Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Tryptases are trypsin-like neutral serine-class proteases which are selectively expressed in mast cells and basophils (for review, see Reference 1). In human mast cells, tryptases are present in extremely high concentrations, accounting for 25% of the total intracellular protein (1). -Tryptase is the major member of this family that is stored in mast-cell granules, and it also is the predominant tryptase in lung tissue (1). Tryptase is released from mast cells by both immunoglobulin E-dependent and -independent mechanisms (1). Once released, it has several potential effects on neighboring cells. One such effect is the ability to stimulate growth (2). In isolated cells in culture, tryptase is a potent mitogen for human and rodent lung fibroblasts (2, 3), airway smooth-muscle (4) and epithelial cells (5), and vascular endothelial cells (6). Therefore, tryptase's properties as a growth factor may contribute to pathologic processes in the lung, such as fibroblast proliferation in interstitial lung diseases, airway smooth-muscle and glandular hyperplasia in asthma and chronic bronchitis (7), and abnormal new-vessel formation in pulmonary vascular disorders.

Little is known about the cellular mechanisms for tryptase-induced mitogenesis and about the early intracellular signals which mediate it. In its heparin-stabilized tetrameric form tryptase has potent catalytic activity (1), and some have assumed that its extracellular actions depend entirely upon its ability to cleave proteins. Of potential relevance are the findings that tryptase-induced mitogenesis in some cells (8) may be mediated via activation of protease-activated receptor (PAR)-2, one member of a new class of G protein-coupled PARs (9). It is not known whether PAR-2 activation accounts for all of tryptase's growth stimulatory effects. There is evidence that it may not. For example, PAR-2 activation has uniformly been associated with brisk increases in phosphoinositide hydrolysis and intracellular calcium concentrations (9). Tryptase did not induce either of these changes in lung fibroblasts in which it had potent mitogenic effects, even under conditions where increases in phosphoinositide hydrolysis and calcium concentrations were easily detectable in response to other mitogenic serine proteases (2). Further, it is increasingly apparent that some mitogenic serine proteases have the capacity to induce growth via nonproteolytic mechanisms (10), and PAR-2 activation by serine proteases is thought to proceed solely via proteolytic mechanisms (9). On the basis of these considerations, it is clear that more information is needed about the proteolytic versus nonproteolytic mechanisms through which tryptase stimulates growth in cells from airways and lung.

With respect to early intracellular signals, an important and unanswered issue is the potential role of mitogen-activated protein (MAP) kinases in mediating tryptase-induced mitogenesis. This family of proteins serves as second messengers for a variety of cellular effects, including proliferation, differentiation, and the responses to stresses such as irradiation, osmotic imbalances, or heat shock (11). MAP kinases include extracellular signal-regulated protein kinases (ERKs), p38 isoforms, and Jun amino-terminal kinase/stress-activated protein kinase proteins, all of which are serine/threonine kinases with extensive sequence homology. In most cells, growth and differentiation factors induce early activation of ERKs via sequential phosphorylation of the protein kinases Raf-1, MAP kinase kinase (MEK) 1 and 2, and ERK1 and 2 (also known as p44 and p42) (11). Whether tryptase activates MAP kinases to induce its growth-stimulatory effects in airway or lung cells is not known.

The major goal of our experiments was to determine whether tryptase activates ERK1 and 2 in cultured dog tracheal smooth-muscle cells, for which tryptase is a potent mitogen (4). Other goals were: (1) to assess the importance of tryptase's enzymatic activity in the mediation of its responses in our cells; (2) to establish the time course of tryptase's effects on ERK1 and 2, because the duration of ERK activation is known to be an important determinant of proliferative responses in airway smooth-muscle cells (12); and (3) to use PD098059, the cell-permeable MEK1 and 2 inhibitor (13), to determine whether activation of ERKs is vital for increased DNA synthesis in response to tryptase. Throughout these experiments, we compared the effects of tryptase with those of platelet-derived growth factor (PDGF), a member of the receptor tyrosine kinase class of growth factors that is a well established stimulator of both ERK activity and mitogenesis in airway smooth-muscle cells (14). Also, because we found significant differences in the time course of tryptase- versus PDGF-mediated activation of ERKs, we examined whether these differences are also apparent in induction of c-fos transcripts, inasmuch as c-fos is an immediate early response gene that is activated after translocation of activated ERK to the nucleus (15).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

Porcine heparin (4 to 6 kD), N-p-tosyl-Gly-Pro-Lys p-nitroanilide (GPK), p-amidino phenylmethanesulfonyl fluoride (p-APMSF), and Triton X-100 were purchased from Sigma Chemical (St. Louis, MO). Other reagents and their sources were PD098059 (CalBiochem, La Jolla, CA); recombinant human PDGF-BB homodimer, protein G-agarose, P-81 phosphocellulose paper, Taq DNA polymerase, and Maloney murine leukemia virus (MMLV) reverse transcriptase (RT) (Life Technologies, Gaithersburg, MD); and serum-free media (cellgro COMPLETE; Mediatech, Inc., Herndon, VA). Antibodies used in these experiments, and their sources, were: phosphospecific anti-ERK antibody kit (New England BioLabs, Beverly, MA), anti-ERK2 antibody for immunoprecipitation (C-14; Santa Cruz Biotechnologies, Santa Cruz, CA), anti-phosphotyrosine antibody 4G10 (Upstate Biotechnologies, Lake Placid, NY), and anti-ERK1 and 2 polyclonal antibody (Zymed Labs, South San Francisco, CA). Goat antimouse horseradish peroxidase (HRP) conjugate was purchased from Dako (Carpinteria, CA), goat antirabbit HRP conjugate from Bio-Rad (Hercules, CA), and reagents for enhanced chemiluminescence (ECL) from Pierce (Rockford, IL). RNAzol B was purchased from Biotecx Inc. (Houston, TX) and RNAsin from Promega (Madison, WI). Sequenase T7 DNA polymerase was from United States Biochemical (Cleveland, OH).

Cell Culture

Primary cultures of dog tracheal smooth-muscle cells were established and maintained as previously described (16). Cells were fed on alternate days and were passaged enzymatically when they approached confluence. We used these preparations of cells because they fulfill established criteria for identification of smooth-muscle cells in culture, including intense and uniform staining for  smooth-muscle actin (at antibody concentrations no more than 2 µg/ml) and a contoured "hill-and-valley" morphology at confluence. Also, these cells possess morphologic features that distinguish them from dog tracheal fibroblasts in culture (16).

Tryptase

Tryptase was isolated from postmortem human lung tissue, as described previously (3). Purity of the tryptase in these isolates has been established using chromatographic, electrophoretic, and immunologic criteria (3).

Catalytic activity of tryptase was measured by assessing its ability to cleave GPK (4). Tryptase, in 1- to 5-µL aliquots, was added to 50 mM Tris buffer, pH 7.7, containing 120 mM NaCl, 20 µg/ml heparin, and 100 µM GPK (final volume: 100 µL). Rates of GPK hydrolysis were determined by measuring the change in absorbance at 405 nm and calculating the amount of cleaved substrate, assuming a molar extinction coefficient of 8,800.

To inhibit catalytic activity, we preincubated tryptase with p-APMSF (17). This compound is a specific, irreversible inhibitor of the class of serine proteases which cleaves substrates after lysine or arginine (17), and tryptase falls into this class (1). In our experiments, tryptase (final concentrations: 10.5 to 27 nM) was incubated with p-APMSF (final concentration: 10 µM) for 100 to 120 min at 4°C in 10 mM BisTris buffer, pH 6.1. As a control, tryptase was also incubated with p-APMSF diluent alone (1:1 acetonitrile/dimethylformamide). Incubates were then diluted 1:5 in serum-free cellgro COMPLETE media, pH 7.4, and incubated on ice for 3 to 5 h before application to cells in 96-well plates. To test for its possible cytotoxic effects, p-APMSF that had been incubated alone was added to wells containing serum-free media, PDGF-BB, and 10% fetal calf serum (FCS), and the effects on DNA synthesis (see the following section) were determined. Before application on cells, p-APMSF-treated tryptase was assayed for completeness of catalytic inhibition using the GPK assay, as described earlier.

DNA Synthesis

To quantify DNA synthesis in the cells, we measured the incorporation of bromodeoxyuridine (BrdU) into cellular DNA using an enzyme-linked immunosorbent assay (ELISA), as we described previously (4). Passages 1-5 were seeded at densities of 10,000 cells/well in 96-well formats. After 24 h in M199 media containing 10% FCS, cells were washed once with phosphate-buffered saline (PBS) and then starved for 24 h in cellgro COMPLETE serum-free media. Mitogens were introduced into the cultures and the BrdU (10 µM) was added 24 to 48 h later. PD098059 is a cell-permeant inhibitor of MEK1 and 2, kinases that activate ERK1 and 2 (13). Therefore, to test the effects of inhibiting this pathway on tryptase-induced increases in DNA synthesis, we added PD098059 40 to 60 min before mitogens in some experiments.

Western Analysis of Phosphorylated ERK1 and 2

After exposing cells to mitogens, we tested for increases in phosphorylation of ERK1 and 2 in lysates prepared from the cells. Different antibodies were used to detect either protein tyrosine phosphorylation in general in the lysates or to detect specific, doubly phosphorylated (on threonine 202 and tyrosine 204) ERK1 and 2 only (18). Cells were grown to confluence in six-well plates, washed once with PBS, and then starved for 24 h in M199 media containing 0.5% FCS. After the indicated times of stimulation, media were removed and cells were washed twice with ice-cold PBS and lysed with 40 µL of Triton X-100 lysis buffer (50 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid [pH 7.5], 100 mM NaCl, 1% Triton X-100, 1.5 mM MgCl₂, 50 mM NaF, 10 mM sodium pyrophosphate, 1 mM ethylenediaminetetraactic acid [EDTA], 10% glycerol, 1 mM sodium vanadate, 1 mM PMSF, 10 µg/ml leupeptin, and 10 µg/ml aprotinin) augmented with 10 mM -glycerophosphate and 0.1% 2-mercaptoethanol. Plates were rocked for 15 min, and the collected lysates were cleared by centrifugation for 15 min at 12,000 rpm. Proteins were separated using 10% polyacrylamide gels and sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). They were then transferred to nitrocellulose for immunoblotting. Nitrocellulose blots were blocked in PBS/0.5% Tween-20/4% nonfat dry milk for 1 h. Tyrosine phosphorylated proteins were detected using 4G10 monoclonal antibody, goat antimouse HRP conjugate, and ECL. Dual phosphorylation of ERK1 and 2 was detected by exposing blots to phospho-ERK1 and 2 polyclonal antibody for 2 h, followed by goat anti-HRP-conjugated antibody for 30 min, and then reagents for ECL. To assess total amounts of ERK1 and 2, regardless of their phosphorylation states, we stripped the same blots and then probed them a second time using an anti-ERK1 and 2 polyclonal antibody, followed by goat antirabbit HRP conjugate and reagents for ECL. The intensity of protein bands on autoradiograms was quantitated using NIH-Image 1.61 software (National Institutes of Health, Bethesda, MD).

ERK2 Activity Assay

To confirm that the newly phosphorylated ERK proteins also possessed increased kinase activity after exposure to mitogens, we immunoprecipitated them from cellular lysates and measured their ability to phosphorylate myelin basic protein (MBP), as described previously (19). The antibody we identified as suitable for immunoprecipitation of dog ERK (C-14; Santa Cruz Biotechnologies) appeared to capture only ERK2 from our lysates (see RESULTS). For these experiments, cells were grown to confluence in 12- or 24-well plates, washed once with PBS, and then starved for 24 h in M199 media containing 0.5% FCS. After stimulation, media were removed and cells were washed twice with ice-cold PBS, then lysed in Triton X-100 lysis buffer. Lysates were cleared by centrifugation at 12,000 rpm to remove insoluble material. Aliquots of cleared lysates, each containing 100 to 150 µg protein, were incubated overnight at 4°C with 3 µg polyclonal goat anti-ERK2 antibody followed by a 2-h incubation with 10 µl of a 50% slurry of protein G-agarose. Immune complexes were washed four times and assayed for kinase activity in 25 mM Tris (pH 7.0) containing 0.7 mM ethyleneglycol-bis-(-aminoethyl ether)- N,N'-tetraacetic acid and 0.33 mg/ml MBP. Reactions were initiated by the addition of 35 mM magnesium acetate and 35 µM adenosine triphosphate (ATP) (1 µCi -³²P ATP). After a 30-min incubation at 30°C, the assay was terminated by pipetting 40 µL of the reaction mixture onto circles of P81 phosphocellulose paper and immersing the papers in 0.5% phosphoric acid. Papers were washed four times with phosphoric acid and once with acetone before drying and counting using liquid scintillation photometry.

Measurement of c-fos Transcript

We used polymerase chain reaction (PCR) to identify and quantify c-fos transcript in our cells. After exposure of the cells to mitogens, we isolated RNA from them, and synthesized complementary DNAs (cDNAs) for use as templates in PCR. For these experiments, subconfluent cells in six-well plates were exposed to mitogens, and media were removed by gentle aspiration before addition of 0.5 ml RNazol B per well. Total (nuclear and cytoplasmic) RNA was isolated as described previously (20) except that the chloroform concentration was increased 3-fold. RNA concentration was determined by absorption at 260 nm. Single-stranded cDNAs were synthesized from the RNA using MMLV-RT. Each reaction was carried out in 20 µl containing 2 µg RNA as template, deoxynucleotide triphosphates (dNTPs) (0.6 mM each per reaction), random hexamers (50 ng per reaction), RNAsin (20 U per reaction), 50 mM Tris (pH 8.3), 3 mM MgCl₂, 75 mM KCl, and 10 mM dithiothreitol. Reverse transcription was at room temperature for 10 min, followed by 42°C for 45 min, with 200 U of MMLV-RT. Subsequent heat inactivation of the transcriptase was for 5 min at 95°C.

The sequences of oligonucleotides used for PCR were obtained from published gene sequences for human fos (21) and rat -actin (22) genes. The c-fos primers were as follows: the 5' primer, complementary to nucleotides 1910 to 1939, lying within exon 3 of the human c-fos gene, was 5'-GAATAAGATGGCTGCAGCCAAATGCCGCAA; and the 3' primer, corresponding to nucleotides 2230 to 2259 of exon 4 was 5'-CAGTCAGATCAAGGGAAGCCACAGACATCT. The predicted amplimer sizes for human cDNA and DNA, respectively, were 236 and 352 base pairs (bp). -Actin primers were as follows: the 5' primer, complementary to nucleotides 3119 to 3136, was 5'-CCGCAAATGCTTCTAGGC; and the 3' primer, corresponding to nucleotides 3774 to 3754, was 5'-GGTCTCACGTCAGTGTACAGG. The predicted product size for rat cDNA was 656 bp.

In preliminary PCR experiments, reactions were optimized for primer pair concentrations, annealing temperatures, and Mg²⁺ concentrations to produce the greatest yield of product with the least loss of specificity. In other preliminary experiments, concentrations of input cDNA were decreased by serial dilution to define a linear range of product amplification. Similarly, reactions were subjected to 20 to 40 cycles of amplification to determine a linear range, which occurred for both primer pairs between 25 and 28 cycles. Therefore, we employed 26 cycles for all subsequent experiments. A total of 15% of the reverse-transcribed cDNA was added to a reaction mixture containing 10 mM Tris (pH 8.0), 50 mM KCl, 1 to 4 mM MgCl₂, dNTPs (0.2 mM each), 5' and 3' primers (0.2 to 0.6 µM each, 10 to 30 pmols per reaction), and 2.5 units Taq DNA polymerase (final volume, adjusted with distilled water: 50 µl). Reaction mixtures were heated to 94°C for 5 min before the addition of Taq DNA polymerase, and then were overlaid with light mineral oil. Incubations in a DNA thermal cycler for 26 cycles of PCR amplification were as follows: denaturation for 100 s at 95°C, annealing for 70 s at 56°C, and extension for 100 s at 72°C. PCR-generated DNA bands were detected by electrophoresis on a 2% agarose gel in 0.04 M Tris acetate containing 1 mM EDTA. Gels were electrophoresed at 10 V/cm for approximately 60 min, then stained with ethidium bromide.

The identity of the c-fos PCR amplimer was confirmed by sequence analysis. The amplified c-fos cDNA was ligated into pCR2.1, which was used to transform Escherichia coli. Several clones were sequenced using the dideoxy chain termination method and Sequenase T7 DNA polymerase.

Photographs of PCR amplimers in agarose gels were made and video images were obtained and stored using a OneScanner gel documentation system and Ofoto software (Apple Computer, Inc., Cupertino, CA). The bands of interest were scanned and peak areas and mean intensities were determined using NIH-Image 1.55 software (Wayne Rasband, National Institutes of Health). Total pixel intensity for each band was determined as the mean intensity multiplied by the area. Background pixel intensity was subtracted from signal-associated total pixel intensity, giving the background-corrected pixel intensity. Background-corrected pixel intensity for c-fos was normalized to the background-corrected pixel intensity for -actin in the analysis of the results of each experiment. The normalized relative intensities are plotted as means ± standard error of the mean (SEM) (n = 3 or more per treatment group). The data were analyzed using an unpaired t test of Sigma STAT (Jandel Scientific, San Rafael, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Concentration Dependence of Mitogen-Induced Increases in DNA Synthesis

Tryptase and PDGF caused potent, concentration-dependent effects on DNA synthesis in our cells (Figure 1). In all subsequent experiments comparing effects of tryptase with those of PDGF, we used concentrations of the two mitogens that were approximately equipotent and induced 30 to 60% increases in DNA synthesis relative to those produced by 10% FCS. The concentrations of the two mitogens inducing these degrees of DNA synthesis differed somewhat in different preparations of cells.

View larger version (12K): [in this window] [in a new window]   Figure 1.   Dose-dependent effects of tryptase (A), compared with those of PDGF (B), on DNA synthesis. Cells were rendered quiescent in serum-free media for 24 h, and DNA synthesis was measured using a BrdU ELISA, as described in MATERIALS AND METHODS. BrdU incorporation is expressed as percent response to 10% FCS (% max stim), and the origin of the y-axis corresponds to BrdU incorporation in wells containing unstimulated cells in serum-free media. Data are means ± SEM; n = 21 and 23 separate experiments, each measured in triplicate, for tryptase and PDGF, respectively.

Attenuation of Tryptase-Induced DNA Synthesis by Irreversible Serine Protease Inhibition

In 10 experiments, pretreatment with p-APMSF (final concentration: 10 µM) always inhibited the amidolytic activity of tryptase (final concentrations: 2.1 to 5.4 nM) for GPK substrate by 99% or more (data not shown). Using the same assay, we demonstrated that this degree of inhibition persisted for at least 24 h under the conditions of the DNA synthesis assay. Treatment with p-APMSF attenuated tryptase-induced increases in DNA synthesis by 76.8 ± 6.4% but did not significantly alter responses to PDGF (Figure 2).

View larger version (30K): [in this window] [in a new window]   Figure 2.   Effects of an irreversible serine proteinase inhibitor, p-APMSF, on BrdU incorporation induced by tryptase and PDGF. Tryptase and PDGF were pretreated with p-APMSF, or its vehicle alone (control), before their application on growth-arrested cells, as described in MATERIALS AND METHODS. Data are means ± SEM; n = 6 and 10 separate experiments for PDGF and tryptase, respectively. *Significant attenuation by p-APMSF of responses to tryptase (P < 0.05, paired t test). The origin of the y-axis corresponds to BrdU incorporation in wells containing unstimulated cells in serum-free media.

Mitogen-Induced Increases in ERK1 and 2 Phosphorylation

Antiphosphotyrosine immunoblots. Exposure to tryptase for 5 or 10 min increased the phosphorylation of two protein bands on immunoblots probed with an antiphosphotyrosine-specific antibody (Figure 3A). These phosphorylated proteins ranged in size from 46 to 48 kD and 42 to 44 kD and comigrated with ERK1 and 2 as identified on the same immunoblots using a polyclonal anti-ERK1 and 2 antibody (data not shown). PDGF treatment increased tyrosine phosphorylation of the same two proteins. PDGF increased tyrosine phosphorylation of many other proteins, as expected, and equipotent concentrations of tryptase had much more limited effects (Figure 3A). Intensities of the ERK1 and 2 bands, expressed as percent increases over controls, are shown in Figure 3B. Concentrations of tryptase and PDGF that were approximately equipotent in terms of their effects on DNA synthesis caused increases in ERK tyrosine phosphorylation which differed in two respects, depending on the mitogen. First, tryptase caused much smaller increases than PDGF. Second, there was a suggestion that the intensity of the bands induced by tryptase was increased at 10 compared with 5 min, whereas the intensity of PDGF-induced bands were either unchanged or decreased at 10 compared with 5 min (Figure 3B).

View larger version (27K): [in this window] [in a new window]   Figure 3.   Tryptase-induced increases in tyrosine phosphorylation of ERK1 and 2. Cells were stimulated for 5 or 10 min with either tryptase (8 to 10 nM) or PDGF-BB (50 to 100 ng/ ml). Lysates were prepared, and proteins were separated using SDS-PAGE as described in MATERIALS AND METHODS. (A) Representative autoradiogram of an immunoblot in which 4G10 antiphosphotyrosine antibody and enhanced chemiluminescence were used for detection. Arrows on the left indicate positions of ERK1 and 2 (p44 and p42). (B) Histograms showing intensities of protein bands corresponding to tyrosine phosphorylated ERK1 (top) and 2 (bottom) on immunoblots prepared from lysates of tryptase- or PDGF-treated cells. Results are expressed as percent increases above intensities of the control bands at the same time points. Data are means ± SEM, n = 4 experiments.

Anti-phospho-ERK immunoblots. MEK maximally activates ERK1 and 2 by phosphorylating both threonine and tyrosine in a TEY sequence of the ERK proteins (11). Using an anti-phospho-ERK antibody developed to detect these phosphorylated residues, we identified two bands corresponding to phospho-ERK1 and 2 in cellular lysates. Tryptase-induced threonine and tyrosine phosphorylation of ERK1 and 2 was reduced after pretreatment of the tryptase with p-APMSF (Figure 4). In five separate experiments, the reduction in phospho-ERK by pretreatment with p-APMSF was 57.6 ± 9.8% for ERK1 and 37.6 ± 6.9% for ERK2 (mean ± SEM).

View larger version (16K): [in this window] [in a new window]   Figure 4.   Effects of serine proteinase inhibition on tryptase-induced dual phosphorylation of ERK1 and 2. (A) Immunoblot, prepared using an anti-phospho-ERK1 and 2 antibody, shows the attenuating effects of pretreating tryptase with the irreversible serine proteinase inhibitor p-APMSF (10 µM) on ERK1 and 2 activation states (ERK1*, ERK2*). In this experiment, tryptase (2.9 nM), treated with p-APMSF or its diluent alone, was applied to cells for 15 min before preparation of lysates. Also shown in (B) are the ERK1 and 2 signals from the same blot but probed with an anti-ERK1 and 2 antibody which detects the proteins independent of their phosphorylation states.

Figure 5 shows the results of time-course experiments in which cells were exposed to tryptase or PDGF for 0 to 120 min before lysis and immunoblotting as shown in Figure 4. At baseline (time 0), ERK2 phosphorylation (Figure 5, bottom) was present even before mitogen addition but ERK1 phosphorylation was not (Figure 5, top). The kinetics of responses to the two mitogens were quite different. For PDGF, responses were near maximal at 5 min. This maximal level of stimulation persisted for 10 to 15 min and then decreased promptly to near control levels by 60 min. For tryptase, the onset of phosphorylation was delayed and did not reach a maximum until 15 min. For the remainder of the 120-min period, responses to tryptase were sustained and remained significantly above control levels, even at 120 min.

View larger version (17K): [in this window] [in a new window]   Figure 5.   Compared time courses of tryptase- versus PDGF-induced dual phosphorylation (on both tyrosine and threonine) of ERK1 and 2. Confluent airway smooth-muscle cells were growth-arrested for 24 h in 0.5% FCS before stimulation with either tryptase (8 to 10 nM) or PDGF-BB (50 to 100 ng/ml). After preparation of lysates and separation of proteins by SDS-PAGE, immunoblots were prepared using an anti-phospho-ERK1 and 2 antibody that detects phosphorylation on both tyrosine and threonine, as described in MATERIALS AND METHODS. The figure shows the magnitude of ERK1 (top) and 2 (bottom) phosphorylation signals from scanned autoradiograms. Signal intensities were normalized to the maximum value in each experiment in response to tryptase (triangles, dashed lines) or PDGF (circles, solid lines), and the results shown are means ± standard deviation from two or three experiments. *Significant difference between band intensities for tryptase versus PDGF at each time point (P < 0.05, paired t test).

Mitogen-Induced Increases in ERK2 Activity and Their Attenuation by MEK1 Inhibition

ERK activity was assessed directly in immunoprecipitates prepared from cell lysates using an antibody recognizing a carboxy terminal epitope of rat ERK2. In several species, this antibody also cross-reacts with ERK1. However, no ERK1 was detected in our immunoprecipitates on blots probed with the anti-ERK1 and -ERK2 polyclonal antibody. Therefore, the precipitating rat ERK2 antibody appeared selective for canine ERK2 and our assay provides a selective measure of ERK2 activity. Treatment of cells with 50 to 100 ng/ml PDGF (approximately 2 to 4 nM) induced 8.1 ± 2.3- and 7.5 ± 2.3-fold increases in ERK2 activity at 5 and 10 min, respectively (Figure 6A). In other experiments, we found that tryptase-induced increases in ERK2 kinase activity were sustained at 30 and 60 min when compared with 15 min (Figure 6B). By contrast, responses to PDGF at 30 and 60 min clearly were decreased compared with 15 min (Figure 6B). Tryptase (3 to 10 nM) increased ERK2 activity by 2.3 ± 0.6- and 3.8 ± 0.7-fold above basal values at 5 and 10 min, respectively (Figure 6A). Because ERK1 and 2 are activated by the dual-specificity kinases MEK1/2 (11), the perturbation of MEKs by PD098059, a specific inhibitor (13) of MEK1 and 2, should attenuate ERK activity in this assay. As shown in Figure 7, pretreatment of cells with PD098059 before their exposure to tryptase or PDGF significantly inhibited basal and mitogen-induced ERK2 activities in lysates prepared from these cells.

View larger version (21K): [in this window] [in a new window]   Figure 6.   Compared effects of tryptase and PDGF-BB on ERK2 kinase activity. Cells were growth-arrested for 24 h in 0.5% FCS before stimulation for 5 or 10 min with either tryptase (3 to 10 nM) or PDGF-BB (100 ng/ml). Lysates were prepared, ERK2 was immunoprecipitated, and its ability to phosphorylate MBP was quantified, as described in MATERIALS AND METHODS. Results shown in A are ERK2 activities after 5- or 10-min periods of stimulation and are expressed as a fold of the activity measured in lysates from control cells not treated with mitogen. Data are means ± SEM; n = 7 and 4 separate experiments for tryptase and PDGF, respectively. Results shown in B are ERK2 activity measured in other experiments after 15-, 30-, or 60-min periods of stimulation with each mitogen. ERK2 activities measured in these experiments are expressed as percentage of responses measured at the 15-min time point for each mitogen. Data are means ± SEM of five and four experiments with tryptase and PDGF, respectively. *Significant differences between ERK2 activities in response to tryptase or PDGF at each time point (P < 0.05, paired t test).

View larger version (24K): [in this window] [in a new window]   Figure 7.   Effects of a MEK inhibitor on mitogen-induced ERK2 activity. Confluent airway smooth-muscle cells were growth-arrested 24 h in 0.5% FCS. A group of cells were preincubated with PD098059 for 45 min before exposure for 10 min to media alone (basal), PDGF-BB (100 ng/ml), or human lung tryptase (4 to 5 nM). The ERK2 activity in immunoprecipitates from four or five experiments is expressed as a fold of basal control. *Significant difference compared with control not exposed to PD098059 (P < 0.05, paired t test).

MEK1/2 Inhibition Reduces Tryptase-Stimulated DNA Synthesis

PDGF-BB and tryptase-induced increases in BrdU incorporation were largely abolished by pretreatment of cells with the MEK1/2 inhibitor PD098059 (Figure 8). Pretreatment of cells with 50 µM PD098059 inhibited BrdU incorporation induced by tryptase (2.3 to 9.9 nM) or PDGF (100 ng/ml) by 102.4 ± 6.1 and 89.8 ± 7.3%, respectively (means ± SEM, n = 6 and 9 experiments for the two mitogens, each performed in triplicate). Responses to 10% FCS were not affected (Figure 8).

View larger version (29K): [in this window] [in a new window]   Figure 8.   Effects of a MEK inhibitor on mitogen-induced BrdU incorporation. Cells were preincubated with 50 µM PD098059 or its diluent alone (0.5% dimethyl sulfoxide) for 30 min before the addition of tryptase (2.3 to 9.9 nM), PDGF-BB (50 ng/ml), or 10% FCS. Pretreatment of cells with PD098059, but not its diluent, significantly attenuated mitogenic effects of tryptase and PDGF, but not 10% FBS. Results shown are means ± SEM, n = 5-9 experiments. The origin of the y-axis corresponds to BrdU incorporation in wells containing unstimulated cells in serum-free media. *Significant difference compared with control (P < 0.05, paired t test).

Mitogen-Induced Increases in c-fos Transcript

Figure 9 shows the electrophoresis of the DNA generated by RT-PCR which produced bands of sizes predicted from the published c-fos and -actin gene sequences (21, 22). Cloning and sequencing of DNA in the band generated with the c-fos primers indicated that, in the 236-bp amplified region of this gene, the dog c-fos product was 98 and 100% identical to human c-fos in nucleotide and amino acid sequence, respectively. To test for differences in RNA integrity and efficiency of the RT reactions, RT-PCR for -actin also was carried out. Assuming that -actin was equally expressed under all conditions and time points, we normalized the RT-PCR-amplified c-fos signal to the -actin signal. PDGF increased the normalized c-fos transcript by greater than 12-fold after 30 min, nearly equivalent to the increases induced by 10% fetal bovine serum (FBS) (data not shown). By comparison, tryptase increased c-fos transcript by only about 2-fold. Further, the peak response to PDGF occurred at 30 min and declined sharply by 75 min (Figure 9B), whereas the response to tryptase was highest at 75 min, remained high at 180 min, and declined toward baseline at 360 min (Figure 9A).

View larger version (29K): [in this window] [in a new window]   Figure 9.   Compared time course of tryptase- versus PDGF-induced increases in c-fos transcript. (A) Representative photographs of ethidium bromide-stained agarose gel electrophoresis of PCR-amplified -actin and c-fos products. Growth-arrested cells were stimulated with 1 nM tryptase or 10% FBS for the indicated times. RNA was isolated and RT-PCR performed as described in MATERIALS AND METHODS. (B) Time course of mitogen-induced c-fos transcript as demonstrated by quantitative RT-PCR. Growth- arrested cells were stimulated with 1 nM tryptase, 50 ng/ml PDGF-BB, or carrier for the indicated periods of time. RT-PCR was carried out and band intensities corresponding to c-fos amplification products were corrected for background and normalized to -actin, as detailed in MATERIALS AND METHODS. The histogram represents the mean ± SEM of three or more experiments carried out at each time point. *Significant increase in transcript compared with control at same time (P < 0.05, unpaired t test).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Thickening of the airway wall is a characteristic finding in asthma that relates in part to hyperplasia of airway smooth-muscle cells (7). This change in the airway wall contributes importantly to the development of bronchial hyperresponsiveness because the thickening gives rise to exaggerated decreases in airway caliber when the smooth muscle shortens (7). Tryptase is released in airways during allergen challenge in atopic subjects (1) and is present in increased concentrations in induced sputum samples obtained from patients with asthma (1). Approximations of in vivo tryptase concentrations in bronchi are possible, based on estimates of mast cell numbers in the smooth-muscle layer and of tryptase concentrations per mast cell (4). Over this range of estimated nanomolar concentrations, we have previously demonstrated that tryptase is a mitogen in cultured dog (4) and human (23) airway smooth-muscle cells. The major new finding of the current study is that tryptase activates the ERK class of MAP kinases in these cells and that these proteins likely are important second messengers for tryptase-induced mitogenesis. In addition, our time-course experiments demonstrated striking differences between tryptase and PDGF in both the early and sustained phases of ERK activation (Figure 5). Tryptase's effects on ERK activation depended in part upon its properties as a protease, but our experiments demonstrated a substantial proportion (40 to 60%) mediated via nonproteolytic mechanisms. Although prior studies have not always shown complete inhibition of tryptase's effects by protease inhibitors (2, 6), our findings constitute the first clear demonstration, to our knowledge, that nonproteolytic actions may contribute importantly to tryptase's activation of cells.

Our experiments provide several different lines of evidence for tryptase-induced activation of ERK1 and 2, including increased tyrosine phosphorylation (Figure 3), specific dual phosphorylation of threonine 202 and tyrosine 204 in ERK1 and 2 (Figure 4), and ERK2 activity in immunoprecipitates (Figure 6). The findings add tryptase to the growing list of airway smooth-muscle mitogens for which ERK activation has been demonstrated, and these include PDGF (14), insulin-like growth factor-1 (12), epidermal growth factor (12), endothelin (24), thrombin (25), and -hexosaminidase (26). Thus, ERKs are likely a part of a pathway shared by multiple smooth-muscle mitogens acting via different receptors.

Despite the relatively similar increases in DNA synthesis ultimately induced by tryptase and PDGF, the time courses of ERK activation in response to tryptase and PDGF differed from one another in several respects over the 2 h they were assessed in our study (Figure 5). The response to tryptase was slower in onset, reaching a peak at about 15 min, as compared with 5 min for PDGF (Figure 5). Similar differences in the onset kinetics were suggested in our analysis of the antiphosphotyrosine immunoblots (Figure 3) and measured ERK2 activities (Figure 6A). In addition, increases in c-fos transcripts were delayed in onset in response to tryptase compared with PDGF (Figure 9), indicating that the differences in ERK1 and ERK2 kinetics were also reflected in the downstream activation of an important early response gene (15).

To our knowledge, few other investigators have reported differences in the time to achieve maximum ERK activation under different experimental conditions in other cells. However, in one study, trypsin induced brisk increases in ERK activity (peaking at 5 min) in rat aortic smooth muscle cells, and this effect of trypsin was mediated via direct activation of PAR-2 (27). In the same study, in bovine pulmonary arterial fibroblasts, peak ERK activation by trypsin was delayed and did not occur until 15 min; the mechanism for trypsin-induced ERK activation in these cells is unknown but was independent of PAR-2. In our experiments, the delay in tryptase-induced ERK activation may be evidence against direct activation of a cell-surface receptor and may reflect, instead, a multiple-step requirement for cellular activation. Examples of such multiple-step processes include proteolytic cleavage of factors in the media, releasing their growth-promoting activity for the smooth-muscle cells, or proteolytic liberation from extracellular matrix of factors, such as transforming growth factor- or fibroblast growth factors, which in turn may activate their own receptors on the smooth-muscle cells. Our findings indicate the need for more information about the earliest events that occur during tryptase's activation of cells in airways and lung.

Another difference in the time courses of ERK phosphorylation by tryptase versus PDGF in our experiments was the more sustained response to tryptase. Thus, at 60 to 120 min after addition of the mitogens, ERK1 and 2 phosphorylation in response to PDGF had returned nearly to baseline levels, whereas responses to tryptase still were elevated well above those levels (Figure 5). Measurements of ERK2 kinase activity confirmed our observations that responses to tryptase were more sustained than to PDGF (Figure 6B). We can only speculate as to possible mechanisms contributing to these differences. The ERK activation state at any time is the net effect of activating kinases and inactivating protein phosphatases. Tyrosine- and threonine-phosphorylated sites on ERK1 and 2 are susceptible to the actions of phosphatases in cytosolic and nuclear compartments (28). Some immediate early genes encode dual specificity MAP kinase phosphatases (MKPs) that inactivate ERK by dephosphorylating both threonine and tyrosine residues. In fibroblasts, MKP-1 messenger RNA was induced in 15 min in response to serum (29). Thus, in our cells, it is possible that PDGF exposure leads to early increases in ERK phosphatase activity but that tryptase either fails to induce these effects or does so more slowly. Inhibitory feedback mechanisms may also modulate the kinetics of intracellular signaling, and activated ERK has been shown to phosphorylate Son of sevenless (SOS) and to disassociate the Grb2-SOS complex involved in Ras activation (30). In our experiments, differences in the importance of such feedback regulation of the Grb2-SOS complex and Ras activation during PDGF- and tryptase-induced ERK activation might account for the observed differences in kinetics. In a Chinese hamster ovarian fibroblast cell line, angiotensin II induced a sustained phase of ERK activity via 12-lipoxygenase activation (31). Whether such a mechanism accounts for tryptase's sustained activation is unknown.

In some previous studies, sustained ERK activation appeared vital for growth stimulation. For instance, in CCL39 cells, thrombin caused both peak and sustained activation of ERK1 (32) and the sustained activation correlated best with the degree of mitogenicity. However, in other cells, thrombin induced a brisk and only transient activation of ERK that was followed by potent mitogenic effects (33). The finding in our experiments that concentrations of tryptase and PDGF which induced comparable increases in DNA synthesis caused quite different degrees of late-phase ERK activation would suggest that sustained ERK activation was not a requirement for induction of potent mitogenic effects in our cells. However, the 2-h period of our observations is not sufficient to make firm conclusions about this issue. Also, the fact that PDGF-stimulated levels of ERK1 and 2 phosphorylation at 60- and 120-min time points still were slightly above basal (Figure 5) may have been an important determinant of growth responses to this mitogen.

Our conclusion that ERK activation is required for tryptase-induced mitogenesis rests on the observed inhibitory effects on tryptase-induced BrdU incorporation of PD098059 (Figure 8). This compound is a cell-permeable inhibitor of MEK1 and 2 (13), the kinases just upstream from ERKs that dually phosphorylate and activate them (11). The strength of this evidence depends on the specificity of PD098059 effects; and support for a high degree of specificity, in the concentrations employed in our experiments, is emerging on the basis of findings in both airway smooth-muscle cells (14) and other systems (13). However, the definitive demonstration of a requirement for ERK activation would necessitate using other experimental approaches, including transient transfections with cDNAs encoding dominant-negative MEK1 or ERKs (14). Also, our experiments provide no information about the degree to which ERK activation alone is sufficient in mediating tryptase's growth-stimulatory effects and no information about the potential importance of other pathways, such as activation of phosphatidylinositol 3-kinase, in mediating these effects. The failure of PD098059 to completely inhibit the mitogenesis induced by 10% FBS in our experiments (Figure 8) is consistent with the compound's reported inability to attenuate extremely potent degrees of MEK1 activation in intact cells (13, 24), perhaps because of its low solubility in aqueous solutions making it difficult to achieve high intracellular concentrations (13).

Our experiments with the irreversible serine protease inhibitor p-APMSF demonstrated that substantial portions of the responses to tryptase were mediated via proteolytic cleavage by its catalytic site. On the other hand, approximately 25% of tryptase-induced increases in DNA synthesis (Figure 2) and 40 to 60% of the peak increases (at 15 min) in ERK1 and 2 phosphorylation (Figure 5) persisted after treatment of the tryptase with p-APMSF. It is possible but unlikely that these latter effects related to the 1% or less residual catalytic activity in p-APMSF-treated tryptase. Also, it is important to note that although the majority of thrombin's mitogenic effects are mediated via proteolytic mechanisms (33), there is strong evidence that noncatalytic portions of the thrombin molecule also can induce mitogenesis (10). Similarly, it is possible that tryptase's effects in our cells related to both catalytic and noncatalytic effects of the molecule. An interaction of tryptase's glycosyl residues (34) with glycoprotein receptors on the smooth-muscle cell is one noncatalytic effect that might lead to growth stimulation, particularly because glycoprotein receptor activation clearly can induce mitogenesis in airway smooth-muscle cells (26). Also, we cannot exclude with absolute certainty the presence of a contaminant in our preparations of human lung tryptase that serves as an important cofactor with tryptase during ERK activation. However, the purity of the human lung tryptase preparations used in our experiments has been characterized extensively using chromatographic, electrophoretic, and immunologic criteria (3). Tryptase from another source, HMC-1 cells (35), also has potent mitogenic effects in our cells (current authors' unpublished observations), and it is somewhat unlikely that an important contaminating cofactor would be common to two preparations isolated by such different techniques.

In summary, our experiments demonstrate that tryptase activates MAP kinases of the ERK class in airway smooth-muscle cells and that this signaling pathway is likely important in stimulating DNA synthesis in these cells. We have developed a new technique for irreversible inhibition of tryptase's catalytic activity using p-APMSF. Our findings indicate that, although full degrees of ERK activation and increased DNA synthesis in response to tryptase require an intact catalytic activity, substantial portions of these responses likely are mediated via noncatalytic effects. Our experiments also demonstrated striking differences in the kinetics of tryptase- versus PDGF-induced ERK activation. These findings suggest the need for more information about the earliest events at the cell membrane which mediate tryptase's mitogenic effects and about the significance of transient versus sustained ERK activation in response to different mitogens.