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

J. Biol. Chem., Vol. 277, Issue 1, 75-86, January 4, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/1/75    most recent M103268200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sawhney, R. S. Articles by Brattain, M. G. Search for Related Content PubMed PubMed Citation Articles by Sawhney, R. S. Articles by Brattain, M. G. Social Bookmarking                      What's this?

Differences in Sensitivity of Biological Functions Mediated by Epidermal Growth Factor Receptor Activation with Respect to Endogenous and Exogenous Ligands^*

Rajinder S. Sawhney, Guo-Hao K. Zhou^§, Lisa E. Humphrey, Paramita Ghosh^¶, Jeffrey I. Kreisberg^¶, and Michael G. Brattain^**

From the  Department of Pharmacology & Therapeutics, Roswell Park Cancer Institute, Buffalo, New York 14263, the ^¶ Department of Surgery, University of Texas Health Science Center, San Antonio, Texas 78229, and the ^§ Department of Pharmacology, Baylor College of Medicine, Houston, Texas 77030

Received for publication, April 12, 2001, and in revised form, October 4, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Despite constitutive expression of autocrine transforming growth factor- (TGF-) in growth factor-independent colon carcinoma cells, the epidermal growth factor receptor (EGFr) is not saturated and can be further activated by exogenous EGFr ligand. Given that the activation of EGFr by exogenous growth factor has no further effect on DNA synthesis, the question arises as to what function this additional EGFr activation might have. We report that EGF induces integrin ₂ expression, integrin-mediated adhesion, and micromotility of HCT116 cells. The stimulatory effect of ligand on these biological functions is abrogated by treatment with AG1478- and EGFr-blocking monoclonal antibody. This provides evidence that the biological responses are EGFr-mediated and EGFr is located upstream of integrin ₂ expression. Therefore, although exogenous EGF has no effect on DNA synthesis beyond that induced by autocrine TGF- (at subsaturating levels of EGFr occupation) exogenous growth factor does induce integrin ₂ expression, cell adhesion, and micromotion. An important finding revealed by this study is the documentation of biological responses of EGFr-mediated functions, including DNA synthesis, cell adhesion, and micromotion, which differ in sensitivity with respect to different degrees of EGFr activation at the basal state and in response to exogenous ligand.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The human colon carcinoma cell line HCT116 is aggressively tumorigenic, invasive, undifferentiated and growth factor-independent (1-5). The HCT116 cell line is representative of growth factor-independent carcinoma cell lines (2, 6). Constitutive expression of a full-length TGF-¹ antisense cDNA has shown that the basis for the growth factor independence of these cells is the constitutive expression of TGF- and, consequently, a low level constitutive activation of EGFr even when the cells are growth-arrested in G₀ (6, 7). We hypothesize that the relatively low level of EGFr activation resulting from autocrine TGF- may be sufficient to fulfill one or more highly sensitive responses to EGFr signal transduction but would require augmentation by exogenously activated EGFr to optimally enable less sensitive functions. In HCT116 cells DNA synthesis from the G₀ state is fully activated by autocrine TGF- such that exogenous EGF (or other growth factors) has no effect on this EGFr function despite the availability of unoccupied EGFr and signal transduction intermediates such as ERK (2, 7). Thus, the objective of this work was to identify whether there are any important functions mediated by the unoccupied EGFr in response to exogenous EGF and to determine whether this function is shared in part by autocrine TGF- as well. We report that EGFr activation by exogenous ligand results in increased integrin expression, cell adhesion, and cell micromotion with less sensitivity than that exhibited by the optimal mitogenic function stimulated through relatively low level receptor occupation generated by autocrine TGF- activity. In contrast, low level receptor occupation of EGFr by autocrine TGF- resulted in a relatively low level of basal integrin expression and biological function relative to that obtained with high levels of receptor occupation generated by exogenous EGF or TGF-.

Growth factors are important effectors of cell adhesion, cell motility, and integrin expression, although the underlying mechanisms are still unclear (8-10). It has been demonstrated that EGF can induce ₁ integrin mRNA expression in quiescent mouse 3T3 cells (11), but it was not determined whether EGF-induced changes in ₁ subunit mRNA expression led to changes in cell surface protein levels as well as functional alterations in cell adhesion. Fujii et al. (12) showed that EGF induced HSC-1 human cutaneous squamous carcinoma cell interaction with type I collagen by up-regulation of integrin ₂₁ but not by ₃₁, ₅₁, or _v3 expression. Recently, Moro et al. (13) showed that, in normal human skin fibroblasts and FCV 304 endothelial cells, integrin-dependent EGFr activation was associated with cell survival and proliferation in response to ECM. These reports suggested that EGFr activation of integrin expression might represent an additional function to that of mitogenesis by EGFr signaling in HCT116 cells.

EGFr-mediated control of integrin expression is important, because integrins and their ligands have significant roles in tumor cell biology (14, 15). For example, transformation of epithelial cells to the malignant state is often accompanied by quantitative changes in integrin expression, which in turn may control cell proliferation and cell metastasis (16, 17). Moreover, alteration of integrin expression can in turn lead to cross-talk with growth factor signaling by the insulin and the transforming growth factor  system (18, 19).

We report that exogenous EGF up-regulates cell surface integrin ₂ expression, cell adhesion, and cell micromotion on ECM protein irrespective of its inability to induce DNA synthesis above that of the basal EGFr activation in these cells induced by autocrine TGF-. However, autocrine TGF- is also responsible for basal levels of integrin ₂, because disruption of autocrine TGF- signaling by EGFr antibodies or chemical inhibitors of EGFr activation inhibits basal cell adhesion, cell locomotion, and integrin ₂ expression. This indicates differences in sensitivity of EGFr responses in which the signal transduction pathway leading to DNA synthesis is fully saturated by a relatively low level of EGFr activation whereas the signal transduction pathway leading to integrin expression is not saturated by a low level of EGFr activation, instead showing increased response with increasing receptor saturation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Collagen type IV (CN IV) and bovine serum albumin (BSA) were purchased from Sigma Chemical Co. (St. Louis, MO). A polyclonal antibody specific for integrin ₂ subunit (Ab1936) was procured from Chemicon International Inc. (Temecula, CA), and monoclonal antibodies PIE6 (₂), PIB5 (₃), and PID6 (₅) were purchased from Invitrogen (San Diego, CA). EGFr monoclonal blocking antibodies, mAb528 and mAb225, were obtained from Oncogene Science (Manhesset, NY), and tyrphostin AG1478 was purchased from Calbiochem. Anti-EGFr (activated) and anti-actin monoclonal antibodies were purchased from Transduction Laboratories (San Diego, CA) and Santa Cruz Biotechnology (Santa Cruz, CA), respectively. McCoy's 5A medium, transferrin, and insulin were obtained from Sigma, whereas EGF was purchased from R&D Systems (Minneapolis, MN). Arrays of gold-film-coated electrodes for cell motility experiments were purchased from Applied Biophysics Inc, New York.

Cell Culture and Adhesion Assay-- HCT116 cells and TGF- antisense mRNA-expressing HCT116 clones 33 were previously established in tissue culture and extensively characterized (5, 7). Cells were maintained at 37 °C in a humidified incubator with 5% CO₂ in chemically defined serum-free medium consisting of McCoy's 5A medium supplemented with 4 µg/ml transferrin (T) and 20 µg/ml insulin (I) either in the absence or presence of EGF (E) (10 ng/ml) depending upon experimental design.

For adhesion assays, 96-well tissue culture clusters were coated with CN IV by allowing 0-2.5 µg/ml collagen in 0.25 M acetic acid to bind to culture plates at room temperature overnight, followed by blocking with 3% BSA for 3 h at room temperature. Subsequently, the methylthiazole tetrazolium (MTT) procedure was followed as described previously (20, 21).

For the inhibition of adhesion by specific antibodies, 96-well tissue culture clusters were coated with CN IV as described above. Mouse ascites anti-integrin ₂, ₃, and ₅ subunit monoclonal antibodies were added to the plates as 1:50 to 1:500 dilutions, and cells were incubated in the presence or absence of antibody for 30 min at 37 °C. Similarly, EGFr antibody was used to determine the effects of blocking autocrine TGF- on cell adhesion. After trypsinization, cells were incubated at 37 °C with AG1478 for 3 h as an additional approach to determine autocrine TGF--mediated cell adhesion functions. Subsequently, adhesion assays were performed as given above.

Cell Surface Radiolabeling and Immunoprecipitation-- The iodination of cell surface proteins was carried out using suspended cells. Cells at 80% confluency were washed twice with phosphate-buffered saline, suspended by scraping in Tris buffer (125 mM NaCl, 5 mM KCl, 1 mM Tris, 1 mM EDTA, pH 7.4), centrifuged, and resuspended in 0.5 ml of buffer solution containing 125 mM NaCl, 5 mM KCl, 1 mM CaCl₂, and 25 mM HEPES, pH 7.4. Cell surface iodination was carried out by addition of 1 mCi/ml Na ¹²⁵I (Dupont, 17 Ci/mg), 0.2 mg/ml lactoperoxidase (Sigma), and 0.001% H₂O₂ (Sigma) for 10 min at 4 °C. The reaction was stopped by centrifugation, and cells were washed four times in the same buffer. The cell pellet was solubilized by vortexing in an ice-cold buffer consisting of 125 mM NaCl, 1 mM MgCl₂, 25 mM Tris, pH 7.5, and 100 mM n-octal--D-glycopyranoside (Sigma) for 30 min. The insoluble material was removed by centrifugation at 13,000 × g for 10 min. The supernatant protein content was determined by the Bio-Rad assay, and radioactive incorporation was calculated.

Equal amounts of supernatant protein were treated with Triton X-100 (0.5% v/v) and BSA (0.5 mg/ml), precleared by incubation with 50 µl of packed protein A-agarose beads (Oncogene Science, Manhesset, NY) for 2 h, and centrifuged. Precleared supernatants containing equal amounts of protein from each sample were incubated with monoclonal antibodies against integrin ₂ at 1:50 dilution for 2 h at 4 °C with constant rotation, followed by rabbit anti-mouse IgG (ICN, Costa Mesa, CA) at 1:20 dilution for 2 h at 4 °C. The use of equal amounts of protein from each sample ensured that changes in integrin expression were selective rather than a reflection of overall changes in protein synthesis. Immune complexes were precipitated by protein A-agarose beads for 2 h at 4 °C, washed four times with 1% Triton X-100, 25 mM Tris, and 1 mg/ml BSA and once with 150 mM NaCl and 25 mM Tris, at pH 7.4. Laemmli buffer was added, samples were heated at 100 °C for 3 min, and proteins were analyzed by electrophoresis on 7.5% SDS-PAGE gel, followed by Coomassie Blue staining, destaining, gel drying, and autoradiography.

Biotinylation and Western Blot Analysis-- Subconfluent cultures of cells were treated with Joklik's EDTA for 8 min at room temperature, and, subsequently, cells were scraped into a tube and kept on ice. The culture dish was rinsed with cold PBS, and cells were pooled with the Joklik's EDTA fraction. Cells were pelleted by centrifugation in a clinical centrifuge for 1-2 min at 800 × g. The pellet was washed twice with cold PBS, and cells were biotinylated in suspension with NHS-LC-Biotin (Pierce), 0.1 mg/ml in Me₂SO at room temperature for 1 h. Cells were washed three times with PBS and lysed in buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 1 mM CaCl₂, 1 mM MgCl₂ and 1% Nonidet P-40) by shearing them through a 26-gauge needle and centrifuging at 16,000 × g for 20 min at 4 °C in a microcentrifuge. The supernatant was analyzed for protein content by the Bio-Rad protein assay. Equal amounts of protein from treated and untreated (control) cell lysates were incubated with streptavidin-agarose (Invitrogen) for 90 min at 4 °C. Agarose beads were pelleted by centrifugation at 4 °C and subsequently washed five times with lysis buffer containing phenylmethylsulfonyl fluoride. The beads were boiled in 2× Laemmli buffer containing 4% -mercaptoethanol for 10 min, and the supernatant was filtered through Bio-Rad columns (22) and applied in 7.5% SDS-PAGE. The proteins were transferred to nitrocellulose membranes (Hybond) by electroblotting using a mini-Bio-Rad Transblot apparatus. The membrane was blocked for at least 1 h with 5% nonfat dry milk in Triton Tris-buffered saline (TTBS) and subsequently incubated overnight at 4 °C with appropriate primary antibody. After washing the membrane with TTBS, it was incubated for 1 h at room temperature with horseradish peroxidase-conjugated rabbit or mouse secondary antibody. The membrane was washed, and detection of specific binding was achieved by using enhanced chemiluminescence (ECL) reagent (Amersham Biosciences, Inc.).

RNA Isolation and Analysis by RNase Protection Assay-- Total cellular RNA was isolated by lysing cells with guanidine isothiocyanate-EDTA and fractionating the resulting extract through a cesium trifluoroacetic acid gradient (23). Equivalent amounts (40 µg) of RNA samples were used in RNase protection assays. The ₂ subunit template was constructed by subcloning a 292-bp EcoRV-HincII fragment of the human ₂ subunit cDNA into plasmid PBSK(). A high specific activity ₂ subunit riboprobe was synthesized by T₇ RNA polymerase, whereas actin an antisense probe was prepared by S6-RNA polymerase in presence of [³²P]UTP (3000 Ci/mmol; Amersham Biosciences, Inc.). High specific activity ₃, ₅, and ₁ subunit riboprobes were synthesized by T3 polymerase in presence of [³²P]UTP. Normalization of sample loading was assessed as previously described (4), and quantitation of protected fragments was achieved by densitometry (Alpha Imager 2000).

Cell Motility Measurements by the Electrical Cell Impedance Sensor (ECIS) Technique-- Cell motility was quantitated by the micromotion detection method using the ECIS technique (24-26). Cells were plated on small gold electrodes (area 5 × 10⁴ cm²) etched by photolithographic procedures on the bottom of tissue culture wells (area 0.5 cm²) (Applied BioPhysics, Troy, NY). A 1-µA, 4000-Hz AC signal from a constant current source was applied between the small electrode and a much larger counter electrode, while the culture medium acted as an electrolyte. This signal was not large enough to disturb the cells or to change cell behavior (27). The voltage of the system was monitored by a lock-in amplifier (Model SR 830, Stanford Research Systems, Sunnyvale, CA) interfaced with a computer that controlled amplifier settings as well as stored the data collected by the amplifier. The in-phase and out-of-phase voltage across the electrode were recorded by the lock-in amplifier once every second for measuring micromotion and once every 2 min for measuring cell attachment. The ECIS software (Applied BioPhysics, Troy, NY) calculated the resistance and capacitance values of the electrode over this period of time. Attachment and movement of the cells on the electrode changed the flow of the current, resulting in fluctuations in the electrode resistance and capacitance. These cellular movements were called micromotion (25) and were a measure of the motile ability of the cell being measured. As the cells moved on the electrode, the sensitive nature of the lock-in amplifier detected the fluctuations in the resistance and capacitance values (24). These fluctuations were then statistically analyzed using the ECIS software to reveal the percentage variation in resistance, which in turn was a reflection of cellular micromotion on the electrode (25).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Determination of Cell Surface EGFr-- A multipoint binding assay was performed on HCT116 cells grown in serum-free medium lacking EGF to measure the number of cell surface EGFrs in the basal state without exogenous growth factor. Using a computer Scatchard analysis program (EBDA/LIGAND), it was determined that HCT116 cells have ~6.8 × 10⁴ EGF cell surface receptors with an apparent dissociation constant (K_d) of ~10 nM, and the B_max was 110 fmol/10⁶ cells. There is only one class of receptors based on Scatchard analysis.

Effect of ECM Concentration on Adhesion of HCT116 Cells Maintained in the Presence or Absence of EGF-- HCT116 cells have been extensively characterized in previous studies. This work has shown that exogenous EGF (as well as other growth factors) does not influence proliferation or induction of DNA synthesis in the HCT116 colon carcinoma cell line due to autocrine activation of EGFr by TGF- (2, 4-7, 28, 29). Despite saturation of mitogenesis by autocrine TGF-, additional EGFr activation and downstream signaling is observed upon addition of exogenous EGFr ligands. This raised the question as to the function of this response to exogenous EGFr ligand. We hypothesized that EGFr activation resulted in the modulation of  integrin expression.

The maintenance of cells in serum-free chemically defined medium allowed us to determine the effects of the long term (~5 days) presence or absence of EGF on cellular adhesion to tissue culture plates coated with basement membrane type IV collagen (0-2.5 µg/ml). HCT116 cells showed a 2-fold higher level of adhesion to this extracellular matrix (ECM) protein than on BSA coating in the absence of EGF. Maintenance of the cells in an EGF-containing medium resulted in approximately a 6-fold increase in adhesion to CN IV (2.5 µg/ml) relative to adhesion in the absence of EGF (Fig. 1A). In contrast, withdrawal of EGF from the culture medium for 48 h resulted in lower attachment to CN IV (Fig. 1B). HCT116 cells adhere well to other ECM proteins fibronectin (FN) and laminin (LN) in the absence of EGF because they showed 2-fold higher levels of adhesion on FN and LN than on BSA coating (Fig. 1C). Maintenance on EGF-containing medium resulted in more than a 4-fold increase in adhesion on LN and about a 3-fold increase in adhesion on FN (Fig. 1D) relative to adhesion in the absence of EGF. Thus, HCT116 cells exhibit differential adhesion with respect to specific ECM proteins. Higher cell adhesion was observed on CNIV than on LN and FN. It is also noteworthy that higher concentrations (10 µg/ml) of FN and LN were required for optimal cell adhesion as compared with lower concentrations of CNIV (2.5 µg/ml). Approximately, 2.5 µg (20 pmol) of CN IV are equivalent to 4.4 µg (20 pmol) of FN and 8.8 µg (20 pmol) of LN, whereas the molarities of the EGFr ligands used are equivalent. Thus, after adjustment for molar concentration, higher cell adhesion was observed on CN IV than on FN and LN. At 20 pM concentration of CN IV, cell adhesion in the presence of EGF is 5.5-fold over that in the absence of EGF. Moreover, adhesion to 20 pg of CN IV is 2.25-fold higher than that of bovine serum albumin in the absence of EGF. In comparison to CN IV, the cell adhesion on 20 pM LN in the presence of EGF is ~3.5-fold over that in the absence of EGF. Adhesion to 20 pM FN is even less than that of LN. The role of TGF- as a promoter of cell adhesion in HCT116 cells was also determined (Fig. 1E). The effect of exogenously added TGF- (the ligand responsible for autocrine control of these cells) was essentially the same as that of EGF.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   The effect of ECM protein concentration on adhesion of HCT116 cells continuously maintained in the absence or presence (10 ng/ml) of exogenous EGF or TGF-. 96-well tissue culture plates were coated with collagen IV at 0, 0.025, 0.1, 0.25, 1.0, and 2.5 µg/ml or with FN and LN at 0, 0.1, 0.5, 1.0, 5.0, and 10.0 µg/ml for overnight at room temperature; nonspecific sites were blocked with BSA (3%) for 3 h, and subsequently, wells were washed once with PBS. Subconfluent cell cultures were trypsinized and seeded at 6 × 104 cells/well onto ECM and BSA-coated plates and incubated for 90 min at 37 °C. Adhesion assays were carried out as described under "Experimental Procedures." The relative number of attached cells was expressed as a percentage increased over BSA. A, cells maintained for 5 days in the EGF-deficient medium were changed to the same fresh medium 48 h prior to assay (open circle); cells maintained for 5 days in EGF-deficient medium changed to medium supplemented with EGF 48 h prior to assay (closed circle). B, cells maintained for 5 days in EGF-supplemented medium changed to the same fresh medium 48 h prior to assay (open box); cells maintained for 5 days in EGF-supplemented medium changed to fresh medium without EGF 48 h prior to assay (closed box). HCT116 cells maintained in the absence (C) or presence (D) of EGF on FN (closed circles) or LN (open circles). E, cells were maintained for 5 days in the absence of exogenous growth factor. TGF- was added for 48 h prior to adhesion assay; HCT116 cells in the absence (open squares) or presence (closed squares) of TGF-. Error bars represent the standard error of four experiments performed in triplicate.

Adhesion of HCT116 Cells to CN IV Is Mediated Predominately by ₂ Integrin-- The specificity of ₂ integrin in mediating adhesion of HCT116 cells was determined by treatment with specific functional blocking antibodies to inhibit binding to CN IV. Monoclonal anti-₂ integrin antibody was highly effective in preventing HCT116 cell adhesion to CN IV both in the presence and absence of EGF. Inhibitory levels ranged from 65 to 8% in the absence of EGF and 80 to 45% in the presence of EGF at antibody dilutions ranging from 1:50 to 1:500) (Fig. 2A). Antibody to the integrin ₅ subunit had no effect on HCT116 cell adhesion to CN IV. However, antibody to the integrin ₃ subunit was only slightly inhibitory (5-10% at 1:50 dilution) to HCT116 cell adhesion to CN IV. Integrin subunits ₃ and ₅ are the predominant cell adhesion receptors for laminin and fibronectin, respectively, on these cells (Fig. 2, B and C).

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Inhibition of HCT116 cells adhesion to CN IV by antibodies to integrin receptors in the absence or presence of EGF. 96-well tissue culture plates were coated with (A) CN IV (5 µg/ml), (B) LN (10 µg/ml), and (C) FN (10 µg/ml). Monoclonal antibodies (Ab) to anti-integrin 2, 3, and 5 subunits were added at different dilutions as indicated. Adhesion assays were performed as detailed under "Experimental Procedures." The 3 and 5 monoclonal antibodies were used as negative controls at the highest concentrations (1:50). Cells were maintained in the absence or presence of 10 ng/ml EGF. A: lane 1, without Ab; lanes 2-4, 2 Ab; lane 5, 3 Ab; lane 6, 5 Ab. B: lane 1, without Ab; lanes 2-4, 2 Ab; lanes 5-7, 3 Ab; lane 8, 5 Ab. C: lane 1, without Ab; lane 2, 2 Ab; lane 3, 3 Ab; lanes 4-6, 5 Ab. Each binding value represents the mean of two individual experiments performed in triplicate.

Antibody to ₃ subunit inhibited binding of HCT116 cells to LN-coated plates in a concentration-dependent manner (Fig. 2B). Inhibitory levels ranged from 80 to 15% in the absence of EGF and 90 to 40% in the presence of EGF at antibody dilutions ranging from 1:50 to 1:500. Although the integrin ₂ predominately mediates adhesion to CN, it has been shown that it is a receptor for LN on endothelial cells and some types of tumor cells (30, 31). Consequently, it was not surprising that antibody to integrin ₂ subunit was also capable of inhibiting HCT116 cell adhesion to LN (Fig. 2B). However, inhibition by antibody to ₂ subunits was not as effective as anti-₃ subunit. Antibody to the integrin ₅ had no effect on HCT116 cell adhesion to LN (Fig. 2B), whereas the antibody was extremely effective in preventing adhesion to FN. Inhibitory levels ranged from 90 to 70% both in the presence and absence of EGF at dilutions ranging from 1:50 to 1:500 (Fig. 2C). These results showed that cell adhesion to FN was primarily through integrin ₅. Neither antibody to the ₂ nor the ₃ subunit affected adhesion of HCT116 to FN.

Kinetics of Cell Adhesion to CN IV-- The experiments described above were performed with cells continuously maintained in the presence or absence of EGF. Therefore, they could reflect a steady-state situation in which cells may have made adaptations leading to differences in cell adhesion and integrin expression, which were not related to control by EGFr activation. Thus it was necessary to determine whether short term changes in EGF exposure could alter cell adhesion and integrin expression. Therefore, we determined the kinetic effects on cell adhesion of removal of EGF from the medium of HCT116 cells adapted to growth in the presence of the polypeptide as well as determining the kinetic effects of addition of EGF to the medium of HCT116 cells adapted to growth in EGF-deficient medium. Subsequent characterizations were performed with CN IV and integrin ₂, because this pair appeared to enhance adhesion to a greater extent than the other integrin-ECM combinations investigated.

Significant enhancement of adhesion to CN IV was observed within 6 h of EGF addition, and, by 12 h, cell adhesion was increased by 2-fold rising to nearly 4-fold by 24 h and 6-fold by 48 h (Fig. 3). Withdrawal of EGF from the medium of EGF-adapted HCT116 cells would be predicted to generate the opposite results. Within 6 h of EGF removal from the EGF-adapted cells, adhesion to CN IV was reduced by 15% and was further reduced by nearly 35% to 70% between 12 and 48 h.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Kinetics of effects of addition (A) or withdrawal (B) of EGF on HCT116 cell adhesion to CN IV. Cells maintained in the absence or presence of 10 ng/ml EGF were plated into 6-well tissue culture plates (5 × 104 cells/well) and allowed to grow for 3-5 days. At this point EGF-containing medium was replaced with medium devoid of EGF whereas medium lacking EGF was replaced with EGF-containing medium for various periods of time (2, 6, 12, 24, 36, and 48 h). Control cells received the same fresh medium on which they were originally grown at the same times. Cells were harvested and plated at 6 × 104 cells per well into 96-well tissue culture plates coated with BSA (open circle) or CN IV 5 µg/ml (closed triangle), incubated for 90 min at 37 °C and assayed for attachment by the standard procedure as described under "Experimental Procedures." Data are reported as the percentage of either EGF-added or EGF-withdrawn adherent cells relative to the attached cells maintained in the original medium. Error bars represent the standard error of the average of two experiments performed in triplicate.

Kinetics of Cell Surface Integrin ₂ Expression and Steady-state Levels of ₂ mRNA-- The results of integrin antibody blocking experiments were consistent with a role for EGF in controlling cellular adhesion to CN IV by controlling cell surface expression of the integrin₂ subunit. Therefore, we determined cell surface expression of integrins in colon carcinoma cells continuously maintained in the absence or presence of EGF in chemically defined tissue culture medium by immunoprecipitation of ¹²⁵I-labeled cell surface integrins. As expected from integrin blocking antibody experiments, HCT116 cells showed cell surface expression of integrin ₂ subunit when it was maintained in the absence of EGF. Increased adhesion by HCT116 cells continuously maintained in EGF-supplemented medium suggested that integrin expression would be increased under these conditions. The kinetics of cell adhesion to CN IV should also reflect changes in cell surface functional integrin ₂ expression. Increased integrin ₂ subunit expression was observed within 6 h of EGF addition to EGF-deficient cells. The 6-h level increased by 3-fold at 12 h and about 5-fold at 24 and 48 h (Fig. 4A, left upper panel). EGF withdrawal resulted in a significant reduction of cell surface ₂ subunit expression within 12 h with more than 50% reduction in expression by 24 h. There was a continued decline in integrin ₂ expression up to 48 h post EGF withdrawal (Fig. 4A, left bottom panel). The quantitation of the kinetics of integrin ₂ expression in the presence or absence of EGF are shown in Fig. 4A (right upper and lower panels, respectively). In contrast to integrin ₂, the integrin ₁ protein is not modulated by EGF, although it is expressed in HCT116 cells (Fig. 4B), thus indicating the selectivity of EGF-dependent changes in integrin expression by HCT116 cells. Furthermore, exogenous EGF did not have any effect on actin levels. Transforming growth factor- was equivalent with EGF in terms of its ability to induce integrin expression (data not shown).

View larger version (36K): [in this window] [in a new window]   Fig. 4.   A, kinetics of EGF modulation of cell surface integrin 2 subunit protein expression in HCT116 cells. Cells were plated as described in Fig. 3. At 3-5 days, cells maintained in the absence or presence of EGF (10 ng/ml) were changed to medium containing or lacking EGF for 0, 6, 12, 24, and 48 h. All cells received a change of fresh medium at the same time and were then harvested and iodinated at the indicated times after medium changed as described under "Experimental Procedures." Detergent extracts of surface-labeled cells from adherent cells were treated with monoclonal antibodies against the human integrin 2 subunit followed by complexing with rabbit anti-mouse IgG. Immune complexes were precipitated by protein A-agarose beads and analyzed by electrophoresis on 7.5% polyacrylamide gel under reducing conditions. Actin was used as a control. The right panels show densitometry quantitation of integrin 2. B, kinetics of EGF modulation of integrin 1 protein expression in HCT116 cells. Cells were maintained in the absence of EGF for 5 days, and EGF (10 ng/ml) was added to the medium for 0, 12, 24, and 48 h. The cells were harvested and biotinylated as described under "Experimental Procedures." Cell lysates were analyzed by Western blot using polyclonal Ab1934 (Chemicon) against integrin 1. Actin was used as a control.

To determine the effect of EGF on the expression of integrin ₂, total RNA (40 µg) was analyzed by RNase protection assays from HCT116 cells, which were treated for different time periods with EGF. As shown in Fig. 5A, EGF enhances levels of integrin ₂ mRNA (2- to 4-fold) in a temporal fashion over the course of 48 h post addition of the growth factor. The EGF-mediated kinetics of integrin ₂ mRNA expression were similar to those found for protein levels. The quantitation of kinetics of integrin ₂ mRNA expression is shown in Fig. 5A (lower panel). Furthermore, we have determined that EGF affects the expression of integrin ₃ (Fig. 5B) and integrin ₅ genes (Fig. 5C) but the increases in expression levels of these  subunits are lower than those on the expression of integrin ₂ gene. An RNase protection assay showing the induction by EGF on expression of integrin ₁ mRNA is shown in Fig. 5D. The effects of exogenous EGF on cell surface integrin expression suggested that autocrine TGF- may have a role in determining functional integrin expression and cell adhesion as well.

View larger version (37K): [in this window] [in a new window]   Fig. 5.   A, temporal expression of integrin 2 subunit mRNA levels detected by RNase protection assay in HCT116 cells. Total RNA (40 µg), isolated from the EGF (10 ng/ml)-treated cells for 0, 6, 12, 24, and 48 h, was hybridized with 32P-labeled RNA probes of the integrin 2 subunit (0.5 × 106 cpm) and actin (8000 cpm) simultaneously according to the details given under "Experimental Procedures." The sizes of the protected fragments on urea-polyacrylamide gel electrophoresis are indicated by the arrows. Actin mRNA levels are shown for normalization of sample loading. Yeast tRNA was used as a negative control. In A, the lower panel shows densitometry quantitation of integrin 2 mRNA. B-D, effect of EGF on expression of integrin 3, 5, and 1 subunit mRNA levels detected by RNase protection assay in HCT116 cells. Total RNA (40 µg), isolated from the untreated or EGF (10 ng/ml)-treated cells for 48 h, was hybridized with 32P-labeled RNA probes of the integrin 3 subunit (left panel) (0.5 × 106 cpm), 5 subunit (middle panel), 1 subunit (right panel), and actin (8000 cpm) simultaneously according to the details given under "Experimental Procedures." The sizes of the protected fragments on urea-polyacrylamide gel electrophoresis are indicated by the arrows. Actin mRNA levels are shown for normalization of the sample loading. Yeast tRNA was used as a negative control.

Antibody to the EGFr Blocks HCT116 Cell Adhesion to CN IV, Integrin ₂ Expression and EGFr Activation-- The experiments described above indicated a role for exogenous EGF in ₂ integrin-mediated cell adhesion to CN IV via EGFr by HCT116 cells. HCT116 cells have an active autocrine TGF- loop, which is responsible for their growth factor independence (4, 7). If this TGF- also acted in an autocrine manner to affect cell adhesion, it would be expected that addition of anti-EGFr antibody would block cell adhesion of cells maintained in the absence of exogenous EGF. Fig. 6A shows that the anti-EGFr antibody mAb528 was effective in blocking adhesion to CN IV by HCT116 cells maintained in the absence of EGF. A similar experiment was performed with cells maintained in the presence of EGF by using a 50-fold excess of EGFr antibody. Treatment of HCT116 cells under these conditions was effective in blocking 90% of the adhesion to all CN IV-coated plates (Fig. 6B). HCT116 cells were treated with EGFr-blocking mAb528, and its effect on integrin ₂ levels was observed by Western blot analysis. Fig. 6C shows that mAb528-treated cells showed inhibition of expression of integrin ₂ (lane 1) relative to control HCT116 cells (lane 2). Loss of integrin ₂ expression and adhesion were directly correlated with reduced EGFr activation by the mAb.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   EGF receptor antibody blocks HCT116 cell adhesion to CN IV, integrin 2 expression, and EGFr activation. Cells maintained either in the absence or presence of EGF were treated with 10 µg/ml EGF receptor blocking monoclonal antibody (mAb528) for 48 h. Cells were then trypsinized and inoculated at 6 × 104 cells per well into BSA- and CN IV-coated plates and incubated at 37 °C for 90 min in the absence or presence of monoclonal antibody as indicated. A, experiment was performed in the absence of EGF, whereas in B, the experiment was performed in the presence of EGF. Non-adherent cells were washed off, and adherent cells were determined by MTT assay as described under "Experimental Procedures." Error bars represent the standard error of two experiments performed in triplicate. C, cells cultured in a 6-well plate were either treated with 15 µg/ml EGFr blocking mAb528 (lane 1) or with mouse IgG (lane 2) for 48 h. Cells were biotinylated for integrin 2 and lysed, and equal amounts of protein were analyzed by Western blot as described under "Experimental Procedures." Actin was used as a loading control.

TGF- Antisense-transfected Cells Show Attenuation of Cell Adhesion, Integrin ₂ Expression, and EGFr Activation-- The results from above indicate that, in addition to its effect on growth factor-independent mitogenesis, autocrine TGF- contributes to adhesion to CN IV by HCT116 cells. If TGF- were acting in an autocrine manner to affect cell adhesion, it would be expected that anti-TGF- transfected cells would inhibit integrin-mediated cell adhesion. TGF- antisense-transfected cells showed reduced adhesion to CN IV relative to control parental cells. The percentage of reduction in adhesion of the TGF- antisense clone varied from 80 to 47% depending on the concentration of coated CN IV when compared with HCT116 cells, under identical conditions (Fig. 7A). These data are consistent with the effects of EGFr blocking antibody on CN IV-mediated adhesion by HCT116 cells. Furthermore, the reduction in cell adhesion of the TGF- antisense clone is correlated with a reduction in the expression of integrin ₂ protein. Lane 1 in Fig. 7B shows levels of integrin ₂ in HCT116 cells, whereas comparison with lane 3 shows reduced levels of integrin ₂ in the TGF- antisense-transfected cells. This indicated that autocrine TGF- plays a role in the steady-state expression of integrin ₂. The expression of integrin ₂ was rescued by treating antisense cells with exogenous EGF (lane 4), indicating that reactivation of EGFr rescues the antisense effect. Lane 2 shows induction of integrin ₂ protein by exogenous EGF in HCT116 cells. Actin levels were not altered. Expression of integrin ₂ mRNAs paralleled the protein expression described above (data not shown).

View larger version (26K): [in this window] [in a new window]   Fig. 7.   A, comparison of adhesion of HCT116 neo control (open bars) and HCT116 TGF- antisense transfected (HCT116-A.S.33) cells (dark bars) to CN IV. Substrates were prepared by coating tissue culture 96-well plates with CN IV at concentrations of 0, 0.25, and 0.50 µg/ml overnight at room temperature. Cells were seeded at 6 × 104 cells/well onto coated plates and incubated for 90 min at 37 °C. The relative numbers of the attached cells were determined by MTT assay as described under "Experimental Procedures." Optical density values at 595 nm on BSA-coated wells were subtracted. B, comparison of integrin 2 expression of HCT116 and HCT116 TGF- antisense-transfected cells. Cells were maintained in the absence of EGF for 5 days. EGF was added to the medium 48 h prior to harvesting. Cells were biotinylated and lysed, and equal amounts of protein were analyzed by Western blotting as given under "Experimental Procedures." Lanes 1 and 3 show basal levels of integrin 2 in HCT116 neo and antisense cells, respectively; lanes 2 and 4 show stimulation of integrin 2 by EGF in HCT116 neo and antisense cells, respectively. Actin was used as a loading control.

Attachment of HCT116 Cells to Electrodes Precoated with Either BSA or CN IV as a Function of Time-- To explore the biological role of increased cell adhesion and expression of integrin ₂ by EGF, we determined the role of the EGF signaling pathway on HCT116 cell motility using CN IV as a substrate. The ECIS technique (24-26) was used to quantitate cell motility. A small AC signal was applied across the gold electrode on which cells were plated, while the resistance and the capacitance of the electrode were measured over time. Within the first 2 h at 37 °C, there was a notable rise in resistance of the electrodes coated with CN IV but not in those coated with BSA (Fig. 8). This increase in resistance in collagen-coated electrodes is due to initial cell attachment, pH, and temperature changes which result in an increase in area covered by cells on the electrode (32). Of note, however, is that the CN IV-coated electrodes (E3-E5) displayed a marked fluctuation in resistance, whereas those electrodes coated with BSA were smooth (E1 and E2). These fluctuations in resistance represent the presence of viable attached and spread cells on these electrodes, and are caused by the small movements of these cells on the electrodes (25). These small cellular movements, termed micromotion (24), result in constant changes in cell-cell or cell-substrate interactions, which accordingly changes the rate of current flow across the cell layer (25). In contrast, the BSA-coated electrodes did not display these fluctuations in resistance indicative of the absence of attached cells on these electrodes. It is noteworthy that cell attachment to and cell motility on the extracellular matrix are biologically related processes (33-36). If fewer cells are attached to the coated electrodes, the total voltage across the electrode is reduced. Consequently, the attachment curves will show a reduced gain as in Fig. 8 (E1 and E2 versus E3-E5). Micromotility is the additional fluctuation (as a percentage calculated by the ECIS software) that results from cell movement above the total voltage as shown in Figs. 9-11.

View larger version (19K): [in this window] [in a new window]   Fig. 8.   Attachment of HCT116 cells to electrodes precoated with either BSA or CN IV as a function of time. Each electrode well was inoculated with 2.0 × 104 cells/400 µl medium as given under "Experimental Procedures." Electrodes E1 and E2 were coated with BSA (3%), whereas electrodes E3 through E5 were coated with CN IV (5 µg/ml). The cell attachment was recorded for 23 h.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Functional blocking monoclonal antibody to integrin 2 decreases cell micromotion in a concentration-dependent fashion. HCT116 cells (6 × 104 cells/condition) in serum-free medium were incubated with either mouse IgG (1:50 dilution; top panel) or with integrin 2 blocking mAb (clone PIE6) at dilutions 1:50 (middle panel) or 1:150 (bottom panel) for 30 min at 37 °C. Subsequently, cells were transferred to electrode wells precoated with CN IV. After 3 h of cell attachment to electrodes, cell micromotion was recorded for 70 min.

Functional Blocking mAb to Integrin ₂ Decreases Cell Micromotion in a Concentration-dependent Fashion-- To establish that cell motility on CN IV was specifically integrin ₂-mediated, cells were preincubated either with mouse IgG or with different concentrations of a function blocking mAb (clone PIE6) before recording micromotion. The percent variations in resistance observed were 3.192% (control IgG), 0.365% (mAb 1:50 dilution), and 1.105% (mAb 1:150 dilution). As shown in Fig. 9, the decrease in cell motility (resistance) of HCT116 cells by mAb at 1:50 dilution was 89% (middle panel), and at 1:150 dilution of mAb, the decrease in locomotion observed was 65% (bottom panel) as compared with control IgG-treated cells (top panel).

EGF Enhances CN IV-induced Cell Motility Whereas Tyrphostin AG1478 Abrogates EGF Effects-- The role of EGFr activation in micromotion was then investigated using tyrphostin AG1478, which is a highly selective EGFr inhibitor (37). In untreated HCT116 cells, the percent variation in resistance measured over a period of about 70 min was found to be 4.130% (Fig. 10A). Addition of EGF (10 ng/ml) to the cell medium increased the fluctuations, indicating an increase in cell motility, such that the percent variation in resistance was now 7.103% (Fig. 10B). To ensure that this effect of EGF was indeed due to the stimulation of the EGFr, we then further treated these cells with AG1478 (10 µM). Addition of AG1478 to the EGF-stimulated cells abrogated the increase in cell motility caused by EGF (percent variation in resistance 3.119%; Fig. 10C). The basal micromotion percent variation in resistance value (4.13%) was slightly higher than the micromotion observed in the presence of AG1478, probably reflecting autocrine TGF- contributions to micromotion as well. These observations were confirmed by results from three experiments shown in Fig. 10D.

View larger version (22K): [in this window] [in a new window]   Fig. 10.   EGF enhances cell motility whereas tyrphostin AG1478 abrogates EGF effect. Electrode arrays after precoating with CN IV (5 µg/ml) were used in these experiments. HCT116 cells were plated in serum-free medium at a density of 2 × 104 cells/electrode/well. At 24 h, the serum-free medium was replaced by McCoy's 5A medium containing transferrin and insulin (see "Experimental Procedures"). Subconfluent (70-80%) cultures were either (A) not treated or (B) treated with EGF (10 ng/ml) and (C) treated with EGF + AG1478 (10 µM). Cells growing on collagen-coated gold electrodes were monitored for cell attachment for 20 h followed by micromotion (resistance) over a period of 70 min. The effects of EGF and AG1478 on cell motility representing three experiments are shown in Fig. 10D.

HCT116 Cell Micromotion Is Attenuated in TGF--transfected Antisense Cells-- To define the role of autocrine TGF- in cell motility, micromotion of HCT116 cells was compared with that of HCT116 antisense cells under identical conditions (Fig. 11). In the upper left panel, the percent variation in resistance (micromotion) of control HCT116 cells was determined to be 2.939%, and, upon addition of EGF, the resistance was increased to 5.370% (83% increase, upper panel, right). The lower left panel shows a percent variation in resistance of 1.741% in control antisense cells, which is about 40% lower than that of parental HCT116 cells, thus showing that autocrine TGF- contributes to cell micromotion. Furthermore, addition of EGF to antisense cells increases percent variation in resistance to 2.995% (lower right panel), thus showing that exogenous EGF is capable of rescuing micromotion in antisense cells. Taken together our results point to a significant role of autocrine TGF- in controlling steady-state cell micromotion, cell adhesion, and integrin ₂ expression in addition to controlling mitogenesis.

View larger version (30K): [in this window] [in a new window]   Fig. 11.   Comparison of micromotion of HCT116 and HCT116 antisense-transfected cells. Precoated electrode arrays with CN IV (5 µg/ml) were used in these experiments. HCT116 and HCT116 antisense cells were plated in serum-free medium at a density of 2 × 104 cells/electrode/well. At 24 h, the serum-free medium was replaced by McCoy's 5A medium containing transferrin and insulin (see "Experimental Procedures"). A, subconfluent (70-80%) HCT116 cultures were either not treated (upper left panel) or treated (upper right panel) with EGF (10 ng/ml) and (B) subconfluent (70-80%) HCT116 antisense cells were either not treated (lower left panel) or treated (lower right panel) with EGF (10 ng/ml). Cells growing on collagen-coated gold electrodes were monitored for cell attachment for 20 h followed by micromotion (resistance) over a period of 70 min.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Earlier we published a general profile of adhesion to distinct ECM proteins in the absence of growth factor by several colon cancer cell lines, including HCT116 cells (21). We have now characterized the role of EGFr activation (both endogenous and exogenous) on cell adhesion. HCT116 cells continuously maintained in EGF were compared with cells devoid of EGF for adhesion to CN IV. Maintenance of cells in EGF enhanced adhesion of HCT116 cells to the ECM protein (Fig. 1). The cell adhesion to CN IV was higher than on FN or LN, thus showing differential adhesion of HCT116 cells on ECM proteins. These results indicated that a known mitogen, EGF, could stimulate cell adhesion, but that the stimulation was unrelated to the mitogenicity of EGF, because HCT116 cells do not respond to exogenous EGF with increased cell proliferation (2, 4). The increase in cell adhesion and integrin production in response to EGF in cells continuously maintained on the growth factor could be due to a steady-state situation in which cells may have made adaptations to culture conditions in different media. We investigated this by determining kinetic effects on cell adhesion to ECM protein following addition or withdrawal of EGF from cells that were adapted to grow in the absence or presence of EGF, respectively. The results indicated a relatively rapid modulation of cell adhesion in response to addition of EGF or TGF- to EGF-deficient cells. We demonstrated that EGF enhanced cell adhesion by stimulating expression of functional integrin ₂. This was further confirmed by treating HCT116 cells with specific monoclonal antibodies to integrin ₂, which blocked adhesion of HCT116 cells to CN IV in a concentration-dependent fashion. Inhibition of cell adhesion by antibodies was more effective on cells grown in EGF suggesting that a higher proportion of the binding to substrate protein in EGF-maintained cells was due to integrin binding than in cells devoid of EGF (Fig. 2). Adhesion of HCT116 cells was significantly enhanced before 24 h and continued to increase up to 48 h after treatment with EGF (Fig. 3A). Removal of EGF from HCT116 cells had the opposite effect as adhesion levels decreased within 24 h. Interestingly, removal of EGF did not affect adhesion as extensively as addition within the 48-h period in which experiments were performed (Fig. 3B). It may be that autocrine TGF- levels in these cells help counteract the immediate effects of EGF removal from the medium so that kinetic effects on cell adhesion are slowed. In addition, the half-life of integrin ₂ may be quite long. Immunoprecipitation of integrin subunit ₂ with specific antibodies indicated that HCT116 cells maintained in EGF expressed severalfold higher amounts of cell surface integrin than cells devoid of EGF (Fig. 4A). Under these conditions, levels of actin and ₁ protein did not change (Fig. 4B). Similarly, RNase protection assays revealed that exogenous EGF up-regulated (2- to 4-fold) mRNA levels of integrin ₂ within 48 h (Fig. 5A). The increases in expression in integrin ₃ and ₅ mRNAs by EGF were relatively small. The kinetic changes of cell surface integrin subunit ₂ expression reflect the differences in kinetics of cell adhesion observed with EGF addition or withdrawal.

We used different approaches to confirm cell adhesion and integrin expression as a less sensitive function of EGFr than mitogenesis. These included the use of specific blocking antibodies, and TGF- antisense-transfected cells. If the increase in adhesion of cells maintained in EGF was in fact due to EGF interaction with its receptor, then antibodies that blocked the receptor should be capable of blocking adhesion. An EGFr antibody (designated mAb528) is known to recognize the EGF/TGF- binding site, competes for EGF/TGF- binding, and blocks EGF/TGF--induced receptor autophosphorylation (10, 38). Monoclonal antibody 528 was capable of inhibiting 90% of HCT116 cell adhesion to CN IV, abrogated expression of integrin ₂ protein, and inhibited activation of EGFr, thus confirming that EGF mediates cell adhesion and integrin ₂ expression via receptor irrespective of its mitogenic properties (Fig. 6). These data confirmed that changes in CN IV adhesion and integrin ₂ expression in HCT116 cells occurred via the EGFr signaling pathway. The role of TGF- in cell adhesion and integrin expression is demonstrated by the use of TGF- antisense-transfected cells (Fig. 7, A and B). The antisense cells showed attenuated cell adhesion and integrin ₂ expression. Expression of integrin ₂ is rescued by exogenous EGF, thus exhibiting the specificity of the effect on EGFr by TGF- antisense. The exact site(s) of the tyrosine residue(s) involved in the activation of the EGFr in integrin expression and cell adhesion remain(s) to be determined. In general, the EGFr autophosphorylates at least five tyrosine residues in the cytoplasmic tail in response to EGF (39, 40). The stoichiometry of the tyrosine autophosphorylation sites of the EGFr in mammalian cells is not known. The hierarchy of autophosphorylation sites may offer different regulatory roles in the EGFr function. Epidermal growth factor receptor uniquely binds at multiple clustered tyrosine sites with adaptor proteins containing a single SH2 domain. It has been reported by Batzer et al. (40) that both Grb 2 and Shc adaptor proteins have high affinity and low affinity binding sites on the EGFr. It was revealed that Grb 2 primarily binds to activated tyrosine 1068 and with low affinity to tyrosine 1086, whereas Shc primarily binds to tyrosine 1173 and binds to tyrosine 992 in a less sensitive fashion. However, the functional significance of secondary sites in the intracellular domain of EGFr has yet to be elucidated. One hypothesis is that, in HCT 116 cells, autocrine TGF- saturates primary high affinity docking sites by adaptor proteins, whereas the less sensitive function of EGFr may be due to the occupancy of one or more secondary tyrosine autophosphorylation sites by exogenous EGF, thus showing the functional regulatory role of minor autophosphorylated EGFr sites via an auxiliary mechanism.

It has been reported that, in human A431 epidermoid carcinoma cells, EGFr activation leads to inhibition of cell growth through induction of p21^cip1/WAF1 at high levels of receptor occupation (41, 42). In contrast, p21^cip1/WAF1 is not induced at low levels of EGFr activation. This phenomenon, however, is somewhat different from the situation we observe in HCT116 cells. In the case of A431 cells, it is likely a matter of the high versus low affinity receptors typically seen in cells with amplified EGFr (43). In contrast to the delay of induction of p21^cip1/WAF1 until a relatively high receptor occupancy is obtained in A431 cells, we examined the situation in which the DNA synthesis response is saturated by the relatively low EGFr occupation level resulting from autocrine TGF-. HCT116 cells have ~6.8 × 10⁴ EGF cell surface receptors with an apparent dissociation constant (K_d) of ~10 nM, and the B_max was 110 fmol/10⁶ cells. Unlike A431 cells, there is only one class of receptors, expressed by HCT116 cells based on Scatchard analysis. It is difficult to say what proportion of EGFrs is endogenously bound. Acid treatment to remove receptor occupation prior to Scatchard experiments does not indicate large amounts of endogenous occupation relative to the 68,000 receptors seen on the cell surface. We believe this is a result of the intracellular activation of these receptors as based on the inability of exogenous EGFr and TGF- blocking antibodies to inhibit cell growth and inhibition of DNA synthesis by these cells (5, 7). Alternatively, it may be due to occupation of a proportion of the cell surface receptors by a transmembrane-bound TGF- precursor, which is not labile to the acid, as we have described previously (44).

The expression level of integrin ₂ shows a wide window of response, relative to mitogenesis, ranging from low EGFr occupation by autocrine TGF- to saturation by exogenous EGF or TGF-. It is initiated at low level receptor occupation as evidenced by its diminution by treatment with EGFr blocking antibody, which inhibits basal EGFr activation resulting from autocrine TGF-. Moreover, stable transfection with a full-length TGF- antisense cDNA inhibits TGF- expression (5, 7) and basal EGFr activation in these cells (data not shown). Addition of exogenous EGF or TGF- results in further EGFr activation (data not shown), which is associated with higher expression of integrin ₂ (Fig. 4). These changes in turn lead to alteration in cell adhesion (Fig. 1) and cell micromotion (Fig. 10). The reduction in the low levels of basal integrin ₂ by EGFr antibody (Fig. 6C) and TGF- antisense on cell adhesion (Fig. 7A) and cell micromotion (Fig. 11) shows that this is an expanded window of response relative to mitogenesis. Thus, there is a difference in the response windows based on extent of EGFr activation. They may not be strictly hierarchical in the sense of cellular priorities or due to different subsets of receptors but, rather, in the sense of degree of saturation of response at a given level of EGFr activation.

Results from this study are consistent with such a role for autocrine TGF-, because the EGFr blocking antibody was able to inhibit cell adhesion in HCT116 cells devoid of EGF in the medium. These results were further supported by adhesion assays in which the adhesion of HCT116 cells was directly compared with adhesion of TGF- antisense-transfected cells. The cell adhesion was markedly reduced in TGF- antisense-transfected cells as compared with the control HCT116 neo clones (Fig. 7A). Thus, showing that autocrine TGF- contributes to basal levels of integrin ₂ expression and cell adhesion. The presence of a strong TGF- loop in HCT116 cells is one of the salient features contributing to the highly malignant properties to this cell line, and antisense TGF- transfectants show loss of basal EGFr activation as well as a requirement for exogenous EGF for optimal mitogenesis (5). It has been proposed that the major growth advantage of autocrine TGF- in malignant cells may be due to the increased ability of cells to re-enter the cell cycle (6). Similarly, it is conceivable that cancer cells involved in metastasis will derive an advantage from a strong autocrine TGF- loop in both growth and motility, because initially the number of cells contributing to metastatic behavior is very small.

Cell locomotion may have significance in colon cancer metastasis (45). We used a cell-substrate electrical cell impedance sensor (ECIS) technique to determine the effects of EGF, EGFr inhibitor AG1478, and integrin ₂ blocking antibody on HCT116 cell motility. Using this technique, cell motion may be quantitatively measured at the nanometer level (micromotion) (25). In any dynamic cell system the cell-cell and cell-substrate interactions are constantly changing due to regular metabolic processes (26). As such, the physical spaces between two cells or between the cell and the surface on which it is growing, changes as well. This results in small cellular movements termed micromotion, which occur at the nanometer scale and cannot be detected in a regular microscope. As the gap between the cells or between the cell and its substrate fluctuates, so does the current flowing across the cell layer (25, 26). The sensitive nature of the lock-in amplifier detects the changes in this current and voltage and translates them into resistive and capacitive units as presented here. Micromotion detected by ECIS technique is directly related to conventional cell motility (46). Drugs that inhibit cell migration and motility, such as cytochalasin D, also inhibit micromotion (47). Micromotion detected by the ECIS technique has been successfully used to detect cell migration and morphological changes in a variety of systems (47, 48). In this study, we have shown that the basement membrane CN IV-mediated adhesion and exogenous EGF significantly enhance micromotion independent of cell growth in human colon cancer cells. The cell locomotion was not observed on BSA-coated electrodes (Fig. 8). Therefore, these cells must have an appropriate ECM for growth factor to affect cell motility, suggesting that the signaling cascades for integrins and growth factors are linked. The cell motility on CN IV-coated electrodes was mediated by integrin ₂ as demonstrated by using a blocking monoclonal antibody (PIE6), which markedly reduced cell micromotion (Fig. 9). The increase in amplitude of fluctuations (resistance) on CN IV caused by EGF (Fig. 10B), as compared with control (Fig. 10A), was reduced by AG1478 (Fig. 10C). These locomotion fluctuations are typical of a cell phenotype and may be considered the cell signature of a particular cell phenotype (26, 27). Based on the evidence that AG1478 abrogates induction of integrin ₂ by EGF (data not shown), the reduction in micromotion may be attributed to the lower expression of integrin caused by tyrphostin AG1478. This indicates that EGFr is a transducing element in the control of cell locomotion. Cell populations with higher expression levels of integrins (in the presence of EGF) exhibit increased cell micromotility (Fig. 10B) as compared with control cells (Fig. 10A). These observations are consistent with evidence of a cause and effect relationship between integrin-mediated adhesion and motility on extracellular matrix that is tightly controlled by ligand density, integrin expression levels, and integrin affinity or avidity (34-36). Optimal levels of adhesion propel migration through a process in which adhesion molecules at the leading edge of the cell form complexes with matrix while molecules at the trailing edge release the substrate, allowing cell movement (35). Thus, higher integrin concentrations may lead to higher rates for this process at cellular interfaces involved in movement, especially in response to EGF receptor activation and associated changes in cytoskeletal arrangements. The role of autocrine TGF- in basal steady-state cell micromotion is demonstrated by using TGF- antisense-transfected cells (Fig. 11). The stimulation of HCT116 cell micromotion, cell adhesion, and higher expression of integrin ₂ via the EGFr signaling pathway may be one mechanism by which these cells become metastatic. The enhancement of integrin expression by EGFr activation in HCT116 cells may also contribute toward cell survival.

	ACKNOWLEDGEMENT

We thank Javier Giron for typing the manuscript.

	FOOTNOTES

^* This work was supported in part by National Institutes of Health Grants CA 54807, 34432, 50457, and HL07446, by a Merit Review from the Veterans Administration, and by the Shelby Rae Tengg Foundation.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Supported by the Veterans Administration grant.

^** To whom correspondence should be addressed: Dept. of Pharmacology & Therapeutics, Roswell Park Cancer Institute, Buffalo, NY 14263. Tel.: 716-845-3044; Fax: 716-845-8857; E-mail: michael.brattain@roswellpark.org.

Published, JBC Papers in Press, October 29, 2001, DOI 10.1074/jbc.M103268200

	ABBREVIATIONS

The abbreviations used are: TGF-, transforming growth factor-; EGF, epidermal growth factor; EGFr, epidermal growth factor receptor; CN IV, collagen type IV; FN, fibronectin; LN, laminin; ECM, extracellular matrix protein; ECIS, electrical cell impedance sensor; BSA, bovine serum albumin; T, transferrin; I, insulin; E, EGF; TTBS, Triton Tris-buffered saline; ECL, enhanced chemiluminescence; MTT, methylthiazole tetrazolium; PBS, phosphate-buffered saline; ERK, extracellular signal-regulated kinase; mAb, monoclonal antibody.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				R. S. Sawhney, M. M. Cookson, Y. Omar, J. Hauser, and M. G. Brattain Integrin {alpha}2-mediated ERK and Calpain Activation Play a Critical Role in Cell Adhesion and Motility via Focal Adhesion Kinase Signaling: IDENTIFICATION OF A NOVEL SIGNALING PATHWAY J. Biol. Chem., March 31, 2006; 281(13): 8497 - 8510. [Abstract] [Full Text] [PDF]

				R. S. Sawhney, M. M. Cookson, B. Sharma, J. Hauser, and M. G. Brattain Autocrine Transforming Growth Factor {alpha} Regulates Cell Adhesion by Multiple Signaling via Specific Phosphorylation Sites of p70S6 Kinase in Colon Cancer Cells J. Biol. Chem., November 5, 2004; 279(45): 47379 - 47390. [Abstract] [Full Text] [PDF]

				J. G. Jackson, P. St. Clair, M. X. Sliwkowski, and M. G. Brattain Blockade of Epidermal Growth Factor- or Heregulin-Dependent ErbB2 Activation with the Anti-ErbB2 Monoclonal Antibody 2C4 Has Divergent Downstream Signaling and Growth Effects Cancer Res., April 1, 2004; 64(7): 2601 - 2609. [Abstract] [Full Text] [PDF]

				L. A. Scott, J. K. Vass, E. K. Parkinson, D. A. F. Gillespie, J. N. Winnie, and B. W. Ozanne Invasion of Normal Human Fibroblasts Induced by v-Fos Is Independent of Proliferation, Immortalization, and the Tumor Suppressors p16INK4a and p53 Mol. Cell. Biol., February 15, 2004; 24(4): 1540 - 1559. [Abstract] [Full Text] [PDF]

				V. Petegnief, B. Friguls, C. Sanfeliu, C. Sunol, and A. M. Planas Transforming Growth Factor-{alpha} Attenuates N-Methyl-D-aspartic Acid Toxicity in Cortical Cultures by Preventing Protein Synthesis Inhibition through an Erk1/2-dependent Mechanism J. Biol. Chem., August 8, 2003; 278(32): 29552 - 29559. [Abstract] [Full Text] [PDF]

				B. K. Ko, K. Kawano, J. L. Murray, M. L. Disis, C. L. Efferson, H. M. Kuerer, G. E. Peoples, and C. G. Ioannides Clinical Studies of Vaccines Targeting Breast Cancer Clin. Cancer Res., August 1, 2003; 9(9): 3222 - 3234. [Abstract] [Full Text] [PDF]

				R. S. Sawhney, B. Sharma, L. E. Humphrey, and M. G. Brattain Integrin {alpha}2 and Extracellular Signal-regulated Kinase Are Functionally Linked in Highly Malignant Autocrine Transforming Growth Factor-{alpha}-driven Colon Cancer Cells J. Biol. Chem., May 23, 2003; 278(22): 19861 - 19869. [Abstract] [Full Text] [PDF]

				H. Xu, M. Inouye, E. R. Hines, J. F. Collins, and F. K. Ghishan Transcriptional regulation of the human NaPi-IIb cotransporter by EGF in Caco-2 cells involves c-myb Am J Physiol Cell Physiol, May 1, 2003; 284(5): C1262 - C1271. [Abstract] [Full Text] [PDF]

				H. Kanno, Y. Horikawa, R. R. Hodges, D. Zoukhri, M. A. Shatos, J. D. Rios, and D. A. Dartt Cholinergic agonists transactivate EGFR and stimulate MAPK to induce goblet cell secretion Am J Physiol Cell Physiol, April 1, 2003; 284(4): C988 - C998. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/1/75    most recent M103268200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sawhney, R. S. Articles by Brattain, M. G. Search for Related Content PubMed PubMed Citation Articles by Sawhney, R. S. Articles by Brattain, M. G. Social Bookmarking                      What's this?