INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A GREAT DEAL OF INFORMATION about regulation of the extracellular receptor kinase (extracellular signal-regulated kinase, ERK) subfamily of mitogen-activated protein (MAP) kinases has been obtained using isolated cells and cells in culture. A clear link between ERK (ERK1 and ERK2) activation and cell proliferation and differentiation has been established (30). However, relatively less data has been acquired on the effects of ERK on differentiated cells, especially arterial smooth muscle.

In differentiated smooth muscle, ERK activation has been implicated in the regulation of contraction (21) and in muscle mechanotransduction (3). Tyrosine kinase inhibitors have been shown to be more selective inhibitors of agonist- than KCl-induced contractions (11, 12), suggesting that tyrosine phosphorylation plays a role in G protein-coupled receptor (GPCR)-induced contractions. GPCR stimulation is known to activate MAP kinase (25, 30, 41), and the selective inhibitor of MAP kinase kinase (MAP or ERK kinase, MEK) PD-98059 has been shown to reduce serotonin-induced contractions in rat arteries (42) and angiotensin II-induced contractions in smooth muscle cells isolated from human resistance arteries (40). Two smooth muscle contractile protein regulatory proteins are ERK substrates: the thin filament protein caldesmon (2, 6) and myosin light chain kinase (23, 28). Activation of ERK by GPCRs may facilitate contractions by enhancing myosin light chain kinase activity (28) or by relieving cross-bridge inhibition (10, 13), although some studies do not support the latter hypothesis (14, 31).

GPCR activation may not only contract smooth muscle cells but may also play other regulatory roles that modulate the contractile state. For example, ₁-adrenergic receptor activation leads to hypertrophy (45), increased DNA synthesis (9), and production of nerve growth factor (7). GPCR activation also may lead to subsequent downregulation of contractions (34). The precise role MAP kinase activation plays in processes not directly related to contraction but important in maintenance of the differentiated contractile state remains to be determined. Thus characteristics of MAP kinase activation in smooth muscle may reveal insights into vascular smooth muscle contractile homeostasis. The present study was designed to quantify the degree to which ERK is active at rest and during contraction in freshly isolated (i.e., differentiated) arterial smooth muscle. The hypotheses tested were that 1) ERK is basally active in a resting muscle that is not stimulated by a contractile stimulus or by application of muscle stretch to impose a passive load, 2) GPCR activation can produce sustained ERK activation, and 3) the degree of steady-state GPCR-stimulated ERK activation correlates with force in differentiated arterial muscle.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tissue preparation. Tissues were prepared as described previously (33). Femoral arteries from adult female New Zealand White rabbits were cleaned of adhering tissue and stored in cold (~4°C) physiological saline solution composed of (in mM) 140 NaCl, 4.7 KCl, 1.2 MgSO₄, 1.6 CaCl₂, 1.2 Na₂HPO₄, 2.0 morpholino-propanesulfonic acid (adjusted to pH 7.4 at either 0°C or 37°C as appropriate), 0.02 EDTA (to chelate trace heavy metals), and 5.6 D-glucose. High-purity (>17 M distilled and deionized) water was used throughout. In some studies, before arterial muscle rings were secured between a micrometer and isometric force transducer (model 52, Harvard Apparatus; South Natick, MA) in water-jacketed muscle chambers (Radnoti Glass Technology; Monrovia, CA), arterial medial preparations were obtained by removing the endothelium and adventitia by gently rubbing the intimal surface with a metal rod and microdissection, respectively.

Isometric force. Contractile force was measured as described previously (33). Tissues were allowed to equilibrate at 37°C for 1 h, and the muscle length at which active force was maximum (L_o) was then determined for each tissue with K⁺ as the agonist (110 mM KCl substituted isosmotically for NaCl) using an abbreviated length-tension curve (16, 33). Once tissues were stretched to L_o, no further length changes were imposed unless indicated. In some studies, tissues were neither stretched nor contracted with KCl during the equilibration period but remained at their slack length (zero preload and afterload) throughout the experiment. For each preloaded tissue, the degree of steady-state contractile force (F) produced at L_o by incubation for 5-10 min in 110 mM KCl was equal to the optimal force for muscle contraction (F_o), and subsequent contractions were calculated as F/F_o. Tissues contracted with KCl were incubated with 1 µM phentolamine to block potential -adrenergic receptor activation caused by release of norepinephrine from periarterial nerves. Phentolamine also was added to tissues stimulated with the -adrenergic receptor agonist isoproterenol. To perform noncumulative concentration-response curves, phenylephrine (PE) was added until steady-state contraction occurred (30 min).

ERK activation. The degree of ERK activation was determined by measuring the degree of phosphorylation of ERK2 using anti-active ERK antibody (14, 46). Artery rings were quick-frozen in an acetone-dry ice slurry, thawed, homogenized (33) in 1% SDS, 10% glycerol, 20 mM dithiothreitol, 25 mM Tris · HCl (pH 6.8), 5 mM EGTA, 1 mM EDTA, 50 mM NaF, 1 mM sodium orthovanadate, 20 µg/ml leupeptin, 2 µg/ml aprotinin, and 20 µg/ml (4-amidino-phenyl)-methane-sulfonyl fluoride, heated 10 min at 100°C, clarified by centrifugation at 5,000 g for 10 min, and stored at 70°C. Thawed homogenates were assayed for protein concentration (NanoOrange, Molecular Probes; Eugene, OR), and proteins were separated (SDS-PAGE) on 12% polyacrylamide gels (12 µg of protein per well) followed by Western blotting onto Immobilon-P membranes (Millipore; Bedford, MA). Active (i.e., doubly phosphorylated) ERK2 was identified using anti-active MAP kinase (ERK) antibody (Promega; Madison, WI) and detected using an horseradish peroxidase-labeled secondary antibody and enhanced chemiluminescence (ECL) and ECL film (Amersham). Quantification of visualized bands was obtained by digitizing the images using a standard camcorder connected to a frame grabber (Snappy, Play; Rancho Cordova, CA) and digital image analysis software. For each sample, only the low-molecular-weight band [phosphorylated ERK2 (phospho-ERK2)] was quantified (see example in Fig. 1). To compensate for gel-to-gel variabilities in efficiencies of Western blotting, antibody labeling, ECL reaction, and film development, a control sample (basal) was included in one lane of each gel, and band intensities from other lanes were reported as the degree of change from basal. To determine "control" ERK phosphorylation, two basal samples, each from a different femoral artery, were included in the gel, and one was normalized to the other. Occasionally, samples were labeled with ERK2 primary antibody (Santa Cruz Biotechnology; Santa Cruz, CA) to determine whether protein loading was consistently uniform across all lanes of the gel (ERK2; Fig. 1).

View larger version (38K): [in this window] [in a new window]   Fig. 1.   Example of bands of phosphorylated extracellular signal-regulated kinase 1 and 2 (phospho-ERK1 and phospho-ERK2, respectively; A) and nonphosphorylated ERK2 (B) obtained by separating proteins from tissue homogenates using SDS-PAGE (12% acrylamide) followed by Western blotting and enhanced chemiluminescence (see METHODS). Tissues were stimulated with phenylephrine (PE) or KCl as indicated in each lane. To quantify the degree of ERK2 phosphorylation produced by a stimulus, data were normalized by dividing each band intensity by the basal band intensity that was included in each gel.

Protein kinase G site-specific vasodilator-stimulated phosphoprotein phosphorylation. The degree of phosphorylation of vasodilator-stimulated phosphoprotein (VASP) at Ser²³⁹ is used to evaluate the degree of protein kinase G (PKG) activity in intact cells and tissues (32, 38). PKG site-specific VASP phosphorylation in rabbit femoral arteries was determined as described for detection of ERK phosphorylation using PKG site-specific anti-VASP(Ser²³⁹) antibody kindly supplied by Dr. Ulrich Walter (Wurzburg, Germany).

Chlorethylclonidine treatment. To inhibit _1B-adrenergic receptors, tissues were incubated for 30 min at 37°C with 10 µM of the irreversible _1B-adrenergic receptor alkylating agent chlorethylclonidine (CEC) and then washed three times for 30 min before addition of the nonselective ₁-adrenergic receptor agonist PE (17, 29).

Statistics. The null hypothesis was examined using one-way ANOVA. To determine differences between groups, the Student-Newman-Keuls post hoc test was used. When comparing two groups, the Student's t-test was used. In all cases, the null hypothesis was rejected at P < 0.05. For each study described, the n value was equal to the number of rabbits from which arteries were obtained.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Time-dependent activation of ERK. Five distinct smooth muscle stimuli were examined to determine whether they produced an increase in ERK phosphorylation early (5 min) and/or at the steady state (30 min) of contraction or stimulation. The -adrenergic receptor agonist isoproterenol (10 µM) did not increase ERK phosphorylation at 5 or 30 min (Fig. 2, C and D). As expected, isoproterenol, a relaxant, did not cause contraction (Fig. 2, A and B). PE (1 µM), an ₁-adrenergic receptor agonist and strong contractile stimulus (Fig. 2, A and B), produced a strong increase in ERK phosphorylation of over 2-fold basal at 5 min (Fig. 2C), which was sustained at ~1.5-fold basal by 30 min (Fig. 2D). Another GPCR stimulus, angiotensin II (1 µM), produced a strong contraction that declined in strength from 5 to 30 min (Fig. 2, A and B). Concomitant with this pattern, angiotensin II stimulated an increase in ERK phosphorylation at 5 min (Fig. 2C), but ERK phosphorylation returned to the control basal level by 30 min (Fig. 2D). KCl, a stimulus that bypasses receptors (contracting arteries by causing membrane depolarization), produced a sustained increase in force (Fig. 2, A and B). Moreover, the degree of force sustained at 30 min by K⁺ depolarization was significantly greater than that sustained at 30 min by 1 µM PE (Fig. 2B). However, although K⁺ depolarization produced an increase in ERK phosphorylation at 5 min (Fig. 2C), ERK phosphorylation was not elevated above the control basal level by 30 min (Fig. 2D; see also phospho-ERK2 examples in Fig. 1, comparing lanes 5 and 6). The comparison between PE and KCl indicates that the degree of force production did not necessarily correlate with the degree of ERK activation. The final stimulus was to induce, by a single step increase in length (stretch, ~1.3 L_o), a strong increase in passive muscle force of nearly 1.0-fold F_o at 5 min (Fig. 2A), which was sustained at ~0.5-fold F_o for 30 min (Fig. 2B). The decline in force from 5 to 30 min was due to the well-known phenomenon of stress-relaxation. The stretch stimulus produced a weak increase in ERK phosphorylation at 5 min (Fig. 2C), which returned to the control basal level by 30 min (Fig. 2D).

View larger version (51K): [in this window] [in a new window]   Fig. 2.   Time-dependent activation of force (A and B) and ERK (C and D). A and C: 5-min stimulation; B and D: 30-min stimulation. Data are means ± SE; n = 9 control and 3 isoproterenol (IP)-, 4 PE-, 3 ANG II-, and 8 KCl-treated arteries and 5 arteries with stretch to 1.3-fold of the length where active force is maximum (Lo). *P < 0.05 compared with control. The asterisk between the brackets indicates that KCl-induced steady-state contraction was significantly greater than that produced by PE. F/Fo, ratio of steady-state contractile force to optimal force for muscle contraction.

PE-induced 5-min concentration-dependent effect on ERK activation. To determine the concentration dependence of a PE-induced early (5 min) increase in ERK activation, tissues were stimulated with 0.03, 0.1, 1, and 10 µM PE for 5 min, and force and ERK phosphorylation were measured. At the lowest concentration examined (0.03 µM), force was significantly increased to ~13% of that produced by 10 µM PE (0.15-fold F_o; Fig. 3A). Although ERK phosphorylation was not significantly increased compared with the control basal level, a small increase in mean ERK phosphorylation of a comparable percentage (14% of that produced by 10 µM PE) was detected (Fig. 3B). PE at a concentration of 1 and 10 µM produced similar increases in force and ERK phosphorylation, and 0.1 µM PE produced intermediate increases (Fig. 3).

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Early (5-min) PE concentration-dependent effect on force (A) and ERK (B) activation. Data are means ± SE; n = 3-6 arteries. *P < 0.05 compared with control.

PE-induced, 30-min, concentration-dependent effect on ERK activation and effect of mechanical disruption and removal of endothelium and adventitia, respectively. The endothelium and adventitia contain cell types that release vasoactive compounds. In particular, the endothelium releases vasodilator compounds, and the adventitia releases norepinephrine as well as vasodilators. To determine whether disruption of the endothelium and removal of adventitia affected both steady-state force and ERK activation, whole artery and media (endothelium disrupted, adventitia removed) preparations were contracted with 0.1, 1, and 10 µM PE for 30 min, and force and ERK phosphorylation were measured. This is the only experiment in this study in which a media preparation was used. The basal level of ERK phosphorylation was significantly elevated in the media (Fig. 4B; control) compared with the whole artery (Fig. 4D; control) preparation, although basal force was not different (Fig. 4, A and C). PE at a concentration of 0.1 and 1 µM produced significantly greater increases in force in the media preparation (Fig. 4A) than in the whole artery (Fig. 4B), as was expected because of a reduction in basal vasodilator influence in the media preparation. Interestingly, an increase in ERK phosphorylation at 0.1 µM PE did not occur (Fig. 4B), despite the large increase in force of approximately sevenfold F_o in the media preparation (Fig. 4A).

View larger version (46K): [in this window] [in a new window]   Fig. 4.   Steady-state (30 min) PE concentration-dependent effect on force (A and C) and ERK activation (B and D) in media (A and B) and the whole artery (C and D). Data are means ± SE; n = 4 arteries. *P < 0.05 compared with control; P < 0.05, media control compared with the whole artery control.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Data from Fig. 4 replotted to display correlation between ERK phosphorylation and steady-state force produced by PE in the whole artery ( and solid line) compared with media ( and dashed line).

Effect of chemical denudation by 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one on ERK activation. The lower basal (control) ERK phosphorylation produced in the whole artery compared with the media preparation (compare Fig. 4, B and D) may be due to nitric oxide (NO)-induced inhibition of ERK activation. To determine whether NO released by the endothelium caused a lower basal (control) ERK phosphorylation by activation of soluble guanylyl cyclase (sGC), leading to generation of cGMP and activation of PKG, tissues were exposed to the sGC inhibitor 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) (43), and both ERK activation and VASP(Ser²³⁹) phosphorylation were measured. VASP(Ser²³⁹) phosphorylation is a measure of PKG activity (39). Under basal conditions, VASP was highly phosphorylated at Ser²³⁹, and 30 µM ODQ abolished this phosphorylation (Fig. 6, A and C). Moreover, endothelial denudation by mechanical means also dramatically reduced the level of VASP(Ser²³⁹) phosphorylation (see Fig. 6A). In the presence of ODQ, 8-bromo-cGMP, an agent that bypasses sGC to directly activate PKG, strongly increased the degree of VASP(Ser²³⁹) phosphorylation (Fig. 6, A and C). These data suggest that PKG is basally active in the endothelium-intact artery and that ODQ inhibited PKG activity. Chemical denudation by ODQ did not produce an increase in ERK activation (Fig. 6, B and D). This is unlike that seen during mechanical denudation, in which ERK activation is modestly but significantly increased (compare control values in Fig. 4, B and D, and see Fig. 6B, Endo). Moreover, 8-bromo-cGMP did not reduce the level of ERK activation but instead weakly (but significantly) increased ERK activation (Fig. 6, B and D). These data indicate that it is very unlikely that the lower basal (control) ERK phosphorylation produced in the whole artery compared with the media preparation is due to inhibition of ERK activity by a NO/PKG mechanism.

View larger version (61K): [in this window] [in a new window]   Fig. 6.   Effect of the inhibitor of soluble guanylyl cyclase 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ; 30 µM) on vasodilator-stimulated phosphoprotein [VASP(Ser239)] phosphorylation (a measure of protein kinase G activation; A and C) and ERK phosphorylation (B and D). The effect of 8-bromo-cGMP (8bcGMP) on VASP(Ser239) phosphorylation (A and C) and ERK phosphorylation (B and D) is also shown. For comparison, the effect of mechanical endothelial denudation (Endo) is included in A (n = 1). Data in A and B are representative Western blots. Data in C and D are means ± SE; n = 3-4 arteries. *P < 0.05 compared with ODQ. The dotted line indicates that the basal level is 1.

Comparison of time course of ERK activation and inactivation with contraction and relaxation. If ERK activation causes an increase in force, then ERK phosphorylation should precede force development, and ERK inactivation should precede relaxation. Contraction (Fig. 7A, solid line) induced by 10 µM PE was maximal within 1 min (Fig. 7A, open bars). However, ERK phosphorylation was not significantly elevated at 1 min but was maximally elevated by 5 min (Fig. 7A, crosshatched bars). As shown in Fig. 4, both force and ERK phosphorylation remained elevated for at least 30 min (Fig. 7A). In tissues contracted with 10 µM PE for 5 or 30 min and then washed in the presence of phentolamine to cause relaxation, complete relaxation (Fig. 7B, solid lines) occurred within 10 min of the onset of agonist washout (Fig. 7B, open bars). However, ERK phosphorylation remained elevated well above the control basal level (Fig. 7B, crosshatched bars). These data indicate that PE-induced contraction preceded ERK phosphorylation and that ERK phosphorylation remained elevated even when PE had been washed from the tissue and relaxation was complete.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   A: comparison of time course of ERK activation (cross-hatched bars) with contraction (open bars) produced by 10 µM phenylephrine. B: comparison of time course of ERK inactivation (crosshatched bars) and relaxation (open bars) produced by washout of PE in the presence of 1 µM of the -adrenergic receptor antagonist phentolamine. C: dependence of ERK phosphorylation on extracellular Ca2+ (tissues were exposed to a Ca2+-free solution containing 1 mM EGTA 10 min before and during activation with 10 µM PE). Data are means ± SE; n = 3-14 arteries. *P < 0.05 compared with control. Solid lines are representative force tracings shown as an example of the contractile responses.

Dependence of ERK phosphorylation on extracellular Ca²⁺. Tissues were incubated in a Ca²⁺-free solution (0 mM Ca²⁺ plus 1 mM EGTA to chelate Ca²⁺) for 10 min and then stimulated with 10 µM PE, and force and ERK phosphorylation were recorded at 5 and 30 min. Contractions in the Ca²⁺-free solution were transient; force developed to a high level but, after ~2 min, fell to a level <0.5-fold F_o at 5 min and reached the basal level within 30 min (Fig. 7C, solid line and open bars). ERK phosphorylation reached a level not different from that produced in a Ca²⁺-containing solution at 5 min (compare Figs. 7, A or B, with C, 5 min, cross-hatched bars). However, by 30 min, ERK phosphorylation was not different than the control basal value. Thus extracellular Ca²⁺ was required for PE-induced maintenance of ERK phosphorylation.

Effect of inhibitors on PE-induced force and ERK phosphorylation. The specific inhibitor of MEK activation PD-98059 (10 µM) (4, 14) significantly inhibited basal and 10 µM PE-induced 30-min ERK phosphorylation (Fig. 8B). Under both conditions, ERK phosphorylation was reduced by ~50%. However, PD-98059 had no effect on basal or 10 µM PE-stimulated force (Fig. 8A). The Ca²⁺/calmodulin blocker trifluoperazine (TFP) (44), an agent that inhibits Ca²⁺ entry, Ca²⁺/calmodulin-dependent myosin light chain phosphorylation, and other Ca²⁺/calmodulin-dependent cross-bridge activation processes (5, 18, 27), completely prevented PE-induced increases in force and ERK phosphorylation (Fig. 9). These data support the finding that the PE-induced increase in ERK phosphorylation depended on extracellular Ca²⁺ (see Fig. 7C) and further suggest that Ca²⁺/calmodulin plays a role. In cultured rat aortic vascular smooth muscle cells, 30 µM KN-93, a specific Ca²⁺/calmodulin kinase II inhibitor, nearly completely abolished ionomycin-induced increases in ERK phosphorylation without affecting ionomycin-induced increases in intracellular Ca²⁺ concentration ([Ca²⁺]_i) (1). In the present study using differentiated vascular smooth muscle, incubation for 30 min with 32 µM KN-93 (22) did not reduce the PE-stimulated increase in ERK phosphorylation (Fig. 9). These data support the hypothesis that PE-induced ERK phosphorylation was due, in part, to Ca²⁺/calmodulin activation of MEK and that Ca²⁺/calmodulin kinase II did not play a role.

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Effect of the specific [mitogen-activated protein or ERK (MEK)] inhibitor PD-98059 (PD) on steady-state PE-induced force (A) and ERK phosphorylation (B). Data are means ± SE; n = 3-5 arteries. *P < 0.05.

View larger version (22K): [in this window] [in a new window]   Fig. 9.   Effect of the calmodulin inhibitor trifluoperazine (TFP) and the Ca2+/calmodulin kinase II inhibitor KN-93 on 5-min PE-induced force (A) and ERK phosphorylation (B). Data are means ± SE; n = 3-7 arteries. *P < 0.05 compared with control. P < 0.05 compared with 10 µM PE.

Effect of -adrenergic receptor antagonists on force and ERK activation Phentolamine (1 µM), a nonselective -adrenergic receptor antagonist, abolished PE-induced force (Fig. 10C) and nearly abolished PE-induced ERK phosphorylation (Fig. 10D) produced at 5 min. CEC (10 µM; see METHODS), an inhibitor of _1B-adrenergic receptors (17, 29), produced no effect on force or ERK phosphorylation, whereas selective inhibition of _1A-adrenergic receptors by 0.1 µM WB-4101 (17, 29) abolished PE-induced ERK activation and reduced PE-induced force by 70% (Fig. 10).

View larger version (31K): [in this window] [in a new window]   Fig. 10.   Effect of selective -adrenergic receptor blockade by chlorethylclonidine (CEC) and WB-4101 and nonselective -adrenergic receptor blockade by phentolamine (PT) on 10 µM 5-min (5') PE-induced increases in force (C) and ERK phosphorylation (A, B, and D). Data in A and B are representative Western blots. Data in C and D are means ± SE; n = 4-5 arteries. *P < 0.05 compared with drug alone. P < 0.05 compared with 10 µM PE.

Effect of zero preload and afterload on ERK phosphorylation. The degree of basal ERK phosphorylation produced by artery rings maintained at zero passive force (zero preload at slack length) was identical to that produced by tissues that were preloaded (stretched to L_o; Fig. 11). Also, K⁺ depolarization produced the same pattern of ERK activation in slack tissues (zero load contraction) as that produced in tissues stretched to L_o and permitted to develop isometric force (afterloaded contraction; compare Fig. 2, C and D, horizontally hatched bars, with Fig. 11, vertically and horizontally hatched bars). These data indicate that preload was not required for development of basal ERK phosphorylation and suggest that the ability of stimuli to induce further ERK phosphorylation was not afterload dependent.

View larger version (22K): [in this window] [in a new window]   Fig. 11.   ERK phosphorylation in muscle at zero preload (slack length) compared with tissues preloaded by stretching to Lo. Also shown is the ERK phosphorylation produced by KCl [for 5 (5') and 30 min (30')] in muscles contracted at zero afterload (slack). Data are means ± SE; n = 3-6 arteries. *P < 0.05 compared with control.

Effects of the tyrosine phosphatase inhibitor pervanadate and hypertonic sucrose on force and ERK phosphorylation. A 30-min stimulation period with 3.2 µM pervanadate, a tyrosine phosphatase inhibitor (19), did not produce a significant elevation in force in the whole artery preparation but increased ERK phosphorylation to 2.72-fold basal, a value 138% of that produced by a 30-min stimulus with 10 µM PE (Fig. 12). These data support the contention that MEK is active in the resting artery.

View larger version (25K): [in this window] [in a new window]   Fig. 12.   Effects of the tyrosine phosphatase inhibitor pervanadate (PV) and hypertonic sucrose on force (A) and ERK phosphorylation (B). Data are means ± SE; n = 4-6 arteries. *P < 0.05 compared with control. NS, not statistically significant. The asterisk between the brackets indicates that KCl produced a weak (but statistically significant) contraction in the presence of hypertonic sucrose.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The selective MEK inhibitor PD-98059 reduced basal levels of ERK phosphorylation, and the tyrosine phosphatase inhibitor pervanadate produced a greater increase in ERK phosphorylation within 30 min than did 10 µM PE. These data indicate that MEK was active in the nonstimulated whole artery. Arterial wall stress did not appear to be the stimulus for basal MEK activity because the degree of ERK phosphorylation in preloaded muscles (stretched to L_o) and those that remained at their slack length (zero preload) were identical. Thus these data indicate that MEK is active in resting differentiated vascular smooth muscle.

The arterial media preparation, in which the endothelium was denuded mechanically, displayed a significantly elevated basal ERK phosphorylation compared with the whole artery. Thus it is possible that the degree of basal MEK activity was tonically inhibited by agents released by the endothelium. Indeed, it has been suggested that agents that elevate cyclic nucleotides exert a negative influence on ERK activation (8, 15). However, unlike mechanical denudation, chemical denudation using ODQ did not elevate ERK activity. By inhibiting sGC, ODQ prevents NO-induced activation of PKG (43). Because ODQ abolished VASP (Ser²³⁹) phosphorylation, a measure of PKG activity (32), these data suggest that mechanical denudation did not cause an elevation of ERK activity by reducing endothelial NO production. Moreover, 8-bromo-cGMP, a cell-permeable activator of PKG, did not reduce ERK activity. Rather, ERK activity was increased (see Fig. 6). These data support the recent finding that cell-permeable cGMP elevates ERK activity (24). The endothelium also releases prostaglandins and endothelium-dependent hyperpolarizing factor. Whether these factors played a role in reducing ERK activity in the whole artery remains to be determined.

Several studies (3, 10, 11, 13, 21, 40, 42) suggest that ERK phosphorylation may play a role in regulation of contraction, whereas others (14, 31) suggest it does not. In the present study, PE-induced steady-state force was strongly correlated with ERK phosphorylation in the whole artery. If an increase in -adrenergic receptor stimulation caused an increase in ERK phosphorylation, which, in turn, caused an increase in force, then perturbations that change the degree of ERK phosphorylation should result in concomitant changes in force. The present study provides evidence that this does not always occur, suggesting that ERK phosphorylation does not cause force production in the rabbit femoral artery. PE produced maximum force within 1 min, but PE-induced ERK phosphorylation was not statistically increased above the basal level at this time. Ten minutes after PE washout, a time when [Ca²⁺]_i, myosin light chain phosphorylation (35), and force have returned to their basal prestimulus levels, ERK phosphorylation remained elevated well above the basal level. In the endothelium-denuded artery, 0.1 µM PE produced a strong contraction (~0.75 F/F_o) but did not significantly elevate ERK phosphorylation. K⁺ depolarization produced sustained contractions but not sustained increases in ERK phosphorylation. The selective MEK inhibitor PD-98059 reduced PE-induced ERK phosphorylation to the basal level but did not inhibit PE-induced force. Hypertonic sucrose produced an increase in ERK phosphorylation comparable with that produced by 10 µM PE but produced only a weak increase in force. The _1A-antagonist, WB-4101 abolished PE-induced ERK phosphorylation, but 30% of PE-induced force was retained. Lastly, pervanadate (3.2 µM) produced a stronger increase in ERK phosphorylation than did 10 µM PE but did not significantly elevate force. However, these data cannot rule out the possibility that very low levels of ERK phosphorylation provide a permissive role in generation of force or that ERK phosphorylation plays a modulatory role in regulating force.

Alternatively, the correlation found between steady-state force and ERK phosphorylation may indicate that the degree of activation of the cell signaling system, ERK phosphorylation, reflects the level of -adrenergic receptor-stimulated force. That is, the cell may be continuously "notified" about the state of -adrenergic receptor-stimulated contraction by the degree of ERK activation. Such "notification" may participate in decision making related to cell homeostasis. For example, -adrenergic receptor-induced ERK activation may participate in feedback regulation to orchestrate receptor or postreceptor desensitization or changes in cell growth appropriate to the degree of receptor stimulation.

Although ₁-adrenergic receptor stimulation produced sustained increases in ERK phosphorylation, this was not true for all other stimuli examined, including K⁺ depolarization, muscle stretched passively (preloaded) to ~1.3-fold L_o, or angiotensin II. The complete inhibition of 10 µM PE-induced steady-state force by the calmodulin antagonist TFP (50 µM) and by a Ca²⁺-free solution indicates that an increase in [Ca²⁺]_i was necessary for the sustained ERK phosphorylation produced by PE. Interestingly, however, K⁺ depolarization produces greater increases in [Ca²⁺]_i than does PE for a given level of force (20), and, although K⁺ depolarization produced an increase in ERK phosphorylation at 5 min, it did not sustain an increase by 30 min. These data support the hypothesis that an increase in [Ca²⁺]_i may be necessary, but not sufficient, for maintenance of elevated ERK activation in whole arterial muscle. These data imply that ₁-adrenergic receptor stimulation activates a second cell signaling system that, along with increased [Ca²⁺]_i, produced sustained increases in ERK activation.

Lefkowitz and colleagues (reviewed in Ref. 25) proposed that internalized GPCRs may activate ERK. In rat 1a fibroblast cells, hypertonic sucrose reduces receptor internalization and receptor-induced activation of ERK (26). In the present study, hypertonic sucrose alone produced a strong increase in ERK phosphorylation (~2-fold basal), a weak increase in force (~0.2 F/F_o), and completely blocked the ability of 10 µM PE to produce additional ERK activation or contraction. The fact that pervanadate induced a greater increase in ERK phosphorylation than did hypertonic sucrose (see Fig. 12) suggests that the degree of ERK activation by the hypertonic solution was not maximal (i.e., additional activation by PE may have been possible). A recent study (17) indicates that _1A-adrenergic receptors may reside largely within cells, whereas _1B-adrenergic receptors reside on the cell surface. CEC, a cell surface _1B-adrenergic receptor alkylating agent, reduced neither PE-induced ERK activity nor force, whereas WB-4101, an _1A-adrenergic receptor antagonist, abolished PE-induced ERK activity and reduced but did not abolish force. These data are consistent with the hypothesis that internalized _1A-receptors caused ERK activation in the rabbit femoral artery.

In summary, in differentiated arterial muscle, MEK was basally active, and ₁-adrenergic receptor activation produced a concentration-dependent increase in ERK phosphorylation. Although this stimulus was not unique in its ability to activate ERK in this muscle, it was unique among the stimuli examined in its ability to produce a sustained increase in ERK phosphorylation lasting at least 30 min. An increase in [Ca²⁺]_i appeared to be necessary but not sufficient for ₁-adrenergic receptor-induced sustained ERK phosphorylation, because K⁺ depolarization, a stimulus known to produce sustained increases in [Ca²⁺]_i (36, 37), did not produce sustained increases in ERK phosphorylation. Moreover, a sustained increase in preload (via stretching the muscle) or afterload (via K⁺ depolarization) did not produce a sustained increase in ERK phosphorylation, suggesting that ₁-adrenergic receptor-induced sustained ERK activation was not due to mechanical activation of focal adhesions alone. Although results from this study do not support the hypothesis that stimulus-induced increases in ERK phosphorylation cause concomitant increases in contraction, a correlation between ₁-adrenergic receptor-induced steady-state force and ERK phosphorylation supports the hypothesis that ERK plays a role in transmission of information to the cell about the degree of ₁-adrenergic receptor-induced contraction in differentiated arterial muscle.