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

J. Biol. Chem., Vol. 280, Issue 8, 6802-6815, February 25, 2005

This Article Abstract Full Text (PDF) All Versions of this Article: 280/8/6802    most recent M409180200v1 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 Ding, Q. Articles by Gladson, C. L. Search for Related Content PubMed PubMed Citation Articles by Ding, Q. Articles by Gladson, C. L. Social Bookmarking                      What's this?

p27^Kip1 and Cyclin D1 Are Necessary for Focal Adhesion Kinase Regulation of Cell Cycle Progression in Glioblastoma Cells Propagated in Vitro and in Vivo in the Scid Mouse Brain^*

Qiang Ding, J. Robert Grammer, Mark A. Nelson¶, Jun-Lin Guan||, Jerry E. Stewart, Jr., and Candece L. Gladson**

From the Departments of Pathology and ^**Cell Biology, the University of Alabama at Birmingham, Birmingham, Alabama 35294, the ^¶Department of Pathology, University of Arizona, Tucson, Arizona 85724, and the ^||Department of Molecular Medicine, College of Veterinary Medicine, Cornell University, Ithaca, New York 14853

Received for publication, August 11, 2004 , and in revised form, November 10, 2004.

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
We have reported previously that the expression of focal adhesion kinase (FAK) is elevated in glioblastomas and that expression of FAK promotes the proliferation of glioblastoma cells propagated in either soft agar or in the C.B.17 severe combined immunodeficiency (scid) mouse brain. We therefore determined the effect of FAK on cell cycle progression in these cells. We found that overexpression of wild-type FAK promoted exit from G₁ in monolayer cultures of glioblastoma cells, enhanced the expression of cyclins D1 and E while reducing the expression of p27^Kip1 and p21^Waf1, and enhanced the kinase activity of the cyclin D1-cyclin-dependent kinase-4 (cdk4) complex. Transfection of the monolayers with a FAK molecule in which the autophosphorylation site is mutated (FAK397F) inhibited exit from G₁ and reduced the expression of cyclins D1 and E while enhancing the expression of p27^Kip1 and p21^Waf1. Small interfering RNA (siRNA)-mediated down-regulation of cyclin D1 inhibited the enhancement of cell cycle progression observed on expression of wild-type FAK, whereas siRNA-mediated down-regulation of cyclin E had no effect. siRNA-mediated down-regulation of p27^Kip1 overcame the inhibition of cell cycle progression observed on expression of FAK397F, whereas down-regulation of p21^Waf1 had no effect. These results were confirmed in vivo in the scid mouse brain xenograft model in which propagation of glioblastoma cells expressing FAK397F resulted in a 50% inhibition of tumor growth and inhibited exit from G₁. Taken together, our results indicate that FAK promotes proliferation of glioblastoma cells by enhancing exit from G₁ through a mechanism that involves cyclin D1 and p27^Kip1.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
FAK¹ is a nonreceptor cytoplasmic tyrosine kinase. It has been shown to be activated on clustering of integrin receptors in the cell membrane or by the ligation of multiple growth factor receptors (1–3). Activation occurs through phosphorylation of Tyr³⁹⁷, either through autophosphorylation or Src-induced phosphorylation (4). Clustering of integrin receptors occurs during cell attachment, and the receptors localize to focal contacts (early adhesion complexes) where, in a temporally related manner, a signaling complex is assembled which includes FAK as well as cellular Src and other kinases, cytoskeletal proteins, and adaptor molecules (3, 5, 6). In nonmalignant and nontransformed cells, this localization of FAK to focal contacts or focal adhesions (mature adhesion complexes) has been shown to be necessary for its activation, indicating that FAK activation is regulated by integrin receptor engagement or cell adhesion in these cells. In contrast, in malignant or transformed cells, the activation of FAK does not appear to be regulated solely by cell adhesion or integrin receptor engagement. For example, we have shown the activation of FAK (phosphorylation of Tyr³⁹⁷) in glioblastoma cells propagated in suspension, suggesting that the mechanisms governing the activation of FAK in malignant or transformed cells may differ from those utilized by normal cells (7).

Previous studies examining the role of FAK in proliferation and regulation of the cell cycle have focused largely on NIH3T3 fibroblasts (8–10). In these cells, overexpression of wild-type FAK promotes cell proliferation, as assessed by bromodeoxyuridine (BrdUrd) labeling, and this is associated with promotion of the exit from G₁ by enhancement of the transcriptional activation of cyclin D1 (9). Specifically, FAK promotes the binding activity of an Ets transcription factor to the Ets B element in the cyclin D1 promoter and induces expression of the transcription factor, kruppel-like factor 8 (KLF8), which binds a GT box in the cyclin D1 promoter (9, 10). In addition, the activation of ERK was shown to be necessary for FAK-promoted cyclin D1 transcription in the NIH3T3 cells, because the MEK inhibitor PD98059, but not the JNK inhibitor curcumin, inhibited cyclin D1 transcription in cells overexpressing wild-type FAK (9). In contrast, expression of a truncated FAK (FAK-C14), which because of the lack of the carboxyl-terminal 14 residues does not localize to focal adhesions, inhibits BrdUrd labeling. Although the expression of FAK-C14 in the NIH3T3 cells was found to enhance the expression of the cyclin-dependent kinase (cdk) inhibitor p21^Waf1, p21^Waf1 was not necessary for FAK-C14 inhibition of cell cycle progression in these cells. The expression of p27^Kip1 was unaltered by the expression of FAK-C14 in the NIH3T3 cells.

FAK protein expression is elevated in a number of malignant tumors, including breast and colon cancer (11, 12). We and others have reported previously that FAK protein and activity (based on phosphorylation of Tyr³⁹⁷) are elevated in grades III and IV malignant astrocytoma tumors (anaplastic astrocytoma and glioblastoma, respectively), compared with the normal brain (7, 13). We have also shown that the overexpression of wild-type FAK promotes soft agar growth of glioblastoma cells (14), the expression of the mutant FAK397F inhibits soft agar growth of these cells (15), and that FAK is sufficient to promote Ras activity in glioblastoma cells propagated in aggregate suspension or as a monolayer (15). Because constitutively active Ras is known to promote soft agar growth of multiple cell types (16, 17), the ability of FAK to promote soft agar growth is likely caused, at least in part, by its activation of Ras.

Because FAK is known to promote proliferation by affecting the cell cycle in NIH3T3 cells, and as it is well established that the expression of molecules that regulate the cell cycle in cancer cells is frequently altered (18, 19), it is possible that the mechanism(s) by which FAK regulates the cell cycle is altered in cancer cells. Here, we report that wild-type FAK promotes the cell cycle progression of glioblastoma cells propagated both in vitro as a monolayer and in vivo in the C.B.17 scid mouse brain. Our results demonstrate that overexpression of wild-type FAK promotes exit from G₁ in glioblastoma cells propagated both in vitro as a monolayer and in vivo in the scid mouse brain. In common with the previous findings examining the effects of expression of FAK on normal cells, we found that FAK enhancement of cell proliferation in the glioblastoma cells requires ERK activity, and the promotion of the exit from G₁ requires cyclin D1; however, in glioblastoma cells we found that the FAK-mediated regulation of the exit from G₁ also involves p27^Kip1.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Reagents—PD98059, curcumin, and SB202247 were purchased from Calbiochem. The following purified rabbit polyclonal antibodies were purchased: anti-integrin _v directed toward the cytoplasmic tail (Chemicon, Temecula, CA); anti-FAK (Upstate Biotechnology, Lake Placid, NY); and anti-phospho-ERK (Cell Signaling-New England Biolabs, Beverly, MA); anti-cyclin A, anti-cyclin B1, anti-p27^Kip1, anti-p21^Waf1, anti-p57^Kip2, and anti-ERK IgG (Santa Cruz Biotechnology, Santa Cruz, CA). The following monoclonal antibodies (mAbs) were purchased: anti-actin (Sigma); anti-PCNA (Dako Corp., Carpinteria, CA); anti-c-Myc (Oncogene Research Products, Cambridge, MA); anti-glyceraldehyde-3-phosphate dehydrogenase (G3PDH) (Research Diagnostics, Inc., Flanders, NJ); anti-cdk4 (Chemicon International, Temecula, CA); anti-cyclin D1, anti-cyclin E, and anti-phospho-JNK IgG (Santa Cruz Biotechnology). The anti-KLF8 antiserum has been described previously (10). Rabbit anti-phosphospecific retinoblastoma (Rb) [pSer⁷⁹⁵] IgG and anti-Rb IgG directed toward the carboxyl terminus were purchased (Cell Signaling).

Cells and Constructs—U-251MG human glioblastoma cells stably overexpressing wild-type FAK in the tetracycline (TET)-inducible system (OFAK5) and the U-251MG cells stably expressing mutant FAK397F (OFAK397F-clones 18 and 3) in the TET-inducible system, as well as the empty vector clones TRE2, have been described previously (14, 15).

Analysis of Mouse Intracerebral Xenograft Tumors—C.B.17 scid mice obtained from the Frederick Cancer Center at the National Cancer Institute were injected intracerebrally with stereotactic assistance at 6 weeks of age with 1 x 10⁶ cells of the OFAK5, OFAK397F-18, or the OFAK397F-3 cell clone, and the injected animals were divided randomly into two groups, as described previously (14). Doxycycline (2 mg/ml in saline) or vehicle alone (1 ml of saline) were administered daily by intraperitoneal injection. At 8 days postinjection, mice were euthanized, and the brains were harvested and processed for either FACS analysis or isolation of human tumor cells by magnetic beads; snap frozen for homogenization and immunoblotting; or fixed, as described previously (14). Briefly, a single brain cell suspension for FACS analysis was prepared as follows. Xenograft brains were harvested, placed into Dulbecco's modified Eagle's medium with 1% bovine serum albumin and the following inhibitors: 100 µM phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 100 µM sodium vanadate, and 20 µg/ml TLCK, minced thoroughly (30 min/brain), passed through a 170-µm mesh, resuspended in the above buffer with fresh protease inhibitors; 500,000 cells were incubated with FITC-conjugated mAb anti-human specific MHC Class I IgG (Pharmingen) or with FITC-conjugated mouse IgG isotype control (Pharmingen) (30 min, 4 °C), washed, and subjected to FACS analysis. The mouse IgG isotype control was used to determine the background fluorescence. Magnetic bead isolation of human tumor cells was performed as follows. A single cell suspension of xenograft brain was prepared as above and incubated with mAb anti-human specific MHC Class I IgG that had been coupled to magnetic polystyrene Dynabeads, as recommended by the manufacturer for 30 min at 4 °C (Cellection Pan Mouse IgG kit, Dynal A.S., Oslo, Norway). The beads were washed, and the human tumor cells were harvested by DNase treatment, which cleaves the linker molecule, lysed in RIPA lysis buffer 1% deoxycholate, 1.0% Triton X-100, and 0.1% SDS in 0.01 M Tris base (pH 7.4) with 0.15 M NaCl) with the above protease inhibitors, centrifuged (35,000 rpm, 4 °C, 1 h), and the protein concentration determined in the supernatant. Brains were snap frozen in liquid nitrogen for immunoblot analysis of total brain lysates and then homogenized using a Polytron (Kinematica, Luzern, Switzerland) in 2 ml of RIPA lysis buffer with the above protease inhibitors, the lysates centrifuged (35,000 rpm, 4 °C, 1 h), and the concentration of the protein in the supernatant determined. For histologic analysis, brains were fixed in 4% buffered paraformaldehyde (4 h, 4 °C) followed by immersion in 6% sucrose overnight, then frozen and maintained at –70 °C until sectioning. Tumor volume was assessed as described previously (14); briefly, all brains were serially sectioned, and 8-µm sections were stained every 40 µm with mAb anti-human specific MHC Class I IgG (10 µg/ml, 2 h, 4 °C) (mAb W6/32, Serotec, Inc., Raleigh, NC), followed by a multilink secondary antibody conjugated to biotin and horseradish peroxidase, followed by incubation with streptavidin, and then 3,3'-diaminobenzidine substrate, as per the instructions in the kit from BioGenex (San Ramon, CA). Digital images were imported into Adobe Photoshop 5.0, stained areas in each section were quantitated as a pixel number, and pixel numbers for all sections from each brain were summed to obtain a total pixel number. The proliferation index or PCNA index was determined as described previously (14); briefly, two sections of each brain were incubated with 1 µg/ml PCNA mAb (1 h, 22 °C), followed by the secondary antibody. The detection procedure was as described above. The percentage of dead or apoptotic cells in each tumor was determined using a TUNEL assay, as described previously (14). Briefly, two tissue sections from each brain were reacted with biotinylated dNTP, plus terminal deoxynucleotidyltransferase enzyme, followed by biotin, streptavidin-horseradish peroxidase, and the 3,3'-diaminobenzidine substrate, according to the instructions provided with the terminal deoxynucleotidyltransferase-Frag EL DNA Fragmentation Detection kit (Oncogene Research Products).

Cell Cycle Analysis—Cells were harvested with trypsin, washed with cold phosphate-buffered saline, fixed in 70% ethanol, and kept at 4 °C overnight. Fixed cells were washed with phosphate-buffered saline and resuspended in 40 µg/ml RNase A (Sigma) and 40 µg/ml propidium iodide (Sigma) in phosphate-buffered saline (30 min, 22 °C). The cell suspension was then analyzed using a FACScan for DNA content (BD Biosciences, FACS Calibur).

BrdUrd Labeling—Cells were treated with doxycycline or vehicle for 4 days and then replated onto vitronectin-coated glass coverslips in serum-containing medium, plus/minus doxycycline and labeled overnight (18 h) with 10 µM BrdUrd. The next morning, cells were fixed, reacted with mAb anti-BrdUrd followed by a secondary horseradish peroxidase-conjugated antibody and substrate according the manufacturer's instructions (Oncogene Research Products). The percent of cells with dark brown nuclei were counted in six fields at 10x magnification.

Annexin V Apoptosis Assay—The Annexin V assay with propidium iodide labeling was performed on cells in culture as per the manufacturer's instructions (Pharmingen).

Immunoblotting—Cells were lysed in 1% Nonidet P-40 lysis buffer containing 137 mM NaCl, 2 mM EDTA, 10% glycerol, 1% Nonidet P-40, and 20 mM Tris-HCl, pH 8.0, with the above protease inhibitors (4 °C, 60 min), centrifuged (35,000 rpm, 1 h, 4 °C), and the protein concentration determined in the supernatant (20). Equivalent µg of protein from each lysate were electrophoresed on a disulfide-reduced 7.5% SDS-PAGE, transferred to Immobilon-P membrane (Millipore Corp., Bedford, MA), blocked with 5% buffered bovine serum albumin (3 h, 22°), and then reacted with primary antibody overnight at 4 °C, followed by 0.025 µg/ml horseradish peroxidase-conjugated secondary antibody (1 h, 22 °C), and the membrane was developed with the ECL system (Amersham Biosciences). For semiquantitative analysis of band intensity, a specific band on the autorad from three different exposures was subjected to densitometric analysis, and the densitometric readings were averaged; the background densitometric reading on the autorad was subtracted.

Soft Agar Growth Assays—Anchorage-independent growth assays were performed as described previously (14). Briefly, cells were treated with doxycycline or vehicle for 4 days, harvested and resuspended in a 0.3% agar (plus/minus doxycycline), poured onto plates containing a layer of 0.5% agar (plus/minus doxycycline), and then maintained at 37 °C, 5% CO₂ for 14 days. The resulting colonies were counted with a colony being defined as >10 cells. Samples were assayed in replicas of three, and the experiment was repeated twice.

siRNA Studies—Commercially available siRNA duplexes directed toward p27^Kip1, p21^Waf1, p57^Kip2, as well as cyclin E were purchased from Santa Cruz Biotechnology. siRNA duplexes directed toward cyclin D1 were purchased from Ambion (Austin, TX). Control siRNA duplexes were purchased from Dharmacon (Lafayette, CO). Lamin A/C siRNA (Dharmacon) was used to optimize the transfection conditions and efficiency, as described previously (21). Additional controls of nonspecific siRNA duplex or vehicle alone were evaluated and found not to alter p27^Kip1, p21^Waf1, p57^Kip2, cyclin D1, or cyclin E mRNA, and protein expression at the concentration of siRNA found to reduce the specific target message and protein optimally.

Immunofluorescence Analysis—The cellular localization of the mutant FAK397F was assessed using immunofluorescence analysis as described previously (21). Briefly, cells were propagated in complete medium with 2 µg/ml doxycycline or vehicle for 4 days, harvested with buffered EDTA, resuspended in adhesion assay buffer with 1 mM MgCl₂ and 100 µM MnCl₂ with 1% bovine serum albumin, and 40,000 cells were plated onto a coverslip coated previously with 5 µg/ml vitronectin (5 h, 37 °C), washed, fixed in 4% buffered paraformaldehyde, permeabilized with 0.3% Triton X-100, blocked, and reacted sequentially with both primary antibodies, followed by both secondary antibodies. Cells were viewed and photographed using a Nikon confocal microscope.

Real Time RT-PCR—Total RNA was extracted from cells using the RNeasy Mini Kit (Qiagen, Inc., Valencia, CA), and 1 µg of total RNA was reverse transcribed to cDNA, as described previously (21). All primers were commercially available and were purchased from Santa Cruz Biotechnology. Quantitative RT-PCR analysis was carried out with the SYBR® Green PCR Master Mix (Applied Biosystems, Foster City, CA) using the GeneAmp 5700 sequence detection system (Applied Biosystems), according to the manufacturer's instructions. The quantitative RT-PCR was performed using 4 µl of the synthesized cDNA, 12.5 µl of SYBR® Green PCR Master Mix, 1 µl of primer, and 7.5 µl of water in a final 25-µl volume. Samples were assayed in triplicate, and the values were normalized to the relative amounts of actin.

cdk4 Immunoprecipitation Kinase Assay—This was performed essentially as described previously (22). Cells were lysed in 50 mM Hepes (pH 7.5), 150 mM NaCl, 1 mM EDTA, 0.1% Nonidet P-40 detergent with 10% glycerol, 100 µM phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 100 µM sodium vanadate, and 20 µg/ml TLCK (1 h, 4 °C), centrifuged, and 700 µg of lysate was immunoprecipitated with mAb anti-cdk4, mAb anti-cyclin D1, or mouse IgG (overnight, 4 °C). Subsequently, the immunoprecipitates were washed twice with the Rb kinase buffer (50 mM Hepes (pH 7.5), 1 mM EGTA, 10 mM KCl, 10 mM MgCl₂, and 1 mM dithiothreitol), resuspended in the Rb kinase buffer with the addition of 10 mM ATP and 0.5 µg of recombinant Rb protein (QED Biosciences, Inc., San Diego), and incubated with gentle shaking (30 min, 30 °C). The kinase assay was stopped with the addition of SDS sample buffer, subjected to 12% SDS-PAGE, and immunoblotted with rabbit anti-phosphospecific Rb [pSer⁷⁹⁵] IgG, stripped, and reprobed with rabbit anti-Rb IgG.

Statistical Analysis—Data were analyzed using an unpaired t test (SigmaPlot 2000, SPSS, Inc.). A p value <0.05 was considered statistically significant. All experiments were repeated two to three times with consistent results.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Wild-type FAK Promotes Cell Cycle Progression and BrdUrd Labeling of Glioblastoma Cells Propagated in Vitro as a Monolayer—To investigate the effects of FAK on cell cycle progression of glioblastoma cells, we analyzed the effects of transfection with doxycycline-inducible wild-type FAK (OFAK5) and doxycycline-inducible mutant FAK (FAK397F) of cells grown in monolayer cultures. The percentage of cells in each phase of the cell cycle was estimated using FACS analysis of propidium iodide-stained cells. These studies were initially performed after serum starvation to synchronize the cell cycle. Under conditions of serum starvation, in which the cells were starved of serum for 2 days (0.4% fetal bovine serum) prior to doxycycline or vehicle administration and then cultured for 4 days in reduced serum (4% fetal bovine serum), the percentage of cells in S phase was higher in doxycycline-induced cells expressing wild-type FAK (26%) than vehicle controls (21%), and the percentage of cells in G₀/G₁ was lower (48% versus 54%) (Fig. 1, A and B). Under conditions of serum starvation, the percentage of cells in S phase was lower in doxycycline-induced cells expressing FAK397F (10% versus 22%), and the percentage of cells in G₀/G₁ was higher (74% versus 56%) (Fig. 1, C and D). Similar results were observed when the experiments were performed in complete medium (10% fetal bovine serum). The percentage of cells in S phase was higher on expression of wild-type FAK (46% versus 33%), and the percentage of cells in G₀/G₁ was lower (40% versus 49%) (Fig. 2, A and B), whereas the percentage of cells in S phase was lower on expression of FAK397F (11.5% versus 31.5%) and the percentage of cells in G₀/G₁ was higher (73% versus 52%) (Fig. 2, C–F; values for the two mutant FAK397F clones averaged for the histogram in E and F). The percentage of cells in G₂/M was not altered on expression of either wild-type or mutant FAK in cells cultured under serum-starved conditions or in complete medium (Figs. 1 and 2). These results are consistent with those of Zhao et al. (8), who also found that serum starvation did not significantly alter the effect of FAK on cell cycle progression in NIH3T3 cells and indicate that FAK promotes exit from G₁ in glioblastoma cells in monolayer culture, This conclusion was supported by the results obtained on analysis of BrdUrd labeling of cells cultured in complete medium, which showed that expression of wild-type FAK resulted in a 58% higher level of BrdUrd labeling than in cells that were not treated with doxycycline (p = 0.001) (Fig. 3) and that expression of FAK397F resulted in a 70% reduction in BrdUrd labeling compared with the cells not treated with doxycycline (p < 0.001) (Fig. 3).

View larger version (17K): [in this window] [in a new window]   FIG. 1. FAK promotes cell cycle progression of glioblastoma cells propagated in vitro as a monolayer. OFAK5 and OFAK397F-18 clones were serum starved in 0.4% serum-containing medium for 2 days, and then parallel cultures were treated with 2 µg/ml doxycycline (Dox) or vehicle (saline) in reduced serum-containing medium (4%) for 4 days, followed by propidium iodide labeling and FACS analysis for DNA content using a BD Biosciences, FACS Calibur, as described under "Experimental Procedures." Note that the scale for cell number differs in A and C, plus/minus doxycycline administration. In B and D, the percentage of cells in G0/G1, S, and G2/M phases is plotted as a histogram, and the statistical significance of differences in each phase when cells were doxycycline treated versus vehicle treated is indicated by the p value; p < 0.05 was considered significant.

View larger version (34K): [in this window] [in a new window]   FIG. 2. FAK promotes cell cycle progression of glioblastoma cells propagated as a monolayer in complete medium. OFAK5, OFAK397F-3, and OFAK397F-18 clones propagated in complete medium were replated, and parallel cultures were administered 2 µg/ml doxycycline (Dox) or vehicle in complete medium for 4 days. Cells were then labeled with propidium iodide and analyzed for DNA content by FACS analysis using a BD Biosciences FACS Calibur, as described under "Experimental Procedures." Note that the scale for cell number differs in A and C, plus/minus doxycycline administration. In B, E, and F, the percentage of cells in G0/G1, S, and G2/M phases is plotted as a histogram, and the statistical significance of differences in each phase when cells were doxycycline treated versus vehicle treated is indicated by the p value; p < 0.05 was considered significant.

View larger version (14K): [in this window] [in a new window]   FIG. 3. FAK promotes BrdUrd incorporation into glioblastoma cells. OFAK5, OFAK397F-18, and TRE4 (control) clones propagated in complete medium were harvested and replated in complete medium, and parallel cultures were administered 2 µg/ml doxycycline or vehicle for 4 days and then harvested and replated onto vitronectin-coated coverslips and labeled with 10 µM BrdUrd (BrdU) overnight. Subsequently, the cells on the coverslips were washed, stained with mAb anti-BrdUrd, followed by a horseradish peroxidase-labeled secondary antibody, coverslipped, and the cells with dark brown-labeled nuclei counted in six 10x magnification fields/coverslip.

View larger version (56K): [in this window] [in a new window]   FIG. 4. c-Myc-tagged mutant FAK397F localizes to focal adhesions. The OFAK397F-18 clone propagated in complete medium was replated, and parallel cultures were administered 2 µg/ml doxycycline (Dox) or vehicle in complete medium for 4 days, plated onto coverslips coated with vitronectin overnight in complete medium, permeabilized, and reacted sequentially with mAb anti-c-Myc (B and D), Alexa 488-conjugated anti-mouse IgG, anti-integrin v IgG (A and C), and Alexa 594-conjugated anti-rabbit IgG, followed by Hoescht nuclear stain. Cells were analyzed and photographed using a Nikon laser confocal microscope. The arrows denote focal adhesions. Magnification, x100.

View larger version (62K): [in this window] [in a new window]   FIG. 5. Modulation of the levels of the cdk inhibitors p27Kip1 and p21Waf1, as well as cyclins D1 and E, by the expression of wild-type FAK or mutant FAK397F in glioblastoma cells. Parallel cultures of TRE2 (control), OFAK397F-3, OFAK397F-18, and OFAK5 clones were administered 2 µg/ml doxycycline (Dox) or vehicle in complete medium for 4 days, lysed, and Western blotted with the indicated antibodies, as described under "Experimental Procedures."

View larger version (38K): [in this window] [in a new window]   FIG. 6. Down-regulation of p27Kip1 blocks the inhibitory effect of mutant FAK397F on the cell cycle progression of glioblastoma cells. Parallel cultures of the OFAK397F-18 clone propagated in complete medium were administered 2 µg/ml doxycycline (Dox) or vehicle for 4 days and then transfected with duplex siRNA directed toward p27Kip1, p21Waf1, or p57Kip2, as described under "Experimental Procedures." A, post-transfection (48 h) the cells were harvested and RNA extracted for real time RT-PCR using forward and reverse primers directed specifically toward p27Kip1, p21Waf1, or p57Kip2. The relative expression of these mRNAs was normalized to actin. B, cells were detergent lysed and the lysate subjected to immunoblotting with the indicated antibodies. C and D, the cells were harvested, propidium iodide labeled, and the DNA content analyzed by FACS analysis as described in the legend for Fig. 1. In D, the percentage of cells in G0/G1, S, and G2/M phases is plotted as a histogram, and the statistical significance of differences in each phase when cells were doxycycline treated versus vehicle treated is indicated by the p value; p < 0.05 was considered significant.

View larger version (41K): [in this window] [in a new window]   FIG. 7. Down-regulation of cyclin D1 inhibits the promotion of cell cycle progression observed with the expression of wild-type FAK in glioblastoma cells. Parallel cultures of the OFAK5 clone were propagated in complete medium and administered 2 µg/ml doxycycline (Dox) or vehicle for 4 days and then siRNA was directed toward cyclin D1, E, or scrambled duplex oligonucleotides transiently transfected into the cells, as described under "Experimental Procedures." A, post-transfection (48 h), the cells were harvested and RNA extracted for real time RT-PCR using forward and reverse primers directed specifically toward cyclin D1 or E, and the relative expression of these mRNAs were normalized to actin. B, the cells were detergent lysed for Western blot analysis with the antibodies indicated. C and D, the cells were harvested, propidium iodide-labeled, and FACS analysis for DNA content was performed as described in the legend for Fig. 1. In D, the percentage of cells in G0/G1, S, and G2/M phases is plotted as a histogram, and the statistical significance of differences in each phase when cells were doxycycline treated or vehicle treated is indicated by the p value; p < 0.05 was considered significant.

View larger version (21K): [in this window] [in a new window]   FIG. 8. Expression of wild-type FAK promotes an enhanced activity of the cyclin D1-cdk4 complex in glioblastoma cells. Parallel cultures of the OFAK5 clone were propagated in complete medium and administered 2 µg/ml doxycycline (Dox) or vehicle for 4 days and then lysed. A and B, equivalent amount of lysate (700 µg) was immunoprecipitated (IP) with mAb anti-cdk4 or mouse IgG, washed, the immunoprecipitated protein subjected to the cdk4 kinase assay, followed by 12% SDS-PAGE and immunoblotting with anti-Rb [pSer795] IgG (A), stripped, and reprobed with anti-total Rb IgG (B), as described under "Experimental Procedures." C, an equivalent amount of lysate (30 µg) was subjected to SDS-PAGE and immunoblotted with mAb anti-glyceraldehyde-3-phosphate dehydrogenase (G3PDH) as a control for the quality of the lysate.

View larger version (39K): [in this window] [in a new window]   FIG. 9. Inhibition of glioblastoma cell proliferation in animals injected intracerebrally with the OFAK397F-18 or -3 clone and administered doxycycline (Dox). Cell clones (1 x 106 cells) were injected intracerebrally into C.B.17 scid mice and allowed to propagate for 8 days, and animals were administered doxycycline or vehicle daily, euthanized, the brains harvested, frozen, and subsequently lysed for immunoprecipitation (IP) and Western blot analysis (A–C). Lysate (400 µg) from each brain was immunoprecipitated with rabbit anti-FAK IgG, subjected to SDS-PAGE, and then immunoblotted with mAb anti-c-Myc (A), stripped, and reprobed with rabbit anti-FAK IgG (B). Whole brain lysate (40 µg) from each animal brain was immunoblotted with mAb anti-actin (C). In another experiment, a single cell suspension was isolated from each brain and the percent tumor cells determined by reaction with a human-specific mAb anti-MHC Class I IgG, followed by FACS analysis (D and E). The percentage of tumor cells in each animal brain (represented as a single dot) was plotted as a scattergram.

To determine whether the decrease in tumor cell number and tumor volume detected on expression of the mutant FAK397F in glioblastoma cells propagated in the scid mouse brain is the result of inhibition of cell proliferation, the number of proliferating cells in two sections from each animal brain was estimated using immunohistochemical staining of PCNA. In animals injected with the OFAK397F-18 clone and administered doxycycline, a 54% reduction in PCNA-positive tumor nuclei was found compared with the percentage of PCNA-positive tumor nuclei in the animals administered vehicle (mean percent positive PCNA-labeled tumor nuclei = 30% ± 3.1% in animals administered vehicle; and mean percent positive tumor nuclei = 14% ± 4.5% in animals administered doxycycline; p = 0.01) (Fig. 10, A and B). A TUNEL assay on two sections from each animal brain indicated no change in the percentage of TUNEL-positive cells in the two groups of animals (mean percent positive cells = 2% ± 0.1% vehicle administration; and mean percent positive cells = 2% ± 0.2% with doxycycline administration; p = 0.77) (Fig. 10, C and D). These data suggest that the decreased tumor cell number and tumor volume found on expression of FAK397F is not caused by a promotion of apoptosis, but rather that expression of FAK397F inhibits the proliferation of glioblastoma cells in vivo.

View larger version (120K): [in this window] [in a new window]   FIG. 10. Decreased mAb PCNA labeling in glioblastoma tumor cells expressing mutant FAK397F and propagated in vivo in the scid mouse brain. C.B.17 scid mice were injected intracerebrally with the OFAK397F-18 clone (1 x 106 cells), the tumors allowed to propagate for 8 days, and the animals administered doxycycline (Dox) or vehicle daily, euthanized, the brains harvested, fixed, frozen and serially sectioned. Two sections of each mouse brain were reacted with mAb PCNA (A and B) or subjected to TUNEL analysis (C and D). A and B, positive tumor nuclei (brown stain) were counted in six 20x magnification fields and analyzed as the percent positive tumor nuclei in each section. Arrows denote mAb PCNA-labeled tumor nuclei. C and D, positive nuclei (brown stain) in six 20x magnification fields from areas of tumor from each section were counted and analyzed as the percent positive tumor nuclei. Arrows denote TUNEL-labeled tumor nuclei.

View larger version (71K): [in this window] [in a new window]   FIG. 11. Decreased expression of p27Kip1 and increased expression of cyclin D1 with the expression of wild-type FAK in glioblastoma cells propagated in the scid mouse brain. C.B.17 scid mice were injected intracerebrally with the OFAK5 clone (1 x 106 cells), the tumors were allowed to propagate for 8 days, and the animals were administered doxycycline (Dox) or vehicle daily, followed by euthanasia, harvesting of the brains, a single cell suspension isolated which was reacted with mAb anti-human specific MHC Class I IgG coupled to magnetic beads, and the human tumor cells lysed in RIPA buffer with protease inhibitors. Equivalent µg of lysate from each animal brain were subjected to disulfide-reduced SDS-PAGE and immunoblotted with the indicated antibodies, as described under "Experimental Procedures."

View larger version (74K): [in this window] [in a new window]   FIG. 12. Increased expression of p27Kip1 and decreased expression of cyclin D1 with expression of mutant FAK397F in glioblastoma cells propagated in the scid mouse brain. C.B.17 scid mice were injected intracerebrally with the OFAK397F-18 clone (1 x 106 cells), the tumors allowed to propagate for 8 days, and the animals were administered doxycycline (Dox) or vehicle daily, euthanized, the brains harvested, a single cell suspension isolated which was reacted with mAb anti-human specific MHC Class I IgG coupled to magnetic beads and the human tumor cells lysed in RIPA buffer with protease inhibitors. Equivalent µg of lysate from each animal brain were subjected to disulfide-reduced SDS-PAGE and immunoblotted with the indicated antibodies, as described under "Experimental Procedures."

View larger version (59K): [in this window] [in a new window]   FIG. 13. Elevated expression of the transcription factor KLF8 with expression of wild-type FAK in glioblastoma cells propagated in vitro and in vivo. A and B, parallel cultures of the following clones OFAK5, OFAK397F-18, and TRE2 propagated in complete medium were administered 2 µg/ml doxycycline (Dox) or vehicle for 4 days, and then detergent lysed. C–F, C.B.17 scid mice were injected intracerebrally with the OFAK5 or the OFAK397F-18 clone (1 x 106 cells), administered doxycycline or vehicle for 8 days, the animals euthanized, the brains harvested, a single cell suspension isolated that was reacted with mAb anti-human-specific MHC Class I IgG coupled to magnetic beads, and the human tumor cells were lysed in RIPA buffer with protease inhibitors. Equivalent µg of lysate were subjected to SDS-PAGE and immunoblotted with a 1:2,500 dilution of anti-KLF8 antisera, stripped, and reprobed with mAb anti-glyceraldehyde-3-phosphate dehydrogenase (G3PDH), as described under "Experimental Procedures."

View larger version (70K): [in this window] [in a new window]   FIG. 14. ERK activity is necessary for FAK promotion of glioblastoma cell proliferation in soft agar. A, parallel cultures of the OFAK5 clone propagated in complete media were administered 2 µg/ml doxycycline (Dox) or vehicle for 4 days, the cells harvested, resuspended in 0.3% agar, and then plated onto 0.5% agar for 14 days (37 °C, 5% CO2), and the colonies were counted. B, parallel cultures of the cell clones (OFAK5 and OFAK397F-18) propagated in complete medium with 2 µg/ml doxycycline or vehicle for 4 days were detergent lysed. Equivalent µg of lysate were subjected to SDS-PAGE and immunoblotted with the indicated antibodies. C and D, the OFAK5 and OFAK397F-18 clones were injected intracerebrally (1 x 106 cells) into C.B.17 scid mice, the animals were administered doxycycline or vehicle for 8 days, euthanized, the brains harvested, a single cell suspension isolated which was reacted with mAb anti-human-specific MHC Class I IgG coupled to magnetic beads, and the human tumor cells lysed in RIPA buffer with protease inhibitors. Equivalent µg of lysate were subjected to SDS-PAGE and immunoblotted with the indicated antibodies.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
In this paper we demonstrate that FAK promotes cell cycle progression of glioblastoma cells (grade IV malignant astrocytoma) when propagated in vitro as a monolayer and in vivo in the scid mouse brain by promoting exit from G₁. Furthermore, we show that FAK regulation of cell cycle progression in these cells requires p27^Kip1 and cyclin D1 and that FAK induces expression of the KLF8 transcription factor, which is known to promote cyclin D1 transcription. As a functional correlate of these observations, we show that FAK promotes the activity of the cyclin D1-cdk4 complex. Lastly, we demonstrate that the promotion of proliferation by wild-type FAK in glioblastoma cells requires ERK activity. These findings confirm prior studies examining FAK regulation of the cell cycle, which have shown that FAK promotes exit from G₁ in NIH3T3 cells by promoting cyclin D1 transcription (8–10) and extend these studies by identifying the involvement of a cdk inhibitory protein and demonstrating FAK promotion of cell cycle progression in vivo.

A requirement for p27^Kip1 in the inhibitory effect of FAK397F on the cell cycle progression of glioblastoma cells is supported by several lines of evidence, including that expression of FAK397F increased the level of p27^Kip1 protein expression, the expression of wild-type FAK reduced the level of p27^Kip1 protein, and down-regulation of p27^Kip1 with siRNA blocked the inhibitory effect of FAK397F. Deregulation of p27^Kip1 protein is a common finding in cancer cells and can be associated with altered proteolytic degradation, mislocalization in the cytoplasm, or altered protein binding (18, 26). Deletion of the p27^Kip1 gene typically does not account for the low levels of p27^Kip1 protein (18, 26). Loss of PTEN function in malignant or cancer cells can affect p27^Kip1 levels, with PTEN regulating the ubiquitin-dependent degradation of p27^Kip1 through the ubiquitin E3 ligase SCF^SKP2 (27). Reexpression of the PTEN gene in glioblastoma cells has been shown to inhibit S phase entry by recruitment of p27^Kip1 into a cyclin E-cdk2 complex (28). Overexpression of p27^Kip1 in U-251MG glioblastoma cells propagated in vitro suppresses cell growth (29), and, in astrocytoma biopsies, the levels of p27^Kip1 protein are correlated inversely with tumor grade and are a strong predictor of survival (30–32). Our finding of an involvement of p27^Kip1 differs from those reported by Zhao et al. (8), who did not find a change in the level of p27^Kip1 protein on expression of wild-type FAK or FAK-C14 in the NIH3T3 cells. Although this difference may suggest some cell type specificity in the mechanism by which FAK regulates the cell cycle, several factors, including mutation of the p53 gene, deletion of the ink4B^(p15) and ink4A^(p16) tumor suppressor genes, and mutation of the PTEN gene impairing function in the U-251MG glioblastoma cells (33–35), could all contribute to differences in the mechanisms whereby FAK regulates the cell cycle in the U-251MG glioblastoma cells compared with the NIH3T3 cells.

Although expression of FAK397F and wild-type FAK altered the expression of p21^Waf1 protein in the glioblastoma cells, p21^Waf1 was not necessary for the inhibitory effect of FAK397F on the cell cycle progression of the glioblastoma cells, as siRNA directed toward p21^Waf1 did not block this inhibitory effect. Similarly, Zhao et al. (8) reported decreased levels of p21^Waf1 protein on expression of wild-type FAK and increased levels on expression of FAK-C14 in the NIH3T3 cells, but they found that p21^Waf1 was not necessary for the inhibitory effect of FAK-C14 on the cell cycle in fibroblasts because expression of this construct in p21-null fibroblasts resulted in inhibition of cell cycle progression.

We also found that cyclin D1 is necessary for FAK promotion of cell cycle progression in the glioblastoma cells because expression of wild-type FAK increased, and expression of the mutant FAK397F decreased cyclin D1 protein levels, and down-regulation of cyclin D1 with siRNA inhibited the ability of wild-type FAK to promote cell cycle progression. Down-regulation of cyclin E had no effect. Other investigators have shown that cyclin D1 is necessary for FAK promotion of cell cycle progression in NIH3T3 cells and that FAK induces the expression of the KLF8 transcription factor and promotes cyclin D1 transcription through the binding of KLF8 to the GT box in the cyclin D1 promoter (10). Based on the latter findings, we examined KLF8 protein levels in the glioblastoma cells and found that the expression of wild-type FAK induced KLF8 expression, and the expression of the mutant FAK397F inhibited KLF8 expression in cells propagated both in vitro and in vivo in the scid mouse brain. These results suggest that FAK likely regulates cyclin D1 transcription in glioblastoma cells in part through the induction of KLF8. FAK also promotes cyclin D1 transcription by inducing binding of the Ets transcription factor to the cyclin D1 promoter in the NIH3T3 cells (9). A focus of future work is to determine whether FAK induction of KLF8 expression in the glioblastoma cells promotes cyclin D1 transcription and the role of the Ets transcription factors. We and others have shown previously that the level of cyclin D1 protein is elevated in anaplastic astrocytoma (grade III malignant astrocytoma) and in glioblastoma (grade IV malignant astrocytoma) tumor biopsies compared with the normal brain (7, 36). The elevation of cyclin D1 protein is consistent with the altered regulation of cell proliferation seen in these tumors and our observation that FAK also promotes the activity of the cyclin D1-cdk4 complex.

FAK signaling in vitro is thought to require its localization to focal contacts/focal adhesions (3). The FRNK construct, which corresponds to the carboxyl-terminal region of FAK and contains the FAT domain, inhibits FAK signaling by competing with endogenous FAK for localization to focal contacts/focal adhesions (37). FRNK expression also disrupts the formation of Src·FAK complexes in cells because of the inhibition of FAK phosphorylation on Tyr³⁹⁷, the site of Src SH2 domain binding (for review, see Ref. 3). We show here that the mutant FAK397F localizes to focal adhesions in the adherent glioblastoma cells, thus, likely the mutant FAK397F competes with the endogenous FAK for focal adhesion localization, thereby inhibiting the activation and signaling of the endogenous FAK.

Expression of the mutant FAK397F in the glioblastoma cells did not induce apoptosis. Down-regulation of FAK expression with antisense oligonucleotides can induce apoptosis in non-glial derived cancer cells (38). In anchorage-dependent p53 wild-type cells FAK promotes cell survival by suppressing p53-mediated apoptosis (39). Furthermore, integrin ₅₁ engagement can send a prosurvival signal through FAK to JNK which does not require ERK in anchorage-dependent, p53 wild-type cells (40). In the latter two studies, expression of the FAT domain of FAK was used to inhibit the endogenous FAK signaling and serum starvation to induce apoptosis. FRNK expression in these cells did not increase apoptosis induced by serum starvation. The FRNK construct includes the proline-rich (PR) regions, PR1 and PR2, whereas the FAT construct does not contain either PR1 or PR2 (41). A docking molecule p130CAS binds to PR1 of FAK through its SH3 domain (42). To determine whether p130CAS and the PR1 region of FAK were important in the different effects seen with expression of FAT or FRNK, the SH3 domain of p130CAS was mutated, and expression of this construct increased apoptosis induced by serum starvation. This along with other data lead to the conclusion that the interaction of p130CAS with FAK through the PR1 domain of FAK may play a role in protecting anchorage-dependent cells from apoptosis induced by serum starvation. Therefore, the absence of apoptosis with expression of the mutant FAK397F in the glioblastoma cells is likely the result of the function of the PR1 domain in the mutant FAK397F, although the anchorage-independence and p53 mutant status of these cells also may contribute to this effect.

FAK promotion of cell cycle progression in NIH3T3 cells requires integrin engagement and ERK activity (9). Integrin engagement can activate ERK through FAK as well as through a FAK-independent mechanism (43–45). We show here that the promotion of glioblastoma cell proliferation with the expression of wild-type FAK in cells propagated in soft agar requires ERK activity, but not JNK activity. In soft agar growth integrin receptors are likely not engaged in cell-matrix interactions; thus, the cell proliferation promoted by FAK and ERK signaling in the soft agar growth of glioblastoma cells is independent of integrin engagement. p27^Kip1 can be regulated by ERK and also by signaling molecules upstream of ERK, such as Ras and MEK (46–48). We have reported recently that FAK promotes Ras activity in the glioblastoma cells when propagated in suspension or as a monolayer (15). Furthermore, we show here that expression of the wild-type FAK increased ERK activity, but not JNK activity, in the glioblastoma cells, whereas the expression of mutant FAK397F decreased ERK activity. In glioblastoma and NIH3T3 cells, FAK promotion of ERK activity appears to be necessary for cell proliferation or cyclin D1 transcription. Yet, p27^Kip1 protein levels are not regulated by FAK in the NIH3T3 cells, suggesting that FAK regulates p27^Kip1 protein levels in the glioblastoma cells through a mechanism that is at least in part independent of ERK, and this is a focus of future work. The increased ERK activity we observed on expression of wild-type FAK in the glioblastoma cells likely promotes binding of an Ets transcription factor to the cyclin D1 promoter because the Ets transcription factors are nuclear effectors of the Ras-mitogen-activated protein kinase pathway (49).

In summary, our results taken together with those reported in the literature suggest a model in which elevated FAK expression in malignant cells promotes Ras activity and thereby ERK activity, reduces p27^Kip1 protein level in a mechanism yet to be defined, and induces expression of the KLF8 transcription factor and thereby promotes cyclin D1 transcription. Together, these effects result in exit from G₁.

	FOOTNOTES

 
^* This work was supported by NCI National Institutes of Health Grants CA59958 and CA97110 (to C. L. G.) and CA70145 (to M. A. N.) and by a National Brain Tumor Foundation grant (to Q. D.). 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.

Present address: Dept. of Medicine, Division of Pulmonary and Critical Care Medicine, University of Alabama at Birmingham, 1900 University Blvd., 215 THT, Birmingham, AL 35294.

To whom correspondence should be addressed: Dept. of Pathology, Division of Neuropathology, University of Alabama at Birmingham, LHRB 567, 701 South 19th St., Birmingham, AL 35294-0007. Tel.: 205-975-7847; Fax: 205-934-7346; E-mail: gladson{at}uab.edu.

¹ The abbreviations used are: FAK, focal adhesion kinase; cdk, cyclin-dependent kinase; BrdUrd, bromodeoxyuridine; ERK, extracellular signal-regulated kinase; FACS, fluorescence-activated cell sorter; FITC, fluorescein isothiocyanate; KLF8, kruppel-like factor 8; mAb, monoclonal antibody; MEK, mitogen-activated protein kinase/extracellular signal-regulated kinase kinase; MHC, major histocompatibility complex; PCNA, proliferating cell nuclear antigen; PR, proline rich; PTEN, PTEN phosphatase; Rb, retinoblastoma protein; Rb [pSer⁷⁹⁵], anti-phosphospecific retinoblastoma protein at Ser⁷⁹⁵; RT, reverse transcription; siRNA, small interfering RNA; scid, severe combined immunodeficiency; TET,