Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Asthma, a chronic disease characterized by airway hyperreactivity, occurs in 5 to 8% of the U.S. population and remains an extraordinarily common cause of pulmonary impairment. Despite considerable research efforts, asthma mortality rates continue to rise and the primary defects that underlie airway hyperreactivity remain unknown, although an intrinsic abnormality of human airway smooth muscle (HASM) has been postulated (1). Increased smooth-muscle mass in airways of patients with chronic severe asthma has been a well documented pathologic finding. Studies have reported that increased HASM mass is due to an increased number of airway myocytes (2, 3).

We postulate that frequent stimulation of HASM by contractile agonists, inflammatory mediators, and growth factors induces chronic adaptive alterations in the airways that result in myocyte proliferation. Such alterations may have important consequences in determining airway caliber and airway smooth-muscle contractility. Although some information is now available about the factors that promote HASM cell proliferation (reviewed in 4), only a few studies have addressed the cellular and molecular mechanisms that regulate HASM cell proliferation (5).

In a number of cell types, activation of Ras proteins has been shown to be essential for agonists to stimulate DNA synthesis (8). The ubiquitous mammalian ras family of proto-oncogenes H-, K-, and N-ras, encode highly conserved 21-kD Ras proteins (H-, K-, and N-p21ras: 188-189 residues) that function as molecular switches (9), cycling between an inactive guanine diphosphate (GDP)-bound state and an active guanine triphosphate (GTP)-bound state. Because the intrinsic GDP/GTP exchange and guanine triphosphatase (GTPase) activity of p21ras are very low, the rate of cycling is regulated by guanine nucleotide exchange factor/s (GEFs) and GTPase-activating factor/s (GAPs) (reviewed in 11). In unstimulated cells, Ras proteins are in the GDP-bound inactive form. Upon agonist stimulation, GEFs act to markedly increase the rate of nucleotide release. GTP, which is more abundant in the cell, then binds to p21ras. Conversely, GAPs increase the rate of GTP hydrolysis to promote formation of the inactive GDP-bound state. In their GTP-bound active state, Ras proteins induce their effects by interacting with downstream effectors. Raf-1, a serine/threonine kinase, is an important Ras effector. The recruitment of Raf-1 to the plasma membrane by GTP-bound Ras and the initiation of the mitogen-activated protein kinase (MAPK) cascade is a relatively well characterized pathway (reviewed in 12) crucial in regulating expression of nuclear transcription factors (12). Recent evidence, however, has shown that p21ras can also utilize a multitude of functionally diverse effector targets that include phosphatidylinositol (PI) 3-kinase (13) and protein kinase  (14). Because these effectors can transduce signals through multiple pathways, both MAPK-dependent and independent (15), Ras signaling has become increasingly complex. Ras proteins have now been shown to be involved in a number of diverse biologic processes including cell proliferation, differentiation, and apoptosis (reviewed in 15).

Ras proteins have been shown to be transiently activated in response to a diverse array of extracellular signals such as growth factors, cytokines, and hormones (reviewed in 15). Because these agonists can act via either receptor tyrosine kinases (RTKs), nonreceptor tyrosine kinases, or G protein-coupled receptors (GPCRs), investigators have postulated that p21ras acts as a point of convergence for these diverse extracellular signal-stimulated pathways (15, 16).

In this study, we test the hypothesis that p21ras is a critical signaling molecule that integrates mitogenic signals from RTK- and GPCR-dependent pathways in a nontransformed cell line. The aims of this study were to: (1) define effects of mitogens that act via either RTK-dependent (epidermal growth factor [EGF]) or GPCR-dependent mechanisms (thrombin) on p21ras expression and activity of p21ras in HASM cells; (2) examine the effect of bradykinin, an agonist that stimulates phosphoinositide turnover and evokes cytosolic calcium increases via GPCR-dependent mechanisms in HASM cells, but not proliferation (17); (3) characterize and compare the p21ras isoform(s) expression of HASM cells with those expressed in ATCC cell lines T24 (bladder carcinoma), SW480 (colorectal adenocarcinoma), and SK-N-SH (neuroblastoma), which have H- (18), K- (19), or N-ras (20) oncogenes as positive controls; and (4) then use single-cell microinjection of Y13-259 (a neutralizing p21ras antibody) to determine whether p21ras is necessary for EGF- and thrombin-induced DNA synthesis in HASM cells, and examine whether RTK- and GPCR-mediated pathways converge to activate p21ras, which in turn regulates mitogen-induced HASM proliferation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

HASM Cell Culture

Human trachea was obtained from lung transplant donors in accordance with procedures approved by the University of Pennsylvania Committee on Studies Involving Human Beings at the University of Pennsylvania. HASM cells were dissected and purified as previously described (21) and cultured in Ham's F12 media supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 0.1 mg/ml streptomycin (GIBCO BRL, Grand Island, NY). HASM cells in subculture during the second through fifth cell passages were studied because in these cell passages the myocytes retain native contractile protein expression, as demonstrated by indirect immunofluorescent staining for smooth muscle-specific actin (17, 21). Further, these cells also retain functional cell-excitation coupling systems as determined by fura-2 measurements of agonist-induced changes in cytosolic calcium (21, 22). A minimum of three different cell lines was used for each experiment.

Unless otherwise specified, all chemicals used in this study were purchased from Sigma Chemical Company (St. Louis, MO).

Other Cell Lines and Cell Culture Techniques

T24 (bladder, carcinoma, human), SW480 (colon, adenocarcinoma, human), and SK-N-SH (neuroblastoma, human) cell lines were purchased from ATCC (Rockville, MD). These cell lines were chosen to act as positive controls for the characterization of HASM p21ras isoforms because these cells have been reported to contain H- (18), K- (19), or N-ras (20) oncogenes, respectively. T24 cells were cultured in McCoy's 5a medium supplemented with 10% FBS. SW480 cells were cultured in Leibovitz's L15 medium with 2 mM L-glutamine and supplemented with 10% FBS. SK-N-SH cells were cultured in Eagle's minimum essential medium with 2 mM L-glutamine and Earle's balanced salt solution (containing 1.5 g/liter sodium bicarbonate, 0.1 mM nonessential amino acids, and 1.0 mM sodium pyruvate) supplemented with 10% FBS.

Ras Expression

Growth-arrested, confluent HASM cells in 100-mm plates were treated for 1 h with 10 ng/ml EGF, 1 U/ml thrombin, or 5 µM bradykinin at 37°C. Agonist concentrations represent a mean effective dose concentration for HASM cell growth based on our previous studies (5, 6, 17, 23). The cells were then rapidly washed in ice-cold Ca²⁺/Mg²⁺-free phosphate-buffered saline (PBS) and lysed by the addition of 500 µl of lysis buffer (50 mM Tris, 20 mM MgCl₂, 150 mM NaCl, 0.5% IGEPAL [tert-Octylphenoxy poly(oxyethylene)] ethanol), 10 µg/ml aprotonin, and 0.5 mM phenylmethylsulfonyl fluoride (pH 7.5), containing 10 µg/ml of rat anti-p21ras monoclonal antibody Y13-259 (Oncogene Research Products, Cambridge, MA). The addition of Y13-259 directly to the lysis buffer inhibits artifactual p21ras-GTP hydrolysis by > 99% (24). In some experiments, as an isotype-matched control, HASM cells were treated in an identical manner with 10 µg/ml rat immunoglobulin (Ig)G in lysis buffer. Cells were lysed for 1.5 h at 4°C and clarified by centrifugation, and the protein concentration was measured (DC Protein Assay; Bio-Rad, Richmond, MA). The average amount of protein lysate obtained from 10⁶ cells was 192.4 ± 3.3 µg. After protein concentration equilibration, 5 mg of Protein G coupled to Sepharose 4B beads was added to the lysates and incubated for 2 h at 4°C. The beads were then collected by centrifugation and washed three times in lysis buffer and three times in PBS before resuspension in a sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer (50 mM Tris, 100 mM dithiothrietol [DTT], 2% SDS, 0.1% bromophenol blue, and 10% glycerol, pH 6.8) and heating at ~ 100°C for 5 min.

HASM cell samples (30 µg protein loaded) were applied to 15% SDS-PAGE homogeneous separating gels (with 5% stacking gel) with broad-range biotinylated protein molecular mass markers (6.5 to 165 kD; New England BioLabs, Beverly, MA). After electrophoresis at 200 V, proteins were transferred to nitrocellulose (0.45 µm; Schleicher & Schuell, Keene, NH) using transfer buffer (48 mM Tris, 39 mM glycine, 20% methanol, and 1.3 mM SDS, pH 9.2) at 400 mA. After transfer, the nitrocellulose was blocked (blocking buffer: 5% skim milk powder, 0.1% Tween 20, 20 mM Tris, and 137 mM NaCl, pH 7.6), then immunoblotted with a mouse antihuman pan p21ras primary antibody (1 µg/ml, IgG₁ monoclonal; Oncogene Research Products) and an antimouse IgG (H + L) secondary antibody conjugated to horseradish peroxidase (HRP) (1/3,000; Boehringer Mannheim, Indianapolis, IN). An antibiotin antibody conjugated to HRP (1/1,000; New England BioLabs) was also added to the secondary antibody solution to detect the biotinylated protein molecular mass markers. Both primary and secondary antibodies were prepared in blocking buffer. Detection of the p21ras bands was performed using an ECL immunoblot detection system (Amersham International, Buckinghamshire, UK). The expression of p21ras was quantitated by scanning densitometry and an NIH Image Analysis program (Version 1.61).

Ras Isoforms

Confluent T24, SK-N-SH, and SW480 cells grown in 100-mm plates were lysed and p21ras immunoprecipitation and SDS-PAGE was performed as described above. In parallel experiments, oncogenic and HASM cells grown on 100-mm plates were trypsinized (T24 and SK-N-SH, 0.25% trypsin in Hanks' balanced salt solution [HBSS]; SW480, 0.25% trypsin, 0.03% ethylenediaminetetraacetic acid [EDTA] in HBSS; HASM, 0.05% trypsin, 0.53 mM EDTA in HBSS [all from GIBCO BRL]) for 5 min at 37°C and the cells were counted (Coulter Counter; Coulter Electronics, Hialeah, FL).

Oncogenic and HASM cell samples were applied to 15% SDS-PAGE with molecular mass markers and a range of concentrations of H-, K-, and N-p21ras^Gly-12 immunoblot standards (Oncogene Research Products). After electrophoresis and transfer, the nitrocellulose was blocked and then incubated with a 1/100 dilution of the relevant primary antibody for each p21ras isoform; either mouse antihuman c-H-p21ras (IgG_2a monoclonal), mouse antihuman c-K-p21ras (IgG_2a monoclonal), or mouse antihuman c-N-p21ras (IgG₁ monoclonal) was used. After densitometric analysis of the H-, K-, and N-p21ras^Gly-12 bands, standard curves reporting the loading concentration of proteins were determined and linear regression equations were established. These equations were then used to semiquantify the p21ras isoforms produced by each cell type; the results are shown as H-, K-, or N-p21ras expression (ng/10⁶ cells).

Ras Activity

Growth-arrested, confluent HASM cells in 100-mm plates were incubated in 5 ml of phosphate-free Dulbecco's modified Eagle's medium (DMEM) with 4.5 g/liter glucose, 0.584 g/liter L-glutamine (pH 7.4), and 0.5 mCi carrier-free [³²P]orthophosphate (specific activity, 800 Ci/mmol; NEN Life Science Products, Boston, MA) at 37°C. After 16 h, the cells were treated at 37°C for the indicated times with either 10 ng/ml EGF, 1 U/ml thrombin, 5 µM bradykinin, or diluent (control). Cells were then rapidly washed in ice-cold PBS, lysed for 1.5 h at 4°C in 500 µl of lysis buffer containing 10 µg/ml Y13-259 (24), then clarified by centrifugation and incubated for 2 h with PBS containing 5 mg of Protein G coupled to Sepharose 4B beads. The beads were collected by centrifugation and washed three times in lysis buffer and three times in PBS before incubating for 20 min at 68°C in 2 mM EDTA, 2 mM DTT, 0.2% SDS, 0.5 mM GTP, and 0.5 mM GDP (pH 7.5) to elute the GTP- and GDP-bound p21ras. Samples were applied to polyethyleneimine thin-layer chromatography (TLC) plates in parallel with 100 nmol of GTP and GDP as unlabeled markers. After elution in 1 M KH₂PO₄ (pH 3.4), the plates were dried and exposed to X-Omat Blue XB-1 X-ray film (Kodak, Rochester, NY) for 5 d at  70°C. After autoradiography, the RF (electrophoretic mobility compared with the solvent front) of the unlabeled markers GTP and GDP were visualized by ultraviolet light, the areas of the TLC corresponding to [³²P]GTP and [³²P]GDP on the autoradiograms were removed, and the radioactivity was quantitated (as cpm) by liquid scintillation counting. Results were expressed as %GTP, i.e., the percentage of [³²P]GTP cpm/([³²P]GTP cpm + [³²P]GDP cpm).

Microinjection and Measurement of DNA Synthesis

Near-confluent HASM cells grown on two-well plastic chamber slides (Nunc, Naperville, IL) were growth-arrested by incubating the monolayers in Ham's F12 with 0.1% bovine serum albumin (BSA) for 48 h (21). Microinjection pipettes (Femtotip I; Eppendorf, Hamburg, Germany) were filled with either antibody dilution vehicle (10 mM NaH₂PO₄ and 70 mM KCl, pH 7.2 [25]) with 5 mg/ml rabbit IgG; 0.5 mg/ml Y13-259 with 5 mg/ml rabbit IgG; or rat IgG₁ (0.5 mg/ml; Zymed Laboratories, San Francisco, CA), as an isotype-matched control for Y13-259, with 5 mg/ml rabbit IgG. Rabbit IgG served as a marker for microinjected cells. In other experiments, the effect of microinjecting higher concentrations of Y13-259 (1.0 and 2.5 mg/ml) was also examined.

Microinjection was performed under ×400 magnification on an inverted microscope (Diaphot; Nikon Corporation, Tokyo, Japan) with Hoffman Modulation Optics (Modulation Optics Inc., Greenvale, NY) using an Eppendorf Model 5171 Micromanipulator. Solutions were injected using the Transjector 5246 (Eppendorf) to deliver pressure to the microinjection pipette (5 psi, 100 msec), injecting an estimated injection volume of 50 fl (26). Approximately 2 h after microinjection, HASM cells were treated with 10 ng/ml EGF, 1 U/ml thrombin, 10% FBS, or diluent (control). At 15 h later, 10 µM of the thymidine analogue 5-bromo-2'-deoxyuridine (BrdU) was added to all wells.

Twenty-four hours after the addition of BrdU to the two-well chamber slides, the cell monolayers were fixed with 70% ethanol (30 min at 20°C), air-dried, then incubated for 1 h at 37°C with 2 µg/ml murine anti-BrdU antibody (Becton Dickinson, San Jose, CA) in 50 mM Tris-HCl (pH 8.0), 5 mM MgCl₂, 10 mM 2-mercaptoethanol, and 0.1 mg/ml BSA. The DNA was denatured by the addition of 200 U/ml exonuclease III (Promega, Madison, WI) to the buffer that contained the anti-BrdU antibody (27). The slides were then treated with 10 µg/ml Texas Red conjugated antimouse antibody for 1 h at 37°C (Jackson ImmunoResearch Laboratories, West Grove, PA) to detect the BrdU-positive cells. To identify microinjected cells, rabbit IgG was detected using tyramide signal amplification (TSA-Direct; NEN Life Science Products) with a fluorescein tyramide, according to the manufacturer's instruction. Nuclei were stained by 2 min incubation in 0.4 µg/ml 4',6-diamidino-2-phenylindole (dihydrochloride; Molecular Probes, Eugene, OR) in 0.9% NaCl at room temperature.

The slides were examined using a fluorescent microscope (Aristoplan; Leica, Wetzlar, Germany) under ×200 magnification with the appropriate fluorescent filters. Results were presented as the mitotic index (i.e., the percentage of BrdU-positive nuclei/number of cells microinjected). Previously we have confirmed that microinjection is a valid technique for the examination of the signaling pathways involved in HASM cell proliferation because microinjection did not alter cell viability or the capability of cells to enter the S phase of the cell cycle (A. J. Ammit, Y. Amrani, S. A. Kane, and R. A. Panettieri, Jr., unpublished data).

Statistical Analysis

One-way or two-way analysis of variance (ANOVA) was used on all data when experiments were of a factorial design, then Fisher's PLSD multiple comparison test was used to compare differences between treatment means. Correlations between two variables were performed using linear regression analysis. For all analyses, effects were considered statistically significant if the probability (P) of the effect being due to chance alone was less than 5%.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

p21ras Isoform Expression in Unstimulated HASM Cells

Unstimulated growth-arrested, near-confluent HASM cells were lysed and immunoprecipitated in the presence of a neutralizing p21ras antibody (Y13-259) or a rat IgG isotype-matched control, then analyzed by SDS-PAGE and immunoblot analysis using a pan p21ras antibody that recognizes H-, K-, and N-p21ras. As shown in Figure 1A, unstimulated HASM cells express p21ras (lane 3). The molecular mass of the resultant band (at ~ 21 kD) was confirmed by comparison with the molecular mass markers shown in lane 1. These results suggest that this method of immunoprecipitation is specific for p21ras because nonspecific binding with the rat IgG isotype-matched control (lane 2) was not evident. In parallel experiments we characterized the p21ras isoforms in HASM cells, using cell lines known to express specific p21ras isoforms as positive controls. Confluent T24, SW480, and SK-N-SH cells underwent lysis and immunoprecipitation with a neutralizing p21ras antibody (Y13-259). These lysates, in comparison with immunoprecipitates from agonist-treated HASM cells, underwent immunoblot analysis using anti-p21ras antibodies specific for either H-, K-, or N-p21ras isoforms. As shown in Figure 1B, HASM cells express K- and N-p21ras, but not H-p21ras. Of the p21ras present in HASM cells, 75.7 ± 11.1% is K-p21ras and 24.3 ± 1.1% is N-p21ras. Although T24 cells contain H-p21ras (1.05 ± 0.09 ng H-p21ras/10⁶ cells) as previously reported (18), T24 cells also have substantial amounts of K- and N-p21ras (Figure 1B). SW480, known to express K-ras oncogene (19), in our hands also expressed high levels of K-p21ras, as shown in Figure 1B (7.89 ± 0.48 ng K-p21ras/10⁶ cells). In addition, SW480 also expresses N-p21ras (1.65 ± 0.14 ng N-p21ras/10⁶ cells). The SK-N-SH cell line, known to expresses the N-ras oncogene (20), also expresses substantial amounts of K-p21ras, as shown in Figure 1B.

View larger version (32K): [in this window] [in a new window]   Figure 1.   HASM cells express K- and N-p21ras, but not H-p21ras. (A) Unstimulated, growth-arrested HASM cells were lysed and immunoprecipitated in the presence of the neutralizing p21ras antibody Y13-259 or rat IgG isotype control, then analyzed by 15% SDS-PAGE (30 µg protein loaded) and immunoblotting using a pan p21ras antibody that recognizes H-, K-, and N-p21ras. Lane 1, protein molecular mass markers (relative kD are indicated); lane 2, immunoprecipitation (IP) with rat IgG isotype control; lane 3, IP with neutralizing p21ras antibody Y13-259. (B) Confluent T24, SW480, and SK-N-SH cells underwent lysis and IP with the neutralizing p21ras antibody Y13-259, then were compared with immunoprecipitates from agonist-stimulated HASM cells. Lysates were analyzed by 15% SDS-PAGE and immunoblotting using antibodies specific to H- (open bars), K- (filled bars), and N-p21ras (striped bars). Data represent means ± standard error (SE) from three to eight replicates.

p21ras Expression and Isoform Pattern in HASM Cells after Treatment with Agonists

To determine whether p21ras expression is modulated by agonists, HASM cells were treated for 1 h with either 10 ng/ml EGF, 1 U/ml thrombin, or 5 µM bradykinin, then p21ras immunoblot analysis was performed. There was no significant difference in the p21ras expression in HASM cells treated with these agonists as compared with those treated with diluent, as shown in Figure 2. In addition, we examined the p21ras isoforms in HASM cells after 1 h stimulation with EGF, thrombin, or bradykinin as compared with control, using antibodies specific for either H-, K-, or N-p21ras. There was no significant difference among the levels of p21ras isoforms after stimulation with any of the agonists (Figure 3).

View larger version (15K): [in this window] [in a new window]   Figure 2.   HASM p21ras expression does not change with agonist stimulation. Growth-arrested, confluent HASM cells were stimulated for 1 h with 10 ng/ml EGF, 1 U/ml thrombin, or 5 µM bradykinin. After lysis and IP using the neutralizing p21ras antibody Y13-259, lysates were analyzed by 15% SDS-PAGE and immunoblotting using a pan p21ras antibody that recognizes H-, K-, and N-p21ras. Data are expressed as percentage change over control (means ± SE from three to 12 replicates). Statistical analysis using one-way ANOVA, then Fisher's PLSD multiple comparison test, found no significant differences from control.

View larger version (22K): [in this window] [in a new window]   Figure 3.   Agonist stimulation does not modify HASM p21ras isoforms after 1 h. Growth-arrested, confluent HASM cells were stimulated for 1 h with 10 ng/ml EGF, 1 U/ml thrombin, or 5 µM bradykinin. After lysis and IP using the neutralizing p21ras antibody Y13-259, lysates were analyzed by 15% SDS-PAGE and immunoblotting using antibodies specific to K- (filled bars) and N-p21ras (striped bars). Statistical analysis using two-way ANOVA, then Fisher's PLSD multiple comparison, found no significant differences between the amounts of K- and N-p21ras expressed in HASM cells after treatment with either agonist, as compared with control. Data represent means ± SE from four to eight replicates.

Agonist-Induced Activation of p21ras

Although the p21ras expression and isoform pattern of HASM cells treated with either EGF, thrombin, or bradykinin were similar, the time courses of p21ras activation after stimulation with these agonists were markedly different. As shown in Figure 4A, EGF rapidly activated p21ras within 30 s and was sustained for up to 30 min (P < 0.05, compared with control at the same time points). Although thrombin also induced a rapid and significant rise (P < 0.05) in p21ras activity after 2.5 min, after 5 min the activity of p21ras was not significantly different from control. In contrast, bradykinin did not activate p21ras. To show specificity of the p21ras activity assay, thrombin-induced p21ras activation was measured after 15 min preincubation with hirudin, a specific thrombin inhibitor. As shown in Figure 5, hirudin preincubation significantly (P = 0.0251) inhibited thrombin-induced p21ras activation.

View larger version (31K): [in this window] [in a new window]   Figure 4.   EGF and thrombin, but not bradykinin, activate p21ras. (A) Growth-arrested, confluent HASM cells were incubated for 14 h with [32P]orthophosphate in phosphate-free DMEM, then stimulated with 10 ng/ml EGF (filled triangles), 1 U/ml thrombin ( filled squares), or 5 µM bradykinin (open circles) for the indicated times. After lysis and IP using the neutralizing p21ras antibody Y13-259, lysates were electrophoresed on polyethyleneimine TLC before autoradiography and quantitation by liquid scintillation counting (* denotes a significant difference [P < 0.05] from the corresponding control time point, using two-way ANOVA and then Fisher's PLSD multiple comparison test). Data represent means ± SE from three to seven replicates. (B) Representative autoradiogram of TLC separation of immunoprecipitated [32P]GDP and [32P]GTP bound to p21ras after 2.5 min stimulation with EGF, thrombin, or bradykinin, as compared with control.

View larger version (21K): [in this window] [in a new window]   Figure 5.   Thrombin activation of p21ras is specific. Growth- arrested, confluent HASM cells were incubated 14 h with [32P]orthophosphate in phosphate-free DMEM, then either left unstimulated (control) (n = 4; open bar); stimulated for 2.5 min with 1 U/ml thrombin (n = 5; filled bar); or stimulated for 2.5 min with 1 U/ml thrombin after 15 min preincubation with 1 U/ml hirudin (n = 3; striped bar). After lysis and IP using the neutralizing p21ras antibody Y13-259, lysates were electrophoresed on polyethyleneimine cellulose TLC before autoradiography and quantitation by liquid scintillation counting. Statistical analysis was performed using one-way ANOVA and then Fisher's PLSD multiple comparison test (* denotes that hirudin preincubation significantly [P = 0.0251] inhibited the thrombin-induced increase in p21ras activity). Data represent means ± SE.

The oncogenic cell lines have single-point mutations in their ras genes (T24 and SW480, G T in codon 12; SK-N-SH, C A in codon 61) that cause p21ras to remain in the active GTP-bound form. Although the p21ras activity of HASM cells treated with EGF is significantly less (P < 0.05) than the intrinsic activity of the oncogenic cell lines, as shown in Figure 6, EGF-treated HASM cells do attain levels of p21ras activation (expressed as percent GTP) comparable to that of unstimulated oncogenic cell lines.

View larger version (39K): [in this window] [in a new window]   Figure 6.   Stimulated HASM have comparable p21ras activity to oncogenic cell lines. Confluent T24, SW480, and SK-N-SH cells were incubated for 14 h with [32P]orthophosphate in phosphate-free DMEM. After lysis and IP using the neutralizing p21ras antibody Y13-259, lysates were electrophoresed on polyethyleneimine cellulose TLC before autoradiography and quantitation by liquid scintillation counting. The activation of p21ras (expressed as %GTP) was compared with the %GTP obtained when HASM cells were either treated with diluent (control) or treated for 2.5 min with 10 ng/ml EGF. Statistical analysis was performed using one-way ANOVA and then Fisher's PLSD multiple comparison test (* denotes that the %GTP obtained in HASM cells stimulated by EGF was significantly [P < 0.05] less than that of each of the oncogenic cell lines). Data represent means ± SE from three to seven replicates.

p21ras Activation Mediates Mitogen-Induced DNA Synthesis in HASM Cells

Using single-cell microinjection, the role of p21ras activation in modulating HASM cell growth was examined by measuring DNA synthesis using indirect anti-BrdU immunofluorescence. HASM cells were microinjected with either antibody dilution vehicle, 0.5 mg/ml Y13-259 (a neutralizing p21ras antibody) or 0.5 mg/ml rat IgG₁ (an isotype-matched control for Y13-259); treated with 10 ng/ml EGF, 1 U/ml thrombin, or 10% FBS for 15 h; and compared with those treated with a diluent control. Cells were then incubated with BrdU and mitotic indices determined 24 h later. Results for each individual experiment are shown in Table 1 and graphically represented as mitotic indices in Figure 7.

View larger version (14K): [in this window] [in a new window]   Figure 7.   Microinjection of HASM cells with Y13-259 (a neutralizing p21ras antibody) significantly inhibits EGF-, thrombin-, and FBS-induced cell proliferation. Near-confluent, growth-arrested HASM cells were microinjected with 0.5 mg/ml Y13-259 (a neutralizing p21ras antibody; filled bars); or 0.5 mg/ml rat IgG1 (an isotype-matched control for Y13-259; open bars). Cells were then treated with either 10 ng/ml EGF, 1 U/ml thrombin, or 10% FBS for 15 h, or diluent alone (control). After BrdU treatment for a further 24 h, DNA synthesis was measured using indirect anti-BrdU immunofluorescence. Results are expressed as mitotic indices. Two-way ANOVA with Fisher's PLSD multiple comparison test was used to compare differences between treatment means (* denotes that the mitotic indices of cells microinjected with Y13-259 were significantly [P < 0.0001] inhibited as compared with those microinjected with rat IgG1). Data represent means ± SE (n = 3).

In cells microinjected with 0.5 mg/ml Y13-259 and then treated with EGF, DNA synthesis was significantly inhibited by 70.2 ± 5.9% (P < 0.0001), as compared with those obtained after microinjection with isotype-matched control. Y13-259 microinjection also significantly inhibited thrombin-induced DNA synthesis, by 73.9 ± 3.8% (P < 0.0001). Although the effect on serum-induced DNA synthesis was less, this was also significantly inhibited by 47.5 ± 7.7% (P < 0.0001) as compared with rat IgG₁-injected cells. There was no significant difference in the mitotic indices obtained in cells microinjected with rat IgG₁ isotype-matched control after stimulation with EGF, thrombin, or 10% FBS, as compared with those microinjected with diluent alone (Table 1).

In additional experiments, the effects of microinjecting higher concentrations of Y13-259 were examined. In these experiments (data not shown), microinjection of 1.0 or 2.5 mg/ml Y13-259 did not inhibit the DNA synthesis induced by EGF, by thrombin, and by 10% FBS, more than that obtained after microinjection of 0.5 mg/ml Y13-259 (Table 1 and Figure 7).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study we examined the expression of p21ras isoforms in HASM cells and the role of Ras proteins in the integration of mitogenic signals from the RTK- and GPCR-dependent pathways. The dominant p21ras isoform expressed in HASM cells appears to be K-p21ras; N-p21ras was also present, however H-p21ras was not detected. The relative amounts of these isoforms did not change after stimulation with EGF or thrombin, or after incubation with the nonmitogenic agonist bradykinin. Interestingly, these agonists activated p21ras in markedly different manners. Only mitogens, EGF, and thrombin induced a rapid activation of p21ras. Microinjection of a neutralizing p21ras antibody (Y13-259) significantly inhibited DNA synthesis in response to both EGF and thrombin. Together, these results suggest that K-p21ras and N-p21ras, but not H-p21ras, are critical signaling molecules that integrate mitogenic signals from RTK- and GPCR-dependent pathways to mediate smooth-muscle cell proliferation.

In this study we investigated the role of Ras proteins in smooth-muscle cell proliferation and, in particular, the role of Ras proteins in stimulating HASM cell growth. These studies are clinically relevant because hyperplasia of airway smooth muscle has been shown to be one of the key pathologic features of chronic severe asthma. To date, studies of HASM mitogenesis have focused mainly upon the identification of agonists capable of inducing myocyte proliferation (reviewed in 4). Recent studies in our laboratory have examined the pathways involved in transducing signals from receptors coupled to RTKs or to GPCRs via MAPK and PI 3-kinase (5, 23). This is one of the first studies to examine p21ras activation in airway smooth muscle, although in vascular smooth muscle, Ras proteins have been shown to play a role in both hyperplasia (28) and hypertrophy (29, 30).

Ras proteins have been shown to be transiently activated in response to a diverse array of extracellular signals such as growth factors, cytokines, and hormones (reviewed in 15). Because these agonists can act via either RTKs, nonreceptor tyrosine kinases, or GPCRs, p21ras has been postulated to act as a point of convergence for these diverse extracellular signal-stimulated pathways (15, 16). Shc, GRB2, and SOS seemingly provide the link among many types of activated cell-surface RTKs and p21ras (11, 15). The signaling pathway that links GPCRs to p21ras is mediated by members of the Src family of nonreceptor tyrosine kinases, which link the -subunits of pertussis toxin (PTX)-sensitive G proteins (31) and the -subunits of PTX-insensitive G proteins (32) to the p21ras-MAPK pathway via phosphorylation of Shc, recruitment of GRB2, and the GEF SOS.

In this study we observed that EGF induced a rapid and sustained induction of p21ras activity, which suggested that signaling via p21ras was a major pathway responsible for the transduction of EGF-induced mitogenesis in HASM cells. Thrombin, an agonist that stimulates a GPCR, also activated p21ras in a specific manner, inasmuch as hirudin, a thrombin antagonist (17), abrogated p21ras activation. In contrast, bradykinin did not activate p21ras. Thrombin and bradykinin have been shown to stimulate receptors linked to G_q and G_i2 in fibroblasts (33). Using 3T3 fibroblasts microinjected with neutralizing antibodies against specific G protein -subunits, LaMorte and colleagues (33) showed that although both thrombin and bradykinin required G_q activation to mobilize cytosolic calcium, thrombin but not bradykinin appeared to require G_i2 in addition to G_q to stimulate cell growth (33). Interestingly, bradykinin does not stimulate mitogenesis in HASM cells (13) and, in the present study, it has been shown that bradykinin does not activate p21ras (34). The finding that stimulation of G_q in vascular smooth-muscle cells induced hypertrophy in a Ras-independent manner (29) may provide insight as to why bradykinin, which induced increases in intracellular calcium by hydrolyzing phosphoinositides to a greater extent than thrombin (17), does not induce mitogenesis in HASM cells (13). Thrombin has also been shown to mediate DNA synthesis in astrocytoma cells via G₁₂ (35). Together, these results in conjunction with our previous studies (13) suggest that mitogenesis of HASM cells is dependent on p21ras activation and may be linked to G_i2 and G₁₂, not G_q.

Although many of the components involved in p21ras-mediated signal transduction have been elucidated, many critical questions remain unresolved. One such question is whether the Ras protein isoforms expressed in mammalian cells (H-, K-, and N-p21ras) serve different functions. The first N-terminal 85 residues of all Ras protein isoforms are identical; this region contains the two "switch" domains (Switch I, Asp³⁰-Asp³⁸; Switch II, Gly⁶⁰-Glu⁷⁶) that are critical for GTP binding and GTPase function (36). The N-terminal region also contains the "effector loop" (Tyr³²-Tyr⁴⁰) (36) responsible for interactions with GAPs and downstream effectors of p21ras. The next 80 amino acids are 95% conserved and would likely not regulate specific Ras protein isoform function. If Ras proteins are to have different functions, then the domains responsible will likely be in the hypervariable region (HVR) (between residues 166 and 185) where sequences differ significantly, with fewer than 15% conserved residues (37). From the HVR to the C-terminus, all Ras proteins terminate in a CAAX motif (C, Cys; A, aliphatic amino acid; X, methionine or serine) that directs post-translational processing and confers Ras protein membrane localization (38). In our study we found that HASM cells express mainly K-p21ras, and some N-p21ras, but no H-p21ras.

Through gene mapping studies, investigators showed that the ATCC cell lines (T24, SW480, and SK-N-SH) have single-point mutations in their ras genes (T24 and SW480, G T in codon 12 [18, 19]; SK-N-SH, C A in codon 61 [20]) that result in mutant p21ras proteins upon transcription (T24, H-p21ras^Val-12; SW480, K-p21ras^Val-12; SK-N-SH, N-p21ras^Lys-61). The single amino acid substitutions at 12 or 61 result in mutant p21ras proteins that are insensitive to GAP inactivation. Consequently, these oncogenic Ras mutant proteins are fixed in the active GTP-bound state, leading to constitutive, deregulated activation of Ras function. In the present study we identified several p21ras isoforms in cell types that were previously reported (18) to express one ras gene exclusively. Whether this difference represents the existence of more oncogenes than previously described or the contribution of normal ras proto-oncogenes remains to be examined further, although it is well accepted that most cells express multiple isoforms and that the level of protein expression may vary (reviewed in 39). Different tumors however, have been shown to have specific ras mutations. N-ras mutations have been detected in 25% of hematopoietic tumors, whereas K-ras oncogenes occur in 50% of colon cancers and 90% of pancreatic tumors (39). H-ras mutations, although the first discovered (18), are uncommon. Consequently, the biologic importance of individual p21ras isoform are likely cell- and tissue-specific.

The neutralizing p21ras antibody Y13-259 (40) has been widely used in studies of p21ras signaling. Y13-259 specifically blocks the serum-induced mitogenic responses of fibroblasts (8) and inhibits morphologic transformation induced by microinjection of mutated K-p21ras proteins (41). Y13-259 binds to an epitope common to all p21ras isoforms (Glu⁶³-Arg⁷³) (42), a region known as loop 4 (43). From the X-ray crystallography and structure activity studies (42, 43), loop 4 has been shown to exist in close proximity to loop 2 (the effector loop). Hence, binding of Y13-259 to p21ras prevents binding of p21ras to GAPs and downstream effectors (42). We now show that microinjection of Y13-259 significantly inhibited both thrombin- and EGF-induced DNA synthesis in HASM cells.

In this study we investigated the role of p21ras in mediating growth factor-induced smooth-muscle cell proliferation in a nontransformed cell line. Such an approach is valuable in characterizing signaling pathways in physiologically relevant cells that are, in part, associated with specific disease entities. We determined that K- and N-p21ras, but not H-p21ras, mediated thrombin- and EGF-induced smooth-muscle cell proliferation. Further studies are necessary to examine whether this critical signaling event may offer a unique site for therapeutic interventions in preventing smooth-muscle cell growth that is characteristic of such diseases as asthma and atherosclerosis.