Introduction

Abstract Introduction Materials and Methods Results Discussion References

Endothelin (ET)-1 is a 21-amino-acid peptide with a striking diversity of important physiologic effects (1). Aside from being the most potent vasoconstrictor yet identified (2), ET-1 is critical for proper development of the head and neck, as indicated by severe craniofacial abnormalities in ET-1 knockout mice (3). Increases in ET are associated with heart failure, which can be treated by the use of ET receptor antagonists (4). Increased levels of ET are found in the bronchoalveolar lavage fluid of asthmatics, suggesting a role in the pathogenesis of asthma (5, 6). Furthermore, ET is a mitogen for airway smooth-muscle (ASM) cells (7, 8) and may be involved in the remodeling that is characteristic of the asthmatic airway (9).

There are two ET receptor subtypes, ETA and ETB, which are members of the superfamily of G-protein coupled receptors (10). ASM cells have been reported to have the ET receptor subtypes at a ratio of 35% ETA:65% ETB (13). ET binding to the ETB receptor triggers several intracellular signaling pathways, including activation of phospholipase C and increases in cytosolic free calcium and inositol phosphates (14). ETB receptor activation is also associated with tyrosine phosphorylation of several cellular proteins (17), and with activation of both the extracellular regulated kinase (ERK) and the c-Jun NH₂ terminal kinase (JNK), subgroups of mitogen-activated protein (MAP) kinases (21). Activation of ERK is important for ET-stimulated ASM cell proliferation, because inhibiting the enzyme blocks proliferation (24).

ERKs are 42- and 44-kD proline-directed, serine/threonine protein kinases activated by diverse stimuli that induce cell growth and/or differentiation in a wide range of species (25). Catalytic activation of ERK results from a cascade of protein phosphorylation events, culminating in phosphorylation of its regulatory tyrosine and threonine residues and subsequent translocation of the enzyme to the nucleus, where it phosphorylates transcription factors such as Elk1 (26). Defined upstream components in the pathway leading to ERK activation include MAP/ERK kinase (MEK) and the protein kinase Raf (27). MEK, the only known activator of ERK, is phosphorylated and activated by Raf-1 (30). Activation of Raf is dependent on membrane localization, involving the guanosine triphosphate-bound form of Ras and poorly defined phosphorylation events (33). For example, phosphorylation of Raf has both positive and negative effects on catalytic activity. Phosphorylation of serine 621 of Raf by the cyclic adenosine monophosphate-dependent protein kinase A (PKA) has been shown to inhibit Raf activity (34). Conversely, phosphorylation by PKC results in activation of Raf (35). In addition, tyrosine phosphorylation may be important for Raf activity because Tyr340 and Tyr341 are required for activation (36).

ET has been demonstrated to activate Raf in cultured ventricular myocytes (37), and Raf activation may be an important step in the proliferation of ASM cells because activating PKA with forskolin, which is known to inhibit Raf by phosphorylation of Ser621, also inhibits ET-stimulated proliferation (8).

ET stimulates a rapid, sustained increase in cytosolic free calcium in ASM cells that may result from influx through L-type calcium channels and from release from internal stores (38). Cytosolic free calcium is known to activate kinase pathways leading to the stimulation of transcription factors. For example, in neuronal PC12 cells, increased calcium activates calcium/calmodulin-dependent kinases which subsequently activate the ERK and JNK MAP kinases (39). The result of this activation cascade is the phosphorylation of CREB and transcription of CRE-dependent genes (40). ET-stimulated calcium increases and MAP kinase activation may be interrelated, because truncation of the ETB receptor cytosolic tail results in loss of MAP kinase activation and in calcium increase (21). However, the potential relationship between MAP kinase activation and calcium in the ET signal transduction pathway is poorly defined.

This study focuses on the upstream regulators of ERK activation in the ET signal transduction pathway. Our findings reveal that in ASM cells, Raf is a key upstream activator of the ERK pathway, and that both ERK and Raf activation require calcium influx.

	Materials and Methods

Abstract Introduction Materials and Methods Results Discussion References

Reagents

ET was purchased from Bachem (Torrance, CA) and maintained as a frozen stock in dilute acetic acid at 80°C. ET was used at a final concentration of 100 nM unless otherwise specified. Nisoldipine and 12-0-tetradecanoylphorbol-13-acetate (TPA) were purchased from Sigma (St. Louis, MO). TPA was used at a final concentration of 1 µM unless otherwise specified, and nisoldipine at 100 nM.

Cells and Transfections

Primary cultures of rat ASM cells were isolated from rat trachea as previously described (41) and maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS). To insure that the differentiated smooth-muscle phenotype was maintained, cells did not exceed the fifth passage. Primary ASM cells were transfected by using a mixture of lipofectamine (GIBCO BRL, Grand Island, NY) and adenovirus (42). Plasmid DNAs (3 to 5 µg total as indicated in figure legends) were added to a mixture of serum-free DMEM (100 µl) containing lipofectamine (30 µl) and 10⁸ particles of replication-defective adenovirus (University of Iowa, Gene Transfer Vector Core, Iowa City, IA). The mixture was incubated at room temperature for 45 min before adding to cells. The ASM cells were grown to approximately 70% confluence and washed twice with serum-free DMEM, and the DNA/lipofectamine/adenovirus mixture was applied to the cells in a total volume of 5 ml. The cells were exposed to this solution for 5 h and the medium was removed and replaced with DMEM containing 10% FBS. The transfected ASM cells were grown for 3 d and serum-starved for the last 16 h in DMEM containing 0.1% FBS. The quiescent cultures were treated with various agonists and the relevant kinases were immunoprecipitated from the lysates as described subsequently. Using adenovirus to shuttle the expression plasmids into the cell resulted in a transfection efficiency estimated to be approximately 25 to 40% on the basis of previous experiments using green fluorescent protein as a marker (data not shown). In addition, the ectopically expressed ERK2 was similar in expression levels to the endogenous ERK2, confirming that the expression level is reasonably high using this method. In transfection experiments in which the effect of a dominant-negative Raf mutant was analyzed, the amount of Raf DNA exceeded the ERK DNA 3:1 to insure that cells expressing transfected ERK were also expressing transfected Raf kinase.

COS cells in 100-mm dishes, approximately 25% confluent, were transiently transfected in serum-free DMEM for 5 h with DNA/lipofectamine (GIBCO BRL) complexes at a ratio of 0.2 µg DNA/1 µl lipofectamine. The medium was replaced with complete DMEM (10% FBS) and the cells were grown overnight. COS cell transfection efficiency was in the range of 40 to 60%. As in the experiments with ASM cells, when the effect of mutant Raf on the ERK pathway was analyzed, the amount of Raf DNA transfected exceeded the ERK DNA 3:1 to insure that all cells expressing ERK also expressed Raf.

Plasmids

Expression vectors for ETB and ERK2 were constructed as previously described (21), as were wild-type and mutant Raf-1 plasmids (35). Raf-1 corresponds to the full-length wild-type kinase; Raf-301 is the kinase-inactive version of Raf containing the site-specific mutation K375W; Raf-BXB is a constitutively active deletion mutant lacking amino acids 26-303; BXB-301 is identical to BXB but is kinase-inactive; and Raf-HCR is the amino terminus of Raf containing the Ras-binding domain.

Immunoprecipitation and Kinase Assays

Myc-epitope-tagged ERK2 was immunoprecipitated with monoclonal antibody (mAb) 9E10, and the activity of ERK2 immune complexes was measured by phosphorylation of myelin basic protein as described (21). Endogenous ERK was immunoprecipitated with C-14 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) in NP-40 IPB (1% NP40, 10 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid [Hepes], 2 mM ethylenediaminetetraacetic acid, 0.1% -mercaptoethanol, 1% aprotinin, 1 mM benzamidine, 0.2 mM sodium vanadate, and 0.05 µg/ml microcystin). Kinase activity of the immune complexes was measured by phosphorylation of myelin basic protein (MBP) in kinase buffer containing 5 µCi of [³²P]ATP, 25 mM Hepes (pH 7.5), 5 mM MgCl₂, 0.2 mM vanadate, and 0.05 µg/ml microcyston, at 30°C for 15 min. Raf-1 was immunoprecipitated with anti-cRaf mAb (Transduction Labs, Lexington, KY) and Raf C12 (Santa Cruz) for 1.5 h at 4°C. The catalytic activity of Raf-1 immune complexes were determined by phosphorylation of catalytically inactive, 5'-p-fluorosulfonylbenzoyladenosine (FSBA)-treated MAP kinase kinase (MKK). FSBA binds to the catalytic site of MKK and prevents its subsequent activation, rendering it catalytically inactive. When immunoprecipitating endogenous Raf from ASM cells, the protein was normalized to cell number. When immunoprecipitating ectopically expressed proteins from ASM cells, the protein level was normalized by Western blot.

Cytosolic Free Calcium Measurements

Measurements of cytosolic free calcium were performed as previously described (21). Briefly, cells loaded with Fura-2 (50 µM) remained at room temperature for 1 h and were then examined under a Nikon microscope that is part of an IMAGE-1/FL quantitative fluorescence PC imaging system (Universal Imaging, Inc., West Chester, PA). ET was applied to the Fura-loaded cells and the calcium concentration was measured once the signal had stabilized, after 2-3 min. Following ET treatment, calcium ionophore (10 µM ionomycin) was added to the bath and the maximal fluorescence was recorded to compare and ensure equal dye-loading among different cell populations.

	Results

Abstract Introduction Materials and Methods Results Discussion References

ET-Stimulated ERK Activation Is Regulated by Raf in ASM Cells

ET is known to activate the Raf kinase in cardiomyocytes (37), and our previous work in ASM cells indicated that inhibiting Raf with forskolin prevented ERK activation by ET (8), further suggesting that Raf activation regulates ERK in ASM cells. In our previous studies we observed very weak activation of the endogenous Raf by ET at 10 min after treatment. Therefore, we have examined the kinetics of Raf activation by ET. Our results demonstrate that Raf is activated within 2 min and declines rapidly, in agreement with the results from cardiomyocytes (37) (Figure 1). This data indicates that Raf is activated by ET, but we were interested in determining whether Raf activation was required for ERK activation, because Raf-independent mechanisms of ERK activation may also exist (43).

View larger version (30K): [in this window] [in a new window]   Figure 1.   Raf is catalytically activated by ET in ASM cells. The kinetics of Raf activation in ASM cells were examined by treating confluent, serum-starved ASM cells with ET or PDGF for various times as indicated. Raf was immunoprecipitated and its catalytic activity measured by immune complex kinase assay using MEK as an in vitro substrate. The results of the experiment are representative of at least three independent experiments.

To examine ET signal transduction, the pathway was reconstituted in COS cells. COS cells were co-transfected with wild-type Raf-1 (WT) or kinase-inactive Raf-301 (301), and the ETB receptor. The cells were treated with ET or TPA for 5 min, and Raf kinase activity was measured by immune complex kinase assay using catalytically inactive MKK as a substrate. ET weakly stimulated endogenous Raf in vector-transfected cells (Figure 2A). In Raf-transfected cells, ET significantly stimulated the catalytic activity of WT, as did TPA (Figure 2A, lanes 5 and 7). This MKK-directed kinase activity can be attributed to Raf and not another ET-stimulated kinase that contaminated the immunoprecipitates, because 301 immunoprecipitated from ET- or TPA-stimulated cells did not have MKK kinase activity (Figure 2A, lanes 6 and 8). Ectopically expressed WT displayed reduced mobility in the ET- or TPA-stimulated cells that was not observed in the 301 transfected cells (Figure 2B). The ability of ET to stimulate the kinase activity of Raf in COS cells suggests that the upstream regulatory features of the ET pathway are similar in ASM cells and COS cells, although the downstream effector function is likely to be significantly different in the two cell types.

View larger version (53K): [in this window] [in a new window]   Figure 2.   Reconstitution of the ET signaling pathway in COS cells. COS cells were cotransfected with the ETB receptor (1 µg), wild-type Raf-1 (WT, 2 µg), kinase-inactive Raf-301 (301, 2 µg), or empty vector (2 µg). The cells were stimulated with ET or TPA for 5 min, and Raf was immunoprecipitated. (A) Raf catalytic activity was measured by immune complex kinase assay using kinase-inactive MKK as a substrate. (B) Expression of Raf was monitored by Western blot. The results shown are representative of two independent experiments.

To determine whether Raf activation was required for ERK activation, the pathway was reconstituted in COS cells by transfecting with plasmids expressing the ETB receptor, myc-tagged ERK2, and various Raf mutants. These included either WT, 301, constitutively active Raf-BXB, kinase-dead Raf-BXB-301, or Raf-HCR, the amino-terminus of Raf containing the Ras-binding domain (35). ERK kinase activity was measured following ET treatment in the various transfected cells. ET treatment rapidly activated ERK2 in the empty vector-transfected cells, and expression of wild-type Raf or 301 had little effect on ET-stimulated ERK2 activation (Figure 3A; compare lane 2 with lanes 4 and 6). Constitutively active Raf (Raf-BXB) stimulated ERK2 kinase activity independent of ligand (lane 7), which was an effect of the catalytic activity of Raf because the kinase-inactive version of this molecule (BXB-301) did not activate ERK2 (lane 9). The amino terminus of Raf containing the Ras-binding domain (Raf-HCR) potently inhibited ET-stimulated ERK2 activation (lane 12), suggesting that the Ras-Raf interaction is required for ET-stimulated ERK activation and that ET-stimulated Raf activation is a Ras-dependent process. Expression of the ectopically expressed proteins was confirmed by Western blot. Analysis of myc-tagged ERK2 (Figure 3B, lower panel) indicates that similar levels of ERK2 were expressed across the experiment, supporting the conclusion that differences in phosphorylation of MBP are due to changes in the catalytic activity of the enzyme and not to differences in protein expression levels. Expression of Raf protein was detected by Western blot (Figure 3B, upper panel). Raf-BXB and Raf-BXB-301 were not recognized by this antibody.

View larger version (44K): [in this window] [in a new window]   Figure 3.   Dominant-negative Raf interferes with ET-induced ERK activation. COS cells were cotransfected with the ETB receptor (1 µg), Myc-tagged ERK2 (1 µg), and either wild-type Raf-1 (2 µg), kinase-inactive Raf-301 (2 µg), constitutively activated Raf-BXB (2 µg), a kinase-inactive version of BXB (2 µg), Raf-BXB-301 (2 µg), or Raf-HCR (2 µg), a dominant-negative form of Raf containing the Ras-binding domain. ET-treated and -untreated cells were immunoprecipitated with 9E10 (anti-myc tag). (A) ERK2 catalytic activity was measured by immune complex kinase assay using MBP as an in vitro substrate. (B) Western blot confirming similar levels of ectopically expressed Raf and ERK2. The results shown are representative of four independent experiments.

To determine whether Raf is a required element in the ERK activation cascade in ASM cells, primary ASM cells were cotransfected with an empty vector, WT, Raf-HCR, or 301, and ERK. Following ET treatment, ERK was immunoprecipitated and its activity was measured by immune complex kinase assay (Figure 4A; compare lane 2 with lanes 5 and 6). When ERK was coexpressed with dominant negative forms of Raf (Raf-HCR, Raf-301) its activation in response to ET was greatly attenuated (Figure 4A). Furthermore, expressing WT augmented ERK activation in response to ET (Figure 4A; compare lanes 1 and 2 with 3 and 4). Together, the data demonstrate that Raf activation is an important upstream feature of the ET signaling pathway that results in ERK activation.

View larger version (57K): [in this window] [in a new window]   Figure 4.   Dominant-negative Raf inhibits ET-stimulated ERK activation in ASM cells. ASM cells were transfected with ERK2 (2 µg) and an empty expression vector (3 µg), wild-type Raf-1 (3 µg), dominant-negative Raf-HCR (3 µg), or Raf-301 (3 µg). ERK2 was immunoprecipitated by means of its myc epitope using mAb 9E10, and its kinase activity was measured by immune complex kinase assay using MBP as an in vitro substrate. The lower panel is a Western blot to document the expression of ERK. The results shown are representative of three experiments.

ET-Stimulated ERK Activation Requires Calcium Influx

Stimulation of the ETB receptor is known to result in increased cytosolic free calcium (10). Furthermore, our previous work with a truncated ETB receptor suggested that the calcium signaling pathway and the MAP kinase pathway are interrelated (21). Therefore, we examined the relationship between calcium influx and ERK activation in ASM cells. The ASM cells were pretreated with ethyleneglycol-bis-(-aminoethyl ether)-N,N'-tetraacetic acid (EGTA) to chelate extracellular calcium, the dihydropyridine (DHP) nisoldipine to block calcium entry through L-type calcium channels, or ryanodine (RYA) to inhibit intracellular calcium release mediated by the RYA receptor. Both chelating calcium with EGTA and blocking calcium entry with nisoldipine inhibited ET-stimulated ERK activation (Figure 5A; compare lane 2 with lanes 3 and 4). Inhibiting intracellular calcium release through the RYA receptor had little effect on ET-stimulated ERK activation (lane 5). However, the expression level of the RYA receptor in primary cells may be decreased, making it difficult to interpret this result clearly. The calcium dependence of ERK activation was specific for ET because neither EGTA nor DHP inhibited phorbol ester-stimulated ERK activation (Figure 5A; compare lane 6 with lanes 7 and 8). Furthermore, depolarization of the ASM cell membrane with 80 mM K⁺ resulted in a DHP- and EGTA-sensitive influx of calcium but did not activate ERK (lane 9), suggesting that calcium entry alone is not sufficient to activate the ERK MAP kinase. Together, the data suggest that calcium entry is a critical step in the ET signal transduction pathway leading to ERK activation. Although the ET signaling pathway appears to require calcium influx for ERK activation, it is unlikely that this is a universal requirement for ERK activation in ASM cells.

View larger version (21K): [in this window] [in a new window]   Figure 5.   Calcium influx is required for ET-stimulated ERK activation in ASM cells. ASM cells were pretreated with EGTA (2 mM), DHP (10 µM), RYA (10 µM), or vehicle for 10 min, followed by ET or phorbol ester treatment for 10 min. (A) Endogenous ERK2 was immunoprecipitated and its catalytic activity measured by immune complex kinase assay using MBP as an in vitro substrate. (B) Cells were pretreated with EGTA, DHP, or RYA for 10 min. The cells were then treated with ET, TPA, or K+, and the cytosolic free calcium was measured following 1 to 2 min of agonist treatment, at which time the signal had stabilized as shown in panel C. (C) Kinetics of the calcium increase in ASM cells following treatment with various agonists. The results shown are representative of three independent experiments. The error bars represent the standard error of the mean of three replicates.

The profile of calcium response in the ET-stimulated ASM cells was examined to confirm that calcium entry was indeed blocked. Treating the cells with ET or TPA stimulated a rapid and sustained increase in cytosolic free calcium (Figures 5B and 5C). Pretreating the ASM cells with EGTA or DHP prevented ET- and TPA-stimulated increase in cytosolic free calcium (Figure 5B; compare lanes 3 and 4 with 7 and 8). Depolarizing the cells with 80 mM K⁺ also stimulated a rapid and sustained increase in cytosolic free calcium (Figures 5B and 5C). Although DHP significantly decreased ET-stimulated calcium increase, a small transient increase was observed, presumably mediated by internal release. Together, the data indicate that preventing calcium influx into the ASM cells inhibited the ability of ET, but not phorbol ester, to stimulate ERK activation. However, calcium influx alone was not sufficient to activate ERK fully.

ET-Induced Raf Activation Requires Calcium Influx

To determine whether the calcium requirement of the ET signaling pathway was upstream of Raf, endogenous Raf was immunoprecipitated from ASM cells pretreated with DHP or vehicle, then treated with ET for 5 min. Blocking calcium influx with DHP inhibited ET-stimulated Raf activation (Figure 6; compare lanes 2 and 5). However, there was no inhibition of phorbol ester-stimulated Raf activation (compare lanes 3 and 6), confirming that ET-stimulated, but not phorbol ester-stimulated, ERK activation is calcium-dependent in ASM cells. Therefore, the calcium dependence of ERK activation may be a feature specific to the ET pathway because other agonists, such as TPA, appear not to require calcium influx for ERK activation.

View larger version (30K): [in this window] [in a new window]   Figure 6.   Calcium influx is required for ET-stimulated Raf activation. ASM cells were pretreated with DHP or vehicle, followed by ET or phorbol ester treatment for 5 min. Endogenous Raf was immunoprecipitated and an immune complex kinase assay was done using kinase-inactive MKK as an in vitro substrate. The results shown are representative of two independent experiments.

	Discussion

Abstract Introduction Materials and Methods Results Discussion References

ET has several important physiologic effects, including vasoconstriction, bronchoconstriction, and mitogenesis. In addition, it is thought to be important in heart failure (4), Hirschsprung's disease (44), and the development of craniofacial structures (3). Several lines of evidence suggest ET may also be important in asthma (5). Two subtypes of G-protein coupled receptors, ETA and ETB, are thought to mediate the biologic effects of ET through the activation of heterotrimeric G-proteins (45).

The ETB receptor, the primary receptor found in ASM, activates the ERK and JNK subgroups of MAP kinases, and stimulates the proliferation of ASM cells (8). Other investigators, using ETA-specific antagonists, have also demonstrated the importance of ETA in mediating proliferation of ASM cells (13). Bogoyevitch and colleagues have demonstrated that Raf is rapidly and transiently activated by ET in cardiac myocytes (37). However, Shapiro and associates found very weak activation of Raf in ASM cells by ET (8). The Raf activation demonstrated by Bogoyevitch and coworkers was maximal at 1 to 2 min and declined by 10 min, whereas the studies by Shapiro and colleagues looked at Raf activation at 10 min, when its activity had presumably declined. In the present study we find that Raf is activated rapidly by ET in ASM cells. Furthermore, the work presented in this report identifies Raf as a critical upstream activator of ERK in the ET signaling pathway. For example, constitutively activated Raf (Raf-BXB) stimulated ERK2 independently of ET, and dominant-negative Raf (Raf-HCR) inhibited the ability of ET to activate ERK (Figure 3). In these experiments, cotransfection of 1 µg of kinase-inactive Raf (Raf-301) had a modest effect on ERK activation. In other experiments, Raf-301 did have a dominant-negative effect but required much greater levels of expression of the kinase-inactive form of Raf (4 to 5 µg of DNA, data not shown). The data indicate that Raf-HCR, containing the amino terminal Ras-binding domain, is a very effective inhibitor of the ERK pathway, presumably by disrupting the Raf-Ras interaction, suggesting that ET activates Raf in a Ras-dependent process. Furthermore, Raf was shown to be rapidly activated in ASM cells treated with ET, and appears to be an important upstream regulator of the pathway in this cell type because dominant-negative forms of Raf interfered with ERK activation in response to ET (Figure 4).

The ET signal transduction pathway involves a rapid and sustained increase in cytosolic free calcium. This calcium increase is thought to activate calcium-dependent kinases such as PKC, and calcium/calmodulin kinase (40). In mesangial cells, the downstream transcriptional activation of c-fos was determined to require calcium influx (46). Therefore, we examined the relationship between calcium influx and ERK activation because ERK is known to be an upstream regulator of various transcriptional responses (26). Our findings suggest that ET-stimulated ERK activation in ASM cells requires calcium influx. Inhibiting the influx of calcium through calcium channels with the DHP calcium channel-blocker nisoldipine inhibited ET-stimulated but not phorbol ester-stimulated ERK activation (Figure 3). Raf activation was similarly inhibited by preventing calcium influx (Figure 4). Our results are consistent with the work of Wang and Simonson in mesangial cells in which ET-stimulated c-fos gene transcription was regulated by calcium influx (46). The data presented in the present report indicate that calcium influx is required for ERK activation in the ET pathway, and that this may be specific for ET because TPA-stimulated ERK activation was unaffected by blocking calcium influx. Furthermore, calcium influx alone is not sufficient to activate ERK because depolarizing the ASM cells with K⁺ stimulated calcium influx but did not activate ERK (Figure 3). Our results and those of Panettieri and coworkers (47) demonstrate that calcium influx alone is insufficient to stimulate ERK activation or ASM cell proliferation.

Although inhibiting calcium influx interfered with ET-stimulated Raf activation, Raf itself is not thought to be calcium-dependent. The most probable explanation of the data is that there are one or more elements upstream of Raf that are calcium sensitive. It may be possible that a calcium-regulated tyrosine kinase, such as Pyk2, is activated by ET and activates Raf (48). However, experiments to date examining Pyk2 in ASM cells have been negative (data not shown). Perhaps another Pyk2-like kinase is activated by ET upstream of Raf. Further work will be required to elucidate the upstream components of the ET pathway.