 µg of GST fusion protein immobilized on the glutathione-agarose beads for 1 h at 4 °C and then washed three times with radioimmune precipitation lysis buffer. Equal amounts of GST fusion proteins were used for each binding assay. Bound proteins were boiled in 2x Laemmli buffer and analyzed by Western blotting.

Dual Luciferase Assay—For luciferase assay, 2 x 10⁵ cells were plated on 6-well plates, cultured overnight, and transfected with the pGL3 plasmids (1 µg/well) using Lipofectamine (Invitrogen) transfection agent according to the manufacturer's protocol. For normalization of luciferase activity, pRL-TK control vector containing the herpes simplex virus thymidine kinase promoter encoding Renilla luciferase was used, resulting in its constitutive expression in a variety of cell types (Promega). The PRL-TK vector was used (0.1 µg/well) together with the pGL3 plasmids for co-transfection. The level of firefly luciferase activity was normalized to that of the Renilla luciferase activity in each experiment. For all experiments, cells were cultured for 24-48 h after transfection and lysed with 1x passive lysis buffer (Promega). Lysates were analyzed using the dual luciferase reporter assay system kit (Promega). Luminescence was measured on Turner TD 20/20 luminometer (Promega). All experiments were performed at least three times.

Apoptosis Assay—SAOS-2 cells were co-transfected with different plasmids plus GFP plasmid. Cells were collected and fixed in 3.7% formaldehyde in 1x PBS solution for the apoptosis assay. In the previous study, we performed simultaneous staining and quantification of apoptotic cells by terminal dUTP nick-end labeling assay with an Apo-Tag fluorescein in situ apoptosis detection kit (Intergen) that produced very similar results (22). In the present study, detection of apoptosis was done with Hoechst 33342 method, as described (22, 33, 47). In brief, Hoechst 33342 in 1x PBS solution (1 µg/ml) was added to the fixed cells for 10 min, and cells were washed twice with 1x PBS, spread evenly on the slide, and examined under a fluorescent microscope. The percentage of apoptotic cells was calculated on a blind basis as a ratio of apoptotic GFP-positive cells with fragmented nuclei/total number of nonapoptotic GFP-positive cells with large nonfragmented, noncondensed nuclei in three independent experiments in several fields with the fluorescent microscope. More than 300 cells were analyzed for each experimental sample. For some experiments, cells were collected 48 h after transfection and sorted by a FACSVantage SE sorter (BD Biosciences). GFP-positive FACS-sorted cells were stained with Hoechst 33342 for detection of apoptosis, as described above. Independent FACS analysis of GFP-positive cells, performed as described in Ref. 77, confirmed the assay above.

siRNA Transfection—1 x 10⁶ cells were transfected with p53 siRNA (SMARTpool p53 reagent) and control siRNA (siCONTROL nontargeting siRNA) (Dharmacon) at 50 nM concentration with Lipofectamine 2000 (Invitrogen), according to the manufacturer's protocol. Cells were collected 72 h after transfection and used for experiments.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Direct Interaction of Human p53 and FAK Proteins in Vitro—To demonstrate the physical interaction of p53 and FAK, we expressed different GST-FAK proteins, corresponding to the N-terminal, kinase, and C-terminal domains and covering full-length FAK, such as GST-FAK-NT protein (GST protein with fused N-terminal domain of FAK, aa 1-423), GST-kinase protein (GST protein with fused catalytic kinase domain of FAK, aa 416-676), and GST-FAK-CD protein (GST protein with fused C-terminal domain of FAK, aa 677-1052) (Fig. 1A, upper panel). A precleared recombinant baculovirus-expressed human p53 protein with GST-glutathione-agarose beads was used for pull-down assays with GST-FAK proteins (Fig. 1A, first left lane). The p53 protein was incubated with GST, GST-FAK-NT, GST-kinase, and GST-FAK-CD proteins and immobilized on glutathione-agarose beads, and the binding of p53 and FAK proteins was detected by Western blot with p53 antibody (Fig. 1A). The pull-down assays showed strong binding of p53 with the FAK-NT domain of FAK, and binding was absent with control GST protein, GST-kinase, and GST-FAK-CD proteins (Fig. 1A, first pull-down panel). To check the specificity of FAK binding with p53, we used one of the p53 homologs with similar structure, a recombinant purified human p73- protein (48), in the same pull-down assay with GST, GST-FAK-NT, GST-kinase, and GST-C-terminal domains of FAK (Fig. 1A, second pull-down panel). We did not detect binding of p73- protein with GST-FAK domain proteins, indicating the specificity of FAK and p53 binding. The GST fusion proteins used for the pull-down assay with recombinant p53 and p73- proteins were analyzed by Coomassie staining and shown in Fig. 1A. Western blot analysis of GST-FAK proteins with anti-FAK antibodies, specific for different domains of FAK, N terminus of FAK (Fig. 1A, clone 77 Ab), kinase domain of FAK (Fig. 1A, BC-2 Ab), and C-terminal domain of FAK (Fig. 1A, C-20 Ab) demonstrated that GST-FAK proteins were specifically recognized by FAK antibodies. Thus, the results indicate that the p53 protein was able to physically interact with the N-terminal domain of FAK in vitro.

To demonstrate that a full-length FAK interacted with p53, we isolated and purified full-length FAK protein using the baculovirus system (Fig. 1B). As a control, we used purified GFP protein expressed with the baculovirus system (Fig. 1B). To analyze the specificity of FAK and p53 binding, we isolated one of the FAK family members, a proline-rich tyrosine kinase-2 that shares high similarity with FAK domains, PYK-2 protein, also known as CADTK, CAK-, RAFTK, and FAK2 (49-51) (Fig. 1B). We then performed a pull-down assay of baculovirus-expressed FAK with GST and GST-p53 proteins (Fig. 1C). We show that full-length FAK binds to GST-p53 and not to control GST (Fig. 1C), whereas PYK-2 did not bind to GST-p53 and GST (Fig. 1D). Control GFP protein also did not bind GST-p53 and GST (Fig. 1E). The GST and GST-p53 proteins used for pull-down assay were analyzed by Western blot with GST antibody (Fig. 1F). Thus, full-length FAK directly and specifically bound p53 in vitro.

View larger version (24K): [in this window] [in a new window]   FIG. 1. A, direct association of p53 with the N-terminal domain of FAK in vitro. Upper panel, the scheme of the GST-FAK constructs used for the pull-down assay. Cloning of FAK-GST constructs was performed by PCR amplification of the N-terminal, kinase, and C-terminal domains of FAK and then subcloning into the pGEX-4T1 GST vector. All sequences were analyzed by automatic sequencing (University of Florida Sequencing Facility). The FAK-GST constructs were full-length GST-N terminus (aa 1-423), GST-kinase domain (aa 416-676), and GST-C-terminal domain (aa 677-1052). All of the constructs covered full-length FAK. Recombinant human p53 protein (0.1 µg) expressed with the baculovirus system (BD Pharmingen) was precleared in radioimmune precipitation-lysis buffer with GST on beads (2 µg) by rotating at 4 °C for 1 h. Precleared p53 (input 10%), loaded on the first lane, was incubated with GST alone, GST-FAK-NT, GST-kinase, or GST-FAK-CD proteins. The p53 protein associated with these GST-FAK proteins was analyzed by Western blot (WB) with p53 antibody. Western blot analysis detected binding of p53 with the GST-FAK-NT protein. The same pull-down assay (above) was performed with purified human p73- protein (American Proteomics). A Western blot with anti-p73- antibody did not detect binding with GST-FAK proteins. Lower panel, the GST proteins used for the pull-down assay were analyzed by SDS-PAGE and Coomassie Blue staining. The GST-FAK proteins are marked by asterisks. The GST-FAK proteins were loaded on parallel SDS-polyacrylamide gel, and Western blotting was performed with antibodies specific to different FAK domains. The GST-FAK-NT protein reacted with anti-FAK antibody specific to the N terminus of FAK (monoclonal antibody) (clone 77; BD Transduction Laboratories). The kinase domain and the N-terminal FAK-NT domain GST proteins were recognized by polyclonal BC2 antibody (see "Experimental Procedures"). The GST-FAK-CD proteins were analyzed by Western blotting with anti-FAK antibody, specific to the C-terminal domain of FAK, C-20 antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Molecular weight marker sizes are shown in the left lanes. B-F, specific and direct binding of full-length FAK with p53 protein. B, we used the baculovirus system (Invitrogen) to isolate and purify a full-length FAK protein and a FAK family member, PYK-2 protein. As a control, we used purified recombinant GFP protein. Coomassie-stained proteins are shown. C-F, pull-down assays of baculovirus proteins with GST and GST-p53 proteins immobilized on glutathione-agarose beads. Proteins expressed with the baculovirus system and used for the pull-down assay were full-length FAK (C), PYK-2 (D), and GFP (E). The GST and GST-p53 proteins were analyzed by Western blotting with GST antibody (F). Full-length FAK directly physically and specifically interacted with p53, whereas PYK-2 and GFP did not bind to p53 protein.

To demonstrate that the binding of FAK and p53 in vivo will depend on the level of endogenous p53 in cells, we treated 293 cells with p53 siRNA and with control siRNA. p53 siRNA specifically and significantly silenced p53 expression in 293 cells, whereas control siRNA did not inhibit p53 expression (Fig. 2B). We performed immunoprecipitation of FAK in control untreated 293 cells and in cells treated with p53 siRNA and then performed Western blot with p53 antibody (Fig. 2B, lower panel). Immunoprecipitation of FAK detected less p53 protein in the complex with FAK in p53 siRNA-treated cells versus untreated cells (Fig. 2B, lower panel). Thus, knockdown of endogenous p53 expression inhibited the physical interaction of FAK and p53 in vivo.

View larger version (27K): [in this window] [in a new window]   FIG. 2. A, association of FAK and p53 proteins in 293 cells in vivo. The lysate of 293 cells was precleared with the A/G-agarose beads and used for immunoprecipitation (IP) with p53 antibody (left panel). The samples were analyzed by Western blot (WB) and probed with FAK antibody. FAK was co-immunoprecipitated with p53. The blot was stripped and probed with p53 antibody to detect immunoprecipitated p53 protein. The reverse experiment was performed with immunoprecipitation using FAK antibody and Western blot using p53 antibody (right panel). The blot was stripped and probed with FAK antibody to detect immunoprecipitated FAK protein. Ten µg of precleared cell lysate is shown in the left lanes. Immunoprecipitation without antibody was used as a control (middle lanes). B, silencing of p53 expression with p53 siRNA decreases the complex of FAK and p53 in 293 cells in vivo. Upper panel, 293 cells were transfected on 100-mm dishes with 50 nM of small interfering p53 siRNA or control siRNA (Dharmacon). Western blot with p53 shows significant inhibition of p53 expression by p53 siRNA compared with control siRNA. Western blot with -actin antibody was used for control of equal protein loading. Lower panel, immunoprecipitation of lysates from untreated and p53 siRNA 293 cells was performed with FAK antibody. Immunoprecipitation of 293 cells without antibody was used as a control, and Western blotting with anti-p53 antibody was performed. The blot was stripped and probed with FAK antibody. p53 siRNA treatment significantly inhibited the complex of FAK and p53 proteins compared with untreated cells. C, association of p53 from 293 cells with the FAK-NT protein. 293 cells were transfected with p53-pCMV4 plasmid or with control vector plasmid. The level of p53 protein in these cells was analyzed by Western blot with p53 antibody, and equal protein loading was analyzed with anti--actin antibodies. (left panel). The 293 cells lysates were used for pull-down assay with GST and GST-FAK-NT proteins, as described in Fig. 1A. GST and GST-FAK-NT proteins were also detected by Western blot with anti-GST antibody. D, association of FAK and p53 proteins in HCT116 cells in vivo. Lysate of colon cancer HCT116 cells was precleared with the A/G-agarose beads (left panel) and used for immunoprecipitation with p53 antibody. The samples were analyzed by Western blot with FAK antibody. FAK was co-immunoprecipitated with p53. The blot was stripped and probed with p53 antibody to detect immunoprecipitated p53 protein. Immunoprecipitation without antibody was used as a control. E, doxorubicin increases p53 level and p53-FAK complex in normal human fibroblasts in vivo. Upper panel, normal human fibroblasts were treated with doxorubicin (0.5 µg/ml) for 24 h, lysed, and analyzed for expression of p53, FAK, and -actin with anti-p53, FAK, and -actin antibodies, respectively. Doxorubicin increased p53 level in normal human fibroblasts. Lower panel, immunoprecipitation of p53 was performed on fibroblasts without and with doxorubicin treatment. The samples were analyzed by Western blot with FAK antibody. FAK was co-immunoprecipitated with p53, and the FAK-p53 complex was significantly increased by doxorubicin. The blot was stripped and probed with p53 antibody to detect immunoprecipitated p53. Immunoprecipitation without antibody was used as a negative control. F, no detection of FAK and p53 complex in FAK-/- MEF cells transfected with p53. FAK-/- and FAK+/+ p53-null mouse embryo fibroblasts (ATCC) MEFs were transfected with p53-PCM4 (1 µg), and immunoprecipitation was performed with p53 antibody. To detect binding with FAK, samples were analyzed by Western blotting with FAK antibody. Then blots were stripped and analyzed with p53 antibody. A negative control with no antibody was included. Upper panels, FAK was not co-immunoprecipitated with p53 in FAK-/- MEF cells, whereas it was co-immunoprecipitated with p53 in FAK+/+ cells (lower panels). Higher exposure images are shown for MEF-/- cells in contrast to MEF+/+ cells to demonstrate the absence of FAK binding with p53 in these cells.

To demonstrate binding of endogenous FAK and p53 in another cell line in vivo, we performed immunoprecipitation of p53 in colon cancer HCT116 cells, followed by Western blot with FAK antibody (Fig. 2D). We detected FAK protein in the complex with p53 protein in HCT116 cells (Fig. 2D). FAK and p53 binding was not detected in control sample without antibody (Fig. 2D).

To demonstrate binding of FAK and p53 in a nontransformed cell line, we performed a co-immunoprecipitation experiment of the normal human fibroblast cell line, NHF-1 (Fig. 2E). This cell line expressed low level of p53 and FAK, so we treated the NHF-1 cells with doxorubicin at dose of 0.5 µg/ml for 24 h and demonstrated significant stress-induced induction of p53 (Fig. 2E, upper panel). Next, we immunoprecipitated p53 in these cells without and with doxorubicin treatment and performed Western blot with FAK antibody (Fig. 2E, lower panel). The FAK-p53 complex was detected in NHF-1 fibroblasts and increased in doxorubicin-treated cells with increased level of p53. In addition, we performed a co-immunoprecipitation experiment of FAK and p53 in FAK^-/^- cells and FAK⁺/⁺ p53-null MEF cells, transfected with p53-pCMV4 plasmid (Fig. 2F). We immunoprecipitated p53 in these cells and performed Western blot with FAK. We did not detect a complex of FAK and p53 in FAK^-/^- cells, whereas it was detected in FAK⁺/⁺ cells (Fig. 2F). The results clearly show binding of FAK and p53 in normal and cancer cell lines in vivo.

Co-localization of FAK-NT and FAK with p53—Previous reports have demonstrated that an amino-N-terminal domain of FAK localized to the nucleus (43, 52, 53). In fact, our previous report in BT474 breast cancer cells have shown that the N-terminal domain of FAK, FAK-NT, localized to the nuclei and caused cell rounding and apoptosis (43). Since p53 binds to this N-terminal fragment of FAK, we sought to determine the intracellular distribution of these proteins in BT474 breast cancer cells. We performed immunohistochemical and confocal laser-scanning microscopy analysis on BT474 breast cancer cells transfected with GFP, GFP-FAK-NT (N-terminal), GFP-FAK-CD (C-terminal), and GFP-full-length FAK constructs (Fig. 3A). The p53 protein was stained with p53 monoclonal antibody and then with rhodamine-conjugated secondary antibodies and was detected both in the nuclei and cytoplasm dots (Fig. 3A). The N-terminal FAK-NT was localized in the nuclei and mainly co-localized with p53 in the nuclei (Fig. 3, arrows), whereas the C-terminal GFP-FAK-CD was mainly localized at the focal adhesion sites (Fig. 3A, arrowheads). GFP-FAK and endogenous FAK mainly localized at the focal adhesions and cytoplasm (Fig. 3A, arrowheads). GFP-FAK co-localized with p53 in the nuclear, perinuclear, and cytoplasmic speckles (Fig. 3A). Endogenous FAK and p53 clearly co-localized in both cytoplasmic and nuclear areas in BT-474 cells (Fig. 3A, lower panel). We detected similar co-localization of FAK and p53 in 293 and MCF-7 cells (not shown). Co-localization of FAK and p53 in the nucleus and cytoplasm suggests that FAK and p53 can undergo nucleocytoplasmic shuttle inside cells, consistent with data for both proteins (25, 54).

To confirm these results, we performed cell fractionation of the cytoplasmic and nuclear components in BT-474 and 293 cells (Fig. 3, B and C). FAK was mainly localized in the cytoplasm in both cell lines, although it was also detected in the nuclear fractions (Fig. 3, B and C). 293 cells had a higher level of p53 in the nuclear fraction (Fig. 3C), whereas the BT-474 cell line had higher level of cytoplasmic p53 (Fig. 3B). The results on detection of FAK in the nucleus and p53 in cytoplasm are consistent with the other reports (52, 53, 55-57). We performed immunoprecipitation of FAK from cytoplasmic and nuclear fractions in these cells and detected binding of p53 protein and FAK proteins in both nuclear and cytoplasmic fractions (Fig. 3, D and E). The complex of FAK and p53 was lower in the nucleus in BT474, since cells expressed lower nuclear p53 (Fig. 3D). The complex of FAK and p53 was higher in the nuclear fraction in 293 cells with higher levels of nuclear p53 (Fig. 3E). Thus, the co-localization of FAK-NT and FAK with p53 (Fig. 3A) and FAK and p53 complexes in the cytoplasmic and nuclear fractions (Fig. 3, D and E) support the nucleocytoplasmic shuttle of both proteins and the link between FAK- and p53-mediated signaling.

The Transactivation Domain of p53 and the N-terminal Domain of FAK Are Involved in the p53/FAK Interaction—To access regions of p53 involved in the interaction with FAK, we prepared the following p53-GST constructs: N-terminal transactivation domain of p53, TAD (called GST-N-p53); central DNA-binding domain (called GST-C-53); tetramerization domain, TD (called GST-T-p53); and combined N plus C domains of p53 (called GST-N+C-p53) (Fig. 4A). We used these constructs for GST pull-down assays to determine the regions of p53 involved in interaction with FAK protein.

View larger version (35K): [in this window] [in a new window]   FIG. 3. A, co-localization of FAK-NT and FAK proteins with p53 protein in breast cancer BT474 cells by confocal microscopy. BT474 cells were transfected with GFP, GFP-FAK-NT, GFP-FAK-CD, and GFP-FAK plasmids. Co-localization of endogenous FAK and p53 was performed in nontransfected cells. Cells were fixed and processed for immunofluorescence analysis by confocal microscopy (see "Experimental Procedures"). Localization of endogenous FAK was detected in nontransfected BT474 cells with FAK polyclonal C-20 antibody (Santa Cruz Biotechnology) and probed with secondary fluorescein isothiocyanate-conjugated anti-rabbit antibody. For detection of p53, immunostaining with p53 mouse monoclonal antibody and with secondary rhodamine-conjugated anti-mouse antibody was performed. The nuclei were stained by Hoechst 33342. GFP staining was cytoplasmic. GFP-FAK-NT was mainly co-localized with p53 in the nuclei in BT474 cells (marked by the white arrows). GFP-FAK-CD was localized in the cytoplasm and in the focal adhesions (marked by the white arrowheads). GFP-FAK mainly localized in the focal adhesions (white arrowheads) and cytoplasm. GFP-FAK co-localized with p53 in the cytoplasmic and nuclear speckles (marked by the white arrows). The same co-localization with p53 was observed with endogenous FAK (lower panel, white arrows). B-E, FAK and p53 associate in cytoplasmic and nuclear fractions. Upper panels, BT474 (B) and 293 (C) cells were fractionated into cytoplasmic (labeled C) and nuclear (labeled N) fractions (see "Experimental Procedures"). 15 µg of each fraction lysate was run on SDS-polyacrylamide gels. Western blots (WB) were performed with anti-FAK antibody to detect FAK protein and with anti-p53 antibody to detect p53 protein. Western blot with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody was performed for detection of the cytoplasmic fraction, and Western blot with anti-histone H3 was performed for detection of the nuclear fraction. D and E, immunoprecipitation (IP) with FAK antibody was performed using cytoplasmic and nuclear fraction of BT-474 (D) and 293 cells (E). No antibody immunoprecipitation was used as a control. Western blot with p53 antibody was performed to detect a complex of p53 with FAK protein. Then blots were stripped and probed with anti-FAK antibody.

To determine more precisely the region of FAK-NT that is involved in interaction with p53, we prepared GST proteins with fused full-length N terminus FAK-NT (aa 1-423), first half of FAK-NT domain (aa 1-205) and second half of FAK-NT domain (aa 206-422) (Fig. 5A). The GST-FAK-NT proteins were analyzed by Coomassie Blue staining (Fig. 5A, middle panel). The GST-FAK-NT proteins were specifically recognized by FAK antibodies (Fig. 5A, lower panel). Western blotting with anti-FAK antibody specific to the second half of FAK-NT (Clone-77) or with anti-FAK antibody specific to first half of FAK-NT (Fig. 5A, lower panel) demonstrated specific FAK-NT binding. Next, we performed pull-down assays with GST-FAK-NT proteins and recombinant baculovirus-expressed human p53 protein (Fig. 5B). The p53 protein binds to the full-length FAK-NT and to the second half of FAK-NT (aa 206-422) and not to the first half of FAK-NT (aa 1-205) (Fig. 5B). Thus, the region of FAK-NT protein between amino acids 206 and 422 of FAK is involved in the interaction with the p53 protein.

FAK Expression Blocks p53-induced Apoptosis—To examine the functional significance of the physical interaction between FAK and p53, we next determined whether FAK could affect p53-mediated apoptosis. It is known that exogenous p53 causes apoptosis in p53-null SAOS-2 cells (58-60). We transfected SAOS-2 cells with wild type p53 plasmid together with or without the FAK plasmid and analyzed apoptosis in SAOS-2 cells. To identify transfected cells, we co-transfected the plasmids with a constant amount of the GFP expression plasmid. Transfected SAOS-2 cells were stained with Hoechst, and the percentage of cells with fragmented apoptotic nuclei among GFP-positive cells was determined by fluorescent microscopy. Introduction of p53 into Saos-2 cells caused apoptosis in 27% of GFP-positive cells (Fig. 6A). Introduction of FAK together with p53 plasmid suppressed p53-induced apoptosis to the background level of 12% in a dose-dependent manner (Fig. 6A). In contrast, introduction of the mutant construct, FAK-NT with a deleted FAK-NT domain of FAK, lacking the ability to interact with p53, did not significantly affect p53-induced apoptosis in SAOS-2 cells (Fig. 6A). The p53 status of SAOS-2 cells was analyzed by Western blot (Fig. 6B, upper panel), and apoptosis was demonstrated with Hoechst 33342 staining. Apoptotic GFP-positive FACS-sorted cells with fragmented nuclei are shown in (Fig. 6B, lower panels, arrows). Thus, FAK expression specifically blocked p53-induced apoptosis in SAOS-2 cells in a dose-dependent manner.

View larger version (31K): [in this window] [in a new window]   FIG. 4. A, scheme of the GST-p53 constructs used for mapping of FAK and p53 binding. Upper panel, all p53 construct; transactivation, TAD (aa 1-92) (called N-p53); DNA binding (aa 102-292) (C-p53); tetramerization (T-p53) (aa 325-393); and N+C (aa 1-292) domains (called N+C-p53) were constructed by PCR and cloned into pGEX-4T1 GST vector. B, the N-terminal transactivation domain of p53 binds FAK-NT protein. Upper panel, human FAK-NT protein eluted by thrombin digestion from recombinant GST-FAK-NT protein (10% input, first left lane) was used for pull-down analysis with GST (control) and GST-p53 proteins: GST-p53, GST-N-p53, GST-C-p53, GST-T-p53, and GST-N+C-p53. The pull-down assays demonstrate that FAK-NT protein binds to the N-terminal domain of p53. Isolated GST-p53 proteins were analyzed by SDS-PAGE and Coomassie Blue staining (lower panel). Recombinant p53-GST and GST proteins are marked by the arrows.

View larger version (18K): [in this window] [in a new window]   FIG. 5. A and B, the FAK-NT (aa 206-422) protein binds p53. A, left upper panel, scheme of GST-FAK-NT proteins used for pull-down assay. We used full-length GST-FAK-NT (aa 1-423), first half of FAK-NT (aa 1-205), and second half of FAK-NT protein (aa 206-422). Middle panel, the GST-FAK-NT proteins were isolated and analyzed by Coomassie staining. The GST and GST-FAK-NT proteins are marked by asterisks. GST-FAK-proteins were analyzed by Western blot with FAK antibody, specific to the second half of FAK-NT (clone 77; BD Transduction Laboratories). The full-length of GST-FAK-NT and terminal GST-FAK-NT (aa 206-422) domain reacted with FAK antibody (clone 77). The proteins were also analyzed with FAK 4.47 antibody (Upstate Biotechnology) specific to the first half of FAK-NT (lower panel). Full-length GST-FAK-NT and GST-FAK-NT (aa 1-205) were effectively recognized with 4.47 antibody. B, pull-down assay with GST-FAK-NT proteins. Baculovirus-expressed p53 protein was used for pull-down analysis with GST-FAK-NT proteins, shown in A. p53 protein binds to the second half of FAK-NT and not to the first half of FAK-NT. All experiments were repeated for three times, and the representative blot is shown. WB, Western blot.

View larger version (12K): [in this window] [in a new window]   FIG. 6. A, FAK expression blocks p53 apoptosis in SAOS-2 cells. SAOS-2 p53-null cells were transfected with p53-pCMV4 plasmid alone (1 µg), FAK-pcDNA3 plasmid alone (at doses of 0.5 and 1 µg), -FAK-NT plasmid alone (at doses of 0.5 and 1 µg), or FAK-pcDNA3 and -FAK-NT plasmids (with deleted N-terminal domain of FAK) at doses (0.5 and 1 µg) together with p53 (1 µg) plasmid. In all transfections, GFP was added to control for transfection efficiency. An equal amount of DNA was used for each transfection. 48 h after transfections, cells were collected for detection of apoptosis. Apoptosis was analyzed by Hoechst 33342 in GFP-expressing cells under a fluorescent microscope. Introduction of FAK plasmid blocked p53-induced apoptosis in a dose-dependent manner, and FAK-NT did not significantly affect the apoptosis in these cells. Apoptosis was analyzed in >300 cells per treatment in three independent experiments. Mean data ± S.E. values are shown. *, p < 0.05 significant inhibition of p53-induced apoptosis by FAK (0.5 µg) compared with cells without FAK expression; **, p < 0.001 significant inhibition of p53-induced apoptosis by FAK (1 µg) versus cells without FAK expression. B, expression of p53 protein and apoptosis in SAOS-2 cells. Upper panel, Western blot was performed for analysis of p53 expression with anti-p53 antibodies. Equal protein loading was analyzed with anti--actin antibodies. Lower panel, cells were co-transfected with p53 plasmid and GFP plasmids, as a control for transfection efficiency. After transfection, GFP-positive cells were sorted by a FACS cell sorter. Hoechst 33342-stained nuclei show p53-induced apoptosis in GFP-positive SAOS-2 cells. Apoptotic nuclei with fragmented DNA in GFP-positive cells are shown and marked by the white arrows.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The transactivation domain is an important functional part of p53, acting as a transcription activation domain (41, 70, 71). It has been shown that Mdm-2 (mouse double minute 2) protein binds to this region and negatively regulates p53 transcription functions (72, 73). p53 and Mdm2 comprise a feedback loop (74), since p53 regulates transcription of the Mdm2 gene (74), and Mdm-2 is a ubiquitin ligase causing proteosome-mediated p53 degradation (75). Similarly, it has been reported that p53 binds to the FAK promoter and affects its transcriptional activity (17). In the present study, FAK overexpression was able to block transcriptional activity of p53. Interestingly, we have shown in this report that FAK expression blocked p53-induced Mdm-2 luciferase activity. It is possible that the feedback regulation loop exists in regulation of FAK-p53-Mdm-2 activity, and this will be addressed in future studies.

View larger version (12K): [in this window] [in a new window]   FIG. 7. Overexpression of FAK inhibits p21-luciferase promoter activity in HCT 116 p53+/+ cells. A, co-transfection of HA-tagged FAK-pcDNA3 plasmid with wild type p21-pGL3 luciferase, containing two p53 binding sites inside the p21 promoter sequence or p21-mut-pGL3 luciferase without p53 binding site constructs, described in Ref. 41, and PRL-Renilla plasmid was performed in HCT p53-/- and HCT p53+/+ cells. An equal amount of DNA was used in each transfection. After 24 h, cells were lysed, and dual luciferase assays were performed, as described (see "Experimental Procedures"). The bars show mean data ± S.E. values of four independent experiments. The -FAK-NT-pcDNA3.1/Xpress plasmid, containing a deletion of the N-terminal FAK-NT fragment (aa 1-416) was co-transfected into p53+/+ cells together with the p21-wild type and p21-mutant-PGL3 luciferase constructs. Overexpression of wild type FAK significantly inhibits p53-mediated p21 transcriptional activity in HCT116 p53+/+ cells, whereas the -FAK-NT did not significantly decrease p21 luciferase activity. *, p < 0.05 compared with vector control p53+/+ HCT116 cells. B, FAK inhibits p21-luciferase promoter activity in HCT 116 p53+/+ cells in a dose-dependent manner. Co-transfection of different amounts (0.25, 0.5, and 1 µg) of HA-tagged FAK-pcDNA3 plasmid with the p21-pGL3 luciferase construct (1 µg) and PRL-Renilla plasmid (0.1 µg) was performed in HCT p53+/+ cells. An equal amount of DNA was used in each transfection. After 24 h, cells were lysed, and dual luciferase assays were performed, as described (see "Experimental Procedures"). The bars show mean data ± S.E. values of three independent experiments. FAK inhibited p21-luciferase promoter activity in a dose-dependent manner in HCT 116 p53+/+ cells. C, expression of HA-FAK and Xpress--FAK-NT proteins in HCTp53+/+ cells. Western blot (WB) analysis with anti-HA tag and Xpress antibodies detected HA-FAK and X-press--FAK-NT proteins, respectively. Equal protein loading was analyzed with anti-vinculin antibodies.

View larger version (12K): [in this window] [in a new window]   FIG. 8. A and B, overexpression of FAK inhibits MDM-2 luciferase promoter activity in HCT 116 p53+/+ cells. A, co-transfection of HA-tagged FAK-pcDNA3 plasmid was performed with MDM-2-pGL2 luciferase construct (see "Experimental Procedures") and PRL-Renilla plasmid in HCT p53-/- and HCT p53+/+ cells, as described in the legend to Fig. 7A. The bars show mean data ± S.E. values of three independent experiments. The -FAK-NT-pcDNA3.1/Xpress plasmid was co-transfected into p53+/+ cells together with the MDM-2 luciferase construct. Overexpression of wild type FAK significantly inhibits p53-mediated MDM-2 transcriptional activity in HCT116 p53+/+ cells, whereas the -FAK-NT did not significantly affect luciferase activity. *, p < 0.05 compared with vector control p53+/+ HCT116 cells. B, FAK inhibits MDM-2-luciferase promoter activity in HCT 116 p53+/+ cells in a dose-dependent manner. Co-transfection of different amounts of HA-tagged FAK-pcDNA3 plasmid (0.25, 0.5, and 1 µg) with MDM-2-pGL2 luciferase construct (1 µg) and PRL-Renilla plasmid (0.1 µg) was performed in HCT p53+/+ cells. An equal amount of DNA was used in each transfection. After 24 h, cells were lysed, and dual luciferase assays were performed, as described (see "Experimental Procedures"). The bars show mean data ± S.E. of three independent experiments. FAK inhibited MDM-2-luciferase activity in a dose-dependent manner in HCT 116 p53+/+ cells.

View larger version (19K): [in this window] [in a new window]   FIG. 9. A and B, overexpression of FAK inhibits p53-induced BAX luciferase promoter activity in HCT116 and SAOS-2 cells. HCT 116 p53-/- cells (A) and SAOS-2 cells (B) were transfected with 1 µg of p53-pCMV4 plasmid alone or together with HA-tagged FAK-pcDNA3 plasmid (1 µg) or -FAK-NT-pcDNA3 (1 µg) plasmids with the BAX-promoter luciferase construct (1 µg). The total amount of DNA was kept constant with pcDNA3 (Vector) DNA. PRL-Renilla plasmid (0.1 µg) was co-transfected together with the above plasmids and used as a normalization control. After 48 h, cells were lysed, and dual luciferase assays were performed, as described (see "Experimental Procedures"). The bars show mean data ± S.E. values of three independent experiments. Overexpression of wild type FAK significantly inhibits p53-mediated BAX-luciferase activity in both HCT116 and SAOS-2 cell lines, whereas -FAK-NT plasmid did not affect BAX-luciferase activity. *, p < 0.05 versus vector control sample.

Several p53 homologues have been discovered, such as p73 and p63 (reviewed in Refs. 82 and 83). These proteins have a similar domain structure: N-terminal transactivation domain, central DNA-binding domain, and C-terminal oligomerization domain (82). Unlike p53, these proteins undergo alternative splicing and have different isoforms for p63 (, , and ) and six isoforms for p73 protein (, , , , , and ) (83). Although p63 and p73 proteins share similarity with p53 structure and some of the p53 functions, the physiological functioning in the cells seems to be different (83). The absence of p73- binding with FAK suggests specificity p53-FAK signaling in the cells that can be explained by the main role of p53 in DNA damage response and FAK in survival pathways.

The N-terminal domain of FAK has homology with the FERM domain of many proteins (51, 84). FERM domains are present in such proteins as erythrocyte band 4.1 protein; ezrin, radixin, moesin (ERM) proteins; and talin (51). The FERM domains mediate protein interactions. The FAK FERM domain binds cytoplasmic tails of -integrin and growth factor receptors and can promote cell motility and signaling (21). One of the FAK family proteins that shares structural similarity with FAK protein is a PYK-2 (85), also known as a CADTK (50), cell adhesion kinase (49), or related adhesion focal tyrosine kinase (86). In contrast to FAK, PYK-2 is mainly expressed in hematopoietic, endothelial, and central nervous system cells (51). Interestingly, we did not detect direct binding of p53 to purified PYK-2, further suggesting specific FAK-p53 signaling. Although PYK-2 was shown to be localized in perinuclear region in fibroblasts (87) and in the nucleus in COS-7 cells (88), the binding partners and functions of FAK and PYK-2 are different in the cells (25).

There is increasing evidence for a distinct function of the N-terminal domain of FAK and for both cytoplasmic and nuclear function of p125^FAK. The N-terminal domain of FAK has also been reported to bind integrins, epidermal growth factor receptor, and platelet-derived growth factor receptor (21) and the death complex protein RIP (24). In addition, the FAK-NT protein caused rounding, detachment, and apoptosis in breast cancer BT-474 cells (43), and this N-terminal portion of FAK was reported to be localized to the nuclei (52, 53). The N-terminal domain of FAK was constitutively present in the nucleus of glioblastoma cells, and increased apoptosis caused aggregation of the N-terminal domain in the nucleus (32). Nuclear localization of FAK and the N-terminal fragment of FAK was also demonstrated in human endothelial cells (55). In these cells, the p50 N-terminal fragment of FAK was increased in the nuclear fraction during apoptosis (55). The N-terminal FAK localization in the nucleus was also reported in other cell lines, including basophilic leukemia cells (53), HEK 293 cells, and epithelial MDCK cells (52). Moreover, the highly conserved nuclear export signal, IALKLGCLEI was detected in the N-terminal domain of FAK (52). In addition, an inhibitor of nuclear export, leptomycin B, increased nuclear localization of exogenously expressed enhanced GFP-FAK, indicating that FAK can shuttle between the nuclear and cytoplasmic fractions (53).

Thus, binding of p53 to the N-terminal part of FAK may also affect binding of FAK with death receptor or growth factor receptor pathways, FAK localization, or cytoplasmic/nuclear transport and survival/apoptotic pathways. Interestingly, the N-terminal domain of FAK also interacts with the protein inhibitor of activated STAT1 (PIAS1), causing FAK sumoylation in the nucleus and increasing its autophosphorylation activity (25). The authors suggested that PIAS1-catalyzed sumoylation of FAK could represent an additional regulatory step controlling FAK functions in the nucleus (25). In addition, PIAS physically interacted and stimulated sumoylation of p53 in the nucleus (89, 90). Thus, it is possible to hypothesize that PIAS1-p53-FAK complexes may have a role in regulating multiple cellular functions, including transcriptional activities, protein localization, and signaling inside cells.

The results of the present study support the previous observations on the link between FAK and p53 in survival signaling (33). The authors found that in the absence of FAK function that transmitted survival signals from extracellular matrix, a p53-regulated pathway was activated through cytosolic phospholipase and one of the protein kinase C isoforms (33). This pathway could be suppressed by Bcl-2 and repressed by C-terminal dominant-negative p53 (33). Other authors supported the link between the FAK/p53 interaction by observing that suppression of FAK increased anoikis and blocking of p53 increased apoptosis in squamous carcinoma cells (37). In addition, overexpression of FAK correlated with overexpression of p53 in endometrial neoplasia (38). This is the first report to demonstrate a direct physical interaction of FAK and p53 and to map domains involved in the interaction and to demonstrate intracellular functioning of both proteins in regulating p53-mediated apoptosis and transcription that can provide a basis for future studies on FAK-p53-mediated signaling/apoptosis.

	FOOTNOTES

 
^* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. of Surgery, Health Science Center, P.O. Box 100286, 1600 SW Archer Rd., Gainesville, FL 32610-0286. Tel.: 352-265-0622; Fax: 352-338-9809; E-mail: cance{at}surgery.ufl.edu.

¹ The abbreviations used are: FAK, focal adhesion kinase; PBS, phosphate-buffered saline; GST, glutathione S-transferase; FAK-NT, FAK N-terminal domain; FAK-CD, FAK C-terminal domain; siRNA, small interfering RNA; MEF, mouse embryo fibroblast; GFP, green fluorescent protein; TRITC, tetramethylrhodamine isothiocyanate; TAD, transactivation domain; PYK, proline-rich tyrosine kinase; CADTK, calcium-dependent tyrosine kinase; PIAS1, protein inhibitor of activated STAT1; aa, amino acids; FACS, fluorescence-activated cell sorting.

	ACKNOWLEDGMENTS