4221
Journal of Cell Science 113, 4221-4230 (2000) Printed in Great Britain © The Company of Biologists Limited 2000 JCS1656
p125 focal adhesion kinase promotes malignant astrocytoma cell proliferation in vivo Dongyan Wang1, J. Robert Grammer1, Charles S. Cobbs2, Jerry E. Stewart, Jr1, Zhiyong Liu1, Rosaline Rhoden1, Timothy P. Hecker1, Qiang Ding1 and Candece L. Gladson1,* 1The
Department of Pathology, Division of Neuropathology and 2The Department of Surgery, Division of Neurosurgery, The University of Alabama at Birmingham, Birmingham, AL 35294, USA
*Author for correspondence (e-mail:
[email protected])
Accepted 3 October; published on WWW 7 November 2000
SUMMARY p125 focal adhesion kinase (p125FAK) is a cytoplasmic tyrosine kinase that is activated upon engagement of integrin cell adhesion receptors, and initiates several signaling events that modulate cell function in vitro. To determine the biologic role of p125FAK in malignant astrocytic tumor cells, U-251MG human malignant astrocytoma cells were stably transfected with p125FAK cDNA using the TET-ON system, and stable clones isolated that exhibited an estimated 5- or 20-fold increase in p125FAK expression on administration of 0.1 or 2.0 µg/ml doxycycline, respectively. In vitro studies demonstrated that induction of p125FAK resulted in a 2- to 3-fold increase in cell migration, increased p130CAS phosphorylation, localization of exogenous p125FAK to focal adhesions, and a 2-fold increase in soft agar growth. To determine the role of p125FAK in vivo, clones were
injected stereotactically into the brains of scid mice. A 4.5fold estimated increase in p125FAK expression was induced by administration of doxycycline in the drinking water. Analysis of xenograft brains demonstrated that, upon induction of p125FAK, there was a 1.6- to 2.8-fold increase in tumor cell number, and an increase in mAb PCNAlabeling of tumor cells in the absence of a change in the apoptotic index. Compared to normal brain, the expression of p125FAK was elevated in malignant astrocytic tumor biopsies from patient samples. These data demonstrate for the first time that p125FAK promotes tumor cell proliferation in vivo, and that the underlying mechanism is not associated with a reduction in apoptosis.
INTRODUCTION
progression of cells through the G1 phase of the cell cycle (Oktay et al., 1999). Phosphatidylinosityl-3-OH kinase (PI3kinase) also binds to activated p125FAK at Y397 and this interaction may be necessary for the activation of PI3-kinase (Chen and Guan, 1994; Ling et al., 1999). The p125FAK/Src complex phosphorylates several substrates, such as paxillin and possibly p130CAS, in fibroblasts (Vuori et al., 1996). Recent in vitro studies have suggested that p125FAK plays an important role in cell proliferation and/or cell survival in vitro. One such study demonstrated that overexpression of wild-type p125FAK resulted in an accelerated G1 to S transition in CHO cells, suggesting a role for p125FAK in the promotion of cell proliferation (Zhao et al., 1998). Several different types of in vitro studies have suggested that p125FAK promotes cell survival; for example (1) antibody or peptide blocking of p125FAK binding to integrins resulted in rapid cell death of fibroblasts (Hungerford et al., 1996), (2) antisense oligonucleotides directed toward p125FAK mRNA resulted in apoptosis of cancer cells (Xu et al., 1996), (3) constitutively active forms of p125FAK rescued cells from programmed cell death (Frisch et al., 1996), and (4) during apoptosis, caspases were found to cleave p125FAK into non-functional FRNK-like polypeptides (Wen et al., 1997; Gervais et al., 1998).
It has been established in vitro that cell contacts with the extracellular matrix that are mediated through the integrin family of cell adhesion receptors serve as gateways to transmit biological signals from the extracellular matrix into the cell (Hynes, 1992). These signals regulate cell proliferation and gene transcription, as well as other biologic functions. One molecule that appears to be involved in early integrin signaling is the cytoplasmic tyrosine kinase, p125focal adhesion kinase (p125FAK) (Giancotti and Ruoslahti, 1999). In vitro, integrin clustering in the cell membrane during cell attachment results in a rapid increase in p125FAK phosphorylation (Kornberg et al., 1992). Auto-phosphorylation of p125FAK at Y397 provides the docking site for another tyrosine kinase, Src, through its SH2 domain (Schaller et al., 1994; Xing et al., 1994). Src subsequently phosphorylates p125FAK at Y925, resulting in the recruitment of the Grb2 adaptor protein to p125FAK and, in some cells, activation of the ras pathway and MAPKinase, which are implicated in the control of cell proliferation (Schlaepfer et al., 1994; Schlaepfer et al., 1999; Schlaepfer and Hunter, 1997). Integrin activation of p125FAK may also signal jun NH2-terminal kinase and result in
Key words: p125FAK, Astrocytoma, Glioma, Glioblastoma, Proliferation, Migration, Vitronectin
4222 D. Wang and others p125FAK also promotes cell migration in vitro (Sieg et al., 1999). This concept was initially suggested when fibroblasts isolated from p125FAK-knockout mice were shown to exhibit reduced cell motility in culture (Ilic et al., 1995), followed by studies demonstrating that p125FAK re-expression in p125FAK null cells restored cell migration (Owen et al., 1999). Furthermore, overexpression of the non-catalytic carboxylterminal domain of p125FAK (known as FRNK) reduced cell migration in vitro (Richardson and Parsons, 1996). More recently, investigators have demonstrated that overexpression of wild-type p125FAK in CHO and COS cells resulted in increased cell migration in vitro (Cary et al., 1996; Klemke et al., 1998). The potential involvement of p125FAK in promotion of cell proliferation and migration in several cell types in vitro suggests that p125FAK could potentially play a role in neoplastic processes in which cell proliferation has escaped control mechanisms. Studies investigating the biologic role of p125FAK in neoplasia in vivo have for the most part been limited to immunohistochemical studies of tumor biopsies; for example, p125FAK expression has been shown to be elevated in colon, breast, and ovarian carcinomas (Weiner et al., 1993; Owens et al., 1995; Judson et al., 1999). This laboratory has been studying the mechanism(s) by which malignant astrocytic tumors proliferate and invade the normal brain. We have shown previously that both in patient biopsies and in an intracerebral xenograft model of human malignant astrocytoma integrins αvβ3 and αvβ5 expression are upregulated on malignant astrocytes and these tumors synthesize a ligand recognized by both integrins, vitronectin (Gladson and Cheresh, 1991; Gladson et al., 1995). Furthermore, we have shown that the attachment of malignant astrocytoma cells to vitronectin in vitro results in activation of p125FAK (Ling et al., 1999). To determine the role of p125FAK in these tumors in vivo, we created stable clones of human malignant astrocytoma cells overexpressing wild-type p125FAK using the TET-ON system. We found that doxycycline induction of p125FAK resulted in increased cell migration and increased anchorage-independent growth in vitro. In our in vivo intracerebral xenograft model, induction of p125FAK resulted in a significant increase in tumor cell proliferation that could not be attributed to decreased apoptosis. These data provide the first in vivo evidence that p125FAK promotes tumor cell proliferation, and thereby contributes to the malignant phenotype of astrocytic tumors. MATERIALS AND METHODS Cells and clones U-251MG human malignant astrocytoma cells were obtained from the ATCC and propagated in complete medium with 10% fetal bovine serum as described (Gladson et al., 1995). Cells were free of mycoplasma contamination based on routine testing. Stable cell lines expressing inducible p125FAK were established using the TET-ON Gene Expression SystemTM purchased from Clontech Laboratories, Inc. (Palo Alto, CA). Briefly, U-251MG cells were transfected consecutively with the regulator plasmid, pTET-ON, and pTRE-FAK. Mouse p125FAK cDNA with a c-Myc tag at the amino terminus (gift from Dr Steve Hanks, Vanderbilt University, Nashville, TN) (Polte and Hanks, 1995), was cloned between EcoRV and XbaI of the polylinker, and the hygromycin gene with the TK
promoter was subcloned between HindIII and PvuII downstream of the polylinker. The DMRIE-C Reagent (Gibco BRL Life Technologies, Inc., Rockville, MD) was used for liposome transfection of the U-251MG cells. Stable double transfectants were selected using 800 µg/ml of G418 (Mediatech, Inc., Herndon, VA) and 200 µg/ml of hygromycin B (Boehringer Mannheim Corp., Indianapolis, IN), and maintained in this selection medium. A clone containing the carboxy-terminal p125FAK (amino acids 752-1052) in the pGEX2TK bacterial expression vector (Schaller et al., 1995) was a kind gift from Dr Wen-Cheng Xiong (University of Alabama at Birmingham). On SDS-PAGE the p125FAK-glutathione fusion protein migrated as a single band at 70-kDa, and on western blot analysis with rabbit anti-p125FAK (Upstate Biotechnology, Lake Placid, NY, catalog #06-543) a single band of the same relative migration was detected. Analysis of intracerebral scid mouse xenografts C.B.17 scid mice, obtained from the Frederick Cancer Center at the National Cancer Institute, were injected stereotactically at 8 weeks of age with OFAK5, OFAK9, or TRE4 cell clones (2×106 cells/injection), as described previously (Gladson et al., 1995). The injected animals were divided randomly into two groups, and doxycycline (2 mg/ml) plus 4% sucrose were added to the drinking water of one group, whereas only 4% sucrose was added to the drinking water of the control group. At 11 days post-injection, mice were euthanized, and the brains were harvested immediately and processed for analysis as follows. A single brain cell suspension for FACS analysis was prepared as follows. Xenograft brains were harvested and immediately placed into DMEM with 1% BSA and the following inhibitors, phenylmethylsulfonyl fluoride (100 µM), aprotinin (10 µg/ml), leupeptin (10 µg/ml), Na vanadate (100 µM), and Nα-p-tosyl-L-lysine chloromethyl ketone (20 µg/ml), thoroughly minced (30 minutes/brain), passed through a 170 µm pore mesh, resuspended in the above buffer with inhibitors, and the single cell suspension counted. Cells (500,000) were incubated with FITCconjugated mAb anti-human specific HLA-A,B,C (Pharmingen, San Diego, CA), FITC-conjugated mouse IgG isotype control (Pharmingen), or with anti-mouse-specific Class I IgG (Pharmingen) (30 minutes, 4°C), washed, and subjected to FACS analysis, as described (Pijuan-Thompson and Gladson, 1997). The mouse IgG isotype control was used to determine the background fluorescence. For isolation of human tumor cells, a single cell suspension of xenograft brain was prepared as above and incubated with mAb antihuman specific HLA-A,B,C that had been coupled to magnetic polystyrene Dynabeads, as recommended by the manufacturer (Cellection Pan Mouse IgG kit, Dynal A.S., Oslo, Norway) (30 minutes, 4°C). Subsequently, the beads were washed, and the human tumor cells harvested by DNase treatment, which cleaves the linker molecule, as recommended by the manufacturer, lysed in RIPA buffer (Gladson and Cheresh, 1991) with the above inhibitors, centrifuged (35,000 rpm, 4°C, 1 hour), and the supernatant frozen at −70°C for immunoblot analysis. The ratio of beads to cells was 2.5 to 1. For immunoblot analysis of total brain lysate, xenograft brains were immediately snap-frozen in liquid nitrogen, and homogenized with a Polytron (Kinematica, Luzern, Switzerland) in 1 ml of RIPA lysis buffer with the above inhibitors. The lysate was centrifuged (35,000 rpm, 4°C, 1 hour), and the supernatant frozen in aliquots at −70°C. Histologic sections were prepared from xenograft brains that had been snap-frozen in liquid nitrogen and then serially sectioned on a cryostat, or fixed in buffered-formalin and serially sectioned. To assess tumor volume, 10 µm serial cryostat sections were prepared and a section at every 300 µm was reacted with mAb anti-human specific HLA-A, B, C (10 µg/ml, 2 hours, 4°C) (mAb W6/32, Serotec, Inc., Raleigh, NC), followed by HRP-conjugated goat anti-mouse IgG and 3,3′-diaminobenzidine substrate. Images were scanned into Photoshop 5.0, and the stained areas quantitated as a pixel number. Pixel numbers for all sections from each brain were summed to obtain
p125 focal adhesion kinase promotes proliferation 4223 a total pixel number. On analysis of one group of 11 animals, we found that serial sections at 150 µm distance doubled the total pixel number for each animal, but did not significantly alter the difference in total pixel number between the different animal groups. Immunoblot analysis Cultured cells were harvested and lysed in 1% NP-40 lysis buffer containing 137 mM NaCl, 2 mM EDTA, 10% glycerol, 1% NP-40, and 20 mM Tris-HCl, pH 8.0, with the above protease inhibitors (4°C, 60 minutes), centrifuged (35,000 rpm, 1 hour, 4°C), and the supernatant stored at −70°C, as described (Ling et al., 1999). Equivalent micrograms of protein from each lysate were electrophoresed on a disulfide-reduced 7.5% SDS PAGE gel, transferred to Immobilon-P membrane (Millipore Corp., Bedford, MA), blocked with 5% buffered BSA (3 hours, 22°C), and reacted with 1 µg/ml of mAb anti-p125FAK (Transduction Labs, Lexington, KY), a 1:1000 dilution of mAb anti-actin ascites (Sigma Chemical Co., St Louis, MO), 1 µg/ml of rabbit anti-p130CAS (Upstate Biotechology, Catalog #06-500), 1 µg/ml of mAb anti-c-Myc peptide (Calbiochem, La Jolla, CA), or 10 µg/ml of rabbit antiphosphospecific p125FAK (Y397) (BioSource International, Camarillo, CA) (overnight, 4°C), followed by reaction with 0.025 µg/ml of HRP-conjugated goat-anti mouse or goat anti-rabbit IgG (Bio-Rad Laboratories, Hercules, CA) (45 minutes, 22°C), and the membrane developed with the Amersham ECL chemiluminescent system (Pharmacia Biotech, Piscataway, NJ). For semi-quantitative analysis of band intensity, specific bands on the autorads were subjected to densitometric analysis, and the background densitometric reading on the autorad subtracted. Immunohistochemical analysis of human brain samples Immunohistochemical analysis was performed after antigen retrieval as described previously (Gladson et al., 1997). Formalin-fixed, paraffin-embedded tissues from biopsies of ten non-tumorous brains, three low-grade astrocytoma tumors, five anaplastic astrocytoma tumors, and six glioblastoma tumors were provided by The University of Alabama at Birmingham Hospital, in accordance with the University Human Tissue Committee policies. Tumors were diagnosed and classified according to the criteria of the revised World Health Organization (WHO) Classification of Brain Tumors (Kleihues et al., 1993). Paraffin sections (6 µ) were cut, frozen at −20°C for no more than two weeks, and immediately prior to use thawed and baked at 37°C for two hours. The primary antibodies used were rabbit anti-p125FAK (Upstate Biotechnology, Catalog #06-543) (10 µg/ml), rabbit anti-GFAP IgG (Sigma Chemical Co.) (10 µg/ml), non-immune rabbit IgG (10 µg/ml), or mAb PCNA (DAKO Corp., Carpinteria, CA) (3.0 µg/ml) (1 hour, 22°C), followed by reaction with an HRP-conjugated goat anti-rabbit IgG and 3,3′-diaminobenzidine substrate. Apoptosis assay of tissue sections (TUNEL assay) Tissue sections were reacted with biotinylated dNTP, plus TdT enzyme, followed by biotin, streptavidin-HRP and the 3,3′diaminobenzidine substrate, according to instructions provided by Oncogene Research Products (Cambridge, MA) in the TdT-Frag EL DNA Fragmentation Detection Kit. Soft agar growth assay Cells previously treated with doxycycline for 4 days, or not, were harvested with trypsin (60 seconds), followed by buffered EDTA, washed, resuspended in 0.3% agar (plus or minus doxycycline), poured onto plates containing a layer of 0.5% agar (plus or minus doxycycline), and then maintained at 37°C, 5% CO2 for 14 days, as described (Chang and Prince, 1993). Subsequently, the colonies were counted; a colony was required to contain greater than 10 cells. Samples were assayed in replicas of three, and the experiment was repeated twice.
Cell migration assay Haptotactic migration assays were performed as described previously (Gladson et al., 1995). Briefly, two-well Boyden-type chambers (Costar, Corning, NY) with 8 µm-pore non-coated polycarbonate filters were coated on the undersurface with 5 µg/ml of vitronectin or 10 µg/ml of fibronectin. Cells were harvested with buffered EDTA (1.5 mM), washed and resuspended in serum-free DMEM with 1% BSA, plated onto coated filters, and allowed to migrate (37°C, 5% CO2, 3 hours). Random migration was determined as the migration toward vitronectin when both the top and the bottom surfaces of the filter were coated with vitronectin. Immunofluorescence Immunofluorescence was performed as described previously (PijuanThompson et al., 1999). Cells were plated onto vitronectin-coated glass coverslips in serum-free medium with 1% BSA overnight. Subsequently, cells were fixed with 3% buffered paraformaldehyde, permeabilized, and reacted sequentially with mAb anti-c-Myc (Upstate Biotechnology) (10 µg/ml, 1 hour, 22°C), Alexa-488conjugated anti-mouse IgG (Molecular Probes, Eugene, OR) (1 µg/ml, 1 hour, 22°C), and then Alexa-594-conjugated mAb antivinculin (20 µg/ml, 1 hour, 22°C), followed by Hoescht nuclear stain, and coverslipped. Alternatively, cells were reacted with a 1:1,000 dilution of Oregon-Green-conjugated phalloidin (Molecular Probes) (30 minutes, 22°C), and then coverslipped. Mab anti-vinculin IgG (Sigma Chemical Co.) was labeled with the Alexa Fluor 594, per the Molecular Probes Protein Labeling Kit.
RESULTS Inducible expression of p125FAK in U-251MG human malignant astrocytoma cells To determine the biologic role of p125FAK in malignant astrocytoma cells, stable clones overexpressing p125FAK with a c-Myc tag at the amino-terminus were created from U251MG human malignant astrocytoma cells using the TETON system. A total of 30 independent clonal cell lines were screened for high level p125FAK expression after doxycycline induction. Two clones that exhibited the strongest induction (designated as OFAK5 and OFAK9) were expanded and used for the studies described here. A representative control cell line, TRE4, was stably transfected with the empty expression vector. Doxycycline treatment (2.0 µg/ml, 4 days) resulted in a dramatic increase in the level of a 125-kDa protein in OFAK5 and OFAK9 cells (an 18-fold and 20-fold increase, respectively) as estimated semiquantitatively by densitometric analysis of immunoblots using mAb anti-p125FAK (Fig. 1, lanes 6 and 4, respectively). A lower dosage of doxycycline (0.1 µg/ml) resulted in a less dramatic induction of p125FAK (a 5-fold estimated increase). The induction of exogenous p125FAK was confirmed by stripping and re-probing with mAb anti-cMyc (Fig. 1B). Expression of the endogenous c-Myc, which migrates on SDS-PAGE at 64-kDa, was not affected by doxycycline treatment (Fig. 1B). The exogenous p125FAK was activated, as stripping and reprobing of the membrane with phosphospecific anti-p125FAK (Y397) IgG (Fig. 1C) also demonstrated a band at 125-kDa. Finally, as a control for protein loading, levels of the cytoskeletal protein actin were evaluated by stripping and reprobing with mAb anti-actin (Fig. 1D), confirming the specificity of doxycycline induction of p125FAK in each clone.
4224 D. Wang and others
Fig. 1. Inducible expression of p125FAK in U-251MG malignant astrocytoma cells in vitro. Cells (clones TRE4, OFAK9 and OFAK5) were administered doxycycline (2 µg/ml, 4 days), detergent lysed, and equivalent micrograms of lysate subjected to SDS-PAGE, transferred to Immobilon, immunoblotted, stripped and re-probed, as described in Materials and Methods. (A) Blotted with mAb antip125FAK; (B) blotted with mAb anti-c-Myc; (C) blotted with rabbit anti-FAK[pY397] IgG; and (D) blotted with mAb anti-actin.
Exogenous p125FAK localizes to focal adhesions To determine whether exogenous p125FAK localized to focal adhesions, immunofluorescent co-staining with mAb antic-Myc and Alexa-594-conjugated mAb anti-vinculin was performed on OFAK5 cells adherent to vitronectin in serumfree medium after doxycycline induction. Cytoplasmic staining was found in greater than 90% of cells at 0.1 or 2.0 µg/ml doxycycline induction, indicating that >90% of cells were expressing the construct. The intensity of the cytoplasmic staining appeared greater after induction with 2.0 µg/ml doxycycline. The exogenous p125FAK co-localized with vinculin to focal adhesions after doxycycline treatment (2 µg/ml) (Fig. 2A and B, respectively). Additional perinuclear cytoplasmic staining of the c-Myc-tagged p125FAK was observed after induction with 2.0 µg/ml doxycycline, which was most likely due to the large amount of protein expressed. No cytoplasmic staining with mAb anti-c-Myc was found in the absence of doxycycline (C), although vinculin localized to focal adhesions (D). Stress fiber formation was not altered after 2.0 µg/ml doxycycline induction of p125FAK (Fig. 2E), as compared to that observed in the absence of doxycycline (Fig. 2F). Induction of p125FAK promotes malignant astrocytoma cell migration and increases p130CAS phosphorylation in vitro To determine if p125FAK promotes migration of malignant astrocytoma cells in vitro, vitronectin and fibronectin-directed haptotactic migration of OFAK5 and OFAK9 cells was
Fig. 2. Exogenous p125FAK localizes to focal adhesions. Cell clone OFAK5 was administered doxycycline (2 µg/ml, A,B,E) for 4 days, or no doxycycline (C,D,F), plated onto vitronectin-coated coverslips in serum-free medium overnight, plus or minus doxycycline, and then subjected to immunofluorescent analysis, as described in Materials and Methods. (A-B and C-D) Cells double labeled with mAb anti-c-Myc and Alexa-594-conjugated mAb anti-vinculin, respectively, followed by Hoescht nuclear strain. (E and F) Cells labeled with Oregon Green-conjugated phalloidin. Arrows denote focal adhesions in A,B,D. Bars, 50 µm. (A,B) ×40; (C-F) ×100.
evaluated (Fig. 3). A 3- and a 2.5-fold increase in migration toward vitronectin was observed for OFAK9 and OFAK5 clones, respectively, and a 2-fold increase in migration toward fibronectin was observed for the OFAK9 and OFAK5 clones, after 2.0 µg/ml doxycycline induction (Fig. 3). Doxycycline treatment had no effect on migration of the control TRE4 clone (Fig. 3), or of the parental U-251MG cells (data not shown). Doxycycline induction of p125FAK in OFAK5 and OFAK9 clones had no effect on random migration. To determine if the increased migration was a result of altered cell adhesion, cell attachment assays of OFAK5 and OFAK9 cells were performed, as described (Pijuan-Thompson and Gladson, 1997). No significant difference in cell attachment was observed with doxycycline treatment (data not shown). Thus, the increase in cell migration found in U-251MG cells overexpressing p125FAK is not due to altered cell adhesion.
p125 focal adhesion kinase promotes proliferation 4225
Fig. 3. Induction of p125FAK promotes U-251MG haptotactic cell migration. Cells (clones TRE4, OFAK9 and OFAK5) were administered doxycycline (2 µg/ml, 4 days), harvested in buffered EDTA, and haptotactic cell migration toward vitronectin and fibronectin were analyzed, as described in Materials and Methods. Random migration was subtracted from the migration toward vitronectin or fibronectin. Samples were assayed in replicas of 4 and the assays were repeated twice. The data is presented as a ratio of cell migration after doxycycline administration divided by cell migration in the absence of doxycycline administration. The mean ± s.e.m. shown is calculated from the ratios averaged over three experiments.
OFAK5
TRE 4
OFAK9
OFAK5
TRE 4
2 1
Fibronectin
Vitronection
p125FAK promotes soft agar growth of malignant astrocytoma cells To determine whether p125FAK promoted anchorageindependent growth, cells treated with doxycycline (0.1 or 1 µg/ml, 4 days) were harvested, resuspended in 0.3% agar (plus or minus doxycycline) and then poured onto a bottom layer of 0.5% agar (plus or minus doxycycline) and maintained at 37°C, 5% CO2, for 14 days after which the colonies 3500 were counted. Doxycycline treatment resulted in a dosedependent increase in the number of colonies (Fig. 4), indicating that p125FAK promotes the growth of these 3000 cells in soft agar.
Fig. 4. p125FAK promotes soft agar growth of malignant astrocytoma cells. Cells (clones TRE4, OFAK9 and OFAK5) were administered 0.1 or 1.0 µg/ml doxycycline for 4 days, harvested, resuspended in 0.3% agar (plus or minus doxycycline), poured onto a 0.5% agar plate (plus or minus doxycycline) and maintained at 37°C, 5% CO2 for 14 days. Colonies were then counted. Samples were assayed in replicas of three and the data are shown as the mean ± s.e.m.
3
0
To determine whether induction of p125FAK resulted in increased phosphorylation of p130CAS, as has been reported for COS and CHO cells in vitro (Cary et al., 1998; Klemke et al., 1998), OFAK5 and OFAK9 clones were evaluated for p130CAS phosphorylation using a semi-quantitative densitometric analysis. We found a 4- and 3-fold estimated increase in tyrosine phosphorylation of p130CAS in OFAK5 and OFAK9 clones, respectively, with 2.0 µg/ml doxycycline induction of p125FAK, and a 2-fold estimated increase in p130CAS phosphorylation after 0.1 µg/ml doxycycline treatment (data not shown). No change in p130CAS phosphorylation was observed in the parental U-251MG cells, nor in the TRE4 control clone (data not shown).
p125FAK expression was induced by the addition of doxycycline (2 mg/ml) to the drinking water. An analysis of total brain lysates from OFAK9 xenograft brains harvested at 11 days and blotted with mAb anti-c-Myc demonstrated a band at 125-kDa in the doxycycline-treated animals (Fig. 5A), indicating that the exogenous p125FAK is expressed in vivo. A band at 66-kDa, consistent with nuclear c-Myc, was detected in all of the xenograft samples. Stripping and reprobing of the membrane with mAb anti-actin demonstrated equivalent protein loading (Fig. 5B). To estimate the induction of p125FAK at 11 days, animals containing OFAK5 xenografts were euthanized, the brains harvested, minced (30 minutes/brain), and passed through a 70-µ mesh filter to obtain a single cell suspension, as described in Materials and Methods. Subsequently, human tumor cells were isolated using mAb anti-human specific HLA-A, B, C IgG coupled to magnetic beads, followed by cell lysis, and immunoblot analysis with mAb anti-p125FAK. An estimated 4.5-fold increase in p125FAK expression was detected in the human tumor cell lysates from doxycycline-treated animals, as
OFAK 5
TRE 4
OFAK 9
2500
Colony #
Doxycycline induction of p125FAK in the intracerebral xenograft model of human malignant astrocytoma To determine whether p125FAK contributes to the malignant behavior of malignant astrocytoma tumors in vivo, OFAK5 and OFAK9 clones were injected stereotactically into the brains of C.B.17 scid mice, as described previously (Gladson et al., 1995). Exogenous
4
OFAK9
Cell Migration (Ratio of + Dox./ - Dox.)
5
2000
1500
1000
500
0 0.0
0.1
1.0
0.0
0.1
1.0
(µg/ml Dox)
0.0
0.1
1.0
4226 D. Wang and others Fig. 5. Doxycycline induction of p125FAK in the intracerebral xenograft model of human malignant astrocytoma. (A and B) Animals were injected with cells (clone OFAK9) and 50% were administered 2 mg/ml doxycycline in the drinking water with 4% sucrose, or sucrose alone. At 11 days animals were euthanized, the brains harvested, snap frozen in liquid N2 and then homogenized into RIPA lysis buffer. The brain lysate from each animal (50 µg) was subjected to SPS-PAGE, transferred to Immobilon, blotted with mAb anti-c-Myc (A), stripped, and reprobed with mAb anti-actin (B). (C-E) Animals were injected with cells (clone OFAK5), and 50% were administered doxycycline in the drinking water plus sucrose, or sucrose alone, as above. At 11 days, animals were euthanized, the brains harvested, followed by mincing of each brain in sterile medium with protease inhibitors (30 minutes/brain), pushed through a 70-µm mesh, resuspended in sterile medium with inhibitors, and the cells counted. Cells (4×106) were then reacted with magnetic beads conjugated with mAb anti-human specific MHC class I IgG (30 minutes, 4°C), followed by pelleting of the cells, and detergent lysis, as described in Materials and Methods. The brain lysate from each animal (40 µg) was subjected to SDS-PAGE, transferred to Immobilon, and immunoblotted with mAb anti-p125FAK (C), stripped and re-probed with anti-phosphotyrosine IgG (D), followed by stripping and re-probing with mAb anti-actin (E).
compared to the lysates from non-treated animals (Fig. 5C), indicating that exogenous p125FAK was induced in the doxycycline-treated mice. Furthermore, the p125FAK was phosphorylated, as stripping and reprobing with antiphosphotyrosine IgG revealed a doublet band migrating at 125kDa (Fig. 5D). Equivalent micrograms of protein lysate were loaded, as stripping the membrane and reprobing with mAb anti-actin demonstrated a band of equivalent intensity migrating at 42-kDa in all animal brain lysates (Fig. 5E). p125FAK promotes malignant astrocytoma cell proliferation in vivo To determine whether p125FAK promotes proliferation of
malignant astrocytoma cells in vivo, single cell suspensions obtained from whole xenograft brains were reacted with FITCconjugated mAb anti-human specific HLA-A,B,C, IgG, followed by FACS analysis, as described in Materials and Methods. Doxycycline-treated animals demonstrated a 2.8-fold increase in tumor cell number with intracerebral propagation of the OFAK5 clone, and a 1.6-fold increase in tumor cell number with intracerebral propagation of the OFAK9 clone (Fig. 6). In contrast, no change in tumor cell number was seen with intracerebral propagation of the wild-type U-251MG cells, or of the control clone TRE4, independent of doxycycline-treatment (Fig. 6). To confirm that the above method accurately detected
30 25
% tumour cells
Fig. 6. p125FAK promotes malignant astrocytoma cell proliferation in vivo. Animals were injected with cells (U251MG and clones OFAK5, OFAK9, or TRE4), and 50% were administered doxycycline, followed by euthanasia at day 11, harvesting of the brains, and mincing and filtering of the brains to obtain a single cell suspension as described in the legend to Fig. 5. Scattergram of percentage of human tumor cells obtained by FACS analysis with FITC-conjugated mAb anti-human specific MHC class I IgG of different xenograft brains. Each point represents one animal. The mean ± s.e.m. was calculated for each group: OFAK5 +Dox, 21.7±1.2 and OFAK5 − Dox, 7.6±1.9, P value
