The phenomenon of site-specific metastatic patterns of

Transcription

1 GENERAL THORACIC Integrin Dependent Protein Tyrosine Phosphorylation is a Key Regulatory Event in Collagen IV Mediated Adhesion and Proliferation of Human Lung Tumor Cell Line, Calu-1 Nishit K. Mukhopadhyay, PhD, David Gilchrist, BS, Gavin J. Gordon, PhD, Chang-Jie Chen, PhD, Raphael Bueno, MD, Michael L. Lu, PhD, Ravi Salgia, MD, PhD, David J. Sugarbaker, MD, and Michael T. Jaklitsch, MD Division of Thoracic Surgery, Department of Pathology, Division of Urology, Brigham and Women s Hospital, Harvard Medical School, Department of Adult Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts Background. The clinical phenomenon of lung cancer metastasis to specific target organs is believed to be a direct interaction between tumor cells and extracellular matrix components. During invasion, tumor cells attach to the basement membrane protein, collagen type IV, degrade it, migrate through blood vessels, and adhere to extracellular matrix proteins. Methods. Four nonsmall-cell lung cancer cells were tested for adhesion, proliferation, migration and morphologic alterations on collagen type IV matrix by immunoprecipitation, Western blotting, phase contrast and time lapse video microscopy. Results. Collagen type IV promoted Calu-1 cell adhesion (< 15 minutes) and motility (< 6 hours) without any significant effect on proliferation. 1 -integrin is essential for collagen type IV adhesion and 8 to 10 fold higher expression of 1 -integrin on the surface of Calu-1 cells was identified. A protein tyrosine phosphatase inhibitor, peroxyvanadate, caused 50% inhibition in the adhesion process within 5 minutes but exposure to 60 mol/l genistein for 72 hours, a protein tyrosine kinase inhibitor, drastically inhibits Calu-1 cell proliferation (> 70%). We examined the influence of 1 -integrin, peroxyvanadate and genistein on the spreading morphogenesis of Calu-1 cells on Collagen type IV. Use of either 1 g of anti 1 -integrin inhibitory antibody (P5D2), 100 mol/l peroxyvanadate or 60 mol/l genistein had dramatic influence on the spreading of Calu-1 cells. Finally, Focal adhesion kinase was identified as a phosphoprotein target. Conclusions. We have characterized an in vitro model of matrix-specific binding of a lung cancer cell line, Calu-1 to Coll IV. Calu-1 cells use primarily a 1 -integrin mediated intracellular tyrosine phosphorylation phenomenon as the key regulatory mechanism for its binding advantage to Coll IV matrix. (Ann Thorac Surg 2004;78:450 7) 2004 by The Society of Thoracic Surgeons The phenomenon of site-specific metastatic patterns of certain cancers (seed/soil theory) has been recognized for decades [1]. Nonsmall-cell lung cancer (NSCLC) has a tendency to metastasize to bone, liver, adrenal glands, lymph nodes, brain, and the opposite lung [2]. We believe this phenomenon is a manifestation of the interaction between the cancer cells and the extracellular matrix (ECM) components. The process of metastasis involves a coordinated series of interactions between the tumor cells and the surrounding ECM. Proteins embedded in both the matrix and the plasma membranes create the phenotype of the normal cell [3]. Alterations in the binding of the cell to the ECM enables tumor cells to proliferate, migrate, invade the matrix, and colonize a distant site. Furthermore, Accepted for publication Jan 28, Address reprint requests to Dr Jaklitsch, Division of Thoracic Surgery, Brigham and Women s Hospital, 75 Francis St, Boston, MA 02115; mjaklitsch@partners.org. organ-specific ECM separates different tissue compartments and may relate to clinical metastatic patterns observed in lung cancer [4]. The ECM and the surrounding microenvironment are capable of influencing the cells by protein receptors, integrins, and adhesion molecules. Activation of these proteins lead to one or more possible intracellular signals resulting in dramatic changes in the cytoskeleton, gene expression, and association of cell membrane-bound proteins. Type IV collagen (Coll IV) is the structural backbone of the basement membrane of several solid organs including lung [5, 6]. This protein has the ability to interact with the cell surface adhesion molecules such as integrin and proteoglycans, and serves as scaffolding for the binding of other basement membrane components [7]. Many complex intracellular signaling pathways activated by ECM components have been studied. However, the intracellular signaling pathway following lung cancer cell binding to Coll IV has not been elucidated. Integrin by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg MUKHOPADHYAY ET AL 2004;78:450 7 INTEGRIN MEDIATED SIGNALING IN LUNG CANCER mediated binding to Coll IV in melanoma, ovarian, and breast carcinoma cells results in the tyrosine phosphorylation of numerous intracellular proteins including both cytoskeleton [8, 9] and signaling molecules [10, 11]. In this study we describe that a lung squamous carcinoma cell line (Calu-1) exhibits an inherent adhesion affinity for Coll IV. In addition, we report the role of tyrosine phosphorylation-dephosphorylation mechanisms in Calu-1 cell adhesion, proliferation, motility, and morphology. Our results indicate that 1 -integrin plays an essential role in this process. Understanding the molecular basis of the lung cancer cell-extracellular matrix interactions could lead to the identification of novel therapeutic strategies. Material and Methods Materials Dulbecco s modified Eagle s medium (DMEM), RPMI 1640 medium, fetal bovine serum (FBS), trypsin, phosphate buffered saline (PBS), penicillin-streptomycin, and glutamate were purchased from Invitrogen Life Technologies (Carlsbad, CA). Collagen type IV (human) was from Research Diagnostics Inc. (Flanders, NJ). Hydrogen peroxide (H 2 O 2 ) and sodium vanadate (orthovanadate) were obtained from Fisher Scientific Co. (Atlanta, GA; www-.fishersci.ca). Proto-gel was from National Diagnostic (Atlanta, GA). Antiphosphotyrosine monoclonal antibody (4G10) was a kind gift from Dr. Thomas M. Roberts of the Dana-Farber Cancer Institute. Polyclonal focal adhesion kinase (FAK), -actin (control antibody), and 1 -integrin Ab (I41720) were from Transduction Laboratory (San Diego, CA). 1 -integrin blocking Ab (P5D2) [12] was from our collaborators; 2 and V blocking antibodies (Ab) are from Cal Biochem (San Diego, CA). Bovine serum albumin (BSA), genistein, Soyabin trypsin inhibitor, mouse IgG, glycerol, Nonidate P-40 (NP40), dithiothretol, phenylmethylsulphonyl fluoride (PMSF), leupeptin, aprotinin, and -mercaptoethanol were purchased from Sigma Chemical Co (St. Louis, MO). Cell Culture The human nonsmall-cell lung cancer lines used in these studies (Calu-1, NCI-520, NCI-H441, and NCI-H23) were kindly provided by Dr. Ian Anderson (Dana-Farber Cancer Institute) but are also available through American Type Culture collection ( All the above cell lines were grown in either DMEM or RPMI supplemented with 10% heat inactivated fetal bovine serum, 1% penicillin-streptomycin, and 2% glutamine at 37 C and 5% CO 2. Cell Adhesion and Proliferation Assay Six-well polystyrene plates (9.6 cm 2 /well) were coated either with Coll IV (5 g per well) or fibronectin (5 g per well) suspended in 1-mL coating buffer containing 100 mmol/l tris (ph 8.0) and incubated at 4 C for 16 hours. The plates were washed with PBS and incubated with 3% BSA in PBS for 2 hours at 37 C in order to block 451 nonspecific binding sites. The plates were then washed twice more with PBS and kept at 4 C until needed. Adhesion assays were performed with serum-starved cells (14 to 16 hours), detached with 0.25% trypsin/edta (ethylene diamine tetra acetic acid) followed by suspension in medium containing DMEM, 20 mmol/l hepes, and 2% BSA. A 1-mL suspension containing 1 to cells was added to each well and incubated at 37 C and 5% CO 2. At the end of each incubation time point, the plates were washed twice with PBS to remove the nonadherent cells, trypsinized, and counted. The data were means of triplicate wells and were expressed as a percentage of deposited cells. Cell proliferation was determined by adding serum-starved cells on either BSA, fibronectin or on Coll IV-coated plates and incubated for several days. The cells were counted every 24 hours as discussed earlier. The data were means of triplicate wells. Immunoprecipitation and Western Blotting Cells from BSA-coated or Coll IV-coated plates were lysed at 4 C in the lysis buffer containing 20 mmol/l Tris-HCl (ph 8.0), 137 mmol/l NaCl, 2 mmol/l EDTA, 10% glycerol, 1% NP-40, 200 mol/l sodium orthovanadate, 1 mmol/l dithiothretol (DTT), 1 mmol/l phenylmethylsulfonyl fluoride (PMSF), 2 mg/ml leupeptin, and 5 mg/ml aprotinin. The lysates were centrifuged at 14,000 rpm for 15 minutes at 4 C, and the supernatants (500 g 1 mg of protein) were used for immunoprecipitations with either antiphosphotyrosine (4G10) or polyclonal anti-fak antibody at 4 C either overnight or for 3 to 5 hours. After incubation for 1 hour with antibody, either protein A-sepharose or protein G-sepharose was added to the lysates and was incubated for required time. The bead bound complexes were pelleted, washed several times with lysis buffer and PBS, and boiled with SDS (sodium dodecyl sulfate) sample buffer for 3 to 5 minutes before loading on SDS-PAGE (polyacrylamide gel electrophoresis, 7.5% to 10%). For Western blot analysis, the proteins were transferred to nitrocellulose membranes after SDS-PAGE and blocked with 5% dry milk in TBST (10 mmol/l Tris-HCl, ph8.0, 150 mmol/l NaCl, 0.05% Tween 20). The blots were incubated with specific primary antibody from 1 hour at room temperature to overnight at 4 C. Membranes were washed briefly with TBST and incubated with the horseradish peroxidase conjugated secondary antibody for 1 hour at room temperature (1:5000 dilution). Following extensive washing, immunoreactive bands were visualized by chemiluminescence (ECL reagent; New England Nuclear, Wellesley, MA). When required, the membranes were stripped in 62.5 mmol/l Tris-HCL (ph 6.8), 2% SDS and 1.0 mmol/l -mercaptoethanol for 30 minutes at 50 C and reblotted again. Time Lapse Video Microscopy The time lapse video microscopy (TLVM) procedure was performed as previously described [10]. Cells were cultured on uncoated, fibronectin-coated or Coll IV-coated tissue culture plates in a temperature and CO 2 controlled GENERAL THORACIC

3 GENERAL THORACIC 452 MUKHOPADHYAY ET AL Ann Thorac Surg INTEGRIN MEDIATED SIGNALING IN LUNG CANCER 2004;78:450 7 chamber in their standard growth media. The cells were observed in a time-lapse fashion using an Olympus IX70 inverted device (Therm-Omega-Tech, Warmington, PA), Optronic Engineering DEI-750 3CCD digital video camera (Optronicsw, Galeta, CA), and Sony SVT-S3100 time lapse S-VHS video recorder (Sony, Tokyo, Japan). For image presentation, video images were captured and printed in a Sony Color Video Printer P-56,000 MD. Densitometry and Statistical Analysis Western blots were quantified by using a densitometricbased analysis performed on scanned fluorograms using Scion Image Beta 4.02 software from Scion Corporation (Frederick, MD; info@scioncorp.com). GraphPad Prism v.3.02 (GraphPad Software, San Diego, CA) was used to assess quantified differences in the number of adherent cells to Coll IV using a two-tailed Students (parametric) t test and the differences were determined to be statistically significant if p less than Adhesion of Calu-1 Cells to Collagen Type IV is Integrin Dependent Cell lysates from all four cell lines were analyzed for the presence of 1 -integrin. Two proteins of apparent molecular mass 120 and 135 kda were immunoreactive in the Western blot by anti 1 -integrin subunit monoclonal antibodies (mab; Fig 2A). Although the 135-kDa 1 - integrin subunit was present in H441, H23 and H520, it was expressed approximately tenfold higher in Calu-1 cells. The 120-kDa band was present in equivalent amounts in H520, H23 and H441 cells, but is less concentrated in Calu-1 cells and the control cell extract. To determine whether 1 -integrin expression on the surface of the Calu-1 cells correlated with its ability to adhere on collagen matrix, the cells in suspension were preincubated with various concentrations of anti 1 - integrin Ab (P5D2) for 60 minutes at 4 C before the adhesion assay. As illustrated in Figure 2B, a complete inhibition of adhesion of Calu-1 cells to Coll IV indicates the predominant role of 1 -integrin in this interaction. As indicated in Figure 2C, no significant adhesion inhibition of Calu-1 cells was observed when either anti 2 or V integrin blocking antibodies were used. Results Adhesion, Proliferation and Motility of NSCLC Cells to Collagen Type IV and Fibronectin Adhesion of four different NSCLC lines were tested, including two adenocarcinomas (NCI-H23, NCI-H441) and two squamous cell carcinomas (NCI-H520, Calu-1) to Coll IV-, fibronectin-, and BSA-coated dishes. Only Calu-1 cells revealed significant adhesion to both extracellular matrix proteins but not to BSA substratum. As illustrated in Figure 1A, Calu-1 cells rapidly attached to human Coll IV-coated dishes, with a plateau in the kinetic curve at 45 minutes. Fibronectin-coated dishes also provided an attachment advantage to Calu-1 cells compared with BSA-coated wells, but did not reach a plateau in the kinetic curve until 90 minutes. There was no substantial difference in the final amount of Calu-1 attachment after 2 hours either on fibronectin or Coll IV, although the early attachment rate was more rapid on Coll IV. Compared with Calu-1 cells, the attachment of H-520, H-23, and H441 cells either on Coll IV or on fibronectin, were 5% to 10%, 6% to 12%, and 10% to 15%, respectively, in 1-hour adhesion assay. There was no significant attachment of Calu-1 cells in the BSA-coated wells even after 4 hours of incubation. The Calu-1 cell attachment advantage to Coll IV did not produce a change in the proliferative rate in our assays (Fig 1B) nor in any other NSCLC cell line (data not shown). Cells were initially resuspended and placed immediately onto the plates coated with the various surfaces and TLVM was performed. As appreciated from Figure 1C, compared with the fibronectin-coated surface, Calu-1 cells on the Coll IV-coated surface had activated morphology, increased membrane ruffling, increased formation, and retraction of lamellipodia and filopodia. Peroxyvanadate-Induced Phosphorylation Negatively Regulates the Adhesion of Calu-1 Cells to Human Collagen Type IV Serum-starved Calu-1 cells were treated with 100 mol/l peroxyvanadate for various time intervals, and an adhesion assay was performed for 1 hour at 37 C with Coll IV-coated plates. Gradual induction of tyrosine phosphorylation induced by peroxyvanadate negatively regulated Calu-1 cell attachment to Coll IV (Fig 3A). More than 50% reduction in adhesion occurred after 3 minutes of treatment and less than 25% of the cells were attached to the Coll IV after 5 minutes of induction. As indicated in Figure 3B, the peroxyvanadate treatment stimulated the tyrosine phosphorylation of several proteins in a time dependent manner ranging from 40 to 200 kda. The level of phosphorylation plateaued within 10 minutes of treatment, although some stimulation was detected as early as 1 minute. Protein-Tyrosine Kinase Inhibitior Prevents Proliferation But Not Attachment of Calu-1 Cells Treatment of Calu-1 cells with 60 mol/l of genistein for 3 days did not alter the attachment rate to Coll IV-coated wells as compared to the cells exposed to DMSO for 3 days (Fig 4A). However, genistein retarded the proliferation of Calu-1 cells severely with 70% to 80% reduction in 48 hours (Fig 4B). Western blot analysis of the untreated and genistein-treated cells revealed that protein tyrosine phosphorylation was globally suppressed by genistein treatment (Fig 4C) Integrin Engagement and Protein Tyrosine Phosphorylation Control Cell Spreading Calu-1 cells developed distinct pseudopodia after adhesion to Coll IV (Fig 5A). Pretreatment of the cells with nonblocking 1 -integrin antibody (I417270) or control IgG antibody maintained this structure (Figs 5B and 5C, respectively). However, pretreatment of the cells with blocking 1 -integrin antibody (P5D2) completely abolished the spreading morphology (Fig 5D). A similar

4 Ann Thorac Surg MUKHOPADHYAY ET AL 2004;78:450 7 INTEGRIN MEDIATED SIGNALING IN LUNG CANCER 453 GENERAL THORACIC Fig 1. (A) Adhesion, (B) proliferation, and (C) morphology of Calu-1 cells when deposited on collagen type IV ( ), fibronectin (E), and bovine serum albumin ( ) coated matrix. The data were expressed as either percentage of control (A, % of total number of attached cells compared to total) or the actual cell number (B). (C) Represents the time-lapsed video microscopy of Calu-1 cells on fibronectin (left) or on Collagen IV (right). Pictures were taken every 3 hours as described in Materials and Methods. The arrow indicates a long actin extension. nonspread, round morphology of cells was seen with either genistein (Fig 5E) or peroxyvanadate treatment (Fig 5F). Adhesion to Collagen Type IV Matrix Leads to Increased Tyrosine Phosphorylation of FAK The lysates of adherent cells were further analyzed by immunoprecipitation with antibodies to FAK followed by Western blotting with 4G10. As illustrated in Figure 6, a significant amount of phosphorylation of FAK is observed within 30 minutes of adhesion of Calu-1 cells to Coll IV. Comment Cell-matrix interactions are a critical factor in tumor invasion and metastasis. Specifically, cell-matrix interactions are involved in the detachment of tumor cells from the primary site, the homotypic interaction between tumor cells and distant matrix after transport in the circulation, and the stimulation of angiogenic growth by regional endothelium at the distant metastatic site. An in vitro model of the interaction between lung cancer cells and the ECM provides an opportunity to dissect the process of metastasis and may identify potential targets for therapeutic intervention. We have identified Coll IV as an ECM ligand that acts as a promoter in integrindependent biological events including adhesion, migration and spreading of NSCLC cells. Coll IV improved adhesion of Calu-1 cells 50% to 70% over control, but did not provide a proliferation advantage nor did it cause inhibition of proliferation. Furthermore, Coll IV has been noted to have disparate effects on adhesion and proliferation in other cell lines. For in-

5 GENERAL THORACIC 454 MUKHOPADHYAY ET AL Ann Thorac Surg INTEGRIN MEDIATED SIGNALING IN LUNG CANCER 2004;78:450 7 Fig 2. (A) Western blot analysis of 1 -integrin subtypes in nonsmall-cell lung cancer cell lines. Calu-1 and H520 are squamous cell lines; H23 and H441 are adenocarcinoma lines. Antibody specific to 1 -integrin (141720) revealed a 135-kDa band expressed approximately eight- to tenfold higher in Calu-1 cells compared with the other three lines, and a 120-kDa band less concentrated in Calu-1 but equivalently expressed in the other three lines. The control lane was loaded with 15 L (equivalent to 15 g) of whole cell extract (from human epidermoid carcinoma cell line supplied by Transduction Laboratories). The blot was stripped and reprobed for p85 to confirm equal loading. (B) Competition assay with increasing concentrations (1 to 10 g) of blocking 1 -integrin antibody (P5D2), for 60 minutes incubation at 4 C. Normal mouse IgG was used as a negative control. (C) Competition assay with 1 to 10 g of blocking 2 (solid bar) and V (open bar) antibodies. The data represent the means of triplicate experiments. (Coll. IV collagen IV; Cont. Ab. control antibody.) stance, adhesion of melanoma cells to Coll IV was enhanced by 50%, but proliferation was inhibited by 40% when compared to control [13]. Proliferation of rat cortical progenitor cells was also inhibited on Coll IV matrix [14]. In the case of hepatoma cells, both adhesion and proliferation increased on Coll IV matrix [15], and Coll IV played a critical role in conditioning glomerular endothelial cells to respond to proliferative stimuli [16]. Fig 3. Effect of peroxyvanadate treatment on adhesion of Calu-1 cells to collagen type IV. Serum-starved Calu-1 cells ( ) were treated with peroxyvanadate (100 mol/l) for time intervals (2.5 to 10 minutes) as described in the Materials and Methods section. (A) The cells were washed to remove the vanadate and an adhesion assay was performed for 60 minutes. (B) Whole cell extract proteins (50 g) from each time point of peroxyvanadate treated cells were loaded on a 10% SDS-PAGE, transferred to nitrocellulose membranes and blotted with 4G10 antibodies to compare the tyrosine phosphorylation pattern. With increasing time of exposure to peroxyvanadate, there is increased tyrosine phosphorylation of multiple intracellular proteins. The phosphotyrosine blot was reprobed for -actin to verify equal loading. Bredin and colleagues [17], and other investigators [18 20], demonstrated that most lung cancer cell lines express a variety of 1 -integrins. In our present study we not only reported that all four NSCLC cell lines examined expressed 1 -integrin, but also demonstrated that high levels of expression in Calu-1 cells differed clearly with respect to the other lung cancer cell lines tested and contributed to its adhesion to Coll IV. The role of 1 -integrin in lung cancer cell adhesion is not clear. We have established that it was the 1 -integrin expression on the surface of the Calu-1 cells that was responsible for interaction with Coll IV. However, other 1 -integrin positive NSCLC lines (including H441, H23,

6 Ann Thorac Surg MUKHOPADHYAY ET AL 2004;78:450 7 INTEGRIN MEDIATED SIGNALING IN LUNG CANCER 455 Fig 4. Effect of genistein on type IV collagen-mediated adhesion and proliferation of Calu-1 cells. (A) There is no difference in the attachment rate of Calu-1 cells after 60-minute exposure to collagen IV matrix with either pretreatment by 60 mol/l genistein ( ) for 3 days or dimethyl sulphoxide exposure (E). Each point in the curve indicates the mean of triplicate counts. (B) Inhibition of Calu-1 cell proliferation by 60 mol/l genistein for 3 days when added to bovine serum albumin (BSA)-coated plates ( ) or collagen IV-coated plates ( ) compared with BSA-coated (E) or collagen IV-coated plates without genistein ( ). Triplicate wells of each time point were counted for cell growth by hemocytometer. (C) Phosphotyrosine blot of whole cell extracts of control and genistein treated cells shows suppressed phosphorylation of several proteins ranging from 30 to 200 kda (as indicated by arrows). -actin was used as a loading control. 4 GENERAL THORACIC and H520) marginally adhere to Coll IV matrix, indicating that either the amount of 1 -integrin on the surface of these cells was not sufficient or other molecules on the surface of these cells besides 1 -integrin also play a role in adhesion. By altering the level of intracellular phosphorylation, we further elucidate the role of tyrosine phosphorylation in the adhesion process. Several reports have indicated that tyrosine phosphatase inhibitors, such as vanadate and peroxyvanadate, activate tyrosine phosphorylation of FAK and paxillin [21 23]. Our data also confirms that the inhibition of tyrosine phosphatase influenced the adhesion process with a delicate balance. Surprisingly, genistein had no effect on Calu-1 cell adhesion at the same concentration that dramatically attenuated its proliferation. It has been reported that the effective inhibitory concentration of genistein in vivo is several fold higher than in vitro, due to differences in intracellular ATP concentration and drug accumulation within cells [24]. The concentration of genistein used in these studies did not produce a large change in protein phosphorylation intensity even after 72 hours, but did affect the proliferative rate of Calu-1 cells, indicating that a fine balance of tyrosine phosphorylation events were important for proliferation in this cell line. Change in morphology is an indication of alteration of intracellular events. Spreading of Calu-1 cells on Coll IV, dramatically altered by treatment with either 1 -integrin blocking antibody, or peroxyvanadate and genistein, indicates that a 1 -integrin dependent phosphorylation event is the control mechanism for the morphologic alteration of Calu-1 cells on Coll IV matrix. In addition to providing a molecular glue essential for tissue organization, integrins are dynamic signaling molecules. Binding and clustering of integrin at the cell surface induces protein tyrosine phosphorylation of intracellular proteins. One of the central phosphorylated molecules in these events is FAK. Accumulating evidence supports a critical role for FAK in promoting cell migration stimulated by different types of cell surface ligand receptors. Activation of FAK following adhesion to Coll IV and integrin engagement has been reported for many different cell types [8, 9]. Our observation of phosphorylation of FAK after Calu-1 cell adhesion to Coll IV might be important for integrin signaling in this lung cancer cell line. It will be important to identify the associated events following phosphorylation and activation of FAK in lung cancer cells. For example, FAK has been reported to phosphorylate and associate with many cytoskeleton and cytoplasmic signaling molecules including paxillin, p130 Cas, C-Src [25], and ERKs [11]. It remains to be established whether signal transduc-

7 GENERAL THORACIC 456 MUKHOPADHYAY ET AL Ann Thorac Surg INTEGRIN MEDIATED SIGNALING IN LUNG CANCER 2004;78:450 7 Fig 5. Effect of peroxyvanadate, genistein, and 1 -integrin antibodies on collagen IV induced morphology of Calu-1 cells. (A) Cells attached to collagen IV for 60 minutes. (B) Cells attached to collagen IV after control antibody (Ab; IgG) treatment (1 g per 10 6 cells). (C) Cells adhered to collagen IV after nonblocking 1 -integrin antibody (I41720) treatment (1 g per 10 6 cells). (D) Cells on collagen type IV matrix following the blocking 1 -integrin Ab (P5D2) treatment (1 g per 10 6 cells). (E) Effect of 60 mol/l genistein treatment for 3 days on Calu-1 cells attached to collagen IV matrix. (F) Effect of 100 mol/l peroxyvanadate exposure for 2.5 minutes on Calu-1 cells attached to collagen IV matrix. tion pathways downstream of the Coll IV receptors participate directly in the tyrosine phosphorylation of FAK in this system. However, we have established that Fig 6. Immunoprecipitation and immunoblotting analysis of Calu-1 whole cell lysates following adhesion to collagen IV matrix. FAK tyrosine phosphorylation increased within 30 minutes of exposure to collagen IV matrix (top). The blot was striped and reblotted for FAK to verify equal loading (bottom). (IP immunoprecipitation; FAK polyclonal focal adhesion kinase; Wb Western blotting.) the phosphorylation of FAK in Calu-1 cells is dependent on 1 -integrin function although the involvement of other integrin subtypes can not be ignored. It is highly likely that additional signaling molecules downstream of FAK are involved in the regulation of proliferation, migration and morphology of Calu-1 cells. In conclusion, we have characterized an in vitro model of matrix-specific binding of a lung cancer cell line to Coll IV. We have demonstrated that Calu-1 cells use primarily a 1 -integrin mediated intracellular tyrosine phosphorylation phenomenon as the key regulatory mechanism for its binding advantage to Coll IV matrix. FAK may be a promising target for future therapeutic intervention strategies in lung cancer. Further understanding of the molecular basis of lung cancer cell adhesion, proliferation and motility with respect to the extracellular matrix could lead to new insights into the molecular mechanism of lung cancer metastsais leading to better therapy. The authors would like to thank Deborah Cole for secretarial support and Anastasia Pappas-Estocin, MSW, MPH, for statistical support. This work was partially supported by grants from the Lowe Center for Thoracic Oncology of the Dana-Farber Cancer Institute and the Brigham and Women s Physician Organization.

8 Ann Thorac Surg MUKHOPADHYAY ET AL 2004;78:450 7 INTEGRIN MEDIATED SIGNALING IN LUNG CANCER References 1. Sugarbaker ED. The organ selectivity of experimentally induced metastases in rats. Cancer 1952;52: Maddaus M, Ginsberg RJ. Lung cancer diagnosis and staging. In Pearson FG, ed. Thoracic surgery. New York: Churchill Livingstone; 1995: Jaklitsch MT, Biro S, Casscells W, Dichek DA. Transduced endothelial cells expressing high levels of tissue plasminogen activator have an unaltered phenotype in vitro. J Cell Physiol 1993;154: Pauli BU, Lee CL. Organ preference of metastasis. The role of organ-specifically modulated endothelial cells. Lab Invest 1988;58: Konomi H, Hayashi T, Nakayasu K, Arima M. Localization of type V collagen and type IV collagen in human cornea, lung, and skin. Immunohistochemical evidence by anti-collagen antibodies characterized by immunoelectroblotting. Am J Pathol 1984;116: Polette M, Thiblet J, Ploton D, et al. Distribution of a1(iv) and a3(iv) chains of type IV collagen in lung tumors. J Pathol 1997;182: Laurie GW, Bing JT, Kleinman HK, et al. Localization of binding sites for laminin, heparan sulfate proteoglycan and fibronectin on basement membrane (type IV) collagen. J Mol Biol 1986;189: Vuori K, Ruoslahti E. Tyrosine phosphorylation of p130cas and cortactin accompanies integrin-mediated cell adhesion to extracellular matrix. J Biol Chem 1995;270: Sakamoto M., Ino Y, Ochiai A, et al. Formation of focal adhesion and spreading of polarized human colon cancer cells in association with tyrosine phosphorylation of paxillin in response to phorbol ester. Lab Invest 1996;74: Salgia R, Li JL, Ewaniuk DS, Wang YB, et al. Expression of the focal adhesion protein paxillin in lung cancer and its relation to cell motility. Oncogene 1999;18: Sanders MA, Basson MD. Collagen IV dependent ERK activation in human Caco-2 intestinal epithelial cells requires focal adhesion kinase. J Biol Chem 2000;275: Wayner EA, Gil SG, Murphy GF. Epiligrin, a component of epithelial basement membranes, is an adhesive ligand for bpositive T lymphocytes. J Cell Biol 1993;121: Han J, Ohno N, Pasco S, et al. A cell binding domain from 457 the alpha3 chain of type IV collagen inhibits proliferation of melanoma cells. J Biol Chem 1997;272: Ali SA, Pappas IS, Parnavelas JG. Collagen type IV promotes the differentiation of neuronal progenitors and inhibits astroglial differentiation in cortical cell cultures. Brain Res Dev Brain Res 1998;110: Ogata R. Type IV collagen and laminin enhance the motility, adhesion, and proliferation of hepatoma cells. Kurume Med J 1998;45: Artishevsky A, Cha DR, Adler S, et al. The effect of transfection of antisense cdna for procollagen alpha 1 (IV) on stimulated proliferation in rat glomerular endothelial cells. J Am Soc Nephrol 1997;8: Bredin CG, Sundqvist KG, Hauzenberger D, Klominek J. Integrin dependent migration of lung cancer cells to extracellular matrix components. Eur Respir J 1998;11: Hirasawa M, Shijubo N, Uede T, Abe S. Integrin expression and ability to adhere to extracellular matrix proteins and endothelial cells in human lung cancer lines. Br J Cancer 1994;70: Suzuki S, Takahashi T, Nakamura S, et al. Alterations of integrin expression in human lung cancer. Jpn J Cancer Res 1993;84: Smythe WR, LeBel E, Bavaria JE, et al. Integrin expression in non-small cell carcinoma of the lung. Cancer Metastasis Rev 1995;14: Ruff SJ, Chen K, Cohen S. Peroxovanadate induces tyrosine phosphorylation of multiple signaling proteins in mouse liver and kidney. J Biol Chem 1997;272: Maa MC, Leu TH. Vanadate-dependent FAK activation is accomplished by the sustained FAK Tyr-576/577 phosphorylation. Biochem Biophys Res Commun ;251: Defilippi P, Retta SF, Olivo C, et al. P125fak tyrosine phosphorylation, and focal adhesion assembly. studies with phosphotyrosine phosphatase inhibitors. Exp Cell Res 1995; 221: Akiyama T, Ishida J, Nakagawa S, et al. Genistein, a specific inhibitor of tyrosine-specific protein kinases. J Biol Chem 1987;262: Richardson A, Malik RK, Hildebrand JD, Parsons JT. Inhibition of cell spreading by expression of the C-terminal domain of focal adhesion kinase (FAK) is rescued by coexpression of Src or catalytically inactive FAK: a role for paxillin tyrosine phosphorylation. Mol Cell Biol 1997;17: GENERAL THORACIC